Knock-for-Knock Indemnities and the Law: Contractual Limitation and Delictual Liability 9781032074085, 9781032074115, 9781003206798

Table of contents :
Cover
Half Title
Series
Title
Copyright
Contents
Detailed Contents
List of contributors
Editors’ preface
Table of cases
Table of statutes and Model Form Contracts
Part I Selected Topics
Chapter 1 The Development of Knock-For-Knock Clauses in the Last 15 Years
1 Introduction
1.1 Definition
1.2 Essential meaning of a K4K clause
1.3 Historical genesis of the K4K clause
2 Issues in using, implementing and enforcing K4K clauses
3 Various uses
3.1 SUPPLYTIME 2005
3.2 Insurance and P&I clubs
3.3 Use in offshore contracts
4 Modern use
4.1 The BIMCO SUPPLYTIME 2017
5 A historical encounter of the K4K clauses in the shipping and in the offshore oil and gas sectors
6 The evolution of K4K clauses in standard contracts and in English case law
6.1 The significance of the Piper Alpha litigation
7 The contribution of other jurisprudence to the historical evolution of K4K clauses
7.1 US jurisprudence
7.2 Deepwater Horizon
7.3 Brazil
7.4 Germany
7.5 Australia
7.6 Nordic countries – the history and validity of K4K clauses from a comparative perspective
7.7 Later judicial trends
7.8 The way forward
8 Conclusive critique
Chapter 2 An Introduction to Risk Allocation in Oil and Gas Contracts From an English Law Perspective
1 Introduction
2 Indemnification and related concepts
2.1 Introduction to the concept of indemnification
2.2 Indemnity and hold harmless clauses
2.3 Mutual indemnity and mutual indemnity and hold harmless clauses
3 Indemnity and hold harmless provisions in the oil and gas context
3.1 Introduction to simple indemnity and hold harmless clauses in oil and gas contracts
3.2 Introduction to mutual indemnity and hold harmless clauses in oil and gas contracts
3.3 The rationale for a mutual indemnity and hold harmless regime in the oil and gas context
3.4 Back-to-back indemnity and hold harmless provisions in oil and gas contracts
3.5 Qualified indemnity and hold harmless provisions
4 Selected further issues in indemnification law and practice in the UKCS
4.1 Statutory control of indemnity and hold harmless clauses
4.2 The law’s default settings: the normal presumptions about the distribution of risk
4.3 The position of third parties
5 Interpreting indemnity and hold harmless clauses
5.1 The traditional approach to interpretation
The general rules
Interpretation contra proferentem
Towards contextualism (and back again)
6 Some known problems of drafting and interpretation
6.1 Contra proferentem and the problem of negligence and breach of statutory duty
6.2 Words delimiting the circumstances in which the indemnity and hold harmless provision will take effect
6.3 Multi-party issues
6.4 “Full and primary”
6.5 Definitional issues
Company groups
Employees and personnel
Property
7 The problem of multiple parties
7.1 Introduction to the problem
7.2 The Industry Mutual Hold Harmless Agreement (IMHH)
The Deed’s core indemnity provisions
Other important provisions
Entry into force
Geographical extent
Order of precedence
Extension of benefits to groups
Waiver of rights of subrogation
Right to defend
Exceptions
Transportation by air excluded
Landward areas
Operators
Commentary and conclusions on the IMHH scheme
The contractual nature of the scheme
Potential dangers of the IMHH
8 Liability for “consequential” loss
8.1 What is “consequential loss”?
8.2 Commentary and conclusions on consequential loss
9 Overall limitation of liability
Chapter 3 Contracting Around Tort Defaults: The Knock-For-Knock Principle and Accident Costs
1 Introduction
2 The knock-for-knock principle
2.1 The industrial setting
2.2 Illustrating the operation of knock for knock
3 Knock for knock, social norms, and private ordering
4 The effect of knock for knock on social welfare
4.1 The perfect separating equilibrium
4.2 The perfect pooling equilibrium
4.2.1 Self-regulation
4.2.2 Repeat interactions
4.2.3 Health and safety regulation
4.2.4 Insurance
5 The hidden perils of knock for knock
5.1 Risk interdependencies
5.2 Litigation externalities
5.3 Harm to environmental interests
5.4 Deterrence versus compensation
6 Conclusion
Chapter 4 On Knock-For-Knock Clauses and Their Optimal Regulation
1 Introduction
2 On the literature
3 When are knock-for-knock clauses useful?
3.1 A model of the use of knock-for-knock contracts
4 Should courts impose carve-outs for gross negligence?
4.1 A framework
5 Conclusion
Chapter 5 Knock-For-Knock Indemnity Provisions and Liability Insurance: Potentially Strange but Always Complicated Bedfellows
1 Introduction
2 K4K provisions and general liability insurance
3 Impact of blanket AI endorsements
4 “Other insurance” clauses
5 Incorporation of indemnity limitations in insurance policies
6 Waiver of subrogation
7 Conclusion
Chapter 6 The Effect of Choice of Law on Knock-For-Knock Clauses
1 Introduction
2 Knock for knock and the full English legal treatment
2.1 The knock-for-knock clause itself
2.2 Simple knock for knock in contractual context
2.3 Knock for knock in its English legal context
3 Compare the United States
3.1 The jurisdictional issue
3.2 Federal maritime law – Deepwater Horizon
3.3 The anti-indemnity statutes
3.3.1 General observation
3.3.2 Louisiana Oilfield Indemnity Act and other state provisions
3.3.3 Texas Oilfield Anti-Indemnity Act and other state provisions
4 Compare Brazil
4.1 General observations
4.2 Knock-for-knock and Brazilian law
5 Conclusion
Part II Knock For Knock in Specific Jurisdictions
Chapter 7 Indemnity Clauses in Fabrication and Construction Contracts in Norway
Chapter 7A Limiting and Channeling Liability Under Offshore Construction Contracts in Norway
1 The problem
2 The contractor’s liability for breach under offshore construction contracts
2.1 The liability
2.1.1 The need for a balanced regime
2.1.2 Types of sanctions against breach
2.1.3 Specific features
2.1.4 Breach of secondary obligations
2.1.5 Obligation to notify – and the role of the variation mechanism
2.2 The limitation of liability
3 The knock-for-knock system
3.1 The call for regulation
3.1.1 The risks of damage
3.1.2 Consequences of no regulation
3.1.3 Potential regulation
3.1.4 The insurance aspects
3.1.5 A global view of risk, liability and insurance
3.2 Provisions on liability
3.2.1 Overview
3.2.2 The family zones
3.2.3 The third-party zone
3.3 Insurance
3.3.1 Considerations
3.3.2 Main components
3.3.3 Insurance arrangements
4 Setting aside provisions limiting exposure resulting from default or damage?
4.1 Authority for censoring provisions limiting exposure?
4.2 Differences and similarities between the two types of liability situations
4.2.1 Liability situations
4.2.2 Liability regulations
5 Are the agreed limitations of liability for breach of contract valid?
5.1 Points of departure
5.2 The contractor’s own gross negligence
5.3 Who is “the contractor”?
5.4 The role of the concept “willful misconduct”
6 Censoring the knock-for-knock regime?
6.1 The question
6.2 “Organized financing of loss”
6.3 Can the waiver of subrogation be maintained?
6.3.1 Waiver of subrogation
6.3.2 The co-insured
6.3.3 Are the arrangements “reasonable”?
6.3.4 The conclusion
7 Do the NTK 15 provisions on exclusion and limitation of liability hold good?
Chapter 7B The Obligation Under Norwegian Offshore Contracts to Insure the Contract Object as Part of a Knock-For-Knock Agreement
1 Introduction and background
1.1 Indemnification, liability and insurance
1.2 Project all risk insurances
2 The knock-for-knock arrangement
2.1 Risk zones
2.2 Waivers of liability, indemnification and waiver of subrogation
2.3 Damage to the contract object
2.4 Waivers and indemnities shall apply regardless of cause
3 The offshore project insurances
3.1 Introduction
3.2 Insured perils
3.3 Basis for recovery
4 Summary
Chapter 8 Liability and Insurance Clauses in Contracts for Vessel Services in the Norwegian Offshore Sector: The Knock-For-Knock Principle
1 Introduction
2 Overview of the relevant tort law and insurance legislation
2.1 Tort law
2.2 Insurance law
3 The content and structure of the knock-for-knock principle
3.1 Type of loss and basis for liability
3.2 Who is included in the liability provisions – “the group concept”
3.3 Freedom of liability, indemnity and subrogation
3.4 The insurance regulation
4 The rationale for the knock-for-knock principle
4.1 The need for contractual control of the liability risk
4.2 Efficient insurance coverage
4.3 Loss prevention?
4.3.1 Loss prevention and efficient liability rules
4.3.2 The efficient basis for liability in contractual relationships
5 The validity of the regulation
5.1 Some starting points
5.2 NL 5–1–2
5.3 The Contract Act § 36
5.3.1 Overview and some starting points
5.3.2 The content of the agreement
5.3.3 The insurance clauses
Chapter 9 Statutory Liability Regulation Versus Contractual Risk Allocation in Upstream Oil and Gas: The Norwegian Case
1 Introduction
2 Statutory regulation of risk under Chapter 7 of the Norwegian Petroleum Act
2.1 Introduction
2.2 Liability for petroleum pollution damage
2.2.1 Unlimited no-fault liability
2.2.2 Pollution damage
2.2.3 Channelling liability to the licensee and licensee’s recourse
3 Contractual allocation of risk on oil and gas contracts in Norway
3.1 Introduction
3.2 Industry-negotiated model contracts
3.3 The risk allocation clause
4 Contractual allocation of risk breaches with Chapter 7
Chapter 10 Applying Knock-For-Knock in Germany
1 Introduction
2 Use of knock-for-knock clauses in the German market
3 Validity issues under German law
3.1 Court review of general terms and conditions
3.1.1 General principle
3.1.1.1 Definition of general terms and conditions
3.1.1.2 Content test
3.1.1.3 What is “individually negotiated”?
Negotiation of terms demands more than debating
Additional security by a “caveat” clause?
3.2 Are knock-for-knock clauses general terms and conditions?
3.3 Applying the reasonableness test
3.3.1 Commercial contracts versus consumer contracts
3.3.2 Fairness test – what is market practice?
3.3.3 The “risky” terms
3.4 Invalid terms under German civil law – no exclusion of liability for intent
3.4.1 General principle of liability under sec. 276 BGB
3.4.2 No agreements to the detriment of third parties
3.4.3 Consequences for knock-for-knock clauses under German law
4 Available case law
5 Practical consequences and advice for use of knock-for-knock clauses under German law
Chapter 11 Knock for Knock Under Brazilian Law
1 Introduction
2 Overview of civil liability regime in Brazil
2.1 Contractual liability and tort liability
2.2 Fault-based liability and strict liability
2.3 Indemnifiable losses
3 Limitation of liability clause
4 Knock-for-knock clauses in Brazil
5 Conclusion
Chapter 12 The Implications of Litigation Post Deepwater Horizon on Knock-For-Knock Clauses in Us Law
1 Introduction
2 Applicable law
2.1 Federal court jurisdiction
2.1.1 Federal question jurisdiction
2.1.2 Diversity jurisdiction
2.1.3 Admiralty jurisdiction
2.2 Applicable law
2.2.1 Outer Continental Shelf Lands Act
2.2.2 State waters
2.2.3 Maritime law
2.2.3.1 Maritime contracts
2.2.3.2 What is a vessel?
2.2.4 Maritime law incomplete, supplemented with state law
2.2.5 Tension between OCSLA and maritime law
3 Indemnity in maritime
3.1 Enforceability of indemnity provisions in a maritime contract
3.2 Indemnity under the Longshore and Harbor Workers’ Compensation Act
3.3 Insurance obligations and additional assured status
4 Application of state law
4.1 The Louisiana Oilfield Indemnity Act
4.2 Texas Oilfield Anti-Indemnity Act
5 Conclusion
Appendix A
Chapter 13 Mutual Hold Harmless Clauses in France and Francophone Civil Law Systems
1 Introduction
1.1 Francophone civil law systems
1.1.1 The French Civil Code of 1804
1.1.1.1 French law of obligations: contract law and civil liability
1.1.1.2 Francophone African legal systems
1.1.2 Recent developments
1.1.2.1 French law
1.1.2.2 Francophone African legal systems
1.2 Mutual hold harmless clauses: a contractual re-allocation of liability
1.2.1 Origins and growing prevalence in French law
1.2.2 A note on English terminology
2 The legal nature of mutual hold harmless clauses under French law
2.1 The legal framework for mutual hold harmless clauses
2.1.1 Principles of the law on civil liability under French law
2.1.2 Contractual mechanisms for allocating risk between the parties
2.1.3 Contractual mechanisms for allocating third-party liability
2.2 Legal characterisation of mutual hold harmless clauses under French law
2.2.1 Allocation of risk between the parties
2.2.2 Allocation of third-party liability
2.2.2.1 Related third parties
2.2.2.2 Unrelated third parties
3 The legal regime governing mutual hold harmless clauses under French law
3.1 Exclusions of liability and waivers of recourse
3.1.1 Validity and exceptions
3.1.1.1 Exceptions linked to the law of obligations
3.1.1.1.1 Limiting non-contractual liability
3.1.1.1.2 Essential obligations
3.1.1.2 Specific legislative exceptions
3.1.2 Effects of waiver of recourse and exclusions of liability clauses under French law
3.1.2.1 Strict interpretation
3.1.2.2 Exceptions linked to the nature of the fault
3.1.2.3 Exceptions linked to the nature of the loss
3.2 Indemnification of third-party losses
3.2.1 Validity of clauses indemnifying third-party losses
3.2.2 Limitations
Chapter 14 Lessons from the Application of Knock-For-Knock Clauses Under Malaysian Law
1 Introduction
1.1 General overview of the Malaysian legal system
1.2 The law of contract in Malaysia
1.2.1 Statutory and judicial approach on indemnity clauses in Malaysia
1.3 Empirical study of oilfield service contracts in Malaysia
1.4 Research design
1.4.1 Research process
2 Description of the case studies
2.1 Findings
2.2 Perception of the contractual formation process
2.3 Perception of risk allocation and indemnity clauses
3 Analysis of the indemnity and hold harmless clauses of Operators A, B and C
3.1 Liability regarding personal injury or death of employees and loss of property of the parties
3.1.1 Under LOGIC
3.1.2 Under FIDIC
3.1.3 Under Operator A’s contract
3.1.4 Under Operator B’s contract
3.1.5 Under Operator C’s contract
3.2 Claims by a third party
3.2.1 Third-party claims under LOGIC
3.2.2 Third-party claims under FIDIC
3.2.3 Third-party claims under Operator A’s contract
3.2.4 Third-party claims under Operator B’s contract
3.2.5 Third-party claims under Operator C’s contract
3.3 Liability regarding pollution
3.3.1 Pollution liability under LOGIC
3.3.2 Pollution liability under FIDIC
3.3.3 Pollution liability under Operator A’s contract
3.3.4 Pollution liability under Operator B’s contract
3.3.5 Pollution liability under Operator C’s contract
3.4 Insurance coverage
4 Conclusion
Index

Citation preview

KNOC K-F OR-KNOC K IN D EMN ITIES AND T HE LAW

CONTEMPORARY COMMERCIAL LAW Maritime Law in China: Emerging Issues and Future Developments Edited by Johanna Hjalmarsson and Jenny Jingbo Zhang Illegality in Marine Insurance Law Feng Wang Insurance Law Implications of Delay in Maritime Transport Ayşegül Buğra Online Arbitration Faye Fangfei Wang Double Insurance and Contribution Nisha Mohamed The Law and Autonomous Vehicles Matthew Channon, Lucy McCormick and Kyriaki Noussia FIDIC Yellow Book: A Commentary Ben Beaumont The Contract of Carriage: Multimodal Transport and Unimodal Regulation Paula Bäckdén FIDIC Red Book: A Commentary Ben Beaumont

Blockchain Technology and the Law: Opportunities and Risks Muharem Kianief Third Party Protection in Shipping Carlo Corcione Multi-sided Music Platforms and the Law: Copyright, Law and Policy in Africa Chijioke Ifeoma Okorie Reinsurance and the Law of Aggregation: Event, Occurrence, Catastrophe, Cause Oliver D. William Comparative Analysis of Interim Measures: Interim Remedies (England & Wales) v Preservation Measures (China) Vivek Jain, Thomas Macey-Dare QC and Shengnan Jia Knock-for-Knock Indemnities and the Law: Contractual Limitation and Delictual Liability Edited by Kristofer Svendsen, Endre Stavang and Greg Gordon Managing International Trade Risk: Customs, Revenue and VAT Compliance Mark Rowbotham

For more information about this series, please visit: www.routledge.com/Contemporary-Commercial-Law/book-series/CCL

K N OCK - F OR -KNO C K INDE M N IT IE S AN D T HE L AW C ON T RAC T UAL LI MI TATI O N A N D D E L I C T UAL LI A BI LI TY EDITED BY K R I S T O F F E R S V E N D S E N, E N D R E S TAVA N G A N D G R E G G O R D O N

Cover credit: © Panuwat Dangsungnoen / Getty Images First published by Informa Law from 2023 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Informa Law from Routledge 605 Third Avenue, New York, NY 10158 Informa Law from Routledge is an imprint of the Taylor & Francis Group, an informa business © 2023 selection and editorial matter, Kristofer Svendsen, Endre Stavang and Greg Gordon; individual chapters, the contributors The right of Kristofer Svendsen, Endre Stavang and Greg Gordon to be identifed as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identifcation and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-032-07408-5 (hbk) ISBN: 978-1-032-07411-5 (pbk) ISBN: 978-1-003-20679-8 (ebk) DOI: 10.4324/9781003206798 Typeset in Times New Roman by Apex CoVantage, LLC

C ON T E N TS

List of contributors Editors’ preface Table of cases Table of statutes and Model Form Contracts PART I CHAPTER 1

CHAPTER 2

CHAPTER 3

CHAPTER 4

CHAPTER 5

CHAPTER 6

xix xxi xxv xxxiii

SELECTED TOPICS

THE DEVELOPMENT OF KNOCK-FOR-KNOCK CLAUSES IN THE LAST 15 YEARS Kyriaki Noussia and Hanieh Bolourian AN INTRODUCTION TO RISK ALLOCATION IN OIL AND GAS CONTRACTS FROM AN ENGLISH LAW PERSPECTIVE Greg Gordon CONTRACTING AROUND TORT DEFAULTS: THE KNOCK-FOR-KNOCK PRINCIPLE AND ACCIDENT COSTS Gideon Parchomovsky and Endre Stavang ON KNOCK-FOR-KNOCK CLAUSES AND THEIR OPTIMAL REGULATION Henrik Lando KNOCK-FOR-KNOCK INDEMNITY PROVISIONS AND LIABILITY INSURANCE: POTENTIALLY STRANGE BUT ALWAYS COMPLICATED BEDFELLOWS Jay R. Sever and Lauren E. Burk THE EFFECT OF CHOICE OF LAW ON KNOCK-FOR-KNOCK CLAUSES Uisdean Vass v

1

20

65

87

98

106

contents PART II CHAPTER 7

KNOCK FOR KNOCK IN SPECIFIC JURISDICTIONS INDEMNITY CLAUSES IN FABRICATION AND CONSTRUCTION CONTRACTS IN NORWAY

CHAPTER 7A LIMITING AND CHANNELING LIABILITY UNDER OFFSHORE CONSTRUCTION CONTRACTS IN NORWAY Knut Kaasen CHAPTER 7B THE OBLIGATION UNDER NORWEGIAN OFFSHORE CONTRACTS TO INSURE THE CONTRACT OBJECT AS PART OF A KNOCK-FOR-KNOCK AGREEMENT Erik Brannsten CHAPTER 8

CHAPTER 9

LIABILITY AND INSURANCE CLAUSES IN CONTRACTS FOR VESSEL SERVICES IN THE NORWEGIAN OFFSHORE SECTOR: THE KNOCK-FOR-KNOCK PRINCIPLE Trine-Lise Wilhelmsen STATUTORY LIABILITY REGULATION VERSUS CONTRACTUAL RISK ALLOCATION IN UPSTREAM OIL AND GAS: THE NORWEGIAN CASE Kristofer Svendsen

117

118

153

164

182

CHAPTER 10 APPLYING KNOCK-FOR-KNOCK IN GERMANY Eckehard Volz and Anna-Sophie Waldmann

193

CHAPTER 11 KNOCK FOR KNOCK UNDER BRAZILIAN LAW Felipe T. Boechem and Márcio Opromolla

204

CHAPTER 12 THE IMPLICATIONS OF LITIGATION POST DEEPWATER HORIZON ON KNOCK-FOR-KNOCK CLAUSES IN US LAW Cindy Matherne Muller, Jeanne Amy, and Jennifer David CHAPTER 13 MUTUAL HOLD HARMLESS CLAUSES IN FRANCE AND FRANCOPHONE CIVIL LAW SYSTEMS Bertrand Montembault and Paul Morton

vi

214

238

contents CHAPTER 14 LESSONS FROM THE APPLICATION OF KNOCK-FOR-KNOCK CLAUSES UNDER MALAYSIAN LAW Wan Mohd Zulhafz Wan Zahari

251

291

Index

vii

D E TAI L E D C O N TEN TS

List of contributors Editors’ preface Table of cases Table of statutes and Model Form Contracts PART 1 CHAPTER 1

1

2 3

4 5 6 7

xix xxi xxv xxxiii

SELECTED TOPICS

THE DEVELOPMENT OF KNOCK-FOR-KNOCK CLAUSES IN THE LAST 15 YEARS Kyriaki Noussia and Hanieh Bolourian

Introduction 1.1 Defnition 1.2 Essential meaning of a K4K clause 1.3 Historical genesis of the K4K clause Issues in using, implementing and enforcing K4K clauses Various uses 3.1 SUPPLYTIME 2005 3.2 Insurance and P&I clubs 3.3 Use in ofshore contracts Modern use 4.1 The BIMCO SUPPLYTIME 2017 A historical encounter of the K4K clauses in the shipping and in the ofshore oil and gas sectors The evolution of K4K clauses in standard contracts and in English case law 6.1 The signifcance of the Piper Alpha litigation The contribution of other jurisprudence to the historical evolution of K4K clauses 7.1 US jurisprudence 7.2 Deepwater Horizon 7.3 Brazil 7.4 Germany 7.5 Australia 7.6 Nordic countries – the history and validity of K4K clauses from a comparative perspective ix

1 1 1 1 2 3 3 3 4 4 5 5 6 7 9 10 10 11 11 12 13 13

detailed contents 7.7 Later judicial trends 7.8 The way forward 8 Conclusive critique CHAPTER 2

15 16 17

AN INTRODUCTION TO RISK ALLOCATION IN OIL AND GAS CONTRACTS FROM AN ENGLISH LAW PERSPECTIVE Greg Gordon

1 2

Introduction Indemnifcation and related concepts 2.1 Introduction to the concept of indemnifcation 2.2 Indemnity and hold harmless clauses 2.3 Mutual indemnity and mutual indemnity and hold harmless clauses 3 Indemnity and hold harmless provisions in the oil and gas context 3.1 Introduction to simple indemnity and hold harmless clauses in oil and gas contracts 3.2 Introduction to mutual indemnity and hold harmless clauses in oil and gas contracts 3.3 The rationale for a mutual indemnity and hold harmless regime in the oil and gas context 3.4 Back-to-back indemnity and hold harmless provisions in oil and gas contracts 3.5 Qualifed indemnity and hold harmless provisions 4 Selected further issues in indemnifcation law and practice in the UKCS 4.1 Statutory control of indemnity and hold harmless clauses 4.2 The law’s default settings: the normal presumptions about the distribution of risk 4.3 The position of third parties 5 Interpreting indemnity and hold harmless clauses 5.1 The traditional approach to interpretation The general rules Interpretation contra proferentem Towards contextualism (and back again) 6 Some known problems of drafting and interpretation 6.1 Contra proferentem and the problem of negligence and breach of statutory duty 6.2 Words delimiting the circumstances in which the indemnity and hold harmless provision will take efect 6.3 Multi-party issues 6.4 “Full and primary” 6.5 Defnitional issues Company groups Employees and personnel Property x

20 20 21 21 22 24 25 25 26 28 30 31 34 34 35 36 37 37 37 38 39 40 40 46 47 48 49 49 50 50

detailed contents 7

8 9

The problem of multiple parties 7.1 Introduction to the problem 7.2 The Industry Mutual Hold Harmless Agreement (IMHH) The Deed’s core indemnity provisions Other important provisions Entry into force Geographical extent Order of precedence Extension of benefts to groups Waiver of rights of subrogation Right to defend Exceptions Transportation by air excluded Landward areas Operators Commentary and conclusions on the IMHH scheme The contractual nature of the scheme Potential dangers of the IMHH Liability for “consequential” loss 8.1 What is “consequential loss”? 8.2 Commentary and conclusions on consequential loss Overall limitation of liability

CHAPTER 3

1 2 3 4

5

6

CONTRACTING AROUND TORT DEFAULTS: THE KNOCK-FOR-KNOCK PRINCIPLE AND ACCIDENT COSTS Gideon Parchomovsky and Endre Stavang

Introduction The knock-for-knock principle 2.1 The industrial setting 2.2 Illustrating the operation of knock for knock Knock for knock, social norms, and private ordering The efect of knock for knock on social welfare 4.1 The perfect separating equilibrium 4.2 The perfect pooling equilibrium 4.2.1 Self-regulation 4.2.2 Repeat interactions 4.2.3 Health and safety regulation 4.2.4 Insurance The hidden perils of knock for knock 5.1 Risk interdependencies 5.2 Litigation externalities 5.3 Harm to environmental interests 5.4 Deterrence versus compensation Conclusion xi

50 50 52 53 54 54 55 55 55 55 56 56 56 57 57 57 57 58 59 59 61 62

65 65 68 69 69 71 73 74 74 75 76 77 78 79 80 80 82 84 85

detailed contents CHAPTER 4

1 2 3 4 5

Introduction On the literature When are knock-for-knock clauses useful? 3.1 A model of the use of knock-for-knock contracts Should courts impose carve-outs for gross negligence? 4.1 A framework Conclusion

CHAPTER 5

1 2 3 4 5 6 7

3

4 5

KNOCK-FOR-KNOCK INDEMNITY PROVISIONS AND LIABILITY INSURANCE: POTENTIALLY STRANGE BUT ALWAYS COMPLICATED BEDFELLOWS Jay R. Sever and Lauren E. Burk

Introduction K4K provisions and general liability insurance Impact of blanket AI endorsements “Other insurance” clauses Incorporation of indemnity limitations in insurance policies Waiver of subrogation Conclusion

CHAPTER 6

1 2

ON KNOCK-FOR-KNOCK CLAUSES AND THEIR OPTIMAL REGULATION Henrik Lando

THE EFFECT OF CHOICE OF LAW ON KNOCK-FOR-KNOCK CLAUSES Uisdean Vass

Introduction Knock for knock and the full English legal treatment 2.1 The knock-for-knock clause itself 2.2 Simple knock for knock in contractual context 2.3 Knock for knock in its English legal context Compare the United States 3.1 The jurisdictional issue 3.2 Federal maritime law – Deepwater Horizon 3.3 The anti-indemnity statutes 3.3.1 General observation 3.3.2 Louisiana Oilfeld Indemnity Act and other state provisions 3.3.3 Texas Oilfeld Anti-Indemnity Act and other state provisions Compare Brazil 4.1 General observations 4.2 Knock-for-knock and Brazilian law Conclusion xii

87 87 88 89 90 92 94 97

98 98 99 100 102 103 104 105

106 106 107 107 108 109 111 111 111 113 113 113 114 114 114 115 116

detailed contents PART II CHAPTER 7

KNOCK FOR KNOCK IN SPECIFIC JURISDICTIONS INDEMNITY CLAUSES IN FABRICATION AND CONSTRUCTION CONTRACTS IN NORWAY

CHAPTER 7A LIMITING AND CHANNELING LIABILITY UNDER OFFSHORE CONSTRUCTION CONTRACTS IN NORWAY Knut Kaasen 1 The problem 2 The contractor’s liability for breach under ofshore construction contracts 2.1 The liability 2.1.1 The need for a balanced regime 2.1.2 Types of sanctions against breach 2.1.3 Specifc features 2.1.4 Breach of secondary obligations 2.1.5 Obligation to notify – and the role of the variation mechanism 2.2 The limitation of liability 3 The knock-for-knock system 3.1 The call for regulation 3.1.1 The risks of damage 3.1.2 Consequences of no regulation 3.1.3 Potential regulation 3.1.4 The insurance aspects 3.1.5 A global view of risk, liability and insurance 3.2 Provisions on liability 3.2.1 Overview 3.2.2 The family zones 3.2.3 The third-party zone 3.3 Insurance 3.3.1 Considerations 3.3.2 Main components 3.3.3 Insurance arrangements 4 Setting aside provisions limiting exposure resulting from default or damage? 4.1 Authority for censoring provisions limiting exposure? 4.2 Diferences and similarities between the two types of liability situations 4.2.1 Liability situations 4.2.2 Liability regulations 5 Are the agreed limitations of liability for breach of contract valid? 5.1 Points of departure 5.2 The contractor’s own gross negligence 5.3 Who is “the contractor”? 5.4 The role of the concept “willful misconduct” xiii

117

118 118 120 120 120 121 122 123 123 124 125 125 125 126 126 127 128 128 128 130 131 131 131 132 133 134 134 135 135 135 137 137 138 141 143

detailed contents 6

Censoring the knock-for-knock regime? 6.1 The question 6.2 “Organized fnancing of loss” 6.3 Can the waiver of subrogation be maintained? 6.3.1 Waiver of subrogation 6.3.2 The co-insured 6.3.3 Are the arrangements “reasonable”? 6.3.4 The conclusion 7 Do the NTK 15 provisions on exclusion and limitation of liability hold good? CHAPTER 7B THE OBLIGATION UNDER NORWEGIAN OFFSHORE CONTRACTS TO INSURE THE CONTRACT OBJECT AS PART OF A KNOCK-FOR-KNOCK AGREEMENT Erik Brannsten 1 Introduction and background 1.1 Indemnifcation, liability and insurance 1.2 Project all risk insurances 2 The knock-for-knock arrangement 2.1 Risk zones 2.2 Waivers of liability, indemnifcation and waiver of subrogation 2.3 Damage to the contract object 2.4 Waivers and indemnities shall apply regardless of cause 3 The ofshore project insurances 3.1 Introduction 3.2 Insured perils 3.3 Basis for recovery 4 Summary CHAPTER 8

LIABILITY AND INSURANCE CLAUSES IN CONTRACTS FOR VESSEL SERVICES IN THE NORWEGIAN OFFSHORE SECTOR: THE KNOCK-FOR-KNOCK PRINCIPLE Trine-Lise Wilhelmsen

1 2

Introduction Overview of the relevant tort law and insurance legislation 2.1 Tort law 2.2 Insurance law 3 The content and structure of the knock-for-knock principle 3.1 Type of loss and basis for liability 3.2 Who is included in the liability provisions – “the group concept” 3.3 Freedom of liability, indemnity and subrogation 3.4 The insurance regulation xiv

145 145 147 148 148 148 149 150 150

153 153 154 155 156 156 156 156 158 158 158 160 162 163

164 164 165 165 166 166 166 167 169 170

detailed contents 4

The rationale for the knock-for-knock principle 4.1 The need for contractual control of the liability risk 4.2 Efcient insurance coverage 4.3 Loss prevention? 4.3.1 Loss prevention and efcient liability rules 4.3.2 The efcient basis for liability in contractual relationships 5 The validity of the regulation 5.1 Some starting points 5.2 NL 5–1–2 5.3 The Contract Act § 36 5.3.1 Overview and some starting points 5.3.2 The content of the agreement 5.3.3 The insurance clauses CHAPTER 9

STATUTORY LIABILITY REGULATION VERSUS CONTRACTUAL RISK ALLOCATION IN UPSTREAM OIL AND GAS: THE NORWEGIAN CASE Kristofer Svendsen

1 2

171 171 172 173 173 174 175 175 176 178 178 179 181

182

Introduction Statutory regulation of risk under Chapter 7 of the Norwegian Petroleum Act 2.1 Introduction 2.2 Liability for petroleum pollution damage 2.2.1 Unlimited no-fault liability 2.2.2 Pollution damage 2.2.3 Channelling liability to the licensee and licensee’s recourse 3 Contractual allocation of risk on oil and gas contracts in Norway 3.1 Introduction 3.2 Industry-negotiated model contracts 3.3 The risk allocation clause 4 Contractual allocation of risk breaches with Chapter 7

182

CHAPTER 10 APPLYING KNOCK-FOR-KNOCK IN GERMANY Eckehard Volz and Anna-Sophie Waldmann 1 Introduction 2 Use of knock-for-knock clauses in the German market 3 Validity issues under German law 3.1 Court review of general terms and conditions 3.1.1 General principle 3.1.1.1 Defnition of general terms and conditions 3.1.1.2 Content test 3.1.1.3 What is “individually negotiated”? Negotiation of terms demands more than debating Additional security by a “caveat” clause? 3.2 Are knock-for-knock clauses general terms and conditions?

193

xv

182 182 184 184 185 186 187 187 188 189 192

193 193 194 194 194 195 195 196 196 196 197

detailed contents 3.3 Applying the reasonableness test 3.3.1 Commercial contracts versus consumer contracts 3.3.2 Fairness test – what is market practice? 3.3.3 The “risky” terms 3.4 Invalid terms under German civil law – no exclusion of liability for intent 3.4.1 General principle of liability under sec. 276 BGB 3.4.2 No agreements to the detriment of third parties 3.4.3 Consequences for knock-for-knock clauses under German law 4 Available case law 5 Practical consequences and advice for use of knock-for-knock clauses under German law

197 198 198 198

CHAPTER 11 KNOCK FOR KNOCK UNDER BRAZILIAN LAW Felipe T. Boechem and Márcio Opromolla 1 Introduction 2 Overview of civil liability regime in Brazil 2.1 Contractual liability and tort liability 2.2 Fault-based liability and strict liability 2.3 Indemnifable losses 3 Limitation of liability clause 4 Knock-for-knock clauses in Brazil 5 Conclusion

204

CHAPTER 12 THE IMPLICATIONS OF LITIGATION POST DEEPWATER HORIZON ON KNOCK-FOR-KNOCK CLAUSES IN US LAW Cindy Matherne Muller, Jeanne Amy and Jennifer David 1 Introduction 2 Applicable law 2.1 Federal court jurisdiction 2.1.1 Federal question jurisdiction 2.1.2 Diversity jurisdiction 2.1.3 Admiralty jurisdiction 2.2 Applicable law 2.2.1 Outer Continental Shelf Lands Act 2.2.2 State waters 2.2.3 Maritime law 2.2.3.1 Maritime contracts 2.2.3.2 What is a vessel? 2.2.4 Maritime law incomplete, supplemented with state law 2.2.5 Tension between OCSLA and maritime law 3 Indemnity in maritime 3.1 Enforceability of indemnity provisions in a maritime contract 3.2 Indemnity under the Longshore and Harbor Workers’ Compensation Act xvi

199 200 201 201 202 202

204 205 205 206 207 209 211 212

214 214 214 215 215 215 216 217 217 219 219 219 221 223 223 224 224 226

detailed contents 3.3 Insurance obligations and additional assured status Application of state law 4.1 The Louisiana Oilfeld Indemnity Act 4.2 Texas Oilfeld Anti-Indemnity Act 5 Conclusion Appendix A

4

CHAPTER 13 MUTUAL HOLD HARMLESS CLAUSES IN FRANCE AND FRANCOPHONE CIVIL LAW SYSTEMS Bertrand Montembault and Paul Morton 1 Introduction 1.1 Francophone civil law systems 1.1.1 The French Civil Code of 1804 1.1.1.1 French law of obligations: contract law and civil liability 1.1.1.2 Francophone African legal systems 1.1.2 Recent developments 1.1.2.1 French law 1.1.2.2 Francophone African legal systems 1.2 Mutual hold harmless clauses: a contractual re-allocation of liability 1.2.1 Origins and growing prevalence in French law 1.2.2 A note on English terminology 2 The legal nature of mutual hold harmless clauses under French law 2.1 The legal framework for mutual hold harmless clauses 2.1.1 Principles of the law on civil liability under French law 2.1.2 Contractual mechanisms for allocating risk between the parties 2.1.3 Contractual mechanisms for allocating third-party liability 2.2 Legal characterisation of mutual hold harmless clauses under French law 2.2.1 Allocation of risk between the parties 2.2.2 Allocation of third-party liability 2.2.2.1 Related third parties 2.2.2.2 Unrelated third parties 3 The legal regime governing mutual hold harmless clauses under French law 3.1 Exclusions of liability and waivers of recourse 3.1.1 Validity and exceptions 3.1.1.1 Exceptions linked to the law of obligations 3.1.1.1.1 Limiting non-contractual liability 3.1.1.1.2 Essential obligations 3.1.1.2 Specifc legislative exceptions 3.1.2 Efects of waiver of recourse and exclusions of liability clauses under French law 3.1.2.1 Strict interpretation 3.1.2.2 Exceptions linked to the nature of the fault xvii

227 229 229 233 234 235

238 238 238 239 239 239 240 240 240 241 241 241 242 242 242 243 243 244 244 244 244 245 246 246 246 246 246 246 247 247 247 248

detailed contents 3.1.2.3 Exceptions linked to the nature of the loss 3.2 Indemnifcation of third-party losses 3.2.1 Validity of clauses indemnifying third-party losses 3.2.2 Limitations CHAPTER 14 LESSONS FROM THE APPLICATION OF KNOCK-FOR-KNOCK CLAUSES UNDER MALAYSIAN LAW Wan Mohd Zulhafz Wan Zahari 1 Introduction 1.1 General overview of the Malaysian legal system 1.2 The law of contract in Malaysia 1.2.1 Statutory and judicial approach on indemnity clauses in Malaysia 1.3 Empirical study of oilfeld service contracts in Malaysia 1.4 Research design 1.4.1 Research process 2 Description of the case studies 2.1 Findings 2.2 Perception of the contractual formation process 2.3 Perception of risk allocation and indemnity clauses 3 Analysis of the indemnity and hold harmless clauses of Operators A, B and C 3.1 Liability regarding personal injury or death of employees and loss of property of the parties 3.1.1 Under LOGIC 3.1.2 Under FIDIC 3.1.3 Under Operator A’s contract 3.1.4 Under Operator B’s contract 3.1.5 Under Operator C’s contract 3.2 Claims by a third party 3.2.1 Third-party claims under LOGIC 3.2.2 Third-party claims under FIDIC 3.2.3 Third-party claims under Operator A’s contract 3.2.4 Third-party claims under Operator B’s contract 3.2.5 Third-party claims under Operator C’s contract 3.3 Liability regarding pollution 3.3.1 Pollution liability under LOGIC 3.3.2 Pollution liability under FIDIC 3.3.3 Pollution liability under Operator A’s contract 3.3.4 Pollution liability under Operator B’s contract 3.3.5 Pollution liability under Operator C’s contract 3.4 Insurance coverage 4 Conclusion

248 249 249 249

251 251 251 252 255 261 262 262 263 265 265 268 274 276 276 277 278 280 281 281 282 283 284 284 285 285 285 286 287 287 288 288 289 291

Index xviii

C ON T RI BUTO R S

Jeanne Amy, Trial Attorney in Aviation, Space and Admiralty Litigation, Torts Branch, Civil Division, US Department of Justice, [email protected]. Felipe T. Boechem, Partner in the Energy and Infrastructure Department and Head of the Oil and Gas practice at Lefosse, [email protected]. Hanieh Bolourian, Legal Research Intern at the University of Exeter, haniehbln@ gmail.com. Erik Brannsten, Attorney at law at Bane NOR (state-owned company responsible for the Norwegian national railway infrastructure), [email protected]. Lauren E. Burk, Of Counsel: Baker, Donelson, Bearman, Caldwell & Berkowitz, PC. Jennifer David, Partner in the New Orleans ofce of Broussard + Williamson, jen@ bwtriallawyers.com. Greg Gordon, Head and Professor of Law at Aberdeen University School of Law, Fellow at the Higher Education Academy, [email protected]. Knut Kaasen, Professor of Law at the Scandinavian Institute of Maritime Law, University of Oslo; Moderator in negotiations between construction industry and oil companies/owners to establish agreed standard contracts for ofshore and onshore development projects; and Arbitrator and Mediator in commercial disputes, knut. [email protected]. Henrik Lando, Professor of Law and Economics at the Copenhagen Business School, [email protected]. Bertrand Montembault, Partner in the energy, resources and infrastructure team in Herbert Smith Freehills’ Paris Ofce, [email protected]. Paul Morton, Of Counsel in the energy, resources and infrastructure team in Herbert Smith Freehills’ Paris Ofce, [email protected]. Cindy Matherne Muller, Partner in the construction, energy, environmental and natural resources, and maritime industry teams at Jones Walker, cmuller@joneswalker. com. Kyriaki Noussia, Associate Professor of Law at the University of Reading, and Arbitrator, [email protected]. xix

contributors Márcio Opromolla, Partner in the dispute resolution practice at Lefosse, marcio. [email protected]. Gideon Parchomovsky, Professor of Law at Hebrew University of Jerusalem Faculty of Law and University of Pennsylvania Carey Law School, [email protected]. ac.il. Jay R. Sever, Partner, Regional Practice Coordinator, and Head of the insurance and reinsurance practice group in the New Orleans Ofce of Phelps Dunbar, jay. [email protected]. Endre Stavang, Professor of Law at the University of Oslo, Department of Private Law, [email protected]. Kristofer Svendsen, Associate Professor of Law at the University of Stavanger and Kristiania University College; Senior Research Fellow at Tulane Center for Energy Law, Tulane Law School; and Associate Member at Aberdeen University Centre for Energy Law, [email protected]. Uisdean Vass, Managing Director at VassPetro Limited, [email protected]. Eckehard Volz, Partner and Head of the Hamburg admiralty and energy and ofshore team at Clyde & Co, [email protected]. Anna-Sophie Waldmann, Senior Associate at Clyde & Co, anna-sophie.waldmann@ clydeco.com. Trine-Lise Wilhelmsen, Professor of Law at the Scandinavian Institute of Maritime Law, University of Oslo; Chairman of the Standing Revision Committee for the Nordic Marine Insurance Plan; Chairman of the board of Norsk Legemiddelansvarsforsikring (drug liability insurance); Chairman of the board of Finansklagenemnda, skade; Chairman of the Petroleum Normprice Board; and Member of the board in Nordic Ofshore and Maritime Arbitration Association, [email protected]. Wan Mohd Zulhafz Wan Zahari, Deputy Dean (Academic and Internationalisation) and Assistant Professor of Law at Ahmad Ibrahim Kulliyyah of Laws, International Islamic University Malaysia (IIUM), and IIUM Company Secretary, [email protected].

xx

E D I T ORS ’ P REFAC E

The knock-for-knock indemnity: concepts, efects and questions Knock-for-knock indemnity clauses are a well-established feature of contracts in the oil and gas service sector. In using such clauses, the parties agree that for certain forms of potential liability – typically property damage, personal injury to employees, and sometimes other heads of claim such as consequential loss – any loss arising will be absorbed by the party who sufers it: “you look after your losses, I’ll look after mine.” It is an apparently simple, pragmatic and neat solution to the question of who bears liability: a risk allocation model so straightforward that it was described by one experienced English judge as “crude”.1 At frst sight, one might wonder why such a simple concept might need to have a book written about it. However, as so often in life, the knock-for-knock indemnity clause gives rise to more – and more important – issues than might at frst appear. The frst clue that there is more to this concept than might meet the eye lies in the length of some of the knock-for-knock indemnities commonly used in the oil and gas industry. For instance, the indemnity clause in LOGIC’s General Conditions of Contract for Ofshore Decommissioning2 extends to 11 sub-clauses spanning a little under two full pages and making use of no fewer than eight defned terms. A further half page is given over to a separate mutual indemnity for consequential loss.3 Why does an apparently crude risk allocation require so much precision and exposition? There are many reasons. Firstly, and perhaps most importantly, these clauses are of huge practical importance. In almost any oil and gas contract, there exists a remote but ever-present risk that an error in the undertaking of the contracted work could result in a disaster where the potential liabilities could easily run to hundreds of millions worth of dollars.4 If the parties are going to use knock-for-knock indemnities to distribute liability, then they are going to want to be as sure as possible that 1 Smit International (Deutschland) GmbH v Josef Mobius Bau-gesellschaft GmbH [2001] CLC 1545 at [19]–[20] per Mr Justice Morison. 2 LOGIC, General Conditions of Contract (including Guidance Notes) for Ofshore Decommissioning (2018) available online from www.logic-oil.com/content/standard-contracts0, Clause 22, pp. 25–27. 3 Ibid., Clause 25, p. 29. 4 Consider, for example, the Piper Alpha disaster which occurred on 6 July 1988, which led to the loss of 167 lives, the complete destruction of the Piper Alpha installation and the shutting-in of a signifcant volume of production capacity upstream of the destroyed platform. So great was the loss of production that the United Kingdom as a whole went from being a net oil exporter to a net oil importer for more than a year after the accident.

xxi

editors’ preface these clauses are going to work as intended at the critical moment when they are required.5 This requires clarity in drafting. It also requires the parties to anticipate any objections that the legislature or the courts may have to the use of such clauses, and to use the contractual wording to ensure that they comply with legislative or judicial barriers to enforcing the clause as drafted. Why should such legislative or judicial barriers exist? Because, in addition to being important, these clauses are controversial. The knock-for-knock indemnity is a radical instrument. These clauses do not tinker at the periphery of liability by, e.g., imposing a mere limitation of liability for economic damages payable under contract in the event of defective performance of the work obligation. They involve a fundamental re-allocation of risk, using the parties’ contract as a means of cutting across both contractual and delictual liability, divorcing liability from fault and wedding it instead to the parties’ conception of efciency, found in what one of the authors describes as the identifcation principle.6 This may be efcient, but what is the cost of severing the link with fault? Is it morally right? Is it safe? Legislatures in some jurisdictions have taken specifc steps to prohibit or at least control the use of knock-for-knock indemnities.7 And while other jurisdictions have taken no specifc steps to regulate the use of these clauses, they may nevertheless have to comply with the requirements of more general unfair contract legislation.8 And even in jurisdictions which adopt a relatively relaxed approach to legislation, the courts may still play a signifcant role in how or if these clauses can be enforced. Should a clause which reallocates liability in a way that has no regard to fault in danger of being struck down as contrary to public policy? If so, is this rule triggered when a party seeks to be excused from the consequences of its own negligence, or only if it claims an indemnity against the consequences of its own gross negligence or wilful misconduct? And what if the indemnifed and the indemnifer are both at fault? Diferent jurisdictions might answer these questions in diferent ways. And even jurisdictions in which the courts hold that there is no bright-line rule against the enforceability of knock-for-knock indemnities may apply technical or strained rules of interpretation to knock-for-knock indemnities, not prohibiting, but in a sense limiting, the manner in which the clauses may be used. It is not entirely clear if English law would at present take such an approach,9 but it plainly has in the past, and this has left a legacy upon the way in which knock-for-knock indemnities are drafted: the explanations, repeatedly found in knock-for-knock indemnities,10 that the clause is intended to provide an indemnity against the consequences of the 5 Or at least, the party that, on the facts of the case, has a claim under the indemnity will want to be sure that it will work as intended. As we shall see, the party called upon to pay under the indemnity (and perhaps more particularly, its insurers) may feel it has a strong fnancial incentive to go looking for problems with the drafting, meaning that the indemnity cannot be used against it. 6 See Chapter 2 of this book, p. 27. 7 See, e.g., the discussion of the principal US oil and gas jurisdictions contained in Chapter 12 of this book. 8 See, e.g., the discussion of the Unfair Contract Terms Act 1977 in Chapter 2 of this book, p. 34. 9 Arnold v Britton [2015] UKSC 36, discussed further in Chapter 2 of this book, p. 40. 10 See, e.g., LOGIC, Ofshore Decommissioning, ibid. n. 2, Cl 22.7: indemnities to apply “irrespective of cause and notwithstanding the negligence or breach of duty (whether statutory or otherwise) of the indemnifed party”.

xxii

editors’ preface indemnifed party’s own negligence are not there by accident, but in response to mishaps in previous cases.11 The editors hope that this brief introductory preface has demonstrated that the knock-for-knock indemnity clause gives rise to a broad range of signifcant legal questions that merit careful thought. Throughout the remainder of this work, these questions will be addressed from a variety of perspectives. The majority of the chapters address the questions from the perspective of a particular jurisdiction, and there is coverage of a broad and representative sample of jurisdictions drawn from around the world. Other chapters are concerned with public international law and legal history. However, it is perhaps fair to say that it is the law and economics perspective adopted by Parchomovsky and Stavang that lies at the heart of this book. It was this paper, frst discussed at a round-table event hosted by the Institute of Advanced Legal Studies in 2016, and the discussion which followed it, which gave the editors confdence that the humble knock-for-knock indemnity could sustain a full edited collection. We hope the reader will agree that that confdence was not misplaced. The editors would also like to thank our editorial assistants, Tulane Law students Andrew J. Pomaville and Jack C. Foster. Kristofer Svendsen, Endre Stavang and Greg Gordon

11 E. E. Caledonia v Orbit Valve [1995] 1 All ER 174.

xxiii

TAB L E OF CA SES

France Cass. 1e civ, 2 août 1951, JCP 1951, II, 6592 ......................................................................245 Cass. 1e civ. 15 avril 1961, JurisData no. 1961–000198........................................................249 Cass. 3e civ., 2 mars 2010, no. 09–12.272.............................................................................247 Cass. 1e civ., 31 mars 1987, no. 79–17.129 ...........................................................................248 Cass. Ch. mixte, 22 avr. 2005, no. 03–14112 ........................................................................248 Cass. Civ, 3, 27 juin 2001, no. 99–21.017 99–21.284 ............................................................248 Cass. Civ. 2e, 19 November 1964 .........................................................................................242 Cass. Civ, 2, 28 février 2013, no. 12–12.813 .........................................................................248 Cass. Civ, 2, 10 novembre 2021, no. 19–12.660....................................................................248 Cass. Com, 29 juin 2010, no. 09–11841 ...............................................................................248 Cass. Com. 24 October 2018, no. 17–25.672 ........................................................................242 Cass. Req. 19 janv. 1863, DP1863, 1, p. 248 ........................................................................246 Civ. 3e, 10 January 1978, no. 76–11.111...............................................................................249 Cour de Cassation, 11 January 1989, 86–17.323 ..................................................................242 Germany BAG, judgement dated 12. 11. 1998–8 AZR 221–97 = NJW 1999, 966 .............................201 BeckOGK/Mäsch, 1.1.2021, BGB § 328 Rn. 123 ................................................................201 Beck OK BGB/Janoschek, 58. Ed. 1.5.2021 Rn. 5, BGB § 328 Rn. 5 ................................201 Beck OK BGB/H. Schmidt, 58. Ed. 1.5.2021 Rn. 6, BGB § 307 Rn. 96 ............................196 Beck OGK/Schaub,1.3.2021,BGB§276Rn.4 .........................................................................200 BGH, judgement dated 13. 11. 1953 – I ZR 140/52 _ NJW 1954, 229 ...............................200 BGH, judgment dated 20.3.2014 – VII ZR 248/13 = NJW 2014, 1725 ..............................197 BGH, judgment dated 20.3.2014 – VII ZR 248/13 = NJW 2014, 1725 ..............................196 BGH order dated 20.11.2012 – VIII ZR 137/12, BeckRS 2013, 5597 Rn. 7 .......................196 BGH, order dated 19.3.2019 – XI ZR 9/18, NJW 2019, 2080 ............................................196 BGH, judgement dated 12.11.1980 – VIII ZR 293/79 = NJW 1981, 275 ...........................201 BGH, judgement dated 02.10.1985 – IV a ZR 18/84 = NJW 1986, 1100 ...........................201 BGH, judgement dated 09.07.1992 – VII ZR 7/92 = NJW 1992, 3158 ...............................200 BGH, judgement dated 18.12.1996 – IV ZR 321/95 = NJW 1997, 1012 ............................201 BGH, judgement dated 20. 11. 2012 – VI ZR 268/11 = NJW-RR 2013, 550 .....................200 BVerfG, judgement dated 23. 04. 1986–2 BvR 487/80 = NJW 1987, 827 ...........................201 HK-BGB/Reiner Schulze, 10. Auf. 2019, BGB § 276 Rn. 6 ...............................................200 Jauernig/Stadler, 18. Auf. 2021, BGB § 276 Rn. 1 ..............................................................200

xxv

table of cases MüKoBGB/Basedow, 8. Auf. 2019, BGB § 305 Rn. 35, § 310 Rn. 11 ...............................196 MüKoBGB/Grundmann, 8. Auf. 2019, BGB § 276 Rn. 1–2 ..............................................200 MüKoBGB/Gottwald, 8. Auf. 2019, BGB § 328 Rn. 261 ...................................................201 MüKoBGB/Wagner, 8. Auf. 2020, BGB § 823 Rn. 47 ........................................................200 Smit International (Deutschland) GmbH v Josef Mobius Baugesellschaft GmbH [2001] CLC 1545 ..................................................................... xxi Wolf/Lindacher/Pfeifer/Pfeifer, 7. Auf. 2020 Rn. 36, BGB § 305 Rn. 36 ..........................196 Malaysia American International Assurance Co Ltd v Koh Yen Bee (f) [2002] 4 MLJ 301 ......254, 290 Amman Singh v Vasudevan [1972] 1 LNS 7 ........................................................................255 Anderson v Fitzgerald (1853) 4 HLC 484; (1853) 10 ER 551 .............................................258 Ayer Hitam Tin Dredging Malaysia Bhd v YC Chin Enterprises Sdn Bhd [1994] 2 MLJ 754, 755 .............................................................................................................257 Azman bin Mahmood & Another v SJ Securities Sdn Bhd [2012] MLJU 660...................260 Chin Hooi Nan v Comprehensive Auto Restoration Service Sdn Bhd [1995] 2 MLJ 100.....................................................................................................................258 CIMB Bank Bhd v Maybank Trustees Bhd [2014] MLJU 117 ...........................................257 Fui Lian Credit & Leasing Sdn Bhd v Kim Leong Timber Sdn Bhd[1991] 1 CLJ 522.........................................................................................................................253 Hotel Anika Sdn Bhd v Majlis Daerah Kluang Utara [2007] 1 MLJ 248 ..........................258 Lee Kam Wah v Associated Asian Securities (Pte) Ltd [1991] 3 MLJ 286 .........................257 Macon Works & Trading Sdn Bhd v Phang Hon Chin & Anor [1976] 2 MLJ 177, HC.............................................................................................................255 Ooi Boon Leong v Citibank N.A. [1984] 1 MLJ 222 ..................................................252, 272 Polygram Records Sdn Bhd v The Search [1994] 3 MLJ 127 ..............................................254 Saad Marwi v Chan Hwan Hua & Anor [2001] 3 CLJ 98 ..........................................254, 290 Sabah Shell Petroleum Co Ltd & Anor v The Owners of and/or any other Persons Interested in The Ship or Vessel (The Borcos Takdir) [2012] 5 MLJ 515, 565 ..........................................................................................256, 257 Sekawan Guards Sdn Bhd v Thong Guan Sdn Bhd [1995] 1 MLJ 811 ...................... 259–261 Siong Electronic Industries (1981) Sdn Bhd v Sanyo Sales & Service Sdn Bhd [1997] MLJU 162 ...........................................................................................260 South East Asia Insurance Bhd v Nasir Ibrahim [1992] 2 MLJ 355; [1993] 1 SCR 89, SC ................................................................................................................255 Syarikat Uniweld Trading v The Asia Insurance Co Ltd [1996] 2 MLJ 160 ......................258 Sze Hai Tong Bank Ltd v Rambler Cycle Co Ltd [1959] MLJ 200 ....................................259 Norway Arbitral Award 20 August 1973 (referred by Knudtzon in Nordisk Skibsrederforening Medlemsblad Spesialnummer A 1984, s. 66) .................................141 Arbitral Award “Mørland 7” (ND 1988 p. 263 f.) .......................................................142, 146 Case 2013–06–25. LG-2012–077280 .....................................................................................158 “Chainco” judgement (ND 1991 p. 180 ...............................................................................146 Rt-2004–1545, para. 52 (Norwegian Supreme Court)..........................................................137 Supreme Court decision in Rt. 1994 p. 626 .........................................................................158 “Trans Tind” Arbitral Award (ND 1984 p.404) ...........................................................144, 145

xxvi

table of cases United Kingdom A Turtle Ofshore SA v Superior Trading Inc [2008] EWHC 3034 (Admlty) [2008] 2 CLC 953 ........................................................................................................... 22 Alderslade v Hendon Laundry [1945] KB 189 ..................................................................... 14 Addax v Acadia [2000] 1 Lloyd’s Rep 493 .............................................................................. 8 Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd [1983] 1 WLR 964; [1981] UKHL 12 (1981) ............................................................................................. 9, 45 Andar Transport Pty Ltd v Brambles Ltd (2004) 217 CLR 424 .......................................... 13 Arnold v Britton [2015] UKSC 36 ...................................................xxii, 37, 38, 40, 45, 48, 60 BHP Petroleum Ltd v British Steel plc and Dalmine SpA [2000] 2 Lloyd’s Rep 277.........................................................................................................59, 61 British Sugar P.l.c. v N.E.I. Power Projects Ltd (1997) 87 BLR 42; (1998) 87 BLR 42 CA ................................................................................................................ 8, 60 Caledonia North Sea Ltd v London Bridge Engineering Ltd (London Bridge) [2000] SLT 1123 (Ct of Sess) ............................ 6, 18, 24, 27–29, 33, 35, 38, 41, 43, 44, 46, 48, 49, 59–61 Caledonia North Sea Ltd. v British Telecommunications Plc; Same v Kelvin International Services Ltd.; Same v London Bridge Engineering Ltd.; Same v Norton (No. 2) Ltd (In Liquidation); Same v Pickup No. 7 Ltd.; Same v Stena Ofshore Ltd. same v Wood Group Engineering Contractors Ltd. –[2002] UKHL 32; [2002] 1 Lloyd’s Rep 553; [2002] SC (HL) 117; [2002] 1 All ER (Comm) 321 ......................................................... 27–30, 33, 35, 38, 43, 44, 46, 48, 49, 60, 61, 110, 270 Caledonia North Sea Ltd v London Bridge Engineering Ltd and Others (The Piper Alpha) [2002] UKHL 4, 32; [2002] 1 Lloyd’s Rep 553, 562, 563 (HL) .......................................................................... 3, 4, 9, 10, 29, 35, 61, 172, 277 Campbell v Conoco (UK) Ltd [2003] 1 All ER (Comm) 35 .....................................29, 46, 47 Canada Steamship Lines v The King (Regem) [1952] 1 All ER 305; [1952] AC 192 .........................................................................................................14, 22, 41– 45 Castellain v Preston (1883) 11 QBD 380 .............................................................................. 21 Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38; [2009] 1 AC 1101 ................. 40 Croudace v Cawoods [1978] 2 Lloyd’s Rep 55 ........................................................................ 8 Currie v Misa (1875) LR 10 Exch 153 .................................................................................255 Czamikow v Koufos [1969] 1 A.C. 350 ................................................................................... 7 Deepak Fertilisers and Petrochemicals Corporation v I.C.I. Chemicals & Polymers Ltd (1999) 1 TCLR 200 ........................................................................................................ 60 Donoghue v Stevenson [1932] AC 562 .................................................................................. 36 E. E. Caledonia Ltd (formerly Occidental Petroleum (Caledonia) Ltd) v Orbit Valve Co Europe Plc [1995] 1 All ER 174, afrming [1994] 1 WLR 221; [1993] 4 All ER 165; [1993] 2 Lloyd’s Rep. 418 QBD (Commercial Court)...................................................................... xxiii, 14, 18, 41–44, 46, 47, 109 Farstad Supply A/S v Enviroco Ltd [2011] UKSC 16; (2011) 15 EdinLR; [2010] UKSC 18; 2010 SCLR 379; (2010) 14 Edin LR 102 ......... 21–23, 27, 31, 34–36, 49, 110 Firma C-Trade SA v Newcastle Protection and Indemnity Association (The Fanti) (No 2) [1991] 2 AC 1 (HL) ......................................................................... 22 Gillespie Bros & Co Ltd v Roy Bowles Transport Ltd [1973] QB 400 ................................. 41

xxvii

table of cases Glencore Energy U.K. Ltd v Cirrus Oil Services Ltd [2014] EWHC 87 (Comm)................ 60 Glen’s Traders v Lancashire and Yorkshire Accident Insurance Co Ltd (1906) 8 F 915................................................................................................................ 37 Hadley v Baxendale (1854) 9 Exch 341 ...........................................................7, 8, 60, 61, 109 Hart v O’Connor [1985] A.C. 1000 ..............................................................................253, 254 HIH Casualty & General Insurance Ltd v Chase Manhattan Bank [2003] UKHL 6 [2003] 2 Lloyd’s Rep 297 ..................................................................... 39, 44–46 Hinton v Dibbin (1842) 2 QB 253 .......................................................................................109 Hollier v Rambler Motors (AMC) Ltd [1972] 2 QB 71 ....................................................... 45 Hotel Services Ltd v Hilton International Ltd [2000] 1 All E.R. (Comm.) 750 ................... 60 ICS. See Investors Compensation Scheme Ltd. v West Bromwich Building Society— Investors Compensation Scheme Ltd. v West Bromwich Building Society [1997] UKHL 28; [1998] 1 WLR 896 .......................................................................37, 39 Jodrell, Re (1890) 44 Ch D 590............................................................................................. 37 Karsales (Harrow) Ltd v Wallis [1956] 2 All ER 866, 869 ..........................................257, 258 London Bridge. See Caledonia North Sea Ltd— London Bridge & Others British Telecommunications plc. See Caledonia North Sea Ltd. v British Telecommunications Plc— Lloyds Bank v Bundy [1975] QB 326 ...................................................................................254 Markerstudy Insurance Co Ltd v Endsleigh Insurance Services Ltd [2010] EWHC 281 (Comm) ...................................................................................................... 60 McCain Foods G.B. Ltd v Eco-Tec (Europe) Ltd [2011] EWHC 66 (TCC) ........................ 60 Millar’s Machinery Co v David Way & Son (1935) 40 Com. Cas. 204 ................................ 60 Motours Ltd v Eurobell (West Kent) Ltd [2003] EWHC 614 (QB) ..................................... 59 Multiservice Bookbinding Ltd v Marden [1979] Ch. 84 ......................................................253 Nelson v Atlantic Power and Gas Ltd, 1995 SLT 102 .................................................... 41–43 Pentecost and Anr. v London District Auditor & Anr [1951] 2 KB 759, 763.....................109 Photo Production Ltd v Securicor Transport Ltd [1980] AC 827 ........................................ 59 Polypearl Ltd v E. On Energy Solutions Ltd [2014] EWHC 3045 (QB) .............................. 60 Prenn v Simmonds [1971] 1 WLR 1381 ................................................................................ 39 Reardon Smith Line Ltd v Hansen-Tangen [1976] 1 WLR 989 ........................................... 39 Robb v Salamis M & I Ltd, 2007 SLT 158 ......................................................................46, 47 Saga Cruises BDF Limited & anor v Fincantieri SPA [2016] EWHC 1875 (Comm) .......... 60 Saint Line v Richardsons [1940] 2 KB 99............................................................................. 60 Scottish & Newcastle plc v G D Construction (St Albans) Ltd [2003] EWCA Civ 16 ........ 22 Scottish Power UK plc v BP Exploration and Drilling [2015] EWHC 2658 ........................ 60 Slessor v Vetco Gray, unreported, 7 July 2006, (Ct of Sess, Outer House); [2007] SLT 400........................................................................... 25, 29, 38, 46–48 Smith v South Wales Switchgear Co Ltd [1978] 1 WLR 165 ..........................................45, 46 Smith v UMB Chrysler (Scotland) Ltd 1978 SC (HL) 1 ...................................................... 41 Starsin, The [2003] UKHL 12; [2004] 1 AC 715 ................................................................... 40 Thompson v T Lohan (Plant Hire) Ltd [1987] 1 WLR 649 ................................................. 35 Transocean Drilling UK Ltd v Providence Resources plc [2016] EWCA Civ 372 ...................................................................................................16, 17, 60 Victoria Laundry v Newman [1949] 2 KB 528 ....................................................................... 7 Walters v Whessoe Ltd and Shell Refning Co Ltd [1968] 2 All ER (All England Law Reports) 816 ........................................................................................................... 14 Wessanen Foods Ltd v Jofson Ltd [2006] EWHC 1325 (TCC) ............................................ 59 WesternGeco Ltd v ATP Oil and Gas (UK) Ltd [2006] EWHC 1164 (Comm) ................... 63

xxviii

table of cases United States of America Am. Home Assurance Co. v Chevron, Inc 400 F.3d 265, 269 (5th Cir. 2005) ....................113 A.M.C. Lifeboats Inc. v Apache Corp., 2008 WL 217177 (E.D. La. 2008) ........................105 Amoco Prod. Co. v Lexington Ins. Co., 98–1676 (La. App. 1 Cir. 9/24/99); 745 So. 2d 676, writ denied, 99–3553 (La. 2/25/00); 755 So. 2d 253 ...................................231 Baker v Director, Ofce of Workers’ Compensation Programs 834 F.3d 542 (5th Cir. 2016) .......................................................................................................222, 223 Barker v Hercules Ofshore, Inc., 713 F.3d 208, 2013 AMC 946 (5th Cir. 2013) ........222, 223 Barnett v Am. Constr. Hoist, Inc., 2011 1261 (La. App. 1 Cir. 02/10/12); 91 So. 3d 345 ..........................................................................................Appendix A, 236 Barrios v Centaur, LLC, 942 F.3d 670 (5th Cir. 2019) ................................................220, 221 BW Ofshore USA, LLC v TVT Ofshore AS, Case No. 14–1052, 2015 U.S. Dist. LEXIS 153840 (E.D. La. Nov. 13, 2015) .....................................................222, 223 Champagne v Tetra Applied Technologies, Inc., No. 05–299, 2006 U.S. Dist. LEXIS 37457, *13–14 (S.D. Tex. Feb. 6, 2006) ............................................................219 Chevron Oil Co. v Huson, 404 U.S. 97 (1971) .............................................................218, 219 Cimarex Energy Co. v CP Well Testing, L.L.C., 26 F.4th 683 (5th Cir. 2022) ....................104 Cleere Drilling Co. v Dominion Expl. & Prod., Inc., 351 F.3d 642, 647 (5th Cir. 2003) ...............................................................................................................233 Daughdrill v Ocean Drilling & Exploration., 665 F. Supp. 477, 481, 482 (E.D. La. 1987) ...............................................................................................................112, 225 De Lovio v Boit, 7 Fed. Cas. 418 (C.C.D. Mass. 1815) (No. 3776) ....................................216 Deepwater Horizon, Re, 470 S.W.3d 452 (Tex. 2015), opinion after certifed question answered, No. 12–30230, 2015 WL 13918242 (5th Cir. June 9, 2015) ...........103 Demette v Falcon Drilling Co., 280 F.3d 492, 2002 AMC 686 (5th Cir. 2002) ...........222, 223 Emery Waterhouse Co. v Lea, 467 A.2d 986 (Me. 1983) ..............................Appendix A, 236 Empire HealthChoice Assur., Inc. v McVeigh, 547 U.S. 677, 689–690 (2006) ....................215 Engel v Davenport, 271 U.S. 33, 37 (1926) ..........................................................................215 Erie R.R. Co. v Tompkins, 304 U.S. 64 (1938) ....................................................................216 Exxon Corp. v Cent. Gulf Lines, Inc., 500 U.S. 603 (1991) ................................................216 Fields v Pool Ofshore, 182 F.3d 353 (5th Cir. 1999) ...........................................................222 Fontenot v Mesa Petroleum Co, 791 F.2d 1207. (5th Cir. 1986); 493 A.2d 946 (Del Sup Ct. 1985) ................................................................................................................. 29 Gulf Ofshore Co., Div. of Pool Co. v Mobil Oil Corp., 453 U.S. 473, 478 (1981) ............215 Gunka v Consolidated Papers, Inc., 508 N.W.2d 426 (Wis. Ct. App. 1993)........................................................................................................Appendix A, 237 Horizon, Re, 470 S.W.3d 452 (Tex. 2015).............................................................................229 Houston Exploration Co. v Halliburton Energy Services Inc., 269 F.3d 528 (5th Cir. 2001) ...............................................................................................................112 J.A.R., Inc. v M/V LADY LUCILLE, 963 F.2d 96 (5th Cir. 1992) ....................................216 Ken Petr. Corp. v Questor Drilling Corp., 24 S.W.3d 344, 346 (Tex. 2000) ........................233 Klaxon Co. v Stentor Elec. Mfg. Co., Inc., 313 U.S. 487 (1941).................................216, 219 Kokkonen v Guardian Life Ins. Co. of Am., 511 U.S. 375, 377 (1994) ..............................215 Kossick v United Fruit Co., 365 U.S. 731, 735 (1961) .......................................................216 Laredo Ofshore Constructors, Inc. v Hunt Oil Co., 754 F.2d 1223, 1230 (5th Cir. 1985) ...............................................................................................................219 Larry Doiron, Inc., Re 879 F.3d 568 (5th Cir. 2018) (en banc), cert. denied, 138 S. Ct. 2033 (2018) ................................................................................... 219–221, 223

xxix

table of cases Lightering LLC v Teichman Group, LLC, 328 F. Supp. 3d 625, 627–629 (S.D. Tex. 2018).............................................................................................................220 Livings v Service Truck Lines of Texas, Inc., 467 So. 2d 595, 599 (La. App. 3 Cir. 1985) ...................................................................................................230 Lloyds of London v Transcontinental Gas Pipe Line Corp., 38 F.3d 193, 196 (5th Cir. 1994) ......................................................................................................................230 Longmire v Sea Drilling Corp., 610 F.2d 1342 (5th Cir. 1980) ..........................................227 Lozman v City of Riviera Beach, 568 U.S. 115, 121 (2013) ........................................221, 222 Marcel v Placid Oil Co., 11 F.3d 563, 569–570 (5th Cir. 1994) ............. 231, Appendix A, 236 Martin v Fab-Con, Inc., 9 F. Supp. 3d 642, 649 (E.D. La. 2014) .......................................222 Matte v Zapata Ofshore Co., 784 F.2d 628 (5th Cir. 1986) ................................................218 Meloy v Conoco, Inc., 504 So. 2d 833 (La. 1987) ...............................................................230 Mendez v Anadarko Pet. Corp. 466 F. App’x 316 (5th Cir. 2012) ......................................222 Merrell Dow Pharmaceuticals, Inc. v Thompson, 478 U.S. 804, 808 (1986) .......................215 Mobil Oil Corp. v Schlumberger, 598 So. 2d 1341 (Ala. 1992) .....................Appendix A, 235 Mooney v W&T Ofshore, Inc., 2013 AMC 1480 (E.D. La. 2013) .............................222, 223 M/V Lady Lucille, 963 F.2d at 98 ........................................................................................219 Norfolk Southern Ry. Co. v James N. Kirby, Pty Ltd., 543 U.S. 14, 25 (2004) (citing Exxon Corp. v Cent. Gulf Lines, Inc., 500 U.S. 603 (1991)) ...................216, 219, 220 Ofshore Co. v Robinson, 266 F.2d 769 (5th Cir. 1959)...............................................222, 223 Ofshore Logistics, Inc. v Tallentire, 477 U.S. 207, 222–223 (1986) .....................................215 Ogea v Lofand Bros., 622 F.2d 186 (5th Cir. 1980) ...................................................227, 228 Oil Spill by the Oil Rig “Deepwater Horizon,” Re, 841 F. Supp. 2d at 988, 1000, 1003–4, 6 (E.D. La. 2012) ................................................................... 111, 224, 225 Oil Spill by the Oil Rig “Deepwater Horizon,” Re, 745 F.3d 157 (5th Cir. 2014) .......222, 223 Owen Equipment & Erection Co. v Kroger, 437 U.S. 365, 374 (1978) ................................216 Riley v Alexander/Ryan Marine Servs. Co., 983 F.Supp.2d 884, 887 (S.D.Tex.2013) ..........221 Roberts v Energy Development Corp., 104 F.3d 782, 784 (5th Cir. 1997) ..................219, 230 Rupp v Am. Crystal Sugar Co., 465 N.W.2d 614 (N.D. 1991) ......................Appendix A, 237 Ryan Stevedoring Co. v Pan Atlantic Steamship Corp., 350 U.S. 124 (1956) .....................227 Seas Shipping Co. v Sieracki, 328 U.S. 85 (1946) ................................................................227 Stewart v Dutra Constr. Co., 543 U.S. 481, 489 (2005) (citing 1 U.S.C. § 3) ......................221 Stewart v Grand Isle Shipyard, Inc., Case No. 10–4016, 2011 U.S. Dist. Lexis 148603 (E.D. La. December 23, 2011) ..........................................................................224 Storage & Processors, Inc. v Reyes, 134 S.W.3d 190, 192 (Tex. 2004) .................................233 Tateosian v Vermont, 945 A.2d 833 (Vt. 2007)..............................................Appendix A, 237 Texas Co. v Gianfala, 222 F.2d 382 (5th Cir.), reversed on other grounds, 350 U.S. 879 (1955) ......................................................................................................222, 223 Theriot v Bay Drilling Corp., 783 F.2d 527 (5th Cir. 1986) ................................................224 Tullier v Halliburton Geophysical Services, Inc., 81 F.3d 552, 1996 AMC 1561 (5th Cir. 1996) ...............................................................................................................228 Union Pac. Resources Co. v Dolenc, 86 P.3d 1287 (Wyo. 2004) ..........................Appendix A, 237 Union Texas Petroleum Corp. v PLT Engineering, Inc., 895 F.2d 1043, 1050 (5th Cir. 1990) .......................................................................................................218, 223 Verdin v ENSCO Ofshore Co., 104 F. Supp. 2d 682, 690 (W.D. La. 2000), af’d, 255 F.3d 246 (5th Cir. 2001) .........................................................................................230 Voisin v O.D.E.C.O. Drilling Co., 744 F.2d 1174, 1175, 1176 & n. 1, 1177, 1179 (5th Cir. 1984) .......................................................................................................226, 227

xxx

table of cases Walter Oil & Gas Corp. v NS Group, Inc., 867 F. Supp. 549, 553 (S.D. Tex. 1994) ..........218 Warrior Energy Servs. Corp. v ATP Titan, 551 F. App’x 749 (5th Cir. 2014) ............222, 223 Wilburn Boat v Fireman’s Fund Ins., 348 U.S. 310, 1955 AMC 467 (1955) .......................223 Other jurisdictions Brazillian Special Appeal No. 1.076.465–SP (2008/0160567–4), published on 21 November 2018 (Brazil) ................................................................................................115 Darlington Futures Ltd v Delco Australia Pty Ltd (1986) 161 CLR 500 ............................ 13 Guardian Assurance Co Ltd v Condogianis (1919) 26 CLR 231 ........................................258

xxxi

TA B LE OF S TAT U T E S A N D MO D EL F ORM C ON TR AC TS

Brazil Civil Code (BCC) ...........11, 115, 207, 210 Arts. 186, 187 ................................ 115 Arts. 393, 398, 402, 493 ................. 206 Art. 432 ......................................... 115 Art. 475 ......................................... 206 Art. 884 ......................................... 207 Art. 927 ................................... 11, 207 Art. 931 ......................................... 207 Code of Civil Procedure: Federal Law no. 13.105 of 16 March 2015 .................................................. 207 Art. 373, I ...................................... 207 Federal Constitution............................ 211 Art. 5, II ........................................ 209 Federal Law no. 10.406 of 10 January 2002 .................................................. 206 Hydrocarbons Law No.9.478 of 6 August1997....................................... 114 Law No. 6,938/81 on Environmental Matters ............................................. 207 Law No. 8,078/1990 on Consumer Rights ............................................... 207 Law No. 13,874/2019 on the Validity of Limitation of Liability provisions ......................................... 210 New Civil Code (BCC) 2002 ............... 207 Pronouncement No. 444, V Jornada de Direito Civil ................................ 208 Denmark Danish Contract Act (Aftaleloven) s. 36(1) ............................................. 15 Danish Product Liability Act ................ 14 France Amended law no. 63–62 of 10 July 1963 .................................................. 240

xxxiii

Civil Code (Napoleonic Code) 1804 ...........................................239, 240 Book III......................................... 240 Title III .......................................... 239 Title IV .......................................... 240 Title V ........................................... 239 Art. 16–1........................................ 248 Art. 1170.................................246, 247 Art. 1217 et seq ............................. 240 Arts. 1231–1, 1240–1244 ................ 242 Arts. 1382, 1383 ............................. 245 Ordinance no. 2016–131 of 10 February 2016, in force 1 October 2016 ......... 240, 244, 246 Code of Commerce Art. 133–1 ...................................... 247 Law No. 2018–287 of 20 April 2018, ratifying Ordinance no. 2016–131 .......................................... 240 Law no. 2019/035/AN of 4 July 2019 (replacing the 1983 Civil Code) ....... 240 Transportation Code Art. L. 5422–12, 15 ....................... 247 Uniform Act on General Commercial Law .................................................. 241 German Civil Code (BGB) ............ 12, 13, 194, 199, 200, 203 s. 276 .............................................. 200 s. 276(1), sent.1 .......................200, 201 s. 276(2) ......................................... 200 s. 276(3) ..................................200, 201 s. 305 et seq ....................194, 195, 197 s. 305(1) ......................................... 195 s. 305(1), sent.1 .............................. 195 s. 305(1), sent.3 .......................196, 197 s. 307 .......................................195, 198 s. 307(1), (2) ................................... 195 s. 308 ...............................195, 198, 199

table of statutes and model form contracts s. 309.......................................195, 198 s. 309(7) ..................................199, 199 s. 310(1), sent.2 .............................. 195 Guinea Civil Code............................................ 240 Malaysia Application of Laws Ordinance (Sabah) 1951 .................................... 252 Application of Law Ordinance (Sarawak) 1949................................. 252 Civil Law Act (formerly the Civil Law Ordinance last revised 1972) 1956.................................................. 252 ss. 3, 5 .....................................252, 290 Civil Law Ordinance 1956 ................... 252 Consumer Protection (Amendment) Act 2010 (2013) 1 MLJ 1................. 253 Contract Act 1950 ............................... 289 Part III........................................... 253 ss. 77, 78 .........................255, 260, 289 Federal Constitution.....................251, 252 Arts. 4, 74, 121 .............................. 252 Oilfeld Anti-Indemnity Act’ (2020)..... 256 Petroleum Act...................................... 274 Straits Settlements (fnal version of the Civil Law Ordinance) 1878 ........ 251 Mali Malian law no. 87–31/AN-RM of 29 August 1987 establishing the General Regime on Obligations..................... 240 Norway Compensation for Damages Act (relating to compensation in certain circumstances; 13 June 1969 No. 26) .............................165, 185 s. 2–1.......................................130, 165 Construction All Risks (CAR) ...................... 132, 133, 137, 140, 149, 155,157, 159–163 s. 29(2) ........................................... 149 Sch. B .....................................161, 162 Contracts Act , 31 May 1918, No. 4 .........................................150, 176 s. 4–9.............................................. 150 s. 4–9(2) ......................................... 150

xxxiv

s. 36 (1983) ............ 119, 122, 134–130, 137–140, 143, 145, 146, 148–150, 176, 178–181 s. 36, ND 1989.225 ........................ 180 s. 36, ND 1990.204 ........................ 179 s. 36, ND 1991.180 ........................ 180 s. 36, ND 2000.240 ........................ 178 s. 36, NL 5–1–1 ............................. 176 s. 36, NL 5–1–2 ...... 176–178, 180, 181 s. 36, Rt-906.717 ............................ 176 s. 36, Rt-1994.626 .......................... 180 s. 36, Rt-2000.806 .......................... 179 s. 36, Rt-2001.603 .......................... 179 s. 36, Rt-2002.1155 ........................ 179 s. 36, Rt-2008.969 .......................... 179 s. 36, Rt-2010.1345 ........................ 179 s. 36, Rt-2013.388 .......................... 179 s. 36, Rt-2013.769 .......................... 179 s. 36, Rt-2014.351 .......................... 179 s. 36, Rt-1915.840 .......................... 176 s. 36, Rt-1926.712 .......................... 176 s. 36, Rt-1948.370 .......................... 176 s. 36, Rt-1948–3700 ....................... 138 s. 36, Rt-1994–626, (“quay- inspector”)..... 138, 141, 146, 176, 180 s. 36, Rt-1959–849 (“the apprentice”) ...................... 141 s. 36, Rt-1999.922 .......................... 178 s. 36, U 1993.851 ........................... 180 s. 36, U 2005.................................. 180 s. 36, U 2005.243, 632, 243............ 180 s. 36, U 2006.632 ........................... 180 Norwegian Contracts Act , 1989— s.784 ............................................... 176 Damages Compensation Act (Lov om skadeserstatning 13 June 1969 Nr. 26).............................................. 184 Fabrication Contract 1987 (NF 87) .............................120, 128, 189 Fabrication Contract 2007 (NF 07)..... 191 Art. 1.10 ........................................ 190 Fabrication Contract 2015 (NFC 15)..........................120, 153, 164, 188, 190 Part 1— Art. 1.14 ........................................ 167 Art. 1.16 ........................................ 168 Art. 1.22 ........................................ 167

table of statutes and model form contracts Art. 1.26 ........................................ 168 Art. 23 ........................................... 190 Art. 23.5 ........................................ 191 Art. 30 ............................189, 190, 192 Art. 30.1 ......................... 167, 189–191 Art. 30.2 ................. 167, 169, 189–191 Art. 30.2(2) .................................... 169 Art. 30.3 .........................169, 190, 191 Art. 30.3(b) .................................... 191 Art. 31.2(a), (b) ............................. 191 Hagstrøm, Ansvarsfraskrivelse (TfR) 1996— ss. 422–518 ..................................... 134 Insurance Contracts Act (ICA) (No. 69/1989) s. 4–9(1) ......149, 171, 177 s. 1–3(1) ......................................... 166 s. 1–3(2)(c), (d)............................... 166 s. 4–9(2) ......................................... 177 s. 7–1.............................................. 171 Maritime Act ....................................... 142 Maritime Code 1994 no. 39 (MC) .........................................165, 183 Ch. 10 ............................................ 185 PA §. 7–3........................................ 165 PA §. 7–4........................................ 166 PA §. 7–4 (a) ................................. 166 MC §. 191 ...................................... 165 MC §. 193 ...............................166, 187 MC §. 193 (c) ............................... 166 §. 507 ............................................. 183 Norske Lov 5–1–2 ............................... 139 Onshore Construction Standard) NS 8407— cll. 40.3, 42.5.................................. 137 Petroleum Act (repealed) 22 March 1985 Nr. 11 ...................................... 183 Petroleum Act 1996 no. 72 (PA) ........................... 154, 165, 183–186 Ch. 7 ....................... 182, 183, 185, 192 Ch. 8 .......................................183, 185 s. 1–6a, d.................................185, 186 s. 5–4 .............................................. 183 s. 7–1 ...............................183, 185, 186 s. 7–2 .............................................. 185 s. 7–3 .......................................184, 186 s. 7–4 ...............................186, 187, 191 s. 7–5 ....................... 187, 188, 191, 192 s. 7–5(c) .......................................... 186 s. 7–7 .............................................. 184 s. 7–8 .............................................. 184 s. 8–1 .......................................183, 185

xxxv

s. 10–1(1)........................................ 132 s. 10–8 ............................................ 169 s. 10–9............. 154, 165, 167, 170, 183 Petroleum Regulations (FOR-1997–06– 27–653) ......................................132, 154 s. 72................................................ 154 s. 73.................................132, 154, 158 s. 73(2) ........................................... 132 Pollution Control Act 13 March 1981 Nr. 06 ...............................183, 184 §. 6 ................................................. 186 §. 53 ............................................... 183 §. 61 ............................................... 186 Sale of Goods Act 1988 s. 13 (1) .......................................... 157 s. 23 (1) .......................................... 135 Subsea Contract 2005 (NSC 05) ............................. 120, 153, 188–190 Art. 1(i) .......................................... 153 Art. 23 ........................................... 190 Art. 23.5 ........................................ 191 Art. 30 ............................189, 190, 192 Art. 30.1 ................................. 189–191 Art. 30.2 ................................. 189–191 Art. 30.3 .................................190, 191 Art. 30.3(b) .................................... 191 Art. 31.2(a), (b) ............................. 191 Tort Act s. 4–2 .............................................. 148 s. 4–3 .............................................. 148 Total Contract 2015 (NTK 2007) ........ 124 Art. 24.2(2), (3).............................. 124 Art. 25.4(1) .................................... 124 Appendix B.................................... 124 5MNOK ........................................ 131 Total Contract 2015 (NTK 15)........... 120, 128, 131, 132, 134–136, 139, 145, 146, 151–154, 159, 188 Arts. 1.12, 14, 15 ........................... 153 Art. 1.16 ........................................ 156 Art. 1.21 ........................................ 156 Arts. 1.23, 1.24 .............................. 153 Art. 1.26 ........................................ 130 Arts. 2, 3 ........................................ 120 Art. 3.1 ...................................142, 158 Art. 8.1 .......................................... 123 Art. 8.4 .......................................... 158 Art. 8.5 .......................................... 127 Art. 8.5b ........................................ 131

table of statutes and model form contracts Arts. 9.2, 14.1 ................................ 123 Art. 15 ........................................... 131 Art. 15.1 ........................................ 124 Art. 16.1(2) .................................... 123 Arts. 17.1, 3, 5 ............................... 121 Art. 19.1 ........................................ 157 Art. 22.1 .................................123, 149 Art. 22.2 ........................................ 123 Art. 23 ........................................... 190 Art. 23.5 ........................................ 191 Art. 24.2 .........................122, 124, 137 Art. 24.3 .................................121, 124 Art. 24.4 ........................................ 125 Art. 25.1, 2 .................................... 137 Art. 25.3 .................................121, 122 Art. 25.3(2) .................................... 122 Art. 25.4 ........................................ 124 Art. 25.5 .................................122, 125 Art. 26.2 .................................121, 123 Art. 27 ........................................... 121 Art. 29 ............................128, 129, 156 Art. 29.1 .........................125, 149, 157 Art. 29.1(1) .................................... 135 Art. 29.2 ................. 125, 135, 157, 160 Art. 30 ............ 128, 143, 189, 190, 192 Art. 30.1 ........ 131, 134, 146, 156, 158, 189–191 Art. 30.1(1) .................................... 136 Art. 30.2 ................ 131, 134, 156, 158, 189–191 Art. 30.2(1) .................................... 136 Art. 30.3 ................ 131, 134, 156, 159, 190, 191 Art. 30.3(b) .................................... 191 Art. 30.3(3) .................................... 136 Art. 31 ........................................... 132 Art. 31.1 .................................158, 159 Art. 31.1(b), (c).............................. 159 Art. 31.1(3) .................................... 148 Art. 31.2 .................................132, 158 Art. 31.2(a), (b) ............................. 191 Art. 31.2(3) .................................... 148 Art. 31.2(4) .................................... 125 Art. 32.1 .................................121, 125 Art. 32.2 ........................................ 124 Art. 34.2 ........................................ 123 Art. 24.2 ........................................ 122 Total Contract (Modifcation; 2015 (NTC (MOD) 15) ............................ 188 Art. 23 ........................................... 190 Art. 23.5 ........................................ 191

Art. 30 ............................189, 190, 192 Arts. 30.1, 2 ............................ 189–191 Art. 30.3 .................................190, 191 Art. 30.3(b) .................................... 191 Art. 31.2(a), (b) ............................. 191 Total Contract (Modifcation; 2015 (NTK (MOD) 15) ............................ 120 WELCAR 2001 ................................... 160 WELCAR 2012 ............................140, 155 Senegal Senegalese Code of Civil and Commercial Obligations................... 240 Sweden Swedish Contract Act .......................... 146 United Kingdom Consumer Rights Act 2015 ............. 23, 34 Contract (Third Party Rights) (Scotland) Act 2017 ........................... 36 Contracts (Rights of Third Parties) Act 1999 ....................................... 36, 55 Law Reform (Miscellaneous Provisions) (Scotland) Act 1940......... 48 s. 3 ................................................... 48 Marine Insurance Act 1906 s. 1 ................................................... 21 Unfair Contract Terms Act (UCTA) 1977 ...........xxii, 23, 27, 34–36, 59, 62, 63, 110, 255, 268 s. 2(1) ......................................... 34, 35 s. 2(2) ....................................... 35, 110 s. 2(3) ............................................... 35 s. 2(4) ......................................... 23, 34 s. 2A............................................... 110 s. 4 ................................................... 34 s. 4.1................................................. 63 ss. 12, 16(1) ...................................... 34 s. 16(1)(a) ................................... 34, 35 s. 16(1)(b), 17 ................................... 35 s. 18.................................................. 34 Unfair Terms in Consumer Contracts Regulations 1994, SI 1994/ 3159 ........ 59 Unfair Terms in Consumer Contracts Regulations 1999, SI 1999/ 2083 ........ 59 United States Anti-Indemnity Statutes, by State (Mostly Appendix A)....... 235–237

xxxvi

table of statutes and model form contracts Alaska. Stat. §. 45.45.900..................... 235 Arizona. Rev. Stat. §§. 32–1159, 34–226, 41–2586 ............................... 235 Arkansas. Code §. 4–56–104, 104(b), § 22–9–214, 214(b) ........................... 235 California. Civ. Code §. 2782............... 235 Colorado. Rev. Stat. §§. 13–50, 50.5, 5–102, 102(8)(a), 13–21–111.5, 111.5(6)(b) ........................................ 235 Connecticut. Gen. Stat. §. 52–572k, 572k(a) ............................................. 235 Delaware. Code. Ann. 6 §. 2704, 2704(a) ............................................. 235 Florida. Stat. Ann. §. 725.06, 725.06(1)........................................... 235 Georgia. Code Ann. §. 13–8–2(b) ........ 235 Hawaii. Rev. Stat. §. 431:10– 222......... 235 Idaho Code Ann. §. 29–114................. 235 Illinois. 740 ILCS §. 35/1 ..................... 235 Indiana. Code Ann. §. 26–2–5 ............. 235 Iowa. Code Ann. §. 537A.5 ................. 236 Kansas. Stat. §. 16–121, 121(b) ............ 236 Kentucky. Rev. Stat. §. 371.180............ 236 Louisiana Civil Code Article 2004 .................................................. 113 Louisiana Oilfeld Indemnity Act (LOIA) La. Rev. Stat. ........ 229–231 R. S. §. 9:2780.................113, 229, 236 R. S. §. 9:2780.1 ......................232, 236 R. S. §. 9:2780.1(A)(1) ................... 231 R. S. §. 9:2780.1(A)(2) ................... 231 R. S. §. 9:2780.B .....................113, 230 R. S. §. 9:2780.F............................. 113 R. S. §. 9:2780.1(I) ..................232, 236 R. S. §. 9:3306................................ 231 R. S. §. 38:2216(G) (Public Contracts) .................................. 236 Maryland. Code Ann. §. 5–401, 401(a)(1)–(2) ..................................... 236 Massachusetts. Gen. Laws Ann. ch. 149 § 29C ................................... 236 Michigan. M.C.L.A. § 691.991 ............ 236 Minnesota. Stat. §§. 337.01, 337.02 ..... 236 Mississippi. Code. Ann. § 31– 5–41 ..... 236 Missouri. Rev. Stat. §. 434.100 ............ 236 Montana. MCA §. 28–2–211 ............... 236 Nebraska. Rev. Stat. §. 25–21, 187, 187(1) ............................................... 236 Nevada. N.R.S. §. AB 125, §. 2 (applies to residential construction contracts) ........................................ 236

xxxvii

New Hampshire. Rev. Stat. § 338-A:1, 338-A:2 ............................................. 236 New Jersey. Stat. Ann. §. 2A:40A-1..... 236 New Mexico. N.M. Stat. Ann. §. 56–7–1, 56–7–1(1) ......................... 236 New Mexico Oilfeld Anti-Indemnity Act (NMOAIA), N.M. Stat. Ann. §. 56–7–2 ........................................ 229 New York. Gen. Oblig. Law §. 5–322.1, 1(1)–(2) .................113, 237 North Carolina. Gen. Stat. Ann. §. 22B-1 ............................................ 237 Ohio. Rev. Code Ann. §. 2305.31 ........ 237 Oklahoma. Stat. title 15 §. 221 ............ 237 Oregon. Rev. Stat. §. 30.140, 140(1)–(2) ......................................... 237 Rhode Island. Gen. Laws §. 6–34–1 .... 237 South Carolina. Code Ann. §. 32–2–10......................................... 237 South Dakota. Codifed Laws § 56–3–18.......................................... 237 Tennessee. Code Ann. §. 62–6–123 ....................................... 237 Texas Construction Anti-Indemnity Act (TCAIA).................................... 234 Tex. Ins. Code §. 151.102........234, 237 Tex. Ins. Code §. 151.103........217, 237 Tex. Ins. Code §. 151.104............... 234 Tex. Ins. Code §. 151.103.............. 234 Tex. Ins. Code §. 151.105............... 234 Texas Oilfeld Anti-Indemnity Act (TOAIA) 1973– 2001 IELTR 18 ............................................... 34, 233 Tex. Civ. Prac. & Rem. Code §. 127.003 ................................... 233 Tex. Civ. Prac. & Rem. Code §. 127.005 ................................... 233 Tex. Civ. Prac. & Rem. Code §. 127.005(b), ............................. 233 Tex. Civ. Prac. & Rem. Code §. 127.005(c) ................................... 233 Tex. Civ. Prac. & Rem. Code §. 127.001(4)................................... 114 Tex. Civ. Prac. & Rem. Code §. 127.002(5)................................... 114 Tex. Civ. Prac. & Rem. Code §. 127.003 ....................................... 229 Tex. Civ. Prac. & Rem. Code §. 127.003 ............................113, 114 Tex. Civ. Prac. & Rem. Code §. 127.004(1), (2), (3)...................... 114

table of statutes and model form contracts Utah. Code Ann. §. 13–8–1 ................. 237 Virginia. Code Ann. §. 11–4.1 ............. 237 Washington. Rev. Code. §. 4.24.115, 115(1)(b) ........................................... 237 West Virginia. Code Ann. §. 55–8–14......................................... 237 Wyoming Oilfeld Anti-Indemnity Act (WOAIA), Wyo. Stat. §. 30–1–131 .....................229, 237 Carriage of Goods by Sea Act, 46 U.S.C.S. §. 30701 .......................................... 216 Art. III........................................... 215 Art. III, §. 2 ................................... 215 Art. III, §. 2, cl. 1 .......................... 215 Clean Water Act (CWA) 33 U.S.C. ...... 94, 224, 225 §§. 1251 et seq. ............................... 224 §. 1321(b)(3), (4), (7)...................... 225 Commercial Instruments and Maritime Lien Act (formerly Ship Mortgage Act 1920), 46 U.S.C ......................... 216 § 31301 et seq. ............................... 216 Constitution ......................................... 215 1 U.S.C. §. 3................................... 221 28 U.S.C. §. 1331 ........................... 215 28 U.S.C. §. 1332(a) ....................... 215 28 U.S.C. §. 1333 ........................... 215 Energy Policy Act of 2005................... 217 Harter Act 1893, 46 U.S.C. ................. 216 §§. 30104, 30105 ............................. 215 §§. 30701–30707 ............................. 216 Lands Act §. 1333 (a)(2) .................................. 218 Longshore and Harbor Workers’ Compensation Act (LHWCA) 33 U.S.C ....................................226, 227 §. 901 et seq. .................................. 226 §. 905(b) ..................................226, 227 §. 905(c) ......................................... 227 National Defense Authorization Act for 2021 ............................................ 217 Outer Continental Shelf Lands Act (OCLSA), 43 U.S.C...........111, 214, 217–219, 223 §. 4 ................................................. 227 §. 1331 et seq ................................. 214 §. 1333 (a)(1) ...........................111, 217 §. 1333(a)(2)(A) .............................. 217 Ship Mortgage Act of 1920. See United States Commercial

Instruments and Maritime Lien Act 30 C.F.R. § 585.100 ....................... 217 International Statutes and Model Form Contracts BARGEHIRE .......................................... 4 BIMCO ................................ 2, 4, 5, 11, 12 BIMCO WINDTIME ............12, 195, 197 BIMCO LOGIC .................................... 12 BIMCO Newbuildcon.......................... 120 Conditions of Contract as a Model for an International Construction Contract’ (2011) 1 BMR 32 ............. 271 cl. 4.18 ........................................... 286 Convention on Civil Liability for Oil Pollution Damage Resulting from Exploration for and Exploitation of Seabed Mineral Resources, London (1 May 1977) ...................... 182 FIDIC Conditions of Contract for EPC Turnkey Projects (Silver Book), 1999 .......................... 275 FIDIC Federation International Des Ingenieurs-Conseils. ..........195, 275, 277, 279, 286, 289 cl. 17 .............................................. 288 cl. 17.1 ....................................277, 278 cl. 18.1 ........................................... 289 cl. 18.3 ........................................... 278 cl. 18.3(d)(i)–(iii) ............................ 278 General Contracting Principles (FAIR Principles)–International Marine Contractors Association (IMCA) 2011............................275, 276, 281, 285 HEAVYCON ....................................... 4, 6 2007 Revision ....................................... 7, 8 IADC ..................................................... 19 IMHH Deed. See Mutual Indemnity and Hold Harmless Deed IMCA General Contracting Principles (Rev. 1) (n199) ............................................. 282 (n628) ............................................. 285 LOGIC............... 15, 19, 23, 24, 26–28, 33, 45, 47, 49, 52–54, 57–59, 63, 108, 109, 130, 195, 197, 275, 276, 279, 282, 283, 285, 289

xxxviii

table of statutes and model form contracts LOGIC Construction Contract, 2nd Edition, 2003.................................... 275 cl. 1.2 ............................................. 283 cl. 1.4 ............................................. 130 cl. 1.9 ......................................130, 283 cl. 17(1)(a)...................................... 284 cl. 18.3 ........................................... 283 cl. 18.3(d)(i)–(iii) ............................ 283 cl. 19 .......................................108, 109 cls. 19.1, 2 ...................................... 108 cl. 19.2c.......................................... 130 cls. 19.3, 4, 5, 6, 9, 10, 11.............. 108 cls. 20, 21.1 .................................... 109 cl. 21.2 ........................................... 108 cl. 22 .............................................. 277 cl. 22.1 ........................................... 286 cl. 22.1(a), (b) ................................ 286 cl. 22.1(c) ........................277, 282, 283 cl. 22.2 ........................................... 286 cl. 22.2(a), (b) ................................ 286 cl. 22.2(c) ................................277, 283 cl. 22.2(d)....................................... 283 cls. 22.3, 4 ...................................... 286 cl. 22.5(b) ................................277, 286 cl. 22.6 ........................................... 277 cl. 25 .............................................. 277 cls. 35, 36 ....................................... 137 Mutual Indemnity and Hold Harmless Deed (IMHH Deed) 2022.........................................52–59, 62 cl. 1 .................................................. 62 cl. 1.1 ................................... 52, 55, 57 cl. 2 ...................................... 53, 55, 56 cl. 2.1 ......................................... 53, 55 cl. 2.2 ............................................... 53 cl. 2.3 ............................................... 54 cl. 3.2 ............................................... 56 cl. 3.2(ii), (iii) ................................... 56 cl. 3.4 ............................................... 56 cl. 4.1 ................................... 53, 54, 57 cl. 4.2 ......................................... 53, 54 cl. 4.3 ............................................... 53 cl. 4.4 ......................................... 52, 53 cl. 5 .................................................. 54 cls. 5.1, 5.2 ....................................... 55 cl. 5.3 ............................................... 56 cl. 6.1(i)............................................ 55 cl. 6.1(ii)........................................... 53 cl. 7.1 ......................................... 52, 54 cls. 7.2, 7.3 ....................................... 56

xxxix

cl. 11 .......................................... 54, 55 cl. 23 ................................................ 59 Sch. 1 ............................................... 52 Sch. 2 ............................................... 54 New York Anti-Indemnity Law........... 113 Nordic Countries Contract Act............. 15 s. 36 (General Clause)...................... 15 Nordic Freight Forwarder Agreement (NSAB).......................... 180 NSAB 1975, s. 25 .......................... 180 NSAB 2000, s. 22(5) ...................... 180 NSAB 2015, s. 23 .......................... 180 Nordic Marine Insurance Plan 2013 Version 2019 (NP) ............166, 171 cl. 3–33 .......................................... 171 cl. 5–14 .......................................... 171 cls. 7–1, 8–1 ................................... 171 cl. 18–1 (i)(1), (2) ........................... 171 Oil & Gas UK Joint Operating Agreement (JOA) [standard form] ................................ 110 cl. 6.2.4(b)...................................... 110 OFL Proposal ............................... 167–169 1.1 .................................................. 167 1.1(3), (10) ..................................... 168 8.1 .................................................. 167 8.1(a), (b) ....................................... 166 8.2 .................................................. 167 8.2(a), (b) ....................................... 166 8.3 .................................................. 169 8.4 ...........................................167, 169 8.5 .................................................. 169 8.6–8.9 ........................................... 167 8.12(1–4) ........................................ 170 8.12(7)............................................ 171 PROJECTCON........................................ 4 SUPPLYTIME (BIMCO) 2005 ......2, 5, 6, 195, 197 Supplytime Part II Defnitions ...... 168 Boxes 2, 3 ...................................... 167 1989 Revision................................. 7, 8 2005 Revision................... 7, 8, 12, 164 2017 Revision....................... 6, 12, 164 Subclauses 9(c), (e) ............................ 5 Clause 14 ............................... 3, 4, 199 Subclause 14(a).................................. 5 Subclauses 14(a)(i), (ii) ...... 5, 166, 167 Subclause 14(c) .................................. 5 Subclause 14(d)...................... 5, 8, 167 Subclause 15 .................................. 167

table of statutes and model form contracts Treaty with Great Britain on Marine Transportation and Litigation, U.S.-Gr. Brit., Dec. 4, 1942, 56 Stat. 1780 (1942) ..... 68 TUGHIRE............................................... 8 UKOOA, IMCA, OCA and WSCA...... 58 Uniform Intermodal Interchange and Facilities Access Agreement ............. 231

Subclause 16 ...................................... 5 Subclause 17(a) .............................. 170 Subclause 17(a)(ii) ......................... 171 Subclause 18(c) .................................. 5 SUPPLYTIME 2005 Charter Party TOWCON ....................................2, 5, 6, 8 cl. 18 ................................................ 22 TOWHIRE ......................................2, 5, 6

xl

PART I CHAPTER 1

The development of knock-for-knock clauses in the last 15 years Kyriaki Noussia and Hanieh Bolourian

1 Introduction 1.1 Defnition “Knock-for-knock” (hereinafter K4K) clauses operate in various forms. In its most basic form, the clause provides that both parties will cover their own losses, i.e., the “knocks”, that are sufered by their respective property or personnel. As such, neither party can sue the other for the loss sufered. This arrangement minimises the costs that such “knocks” would otherwise incur, by saving the parties the valuable time and money that would typically be spent litigating fault-contingent liability. 1.2 Essential meaning of a K4K clause Under a K4K scheme, the loss lies where it falls, irrespective of fault and without recourse to other parties.1 It is often accompanied by a series of mutual indemnity clauses, wherein parties agree to indemnify the other contracting parties against injury to, or death of, their own personnel, loss or damage to their property, and any other specifed losses. The primary advantage of K4K clauses is that liability is predetermined, thus saving insurance costs and time in attributing fault, from both a factual and legal perspective. This leads to faster compensation to injured parties, and thus makes K4K clauses very attractive to industry, especially where multiple parties are involved. Essentially, K4K clauses underpin a simple, consensual scheme of mutual risk allocation. Moreover, they function as contractual exclusion clauses, with each party seeking (1) to exclude its own liability, even if caused by its own fault; and (2) to obtain an indemnity from other parties for any liability to which it may be exposed.2 Hence the need for their terms to be reciprocal, and to cover similar liabilities, as many jurisdictions prohibit one-way liability exclusions, most notably US maritime law.2

1 A. McCooke, Shipowners: The False Economy of Amending Knock for Knock Clauses, www. shipownersclub.com/the-false-economy-of-amending-knock-for-knock-clauses/, accessed 10 August 2021. 2 Ibid.

DOI: 10.4324/9781003206798-1

1

kyriaki noussia and hanieh bolourian 1.3 Historical genesis of the K4K clause The phrase and general meaning of “knock for knock” originated in the US car insurance industry during the early 1900s.3 Early automobile insurers commonly found themselves trapped in the nonideal position of having to pay for their customers’ car repairs frst, and then having to engage in costly, drawn-out litigation, sometimes spending years arguing as to which party was at fault. To avoid losing the time and money required to litigate, they began to enter agreements with one another that, after a collision, each insurer would simply pay their policyholder’s losses and seize the continuation of the claim. K4K clauses were efectively devised and designed as a contractual tool to limit the uncertainty of liability exposure stemming from inherently hazardous activities.4 Additionally, K4K was brought into the maritime insurance context in the United Kingdom during World War II as a mechanism for reducing litigation costs arising from frequent naval accidents.5 Responding to the threat of German submarines, English ships sailed in the dark, with all lights switched of, in a very tight cluster. This tactic reduced their exposure to German submarines but increased the overall rate of collisions, hence the parties decided to subject themselves to the K4K principle, essentially agreeing that each party will bear its own costs.6 K4K clauses entered the energy sectors in the 1960s with the commencement of oil and gas exploration in the North Sea.7 This method of risk allocation was also common in towage agreements in the United States, contractually restricting the allocation of liability between the two parties.8 Each party, i.e., the tug and the tow, would bear their own risk arising from the towing operation. Additionally, such a contract formulation was viewed as more likely to be able to be relied upon in court.9 By the 1970s, K4K clauses were being used in some ocean or international towage agreements. In 1985, BIMCO, one of the largest international shipping organisations, and an authority on industry best-practices, released two standard form towage agreements, TOWCON and TOWHIRE, with K4K at the heart of their liability regime. Moreover, SUPPLYTIME, a BIMCO-drafted time-charterparty contract designed for supply vessel and ofshore work, had a K4K clause added as standard in its 1989 revision. Other standard form contracts used in the ofshore industry, and many charterers’ in-house charterparties, also apply a K4K allocation of liabilities. This industry-wide use, along with widespread judicial acknowledgement of its enforceability, has efectuated the K4K clause’s predominance as a mechanism by which coventurers can streamline the allocation of risk and liability in ofshore operations.10

3 P. Saraceni, N. Summers, “Reviewing Knock for Knock Indemnities: Risk Allocation in Maritime and Ofshore Oil and Gas Contracts” (2016) 30 ANZ Mar LJ, 28–43, 28. 4 Ibid. 5 See Parchomovsky and Stavang, Chapter 3 of this book. 6 Ibid. 7 P. Saraceni, N. Summers, “Reviewing Knock for Knock Indemnities: Risk Allocation in Maritime and Ofshore Oil and Gas Contracts” (2016) 30 ANZ Mar LJ, 28–43, 28. 8 A. McCooke, Shipowners: The False Economy of Amending Knock for Knock Clauses, www.shipownersclub.com/the-false-economy-of-amending-knock-for-knock-clauses/, accessed 10 August 2021. 9 Ibid. 10 Ibid.

2

development of knock-for-knock clauses in the last 15 years 2 Issues in using, implementing and enforcing K4K clauses In accordance with general professional liability standards, under both contract and tort, negligent performance of a professional service typically leads to liability for damages on behalf of the negligent party. The main issue in connection with the K4K regime is whether the professional liability for negligently conducted or omitted provisions of service – this being the fundamental obligation of a contract for services – can be limited, or exempted, by an indemnity clause and, if so, to what extent. The efect of a contractual K4K scheme is to reverse the commonly accepted principle whereby the at-fault party is held liable for the ensuing damage. As the K4K provision creates a novel legal relationship, departing from centuries of default tort law, the contract absolving parties of this liability must do so properly, or it may not be readily enforceable. As a general matter, courts do not typically favour standard form contracts, or worse, contracts of adhesion, and will interpret such provisions against the drafting party, under contra proferentem. Nevertheless, K4K clauses are regularly included in standard form contracts in the maritime law domain, in various types of charterparties and ofshore contracts.11 Such ubiquitous use of K4K clauses in the wider maritime sector makes logical and economic sense to frms, as they limit the parties’ exposure to unpredictable liability.12 Furthermore, participants in the oil and gas industry, specifcally, regularly choose to contract out of standard tort liability, suspending the rules of negligence and strict liability in their dealings with one another, and embrace the K4K rule instead. Conclusively, the industry’s adoption of K4K eliminates liability for harms incurred through the course of their inter se relationships. Hence parties internalise the cost of the harms they may sufer at the hands of other contractual parties. As for insurance, K4K operates in terms of frst-party insurance, an intriguing feature that runs contrary to contemporary theory and practice.13 3 Various uses 3.1 SUPPLYTIME 2005 K4K clauses express the parties’ contractual language. Clause 14 of the SUPPLYTIME 2005 charter party excludes from the K4K indemnity any undeclared dangerous cargo or hazardous or noxious substances shipped by the charterers on board the vessel, any pollution claims and any general average. Such claims must be resolved under traditional fault-based liability regimes. As such clauses are exclusion clauses, there must be unambiguous wording to exclude one’s liability. The scope of K4K clauses extends to protect the indemnifed party from gross negligence, wilful

11 See, e.g., Caledonia North Sea Ltd v London Bridge Engineering Ltd and Others (The Piper Alpha) [2002] UKHL 4; [2002] 1 Lloyd’s Rep 553 (HL). 12 M. Mudrić (2015). “The Guardcon Contract, Knock-for-Knock Clauses, DCFR and Unfair Terms (Part I)”, JIML 21 (2015), 51–62, 55. 13 See Parchomovsky and Stavang, Chapter 3 of this book.

3

kyriaki noussia and hanieh bolourian misconduct, material breach of contract, statutory or strict liability, consequential loss and proportionate and concurrent liability.14 3.2 Insurance and P&I clubs In the insurance sector, P&I Clubs will often review and, if appropriate, approve K4K clauses if they are balanced and mutual and so long as the P&I Club member has not waived any right to limit liability under any applicable law. Unbalanced K4K provisions are not “poolable” in respect to any liabilities of which the member would not have been exposed in the absence of the contract. Also, for the K4K liabilities to be “poolable”, they must incorporate indemnities, protecting members if they are sued by a third party who is not bound by the contract, and members must contract on terms that do not expose them to disproportionate liabilities.15 K4K clauses are a fundamental part of the maritime and ofshore oil and gas sectors. Recognised in English and Australian jurisprudence, as well as in other legal systems, and in various bodies of case law, such hold harmless and indemnity clauses play a signifcant role in commercial operations within those sectors. Also, courts carefully examine the wording of K4K clauses in an efort to determine their meaning, scope, operation and efciency; hence clarity in drafting such clauses is absolutely necessary to make sure that they serve the scope for which they have been drafted.16 3.3 Use in ofshore contracts The arrangement provided by K4K clauses has been used extensively in upstream oil exploration activities, as such operations typically involve many contractors and subcontractors. Furthermore, in the case of a potential loss – e.g., the Deepwater Horizon incident in the Gulf of Mexico in 2010 – multiple potential defendants would be involved. To avoid such complexities, the parties agree to accept responsibility for loss or damage to their own property and for injury or death to their own employees, irrespective of blame. Such arrangements were described by Lord Bingham in the Piper Alpha17 case as a market practice that has developed in order to accommodate the peculiar features of ofshore operations. Additionally, K4K clauses are often used in downstream oil and gas contracts, as the parties to such contracts are also usually parties to the contracts upstream or are subcontractors who serve in the same role that they serve in the upstream contracts. BIMCO has produced a series of specialised charter forms that are designed to cater for the specialised requirements of the ofshore industry, such as HEAVYCON, BARGEHIRE, PROJECTCON,

14 P. Saraceni, N. Summers, “Reviewing Knock for Knock Indemnities: Risk Allocation in Maritime and Ofshore Oil and Gas Contracts” (2016) 30 ANZ Mar LJ, 28–43, 32–33. 15 Ibid., 41. 16 Ibid., 43. 17 Caledonia North Sea Ltd v. London Bridge Engineering Ltd and Others (The Piper Alpha) [2002] 1 Lloyd’s Rep 553 (HL).

4

development of knock-for-knock clauses in the last 15 years TOWCON, TOWHIRE, and SUPPLYTIME, all of which contain K4K clauses, the commercial purpose of which is to make it clear who is to bear the risk.18 4 Modern use Today, the K4K principle is acknowledged as being at the very core of the SUPPLYTIME form-contract, and K4K clauses in SUPPLYTIME mean that each party should bear responsibility for any damage or loss to its own property, or injury to its own personnel, even if the damage, loss or personal injury is caused by the act, neglect or breach of duty (whether statutory or otherwise) of the other party. When the other party compensates the claimant, even if the loss of compensation fled by the claimant is caused by the other party’s act, neglect or breach of duty – be it statutory or otherwise – the party shall still indemnify the other party for the compensation submitted by the claimant, and the compensation made by the party against the other party shall be called K4K indemnity. Again, the essence of K4K clauses is that the parties, based on simple apportionment of risks and liabilities, replace the original fault liability by mutual agreement for several exempted and non-exempted items.19 4.1 The BIMCO SUPPLYTIME 2017 In 2017, the Documentary Committee of BIMCO adopted the revised SUPPLYTIME 2017, bringing it more in line with current practices in the ofshore sector. The objective of BIMCO’s 2017 revision of SUPPLYTIME was to better balance the interests of the owners, and the interests of the charterers, in relation to the K4K liability regime. Apart from that, there were also several smaller changes from the previous iteration of the contract. In relation to the K4K clause (Clause 14) and the liability regime, this has been strengthened in the 2017 edition, because the defnitions of the “Charterers’ Group” and of the “Owners’ Group”, which afect the operation of the K4K liability regime, have been improved and moved to the “Defnitions” section, so that the usage is consistent throughout the charterparty. Subclause 14(a)(i) describes the losses that fall on the owner, as any loss or damage to property belonging to the Owners’ Group, or personal injury and death of anyone in the Owners’ Group, are to be borne by the owners, even if they are caused by the negligence or default of the Charterers’ Group, subject to three exceptions stated in subclauses 9(e), 14(c) and 18(c). Subclause 14(a)(ii) describes the losses that fall on the charterers. It is not identical to the subclause governing the owners, and this is due to the diferences in their respective obligations. Nevertheless, there is an equivalent strengthening of the K4K regime for charterers, subject to two exceptions in subclauses 9(c) and 16.

18 R. Williams, “Knock for Knock Clauses in Ofshore Contracts: The Fundamental Principles”, 53–67, in: B. Soyer, A. Tettenborn, (Eds) (2014) Ofshore Contracts and Liabilities, Informa Law. 19 Y. Han, “On Knock-for-Knock Principle: Analysis of SUPPLYTIME 2017 Clause 14(a)”, 2019 China Oceans L. Rev. (2019) 4, 128–149, 128; S. Rainey, “The Construction of Mutual Indemnities and Knock-for-Knock Clauses” 68–107, 70–71, in: B. Soyer, A. Tettenborn, (Eds) (2014) Ofshore Contracts and Liabilities, Informa Law.

5

kyriaki noussia and hanieh bolourian Notably, the description of the included causes of loss has been expanded to include any loss arising out of, or in any way connected with, the performance or non-performance of this Charter Party whatsoever, and in any circumstances, even if such loss, damage, or personal injury or death is caused wholly or partially by the act, neglect, breach of duty, be it statutory or otherwise.

Hence SUPPLYTIME 2017 has retained the key features that made this particular standard form one of the most popular in the ofshore sector, while making several changes to modernise it, and to ensure that it keeps up with the changing market practice; strengthening the K4K liability regime by balancing liabilities and indemnities; and by reducing of the number of exceptions to the regime.20 5 A historical encounter of the K4K clauses in the shipping and in the ofshore oil and gas sectors Early on, K4K fourished in contracts for the shipping industry, and the ofshore oil and gas industry, in exploring the North Sea. With time, certain risks were excluded from the K4K regime, as evidenced by the evolution of the BIMCO SUPPLYTIME form. When frst published in 1975, the form represented a contract where all intricacies of the sector were therein depicted, and as a result, many aspects of the regime favoured owners, especially in regard to a charterer’s liability for loss or damage negligently caused to the vessel or to her owners. Such provisions were only looked upon and amended in 1989, when the SUPPLYTIME form was redrafted to encompass fairer provisions, and to achieve balance between the interests involved. The K4K regime provisions for mutual risk allocation were popular in the ofshore industry forms of contract such as HEAVYCON, TOWCON and TOWHIRE, drafted by BIMCO; the LOGIC Construction Contract; and the Norwegian Subsea Contract 05 (known as NSC 05). Under those contracts, K4K operates fairly, as liability and indemnity provisions are balanced. As oil-and-gas activities involve risky undertakings at virtually all stages, the use of the clause has become popular for a number of reasons. Fault, though traceable through the traditional allocation of liability under the common law, becomes a matter of contractual stipulation for who bears the burden under K4K. Allocation of liability therefore becomes simpler. The commercial or economic imperative for mutual indemnifcation was indeed recognised in the case of Caledonia North Sea Ltd v London Bridge Engineering Ltd (London Bridge).21 There, Lord President Rodger noted as relevant evidence from one of the witnesses, that the practice is fundamental economics of the business, and thus, to the North Sea operation. The use of the clause, however, is not a foolproofy guard against litigation for the purpose of tracing which party is liable in the event of damage or loss, as experience indicates. A possible corollary is that parties could assess risk and accept liabilities more easily. Indemnity clauses also reduce the scope and 20 Kennedys Law, “Notes from the Bar: BIMCO SUPPLYTIME 2017, 18 October 2017, https://kennedyslaw.com/thought-leadership/article/notes-from-the-bar-bimco-supplytime-2017/. 21 Caledonia North Sea Ltd v London Bridge Engineering Ltd [2000] S.L.T. 1123 at 1150 (London Bridge).

6

development of knock-for-knock clauses in the last 15 years cost of insurance available for parties. The extent of possible risk instances that may occur, e.g., in an oil platform due to the act or omission of a contractor, is infnite.22 6 The evolution of K4K clauses in standard contracts and in English case law In cases where contracts contain a K4K regime which either adopts standard industry wording or forms part of a standard form contract, courts have shown a willingness to give efect to the object and purpose of K4K clauses. By conforming to a K4K clause in ofshore contracts, in the event of loss and damage to respective parties’ property, or the subsequent death of employees, companies uphold personal liabilities.23 Under the BIMCO SUPPLYTIME revisions of 1989 and 2005 formats, and the revised HEAVYCON 2007 contract, in the event of “consequential loss” the exclusion clause becomes a provision of the K4K clause. The term “consequential loss” has been defned in a contract setting, whereby the normal loss can be stated as the market value of the property, as the money or services that the claimant should have received under the contract, less either the market value of what he does receive or the market value of what he would have transferred, but for the breach having occurred. Indeed, consequential losses have been perceived to encompass anything beyond such a normal measure, e.g., profts lost, or expenses incurred, through the breach, and they are recoverable if not too remote. Because misassumption in the classifcation of losses has instigated contractual difculties in energy construction and supply ofshore contracts, a solution was sought. Recoverable losses in contracts were started from the “limb one” and “limb two” categories of losses established by the rules adopted in the seminal case of Hadley v Baxendale.24 The English court rejected the mill’s loss of proft claim for late delivery of a mill part, because of the lack of clarity in the mill’s subsequent loss of proft in the event of a late restart in production. Following this case, the ruling was re-stated in Victoria Laundry v Newman25 and in Czamikow v Koufos26 for which the Victoria Laundry ruling27 explained the type of knowledge that parties must possess in order to recover losses. It has been stated that knowledge “possessed” is of two kinds, one imputed, and the other actual, and that the reasonable person is taken to know the ordinary course of things and, consequently, what loss is likely to result from a breach of contract in that ordinary course. Indeed, this was taken to be the subject matter of the “frst rule” in Hadley v Baxendale28 or any other case containing knowledge of special circumstances outside the ordinary course of things, such that a breach given those special circumstances would be likely to cause more loss. Finally, it was remarked that such a case instigated a “second rule” such that additional losses are recoverable.

22 C. Ugwuanyi, “Examining the Exclusionary Nature of Oil and Gas Contract Mutual Indemnity Hold Harmless Clauses”, I.E.L.R. 2012, 4, 136–146. 23 A. Iyer, “Excluding ‘Consequential Loss’ in Ofshore Contracts”, 23 October 2020, www.iylegal.com/ excluding-lsquo-consequential-loss-rsquo-in-ofshore-contracts. 24 Hadley v Baxendale (1854) 9 Exch 341. 25 Victoria Laundry v Newman [1949] 2 K.B. 528. 26 Czamikow v Koufos [1969] 1 A.C. 350. 27 Victoria Laundry v Newman [1949] 2 K.B. 528. 28 Hadley v Baxendale (1854) 9 Exch 341.

7

kyriaki noussia and hanieh bolourian Recoverable indirect losses, or “limb two” losses, under English law must stem from the breaching party’s actual knowledge of the loss, independent of the scale, which the party was aware of at the time of contract that such a loss would result from a breach. The difculty is that this distinction between consequential losses and all other losses is not the same as the distinction between the “frst” and “second” “limbs” in the Hadley v Baxendale29 rule, as consequential loss may well fall within the “frst limb” as a direct loss which was a natural consequence of the breach. Hence it may not be entirely accurate to say that the Hadley v Baxendale30 “second limb” covers only, or every, consequential loss. Under English law, the scope of certain K4K clauses can turn upon the interpretation of the term “consequential loss” provided in exclusion clauses. In ofshore contracts, wherein a loss is deemed to be incurred as a “direct loss” following a breach of contract, the parties can still be found liable to meet any number of “consequential losses”. An example of such could be a loss of proft. English courts will construe exclusion and limitation clauses strictly against the party seeking to rely upon them, pursuant to the contra proferentem rule, which may – in the context of consequential loss clauses in ofshore contracts – bring other consequences.31 The widely cited case Croudace v Cawoods32 determined that the use of the word “consequential” did not cover any loss which directly and naturally resulted in the ordinary course of events from late delivery; and in Addax v Acadia,33 non-consequential losses associated with incurred “hedging costs” were found recoverable as “direct” losses. In British Sugar Plc v NEI Power Projects,34 regarding increased production costs and loss of profts for the claimant, it was held that since the faulty power station equipment supplied by the defendant had directly led to such losses, those losses would not be encompassed within an exclusion of liability for “consequential” losses and could thus be recovered as direct losses. By itself, inserting the term “consequential losses” into the exclusion clause of an ofshore contract may be insufcient to protect the owner from costly claims for loss of proft, production or business interruption. Nevertheless, the main hiring contracts (TUGHIRE, TOWCON, SUPPLYTIME 89 [and revised 05] and HEAVYCON 07) each have consequential loss provisions which purport to protect the parties from claims for “consequential losses” specifcally. Clause 23 of the 2007 revision of HEAVYCON (HEAVYCON 07) marked the frst introduction of a consequential damages clause into HEAVYCON, identical to clause 14(c) of SUPPLYTIME 05. In interpreting the language of such a clause, courts may likely rule that all loss of proft, whether foreseeable or not, must necessarily include losses falling within the frst limb of Hadley v Baxendale35 as well as those falling within the “second limb”, and that the lack of any reference to any other direct losses 29 Ibid. 30 Ibid. 31 A. Iyer, “Excluding ‘Consequential Loss’ in Ofshore Contracts”, 23 October 2020, www.iylegal.com/ excluding-lsquo-consequential-loss-rsquo-in-ofshore-contracts. 32 Croudace v Cawoods [1978] 2 Lloyd’s Rep 55. 33 Addax v Acadia [2000] 1 Lloyd’s Rep 493. 34 British Sugar Plc v NEI Power Projects (1998) 87 BLR 42 CA. 35 Hadley v Baxendale (1854) 9 Exch 341.

8

development of knock-for-knock clauses in the last 15 years would tend towards such an interpretation. However, there is still a risk that, if the specifc term used in those clauses is “consequential loss”, courts may interpret the clause as excluding only any “second limb” losses and not any losses falling under “limb one”, assuming they are “direct” or “natural losses”.36 A distinction also appears to be drawn between the construction of exclusion clauses on the one hand and limitation clauses on the other. In Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd,37 the House of Lords held that a limitation clause need not be construed in accordance with the same principles that apply to exclusion or indemnity clauses, and that the actual absence of a reference to negligence in the limitation clause did not prevent it from protecting against the contractor’s liability for negligence. However, in E. E. Caledonia Ltd v Orbit Valve Co Europe,38 wherein the indemnity clause was expressed generally and without reference to negligence, the court found that the parties’ right to sue each other for negligence had been preserved. For K4K to operate, an indemnity clause should expressly and clearly refer to negligence; as this was not the case in Orbit Valve, the K4K regime did not extend to liability caused by a party’s own negligence.39 6.1 The signifcance of the Piper Alpha litigation The basis of “mutual hold harmless” or K4K clauses can be seen in the House of Lords’ landmark decision in the case of the Piper Alpha litigation.40 The Piper Alpha disaster of 6 July 1988 resulted in 167 dead, 62 injured, and total destruction of the Piper Alpha oil platform of the coast of Scotland. The operator of the platform, Caledonia North Sea Ltd (formerly Occidental Petroleum (Caledonia) Ltd), along with its co-venturers, and its insurers, settled the fatal accident and personal injury claims using a formula agreed upon within months of the disaster. In the House of Lords’ decision, Lord Mackay reinforced the view in relation to primary liability on the contract in this case by stating that this was not entirely typical, as the relevant clause required the contractor to indemnify the operator in respect of injury to or death of persons employed by the contractor, irrespective of any contributory negligence, whether active or passive, of the party to be indemnifed, unless such injury, death, damage, loss or destruction was caused by the sole negligence or wilful misconduct of the party which would otherwise be indemnifed. This case demonstrates that an operator may voluntarily choose to insure against claims for which it has a contractual right of indemnity, and that choice will not operate to the advantage of the party contractually obliged to pay the indemnity.41 Accordingly, this case is notable

36 A. Iyer, “Excluding ‘Consequential Loss’ in Ofshore Contracts”, 23 October 2020, www.iylegal.com/ excluding-lsquo-consequential-loss-rsquo-in-ofshore-contracts. 37 Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd [1983] 1 WLR 964. 38 E. E. Caledonia Ltd v Orbit Valve Co Europe [1994] 1 WLR 221. 39 P. Saraceni, N. Summers, “Reviewing Knock for Knock Indemnities: Risk Allocation in Maritime and Ofshore Oil and Gas Contracts” (2016) 30 ANZ Mar LJ, 28–43, 30–31. 40 Caledonia North Sea Ltd v. London Bridge Engineering Ltd and Others (The Piper Alpha) [2002] 1 Lloyd’s Rep 553 (HL). 41 T. Hewitt, “Who Is to Blame? Allocating Liability in Upstream Project Contracts”, Journal of Energy & Natural Resources Law, 26:2, 177–206.

9

kyriaki noussia and hanieh bolourian for its contribution to the emergence of the mutual hold harmless indemnity regime. The fact that the operators had indemnity did not stop the court from holding the contractors responsible for indemnifcation of their staf and crew members. Hence coming to this point in law, where there is contributory negligence – versus sole negligence on the part of the operator – the contractor needs to apply its own share of indemnity, regardless of the operator’s contributory negligence.42 The insurers, therefore, would have been subrogated to the claims of the operator against the contractors, as the House of Lords upheld in the previous judgment. The indemnity was to have applied in favour of the operator, regardless of any negligence in the part of the operator; and death or injury to the contractor’s employees was to be covered by the mutual hold harmless regime rather than the standard fault-based regime. 7 The contribution of other jurisprudence to the historical evolution of K4K clauses 7.1 US jurisprudence K4K clauses, when present in an oilfeld agreement, must be balanced and comprehensive for several reasons, as, but for their existence, it would be extremely difcult to determine who holds fault in drilling site accidents.43 Some US states (e.g., Louisiana, Texas, New Mexico and Wyoming) have Limitation Acts prohibiting certain indemnity agreements. This means that any contractual clause that liberates the contractor or the indemnifed party from taking responsibility for loss or damages caused by its own negligence or fault, such as K4K, is to be considered void. Some American courts have also issued rulings rejecting contractual exculpation from responsibility for loss or damage caused by gross negligence or wilful misconduct.44 However, it remains a question as to how exactly the K4K regime works out in jurisdictions that are weary of broad liability limitation, as it may not be a practical insurance tool in onshore operations, where there is more clarity in terms of potential risks as compared to ofshore sites.45 Pursuant to such terms, K4K clauses need to specify when laying out health and safety guidelines, and each party’s responsibilities and liabilities, in light of those guidelines. There also should be no carve-out for “gross negligence” or “wilful misconduct” as they can lead to less certainty in allocation of liabilities unless such clause is thoroughly defned. The K4K regime must also apply to all the parties involved, such as other contractors and subcontractors, to promote fairness and equality in the legal treatment provided.46 In some American states (e.g., Texas and Louisiana), the anti-indemnity legislation is widely seen as a reaction to oil companies’ attempts to force local service and product providers to accept all liabilities, even if 42 Caledonia North Sea Ltd v London Bridge Engineering Ltd [2002] UKHL 32; [2002] 1 Lloyd’s Rep 562, 563, HL. 43 T. Middis, “Knock for Knock Indemnities – Are They Appropriate for On-Shore Infrastructure Projects?” 7 May 2015, Addisons, 1–3. 44 Ibid. 45 Ibid. 46 Ibid.

10

development of knock-for-knock clauses in the last 15 years the loss stems from the oil company’s fault. However, an enforceable K4K clause in an oil or drilling contract can be achieved, despite such statutory limitations, given that the language of the contract is well constructed and clear. Summarily, contractual agreements including K4K clauses are generally enforceable in the United States, absent any statutory or judicial precedent to the contrary.47 7.2 Deepwater Horizon On 20 April 2010, the Macondo exploratory well was being drilled ofshore by the Deepwater Horizon platform when it sufered a blowout, causing a fre and resulting in 5 million barrels of oil spilled into the Gulf of Mexico. Thousands of legal suits were fled against BP, and on 20 May 2015, BP settled the multibillion-dollar lawsuits with its co-venturers Transocean, Halliburton and Cameron. Importantly, Transocean and its contractors relied on K4K clauses in the respective contracts to exclude their liability for loss and damage other than to their own employees and property.48 7.3 Brazil In Brazil, contractual terms in the ofshore sector would fall under two main contractual regimes: (1) international contract forms, such as BIMCO standard charter contracts, incorporating the K4K clause principle; or (2) Petróleo Brasileiro SA (Petrobras). The latter, a Brazilian state oil contract representing 95% of Brazilian ofshore contracts, instigates its own contractual terms under the Brazilian Civil Code. Pursuant to Brazilian civil law, Petrobras contracts avoid K4K indemnity through provisions limiting indemnifcation amounts and excluding indirect damages or loss of earnings. A basic principle of Brazilian civil law is that any person who causes damage, whether through employees or subcontractors, to another must indemnify the aggrieved party in a form proportional to the damage sufered. Under Section 927 of the Brazilian Civil Code, this obligation arises regardless of fault, in the circumstances specifed by the Code; or when the activity that caused the damage included a risk to the environment, or to third parties. Nevertheless, under Brazilian law, the addition of a K4K clause is possible since terms and conditions of contracts and clauses are negotiable between parties so long as they do not contravene public order and third-party interests. For example, a limitation of liability clause under a contract of carriage is considered by the Brazilian courts as contrary to Brazilian law and therefore null and void. It must be noted that in the context of carriage of goods contracts, for instance in relation to standard bills of lading, the Brazilian courts’ interpretation, and position on the limitation of liability clause included in the bill of lading, is considered as onerous to the receiver and thus not valid. Also, considering the nature of ofshore contracts in the oil sector – the equal bargaining strength, and fexibility for more open and reciprocal negotiation of contractual terms – 47 L. Lambert, “Knock-for-Knock Contracts Are Enforceable in the US”, Standard Bulletin: Ofshore Special Edition, November 2011. 48 P. Saraceni, N. Summers, “Reviewing Knock for Knock Indemnities: Risk Allocation in Maritime and Ofshore Oil and Gas Contracts” (2016) 30 ANZ Mar LJ, 28–43, 30–31.

11

kyriaki noussia and hanieh bolourian dissipates the Brazilian civil courts’ position regarding the potential for unfair ofshore contracts, and thus could potentially yield to the inclusion of a K4K clause. Such a clause has yet to be challenged in a Brazilian court, which considers the clauses an adaptation of foreign law and jurisdiction regulations.49 7.4 Germany The growth in recent years of ofshore wind farm projects in the Exclusive Economic Zone (EEZ) of Germany has given rise to the possibility of potential disputes related to their underlying contracts in German courts and arbitral tribunals. German courts generally lack practical experience in interpreting contractual forms such as the BIMCO SUPPLYTIME 2005 or 2017, the BIMCO WINDTIME or LOGIC contracts, all of which have been developed against the background of English law. The validity of K4K clauses under German law would depend on the court’s evaluation of whether (1) the clause is part of general terms and conditions; (2) it can be considered reasonable; and (3) irrespective of general terms and conditions, whether it contains specifc liability exclusions disallowed under German Law. Under German law, K4K clauses in standard form industry contracts, such as BIMCO SUPPLYTIME, will be considered as general terms and conditions unless the particular clause has been individually negotiated between parties. The term “negotiating” as per the German Civil Code requires the user of general terms and conditions to allow the contractual partner to actively protect his own interests and demonstrate his willingness to negotiate terms in principle. Previously, “rider clauses”, pursuant to a contractual agreement, including a reasonable and fair allocation of rights and duties, and highlighting the known nature of every clause as part of the individual negotiations, were accepted in German jurisprudence. Later case law shows an understanding, in the German federal courts, that the inclusion of a rider clause does not automatically prevent a contract from still being regarded as general terms and conditions, because whether a contract is agreed on the basis of general terms and conditions or individually is a question of fact, which cannot be contractually regulated by the parties; hence such a rider may still be considered to be part of the general terms and conditions.50 The liability regime of a K4K clause is contrary to any German civil liability regime and thus considered foreign by many courts. Such contracts containing a K4K clause must ensure fairness between the parties to keep these particular clauses from being invalidated by German courts. The main deterring factor in a court’s decision would be the consideration of the practices and customs that apply in general German business dealings, for which the ofshore wind industry’s rather young “status quo” would likely not be a determinative factor. Hence it is likely that a German court or arbitral tribunal will render a K4K clause invalid if it attempts to exclude liability for damages arising from personal injury to life and body or health, wilful or intentional

49 G. Mendes Vianna, J. Furtado Senna, “Knock for Knock Clauses under Brazilian Law” (ofshore charter contracts), Standard Bulletin: Ofshore Special Edition, November 2011. 50 E. Volz, A. Waldmann, “Knock-for-Knock Liability – Does It Work in the German Market?”, 3 September 2019, https://www.mondaq.com/uk/contracts-and-commercial-law/842380/knock-for-knockliability--does-it-work-in-the-german-market, accessed 18 October 2022.

12

development of knock-for-knock clauses in the last 15 years misconduct and/or gross negligence. Consequentially, in such circumstances, a court or arbitral tribunal may not only consider the K4K clause but rather the entire clause to be invalid, even if only a part of such clause is afected. In efect, in case of any damage or loss, as per the German statutory liability regime of the German Civil Code (which adopts a fault-based liability regime), the party claiming damages would be entitled to damages without any limitation, provided the other party was at fault. Under German law, certain restrictions apply to all contract clauses, providing for the provision on civil liabilities. If the exclusion of liability is for damages resulting from wilful or intentional misconduct, even if the clause has been individually negotiated, the clause will be invalidated, unless a carve-out regarding exclusions and limitations of liability is in place to be used for this particular type of damages.51 7.5 Australia In Australia, in commercial contracts, K4K clauses are construed in accordance with ordinary rules of contractual construction. In Darlington Futures Ltd v Delco Australia Pty Ltd,52 the High Court of Australia held that an exclusion clause is to be construed according to its “natural and ordinary” meaning; and in Andar Transport Pty Ltd v Brambles Ltd,53 the High Court applied the traditional rule of construction: that indemnities should be construed strictly against the guarantor. 7.6 Nordic countries – the history and validity of K4K clauses from a comparative perspective In Nordic countries, the validity of K4K clauses, and the liability exclusions that they contain, is often analysed comparatively, between the law of their tradition of origin, i.e., common law, especially English law; and Nordic civil law, especially Norwegian and Danish law, where such agreements are also frequently used, namely in the context of oil extraction activities in the North Sea.54 The subjective characteristics of any K4K arrangement depends on each jurisdiction’s respective stance, hence this becomes one of the most important issues for the contracting parties during their contractual negotiations, as the establishment of which national law applies will also determine the applicability parameters of the contracts, and the extent and treatment of liability. Standard K4K clauses often establish that the K4K liability allocation applies unless the party causing the loss has acted with gross negligence or wilful misconduct. Furthermore, the insurer might be able to exclude such qualifed forms of faulty behaviour from the ofered coverage, so that the party causing the damage is contractually exempted from liability but cannot recover its loss from the insurer or require the claiming party to frst pay a deductible.55 Also, certain types of damages, such as 51 Ibid. 52 Darlington Futures Ltd v Delco Australia Pty Ltd (1986) 161 CLR 500. 53 Andar Transport Pty Ltd v Brambles Ltd (2004) 217 CLR 424. 54 S. Cavaleri, “The Validity of Knock-for-Knock Clauses in Comparative Perspective” (3 October 2017), European Review of Private Law, Vol. 25, 2017, University of Copenhagen Faculty of Law Research Paper No. 2017–50, https://ssrn.com/abstract=3063280, accessed 27 May 2021. 55 See Parchomovsky and Stavang, Chapter 3 of this book.

13

kyriaki noussia and hanieh bolourian pollution damages, may warrant exclusion from the K4K regime, and may simply be governed by the usual principles of tort liability, since allocation of liability for those certain types of damages is imposed by mandatory law.56 Contract validity is dependent on whether the parties had intended that a certain behaviour of the responsible party, or a certain type of loss, is to be covered by the exclusion. Furthermore, the liability exclusion or limitation must be compatible with statute and public policy. English courts have traditionally assumed that, unless the contract contains clear wording to the contrary, an exclusion or limitation of liability does not cover negligence, and have ruled that it is “established that indemnity will not lie in respect of loss due to a person’s own negligence or that of his servants unless adequate and clear words are used or unless the indemnity could have no reasonable meaning or application unless so applied”.57 In Canada Steamship Lines v Regem,58 Lord Morton developed a three-step contract interpretation test in determining the criteria for parties to be relieved of their negligence or the negligence of their “servants”. The applicability of this three-step test was later confrmed by the Queen’s Bench Division of the High Court of England and Wales in their interpretation of the ofshore indemnity clauses in E. E. Caledonia Ltd v Orbit Valve Co Europe.59 This case concerned the question of whether a contract containing a mutual indemnity clause entitled the operators of an oil platform to an indemnity for the compensation paid to the descendants of an engineer who had died because of a fre on the platform caused, at least in part, by the negligence of the platform operators. The court held that the operators were not entitled to indemnifcation because the indemnity clause did not include a reference to negligence or any similar word, and that the mere reference in the disputed clause to “any” claim or liability was not sufcient. In Danish law, with regards to liability exclusions and limitations, the validity of a contract depends on a court’s interpretation of four contractual factors (1) whether a clause has not been validly adopted by the parties;60 (2) whether a clause is restrictively interpreted, sometimes to the efect that the clause does not apply at all (e.g., contra proferentem, wherein the party who drafted the clause must sufer the consequence of its lack of clarity);61 (3) whether the theory dictating the consequence of fundamental changes of circumstances (forudsætningslære) applies; and (4) whether circumstances justifying the application of the clause are no longer present, or that it would be contrary to the parties’ expectations to apply the clause in accordance with its wording.62 The Danish Product Liability Act contains a prohibition against clauses purporting to exclude or reduce liability for death or personal injuries falling 56 T-L. Wilhelmsen, “Liability and Insurance Clauses in Contracts for Vessel Services in the Norwegian Ofshore Sector – the Knock for Knock Principle”, SIMPLY 2012, 88. 57 Alderslade v Hendon Laundry [1945] KB 189 at 192; see also Walters v Whessoe Ltd and Shell Refning Co Ltd [1968] 2 All ER (All England Law Reports) 816 per Sellers SJ. 58 Canada Steamship Lines v Regem [1952] 1 All ER 305, § 16 to 19. 59 E. E. Caledonia Ltd v Orbit Valve Co. Europe [1994] 1 WLR (Weekly Law Reports) 221. 60 B. Gomard, “Obligationsret” 2. del, 4th ed., 2011, 311–312; MB. Andersen, J. Lookofsky, “Lærebog i obligationsret I”, 4th ed, 2015, 411. 61 B. Gomard, “Obligationsret” 2. del, 4th ed., 2011, 313–314; MB. Andersen, J. Lookofsky, “Lærebog i obligationsret I”, 4th ed, 2015, 411. 62 MB. Andersen, J. Lookofsky, “Lærebog i obligationsret I”, 4th ed, 2015, 413–415; V. Hagstrøm, “Obligasjonsrett”, Universitetsforlag, 2nd ed. 2011, 658.

14

development of knock-for-knock clauses in the last 15 years within the scope of application of this Act, i.e., caused by products within the meaning of the Act. Furthermore, common to all Nordic countries, a clause can be invalidated on the ground of unreasonableness pursuant to § 36 (the “general clause”) of the Contract Act (CA) which states that a contract can be modifed or set aside, in whole or in part, if it would be unreasonable or incompatible with the principle of good faith to enforce it.63 The validity of a clause, and the gravity of the fault leading to damage (gross negligence or intentional breach), are the main focus of relevant Danish court decisions; whereas Norwegian court decisions also place an emphasis on the hierarchical position of the person who caused the damage, as gross negligence by a member of management can be treated diferently than similar behaviour by a subordinate employee.64 7.7 Later judicial trends Especially from the early 2010s, oilfeld operators have shown a real tendency to chisel away at the K4K indemnity model, or abandon it altogether, to allocate more responsibility to the involved service companies for all sorts of well damages, e.g., pollution, blowouts/wild well events, loss of oil/gas, to the extent that such damages are caused by the negligence of the service company. Arguably, one way to do this, i.e., to expand the range of service companies’ liabilities, is to embed a “gross negligence” exception provision in mutual hold harmless indemnity agreements.65 In terms of contracts, many industry players use model form contracts produced by not-for-proft industry associations such as the LOGIC standard form model contract. LOGIC publishes a set of “standard” contracts, and in most of these contract models, the “mutual hold harmless” clause is incorporated.66 Often, the “mutual hold harmless” clause and the allocation of liability vary between diferent types of contracts and may serve one party more than the other, depending on the circumstances under which the contract terms were negotiated. International and state oil companies usually have certain global policies for insurance in place, which exceed the minimum requirements set by applicable jurisdictions. However, the beneft of such “mutual hold harmless” clauses is – given that there is a clearer and simpler allocation of liability in K4K – that liability becomes a matter of contract. Moreover, there is no need for an investigation or assignment of fault, allowing parties to avoid lengthy litigation. In the absence of such a regime, each contractor will need to insure against all risks, e.g., destroying the whole facility or injuring each one of the personnel. Often, one aspect to consider while drafting the “mutual hold harmless” indemnity clause is to defne whose property is intended to be covered, i.e., whether it is property

63 Section 36, section 1 of the Danish Contract Act (Aftaleloven), free translation. “36.-(1) A contract may be modifed or set aside, in whole or in part, if it would be unreasonable or at variance with the principles of good faith to enforce it. The same applies to other juristic acts.” https://pure.au.dk/ws/fles/32350272/ Danish_20Contracts_20Act.pdf, accessed 18 October 2022. 64 V. Hagstrøm, “Obligasjonsrett”, Universitetsforlag, 2nd ed. 2011, 658. 65 S. Johanson, “Is It Really Knock-for-Knock?” 21 June 2019. Boyar Miller, www.boyarmiller.com/ is-it-really-knock-for-knock/, accessed 27 May 2021. 66 Ibid.

15

kyriaki noussia and hanieh bolourian or equipment rented to the operator by the contractor, or if it is under the contractor’s or supervisor’s control but it is operated by the employees of the contractor.67 As of late, the “mutual hold harmless” regime is less commonly used due to changes in the infrastructure of oil and gas operations, and developments in insurance attempting to adapt to the complexity of oilfeld operations and their inherent risks. Historically, a key factor in the emergence of the mutual indemnity regime has been access to insurance. Previously, insurance policies were not well suited for the complexity and risks associated with oil and gas exploitation. However, as the insurance market developed a deeper understanding of the oil and gas industry, insurance policies became more comprehensive, and as a result, afected the industry’s preferred allocation of liability. Moreover, the recent shift in collaboration between operators and contractors has led to the necessity for a diferent approach to risk and liability allocations. With collaborative projects, tasks are tackled through bringing in specifc skilled contractors, and if using a diferent reward system, rather than basic reimbursement, allocation of liability would be diferent. Cases such as Transocean Drilling UK Ltd v Providence Resources PLC68 highlight the importance of tailoring the indemnity regime to a specifc project, rather than using a standardised version of the liability system in use.69 7.8 The way forward The 2016 Court of Appeal (EWCA) decision in Transocean Drilling UK Ltd v Providence Resources PLC70 brought about signifcant discussion on K4K provisions in oil and gas contracts. The case turned on the interpretation of the mutual indemnity provisions, covering third-party losses, and consequential loss carve-outs within the contract. The provisions in dispute were relatively typical in such contracts and have been used in the ofshore industry for well over 25 years. Originally, these contracts were used in recognition of the fact that costs from certain loss events, such as the destruction of an entire oil platform, could not be borne by a single contractor. In addition, the manner in which ofshore work has been carried out, historically, has meant a signifcant overlapping of roles and responsibilities among groups of contractors. Furthermore, a key justifcation in introducing mutual indemnity provisions was the availability of insurance, and where such insurable risks lay. In April 2011, Providence had contracted with rig owner Transocean for the hire of a semi-submersible drilling rig to be deployed in the Barryroe Field, Ireland, to drill an appraisal well. The High Court held that the rig had not been delivered in good working condition, and that Transocean was in breach of contract for the resulting delays. Transocean’s appeal, however, focused solely on the High Court’s decision to allow Providence to recover the “spread costs” incurred as a result of the delay, which were the usual wasted costs of third-party personnel, equipment and services. Transocean contended that their liability for such costs was excluded under the contract’s standard

67 Ibid. 68 Transocean Drilling UK Ltd v Providence Resources PLC [2016] EWCA Civ 372. 69 P. Murray, “Is It Time for ‘Knock-for-Knock’ to be Knocked Out?” 12 May 2018, News and Views, Ledingham Chalmers Solicitors. 70 Transocean Drilling UK Ltd v Providence Resources PLC [2016] EWCA Civ 372.

16

development of knock-for-knock clauses in the last 15 years consequential loss clause. The Court of Appeal reviewed the risk management and liability provisions in the contract, focusing on the complexity of the series of indemnities, noting that such provisions were clearly designed to complement each other and were efectively a detailed and sophisticated scheme for apportioning responsibility for loss and damage of all kinds, backed by insurance. The Court also closely examined the wording of the disputed language, including critical words such as “loss of use” (including, without limitation, loss of use or the cost of use of property, equipment, materials and services including without limitation, those provided by contractors or subcontracts of every tier or by third parties). The Court also found that the use of the phrase “without limitation” twice within the clause clearly indicated the parties’ intention to emphasise the width of the limitation, and that this wording was “plainly apt” to cover the “spread costs” claimed by Providence. The Court of Appeal disagreed with the High Court’s reasoning in reaching its decision and went on to note that a court can reinterpret the contract, so long as the parties’ intent is preserved, giving the words that they have used their ordinary and natural meaning. In so ruling, the court repeatedly emphasised the choice of the parties to accept responsibility for losses that might have otherwise been recoverable as damages for breach of contract. In viewing the contract as a whole, the court stated that, in their view, there is no reason to not give efect to the agreement. The Court of Appeal reafrmed that parties can be rightfully bound by properly executed K4K agreements. As a consequence, parties must be prepared to accept that, by the operation of these contracts, claims which may have otherwise been recoverable at law may well be excluded, as was the matter in this case.71 It is also important to note that over the last few years, there have been signifcant changes in the ownership of key pipelines and terminals moving to independent ownership. Consequentially, there has been a shift in operators’ willingness to undertake certain risks with some commentators predicting the end of the K4K regime. However, this demise will likely not happen anytime soon, and K4K clauses remain viable, so long as proper drafting and insurance arrangements can be made between parties, negotiating in good faith.72 8 Conclusive critique It is an established rule that clauses which tend to exclude liability must be clear and precise; courts can also hold that such clauses, where they are exclusionary, must employ specifc and express terms. Limitation clauses are distinguishable from exclusion clauses, and so should indemnity clauses be too, even if backed by emphatic words, but especially where there is established usage/practice. The word “indemnify” logically suggests initial liability, usually to a third party. “Hold harmless” goes further; an explicit statement of what the party obliged to indemnify the party at loss is to do (hold them harmless). This chapter is meant to highlight the concept of mutual indemnifcation in UK ofshore oil industry contracts. The use of indemnity 71 C. Kumarasinghe, “Case Update: Transocean v Providence”, May 2016, HFW, www.hfw.com/Transocean-Drilling-UK-Ltd-v-Providence-Resources-PLC-May-2016, accessed 27 May 2021. 72 P. Murray, “Is It Time for ‘Knock-for-Knock’ to Be Knocked Out?” 12 May 2018, News and Views, Ledingham Chalmers Solicitors.

17

kyriaki noussia and hanieh bolourian clauses has become underscored with words such as “save”, “defend”, “hold harmless” and so forth. Nevertheless, it is the position argued here that the meaning of the clause among industry parties remains: the reciprocal indemnifcation of liability in accordance with contract terms. This becomes more important when the interests of a third party to the contract are involved. Where parties intend to exclude liability, then none exists ab initio; fault will lie where it falls, and there will be no need to employ the word “indemnify”.73 The contractual practice of knock-for-knock indemnities has been reviewed by the English courts. In a number of cases, they have examined the meaning of key concepts such as “negligence”, “gross negligence” and “wilful misconduct” in the context of the enforcement of liabilities. Their analysis of such contracts begins with the presumption that a party to a contract is unlikely to intend to absolve another party entirely from the consequences of the latter party’s negligence. It is therefore important to provide an indemnity that expressly covers liability in negligence. In E. E. Caledonia Ltd (formerly Occidental Petroleum (Caledonia) Ltd) v Orbit Valve Co Europe74 and in Caledonia North Sea Ltd v London Bridge Engineering Ltd,75 the contractor’s liability arising from negligence had to be clearly provided for in the indemnity clause or it would not be so construed. Although the K4K indemnity structure is produced by contract parties, it has been noticed by the courts within the context of ofshore drilling contracts. In Caledonia North Sea Ltd v London Bridge Engineering Ltd,76 the court accepted this as an industry practice which was known to, and accepted by, the courts. In essence, the House of Lords noted that the indemnity said nothing about requiring the contractors to be liable to their employees in order for the indemnity to operate; and that it imposed a general liability to indemnify, with an exception only in cases where the accident was attributable to the sole negligence or wilful misconduct of the operator. Otherwise, courts have had little difculty in upholding indemnity provisions which were clearly worded, and in which the parties had made express provisions for the allocation of liability. In justifying this position, the court approached the mutual indemnity structure, not just as a risk allocation mechanism, but also as a tool through which the real intentions of the contract parties could be discerned, and upheld, especially when the wording of the indemnity provisions is clear and unambiguous.77 It is also noted that the wording of indemnity clauses is notoriously uncertain with respect to their enforceability. Insurance can also be inadequate or even non-existent. The result is that substantial residual liability can rest with the contractor. The insurance market post–Deepwater Horizon appears to be nervous about perceived contractor liability risk, especially for gross negligence, and capacity is less than $1 billion and increasingly expensive. Insurance also carries general exclusions for catastrophic

73 C. Ugwuanyi, “Examining the Exclusionary Nature of Oil and Gas Contract Mutual Indemnity Hold Harmless Clauses”, I.E.L.R. 2012, 4, 136–146. 74 E. E. Caledonia Ltd (formerly Occidental Petroleum (Caledonia) Ltd) v Orbit Valve Co Europe Plc [1994] 1 W.L.R. 221; [1993] 2 Lloyd’s Rep. 418 QBD (Commercial Court). 75 Caledonia North Sea Ltd v London Bridge Engineering Ltd [2000] S.L.T. 1123. 76 Ibid. 77 P. Cameron, “Liability for Catastrophic Risk in the Oil and Gas Industry”, I.E.L.R. 2012, 6, 207–219.

18

development of knock-for-knock clauses in the last 15 years risks from the well, such as blowout and pollution, which are traditionally viewed as operator risks, indemnifed to the contractor. With regards to the oil and gas industry, in practice, the latter has what is essentially a global template for the allocation of liability, evident in much of the standard documentation in the industry, e.g., LOGIC and IADC standard contracts. Many internationally operating oil companies have master services agreements with contractors, which can often be global in character. The principal exception to this is the practices of national oil companies. Such model contract templates provide a balance between risk and reward and “command and control” realities. However, it is of note that recent regulatory actions against contractors are throwing this balance substantially of.78

78 Ibid.

19

CHAPTER 2

An introduction to risk allocation in oil and gas contracts from an English law perspective1 Greg Gordon

1 Introduction The oil and gas industry is a hazardous one. The activities involved in exploring for, producing, transporting and processing volatile hydrocarbons are attended by a whole host of risks: to people, property, the environment and the valuable commodity itself. These activities also frequently require co-operative working by large numbers of contractors whose interests may be adversely afected in the event of an accident causing damage to property or personnel, or economic loss, and the commercial framework of contracts governing these operations is complex. Not all of these contractors will be in a direct contractual relationship with each other, and not all of them will necessarily be in a direct contractual relationship with the operator. The oil and gas industry has developed a number of contracting practices to allow it to manage these physical2 risks. Generally speaking, up- and midstream oil and gas contracts seek to depart quite radically from the common law’s presumptions about how such risk should be allocated.3 As Parchomovsky and Stavang note in Chapter 3 of this work, “the gas and oil industry has largely opted out from standard tort liability”. Three vehicles are commonly used to achieve this re-allocation of risk: (1) indemnity and hold harmless clauses; (2) clauses which exclude or limit liability for what are commonly, if rather loosely, described as “consequential losses”; and (3) overall limitations on liability. Each will be discussed in turn.

1 This chapter is based upon G Gordon, “Risk Allocation in Oilfeld Service Contracts”, in G Gordon, J Paterson and E Usenmez, UK Oil and Gas Law: Current Practice and Emerging Trends Volume II: Commercial and Contractual Aspects, 3rd Edition, Edinburgh University Press, 2018, pp. 175–234, and is reproduced with kind permission of Edinburgh University Press. 2 This chapter is concerned with risk of accident and is not concerned with other risk factors such as political or geological risk. 3 The terminology used to describe these provisions varies. Sometimes they are called “risk allocation” provisions and sometimes “liability allocation” provisions. Little turns on this, but throughout this chapter, the expression “risk allocation” shall be preferred simply because at the point when the contracts are drafted and entered into, the parties are looking at the matter prospectively and are concerned with the potential consequences of events which may or may not come to pass. While the risk of future liability is inevitable, actual existence of liability is not; this only crystallizes after a loss-causing event has occurred some time after the contract entered into force.

20

DOI: 10.4324/9781003206798-2

risk allocation in oil and gas contracts in english law 2 Indemnifcation and related concepts 2.1 Introduction to the concept of indemnifcation Courtney defnes a contractual indemnity as “a promise to protect another against loss from an event or events, or set of circumstances”.4 An alternative formulation previously ofered by the present author is that a contractual indemnity5 is “a contractual provision whereby the indemnifer agrees to make a payment to the indemnifed party in the event that the indemnifed party sufers loss as a result of the occurrence of a specifed event”.6 Although there is considerable commonality between the two defnitions, they difer in at least two respects. Courtney’s defnition relies upon promissory theory, whereas the alternative does not;7 and Courtney suggests that the obligation is to “protect” against the fnancial losses associated with an event. This may involve the payment of a sum or money by the indemnifer to the indemnifed, but may also embrace some other form of prevention: for instance, the waiver of a claim that would otherwise have existed. The present author’s formulation, by contrast, is focused upon payment, meaning that if some form of protection beyond payment is required, this must be obtained by a device other than indemnity. It is submitted that although Courtney’s defnition is consistent with the way that indemnity has tended to be expressed, the present author’s formulation is more consistent with the approach adopted in the leading UK Supreme Court case on the specifc risk allocation provisions used in the oil and gas industry.8 If this is correct – and it must be acknowledged that the conclusion builds out from only one case, albeit a Supreme Court one – then it means that in order for a risk allocation clause to operate so as to deal with both waiver and payment, the obligation needs to be expanded out from a mere “indemnity” by using words such as “indemnify and protect” or (as is more common in the oil and gas industry) “indemnify and hold harmless”. The signifcance of this will be discussed further below. As we shall see below, indemnifcation is a contractual device which the courts have at times treated with considerable suspicion. The fact that the device subverts – in fact, frequently inverts – the common law’s default approach to liability has caused concern.9 It is nevertheless a commonly encountered legal concept, by no means particular to oil and gas contracts. Indemnifcation lies at the heart of the law of marine10 and fre11 insurance. It also features in construction contracts, where contractors will

4 W Courtney, Contractual Indemnities, Hart, 2014, para 1–2. 5 Not all indemnities are contractual: some arise as a result of operation of law. See Courtney, above n 4, paras 1.5–1.8. 6 G Gordon, “Risk Allocation in Oil and Gas Contracts”, in G Gordon, J Paterson and E Usenmez, Oil and Gas Law: Current Practice and Emerging Trends, 2nd Edition, Dundee University Press, 2011, para 14.3. 7 For a theoretical account of contract which recasts traditional promissory theory in terms of consent, see R. Barnett, “Contract Is Not Promise: Contract is Consent”, in G Klass, G Letsas and P Saprai (eds), Philosophical Foundations of Contract Law, Oxford, 2014, 42–57. 8 Farstad Supply AS v Enviroco Ltd [2010] UKSC 18. 9 I refer here to the common law simply because English law is the focus of this chapter. It is not intended to imply that civilian jurisdictions are less likely to exercise scrutiny of a knock-for-knock indemnity. See, e.g., the discussion of German law in Chapter 10 of this book. 10 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.” 11 Castellain v Preston (1883) 11 QBD 380, per Brett LJ at 386.

21

greg gordon commonly provide their employer with indemnities against personal injury or death, or damage to property, in any way associated with the work contracted for,12 as well as in the aviation and shipping industries.13 The concept will also be familiar to the company lawyer: in a corporate acquisition and disposal, the seller will frequently be asked to provide an indemnity in respect of liabilities incurred by the company in the period between the deal’s conclusion and its completion.14 Many other examples could be given.15 For the remainder of this chapter, the one-sided16 indemnity clauses just described will be referred to as simple indemnity clauses in order to diferentiate them from mutual indemnity clauses, which will be discussed in greater detail below. 2.2 Indemnity and hold harmless clauses As we shall see in greater detail below, the oil and gas industry makes extensive use of a particular form of wording in its risk allocation clauses. A party will very often ofer not merely to indemnify but to “indemnify and hold harmless” the other party.17 Before the UK Supreme Court decision in Farstad Supply AS v Enviroco Ltd,18 commentators on the industry’s risk allocation practice considered that the words “and hold harmless” added little to the content of such clauses, which were often described simply as “indemnity clauses”.19 It is submitted that this approach was both rooted in the long-standing legal usage of the terms “indemnity” and “hold harmless”20 and accurately mirrored that 12 See the Joint Contracts Tribunal Standard Form of Contract With Quantities available for download from www.jctltd.co.uk/product/standard-building-contract-with-quantities, (accessed 15 December 2022) cll 6.1 and 6.2 respectively. For an illustration of cl 6.1.2 in operation, see Scottish & Newcastle plc v G D Construction (St Albans) Ltd [2003] EWCA Civ 16. 13 In this connection, see, e.g., S Rainey, The Law of Tug and Tow and Ofshore Contracts, 3rd Ed, Informa, 2011, pp. 155–196. 14 See, e.g., J Young and J Kitching, “Buying and Selling a Business: Warranties and Indemnities” 1995 6(10) ICCLR 336. 15 For instance, one of the leading cases on the construction of indemnities is concerned with a lease: Canada Steamship Lines v The King [1952] AC 192. 16 The term “one-sided” is here used not to suggest that simple indemnity clauses are necessarily unfair or biased, merely to denote the fact that in a simple indemnity clause, the indemnity travels in one direction only – from the indemnifer to the indemnifed party. No reciprocal or mutual indemnity is provided. The structure is simple in that (as distinct from a mutual indemnity clause) one can tell in advance who will be the indemnifed and who the indemnifer, in the event that the clause is triggered. Such certainty is not possible in the case of a mutual indemnity, where the roles of the respective parties will not be known until after an accident has happened and where the losses lie has become evident. 17 That said, clauses which use the word “indemnify” are not unknown in an oil and gas context, particularly in contracts relating to maritime matters such as towage: see, e.g., TOWCON cl 18 referred to in A Turtle Ofshore SA v Superior Trading Inc [2008] EWHC 3034 (Admlty) [2008] 2 CLC 953, a case concerning the towage of an ill-fated drilling rig. 18 [2010] UKSC 18, 2010 SCLR 379 (hereinafter Farstad v Enviroco). 19 See, e.g., G Gordon, UK Oil and Gas Law (Dundee, 2007), chapter 13; G Gordon, “Indemnifcation and Contribution: Farstad Supply AS v Enviroco Ltd” (2010) 14 Edin LR 102; T Hewitt, “Who Is to Blame? Allocating Liability in Upstream Project Contract” (2008) 26 JENRL 177 at 182; T Daintith, G Willoughby and A Hill, United Kingdom Oil and Gas Law (3rd edn, looseleaf, 2000–date), para 1–845; D Sharp, Ofshore Oil and Gas Insurance (1994) at p. 108. 20 See, for instance, Firma C-Trade SA v Newcastle Protection and Indemnity Association (The Fanti) (No 2) [1991] 2 AC 1 (HL) per Lord Gof at 35: “a promise of indemnity is simply a promise to hold the indemnifed person harmless against a specifed loss or expense.” Courtney, for example, considers the expressions “indemnity”, “save harmless” and “keep harmless” to all fulfl the same function within the indemnity clause. See further Courtney, para 8–11.

22

risk allocation in oil and gas contracts in english law of the oil and gas industry. The oil and gas industry routinely used the terms “indemnify”, “hold harmless” and “indemnify and hold harmless” interchangeably in its risk allocation arrangements. Thus, in the LOGIC Standard Contracts for the Oil and Gas Industry, clauses by which a party indemnifes and holds harmless the other are described simply as “Indemnities”.21 Similarly, while the industry’s attempt to put in place a contractual risk allocation regime between ofshore contractors, who would not otherwise have a contractual relationship, is known throughout the industry as the Industry Mutual Hold Harmless Deed, the Deed is formally titled the “Mutual Indemnity and Hold Harmless Deed”.22 Although the Deed’s central risk allocation clause uses the wording “indemnify and hold harmless”, the clause is titled “Indemnities by the Signatories”23 and the Deed is itself referred to in the Deed of Adherence (by which parties other than the original signatories can enter the scheme) as the “Indemnity Deed”. However, in Farstad v Enviroco, the Supreme Court held that a clause whereby the owner of a vessel under charter agreed to “indemnify and hold harmless” the charterer against all liability resulting from loss of or damage to the vessel was not a mere indemnity clause but a mixed risk allocation provision containing elements of (on the one hand) indemnity and (on the other) an exclusion or waiver of liability.24 Whether it operated as an indemnity or an exclusion would depend upon whether the clause sought to determine who was to bear responsibility for “third-party exposure” (in which case the clause would be an indemnity) or whether it resolved “direct exposure to the other contracting party” (in which case the “hold harmless” dimension would be activated, and it would be an exclusion or waiver of liability).25 On the facts of the case in question, the owner had sufered damage to his own property. The case was therefore seen by the Supreme Court as one of “direct exposure”; hence the clause was, on this occasion, to be seen as an exclusion of liability clause.26 This would suggest that there will be occasions when such clauses will have to comply with the provisions of the Unfair Contract Terms Act 1977.27

21 See, e.g., LOGIC, Supply of Major Items of Plant and Equipment (3rd edn), available online at www. logic-oil.com/content/standard-contracts-0 (accessed 27 April 2017), cl 21. 22 LOGIC, Mutual Indemnity and Hold Harmless Deed, available for download from www.logic-oil. com/imhh/documents (accessed 2 April 2022). 23 Ibid., cl 2. 24 In so holding, their Lordships laid considerable emphasis upon the fact that the parties to the contract had titled their clause “Exceptions/Indemnities”. This, thought Lord Clarke (delivering a speech concurred in by Lord Phillips), was a feature of “particular importance”, and strong evidence of the parties’ intentions (Farstad v Enviroco per Lord Clarke at para 22; see also Lord Mance at para 56). However, their Lordships seem to have failed to notice that there were a number of indications of a contrary intention within the clause, including the obligation to exchange “mutual hold harmless indemnities” with other parties in certain circumstances. This wording might tend to suggest that the words “hold harmless” were intended only to describe a particular type of indemnity clause. Speculation is to an extent idle, but given the importance apparently attached to the title of the clause, one cannot help but wonder how their Lordships’ decision would have difered had the parties followed the form of the LOGIC contracts and titled their clause “Indemnities”. 25 Farstad v Enviroco per Lord Mance, para 59. 26 Ibid. Lord Clarke at para 29 and per Lord Mance at para 59. 27 Although the Unfair Contract Terms Act 1977 has been superseded by the Consumer Rights Act 2015 in the context of consumer contracts, UCTA continues to govern the regulation of exemption clauses in the business to business context. See, e.g., UCTA s.2(4).

23

greg gordon These clauses will frequently be supplemented by a provision stating that the indemnifer28 will not just indemnify and hold harmless the indemnifed party, but also defend claims taken against the indemnifed party.29 This has the efect of imposing upon the indemnifer the burden of conducting the defence of any litigation that may arise, but also of conferring upon the indemnifer the right to control the manner in which the defence is conducted. Many indemnifers consider that the beneft of the right to control the conduct of the defence outweighs the burden of conducting it. Best practice is now thought to be not to rely solely upon the word “defend” but to include a conduct of claims clause expressly stipulating the way in which claims are to be handled. Such a clause is absent in the present draft of the LOGIC standard form contracts, but is commonly revised into contracts based upon the LOGIC standard forms. 2.3 Mutual indemnity and mutual indemnity and hold harmless clauses A mutual indemnity – sometimes also called a “reciprocal indemnity”, a “crossindemnity” or a “knock-for-knock” indemnity – is a contractual device where the parties with the one hand give and with the other hand take an indemnity in respect A accepts liability for its (A’s) people & property and grants B an indemnity & hold harmless provision in respect of any losses B may suffer in respect of those people and property, howsoever caused

A

B B accepts liability for its (B’s) people & property and grants A an indemnity & hold harmless provision in respect of any losses A may suffer in respect of those people and property, howsoever caused

Figure 2.1 Diagrammatic representation of a mutual indemnity and hold harmless provision relative to personal injury. 28 Although, as we have seen, the Supreme Court held in Farstad v Enviroco that the words “hold harmless” add an additional element to an indemnity clause, meaning that indemnity and hold harmless clauses will in certain circumstances operate not merely as indemnities but also as exclusion clauses, for the sake of brevity the parties giving and receiving these clauses will be described as the “indemnifer” and the “indemnifed” throughout this chapter. 29 See the observations by Lord President Rodger in the Inner House phase of Caledonia North Sea Ltd v London Bridge Engineering Ltd 2000 SLT 1123 at 1155. The standard contracts for the oil industry developed as part of the CRINE initiative and now maintained by LOGIC contain such a provision: see, e.g., LOGIC, General Conditions of Contract for Services (On- and Of-Shore) (3rd edn, 2014), available for download from www.logic-oil.com/content/standard-contracts-0 (accessed 27 April 2017) (hereinafter LOGIC, Services) at cl 19.1: “The contractor shall . . . Save, indemnify, defend and hold harmless”.

24

risk allocation in oil and gas contracts in english law of related species of loss. A simple example relative to liability for loss arising from personal injury to employees is given in Figure 2.1. A mutual indemnity therefore difers from a simple indemnity, where one party consistently has the burden of giving the indemnity (i.e., is the indemnifer) and the other party consistently has the beneft of being indemnifed. In a mutual indemnity, each party is simultaneously both an indemnifer (in relation to one species of loss) and the indemnifed (in relation to a diferent, but related, species of loss). In the example given above, both parties are potentially the benefciaries and the recipients of an indemnity relating to loss arising from personal injury to employees. Which (if any) of the parties eventually bears liability as the indemnifer or which claims as indemnifer will depend upon (1) whether an accident occurs and (2) whose employee is hurt.30 It is important to appreciate that, to be efective, a mutual indemnity or mutual indemnity and hold harmless clause must not be drawn so as to provide that each party indemnifes (or indemnifes and holds harmless) the other against the occurrence of exactly the same loss. To illustrate the point by way of example, let us imagine that A and B enter into an arrangement where A grants B an indemnity against B’s house burning down, and B also grants A an indemnity against B’s house burning down. In such a situation, all the clause succeeds in achieving is a position where the losses arising if the house burns down is passed from one party to another ad infnitum. Such a clause (sometimes described as a “circular indemnity”31) is inefectual32 and leaves the risk it purports to allocate to be borne by the parties in the way provided for by the law at large. If, however, the example is altered so that A indemnifes B in respect of the losses incurred by B if B’s house burns down, and B in turn indemnifes A against the losses that A sufers if A’s house burns down, circularity is avoided. This is because the species of loss in respect of which the indemnity is given, although conceptually related (both pertain to the losses sufered when houses burn down), are not exactly the same: A agrees to accept the losses to B’s property, and B agrees to accept the losses to A’s property. 3 Indemnity and hold harmless provisions in the oil and gas context 3.1 Introduction to simple indemnity and hold harmless clauses in oil and gas contracts Simple indemnity clauses are used in oil and gas contracts in at least two ways. First, the petroleum industry sometimes just provides the commercial context for

30 In the oil and gas context, it is usual for the parties to enter into not just mutual indemnity provisions but into mutual indemnity and hold harmless clauses, thus bringing into play the further conceptual issues described at section 2.3, above. 31 Slessor v Vetco Gray, unreported, 7 July 2006, Court of Session, Outer House, available for download from www.scotcourts.gov.uk/search-judgments/judgment?id=2fbb86a6-8980-69d2-b500-f0000d74aa7 (accessed 15 December 2022). See the submissions of counsel summarised by Lord Glennie at para 6. On the facts, the court rejected the argument that the indemnity was circular. 32 It is also commercially unrealistic: why on earth would B indemnify A against the loss of B’s own house? However, the example is given because in practice one does, from time to time, encounter circular indemnities – almost invariably they arise by accident, when something has gone wrong in the drafting of the indemnity clause.

25

greg gordon the kind of situation where an indemnity would be commonplace.33 Oil and gas contracts therefore commonly contain a number of simple indemnity clauses of a type no diferent to those routinely found in commercial agreements. In addition, some, but by no means all, up- or midstream oil and gas contracts will also contain one or more simple indemnity and hold harmless clauses designed to allocate between the parties some of the risk factors specifc to the petroleum industry. In drilling34 or well services contracts,35 the indemnity and hold harmless provisions are often more complex than those associated with other operator-to-contractor contracts. The operator will frequently provide a simple indemnify and hold harmless provision in favour of the contractor, protecting against risks such as loss of or damage to the hole,36 blowout, fre, the well becoming uncontrollable, or damage to the reservoir, geological formation or underground strata, howsoever caused. The courts have sometimes viewed indemnity clauses with suspicion, on the basis that when they are found in a contract, this is because a dominant party has imposed them upon a weaker one.37 However, in UKCS operator-to-contractor agreements, one-sided indemnity and hold harmless clauses are most commonly granted by the operator to the contractor.38 Such indemnities are not given because the operator is weak, but because it is strong. The losses that could accrue in the event that the well is lost or damaged are potentially very substantial – so large that it might not be economic, or perhaps even possible, for contractors to obtain insurance against these contingencies. However, the operator requires the well to be drilled if he is to produce from the discovery, and is – or traditionally has been; given the changing face of the industry, this may no longer be universally true – of a sufcient size to absorb the losses if they come to pass.39 It is (or has been) therefore willing to accept them. 3.2 Introduction to mutual indemnity and hold harmless clauses in oil and gas contracts As has already been noted, the oil and gas industry is by no means alone in making use of indemnity clauses. However, the oil and gas industry utilises indemnity and hold harmless clauses in a more thoroughgoing way than most other industries. This 33 See the discussion at section 2.1, above. 34 See, e.g., LOGIC, General Conditions of Contract for Mobile Drilling Rigs (2002), available for download from www.logic-oil.com/content/standard-contracts-0 (accessed 15 December 2022) (hereinafter LOGIC, Mobile Drilling Rigs) at cl 18. 35 Ibid. at cl 19. 36 However, these indemnities may difer somewhat in their precise content. Note, for instance, that in LOGIC, Well Services, cl 19.9(a), the operator ofers a full indemnity in respect of loss of or damage to the hole, while in LOGIC, Mobile Drilling Rigs, cl 18.6(a), an indemnity is given subject to a (quite tightly confned) carve-out provision in respect of damage to hole caused by contractor’s negligence. 37 See, e.g., the discussion of the Orbit Valve case at section 6.1, below. 38 Perhaps the major exception to this is in the case of ofshore construction works, where the contractor will commonly be asked to provide a one-sided indemnity to the operator in respect of the recovery, removal or marking of any wreck or debris associated with the work under the contract. 39 As the UKCS matures and a more diverse set of companies become operators, there will be a greater need for operators to carry insurance against such risks.

26

risk allocation in oil and gas contracts in english law is borne out by, for example, the amount of time invested by the court in Caledonia North Sea Ltd v London Bridge Engineering Ltd (London Bridge)40 in examining the particular features of the oil industry which give rise to what is still viewed as an unusual and rather counter-intuitive practice.41 At bottom, the commercial purpose of an indemnity and hold harmless clause is quite simple. The parties are allocating (more properly, re-allocating42) between themselves the risk of loss associated with the occurrence of a particular event. As we have already seen, in the context of a simple indemnity and hold harmless clause, one party is agreeing that (within the parameters of this particular contract) it is better placed than the other to bear the risk of a particular type of loss. By contrast, in the case of a mutual indemnity and hold harmless clause, the parties are generally saying that neither of them should have sole responsibility for a particular species of risk – for instance, the risk that people engaged on the contract may be injured or killed – but that it is appropriate to divide between themselves the responsibility for that type of risk. The division does not follow faultbased principles, but instead follows what might call the identifcation principle, with the loss being allocated to the party most closely identifed with it. The clause will therefore commence with the party identifying the aspect(s) of a given type of loss for which it is willing to take responsibility, and those in respect of which it is not. Each party then agrees to indemnify and hold harmless the other in respect of the element of the potential loss that it has accepted, and in return receives the beneft of an indemnity and hold harmless provision relative to the aspect of the potential loss accepted by the other party. So if A and B are respectively an oil company and a contractor who have entered into a contract, in a typical mutual hold harmless indemnity provision pertaining to the risk that personal injury or death will befall one or another of the parties’ personnel while engaged on the contract,43 A will confrm that it accepts responsibility for any injuries or fatalities sufered by A’s own personnel, however caused, and that it will indemnify and hold harmless B in respect of that category of loss. B agrees the converse: that it will accept responsibility for any injuries or fatalities sufered by B’s own personnel, irrespective of how these were caused, and that it will indemnify and hold harmless A in respect of such loss.44 40 Rather confusingly, as a number of defenders settled the claims against them and dropped out of the case in the period between the Inner House appeal and the case’s hearing in the House of Lords, the House of Lords phase of the case is reported as Caledonia North Sea Ltd v British Telecommunications plc. In the interests of consistency the case will be referred to as “London Bridge” throughout. 41 In the Inner House of the Court of Session (London Bridge, 2000 SLT 1123) see, e.g., section 2.5 of Lord Rodger’s speech, from 1150, Lord Sutherland at 1174E–F and L, and Lord Gill at 1213F–H. In the House of Lords (London Bridge, orse British Telecommunications plc 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321), see Lord Bingham at paras 7–9 and Lord Hofmann at paras 81–82. 42 This exercise does not take place in a vacuum; the law has a pre-existing view on how, in the absence of agreement, such risks should be borne. 43 Such as may be found throughout the suite of LOGIC Standard Conditions: see, e.g., LOGIC, General Term and Conditions of Contract for Supply of Major Items of Plant and Equipment (3rd edn, 2015), available for download from www.logic-oil.com/content/standard-contracts-0 (accessed 15 December 2022) (hereinafter LOGIC, Supply of Major Items), cll 21.1(b) and 21.2(b). For a discussion on the potential impact, following Farstad v Enviroco, of the Unfair Contract Terms Act 1977 upon such clauses, see section. 4.1, below. 44 For a discussion of drafting issues relative to “personnel” and cognate phrases, see the discussion at section 6.5, below.

27

greg gordon Mutual indemnity and hold harmless provisions are not used only for cases of personal injury but will also commonly be agreed in relation to other categories of loss. An operator-to-contractor agreement will typically also contain such a clause in respect of loss of or damage to property, where A confrms that it accepts responsibility for loss of or damage to A’s property, however caused, and that it will indemnify and hold harmless B in respect of that category of loss; and B agrees to accept responsibility for loss of or damage to B’s property, howsoever caused, and indemnifes and holds harmless A relative to such losses.45 Pollution risk will also sometimes be divided up along similar lines, with the contractor accepting certain kinds of pollution (commonly, that emanating from its own equipment) and the operator accepting other kinds (typically, all other instances);46 however, it is important to note that indemnities for pollution risk are more likely than those already discussed to be cut into by a qualifcation. Consequential losses are sometimes dealt with as by way of exclusion or limitation of liability clauses, but can alternatively be the subject of indemnity and hold harmless provisions.47 3.3 The rationale for a mutual indemnity and hold harmless regime in the oil and gas context A number of factors have been advanced as the reason for the oil and gas industry’s use of mutual indemnity and hold harmless clauses. In the leading work on UK ofshore oil and gas insurance, the rationale for is presented thus: If an individual is injured he will expect to have a right to sue any party who may have been guilty of negligence leading to the circumstances which caused the injury. This party may be another contractor, the Principal or his employer, or any combination of all three. The issue can become complicated by reason of contributory negligence. Determining liability and awarding costs can be a lengthy process in these circumstances, and this can only add to the anguish of the injured party, or the dependents of the deceased who may have been the sole breadwinner. The employer therefore accepts a responsibility to provide for his employees and will generally give the party with whom he is contracting a full indemnity in respect of any suit or action brought against that other party.48

Sharp’s justifcation was accepted by several of the judges in London Bridge,49 the main piece of litigation to arise out of the Piper Alpha disaster.50 However, it is only 45 For an example of such sub-clause, see LOGIC, Supply of Major Items, cll 21.1(a) and 21.2(2). For a discussion of drafting issues concerning the defnition of “property”, see section 6.5, below. 46 See LOGIC, Mobile Drilling Rigs, cll 18.3 and 18.4. 47 See, for instance, LOGIC, General Conditions of Contract (including Guidance Notes) for Ofshore Decommissioning (2018) available online from www.logic-oil.com/content/standard-contracts0, Clause 25, p. 29. 48 D Sharp, Ofshore Oil and Gas Insurance (1994) at p. 108. London, Witherby. 49 In the House of Lords phase of the case, reported at 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321, see the dictum of Lord Bingham at para 7 and that of Lord Hofmann at para 82. In the earlier Inner House phase, reported at 2000 SLT 1123, see LP Rodger (who is rather more agnostic about Sharp’s justifcation than his colleagues) at 1150L–1151B, Lord Sutherland at 1174F, Lord Coulsfeld at 1202K and Lord Gill at 1213J–K. 50 The Piper Alpha installation was destroyed in a series of fres and explosions on 6 July 1988 with the loss of 167 lives. For a detailed account of the cause of the disaster, and its impact upon regulatory law in the United Kingdom, see J. Paterson, “Health and Safety Law Ofshore”, in G Gordon, J Paterson and

28

risk allocation in oil and gas contracts in english law partially convincing. By defnition, it can only serve to explain why the industry adopts such an approach in relation to personal injury; it cannot explain why the industry takes a virtually identical approach in relation to damage to property,51 or a broadly similar approach in relation to other matters such as pollution costs and consequential loss. It is certainly true that by and large the industry prefers swift and certain resolution to its disputes, and that it does not generally favour time-consuming and costly litigation.52 But in so far as Sharp suggests that the primary reason for the existence of the mutual indemnity regime is the industry’s desire to give an efectual remedy to, or diminish the anguish of, the injured party or his dependants, he would seem to overstate his position. If that truly were the industry’s intention, it has not implemented it very efectively. The mutual indemnity and hold harmless regime is not a system of strict liability so far as the injured worker is concerned; while it determines who will ultimately pick up the bill for a personal injury claim, for there to be a personal injury claim in the frst place, the injured worker must establish that someone is legally liable for the injury. Liability must be established before it is ported about to its fnal resting place. It is, moreover, unrealistic to suggest that such altruistic concerns lie at the very heart of the industry’s approach to risk allocation. Diminution of levels of anguish is more likely to be a fortunate side efect of the practice than its raison d’être. The principal reasons for the mutual indemnifcation regime are far more likely to be business ones. This was recognised by Lord President Rodger in the Court of Session phase of London Bridge when he noted that the practice of indemnifcation was “fundamental to the economics of the North Sea operation”.53 Insurance (and the broader but related concept of risk management) is the economic driver that makes this so.54 It may at frst sight be surprising that something which E Usenmez, UK Oil and Gas Law: Current Practice and Emerging Trends Volume I: Resource Management and Regulatory Law, 3rd Ed, Edinburgh University Press, 2018, pp. 187–230. 51 While, tragically, it makes perfect sense to talk of the anguish of the family in the context of a fatal injury, it makes no sense at all to describe property losses in these terms. When an oil tool is lost over the side of a vessel, it does not leave a grieving spouse and family behind. 52 See, e.g., D Peng, “Mutual Indemnities in North Sea Contracts – Liability and Insurance Clauses” in D Peng, Insurance and Legal Issues in the Oil Industry, 1993, p. 157, University of Dundee, Centre for Petroleum and Mineral Law and Policy. Among the main reasons given for the practice are “that it permits the parties to assess and accept the risks more easily” and that “it avoids delays in claim settlement and it reduces the fghting of lawsuits”. However, it is not immediately apparent to the present author that indemnifcation reduces disputes. Clauses are scrutinised carefully before claims are accepted and, if there is disagreement between the parties (or more particularly, between the parties’ insurers) about the proper construction of the clause, litigation will follow which may prove to be time-consuming and costly: see, e.g., London Bridge. The Piper Alpha disaster occurred on 6 July 1988. The proof began on 3 March 1993; in all, 391 days of evidence were heard. The case was not fnally concluded until judgment was handed down in the House of Lords on 7 February 2002. See also the comments of Circuit Judge Brown in Fontenot v Mesa Petroleum Co, quoted by Lord President Rodger in London Bridge, 2000 SLT 1123 at 1151C–F. It may be that in some cases the fact of indemnifcation brings a quicker resolution to the claim of the injured party: this seems to have occurred in both London Bridge and Campbell v Conoco (UK) Ltd [2003] 1 All ER (Comm) 35 at para 6. Even this, however, does not seem to be a universal truth: see the experience of the pursuer in Slessor v Vetco Gray. The pursuer sufered severe injuries in an accident in May 2003. Liability was in principle established on 23 March 2007 (see 2007 SLT 400) but even then a number of issues remained outstanding, among them the construction of the contractual indemnity clause, discussed further at section 6.3, below. 53 London Bridge 2000 SLT 1123 at 1150I. 54 See T Daintith, G Willoughby and A Hill, United Kingdom Oil and Gas Law (3rd edn, looseleaf, 2000– date), para 1–845: “The client will, any event, normally carry insurance cover for his own employees and his own property and the cost of this insurance would not be reduced if the particular contractor was also required to be insured against the same risks. It is thus normal for the client and the contractor to assume

29

greg gordon seems to be an ancillary matter should be so fundamental. However, in a high-risk endeavour such as the ofshore oil and gas industry, insurance premia are not marginal costs but major expenses, made all the more so by the involvement of many contractors and subcontractors.55 Oil platforms are not stafed wholly, or even mainly, by the operator’s own personnel. At any given moment in time, one can reasonably expect there to be representatives from upwards of 20 other companies on board.56 If a large proportion of these companies were required to carry insurance against the fairly remote, but potentially catastrophic, risk that they might cause or contribute towards the destruction of the platform and/or widespread injury or loss of life among those on board,57 then, always assuming that such insurance cover could be obtained, the cumulative cost of doing so would be very considerable. In addition, parties’ separate policies would simply run in parallel, and in the event of a catastrophic event (assuming that the cause of the calamity could be identifed, and was attributable to one contractor) only one policy would be claimed upon, and the remaining contractors’ policies, and that of the operator, would prove to have been surplus to requirements. This would add considerably to the cost of operations without adding any value to them. Counter-intuitive as it may at frst appear, on analysis it can be seen that there are sound economic reasons for the practice of mutual indemnifcation. 3.4 Back-to-back indemnity and hold harmless provisions in oil and gas contracts Although at any given time there may be upwards of 20 to 50 contractors on a producing platform, only a handful of these parties will be in a direct contractual relationship with the operator. So for instance, in the production phase of a platform’s life, the operator will ordinarily enter into a handful of lead contracts through which it will entrust important parts of the platform’s functions to three or four contractors – typically a rig services manager, a well services supervisor, and a drilling company. Most of the other “contractors” represented on the platform will be in a direct contractual relationship not with the operator, but with either the layer of contractors just described, or with their subcontractors. Viewed from the operator’s perspective, these parties will be subcontractors, sub-subcontractors, and so on. Thus, at any given time,58 there will be a number of chains of contractual relationships in place.

full liability, and give each other mutual indemnities, for claims arising out of death of or injury to their own employees and for loss or damage to their own property . . ., regardless of any negligence or default on the part of the other party or its employees, agents or sub-contractors.” 55 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321 per Lord Bingham of Cornhill at para 2. 56 For instance, of the 165 people on the platform who lost their lives in the Piper Alpha disaster (167 people died in all; 2 were crew of the fast rescue ship Westhaven), 31 were employed by the operator. The remaining 134 were employed by 24 diferent contractors: see London Bridge 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321 per Lord Bingham of Cornhill at para 2. 57 Not all contractors would necessarily be in a position to cause catastrophic loss. It is hard to imagine, for example, that the catering contractor could cause the total loss of a platform. 58 Although the example focuses on the case of a producing platform, the position is similar in other phases of the platform’s life. When it is being constructed, overhauled or decommissioned, the usual position

30

risk allocation in oil and gas contracts in english law The signifcance of the above is that every link in the chain is a contract in which the parties have to agree how to allocate risk as between themselves. The full economic benefts of instituting an indemnity and hold harmless regime do not accrue if only some of the parties are included in it; moreover, if some parties are part of the regime and others are not, there is a serious risk of misunderstandings as to who bears which risk, and of accompanying litigation, gaps in insurance cover and uninsured losses.59 The general practice is therefore for each of the contracts within the chain to contain so-called back-to-back provisions. The overall intent of such a set of provisions is generally that, when all the clauses are read together across the set of contracts, they should have the efect that, in respect of the risk element(s) with which the indemnity clauses deal, each party in the chain bears the loss or damage directly identifed with it, and such loss only. However, achieving this result requires something of a leap of faith. To make the losses migrate to the appropriate point in the chain, a contracting party has to, in the anterior contract, assume responsibility not just for the losses identifed with itself, but also for the losses of the parties below it on the chain. However, in the posterior contract, it will require to be indemnifed and held harmless by its subcontractor in respect of all losses identifed with the subcontractor, and any subcontractors lying down the chain of whatever level. Equally, in the posterior contract, the subcontractor will demand an indemnity and hold harmless clause from the contractor in respect not just of the contractor’s losses, but also those of the parties lying above him in the chain. The contractor will give that in the knowledge that in the anterior contract, he should already have obtained an indemnity in respect of the losses identifed with the parties above him in the chain. Thus if a set of back-to-back mutual hold harmless indemnities pertaining to personnel and property operate as the parties intended, the operator will ultimately carry the risk of injury to or death of its own personnel and loss of or damage to its own property, but not any like losses sufered by the lead contractor or its subcontractor; the lead contractor will bear the risk of injury or death in respect of its own personnel and loss of or damage to its own property, but not any like losses sufered by the operator or the subcontractor; and likewise the subcontractor will accept risk in relation to its own property and personnel only. This is shown diagrammatically in Figure 2.2. 3.5 Qualifed indemnity and hold harmless provisions As has already been noted, when parties agree to a comprehensive mutual indemnity and hold harmless regime they agree to bear the risk of loss not on the basis of who was at fault, but on the basis of who is best placed to insure against the loss or otherwise absorb it. Sometimes, however, the parties will deviate from this paradigm and one or more of their provisions will be made subject to qualifcations or so-called

is for the operator to contract with a limited number of lead contractors and for them to let out parcels of work to appropriate subcontractors. 59 Farstad v Enviroco demonstrates the dangers which can be posed by the interaction between the contractual risk allocation provisions and the statutory law of contribution: see G Gordon, “Indemnifcation, Exclusion and Contribution: Farstad in the Supreme Court” (2011) 15 EdinLR, pp. 259–265.

31

greg gordon Contractual provision

Intended Effect Operator O On the facts given (see Note, below) O carries risk in relation only to its own personnel and is at the head of the contractual chain.

O offers C a hold harmless indemnity in respect of personal injury or death of O’s personnel C offers O a hold harmless indemnity in respect of personal injury or death of C’s personnel and that of C’s subcontractors of whichever level

Contractor

C offers SC a hold harmless indemnity in respect of personal injury or death of C’s personnel and that of O SC offers C a hold harmless indemnity in respect of personal injury or death of SC’s personnel and that of SC’s subcontractors of whichever level

SC offers SSC a hold harmless indemnity in respect of personal injury or death of SC’s personnel and that of O and C

Subcontractor

Sub-subcontractor

SSC offers SC a hold harmless indemnity in respect of personal injury or death of SSC’s personnel and that of SSC’s subcontractors (of which, on the facts given, there are none).

C Under this contract, C carries risk in relation to its own personnel and also that of all subcontractors. Thus in any dispute between O and C, O will expect C to bear any losses pertaining to those personnel. C, however, has the opportunity, in its contract with SC, to take an indemnity back in respect of the sub-contractors, thus attempting to move ultimate liability down the contractual chain.

C: Under this contract, C takes the hold harmless indemnity from SC referred to above; thus when the two contracts are read together he should be carrying liability only in respect his own personnel, not the subcontractors’. However, as regards any dispute between himself and SC, he has had to assume risk in relation to his own personnel and those of O. However the risk to him of so doing is reduced by the terms of the hold harmless indemnity he has entered into with O in the contract above. SC: The fact that SC obtains an indemnity from C in relation to injury to both C and O’s personnel means he is able to accept these losses in the contract below as, if SSC claims, he may pass these losses up the chain to C. However, under this contract, in any matter between SC and C, C is made to carry risk in relation to his own personnel and those of all SSCs. He therefore needs to ensure that his agreement with SSC contains a valid hold harmless indemnity from SSC in relation to this risk.

SC: In any matter between SC and SSC, SC bears the risk in relation to his own personnel; also that of O and C. To avoid ultimate responsibility for these losses SC must ensure that it has the benefit of an effectual hold harmless indemnity from C in relation to C and O’s potential losses in its contract with C, above. SSC carries risk in relation to his own personnel only as, on the facts, there are no sub-contractors beneath him in the chain. If there were, SSC would need to enter into a further back to back hold harmless indemnity with them.

Figure 2.2 Simplifed example of a set of back-to-back mutual hold harmless indemnity provisions in respect of personal injury. Note: This is a simplifed representation of one quite short contractual chain. With regards to any given producing platform, there will be more than one such chain: the operator is likely to have entered into direct contracts with at least a handful of parties. The relationship between the parties within this chain and those in other chains gives rise to further complications discussed at section 7, below.

32

risk allocation in oil and gas contracts in english law carve-outs. Qualifcations are commonly encountered in the provisions pertaining to responsibility for injury to or death of third-party personnel, or damage to thirdparty property. Here, the parties often state that the indemnity and hold harmless provision will be ofered only to the extent that the injury, death or damage was caused by the negligence or breach of duty of the indemnifying party.60 At frst sight this is a major deviation from the standard indemnity and hold harmless regime. It is, however, justifable in the case of third-party liability, as unlike the situation where one takes responsibility for one’s own people and property come what may, neither party has a close association with a true third party such as would justify a deviation from the law’s default position on how risk should be allocated. At the level of contracts between oil companies, it is common for such mutual indemnity and hold harmless provisions as are granted to be qualifed by the statement that they will not apply in the case of wilful misconduct or gross negligence. The objective of such a clause is to protect the company from acts of deliberate sabotage or conduct which falls well below the standard of care which would ordinarily be expected in such operations. Neither “wilful misconduct” nor “gross negligence” has a wholly settled meaning in English or Scots law; as a result it is prudent, when the terms are used, to defne them in the agreement.61 Moreover, some companies (particularly some in the United States) have a corporate policy of not accepting indemnity and hold harmless provisions which operate in such a way as to permit a contractor to escape from the consequences of its gross negligence or wilful misconduct. Historically, some companies sought to exclude from the ambit of the indemnity and hold harmless provisions losses attributable to the “sole negligence”62 of the other party or parties, but this particular carve-out appears to be less common in current practice. Given the complexity of oil and gas operations and the interdependent way in which the various parties work, it is, in practice, quite rare for one party’s actions to be the sole cause of an accident; and even when it is, the time, efort and money which may have to be expended in order to establish that fact may be very considerable. Moreover, prior to the occurrence of the incident, at the point when parties are mapping their potential liabilities and purchasing the necessary insurance cover, it is impossible to know if it will be caused by the sole negligence of one or another of the parties. Thus sole negligence clauses would seem to increase the prospect of uninsured losses. Qualifying indemnity provisions is not without its benefts. One can readily understand why an operator would wish an obligation to re-drill to be “carved out” of the general indemnity and hold harmless provision which a drilling contractor will usually enjoy relative to loss of hole. However, the practice also has drawbacks. It adds considerably to the complexity of what are already rather awkward clauses. It also undercuts the economic benefts provided by the clause: the greater the number of exceptions carved out of the indemnity and hold harmless regime, the greater the

60 See, e.g., LOGIC, Supply of Major Items, cll 21.1(c) and 21.2(c). 61 Among the matters to be dealt with in the defnition is, e.g., the issue of whose gross negligence or wilful misconduct is relevant to the clause: for instance, all personnel or senior management only? 62 The indemnity clauses litigated in the London Bridge case were in such terms: see the extracts from the relevant contracts reproduced at London Bridge, 2000 SLT 1123 at 1126–1129.

33

greg gordon risk the contractor is exposed to, and the more insurance cover it must purchase. Deviating from standard practice can also lead parties into error concerning the precise forms of insurance cover which they require on this particular project. But ultimately, the purpose of the risk allocation regime in any contract is to express the parties’ intent, and the relative importance of these various factors, are matters for the parties to themselves. 4 Selected further issues in indemnifcation law and practice in the UKCS 4.1 Statutory control of indemnity and hold harmless clauses Unlike some other jurisdictions – for instance, several of the petroleum-producing states of the United States63 – the United Kingdom has not imposed any specifc statutory controls on the use of indemnity and hold harmless clauses in the oil and gas industry. So far as the general body of commercial law statutes is concerned, the Unfair Contract Terms Act 1977 as amended (hereinafter UCTA) imposes some relevant restrictions.64 The restrictions imposed by UCTA on the use of indemnity clauses apply only when the indemnifying party deals as a consumer.65 This, together with the belief that indemnity and hold harmless clauses were mere indemnities, led the present author to state in the frst edition of this work that “there are therefore no statutory controls in force in the United Kingdom which impact upon the parties’ ability to make use of indemnity clauses in commercial oil and gas contracts”. Following Farstad v Enviroco, however, this claim can no longer be made. As indemnity and hold harmless clauses have now been held to operate as exclusions when they operate in the context of “direct exposure to the other contracting party”66 (as opposed to third-party losses), then the restrictions imposed by UCTA relative to exclusion clauses also need to be considered. Of greatest concern to the oil and gas industry would be the rule contained in ss 2(1) and 16(1)(a) UCTA,67 that any attempt by a party to restrict its liability for death or personal injury resulting from negligence will be inefectual. At frst sight, this provision would seem to be triggered by an indemnity and hold harmless clause which pertains to losses associated with personal injury or death and which applies irrespective of negligence; and as Hewitt notes, “[t]he application of the section appears on its face to be strict and it does not appear

63 So-called anti-indemnity statutes have been enacted in Texas, Louisiana, New Mexico, Wyoming and Oregon: see P Gerald and H Williams, “Injuries to Third Parties Arising From Oil and Gas Operations: An Analytical Framework for Examining Indemnity and Additional Insured Issues”, 15 J Nat Resources & Envtl L 21 at 28–31. For a detailed discussion of the position in Texas, see T Fox, “Return to Certainty in Risk Assessment, Management and Transfer: The Journey of the Texas Oilfeld Anti-Indemnity Act”, 2001 IELTR 18. See also Chapter 12 of this book. 64 The more extensive provisions of the Consumer Rights Act 2015 can be discounted as they apply only to consumer contracts. UCTA continues to govern exemption clauses in business-to-business contracts: see, e.g., UCTA s.2(4). 65 1977 Act, ss 4 (for English law and that of Wales and Northern Ireland) and 18 (Scots law). A party deals as a consumer when he does not contract in the course of a business but the other party does: see s 12. 66 Farstad v Enviroco per Lord Mance at para 59. 67 Section 2(1) applies in English, Welsh and Northern Irish law and s 16(1)(a) in Scots law.

34

risk allocation in oil and gas contracts in english law to be possible for the parties to contract out of the Act”.68 However, Hewitt goes on to note: indemnities concerning death and personal injury in the context of the oil and gas industry (even where they are caused by negligence) have been upheld, notably in London Bridge. This may be because the clauses concerned in London Bridge were not construed as exclusions of liability for death or injury but rather exclusions of liability for claims by third parties in respect of the same and were therefore not covered by the strict prohibition in section 2(1).

Hewitt is, it is submitted, wrong to characterise the mutual hold harmless and indemnity clauses in London Bridge as clauses which involved an exclusion of liability. It is submitted that the clauses were, on their true construction, concerned not with extinguishing liability, but with its re-allocation.69 He is, however, surely correct in conceptualising the London Bridge clauses as being essentially concerned with third-party claims. This would seem to mean that they do not seem to operate so as to, using Lord Mance’s Farstad v Enviroco formulation, “direct exposure to the other contracting party”. Indeed, it would seem to be possible to go further. It would not seem that any indemnity and hold harmless clause appertaining to liability for personal injury or death could ever be classifed as “direct exposure”. A company, being an incorporeal corporation, can never itself sufer personal injury or death. Thus it would appear that, by defnition, all deaths and personal injuries sufered as a result of the negligence of a company must be third-party losses for the purposes of Lord Mance’s Farstad v Enviroco formulation. Thus, in this context of personal injury, indemnity and hold harmless clauses would seem to inevitably operate as indemnities, not exclusions. Where, however, the case where the clause does seek to regulate direct exposure to the other contracting party – as in Farstad itself, where the claim pertained to property belonging to one of the parties – the clause would seem to have exclusionary efect, and UCTA (although not argued in Farstad v Enviroco) would appear to be relevant. However, the provisions engaged would not be the bright-line prohibition upon exclusion contained in s 2(1), but the (less strict) reasonableness requirements contained elsewhere in the Act.70 The party having the beneft of the indemnity and hold harmless clause has less to fear from these provisions than from ss 2(1) and 16(1)(a). Given the widespread use of indemnity and hold harmless clauses within the industry and the economic benefts of the practice, one would not expect such a clause to be struck down by the court other than in very unusual circumstances. 4.2 The law’s default settings: the normal presumptions about the distribution of risk The risk distribution exercise that the parties carry out by entering into an indemnity and hold harmless regime does not take place in isolation. It occurs against the background of the assumptions which, in the absence of specifc agreement, the law of 68 T Hewitt, (n. 19, above), at p. 205. 69 Thompson v T Lohan (Plant Hire) Ltd [1987] 1 WLR 649. See also Lord Mance’s “direct exposure” formulation in Farstad v Enviroco, discussed later in this section. 70 I.e. ss 2(2) and 3 (in English, Welsh and Northern Irish law) and ss 16(1)(b) and 17 (in Scots law).

35

greg gordon contract and tort/delict71 makes about how certain risks are to be allocated. In contract law, the broad expectation, under both Scots law and the law of England and Wales, is that, subject to considerations such as remoteness of damage and the need for the nonbreaching party to take reasonable steps to mitigate its loss, the party in breach is obliged to make good the losses sufered by the non-breaching party.72 This may have the efect of liability greatly exceeding the commercial value of the contract. If a loss occurs without either party breaching the contract, then, viewed from the perspective of contract law, the loss will lie where it falls. In tort/delict, the breach of a statutory duty that causes a party loss may in some circumstances found an action in reparation. And the founding principle of the tort/delict of negligence is that where a person owes another a duty of care and breaches that duty, causing a loss, he is under an obligation to make a payment of compensatory damages to make good that loss.73 But if there is no breach of statutory or common law duty, neither party can sue and the loss lies where it falls. It can therefore be seen that the general law’s default position is that liability follows breach of contract or breach of duty. Under this model, liability is wedded to fault. By contrast, as we have seen, carve-outs apart, the indemnity and hold harmless regime is predicated on the basis that liability should not fow along these lines, but that it should be accepted by the party best placed to insure against or otherwise absorb that particular type of loss. Thus the common law and the contractual regime that the industry creates to govern such matters are not closely aligned. The signifcance of this is that when the contractual regime fails for some reason – typically, because a clause is not sufciently clearly drafted, but perhaps, post-Farstad, because it is deemed to be an exclusion clause and fails one of the tests set out in UCTA – and risk fails to be allocated under the general law, it will likely be distributed in a way which is very diferent from that which the parties intended. This may very well mean that one or both of the parties will fnd themselves facing losses against which they are not insured. The stakes are therefore high when one is drafting an indemnity and hold harmless clause. 4.3 The position of third parties While the law of both England and Scotland now permits the beneft of a contract to be extended to a third party,74 neither jurisdiction permits a contract to impose obligations upon someone who is not party to it. It is therefore not strictly correct to say that a welldrafted indemnity and hold harmless clause modifes or overrides the law’s conventional approach to risk allocation. It does so only so far as the parties to the contract are concerned. A stranger to the contract – usually described as a third party – who is injured or otherwise 71 The Anglo–American expression “tort” is not used in Scots law, which prefers “delict”. However, at least in the feld of the tort/delict of negligence, there is little to distinguish the laws of England and Scotland. 72 For the position in England and Wales, see, e.g., G Treitel, The Law of Contract (14th edn, 2015) (hereinafter Treitel, Contract) at 926 f. For the Scots position, see W McBryde, The Law of Contract in Scotland (3rd edn, 2007), Scottish Universities Law Institute / W. Green and Son, chapter 22. 73 See Donoghue v Stevenson [1932] AC 562. 74 In England, where the doctrine of privity of contract was traditionally very strong, this is as a result of the Contracts (Rights of Third Parties) Act 1999. In Scotland the long-standing common-law doctrine of jus quaesitum tertio permitted third-party rights, albeit subject to some rather onerous qualifying conditions. Following a review by the Scottish Law Commission, the Scots law of third-party contractual rights was placed on a statutory footing by the Contract (Third Party Rights) (Scotland) Act 2017.

36

risk allocation in oil and gas contracts in english law sufers a loss as a result in the course of the execution of the contract works will seek compensation through the time-honoured route of suing the person or persons whose negligence and/or breach of statutory duty caused the loss. Sometimes the person against whom the claim is directed will, by pure coincidence, happen to be the indemnifying party. Where, however, the third party sues not the indemnifer, but the indemnifed party, a claim will be made by the indemnifed party under the indemnity and hold harmless clause. So, if the standard mutual indemnity and hold harmless provision is in place between companies A and B and C, one of company A’s employees is injured by the negligence of one of contractor B’s employees, he can be expected to sue B for negligence and/or breach of statutory duty.75 B cannot defend C’s claim on the basis that it has an indemnity, as that is irrelevant so far as C, a third party, is concerned. However, B can make a claim against A as a result of their contractual arrangements. The fact that the third party is disinterested in the indemnity and hold harmless clause has the potential to have serious implications for the parties, particularly in the event of insolvency. Unless it is fortifed either by meaningful guarantees or by adequate levels of insurance, a right to be indemnifed is only as strong as the fnancial covenant of the company providing it.76 5 Interpreting indemnity and hold harmless clauses 5.1 The traditional approach to interpretation The general rules Although the precise formulation of the tests used in the jurisdictions traditionally difered, it is possible to say that both English and Scots law have traditionally taken an objective approach to the interpretation of contracts.77 Although objective, neither jurisdiction has been wholly literal in approach:78 both jurisdictions require the clause or phrase under discussion to be interpreted not in isolation, but in the light of the written document as a whole.79 Both have permitted evidence as to the objective factual background to be adduced with a view to allowing the court to determine the aim or thrust of the agreement,80 albeit there are diferences between the two systems surrounding precisely when this is to be permitted.81 75 Depending on the facts, the injured employee may in addition be able to raise a case directly against A in respect of A’s breach of its non-delegable “duty to take reasonable care to see that its employees are safe”. See J Steele, “Employers Liability” in M Jones et al, Clerk and Lindsell on Torts, Sweet and Maxwell, 23rd Ed, 2020, para 12.14. 76 See, e.g., T Taylor, Knock for Knock Revisited, Blog-Post, Clyde and Co website, 20 February 2013, available at www.clydeco.com/blog/energy/article/knock-for-knock-revisited; P. Roberts, Petroleum Contracts: English Law and Practice (2nd Ed) Oxford, 2016 para 13.11. 77 For the traditional position in England, see K Lewison, The Interpretation of Contracts (2nd edn, 1997) (hereinafter Lewison, Interpretation) at para 1.05. For the position in Scotland, see Scottish Law Commission, Scot Law Com No 160, Report on Interpretation in Private Law, available for download from: www.scotlawcom.gov.uk/ fles/1512/7989/6878/rep160.pdf (accessed 27 April 2017) (hereinafter SLC, Report on Interpretation) at para 2.3. 78 SLC, Report on Interpretation, para 2.1. 79 See, e.g. (in the law of England and Wales), Re Jodrell (1890) 44 Ch D 590; for a Scots example, see Glen’s Trs v Lancashire and Yorkshire Accident Insurance Co Ltd (1906) 8 F 915. 80 See Lewison, Interpretation at para 2.10. 81 In England, even before Lord Hofmann’s restatement in Investors Compensation Scheme Ltd. v West Bromwich Building Society [1997] UKHL 28 (discussed at p. 39 below), and particularly thereafter, a

37

greg gordon Interpretation contra proferentem In addition to the generalities expressed above, both English and Scots law have traditionally deployed the contra proferentem rule to interpret exemption, limitation and indemnity clauses.82 An ongoing problem in the law of contractual interpretation is the extent to which individual canons of construction, such as contra proferentem, have continued to survive the upheavals in the general law of contractual interpretation. For present purposes, a precautionary approach will be adopted which assumes that such rules continue to be in efect and that contractual draftsmen should continue to draft as if such rules remain in efect. However, this is intended as an indication of prudent drafting practice, rather than an indication that it is the optimal way of interpreting indemnity and hold harmless clauses. The current author would argue for a more naturalistic approach which recognises the normalcy of indemnity and hold harmless clauses in the oil and gas context and does not call for artifcial or strained interpretations of such clauses, at least in that particular context. Contra proferentem means that “[a]mbiguous words in exemption clauses [will be] construed in the way least favourable to the party relying on them”.83 In the context of a simple indemnity, the party relying on the clause will be the one who stands to receive the beneft of it – i.e., the indemnifed party. But who is the proferens when the clause is mutual? The prevailing view appears to be that whichever party has the misfortune to have a claim directed towards it (and therefore requests indemnifcation under the contractual risk allocation provisions) is to be treated as the proferens.84 This, however, reduces the question to a matter of happenstance. The arrangement is mutual – indemnity and hold harmless provisions travel in either direction – and had the facts been diferent the indemnity might have run in the opposite direction. It seems to be artifcial, in such a situation, to contend that there is a proferens at all. This fact was recognised in the Scottish Outer House case of Slessor v Vetco Gray, where Lord Glennie stated: I accept Mr Armstrong’s submission that the contra proferentem approach, which in any event only applies in a case of ambiguity, has much less impact where the exemptions and indemnities are mutual or reciprocal. Both parties are, in a sense, the proferens; and it makes little sense to construe the clause against each one of them leaving the possibility of a hole in the middle.85

willingness to admit “matrix of fact” evidence can be seen: see Lewison, Interpretation (n. 77, above), para 2.10. In Scotland, evidence as to surrounding circumstances is admissible only in the case of ambiguity, or if the contract was unintelligible without it: SLC, Report on Interpretation, para 2.3; see in particular the materials cited in n 7 therein. In Arnold v Britton [2015] UKSC 36, which has supplanted ICS as the leading English case in this area, the Supreme Court clearly wishes the courts to focus primarily upon the natural meaning of the words used by the parties and to have recourse to external factors such as commercial common sense only in “unusual” cases where the wording is not otherwise intelligible: Arnold v Britton, per Lord Neuberger at para 17. Thus it may be that the efect of Arnold v Britton will be to close the gap between the Scots and English approach. 82 For an excellent account of this area, see E Peel, “Whither Contra Proferentem?”, in A Burrows and E Peel, Contract Terms (2007), pp. 61–75, Oxford University Press. 83 Treitel, Contract (n. 72, above), p. 221. 84 Orbit Valve per Steyn LJ at 182g–h; London Bridge, (Inner House) 2000 SLT 1123 per LP Rodger at 1148C–L and per Lord Sutherland at 1174L. The House of Lords passed no concluded view upon the matter as their Lordships took the view that the clauses were clear and unambiguous: London Bridge (House of Lords) 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321) per Lord Mackay of Clashfern at para 43. 85 Slessor v Vetco Gray at para 12.

38

risk allocation in oil and gas contracts in english law It is respectfully suggested that this approach is logical and takes cognisance of the realities of the parties’ contractual arrangements. It therefore has much to commend it. However, on the balance of current authority, it cannot be said to be the established view. Towards contextualism (and back again) The traditional approach referred to above requires now, at least in England, and possibly also in Scotland,86 to be considered in the light of subsequent developments. In Investors Compensation Scheme Ltd v West Bromwich Building Society,87 Lord Hofmann ofered a dictum which has come to be known as Lord Hofmann’s restatement of the English law of contractual interpretation. A comprehensive examination of the restatement is outside the scope of this chapter. However, putting the matter shortly, Lord Hofmann suggested that two cases decided by the House of Lords in the 1970s88 had efected a quiet revolution: a shift away from formalism towards contextual interpretation. He considered the efect of these cases had been to discard “[a]lmost all of the old intellectual baggage of ‘legal’ interpretation”.89 In particular, Lord Hofmann confrmed that contractual interpretation was still in a sense an objective exercise, in that it continued to be a matter of ascertaining what a reasonable and objective bystander would make of the parties’ communings. However, he added that such a bystander should be presumed to be aware of at least most of90 the background knowledge and circumstances (otherwise the “matrix of fact”) which comprised the setting in which those communings took place. Moreover, as “the meaning of the document is what the parties using those words against the relevant background would reasonably have been understood to mean”, the bystander’s knowledge of the background circumstances might from time to time prompt him to conclude that the parties must have used the wrong words. Such a conclusion would be reached rarely, as the courts would not readily accept that the parties had made linguistic mistakes; however, when an appropriate case did arise, the reasonable bystander should not attribute to the parties an intention which they plainly could not have had.91 This test, as expressed in ICS, involved a greater emphasis upon matrix of fact evidence than had previously been seen. In ICS itself, the approach led the court to conclude that something had gone wrong with the drafting of the clause in question, as a result of which the court was willing to re-order the wording of the clause in order to give it greater commercial sense. Other striking examples of what might be described as remedial

86 The extent to which the restatement has been received into Scots law is presently unclear: see D Cabrelli, Commercial Agreements in Scotland: Law and Practice (2006) at paras 2.03–2.24 and McBryde, Contract (n. 72, above) (3rd edn), paras 8.25–8.27. 87 Investors Compensation Scheme Ltd v West Bromwich Building Society (hereinafter ICS) [1998] 1 WLR 896. 88 I.e., Prenn v Simmonds [1971] 1 WLR 1381 and Reardon Smith Line Ltd v Hansen-Tangen [1976] 1 WLR 989. 89 At frst sight, this might appear to be a reference to formal legal rules such as contra proferentem: see E McKendrick, Contract Law: Cases, Text and Materials (7th edn, 2016) at 381; Peel, “Whither Contra Proferentem?”, p. 61, (n. 82, above). However, as McKendrick notes, appearances can be deceptive: see further the discussion of HIH Casualty & General Insurance at section 6.1, below. 90 Previous negotiations and declarations of the parties’ subjective intent continue to be excluded from consideration. 91 Investors Compensation Scheme [1998] 1 WLR 896 per Lord Hofmann at 912–913.

39

greg gordon re-drafting were seen in The Starsin92 and in Chartbrook Ltd v Persimmon Homes Ltd.93 In the now-leading case of Arnold v Britton,94 the Supreme Court took the opportunity to restate the restatement. While Arnold v Britton does not render matrix of fact evidence irrelevant in all circumstances, the court strongly emphasised the centrality to the interpretative exercise of the words used by the parties themselves, and relegated the role of external factors such as commercial context to supplementary criteria to which regard could only be had in “very unusual” circumstances.95 In spite of Lord Hofmann’s apparent belief that his dictum was not a radical departure, but merely a convenient synthesis of what the law already said, the restatement proved to be controversial, and there can be no doubt that Arnold v Britton represents a conscious reining-in of the contextual approach. It should be noted, however, that what Arnold v Britton promotes instead of contextualism is literalism – the fnding of meaning primarily from the words used by the parties themselves – rather than canonical interpretation. Thus, while following Arnold v Britton, a court may be slower than previously to have regard to factors such as commercial common sense when interpreting an indemnity clause, Arnold v Britton provides no obvious support for the practice of adopting strained or artifcial constructions of clauses in order to comply with the old canons of construction. Yet neither are these old rules expressly swept away. In these circumstances, a penumbra of doubt remains, and the only prudent approach is for contractual draftsmen charged with the task of drafting indemnity and hold harmless clauses to attempt to cross the highest barrier to implementation that they face, and to proceed on the basis that indemnity and hold harmless clauses will continue to be construed contra proferentem. Drafting in this manner should minimise the prospect of a dispute arising as to the true meaning of the clause. However, in the event that an indemnity clause is challenged on the ground that it does not satisfy one of the tests established by the traditional canons of interpretation, very careful consideration should be given to advancing the argument that all that is needed for the clause to operate is for this to be the ordinary consequence of the wording used by the parties.96 In addition, if the contract in question does not accurately express the parties’ intentions, advice should be taken on the possibility of rectifying the deed.97 6 Some known problems of drafting and interpretation 6.1 Contra proferentem and the problem of negligence and breach of statutory duty As has already been noted, when the parties to an oil and gas contract opt to include an indemnity and hold harmless regime in their contract, they are choosing to

92 [2003] UKHL 12; [2004] 1 AC 715. 93 [2009] UKHL 38; [2009] 1 AC 1101. 94 [2015] UKSC 36. 95 Per Lord Neuberger at para 17. 96 The recent evolution of the rule of contra proferentem is further considered immediately below. 97 For a general discussion on rectifcation of deeds, see (from the standpoint of English law) Lord Mackay of Clashfern (General Editor), Halsbury’s Laws of England (4th edn, 2003 Reissue) (hereinafter Halsbury), vol 32, paras 52–69. For the position under Scots law, see J Murray and S Wolfe, “Judicial and Other Remedies”, The Laws of Scotland: Stair Memorial Encyclopaedia, vol 13 (1992), paras 69–70.

40

risk allocation in oil and gas contracts in english law superimpose their own views on risk allocation upon the one provided by the common law. As we have seen, the common law generally expects liability to follow breach of contract or breach of duty. The courts have therefore considered it to be “a fundamental consideration in the construction of contracts of this kind that it is inherently improbable that one party to a contract should intend to absolve the other from the consequences of his own negligence”,98 and, have considered it to be, if anything, even less likely that one party would go so far as to accept liability for another party’s negligence.99 While the courts stop short of saying that the parties cannot use the terms of their contracts to re-allocate risk in the way in which they see ft, they have traditionally invoked the contra proferentem rule to say that the parties must use clear language when they do.100 Canada Steamship Lines Ltd v The King101 is authority for the proposition that efect will be given to a term which expressly states that the indemnifed party is to be relieved from the consequences of its own negligence. This could be described as the primary route to enforceability under Canada Steamship – express provision.102 However, the same case also provides that, if the clause does not expressly indemnify in respect of negligence or some synonym for it, it will operate to relieve the indemnifed party from the consequences of its own negligence only if (1) the words used in the clause are “wide enough in their ordinary meaning to cover negligence” on the part of the indemnifed party, and (2) no alternative ground of liability which is neither too remote or too fanciful to be in the parties’ contemplation might exist. A party able to satisfy both elements of this test could be described as having identifed a secondary route to enforceability under Canada Steamship. The question of whether or not an indemnity provision will protect the indemnifed against the consequences of its own negligence is of considerable practical importance. Unless there have been clear instructions that the indemnifed party’s own negligence is to be excluded from the ambit of the indemnity and hold harmless clause, a provision which is inefectual against the indemnifed party’s own negligence is unlikely to be what a party to an oil and gas industry contract wanted. Worse, it is unlikely to match with the risks that the party has insured against. The Orbit Valve case demonstrates the risks involved, while the Nelson case illustrates an alternative approach to this problem. In Orbit Valve, a case determined under English law, the employers of a man killed in the Piper Alpha disaster were sued by the operator, who sought reimbursement of the compensation which the operator had paid to settle a claim by the deceased’s family. The operator, who accepted that the negligent actions of one of its employees had at least contributed towards the disaster, claimed that it was entitled to recover under the contractual

98 Gillespie Bros & Co Ltd v Roy Bowles Transport Ltd [1973] QB 400 per Buckley LJ at 419. 99 Smith v UMB Chrysler (Scotland) Ltd 1978 SC (HL) 1 per Viscount Dilhorne at 7. 100 See, e.g., London Bridge (Inner House) 2000 SLT 1123 per LP Rodger at 1148K–L. See also Orbit Valve [1993] 4 All ER 165 (afrmed by the Court of Appeal: see [1995] 1 All ER 174) per Hobhouse J at 173f: “The parties are always able, by the choice of appropriate language, to draft their contract so as to produce a diferent legal efect. The choice is theirs. In the present case, there would have been no problem in drafting the contract so as to produce the result for which the plaintifs have contended; however, the contract was not so drafted and contains only generally working and is seriously lacking in clarity.” 101 Canada Steamship Lines Ltd v The King [1952] AC 192 (hereinafter Canada Steamship). 102 Ibid. per Lord Morton of Shuna at 208.

41

greg gordon indemnity and hold harmless clause. However, the contractor contended that the clause was inefectual: the operator was seeking to be reimbursed for a loss incurred at least in part as a result of its own negligence; however, the clause did not expressly state that it would operate in these circumstances. Applying Canada Steamship, the Court of Appeal held that the clause contained words which were wide enough to have the efect of conferring an indemnity upon the indemnifed party relative to the consequences of its own negligence. However, it refused to give efect to the clause on the basis that the second limb of the Canada Steamship test was not satisfed. The words were not necessarily directed towards the indemnifed party’s negligence, but could just as easily have been intended to exclude another head of claim, namely breach of statutory duty.103 This head of claim was not either too remote or too fanciful to be in the parties’ contemplation. The claim therefore failed. The result in Orbit Valve has received some support. Rainey, a leading commentator and practitioner in the feld, recently endorsed the question posed by Lord Steyn in Orbit Valve: “Why do draftsmen not take note of the impact of a clear and consistent line of judicial decisions?”104 That is a fair question. It would certainly make life easier if draftsmen were aware of these rules and followed them. However, it could be argued that the more fundamental questions to arise from Orbit Valve pertain to the reasoning of the court, rather than the drafting of the parties. Is it really appropriate for the court to take a clause that is, in its own terms, perfectly clear, deem it ambiguous because it does not expressly deal with a matter likely to be of no signifcance to the parties in question, and thereby render the clause unusable? How are the interests of justice served when an artifcial construction is applied to perfectly clear wording, leading to an unexpected result and uninsured losses? Does the occasional and highly specifc use of canonical construction really make the law clearer or fairer when the law generally favours an approach based upon textual analysis? In the Scots case of Nelson,105 the court was faced with a factual situation and contractual clause that was very similar to that in Orbit Valve. A worker employed by a contractor was injured on an ofshore installation. The operator settled the employee’s personal injury claim and intimated a claim under the indemnity against the employer of the injured worker. The indemnity and hold harmless provision envisaged that the employer would take responsibility for “any and all losses, claims, suits, demands . . . and causes of action in respect of death or of injury to [the employer’s] personnel . . . howsoever caused”. The employer argued that the indemnity did not apply in circumstances where the indemnifed party was seeking to be protected against its own negligence. However, the court rejected this argument. At frst instance, the judge attached great weight to the fact that the integrity of the reciprocal arrangement of risk allocation would be seriously undermined if the employer’s argument were successful; there was no evidence that the parties considered the issue of negligence to be especially important in their system of risk allocation, and to give primacy to this issue would disrupt the parties’ own carefully considered scheme. On appeal, the 103 Orbit Valve [1995]1 All ER 174 per Steyn LJ at 181h–j. 104 S Rainey, “Construction of Mutual Indemnity and Knock-for-Knock Clauses”, in B Soyer and A Tettenborn (eds), Ofshore Contracts and Liabilities (2015), 67–107 at 107, Informa. 105 Nelson v Atlantic Power and Gas Ltd, 1995 SLT 102.

42

risk allocation in oil and gas contracts in english law Inner House of the Court of Session upheld his judgment. The Inner House endorsed the reasoning of the Lord Ordinary, and considered that it fortifed its view that the clause should be given efect to.106 However – while there is an intriguing hint that the court would have been willing to distinguish Canada Steamship and decide on the basis of the Lord Ordinary’s analysis had this been necessary107 – the primary basis for the Inner House’s decision was its interpretation of the requirements of the Canada Steamship test.108 The Inner House considered that where (as here) the indemnifed party’s actings causing the loss could be characterised both as negligent and a breach of statutory duty, then while Canada Steamship might prevent the operation of the indemnity in relation to negligence, it did not preclude the indemnity from taking efect in relation to the breach of statutory duty.109 This is a markedly diferent approach to Canada Steamship than that taken in Orbit Valve. The House of Lords had the opportunity to consider the construction of oil and gas risk allocation clauses in London Bridge, a Scots case which (like Orbit Valve) arose out of the Piper Alpha disaster. Unfortunately, however, the court did not have the opportunity to settle which of the two competing approaches was to be preferred, as the wording of the contractual risk allocation clauses was materially diferent to those discussed in Nelson and Orbit Valve. In London Bridge, the operator sought to use indemnity and hold harmless provisions in order to recover from the various employers of personnel injured or killed in the disaster sums in respect of damages paid in order to settle the personal injury claims of the injured and the families of the deceased. After proof, it was held that the accident had occurred as a result of both the negligence of the operator and that of an employee of a specialist valve contractor.110 The operator was again met with the argument that the clause did not provide it with a remedy as it required to be interpreted contra proferentem,111 and did not deal sufciently clearly with the issue of negligence to satisfy the Canada Steamship requirements. As noted above, this time, the indemnity clauses112 were drafted diferently to the clauses in Orbit Valve and Nelson. The clauses were rather inelegantly expressed and did not address the question of negligence as expressly as one might have desired. However, they did make express provision for how issues of contributory negligence were to be resolved.113 This fact was sufcient to permit the court to conclude that the parties had intended that the mutual indemnity and hold harmless provision would take efect even in circumstances where one of the parties was seeking protection from

106 Ibid. per the Opinion of the Court at 104. 107 Ibid. 108 Ibid., pp. 102–103. 109 Ibid., 103. 110 This fnding was signifcant as the relevant clause contained a “sole-negligence” carve-out, and had liability not been at least partially the fault of another party, the pursuer’s claim would have failed: London Bridge (Inner House) 2000 SLT 1123 per LP Rodger at 1131. 111 See also the discussion at section 6.1, above. 112 The case involved a multiplicity of parties; seven separate sets of indemnity provisions were under consideration. 113 See the extracts from the relevant contracts reproduced at London Bridge 2000 SLT 1123 at 1126–1129.

43

greg gordon consequences of its own negligence.114 The claim for indemnity therefore succeeded. Although London Bridge was concerned with a clause that was markedly diferent from that in Orbit Valve, it can still be noted that the overall tenor of London Bridge difered from that in Orbit Valve, with the court in London Bridge being led more by the wording of the clause itself and showing markedly less concern that the clause was a trap for the unwary. Away from the oil and gas context, the House of Lords has had an opportunity to consider the continued role for the rules in Canada Steamship following Lord Hofmann’s restatement. HIH Casualty & General Insurance Ltd v Chase Manhattan Bank115 (HIH Casualty & General Insurance) was concerned with the interpretation of an insurance policy designed to pay out to the insured (who were investors in flms) in the event that the flms in which they had invested did not make enough money to permit the repayment of the investor’s loans. This rather specialised “high risk, high-premium”116 insurance product was designed and marketed by a frm of brokers who also acted as the insured’s agent when presenting the proposal to the insurers. The policy contained a “truth of statement” clause which, put shortly, provided that the insured (i.e., the investors) “would have no liability of any nature to the insurers for any information provided by any other parties”, and in particular that any information either provided or not disclosed by the brokers would not be a ground by which the insurers could avoid or escape their liability to make payment under the policy. This clause was included in the policy in order to protect the insured against the risk of unwittingly becoming responsible for anything said or known by the many other players involved in the procurement of fnance for any of the flms in question.117 In the event, there were substantial shortfalls in the flms’ revenue and the investors claimed under the policies. The insurers repudiated liability on the grounds of misrepresentation and non-disclosure, either fraudulent or negligent, on the part of the broker. The insurers contended that the wording of the “truth of statement” clause was not sufciently specifc to meet the terms of the Canada Steamship rules in that although the wording was sufciently wide to extend to negligence or fraud, it could just as readily be intended to exclude other causes of action which it was not fanciful to imagine the parties to have had in contemplation, such as innocent misrepresentation or non-disclosure. The House of Lords agreed relative to fraud, holding, by majority, it to be “a thing apart”,118 a species of liability which “must be 114 See London Bridge (House of Lords) 2002 SC (HL) 117 [2002] 1 All ER (Comm) 321 per Lord Mackay of Clashfern at paras 40–43. 115 [2003] UKHL 6 [2003] 2 Lloyd’s Rep 297. For an excellent summary, see E Peel, “Whither Contra Proferentem?” (n. 82, above), at pp. 61–64. 116 Lord Hofmann at para 25. This paragraph contains an evocative description of the flm investment industry which the author commends to the reader. 117 A multiplicity of parties were involved, including the broker: see HIH Casualty & General Insurance per Lord Hofmann at paras 26–33. The insured needed to be protected against the risk (not a fanciful one, given the manifold duties generally imposed by the law upon an insured, the contract being one of utmost good faith) of being deemed to have constructive knowledge of things known by persons deemed to be the insured’s agent. 118 HIH Casualty & General Insurance per Lord Bingham at para 15. His Lordship continued: “Parties entering into a commercial contract will no doubt recognise and accept the risk of errors and omissions in the preceding negotiations, even negligent errors and omissions. But each party will assume the honesty and good faith of the other; absent such an assumption they would not deal.”

44

risk allocation in oil and gas contracts in english law excluded in clear and unmistakeable terms on the face of the contract”.119 Importantly, however, the House of Lords unanimously rejected the insurers’ argument relative to negligence. Although, as Peel notes, “[t]here was no suggestion from their Lordships that [the possibility that the parties might have had in contemplation innocent misrepresentation or non-disclosure] was regarded as fanciful”,120 the so-called Canada Steamship rules were relegated to the status of helpful but non-determinative guidance121 as opposed to a code. As Lord Bingham put it: The passage does not provide a litmus test which, applied to the terms of the contract, yields a certain and predictable result. The courts’ task of ascertaining what the particular parties intended, in their particular commercial context, remains.122

In turning to that task, the House of Lords had no hesitation in holding that the parties had used “comprehensive language, clearly chosen to give [the investor] an extended immunity”.123 It found “nothing commercially surprising in this interpretation”,124 as in a transaction of this kind “the possibility that [the broker] might make and fail to correct a representation which was later held to be both untrue and negligent would be very real”.125 Thus, while it would appear that contra proferentem has survived the purge of “[a]lmost all of the old intellectual baggage of ‘legal’ interpretation”, it has not emerged unchanged from its brush with Lord Hofmann’s restatement. Fraud continues to be “a thing apart”; negligence, it would seem, is not. As HIH was decided at the height of the contextual era, its authority might be thought to be questionable following Arnold v Britton. However, while the case does, at points, advert to considerations of commercial common sense, the judgment is fundamentally a close textual analysis of the wording of the insurance policy in question. As such, it is submitted that it is compatible with Arnold v Britton. The indemnity and hold harmless clauses presently in wide use within the oil and gas industry make express reference to the parties’ intention that the clauses will be efectual irrespective of negligence or breach of duty, whether statutory or otherwise.126 It would seem to be foolhardy for contractual draftsmen deliberately to deviate from that practice. However, in the event that a clause is, upon examination following intimation of a claim, found to have been drafted in a way which is less explicit on this

119 Ibid. per Lord Bingham at para 16. 120 E Peel, “Whither Contra Proferentem?” (n. 82, above), at 62. 121 As Lord Hofmann notes at paras 61–63, they had also been characterised as mere guidelines not to be mechanistically followed in Smith v South Wales Switchgear Co Ltd [1978] 1 WLR 165; Hollier v Rambler Motors (AMC) Ltd [1972] 2 QB 71; and Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd [1983] 1 WLR 964. 122 HIH Casualty & General Insurance per Lord Bingham at 11. 123 Ibid. per Lord Bingham at 12. 124 Ibid. See also Lord Hofmann at 67: “There is no inherent improbability in such an intention. As Rix LJ said, in a case like this the question of negligence can never be all that far from the contemplation of the parties.” 125 HIH Casualty & General Insurance per Lord Bingham at 13. See also Lord Hofmann at 67: “And it seems to me that the commercial objective of the Truth of Statement clause would be substantially undermined if Chase’s right to the policy monies depended upon an inquiry into whether Heaths had or had not taken reasonable care in checking the truth of representations or deciding which facts should be disclosed.” 126 See, e.g., LOGIC, Mobile Drilling Rigs at cl 18.8; see also the discussion the IMHH Deed at section 7.2, below.

45

greg gordon point than one might wish, London Bridge and HIH Casualty & General Insurance suggest that the courts may be willing to take a somewhat more forgiving approach than that seen in Orbit Valve. More radically – and more speculatively – if Lord Glennie’s dictum in Slessor127 fnds favour, then the contra proferentem rule will be found to be wholly irrelevant in the case of mutual indemnity and hold harmless clauses.128 6.2 Words delimiting the circumstances in which the indemnity and hold harmless provision will take efect Another important issue which arises in the drafting of an indemnity and hold harmless clause is the need to delimit the set of circumstances in which the indemnity is to take efect. The indemnity and hold harmless provision will not be being given generally, for all times and for all circumstances. Instead, it is given because it has been rendered necessary by the fact that one party intends to carry out work under a particular contract. Words need to be included in the clause which make this clear and state the circumstances in which the provision is to take efect. If this is not done, the provision is in danger of being either struck down as a result of its indeterminate scope, or of applying in circumstances not intended by the parties. However, and again, exercising prudence as a result of the possible continued application of the contra proferentem rule, draftsmen must give careful thought to the words they use. It is dangerous to draw too narrow a connection between the operation of the indemnity and the scope of work under the contract. This is so because there are many circumstances in which property may be lost or personnel injured or killed while they are on an operator’s platform. Taking the example of an injury on an ofshore platform, such an event may befall a worker as a result of an accident directly connected to the scope of work to be done under the contract. However, it may also occur as a result of an accident occurring while he was carrying out work, but for reasons wholly unconnected to it, or while he is on the platform but not engaged in this work.129 Smith v South Wales Switchgear is authority for the proposition that if an indemnity is drawn so that it takes efect only if injuries are sufered during the “execution of this Order”, it will capture only injuries occurring as a result of the doing of the contractual work.130 The dangers that this judgment poses to the party claiming the indemnity are clear. In Campbell v Conoco (UK) Ltd,131 it was argued that a clause which provided that the subcontractor would indemnify and hold harmless the lead contractor132 against all injuries sufered by the subcontractor’s employees “as a result 127 Discussed at section 5.1, above. 128 Lord Glennie’s dictum would not prevent the rule from continuing in operation in the case of simple indemnities. 129 A worker may, for example, be at rest in the accommodation module when a fre breaks out, as happened to many of those killed or injured in the Piper Alpha disaster. Or the injury or illness may be the result of a more mundane accident. For instance, the employee may be injured falling from his bunk (as occurred in Robb v Salamis M & I Ltd, 2007 SLT 158, albeit no indemnifcation issue was litigated here). 130 Smith v South Wales Switchgear [1978] 1 WLR 165 per Lord Keith at 178. 131 Campbell v Conoco (UK) Ltd [2003] 1 All E R (Comm) 35 (hereinafter Campbell). 132 The terminology used throughout the case is rather confusing: because the case concerns a subcontract which incorporated the terms of the lead contract mutatis mutandis, the contract describes the parties

46

risk allocation in oil and gas contracts in english law of or arising out of or in connection with the performance or non-performance of the contract” did not take efect in circumstances where the employee sufered injuries after being struck in the back by a blast of compressed air discharged from equipment which had nothing to do with the job he was doing, but which happened to be positioned close to the area in which he was working. The subcontractors contended that the indemnity would be triggered only if the injuries were directly attributable to the performance of work under the contract.133 The Court of Appeal held that the wording of that particular clause did not bear any such construction, and that the phrase in question was wide enough to encompass at least a situation where the injury occurred while the worker was engaged in his work,134 and may well have been wide enough to encompass a situation where injury occurred while the worker was at rest.135 However, the court also observed that “each contract depends on its own wording and context”. Thus, if they wish the clause to be efective for the whole period when the worker covered is on the platform, drafters of indemnity and hold harmless clauses must be careful not to use words which tie the triggering of the clause too closely to the performance of the contract or its works.136 6.3 Multi-party issues Sometimes, for instance, in a consortium or alliancing agreement, there will be more than two parties to the contract. This does nothing to alter the central concept of the indemnity and hold harmless clause. However, the addition of further parties does add a further layer of complexity to the drafting exercise. Slessor v Vetco Gray illustrates the difculties that can be caused by attempting to provide for an unfamiliar situation. This case concerned the interpretation of an indemnity clause contained not in a standard services contract but within a multi-party consortium agreement. The clause was in the following terms: “The Parties hereto mutually and irrevocably undertake to release, defend and indemnify each other for damage to any property and/or injury to/or death of the personnel of the others, arising out of or in connection with the Work, howsoever caused”.137 At frst glance, this may look very like a fairly standard mutual indemnity and hold harmless clause (albeit one in which the parties use the expression “release” rather than “hold harmless”). However, it difers from such a clause in one crucial respect. Ordinarily, A will indemnify B in respect of any injuries sufered by A’s personnel. This clause was held to achieve something quite diferent: A here indemnifes B and as operator and contractor. However, in context this is to be read respectively as contractor (Amec) and subcontractor (Salamis): Campbell at paras 6–7. 133 Ibid. at para 10. 134 Ibid. per Rix L J at paras 18–19. 135 This point was left open as it was not necessary for the court to decide it: see Campbell per Rix LJ at para 24. In Orbit Valve, similar wording was held not to limit the indemnity only to occasions when the injured party was actively carrying out work under the contract but to extend also to injuries occurring while the injured party was on the platform but at rest: Orbit Valve [1995] 1 All ER 174 at 186. 136 The modern LOGIC standard contract wording, “arising from, relating to or in connection with the performance of or non-performance of the CONTRACT”, is a good example of a broadly drawn formulation: see, e.g., LOGIC, Services, para 19.2(b). 137 Slessor v Vetco Gray (n. 31, above), at para 3.

47

greg gordon C for any injury sufered by B and C’s personnel while B gives a like indemnity to A and C in respect of injury or death to their personnel, and party C similarly indemnifes A and B. The court declined to interpret the clause contra proferentem, for the reasons which were given above, and recognised that it would be unusual for the parties intentionally to draft a clause in these terms.138 However, the court felt that the words used by the draftsman were clear and unambiguous, and that efect must be given to them.139 Given that the wording achieved the diametric opposite of one what would ordinarily expect an indemnity clause to be seeking to achieve, one cannot help but wonder if this may have been one of the occasions referred to by Lord Hofmann, where “the language has gone wrong”. However, this was a Scottish case, and no attempt was made to present an argument in accordance with Lord Hofmann’s restatement. Moreover, as we have already seen, following Arnold v Britton,140 English law has retreated from the high water mark of Lord Hofmann’s approach and would seem to be broadly in sympathy with the Scots approach to interpretation. 6.4 “Full and primary” In London Bridge, one of the arguments advanced in the attempt to defeat the operators’ claim under the indemnity was that, as the operators’ insurers had already settled the claims of the contractor’s employees or their representatives, the operators had sufered no loss in respect of which they required to be indemnifed. It was also argued that the operators’ insurers could not claim under the indemnity either, on the basis that both the insurers and the contractual indemnifers had ofered primary indemnities, and the insurers were therefore not entitled to rights of subrogation, but only to contribution;141 the argument ran that subrogation rights were enjoyed only by a secondary indemnifer who has paid “out of order”, to allow it to recover from the primary indemnifer who has a stronger obligation to settle the claim. The contract was silent on this aspect of the nature of the indemnity. On the particular facts of London Bridge, it was held that the contractual indemnity was primary, and the insurers were secondary indemnifers entitled to their right of subrogation. Thus, they could claim to be reimbursed in full. However, it must be emphasised that this decision does not settle all such debates once and for all; for instance, the decision was based at least in part upon the fact that there was no contractual obligation

138 Slessor v Vetco Gray per Lord Glennie at para 11. The court observed that the usual regime could easily have been put in place by substituting the contentious phrase with “their own personnel”. A yet safer way of drafting the clause might have been to individually list out each party’s responsibilities to the others. This can be cumbersome, particularly where multiple parties are concerned, but in such matters elegance is less important than clarity. Inverting the intentions of the parties, which seems to be what occurred in Slessor, may seem an extreme drafting error, but there is always a risk, when trying to draft indemnities in too concise a form, of inadvertently creating ambiguity, or of failing to consider which result would obtain under the clause in any given factual situation. 139 Slessor v Vetco Gray at para 11. 140 [2015] UKSC 36. 141 The Scots law of contribution is concerned with the respective liabilities of joint wrongdoers and is contained in s 3 of the Law Reform (Miscellaneous Provisions) (Scotland) Act 1940.

48

risk allocation in oil and gas contracts in english law upon the operators to insure.142 How the court would have decided the matter if there had been such an obligation is therefore an open question. Best practice (which does not appear to have been followed in the LOGIC Standard contracts, but which is in evidence in the Industry Mutual Hold Harmless Agreement) is therefore to expressly state that the indemnities are “full and primary”. 6.5 Defnitional issues Company groups In practice, it is common for the parties to intend the efect of the indemnity clause to extend beyond the immediate parties so as to include their respective groups. Thus, in an operator-to-contractor contract, a contractor will commonly agree to indemnify and hold harmless the operator’s group against loss of or damage to the contractor group’s property, and the injury or death of the contractor group’s personnel; likewise the operator143 will usually agree to indemnify the contractor group against loss of or damage to the operator group’s property, and the injury or death of the operator group’s personnel. Typically, an operator group will be defned so as to include the operator itself, its co-ventures, its and their respective afliates, and its and their respective directors, ofcers and employees and personnel.144 A contractor group will generally include the contractor, its subcontractors and its subcontractors’ own subcontractors of whatever level, and its and their respective afliates and its and their respective directors, ofcers and employees and personnel.145 This is done in recognition of the fact that contractors and operators will commonly use other members of their groups to carry out the activities and operations envisaged by the contract, and, if the benefts of the indemnity regime discussed at paras 6.16–6.18 are to accrue fully, it is necessary for the efect146 of the indemnity provisions to extend to all parts of the group. It will usually be the intention of the parties to establish an indemnity regime on a grouped basis. However, parties need to be mindful in such cases not just to scrutinise the indemnity clause but also carefully to check the defnition of “group” or whatever cognate expression is being used in the contractual documentation, and to ensure that the implications of accepting those indemnity provisions, read together with that defnition, are what is truly intended.147

142 London Bridge per Lord Hofmann at para 97. 143 The operator is generally referred to, in such contracts, as the “company”, and its group as the “company group”. However, in the interests of keeping terminology consistent the term “operator” is used throughout this chapter. 144 See, e.g., LOGIC, Services, cl 1.2. 145 See, e.g., ibid., cl 1.6. 146 The term “efect” is used because persons who are not party to a contract cannot be directly bound by it; the grouping provisions do not directly bind the non-signatory members of the group, but achieve their efect circuitously. If there is a contract between A and B, and a dispute arises between A and a company (B2) which is part of B’s group, if A and B’s contract states that neither will be liable for the relevant species of loss and provide each other with indemnities in respect of it, then, while B2 is still entitled to sue A if B2 sufers such a loss, A can claim an indemnity in respect of it from B. 147 A second Farstad v Enviroco case arising from the same incident as the frst but argued in the English courts and pertaining to the defnition of “subsidiary” of a parent company group is Farstad Supply A/S v Enviroco Ltd [2011] UKSC 16.

49

greg gordon Employees and personnel As we have seen, many indemnity clauses are concerned with dividing up responsibility for injuries, illnesses or fatalities sustained by people during the course of the project. The expectation will ordinarily be that each party takes responsibility for its own people. Great care needs to be taken in how this intention is expressed. In particular, it should not be expressed too specifcally or narrowly, by using a word which has a technical legal meaning, such as “employee” or even “worker”. Many people working on a platform will not be “employees” in the formal legal sense, but contractors. If the indemnity refers only to “employees” it is most unlikely to refect the will of the parties. The current general practice is to frame the indemnity using a less technical term, such as “personnel”, and then to defne that term broadly. Property Similarly, many indemnity clauses will be concerned with apportioning responsibility for damage to property. Again, it is important for the parties – most particularly, contractors and any subcontractors entrusted with expensive pieces of equipment which they do not own – not to be lured into a false sense of security by observing that the contract contains a clause of this nature, but to give critical thought as to its precise terms, and the terms of any accompanying defnition. Contractors will usually want “company property” to be given an expanded defnition so that it encompasses not just items owned by the company but all items on the platform which the contractor might be called upon to use and/or which might reasonably be afected by the works carried out by the contractor. The operator will not necessarily own all the expensive or important items of property on the platform: some may be hired, or subject to retention of title clauses, or may be the property of other contractors, for instance the drilling contractor. 7 The problem of multiple parties 7.1 Introduction to the problem As we have already seen, at any given time there is likely to be a multiplicity of parties with personnel and/or property on a production platform. We have already discussed the special difculties caused by ensuring that all subcontractors within a given contractual chain have appropriate risk allocation terms in place. This, however, is only part of the picture. Although the great majority of the undertakings with personnel on a platform at any given time are likely to be in a contractual relationship with someone, they will not all be in a contractual relationship with each other. They will all be involved in a contractual chain; however, they will not all be involved in the same contractual chain. This situation is demonstrated in Figure 2.3. If, for example, subcontractor D2 negligently causes an explosion which injures personnel from D, A, D1A and B2, then there will be a direct contractual link (and presumably an indemnity and hold harmless provision) between D2 and D. Moreover, it is likely that back-to-back provisions will exist which, issues such as insolvency and drafting errors apart, should have the efect of apportioning risk throughout D2’s contractual chain, i.e., from A, through D and D2 down to D2A. However, there is no contractual connection, and therefore no contractual indemnity, between D2 and 50

risk allocation in oil and gas contracts in english law Operator

A Lead contractor

B

C

D

Subcontractor

B1

B2

C1

C2

D1

D2

Sub-subcontractor

B1A

B2A

B2B

C1A

D1A

D1B

D1C

D2A

Figure 2.3 Simplifed example of contractual relations on a production platform.

D1A, or between D2 and B2. Unless the parties are willing take the view that such risk will be dealt with by the law at large – a decision that would be surprising, given that it would undercut much of the economic benefts of the indemnity regime – a means needs to be found of “flling in the blanks” in the indemnity regime. How can this be achieved? A system whereby the parties entering into individual mutual hold harmless agreements with all other contractors with whom they would otherwise have no contractual link is a theoretical possibility, but it would require signifcant administrative efort, expense and co-ordination and would have the potential to go awry, leaving dangerous gaps in insurance cover if, for instance, a new contractor or subcontractor not previously known to operate on the platform was awarded a contract and information about this was not promptly disseminated among the contracting community on that platform. This potential solution is therefore unattractive in practice. Until the early 2000s, the practice in the UKCS was generally that the operator and contractor would grant indemnities on a grouped basis, with the operator’s group defned so as to include “other contractors”148 and the contractor’s

148 This term would be defned so as to include all contractors involved in operations other than the contractor’s own subcontractors of whatever level. If one refers to Figure 2.3, if the operator (A) was contracting

51

greg gordon group defned so as to include its afliates and all subcontractors of whatever level. The operator provided the contractor with an indemnity in respect of all claims relative to the people and property not just of the operator, but also of all contractors with whom the contractor did not itself have a contractual relationship, and by the contractor providing an indemnity in respect of its group’s people and property not just to the operator, but also to the other contractors. The risks would then be further broken up and re-allocated along the back-to-back principles above. Thus the problem of multiple parties was dealt with by the operator acting as a fulcrum in the contractual arrangements. This provided a workable solution to the problem. However, the practice was difcult and time-consuming for operators to administer. It was also risky: if, for some reason,149 the back-to-back provisions broke down, then the operator would commonly fnd itself responsible for a greater degree of liability than it had anticipated or intended. For both of these reasons a more ambitious solution, modelled upon what had become the practice of certain operators,150 was instituted in 2002. This is discussed further below. 7.2 The Industry Mutual Hold Harmless Agreement (IMHH) The problems referred to above, and the fact that the industry perceived the potential for efciencies and savings,151 led to the creation by LOGIC’s Standard Contracts Committee of an Industry Mutual Hold Harmless sub-committee which, after extensive consultation, recommended the institution of a Mutual Hold Harmless Scheme and prepared a draft of the relevant Deed.152 This scheme was initially efected by the Mutual Indemnity and Hold Harmless Deed (the “IMHH Deed”)153 which had been signed by 93 parties154 when the scheme became efectual on 1 July 2002.155 A signifcant number of other contractors have subsequently entered the scheme. The scheme is presently administered by LOGIC156 (which is itself the benefciary of extensive simple indemnity provisions relative to losses which might arise as a result of the manner in with contractor D on this model, A would defne its Group so as to include all contractors of whatever level within the chain of contractual relationships fowing from its contracts with B and C. 149 E.g., drafting error such that the frst level of indemnifcation worked but the back-to-back provision was inefectual; or a failure to ensure that a back-to-back arrangement was entered into. 150 E.g. Shell. 151 See LOGIC, IMHH – Introduction and Background, available for download from: https://www. logic-oil.com/imhh (accessed 20 November 2022): “The IMHH Scheme has been created to manage the risks inherent in the industry in a much more comprehensive and efective manner. It operates on the premise that a company is in a much better position to protect and insure their own people and equipment. In doing so, companies can be more certain of the risks they need to insure and this, in turn, reduces multiple policies insuring the same risk.” 152 LOGIC, About: see text under the heading, “Introduction”. Note that although the word “indemnity” is not used by LOGIC in its introductory discussion of the Deed, as we shall see, the core risk allocation provision of the Deed is clearly a mutual hold harmless and indemnity clause. 153 Although the Deed bringing the scheme into existence is clearly titled the “Mutual Indemnity and Hold Harmless Deed”, it is known throughout the industry as the “IMHH Deed”. This chapter will follow the prevailing industry practice. The Deed is available for download from: https://www.logic-oil.com/sites/default/ fles/documents/IMHH%20DEED%202022%20FOR%20WEBSITE_1.pdf (accessed 20 November 2022). 154 Listed in Sch 1 to the IMHH Deed. 155 IMHH Deed, cl 7.1, read together with the date of the Deed. 156 IMHH Deed, cl 1.1, defnition of “Administrator”. The Administrator may be replaced when a majority of signatories believe it has carried out its duties incompetently: see IMHH Deed, cl 4.4.

52

risk allocation in oil and gas contracts in english law which it carries out its administrative function: see IMHH Deed, cll 4.2 and 4.3 read together with cl 6.1(ii)).157 LOGIC’s primary functions are to act as the signatories’ attorney in executing Deeds of Adherence with new parties, receive and intimate to the remaining parties any notices of withdrawal, ascertain and act upon the will of the signatories upon the occurrence of any of the events which allow signatories to vote on whether to terminate the scheme, and to maintain the IMHH website. The initial IMHH Deed expired at the end of 31 December 2011. A replacement 2012 Deed entered into force immediately upon the expiry of the 2002 Deed. It has since been superseded by a further replacement, the 2022 Deed, which entered into force in January 2022. All references to the IMHH Deed are to the 2022 version. The Deed’s core indemnity provisions The central purpose of the Deed is to provide an efcient and practical means of bringing into existence an efectual set158 of hold harmless indemnities between parties who, save for the purposes of entering into an agreement about the allocation of physical risks, would have no commercial imperative to contract with each other. The IMHH Deed achieves this by providing: the Signatories shall be solely responsible for and shall defend, indemnify and hold harmless the other Signatories and the other members of their respective Groups against all Claims arising from, out of, or relating to the Services in connection with: (i) personal injury to or sickness, disease or death of Personnel of the Indemnifying Signatory or any other members of its Group; and (ii) loss of, recovery of, or damage to any Property of the Indemnifying Signatory or any other members of its Group; and (iii) Consequential Loss sufered by the Indemnifying Signatory or any other members of its Group.159

The “indemnities given pursuant to this Deed”160 are expressed as being “full and primary” and to apply “irrespective of cause and notwithstanding the negligence or breach of duty (whether statutory or otherwise) of the Indemnifed Party and shall apply irrespective of any claim in tort, under contract or otherwise at law”.161 Taking the main indemnity and hold harmless clause phrase by phrase, in connection with all claims arising from, out of, or relating to the services that it covers,162 it provides a full and primary mutual indemnity and hold harmless regime drawn 157 LOGIC, is provided with a right to outsource the performance of its administrative duties: cl 4.4 states that the administrator shall be “entitled to discharge any of its obligations and/or duties under this Deed by procuring that such obligations or duties are performed on its behalf by another person”, to be known as the Service Provider. The service provider, like the administrator, may be replaced if the majority of the parties to the Deed so wish. The service provider does not, through the operation of the IMHH Deed itself, obtain a right to share in the simple indemnity and hold harmless provision granted to but one would presume that any person agreeing to act in that capacity would insist upon obtaining the beneft of a backto-back indemnity from the administrator in any contract appointing them to the position. 158 As we shall see, the Deed does not provide for a wholly comprehensive set of indemnities: matters such as pollution are not included. 159 IMHH Deed cl 2.1. 160 Note that this term is habile to include not just the main cl 2.1 indemnity but also others that may exist throughout the Deed – e.g. the indemnities given to the Administrator by cl 4.2 and 4.3. 161 IMHH Deed cl 2.2. 162 Note the important exceptions discussed below.

53

greg gordon on a grouped basis, operating irrespective of negligence or breach of duty, whether statutory or otherwise, and covering (1) personal injury, sickness, disease or death of an expansively drawn set of people defned as “Personnel”; and (2) loss of, recovery of, or damage to “Property” and (3) a class of loss defned as “Consequential Loss”. Thus, the IMHH Deed provides indemnities and a hold harmless provisions in respect of many, but by no means all, of the areas where the industry commonly uses mutual indemnity clauses.163 The indemnity clause is supported by a clause providing for a waiver of rights of subrogation164 and is a meticulously drafted clause which avoids the errors and ambiguities of many of the clauses which have been litigated in the past. Other important provisions The IMHH contains a number of other signifcant provisions which will be discussed briefy in turn. Entry into force As has already been noted, the 2022 IMHH Deed entered into force as between its original signatories on 1 January 2022.165 The Deed becomes efective as between those parties from that date.166 However, many parties have entered the IMHH scheme since it was instituted. Such parties do so by entering into a Deed of Adherence167 between themselves and all the other current members of the scheme, who sign per the Administrator as attorney.168 Such parties are known as “New Parties”169 and become members of the scheme only from the date on which a Deed of Adherence signed by that Signatory has been dated by the Administrator”.170 The IMHH website displays the date when each party to the Deed became a member of the scheme.171 The Deed expressly states that the core indemnity provisions contained in cl 2 do not apply and are not enforceable “in respect of any Claims arising out of events occurring prior to the date on which that Signatory became a Signatory”.172

163 The Deed does not cover, for instance, pollution: this is because “[t]he prevailing consensus in the industry was that the scheme should apply to personal injury, property damage and consequential losses only. Other areas of risk such as pollution were considered, but were ultimately discounted. One of the main reasons for having a mutual hold harmless arrangement in respect of a Signatory’s own property, personnel and consequential loss is that the Signatory is best placed to assess the value at risk and, if required, make the appropriate insurance arrangements. Pollution risks are less quantifable and hence would have created a complication to the IMHH provisions which was not widely welcomed”. See LOGIC, IMHH General Guidance, available for download from: www.logic-oil.com/imhh/general-guidance (accessed 27 April 2017) (hereinafter LOGIC, General Guidance). 164 IMHH Deed, cl 5. 165 IMHH Deed, cl 7.1, read together with the date of the Deed. 166 Ibid., cl 11. 167 Ibid., cl 4.1; see also Sch 2 to the Deed for the prescribed for of the Deed of Adherence. 168 IMHH Deed, cl 4.2. 169 Ibid., cl 4.1. 170 Ibid., cl 11. 171 See www.logic-oil.com/imhh/signatories (accessed 27 April 2017). To view the dates, click on the hyperlink embedded within the name of each of the parties. 172 IMHH Deed, cl 2.3.

54

risk allocation in oil and gas contracts in english law Geographical extent The Deed applies not only to services provided on or in the UK Continental Shelf and/or between the United Kingdom low water mark and the innermost boundary of the UK Continental Shelf, but also to those carried out within Irish territorial waters and upon the Irish Continental Shelf.173 As further noted below, the Deed does not apply to landward areas. Order of precedence The Deed expressly provides that it “shall not take precedence over, amend, modify, or apply to the terms of any agreement between Signatories entered into prior to, on or after this Deed becoming efective in relation to such Signatories”.174 It is worth emphasising that the Deed refers to “any” agreement. So where a particular incident involves a dispute between signatories to the Deed who are at the material time in a contractual relationship germane to the provision of Services as understood by the Deed,175 the specifc contract will take precedence over the general Deed. This is so even where the specifc contract is silent – either by accident or design – on the question of risk allocation. Thus, the IMHH Deed will not ride to the rescue if signatories to it are in a direct contractual relationship and omit to include indemnity provisions, or to include provisions which are inefectual. This is to provide signatories with the freedom to enter into diferent liability arrangements if they so wish. Extension of benefits to groups In common with many oil and gas indemnity clauses in the UKCS, the indemnity and hold harmless provisions in the IMHH Deed are drawn so as to be for the beneft only of the Deed’s immediate signatories but also to “the other members of their respective Groups”.176 The Deed defnes “Group” broadly so as to include inter alia “the Signatory in question and its respective Afliates, the Personnel of all of the foregoing and their Invitees”,177 and cl 6.1(i) utilises the provisions of the Contracts (Rights of Third Parties) Act 1999 to extend the benefts of the core, cl 2 indemnity provision, and also the waiver of subrogation clause, discussed immediately below, to all members of a Signatory’s Group. Waiver of rights of subrogation The IMHH Deed provides that the signatories thereto shall procure that their insurers shall waive their rights of subrogation or any other rights they may have to proceed against the other signatories or members of those signatories’ groups in relation to the matters covered by the Deed.178 Signatories are entitled to require each other to exhibit evidence of the waiver.179 The Deed also states that, in the event that 173 IMHH Deed cl 1.1, under the defnition of “Services”. 174 Ibid., cl 11. 175 See the defnition in the IMHH Deed, cl 1.1. 176 IMHH Deed, cl 2.1. 177 For the full defnition, see IMHH Deed, cl 1.1, “Group” read together with the other defned terms which themselves feature within that defnition. 178 IMHH Deed, cl 5.1. 179 Ibid., cl 5.2.

55

greg gordon the parties fail to procure the relevant waiver, the failing party’s rights to enforce its rights and benefts under the IMHH Deed shall be suspended.180 Right to defend As has already been noted, the obligation to indemnify and hold harmless is separate and distinct from the right and obligation to defend a claim; however, it is relatively common within the UKCS for the indemnifer to accept (or demand) responsibility for conducting the defence of any claim made against the party it is indemnifying.181 Under the IMHH Deed, the indemnifying signatory (i.e., the party who is, in the event of a claim, liable under the Deed to indemnify whosoever the claim is initially directed towards by the party sufering a loss) is entitled, but not obliged, to take over the conduct of the defence to any such claim.182 Where this option is exercised, the indemnifying signatory accepts certain obligations concerning the provision of information to,183 and consulting with,184 the indemnifed signatory, and is in turn entitled to reasonable assistance from the indemnifed signatory in the conduct of the defence. Parties to the IMHH Deed enter into a contractual relationship which radically reallocates the legal liabilities they carry. Where a company is a member of the IMHH scheme, other members of the scheme are likely to feel entitled to refrain from entering into mutual hold harmless indemnity contracts, and/or to purchase less insurance cover which they would otherwise have been likely to believe they required. It is therefore imperative that signatories to the Deed know with certainty which parties are members of the scheme at any given time. For this reason, the right to withdraw from the scheme is quite heavily regulated. Upon signature, the parties are locked in to membership of the scheme until 31 December 2031,185 subject only to potential escape points occurring on 31 December 2024 and 2027, but only in the event that they have provided the Administrator with 60 days’ notice of this intention.186 Exceptions A number of exceptions are carved out of the IMHH Deed’s core indemnity and hold harmless provisions or introduced by way of the Deed’s defned terms. Each of these has a very signifcant practical impact upon the Deed’s scope. Each is discussed below. Transportation by air excluded None of the cl 2 indemnities applies or is given in relation to “any Claims arising from, out of or relating to the transportation by air of any member of a Signatory’s Group”. Thus, transportation by air falls wholly outside the scope of the IMHH Scheme.

180 Ibid., cl 5.3. 181 See para 14.8. 182 IMHH Deed, cl 3.2. Where it does not take up this option, the Indemnifed Party assumes an obligation not to settle the case without the consent of the Indemnifying Signatory: IMHH Deed, cl 3.4. 183 IMHH Deed, cl 3.2(ii). 184 Ibid., cl 3.2(iii). 185 IMHH Deed, cl 7.2. 186 IMHH Deed, cl 7.3. The Administrator is obliged by the same clause to notify all other parties in writing.

56

risk allocation in oil and gas contracts in english law Landward areas Despite the fact that, in practice, onshore works will quite commonly be the subject of indemnity provisions similar to those governing ofshore,187 the IMHH Deed applies only to services and/or supplies “carried out on or in the United Kingdom Continental Shelf and/or between the United Kingdom low water mark and the innermost boundary of the United Kingdom Continental Shelf ”.188 Thus, services or supplies carried out on the landward side of the UK low water mark are excluded. This has the efect of excluding onshore services, and will also have the efect of excluding any services or supplies carried out in the part of the territorial sea which lies on the landward side of the low water mark.189 It was decided to so limit the efect of the IMHH Deed because The general industry view was that extending the IMHH to cover onshore activities as well as ofshore activities could have inadvertently led to the application of the IMHH in circumstances where it was not intended (e.g. motor vehicle accidents) and, as such, in order to avoid any ambiguity it was limited to ofshore activities only.190

Operators Although operators played a substantial part in bring it into being, the IMHH Deed is designed for, and signed by, the contracting sector. It is not well suited to operatorto-operator situations191 and, given that the operator sits at the top of the contractual pyramid, operators should already be in a direct or at least indirect contractual relationship with anyone who steps onto their platform for commercial purposes, and therefore in a position to negotiate whatever contractual risk allocation regime the parties to those contracts choose.192 Commentary and conclusions on the IMHH scheme Some brief concluding words on the scheme are ofered below. The contractual nature of the scheme It is important to appreciate that the Deed is not a document in the nature of a piece of legislation, imposing conferring rights and imposing obligations by virtue of the mere fact that it exists. It is a contractual solution to a problem; it is just not the same contractual solution that the oil and gas industry previously deployed. It is axiomatic, therefore, that it takes efect only between those parties who have executed it, or those who have entered into a Deed of Adherence. It needs to be emphasised that it cannot simply be assumed that all oil and gas contractors are adherents to the Deed. In particular, not all drilling contractors have signed the deed, although

187 LOGIC, Services is intended to be applied both on- and ofshore. Note also that the pursuer in Slessor v Vetco Gray was working onshore when he was injured. 188 IMHH Deed, cl 1.1, defnition of “Services”. 189 A surprisingly large area, although not one in which large amounts of oil and gas operations have been undertaken: see para 4.1. 190 See LOGIC, General Guidance. 191 The possibility of an operator-to-operator IMHH Deed is discussed at para 14.65. 192 See LOGIC, General Guidance, under the heading “Can Operators Sign?”

57

greg gordon the numbers to have done so are gradually increasing. It is understood that the drillers’ general reluctance to adhere to the Deed is because the potential losses which may be sufered by drilling contractors, whose equipment is extremely valuable, and who will commonly have relatively large numbers of personnel on a rig or platform, is substantially larger than most contractors, and the drilling community have, by and large, yet to be persuaded of the value of the scheme. Other contractors will need to take appropriate steps to accommodate the risk caused by this refusal, either by seeking an indemnity from the operator, entering into a mutual hold harmless and indemnity arrangement directly with the driller, or simply by accepting the exposure and insuring accordingly. In addition, and as noted above, the Deed’s order of precedence clause means that the Deed is rendered irrelevant if there is any direct contractual relationship bearing upon “Services” as understood by the Deed. Potential dangers of the IMHH The IMHH was launched with considerable fanfare and broad industry backing193 and has quickly been accepted by the majority of the contracting community in the UKCS. It provides a solution to a problem which has caused considerable difculties and inefciencies for some time. There is no reason to believe that in general terms the scheme has been, or will continue to be, anything other than a success. However, the IMHH initiative – or perhaps more properly, the initiative’s application in practice – is not without its potential pitfalls. The Deed, while by no means long, is of necessity relatively complex. While many of the contractors who operate within the UKCS are large and sophisticated companies possessing the understanding and resources to ensure that their personnel understand the full import of the scheme, this is not true of all contractors. It is least possible that, in spite of the eforts which have been made by LOGIC to promote an understanding of the Deed, not all signatories to the scheme are fully conversant with all its subtleties. Although experienced oilfeld contractors should be aware of the Deed’s provisions, new contractors, some of whom work in the ofshore feld only semi-regularly, are signing up all the time. Moreover, the natural turnover of staf within even the more established players means that ongoing training and education is essential. In addition, and rather perversely, as the Deed becomes more broadly accepted, there is the danger that familiarity may breed contempt. Procedures may become sloppy, and parties may simply assume that everyone working on a particular rig or platform with whom they are not in a contractual relationship is a subscriber to the IMHH Deed. As we have already seen, however, this is unlikely to be true of the driller, and may not be true of all other contractors. If the contractor is to minimise risk it is incumbent upon it to ascertain whose property and personnel is or will be on the platform at the material time and to check that those on board fall within the ambit of the IMHH scheme, or to obtain a binding contractual undertaking from the operator that – apart from true third parties – only subscribers to the IMHH will be permitted onto the platform. None

193 It was launched under the auspices of LOGIC and with the support of UKOOA, IMCA, OCA and WSCA.

58

risk allocation in oil and gas contracts in english law of this is intended to attack the principles underlying the IMHH Deed but simply to urge due care in the scheme’s application. 8 Liability for “consequential” loss In addition to containing the indemnity and indemnity and hold harmless provisions described above, a wide variety of oil and gas contracts will also include clauses which exclude, limit194 or put in place an indemnity and hold harmless regime195 in respect of liability for so-called consequential or indirect losses.196 Such provisions are commonplace in operator-to-contractor contracts197 and are seen in some – but by no means all – contracts between oil companies.198 Where such clauses are used, their purpose is to protect parties from the full consequences of causing certain types of loss, typically delays in or loss of production, loss of use, loss of profts, and business interruption. The quantum of such losses is generally difcult to anticipate in advance and may be hard to quantify even retrospectively.199 However, they tend to accrue on a daily basis and, if a particular feld, facility or critical piece of infrastructure is “knocked out” for a sustained period of time, these losses can rise to extraordinary and alarming levels. As such, they are frequently either excessively expensive or simply impossible for a contractor to insure against, and the industry has by and large accepted that such losses should lie where they fall. Thus, we can see that economic factors very similar to those that drive the indemnity regime underpin the exclusion or limitation of consequential losses. 8.1 What is “consequential loss”? There is considerable uncertainty over what, if any, fxed meaning the term “consequential loss” conveys. The prevailing interpretation under English law has, for some 194 In so far as these contracts “exclude” or “limit” such loss they are subject to the provisions of the Unfair Contract Terms Act 1977, but not the 1999 Regulations. However, the courts have repeatedly held that, while they will always consider full circumstances of the case (see, e.g., Motours Ltd v Eurobell (West Kent) Ltd [2003] EWHC 614 (QB)), it will be only in unusual circumstances that they will hold a business – business exclusion of consequential loss clause unreasonable: see Wessanen Foods Ltd v Jofson Ltd [2006] EWHC 1325 (TCC); Photo Production Ltd v Securicor Transport Ltd [1980] AC 827. 195 A carefully drafted mutual indemnity regime is thought necessary where the intention is to extend the efect of the consequential loss provisions to groups: if there is a contract between A and B, and a dispute arises between A and a company (B2) which is part of B’s group, a simple exclusion or limitation of consequential loss will not be binding upon B2. However, if A and B state that neither will be liable for the other group’s consequential losses and provide each other with indemnities and hold harmless provisions in respect of such loss, then, while B2 is still entitled to sue A if B2 sufers a consequential loss, A can claim an indemnity back from B. 196 The precise terminology varies: sometimes “consequential loss” or “indirect loss” is used on its own. Sometimes “damages” is used instead of “loss”. See the discussion in London Bridge (Inner House) 2000 SLT 1123 per LP Rodger at 1156D–E. 197 They appear throughout the LOGIC standard contracts series. See, e.g., LOGIC, Supply of Major Items at cl 23. 198 See A Jennings, “FPSO Agreements” and J Izod, “Satellite Tie-Back and Processing Agreements”, both in M David (ed), Oil and Gas Infrastructure and Midstream Agreements (1999), at 209f and 170, respectively. London, Langham Legal Press. 199 Note the difculties which occurred in this regard in BHP Petroleum Ltd v British Steel plc and Dalmine SpA [2000] 2 Lloyd’s Rep 277 (hereinafter BHP Petroleum).

59

greg gordon time, been that the expression is almost a term of art, which usually denotes a reference to the category of losses encompassed by the second limb of the rule in Hadley v Baxendale.200 On this analysis, if a contract excluded liability for consequential loss, the parties would still be able to recover the sort of damages envisaged by the frst limb of Hadley v Baxendale, which is to say, losses “such as may be fairly and reasonably be considered . . . [as] arising naturally, i.e. according to the usual course of things, according to the breach itself ”. However, the parties would not be entitled to recover in respect of the second limb, which is to say, for damages such as may reasonably be supposed to be in the contemplation of the parties at the time they made the contract as the probable result of the breach. Now if the special circumstances under which the contract was actually made were communicated by the plaintifs to the defendants, and thus known to both parties, the damage resulting from the breach of such a contract, which they would reasonably contemplate, would be the amount of injury which would ordinarily follow from a breach of contract under those special circumstances.201

If this line of reasoning was ever sound, one could question whether it is consistent with the modern approach to contractual interpretation. Imposing a fxed meaning to a phrase smacks of the application of the canonical approach of an earlier age, rather than the approach currently ascendant in English law, which is fundamentally directed towards the import of the text used within its contractual (but not necessarily commercial) context.202 There are some – albeit equivocal – signs that the English judiciary may be preparing to adopt a diferent view to this question.203 If so, it may come to follow the approach to this question taken in Scotland, which rejected the “consequential loss as second limb” approach in London Bridge.204 In London Bridge, the court held that it was not appropriate to attach “a special, technical meaning” to the phrase “indirect or consequential loss” so that the phrase would be “interpreted as referring to losses which would be regarded as indirect or consequential under 200 (1854) 9 Ex 341. The line of reasoning sprang from Millar’s Machinery Co v David Way & Son (1935) 40 Com. Cas. 204 and Saint Line v Richardsons [1940] 2 K.B. 99 and has subsequently been consolidated by decisions such as British Sugar P.l.c. v N.E.I. Power Projects Ltd (1997) 87 B.L.R. 42; Deepak Fertilisers and Petrochemicals Corporation v I.C.I. Chemicals & Polymers Ltd (1999) 1 T.C.L.R. 200; Hotel Services Ltd v Hilton International Ltd [2000] 1 All E.R. (Comm.) 750; McCain Foods G.B. Ltd v Eco-Tec (Europe) Ltd [2011] E.W.H.C. 66 (T.C.C.); Markerstudy Insurance Co Ltd v Endsleigh Insurance Services Ltd [2010] E.W.H.C. 281 (Comm.); Glencore Energy U.K. Ltd v Cirrus Oil Services Ltd [2014] E.W.H.C. 87 (Comm.); Polypearl Ltd v E. On Energy Solutions Ltd [2014] E.W.H.C. 3045 (Q.B.). 201 (1854) 9 Ex 341, 355–356 per Alderson B. 202 See the discussion of Arnold v Britton at section 5.1, above. 203 In Scottish Power UK plc v BP Exploration and Drilling [2015] EWHC 2658, Mr Justice Leggatt stated obiter (at para 180) his concurrence with a leading commentator’s critical analysis of the “consequential loss as second limb” approach. In Transocean Drilling UK Ltd v Providence Resources plc [2016] EWCA Civ 372, Lord Justice Moore-Black, delivering a judgment concurred in by Lord Justices McFarlane and Bridges, noted (at para 15) that “the expression ‘consequential loss’ has caused a certain amount of difculty for English lawyers”, and thought it “questionable” whether some of the “consequential loss as second limb cases” would be decided in the same way today. In Saga Cruises BDF Limited & anor v Fincantieri SPA [2016] EWHC 1875 (Comm), Deputy High Court Judge Sara Cockerill QC considered (at para 198) “the consequential loss as second limb” line of authority to be “well-established”, although she did qualify this by adding the rider “on the current state of the law”. 204 As we shall see, the detailed reasoning is contained in the Inner House phase of the case. However, this reasoning was expressly approved of by Lord Mackay of Clashfern in the House of Lords: 2002 SC (HL) 117 [2002] 1 All E.R. (Comm) 321 at para 69. Lord Mackay’s judgment was in turn concurred in by all the other judges in the House.

60

risk allocation in oil and gas contracts in english law the rules in Hadley v Baxendale and subsequent cases”.205 Instead, while prior court decisions were certainly a relevant consideration, they were not to be thought of as determinative. The court’s primary task was to interpret the words in the context in which they were used.206 When such an approach was taken, the Court of Session held, and the House of Lords confrmed, that the disputed loss in London Bridge207 did not fall to be considered as “indirect or consequential” but was a direct loss.208 8.2 Commentary and conclusions on consequential loss It is wholly understandable that the parties to oil and gas contracts should seek to exclude, place limits upon the recovery of, or seek indemnities in respect of, certain types of potentially exorbitant losses. It is, however, perhaps unfortunate that the term “consequential loss” was ever adopted as the vehicle for attempting to do so. Even when the term was thought to be a synonym for the losses envisaged by the second leg of the Hadley v Baxendale test, it created difculties. It is not always easy to ascertain whether a particular type of loss is going to fall within limb one of the test or limb two. There was a clear danger that the contractual draftsman would assume that the expression was a convenient shorthand for all of the types of potentially exorbitant or hard to insure losses that he wished to exclude. However, it was not. Many of the most onerous losses which might be sufered during oil and gas operations do not clearly fall under the rubric of the second limb of Hadley v Baxendale, but may instead be captured by the frst limb of the test. This can be illustrated by considering the factual circumstances described in BHP Petroleum. In this case, a contract was entered into for the procurement of steel for a gas reinjection pipeline. The purpose of the project was to improve production from a producing oilfeld. The steel was in due course found not to conform to specifcation. It required to be replaced and the gas reinjection programme was delayed. In such circumstances, it is not just the cost of replacing the pipe that “[arises] naturally, i.e. according to the usual course of things, according to the breach itself ”. It is also plainly apparent that production from that oil feld will be delayed for a period of time while these works are carried out. Production lost during that period of time would therefore appear to be a direct loss, not a “consequential or indirect” one. So, a clause which – without saying more – excludes “consequential or indirect” loss is unlikely to exclude the loss described. And while the London Bridge approach is, it is submitted, a just

205 Lord Rodger at 1154. 206 London Bridge (Inner House) 2000 SLT 1123 per LP Rodger at 1156. 207 The circumstances were rather unusual. Because the operator had a material connection to Texas, they feared (with considerable justifcation) that if the damages claims of the persons injured and those of the families killed in the disaster were not settled on a generous basis, they might be sued in Texas, where the level of awards of damages were signifcantly higher than in Scotland. The claims of the injured parties and the families of the deceased were therefore settled on a “mid-Atlantic” basis, which is to say, at a level somewhere between traditional Scots awards and those which could be expected in the state of Texas. The defenders accepted that the element of the Scots element of the damages award was a direct loss, but that the “mid-Atlantic” augmentation was an indirect or consequential loss. 208 For a fuller account see G Gordon, “The Exclusion of Consequential Loss in the Piper Alpha Case: A Study of Lord Rodger’s Method of Textual Analysis”, in Simpson, Styles, Wilson and West, Pragmatism, Continuity and Change: Essays in Honour of Prof Angelo Forte, 2016 Aberdeen University Press pp. 434–465.

61

greg gordon and sensible one, it does little to make the situation more straightforward. Although, as with the contextual approach more generally, it may very well make the outcome of any given litigation fairer, the uncertainty that it engenders arguably increases the chances of litigation occurring in the frst place and may very well make the scope of enquiries of any litigation which does result more wide-ranging. Against this rather shifting background, the safest way in which to construct a clause excluding or limiting the type of losses which the oil industry has tended to design as “consequential or indirect” is to make the working presumption that those words will of themselves convey very little reliable protection, and to defne all essential content into them. This means that the draftsman must give consideration to the types of uninsurable or potentially exorbitant losses that might befall the project and must list them out in the clause. This is the approach taken by most of the contractual documents in broad use in the industry just now. There is some disagreement on whether the best practice is to defne consequential loss by reference to a closed list, or to state that consequential loss “includes but is not limited to” the potential heads of loss items listed. Jennings favours the former approach on the basis of clarity,209 but the latter approach is the one which is used more commonly throughout the industry’s standardised contractual documentation. Thus, the Industry Mutual Hold Harmless Deed defnes consequential loss as: (i) consequential loss under applicable law; and (ii) loss and/or deferral of production, loss of product, loss of use and loss of revenue, proft or anticipated proft (if any) whether direct or indirect to the extent that these are not included in (i), whether or not foreseeable at the date of execution of this Deed.210

Finally on this topic, where, as will frequently be the case, the contract contains both an indemnity regime and either an exclusion, limitation, or an indemnity in respect of consequential loss, thought needs to be given to the interaction between these provisions. They may very well overlap. It will generally be intended that the consequential loss clause will take precedence over the indemnity provisions.211 Whatever the parties intend, clear words should be used to express their wishes. 9 Overall limitation of liability An overall limitation of liability212 – sometimes described as a “liability cap” – is a clause that seeks to limit a party’s liability not by reference to particular species of loss, but by reference to a total sum payable. Such caps have the beneft of providing at least a level of certainty as to overall liability213 and facilitate the obtaining of insurance.214

209 A Jennings, “FPSO Agreements” in M David (ed), Oil and Gas Infrastructure and Midstream Agreements (1999) at p. 210, (n. 198, above). 210 See the IMHH Deed, cl 1. 211 See, e.g., LOGIC, Supply of Major Items, at cl 23. 212 A straightforward liability cap is of course a limitation clause and is therefore subject to a number of the controls imposed in UCTA, discussed at paras 14.26 and 14.66. For the reasons given at para 14.66, it is thought unlikely that the courts would decline to enforce a cap on liability in all but the most unusual circumstances. However, it is also possible to draft these clauses as indemnities: for reasons similar to those given in para 14.63, this is necessary if the intention is to extend the efect of the cap to groups. 213 See the discussion below relative to third-party issues. 214 P Roberts, (n. 76, above), para 13.61.

62

risk allocation in oil and gas contracts in english law In any given clause, the limit or cap may be expressed in one of a number of diferent ways: either as a fxed sum of money (often the maximum amount of insurance available at an economically viable rate) or as a proportion (or multiple) of the sum payable for the job. However it is expressed, the level at which the cap is to be set is essentially a commercial matter to be agreed between the parties, albeit if the cap is set too low, there is a risk that it could be deemed unreasonable for the purposes of the Unfair Contract Terms Act 1977.215 Liability caps are commonplace in oil and gas contracting practice.216 Such clauses are less frequently written about, simpler to understand and somewhat less technically demanding to draft than either indemnity or consequential loss provisions.217 However, they are not wholly without legal difculty. As we have already seen, an inappropriate level of liability could lead to difculties with enforcement under UCTA 1977; and it would be prudent for the clause to clarify if it is intended to take efect even against the proferens’ own negligence. Moreover, consideration needs to be given to the interaction between the liability cap and other risk allocation provisions, and in particular, third-party claims. In WesternGeco Ltd v ATP Oil and Gas (UK) Ltd,218 the liability cap was held to apply only internally to the parties and not to have the efect of limiting the amount payable as a result of an indemnity granted against third-party claims. Contractors therefore need to understand that if the liability cap is intended to take efect against an indemnity against third-party claims, very clear and specifc wording will have to be used. Liability caps are particularly important in circumstances where the parties are not prepared to agree to a mutual hold harmless indemnity and/or consequential loss regime.219 In these cases, the liability cap is the only clause which stands between a party and the prospect of unlimited – and potentially exorbitant – liability. The importance of the clause is perhaps most clearly demonstrated in the context of applications for access to infrastructure, where it is the expectation of the industry’s regulator that the infrastructure owner will be prepared to set a reasonable liability cap.220 A liability cap is arguably less important in cases where the parties have agreed upon an indemnity regime and/or exclusion of liability for consequential loss. The combined efect of these clauses is, or should be, to shut out a large number of the heads of claim which may result in losses disproportionate to the gains that a party had in contemplation when commencing on the project. However, even here a contractor should seek to obtain a liability cap for two reasons. First, some oil and gas projects are so valuable that even the potential losses which are unequivocally direct may be more than a contractor can comfortably bear. And second, even if 215. See the discussion of UCTA at section 4.1, above. See also P Roberts, (n. 76, above), para 13–81. 216. See, e.g., LOGIC, Supply of Major Items, cl 35. 217. This is not to say that such clauses create no difculties. One sometimes encounters liability caps which are subject to exclusions – where this is so, proper risk management requires that the efect of the exclusion must be clearly understood before the clause is agreed to. In addition, the interrelationship between the liability cap and the warranties provisions in the contract needs to be closely considered. 218. [2006] EWHC 1164 (Comm). 219. Indemnities are not uniformly accepted in all oil and gas contracts: see R Palmer, “Tie-In Agreements”, in M David, Oil and Gas Infrastructure Agreements, (n. 198, above), p. 231 at 239. 220. See “North Sea Transition Authority, Guidance on Disputes over Third Party Access to Upstream Oil and Gas Infrastructure”, NSTA 2022, para 79.

63

greg gordon the losses which the contractor fears should, strictly speaking, be shut out by the indemnity and/or consequential loss clause, there is always a risk that those clauses might for some reason be susceptible to challenge. A liability cap therefore performs the valuable function of providing a second line of defence against indeterminate or disproportionate liability. Any embarrassment felt by asking for a second line of protection should be swiftly overcome by considering just how dreadful the consequences are likely to be if something has gone wrong with the drafting of the indemnity or consequential loss clause.

64

CHAPTER 3

Contracting around tort defaults The knock-for-knock principle and accident costs Gideon Parchomovsky and Endre Stavang 1 Introduction Unknown to most, the gas and oil industry has largely opted out from standard tort liability. For several decades now, members of the gas and oil industry have contractually suspended the rules of negligence and strict liability in their dealings with one another. The industry has endorsed the knock-for-knock rule under which losses lie where they fall and govern their internal interactions. As a result, gas and oil companies are subject to a dual liability regime: vis-à-vis third parties, industry participants are subject to standard tort liability. Vis-à-vis each other, however, they operate under the knock-for-knock principle. The knock-for-knock principle eliminates the option of bringing private tort suits for harms incurred by gas and oil companies in their inter se relationships. Instead, it forces parties to incur the cost of the harms they sufered at the hands of other contractual parties. From an insurance perspective, the knock-for-knock principle relies exclusively on frst-party insurance, an intriguing feature that runs contrary to contemporary theorizing and practice.1 Given the moral hazard problem generated by the knock-for-knock regime, one would not expect it to see it in the real world. Yet, it exists. At frst glance, the gas and oil industry’s adherence to the knock-for-knock system is peculiar. Economically-minded tort theorists would expect the knock-for-knock system not to come into existence or unravel the moment a major accident occurs. By its very design, the knock-for-knock rule gives rise to a serious moral hazard problem. Left to their own devices, proft-maximizing actors who operate under this system are expected to underinvest in precautions and fall short of the optimal standard of care. Foreseeing this problem, rational actors would be expected to shun the knock-for-knock principle and subject themselves to the formal tort system while buying third-party insurance. Yet, insiders in the gas and oil industry maintain that

1 See Jefrey O’Connell & John Linehan, Neo No-Fault Early Ofers: A Workable Compromise Between First and Third-Party Insurance, 41 Gonz. L. Rev. 103, 107–108 (tracing the rise in the focus on thirdparty insurance in our tort framework). See also Nora Freeman Engstrom, An Alternative Explanation for No-Fault’s “Demise”, 61 DePaul L. Rev. 303 (2012) on the history of the American no-fault experiment explaining why automobile no-fault faltered as an efective legislative movement shedding light also on tort’s durability.

DOI: 10.4324/9781003206798-3

65

gideon parchomovsky and endre stavang the knock-for-knock system is efcient and vital to the successful operation of the gas and oil sector.2 In this chapter, we explore the content and structure of the knock-for-knock regime. Furthermore, we specify the conditions under which it may be preferable to standard tort liability coupled with third-party insurance. Our ultimate goal, however, is to conduct a comprehensive examination of the welfare implications of the knock-forknock principle. We conclude that despite the gas and oil industry endorsement of the knock-for-knock principle, from a societal standpoint the case is at best questionable. The chapter commences by tracing the historical origins of the knock-for-knock principle and explaining its efects. The discussion then analyzes the knock-for-knock principle from a social perspective, focusing on the potential for negative spillovers. Although gas and oil insiders consider the knock-for-knock system efcient because it economizes litigation costs, it is far from clear that it is also efcient from a social perspective. From a social vantage point, the system is desirable if and only if the private cost savings it afects are not outweighed by the moral hazard problem the system creates.3 Hence we review the various legal and non-legal mechanisms capable of ameliorating the moral hazard problem arising from the knock-for-knock regime. The four principal mechanisms are (1) contractual provisions that exclude intentional harm and harm resulting from ultra-hazardous activities from the knock-for-knock system; (2) deductibles and mandatory precautions requirements in insurance policies; (3) state and federal regulation; and (4) continuous interactions among repeat players that lead to cooperative behavior. We conclude that the knock-for-knock regime would be socially efcient if the activities of the contracting parties that may result in harm could be kept totally distinct from activities that may result in harm to third parties (we coin this condition “the separating equilibrium hypothesis”), or alternatively, if the aforementioned mechanisms were powerful enough to force the contracting parties to adopt the socially desirable level of precautions in all of their activities, both inter se and vis-à-vis third parties (we term this option “the optimal pooling equilibrium hypothesis”). The chapter seeks to contribute to two distinct literatures. First, we hope to enrich tort theory by studying an unusual liability regime that has evaded the searching eyes of tort theorists. It is important to explain in this context that the knock-for-knock 2 See, e.g., Cary A Moomjian, Contractual Insurance and Risk Allocation in the Ofshore Drilling Industry, Drilling Contractor, Jan./Feb. 1999, at 19–21; Trine-Lise Wilhelmsen, Liability and Insurance Clauses in Contracts for Vessel Services in the Norwegian Ofshore Sector – The Knock-for-Knock Principle, Scandinavian Yearbook of Maritime Law (SIMPLY) Conference paper for the Oslo/Southampton/Tulane Conference, 2 October 2012, at 101. Wilhelmsen provides, on the basis of several decades of Maritime Law scholarship in Norway on legal relationships involving knock-for-knock clauses, a very good account of the system and its rationale. Id. at 87–94, 95–101. The leading treatment however is (albeit in Norwegian) Hans Jacob Bull, Tredjemannsdekninger i forsikringsforhold (Oslo 1988). 3 Guido Calabresi famously termed this component “tertiary cost”. Guido Calabresi, The Cost of Accidents (1970); see also infra note 45. On the relevance of various types of administrative costs for legal rule design in this area, see also Ronald H. Coase, The Firm, the Market, and the Law 114–119 (1988); Steven Shavell, Economic Analysis of Accident law, 262–291 (1987); Michael Faure & Göran Skogh, The Economic Analysis of Environmental Policy and Law, 141–239 (2003); William D. Landes & Richard A. Posner, The Economic Structure of Tort Law (1987); and Richard A. Posner, Catastrophe – Risks and Response (2005).

66

contracting around tort defaults system is not a pure no-liability regime. Nor is it a no-liability regime with frstparty insurance. Rather, it is a hybrid, or a dual-liability system under which parties are subject to a no-liability regime and negligence or strict liability regime. Thus, the knock-for-knock system creates unique incentives for the parties that cannot be found in standard liability regimes. It also allows one to test the standard economic models of tort liability in a complex real-world setting. The second line of scholarly inquiry into which this chapter fts is the literature on private ordering or private production of law. Building on Robert Ellickson’s pathbreaking study of the extra-legal norms among neighbors in Shasta County concerning trespass by cattle,4 scholars have set out to study communities that opted out of the legal system and replaced it with a private set of norms. At the risk of mild overgeneralization, it can be said that as a rule those studies endorse private ordering as the superior form of rulemaking. Our case study is unique in several critical respects. To begin with, our case study involves only a partial opt-out from the legal system. As importantly, accidents in the gas and oil industry often afect multiple third parties, unlike previous studies which focused on largely self-contained communities or interactions, such as disputes among neighbors in an isolated community,5 or among merchants in the diamond industry,6 or fshermen in Massachusetts.7 One need only recall the British Petroleum (BP) oil spill in the Gulf of Mexico to recall this point. Relatedly, the harm resulting from an accident in the oil industry may be enormous and irreparable. Finally, the oil industry constantly expands its operations to new places. At the time of this writing, new sites are being developed of the coasts of Mexico, Brazil and north of the border between Russia and Scandinavia. Every such expansion extends the knock-for-knock regime to new territories, efectively transplanting it into new legal systems. At the end of the day, our analysis sounds a cautionary note as to the social desirability of the knock-for-knock principle. And while we stop short of calling for a legislative or regulatory ban, we believe that there is a strong prima facie case against extending the knock-for-knock principle to other industrial settings. Structurally, the chapter unfolds in four main sections. In section 1, we introduce the knock-for-knock principle and explore its historical origins. In section 2, we position the knock-for-knock principle within the theoretical literature on private ordering and social norms. In section 3, we specify the conditions under which a liability regime, such as knock-for-knock, that requires parties to bear their own loss may be socially desirable. In section 4, we illuminate the hidden costs of the knock-forknock rule and explain why it is unadvisable to extend it beyond its present scope. A short conclusion ensues.

4 Robert C. Ellickson, Of Coase and Cattle: Dispute Resolution among Neighbors in Shasta County, 38 Stan. L. Rev. 623 (1986). 5 See, e.g., Glen O. Robinson, Communities, 83 Va. L. Rev. 269, 306–309 (1997). 6 Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992). 7 André Verani, Community-Based Management of Atlantic Cod by the Georges Bank Cod Hook Sector: Is It a Model Fishery?, 20 Tul. Envtl. L.J. 359, 369–370 (2007).

67

gideon parchomovsky and endre stavang 2 The knock-for-knock principle George Fletcher famously theorized that tort liability is predicated on the principle of average reciprocity of risk.8 On this principle, tort liability should not attach to actors who accidentally expose each other to risks roughly similar in their gravity, while making reasonable eforts to avoid risky behavior. According to Fletcher, such risks are reciprocal in nature and accordingly their materialization should not trigger tort liability.9 In contrast, if an actor exposes another to a risk to which she is not herself exposed and harm follows, tort liability should be imposed because the risk was not reciprocal.10 Of course, whether a risk is reciprocal depends on the parties relevant positions towards one another and the nature of their activities. For example, an accident’s risk is reciprocal between two drivers, but non-reciprocal between a driver and a pedestrian.11 Hence if an accident occurred in the former case there would be no liability, whereas in the latter case liability would attach to the injurious driver.12 As it was frst developed, the knock-for-knock system seemed to perfectly embody Fletcher’s “reciprocity” principle. It was developed in London during World War II to reduce litigation costs from frequent naval accidents. Responding to the threat of German submarines, English ships sailed in the dark with their lights switched of in tight clusters.13 This reduced ship exposure to German submarines but increased the rate of collisions. Instead of engaging in expensive and prolonged litigation over those harms, the parties decided to subject themselves to the knock-for-knock principle, where each party agreed to bear its own costs.14 In this original setting, the risk each ship exposed to the other was reciprocal in nature, albeit not identical.15 The knock-for-knock principle was extended to other settings. Until the 1960s, the knock-for-knock principle governed automobile accidents in the United Kingdom,16 forcing drivers to buy frst-party insurance against accidents or bear their own losses. This system was subsequently abolished.17 The knock-for-knock principle was also extended to gas and oil companies and to diferent segments of the general shipping industry. In both industries, the knock-for-knock principle continues to persist. However, only in the gas and oil industry it became ubiquitous.18 8 George Fletcher, Fairness and Utility in Tort Theory, 86 Harv. L. Rev. 537 (1972). 9 Id. at 542. 10 Id. at 548. 11 See Gideon Parchomovsky, Fair Use, Efciency and Corrective Justice, 3 Legal Theory 347 (1997) (using this example as an illustration of a reciprocal risk). 12 Id. 13 See generally Frank H. Shaw, The Convoy Goes Through (1942), available at www.convoyweb.org. uk/extras/index.html (recounting frsthand experiences on World War II naval convoys). 14 See, e.g., Treaty with Great Britain on Marine Transportation and Litigation, U.S.-Gr. Brit., Dec. 4, 1942, 56 Stat. 1780 (1942). This is perhaps the best example of Fletcher’s “average reciprocity of harm.” See Fletcher, supra note 8, at 542. 15 The precise risk to which each ship was exposed depended on its position in the feet. Ships that sailed at the center of the feet were exposed to a greater risk. 16 Richard Lewis, Insurer’s Agreements Not to Enforce Strict Legal Rights: Bargaining with Government and in the Shadow of the Law, 48 Mod. L. Rev. 275, 285–286 (1985). 17 Insurance companies, however, still use knock-for-knock arrangements in their inter se relationships to economize on administrative costs. Id. 18 Wilhelmsen, supra note 2.

68

contracting around tort defaults 2.1 The industrial setting Mature petroleum provinces, such as the North Sea and the Gulf of Mexico, contain various patterns or mixes of gas and oil deposits and intricate infrastructure for extracting and transporting the resource. The rig is the system’s heart and is used for drilling. Surrounding the rig is a series of complex pipelines that transports the extracted gas and oil. Ships are often used as an alternative means of resource transportation, as well as a means for transporting workers and equipment. This interconnected infrastructure is connected to the outside world through ports and onshore pipeline endpoints, typically with some large industrial facility for handling, treatment and further transportation on land. The gas and oil industry has various positive and negative economic efects on its surroundings. On the positive side, gas and oil platforms create jobs and generate revenues for local businesses and communities. Furthermore, the gas and oil industry often directly invests in the local communities. On the negative side, however, the operation of gas and oil sites may cause adverse externalities. For example, it can harm the environment through degradation of marine resources, discourage tourism and possibly dislocate local populations. The ownership interests of pipeline infrastructure are managed by joint venture agreements between two or more companies.19 Typically, there are separate joint venture agreements for the feld and for each pipeline, though in Norwegian waters most of the pipeline infrastructure is managed through a single comprehensive joint venture agreement among all the oil companies with an ownership stake in the pipelines. The knock-for-knock system requires joint venture–related agreements to contain knock-for-knock clauses that bind all the parties involved in the venture. 2.2 Illustrating the operation of knock for knock As stated, the general thrust of knock-for-knock is to expose the contracting parties to as little tort liability as legally possible and replace it with frst-party insurance.20 To illustrate the efect of the knock-for-knock rule, let’s assume that a drilling company contracts to perform drilling services for an oil company, as was the case in the Deepwater Horizon blowout that occurred in 2010. Both the drilling company and the oil company had insurance against harm. Each company used subcontractors, who employed other subcontractors. At any given time, there was a mix of the drilling company’s and oil company’s employees, as well as employees of the various subcontractors on the drilling platform. Thus, an accident will sometimes initially 19 The joint ventures are typically not deemed to be separate legal entities. 20 Specifcally, a typical agreement would employ the following language: “Contractor shall indemnify Company Group from and against any claim concerning: a) personal injury to or loss of life of any personnel of Contractor Group, b) loss of or damage to any property of Contractor Group”, etc. Sometimes these wordings would exclude liability and order indemnifcation altogether without any exceptions. At other times, the clauses would include exceptions for intentional harms and gross negligence by the employees with whom the company in question could be identifed with, i.e., top management people. See sources indicating such capping and restriction of the scope for knock-for-knock agreements. For a current exposition of knock-for-knock clauses in contracts for vessel services in the Norwegian ofshore sector, see Wilhelmsen, supra note 2, at 87–94.

69

gideon parchomovsky and endre stavang impose costs on the oil company or its subcontractors, and at other times the cost will most immediately fall on the drilling company or its subcontractors. In this setting, the knock-for-knock systems accomplishes two things. First, if the oil company’s employees or property are harmed, the knock-for-knock clauses exclude tort liability, preventing the oil company from suing the drilling company in torts. Since the oil company’s insurance provider (under a frst-party contract) cannot obtain a subrogated claim against the drilling company, no tort claim arises.21 Losses are fnanced through the oil company’s insurance contract. The same holds true if the drilling company’s employees or property are harmed. Second, in the case of harm to the oil company’s employees or subcontractors, the oil company undertakes to indemnify the drilling company in the event of a lawsuit against it by a “member” of the oil company family.22 Thus, the drilling company does not have to call on its liability insurance. In this case, too, no lawsuit would be brought, either by the drilling company or by its insurance company (based on subrogation). The same holds true if a member of the drilling company’s family sues the oil company or one of its members. Therefore, knock-for-knock clauses not only bar direct tort suits by the contracting parties, but prevent the enforcement of subrogated claims that would produce much of the same fnancial efects as tort claims.23 The function of the knock-for-knock principle is to allocate harm to the “victim” ex ante while spreading losses through frst-party insurance.24 By efecting cost savings through the elimination of litigation and cheaper insurance,25 the knock-for-knock system increases gas and oil industry’s profts, thus it has become the darling of industry.26 But the picture painted by industry insiders is too rosy. Knock-for-knock arrangements give rise to a moral hazard problem.27 By exempting industry participants from liability for harm, they take away their incentive to act responsibly. Hence from a social perspective, a knock-for-knock arrangement is welfare-enhancing if it does not raise the number or severity of accidents or if the deterrence defcit it efects is smaller than the cost-savings it generates.28 The knock-for-knock principle only applies to the internal relationship among the contracting parties. It does not extend to harm caused to outsiders.29 In their interactions with outsiders, the contracting parties are subject to the standard rules of the tort system.30 Members of the gas and oil industry cannot unilaterally negate or modify their liability through their contractual arrangements toward third parties, 21 Id. at 93–94. 22 Id. at 91–93. 23 Id. at 85–86. 24 Id. at 83. 25 Id. at 96. 26 In Wilhelmsen’s opinion, such clauses should be fully upheld by courts during any contract enforcement in the wake of any accident. Id. at 102–111. 27 Id. at 98–99. 28 Id. at 101. 29 Id. at 90–91. Even though each party may still be liable to damages negligently inficted upon third parties, employees of the contracting parties are not considered third parties and still fall within the scope of the agreement: “[T]he Owner/Company or Charterer/Contractor not only agrees to be responsible for any damage that befalls the property of the company or the property or the persons of the employees, but also assumes responsibility for such damages throughout the group.” Id. at 89. 30 Id. at 83.

70

contracting around tort defaults such as fshermen, local governments, or communities. Accordingly, the activities of industry participants that afect third parties are governed by either strict liability or negligence. In efect, therefore, members of the gas and oil industry voluntarily subject themselves to a dual liability regime. Vis-à-vis each other, the knock-forknock principle provides that each contractual party must bear the harm inficted upon it by other contractual parties even when they behaved negligently. Vis-à-vis third parties, industry members must abide by the rules of the tort system or else face the consequences.31 3 Knock for knock, social norms and private ordering At frst glance, it could be concluded that the knock-for-knock system is another example of private ordering superiority over public ordering. This is because it constitutes another example where a private system contractually produced a better liability regime than provided by tort law. Yet, we are not at all convinced this claim is demonstrative. The “order without law” literature pioneered by Robert Ellickson32 might suggest that the knock-for-knock regime is just another example of efcient private ordering. In his famous study of Shasta County, Ellickson studied how farmers resolved trespass disputes. He reported that his study’s participants bypassed the legal system and followed a “live and let live” norm.33 In another important study of the diamond industry, Lisa Bernstein found that diamond merchants opted out of the legal system and developed their own set of courts and substantive rules.34 Bernstein argued that the private rules are clearly superior to the defaults provided by the legal system. Similarly, Eric Feldman, who studied the norms of dispute resolution among tuna fshermen in Japan, discovered that members of the fshing industry adopted a unique system of norms and institutions that allows for an expedient resolution of disputes.35 Yet there is a major diference between our study and those of Ellickson, Bernstein and Feldman. While the other studies pointed to the superiority of private ordering over state production of legal rules, which celebrated the advantage of private enterprise and the prevalence of social norms over formal legal rules, our study presents a much more challenging case. In the case of the knock-for-knock principle, it is impossible to unequivocally conclude that it is welfare-enhancing. This is because the operations of the gas and oil industry give rise to myriad efects that are borne by other parties. The scale of the operations of the gas and oil industry is enormous, and it is not nearly as self-contained as the diamond industry, the ranchers in Shasta County or the tuna fshermen in Japan. Moreover, accidents in the gas and oil industry have far-reaching consequences that go beyond the interests of other industry

31 Id. 32 Ellickson, supra note 4. 33 Id. at 15 et seq. 34 Bernstein, supra note 6. 35 Eric Feldman, The Tuna Court: Law and Norms in the World’s Premier Fish Market, 94 Cal. L. Rev. 313 (2006).

71

gideon parchomovsky and endre stavang participants. An oil spill, unlike a wandering cow or a dispute about a diamond or a tuna fsh, can devastate entire populations of people. Hence the knock-for-knock principle tests the outer limits of private ordering and must therefore be subject to careful scrutiny. Furthermore, it is clear that Lisa Bernstein’s explanation why the diamond industry opted out of the legal system does not apply in our case. Bernstein suggested that the diamond industry’s reluctance to subject itself to judicial review was largely motivated by concealment considerations.36 Industry participants wanted to keep the industry and its practices away from the public and therefore shy away from court proceedings that would lead to disclosure. Concealment is not a strategy that society should condone, let alone discourage, in oil industry. Accidents in the gas and oil industry often give rise to casualties and substantial environmental harms. Consequently, the operations of gas and oil companies ought to receive – and, in fact, do receive – close scrutiny from environmental and other regulatory agencies both ex ante and ex post. Furthermore, when an accident occurs it will be thoroughly investigated by the police and state or federal prosecutors. We, therefore, proceed on the assumption that the knock-for-knock principle maximizes gas and oil industry profts. The accepted rationale for the knock-forknock regime is that the savings generated in insurance and litigation costs exceed the possible increase in the cost of accidents because state regulation of health, safety and the environment provides an adequate deterrent against risky behavior.37 By adopting the principle, the contracting parties can then share a surplus, making knock-for-knock mutually advantageous.38 This does not mean, of course, that the preference of the gas and oil industry is optimal from a social perspective. The savings in tertiary costs39 may be overstated, the costs of harm may be underestimated, and the regulatory burden resulting from this system may be too heavy. A theoretically inclined reader may fnd an afnity between the knock-for-knock rule in the context of accidents and the “live-and-letlive” rule in nuisance.40 But just like the belief that nuisance law is growing organically, harmoniously and along a socially optimal path all by itself is not self-evident, the knock-for-knock’s contractual underpinnings have to be carefully evaluated from a

36 Bernstein, supra note 4 at 148 (noting that diamond dealers prefer arbitration to adjudication because it is, inter alia, more secret). 37 Bull, supra note 2, 333–357; Wilhelmsen, supra note 2, 95–101. 38 From the perspective of economic theory both strict liability as well as the negligence rule may be the basis for a preferable liability regime. But obviously a no liability regime is preferable if liability results in relative small deterrence efect compared to costs of operating the regime. See, e.g., R.H. Coase, The Problem of Social Cost, 3 J. L. & Econ. 1, 44 (1960); Louis Kaplow & Steven Shavell, Fairness Versus Welfare, 114 Harv. L. Rev. 961, 1011–1017 (2001). 39 Guido Calabresi defnes “tertiary costs” as those arising from administering an accident liability regime. These include the costs of reducing “primary costs” of accidents, which are the direct harm to victims and costs of accident avoidance, and “secondary costs”, the costs of spreading risks in an economically desirable way, such as through insurance. Calabresi, supra note 3, at 24 et seq. 40 See, e.g., Richard A. Epstein, Nuisance Law: Corrective Justice and its Utilitarian Constraints, 8 J. Legal Stud. 49, 82–87 (1979); Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes (1991).

72

contracting around tort defaults social welfare point of view.41 It is also not the case that the social norm literature has successfully demonstrated how private law can be replaced partly or in whole by private ordering through informal norm formation.42 In sum, the stakes in oil provinces like the Gulf of Mexico and the North Sea are so high that the extant academic attention given to this issue seems unreasonably low.43 4 The efect of knock for knock on social welfare Economic models of tort liability teach that a no-liability regime places the burden of investing in precautions solely on the potential victim.44 As a result, the injurer has no incentive to expend resources on accident avoidance or on mitigation of harm because they do not bear the cost of harms to third parties. The injurer will invest in precautions only to prevent self-harm and only if the expected harm is greater than the cost of the precautions.45 Any expenditure that exceeds the magnitude of the expected self-harm is a pure waste.

41 For recent nuisance scholarship taking a social welfare perspective, see, for example, Keith N. Hylton, The Economic Theory of Nuisance Law and Implications for Environmental Regulations, 58 Case W. Res. L. Rev. 673 (2008), Henry E. Smith, Exclusion and Property Rules in the Law of Nuisance, 90 Va. L. Rev. 965 (2004). 42 Robert Cooter, Structural Adjudication and the New Law Merchant: A Model of Decentralized Law, 14 Int’l Rev. L. Econ. 215, 226 (1994); Eric Posner, Law and Social Norms, 176, 179 (2000); and Robert C. Ellickson, The Market for Social Norms, 3 Am. L. Econ. Rev. 1, 30–36 (2001). 43 Most articles directly addressing the benefts and drawbacks of the knock-for-knock rule are still relegated to less-prominent journals and draw little attention as of yet. See, e.g., Nick Kangles et al., Risk Allocation Provisions in Energy Industry Agreements: Are We Getting It Right?, 49 Alta. L. Rev. 339 (2011); Christopher L. Evans & F. Lee Butler, Reciprocal Indemnifcation Agreements in the Oil Industry: The Good, the Bad, and the Ugly, 77 Def. Counsel J. 226 (2010). Even three years after the Deepwater Horizon explosion, only one article analyzing the knock-for-knock rule in light of the disaster has been written. Chidi Egbchue, Reviewing Knock-for-Knock Indemnities Following the Macondo Well Blowout, 7 Const. L. Int’l 7 (2013). 44 For the canonical expositions, see Shavell, supra note 3, and John Prather Brown, Toward an Economic Theory of Liability, 2 J. Legal. Stud. 323 (1973). Calabresi, supra note 3, is the main prior contribution. However, more than a century ago, the Austrian lawyer-economist Victor Mataja ofered the following observations on tort liability for, e.g., environmental harms. First, he pointed out that, a negligence rule is sometimes insufcient and strict liability is desirable on the basis of cost-internalization considerations. Second, he noted that a system of strict liability must rely on a broad conception of compensable harm is required, one that goes beyond pecuniary harm. Only in this case will it be safe to assume that cost internalization leads to a powerful enough incentive to take precautions. See Izhak Englard, Victor Mataja’s Liability for Damages from an Economic Viewpoint: A Centennial to an Ignored Economic Analysis of Tort, 10 Int’l R. L. Econ. 173 (1990). The interdependence of the issues of strict liability and compensable harm illustrated by Mataja with the help of the welfare economics of his day is confrmed by the now standard economic model of precaution, harm, and liability. See, e.g., Endre Stavang, Explaining Welfare-Based Torts, 4 Global Jurist Topics 1 (2004). Finally, “the economic model of precaution and liability” may be seen as a corollary of Aristotle’s famous quip about how un-owned resources are least well managed. In a similar vein, Dari-Matiacci and Parisi remark: “[A]lthough the cost-beneft calculus is a tool of modern economists, lawmakers have long understood the need to balance the costs and benefts produced by desirable activities. The intellectual roots of tort law can be traced back as far as the Roman principle that a party reaping the benefts (commoda) of an activity should also bear its costs (incommode). And in a broad sense, the economic analysis of tort law can be seen as a mere formalization of that pre-theoretical intuition.” Amsterdam Law School Legal Studies Research Paper No. 2013–09/Amsterdam Center for Law & Economics Working Paper No. 2013–01, 2. In sum, the model crystallizes sustainable (contingent) normative insights. 45 Robert Cooter & Ariel Porat, Does Risk to Oneself Increase the Care Owed to Others – Law and Economics in Confict, 29 J. Legal Stud. 19, 21–22 (2000).

73

gideon parchomovsky and endre stavang This means accident rates and their order of magnitude should be greater under the knock-for-knock rule than either a negligence or a strict liability rule – the two standard default settings for tort systems. Hence, the knock-for-knock rule would be welfare-enhancing only if one of the following conditions obtains: (1) “the perfect separating equilibrium”, where the inter se activities of the contracting parties are completely separate from the activities that afect the rest of society, such that the harms resulting from these activities will not afect third parties; (2) the expected private harm to the contracting parties from accidents exceeds the expected social harm and the parties contractually adopted optimal measures to address the private harm (we term this option the “perfect alignment hypothesis”); (3) the knock-for-knock rule creates negative externalities for third parties, but those are outweighed by the gains to the contracting parties. Otherwise, the knock-for-knock system is inefcient. We therefore proceed to discuss the likelihood that one of the previous conditions exists. 4.1 The perfect separating equilibrium In an ideal setting, the knock-for-knock rule could be efcient when there is perfect separation between the activities undertaken by the contracting parties toward each other and their activities that afect third parties. The perfect separation condition implies the existence of two distinct sets of relationships: one between the parties to the joint venture contract and another between those parties and the rest of the world. It also requires that there are no spillovers between the two relationships. Under these conditions, the erosion of the parties’ motivation to optimally invest in precautions would not adversely afect third parties. And if one assumes – as gas and oil insiders do – that the knock-for-knock system is internally efcient, then it is also socially acceptable because the external risks created by the contracting parties are governed under the tort system. Although the perfect separating equilibrium is a theoretical possibility, it is hard to see how it can exist. Gas and oil operations are complex, and it is unrealistic for contracting parties to separate activities that come under the knock-for-knock system from those that may afect third parties. In fact, the opposite is true. Experience teaches us that any accident potentially afects insiders and outsiders alike. This is clearly true for large-scale accidents, such as the BP leak, but it is also true for accidents whose efects are more limited. This does not mean, of course, that all accidents in the gas and oil industry generate third-party efects. However, this is not the relevant inquiry. All it takes to refute the perfect separating equilibrium hypothesis is to show that some accidents give rise to adverse third-party efects and that these accidents occur as result of the lower level of care from the knock-for-knock system. 4.2 The perfect pooling equilibrium The knock-for-knock system may also be socially desirable if it does not lower the contracting parties’ investment in precautions. Their precaution regarding investments may not change because various internal and external mechanisms – some of them private and some of them public – that apply to them. There are four possible mechanisms that may induce contracting parties to adopt the socially optimal level 74

contracting around tort defaults of care, notwithstanding the knock-for-knock system: (1) self-regulation; (2) repeat interactions; (3) health and safety regulation; and (4) mandatory provisions in insurance policies. A rich literature suggests that each of these mechanisms may ameliorate the moral hazard problem to which the knock-for-knock system gives rise. We examine each mechanism in turn. 4.2.1 Self-regulation Left unchecked, the knock-for-knock principle creates severe moral hazard problems. This is because each party bears none of the cost it imposes on other contracting parties but obtains the full cost of precautions. Thus, parties operating under the knock-for-knock principle have an inherent incentive to underinvest in precautions and take on an excessive level of risk. To address this problem, the contracting parties and insurance companies introduced contractual measures to ensure parties do not abuse the knock-for-knock system. In the main, those measures fall into two categories: exclusions and deductibles.46 As their name implies, exclusion clauses deny coverage to parties in cases of harm resulting from intentional acts and gross negligence. Exclusions provide an efective mechanism for controlling the insured’s behavior when the insurer can set the proper standard of behavior and can monitor the activities of the insured at a reasonable cost.47 Although the degree of an actor’s negligence cannot always be readily established, insiders in the gas and oil industry have developed shared conceptions of behaviors that fall outside ordinary negligence. The second standard mechanism employed to counter opportunistic behavior is deductibles.48 Deductibles are a standard feature in all insurance contracts.49 The purpose of deductibles is to prod the insured to invest in precautions up to the deductible amount. Deductibles should ideally be set to equal the cost of optimal precautions.50 In the real world, however, insurers often do not have the information necessary to achieve such accuracy.51 Yet, even if a deductible is set too high, it would induce the insured to take precautions.52 Although exclusions and deductibles operate within the knock-for-knock system’s framework, they also reduce the risk of harm to third parties. Accidents resulting in harm to third parties may arise from negligent acts, gross negligence, or intentional acts. It is reasonable to hypothesize that, on average, harm is greater in cases of gross negligence or intentional behavior. If this is true, then contractual exclusions dramatically lower the risk of harm to third parties. Furthermore, because it is difcult to distinguish between “merely” negligent acts and acts that constitute gross negligence, it is possible that the contracting parties

46 See, e.g., Jeffrey W. Stempel, Stempel on Insurance Contracts, 2–95 et seq. (2007). 47 Id. at 2–96. 48 Eugene R. Anderson et al., Insurance Coverage Litigation, 1–59 (2004). 49 Id. 50 B. Peter Pashigian et al., The Selection of an Optimal Deductible for a Given Insurance Policy 39 J. Bus. 35 (1966). 51 Id. at 42.; Ralph A. Winter, The Liability Insurance Market, 5 J. Econ. Persp. 115, 122–124 (1991). 52 Given that the underlying legal system constitutes the right incentives as in Shavell, supra note 3, at 5–47.

75

gideon parchomovsky and endre stavang would prefer cautious conduct and refrain from questionable behavior in their internal interactions.53 But irrespective of the veracity of these conjectures, because exclusions and deductibles lower the probability of inter se accidents, the likelihood of harm to the environment is reduced. Moreover, said contractual mechanisms also obviate the need for regulatory oversight to some extent, and thereby save societal resources. 4.2.2 Repeat interactions The moral hazard problem is further mitigated because members of the gas and oil industry continuously interact with each other. As Robert Axelrod54 and others55 established, parties engaged in repeated interactions tend to follow the “tit-for-tat” strategy. Specifcally, Axelrod demonstrated that in repeat interactions the dominant strategy is to cooperate as long as the other party cooperates and to punish defection by defection.56 An important implication is that long-term relationships between the same actors tend to foster cooperative behavior.57 Historically, the gas and oil industry constituted a near-perfect setting for a titfor-tat strategy. The industry comprises a relatively small group of members that interact with each other continuously. The companies in the gas and oil industry are also familiar with one another, and the reputational stakes they face are high. Consequently, members have an incentive to maintain a good relationship with their peers and not “defect” and risk retaliation. At least historically, the industrial setting in the gas and oil sector has been similar to the setting Ellickson found in Shasta County. And as the setting in Shasta County led to the adoption of a cooperative norm of “live and let live”, the conditions that prevail in the gas and oil industry are likely to induce a similar cooperative approach among industry members. Specifcally, we expect repeat interactions among members to exert a disciplinary efect on their members’ behavior through inter se relationships. Members are likely to pass small or short-term gains and avoid precautionary investments to reap long-term gains resulting from cooperation. If, indeed, the contracting parties are guided by long-term cooperation, their investment in precautions, even in their inter se relationships, may not fall short of the socially desirable standard. Consequently, the fear that the knock-for-knock regime gives rise to negative externalities could be much smaller than one would think. That said, one cannot rule out the possibility that some industry participants may not be adequately informed about the risks involved in gas and oil operations, which may lead to distortions in the pricing of

53 See, e.g., Robert D. Cooter & Thomas S. Ulen, Law and Economics (4th ed.) (2003); John E. Calfee & Richard Craswell, Some Efects of Uncertainty on Compliance with Legal Standards, 70 Va. L. Rev. 965 (1984). 54 Robert Axelrod, The Evolution of Cooperation: Revised Edition (2009). 55 See, e.g., Paul G. Mahoney & Chris William Sanchirico, Norms, Repeated Games, and the Role of Law, 91 Cal. L. Rev. 1281, 1289 (2003); George J. Mailath & Larry Samuelson, Repeated Games and Reputations: Long-Run Relationships 3–4 (2006); Robert E. Scott, Confict and Cooperation in Long-Term Contracts, 75 Cal. L. Rev. 2005, 2026 (1987). 56 Axelrod, supra note 54. 57 See, e.g., Mailath & Samuelson, supra note 55; Jane Sell & Rick K. Wilson, Maintenance of Cooperation: Expectations of Future Interaction and the Trigger of Group Punishment, 77 Soc. F. 1551, 1552–1554 (1999); Rachel E. Kranton, The Formation of Cooperative Relationships, 12 J. L. Econ. & Org. 214, 217–219 (1996).

76

contracting around tort defaults risks.58 Furthermore, the desire of regulators to make the industry more competitive by increasing the number of active frms in an oil producing locale may exacerbate knock-for-knock–induced moral hazards without ever reaching the elusive goal of perfect competition. 4.2.3 Health and safety regulation The presence of health and safety regulation is another factor that ameliorates the knock-for-knock regime’s potential moral hazard problem. Gas and oil companies operate within a regulatory framework that afects every facet of their activities. As a rule, regulatory standards are immutable and are not afected by private ordering. The presence of health and safety regulations forces gas and oil companies to comply with their content and ensure that they do not fall below the regulatory standard. Hence if the relevant regulatory standards are set appropriately, they should guarantee that the operations of oil and gas companies do not pose a risk to third parties. Critically, the need to comply with regulatory standard should also afect the level of care taken by gas and oil companies in their internal relations. Assuming, as we do in this section, that it impossible to separate activities that may cause harm to other contracting parties from those that jeopardize outsiders, the measures taken to comply with regulatory standards should also yield benefts to contracting parties. In other words, the reduction in the external risk brought about by regulation should also diminish the internal risk faced by the contracting parties. Consider, for example, safety regulations for the construction and operation of drilling rigs. The regulations are designed to protect the public at large. But frst and foremost, they protect the employees of the companies that operate the rig and perform the drilling. Of course, accidents may, and unfortunately do, happen, but clearly there are positive spillovers from the regulation for the contracting parties. Clearly, the more comprehensive the regulation, the lesser the scope of the moral hazard problem.59 In the extreme, if regulations covered every aspect of the contracting parties’ operations and were perfectly enforced, there would be no moral hazard problem. In this case, the regulation would force the parties to adopt the socially optimal level of care and the knock-for-knock regime’s sole efect would be to economize litigation costs. Yet, in the real world, not all aspects of the operations of gas and oil companies are subject to regulation. It must also be recognized that not all regulatory standards are set properly and that industry participants do not always comply with them. The administrative and information costs that are endemic to the regulatory process often create regulatory lacunas or gaps.60 This, in turn, makes it rational for efciency-minded regulators to rely on courts to fll those gaps.61

58 Wilhelmsen, supra note 2, at 101. 59 See, generally, on the comparative analysis of approaches to the control of risk, Shavell, supra note 3, at 277–291. 60 Shavell, supra note 3, at 282. 61 Susan Rose-Ackerman, Public Law versus Private Law in Environmental Regulation: European Union Proposals in the Light of United States and German Experiences, in Law and Economics of the Environment 13, 19–22 (Erling Eide & Roger van den Bergh, eds. 1996). For economic modeling of the argument that regulation and liability should be used jointly because neither policy used alone incentivizes frms to choose the efcient level of care, see Edward L. Glaeser & Andrei Schleifer, The Rise of the Regulatory State,

77

gideon parchomovsky and endre stavang Notwithstanding this observation, regulation is only as efective as its enforcement. Even the best regulatory standards would fail to yield socially optimal results if they were not properly enforced. The dynamic and highly complex operations of the gas and oil industry call for continuous monitoring and enforcement. Naturally, doing so is time-consuming and cost prohibitive, and even the best regulators may not live up to this standard. Tellingly, in its fnal report, the National Commission on the BP Deepwater Horizon Oil Spill concluded that the accident was avoidable and that the accident’s responsibility should be apportioned between the private companies that operated the site, “in the frst instance by BP, Halliburton and Transocean” and “government ofcials who, relying too much on industry’s assertions of the safety of their operations, failed to create and apply a program of regulatory oversight that would have properly minimized the risk of deep water drilling.”62 In the real world, the scope and substance of the regulation imposed on the gas and oil industry varies dramatically among countries. In Norway, for example, regulation of maritime and petroleum safety is largely based on self-assessment and self-regulation.63 The particular regulatory standards prescribe broad end-result while containing no reference to specifc behavior. It is therefore up to the industry to choose how to reach the performance requirements. If an accident occurs, sanctions may be imposed on the company. This, in turn, may lead to under-deterrence on the individual level. Moreover, the informational ex ante advantage enjoyed by frms in assessing the costs and benefts of rules aimed at inducing safety can be used by them to subvert the regulation and in some cases may even result in industry-wide collusions. 4.2.4 Insurance The economic and legal literatures suggest that insurance contracts signifcantly afect the behavior of the insured.64 Multiple insurance scholars developed this theme and termed the phenomenon “insurance as governance.”65 Insurance contracts often require policyholders to adopt various precautionary measures on the insured party as a precondition for collecting the insurance money. Furthermore, insurance policies 41 J. Econ. Lit. 401 (2003); Charles D. Kolstad, Thomas S. Ulen, & Gary V. Johnson, Ex Post Liability for Harm vs. Ex Ante Safety Regulation: Substitutes or Complements?, 80 Am. Econ. Rev. 888 (1990); Patrick W. Schmitz, On the Joint Use of Liability and Safety Regulation, 20 Int’l Rev. L. Econ. 371 (2000); Steven Shavell, A Model of the Optimal Use of Liability and Safety Regulation, 15 Rand J. Econ. 271 (1984). For a study contextualizing these and other theoretical contributions through analysis of Norwegian law, see Ekaterina Denisova, The Role of Environmental Civil Liability in Regulation of Marine Oil Pollution in Norway, 383 MarIus (Scandinavian Institute of Maritime Law) 1 (2009). 62 John M. Broder, Blunders Abounded Before Gulf Spill, Panel Says N.Y. Times, 6 January 2011 at A14, available at www.nytimes.com/2011/01/06/science/earth/06spill.html?_r=3&ref=gulfofmexico2010&. For broader discussions of the regulatory aspects of Macondo, see, e.g., Rebecca M. Bratspies, A Regulatory Wakeup Call: Lessons from BP’s Deepwater Horizon Disaster 5 Golden Gate U. Envtl. L.J. 7 (2011); Mark A. Cohen et. al. Deepwater Drilling: Law, Policy, and Economics of Firm Organization and Safety, Resources for the Future Discussion Paper 10–65 (2011); Sandra Zellmer, Joel A. Mintz, & Robert Glicksman, Throwing Precaution to the Wind: NEPA and the Deepwater Horizon Blowout 2 J. Energy Env. L. 62 (2011). 63 Anita Ronne, Alternative Approaches to Regulatory Agency Structures and Powers: Eastern and Western Europe, 15 J. Energy & Nat. Resources L. 41, 44 (1997). 64 Steven Shavell, On Liability and Insurance, 13 Bell J. Econ. 120 (1982). 65 See, e.g., Kenneth Abraham, Four Conceptions of Insurance, 161 U. Pa. L. Rev. 653, 683–697 (2013); Richard V. Ericson & Aaron Doyle, Uncertain Business: Risk, Insurance and the Limits of Knowledge (2004); Richard v. Ericson et al., Insurance as Governance (2003).

78

contracting around tort defaults employ exclusions and deductibles to ameliorate adverse selection and moral hazard problems that plague insurance markets. Consequently, insurance markets can supplement or even supplant private litigation, regulation, and social norms as a means for guiding behavior.66 We do not dispute the general observation that insurance companies can curb risky behavior. That said, the insurance market plays a very limited role in regulating the operations of large gas and oil companies around the world. To begin with, frst-party insurance contracts only target the behavior of accident victims. Obviously, frst-party insurance contracts cannot signifcantly infuence other companies’ level of care in joint ventures. One could nonetheless suggest that if the same insurance company were to insure all industry members, it could afect the behavior of all parties involved and induce them to adopt the appropriate level of care. This is not the case in the real world, however. Oil companies and many of their contractors sometimes self-insure and more typically own or control small insurance companies that provide them with frst-party insurance. The insured risks then get traded on reinsurance markets. Our conjecture is that this approach treats risk more as a basis for bets and fnancial calculations in the “Protection and Indemnities (P&I)” market67 and does not necessarily lead to a governance scheme, which is more likely to arise under a tort liability regime. 5 The hidden perils of knock for knock Thus far, our analysis has established that the parties’ contractual measures coupled with the regulations and norms born from repeat interactions do not ensure optimal investment in precautions; rather, the combined efect of these measures is to ameliorate the moral hazard problem to some extent. In light of this conclusion, it is impossible to say, in the abstract, that the private gains to the contracting parties from the knock-for-knock rule may be greater than the marginal reduction in deterrence afected by the rule. Hence, based on our analysis, we cannot assert with any degree of certainty that – as industry insiders insist – the knock-for-knock rule is efcient. In this section, we highlight four additional factors that cast further doubt as to the desirability of knock-for-knock arrangement in the gas and oil industry. The frst is the assumption introduced in section 4.2, namely risk interdependencies. We show that because the gas and oil industry has a high level of risk interdependencies, relatively minor oversight can rapidly evolve into major accidents. This means that it is critical to prevent even the smallest risks from arising. The second factor is what we call “litigation externalities.” As we will explain, because knock-for-knock clauses bar suits by other contractual parties, they efectively eliminate the most natural plaintifs, who are best positioned to sue for deviations from socially acceptable practices. The third factor we highlight in this part is harm to environmental interests. As we will

66 Faure & Skogh, supra note 3, 263, 264. 67 Remarks, although not as skeptical as here, on P&I clubs, are made in Jan C. Bongaerts & Aline F.M. De Bièvre, Insurance for Civil Liability for Marine Oil Pollution Damages, 12 Geneva Papers on Risk and Insurance 145 (1987); Faure & Skogh, supra note 3, at 273–274; Gary D. Libecap & Steven N. Wiggins, Contractual Responses to the Common Pool: Prorationing of Crude Oil Production, 74 Am. Econ. Rev. 87 (1984).

79

gideon parchomovsky and endre stavang demonstrate that the knock-for-knock rule exacerbates the risk of environmental harm relative to standard liability regime. Fourth, we will explain why ex post compensation does not provide an adequate solution for the erosion in deterrence resulting from knock-for-knock clauses. The harms resulting from accidents in the gas and oil industry are so grave that compensation after the fact would not be an efective remedy. In the case of large-scale accidents, the damage amounts are so substantial that the responsible party would prefer to fle for bankruptcy protection or pledge for government intervention to cap its liability to victims. Ironically, perhaps, the gravity of the harm prevents society from relying on ex post compensation, requiring instead that society puts the premium on ex ante deterrence. 5.1 Risk interdependencies As we explained in section 4.1, if the contracting parties agree to the knock-for-knock arrangement and their interactions can be kept distinct from actions that afect third parties, then society has no reason to worry about the efect of the regime on third parties. By contrast, if the contracting parties’ actions, especially their precautionary investment, adversely afect the public at large, then knock-for-knock clauses present a serious cause for concern. In the latter case, the contracting parties’ decision to forgo certain precautions on account of the knock-for-knock regime increases the risk to third parties. In the gas and oil industry, the degree of interdependence among risks is high. Relatively minor errors in the operation of oil rigs and drilling equipment may have disastrous consequences for the public at large. The Gulf of Mexico oil spill, also known as the “Macondo blowout”, provides a gruesome illustration. A White House commission reported and placed the accident’s blame on the cost-cutting policies adopted by British Petroleum and its partners. Specifcally, it stated that “many of the decisions that British Petroleum, Halliburton, and Transocean made that increased the risk of the Macondo blowout clearly saved those companies signifcant time (and money).” Furthermore, the report cautioned that “absent signifcant reform in both industry practices and government policies, [a blowout] might well recur.” While these fndings do not directly link the knock-for-knock regime to the oil spill, it does suggest that cutbacks on investment in precautions can have serious adverse efects on third parties. And although we cannot prove knock-for-knock clauses caused the insufcient precautions adopted by British Petroleum and its partners, our earlier analysis demonstrates a characteristic of knock-for-knock clause logic is that they are liable to efect a reduction in the level of care relative to standard tort liability. 5.2 Litigation externalities The deterrent efect of legal rules critically depends on enforcement. In the present context, liability is largely governed by private enforcement. Both standard tort liability and contract liability require the fling of private enforcement suits by victims.68 If private lawsuits are not brought, under-deterrence would result. Once litigation is

68 Shavell, supra note 3, at 262–277.

80

contracting around tort defaults incorporated into our analysis, another peril of knock-for-knock clauses comes to the fore. Knock-for-knock clauses “eliminate” the plaintifs who are most likely to sue – namely, the other participants in the joint venture. The other members of the joint venture are not only the most likely victims but also the parties who can most cost-efectively sue. The private parties involved in the venture are best situated to prove deviations from the socially desirable standard of behavior. Their daily presence on gas and oil sites coupled with their familiarity with safety standards puts them in a unique position to monitor the actions of their partners and collect evidence of sub-standard behavior. Hence they are the most natural private enforcers. The elimination of suits for negligence by other contracting parties leaves the lion’s share of the enforcement efort to third parties, who are not bound by knock-forknock clauses. However, in many cases the harm to third parties is too low to justify legal action. This is especially true in jurisdictions that do not recognize class actions and whose laws do not incorporate other procedural mechanisms for aggregating small claims. In this context, it behooves us to remind our readers that class actions are largely unrecognized in Europe,69 as well as in most other countries around the world.70 Furthermore, even in those cases in which the harm to third parties may be substantial, it may be too difcult or costly for them to prove negligence in court. Third parties wishing to sue for harms they sufered at the hands of gas and oil companies would be forced to expend signifcant resources to substantiate their claims. The high cost of litigation means that only parties who incurred a severe harm – that is, a harm greater than the expected cost of fling suit – should be expected to sue. The upshot is that many small and even medium-sized harms may be left unredressed. It is likewise critical to understand that litigation costs are not symmetrical.71 Furthermore, there is a substantial heterogeneity among defendants as far as litigation costs are concerned. The cost of litigation is higher for individuals who are not repeat players.72 Such parties must build their cases from scratch and incur expenditures that repeat players avoid on account of economies of scope and scale.73 To give a concrete example, compare the cost of defense for a gas and oil company that gets sued for negligence to the cost of a frst-time plaintif. The company typically has a legal department that is well versed in such claims after handling multiple similar suits. The plaintif, by contrast, would have to search for an attorney, who would have to educate herself about the facts of the case and the applicable law. This means that defendants enjoy an inherent cost advantage over most individual defendants.

69 See generally, Deborah R. Hensler, Goldilocks and Class Actions (response to Margaret H. Lemos, Aggregate Litigation Goes Public: Representative Suits by State Attorneys General, 126 Harv. L. Rev. 486 (2012)), 126 Harv. L. Rev. F. 56, 56–7 (2012) (noting that this issue is the subject of vigorous debate in the European Union, where advocates for private enforcement of antitrust and consumer protection law have struggled against those who champion traditional European reliance on public enforcement and deride proposals for “American-style class actions”.). 70 Id. at 57 (“[T]wenty-odd countries outside the United States that have adopted class actions, most have limited standing to represent a class to public ofcials or nonproft organizations vetted or approved by the government.”). 71 See generally Gideon Parchomovsky & Alex Stein, The Relational Contingency of Rights, 98 Va. L. Rev. 1313 (2012) (noting the phenomenon and analyzing its implications). 72 Id. at 1344 (explaining the cost advantage of repeat litigators). 73 Id. at 1318 (discussing the efect of economies of scale on scope on litigants).

81

gideon parchomovsky and endre stavang The corporations that participate in the joint venture could level the litigation playing feld to some extent on account of their being repeat players who are familiar with the legal machinery of litigation. But for the knock-for-knock clauses, those companies could bestow a positive externality on third parties and remediate to some degree the inherent under-deterrence problem. However, the knock-for-knock regime prevents this from happening. 5.3 Harm to environmental interests The environmental harm caused by various gas and oil companies are well-documented, and one may surmise that there are other long-term harms that have yet to materialize. Let us be clear: we do not dispute the evidence.74 But this is not the point. The critical question is whether the knock-for-knock principle contributes to these harms. Or, to state the question somewhat diferently: would those harms not occur under a negligence or strict liability regime? Although standard liability regimes are highly imperfect in protecting the environment, our analysis suggests that the knockfor-knock system exacerbates the risk of environmental harm. Environmental interests may be protected under a strict liability or a negligence rule. At least in theory, both negligence and strict liability can lead to desirable results, depending upon how the liability rules are operationalized by the relevant decision-makers.75 It is important that a negligence rule set the right legal standard: adopting either an overly lax or stringent standard would distort actors’ investment decisions, leading to too few or too many precautions. Under a strict liability regime, it is crucial to carefully calibrate the sanction to the harm caused by the activity.76 If the assessment of harm is too lax or too stringent, actors would be incentivized to underinvest or overinvest in precautions.77 Legislators around the world, often under the rhetoric of the so-called “polluter pays” principle,78 have increasingly gravitated toward strict liability as the main tort regime in this context.79 A major rationale for adopting strict liability concerns information asymmetries. For a negligence rule to function properly, the court weighs the marginal social costs and benefts of various courses of action and then adopts the optimal behavior standard. To perform this task, courts must consider (and monitor) all the actions tortfeasors can take. If, however, certain aspects of the tortfeasor’s

74 See, e.g., James Hansen & Makiko Sato, Greenhouse Gas Growth Rates, 101 Proc. Nat. Acad. Sci. U.S.A. 16109, 16109 (2004); Andrew K. Jorgenson, Global Warming and the Neglected Greenhouse Gas: A Cross-National Study of the Social Causes of Methane Emissions Intensity, 1995, 84 Social Forces 1779, 1981–1982; Max Blumer & Jeremy Sass, Oil Pollution: Persistence and Degradation of Spilled Fuel Oil, 176 Sci. 1121, 1122 (1972). 75 Shavell, supra note 3 at 121. Faure & Skogh, supra note 3, at 247–248. 76 Shavell, supra note 3, at 9, 128. 77 See supra Part I.C. 78 Hans Chr Bugge, The Principles of “Polluter-Pays” in Economics and Law, in Law and Economics of the Environment 53 (Erling Eide & Roger van den Bergh eds. 1996); Hans Chr Bugge, The Polluter Pays Principle: Dilemmas of Justice in National and International Contexts, in Environmental Law and Justice in Context 411 (Jonas Ebbesson & Phoebe Okowa eds. 2008); Faure & Skogh, supra note 3, at 26–30. 79 See, e.g., Ursula Kettlewell, The Answer to Global Pollution – A Critical Examination of the Problems and Potential of the Polluter-Pays Principle,3 Colo. J. Int’l Envtl. L. & Pol’y 429, 438–441 (1992).

82

contracting around tort defaults behavior cannot be monitored under a negligence rule, strict liability will tend to be superior. The reason is that strict liability dictates the behavior, which allows the actor to adjust any aspect of their controlled behavior, or about which they may have superior information.80 Another problem with negligence as the sole basis for imposing liability (as opposed to a hybrid system of negligence with partial strict liability regime, as is in the case of liability for ultra-hazardous activities) is that negligence determinations may in practice be reduced to enforcement of settled industry practices81 or regulatory standards.82 These sources may not necessarily represent the social ideal but be adopted by courts nonetheless, owing to institutional constraints and incomplete information. Unfortunately, although many national legal systems are improving their strict liability metrics concerning environmental harm, they are still far of the theoretical mark. This is because the legal system fails in defning compensable harm, both conceptually and institutionally.83 Moreover, insolvency84 and damage caps85 further complicate the task of “pricing” environmental risk through the tort system. More generally, there are so many loopholes and imperfections in strict liability’s current design that there is little reason to believe the tort system could fully internalize environmental risks.86 In practice, therefore, neither negligence nor strict liability provides optimal protection to environmental interests.

80 To elaborate, one can distinguish between two forms of precautions: level of care and level of activity. An example, in the context of noise from trafc, is the distinction between protective measures such as thicker walls and double paned windows (care) and the lowering of the amount of trafc (activity). Another example in the context of wind energy is the distinction between the improvement of windmill technology (care) and the number of hours of operation (activity). Finally, in the context of toxic releases from ships in fjords, bays and harbors, the distinction lies between the choice of chemicals and ship paint (care) and the number of trips (activity). In these examples, inclusion of the activity level in the determination of negligence may be hampered for two reasons. First, such inclusion requires precise information about the use that the actor obtains from carrying out the activity. Yet, the court may lack such information. Secondly, such negligencebalancing requires the fnding of a certain level of activity as a matter of fact, and that factual determination may be hard to make in practice. For these reasons, strict liability enjoys an advantage over negligence in certain pollution contexts. For more refned analysis adding more, see esp. Steven Shavell, Strict Liability Versus Negligence, 9 J. Legal Stud. 1 (1980). 81 Gideon Parchomovsky & Alex Stein, Distortionary Efect of Evidence on Primary Behavior, 124 Harv. L. Rev. 518, 545–546 (2010). 82 Endre Stavang, The European Court of Justice and the Environmental Liability Directive, 5 Environmental Liability 198 (2010). 83 Donald N. Dewees, The Role of Tort Law in Controlling Environmental Pollution, 28 Can. Pub. Pol’y 425, 430–431 (1992); Claus Ott & Hans-Bernd Schäfer, Widening the Scope of Environmental Liability, in Law and Economics of the Environment 91, 99–100 (Erling Eide & Roger van den Bergh eds. 1996); Faure & Skogh, supra note 3, at 284. On the new and so-called Ecosystem Services Approach and the worry of its lacking legal implementation in the Macondo context, see Carrie Presnall, Laura López-Hofman & Marc L. Miller, Can the Deepwater Horizon Trust Take Account of Ecosystem Services and Fund Restoration of the Gulf? 40 ELR 11129 (2010). 84 Faure & Skogh, supra note 3, at 267–270, 276–277. 85 Michael G. Faure & Hui Wang,Financial Caps for Oil Pollution Damage: A Historical Mistake?32 Marine Pol’y 592 (2008). 86 See, e.g., Debra L. Baker, Bankruptcy – The Last Environmental Loophole, 34 S. Tex. L. Rev. 379, 387, 397 (1993); Catherine Tinker, Strict Liability of States for Environmental Harm: An Emerging Principle of International Law, 3 Touro J. Transnat’l L. 155, 160 (1992); Palma J. Strand, The Inapplicability of Traditional Tort Analysis to Environmental Risks: The Example of Toxic Waste Pollution Victim Compensation, 35 Stan. L. Rev. 575, 590 (1983).

83

gideon parchomovsky and endre stavang Enter knock-for-knock. The knock-for-knock system does not take full account of environmental interests. Hence to some degree it widens the gap between the optimal level of protection and that taken in practice. This implies that knock-for-knock’s introduction marginally worsens the fate of environmental interests. However, considering the risk interdependencies introduced in section 3.2 and discussed in section 4.1, and the fact that they may create massive environmental harms that are not always measurable and may not even be readily observed, there is reason to believe that the knock-for-knock system presents a real threat to environmental interests. At this point, one may invoke the precautionary principle to call for a ban on knock-for-knock clauses. The precautionary principle maintains that if an activity threatens environmental harm, measures should be adopted to control it even if the precise causal relationship has yet been established.87 To rephrase, the precautionary principle embodies the idea that regulators should err on the side of safety. Although we are sympathetic to the precautionary principle, we do not think that its application in this case necessitates an outright ban on knock-for-knock clauses. Properly understood, the precautionary principle requires the imposition of regulation on the operations of the gas and oil industry. Of course, such regulation already exists. The precautionary principle does not prescribe the precise nature, or even the scope, of the regulatory measures that need to be adopted. Given the highly incomplete data on the relationship between knock-for-knock clauses and environmental harm, we feel that currently there is not enough evidence to legally ban knock-for-knock arrangements. But given the genuine uncertainty surrounding the social desirability of knock-for-knock clauses, it is not surprising that in some jurisdictions standard rules of contract enforcement are not fully applicable to such clauses. 5.4 Deterrence versus compensation One might argue that the erosion in deterrence that results from knock-for-knock does not pose a serious concern as long as injured parties are compensated after the fact. Extant theorizing suggests that compensation serves a dual purpose: it restores victims to their pre-accident state, and it deters tortfeasors from deviating from socially accepted standards of behavior.88 Corrective justice scholars emphasize the former purpose,89 while law and economics scholars put the premium on the latter.90 However, ex post compensation cannot always be relied on to afect the socially desirable level of deterrence. Paradoxically, after-the-fact compensation is especially inefective with large-scale catastrophes. In such cases, the aggregate harm typically 87 Stephanie Joan Mead, The Precautionary Principle: A Discussion of the Principle’s Meaning and Status in an Attempt to Further Defne and Understand the Principle, 8 N.Z. J. Envtl. L. 137, 138 (2004). 88 See Michael G. Faure, Calabresi and Behavioural Tort Law and Economics, 1 Erasmus L. Rev. 75, 94 (2008) (“When [as a result of liability] an enterprise is held to compensate the costs its activity generates, dangerous activities will become more expensive and the enterprise will, as a result of market forces, have an incentive to increase safety.”). 89 See Mark R. Reiff, Punishment, Compensation, and Law: A Theory of Enforceability 170 (2005) (“Most corrective justice theorists . . . contend that a person who wrongfully injures another has a moral obligation to compensate the injured party for his loss.”). 90 See Richard Posner, Frontiers of Legal Theory 266 (2001) (arguing that deterrence is more important than compensation).

84

contracting around tort defaults exceeds the fnancial resources of the responsible parties. And although criminal law is supposed to address the deterrence defcit, it often falls short of achieving this goal when corporate entities are involved.91 Obviously, criminal prosecution, even when successful, ofers little solace to tort victims. The key point for the purpose of this discussion is that after-the-fact compensation cannot be counted on to prompt members of the gas and oil industry to adopt the socially optimal level of precautions. In this context, the emphasis should be put on preventing accidents ex ante. Ex ante prevention is typically the domain of regulation.92 But given the inherent difculty in enforcing regulatory standards and monitoring compliance and because relatively minor failures may evolve into largescale accident, we believe that it would be a mistake from a societal perspective to allow the members of the gas and oil industry to contract into a liability regime whose efect is to cause a drop in the level of care. 6 Conclusion Can private ordering successfully replace standard tort liability in an industrial setting? In this chapter, we looked at this question by discussing the knock-for-knock regime that formally only governs the inter se relationships of industry participants in the gas and oil sector. Although industry insiders are convinced that the knock-for-knock rule is optimal for the gas and oil industry and that there is no real social downside to its adoption,93 our analysis provides a more cautionary note. The knock-for-knock regime lowers operation costs for gas and oil companies, but only at the cost of creating a serious moral hazard problem. From a societal perspective, the knock-for-knock rule is efcient only if the private cost savings it generates are greater than the increase in the rate and severity of accidents that result from its adoption. Our analysis suggests that the moral hazard problem to which the knock-for-knock rule gives rise can be efectively addressed only by far-reaching and wide-ranging regulation. In this regard, we think it is prudent to consider that both the informational asymmetries and the tendency to exploit it may vary greatly depending on the ownership structure of the operating companies. For example, in Norway, 83% of the largest oil company (Statoil) is owned by the Kingdom of Norway. This should be expected to infuence safety when compared to felds like Macondo in the Gulf of Mexico or felds in the developing world, where there is no similar government involvement. More generally, although the operations of gas and oil companies are regulated everywhere in the world, the scope and nature of the regulation varies dramatically from one country to another. The extent of government ownership of gas and oil companies can also afect the risk to which third parties are exposed. Where the government is the controlling shareholder of gas and oil companies, one can fnd, 91 Assaf Hamdani & Alon Klement, Corporate Crime and Deterrence, 61 Stan. L. Rev. 271 (2008) (challenging the conventional view concerning the deterrence value of corporate criminal liability and “demonstrating that harsh entity-level penalties might discourage monitoring for misconduct and undermine compliance incentives within professional frms.”). 92 See Yael Aridor Bar-Ilan, Justice: When Do We Decide, 39 Conn. L. Rev. 923, 932–938 (2007) (summarizing the role of ex ante regulation in the law). 93 Bull, supra note 2.

85

gideon parchomovsky and endre stavang on average, stricter regulatory standards and a higher degree of industry compliance. It should be noted, though, that it is difcult to fnd gas and oil sites where the previous conditions fully obtain. On the other hand, it is easy to identify gas and oil provinces in diferent parts of the globe where regulatory oversight is lacking, and government ownership of the venture is low to non-existent. At the end of the day, then, one can see the glass as half full or half empty. One fnal note is in order. Given the great variation in institutional quality and regulatory cultures around the world, a uniform approach to knock-for-knock clauses may be misguided. We conjecture that knock-for-knock clauses produce the greatest savings for gas and oil companies in developed countries, such as the United States, Canada and Europe, where litigation costs are very high. These countries have frst-rate regulatory institutions and adequate regulatory capabilities. Consequently, the risk posed by knock-for-knock clauses is relatively moderate. By contrast, the regulatory infrastructure in developing countries is often inadequate, and the use of knock-for-knock clauses poses a much greater risk of harm to the environment and third parties. This risk is compounded by the fact that the legal systems of many developing countries do not recognize class actions or a comparable mechanism for aggregating small claims. Correlatively, the cost savings enjoyed by gas and oil companies as a result from knock-for-knock clauses are more modest. If we are right, a tentative conclusion emerges: knock-for-knock clauses can be tolerated in developed countries where they produce the highest private gains while arguably posing a manageable public risk, and they should be banned in developing countries where, from our analysis, they yield modest private gains while posing a signifcant risk of harm. Acknowledgements We are grateful to Ian Ayres, Tom Baker, Shyam Balganesh, Derek and Jane Bambauer, Abraham Bell, Miriam Bitton, Ellen Bublick, Hans Christian Bugge, Fabrizio Cafaggi, Guido Calabresi, Alon Cohen, Simon Deakin, Erling Eide, John Goldberg, Sharon Hannes, Olav Hasaas, Keith Hylton, Jacob Nussim, Ariel Porat, Christopher Robertson, Carol Rose, Erik Røsaeg, Peter Siegelman, Henry E. Smith, Stephen Smith, Roy Spece, Alex Stein, and participants at the law and economics workshops at Bar Ilan and Tel Aviv faculties of law and the frst annual meeting of the Private Law Consortium for invaluable comments and contributions. For outstanding research assistance, we would like to thank Levi Morris and Ananth Padmanabhan.

86

CHAPTER 4

On knock-for-knock clauses and their optimal regulation Henrik Lando

1 Introduction The knock-for-knock system is a contractual system in which the parties to the system agree to cover all their losses through frst-party insurance and to dispense with tort liability. The system ensures that each party’s insurance contracts cover losses to the party’s own assets and own employees, and that the frst-party insurers cannot seek redress under tort law against the other parties. Moreover, while the system cannot dispense with tort claims raised by third parties, the system can allocate such risks to the better insured party (e.g., the operator of an oil rig rather than the contractors) through indemnity clauses. Liability exclusions are not at all uncommon in the business world but the socalled unqualifed knock-for-knock contracts exclude liability and indemnifes also for grossly negligent acts or even sometimes for losses caused by willful misconduct.1 Such exclusions are thought by some academics and judges to be unjust and/or to undermine incentives for precaution. This concern has been heightened after the tragic Macondo and Alpha Piper accidents. Whether for fairness reasons or out of a concern for precaution, courts have sometimes struck down unqualifed knock-forknock clauses. For instance, in a case where a vessel through gross negligence collided with an oil-platform, a Norwegian court set aside the exclusion of liability for gross negligence, arguing that “the lack of precaution was blameworthy and grave . . . and the lack of precaution could have had far worse consequences”. The same concern for proper precaution led to Parchomovsky and Stavang’s study in Chapter 3. Their study adopted a skeptical attitude towards unqualifed knockfor-knock contracts, especially in countries where it cannot be assumed that public regulation substitutes for tort law in securing a proper incentive for precaution. This chapter addresses two topics in relation to knock-for-knock contracts. The frst topic is the nature and purpose of knock-for-knock clauses, and why we observe them in some sectors of the economy but not in others. Why, for instance, do we observe knock-for-knock contracts mainly in the ofshore rather than in the onshore oil and gas sector? I will argue that the answers in the literature are not satisfactory, although they contain elements for an explanation. Instead, I suggest a simple model can be used to think about when liability exclusions are optimal.

1 Gross negligence is more often applied than willful misconduct and we shall focus on gross negligence.

DOI: 10.4324/9781003206798-4

87

henrik lando The other topic was raised by Parchomovsky and Stavang: when should courts uphold unqualifed knock-for-knock contracts? I broaden this question by inquiring not only what courts should do but also what lawmakers and public regulators should do to implement an optimal regulatory system. My point will be that civil fnes are meant to exactly counter negative efects on third parties. Instead of striking down unqualifed knock-for knock-clauses when an injurer has acted with gross negligence, I suggest raising fnes for such gross negligence. Fines do not jeopardize the signifcant savings of dispute costs that are associated with certainty of enforcement of knock-for-knock clauses. First, however, it is worth positioning the discussion in the context of the literature on exclusions of liability, part of which addresses knock-for-knock contracts. 2 On the literature Knock-for-knock contracts are essentially mutual liability exclusions, and hence related to to the literatures on tort law and on liability exemptions and exclusions. Several articles address whether tort law’s deterrent efect warrants its administrative costs. Part of that literature is empirical and part is theoretical. An example of the empirical literature is from Sugarman. As for theoretical literature, one contribution is that of Polinsky and Shavell who analyze when the deterrence benefts of product liability claims outweigh their dispute costs. They fnd that for certain categories of goods that are produced for mass markets and that are monitored for their quality by consumer organizations and others, the case for liability is at best tenuous. Dispute costs are high, and the efect of liability on deterrence appears small in a market where quality is constantly monitored and where reputation works as a strong incentive. When liability exclusions are efcient is analyzed, e.g., in Polinsky’s textbook. A major point is that when the prospect of being found liable diminishes the risk of an accident, dispute costs are also hereby diminished, as the dispute costs are incurred only when there is an accident. However, the law and economics piece most related to this chapter is from Parchomovsky and Stavang in Chapter 3 of this book. They maintain the knock-for-knock system is likely to diminish precautions, and while this may be an acceptable consequence to the parties involved in the knock-for-knock system, since it is outweighed by their administrative savings, such may not be the case for third parties who may not be fully compensated in case of an accident. While third parties (that are not parties to the knock-for-knock contracts) can claim compensation for economic loss and for physical injury, a higher likelihood of accidents will harm them to the extent that some of their losses are not covered (or not documentable) under tort law. On this basis, Parchomovsky and Stavang advocate a skeptical attitude towards knockfor-knock clauses, especially in countries where public regulation cannot be trusted to maintain deterrence. In the legal literature, Bull and also Wilhelmsen explain and discuss the advantages and disadvantages of the knock-for-knock system. Bull is the standard legal reference, while Wilhelmsen includes a broader economic perspective by taking account of the role of deterrence. 88

on knock-for-knock clauses and their optimal regulation Lando and Mortensen discuss when knock-for-knock clauses will be enforced by Danish courts, and when they should be from an efciency perspective. The present chapter goes further by providing a simple model for when knock-for-knock clauses are used and by considering knock-for-knock clauses in a broader perspective of optimal regulation. Finally, Mielcarek discusses the advantages and disadvantages of gross negligence and willful misconduct liability exclusions and stresses the role of dispute and insurance costs. Mielcarek also makes the claim that liability exclusions, by excluding disputes over claims, are likely to enhance a spirit of cooperation among the parties. We now turn to our frst issue, which considers the basic question that when parties agree to knock-for-knock contracts, when are they useful? 3 When are knock-for-knock clauses useful? Although exclusions of liability occur in contracts of diverse kinds and in many diferent parts of the economy, we observe the knock-for-knock contracts predominantly but not exclusively2 in oil and gas extraction or in the construction of wind turbines. Moreover, as a stylized fact, we observe knock-for-knock arrangements occur more often when these activities are carried out ofshore than onshore. To explain these patterns, the following factors are often evoked: • The liability system involves double insurance, both frst-party insurance and third-party (liability) insurance. The knock-for-knock system dispenses with double coverage.3 • Liability involves administrative cost; conficts are expensive in terms of time and money. They may also sour relationships. • Public regulation is intense in the oil and gas industry and can ensure that the parties exercise efort to lower risks of accidents. • Liability insurance contracts are costly to draft, especially for small contractors. • The parties’ concern for reputation maintains their incentive for precaution. However, these explanations apply to many sectors of the economy. The liability system usually involves the risk of double insurance and always includes administrative costs. Public regulation is widely present in sectors that do not rely on knockfor-knock clauses; liability insurance is often costly to draw up, and incentives created through fear of loss of reputation are ubiquitous in the economy. Also, the explanations seem applicable both onshore and ofshore. We need a model of the use of liability exclusions that can account for when exactly they are advantageous.

2 The exact extent to which mutual liability exclusions for gross negligence are employed in the economy I do not know, but it should be noted that similar arrangements can arise as when contractors and owners together draw up an insurance that covers all damage and losses that may occur at a worksite, as in a Contractors’ All Risks (CAR) Insurance. 3 Note that this argument can be criticized: due to competition between the insurers, one would expect insurance premia in the long run to refect the actual payouts made by the insurers to the parties.

89

henrik lando 3.1 A model of the use of knock-for-knock contracts In economic terms, liability rules fnd their rationale in their deterrent efect. If the parties can create frst-party insurance, this is a better way for the parties of allocating risk when there is no deterrent efect, as frst-party insurance dispenses with the costs that arise when a frst-party insurer raises claims against the injurer’s liability insurer. Hence if rational contracting parties agree on a liability rule it must be because they believe in its deterrent efect. Alternatively, the parties could be concerned with fairness; however, if losses are placed primarily on insurance companies, then the role of fairness is reduced. On this basis, one might conclude that the deterrent efect is small in sectors that employ knock-for-knock contracts; however, this is not a satisfactory explanation because it fails to explain why the deterrent efect would be smaller in some sectors than in others, or why it would be smaller ofshore than onshore. Both ofshore and onshore activities are subject to public safety regulation. Based on the above logic, we can conclude that since liability is applied in many sectors of the economy, it must be the case that liability (often) does increase incentives for precaution. Most likely, this is due to the activities of insurers who may require certain precautions, graduate premia over time, or set deductibles that encourage the injurer to be cautious. Let us therefore assume, in a simplifed model of knock-for-knock clauses where only one party acts, that the probability of an accident caused by that party under a negligence rule, p, is greater than the probability when the party cannot be held liable for any kind of negligence (as under an unqualifed knock-for-knock system), r. Let us also assume there are costs of taking greater precautions, such as when employees must undergo training or procedures for precaution must be put in place, and so forth. Denote this cost by c, and let L be the monetary cost of an accident. Furthermore, we assume that if an accident occurs, the probability of a conflict is q and the average cost of resolving a potential conflict over who was at fault is K. Often, part of this cost is borne by the insurance companies on either side, but eventually this cost will be carried by the parties through the insurance premium. Based on this notation, we can derive when liability is better than exclusion from liability if we assume that the parties wish to minimize the sum of their expected costs. Note that rational parties will wish to do so, since they can share the saved costs between themselves when they agree on the price of the contract. If there is a rule of liability that leads the injurer to take precautions, total expected costs for the two parties combined amount to: c + p(L + qK ) If there is a knock-for-knock exclusion from liability, the expected total costs for the parties’ amount to rL. Hence, exclusion of liability is preferable when: c + p(L + qK ) > rL

(1) 90

on knock-for-knock clauses and their optimal regulation This equation brings out the essential factors: how efective are the extra precautions taken under liability (p − r)L compared with their costs c, and compared with the expected confict costs incurred under a liability rule: pqK? This simple calculation casts light on why the knock-for-knock system is more often employed ofshore than onshore. An essential diference between ofshore and onshore activities is their inherent dangerousness. Let us catch this aspect by letting the probabilities be systematically smaller onshore, that is p and r are both equally reduced by some factor ∂. Here ∂ expresses the general operational dangerousness of the operations, and so subtracting ∂ represents onshore activity. If the condition (1) above is less likely to be fulflled when probabilities are reduced by ∂, we have a simple explanation of why ofshore activities more often than onshore activities use exclusions of liability. When probabilities are so reduced, we obtain the condition: c + ( p − ∂)(L + qK) > (r − ∂)L This can be rewritten as: c + p(L + qK) − ∂qK > rL

(2)

This equation is less often fulflled than (1), since ∂qK must be subtracted from the left-hand side. Hence, exclusion is more rarely optimal when the activity is less inherently dangerous, as it is onshore. The logic of this result can be explained as follows. If the activity is dangerous in the sense that accidents occur with a signifcant probability even when the parties take care, and if there is a signifcant risk that there will be litigation even though care was actually taken (i.e., qK is non-trivial), there will be signifcant costs of confict when liability is not excluded. The high costs of confict may render it optimal to dispense with liability. Hence, the inherent dangerousness of the activity, as measured by the probability of accidents when the parties take proper care, may help to explain the pattern of use of knock-for-knock contracts. The model can be interpreted in two ways. We can either interpret the occurrence of accidents despite due care as involving exogenous events, and the dispute between the parties can then be about whether these events or some alleged negligence on the part of the injurer caused the accident. Or we can think of accidents occurring due to lapses, e.g., momentary inattention on the part of management or employees, or perhaps inadvertent miscommunications in large organizations. Such mistakes can occur even when parties have shown a proper precautionary attitude, and when they do occur, they can cause accidents and disputes about whether the lapse occurred and whether it amounted to negligence. Another factor that, beside the inherent dangerousness of the activity, can give rise to the knock-for-knock system is uncertainty about liability. This is the role of qK; the higher this number, the more attractive the knock-for-knock system. One fact that springs to mind here is that when many workers coordinate their acts on a site, doubt can easily arise about who of many potentially culpable actors was (most) to blame for the accident. This can also help explain the greater use of knock-for-knock clauses ofshore as workers are forced into smaller spaces. Both Bull and Wilhelmsen stress this latter factor, the difculty of ascertaining fault, as important in explaining the knock-for-knock system. 91

henrik lando Moreover, the inherent dangerousness and the difculty of apportioning liability can interact to create a rationale for knock-for-knock clauses: When qK is high, the inherent dangerousness of the activity becomes more important, as is clear from equation (2). Then, even if difculties of ascertaining who was negligent are the same ofshore, they can still account for the knock-for-knock system, since the difculties arise more often ofshore due to the inherent dangerousness of ofshore activities. A caveat to the explanation ofered in terms of inherent dangerousness is worth noting. It seems plausible that care may be more important when an activity is inherently dangerous. If so, one might not expect the knock-for-knock system to be used more often ofshore. The explanation based on inherent dangerousness rests on the incentive efect being limited and not much greater when the activity is inherently dangerous. The theory must therefore be reformulated: the knock-for-knock system is likely to arise under inherently dangerous circumstances when attribution of fault is difcult, and when incentives for care are established to a signifcant extent through other instruments than tort liability. These explanations of the knock-for-knock system fnd support in the description of ofshore projects that employ knock-for-knock contracts. For instance, Fjaervoll writes: At an ofshore project, the parties involved in the operations work in an environment where there is high risk of causing damages, because high property values are situated in a limited area where many contractors and sub-contractors work at the same time. Considering the nature and size of ofshore projects, the threat that damages will occur is present at all times and the parties involved are therefore exposed to substantial risks for damage, not only caused due to their own actions but also as a result of actions by others operating at the same project.

It is also consistent with this explanation that the knock-for-knock became known during the World War II when the allied countries, to defend against submarine attacks, moved their ships in convoys in which the ships could not avoid colliding. The countries agreed to each pay for their own damage, regardless of fault. In these convoys there would seem to be a high risk of collision even when all actors were careful, and it seems likely it was difcult to determine whether a ship’s collision was due to a negligent act rather than the infuence of the sea and weather but also whose negligence was involved when both parties acted to prevent a collision. We now address the question of whether courts should respect the exclusion of liability for gross negligence. 4 Should courts impose carve-outs for gross negligence? The following discussion will be based on a set of premises that are worth establishing at the outset. First, we will assume a Scandinavian context and not, for instance, the context of developing countries where public regulation and tort law may be much less developed. Second, we shall take for granted that if there are no externalities involved such that all parties external to the contract are fully compensated for the losses arising as a result of accidents that occur within the knock-for-knock contracts, there is no reason for courts to set aside the parties’ agreements. This premise is based on the idea that if the parties to the knock-for-knock system have agreed to a knock-for-knock contract and no 92

on knock-for-knock clauses and their optimal regulation third party is afected, their choice of contract is Pareto-optimal and all parties then stand to beneft in an ex ante sense from court enforcement. This reasoning takes for granted that the parties do not wish courts to interfere in what may be an incomplete contract. If, for instance, the behavior of a party leading to an accident is reckless, the parties would perhaps have carved out such behavior in their contract if they had had the resources to anticipate it and to make reference to it in the contract. Just like courts generally interpret contracts to exclude meanings that the parties cannot have intended, certain types of gross negligence may perhaps justifably be struck down. However, we shall not consider this aspect but assume that the behavior for which the parties would want courts to apply liability is so extreme and occurs so rarely that we can omit it from analysis. Based on this assumption we can maintain that courts should, if at all, only strike down carefully drafted and mutually agreed knock for knock contracts when third parties are adversely afected. More precisely, we shall assume that courts can be justifed in striking down knockfor-knock clauses only if the clauses mute incentives for precaution and if third parties are afected by the lack of precaution because they are insufciently compensated in the event of an accident. There may be extreme instances of misconduct for which the court should not uphold the liability exclusion, but such instances will not be specifcally addressed here. Third, we shall make the point that while employees are generally compensated under worker’s compensation schemes (which are themselves in some countries essentially knock-for-knock systems in that they do not allow insurers to seek redress), such schemes undercompensate physical injury or death by not taking into account the full loss for the injured person or for his or her relatives and friends.4 Therefore, employees (and their families and friends) hold an interest in frms being incentivized to take greater care. Fourth, as far as personal injury to employees is concerned, the knock-for-knock system limits the incentive for care, although to a limited extent. Most economic consequences of such harm are covered by a worker’s compensation system. Thus, whether parties enter a knock-for-knock contract, most of the economic losses to employees will be covered by frst-party insurance. However, tort law supplements worker’s compensation, since not all categories of losses are covered by the worker’s compensation system, and this supplementary tort law will be afected by the knockfor-knock system and by whether it is qualifed or not. Therefore, a court may strike down the indemnifcation of tort claims relating to physical harm to employees and thereby afect incentives for precaution. Fifth, accidents cause harm to third parties who are far from always fully compensated through tort law. For instance, in large oil-spills fsheries, tourist industries and others may see their livelihood afected. In general, they will not all be eligible for compensation under tort law. Losses may, for instance, be spread thin and each afected victim may not have the incentive to sue for compensation or be able to

4 Non-economic harm is part of the worker’s compensation system, but the compensations are low compared with the utility consequences sufered when there is severe physical harm.

93

henrik lando organize a class action suit. Or the doctrine of pure economic loss can bar claims from people who have sufered neither personal injury nor loss of physical assets. Sixth, we imagine a setting where O examines C’s procedures before allowing C to take acts that may jeopardize O’s valuable assets X. Practitioners often mention that such vetting can be extensive, and that there can therefore be an inducement for C to gain a reputation for maintaining safe procedures. It is worth keeping this reality in mind, as already alluded to above, as one might otherwise get an exaggerated impression of the role of marginal incentives created by liability. Under these premises we now consider a framework for understanding the extent to which carve-outs of knock-for-knock contracts afect incentives and whether this efect warrants court intervention. 4.1 A framework Consider two parties, O, the operator, and C the contractor. In oil extraction, O is the oil company (owning the concession), and C is one of often many contractors hired to dig oil from a platform owned by O. Both parties put assets and employees at risk. The assets of O are denoted X, whereas the assets of C are denoted Y. Typically, X is of much greater value than Y. Third parties and employees can sufer losses that in economic terms sum to E. Part of these, γE, are compensated under tort law (such as the harm to privately owned land) while (1 − γE) is not compensated (0 < γ < 1). We imagine that both O and C make economic decisions that afect the safety of the project and that both parties may lapse. Lapses are considered exogenous mistakes, as when employees act without paying attention or thinking rationally about consequences. In mathematical terms, one can think in terms of actually exercised care z being afected by the degree of precaution and a stochastic lapse σ : z = x − σ, where σ is a stochastic variable. This means that gross negligence may occur when a party did not intend for it to occur, but perhaps did not do enough to encapsulate the efects of lapses. This way of considering accidents has implications for how one analyzes incentive efects, as will become clear. The two parties will be insured by IC and IO, respectively. These insurance companies provide frst-party insurance and, if relevant, liability insurance to their respective party. We assume that the insurer, whether as frst-party or as liability insurer, can afect a party’s negligence through requiring certain precautions or through deductibles or adjustments of premia over time in response to accident history. However, these instruments will be assumed to create less than frst-best incentives for precaution. Furthermore, in part due to the incompleteness of tort compensation, we assume that a public regulator sets fnes for unlawful behavior. Often, grossly negligent acts can be fned under rules of public regulation; the size of such fnes is a policy instrument under the discretion of Parliament and to some extent also under the discretion of courts. Notably, in Danish law,5 neither civil nor criminal fnes can be indemnifed under knock-for-knock contracts.

5 In the United States, civil fnes (e.g., for breach of the Clean Water Act) are multiplied by some factor when the act has been grossly negligent.

94

on knock-for-knock clauses and their optimal regulation One can analyze the regulation of knock-for-knock clauses either from the perspective of a court, which must take other instruments (e.g., public fnes) as given, or from the perspective of the overall optimal regime of regulation. I shall adopt the latter perspective and argue that the rationale for imposing a carve-out is then quite weak. Let us frst consider the arguments for imposing a carve-out, i.e., for not allowing the liability exclusion, and then the arguments against. The main argument in favor of a carve-out springs from the fact that the employees of O and C and third parties sufer losses that are not compensated under tort law. As only γE of their losses is compensated, they are interested in a higher level of care from both O and C. This is an externality that parties to the knock-for-knock clauses do not take into account when choosing between the qualifed and unqualifed knockfor-knock system, and they are therefore likely to choose a system that induces too little precaution. The reason the level of precaution is smaller under the unqualifed knockfor-knock system is that when there is liability for gross negligence, it is expected that the parties will take out liability insurance (in order to lower the risk of bankruptcy) and the liability insurers will take steps to ensure a greater level of care. Thus, three factors will increase the level of care by the parties when they are liable for gross negligence: 1 The insurers will require deductibles (or increase the premium) in the event of gross negligence. The frst-party insurers already set deductibles under the unqualifed knock-for-knock system, but there will be an additional efect of the deductibles on the liability insurance, especially for C. 2 The liability insurer will require specifc safety measures as a condition for coverage; these requirements may be more expansive than those required by the frst-party insurers under the qualifed knock-for-knock system. 3 The parties will anticipate added dispute costs when there will be disagreement after an accident about whether the injurer’s level of care constitutes gross negligence. These costs can be rendered less likely by taking greater care. The argument for carving out gross negligence is that these three factors will beneft third parties and employees to the extent that they are undercompensated under tort law and the worker’s compensation system. Note that these incentive efects will be more important the lower the level of care taken under the qualifed knock-for-knock system. If incentives fall short by a wide margin, a marginal increase will be likely to be more important (according to the principle of decreasing marginal return). The extent to which incentives for precaution fall short under the qualifed system is, however, not clear. On the one hand, there are three reasons for expecting incentives to be insufcient as compared with the frst best under the qualifed system. First, the compensated losses under tort law are not the full losses sufered. Only part of E is compensated. Second, simple negligence does not imply liability. As simple negligence is far more common than gross negligence, this is a factor that can seriously lower the incentive for IC to induce precaution through the liability insurance contract. Third, O and C are only incentivized by the deductible in their liability insurances, not by the full loss. On the other hand, public regulation, including the imposition of fnes for breach of safety regulation, and C’s and O’s concerns for their reputation for safety, are 95

henrik lando likely to increase their levels of care. It is therefore difcult to say whether the marginal increase in precaution due to the three factors mentioned above is of real importance. Against these arguments for carving out gross negligence, two main arguments can be put forward. The frst concerns the added dispute cost. As mentioned in the frst part of this chapter, in sectors where the risk of accidents is large, and where lapses are likely to lead to accidents, there will be a considerable number of instances in which it can be claimed that there was negligence, and in some of these cases, especially when a signifcant accident has occurred, there may be a considerable temptation for the frst-party insurer of the victim to claim that the injurer acted with gross negligence. Hence, dispute costs may well be much larger in case of a carve-out, especially because of hindsight bias: when the accident is severe, simple negligence may appear to be gross. The second concerns the need for C to take out a large liability insurance. The drafting of a large liability insurance is said by practitioners to involve signifcant costs, especially when there is considerable asymmetric information about C’s type. Based on these arguments, it can be difcult to tell whether the benefts of carving out gross negligence outweigh the costs. However, the following argument indicates, I believe, that a carve-out is not part of an optimal regulatory system. Most of the benefts and the costs of a carve-out are taken into consideration by the parties themselves when they decide whether to carve out or to exclude liability for gross negligence. The parties themselves must believe that the incentive efects of liability for gross negligence are smaller than the costs of a carve-out. As mentioned, the reason for imposing a carve-out on the parties is that third parties and employees hold an interest in a higher level of care than that which is optimal from the point of view of the parties, since the third parties and the employees are undercompensated in case of an accident involving personal injury or environmental harm. However, this interest in greater incentives for care can be accomplished by setting fnes for breach of regulations; such breach is typically involved when the parties act with gross negligence. Fines increase deterrence without jeopardizing the savings in dispute costs and drafting costs that unqualifed knock-for-knock clauses accomplish. The point is that fnes serve the role of making up for the fact that not all losses are compensated through tort law. They serve to increase the parties’ incentives where third parties (or employees) are less than fully compensated under tort law or workers’ compensation system. The administrative costs of higher civil fnes are unlikely to be of the same order of magnitude as the extra dispute costs that may arise when parties hope for courts to strike down the knock-for-knock clause. One more argument in favor of respecting the unqualifed knock-for-knock system is that it should not be taken for granted that the legal system is an efcient mechanism for resolving complex disputes, such as those arising in ofshore oil extraction. Courts may not be able to accurately discern a cost-efective manner what happened in an accident involving several actors, and who were ultimately to blame. Instead, the parties may rationally decide to rely on reputational mechanisms in which privately held information that is difcult to convey to a court can play a part. 96

on knock-for-knock clauses and their optimal regulation 5 Conclusion I have argued that knock-for-knock clauses emerge mainly in sectors where the risk for the parties of negligently causing accidents through lapses is high even when the parties take adequate precautions, and where it can be hard to tell who, if any, has acted negligently. Under such circumstances, dispute costs can be much reduced by dispensing with tort law. Such dispensing involves an externality, as employees and third parties are generally not fully compensated for their losses. This externality may call for courts to invalidate the exclusion of liability for gross negligence to make room for liability insurers to increase the incentive for care. However, such court policy will force even small contractors to draw up extensive liability insurance, which is typically costly, and is also likely to induce disputes even when the act of the injurer only amounts to simple negligence. The temptation on the part of the victim’s frstparty insurer to argue that the act was grossly negligent can be great. The incentive efect can therefore be created at lower cost by increasing the fnes levied against an injurer who has violated public safety regulation in a grossly negligent manner. In an incentive perspective, such fnes serve the role of exactly increasing care when third parties are not fully covered under tort law or the workers’ compensation system. It may be remarked that the system which appears optimal is hence similar to that which emerged after the Macondo accident. The Louisiana court set very high fnes that were augmented by the fnding of gross negligence, and the court emphasized that fnes could not be indemnifed as they served a deterrent role.6

6 See, e.g., Saraceni’s and Summer’s description of the Macondo accident and its legal repercussions here: https://www.studocu.com/en-gb/document/university-of-london/contract-law/knock-for-knock/2603964

97

CHAPTER 5

Knock-for-knock indemnity provisions and liability insurance Potentially strange but always complicated bedfellows Jay R. Sever and Lauren E. Burk 1 Introduction Although knock-for-knock (K4K) indemnity provisions have become common (if not necessary or legally required) in many energy and marine insurance contracts, in reality their actual application in insurance claims can be complex, as K4K provisions do not always harmonize with basic general liability policy provisions. As discussed in more detail in Chapter 2, K4K indemnity provisions involve reciprocal responsibilities, wherein each party agrees to be responsible, or to indemnify, the other for damage or injury to its own property or employees, regardless of fault for loss. The rationale for K4K provisions starts with the need to avoid delay and disruption. In theory, these provisions will prevent the need for detailed investigation and reduce the scope (and costs) of any dispute or litigation. They can also beneft the business relationships of the contracting parties by reducing the possibility of scorched-earth contract and tort disputes and, theoretically, encouraging the parties to provide a safer workplace. The concept of K4K indemnity is so fundamental to marine energy contracts that such provisions are required by the P&I insurance clubs for certain type of insured entities to be considered in their insurance “pooling” arrangements. In this context, any deviation from the K4K regime by the entities will necessitate their purchase of additional extensions to their coverage that will typically have lower limits and be more expensive. However, outside of the context of P&I insurance clubs, other insurers may be less experienced with understanding the interplay between K4K provisions and their policy language. Indeed, such provisions are somewhat contrary to basic underwriting concepts contained in most liability insurance. These provisions are intended to make indemnifcation obligations reciprocal for certain losses between two contracting parties. When these indemnity obligations are insured, however, they may require an insurer to indemnify a party who is not an insured and whose liabilities and operations may be wholly foreign to the insurer. While it has become common for insurers to make non-contracting parties “additional insureds” (AI), the indemnity scenario presented under K4K provisions can be very diferent than a traditional AI scenario – not least because the non-insured reciprocal indemnitee may never have direct rights under the insurance policy. Especially when combined with “other insurance” provisions, these K4K provisions can present hopelessly complicated scenarios for the contracting parties 98

DOI: 10.4324/9781003206798-5

knock-for-knock indemnity provisions and liability insurance and their insurers. This short chapter will highlight and address some of those actual and potential complications.1 2 K4K provisions and general liability insurance K4K liability provisions are most often used in contracting environments that involve high risk and potentially large, if not catastrophic, potential liabilities, such as oil feld and ofshore construction and operations. Although the parties to these indemnity contracts are sometimes large, well-leveraged, multinational corporations, they are just as often smaller operations that rely heavily upon liability insurance to protect their corporate assets. Indeed, insurance must play a critical role in the success of these indemnity schemes. Ideally, each party In a K4K indemnity contract should be in a position to warrant that it has obtained applicable insurance to cover its reciprocal indemnifcation obligation. Such insurers presumably would be able to respond to any claim made by its named insured for its specifc reciprocal obligations under the indemnity contract. However, the insurers are neither signatories to the indemnity contracts nor, usually, part of the negotiations that result in the K4K obligations, so there can be no guarantees that the insurance liability policies will conform or even harmonize with the indemnity agreements. In fact, insurance policies can potentially complicate the contractual agreements and lead to liability disputes and coverage litigation that K4K provisions were intended to avoid. Indeed, there is an inherent tension between the fundamental purpose of general liability insurance and the enterprise being undertaken in K4K indemnity schemes. General liability insurance is intended to respond to the unforeseen fortuitous responsibilities of the insureds to other third parties to whom the insureds are legally liable. Reciprocal indemnity agreements upend that arrangement inasmuch as they may require an insured to pay for an event or costs for which the insured is not responsible. The insured may have no involvement at all in the event for which it is responsible under the K4K indemnity arrangement. K4K provisions undermine the basic tenet of liability coverage – that there must be a claim by a third-party presenting potential judgment liability to trigger coverage, because with reciprocal indemnity schemes the claim is from the named insured to cover liability it previously agreed to undertake. Many of the standard general liability form conditions and exclusions confict with the intended goals of K4K clauses. For example, one standard condition is that no insured “will, except at that insured’s own cost, voluntarily . . . assume any obligation . . . without [the carrier’s] consent.”2 While this condition is generally only applied to assumptions of obligations after a loss, it is a textual example of the complications of trying to force independently negotiated and separate contracts to work together. An insured’s agreeing to be responsible for a risk is contrary to a standard form general liability policy’s Conditions to coverage. Similarly, the conditions call for the insured to assist the insurer

1 This chapter does not attempt to address every relevant aspect of this subject. For example, it does not analyze anti-indemnity statutes and their impacts on K4K provisions and liability insurance. Rather, the intent of this chapter is to focus generally on the impacts that K4K contracts have when they are applied to standard liability insurance provisions. 2 ISO Form CG 00 01 04 13, Section IV – Commercial General Liability Conditions (2)(d).

99

jay r. sever and lauren e. burk in enforcing the insurer’s rights against any person or organization “which may be liable to the insured”,3 but K4K provisions necessarily require that the insured not enforce rights against the responsible party. And as a consequence, to be marketable, insurance policies often must include waiver of subrogation endorsements whereby the carrier limits is recovery rights to supports its insured’s agreement not to try to seek reimbursement for claims paid by other insurance companies.4 Additionally, standard form general liability exclusions are intended to avoid many of the risks that insureds undertake in K4K provisions. As a key example, the maritime and energy industries frequently involve the use of watercraft, but general liability policies exclude coverage for bodily injury and property damage resulting from the “ownership, maintenance, use or entrustment to others of any .  .  . watercraft owned or operated by or rented or loaned to any insured.”5 The insured’s indemnity obligations of K4K provisions may prevent application of this standard exclusion, resulting in a carrier providing unintended coverage. It is axiomatic that general liability policies are not meant to respond to injuries to the insured’s employees. K4K provisions also usually include an insured’s accepting liability for injury to its employees, which is a clearly excluded risk on a general liability policy. General liability policies exclude such risks to avoid functioning as and overlapping with workers’ compensation coverage.6 Again, by contracting to take responsibility for injuries to its employees and agreeing to indemnify a potential responsible third party for injuries to the insured’s employees, the policy may ultimately provide coverage for the excluded risk of injury to the insured’s employees. Notwithstanding all of these incongruities, insurers clearly have adapted to K4K schemes. They continue to write policies to insure liabilities assumed in reciprocal indemnity agreements. But they do so, to some degree, in an environment that continues to present challenges and potential disputes arising out of the disconnect related to (1) the fact that the insurance underwriters are not involved in the drafting and negotiating of the underlying indemnity contracts that attempt to govern the parties’ liabilities as well as their insurance obligations, and (2) their policies contain language that is often at odds with the undertakings set forth in the underlying indemnity contracts. 3 Impact of blanket AI endorsements Inasmuch as coverage for K4K indemnity obligations is not fully based upon an evaluation of an insured’s specifc business risks, there are some similarities to blanket AI endorsements. Blanket AI endorsements provide automatic AI status to entities that a named insured is required by contract to name as an AI in the named insured’s policy.7 However, blanket AI provisions are meant only to add a contracted AI to 3 Id. at c.(4). 4 Another way insurers undertake their insureds’ K4K provisions are endorsements stating that the coverage aforded is non-contributory and will not seek contribution from other insurance carriers. 5 ISO Form CG 00 01 04 13, Section I – Coverages, 2. Exclusions g. 6 Akin to note 4, supra, insurers now accept their insureds’ K4K provisions by including an exception to the Employer’s Liability Exclusion for “liability assumed by the insured under an ‘insured contract’.” 7 ISO endorsement CG 20 33 is commonly used form, and identifes the additional insureds as follows: any person or organization for whom you are performing operations when you and such person or organization have agreed in writing in a contract or agreement that such person or organization be added as an

100

knock-for-knock indemnity provisions and liability insurance the liability coverage, and they do not provide the kind of frst-party coverage implied in the K4K system. As discussed in Chapter 2, because contracting parties agree to be responsible for losses to their employees and property regardless of liability for the injuries, the K4K system may allow an insurer8 to review and potentially approve its named insured’s known risks. Blanket AI endorsements have the contrary efect. Insurers have limited, or zero, opportunity to fully evaluate the potential risk associated with AIs which stand to gain beneft from blanket AI endorsements, because there is no limitation on the named insured’s ability to enter into contracts and no limitation on the types of businesses that might obtain AI status. The named insured is free to bind the insurer to any entity with which the named insured agrees to name as an AI with no notice to the carrier and without consideration to how such agreement might expand the risk under the policy. Thus, although there is some symmetry between blanket AI endorsements and K4K provisions, that does not mean that AI issues are any less problematic under K4K scenarios. K4K provisions theoretically foster amicable working relationships between the contracting parties by allocating responsibility for losses before they happen. However, when an insurer and its policy terms are added to the mix, and it challenges the agreement between the parties, the benefts of K4K may be lessened or destroyed. Identifying the correct AI form is critical because, as is discussed below, many AI forms preclude coverage for the specifc risks that insureds generally agree to undertake in a K4K clause. Blanket AI endorsements frequently require that liability be tied to the named insured’s acts or omissions, which conficts with the K4K provision’s predetermined allocation. For example, the commonly used AI Form CG 20 10 04 13 only applies to injuries “caused, in whole or in part, by . . . [the named insured’s] acts or omissions . . . in the performance of [the named insured’s] ongoing operations for the additional insured.” Similarly, another commonly used form, CG 20 37 04 13, states that it only applies for bodily injury and property damage “caused, in whole or in part, by ‘[the named insured’s] work’ at the location designated . . . performed for that additional insured.” Such requirements are completely contrary to K4K clauses, which are written such that the insured is responsible for whatever losses it has agreed to accept liability for in the K4K provision (often its employees and equipment) regardless of the insured’s involvement in the loss. Another aspect of K4K provisions that can be complicated by the specifc wording of the AI form at issue arises when the K4K provision’s AI clause encompasses a contracting party’s group, or afliates, requiring AI status for all of those entities. The afliate entities may be unknown to the insured at the time of contracting but additional insured on your policy. . . . A person’s or organization’s status as an additional insured under this endorsement ends when your operations for that additional insured are completed.

IRMI. “Additional Insured Status: Blanket Additional Insured Endorsements.” Available at www.irmi. com/online/crt/ch011/1l11e000/al11e030.aspx (accessed 23 January 2022). 8 See Christopher L. Evans & F. Lee Butler, “Reciprocal Indemnifcation Agreements in the Oil Industry: The Good, the Bad and the Ugly, 77 Def. Couns. J. 226, 228 (2010) (“each party [to a knock-for-knock contractual provision] can better determine the amount and cost of the insurance they will need for the job by simply knowing how many people will be at the work site that fall under the contracting party’s scope of liability.”)

101

jay r. sever and lauren e. burk can be wide-reaching, and the insurance policy’s AI endorsement may not satisfy that requirement, potentially resulting in the insured’s breach of the underlying contract. For example, ISO Form L805 (05/09), an Additional Insured – Primary and Noncontributory – Automatic Status When Required in Contract or Agreement Endorsement (AI Endorsement) provides, in part, that the “Who Is an Insured” provision is amended “to include as an AI any person or organization when you and such person or organization have agreed in writing in a contract or agreement that such person or organization be added as an AI on your policy” (emphasis added). When an insured contracts with Company X and agrees to name X’s afliates as AIs, such AI endorsements may run afoul of the insured’s agreement because the insured has not contracted with any “afliates” of X and arguably no coverage would be aforded to X’s afliates under these terms. Similarly, the AI Endorsement states that “[s]uch person or organization is an additional insured only with respect to liability for ‘bodily injury’ . . . caused, in whole or in part, by your acts or omissions, or the acts or omissions of those acting on your behalf .  .  . [i]n the performance of your ongoing operations for the additional insured.” If the insured is not performing ongoing operations for X’s afliates, no coverage would be aforded to the entity, and the insured would, again, be in breach of the contract. Another potential confict between the underlying contract and the AI endorsement occurs where gross negligence and willful misconduct are excluded by K4K provisions. Putting aside the question of whether gross negligence or willful misconduct is adequately defned in the K4K agreement, the question may arise whether AI status aforded by the contract is nevertheless provided (even if reciprocal indemnity is not). As suggested, the language of the AI provision may control this issue; e.g., if the AI provision contains limitations as to the status of the AI or follows the scope of the indemnity contract, it may not provide AI coverage where the purported AI is allegedly grossly negligent or accused of willful misconduct. There is, nevertheless, the possibility that an entity could be provided AI status even if there is no reciprocal indemnity due to the alleged gross negligence or willful misconduct of either of the contracting parties. In this situation, the AI provision would ultimately provide more coverage than was accounted for by the K4K provision. 4 “Other insurance” clauses One of the most basic questions that arises in K4K scenarios is the following: which insurance is meant to cover a loss where there are reciprocal indemnity obligations and multiple potentially applicable policies? Other insurance clauses attempt to prioritize and “stack” coverages when multiple policies are implicated. In general, such clauses either attempt to make the policy in question either primary to or excess over any other policy(ies) that provide coverage on the same basis. Whether, and how, such clauses work to prioritize and allocate coverages is highly dependent upon the wording and on the laws applicable to such provisions. Unless the order of coverage is specifcally spelled out in the K4K provision, it is possible, if not likely, that more than one liability insurance policy will potentially respond to the claim. This will require some determination as to whether either policy is primary, or whether both policies apply on some pro rata basis. Moreover, even if 102

knock-for-knock indemnity provisions and liability insurance the K4K agreement specifcally attempted to prioritize the insurance obligations, there is a valid question as to whether such contractual obligation would be binding upon the insurer, particularly if the policies’ “other insurance” provisions do not refect the commercial agreement. Thus, unless the polices are written to harmonize with the reciprocal indemnity scheme, “other insurance” provisions are often likely to confict with K4K provisions, particularly when there is the possibility of gross negligence or intentional conduct or the parties are AIs in each other’s policies. Because there is no bright-line legal rule as to whether “other insurance”’ clauses trump the insurance priority provisions in K4K provisions, there will often be confusion as to the priority of coverage, unless there has been some coordination in the underwriting of the potentially applicable policies. 5 Incorporation of indemnity limitations in insurance policies The Deepwater Horizon group of decisions illustrates the potential interconnection between the specifc reciprocal K4K indemnity and AI requirements often contained in K4K indemnity contracts and the application of the actual language in the policies themselves. BP’s position was that it was owed both AI status and coverage under Transocean’s $750 million of primary and excess insurance policies.9 Transocean’s policies contained blanket AI endorsements, granting AI status to “[a]ny person or entity for whom the ‘Insured’ is obligated by oral or written ‘Insured Contract’ to provide insurance such as aforded by [the] Policy.”10 There was no dispute that the drilling contract was an “Insured Contract.” Transocean’s insurers argued that the scope of coverage was governed by the terms of the agreement between Transocean and BP, specifcally limiting coverage to above-surface pollution. BP’s primary argument was that a court could look only to the four corners of the policies themselves to decide whether they cover a particular loss, and that the policies themselves did not limit Transocean’s coverage to pollution coverage caused either above or below the surface.11 However, the Texas Supreme Court eventually concluded that BP was not entitled to unrestrained coverage because the language contained in the BP-Transocean drilling contract actually limited Transocean’s liabilities for sub-service pollution liabilities. In fact, the drilling contract did not require Transocean to procure insurance to cover BP’s liability below the surface. The Texas Supreme Court essentially incorporated the terms of the drilling contract into the insurance policies themselves, holding that Transocean’s policies’ reference to Insured Contract required looking at the drilling contract itself to determine Transocean’s specifc obligations in order to resolve BP’s coverage status. The Deepwater Horizon decision makes clear that if the wording of an insurance policy allows for the incorporation of another contract to determine AI status and coverage, a court must incorporate the other contract to the extent required by the language of both contracts. 9 In re Deepwater Horizon, 470 S.W.3d 452 (Tex. 2015), opinion after certifed question answered, No. 12–30230, 2015 WL 13918242 (5th Cir. June 9, 2015). 10 Id. 11 Id. at 456.

103

jay r. sever and lauren e. burk Another complicated issue where the underlying contract infuences the insurance contract and could result in controversy (if not litigation) involves the setting of “minimum” insurance limits in a K4K agreement. As is the case with indemnity contracts generally, negotiating parties sometimes will choose to require a minimum limit of insurance to be provided by one or more of contracting party’s indemnity contracts (e.g., Master Service Agreements [MSA]) which contain K4K provisions. In a case involving the Texas Oilfeld Indemnity Act (TOIA), the United States Court of Appeals for the Fifth Circuit disagreed that the word “minimum” limits the amount of insurance actually available but ultimately found that the MSA’s insurance requirements capped coverage under the TOIA.12 The MSA in Cimarex Energy provided a K4K indemnity requiring Cimarex Energy (Cimarex) and CP Well Testing (CP) to indemnify each other against their “group.” The MSA further required Cimarex to secure a minimum of $1 million in primary general liability coverage and $25 million in excess, while CP was required to obtain a minimum of $1 million in primary general liability coverage and $2 million in excess. CP actually carried a total of $11 million in primary and excess coverage, however. After an employee connected to CP was injured, Cimarex settled the claim for $4.5 million and looked to CP for reimbursement. CP refused to indemnify Cimarex for more than the “minimum” $3 million required in the contract, causing Cimarex’s insurer to cover the additional $1.5 million. That Cimarex insurer thereafter sued CP to recover the diference, arguing that the use of the word “minimum” does not limit the amount of insurance that a contracting party may obtain, and that CP had ample limits to cover its reciprocal insurance obligation. Both the district court and the Fifth Circuit initially held that, in using the word “minimum”, the MSA agreement did not set a ceiling for the amount of insurance available for CP’s reciprocal obligations but instead only established a foor.13 However, the Fifth Circuit agreed with the district court that while $3 million was not necessarily the ceiling, CP’s additional $8 million in coverage was not obtained “for the beneft of [Cimarex] as indemnity and therefore not available to Cimarex.” The Fifth Circuit specifcally endorsed the district court’s reference to extrinsic evidence, including language in CP’s excess policies, which limited those policies to paying the minimum limits CP agreed to procure in an indemnity agreement. Thus, neither CP nor its excess insurers above $3 million were required to reimburse Cimarex’s insurers. If nothing else, the Cimarex Energy decision, and the dispute underlying that decision, make clear that the parties’ intent when they use words like “minimum” in reciprocal indemnity contracts may not harmonize with either the expectations of the insurers or the language of their respective policies. It is likely that the Cimarex Energy decision will result in a reassessment of such contract language in future K4K agreements. 6 Waiver of subrogation Many K4K provisions require that the insurers of the reciprocal indemnity obligations waive their rights of subrogation. Generally, outside of the context of K4K provisions, an insurers’ right of subrogation is sacrosanct, and any attempt by an 12 Cimarex Energy Co. v CP Well Testing, L.L.C., 26 F.4th 683 (5th Cir. 2022). 13 Id. at *3.

104

knock-for-knock indemnity provisions and liability insurance insured to waive such rights would be viewed as unenforceable or null and void. This is yet another curiosity of the K4K contracting world. The impact of subrogation waivers can be profound. In A.M.C. Lifeboats Inc. v Apache Corp., the court was asked to harmonize two K4K indemnity contracts with the insurance procurement provisions contained therein.14 Importantly, one of the applicable insurance policies contained a full waiver of subrogation. The court found that when insurers include this provision, they and their insureds “cannot recoup from the additional insured any amount they have paid to settle a risk covered by the policy, even on the theory that the recoupment is based on the additional insured’s risks not covered by the policy.”15 The court further found that, read together, the two operative contracts mandated that A.M.C. Lifeboats’ insurance policy was specifcally made primary to the policy issued to Apache. Thus, the claims made by A.M.C. Lifeboats against Apache were disallowed. 7 Conclusion As demonstrated herein, K4K provisions do not easily harmonize with liability insurance policies. This may well relate to the common lack of communication between (1) the parties negotiating reciprocal indemnity agreements and (2) those underwriting the insurance policies meant to fnancially support liability arising from the operations addressed in those agreements. Despite these challenges, many businesses clearly wish to continue purchasing liability insurance policies to support reciprocal indemnity schemes, and many insurers wish to continue selling and issuing them. The bigger question may well be whether insurers and insureds alike will work together to better address some of the uncertainties and incongruities before they result in coverage litigation.

14 A.M.C. Lifeboats Inc. v Apache Corp., 2008 WL 217177 (E.D. La. 2008). 15 Id. at 5.

105

CHAPTER 6

The efect of choice of law on knock-for-knock clauses Uisdean Vass

1 Introduction Writing in the context of the upstream oil industry, I take the broad “knock-forknock” concept to be composed of three elements which are, beginning with the simplest and moving to the most expansive, (1) the basic knock-for-knock clause itself; (2) the wider system of indemnities and exclusions which, along with the knock-for-knock clause, form the fabric of oil service responsibility allocation; and (3) a supportive general system of law which facilitates the application of such contractual devices. In order to look at the efect on knock-for-knock clauses when choice of law changes from one law to another, one must have a default system to compare other laws against. For reasons stated below, I propose to “adopt” English law as being the default or “mother” system for knock-for-knock clauses.1 After having analysed knock-for-knock clauses and relevant supporting English law immediately below, I will go on to analyse what happens, or may arguably happen, when one attempts to use the knock-for-knock modality under other systems of law, specifcally including US maritime law, Texas law, Louisiana law, and Brazilian law. English law allows sophisticated contracting parties very broad latitude to agree through contract their own rules for the allocation of contractual and tortious liabilities. In particular, knock for knock under English law is very common in international oil and gas and international maritime contracts. English law knock for knock has attained this widespread acceptance in certain high-risk industries because of the respect it attaches to the wishes of sophisticated contracting parties expressed in writing, to the certainty it brings to contractual construction, and to the wealth of its body of relevant precedents and judicial commercial expertise.2 1 As a Scottish solicitor, it is worth noting that while Scotland is a separate British legal jurisdiction, along with England and Wales, and Northern Ireland, the diferences between English contract law and Scottish contract law are regarded as being small in substantive terms. Despite the fact that the majority of UKCS oilfelds lie in what might be loosely described as “ofshore Scotland” English Contract Law is the default choice-of-law for contracts involving the ofshore oil and gas industry in the United Kingdom. 2 As the authors Nadorf and Gomes state: “By far the most popular governing law for Wellsite Contracts between IOCs and Contractors in respect of operations to be conducted in a civil law jurisdiction is that of England and Wales. English law is favoured because it is: (i) generally regarded as user-friendly, (ii) particularly well suited to interpreting the nuances of a contract drafted in English, (iii) fexible, pragmatic and commercially minded, seeking to uphold freedom of contract, and (iv) provides a healthy body of oil and gas case law. In addition, English courts are highly respected for their independence, efciency, predictability and probity”. Nadorf & Gomes, “Look before you leap: are your oil patch liability clauses enforceable? (An analysis under civil law jurisdictions with emphasis on Brazil)”, Journal of World Energy Law and Business, 2021, 14, 49, at 52.

106

DOI: 10.4324/9781003206798-6

the effect of choice of law on knock-for-knock clauses In terms of oil industry imperatives, the moral is to think very carefully about how choice of law may afect one’s crucial knock for knock and broader indemnity and insurance regime. 2 Knock for knock and the full English legal treatment 2.1 The knock-for-knock clause itself In the most basic case, two contracting parties (A and B) promise one another that if damage accrues to one or other party arising from the performance of the contract, Party A will bear the damage to its property and people, and Party B will bear the damage to its property and people. For example, see the following language in a Petrobras (Petrobras Clause) standard form drilling contract:3 The CHARTERER undertakes to indemnify PETROBRAS Group for any and all damages involving i. The CHARTERER GROUP personnel. ii. goods or facilities owned by the CHARTERER Group PETROBRAS undertakes to indemnify the CHARTER Group from any and all damages involving i. PETROBRAS Group personnel. ii. goods or facilities owned by PETROBRAS Group.

If we take the Petrobras Clause at face value (and leave out possible controversies about what the word “indemnify” may actually mean),4 the intent expressed seems to be that the Charterer will bear losses to its “people and property” expressed as belonging to Charterer Group. “Charterer Group” will mean the Charterer company and its directors, ofcers, and employees, along with the Charterer’s subcontractors and their directors, ofcers, and employees. In order to “isolate” Petrobras from damages to the people and property of the Charterer Group, the Charterer must agree to three diferent conditions with Petrobras. Firstly, if the employees of the Charterer Group are damaged by the acts of Petrobras, then the legal recourse of these individuals (who are not parties to the Charterer/Petrobras contract) must be against the party at fault, which in this case will be Petrobras. In order to “protect” Petrobras from losses arising from such third-party suits, the Charterer will have to agree to compensate Petrobras for any losses recovered by the employees of the “Charterer Group”. This kind of contractual clause is properly referred to as an indemnity. Secondly, the Charterer must also agree to indemnify Petrobras from damages payable in suits brought by members of the Charterer Group (other than itself) for property losses caused by Petrobras in the performance of the contract. Thirdly, and perhaps most consequentially, the Charterer must waive its own right to sue Petrobras for its property losses caused by

3 Quoted in Nadorf & Gomes, id. at 65. 4 See Chapter 2, section 2.1 of this book.

107

uisdean vass Petrobras in the performance of the contract. This latter type of clause is properly called a waiver, an exclusion, or a release. 2.2 Simple knock for knock in contractual context At least in the oil industry, simple knock-for-knock clauses rarely appear on a standalone basis. These clauses will generally appear as the principled centrepiece of a wide range of specifc indemnities, exclusions, and sometimes limited liability clauses.5 For example, in the (UK) LOGIC Well Services Edition 2, March 2001 (LOGIC), the indemnities and exclusions are largely contained in Clause 19, and the classic knock-for-knock clauses are Sub-Clauses 19.1 (Contractor in favour of Company) and Sub-Clause 19.2 (Company in favour of Contractor). The key knock-for-knock language is “save, indemnify, defend and hold harmless”, which covers both contract and tort. Except where expressly qualifed, this language in favour of the benefciary (whether Company or Contractor) is held to cover “negligence or breach of duty”.6 Further, LOGIC Clause 19 sub-clauses speak to such issues as pollution and contamination from the reservoir;7 pollution emanating from, inter alia, the equipment of Contractor Group while at the wellsite, and above the rotary table or the vessel bottom, and caused by the negligence or breach of duty of Contractor Group up to an agreed limit;8 loss of Contractor Group property, materials, and equipment downhole;9 repair of Contractor Group’s equipment (excluding downhole equipment) for damage beyond fair wear and tear if such damage is caused by corrosion, erosion, or abrasion;10 loss or damage to the well, blowout, or uncontrolled well condition and damage to reservoir;11 and losses arising from the use of radioactive tools downhole.12 In Clause 21.2 the Company agrees to indemnify and hold harmless the Contractor Group for Company Group’s Consequential Losses,13 and the Contractor 5 A limited liability clause simply limits the extent to which contracting party A is required to compensate contracting party B for damages. Indemnities may be qualifed by limitations of liability. See, e.g., LOGIC Well Services Edition 2, March 2001, Sub-Clause 19.4 (Contractor indemnifes Company up to an agreed cap for pollution caused at the wellsite through its breach of duty or negligence, which pollution occurs above the rotary table or above the bottom of the vessel). In the leading case of Ailsa Craig Fishing Co Ltd v Malvern Fishing Co Ltd [1981] UKHL 12 (1981), Lord Wilberforce stated “Clauses of limitation are not regarded by the courts with the same hostility as clauses of exclusion: this is because they must be related to other contractual terms, in particular to the risks to which the defending party may be exposed, the remuneration which he receives, and possibly also the opportunity of the other part to insure”. Ailsa Craig [1981] UKHL. 6 LOGIC, Sub-Clause 19.11. For reasons discussed in section 1.3, we interpret the use of the word “negligence” in Sub-Clause 19.11 to be wide enough to cover the kind of conduct usually thought of as being “gross negligence” but not the kind of conduct usually thought of as being “wilful”. 7 Id. Sub-Clause 19.3. Company indemnifes and holds harmless unless the damage is caused through wilful misconduct. 8 Id. Sub-Clause 19.4. Contractor pays up to agreed limit. 9 Id. Sub-Clause 19.5. Company pays except if the loss is caused by the negligence or breach of duty of Contractor Group. 10 Id. Sub-Clause 19.6. 11 Id. Sub-Clause 19.9. Company pays except if loss is caused by Contractor wilful misconduct. 12 Id. Sub-Clauses 19.10. Contractor pays up to an agreed limit if the losses were caused by Contractor breach of duty or negligence, but Company indemnifes and holds Contractor harmless for all such losses whose value exceeds the mentioned cap. 13 This is a defned term which covers “indirect losses and/or loss of production, loss of product, loss of use, and loss of revenue, proft or anticipated proft . . . and whether or not such losses were foreseeable at

108

the effect of choice of law on knock-for-knock clauses in turn agrees to indemnify and hold harmless the Company Group for the Contractor Group’s Consequential Losses. LOGIC provides a frst-class model of how a knock-for-knock provision lies at the heart of a complex structure of indemnities, exclusions, and limitations of liability clauses in standard oil and gas service contracting. While LOGIC embodies a signifcant range of mutual indemnities, covering both contract and tort liabilities, we should refect that overall, the main benefciary of the complex indemnity structure of Clauses 19 and 20 is clearly the Contractor. This is because the Contractor is the likely party-in-breach as it is the main contract-performing party, and there are certain oil industry risks that Contractors are unable or unwilling to accept. 2.3 Knock for knock in its English legal context My thesis is that English law establishes a strongly supportive foundation for the type of knock-for-knock/indemnity-exclusion structure described above. Let us briefy examine certain pertinent legal principles. At frst sight, English law, through its contra proferentem rule, tends to interpret contractual clauses against their benefciaries. Therefore, if an indemnity or an exclusion does not explicitly mention that it covers negligent acts of the indemnitee, then courts will presume that negligence is not covered.14 However, if knock-for-knock drafters do mention “negligence” (which they now invariably do), then they are using a word which has a legal defnition under English law. The word “negligence” under English law (assuming no contractual language to the contrary) is held to cover any kind of negligent conduct, as opposed to wilful conduct.15 This would include the type of conduct often thought of as being “gross negligence”. “Gross negligence” and “wilful misconduct” have no defned meaning under English law but are frequently used in contracts, often as “carve-outs” for indemnity or exclusion provisions. These concepts should always be specifcally defned in contract, though if they are not so defned, English courts will attempt to construe them according to the wording and context of the contract.16 Provided that the meaning the time of entering into the CONTRACT”. Id. Sub-Clause 21.1. This careful wording is designed to cover all heads of probable “economic loss” whether such loss is deemed to be “direct” or “indirect” under the celebrated holding of the House of Lords in Hadley v Baxendale (1854) 9 Ex. 341. 14 The classic case on this topic is E. E. Caledonia Ltd v Orbit Valve [1994] 1. W.L.R. 221. The Orbit Valve case involved a situation where an employee of an oil service contractor was killed on the Piper Alpha platform in the 1988 disaster. The employee’s heirs successfully brought suit against the operator, oil company E. E. Caledonia Ltd (Occidental) with the operator admitting negligence which caused the death. The oil company in turn sued Orbit Valve for indemnifcation under its oil service contract. While the contract contained an indemnity by Orbit Valve in favour of E. E. Caledonia this did not explicitly mention the word “negligence”. The court found that the law in that case would presume that “negligence” was not covered. 15 As long ago as 1842, Lord Denman opined in Hinton v Dibbin (1842) 2 QB 253 that “it may be doubted whether between gross negligence and negligence merely, any intelligible distinction exists”. In the later case of Pentecost and Anr. v London District Auditor & Anr [1951] 2 KB 759, 763, Lynskey J opined that “gross negligence is not known to the English common law as far as civil proceedings are concerned.” 16 A good recent analysis of English and Scottish case law construing the meaning of “gross negligence” and “wilful misconduct” is contained in Pickavance & Bowling, “Exclusions from Immunity: Gross Negligence and Wilful Misconduct” (a paper presented to the Society of Construction Law at a meeting in London on 5 September 2017). See www.eversheds-sutherland.com/documents/services/construction/ D207-pickavance-bowling.pdf.

109

uisdean vass of gross negligence and wilful misconduct is sufciently defned in the contract, there is little that English courts will do to restrict corporate parties’ ability to limit or eliminate the agreed contractual allocation of a company’s tortious or contractual responsibility for acts of gross negligence or wilful misconduct.17 This is important because gross negligence and (perhaps less frequently) wilful misconduct are used as both carve-outs but also “carve-ins” in commercial indemnity/exclusion structures.18 English courts interpret the words “indemnify and hold harmless” to include both a traditional indemnity and an exclusion.19 Furthermore, English courts do not attach any signifcant policy diference between a traditional indemnity and a traditional exclusion at least insofar as companies are concerned. This is a most important point because, as we shall see below, jurisdictions such as US maritime law draw a sharp policy distinction between indemnities and exclusions. Moreover, unlike a number of other systems, English judges will not consider the existence (or not) of insurance in assessing the enforceability of indemnity and exclusion clauses.20 17 No party may contractually agree to be defrauded, and exclusion clauses may not serve to eliminate or virtually eliminate a party’s entire set of contractual obligations. Any agreement by an individual to waive his or her rights to damages for injury or death based on negligence is void. See the Unfair Contract Terms Act (UNCTA) 1977, Section 2A. There are also UNCTA limitations on exclusion clauses for companies, but these are much less onerous, and the provisions of UNCTA section 2(2) (subjecting other exclusion clauses to a reasonableness test) do not pose a threat to most of the type of commercial exclusions which might be found in commercial knock-for-knock clauses and indemnities/exclusions. Penalty clauses in contracts are outlawed under English law, but liquidated damages provisions are acceptable, so long as they fairly and reasonably estimate actual losses. 18 “Gross negligence” is quite frequently used as a carve-in in oil service contracts, but “wilful misconduct” is rarely, if ever, used as a “carve-in” in such contracts. As a practical matter, it is virtually impossible to insure against acts of wilful misconduct. However, in the case of so-called commercial oil and gas contracts which are usually between oil companies, it is not uncommon to see “wilful misconduct” as a carvein. This is perhaps the ultimate display of the extent to which English law is willing to recognise and apply the well-written indemnity/exclusion provisions of sophisticated commercial parties. To cite one example, Clause 6.2.4(b) of the [standard form] Oil & Gas UK Joint Operating Agreement (JOA) provides that the Participants are required to indemnify the Operator for their share of any Consequential Loss caused by the Operator including such Consequential Loss as is caused by Wilful Misconduct (capitalised terms used in this sentence have their meaning stated in the JOA). 19 This was the holding of the UK Supreme Court in Farstad Supply AS v Enviroco Limited and others [2010] UKSC 18. See also Chapter 2, section 2.1 f. of this book. 20 Perhaps the leading UK authority on the interface between indemnities and insurance is the case of Caledonia North Sea Limited v British Telecommunications & Ors [2002] UKHL 32. Caledonia concerned the tragic Piper Alpha disaster on 6 July 1988, in which some 165 individuals were killed. The majority of the claims by the victims or their heirs were settled and paid by the operator’s insurers at an early stage before the cause of the disaster was ascertained. The operator carried insurance for the death or injury of contractor personnel, even though insurance for such losses was not required by the operator’s contracts with the contractors. The contracts had “knock-for-knock” type clauses under which the contractors were obligated to indemnify the operator for the death or injury of employees belonging to each individual contractor. In the wake of the claim settlements, the operator’s insurers brought suit against the contractors based on the fact that they were subrogated to the indemnity rights of the operator. By the time of the appeal, only one contractor had failed to settle. The (remaining) contractor argued that since the operator’s insurer had paid the relevant damage claims, the insurer had no standing to sue the contractor, as such would constitute a double recovery. The House of Lords rejected this argument holding that an insurer has a right to be subrogated to the rights of its insured, even if the right in question is an indemnity. The mere fortuitous fact that the operator had insurance for the accident claims did nothing to afect its wholly separate right to indemnify against the contractor. In the words of Lord Bingham of Cornhill: “Thus, the existence of such [operator] insurance, prudent though it no doubt was in business terms, is irrelevant to the mutual [indemnity] obligations of the operator and the contractor; in technical language, it was strictly res inter alios acta”. Caledonia North Sea Limited [2002] UKHL at para 13. In other words, indemnities are one thing and insurance is another.

110

the effect of choice of law on knock-for-knock clauses 3 Compare the United States 3.1 The jurisdictional issue Since my primary interest is the application of knock-for-knock to the ofshore oil and gas industry, it is worthwhile to briefy note (as is much more fully explained in Chapter 12 by Cindy Matherne Muller et al.) that, when operating in the Outer Continental Shelf of the United States (OCS), contracting parties do not have the legal authority to choose their own choice of law to govern oil service contracts. Instead, the governing law is efectively established by the provisions of the Outer Continental Shelf Lands Act (OCSLA).21 That statute provides that federal law will mandatorily apply to “artifcial islands” and other “installations” for the purpose of “exploring for, developing or producing resources therefrom”.22 As explained in Chapter 12, this federal law is in the nature of a surrogate state law, meaning that for exploration of the OCS of the coast of Louisiana, Louisiana law will be the applicable federal law without recourse to state choice-of-law rules. The main exception to this rule is where the mineral operations on the OCS are being conducted by a vessel, and in that case, federal maritime law will apply. For a detailed analysis of when a “vessel” will be held to be involved in oil and gas operations on the OCS, see Chapter 12. In practice, this distinction is very important to our knock-for-knock analysis, as both Louisiana and Texas have so-called anti-indemnity statutes (see discussion below) which would severely prejudice English-style knock-for-knock (if found to be applicable). Conversely, federal maritime law is comparatively friendlier to the English model.23 3.2 Federal maritime law – Deepwater Horizon The case In re Oil Spill by the Oil Rig “Deepwater Horizon”24 addressed the potential extent of the liability of driller Transocean for the damages and death occasioned by the infamous Macondo well blowout incident of 20 April 2010, when an explosion and fre occurred on the mobile drilling unit Deepwater Horizon. BP was the operator that had contracted with Transocean to drill the Macondo well. The incident occurred on the OCS and not in state waters, and all parties concurred that federal maritime law (and not surrogate state law as federal law) applied, presumably because the maritime status of the drilling operation was incontestable. While the decision covered a range of matters, we will focus on one main issue. The Drilling Contract at Article 24.2 provided that BP would “protect, release, defend, indemnify and hold harmless” Transocean from pollution or contamination arising below the surface of the water, whether or not the pollution or contamination was 21 Littoral states in the United States have been accorded varied extents of “state waters” of the US coasts. In the case of Louisiana this has been fxed as three miles, but in Texas state waters extend to three marine leagues, which approximates to 12 miles. All waters beyond state waters to the edge of the legal continental shelf area are federal in nature. 22 43 U.S.C. § 1333(a)(1). 23 It might be refected that mandatory choice-of-law rules applicable in the OCS contrast with the broad (but not totally unfettered) freedom enjoyed by contracting parties in the United Kingdom to select their contractual choice of law. See discussion in Chitty, ON CONTRACTS, Paragraphs 33–001 to 33–010. 24 841 F. Supp. 2d 988 (E.D. La. 2012).

111

uisdean vass “caused in whole or in part by the negligence or the fault of ” Transocean.25 Furthermore, the Drilling Contract went on to clarify at Article 25.1 that the words “protect, release defend, indemnify and hold harmless” meant, inter alia, that the indemnity extended to “gross [negligence] or any other theory of legal liability”.26 BP fled for summary judgment arguing that while it was clearly obligated to indemnify Transocean for Transocean’s acts of ordinary negligence which may have caused the fatal pollution and contamination, it was not liable to indemnify Transocean for other acts of fault in connection with the pollution and contamination, including gross negligence and acts giving rise to an award of punitive damages. BP argued that under federal maritime law an indemnity clause which covers gross negligence is invalid as being contra bonos mores. The Court noted that Drilling Contract Articles 24.2 and 25.1 were broad enough to cover both indemnities and releases, but the issue here was indemnifcation. Judge Barbier, who carefully analysed earlier case law, found that there was no controlling case in the Fifth Circuit which examined the issue of whether the inclusion of gross negligence would serve to invalidate an indemnity, as opposed to a release. Interestingly, the Court was crystal clear that under federal maritime law, a contractual release for acts of gross negligence would certainly be against public policy.27 The Court noted a signifcant policy diference between the circumstances of (true) indemnity and (true) release. In the case of an indemnity, the principal issue is who ultimately pays for the damages incurred by a genuinely injured (and compensated) third party. In the case of a release, a genuinely innocent party loses all hope of compensation, and a “guilty” party walks away free from liability. Addressing the indemnity issue as one of frst instance, the Court found that in these circumstances of true indemnity, BP’s contractual obligation to indemnify Transocean for its acts of gross negligence was not contra bonos mores. The Court found that it had to weigh two competing considerations: freedom of contract and “a reluctance to encourage grossly negligent behaviour”.28 The Court ruled that courts should be loath to restrict the agreements of competent parties. It noted that the Drilling Contract also allocated liabilities to Transocean (e.g., for pollution arising above the water level). The system of reciprocal indemnities in the Drilling Contract should arguably have served to deter Transocean from grossly negligent acts. And Transocean and BP were highly sophisticated parties of relatively similar economic strength.29 The Court took a diferent view on whether any indemnity in favour of Transocean for punitive damages payable to third parties would be legally valid. Citing Daughdrill v Ocean Drilling & Exploration,30 Judge Barbier explained that the purpose of punitive damages is to punish and deter wrongdoers. Given that background, any indemnifcation for punitive damages would be against public policy.31

25 Id. 26 Id. 27 The Court cited the Fifth Circuit case of Houston Exploration Co. v Halliburton Energy Services Inc., 269 F.3d 528 (5th Cir. 2001) for this principle. 28 841 F. Supp. 2d 1000. 29 Id. 30 665 F. Supp. 477 (E.D. La. 1987). 31 841 F. Supp. 2d 1003.

112

the effect of choice of law on knock-for-knock clauses 3.3 The anti-indemnity statutes 3.3.1 General observation As detailed in Chapter 12, there exists a societal concern in the United States, refected in varied state legislation, that indemnity provisions in commercial contracts, whether these be in areas such as construction or oil and gas, can be used to force contractors to indemnify “project owners” (whether these be builders, oil companies, or others) against the negligence of the “owner” parties. As Muller, Amy, and David note: “Most US states have anti-indemnity statutes, so when contracting for work to be performed in a particular state, great care should be taken to understand the parameters of each state’s particular provisions”.32 Since this chapter is primarily concerned about knock-for-knock clauses in the context of the oil and gas business, we will briefy look at the Louisiana Oilfeld Indemnity Act (LOIA)33 and the Texas Oilfeld Anti-Indemnity Act (TOAIA).34 Such “anti-indemnity” policy considerations are entirely absent under English law. 3.3.2 Louisiana Oilfeld Indemnity Act and other state provisions LOIA was enacted in 1981. According to the Fifth Circuit in Am. Home Assurance Co. v Chevron, Inc, the “purpose of the legislature, and thus the policy interest of the state, is to protect certain contractors, namely those in oilfelds, from being forced through indemnity provisions to bear the risk of their principal’s negligence”.35 Simply put, LOIA provides that any contractual provision “pertaining to a well for oil, gas or water”36 which purports to require a contracting party to indemnify a counterparty for death or personal injury arising from that counterparty’s negligence is invalid. As indicated in Chapter 12, there is considerable case law on what constitutes “pertaining to a well”, and there are certain exemptions to the scope of the statute. For example, the provision does not apply to bodily injury or death arising from radioactivity or operations to control a wild well.37 The statute does not apply to property damage. LOIA seems to apply as much against relevant indemnities favouring contractors as it does indemnities favouring oil companies. Since LOGIC style knock-for-knock clauses and related indemnities/exclusions arguably principally favour contractors, it is difcult to see the rationale for LOIA-style indemnity invalidation, at least in the ofshore context. The same might be said regarding the TOAIA on indemnity invalidation.38 Additionally, Louisiana Civil Code Article 2004 provides that “Any clause is null that, in advance, excludes or limits the liability of one party for intentional or gross fault that causes damage to the other party”. This language would appear to rule out

32 See Chapter 12 of this book. 33 La. R. S. § 9:2780. 34 Tex. Civ. Prac. & Rem. Code § 127.003. 35 400 F.3d 265, 269 (5th Cir. 2005) (cited in Donaho, “Texas Oilfeld Indemnity Handbook” (Donaho) (Baker Hostetler) (2019). 36 La. R. S. § 9:2780.B. 37 Id. § 9:2780.F. 38 There are precedents in other states for the invalidation of indemnities which are granted by the contractor but not those favouring the contractor. An example is the New York Anti-Indemnity Law which is general, and not oil and gas specifc, in nature. See New York Gen. Oblig. Law § 5–322.1.

113

uisdean vass any contractual release for acts of gross negligence or wilful misconduct. Donaho’s view is that it is unclear whether an indemnity for gross negligence is valid under Louisiana law.39 3.3.3 Texas Oilfeld Anti-Indemnity Act and other state provisions TOAIA was enacted in 1973 and amended in 1989 and 1999. Its purposes are similar to that of LOIA, and its wording is also similar to some extent. In the case of TOAIA, indemnities in an “agreement pertaining to a well for oil, gas, or water or to a mine for a mineral” covering the negligence of a contracting party are void if this negligence causes “personal injury or death”, “property injury”, or any loss which “arises from personal injury, death or property injury”.40 Therefore, unlike LOIA, TOAIA extends also to indemnities covering property damage. TOAIA lists 15 distinct activities which fall under the “well” or “mine” defnition.41 TOAIA also lists injuries and contracts to which it is not applicable, and these include losses from radioactivity,42 pollution43 and reservoir damage,44 and under contracts such as joint operating agreements.45 TOAIA has a legislative “Insurance Coverage Exception”, also referred to as a “Safe Harbor Provision”. What this means is that TOAIA will not invalidate otherwise defcient indemnity provisions if the indemnitor provides insurance for the indemnity obligation to the indemnitee. There are diferent rules for “mutual indemnity obligations” and “unilateral indemnity obligations”.46 Summarily, TOAIA, like LOIA, seems to apply as much against indemnities favouring contractors as it does indemnities favouring oil companies. Also of note, Donaho opines that it is not clear whether an indemnity covering gross negligence is against public policy in Texas or not.47 4 Compare Brazil 4.1 General observations Brazil, a classic “full” civil law jurisdiction, takes its place among the world’s great ofshore oil producers. Importantly, up until the enactment of the Hydrocarbons Law in 1997, the national oil company Petrobras SA held a monopoly over the Brazilian upstream oil and gas industry.48 Despite the creation of a new regime of concessions, Petrobras retained title to its signifcant producing acreage and it remains, far and away, the main player and largest oil producer in Brazil. The Petrobras contracting system is of huge importance for both Petrobras and its supplier community. 39 Donaho, supra note 35, at 46. 40 Tex. Civ. Prac. & Rem. Code Ann § 127.003 (West 2016). 41 Id. § 127.001(4). 42 Id. § 127.004(1). 43 Id. § 127.004(2). 44 Id. § 127.004(3). 45 Id. § 127.002(5). 46 See discussion of the “Safe Harbor” provisions in Donaho, supra note 32, at 8–9. 47 Donaho, supra note 35, at 46. 48 Law No. 9.478 of 6 August 1997. See generally, the wider discussion of knock-for-knock clauses under Brazilian law in Chapter 11 of this book.

114

the effect of choice of law on knock-for-knock clauses As alluded above, Petrobras, in addition to many private operators, is now moving to an indemnity system based on knock-for-knock clauses. Petrobras invariably demands that its oil service and construction contracts are subject to Brazilian law and Brazilian adjudication, though private oil companies do frequently use English choice-of-law clauses. At the time of writing, there is considerable interest in the Brazilian and wider oil industry,49 and among Brazilian legal practitioners, regarding how the knock-for-knock modality functions under Brazilian law.50 4.2 Knock-for-knock and Brazilian law There is little specifc case law on knock-for-knock clauses as such in Brazil. The general rule of the Brazilian Civil Code is that parties must respond for the damage which they have caused.51 However, there is also the competing precept of freedom of contract which can, in principle, allow contracting parties to allocate between themselves responsibility for losses in ways at variance from the provisions of the Civil Code. There is authority to the efect that limitation of liability clauses are valid if they are (1) reasonably related to the prospective damage and (2) they have not resulted from the unequal negotiating power of one contracted party.52 Nadorf and Gomes note that the frst principle (reasonable relation to prospective damage) could sit awkwardly with knock-for-knock clauses since indemnities and exclusions under such clauses must, by their nature, cover unforeseeable future damages. Furthermore, most commentators believe that Brazilian courts would not be inclined to treat classical indemnities and exclusions much diferently, especially where the parties are sophisticated commercial players. The main Portuguese verb used in oil and gas indemnities is indenizar, which most commentators see as covering both classical indemnity and exclusion senses. Last, most Brazilian authorities seem to take the view that contractual parties have the right in commercial contracts to allocate contractual responsibility amongst themselves providing that this allocation does not ofend the rules of good public order.53 Most commentators are clear that indemnities covering wilful misconduct (dolo) would be invalid as violating good order. Arlota grounds this view in the fact that the Brazilian Civil Code at Article 432 imposes a duty of good faith on contracting parties, and wilful misconduct would always be a violation of good faith.54 Accordingly to Arlota, the better, or majoritarian, view is that indemnities covering gross negligence (culpa grave) are also inadmissible. However,

49 This author had extended discussions regarding the issue with a major oil service company based in Aberdeen in November 2021. 50 This author gave a presentation to a Brazilian legal and industry audience on “Knock-for-Knock – A Liberal English Journey” on 1 November 2021, organised by Lefosse Advogados. 51 Brazilian Civil Code Articles 186, 187. 52 Special Appeal No. 1.076.465–SP (2008/0160567–4), published on 21 November 2018. This case is cited and discussed by Nadorf & Gomes, supra note 2, at 63. 53 See Alexandre Sales Cabral Arlota (Arlota), “A Cláusula Knock-for-Knock” (The Knock-for-Knock Clause), 2019 (published in Brazil solely in Portuguese), p. 116. 54 Id. at 165–170. It might be noted that under English law, there is not, as a general proposition, a duty to act in good faith.

115

uisdean vass Arlota accepts that a minority view exists that argues for the admissibility of indemnities for gross negligence.55 5 Conclusion Knock-for-knock clauses and the kinds of quite radical responsibility shifting through indemnities and exclusions common in oil service contracts are likely to receive their widest, most consistent, and strongest support under English law. Providing the activity intended to be indemnifed/excluded is express in the contract (e.g., negligent acts), English law has not, like Louisiana and Texas, any great concern about either indemnities or exclusions for negligent or grossly negligent conduct. Given that an efective knock-for-knock clause between corporations requires both indemnities and exclusions, it is important that English law, at least at the corporate level, makes no great policy distinction between indemnities and exclusions. The opinion of Judge Barbier in the Deepwater Horizon case discussed above highlights how diferent the situation is in that respect under US maritime law. And as we saw, under Louisiana law there can be no contractual exclusion for gross negligence. Under English law, the validity or otherwise of indemnities is not impacted by the presence or otherwise of insurance, which is not the case, say, in Texas. It is doubtful whether indemnities for acts even of gross negligence (or far more, wilful misconduct) are valid in Brazil. And in the other jurisdictions which we reviewed, courts may look beyond what the parties have written and enquire into considerations of individual fairness, bargaining strength, and so forth. The situation is much diferent in England and Wales. Bereft of considerations of legal complexity, doubtful clarity, and moral agonising, oil service contracting under English law is above all simple, certain, and not very sentimental. It might be objected that the Gladstonian market-orientation of English commercial law might serve to encourage irresponsible behaviour by parties covered by broad and morally dubious indemnities. But this does not seem to be the case. The safety record of the UK oil industry, for example, is very highly rated, and no rational contracting party is likely to be encouraged to indulge in risky behaviour by the English knock-for-knock regime as no-one wants the stain of personal casualties, property losses, and pollution.

55 Id. at 171–176.

116

PART II CHAPTER 7

Indemnity clauses in fabrication and construction contracts in Norway

DOI: 10.4324/9781003206798-7

117

CHAPTER 7A

Limiting and channeling liability under ofshore construction contracts in Norway Knut Kaasen

1 The problem The purpose of this chapter is to investigate means and boundaries for regulating two types of liabilities arising from contractual relationships: liability for default and liability for physical damage caused in connection with the work under the contract. The basis for these analyses lies within agreed Norwegian standard conditions of contract for major ofshore construction projects.1 Such projects involve lengthy contract periods, huge costs and complex deliveries due to the extensive research and development often required. Therefore, they normally carry the risk that the contractor is not able to fulfll all of his contractual obligations, thus exposing him to broad liability for contractual breach – potentially exceeding his fnancial capabilities. As such exposure is not considered to be efcient or rational, the parties agree on some kind of limitation to this exposure. This is illustrated by ofshore construction contracts.2 Inherent in major construction projects is also the risk that any work related to the contract may cause damage to life, health or property of one of the contractual parties or a third party. Damage to third parties may result in one of the contractual parties being held liable, implying that a loss is brought into the contractual relationship in the same way as if one party causes direct physical damage to the other. Without provisions to the contrary, any physical damage sufered by a third party or a contractual party would incur liability on behalf of the tortfeasor, potentially 1 Norwegian Total Contract 2015 (NTK 15), Norwegian Fabrication Contract 2015 (NF 15) and more (see note 6 below). NTK 15 (and its “siblings”) may be downloaded at no charge from www.norskindustri.no/ contentassets/69b36f82c6f341a68c1aa7691e4f5e06/versjoner-lagt-inn-des.-2017/norsk-totalkontrakt-2015.pdf (Norwegian and English parallel texts). The sponsors of the NF and NTK standards have changed over the years since the frst edition in 1987, due to organizational changes. The current 2015 editions are agreed between Norsk Industri (The federation of Norwegian Industries) and Norsk Olje og Gass (The Norwegian Oil and Gas Association), representing Norwegian contracting companies and oil and gas companies operating on the Norwegian Continental Shelf respectively. The two organizations “recommend that the standard contract is applied when contracting for the supply of large components for the production of production of petroleum reserves on the Norwegian continental shelf if the contractor will be responsible for engineering, procurement, construction and potentially also installation (EPC (I))” (preface to the standard, a similar recommendation is included in the preface to the other 2015 standards). A brief general presentation of the 2015 standards is given in Roggenkamp et al. (ed.), Energy Law in Europe, third ed. (Oxford 2016) 11.87–11.112. Even contracts that generally are not based on these standards will normally contain clauses to the same efect as those discussed here. 2 Clauses limiting liability for breach in contract (and for damages in tort, see below) are of course to be found in several other types of manufacturing contracts (shipbuilding, on-shore construction, data programming, etc.) both in Norway and internationally. The reason for concentrating on ofshore construction contracts is provided below.

118

DOI: 10.4324/9781003206798-8

limiting and channeling liability in norway exposing him to losses exceeding his fnancial capabilities. According to ordinary tort law criteria for liability, it would also be necessary to establish whether the damage was actually caused by the party, whether he caused it by an act of negligence, whether the party sufering the damage should have avoided (some of) the loss, and so forth. As an alternative, positive contract regulation may simplify the distribution between the contractual parties of both the loss caused by damage sufered by one of them (internal claims) and loss arising from third-party claims being honored by a contractual party (external claims). There is a long tradition for including regulations to the efect that such losses are distributed exhaustively and predictably according to clear criteria. Such arrangements, at the same time, facilitate rational insurance coverage: risks are identifable and expressly channeled to one of the parties, allowing for rational insurance coverage of the risk. By adding provisions to the efect that the other party shall beneft from such insurance (status of co-insured and waiver of subrogation), the cost and hassle caused by double insurance can also be avoided. Hence these contracts combine the channeling of liability for damages with the establishing of an efective insurance coverage. In combination, these provisions constitute the knock-for-knock system. These major construction contracts therefore illustrate systems for limiting the parties’ exposure to two types of contractual risk: the liability for breach is to a large extent limited, and so is liability for damage caused by contract work, but based on a diferent mechanism. While the former is a universal, well-known contractual technique, the latter – the knock-for-knock regime – is not as widely applied or known in its combination of liability, indemnity and insurance provisions. In either respect, contract provisions limiting or excluding liability may be contested: should the agreed risk and liability systems be fully afrmed, or should they be censored based on general rules of contract law – the test of reasonableness?3 The issue of censoring clauses on limitation of liability for breach and clauses disclaiming liability (and indemnifying the other party) in tort are often discussed simultaneously. As we will see, ofshore construction contracts require a more nuanced approach. From a practical point of view, the efect of provisions limiting liability in diferent types of situations are quite similar: the party at fault will not be held liable for any loss actually arising from a situation that he would typically be accountable for based on general rules of law. But in legal terms, liability for breach is fundamentally diferent from liability in tort. The diferences relate to the basis and conditions for liability, how the loss is calculated, the method of limiting the liability, and the considerations potentially resulting in “the limitation being limited”. One would therefore expect these two areas of liability to be kept apart in most contexts. However, despite the diferences, taking a combined view of the liability regimes for breach and damage ofers an opportunity to discuss two general aspects of contractual regimes limiting risk related to construction contracts. First, they illustrate various techniques for limiting diferent kinds of risk exposure, and second, in doing so, they ofer an opportunity to investigate the general reasoning behind setting aside clauses that limit liability – or giving them their intended efect.

3 The test of reasonableness refers to the criteria of the Norwegian Contracts Act sect. 36. See further section 4.1, below.

119

knut kaasen The agreed standard ofshore construction contracts include all of these elements. This is illustrated when the contract object sufers damage prior to delivery. Because title to the contract object passes to the client as the work progresses, such damage is deemed to hit property belonging to the client. At the same time, there is an inherent risk that contract specifcations will not be fulflled at delivery (i.e., the risk of breach). Normally, the loss caused by the damage is covered by the project insurance, which the client is required to establish. The situation thus ofers an interesting example of the link between the two liability situations, including the role of contractually organized insurance coverage. In the following, the means and boundaries of limiting risk for default and damage are considered based on the well-established Norwegian Total Contract 2015 (NTK 15),4 which also has infuenced contract practices and standards far beyond its intended scope of application.5 The obvious starting point is that liability for breach and liability for damages in tort must be handled separately – they are diferent risks that are subject to diferent types of limitation. Therefore, we start by a separate look at each of these types of liability and corresponding limitation. Also, the discussion of whether the clauses can be given full efect may well take diferent routes depending on which type of liability we look at. These diferences may, on the other hand, ofer contrasts that are helpful when discussing the ultimate general issue: to which extent may contractual arrangements defning extent of liability for damage and breach occurring in connection with fulflling the contract be set aside by general rules of contract law that limit the parties’ autonomy to establish such arrangements? First, however, we need to look into the contract provisions on liability for breach (section 2) and physical damage (section 3), both based on the regime of the “NTK family of contracts”.6 2 The contractor’s liability for breach under ofshore construction contracts 2.1 The liability 2.1.1 The need for a balanced regime The provisions on liability for breach must, on the one hand, make it desirable for the contractor to avoid breach. Further, they should hold the company harmless – at least to some extent – from the breach. These aspects are discussed in this section 2.1. On the other hand, the liability should be realistic, providing a workable balance

4 NTK 15 is the (so far) latest in a row of “agreed documents” originating in the Norwegian Fabrication Contract 1987 (NF 87). Since then, there have been fve rounds of revisions, none of them implying major material or editorial changes. The terms have been negotiated and agreed between major oil companies operating on the Norwegian Continental Shelf on the one side and the association of Norwegian mechanical industries on the other. 5 Slightly modifed versions of NTK 15 (or its predecessors dated 2000, 2005 or 2007) are used in several major onshore construction projects, and the standard has put clear marks on agreed contracts dedicated to that area (NS 8405, NS 8407) as well as on shipbuilding contracts (BIMCO’s Newbuildcon) and contracts for data programming projects. 6 In addition to NTK 15 the family comprises the Norwegian Fabrication Contract 2015 (NF 15), the Norwegian Total Contract (Modifcation; in two versions) 2015 (NTK (MOD) 15) and Norwegian Subsea Contract 2005 (NSC 05).

120

limiting and channeling liability in norway between incentive and exposure. Often this is achieved by capping the liability for delay (in the form of liquidated damages) at some percentage of the contract price per day delay and a maximum of, say, 10% of the contract price. However, with very large contract values, this could result in liability exceeding hundreds of millions of Norwegian kroner, which to many contractors would not make any practical sense as a limitation. Consequently, a general cap to liability for delay could be set at a fxed amount (related to the contract price or not), curtailing the contractor’s exposure if the percentage system yields higher limits. Liability for defects could similarly be maximized at a percentage of the contract value in combination with a maximum fxed amount. It can also be limited by excluding certain costs (e.g., related to removing other parts in order to rectify the defective part).7 Some techniques for limiting liability for breach are further presented in section 2.2. 2.1.2 Types of sanctions against breach Contractor’s default gives rise to the traditional consequences: he has to rectify defects and pay liquidated damages upon delay.8 In addition, default has a practical implication: delay (directly or by way of defects having to rectifed) normally entails that payment is withheld or curtailed because payment normally is based on progress of the work. The contractor may also be held liable for the company’s losses caused by the default, except in cases of delay.9 Ultimately, substantial breach may result in termination of the contract (having more drastic consequences than cancellation), but this is utterly unusual.10 The standard does not provide an explicit basis for price reduction in the case of a contractor’s breach. However, a similar result follows from the provisions that apply when a contractor has not fulflled his contractual obligations upon delivery. His claim for consideration is then reduced corresponding to the part of the work which he has not contractually completed, either based on the percentage of original work still remaining or in the form of a “negative VO” deleting the remaining work from the contract. In both cases the economic result is similar to the concept of price reduction because “market 7 The standard contracts of course also contain provisions on the company’s (client’s) liability for default. Default under payment obligations entail liability for interests. Other types of default on company’s side give the contractor the right to request a variation order (VO) on certain conditions, including, inter alia, that the contractor has fulflled his obligation to check company provided items upon receipt and that he has presented the variation order request (VOR) within a prescribed time limit. The practical efect of a VO in this situation is an obligation on company to pay. The company’s liability is not subject to any limitation, except the general provision that neither party can be held liable for the other party’s “indirect losses” (NTK 15 Art. 32.1). However, the company has a discretionary right to cancel the contract at convenience (Art. 17.1), implying that company’s exposure in principle never exceeds the cancellation fee of 4% of the contract price at the time of cancellation (Art. 17.3) plus contractor’s right of consideration for work actually performed at the time of cancellation (Art. 27). This right to cancel also means that it is not an option to require specifc performance (naturaloppfyllelse) from company. As the company’s liability for default is not limited beyond these provisions, this aspect is not discussed in the following. 8 Liability for delay is thus not set at the actual loss sufered by the company due to the delay, but see (c), below. 9 NTK 15 Art. 24.4 and 25.3, para. 3. 10 According to NTK 15 Art. 26.2 termination entails an ex nunc settlement, see (c), below. By contrast, termination (for company’s convenience) takes the form of an “unfnished delivery” and the company shall pay the corresponding part of the contract price plus the lesser of 4% of the contract price and 6% of the remaining part of the contract price as compensation for loss of proft; see Arts. 17.5 and 17.3.

121

knut kaasen value” for this type of performance for most practical purposes is equal to the contract price agreed – there are no other footholds. This is diferent if the remaining work at delivery consists of rectifcation of defects only – as opposed to contract work which has not been performed at all. In such cases, price reduction in the classical sense is not an option. Instead the provisions on the right and obligation to rectify exclusively govern the situation, supplemented by the provisions on liability for certain costs incurred by company.11 If the contract object remains faulty because the contractor does not succeed in rectifying it or is not permitted to try, his liability is positively and negatively defned by these contract provisions,12 no matter how it difers from what would typically follow from the general background law rules on price reduction. 2.1.3 Specifc features Even if the types of consequences of contractor’s breach are traditional, the details are not. Liability for delay is standardized in the form of liquidated damages (LD), which accrue irrespective of whether and to which extent the company has sufered a loss due to the delay. This simplifcation makes the reactions to delay more efective, but at the same time it disregards the tort perspective of ordinary contract law.13 The LD amounts to a fxed amount or percentage of the contract price per day, maximized to a specifed amount or percentage.14 The contractor has an obligation but not a right to rectify defects: the company may leave the rectifcation to another contractor – typically an ofshore “hook up” contractor – even if the contractor himself would be able to carry out the work within a reasonable time. However, in such cases the contractor’s cost exposure is defned by the rules on calculating the work as if it had been variation work under the contract, not by reference to the company’s costs if it had let another contractor carry out the rectifcation.15 Also, under the ofshore standard contracts, the company has the right to take delivery on the agreed delivery date, even if the contractor does not want to deliver because the work has not yet been completed.16 The choice between delay and uncompleted delivery thus rests with the company.

11 NTK 15 Art. 25.5. 12 NTK 15 Art. 25.3. 13 As a general rule, these contracts do not give the company the right to claim damages for delay based on his actual costs. However, there is one modifcation: if liquidated damages have accrued on previous penalty milestones, but the contractor meets the fnal delivery date, the accrued liquidated damages are annulled, except to the extent that the company can show that it has sufered losses by such earlier delays (NTK 15 Art. 24.2). 14 Under Norwegian law there is no basis for setting the agreed LD system aside on the basis that it is deemed to amount to a penalty rather than a standardised compensation; there is no distinction between “penalty” and “liquidated damages” like in some other jurisdictions. However, the efect of the agreed system in the actual situation could in principle fail the test of reasonableness under the Contracts Act § 36, see section 4.1, below. 15 NTK 15 Art. 25.3 para. 2. Rectifcation work will often have to be performed ofshore. Restricted access (transportation, housing and work areas) warrants that company has to remain in charge of setting priorities in a hectic commissioning phase. 16 NTK 15 Art. 24.3. The reasoning behind this equals that of depriving the contractor from a right to rectify defects.

122

limiting and channeling liability in norway Termination due to breach does not imply restitution of the parties’ respective performances. Both parties keep what they have received from the other party under the contract as per the day of termination: the company has title to the contract object as it presently stands and the contractor keeps the consideration that he has received for the work so far.17 The parts of the contract scope which have not been performed, are deleted. Termination thus takes efect ex nunc, not ex tunc as normally is the case outside of construction contracts. 2.1.4 Breach of secondary obligations The standards do not specify remedies for breach of secondary contractual obligations, e.g., the contractor’s duty to keep certain information confdential, ensure opportunity for trade union activity and request company consent prior to subcontracting parts of the work.18 Neither do the standards exclude remedies according to general principles of contract law in such cases. This means that such general principles supplement the standards, implying for example that the contractor can be held liable for company’s losses resulting from the contract object not being free of liens and encumbrances at delivery.19 2.1.5 Obligation to notify – and the role of the variation mechanism Contracts governing major construction projects normally contain several clauses obliging the contractor to notify company of certain facts during the contract period, for example of any defects, discrepancies and inconsistencies discovered in companyprovided specifcations and items. Breach of such duties normally entail that the contractor is held liable for all costs incurred by the company that would have been avoided if the duty to notify had been fulflled. Such breach therefore results in a new payment obligation on the contractor. There is no limitation to this exposure, so the obligation to notify is of no relevance in our context. However, one specifc type of notifcation is relevant when considering the end efect of the contract’s general regulation of liability for breach. By means of the “variation mechanism” the company may (by variation order, art. 14.1) unilaterally change the contract schedule or specifcations, amending the relevant defnition of the contractor’s obligations under the contract and thus redefning what constitutes “breach”. More importantly in our context, the contractor may allege that company has instructed or caused him to perform the work diferently from the contract specifcations or schedule without issuing a variation order, thus at the outset depriving him of the right to corresponding adjustments of the contract price and schedule. If he fails to notify accordingly within a prescribed time limit,20 he loses his right to request a variation and consequently will have to fulfll the new requirements without any additional compensation.

17 NTK 15 Arts. 22.1 and 26.2, respectively. 18 NTK 15 Arts. 34.2, 9.2 and 8.1 respectively. 19 NTK 15 Art. 22.2. 20 Normally 21 calendar days (see NTK 15 Art. 16.1 para. 2), but specifc time limits are applied in certain situations.

123

knut kaasen In these cases, there is no obligation to notify. The loss of rights resulting from a claim not being timely notifed is not a remedy for breach of contractor’s obligations. Lack of notifcation does not introduce obligations that the contractor did not already have21 but deprives him of rights he might otherwise have. However, these provisions on notifcation form an important part of establishing a basis for the efective control of complex and dynamic projects. This purpose is served by putting pressure on the contractor’s wish to contribute to the information fow by submitting notifcations. The pressure is efectively obtained by nullifying claims which have not been presented in due time. 2.2 The limitation of liability If the contractor breaches his contractual obligation, he is subject to remedies as discussed in section 2.1. However, his liability for remedies is limited in several ways, which are briefy presented below. Liability for delay is maximized at a certain percentage of the contract price per day of delay, never cumulatively exceeding a set total percentage or amount (whichever is the lower).22 This kind of arrangement is normally conceived as a way of limiting the liability for delay. However, it can also work the other way: the liquidated damages accrue at the defned level even if the actual loss caused to the company by the delay is lower. The liquidated damages thus constitute a system for compensating a “normal loss”, disregarding the actual loss. The design parameter is less the wish to limit the contractor’s liability than to establish simple rules for assessing loss in a complex setting. The contractor’s obligation to remedy defects is also limited. Under no circumstances does it include costs relating to, inter alia, “dismantling of other objects than the Contract Object to provide access to the Contract Object” or costs relating to certain other activities, e.g., heavy lift operations ofshore.23 Other remedial costs are not limited, except that the contractor’s total liability for breach of contract is limited to a set percentage of the contract price, maximized at a certain amount.24 21 This is true even in case of company’s instructions: admittedly, the contractor has to implement these instructions, but this follows directly from the mere instruction, not from the fact that he has not presented a timely VOR to be compensated for the extra work allegedly implied in the instruction (see Art. 15.1). 22 Until 2015 these percentages were fxed as a part of the agreed standards, at a rate of 0.15% per day and a maximum of 10% of the contract price (NF/NTK 2007 Art. 24.2, paras. 2 and 3). In the 2015 versions, the liquidated damages per day are defned in Appendix B, and thus not made part of the agreed standard. The fxing of the maximum exposure is also the result of individual negotiation, as the percentage and a maximum amount is to be flled in. However, the normal level is discreetly indicated by the fgure of 10% in square brackets, followed by this footnote: “As an example it is inserted the rate that was included in NTK 2007. The rate is determined on a case by case basis when the Contract is applied.” While no ofcial comments have been made to these amendments, it may be assumed that they have to do with the internationalization of the market and the fact that the contractors more frequently are large companies as opposed to relatively minor Norwegian specialized yards previously taking on the obligations. 23 NTK 15 Art. 25.4. 24 NTK 15 Art. 32.2, leaving it for the contracting parties to fll in the fgures, but indicating an appropriate level by putting 25% in square brackets, with a footnote to the same efect as in relation to the open clauses on the level of liquidated damages (see note 22 above). In the predecessor NTK 07, both fgures were a part of the agreed standard. In addition, there was a specifc limitation to the contractor’s liability for costs incurred in remedying defects, set at 15% of the contract price (NF/NTK 07 Art. 25.4 para. 1).

124

limiting and channeling liability in norway These clauses on liability for delay and defects are explicitly the sole remedies open to the company against contractor for these types of breach.25 Finally, overarching these traditional distinctions, the standard contract also contains a diferent type of limitation of liability: each party shall indemnify the other party from indirect losses, including but not limited to “loss of earnings, loss of proft, loss due to pollution and loss of production” (NTK 15 art. 32.1). This applies both to losses in contract and in tort, and therefore includes all indirect losses caused by breach. There are some additional mechanisms resulting in various types of limitations to the contractor’s liability under the contract, but they do not relate to liability for contractual breach. One type is the knock-for-knock regulation: these clauses limit a party’s liability for loss caused to the other party by physical damage but also imply that he must carry the loss resulting from damages to his own property caused by the other party (see section 3). Also, from a practical point of view, the contractor’s right to draw on the company’s insurance coverage may be said to amount to a limitation of his liability.26 As an example, the contractor is obliged – in line with general principles of contract law – to take necessary actions to bring the contract object in line with contractual requirements if it is damaged prior to delivery; in the worst case this may imply having to reconstruct it in case of total loss. However, major parts of the resulting costs can normally be claimed under the company’s “construction all-risk insurance”.27 3 The knock-for-knock system 3.1 The call for regulation 3.1.1 The risks of damage Despite restrictive safety regulations and extensive safety management systems, the work related to ofshore construction contracts obviously entails the risk of injury or damage being caused to personnel and property of the parties, subcontractors or third parties without any contractual link to the parties. The contract object itself may also sufer damage during its construction. Such damage may be caused directly by the work, e.g., scafolding falling down at the contractor’s employee. The link may also be more indirect, e.g., the contract object has a weakness that causes damage to employees of the company after having been in operation for some time. Another distinction can be drawn between damages that are linked to the fact that the parties have a contractual relationship and those that could just as well occur without any such background (damages caused by the contractor’s general activities versus those caused by his performance under the contract). Finally, the damage may result in direct losses (the broken crane has to be replaced), or in indirect losses (loss of proft, loss related to other contractual relationships, etc.). 25 NTK 15 Arts. 24.4 and 25.5. 26 The company has a similar right under the contractor’s insurances. However, this relates to contractual claims against the contractor, ref. NTK 15 Art. 31.2 para. 4. As seen from the company’s side, this does not amount to a limitation of liability, but rather a security for the contractor’s ability to honor his liabilities. 27 NTK 15 Arts. 29.1 and 29.2.

125

knut kaasen Naturally, the contractor is most likely the tortfeasor, as he is carrying out most of the physical activities that are likely to cause damage. However, several activities are normally subcontracted, or even the company may take part in certain work – itself or by hiring other contractors – that may cause damage. It is also conceivable that third parties cause damage or injury to persons or property that are related to the contractual work. 3.1.2 Consequences of no regulation In absence of specifc regulations in the contract, the legal handling of damages related to the contract work would have to be based on the rules of tort law. The tortfeasor will then be held liable for the damage provided that the ordinary conditions are fulflled. In principle, this implies that the party sufering losses is indemnifed from the loss, while the potential tortfeasor has an incentive to avoid causing damage. Handling the situation along these lines of traditional tort law necessitates extensive and individual investigation and evaluation of a number of facts. For example, are there circumstances which may constitute a basis for liability (can the tortfeasor be blamed, or is the damage caused by a kind of activity which may give rise to strict liability?); is there a sufcient causal link between the damage and these circumstances, and is the damage a foreseeable consequence of them? These relevant facts take numerous forms, and the conclusions reached by such investigations and evaluations may be highly discretional, resulting in a lack of predictability. This is a challenge to both parties. To a potential tortfeasor, it will be difcult to foresee to which extent he may be held liable for damages caused by his activities, and the potentially injured party will not know for certain whether he will have a claim against anyone if damage occurs. Such knowledge is a prerequisite for establishing rational, and thereby reasonable, insurance coverage for such risks – including the risk that a liable party will not be able to meet his liabilities. 3.1.3 Potential regulation An alternative to having the assignment of damages hinge on factual and legal discretionary decisions – creating unpredictable legal positions – is to regulate the channeling of the risk. The simplest method is to distinguish between who sufers the loss or damage. The frst class of losses comprises damage sufered by one of the contractual parties themselves, by one contract party causing damage to the other party’s persons or property. Personal injury will normally involve employees. The employers are obviously not in a position to regulate all aspects of such claims, but they may regulate how the claim – once presented by the employee – is to be distributed between the contract parties, not necessarily dependent on who takes the position of employer. On the other hand, property damage may be regulated without regard to such limitations, subject only to possible restrictions on the enforceability of agreements limiting liability (see section 6). Such regulation may therefore identify calculable types of risks (based on, e.g., the type of damage to which kind of property and the resulting amount of loss), paving the way for rational handling of potential liability in terms 126

limiting and channeling liability in norway of insurance coverage. A specifc challenge arises in case of damage to the contract object itself – this is an interesting example of a meeting point between liability in tort and liability in contract.28 Second, the contract could regulate risks of damage to parties who are not contractual parties themselves, but who have a direct or indirect contractual relationship to one of the contract parties. Admittedly, the contract parties do not necessarily have a direct control over this legal relationship (although the main contract may lay down how certain parts of the subcontracts are to be drafted).29 However, similar to the situation where employees sufer damage, the contract parties may nevertheless distribute between themselves any claim which one of them has been held liable for in the frst place, resulting in their respective exposure becoming foreseeable. Such provisions may be supplemented by provisions that would create a contractual obligation for each party to implement liability clauses in their respective subcontracts facilitating the liability regime introduced in the main contract. We will return to the technicalities of such systems in section 3.2. Finally, the contract may distribute the risk of liability for damage caused to third parties not having any direct or indirect contractual relationship to the contract parties. Such third parties may bring a loss into the contractual relationship by holding one of the parties liable based on ordinary tort law. The resulting risk to the parties may be made foreseeable and therefore manageable by prior arrangements in the main contract. 3.1.4 The insurance aspects The need for insurance stems directly from risks of liability and loss. By identifying and defning the risk in the contract, the need for insurance coverage of that risk is also identifed. The parties get a basis for acquiring appropriate coverage for liability, property loss and personal injury. And they can avoid insuring against the same risk – as both a liability issue and, e.g., a property loss under their respective insurances. However, if the adequacy of the insurance coverage was estimated by each contract party unilaterally, there would be a risk that the party would not have sufcient insurance in covering the risk allocated to him by the contract. The contractual channeling of risk would then be inefective – the loss might have to be borne by the other party despite the contract’s liability system. This other party would consequently have a need for insurance covering this risk, undermining the benefts that were to follow from the originally clear distribution of risks. Therefore, there is a need for both obliging the parties to acquire the necessary insurances, and for this coverage to be rational (i.e., not overlapping). These considerations call for the insurance arrangements to be regulated in the contract. There is, however, an additional need: little is gained by predetermining the distribution of risk related to damages if the insurer reintroduces all the problems of tort law by holding the party who caused the damage liable under a claim for recourse.

28 See Mestad p. 267 at note 64, and section 3.2 a), below. 29 See NTK 15 Art. 8.5.

127

knut kaasen If the agreed channeling of risk is to be efective, the resulting insurance coverage must be the fnal word – at least between the parties and their insurers. 3.1.5 A global view of risk, liability and insurance The observations presented in sections 3.1.3 and 3.1.4 indicate that there is a need to evaluate the contract’s provisions on risk, liability, indemnifcation and insurance as a whole. Each element can only be efective if designed in light of the others. The natural starting point is to channel the risk of losses due to damage caused in relation to the contract work as clearly and defnitively as possible to one of the parties. This provides predictability that paves the way for simplifed handling of each incident (because the details of a typical tort action do not have to be investigated) and therefore, for a rational insurance coverage of the risk. This approach implies that traditional tort law concepts of reparation and prevention are signifcantly toned down. The provisions on allocation of risk and liability do not appear to refect them at all. This is the cost resulting from cutting loose from the demand for individual – and thus hardly predictable – evaluation of other aspects of the damage than the basic question of where it hit (plus its link to the contract work and the resulting amount of loss). 3.2 Provisions on liability 3.2.1 Overview The provisions on liability in NTK 15 arts. 29 and 3030 are disconnected from traditional conditions for liability under tort law. Liability for damages arising in connection with the contract work is allocated to the parties disregarding how the damage occurred, whether it was caused by negligence in any form, whether the tortfeasor was an employee of one of the parties or an independent contractor, and (generally) the amount of loss caused by the damage. Rather, the allocation is based on a simply verifable, objective criterion: who has sufered the loss frst-hand? Naturally, the damage must belong to a category included in the contract’s liability regime: the damages and the contract work must be linked factually and time-wise. This condition may, of course, call for difcult analysis. However, these provisions on liability clearly allocate the risks in a less discretionary and more predictable way than would have been the case if ordinary tort law were to govern. These criteria for allocating such losses are more challenging if the loss in the frst place is sufered by someone other than the contractor or the company. The project involves a large number of actors (companies and individuals) having diferent types of direct and indirect contractual relationships – or even no such relationship at all. In a liability regime allocating risks based on whether the damage hits either the contractor or company, damage that in the frst place hits none of them must 30 Hans Jacob Bull, Tredjemannsdekninger i forsikringsforhold (Oslo 1988) p. 333 f. ofers a thorough analysis of the system of liability and insurance provisions of NF 87, the frst agreed standard contract for construction work on the Norwegian Continental Shelf – the “grandfather” of NTK 15. As the clauses of NTK 15 on the relevant points are similar to those of NF 87, Bull’s analysis is still highly relevant and provides basis for several views expressed in the following.

128

limiting and channeling liability in norway in the next round be allocated to one of them in order to be handled efectively by the liability regime of the contract. This brings in the concept of “risk zones”:31 If a certain person, property or party in the contractual hierarchy of the project suffers damage linked to the project, that damage is held to be the risk of either the contractor or the company, thus belonging to said party’s risk zone. This naturally leads to distinguishing between three situations: frst, where the damage hits one of the contract parties; second, where it hits someone else, though contractually linked to a party (including the employees of a party); and third, when damage hits a totally external third party. Based on this distinction, the risk zone of the company and the contractor respectively comprises damages sufered by said party itself or someone it has a contractual relationship to – other than via the contractual relationship to the other primary party. However, since both the company and contractor belong to the contractual hierarchy of the project, there is a need to decide which parts of the hierarchy should be allocated to which zone. We will return to this issue of “nuclear family” or “extended family” in (b) below. Damages caused to third parties are allocated to the two contract parties based on two criteria: when the claim for damages is presented and the amount of loss caused by the damage. All this allocation of risk applies regardless of whether the damage is caused by the other party, possibly even if he has acted with gross negligence. This liability system is often termed the knock-for-knock32 system. As shown, it consist of three main components.33 First, it implies that the parties refrain from claiming damages from each other, even if the normal tort law conditions are met. Second, they waive their right of subrogation against each other in cases where they, according to contract, have covered the loss sufered by a third party, even if there is a basis for subrogation according to ordinary rules. And third, they accept to hold the other party harmless if he has had to cover third-party claims – including claims from his own employees – that according to the contract shall not be borne by him, even if neither he nor the third party has any basis in general law for such claim for being held harmless. The provisions on damage to the contract object (NTK 15 art. 29) constitute a class of their own. In line with general contract law, damage to the contract object prior to delivery is explicitly at the contractor’s risk, as is his unconditional obligation to specifc performance (thus rendering the discussion on the extent of this rule under Norwegian law superfuous). These provisions apply even if the damage is caused by the company under circumstances that would normally entail liability under ordinary tort (or even contract) law. On the other hand, this obligation is closely linked to the company’s insurances. The end efect is that the contractor’s economic exposure by

31 Bull (note 30) p. 347. 32 The historical background for the general idea of this system is said to be an agreement between the United States and its allies during World War II related to ship convoys transporting goods across the North Atlantic, and even the Barents Sea, in midwinter darkness and storms, without carrying lanterns. Damage caused to one Allied ship by another should be borne by the state owning the damaged ship, no matter the circumstances – there were more important issues at hand. This system was, however, of course not as detailed and refned as the one to be found in modern ofshore construction contracts. See further Ørvig, The knock-for-knock agreement, AfS (Arkiv for Sjørett) 3.448 f. 33 Bull (note 30) pp. 346–347.

129

knut kaasen having to repair the contract object in order to make it conform to contract specifcations is normally very limited. 3.2.2 The family zones Since the liability provisions entail that each party carries the risk of damages hitting his part of the contractual hierarchy of the project, the contract must determine which part of the hierarchy shall belong to which party. There are several alternatives in this respect. The company’s risk zone may comprise all actors having a direct or indirect contractual relationship to the company (its own employees; mother, daughter and sister companies; its contractors other than the contractor in question; and all employees of these companies); while the contractor’s risk zone may comprise similar actors having a contractual link to the contractor. As an alternative, the company risk zone may be limited to the company itself; its mother, daughter and sister companies; and all employees of these companies, leaving out contractors and actors contractually linked to them.34 The latter alternative implies that the “third-party zone” ((C) in the quote below) becomes much bigger, in that it includes the whole set of project contractors of any tier, other than the contractor in question, as well as all their employees, in addition to “genuine third parties” having no contractual relationship to company or contractor. Under the frst alternative, the third-party zone is similarly very restricted. Further, no matter the width of the risk zones, the parties are identifed with their zone members only to the extent that the members are involved in the construction project governed by the contract. NTK 15 art. 1.26 introduces the extended company zone.35 In order for the allocation of liability to work as intended, the liability provisions must be supplemented by indemnifcation clauses to take care of situations where the party sufering damage has relied on tort law to hold liable another contractual party, rather than the party who should carry the fnal loss according to the knockfor-knock provisions. If for instance an employee (A) of a subcontractor (B) is injured by an employee of another main contractor (C) than the one who engaged B, the injured employee A may often hold C (who is the employer of the tortfeasor) liable according to ordinary tort law.36 In this way a loss which has hit the family that A is part of has been moved to the family to which C is a member. It is shifted back to the A family by contractually obliging the head of A’s family – the contractor in our contract – to indemnify C (or whoever is held liable) from the loss.

34 Bull (note 30) p. 347 introduces the term “extended family solution”, as opposed to “nuclear family solution” in contracts where the company is not identifed with its other contractors and their contractors and subcontractors and the employees of these companies. 35 In contrast, the UK standard model contract LOGIC (can be obtained against payment from https:// www.logic-oil.com/content/standard-contracts0) is based on a nuclear solution on the company side and an extended solution on the contractor side (Clauses 1.4 and 1.9, respectively). This means that damage caused to another main contractor’s property by the contractor is handled as a third-party damage. This implies that the company is only liable to the extent that the damage “is caused by the negligence or breach of duty . . . of the Company Group” (Clause 19.2 c), also necessitating investigating whether an act of negligence on the part of the company contributed to the damage. 36 See the Norwegian Act relating to compensation in certain circumstances (13 June 1969 No. 26) sect. 2–1.

130

limiting and channeling liability in norway If a contract party is held liable based on ordinary tort law for losses that according to the contract shall be borne by the other family, the head of the other family shall hold him harmless, no matter the basis of liability (negligence, vicarious liability, strict liability, etc.) or the extent of the damage. However, the distribution within the other family is not regulated directly in the main contract. Thus, it is conceivable that the loss stays with the head of the family, simply because he has not arranged any legal basis for transferring it somewhere. Normally, however, the contracts between the family members will contain back-to-back arrangements for such situations. Main contracts based on NTK 15 require both parties to see to it that such arrangements are implemented in their contracts related to the project.37 When this arrangement is in place, the party who frst-hand sufered the damage undertakes to hold the family head harmless from losses related to the damage. When this regime is implemented, there is no point in the injured party holding the tortfeasor liable based on tort law; the “carousel of loss distribution”38 is stopped at its start. 3.2.3 The third-party zone Under NTK 15 the risks for damage inficted on (the very limited group of) third parties are divided between the contract parties based on the magnitude of the loss and the time at which claims are presented. The formal main rule is that the contractor carries this risk, but his liability for loss or damage arising out of each incident is limited to a set amount to be inserted into the general conditions of contract.39 The company shall hold the contractor harmless for liability above this level, no matter the basis for the contractor’s liability. After issue of the acceptance certifcate, the company shall indemnify the contractor group from and against any claims of the kind mentioned in the frst paragraph above, regardless of any form of liability on the part of the contractor group (art. 30.3 last para). 3.3 Insurance 3.3.1 Considerations The provisions on insurance contained in NTK 15 are based on several types of considerations, the fundamental one being that the insurance system should refect the liability regime (see section 3.1(d)). This entails that the insurances should enable each party to stand the fnancial burden following from the contract’s channeling of risk and liability; but at the same time, double insurance coverage should be avoided. Therefore, the contract should state which party should establish which types of insurance coverage. On the other hand, the insurances should not undermine the

37 See Arts. 30.1 last para and 30.2 last para, and – specifcally relating to the contractor’s obligation – Art. 8.5 b). 38 The term is introduced by Bull (note 30) p. 341; see pp. 341–342 on how the concept works. 39 In versions prior to 2015, the amount was fxed at 5 MNOK. Leaving it blank in the standard must be seen in light of the fact that the contractors in these projects frequently are major international construction companies having insurance arrangements in place that allows them to undertake larger risks related to third-party damage.

131

knut kaasen liability regime. To this end the contract should require the parties to include provisions on co-insurance and waiver of subrogation in their insurance policies. A further consideration is the need for rational structuring of the insurance coverage: the parties’ existing insurances should be used, rather than tailoring insurance systems for the individual contract, and the party likely to obtain the most extensive coverage at the lowest price should buy insurance coverage. The ofshore aspect brings in a fnal consideration. Both the Petroleum Regulations40 and the Joint Operating Agreement between the joint holders of a production license impose a duty on the licensee to obtain specifc insurances covering activities under the production license.41 The licensee is also obliged to ensure that the contractors and subcontractors that he engages in his activities hold insurance for their employees to the same extent as the operator insures his own employees.42 The insurance provisions of NTK 15 are partly explained by this background. 3.3.2 Main components The insurance provisions of NTK 15 art. 31 comprise three elements. First, certain specifed types of insurance are to be provided. Second, the other party is to be made co-insured under these policies. Third, the insurer should be barred from pursuing a right of subrogation against the other party, no matter the circumstances of the loss covered by the policy. The core requirement is each party’s obligation to provide certain insurance coverage. Generally, the insurance should cover all liability allocated to the party in question. However, there are modifcations to this. By way of illustration, it would not be of much help to the company that the contractor has the risk for damage to his own equipment if he is not in a fnancial position allowing him to repair or replace the equipment. In this case the company obviously has a selfinterest in having a satisfactory insurance for property damage in place, but the contract does not require the contractor to provide such coverage.43 The contract specifes type and amount of coverage and the period that the insurance should be valid. Obviously, there are diferences between the company and the contractor in these respects, the most important stemming from the fact that the company usually provides Construction All Risks (CAR, ref. (c) below) insurance, which covers the whole development project rather than each individual contract. The contract seeks to exploit the benefts of this by also making the CAR policy the main coverage for the contractor’s fnancial exposure in case of damage to the contract object and

40 The Petroleum Regulations (FOR-1997–06–27–653) sect. 73. 41 The licensee is required to obtain insurance coverage for property damage insurance, wreck removal and personal injury to employees. See further Bull (note 30), at pp. 99–103 (based on the regulations then in force, which on the relevant points are almost equal to the present regulations). He emphasizes (pp. 99–100) that these provisions are the only example of obligatory insurance coverage for damage to property. This fact must be seen considering insurance’s role in the knock-for-knock system. However, it also has some link to the Norwegian petroleum safety regulations, which state that due account shall be taken “of the safety of . . . the fnancial values which the facilities and vessels represent, including also operational availability” (the Norwegian Petroleum Act 1996 sect. 10–1 para. 1). 42 The Regulations sect. 73 para. 2. 43 See NTK 15 Art. 31.2, which admittedly requires all risks, hull and machinery insurance for vessels engaged in the activities, but this is due to the liability coverage under such insurance. See Falkanger, Bull and Brautaset, Scandinavian Maritime Law (4th ed. 2017) pp. 616, 653 f.

132

limiting and channeling liability in norway materials prior to delivery. This is obtained by requiring the CAR policy to name the contractor as co-insured. On the other hand, the contractor is required to ensure that the company has a right of direct action against the insurer under the policies which the contractor has to obtain. Both parties shall ensure that their insurers renounce any right of subrogation against the other party. And fnally, the contract establishes systems for ensuring that required insurance coverage is maintained and for handling incidents covered by insurance. 3.3.3 Insurance arrangements At the core of the insurance system established by the contract we fnd the company’s CAR insurance. The conditions of the policy are normally attached to the contract as an exhibit, albeit often in a preliminary version. The conditions are largely standardized. The CAR policy covers the total project, and because of this, it may appear as overkill in relation to each individual contract. While the whole group of licensees under the license normally provides the CAR jointly, it also happens that individual participants make their own arrangements. Notably, the state will be self-insured for its participating interest in the form of the State’s Direct Financial Interest (SDFI).44 Such arrangements, however, do not afect the fnal distribution of the loss, neither between the participants in the group of licensees (i.e., “the company”) nor in relation to the contractor. The CAR policy provides coverage for both liability and property damage. Depending on the individual conditions (which are generally standardized to a high degree), the property damage coverage includes physical total loss of and damage to the property insured, rescue costs, certain costs unrelated to physical damage, and certain elements of liability. The policy normally covers the contract object itself, components meant for incorporation into the contract object, materials necessary for performing the work without becoming a part of the contract object, and consumables. The coverage extends over a lengthy period of time, including a “maintenance period” following the main period. Further, the CAR insurance covers a wide range of risks; it takes the form of an “all risks” coverage, i.e., property insurance covering loss arising from any fortuitous cause except those that are specifcally excluded. The contractor’s insurances are normally less standardized. They are usually bought in the ordinary market rather than in the “oil market” and are normally general insurances running unrelated to any specifc project. By way of example, the contractor will have occupational injury insurance covering his employees without regard to their involvement in specifc projects, and his subcontractors may have P&I insurance covering the vessel chartered for the project. However, some project-specifc modifcations usually must be made in order to fulfll obligations under the insurance provisions of the contract, e.g., obtaining a right of “direct action” by the company against the contractor’s insurer.

44 For a brief presentation of SDFI, see www.petoro.no/about-petoro/establishment-of-sdf-and-petoro. The Norwegian state-owned company Petoro AS has managed the SDFI portfolio since 2001.

133

knut kaasen 4 Setting aside provisions limiting exposure resulting from default or damage? 4.1 Authority for censoring provisions limiting exposure? All of the provisions mentioned in sections 2 and 3, limiting a party’s exposure when liable for default or physical damage, are clearly worded and agreed. Some of them even state explicitly that they govern “even if the loss or damage is the result of any form of liability, whether strict or by negligence in whatever form” by the party relying on the limitation.45 But even in the absence of such clear wording, there is nothing in the contract indicating that the parties have envisaged that there should be made exemptions to the arrangements in certain situations. These exclusion clauses apply mutually. Hence their enforceability should be evaluated in relation to both parties. However, in practice it is prevailingly the contractor who has the beneft of the clauses limiting his liability and holding him harmless. This is obviously the case in relation to liability for contractual default; but also, physical damage caused in connection with the work is more likely to have been caused by the contractor than by the company. In the following discussion, we will therefore concentrate on the contractor’s right to limit and be held harmless from liability for breach and for physical loss and damage. The starting point is that the parties are at liberty to contract out of the liability regimes contained in the background rules of law applied when a contract is silent. This includes the right to limit the application of such rules. But any form of disclaiming or limiting liability under contract is met by skeptical scrutiny.46 Such arrangements may shake the basic balance of the contractual relationship. It may be conceived as unacceptable, should the arrangement be valid and enforceable in all liability situations, irrespective of the actions of the liable party. Therefore, the traditional view in jurisprudence – referring to, inter alia, pacta sunt servanda – is that agreed disclaimers and limitations may be set aside in certain situations, despite their clear wordings. After the introduction of the Contracts Act sect. 36, it is natural to use this as a base for censoring such provisions. The parties cannot contract out of this provision – it is mandatory background law. The question is therefore whether “it would be unreasonable or confict with generally accepted business practice” to invoke the provisions of NTK 15 that result in limitation or exclusion of the contractor’s exposure in case of loss, damage or default for which he is accountable. In determining this, “account will be taken not only of the contents of the agreement, the positions of the parties and the circumstances prevailing at the time of conclusion of the agreement, but also of subsequent events and circumstances in general.”47

45 See, e.g., frst para. of Arts. 30.1, 30.2 and 30.3. 46 An extensive discussion can be found in Viggo Hagstrøm, Om grensene for ansvarsfraskrivelser, særlig i næringsforhold [On the limits for disclaimers and exculpatory clauses, specifcally in commercial contracts] (TfR 1996 s. 422–518) (hereinafter Hagstrøm, Ansvarsfraskrivelse). 47 The Contracts Act, 31 May 1918, No. 4 sect. 36. Unofcial translation.

134

limiting and channeling liability in norway 4.2 Diferences and similarities between the two types of liability situations 4.2.1 Liability situations The test of reasonableness according to the Contracts Act sect. 36 may depend on whether the issue is the limitation of contractual liability for delay and defects48 or the channeling of liability for losses sufered by the contractual parties or third parties due to damage caused in connection with the contract work. In contrast to the former, the latter situation does not entail limitations to the contractor’s fundamental obligation to fulfll his contractual obligations but concerns the fnal placing of a loss emerging from physical damage caused by the contract work. In between these situations, we have the case of damage to the contract object prior to delivery, i.e., while it is still in the contractor’s custody. The contractor is obliged to deliver in accordance with the contract specifcations – this is at the core of his contractual obligations.49 Failing to do so brings into play the ordinary provisions on defects and delay – including the limitations to the contractor’s exposure in this respect (and hence the question of whether the contractor may rely on these limitations if he willfully refrains from taking necessary actions to remedy the damage). In this respect we are in the frst category mentioned above. However, based on art. 29.2 the contractor may usually claim compensation in addition to the contract price to cover most of the extra costs related to curing the damage. In practice this implies that the risk of damage to the contract object normally does not form part of the contractor’s general performance risk, which is basically covered by the contract price. In this sense, art. 29.2 implies a limitation to the contractor’s unlimited obligation to fulfll his contractual obligations at his own cost. To which extent this right of compensation will be upheld in all situations depends on a somewhat diferent evaluation than the one applied to the question of censoring ordinary exemption clauses related to contractual liability. Because the compensation of costs is based on insurance, the discussion of whether the resulting limitation of the contractor’s exposure can be maintained must be based on the conditions of the insurance policy and the background rules of law on insurance. These questions are discussed in sections 6.2 and 6.3. 4.2.2 Liability regulations When evaluating the enforceability of the provisions limiting the contractor’s exposure under the contract, some diferences between the provisions should be noted. 48 Defned by reference to certain amounts or percentages of the contract price; see section 2.2 above. 49 This general rule of contract law is expressed in NTK 15 Art. 29.1 para. 1, stating that “If loss of or damage to the Contract Object occurs between the start of the Work until the time when the Delivery Protocol has been concluded . . ., Contractor shall carry out necessary measures to ensure that the Work is completed in accordance with the Contract.” There are no explicit modifcations to this strict obligation, as opposed to the rule contained in the Norwegian Sale of Goods Act 1988 sect. 23 (1) (stipulating the buyer’s rights following the seller’s breach of contract): “The buyer may maintain the contract and demand performance. This rule does not apply if there is any impediment which the seller is unable to overcome, or if the performance would entail such considerable inconvenience or cost to the seller as to be substantially disproportionate to the buyer’s interest in the seller’s performance.” The reason that no such reservation is to be found in NTK 15 is undoubtedly the compensation regime stipulated by sect. 29.2 based on the insurance regime established under the contract.

135

knut kaasen First, only the liability clauses explicitly state that they prevail “regardless of any form of liability whether strict or by negligence, in whatever form” on the part of the party to be indemnifed from liability for loss or damage arising in connection with the contract work.50 There is no similar explicit statement in the provisions dealing with liability for breach of contract. However, neither do these provisions contain any modifcations to the clearly implied joint view of the parties: the limitations shall have efect to the extent permitted under the background mandatory rules of law. Second, while the limitations to liability for breach operate to the beneft of the contractor only,51 the provisions on limits to and indemnifcation from liability for loss and damage are in principle mutual: the parties are to indemnify each other from loss and damage inficted by the other party or third parties. However, this reciprocity is not quite realized in practice; there is little doubt that the contractor (and the other parties he is identifed with – “the Contractor Group”) for purely practical reasons is more likely to cause damage to the company’s (and the Company Group’s) personnel and property than the other way around. On the other hand, the contract is often used on a back-to-back basis between the contractor and his subcontractors, and in this relation, the reciprocity is real. Another diference is now, therefore, more signifcant. The reasonableness of provisions on breach of contract may be evaluated separate from most other elements of the contract. Conversely, in discussing the “reasonableness” of provisions limiting exposure in case of physical loss or damage it is necessary to consider the totality of a system comprising elements of liability, indemnifcation and insurance. The risk for such liability is presumed to be covered – and is in fact covered – by insurances for the beneft of all involved parties, not just the one who has bought the coverage. As far as this right is upheld, the party sufering the loss has no need to hold the other party under the contract liable for damage he has caused. This is not just a matter of taking into consideration insurance coverage which happens to be in place; the insurance arrangement is, in all detail, made a mandatory part of the liability regime under the contract. The combination of liability and insurance regimes may be seen as an alternative to traditional channeling of loss based on tort law. There are several examples in Norwegian and Scandinavian law52 that the existence or even possibility of insurance coverage of losses may be relevant when determining the reasonableness of agreed channeling of contract related risks.53 In all these examples, the insurance regimes are more loosely defned and regulated than the regime that is made an integrated part of NTK 15.

50 NTK 15 Art. 30.1 para 1, Art. 30.2 para. 1 and Art. 30.3 para. 3. 51 The company may obviously also breach its obligations, but the liability for breach is in principle not limited; see further note 7 above. 52 The Norwegian legislature has explicitly stated a wish to obtain Scandinavian uniformity in this area, see the preparatory works for the Contracts Act sect. 36: NOU 1979:32 p. 40 and Ot.prp. nr. 5 (1982–83) p. 11 f. 53 See as illustrations Jan Hellner, Speciell avtalsrätt II Kontraktsrätt, 2 häftet Almänna ämnen, tredje upplagan (Stockholm 1996) p. 224, Hagstrøm pp. 630–631 and Erik Røsæg, Lastehåndterings- og forvaringstjenester (MarIus nr. 271 s. 27–48) s. 41. See also Rt-1994–626 (“quay-inspector”) at p. 630.

136

limiting and channeling liability in norway 5 Are the agreed limitations of liability for breach of contract valid? 5.1 Points of departure The provisions limiting the contractor’s liability for breach of contract are clearly drafted and agreed without any explicit reservations. They are part of an agreed standard based on extensive and repeated rounds of negotiations between professional and balanced parties54 disconnected from the entering into of any individual contract.55 These provisions have been included in this type of contract since the beginning of ofshore activities on the Norwegian Continental Shelf, and are to be found also in other standard contracts for major construction work.56 Such clauses appear to amount to standard practice in this sector – they are “generally accepted business practice” (sect. 36). This indicates that the provisions are commercially sensible responses to general needs. There is potential for major losses, and the parties need to place the risk in a predictable way. In dramatic instances, the losses caused by breach are so high that the breaching party is not fnancially able to cover it, which means that exposure to unlimited liability does not make practical sense and thus would misguide the parties. Also, specifcally relating to the contractor’s obligation to deliver according to contract specifcations, the provisions limiting his exposure in this respect must be viewed in conjunction with his right to call on the company’s CAR insurance (see further section 6). Against this backdrop, setting aside the agreed limitations to the contractor’s liability for breach requires strong arguments. The arguments must relate to the contractor being to blame for the breach;57 if not, the limitations cannot be set aside. Some situations may be sorted out at the start: the agreed limitations cannot be invoked if the breach is caused by willful misconduct on the part of the contractor’s leading personnel.58 This liability would apply inevitably if the leading personnel acts with the knowledge that their act or omission is likely to impose loss on the 54 If the parties to an individual contract are not in balanced positions, the contractor would normally be the weaker party, implying that censoring the liability regime would be to the disadvantage to the weaker party – which would not make sense. However, as the construction part of the ofshore industry becomes more global, it is conceivable that the commercial strength of the parties tips the other way. 55 The situation is therefore very diferent from the one contemplated in the preparatory works of the Contracts Act sect. 36 (NOU 1979:32 p. 39) as being at the core of the “reasonable” test. The test according to sect. 36 aims mainly to protect the weaker party against the other party’s abuse of freedom of contract. The Norwegian Supreme Court concluded in Rt-2004–1545 that provisions of an insurance bought by a professional party were not unreasonable, referring, inter alia, to the fact that the provisions were drawn up by a committee in close cooperation with representatives of the interested parties in the relevant business sector (para. 52 of the judgement). 56 See, e.g. (the Norwegian onshore construction standard) NS 8407 cll. 40.3 and 42.5 and (the UK ofshore construction standard) LOGIC cl. 35 and 36. 57 Liability for breach is not contingent on the contractor being to blame in any way for the breach, see NTK 15 art. 24.2 and 25.2 ref. 25.1. 58 See Are Brautaset, Kontraktsreguleringen ved salg av gass (I: Brautaset et al., Norsk Gassavsetning. Rettslige hovedelementer. Oslo 1998) p.  117, observing that an important consideration when deciding whether a limitation to liability for breach is unreasonable according to the Contracts Act sect. 36 is that behind every contractual obligation there should be a party having a certain minimum liability. If exemption from all liability were to be accepted even when the breach is caused willfully by leading personnel, it would mean an exemption from liability for all types of loss covered by the limitation. This, he holds, would entail results that can hardly be accepted and that would be regarded as unreasonable.

137

knut kaasen company and in such a way that their action or omission transpires as a reckless neglect of the company’ interests (see section 5.4).59 Equally clearly, the contractor may invoke the limitations if the breach is caused by his leading personnel’s ordinary negligence – this will not be “unreasonable” according to sect. 36. It is also clear that there is no basis for setting the agreed limitation aside if the breach is solely caused by ordinary or gross negligence on the part of subcontractors or the contractor’s personnel other than the leading personnel.60 Normally the agreed limitations will stand frm even if such personnel cause the breach willfully – this is not the contractor’s “own fault”.61 The relevant criterion being whether the provisions limiting liability for breach are “reasonable”, it will not sufce to consider whether the relevant person caused the breach by an act or omission which itself must be considered as grossly negligent or even willful. Rather, the perspective has to be widened to include elements such as the type of loss created by the breach, its impact on the company, the organizational position of the relevant actor, his insight into the alternatives and consequences and other factors. However, it may be of help to try to analyze such factors separately. 5.2 The contractor’s own gross negligence In light of the discussion in section 5.1, the remaining question is whether an agreed limitation of liability for breach may be invoked when the breach is caused by gross negligence on the part of the contractor’s leading personnel. Based on the criteria contained in the Contracts Act sect. 36, the question is whether invoking the provisions in such circumstances “would be unreasonable or confict with generally-accepted business practice”. The only judgement from the Supreme Court of Norway on the censoring of agreed limitation of liability after the entering into force of sect. 36 (Rt-1994–626, “quayinspector”) touches upon the question but does not resolve it. The inspector acting on behalf of the liable party did indeed cause the damage by gross negligence, but he could not be said to belong to the party’s leading personnel. The court notes that the provision limiting the liability was part of an agreed standard contract developed in negotiations between organizations representative of the parties. It should thus be expected to balance the conficting interests, taking into consideration, among other things, options for assuring the risk and which party would be most apt to take out such insurance. The court also noted (p. 630) that the need for predictability and clearcut, simple solutions was an essential consideration in striking a balance and found it unfortunate to disturb the balance established by the provision. This judgement 59 As an example, the contractor decides to use his limited personnel resources on a competing project, thus inficting delay. This may amount to a basis for revising the agreed limitation or setting it aside, but not necessarily. 60 See Rt-1994–626 (“quay-inspector”). However, in such situations there could also be gross negligence on the part of leading personnel, e.g., in their organizing of the work or choice of personnel to perform or supervise the work, see below in text. 61 See Rt-1948–370. However, modifcations are possible – depending, inter alia, on how the work has been organized; see Brutaset (note 58) p. 116.

138

limiting and channeling liability in norway cannot be understood to indicate that the limitation would not have been accepted if the inspector had been a part of the leading personnel; the question is held open.62 Older decisions from the Supreme Court63 – prior to the introduction of sect. 36 in 1983 – are still of interest. However, they do not provide clear answers. First, they do not apply the same legal standard as provided by sect. 36 – the social norms in general and accepted business standards specifcally may have changed since the older decisions were handed down by the court. Second, none of the decisions concludes whether it is acceptable to exclude liability for losses caused by gross negligence. Finally, the old case law does not illustrate parallels to the mutually agreed liability limitation regime of the present large construction contracts like NTK 15. Consequently, the guidance ofered by this case law is vague. The issue of censoring such clauses must therefore be discussed primarily based on the criteria presented by sect. 36 itself. In this context, the perspective cannot be limited to the limitation provisions alone – it is necessary to consider the distinctive character of the contractual relationship and the nature of the contractual breach.64 Legal theory appears to take as a starting point that there is no room for excluding oneself from liability for one’s own gross negligence.65 However, it is difcult to identify any discussion of whether there is – or even should be – room for making modifcations to this rule.66 The preparatory works of sect. 36 explicitly do not conclude on the question of whether the old provision of Norske Lov 5–1–2 (which was not revoked upon the entry into force of sec. 36) excludes the right to limit liability for one’s own gross negligence.67 When exclusion or limitation of liability follows from express provisions contained in an agreed standard contract developed in the way NTK 15 is, there are strong arguments that the parties’ agreement be honored. However, it can possibly not be excluded that the contractor loses his right to limitation if the contractual breach is caused by gross negligence on the part of leading personnel with the understanding that the company is sufering a loss. Both “gross negligence” and “leading personnel” are fexible criteria providing little predictability if they were to be decisive for the outcome. Basing the result on these criteria would therefore contradict a fundamental feature of the liability regime of the contract, being that liability is to be allocated according to easily identifable criteria without notable elements of discretion or equivocality. It would not be in harmony with these considerations if the allocation of risk for liability due to contractual breach were to be decided according to fexible criteria allowing for extensive discretion. The urge to minimize transaction costs speaks against this solution.68 At 62 Hagstrøm, Ansvarsfraskrivelse p. 486 is of a diferent opinion. 63 A thorough exposé of these judgements is given by Hagstrøm, Ansvarsfraskrivelse p. 448. 64 See in this direction Krüger, Norsk kontraktsrett (Bergen 1989) p. 786. See also the summary of the Swedish preparatory works related to the similar Swedish act (which in light of the prehistory of sect. 36 must be assumed to be of great interest when interpreting the Norwegian Act) in Hagstrøm, Ansvarsfraskrivelse pp. 462–465. 65 See the review in Hagstrøm, Ansvarsfraskrivelse pp. 424–430. 66 See Brautaset (note 58) at pp. 118–119 for a discussion of arguments that supports that the main rule should be as stated. 67 NOU 1979:32 p. 19 (NL 5–1–2 contains the pacta sunt servanda principle). Hagstrøm, Ansvarsfraskrivelser claims that Norske Lov 5–1–2 prohibits such limitations. 68 Hagstrøm, Ansvarsfraskrivelse pp. 480–481 is more reserved.

139

knut kaasen the same time the wish to prevent breach, which generally is a strong argument for setting agreed exclusion clauses aside,69 has little bearing in this context. The risk of losing professional reputation and esteem is likely to be a stronger incentive for the leading personnel of the contractor not to commit willful or grossly negligent breach than is the prospective of losing the right to limitation. Admittedly, this may be diferent in situations where the contractor knowingly puts his interests (all things considered) ahead of the company’s. However, such actions will result in the full exposure for liability if the contractor’s knowledge includes the likely consequence to the company provided that his action on this basis must be considered to imply a reckless disregard of the company’s interests (see section 5.4). On the other hand, willful misconduct is difcult to prove. The wish to avoid cases of “hidden” (unprovable) willfulness may therefore be an argument for setting aside limitation provisions also in case of gross negligence. Certainly, this would imply just moving a difcult borderline. However, admittedly there is an important diference between proving willful misconduct and drawing the line between simple and gross negligence. Proving willful misconduct necessitates performing a challenging assessment of the contractor’s consciousness, while distinguishing between simple and gross negligence mostly requires examination and evaluation of just external facts. This striking diference makes it questionable to equate willful misconduct and gross negligence when deciding whether to censor agreed limitation clauses. As noted, obtainable insurance programs may be relevant in the test of reasonableness under sect. 36. If the contractor’s liability for breach is limited, the consequence may be that loss exceeding the limit has to be carried by the company. Loss due to physical damage may be – and is expected to be – covered by insurance; this is an important element in the knock-for-knock regulation because there is a direct connection between the liability and the insurance regimes of the contract (see section 6.2). To the extent that insurance is also an option for loss induced by breach, insurance may be relevant in determining whether the limitation is unreasonable to the company. However, there is no feasible insurance coverage available to cover losses resulting exclusively from breach as opposed to losses caused by physical damage (which may be caused by breach), and the contract does not assume there is. Originally, the company’s CAR insurance normally covered losses due to damage caused by “faulty design/materials/workmanship”, i.e., cases of classical breach as opposed to a fortuitous event. In practice, this implied that the company could claim against the insurance if it sufered losses because the contractor’s liability for such breach was limited. In recent years, the tendency seems to be that the CAR also covers physical damage to objects included in the property insurance coverage of CAR if caused by “a Defective part, faulty design, faulty materials, faulty or defective workmanship or latent defect”, however, damages to the defective part itself are not covered unless caused by another “insured peril” due to no fault or defect of the part itself.70 This would seem to imply that the CAR policy no longer provides coverage for losses due 69 See, e.g., Hagstrøm, Ansvarsfraskrivelse pp. 479–480. 70 Sited from WELCAR, which normally is basis for CAR insurances on the Norwegian Continental Shelf. WELCAR is developed by Joint Rig Committee (JRC) at Lloyd’s of London. WELCAR 2001 is available in its original form at https://content.sbigeneral.in/uploads/c979f1604f6f4423b15ba81a545c8875.pdf.

140

limiting and channeling liability in norway to limitations to the contractor’s obligation to rectify defects, but possibly for losses resulting from the contract limiting others of the contractor’s liabilities for breach. 5.3 Who is “the contractor”? The contractor clearly loses his right to limit his liability for default if the default is caused by his own willful misconduct (with knowledge of the consequences, see section 5.4), and it cannot be ruled out that the situation will be the same if the default is caused by his own gross negligence (see section 5.2). Because the contractor is always a legal entity (as opposed to physical), it is necessary to determine which physical persons and corporate bodies constitute “the contractor” in this context. The conduct of these persons has the potential, in relation to the contractor’s default, to result in the loss of its right to limitation – provided there is a relevant causal link between their conduct and the contractor’s liability for breach. The basic argument for setting aside agreed limitations if the default is caused by willful misconduct is that behind every contractual obligation there should be a contract party having a minimum liability for fulflling the obligation.71 If the parties were allowed unlimited freedom to contract out of this fundamental prerequisite for the contract being “binding”, the concept of pacta sunt servanda would evaporate. Consequently, a guiding observation may be that “the contractor” should be identifed as the persons and corporate bodies who have been entrusted the task of making decisions or taking actions which directly impact the contractor’s fundamental ability to perform his contractual obligations – as opposed to every single element that would need to be in place in order to fulfll these obligations. By way of illustration, the identity should possibly comprise of those who, on behalf of the contractor, determine the qualifcations of the lead engineer, but should not include the engineer who’s performing the calculations. The contractor’s board of directors is clearly “the contractor” in this respect, as is the CEO.72 This view must be extended beyond the concept of corporate bodies. The technical manager must be identifed with the contractor in relation to activities performed under his supervision,73 even though the “quay-inspector” case74 indicates that this cannot be taken without modifcations (the inspector could be said to be the “technical manager” of the forwarding agent). On the other hand, the agreed limitations clearly are valid if the default or damage is caused by gross negligence (or even willful misconduct) on the part of subordinate employees or

71 Brautaset pp. 17 and 119 and note 58 above. 72 Possibly, there is a limit to this observation: if the contractor’s top management, or other persons that he is liable for, substantially exceed what can be reasonably expected in their role, the agreed limitations to the contractor’s liability for default may be upheld, in line with the views expressed in Rt-1959–849 (“the apprentice”). 73 An example is an arbitral award 20 August 1973 (referred by Knudtzon in Nordisk Skibsrederforening Medlemsblad Spesialnummer A 1984 s. 66) where a yard was held liable for decisions taken by its technical manager, despite agreed limitations. 74 Rt-1994–626.

141

knut kaasen subcontractors. For this purpose, neither of these groups may be identifed as the contract party itself. Beyond these observations, the border is hard to draw.75 The formal position of the employee may serve as a good starting point, but it is not necessarily decisive if it is not combined with real infuence on the matter in question. If a decision by the board of directors does not impact the contractor’s fulfllment of his contractual obligations (because other factors come in between), it would not be to his detriment that the decision in itself would deprive him the right of limitation. The other way around, the contractor would hardly lose his right to limit liability due to the actions of an employee if the employee has not been given the task of managing basic aspects of the contractor’s fulfllment in the relevant area – i.e., has not been given a formal role on a higher organizational level. For this to happen, there must be organizational disorderliness attributable to blameworthy actions by the leadership of the contractor to such an extent that the agreed limitations may not be invoked for that reason. The hierarchical position is thus important, but often the individual circumstances of the situation will be decisive. For example, it cannot be stated generally whether the party’s project director is “the contractor itself ” in this context. It will depend, inter alia, on how independently he is allowed to (and expected to) operate on behalf of the party in managing the project.76 The parties’ “representatives”, appointed in accordance with NTK 15 art. 3.1, normally cannot be identifed with the party in relation to the provisions on limitation of liability. Normally, the independent authority of the representatives is restricted, and they do not belong to a sufciently high level of the party’s organization. However, the contractor’s representative may well also be the project director. In that case, the latter position may be relevant in identifying him with the contractor in relation to limitation of liability. Nevertheless, a person’s hierarchical position is not wholly decisive. The contractor may be identifed with persons having been trusted with the supreme authority within defned areas of the contractor’s business activities, no matter their formal position. Normally this would require that the vested authority cover an independent function in fulflling the contractual obligations in such a way that the person makes independent decisions substantially efecting the fulfllment.77

75 An extensive analysis of the question of identifcation – specifcally in relation to limitations to the shipowner’s liability according to the Maritime Act – is to be found in Ole Lund, Om erstatning for mangler, forsinkelse, utbedringsplikt og prisavslag ved skipsbygging, in AfS 8.313 f. The issue is also discussed comprehensively in the arbitral award “Trans Tind” (ND 1984 p. 404). See also Jan Kaare Tapper, Fabrikasjonskontrakten (in Askheim et al., Kontrakter i petroleumsvirksomheten, Oslo 1983) p. 183 and Brautaset (note 58) pp. 115–116. 76 An illustration in this respect is the arbitral award “Mørland 7” (ND 1988 p. 263). In this case the contractor was identifed (in relation to the liability provisions of the contract) with his project director, who had “the authority to make necessary decisions in relation to the performance of the project”, but who were not authorized to sign (in company law meaning) on behalf of the contractor. It was of no relevance that the manager was only coincidentally present when the decision was taken, nor that the decisions otherwise would have been taken by an employee whom the contractor could not be identifed with. 77 In the same direction, Hagstrøm, Ansvarsfraskrivelse pp. 517–518.

142

limiting and channeling liability in norway 5.4 The role of the concept of “willful misconduct” As we have seen, if contractual breach is due to willful misconduct by the contractor, the enforceability of provisions limiting his liability for the consequences stand at risk. However, the unescapable condition for setting such provisions aside is whether they are considered “unreasonable” under the circumstances (see section 4.1). The question is therefore whether “willful misconduct” will necessarily satisfy this test.78 In my opinion the concept is not quite suitable for the role it has been given in this context. Traditionally, the evaluation of reprehensibility has moved along a continuum from mere negligence79 via gross negligence80 to willful misconduct. Consequently, one has coined sentences like “no-one can exclude liability for consequences of willful misconduct”. This implies a need to decide whether the willfulness has to refer to the efects that the action is likely or certain to have for the other party – in practice inficting a loss – or whether it sufces that the default itself is caused willfully (or by gross negligence, if that is the relevant criterion).81 The concept “willful misconduct” appears on these occasions to be understood in line with the criminal law concept of intent. A key element is to which extent the action or omission is conscious, and possibly whether the actor is conscious about the consequences that the action or omission is likely to have to the other party. This appears to be a sound starting point. A contractual party may breach his contractual obligations willfully without actually knowing or counting on the other party thereby sufering a (substantial) loss – and without there being basis for criticizing him for not considering this possibility. In ofshore contracts, this is not a theoretical situation; frequently the contracts specify methods and actions to be implemented by the contractor in constructing the contract object, but these requirements do not necessarily need to be adhered to in order for the contract object to fulfl the contractual requirements pertaining to the contract object at delivery. Nevertheless, not fulflling the requirements amounts to breach, but not necessarily substantial losses to the other party. The conditions for applying the criteria of sect. 36 for setting agreed limitations aside (the reasonableness test) must at least be that the contractor has breached his contractual obligations, that the company, for that reason, sufers 78 This is simpler in relation to the knock-for-knock regime (art. 30): For the limitations to be set aside, the damage to the other contract party, his “group” (section 3.2 above) or third parties, as a minimum must have been caused by the contractor’s leading personnel’s gross negligence or willful misconduct. This situation is hardly frequent but may occur if the management – typically for economic reasons – exposes the surroundings to an unacceptable substantial risk for physical damage. It is more conceivable that such damage is caused by an employee’s gross negligence. In such cases the provisions on limitation and indemnifcation will be given full efect – this is the kind of situations they are designed for. 79 According to a widely used defnition: the failure to exercise the standard of care that a reasonably prudent person would have exercised in a similar situation. 80 According to a widely used defnition: a high or serious degree of negligence; a very marked departure from the standards by which responsible and competent people habitually govern themselves. 81 The proper answer has been disputed. Per Augdahl, Den norske obligasjonsretts almindelige del, 5. ed. (Oslo 1989) p. 293 og Lund, Standardkontrakter, bilsalg og preseptoriske regler (in LoR 1964 pp. 66–81) on p. 68 appear to put decisive weight on whether the breach itself is gross negligent or even willful, while Per Brunsvig, AfS pp. 1.273, 3.172 and 13.354 argues that it is a prerequisite that the party, either grossly negligently or willfully, has inficted a loss to the other party. The arbitral tribunal in “Trans Tind” (ND 1984 p. 404) endorses Brunsvig’s conclusion.

143

knut kaasen a loss and that the contractor is aware of both of these facts – or at least that his ignorance must be attributed to gross negligence at a sufciently high level of his organization. Such consciousness must be assumed to be a condition for censoring agreed limitations contained in the construction contracts (see section 5.2 on gross negligence). However, it is not sufcient. The contractor may breach his duties with open eyes, knowing that this will cause loss to the company, without necessarily losing the right to limit his liability.82 The starting point is that the limitations are agreed without any modifcations. If these provisions are to be censored, it must be because the contractor, by markedly and fagrantly departing from the standard of conduct of a reasonable person, sets his interests above those of the company.83 Under such circumstances, applying agreed limitations would be “unreasonable”. Neither “willful misconduct” (alternatively “intent”) nor “gross negligence” fts well as an exhaustive criterion for evaluating the relevance of the contractor’s ideas of the consequences of the breach. It may seem more proper to tie in directly to an evaluation of the contractor’s balancing of his own interests against the company’s. The crucial issue is not whether this act of balancing is “grossly negligent” or even “willful” or “intentional” (but as noted, it is in practice a precondition that it must fall into one of these categories). The decisive characterization is whether the balancing is “reckless” or “clearly disloyal” to the company’s interests, or something to that efect.84 If the action or non-action chosen by contractor results in physical damage to the company, as opposed to non-fulfllment of contractual obligations, the situation may be diferent. The characteristics of the action/non-action would then be at the forefront: was the damage caused by gross negligence or even willful misconduct (at a sufciently high level of the contractor’s organization)? In making this determination, it is often relevant to consider to which extent the contractor realized the possible consequences of the action/non-action to the company – the duty of care is infuenced by which consequences may result from lack of care. However, these consequences, and specifcally whether the contractor evaluates them reasonably, are more central in case of breach than if the contractor causes damages. Actions/non-actions leading to breach are normally intentional in the sense that the contractor is well aware of the fact that the action is or is not taken (e.g., delay resulting from not engaging a night shift). The focus is therefore shifted to whether the decisions so made appear to take the company’s interest into account to a sufcient degree. The decisive criterion should be a general evaluation of whether the contractor’s actions emerge as reckless or clearly disloyal towards the company, which is not necessarily the case even if the breach is intended and the contractor knows that it

82 Krüger, pp. 784–785 om “forsettsbegrepet som kontraktsrettslig kvalifkasjonskriterium”. 83 This is illustrated in the “Trans Tind” arbitral award (ND 1984.404), where the tribunal underscores that (my translation) “if the management has violated the rules that it is bound by, there is a willful breach on the part of the management. However, the management may have assumed that this breach would not have any detrimental efects on the fnal result. Only if this assumption must be categorized as grossly negligent the provisions limiting the liability for breach should be censored. What entails such consequences is not the formal breach, but actions that may damage the other party and that for this reason must be characterized as grossly negligent.” 84 I am therefore not quite in line with Hagstrøm, Ansvarsfraskrivelse p. 491 on this point.

144

limiting and channeling liability in norway will cause loss to the company. It is only in such grave situations that it would appear “unreasonable” to uphold the provisions limiting the liability (see the Contracts Act sect. 36). For this reason, it does not appear adequate to base the discussion of these questions solely on the traditional concept of willful misconduct or gross negligence – other elements must be included. 6 Censoring the knock-for-knock regime? 6.1 The question The risk of physical damage to persons and property in connection with the contract work is allocated “regardless of any form of liability whether strict or by negligence, in whatever form”; the decisive criterion is where (and sometimes when) the damage hits, not who causes it and how it is caused. This implies an express exclusion of liability, explicitly without any reservation. Can this arrangement be maintained in light of the general rules on the enforceability of provisions excluding liability? The relevant general rule, also in this context, is the Contracts Act sect. 36. Several authors discuss this question. Bull85 assumes that the courts will not censor such provisions where – as in the case of NTK 15 – the contract’s liability regime is based on reciprocity, except in cases of the tortfeasor’s gross “own fault” (egenfeil). He argues that the basic approach of the courts in relation to this type of liability regimes should be that the regime is accepted as appropriate, adequate and reasonable (p. 393). As will transpire from the following, I concur. Selvig86 discusses the application of sect. 36 in professional relations. See also Sandvik and Ness.87 Wilhelmsen discusses the basis for censoring knock-for-knock clauses of ofshore charter parties (maritime services contracts) and concludes that the legal situation must be considered to be uncertain.88 Zak discusses the validity of liability clauses in drilling contracts under Norwegian, English and American law. The discussion is relevant also to complex liability and insurance regimes contained in other types of contracts, including NTK 15. Referring to the complexity of the regimes and of the factual context to which they apply, she voices objections against applying absolute rules of invalidity in this area.89 The validity of the liability exemption clause of the (then operational) standard ship building contract is discussed in the arbitral award concerning “Trans Tind”.90 The issue has clear resemblance to the question of censoring provisions limiting liability for default (see section 5), but there are also important diferences. The point of departure is the same: the parties may exclude liability for loss and damage caused by subordinate employees and subcontractors, regardless of any form of liability, in whatever form. The parties may also exclude liability for loss and damage caused by 85 Bull (note 30) p. 394 (with further references). 86 Selvig, Kontraktsretten (in: Knophs oversikt over Norges rett, 10th ed. 1993 p. 553 f., particularly at p. 559. 87 Sandvik, xxx pp. 239–240, Ness, Ansvarsfordeling og forsikring i petroleumskontrakter, MarIus nr. 323 (2005) p. 59 f. 88 Wilhelmsen, supra note 4, at pp. 102–111, particularly at p. 110. 89 Zak, Ansvarsregulering i borekontrakter, MarIus nr. 415 (2013) p. 138. 90 ND 1984 p. 404.

145

knut kaasen simple neglect by any person whom they are identifed with (see section 5.3), but not for the consequences of their willful misconduct. The doubt relates to whether a party may also exclude liability for gross negligence on the part of his leading personnel.91 In this discussion, there emerges a major diference between clauses excluding a party from liability for physical damage that he has caused and those clauses limiting his liability for breaching the contract. While the risk of sufering physical damage or being held liable for having caused it can be insured, the availability of insurance coverage for risk related to breach is much more limited. This puts a mark on the discussion of reasonableness of the respective contractual regimes. The combination of liability and insurance regimes may be viewed as an alternative to traditional tort law–based channeling of losses. The relevance of insurance when determining whether agreed risk channeling is “reasonable” can be seen from the “quay-inspector” case, where existing options for obtaining insurance and which of the parties that were closest to buy insurance are elements given weight in the test of reasonableness.92 This is also emphasized in the Swedish preparatory works for the Swedish Contract Act, which contains a provision similar to the Norwegian Contract Act sect. 36.93 Hagstrøm (p.  662) states that the Swedish preparatory works allow for accepting exclusion of liability even for gross negligence if insurance options make it unobtrusive. Røsæg94 fnds that the importance of insurance in relation to the channeling of liability for cargo handling and storage can hardly be overstated: it may be rational to waive claims for damages even in cases of willful misconduct if that results in economic gain, e.g., by obtaining a rebate exceeding the insurance premium. Ness95 provides further overview of the attitude in Nordic jurisprudence regarding the relevance of insurance when determining the reasonableness of clauses excluding liability. In all these examples, the insurance arrangements contemplated are of a looser nature than the one NKT 15 makes an integrated part of the liability regime.

91 This situation is not very theoretical. When the GBS structure (support structure held in place by gravity) of the “Sleipner A” platform sank (August 1991) prior to delivery, it was argued by the client that the loss was caused by the management of the contractor by gross negligence. The argument was that the management did not organize the calculation of the strength of crucial parts of the structure in a proper way. See also the arbitral award “Mørland 7” (see note 76 above; ND 1988 p. 263 f.): a barge was damaged when four steel piles were lifted of from it under the supervision of the ofshore contractor who had chartered the barge. The charterparty contained a knock-for-knock clause corresponding to the one found in NTK 15 art. 30.1. The contractor was acquitted from the claim concerning damages caused by the two frst piles but held liable for the damage caused by the two last. In the latter case there was a willful ofering of the owner’s interest for the beneft of the contractor, in that the contractor had decided that it would be better inficting further damage to the barge than exposing more valuable interests to the risk of extensive damage. (The piles were used to pile a steel jacket to the seafoor, and the jacket would possibly not withstand the inclement weather conditions with just two piles in place.) This balancing of interests was defensible, but the tribunal found that this willful causing of damage implied that the contractor could not rely on the knock-for-knock clause. 92 Rt-1994–626 at p. 630. Insurance arrangements was also considered relevant in the “Chainco” judgement (ND 1991 p. 180, Eidsivating court of appeals). See also Ness (note 85) p. 52. 93 SOU 1974:83. The Norwegian lawmaker explicitly states that Nordic uniformity of law is important regarding sect. 36, se NOU 1979:32 p. 40 and OT.prp.nr. 5 (1982–83) p. 11 f. 94 Røsæg, Lastehåndterings- og forvaringstjenester, MarIus nr. 271 (2011) p. 4. 95 Ness (note 85) pp. 53–57.

146

limiting and channeling liability in norway 6.2 “Organized fnancing of loss” The link between insurance and the contractual liability regime implies that the latter is not just a clear-cut exclusion of liability. The combined efect is that the system establishes an “organized fnancing of loss”96 by means of insurance: the insurer shall, according to the contract, in the last resort carry the risk that is channeled to a party.97 The implication of the liability regime is therefore not that the loss is shifted over to the sufering party, leaving it to him to cover the risk by insurance as he sees ft, which is normally the case. Rather, the liability provisions are part of a larger system where a crucial and compulsory element is a rational organizing of the kind of insurance coverage that the parties naturally need in light of the liability clauses. This has two important consequences: frst, the reasonableness of the clauses excluding a party from liability for loss and damage he has caused to the other party cannot be evaluated as an isolated phenomenon; and second, insurance arrangements are brought into the evaluation. The practical implication is that the contractual channeling of the risk related to physical damage is supplemented by elements of insurance law, where the fundamental question is what extent of protection a contractual party has under the insurances he or the other party has acquired in fulfllment of their obligations under the contract’s insurance clauses. As far as this protection applies, the distribution of risk under the contract can hardly be considered to be unreasonable. The contract, expressly and in detail, links liability clauses to provisions requiring one of the parties to acquire insurance coverage for the risk allocated by the liability rules. This implies that it can, at least, not be easier to censor clauses excluding liability for physical damage than censoring provisions limiting liability for defects. It is therefore not conceivable that the reasonableness test of the knock-for-knock regime will be more extensive than the one discussed in section 5. However, this presumes that the required (and supposedly acquired) insurances actually cover the risk as intended. If they do not, for reasons that are not attributable to the party losing the intended protection, the general existence of insurance can no longer serve as an argument for the liability clauses being reasonable. Consequently, even though we are dealing (like in section 5) with an issue of censoring contracts on the basis of sect. 36, we have to expand the perspective by investigating whether the insurer may refuse to cover the loss and damage as contemplated in the contract. The point of departure is that the insurer (assumedly) has undertaken to cover the risk allocated to a contractual party by the liability regime of the contract, including the risk following from the other party having the right under the contract to deny liability which would otherwise follow from ordinary rules of tort or contract law. In other words, the liability provisions of the contract defne the hazards at risk under the insurance.98 Further, as a main rule, the insurer is bound by this liability regime – he has accepted that they defne the hazards at risk if rules of insurance law do not relieve him from the obligation to cover the loss. In practice, this is a question of whether the insurer has waived his right of subrogation against the party causing the damage in cases where the

96 Bull (note 30) p. 349. 97 Admittedly, there is normally a deduction for own risk. 98 Bull, Forsikringsrett (2008) p. 205 discusses the hazard at risk (farefeltet) of insurances.

147

knut kaasen contract has channeled the loss to the party sufering the damage. According to the contract, the terms of the insurance shall include such waiver.99 6.3 Can the waiver of subrogation be maintained? 6.3.1 Waiver of subrogation Based on the above, the overall assessment of the validity of the knock-for-knock regime partly rests on whether the insurer ultimately has to cover the loss resulting from the damage caused by the contract party. The question is whether the insurance terms on waiver of subrogation can be maintained no matter the party’s actions when causing the damage.100 The starting point is that the insurer explicitly has waived any right of subrogation against the tortfeasor.101 There is hardly any doubt that this is an element in a rational allocation of risk and liability under the contract, established by professional actors in the course of well-considered negotiations resulting in an agreed standard contract which has been continuously in use since the mid-1980s. The issue is, however, not whether these elements of the contract are reasonable, but rather whether the insurer’s waiver applies in all situations (see the Contracts Act sect. 36). This will not be dealt with thoroughly, but some elements of the evaluation will be discussed. 6.3.2 The co-insured In addition to the waiver of subrogation, the insurances required by the contract shall also contain clauses naming the other party as co-insured. These two elements may appear to constitute an overkill, as the co-insured party is protected from claims 99 NTK 15 art. 31.1 para. 3 and art. 31.2 para 3. 100 The background rules of law are basically the Norwegian Tort Act sect. 4–3, stating that an insurer who has paid damages for property damages has a right of subrogation against the tortfeasor to the same extent that the suferer could have claimed the tortfeasor based on the Act sect. 4–2. The main rule under sect. 4–2 is that the suferer cannot claim the tortfeasor if the loss is recoverable under the suferer’s property damage insurance. However, exempted from this are damages caused by the tortfeasor in the course of his trade or economic activity. Most cases of damage caused by one party to the other in connection with the contract work fall into this exemption. Consequently, the practical main rule according to background law is that the insurer has a right of subrogation against the tortfeasor for compensation paid to the sufering party. However, it follows from the main rule of the Act sect. 4–3 that the insurer is bound by clauses excluding the tortfeasor’s liability to the same extent as the clause can be maintained by the tortfeasor against the sufering party. The rules therefore establish the subrogation principle: the insurer steps into the sufering party’s legal position against the tortfeasor. Against this backdrop, the position of the tortfeasor according to a property insurance should depend on whether his exclusion of liability as part of the knock-for-knock regime can be upheld. So far, the situation resembles the issues of censorship discussed in section 5above. However, since the contract requires the parties to obtain insurances where the insurer waives any right of subrogation against the other party, we see that the “reasonableness” – and thus, validity – of the clauses excluding liability for property damage depends on whether this waiver is binding in all situations. Admittedly, the liability clauses are the point of departure, but the fnal allocation of the loss may depend not only on the allocation contemplated in the contract, but also on the scope of the waiver of subrogation contained in the insurance terms. 101 Two examples of such clauses: 1. “It is . . . agreed that all rights of subrogation hereunder shall be waived as respects the Assureds hereon . . ., whenever contracts executed by the Assureds require such waiver of subrogation, provided such waiver is agreed in writing prior to a loss as may be covered hereon.” 2. “Underwriters agree to waive rights of subrogation against any Assured and any person, company or corporation whose interest are covered by this policy.”

148

limiting and channeling liability in norway of recourse to the same extent as the insured party has a claim under the policy.102 However, this overkill paves the way for the following notes on insurance law. According to insurance law, the rights of a co-insured party are subject to some distinct limitations, no matter the seemingly unreserved wording of the waiver and co-insurance clauses. The insurer cannot waive his right of subrogation beyond the restrictions imposed by rules of compulsory background law. According to the Act relating to Insurance Contracts (No. 69/1989) sect. 4–9 frst para., the insurer is not liable if the insured has intentionally brought about the insurance event. This implies that the insurer is not prevented from holding liable a co-insured party who has caused the damage intentionally, despite any agreed waiver of subrogation. Further, it follows from sect. 4–9 second para. of the Act that the liability of the insurer may be reduced or cease to exist if the insured has brought about the insurance event by being grossly negligent. This applies, however, only to an insurance other than liability insurance, that is to property damage insurance. The main rule of the Act is therefore that gross negligence on the part of the insured does not relieve the insurer from the obligation to cover the damage so caused. In the case of property damage insurance, it depends on the degree of blame and the circumstances in general whether the insurer is liable to cover the damage partly or in total. 6.3.3 Are the arrangements “reasonable”? This normal system of the Act carries extensive weight when evaluating whether the insurance system established by the contract is “reasonable”. However, this necessitates deciding if the insurances required by the contract shall be labeled liability or property damage insurances. Only if labeled liability insurance, the Act itself provides sufcient basis for enforcing the clauses on waiver (and co-insurance). The classifcation of the insurances is less obvious than one would expect. The question may be illustrated by the situation where the contractor’s management by acting grossly negligent has caused damage to the contract object prior to delivery. The contractor shall carry out necessary measures to ensure that the work is completed in accordance with the contract (NTK 15. 29.1), but his costs in doing so are limited to an individually agreed maximum amount, inter alia, if the damage is covered by the company’s Construction All Risks (CAR) insurance (art. 29.2). In his capacity as co-insured under this policy, the contractor has an independent right of coverage.103 If the CAR, for this purpose, is considered a liability type of insurance, there can be no room for rejecting the contractor’s claim for coverage referring to the Contracts Act sect. 36. If the CAR is categorized diferently, the situation might be diferent. Several approaches may be taken to this question. One may base the discussion on the fact that the company owns the contract object as it comes into being (art. 22.1), implying that the contractor has to compensate the company’s loss caused to

102 Bull (note 30) pp. 387–388 with further references. It is, however, possible that the waiver of subrogation clause in some situations will provide a wider protection than the one following from the position as insured. See Ness (note 85) p. 86 f. 103 The issue in this situation is therefore not the waiver of subrogation – there is no claim form the company that the contractor may step into. Hagstrøm, Ansvarsfraskrivelse appears to take a diferent view at p. 476 f., especially pp. 87–491.

149

knut kaasen its property – i.e., the insurance works as liability insurance. However, even under this view, the underlying reality is that the contractor carries the risk of the contract object until delivery; the right of ownership is not relevant in this context.104 6.3.4 The conclusion The answer is not certain, and it is not necessarily decisive as to how the insurance is categorized since the contract provisions on insurance do not have to be set aside based on the Contracts Act sect. 4–9 (except in case of willful misconduct on the part of the contractor’s management). The decisive question is whether upholding the required waiver (and co-insurance) is unreasonable under the Contracts Act sect. 36 test in cases where the contractor has caused the insured event by his own management’s gross negligence.105 When evaluating reasonableness, it carries great weight that the contract’s provisions on liability and insurance stand out as rational and appropriate – commercially and legally. Coordinated insurance solutions imply that losses within the sphere of the contract are covered irrespective of who has bought insurance, who has caused the damage or loss and – at least apparently – whether the insurance in question is categorized as liability or property damage insurance. Therefore, great weight should be attached to the fact that the insurer has undertaken an unconditional obligation to exclude the contractor from claims of subrogation and name him as co-insured. It is also important that this arrangement is part of a well-known and well-established contractual liability regime where the coverage at least in reality can be said to take the form of liability insurance (having consequences under the Contracts Act mentioned above). This system deserves acceptance on the additional basis that property damage insurance containing a waiver of subrogation against a party who is also co-insured under the policy, contains an element strongly resembling liability insurance.106 7 Do the NTK 15 provisions on exclusion and limitation of liability hold good? The answer depends on whether the result emerges as “unreasonable” under the test based on the Contracts Act sect. 36. This requires a complex evaluation of a number of elements. The fundamental task of interpreting the contract provisions ofers few challenges. There are no indications that the provisions are meant to imply less wideranging exclusions and limitations than allowed under the background rules of law. On the contrary, in relation to the knock-for-knock regime, the contract specifcally provides that the exclusions and indemnifcations apply regardless of any form of liability, whether strict or by negligence, in whatever form, and without any distinction regarding how, or by, or to whom the damage is caused. 104 The title is regulated for a diferent purpose, being a wish to pave the way for the company’s right to take delivery of the contract object (in its present state) in competition with the contractor’s bankruptcy estate. 105 In this context, it may be of relevance that the “sliding scale” introduced by sect. 4–9 para. 2 opens for a more fexible approach in the handling of grave instances than does the black or white approach resulting from categorizing the insurance as a liability insurance. It may seem ofensive to allow the contractor full coverage in cases of qualifed negligence. On the other hand, all cases of damage except those caused by management willful misconduct is covered under a liability insurance. 106 Ness (note 85) p. 109.

150

limiting and channeling liability in norway Important issues when deciding whether to censor these provisions have traditionally been who has caused the loss (management, subordinate employees or others), what kind of negligence was involved (willful misconduct, gross or ordinary negligence), the party’s insight into the consequences of his actions, and what kind of losses are subject to limitation or exclusion (especially direct/indirect losses). On this basis, it is possible to derive certain clear rules. The contractor cannot rely on clauses limiting or excluding his liability for losses caused by his management’s willful misconduct if the conduct can be categorized as a reckless disregard for the company’s interests. On the other hand, he may normally limit or exclude liability for losses caused by his management’s ordinary negligence or even by gross negligence on the part of his other employees. Further, when deciding whether the contractor may also limit or exclude liability for loss caused by his management’s (i.e., his own) gross negligence, additional factors need to be considered in order to determine the reasonableness. The help ofered by case law is, in my opinion, limited in this respect. Nor does legal theory provide a clear foothold, though it must be said that it, generally, is skeptical to limitation or exclusion of liability in these situations. Both case law and legal theory identify elements that must be considered relevant in evaluating the issue. However, this is of limited help as long as neither of the sources discuss which implications follow from these identifed elements operating in combination, in the way illustrated by the complex construction contracts. Four aspects appear central in this context – separately and especially combined. First, the NTK 15 is an “agreed document”, created by highly professional parties devoting huge resources to the task. Most of the contents – and all of the issues discussed here – have remained unchanged through multiple revisions over a period exceeding 30 years. We are far of from unbalanced, one-sided standards of shady origins and based on doubtful legal understanding. And if there were to be a diference between the parties in terms of economic strength, it typically would imply that the party that would have an interest in censoring the clauses limiting or excluding liability would be the stronger party – the company – although market developments in recent years ofers examples of the opposite. Second, the clauses limiting and excluding liability are, in principle, mutual, albeit the contractor is the one who is – for practical reasons – the more likely benefciary of the system. We do not look favorably upon the one-sided type of regimes that typically call for scrutiny by the legal system. Third, the clauses limiting and excluding liability must be viewed in connection with the insurance regime established (not just contemplated) by the contract. The discussion of reasonableness must take the combined efect of these regimes into account. This is markedly diferent from situations where there is just a more or less relevant possibility of buying insurance coverage, which is not refected in the contract’s risk regime and even less made an integral part of it. Finally, the liability regime appears as a necessary part of a thoroughly prepared totality that seems commercially sensible: an unlimited exposure to liability does not make commercial sense in contractual relations of this type and size. Channeling the risk to insurance appears to be a rational approach, and it is a beneft that the need for evaluating causality and individual loss is reduced. The limitation and exclusion of liability are parts of a complex and complete system, where it must be assumed – based on the 151

knut kaasen history of the establishing of the system – that the parties internally have balanced the pros and cons of each of the elements and of the totality. Even though the starting point in more classical contractual relations is that a party cannot rely on clauses limiting and excluding his liability for loss that he has caused by gross negligence, the aforementioned characteristics of complex construction contracts like NTK 15 are, in my mind, clearly convincing arguments for accepting such clauses in such contracts. This view is applicable to both of the main forms of limitation of liability; limiting and exclusion of liability for breach, and distribution of risk for losses caused by damage to the contract object or other property in connection with the contract work. However, in the latter context the arguments are even stronger: both the reciprocity and the direct link to the insurance come to the forefront, and the knock-for-knock system stands out as a well-founded rational way of handling this type of risks in this type of contracts. Following this, my view is that the clauses of NTK 15 and similar contracts limiting or excluding liability can also be maintained when the breach, damage or loss is caused by the contractor’s own gross negligence. If the limitations or exclusions are not to be respected, it must be because it is proven that the breach, damage or loss was caused by willful misconduct on the part of the leading personnel of the contractor, knowing that the consequence will be a reckless disregard of the company’s interests.

152

CHAPTER 7B

The obligation under Norwegian ofshore contracts to insure the contract object as part of a knock-for-knock agreement Erik Brannsten

1 Introduction and background This chapter concerns the main features of the oil company’s obligation to insure the contract object under an ofshore construction contract compared to the available insurances in the market. Ofshore construction insurances are normally taken out on a project basis to support a project overall knock-for-knock arrangement, such as the knock-for-knock provisions found in the Norwegian ofshore contracts’ arts. 29 and 30, used for construction of production facilities in connection with an ofshore development project on the Norwegian Continental Shelf (NCS). The ofshore contracts most commonly used for this purpose are the Norwegian Fabrication Contract 2015 (NF), the Norwegian Total Contract 2015 (NTK) and the Norwegian Subsea Contract 2005 (NSC). Throughout this article, NF/NTK and NSC will be referred to as the “ofshore contracts”. The parties to the contracts are the “Company”1 and a “Contractor”, (see section NF/NTK art. 1.23/1.24 and NF/NTK art. 1.15). In 2016, a new set of standard purchase conditions were also established, referred to as “Norske Innkjøpsbetingelser/Norwegian conditions for purchase” NIB 16. These are often used for subcontracts for the purchase of materials and equipment. Development of ofshore oil and gas production on the NCS is a complex and hazardous task. Construction of production platforms and subsea installations takes many years and involves large investments, technical challenges and a high risk for loss and damage to personnel and property. When concluding an ofshore contract, a Company’s main objective is to receive timely delivery of the contract object2 which is suitable for the purpose that the Company intends to use it for. Common for the contracts is that the Company will become the owner of the contract object, and that title will incrementally pass to the Company as the work progresses (see art. 22). Should loss or damage to the contract object occur during construction, it is crucial for the Company to be able to fnance, repair the damage and complete the works within the delivery date. Such fnancing is one of the main purposes of the project insurance. 1 The Company will normally be a group of companies which together holds a production license and are thus also often referred to as a licensee. One of the licensees will be nominated as the operator and shall take care of the operational activities of the license group. 2 The term ‘Contract Object’ is defned in NF/NTK 15 art. 1.12 and NSC 05 art. 1 (i) and is a reference to the platform, modules, pipeline etc. which the contractor shall construct and deliver, and/or install. In NTK 15 the contract object and the documentation are together defned as ‘Deliverables’; see art. 1.14.

DOI: 10.4324/9781003206798-9

153

erik brannsten An ofshore project will normally be divided into several parts, each covered by contracts with work executed in sequence and/or in parallel, with interfaces between each other. Under each major project contract, there will be a value chain forming a pyramid of subcontractors, sub-suppliers, sub-subcontractors and so on. Subcontracts are often entered into on “back-to-back”3 terms, so that the rights and obligations between company and contractor are refected in every subcontract tier. As we will come back to, there is a strong incentive for the Company to take out one project insurance where all contractors and suppliers are co-insured, which, in turn, will from an insurance perspective remove the interfaces established by the various project contracts. 1.1 Indemnifcation, liability and insurance As pointed out above, construction activities and execution of a development project will involve a risk for loss and or damage to property and/or personnel, as well as pollution and damage to the environment. This risk stems from the complexity, value, level of activity, and duration of such projects. A loss or damage could be the result of negligence or another legal basis for liability, such as the Company’s liability under § 10–9 of the Petroleum Act.4 In addition, due to the material values and number of persons involved, any accident has the potential to incur a loss greater than a company can manage fnancially, which risks the completion of the project on time, should the contractor go bankrupt. In addition to the incentives mentioned, due to the inherent risks, the Petroleum Regulation5 § 73 also requires that the Company procures certain insurances related to its activities: Section 73. Insurance The activities conducted by the licensee pursuant to the Act Chapters 3 and 4 shall be insured at all times. The insurance must at least cover: a) damage to facilities, b) pollution damage and other liability towards third parties, c) wreck removal and cleanup as a result of accidents, d) insurance of the licensee’s own employees who are engaged in the activities. The licensee shall ensure that contractors and subcontractors engaged in the activities take out insurance for their employees to the same extent as the operator insures his own employees. When taking out insurance as mentioned in the frst paragraph literas a) to c), the licensee shall provide reasonable insurance cover, taking into consideration risk exposure and premium costs. Insurance as mentioned under litera d) shall be taken out as further agreed with the organizations of the employees. The Ministry may consent to the licensee using another form of security arrangement. At the end of each calendar year, the licensee shall inform the Ministry about existing insurance agreements, with an indication of the main terms. The Ministry may require further insurance to be taken out. 3 See Knut Kaasen Tilvirkningskontrakter med kommentarer til NTK 15 og NF 15, Universitetsforlaget 2018, Kaasen 2018 (‘Kaasen 2018’) pp. 39 and 235 with further references. 4 Lov om Petroleumsvirksomhet av 29 November 1996 nr. 72 (Petroleum Act). 5 Forskrift til lov om petroleumsvirksomhet 27 June 1997 nr. 653 (Petroleum Regulation).

154

the obligation under norwegian offshore contracts As indicated above, the Company’s legal insurance obligations are extensive, and the Company has a strong incentive to procure such insurances at minimal cost while at the same time fulflling its legal obligations. To achieve this, the Company should avoid overlapping insurances within the project and reduce the amount of liability insurance. On this background the standard ofshore contracts contain reciprocal indemnities and waivers of liability which in detail clarify how personnel and property losses shall be distributed between the participants in the project, rather than containing provisions distributing potential liability. The knock-for-knock arrangement intends to displace ordinary tort liability and will thus deviate from the rules of liability found in the Norwegian background law. This way each project participant will be able to calculate its needs for insurance coverage and procure its insurances accordingly. Such arrangement is often referred to as a “knock-for-knock” agreement. 1.2 Project all risk insurances The insurances used in connection with a knock-for-knock agreement are commonly known as a Project/Construction/Contractor/Builders All Risk (CAR/BAR) policy. A main feature of such insurances is that they cover the whole project as such on an all-risk basis. An all-risk CAR insurance represents the best cover that the insurance market can ofer and covers all causes for damage, unless an exception is stated in the policy. There are several advantages of this approach. First, the total insurance costs over the project period will be reduced. If all involved parties and participants insured their liability, personnel, and property, then there would be extraneous overlapping coverage. This is unnecessarily expensive and could increase the number of insurance disputes in the project, since it would require discussion between insurers about which insurance policy would cover a loss. Second, when determining where the loss or damage has occurred, claim settlement becomes more efcient, as time-consuming discussions and expensive examinations of liability and causation will be avoided. Traditionally, project all-risk insurances are taken out through the London market through use of international insurance brokers such as Aon, Marsh and Willis Towers Watson. A standard set of insurance terms which are often used as a starting point is the WELCAR (Ofshore Construction Project Insurance Wording), which was developed in 2001.6 WELCAR 2012 wording was fnalised and released for consultation in the market, but work stopped and a new version has still not been fnalised.7 WELCAR was originally developed under English law, however, when used for projects on the NCS the choice of law is changed to Norwegian law, which means that practices evolved under English law may be relevant when applying the policy in Norway.

6 See Kaasen 2018, note 3 above, pp. 774–776. 7 Paul Wordley & Jonathan Bruce, ‘Revision to the ofshore construction policy’ (July 2012) Middle East Insurance Review at www.hfw.com/downloads/HFW-MEIR%20Article%20-%20Ofshore%20Construction%20Policy%20[A4%204pp]%20August%202012.pdf.

155

erik brannsten 2 The knock-for-knock arrangement 2.1 Risk zones A knock-for-knock agreement implies that each party must carry the loss or damage which strikes his own so-called risk zone. In broad terms the four risk zones normally used in the ofshore contracts are: Art. 29 – the contract object and materials prior to being delivered; Art. 30.1 – the property and personnel of Contractor group;8 Art. 30.2 – the property (except for the contract object prior to delivery) and personnel of Company group;9 Art. 30.3 – genuine third parties which are not involved in the project. It is a condition under the ofshore contracts that the legal entity or person of the Contractor group is “participating in the Work” and that the legal entity or person of the Company group is “involved in the [feld name] project”. This requirement corresponds with the scope of the insurances, ref. below. For the ofshore contracts, the Company group defnition is based on a so-called extended family, where the “group” includes the Company, his afliates, other licence group members, their employees and so on, and the Company’s other contractors, their subcontractors, employees and so on. This as opposed to a so-called nuclear family solution, where the Company will only indemnify for losses sufered by himself, his afliates, other licence group members, their ofcers, employees and so on. The efect of using the extended group defnition is that the genuine third-party group is correspondingly reduced to include only parties which are not part of the project, such as fshermen, ofcial inspectors, guests and visitors. This will in turn reduce the need for liability insurance. 2.2 Waivers of liability, indemnifcation and waiver of subrogation To make sure the knock-for-knock agreement will work, the arrangement must further include: 1 That the parties waive their right to hold the responsible (or causing) party liable for losses or damage sufered; 2 That they also waive their right to subrogation in case they have been held liable by a third party within their own group; 3 That a party must indemnify the other (causing) party for loss or damage sufered by anyone within his own group, in case such third party directs his claim against the other (causing) party.10 2.3 Damage to the contract object With the high focus on safety in the industry, losses or damage to property and personnel rarely occur. It is more likely that the contract object is damaged as part 8 Defned in NF/NTK art. 1.16. 9 Defned in NF/NTK art. 1.21. 10 The system is illustrated in Kaasen 2018, note 3 above, pp. 770–772.

156

the obligation under norwegian offshore contracts of the construction process or during installation, testing activities and so on. As mentioned above, the contract object constitutes a separate risk zone, and the risk for loss or damage passes from Contractor to Company upon delivery according to the ofshore contracts (art. 19.1).11 Delivery occurs when the delivery protocol is concluded, which may not necessarily correspond to the Company’s physical takeover of the contract object. As mentioned above, it is the Company’s main objective to receive timely delivery of the contract object. Should a loss or damage to the contract object occur, it is therefore agreed in art. 29.1 of the ofshore contracts that the Contractor will immediately begin to repair, rebuild, and complete the contract object – and in the case of a total loss, even start the construction of a new contract object. This obligation applies regardless of the cause of the loss or damage. A well-known total loss incident on the NCS happened when the substructure for the Sleipner A platform collapsed during testing on 23 August 1991. The occurrence was covered by insurance and eventually led to a massive insurance and subrogation claim.12 Since the collapse happened before delivery, the occurrence was covered by art. 29.1 of the contract and the Contractor had to build and complete a new platform sub-structure. Completion of the Sleipner A platform was essential for Norwegian gas sellers to commence export of gas to the European market in 1993 according to the Troll gas sales agreements.13 Distribution of economic responsibilities between the Company and the Contractor is regulated by art. 29.2. As a starting point, the Contractor is economically responsible for the repair or replacement costs, unless the loss or damage was caused by someone in the Company group or by an event which it is not possible to insure against under a CAR policy. Art. 29.2 specifcally mentions nuclear damage, war and terrorism, which would require separate and more expensive insurances. Furthermore, Contractor’s fnancial liability for repair and replacement costs will be limited to a certain amount, often corresponding to the deductibles under the CAR insurance or as agreed between the parties. A precondition for this limitation is that the CAR covers the loss or damage in questions. If the policy does not provide cover, for instance because the Contractor wilfully has caused the damage, the said limitation will not apply. However, the limitation will provide protection if the policy does not apply, and this is the “result of circumstances for which Company carries the risk”, for instance if the Company has failed to pay the insurance premium. The latter may also be the case if the policy provides sub-limits or other carve-outs and limitations which the Company has chosen, unless the ofshore contract stipulates that the Contractor will be liable for such costs which the policy does not cover. In this respect, it is also important that the Contractor was able to review the policy prior to calculating and submitting his tender (see the principle applicable for frame

11 See art. 29.1. For sale of goods the same follows from Section 13 frst paragraph of the Sale of Goods Act. 12 Wikipedia, ‘Sleipner A’ (19 August 2021) at https://en.wikipedia.org/wiki/Sleipner_A. 13 See Christian Lund: “Sleipner Øst – Fra totalhavari til Nasjonal dugnad” in Industribygging og Rettsutvikling, Juridisk festskrift i anledning Norsk Hydros 100 års jubileum, pp. 91–103.

157

erik brannsten agreements in art. 8.4). This is also often done by using a “dummy” policy at the tender stage (see section 3.1). 2.4 Waivers and indemnities shall apply regardless of cause The waivers and indemnities of the knock-for-knock arrangement shall in principle apply regardless of how the loss or damage was caused. This is often expressed such that the indemnity “applies regardless of any form of liability, whether strict or by negligence, in whatever form, on the part of Company Group.”14 In general, Norwegian courts are likely to accept limitations of liability and waivers which are part of “agreed documents” negotiated by equal parties, which are reciprocal, clear and predictable and which refect a deliberate allocation of risk backed up by available insurances.15 Moreover, it is not likely that the agreed limitation will be upheld in case of wilful misconduct (forsett). Nevertheless, it is still an open question as to how the Norwegian Supreme Court will rule if a loss or damage has been caused by gross negligence (grov uaktsomhet) by the responsible party himself or anyone belonging to his higher management. An example from the courts is found in a judgment by the Gulating Court of Appeal from 2014.16 The case concerned the distribution of liability following a collision between the FSO17 vessel Njord B, owned by the Njord licensees, and the ofoading tanker Navion Hispania, owned by Teekay and chartered by Statoil ASA (which was not participating in the Njord licence). The Navion Hispanie was chartered to lift and transport Njord oil which Statoil had purchased. In this case the court found that the shipowner Teekay had caused the damage to the Njord B vessel through gross negligence and therefore concluded that the limitation of liability was suspended vis-à-vis Teekay, who became responsible towards the Njord licensees for “any loss”. 3 The ofshore project insurances 3.1 Introduction As discussed previously, the agreed indemnities (NTK 15 arts. 31.1 and 31.2) clarify which insurances related to the construction project that the Company must procure and maintain, and which insurances Contractor shall purchase and maintain, throughout the project.18 According to the agreed distribution of risk and losses, the Company’s insurance policy shall include

14 See NTK art. 30.1 para. 1 i.f. and the reciprocal provision in art. 30.2 for the indemnity provided by Company. 15 See the Supreme Court decision in Rt. 1994 p. 626 which upheld a limitation of liability in relation to a damage caused by gross negligence of an employee of the forwarding agent against whom the claim was directed. The employee was not considered part of the management of the forwarding agent. See also the discussion in Kaasen 2018, pp. 776–785. 16 Case 2013–06–25. LG-2012–077280. 17 Floating Storage and Ofoading vessel (lagerskip). 18 See also Section 73 of the Petroleum Regulations (F27.06.1997 nr 653).

158

the obligation under norwegian offshore contracts 1 Coverage for physical damage to the contract object, materials, Company’s materials; 2 Transportation insurance (art. 31.1 (b)); 3 Insurance for (so-called genuine) third-party liability up to NOK 500 million (art. 31.1 (c)). Accordingly, the Contractor is obliged to provide insurance cover for his own personnel, property and liability up to certain limits. The Company’s CAR insurance will often provide an exclusion for the items or costs which, according to the ofshore contract, fall within the Contractor’s obligation to insure.19 Contractor’s obligations are as: 1 All risks, hull, and machinery insurance for vessels; 2 P&I insurance, including oil pollution insurance for vessels; 3 Liability insurance covering Contractor’s liability for damage to property and personal injury under art. 30.3 for (so-called genuine) third-party liability, to the extent such liability has not been covered by the insurance for vessels (ref. above); 4 Personnel insurance which shall cover losses connected with illness, personal injury, or accidental death in Contractor Group. The insurance cover described above is closely linked to the various risk zones of the ofshore contracts as mentioned in section 2.1. Company’s CAR insurance refects the contract requirements in art. 31.1 and usually has two sections: 1 Section I contains the construction all risk and transportation. 2 Section II contains third-party liability insurance. It is common to submit a draft version of the CAR Project Insurance in the tender documents as part of a competition so that the bidders can evaluate what the Company’s policy covers, what the deductibles are, the exclusions, limitations, carve-outs, duration and so forth. Deductible levels can vary greatly depending on market conditions, risks in the project, where the damage occurs and so on. Damage to a subsea cable discovered before transportation and installation will often have a much lower deductible compared to damage occurring during installation or testing of the cable after it has been installed. It is important for the Contractor to be aware of these limitations so that he can evaluate the need to buy additional insurance. The same applies for the duration of the CAR. The intention is that the insurance shall be valid from start of the work until the acceptance certifcate is issued, which means that the CAR will be efective even during the guarantee period of two years from delivery (in the policy this will often correspond to the maintenance period). However, in the NF/NTK 15 versions the duration is linked to the actual duration 19 For instance, as follows: “Section I of this Policy will not respond in respect of: Loss of or damage to contractor’s, sub-contractors’ and suppliers manufacturing facilities, also excluded hereunder are tools or equipment intended for use in other projects’ works, unless the above-mentioned items are specifcally identifed and the value is included in the declared Estimated Contract Price or specifed with a sub-limit in the Schedule.”

159

erik brannsten of the policy which can be shorter, for instance, if the Company decides to only buy one year of discovery and maintenance. This will also typically be something a tenderer should check when calculating his bid. Further, according to art. 31.1 (a), the Company procures a Project Insurance which shall include the Contractors as co-insured and a waiver of subrogation from the insurers (see art. 31.1 third para.). If the insurance company could subrogate against the causing party, the knock-forknock arrangement would not be efective (see section 2.6). In the CAR policy, Company will be nominated as the “Principal Assureds” and the Contractors and/or Sub-Contractors and/or Manufacturers and/or Suppliers as “Other Assureds”, thus implementing the requirement of the ofshore contracts for the Contractor Group to be co-insured. Welcar 2001 also contains a waiver of subrogation which ensures that a loss will be covered by the insurers and cannot be claimed from the party causing the loss or damage: Underwriters agree to waive rights of subrogation against any Principal Assured(s) and/ or Other Assured(s). This waiver is not to apply to an Assured in the event that the said Assured commits an act of gross negligence or wilful misconduct.

The assureds will be covered for their own negligent acts.20 However, to make the waiver of subrogation compliant with the requirements of the ofshore contracts, the last sentence of the waiver of subrogation clause should be deleted. However, as pointed out above, it is not likely that the waiver will be upheld for any damage caused by willful misconduct of the assured. As mentioned in section 2.7, the situation is not clear for damage caused by gross negligence by the assured. 3.2 Insured perils Welcar section 1 provides an all-risk coverage, expressed as follows: Subject to the terms, conditions and exclusions herein, this Policy insures against all risks of physical loss of and/or physical damage to the property covered hereunder, provided such loss or damage arises from an Occurrence within the Policy Period set out in Item 3 of the Declarations. [emphasis added]

“All risk” implies that any cause of loss or damage is covered unless the insurer can prove that there is an exclusion stated in the policy. There are very few exclusions provided in Welcar, normally limited to damage due to war, nuclear damage, or terror. This is refected in NTK 15 art. 29.2, frst para. However, there are buy-back options available. In addition, there are limitations, sub-limits and exclusion clauses to be aware of.21 Insurance is taken out for property located anywhere in the world,22 whether under construction or not. Further, such property must be part of Company’s 20 See also Skadeserstatningsloven (Skl) § 4–3 and Forsikringsavtaleloven (Fal) § 4–9. 21 Two types of exclusions for defective design or workmanship have been developed and are commonly used in CAR policies: (1) the Design Exclusions (DE) written by the London CAR Group and (2) the London Engineering Group (LEG) exclusions. 22 For certain infrastructure projects in Norway, it is common that the CAR policy is geographically limited to Europe.

160

the obligation under norwegian offshore contracts investment and included in the declared contract values stated in Section B of the policy. Property covered will include the contract object, goods, materials, equipment, temporary works and so forth. Welcar expresses the cover as follows: This insurance covers works executed anywhere in the world in the performance of all contracts relating to the Project including (provided they are included in the contract values declared to Underwriters and insured herein) materials, components, parts, machinery, fxtures, equipment and any other property destined to become a part of the completed project or used up or consumed in the completion of the project. This insurance shall also cover (provided they are declared to and agreed by Underwriters) all temporary works, plant, equipment, machinery, materials, outfts, and all property associated therewith, whether such items are intended to form a permanent part of the works or not, including site preparatory work and subsequent operational risks. It is understood and agreed that any insured equipment and/or property that is not for incorporation into the contract works shall be covered whilst it is being utilized in the Project and whilst in transit from the Project site(s) until the earlier of the date of arrival at its fnal destination or the 30th day after its removal from the Project site(s).

The policy provides coverage for “physical loss of and/or physical damage” only. Such loss or damage could typically be caused by fres, collapses, breakage, fracturing, water leakages, electrical short circuits, pollution, spills, fallen objects and so on. A general requirement is, of course, that the damage must be unexpected, sudden and accidental and not be the result of an inherent feature of the materials applied. If, for instance, an environmentally friendly paint is used on the platform and the paint starts to peel after one year, as expected according to the specifcations of the paint, then the situation is neither a guaranteed matter nor a matter for insurance. Design errors by themselves are not covered by the CAR insurance. But if a physical loss or damage has resulted from a design error, then such damage will be covered. If, for instance, a load-bearing structure has been under-dimensioned, and this is discovered before any sign of collapse or physical alteration has materialised, then any replacement or reinforcement will not be covered by the CAR. On the other hand, if the beam breaks, insurance will cover the repair costs. The same is the case for construction errors. If a beam has been installed wrongly, for instance, it was not properly levelled or the contractor used carbon steel when he should have used stainless, then any repair or replacement work will not be covered by insurance. If, however, the steel beam was not dimensioned properly and has suffered cracks, then this will be a damage covered by the insurance. Excluded from the standard coverage is the defective part itself. If, for instance, the defective wiring of an electrical transformer in a machine has started a fre, then it could be argued that the cost of the transformer will not be covered by insurance, but the machine and its damaged surroundings will be covered. On the other hand, the cost of the faulty part may not be big compared to mob/demob cost, access costs and installation costs which will be incurred. In many instances it will not be clear what constitutes the defective part itself despite it being a defned term of the CAR. The defnition reads: “any part of the subject matter insured which is or becomes defective and/or unft or unsuitable for 161

erik brannsten its actual or intended purpose, whether by reason of faulty design, faulty materials, faulty workmanship, a combination of one or more thereof or any other reason whatsoever.” It should be possible to obtain a faulty part buy-back, but this may be costly. 3.3 Basis for recovery The ofshore contracts provide that the CAR insurance shall provide coverage for physical damage to the contract object and materials, which is not very precise. Sometimes it will be stated that the Company will provide insurance coverage according to the attached draft insurance policy. To investigate which costs the policy actually covers, the CAR policy draft attached to the tender documents must be studied carefully. As a starting point, no coverage exists for costs incurred if no physical damage has occurred (see above concerning pure design faults or errors in workmanship). This is sometime expresses such that “any portion of the Property Insured shall not be regarded as damaged solely by virtue of the existence of any defect of material workmanship design plan or specifcation.” Likewise, the insurance does not provide cover for costs incurred by the assureds to improve the original design, plan, specifcation or materials. For instance, the insurance incident may have revealed that a load-bearing structure was insufciently designed and that a much more robust structure should have been used instead. Insurance will not cover the gap between the cost of the original design and the cost of the reinforced design. This is often expressed as follows: “The cost of any alterations, additions or improvements shall not be recoverable under this Policy unless specifcally agreed hereunder.” On the other hand, the policy may contain a clause covering increased costs of unbuilt works, where “the cost of the permanent or temporary works not commenced at the date of the physical loss or damage shall exceed the cost which would have been incurred but for the physical loss or damage.” This would typically cover extra costs incurred if later work would need to be performed out of sequence, during winter instead of summer and so forth. Further, the CAR insurance does not cover maintenance costs or costs due to gradual deterioration. When the CAR insurance responds to a physical loss or damage, the basic cost recovery principle for items repaired or replaced is “new for old”, expressed such that the assured will be compensated “the cost of repairing or replacing or re-erecting the property damaged with new materials of like kind and quality.” In case of a total loss the recoverable amount will be “the cost of reconstruction or replacement of the Property Insured by property substantially the same as, but not more extensive than, that lost or destroyed.” The replacement or reconstruction cost will also apply where the cost of repair is higher. The policy will also respond where damage is not repaired, and the recoverable cost will be “the actual value of the property immediately before the occurrence.” Certain costs related to the loss or damage will also be covered such as costs of extra testing, leak search costs, frefghting expenses, wreck and debris removal and clean up, cost of temporary repair, storage, access costs and so on. Nevertheless, for any single physical damage, recovery will normally be capped at 125% of the latest insured values which are given in Schedule B of the CAR policy. 162

the obligation under norwegian offshore contracts 4 Summary As seen in this chapter, the Company has extensive insurance obligations which are sought to be fulflled by the taking of project all-risk insurances. The insurance policy is a complex document which is not clear on all aspects. In addition, some questions, for instance, regarding limitation of liability have not been solved in background law. Even though the Company shall insure the contract object, the content of the insurance policy and the background law determines the actual cover and protection extended to the Contractor. Thus, the Contractors will need to investigate thoroughly what the limitations, sub-limits, deductibles, durations and exclusions of the CAR policy are, to assess its own risk exposure and its potential need to procure additional insurance.

163

CHAPTER 8

Liability and insurance clauses in contracts for vessel services in the Norwegian ofshore sector The knock-for-knock principle Trine-Lise Wilhelmsen 1 Introduction This chapter covers the Norwegian ofshore sector’s liability and insurance clauses in contracts for vessel services. In this chapter, “contracts for vessel services” means ofshore charter parties and contracts for drill and well services. The contracts referred to in this context are the Supplytime 051 and the OLF2 Proposal “New Conditions of Contract for Drilling and Well Services.”3 A main feature of such liability and insurance clauses is that they establish a systematic liability and insurance system in all contracts that involve a particular project, whereby the risk of damage is fnanced by insurance efected by the contractual party sustaining the damage. The principle that damage should stay where it occurs is called the knock-forknock principle. Though the knock-for-knock system frst and foremost regulates damage to the contractual parties, the liability system may also regulate damage to third parties. The knock-for-knock principle is easy to establish in the contractual relationship between two parties. The parties to a contract are free to regulate the risk of causing damage to each other, including both limiting liability for damage caused to the other party and waiving the right to claim damages from the other party. However, this freedom of contract only applies to agreements between the parties concerning damage to economic interests held by the two parties, that is to say, damage to their goods, loss of income and other losses sustained by those parties. The parties to the contract do not have contractual freedom to regulate the tort position of an injured third party. This position is regulated by tort law governing the relationship between the injurer and the victim, and this tort law position cannot be departed from by a contract to which the victim is not a party. If the knock-for-knock principle is to be extended to apply to damage to personnel employed by the contractual parties, this must therefore be achieved through indemnity and subrogation clauses. The same is true if the principle is to be extended to apply to subcontractors or other cooperative parties on each side 1 Supplytime 2017 Time Charter for Ofshore Service Vessels (Supplytime). 2 Oljeindustriens Landsforening, today Norsk olje&gass. 3 OLF Proposal ‘New Conditions of Contract for Drilling & Well Services’ (Approved by the Development & Operations Committee 26 February 2003) www.norskoljeoggass.no/contentassets/5cd867185b5c4 7fc85720d3d96a6f399/drilling-and-well-services.pdf. Some references will also be made to the Norwegian Fabrication Contract 2015 (NF), which contain a similar regulation, cf. www.norskindustri.no/dokumenter/ leveringsbetingelser/nfntk-standardkontrakter/.

164

DOI: 10.4324/9781003206798-10

liability and insurance clauses in norwegian offshore sector of the contract. The regulation of liability in the knock-for-knock principle is therefore constructed partly as liability clauses and partly as indemnity and recourse clauses. Much has been written about the knock-for-knock principle in Norwegian law.4 The main focus of this chapter is to discuss the validity of the clauses. In order to do that, however, it is necessary to provide an overview of the main content and structure of the clauses, as well as the rationale behind them. Furthermore, a brief presentation of the relevant tort law and insurance law framework is also necessary. 2 Overview of the relevant tort law and insurance legislation 2.1 Tort law According to Norwegian tort law, a person is liable for damage to another person if three conditions are fulflled: there must be a legal basis for liability, there must be an economic loss, and there must be legally relevant causation between the act or omission of the injurer and the loss. These conditions are mainly developed in court practice and are not stated in general legislation. There are three main types of the basis for liability: liability for negligence, strict liability, and vicarious liability. The main rule is that negligence is required to invoke liability for damages. Strict liability will normally require legislative action. Vicarious liability is regulated by the Compensation Act5 § 2–1, which states that an employer is liable for any damage caused by negligence or a deliberate act from their employees. This is therefore a combination of strict liability and negligence. However, in several areas there is legislation providing for non-fault liability. Of interest here is the regulation of petroleum and maritime legislation. The Petroleum Act6 § 10–9 provides for extended liability of the licensees operating on the Norwegian Continental Shelf: If liability in respect of a third party is incurred by anyone undertaking tasks for a licensee, the licensee shall be liable for damages to the same extent as, and jointly and severally with, the perpetrator and, if applicable, his employer.

Thus, if a contractor performs drilling services for a licensee on the Norwegian Continental Shelf and it causes damage while performing the services, the licensee is jointly and severally liable with the contractor. The liability of the contractor is regulated by ordinary tort law as outlined above. Furthermore, both the Petroleum Act and the Maritime Code7 contain rules on strict liability for pollution. A general feature of these rules is that the licensee or shipowner has strict liability for pollution8 and that the liability is specifcally directed against these persons so that claims may not be raised against other persons. Further the licensee or 4 Hans Jacob Bull, Tredjemannsdekninger i forsikringsforhold, Oslo 1988, Del IV, (Bull); Knut Kaasen, Petroleumskontrakter, 2018, Del VIII (Kaasen); Monika Zak, Ansvarsregulering i borekontrakter – Gyldighetssensur i norsk, engelsk og amerikansk rett, Master Paper 19.05.2012 www.duo.uio.no/handle/10852/35075 (Zak). 5 Norwegian Compensation Act (Skadeserstatningsloven) 13 June 1969 no. 26. 6 Petroleum Act 1996 no. 72 (PA). 7 Maritime Code 1994 no. 39 (MC). 8 PA § 7–3, MC § 191.

165

trine-lise wilhelmsen shipowner, having settled the claim, may not raise a subrogated claim against the persons protected against claims, unless any such protected person acted deliberately or with gross negligence.9 However, even if the charterer is protected against a pollution claim from a third party, he is not protected against a claim for subrogation.10 The licensee, on the other hand, may not claim recourse against anyone who by agreement with the licensee or his contractors has performed tasks or work in connection with the petroleum activities.11 The starting point is therefore that the licensee and contractor are jointly liable for damage caused during operations on the Norwegian Continental Shelf, and that liability for pollution is specifcally channelled against the licensee and the shipowner. 2.2 Insurance law Norwegian insurance contracts are governed by the Insurance Contract Act (ICA).12 ICA’s provisions are generally mandatory; however, there are several exceptions for insurance relating to commercial activities.13 The relevant exclusions are for ships and ofshore units and internationally transported goods, which includes transport to and from the Norwegian Continental Shelf.14 As a starting point, therefore, the insurer and parties to the charter or drilling contracts have full contractual freedom in relation to the content of the insurance contract. Nordic marine insurance is regulated by the Nordic Marine Insurance Plan 2013 Version 2019 (NP).15 The NP is an agreed document which may be compared to private legislation, and it is used for most marine insurance contracts in Norway. The NP contains conditions for, inter alia, hull and loss of hire insurance for ocean-going vessels and ofshore units. Hull insurance includes liability for collision, but ordinary liability insurance is regulated by Protection and Indemnity insurance contracted in the P&I Clubs. Of particular relevance here is that NP chapter 18, which regulates insurance for mobile ofshore units, has special rules on subrogation and co-insurance which are relevant for the knock-for-knock regulation. 3 The content and structure of the knock-for-knock principle 3.1 Type of loss and basis for liability Both Supplytime and the OLF Proposal apply to the following groups of damage and loss: 1 Damage to property, including the vessel;16 2 Personal injury or damage;17

9 PA § 7–4, MC § 193. 10 MC § 193 (c). 11 PA § 7–4 (a). 12 Act no. 69 of 16 June 1989 relating to insurance contracts. 13 ICA section 1–3 frst subparagraph. 14 ICA section 1–3 second subparagraphs (c), (d). 15 The Nordic Marine Insurance Plan of 2013, Version 2019, at www.nordicplan.org/The-Plan/. 16 Supplytime 14 (a) (i) and (ii), OFL proposal 8.1 (b) and 8 (2) b. 17 Supplytime 14 (a) (i) and (ii), OLF proposal 8.1 (a) and 8.2 (a).

166

liability and insurance clauses in norwegian offshore sector 3 Consequential damage;18 4 Pollution damage.19 In the OLF Proposal, the principle also applies to loss of or damage to in-hole equipment, loss of hole, blowout, damage to reservoir and use of radioactive tools, and infringement of patents/property rights.20 However, even if the core of the knock-for-knock principle is that each party carries the damage that it has sustained, some losses will nonetheless be channelled against one of the parties or divided between the two parties. This may be necessary to comply with the mandatory pollution regulation.21 The knock-for-knock regulation applies regardless of the basis of liability that may be invoked against the injurer, i.e., it applies to both strict liability and negligence. This means that the regulation applies to liability which is based on either the licensee’s contractor’s liability according to PA § 10–9, on strict liability for pollution according to the petroleum or maritime regulation, or on negligence. Furthermore, the knock-for-knock regulation applies to ordinary negligence, gross negligence, and damage caused by intent. In addition, there is no distinction between faults made by an employee and faults made by a company’s representative. 3.2 Who is included in the liability provisions – “the group concept” The standard liability provisions frst and foremost apply to the parties to the contract. In the charter party, this will be the “Owner”22 and the “Charterer”.23 In the OLF proposal, the parties are the “Contractor”24 and the “Company”.25 However, the standard liability provisions also apply to other parties. Firstly, the provisions address both property damage and personal damage sufered by persons employed by the parties to the contract. Secondly, the provisions apply not only to the parties to the contract, but to others who are in a contractual relationship with these parties. In the OLF Proposal, this follows directly from the clause’s wording, where the Contractor and Company shall “indemnify” not each other, but instead the “Company Group” and “Contractor Group”.26 In Supplytime, the result is more indirect as the clause only states that the Owner and Charterer shall not be liable for damage to the Charter Group/Owner Group, but where according to a Himalaya clause this freedom from liability shall apply also to the beneft of other members of the respective Groups.27 18 Supplytime 14 (b) (ii), OLF proposal 8.10. 19 Supplytime 15 (a) and (b), OLF proposal 8.4 and 8.5. 20 OLF proposal 8.6–8.9. 21 According to OLF 8.4, the Company is liable for pollution from reservoir and property of the company, whereas the Owner according to Supplytime 15 is liable for all pollution from discharge, spills or leaks from the vessel. 22 The Owner is the party stated in Box 2, Supplytime Part II Defnitions. 23 The Charterer is the party stated in Box 3, Supplytime Part II Defnitions. 24 The “Company” means X as operator on behalf of one or more Licence Groups as specifed; cf. OLF proposal General conditions 1.1. Defnitions. 25 The Contractor is identifed by name, cf. OLF proposal General conditions 1.1. Defnitions. See similarly NF Part 1 Art 1.14 (contractor) and 1.22 (Company). 26 OLF proposal 8.1 and 8.2, NF Art. 30.1 and 30.2. 27 Supplytime 14 (a) (i) and (ii); cf. 14 (d).

167

trine-lise wilhelmsen The “group concept” in this context is used to defne risk zones, which are zones for which each party in the contract carry the tort risk. The point here is that the Owner/ Company or Charterer/Contractor not only agrees to be responsible for any damage that befalls the company property or the property or persons of the employees, but also assumes responsibility for such damage throughout the group. The extent of the risk zones has varied over time,28 but both the Supplytime and the OLF Proposal apply a broad group concept, or a “big family group” concept.29 Though the defnitions vary, the primary idea is that it consists of the contractual party and the parties with whom they cooperate on a particular project, including all the employees of such third parties. The “Owner Group” means the “Owners”, “Owners’ Afliates”, “contractors and subcontractor” and “Employees of any of the foregoing”.30 The Company Group means the Licensee Group, each of the participants herein, their afliated companies, the Company’s other contractors and their contractors or subcontractors, the Company’s invitees, and personnel employed in or engaged by the aforementioned corporate entities, and others whose services are used by the Company.31 The Charterers Group is “the Charterers and Charterers clients”. “Co-ventures of the foregoing”, “Afliates”, “contractors and subcontractors” and Employees of any of the foregoing”.32 The contractors Group is the “Contractor, Contractor’s Afliated Companies participating in the Work, its Subcontractors and their contractors and subcontractors, participants in a joint venture or similar partnership involved in the Work, Contractor’s invitees, and personnel employed in or engaged by the aforementioned cooperate entities.33

In the contract between Owner/Company and Charterer/Contractor, the parties may agree to waive the right to make a claim against a third party for liability. This could be for damage to their property or person. This constitutes a promise not to make any claim against the named third party. However, the parties may not, through their contract, require other members of their respective groups to accept a similar waiver of their rights. If matching waiver provisions are not included throughout the contractual chain, a contractual party who has not agreed to a knock-for-knock provision may refuse to accept responsibility for damage to his property and personnel and may instead make a claim against the injurer. This will disrupt the system and can easily lead to a chain of subrogated claims.34 To obtain an overall knock-for-knock governing provision in all the contracts relating to a particular project or work, the Company/Owner and Contractor/Charterer must therefore include equivalent agreed liability terms in all their other contracts tied to the same project or work and induce their contractual partners to do the same. To the extent the same contractual terms are used throughout, this should secure a consistent approach to liability. But such “back to back” regulation is not always agreed. NF therefore includes a duty on the Company and Contractor to “ensure 28 Bull p. 333 f., Sofa Lazaridis, Maritime ofshore contracts Compendium, Sjørettsfondet 2011, p. 51. 29 Bull p. 347. 30 Supplytime Part II Defnitions. 31 OLF proposal 1.1 third paragraph. See similarly NF Art. 1.26. 32 Supplytime Part II Defnitions. 33 OLF proposal 1.1 sub-paragraph 10, see similarly NF Art. 1.16. Cf. further on the group principle in drilling contracts Zak p. 21 f. 34 So-called loss carousel; cf. Bull p. 341.

168

liability and insurance clauses in norwegian offshore sector as far as practicable that other companies” in the group agree to waive their right in an equivalent way.35 Supplytime and the OLF Proposal do not include a duty to secure equivalent provisions in other contracts. The aim of the knock-for-knock principle is thus to allocate the risk for liability for all the parties that are involved in the same project. But the project may also result in damage or loss to a third party, that is to say a party outside the mentioned groups. This is not regulated in Supplytime, but the OLF Proposal 8.3 contains the following provision:36 Subject to clause Article 8.4 – Pollution from reservoir and property of Company, Contractor shall indemnify Company Group from and against any claim arising out of loss or damage sufered by a Third Party in connection with the Work, to the extent that any such loss or damage is caused by the negligence or breach of duty (whether statutory or otherwise) of Contractor Group. Subject to Article 8.5 – Pollution from Contractor’s property, Company shall indemnify Contractor Group from and against any claim arising out of loss or damage sufered by a Third Party in connection with the Work, to the extent that any such loss or damage is caused by the negligence or breach of duty (whether statutory or otherwise) of Company Group.

These provisions mean that each party or group is liable for damage caused by their own negligence. As a starting point this division of liability conforms to ordinary contract law. However, as it follows from PA § 10–8 that the licensee (Company) is severally and jointly liable with the contractor for all damage for which the contractor is liable, it means that this risk is transferred back to the contractor. 3.3 Freedom of liability, indemnity and subrogation The starting point is that parties to the contract may only regulate their own duties and rights in accordance with the contract. They are therefore free to limit their liability against a contractual party, and waive their right to claim any liability in tort from this party.37 As mentioned, they are also free to waive their right to claim any liability from a third party and thus waive such right in regard to the whole group. The parties to a contract may not, however, weaken the rights of a third party to a contract. To the extent the third party is within the risk zones as defned in the contracts, and that this party has included a similar knock-for-knock regulation in their contract, this problem is solved through the matching contractual terms. However, this restrictive approach also applies to third parties who do not have a contractual relationship with the Company/Owner or the Contractor/Charterer. This will be the case for all the employees in the groups, and for third parties outside the groups. Thus, if the Owner/Company harms employee E of the Charterer/Contractor or causes damage to E’s property, this damage shall – according to the contract – be compensated by the Charterer/Contractor. However, E does not have to accept that the Charterer/ Contractor shall pay the claim. He may direct his claim to the Owner/Company instead. If so, the knock-for-knock principle is obtained through a subrogated claim from the Owner/Company against the Charterer/Contractor after E is compensated.38 35 36 37 38

NF Art 30.2 and 30.2 para. 2. Third-party liability is also regulated in NF Art 30.3, but here with a diferent model. Bull p. 346; Zak p. 23. Bull p. 347; Zak p. 23.

169

trine-lise wilhelmsen If E accepts compensation from the Charterer/Contractor, this regulated approach further implies that the Charterer/Contractor does not have a right to claim recourse from the Owner/Company, even if the Owner/Company is at fault. This part of the principle is particularly signifcant for work performed on the Norwegian Continental Shelf. As mentioned, it follows from PA § 10–9 that the licensee is jointly liable for any damage caused by a contractor. Thus, any party who sufers harm in Norwegian petroleum activity may always make a claim against both the contractor and the company. If the claim is raised against the party who is responsible according to the knock-for-knock principle, subrogation according to PA § 10–9 is barred.39 The structure of the knock-for-knock principle is therefore a combination of freedom from liability/acceptance of not making a claim, a basis for recourse from the party having paid the claim according to tort law, but who is not liable according to the contract, and a bar to recourse from the party having paid the claim according to the contract even if he was not liable according to ordinary tort law. 3.4 The insurance regulation The regulation of liability is normally supplemented by a regulation of insurance. The purpose of this is partly to secure that the liability risk of each party is fnanced by insurance. Normally, it will be up to each contractual party to what extent he needs fnancial security through insurance. However, if the knock-for-knock principle is put into efect through indemnifcation after having frst paid the claim, it is important for the party having paid the claim in the frst place that the other party is covered by liability insurance which includes liability according to contract. Further, in the case of joint liability, in particular according to PA § 10–9, if the Contractor cannot pay for the damages, the Company will always be liable. Therefore, the contracts will provide the party with a duty to take out proper insurance protection to cover its liability under the contract.40 Further, in order for the liability system to be carried through the contracts as outlined above, it is important that the division of risk is not disrupted by a subrogation claim from the insurer.41 The starting point according to Norwegian law is that the insurer, after having paid compensation for loss or damage, may claim subrogation from the injurer that caused the loss.42 However, it is clear that the insurer does not have a wider right to subrogation than the injured party could have claimed from the injurer.43 If the injured party has agreed to waive his right to claim damage against the injurer, a recourse claim from the insurer is similarly barred. However, this may in turn mean that the insurer’s liability may be reduced by an amount equal to “that which he is prevented from collecting because the assured has waived his right to claim compensation from a third party, unless the waiver may be considered customary in

39 Bull p. 346; Zak pp. 23–24. 40 OLF proposal 8.12 para. 1–4, Supplytime 17 (a), but duty for the Owner only. 41 Kaasen p. 767; Zak p. 33. 42 Trine-Lise Wilhelmsen, Regress i skadeforsikring, Tidsskrift for erstatningsrett, forsikringsrett og trygderett, 2019 p. 7 at pp. 12–13 with references. 43 Bull p. 489; Knut Selmer, Forsikringsrett, Oslo 1982, pp. 349 and 357.

170

liability and insurance clauses in norwegian offshore sector the trade in question”.44 Therefore, in order to protect the position of the insurer and the assured, it is necessary to secure against subrogation in the contract. Two diferent contractual techniques are used in this context. The frst is the most direct, and simply imposes a duty on the Contractor or Owner to require the insurers to waive all rights of subrogation against the Company or Company group.45 Such waiver of subrogation follows directly from NP § 18–9 for insurance of MOUs and probably also from NP § 5–14 for hull insurance.46 The second technique is more indirect: the party not efecting the insurance shall be named as co-insured under the policy.47 Such a right is automatically included in NP clause 18–1 (i) subclause 2, and may be agreed in hull insurance.48 The main content of co-insurance is that a third party with owner interest, security interest or other economic interest in the insured property is insured for this interest under an insurance efected by the “main owner” or “assured”.49 This is of lesser interest here. But the co-insured also has indirect liability protection. This is not regulated directly either in the NP or in the ICA, but it is presumed in the preparatory documents to the ICA that the co-insured as injurer will have the same protection as the assured, if they cause damage that constitutes an insured event under the casualty insurance efected by the assured.50 This protection means that the co-insured has the same protection against the insurer as he would have had as an assured if he had been responsible for causing an insured event through a breach of the so-called duties of due care.51 According to the NP clause 3–33, the insurer may, in cases where the assured causes the damage through gross negligence, reduce his liability from 0% to 100% depending on the degree of fault and circumstances generally. Among the relevant circumstances taken into account in applying this rule will be the professionality of the assured, the risk involved in the activity and the injurer’s options to avoid the risk. A co-insured will then have the same protection against recourse from the insurer as he would have had if he had caused loss or damage under his own casualty insurance. This is called the co-insured’s indirect liability insurance. It follows from this that the protection for the injurer as a starting point is better with a waiver of recourse clause than with a co-insurance clause in cases where the injurer causes losses through gross negligence. 4 The rationale for the knock-for-knock principle 4.1 The need for contractual control of the liability risk The regulation of liability through the knock-for-knock principle is unique for the ofshore sector. Normally, contractual parties do not regulate the risk for tort law 44 45 46 47 48 49 50 51

NP Cl. 5–14; Bull p. 490; Selmer p. 357. OLF proposal 8.12 para. 7, Supplytime 17 (a) (ii). Bull p. 490. In the NP 2013 the provisions are Cl. 5–14 and Cl. 18–1 (i) sub-clause 1. OLF proposal 8.12 paragraph 6, Supplytime 17 (a) (ii). NP Cl. 8–1. NP Cl. 8–1 and Cl. 7–1, and ICA § 7–1. NOU 1987:24 p. 145; cf. pp. 151–152; Bull 2008 p. 539. Bull p. 487.

171

trine-lise wilhelmsen liability and coverage under the other party’s insurance. However, there are three main features in the ofshore sector that makes contractual control of the liability risk important. One feature is that the risk of causing damage during the operations is more substantial than in land-based projects or ordinary transport contracts.52 The ofshore industry has always been a very risky and hazardous business, and this is more accentuated as the upstream oil and gas sector has gradually moved into deeper and more unsafe regions with adverse weather conditions. A second feature is the huge capital sums invested, which easily results in enormous losses if accidents happen. For example, the catastrophic Deepwater Horizon oil spill is a primary illustration. Therefore, it is important that the allocation of this risk is controlled by contractual regulation. A third feature is the involvement of many contractors and subcontractors, which results in several potential injurers and victims.53 The substantial risk for damage and the number of people involved create a need for foreseeability. Without regulation, liability claims will be handled according to the ordinary tort law system. This requires investigation into which party was at fault and could result in costly litigations, causing substantial economic uncertainty.54 For example the Piper Alpha disaster led to claims against 24 diferent contractors. Among those on board the platform who were killed, 134 were employed by contractors and 31 by the operator. Among those who survived, 55 were employed by contractors and 31 by the operator.55 A clearer defnition of risk allocation between the parties would transfer this uncertainty into a risk that may be calculated more accurately. This will also help prevent difcult discussions between contractual parties who are obliged to work together in long-term projects.56 4.2 Efcient insurance coverage The principle may also be explained in terms of insurance coverage.57 Without risk allocation, each party must purchase liability insurance to cover potential liability for damage during the project. This will then be an addition to casualty insurance covering damage to and loss of property and loss of income, and insurance covering personal damage and death of employees. If an accident occurs, this could easily result in overlapping insurances, where property damage loss is covered both by the injured party’s casualty insurance, and also by the injurer’s liability insurance. Insurance is a costly way to fnance risk; and, as with all ofshore sector’s costs, it is a substantial amount. By channelling the risk for damage to the party where the damage occurs, the need for liability insurance and thus the premium for this insurance will be reduced.

52 53 54 55 56 57

Kaasen p. 765. Ibid. p. 765. Kaasen p. 766, Bull p. 353. Caledonia North Sea Ltd v London Bridge Engineering Ltd [2002] UKHL 4; [2002] 1 Lloyd’s Rep 553, HL. Bull pp. 353. Cf. further Bull pp. 349–352; Kaasen p. 767.

172

liability and insurance clauses in norwegian offshore sector The result of this is, of course, a higher risk exposure under the casualty insurance. However, such insurance will normally be purchased in any case because damage may easily occur without anybody being responsible through the tort law system. Typical examples would be damage caused by natural disasters or mistakes made within the insured’s own organisation. Anyone involved in the petroleum sector therefore needs to make a risk assessment as to how best to handle and fnance these risks. The need for casualty insurance is therefore not reduced even if a normal liability regime applies. It is also generally considered cheaper to channel risk to the casualty insurance than to divide the risk between casualty insurance and liability insurance, with a right for the casualty insurer to claim recourse against the liability insurance. Recourse claims are costly to invoke, and casualty insurance is generally cheaper than liability insurance.58 4.3 Loss prevention? 4.3.1 Loss prevention and efcient liability rules A major argument against the knock-for-knock principle is that it is contrary to the considerations of deterrence which are a key reason for regulation of tort liability. It fails to adequately discourage parties to avoid causing damage in other parties’ risk zones because negligence and faults have no consequences. In the Norwegian discussion on this issue in relation to the knock-for-knock principle in the petroleum sector, it is claimed that this consideration is exaggerated. Control over routines, the relationship with public authorities and the parties’ desire to be qualifed to undertake further projects on the Norwegian Continental Shelf are claimed to be more important.59 As a starting point, because there is difculty in measuring how a lack of deterrence afects a specifc situation, it is difcult to have a meaningful opinion on the issue. However, deterrence considerations can be analysed in terms of law and economics, which can explain the signifcance of this consideration. The approach in this model is to defne how diferent rules may infuence the optimal level of care, in order to minimise the sum of costs for preventive measures and damage. This theory has several main presumptions, which includes that the liable party or injurer is acting rationally by seeking to minimise his own costs in relation to damage, that he has full information regarding tort rules, and that the rules are enforced.60 As a starting point, the theory demonstrates three attributes: that no liability is never optimal, that no-fault liability will induce the injurer to choose a level of care that minimises the sum of prevention costs and expected damage, and that liability for negligence will induce the injurer to choose the level of care necessary to avoid liability so that he is only liable for loss prevention costs.61 This model conforms to the traditional thinking on the signifcance of deterrence. However, the conclusions may be diferent when there is a contractual relationship between the injurer and the victim.

58 59 60 61

Bull pp. 349–350. Ibid. p. 355. Erling Eide and Endre Stavang: Rettsøkonomi, 2. ed., Oslo 2018, (Eide and Stavang) pp. 243–244. Eide and Stavang pp. 246–249 and Stephen Shavell, Economic Analysis of Accident Law, 1987, p. 8.

173

trine-lise wilhelmsen 4.3.2 The efcient basis for liability in contractual relationships In the event of a contractual relationship, for instance between an owner and a charterer, the model must be supplemented by an assessment of the charterer’s willingness to pay for the owner’s services. The charterer’s willingness to pay will depend on how he assesses the risk involved in chartering the ship.62 The injurer’s/owner’s inducement to prevent damage will therefore depend both on the risk of damage and the possibility of charging more for the service.63 The model builds on a presumption that the owner maximises his own proft. It is further presumed that he acts in a market with perfect competition, which means that the price of the service provided is equal to the total costs necessary to produce the service, including the costs connected to liability against the customers/ charterers.64 A third presumption is that the risk of damage depends on negligence or non-negligence. With these presumptions, Table 8.1 – showing level of care, cost of taking such precautions, percentage of probability of an accident and expected accident losses – may be used as a basis for discussion.65 The owner’s costs in producing one unit of transport = 10. This amount does not include costs in relation to care or liability against the charterer. Based on these fgures, the optimal level of care may be discussed depending on the charterer’s information about the risk. In the frst case, the charterer has full information about the risk of damage. If the owner does not face liability for damage, he will not have any costs connected to duty of care or expected accident losses.66 In this case, an owner O-1 may choose to sell one unit of transport for 10, which is the production cost. A charterer, who has full information about the inherent risk in the transport, will know, however, that the transport will in fact cost him 10 + 9 = 19 due to the accident costs. Another owner, O-2, chooses to take due care. The cost per unit of transport will then be raised by 2 to 12. At the same time, the charterer’s risk will be reduced to 3. The charterer’s total cost will therefore be 15. A cost-minimising charterer with full information about the risk of damage will select O-2 with total costs of 15 instead of O-1 with total costs of 19. The result is that the less careful O-1 will lose his customers to the more careful O-2, even if no liability is imposed. Table 8.1 The Optimal Level of Care Level of care

Cost of care

Accident probability 

Expected accident losses

None Care

0 2

9% 3%

9 3

62 63 64 65 66

The model is based on Shavell p. 47 f. Shavell p. 47 and pp. 51–52. Shavell p. 47. Taken from Shavell p. 49. Shavell p. 52.

174

liability and insurance clauses in norwegian offshore sector If the owner is facing strict liability, he will be liable for the cost of care and expected accident costs. As 2 + 3 is less than 9, he will choose to take due care. If we presume that the norm for negligence follows the optimal care in the ordinary tort law model, the result is the same for liability for negligence. The result is therefore that the owner will choose to take care regardless of liability for damage.67 In the second case, the charterer does not have sufcient information about the probability of damage to calculate the correct price for the service from diferent service providers.68 Under this presumption, the owner cannot expect the charterer to pay extra because the owner takes care, since the charterer does not know about the probability of damage. Without any potential liability, the owner will therefore not take due care. But if the owner is liable for negligence he will take optimal care in relation to the evaluation of negligence.69 However, a negligence rule corresponding to the optimal level of care presumes that the judge has sufcient knowledge to evaluate the carrier’s activity. If this proves difcult, the evaluation of negligence may well lead to an over- or underestimation.70 In case of no-fault liability, on the other hand, the owner will choose to take due care as long as the costs of taking such care are lower than the potential liability.71 The model demonstrates that if services are provided in a market with perfect competition and full information, the optimal level of care is not infuenced by liability rules. This implies that if the parties to the contracts are professionals and have suffcient volume of activity to establish their own statistics relating to the probability of accidents and accident losses, liability for the damage caused by each party is not necessary to obtain an optimal level of care. At the same time, a rule of liability for negligence will result in transaction costs in relation to the settlements, which must be calculated into the price of the transport.72 To the extent that the ofshore market satisfes these presumptions, it may be argued that the knock-for-knock principle conforms to economic efciency and that a liability regime is not necessary to obtain the optimal level of care. On the other hand, in regard to less professional or smaller market participants with limited knowledge of the risk of damage, the knock-for-knock principle is not defended by this model. 5 The validity of the regulation 5.1 Some starting points Norwegian legislation does not contain a special rule prohibiting parties from freeing themselves from liability for damage, or from waiving their right to claim for damages in tort. Therefore, the general starting point is that the parties are free to agree

67 68 69 70 71 72

Ibid. pp. 52–53. Ibid. pp. 53–54. Ibid. pp. 53–54. Ibid. p. 56. Ibid. p. 54. Trine-Lise Wilhelmsen, Rett i havn, Oslo 2006, p. 337.

175

trine-lise wilhelmsen to such freedom from liability, regardless of the basis for this liability. Further, the main rule is that contracts shall be fulflled as agreed.73 However, there are two general mandatory restrictions that are applicable to contracts. The frst is NL 5–1–2,74 which prohibits contracts that are against law and morality. The second is the Contract Act § 36,75 which states that contracts that are unfair may be set aside partly or in full. Based on these rules, the legal theory assumes that a contractual party cannot avoid liability for damage they cause deliberately.76 Further, it is a general view that liability limitation is accepted for acts or omissions committed by ordinary employees, in contrast to those committed by company leadership. This is also true if the act is made deliberately or with gross negligence.77 What is less certain is the extent to which freedom from liability for gross negligence by the company itself may be valid. This must therefore be discussed based on the two rules mentioned. It is also necessary to address the insurance clauses according to these rules. 5.2 NL 5–1–2 NL 5–1–2 prohibits contracts that are contrary to the law or morality. The rule’s latter portion regarding morality is relevant to the knock-for-knock principle. The expression “contrary to morality” means where the contract is outside generally accepted moral norms.78 The concept of “generally accepted moral norms” can be static or dynamic. If the concept is static, previous Supreme Court judgments would be decisive regardless of their occurrence. There are older Supreme Court judgments79 that implicate freedom from liability for gross negligence by the company is void. Based on these judgments, it is claimed in legal theory that limitation from liability for the company’s own gross negligence is invalid as an absolute rule.80 With this interpretation, the knock-for-knock principle cannot be applied in cases where damage is caused by gross negligence by persons acting on behalf of the Owner/Company or Charterer/Contractor. Expanding on the concept of “general accepted moral norms” also includes a dynamic interpretation. This presumes a reference to the time of the evaluation and that caution in the use of old judgments is required.81 There is some support for this view in the preparatory documents to the Contract Act § 36, which claims that the interpretation of NL 5–1–2 is uncertain in regard to limitation of liability.82 It 73 Kong Cristian Den Femtis Norske Lov av 25. april 1687 (NL) 5–1–1. 74 Ibid. 5–1–2. 75 Act 31. mai 1918 no. 4 on “avslutning av avtaler, om fuldmagt og om ugyldige viljeserklæringer (Contracts Act). 76 Kai Krüger: Norsk Kontraktsrett, 1989 s. 784 with further references, Bull p. 394 and note 151 with references, and Kaasen p. 777. An illustration of this principle may be found in ND 1988.263 “Mørland 7” NA. 77 Rt. 1994.626, Rt. 1948. 370, Rt. 1915.840, ND 1989.225 NA, ND 1991.180 Eidsivating, NOU 1979:32 Formuerettslig lempningsregel (NOU) p. 19, Bull p. 394 and note 152; Kaasen p. 777; Zak pp. 36–37 and p. 41. 78 Zak p. 36 with further references in note 157. 79 Rt. 916.717 and Rt. 1926.712. 80 NOU p. 19 and references in note 1; Viggo Hagstrøm, “Om grensene for ansvarsfraskrivelse, særlig i næringsforhold”, Tidsskrift for rettsvitenskap 4/1996, pp. 464 and 475. 81 Zak p. 36. 82 NOU p. 19.

176

liability and insurance clauses in norwegian offshore sector is also argued that invalidity is merely a guiding principle, where there is room for exceptions depending on the circumstances.83 If it is accepted that NL 5–1–2 only provides a guiding principle that gives room for exclusions, the question is whether such exclusion is justifed for the knock-forknock principle in the ofshore sector. In this context, it is natural to analyse the “morality” aspect of no liability in regard to the reasoning behind the liability rules. Here, the most “immoral” aspect seems to be retribution, i.e., that a lack of retribution in cases of gross negligence by the company is in itself against “morality”. This may be explained by a requirement for corrective justice in the relationship between the parties. But a requirement for corrective justice is difcult to reconcile with the development in the use of liability insurance. Further, it follows from the ICA84 that the liability insurer will also cover liability caused by gross negligence by the assured, which in this case will be the insured company. If the liability is insured, the retribution aspect is therefore reduced to an insurance premium payment. This payment is made before the damage resulting in liability is caused. The only actual retribution will therefore be that the liability insurer may raise the premium for the next insurance period. And if they do, the assured may refuse renewal and enter into a contract with another company. Therefore, there is not much left of corrective justice when the liability is covered by liability insurance. The knock-for-knock principle also means that the contractual partner as a victim does not obtain repair of damage from the injurer. However, the knock-for-knock principle assumes that repair of damage to the victim’s interests is fnanced through casualty insurance covering loss of or damage to property and income, to the extent that such fnancing is needed. Accidents to employees will be separately covered through employment casualty insurance, which is mandatory in Norway.85 Repair of damage is therefore secured through insurance. The last consideration behind liability rules is deterrence. It may be argued that it is immoral not to have rules that are aimed at preventing damage. However, deterrence’s efect on liability must be seen in conjunction with developments in public safety regulations and requirements, and the attitude towards safety issues within companies. This may imply that deterrence through liability is less needed. Furthermore, if it can be demonstrated that liability is not needed because the market will secure optimal care, it is difcult to see why a notion of morality should prevent an efcient development of risk sharing. It may therefore be argued that the earlier view of morality in court practice and legal theory is outdated by recent developments in insurance and security legislation, as well as by the economic models on the deterrent efect of liability where there is a contractual relationship between the parties. However, since there are no decisions from the Norwegian Supreme Court, the conclusion is uncertain. A possible underlying reason for disagreements on this issue is a more general confict between corrective justice as the raison d’être for liability rules, and the legal 83 Lasse Brautaset, “Kontraktsreguleringen ved salg av gass”, Norsk Gassavsetning. Rettslige hovedelementer, Sjørettsfondet 1998, pp. 117–118; Zak pp. 37–38. 84 ICA § 4–9 para. 2. 85 Yrkesskadeforsikringsloven 1989 no. 65 § 3 cf. § 1.

177

trine-lise wilhelmsen and economics-based approach where liability rules are analysed in terms of efciency and optimal care.86 The goal of maximising wealth in the legal and economics-based approach is considered “immoral” because it does not consider distributive and corrective justice.87 But even if such more fundamental considerations of justice are relevant in relation to liability regimes, they appear less useful in a professional and well-organised contractual setting. 5.3 The Contract Act § 36 5.3.1 Overview and some starting points The Norwegian Contract Act § 36 provides: An agreement may be wholly or partially set aside or amended if it would be unreasonable or confict with generally accepted business practice to invoke it. The same applies to a unilaterally binding disposition. When making a decision, account will be taken not only of the contents of the agreement, the position of the parties and the circumstances prevailing at the time of conclusion of the agreement, but also of subsequent events and circumstances in general.

The assessment according to this provision is therefore diferent from that according to NL 5–1–2. A contract may be invalid due to immorality even if it is totally fair between the parties. This may, for example, be the case if the party claiming freedom from liability has paid the other party a fair price for obtaining this right, and this payment is used to buy alternative fnancing for costs of damage through insurance. On the other hand, a contract may be unfair without being immoral, for instance, if a change of circumstances has resulted in an altered level of risks for damage being incurred, making the knock-for-knock agreement unfair for one of the parties. The Contract Act § 36 primarily applies to consumer contracts, but it may also be applied to professional contracts.88 However, the threshold for applying this rule is higher in such contracts than in consumer contracts.89 In the oil and gas industry, it is generally recognised that adjusting and rewriting contracts according to the Contract Act § 36 can only occur under special circumstances.90 However, it follows from the Contract Act § 36’s preparatory documents that a primary reason for establishing this rule was clauses used in the oil and gas sector; namely, the clauses giving the company a right to the beneft of technological developments made by the contractors during building projects.91 Even if contractual revision is not out of the question, it seems that the attitude towards contractual revision has hardened over the last 20 years. Generally, the change 86 Cf. for instance Mårten Schultz, Kausalitet. Studier i skadestandsrettslig argumentation, Stockholm 1992 p. 101 f., and Trine-Lise Wilhelmsen, Årsakssammenheng i erstatningsretten, Oslo 2011, p. 15–19. 87 Schultz p. 127, p. 131 f. and p. 141 f., Wilhelmsen 2011 p. 18. 88 NOU pp. 47 and 61, Ot. Prp. no 5 (1982–1983) Om lov om endringer i avtaleloven 31. mai 1918 nr. 4 m.m. (Generell formuerettslig lempningsregel) (Ot. Prp.) p. 33, Generalklausul i förmögenhetsrätten, SOU 1974:82 p. 111, Regjeringens propositioner 247/1981 p. 14, Kai Krüger, Kontraktsrett 1989, p. 422, Viggo Hagstrøm, Obligasjonsrett, Oslo, 2. ed. 2011, p. 290. 89 Rt 1999.922 at p. 932; Zak p. 42. 90 ND 1990.204 NA Ula and ND 2000.240 NA Troll. 91 NOU pp. 47 and 61.

178

liability and insurance clauses in norwegian offshore sector of attitude is because courts have favored predictability at the cost of fairness, particularly in relation to later events resulting in more extensive losses than expected. There are very few cases where an agreement is set aside according to the Contract Act § 36 over the last 20 years, and in those cases the reasoning is inapplicable in our context.92 This process conforms to the Supreme Court’s development and practice of giving more weight to the objective understanding of the wording in the contract where the parties are professional.93 The evaluation according to the Contract Act § 36 is a broad evaluation of the circumstances listed in the provision in relation to the individual contract. However, not all the circumstances are relevant for this paper. The argument “the position of the parties” refers to the situation where one of the parties lacks the competence or ability to enter the agreement, or lacks the knowledge and experience to understand it, or there is a clear inequality between the parties in regard to the contract. As a general consideration this seems less relevant for the types of contracts discussed here. The same is true for the argument “the circumstances prevailing at the time of conclusion”, which refers to duress, misuse of negotiation power, exploitation and information failure. Therefore, the argument that is more generally relevant is that of the content of the agreement. 5.3.2 The content of the agreement The content of the agreement as a reason for setting aside this kind of liability provision means that the knock-for-knock principle applied in cases of gross negligence by the company itself would be unfair. The preparatory documents to the Norwegian Contract Act § 36 provision sheds little light on limitation of liability clauses in general. However, the equivalent Swedish rule and their preparatory documents contains relevant remarks.94 In those preparatory documents, it is stated that the application of limitation of liability clauses shall not be limited to a specifc degree of fault, but instead it depends on a total evaluation of the specifcs of the actual contract. In cases where freedom from liability is tied to fnancing through insurance, the main purpose will be to limit recourse from the insurer, and a convenient liability and insurance regime limiting the costs of recourse processes should not be denied through strict principles of fairness.95 Based on these statements, Swedish legal theory has assumed that limitation of liability clauses in professional contracts should be treated diferently from other contracts, in particular if they are combined with insurance.96

92 Rt. 2014.351 (energy company freed from a contract where they would have to pay revenue to the landowner for 50 years against no performance from the landowner), Rt. 2013.769 (revision of insurance settlement with no cover for income disability after such disability proved to be 50%), Rt. 2013.388 (consumer investment in structured saving products set aside because misleading information of central parameters for the investment and the risk for loss gave a wrong picture of the possibility of revenue on the investment), Rt 2008.969 (Fraud or misleading information), Rt 2001.603 (a continuation of an agreement seemed meaningless). 93 Cf. for instance Rt. 2010.1345, Rt. 2002.1155, Rt. 2000.806. 94 Relevant for the Norwegian Contract Act § 36 also because this paragraph is a result of Nordic legislative cooperation with identical rules in all the Nordic countries, cf. Wilhelmsen, Avtaleloven § 36 og økonomisk efektivitet, Tidsskrift for rettsvitenskap 1995 no. 1, pp. 13–14; Hagstrøm (2011) p. 287 f.; Zak p. 43. 95 SOU 1974:83 pp. 180–181. 96 Claes-Robert von Post, Studier kring 36 § avtalslagen med inriktning på rent kommersiella förhållanden, Stockholm 1999, p. 207; Jan Ramberg og Christina Ramberg, Allmän avtalsrätt, Åttonde upplagan,

179

trine-lise wilhelmsen A similar focus on professionalism and insurance is found in court practice concerning the Nordic Freight Forwarder Agreement (NSAB), which is an agreed standard contract with a long tradition. NSAB states that the freight forwarder’s liability for damage is limited regardless of fault.97 In U 1993.851 this clause was set aside by the Danish Supreme Court when the freight forwarder negligently failed to follow its own established practise for delivery of goods.98 However, the limitation was accepted in U 2005.243 and U 2006.632. In U 2005.243 the Danish Supreme Court simply stated that the clause must equally be accepted as written even in cases of gross negligence. In U 2006.632, the situation was that the company failed in the planning and performing of the service. It is not directly stated that the failure was grossly negligent, but this seems to be presumed in the lower court, which set the limitation aside. The Danish Supreme Court referred to U 2005.243, and stated that the limitation could not be set aside, according to the Contract Act § 36, in cases of gross negligence. The main arguments were that the freight forwarder contract is an agreed standard contract where the limitation is part of a total liability regime, which presumably rests on a total evaluation where considerations of efcient insurance play a central role. This result conforms to Norwegian practice, but here the negligence is tied to the employee, and not to the company. In Rt 1994.626, the same clause was accepted with similar reasoning found in U 2006.632. A similar view is found in previous arbitration and appeal cases with towing contracts.99 The Norwegian theoretical discussion on this issue is divided into two factions. One faction argues that such clauses should be set aside due to traditional considerations of fairness similar to those following from the discussion on NL 5–1–2.100 Arguments in relation to the knock-for-knock principle in the Norwegian Fabrication Contract are that the activity constitutes a risk for personal safety and pollution, and that transaction costs are presumably small compared to the importance of these interests. But even if they are not, efciency considerations must be given less weight than the need for a liability regime to protect personal safety and environment.101 This argument seems to overlook the fact that the risk for pollution damage is regulated by mandatory regulation in the MC and the PA, and that this regulation is adhered to in the contracts. Furthermore, under certain market conditions considerations of deterrence may not necessitate a liability regime. The other faction argues that the principle should be accepted as it is described.102 Main arguments here are that the knock-for-knock system is supported by both

Stockholm, 2010, p. 215; Thorsten Lundmark, Friskrivingsklausuler giltighet og räckvidd, Uppsala, 1996, p. 133; Zak p. 44. 97 NSAB 1975 § 25; NSAB 2000 § 22 cf. § 5 (no limitation in case of deliberate acts); NSAB 2015 § 23. 98 Cf. further Hagstrøm (1996) p. 435. 99 ND 1989.225 NA and ND 1991.180 Eidsivating. 100 Hagstrøm (1996) p. 422, p. 478 f.; Jo Hov and Alf Petter Høgberg, Alminnelig avtalerett, Oslo 2009, p. 402; Lasse Simonsen, ”Kreditors mangelsbeføyelser – særlig for tilvirkningskontraktene”, Jussens Venner 1999 p. 305 at p. 380; Zak p. 41. 101 Hagstrøm (1996) p. 481. 102 Kaasen p. 776 f., Bull p. 391 f., Erik Røsæg, “Lastehåndterings og forvaringstjenester”, MarIus no. 271, p. 41.

180

liability and insurance clauses in norwegian offshore sector parties to the contracts103 and that freedom of liability is closely tied to an insurance regime securing the interests of the victim.104 What can be concluded here is that freedom from liability in cases where the damage is caused by gross negligence by the company itself can only be achieved if some minimum requirements are fulflled: the contract should be agreed to secure involvement and acceptance by both parties, the freedom from liability should be tied to a systematic insurance regulation to secure that all potential victims are compensated, the liability and insurance system should refect a thorough analysis of what combination of liability insurance and casualty insurance is most convenient for the parties, and the system should reduce transaction costs. But even when these conditions are fulflled, acceptance by the court is still uncertain. 5.3.3 The insurance clauses The knock-for-knock principle is, as mentioned, combined with waiver of subrogation clauses and co-insurance clauses in the insurance policies. The waiver of subrogation clause is similar to the indemnity clause because it is not tied to any degree of fault. However, the position of the co-insured will mean that the co-insured party obtains an indirect liability cover for ordinary negligence, but in the case of gross negligence the result may be a reduction in the compensation. It must be presumed that these insurance clauses follow the same mandatory regime as the indemnity clauses. If the indemnity clause is deemed invalid according to NL 5–1–2 or unfair according to the Contract Act § 36, but the waiver of subrogation or co-insured’s protection is upheld, the position of the party claiming to be indemnifed will difer according to whether the injured party makes the claim against the injurer, or prefers to claim coverage from his own insurer. In the frst case, the injurer must compensate the injured party. In the second case, he will be free from liability. Such an arbitrary result is inconsistent and contrary to considerations of fairness.105 The result must therefore be that the mandatory rules apply similarly to the waiver of subrogation clauses and the co-insured’s indirect liability protection.

103 Bull p. 393; Kaasen p. 778. 104 Kaasen pp. 778–780; Bull pp. 393–394. 105 Hagstrøm (1996) pp. 485–486. See also Selmer p. 130; Bull p. 319.

181

CHAPTER 9

Statutory liability regulation versus contractual risk allocation in upstream oil and gas The Norwegian case Kristofer Svendsen 1 Introduction This chapter looks at the interaction between the statutory regulation of liability for damage or loss caused by petroleum spills from ofshore installations, as set forth in Chapter 7 of the Norwegian Petroleum Act,1 and allocation of risk in oil and gas contracts achieved through the use of knock-for-knock clauses. The chapter frst sets forth the liability regime under Chapter 7, then examines the contractual allocation of risk in model oil and gas contracts used on the Norwegian Continental Shelf (NCS), before determining the validity of said clauses and its efects when interacting with Chapter 7 liability. It is important to remember, however, that knock-for-knock clauses cover many more situations than just possible Chapter 7 pollution damage. In other words, situations arising under Chapter 7 of the Petroleum Act are only a part of what these risk allocation clauses cover. 2 Statutory regulation of risk under Chapter 7 of the Norwegian Petroleum Act 2.1 Introduction The Norwegian authorities started examining delict liability for pollution damage caused by oil spills from ofshore installations as early as 1970, which resulted in a committee report in 1973.2 The Norwegian government’s work on the issue was temporarily halted due to a belief that the work leading up to and the fnalized Convention on Civil Liability for Oil Pollution Damage Resulting from Exploration for and Exploitation of Seabed Mineral Resources3 would take efect. When the convention did not receive necessary backing from important countries, such as the United Kingdom, the Norwegian government appointed another committee to suggest regulation for compensation of pollution damage from ofshore installations. The committee

1 Lov om petroleumsvirksomhet 29 November 1996 Nr. 72. 2 NOU: Erstatningsansvar for forurensingsskader. Om erstatningsansvar m.v. for forurensingsskader i forbindelse med undersøkelse etter og utvining av undersjøiske naturforekomster (1973:8). 3 Convention on Civil Liability for Oil Pollution Damage Resulting from Exploration for and Exploitation of Seabed Mineral Resources, London (1 May 1977).

182

DOI: 10.4324/9781003206798-11

regulation versus risk allocation in upstream oil and gas gave its report in 1981,4 which was afrmed, with minor changes, in Chapter 7 of the Petroleum Act.5 The frst edition of the Petroleum Act came into force in 19856 and received an overhaul in 1996.7 The Petroleum Act of 1996 is currently in force and only minor changes took place in Chapter 7 of the 1996 edition. The Petroleum Act contains four liability regimes exclusively dealing with petroleum activities. Chapter 7 of the Petroleum Act is the main legislation regulating liability for petroleum pollution damage.8 Chapter 8 regulates compensation of fnancial losses sufered by Norwegian fshermen as a result of the petroleum activities occupying fshing, causing pollution and waste, or inficting damage by a facility or actions in connection with the placing of a facility.9 The regulation of Chapter 8 damages does not apply to petroleum pollution damage under Chapter 7. Section 5–4 of the Petroleum Act regulates liability in the abandonment phase, while section 10–9 of the Petroleum Act regulates liability of the licensee for damage caused by a legal entity of physical person who performs work for the licensee. Section 10–9 does not, however, apply to losses resulting from petroleum pollution damage under Chapter 7 or losses sufered by Norwegian fshermen as a result of petroleum activities under Chapter 8. The chapter focuses on the efect of Chapter 7 on contractual risk allocation. The Maritime Code10 defnes the outer limits of Chapter 7, and more generally the Petroleum Act. The Code mainly regulates various aspects of vessels and shipping and includes provisions on liability of oil pollution from vessels, among others. Importantly, the Maritime Code regulates ‘drilling platforms and similar mobile constructions’ when these units are in motion and therefore categorized as vessels.11 The Pollution Control Act12 also assists in defning the outer limits of Chapter 7, and the Petroleum Act. The Pollution Control Act regulates compensation for pollution damage, unless regulated by other legislation, contract, or other things.13 The Pollution Control Act does not apply to situations specifcally regulated by Chapter 7 of the Petroleum Act.14 However, Chapter 7 is not exhaustive regarding the defnitions

4 NOU: Erstatningsansvar for forurensningsskade som følge av petroleumsvirksomhet på norsk kontinentalsokkel (1981:33). 5 Ot. prp.nr. 72. Lov om petroleumsvirksomhet (1982–1983). 6 Lov om petroleumsvirksomhet (repealed) 22 March 1985 Nr. 11. 7 Lov om petroleumsvirksomhet 29 November 1996 Nr. 72. 8 Id. at § 7–1. 9 Id. at § 8–1. 10 Lov om sjøfarten The Norwegian Maritime Code is translated into English in MarIus No. 393 (24 June 1994 No. 39). 11 Id. at § 507. 12 Lov om vern mot forurensninger og om avfall (Forurensningsloven) 13 March 1981 Nr. 06. 13 Id. at § 53. 14 The preparatory works states that the Pollution Act supplements or complements the special rules ‘as far as the rules ft’. Ot.prp.nr.33 (1988–1989) Om lov om endringer i lov 13 mars 1981 Nr. 6 om vern mot forurensninger og om avfall (forurensningsloven) m.v (Erstatningsansvar ved forurensningsskade) (1988–1989) pp.  99–100 and 104. See also a discussion in this topic in Hans Christian Bugge, Forurensningsansvaret (1999) (Tano Aschehoug, Oslo) para. 6.3.2.2.

183

kristoffer svendsen of pollution damage, which is seen below.15 Finally, the general rules, lex generalis, of delict law apply when the special rules, lex specialis, of delict law come short.16 2.2 Liability for petroleum pollution damage 2.2.1 Unlimited no-fault liability The Petroleum Act places unlimited no-fault liability on the licensee for pollution damage caused by petroleum spills from ofshore installations.17 Usually there are several licensees under a license, of which one is the operator. In these situations, claims for compensation should be directed to the operator. If the operator does not pay claims when due, the licensees pay the claims according to their participating interest in the license.18 If a licensee fails to pay its share, that share is allocated proportionally between the remaining licensees. Interestingly, the Act continues with a very practical approach to the procedural aspect of mass claims requiring the operator without undue delay, by public announcement, to provide information regarding the party to whom claims for compensation for pollution damage should be directed and of the period of limitation.19 The Ministry of Petroleum and Energy decides where actions should be brought when questions of venue arise.20 The Act places unlimited no-fault liability on the licensee because the licensee has the authorization to conduct the activity, the fnal say, obtains the profts from the activity (high fnancial reward), holds bargaining power (insurance, contracts, etc.), and therefore should hold the risk and the consequences of oil pollution damage.21 The equitable policy consideration ‘prevention’ is also used to justify strict liability, based on which the licensee will take better precautions to hinder an accident as he has a direct fnancial interest in the matter.22 Furthermore, the licensee must obtain liability insurance, of which the Ministry of Petroleum and Energy receives summaries every year and ensure that appropriate insurances are maintained.23 The 15 The Pollution Act covers all non-petroleum pollution damage such as other chemicals and wasterelated petroleum pollution damage. Bugge states that it is reasonable to apply the Pollution Control Act to supplement and complement the Petroleum Act (a) for damages sufered by other groups of people then Norwegian fshermen as a result of other types of pollution then for petroleum pollution, and (b) in situations where other types of pollution than petroleum pollution infict personal harm on Norwegian fshermen or a reduction in fshing abilities. Bugge, Forurensningsansvaret (1999), para. 6.3.2.2, pp. 263–264. 16 The general rules of the law of compensation are compiled in the Damages Compensation Act (Lov om skadeserstatning 13 June 1969 Nr. 26) as well as in case law, preparatory works, legal theory, and by applying and balancing the equitable policy considerations (reelle hensyn). 17 Lov om petroleumsvirksomhet (29 November 1996 Nr. 72). § 7–3. 18 Id. at § 7–3. 19 Id. at § 7–7. 20 Id. at § 7–8. The Ministry decides where the action shall be brought if (a) the efuence or discharge has taken place or the damage has been caused outside the area of any court district; (b) it cannot be demonstrated within which court district the efuence or discharge has taken place or damage has been caused; (c) the efuence or discharge has taken place in one court district and the damage is caused in another court district (d) damage has been caused in more than one court district. 21 NOU: Erstatningsansvar for forurensningsskade som følge av petroleumsvirksomhet på norsk kontinentalsokkel. 1981:33, p. 21. 22 Id. at p. 25. 23 Erik Røsæg, The Norwegian perspective with regard to liability regimes concerning oil rigs and installations, in Ofshore contracts and liabilities (Baris Soyer & Andrew Tettenborn eds., 2015), p. 281.

184

regulation versus risk allocation in upstream oil and gas average insured amount appears to be 250 million USD for each licensee covering its pro rata share of the liability.24 The geographical scope of Chapter 7 is subject to a special regulation in section 7–2, which is diferent from all other liability regimes. The section applies lex loci damni when pollution damage ‘occurs in Norway or inside the outer limits of the Norwegian continental shelf or afects a Norwegian vessel, Norwegian hunting or catching equipment or Norwegian facility in adjacent sea areas.’25 The consequence of the scope of Chapter 7 is a unilateral extension of protection in delict law to Norwegian interests harmed outside of Norway. This privilege has resulted in discrimination against only Russian interests, which do not receive any judicial remedy on pollution damage caused by a Norwegian operator sufering an oil spill on the Norwegian side of the sea border in the Barents Sea, and that oil spill inficts pollution damage on the Russian side of the sea border in the Barents Sea.26 A Russian injured party forced to pursue a legal claim against a Norwegian licensee without assets in Russia may receive no compensation, because there is no agreement about recognition and enforcement of foreign court judgments between Norway and Russia.27 2.2.2 Pollution damage The term ‘pollution damage’ is defned as ‘damage or loss caused by pollution as a consequence of efuence or discharge of petroleum from a facility, including a well, and costs of reasonable measures to avert or limit such damage or such loss, as well as damage or loss as a consequence of such measures.28 Damage includes personal injury,29 and damage to fshermen.30 Chapter 7 does not cover discharge of petroleum from vessels transporting petroleum, which is covered by Chapter 10 of the Maritime Code. The Act defnes ‘pollution damage’ in relation to events for which liability under Chapter 7 of the Act may arise, but does not assist in defning compensable damage. The main delict statute, the Compensation for Damages Act,31 and other delict legislation do not notably assist much in defning compensable damage.32 The preparatory works

24 Id. at p. 281. 25 Lov om petroleumsvirksomhet 29 November 1996 Nr. 72. § 7–2. 26 Kristofer Svendsen, Compensable damage ex delicto as a result of harm in the Barents Sea caused by petroleum spills from ofshore installations. A Norwegian and Russian comparative legal analysis of confict of laws, the concept of harm, losses sufered by third parties, and environmental damage and its valuation and calculation, caused by petroleum spills from ofshore oil rigs and installations in the Barents Sea (2015) Tromsø Ph.D. in law, Faculty of Law, UiT – Arctic University of Norway). 27 No other sea bordering countries to Norway are afected due to the Convention on jurisdiction and the enforcement of judgments in civil and commercial matters, Lugano (30 October 2007). 28 Lov om petroleumsvirksomhet 29 November 1996 Nr. 72. § 7–1. Petroleum is defned in § 1–6a, and facility is defned in § 1–6d. 29 Ot. prp.nr. 72. Lov om petroleumsvirksomhet. 1982–1983, p. 70. 30 Chapter 8 damage (diferent from Chapter 7 damage) is specifc damage to Norwegian fshermen caused by petroleum activities (and maybe non-petroleum substances). Norwegian fshermen may be compensated for their economic loss caused by pollution damage or waste from petroleum activities. Compensation includes loss of fshing opportunities and equipment due to the aforementioned pollution. Lov om petroleumsvirksomhet 29 November 1996 Nr. 72, § 8–1. 31 Lov om skadeserstatning 13 June 1969 Nr. 26. 32 Sören Koch, Det erstatningsrettslige skadebegrepet – en sammenligning mellom tysk og norsk rett, Tidsskrift for erstatningsrett, forsikringsrett og velferdsrett pp. 250–281 (2010), p. 255. Koch summarises the understanding of damage in legal literature on pages 256–261.

185

kristoffer svendsen give examples of certain types of damage that would qualify as pollution damage, such as harm inficted to wildlife in the sea and on land, the soiling of beaches and fshing gear, the closure of a water area as an obstacle for fshing and shipping, the soiling of real estate and objects, and costs incurred by the public or others for cleaning up soiled beaches.33 The preparatory works confrm that compensable pollution damage must fulfll the prerequisites for every type of damage compensable in delict law.34 Section 6 of the Pollution Control Act defnes pollution through four sentences. The frst sentence qualifying for pollution is ‘the introduction of solids, liquids or gases to air, water or ground.35 Efuence or discharge of petroleum would be within the wording of section 6. The fnal part of each of the four sentences is ‘which cause or may cause damage or nuisance to the environment. 36 The wording ‘damage or loss caused by pollution’ in section 7–1 fulflls the fnal part of the frst sentence of section 6. Thus, the defnition of pollution in the Pollution Control Act defnes ‘pollution’ in section 7–1,37 while the Petroleum Act defnes petroleum,38 efuence and discharge,39 and facility.40 2.2.3 Channelling liability to the licensee and licensee’s recourse The Petroleum Act channels unlimited liability for pollution damage to the licensee. The courts have discretionary power to reduce this unlimited liability partly or completely upon (1) an inevitable event of nature, (2) an act of war, (3) the exercise of public authority, or (4) a similar force majeure event that has contributed to a considerable degree to the damage or its extent under circumstances which are beyond the control of the liable party.41 Liability can only be reduced to the extent it is reasonable, with particular consideration to the scope of the activity, the situation of the party that has sustained damage and the opportunity for taking out insurance on both sides.42 The channeling provisions in section 7–4 bar claims for petroleum pollution damage to be directed towards licensee’s contractors, manufacturers or suppliers of equipment, anyone who undertakes measures to avert or limit pollution damage,43 and 33 NOU: Erstatningsansvar for forurensningsskade som følge av petroleumsvirksomhet på norsk kontinentalsokkel. 1981:33, p. 35. 34 Id. at p. 35 agreeing with NOU: Erstatningsansvar for forurensingsskader. Om erstatningsansvar m.v. for forurensingsskader i forbindelse med undersøkelse etter og utvining av undersjøiske naturforekomster. 1973:8. 35 Lov om vern mot forurensninger og om avfall (Forurensningsloven) (13 March 1981 Nr. 06). § 6 1. 36 Id. at § 6. 37 U. Hammer, et al., Petroleumsloven (Universitetsforlaget, Oslo. 2009). p. 532. 38 § 1–6 (a) of the Petroleum Act. 39 The preparatory works to the Petroleum Act do not expressly distinguish between “efuence” (utstrømming) and “discharge” (utslipp). The legislator uses the word “efuence” when describing a leak from an oil pipeline. Ot. prp.nr. 72. Lov om petroleumsvirksomhet. 1982–1983, p. 70. 40 Lov om petroleumsvirksomhet 29 November 1996 Nr. 72, § 1–6d. 41 Id. at § 7–3. 42 Id. at § 7–3. 43 The full text reads ‘anyone who undertakes measures to avert or limit pollution damage, or to save life or rescue values which have been endangered in connection with the petroleum activities, unless the measures are performed in confict with prohibitions imposed by public authorities or are performed by someone other than public authorities in spite of express prohibition by the operator or the owner of the values threatened.’ Id. at § 7–5(c).

186

regulation versus risk allocation in upstream oil and gas employees of the licensee or employees of any of the above groups.44 The same parties shielded from liability in section 7–4 are shielded from indemnity in section 7–5. The licensee can only seek recourse from the parties listed in section 7–4 if ‘the person in question or someone in his service has acted willfully or by gross negligence.’45 Similarly, a shipowner is also strictly liable for oil pollution damage, with a similar group46 protected through channeling of liability except when a person caused damage with intent or through gross negligence, and with the knowledge that such damage would probably result.47 In summary, injured parties cannot direct claims towards these parties, and the licensee cannot demand that these parties accept liability for pollution damage, unless intent or gross negligence. This begs the question whether licensees on the NCS are solid enough to carry the liability of a potential oil spill. Importantly for our current discussion, section 7–5 states: Any agreement on further recourse in respect of those against whom liability cannot be claimed pursuant to Section 7–4, second paragraph, shall be invalid.

The parties can, however, further limit access to recourse,48 but not extend recourse in contravention with section 7–5. This would indicate that the shielded parties could contractually be protected against pollution damage inficted by them through, for example, gross negligence. However, it is doubtful in Norwegian legal theory whether a party can contractually agree to avoid liability for gross negligence.49 3 Contractual allocation of risk on oil and gas contracts in Norway 3.1 Introduction The contractual allocation of risk in oil and gas contracts is often achieved through the careful wording of clauses describing which parties should pay which parties for damage or harm arising out of diferent situations as a result of physical risks materializing. Gordon describes three vehicles used to regulate and manage these physical risks in the oil and gas industry: (1) indemnity and hold harmless clauses, (2) clauses which exclude or limit liability for ‘consequential losses’, and (3) overall limitations of liability.50 Gordon takes a further dive into indemnity and hold harmless clauses in Chapter 2 of this book. This part of the chapter looks at the main risk allocation 44 Id. at § 7–4. 45 Id. at § 7–5. 46 (a) The servants or agents of the owner or the members of the crew; (b) the pilot or any other person who performs services for the ship; (c) the ship operator (reder) or manager or operator of the ship, the charterer, the consignor, shipper, owner of cargo or consignee; (d) any person performing salvage operations with the consent of the owner or on the instructions of the public authority; (e) any person taking preventive measures; (f) all servants or agents of persons mentioned in subparagraphs (c), (d) and (e). 47 Lov om sjøfarten The Norwegian Maritime Code is translated into English in MarIus No 393 24 June 1994 No. 39, § 193. 48 NOU: Erstatningsansvar for forurensningsskade som følge av petroleumsvirksomhet på norsk kontinentalsokkel. 1981:33, p. 41. 49 Viggo Hagstrøm, Om grensene for ansvarsfraskrivelse, særlig i næringsforhold, 1996 Tidsskrift for rettsvitenskap pp. 421–518 (1996); see Wilhelmsen in Chapter 8 and Kaasen in Chapter 7a of this book. 50 Greg Gordon, Chapter 14, Risk Allocation in Oil and Gas Contracts, in Oil and Gas Law: Current Practice & Emerging Trends (Greg Gordon, et al. eds., 2011), p. 14.2 f.

187

kristoffer svendsen clause used in model oil and gas contracts in Norway, and whether they make reference to or take into account the above-mentioned section 7–5 of the Petroleum Act. 3.2 Industry-negotiated model contracts The NCS is well known for the wide use of industry-negotiated model contracts (also called agreed documents) setting forth a set of standard conditions developed for contracting on the NCS. These model contracts are not mandatory to use but are widely used in engineering, procurement, and construction on the NCS. The following contracts are the main industry-negotiated model contracts: • The Norwegian Total Contract of 201551 (NTC 15); • The Norwegian Total Contract of 2015 Modifcation52 (NTC 15 Mod); • The Norwegian Total Contract of 2015 for module and modifcation (modifcation with single delivery of the entire contract object);53 • The Norwegian Total Contract of 2015 for module and modifcation (modifcations with separate delivery of module, prefabricated items, and ofshore permanent works);54 • The Norwegian Fabrication Contract of 201555 (NFC 15); • The Norwegian Subsea Contract of 200556 (NSC 05). The NTC 15 is recommended to regulate contracts for the delivery of larger components for production of hydrocarbons on the NCS where the contractor is responsible for engineering, procurement, construction, and possibly installation (EPC(I)). NTC 15 Mod is recommended to regulate contracts for larger modifcations of platforms on the NCS where the contractor is responsible for EPC(I). The NTC 15 for module and modifcations is recommended when the delivery also includes a new module, and the contractor is responsible for EPC(I). This standard contract is issued in two versions, the frst version for which the contract object is delivered collectively, while the second version caters for separate delivery of the module, prefabricated items, and ofshore permanent works. The NFC 15 is recommended for larger contracts for delivery of fabrication, such as machinery and mechanical equipment, to the NCS. Finally, the NSC 05 is intended to regulate contracts covering ‘marine operations such as installation of pipelines, cables, umbilicals and other subsea structures and related subsea

51 The contract can be retrieved at www.norskindustri.no/contentassets/69b36f82c6f341a68c1aa7691e4 f5e06/versjoner-lagt-inn-des.-2017/norsk-totalkontrakt-2015.pdf 52 The contract can be retrieved at www.norskindustri.no/contentassets/69b36f82c6f341a68c1aa7691e4 f5e06/versjoner-lagt-inn-des.-2017/norsk-totalkontrakt-2015-modifkasjon.pdf. 53 The contract can be retrieved at www.norskindustri.no/contentassets/69b36f82c6f341a68c1aa7691e 4f5e06/versjoner-lagt-inn-des.-2017/norsk-totalkontrakt-2015-modul-og-modifkasjon-samlet-levering.pdf. 54 The contract can be retrieved at www.norskindustri.no/contentassets/69b36f82c6f341a68c1aa7691e4 f5e06/versjoner-lagt-inn-des.-2017/norsk-totalkontrakt-2015-modul-og-modifkasjon-separat-levering.pdf. 55 The contract can be retrieved at www.norskindustri.no/contentassets/69b36f82c6f341a68c1aa7691e4 f5e06/versjoner-lagt-inn-des.-2017/norsk-fabrikasjonskontrakt-2015.pdf. 56 The contract can be retrieved at www.norskoljeoggass.no/drift/publikasjoner/hms-og-drift/ norwegian-subsea-contract/.

188

regulation versus risk allocation in upstream oil and gas construction work where the use of vessels is involved.57 The NSC 05 captures both installation-only contracts as well as full EPC(I)-type contracts and addresses specifc risks in connection with subsea work and the operation of vessels.58 Large industrial actors, namely the predecessor of the Federation of Norwegian Industries (Mechanical Industry Association) and the Norwegian Union of Iron and Metalworkers (now part of the Norwegian United Federation of Trade Unions) started the negotiations of these model contracts, the NTC and the NFC, in the 1970s, with the frst draft presented in 1983. Other large industrial parties, such as Hydro, Statoil, and Saga, joined in and the frst edition of the model contract was presented in 1987.59 The current editions are a result of eforts between the Federation of Norwegian Industries, the Norwegian Oil and Gas Association, and companies in the industry. The predecessor of the Norwegian Oil and Gas Association, OLF, initiated the preparation of the model contract NSC 05 with participation of Statoil, Stolt Ofshore, Subsea 7, and Technip Ofshore Norge. 3.3 The risk allocation clause These model contracts were negotiated to implement a standard for a developing industry to rely on, even the commercial playing feld through balancing commercial contracting provisions, efectively train staf, and minimize transactional costs. In all the model contracts above, article 30 contains the provision titled ‘Exclusion of liability. Indemnifcation’, which is the preferred method of the industry to allocate physical risk. All the above-mentioned model contracts, except for a couple of words in the NSC 05, contain the exact same wording of their exclusion of liability and indemnifcation clause. This is not particularly strange as exclusion of liability and indemnifcation work best in the contractual chain when all parties use the same clause, ‘back-to-back’. Article 30 of the above-mentioned contracts sets forth a mutual indemnity clause60 for loss or damage to Contractor and Company Group, Contractor’s indemnifcation obligation for third-party claims and its limitation, indemnifcation for industrial property infringements, and notifcation procedures. In other words, article 30 consists of a combination of Gordon’s three vehicles. An mutual indemnity clause means ‘a contractual device where the parties with the one hand give and with the other hand take an indemnity in respect of a species of loss which, if the indemnity is to avoid circularity, must not be identical to each other, but which are usually closely related61 Articles 30.1 and 30.2 set forth a mutual indemnity clause for personal injury, death, and loss of or damage to property in relation to the subject matter of the contract applying when sufered by the parties to the contract, the Contractor and the Company Group. The identical 57 See OLF, Norwegian Subsea Contract, NSC 05, Conditions of Contract (2005). ‘Introduction to NSC 05.’ 58 Id. 59 Knut Kaasen (2006) Petroleumskontrakter med kommentar til NF 05 og NTK 05 (Universitetsforlaget, Oslo), pp. 23 f. 60 A mutual indemnity is also called ‘reciprocal indemnity’, ‘cross indemnity’, or a ‘knock-for-knock’ indemnity. Gordon, Chapter 14, Risk Allocation, in Oil and Gas Contracts, 2011. p. 14.5. 61 Id.

189

kristoffer svendsen mutual indemnity clause between the Contractor and Company Group reads as follows (identical in the NSC 05): 30.1 Contractor shall indemnify Company Group from and against any claim concerning: a) personal injury to or loss of life of any employee of Contractor Group, and b) loss of or damage to any property of Contractor Group, and arising out of or in connection with the Work or caused by the Contract Object in its lifetime. This applies regardless of any form of liability, whether strict or by negligence, in whatever form, on the part of Company Group. Contractor shall, as far as practicable, ensure that other companies in Contractor Group waive their right to make any claim against Company Group when such claims are covered by Contractor’s obligation to indemnify under the provisions of this Art. 30.1.62 30.2 Company shall indemnify Contractor Group from and against any claim concerning: a) personal injury to or loss of life of any employee of Company Group, and b) loss of or damage to any property of Company Group, except as stated in Art. 29, and arising out of or in connection with the Work or caused by the Contract Object in its lifetime. This applies regardless of any form of liability whether strict or by negligence, in whatever form, on the part of Contractor Group. Company shall, as far as practicable, ensure that other companies in Company Group waive their right to make any claim against Contractor Group when such claims are covered by Company’s obligation to indemnify under the provisions of this art. 30.2.63

The parties, the company group and the contractor, agree to indemnify each other from and against any claim concerning their own employees’ personal injury or death and loss of and damage to any of their own property. The wording ‘from and against any claim’ would mean a loss or damage independent of amount and injured party.64 Articles 30.1 and 30.2 apply to the internal relationship among the contracting parties. In article 30.3, however, the contracting parties regulate their interaction with public authorities and third parties. Article 30.3 reads: Until the issue of the Acceptance Certifcate, Contractor shall indemnify Company Group from: a) costs resulting from the requirements of public authorities in connection with the removal of wrecks, or pollution from vessels or other foating devices provided by Contractor Group for use in connection with the Work, and b) claims arising out of loss or damage sufered by anyone other than Contractor Group and Company Group in connection with the Work or caused by the Contract Object, even if the loss or damage is the result of any form of liability, whether strict or by negligence in whatever form by Company Group.

The Acceptance Certifcate is a document issued by Company when the work has been completed according to the contract.65 Article 23 discusses, amongst other things, the Acceptance Certifcate and requires the Company to the Acceptance 62 The wording of the NFC 15. 63 Id. 64 Hans Jacob Bull (1988) Tredjemannsdekninger i forsikringsforhold: en studie av dekningsmodeller, med basis i sjøforsikringsretten og i petroleumskontraktenes ansvars- og forsikringsregulering (Sjørettsfondet, Oslo), p. 394. 65 See Norwegian Fabrication Contract 2007 (NF 07), art. 1.10.

190

regulation versus risk allocation in upstream oil and gas Certifcate when the work has been completed under the contract, deeming the certifcate to be issued 30 days after either the guarantee period has ended.66 This part of article 30.3 does move a possible heavy fnancial burden over on the Contractor as the Company Group requires unilateral indemnity from the Contractor for removal and pollution expenses from vessels or other foating devices provided by the Contractor Group for use in connection with the work. More importantly is the sub-paragraph b) of article 30.3 requiring the Contractor to unilaterally indemnify the Company Croup from third-party claims for loss or damage sufered in connection with the Contractor’s work or caused by the object of the contract, even if the loss or damage is the result of any form of liability, whether strict or by negligence in whatever form by Company Group. Sub-paragraph (b) is in direct breach of articles 7–4 and 7–5 of the Petroleum Act, which expressly shields the contractor from liability for petroleum pollution damage and bans the licensee from seeking any indemnity from the contractor. Continuing in article 30.3, a separate limitation of contractor’s liability follows this paragraph: Contractor’s liability for loss or damage arising out of each accident shall be limited to NOK .  .  .  million. This does not apply to Contractor’s liability for loss or damage for each accident covered by insurances provided in accordance with Art. 31.2.a) and b), where Contractor’s liability extends to the sum recovered under the insurance for the loss or damage. Company shall indemnify Contractor Group from and against claims mentioned in the frst paragraph above, to the extent that they exceed the limitations of liability mentioned above, regardless of any form of liability, whether strict or by negligence, in whatever form, on the part of Contractor Group. After issue of the Acceptance Certifcate, Company shall indemnify Contractor Group from and against any claims of the kind mentioned in the frst paragraph above, regardless of any form of liability, whether strict or by negligence, in whatever form, on the part of Contractor Group.

This limitation of contractor’s ‘third-party liability’ to a set amount does not negate article 30.3’s infringement on articles 7–4 and 7–5 of the Petroleum Act. Previous editions of this part of article 30.3 contained a specifc amount of million NOK, which have increased through negotiation rounds throughout the years, the latest being 5 million NOK (approx. 530,000 EUR) in the 2007 edition.67 The current edition’s ability to set your own limitation might increase the licensee’s risk of having to cover a certain amount of money for petroleum pollution damage without the amount being insured. That said, many companies self-insure so the insurance aspect might be a non-issue.68 It is interesting to note that the Contractor in article 30.3, and the contracting parties in articles 30.1 and 30.2, is obligated to indemnity the Company Group ‘regardless of any form of liability, whether strict or by negligence, in whatever form’ on

66 Id. art. 23.5. 67 See Norwegian Fabrication Contract 2007 (NF 07). 68 Neil A. Doherty & Cliford W. Smith Jr., Corporate Insurance Strategy: The Case of British Petroleum, 6 (3) Journal of Applied Corporate Finance pp. 4–15 (1993).

191

kristoffer svendsen the part of any of these two parties. This means that article 30 applies to, amongst other, no-fault liability for petroleum pollution damage. According to Norwegian contract law, a party cannot contractually avoid liability for intent, but can arguably avoid liability for gross negligence (see more in Kaasen, Chapter 7a, section 5.2). This quick look at the model contracts shows no reference to section 7–5 of the Petroleum Act, or any indication that the model contracts incorporated the wording of section 7–5. 4 Contractual allocation of risk breaches with Chapter 7 The contracts consist of three main elements: (1) the parties to the contract waiver the ability to claim compensation from each other even if the elements to claim compensation is fulflled; (2) the parties to the contract waiver the ability to claim recourse from each other when a third-party damage or a damage to own employees are covered according to the agreement, even though a recourse action could take place; and (3) the parties commit to hold the other party harmless, even though one party might have to pay a third-party claim or an employee claim that should not have been paid by the party in another legal sense.69 The Norwegian model contract contains a mutual indemnity clause for damage or loss among the contracting parties, a unilateral indemnity clause for third-party damage to the beneft of the Company Group, and a fxed-amount limitation of this unilateral indemnity obligation left to the contracting parties to decide. However, the model contracts do not consider section 7–5 of the Petroleum Act, which states that there cannot be any changes to the channeling of liability for pollution damage set forth in Chapter 7 of the Petroleum Act. As such, Norwegian law does not allow for an indemnity clause shifting the liability for pollution damage away from the licensee or creating a risk allocation model diferent from the statutory regulation of liability for pollution damage in Chapter 7. Thus, mutual indemnity clauses would be invalid for pollution damage under Chapter 7.

69 Knut Kaasen, Ansvarsbegrensning i fabrikasjonskontrakter, in Industribygging og rettsutvikling. Juridisk festskrift i anledning Hydros 100-årsjubileum 2005 (Odd Ivar Biller, et al. eds., 2005), p. 241.

192

CHAPTER 10

Applying knock-for-knock in Germany Eckehard Volz and Anna-Sophie Waldmann

1 Introduction Conducting constructions, operations and maintenance ofshore poses a large number of challenges for developers, operators and contractors. Typically, there are many parties involved, very large sums of money at stake, and constantly evolving and therefore potentially dangerous work environments to contend with. Consequently, the allocation of risk between the parties is a fundamental issue for consideration at the outset of contract negotiations. In the ofshore oil and gas sector, the so-called knock-for-knock regime has been practiced for decades. In the German ofshore wind farm sector, although many parties are keen to contract on a knock-for-knock basis, there has been and remains resistance from some. This is because the knock-for-knock regime is fundamentally diferent from the liability regime and fault-based concept of German law. Consequently, German companies have been initially hesitant to accept these clauses, which were developed in the context of English law. The clauses, and their validity, are still not fully tested by German courts, so there are several legal uncertainties in this regard. In any case, the standard wording of knock-forknock clauses needs to be adapted to be valid under German law. This chapter means to illustrate the common legal issues regarding the knock-forknock clause under German law, because, in contrast to English law, German law signifcantly limits the parties’ right to exclude and/or limits their liability, especially in circumstances where the respective clause will most likely be treated as general terms and conditions (Allgemeine Geschäftsbedingungen). First, we will give an overview on the use and prevalence of knock-for-knock clauses in the German market (section 2) and then continue to illustrate the German approach to clauses that are deemed general terms and conditions and whether knock-for-knock clauses qualify as such (sections 3.1 and 3.2). Next, we will focus on the fairness standards a knock-for-knock clause has to meet and the prohibition to exclude liability for wilful/intentional misconduct (sections 3.3 and 3.4). Finally, we will shortly review of available case precedent (section 4) and then give a general recommendation as to how to conduct contract negotiations and amend the respective clauses in order to avoid disputes and minimise legal risks (section 5). 2 Use of knock-for-knock clauses in the German market As already mentioned in the introduction, knock-for-knock clauses originated in the English insurance market. German developers and contractors only encountered this DOI: 10.4324/9781003206798-12

193

eckehard volz and anna-sophie waldmann legal concept recently, once they became more heavily involved in the ofshore sector, particularly the ofshore wind sector. The frst German ofshore wind farm Alpha Ventus was commissioned in 2010, many of the frst projects still had very few international connections, and most if not all the contracting parties involved were German companies operating under the familiar German law and liability regime. Consequently, knock-for-knock clauses have only become more frequent and common during about the last fve years. While there were initially many companies in the market that refused to accept this liability regime, it became rapidly clear that German developers, wanting to attract international contractors, and German contractors, tendering for international projects, would need to adapt. Many quickly became used to the concept, which, admittedly, has many advantages as it makes liability risks more foreseeable, bankable, and insurable. Today, many German companies operating in the ofshore wind sector will accept or ask for knock-for-knock clauses to be part of their contracts. Outside of this industry, however, the concept remains largely unknown. 3 Validity issues under German law Due to some elements of German law and since knock-for-knock clauses have their origin in a diferent jurisdiction, there are elements of the liability regime that are invalid and some that are potentially invalid under German law. Therefore, the knockfor-knock clause needs to be adapted to work under a German contract. Carve-outs from the general concept that each party shall cover its own losses need to be made. In the following, we will explain the background on why these adaptations are necessary and what exactly needs to be amended. 3.1 Court review of general terms and conditions Under German law, all general terms and conditions are subject to court review. While in principle there is freedom of contract, court review serves to protect the party that has not drafted the general terms and conditions – particularly consumers, but also experienced commercial parties – from unfair treatment and unexcepted provisions. In the following, we will frst illustrate the general principle of court review of general terms and conditions and then go on to apply this principle to knock-for-knock clauses. 3.1.1 General principle In respect to general terms and conditions, the German Civil Code (BGB) contains rules in sec. 305 et seq. BGB to ensure the fairness and appropriateness between the parties, which must be complied with to avoid those particular clauses that are considered to be invalid by German courts. Accordingly, general terms and conditions are subject to supervision by the German courts as regards their enforceability. Such supervision predominantly intends to protect consumers from unreasonable and surprising clauses. However, most of these rules also apply to business contracts. German courts will therefore examine whether any of these clauses are unfair to the detriment of one of the parties. As far as particular clauses of the general terms 194

applying knock-for-knock in germany and conditions do not comply with the regulations set out in sec. 305 et seq. BGB, the relevant clause will be deemed invalid and consequently as non-binding between the parties of the contract. If a particular clause is considered invalid, this afects the whole clause, not only the invalid part. Thus, special attention must be paid during the drafting process and in case of a liability regime, even a minor drafting error can undermine the whole concept of the liability in a contract. 3.1.1.1 Definition of general terms and conditions Sec. 305 (1) sent. 1 BGB reads: Standard business terms are all contract terms pre-formulated for a multitude of contracts, which one party to the contract (the user) presents to the other party upon the entering into of the contract.

From a legal perspective any standard form contract, such as the BIMCO SUPPLYTIME 2005, WINDTIME, FIDIC, LOGIC or any other standard business terms, regardless of the actual frequency of its use by your company or others, will be deemed general terms and conditions under German law pursuant to sec. 305 (1) BGB, unless (1) the terms of the contract were not pre-formulated for a multitude of contracts – which will be very difcult to prove as generally standard forms are used as the contractual basis – or (2) the clause has been individually negotiated (see below) between the parties. 3.1.1.2 Content test In addition, any clause from standard business terms, if not individually negotiated, is subject to the so-called content test set out in sec. 307 (1) and (2) BGB, which read: (1) Provisions in standard business terms are inefective, if they unreasonably disadvantage the other party to the contract with the user contrary to the requirements of good faith. An unreasonable disadvantage may also arise from the provision not being clear and comprehensible. (2) An unreasonable disadvantage is, in case of doubt, to be assumed to exist, if a provision 1. is not compatible with the essential basic principles of the statutory provision from which it deviates, or 2. limits essential rights or duties inherent in the nature of the contract to such an extent that attainment of the purpose of the contract is jeopardised.

Whether or not a standard knock-for-knock clause is to the unreasonable disadvantage of a party will be determined by the relevant court or arbitral tribunal within the scope of a test of the reasonableness of content (Inhaltskontrolle) by looking at the principles set out in secs. 307, 308 and 309 BGB (Content Test). This Content Test therefore has two legs: frst, general principles of fairness must be observed; second, secs. 308 and 309 contain a list of terms that are prohibited in any case and of terms that can either be valid or invalid subject to interpretation of the specifc wording of the term and whether it allows for an interpretation in a way that is not unfair or grossly detrimental. Strictly speaking secs. 308 and 309 BGB are only directly applicable for consumer contracts. Nevertheless, the German courts take these provisions into account as 195

eckehard volz and anna-sophie waldmann guidelines to business contracts, however certainly also looking at the common business practice in a particular industry.1 3.1.1.3 What is “individually negotiated”? Thus, for business-to-business contracts it is advisable to prevent the courts or an arbitral tribunal from conducting the Content Test on any clause, i.e., to prevent clauses of the contract to be regarded as general terms and conditions at all. Under German law pursuant to sec. 305 para. 1 sent. 3 BGB any clauses, which are individually negotiated – this may include clauses that were originally general terms and conditions provisions – are not regarded as being general terms and conditions. However, the notion of what is individually negotiated (im Einzelnen ausgehandelt) has once more been subject to several court decisions by the German Federal Supreme Court (BGH): NEGOTIATION OF TERMS DEMANDS MORE THAN DEBATING

According to the BGH negotiating (aushandeln), in the sense of sec. 305 para. 1 sent. 3 BGB, demands more than simply debating or disputing (verhandeln) the terms of a contract. In fact, the user of general terms and conditions, which frst need to contain provisions that amend or add essential content to statutory provisions, must additionally put the terms to serious disposition and allow the contractual partner to protect his own interests by being able to infuence and modify the content of the contract.2 Therefore, the user must clearly and seriously agree to amend respective clauses, if requested by the contractual partner. However, a general reference that all terms of the contract were open to discussion is not considered to be sufcient to establish the necessary negotiation in respect of the essential content of individual clauses. The same applies for protocols of the negotiations, which contain wordings like “all terms of the contract have been thoroughly and seriously negotiated”; this is equally considered to be insufcient for proving that individual negotiations in the sense of sec. 305 (1) sent. 3 BGB were conducted. ADDITIONAL SECURITY BY A “CAVEAT” CLAUSE?

Until recently, as a means of additional security, it was possible and advisable to insert a rider clause aiming to change the character of the contract for the knockfor-knock clause into an ordinary contract clause. This was to prevent the abovementioned strict limitation rules for German general terms and conditions by suggesting that the draft contract was individually negotiated between the parties. Such a “caveat” clause would have had to state that both parties considered the entire agreement as in line with the common industry standard and as a reasonable and fair allocation of rights and duties, highlighting that each clause was known and

1 BeckOK BGB/H. Schmidt, 58. Ed. 1.5.2021 Rn. 6, BGB § 307 Rn. 6; BeckOK BGB/H. Schmidt, 58. Ed. 1.5.2021, BGB § 307 Rn. 96; MüKoBGB/Basedow, 8. Auf. 2019, BGB § 310 Rn. 11. 2 BGH, judgment dated 20.3.2014 – VII ZR 248/13 = NJW 2014, 1725; BGH order dated 20.11.2012 – VIII ZR 137/12, BeckRS 2013, 5597 Rn. 7; BGH, order dated 19.3.2019 – XI ZR 9/18, NJW 2019, 2080; MüKoBGB/Basedow, 8. Auf. 2019, BGB § 305 Rn. 35; Wolf/Lindacher/Pfeifer/Pfeifer, 7. Auf. 2020 Rn. 36, BGB § 305 Rn. 36.

196

applying knock-for-knock in germany part of the individual negotiations. The following additional wording was added at the end of the contract, if made subject to German law: This contract was in its entirety freely negotiated by both parties and both parties have duly noted all its contents and acknowledge that all the terms of this contract constitute a fair and balanced allocation of rights and duties which properly refect the . . . industry’s customs and practices.

Of course, parties never entirely discounted the risk that a court or arbitral tribunal might overrule this clause and still qualify such clause and with it any other clause of the contract, although negotiated and changed, as general terms and conditions. But it at least supported and may in future still support the argument of the parties that the contract is not constituted of general terms and conditions, including the implications this may have, but was indeed negotiated openly. Nevertheless, as mentioned above, the BGH has made clear that such clauses do not automatically prevent a contract from still being regarded to be general terms and conditions.3 Whether a contract is agreed based on general terms and conditions or individually is a question of fact, which cannot be contractually regulated by the parties, because this corresponds to the attempt to waive the general terms and conditions provisions. Being able to waive these provisions individually would exceed the scope of sec. 305 para. 1 sent. 3 BGB. The BGH is of the opinion that an economically superior party of the contract could then force such an “individual clause” into the contract to annul the protective purpose of German general terms and conditions provisions for the entire contract. For this reason, even in the commercial business area, a “caveat” clause will be regarded as legally non-binding, because sec. 305 et seq. BGB do not underlie the disposition of the parties to the contract, but instead have legally binding force. 3.2 Are knock-for-knock clauses general terms and conditions? Most knock-for-knock clauses used in practice are part of standard contract forms, such as BIMCO SUPPLYTIME, WINDTIME, LOGIC or other pre-formulated terms used in the industry. Under German law, these standard forms will be considered general terms and conditions, which then by extension applies to the knock-for-knock clause contained therein. If a tailor-made contract is drafted, copying the wording from one of these standard forms into the tailor-made contract without any amendments or negotiations, it is very likely that a court would also consider the knock-for-knock clause part of general terms and conditions. As illustrated above, individual and earnest negotiations about the content and wording of the knock-for-knock clause can make it an individually agreed term, but ultimately this will be decided by German courts or arbitral tribunals on a case-by-case basis, taking into account the circumstances of the individual matter. 3.3 Applying the reasonableness test Having established the basic principles in section 3.1 and whether knock-for-knock clauses are considered as such, we will now look at the standards for validity that 3 BGH, judgment dated 20.3.2014 – VII ZR 248/13 = NJW 2014, 1725.

197

eckehard volz and anna-sophie waldmann are applied by German court. As explained above, the Content Test has two legs, the general fairness and reasonableness test specifed in sec. 307 BGB and the more specifc list of terms not allowed as per secs. 308 and 309. 3.3.1 Commercial contracts versus consumer contracts As mentioned previously, the second leg of the Content Test only applies, in principle, to consumer contracts. Since knock-for-knock clauses are not generally used vis-á-vis consumers, this might therefore appear irrelevant. However, it is important to know that German courts use the list of terms prohibited in secs. 308 and 309 as indicators for what could be considered unreasonable or unfair also in a strictly commercial contract between businesspeople. Therefore, case precedent and legal commentary on the list of prohibited terms are signifcant also in a strictly commercial context and need to be considered during the drafting process for knock-for-knock clauses. 3.3.2 Fairness test – what is market practice? Sec. 310 para. 1 sent. 2 BGB states that when interpreting and applying sec. 307 BGB in connection with business-to-business contracts “reasonable account must be taken of the practices and customs that apply in business dealings” by the courts. Accordingly, a court would have to consider the ofshore wind industry’s customs and practices. Given that the wind industry is young in the context of established legal practices, it is barely foreseeable to what extent a court or arbitral tribunal would respect the ofshore wind industry’s practices and customs in this context. That said, German courts and arbitral tribunals will consider that the industry is dominated not only by German companies but companies with an international background, which use knock-for-knock clauses frequently. Should a knock-forknock clause ever be tested in front of a German court, it is to be expected that the court will not only look towards the use of knock-for-knock clauses in Germany, but to the industry, if the parties of that hypothetical case have an international background. Finally, as we will elaborate on in more detail below, it is important to keep in mind that there is as yet no court decision on the validity of knock-for-knock clauses. Therefore, the question whether German courts will accept them as common industry practice remains unanswered and it is quite difcult to predict how German courts will resolve the confict between the ofshore wind industry’s customs and practices on the one hand and the guidance taken from the principles contained in secs. 308 and 309 BGB on the other hand, when ruling on whether a knock-for-knock clause is considered to be unreasonably disadvantageous pursuant to sec. 307 BGB and thus invalid or not. It will largely depend on the individual circumstances of the matter. How those circumstances may be infuenced in favour of the validity of the clause is discussed below. 3.3.3 The “risky” terms As mentioned above, sec. 309 BGB, inter alia, is not directly applicable to commercial agreements between professional businesses but the German courts consider them as a guideline for business contracts. With regard to knock-for-knock clauses 198

applying knock-for-knock in germany (e.g., in Clause 14 BIMCO SUPPLYTIME 2005) sec. 309 No. 7 BGB is most relevant: Even to the extent that a deviation from the statutory provisions is permissible, the following are inefective in standard business terms: . . . a) any exclusion or limitation of liability for damage from injury to life, body or health due to negligent breach of duty by the user or intentional or negligent breach of duty by a legal representative or a person used to perform an obligation of the user; b) any exclusion or limitation of liability for other damage arising from a grossly negligent breach of duty by the user or from an intentional or grossly negligent breach of duty by a legal representative of the user or a person used to perform an obligation of the user.

It follows that a German court or arbitral tribunal may render a knock-for-knock clause invalid, since it attempts to exclude liability for damages arising from (1) personal injury to life and body or health, (2) wilful/intentional misconduct (not only by representatives and employees, but the parties themselves), and (3) gross negligence. There is a considerable risk that this could happen because German courts tend to view these liability exclusions as highly detrimental and thus potentially unfair or unreasonable even in a purely commercial contract. Consequently, in such circumstances a court may consider the entire clause to be invalid, even if only a part of such clause is afected. This means – in case of any damage or loss – that the German statutory liability regime of the German Civil Code, which is generally a fault-based liability regime, would apply and the party claiming damages would be entitled to damages without any limitation, provided the other party was at fault. It follows, that the following elements of any knock-for-knock clause and the overall liability regime of a contract containing such knock-for-knock clause need to be carefully considered and potentially adapted in a German law context: • Wording that attempts to limit or exclude liability for bodily injury or death; • Wording that attempts to limit or exclude liability for gross negligence; • Wording that attempts to limit or exclude liability for intentional misconduct of employees or legal representatives of the parties; • Wording that attempts to put in place an overall liability cap for the abovenamed types of damages or the overall damages payable under a contract. It needs to be emphasised that the above limitations are not generally invalid or prohibited. However, they are terms listed under secs. 308 and 309 that will be under scrutiny of a German court. Even commercial contracts need to be fair and reasonable to some extent and a court will look at the overall liability provisions to establish whether the whole of the contract – on balance – could be to the clear detriment of one party. Including all the above-mentioned liability limitations or even exclusions and low overall caps could tip the balance towards unreasonableness and make a knock-for-knock clause fail the Content Test. 3.4 Invalid terms under German civil law – no exclusion of liability for intent Aside from the Content Test applied to general terms and conditions, there are two elements of a typical knock-for-knock clause that are invalid in any case, irrespective 199

eckehard volz and anna-sophie waldmann of whether it is individually negotiated or not and whether the overall balance of the contract is achieved or not. Sec. 276 para. 3 BGB does not allow the exclusion or limitation of liability for wilful or intentional misconduct (Vorsatz) of a party to a contract. And German law also does not allow contractual agreements to the detriment of third parties that are not signatories to the contract at hand. To illustrate this, we will briefy explain the general concept of liability for fault under German law. 3.4.1 General principle of liability under sec. 276 BGB The central civil law norm in Germany for fault-based liability for breaches of both contractual and statutory obligations is sec. 276 BGB: Sec. 276 Liability of the debtor (1) (2) (3)

The debtor is responsible for intent and negligence if a stricter or milder liability is neither stipulated nor can be inferred from the other content of the obligation, in particular from the assumption of a guarantee or a procurement risk. . . . Any person who fails to exercise the due care required in the course of ordinary business shall be guilty of negligence. Liability for intent may not be waived in advance for the debtor.

Sec. 276 thus contains an attribution rule, which determines under which circumstances non-performance, bad performance, delay or breaches of statutory duties (comparable to tort) are attributed and claims are established. However, it is not a specifc basis for a claim.4 Sec. 276 para. 1 sent. 1 BGB contains the legislative decision that the principle of fault requires (simple) fault,5 which results in a fundamentally fault-based liability regime of German statutory law. Fault-based liability means the obligation to pay damages based on conduct that is not only objectively unlawful but also personally culpable. The principle of fault, which is the basis of sec. 276 (1) sent. 1 BGB, binds liability to the prerequisite that the damage or unlawful condition has been caused by a reproachable conduct of the person against whom a claim is made.6 The BGB itself does not explicitly defne the term “fault”. According to the history and system of the law, it is the generic term for the two forms of fault mentioned in para. 1 sent. 1, namely intent and negligence. In both forms, fault is an expression of subjective reproachability. Intent means the knowledge and intention of the realisation of the facts, i.e., the deliberate breach of duty. The law does not provide a defnition of intent. The so-called intent theory defnes it for private law by an intellectual and a voluntary element, each of which must extend to the unlawfulness of the result: intent is the knowledge and intention of the unlawful result.7 Negligence, on the other hand, according to the legal defnition of sec. 276 para. 2 BGB, means disregarding the care required in the ordinary course of business. In civil law, a distinction is made between two diferent forms of negligence: simple negligence and gross negligence. To determine the degree of care required in the 4 BGH, judgement dated 13. 11. 1953 – I ZR 140/52 _ NJW 1954, 229; MüKoBGB/Grundmann, 8. Auf. 2019, BGB § 276 Rn. 1–2; Jauernig/Stadler, 18. Auf. 2021, BGB § 276 Rn. 1. 5 BeckOGK/Schaub, 1.3.2021, BGB § 276 Rn. 4; MüKoBGB/Grundmann, 8. Auf. 2019, BGB § 276 Rn. 1–2. 6 BGH, judgement dated 09.07.1992 – VII ZR 7/92 = NJW 1992, 3158. 7 BGH, judgement dated 20. 11. 2012 – VI ZR 268/11 = NJW-RR 2013, 550; HK-BGB/Reiner Schulze, 10. Auf. 2019, BGB § 276 Rn. 6; MüKoBGB/Wagner, 8. Auf. 2020, BGB § 823 Rn. 47.

200

applying knock-for-knock in germany ordinary course of business, the respective circles/groups of the public must be considered, considering the special features of the type of transaction and the groups of persons typically involved in it.8 Gross negligence requires that the care required in the ordinary course of business is violated to a particularly serious degree. This is to be assumed if something was not observed that should have been obvious to anyone in the given situation,9 obvious and simple considerations were not considered or concerns that should have been obvious due to typical indications of the danger were recklessly disregarded. According to sec. 276 para. 1 sent. 1 BGB, liability for intent and negligence only arises if a stricter or milder liability is neither stipulated by law or contract nor can be inferred from the other content of the obligation. This means that the law itself exceptionally provides for strict liability or allows it by contract. Contractual extensions of liability are permissible by individual agreement within the limits of secs. 138, 242 BGB. These prohibit immoral agreements and prescribe conduct that is guided by the principle of good faith. Examples of this are the guaranteed agreement or the assumption of strict liability. Finally, as indicated already above, sec. 276 para. 3 BGB prohibits the waiver or exclusion of liability for intentional misconduct in prior to the damage having occurred. 3.4.2 No agreements to the detriment of third parties As a general concept of German civil law, contractual agreements to the detriment of third parties that are not part of the agreement itself, are invalid.10 This afects knock-for-knock clauses where they establish a “Contractor’s Group” and “Employer’s Group” concept, where subcontractors and afliates companies are included in the knock-for-knock liability regime. The intention of the knock-for-knock regime and the concept of establishing “groups” of companies that fall on either side of the liability spheres is easy to understand: it means to be a catch-all clause to establish a back-to-back liability in a multi-party project contract, with several subcontractors and supplies and service providers involved during all stages of the project. From a German law point of view, however, including third parties in such an agreement is highly problematic, because it attempts to take away title to sue or claim damages from such third parties. Therefore, a German court would consider this an agreement to the detriment of a third party, which is invalid. 3.4.3 Consequences for knock-for-knock clauses under German law Because of sec. 276 para. 3 BGB, even if the clause has been individually negotiated prior to the signing of the contract, a knock-for-knock clause will be deemed invalid under German law if it aims to exclude liability for wilful or intentional misconduct. 8 BGH, judgement dated 02.10.1985 – IV a ZR 18/84 = NJW 1986, 1100. 9 BAG, judgement dated 12. 11. 1998–8 AZR 221–97 = NJW 1999, 966; BGH, judgement dated 18.12.1996 – IV ZR 321/95 = NJW 1997, 1012. 10 BGH, judgement dated 12.11.1980 – VIII ZR 293/79 = NJW 1981, 275; BVerfG, judgement dated 23. 04. 1986–2 BvR 487/80 = NJW 1987, 827; BeckOGK/Mäsch, 1.1.2021, BGB § 328 Rn. 123; BeckOK BGB/Janoschek, 58. Ed. 1.5.2021 Rn. 5, BGB § 328 Rn. 5; MüKoBGB/Gottwald, 8. Auf. 2019, BGB § 328 Rn. 261.

201

eckehard volz and anna-sophie waldmann Hence a valid knock-for-knock clause under German law must always contain a “carve-out” for this damages. In contrast, the attempt to govern rights and claims of third parties that are part of the Contractor’s Group or Employer’s Group would likely not as such render the knock-for-knock clause invalid, because it does not directly afect either of the parties to the contract. But it is important to note, that neither of the parties to the contract would be able to refer to the knock-for-knock clause between them, if a third party were to make claims against either of them. 4 Available case law Knock-for-knock clauses have not yet been the subject of any published German court case or arbitral award, so it remains to be seen how a German court would treat the above illustrated legal issues. There are many uncertainties in this regard, starting with the question of whether the clause will be considered part of general terms and conditions, to the question to what extent the court will accept the clause as common industry practice and fnally what limits will be placed on the efects of the clause. 5 Practical consequences and advice for use of knock-for-knock clauses under German law Considering the above and the uncertainty given that the above addressed issues have not been decided on in the context of a knock-for-knock clause subject to German law by German courts or tribunals, it is very difcult to give fully reliable advice and predict a precise outcome to its validity under German law. Nevertheless, if the contract shall be governed by German law, the best approach seems to be to avoid agreeing on knock-for-knock clauses by way of standard terms, but rather to enter into an individual agreement. To do that, it would be advisable to: 1 Never refer to standard contracts only, but to always customise a contract for the specifc purposes. 2 Before entering into a contract under German law, allow the other side to propose and negotiate the wording of the knock-for-knock clause. By doing so, it is more likely that an individual agreement as described above will be assumed. When negotiating a contract the drafting party should always express its willingness to really negotiate, i.e., no clause of the contract, in particular the clauses on liability, should be “set in stone” from the beginning and throughout the proceedings of negotiations. 3 Document that negotiations took place, ideally by way of clear written correspondence between the parties. One could refer to “common custom” and the advantages of a knock-for-knock clause to demonstrate that both parties really wanted to implement this particular liability regime and be a part of the contract. 4 If necessary, change the knock-for-knock clause to be compatible with German law or, alternatively, choose English law to govern the contract as a whole or parts of the contract, the liability regime (if so, preferably as a 202

applying knock-for-knock in germany whole) in particular. A partial choice of law clause is permissible under German law, if the division is clear and does not result in contradicting provisions within the contract as a whole. As far as the diferent sub-questions for the chosen law regimes, however, cannot be consolidated with each other in an appropriate manner, i.e., are deemed contradictable, the applicable law will be determined by reverting to objective connecting factors. Please note that these recommendations will not automatically make a knock-forknock clause contained in standard terms valid under German law, and legal counsel should always be sought. However, it will strengthen the position to argue that an individual contract was entered into, and that for this reason the strict rules of the German Civil Code do not apply.

203

CHAP TER 11

Knock for knock under Brazilian law Felipe T. Boechem and Márcio Opromolla

1 Introduction Historically, knock-for-knock clauses have not deserved a lot of attention from Brazilian scholars. This is not due to the negligence of Brazilian scholars, but rather a consequence of the very limited (not to say rare) historical use of knock-for-knock clauses in Brazil, which also leads to very little case law on this matter. The adversarial culture consolidated in contractual relations in Brazil also plays a relevant role in the resistance to the type of risk allocation inherent to the nature of such clause. Even though Brazil is a relevant global player in the ofshore oil and gas exploration and production industry, where the knock-for-knock liability regime is more commonly adopted worldwide, it has not been a practice to adopt such kind of risk allocation regime in Brazil. Apparently, the main reason for this is that Petrobras has maintained a de facto monopoly over the Brazilian oil and gas exploration and production industry and has historically opted to implement fault-based liability in its services contracts with all kinds of suppliers, including drilling services agreements and O&M agreements. To better illustrate the prominent position of Petrobras in the Brazilian upstream industry, it operated 93% and owned 73% of all Brazilian oil and gas production in November 2021,1 even after 20 years from the opening of the market. More recently, though, this scenario started to change as Petrobras decided to implement knock-for-knock provisions in some of its services contracts, which has been a long-time request from services providers. In this sense, the debate around knock-for-knock clauses has gained momentum and tends to be further developed in the coming years. The fact that Petrobras has been conducting a relevant divestment program, which includes several upstream assets, is also opening the market for new operators that tend to drive an increase in the use of knock-for- knock provisions. Although Petrobras’ decision to adopt the knock-for-knock regime has been well received by services providers in Brazil, the risk allocation refected in the contracts ofered by Petrobras in its bid processes has received signifcant critics from the market. In part, such critics are related to the fact that knock-for-knock provisions appear as the centrepiece of a wider range of specifc indemnities, exclusions and limited

1 See Ministry of Mining and Energy, Brazilian oil and gas production report, prepared yearly (Agência Nacional de Petróleo, Gás Natural e Biocombustíveis – ANP), available at www.gov.br/anp/pt-br/ centrais-de-conteudo/publicacoes/boletins-anp/boletim-mensal-da-producao-de-petroleo-e-gas-natural.

204

DOI: 10.4324/9781003206798-13

knock for knock under brazilian law liability clauses which, depending on the wording, may undermine the cornerstone of knock-for-knock clause. Therefore, the purpose of this chapter is to contribute to a better understanding on the nature of knock-for-knock clauses and how it fts in the Brazilian legal framework with special focus on the Brazilian oil and gas industry. Section 2 of this chapter presents the civil liability regime in Brazil to get an overview of the diferent types of liability recognized by Brazilian law, to better understand the issues surrounding knock-for-knock provisions such as (1) contractual liability and torts liability, (2) fault-based liability and strict liability and (3) the concept of damages. Section 3 describes how limitation of liability clauses are regulated in Brazil with special focus on issues around its validity and efectiveness. For purposes of this chapter, we adopted a wide concept of limitation of liability clauses, including clauses that restrict or fully release a party from a liability it would bear should the contract was silent. Section 4 departs from the classifcation of knock-for-knock provisions as a limitation of liability clause to discuss some relevant issues arising from its application under Brazilian law. We also analyse the link between knock-for-knock provisions and parties’ potential insurance strategy, forming the so-called knock-for-knock system. Section 5 is dedicated to two relevant conclusions. Firstly, Brazilian law, in general, provides a supportive backdrop to knock-for-knock clauses. Secondly, knock-forknock clauses are well suited for some operations, especially in the oil and gas industry, which may lead to their increasing use by Petrobras as well as by the increasing number of independent operators in the Brazilian upstream industry. 2 Overview of civil liability regime in Brazil Knock-for-knock provisions are deemed as limitation of liability clauses in Brazil, which aim to contractually limit or release the liability of a party that would otherwise bear such liability. Before entering a more detailed analysis of knock-for-knock provisions, it is relevant to set some important considerations regarding key aspects of civil liability under Brazilian law. 2.1 Contractual liability and tort liability The obligation to indemnify arises when a duty is breached. The basis for the duty may either be a pre-existing and voluntary legal relationship (a contract) or a general legal duty (the law).2 As many other jurisdictions of both civil law and common law traditions, Brazilian law recognizes two diferent categories of civil liability in connection with the basis of such liability: (1) contractual liability, when the liability arises from the non-fulflment of a contractual obligation and (2) torts liability or non-contractual liability, when the liability arises from the breach of duty provided by law.

2 Sergio Cavalieri, Programa de Responsabilidade Civil (2021) p. 16.

205

felipe t. boechem and márcio opromolla Contractual obligations are categorized between (1) obligation of means and (2) obligation of result. Where there is an obligation of a result, liability arises if a party fails to achieve or deliver the result as contractually agreed. For example, a seller is liable for delivering the goods which was sold; a constructor must deliver the construction according to the schematics and perfectly usable conditions and so forth. Where there is an obligation of means, no liability arises for lack of delivery of the result. There is only liability for failure to take the steps reasonably required to achieve the expected result. Some examples of obligations of means are typically those of physicians (to cure a patient) and of lawyers (defending their clients’ interests) – both must employ their best eforts and technical expertise to achieve the intended result but are not liable (except for a failure to employ best eforts and technical expertise) for the outcome. Non-contractual liability is divided between (1) subjective liability and (2) strict liability, each of those containing its own subdivisions3 (which will be described in further detail below). Brazilian scholars usually highlight two key diferences between contractual and torts liability. First, in contractual liability, one party breaches a “positive” duty (one which requires action or omission), while in tort liability the ofender breaches a “negative” duty of not causing harm (e.g., a duty of care). Second, in contractual liability, as a general rule, (1) guilt is presumed for obligations of result, so the defaulting party need to prove that another rule exempted them from performance to avoid or mitigate its indemnifcation obligation; and (2) guilt must be proven by the creditor in case of obligations of means. In torts liability, as a general rule, guilt is not presumed and must be proven by the creditor/victim of non-contractual obligation.4 The distinction above, however, has nuances, as there are several rules that apply to both contractual and non-contractual liability (arts. 393,5 4026 and 4037 of the BCC). Moreover, there are peculiarities concerning consumer relations under Brazilian Consumer Law (which has its own set of rules and diferent principles), which is not within the scope of this chapter. 2.2 Fault-based liability and strict liability There are certain requirements that must be met for liability to arise whether under contractual liability or non-contractual liability. These requirements are slightly 3 Cavalieri, p. 21. 4 Important exceptions pertaining to strict liability are applicable, as will be addressed in in this chapter. 5 “Art. 393. The debtor is not liable for losses resulting from a fortuitous event or force majeure, unless he has expressly made himself liable therefor.” 6 “Art. 402. Except where otherwise expressly provided for by law, the losses and damages owed to the creditor cover, in addition to what he efectively lost, that which he reasonably failed to proft.” 7 “Art. 403. Even where non-performance results from the debtor’s wrongful conduct, losses and damages only include efective losses and lost proft that are the direct and immediate efect of non-performance, without prejudice to the provisions of the legislation governing procedure.”

206

knock for knock under brazilian law diferent for each but follow essentially the same structure: liability arises when one proves that (1) a damage occurred and (2) a conduct breached either law or contract (give or take additional requirements). Notwithstanding, liability can also be divided between (1) fault-based liability or subjective liability (responsabilidade subjetiva), which has traditionally been the general liability regime under BCC; and (2) strict liability (responsabilidade objetiva), which, although exceptional, has been gaining more applicability since the new BCC was enacted in 2002. Fault-based liability takes place when the following criteria are identifed: (1) a “guilty” conduct (intention or negligence), (2) causation, and (3) a damage. The burden of proof of these elements is generally placed on the plaintif (creditor/injured party) (art. 373, I8 of the Brazilian Code of Civil Procedure9). As to strict liability, Brazilian law adopts the “theory of risk”, whereby agents are deemed to have assumed the risk embedded in their activity and are therefore liable even when there is no “guilt” (intention or negligence; e.g., art. 927, sole paragraph,10 and 93111 BCC). Therefore, one need only to prove the damage and causation to entail liability. In this sense, strict liability is triggered by (1) evidence of the act or omission (i.e., evidence of its occurrence, and not of the fault or intention motivating it); (2) the damage allegedly sufered by the party; and (3) the link of causation between the conduct and the damage. Strict liability applies to cases specifed in the law such as consumer rights under Law No. 8,078/1990 and environmental matters pursuant to Law No. 6,938/81) or when the ofender’s activity, by its nature, implies inherent risks to third parties (e.g., a carrier of hazardous or fammable substances). 2.3 Indemnifable losses The BCC sets out the general legal regime regarding indemnifcation of damages in connection with both contractual liability and torts liability. Accordingly, only actual damages (danos emergentes) and loss of profts (lucros cessantes) supported by the aggrieved party which are considered a direct and immediate efect (por efeito dela direto e imediato) of the breach of a legal or contractual obligation by the other party are indemnifable. Therefore, hypothetical damages (meaning those that are not a direct consequence of the conduct) are not indemnifable under Brazilian law. Allowing indemnifcation for hypothetical damages would ultimately create a situation of unjust enrichment (art. 884 BCC12).

8 “Art. 373. The burden of proof is on: I – the plaintif, as to the constitutive facts of his or her right.” 9 Federal Law no. 13.105 of 16 March 2015. 10 “Art. 927. Anyone who, through an illicit act (arts. 186 and 187), causes damage to another is obligated to repair it. Sole paragraph. The obligation to repair the damage will exist, regardless of fault, in the cases specifed by law or when the activity normally carried out by the person who caused the damage entails, by its nature, risk to the rights of others.” 11 “Art. 931. Save for other cases provided for in special legislation, sole proprietors and companies are liable for the damage caused by the products put into the market, regardless of fault.” 12 “Art. 884. Anyone who, without just cause, enriches themselves at the expense of others, will be obliged to repay the unduly earned, after updating the monetary values.”

207

felipe t. boechem and márcio opromolla Actual damages or positive damages consist of the efective reduction or loss of assets sustained by the aggrieved party. A typical example is the damage sustained by a car in the case of a trafc accident (the damages payable being the amount needed to repair the vehicle). Loss of profts (lucros cessantes) or negative damages consist of values that an injured party fails to receive/earn due to the illicit act. For example, in the case of the trafc accident, the taxi driver involved in the accident may claim loss of profts for the revenue lost due to inactive days. Loss of profts may also be present when a company loses revenues due to a breach of contract, for example. As one can note, the concepts of “direct damages”, “indirect damages” and “loss of profts” are relatively broad and are not further detailed in the law. Therefore, the defnition of which damages are indemnifable will need to be assessed on a case-bycase basis considering the facts and evidence of each specifc case. It is worth highlighting that the expression “consequential loss” is not used by Brazilian law. In this sense, if the expression “consequential loss” is used in a contract – which is often the case in the Brazilian oil and gas industry – it is recommended to include a defnition of its content in the contract to mitigate disputes around its meaning. Indemnifable damages under Brazilian law can also be classifed in the following broad categories: material, moral, aesthetic, loss of a chance, and social or difuse. Other categories exist but are seldomly applied. Material damages (danos materiais) are losses that afect the physical property of a person, legal entity, or depersonalized entity. Moral damages (danos morais) are due when there is a breach of personal rights, such as the right to privacy, freedom, sexual or religious preferences and so forth. The monetary indemnifcation for this kind of damage seeks to partially mitigate the distress and emotional consequences of an immaterial harm which is caused to an injured party. Aesthetic damages (danos estéticos) are treated as a separate form of moral damage. As a rule, these damages occur when a person sufers wounds, scars, cuts, injury or loss of organs, amputations, or other anomalies that afect human dignity. It concerns an existential scope of the aggrieved party and aims at repairing the dread stemming from this sort of damage. Such damages are presumed (in re ipsa) and need no proof other than the demonstration of the damage. The loss of a chance (perda de um chance) occurs when an aggrieved party sufers a frustration of a reasonable expectation or of a future opportunity (that would take place to a degree of certainty) due to the illicit act. Had circumstances followed their normal course, chances would not have been lost and the aggrieved party would reasonably and to a degree of certainty be in a diferent fnancial situation. It is a case law construction and is not frequently relied upon by plaintifs. Therefore, for this sort of damages to be considered indemnifable, the relevant chance must be serious and real13 (proof of which must be made by the plaintif). Therefore, liability for loss of a chance is contingent on the degree of certainty of the 13 “Civil liability for loss of chance is not limited to the category of moral damages, since, depending on the circumstances of the specifc case, the lost chance may also have the legal nature of material damage. The chance must be serious and real, not being restricted to a priori percentages” (Pronouncement No. 444, V Jornada de Direito Civil).

208

knock for knock under brazilian law event that was frustrated due to the illicit act. For this reason, the lost chance must represent much more than a simple subjective hope, and proof is usually complex. Social damages are difuse/collective, involving rights in which the victims are indeterminate or indeterminable. These damages are usually related to consumer issues, environmental and other aspects of collective nature. Difused rights are those that deserve special protection, as they do not afect anyone in particular and, at the same time, everyone. Examples of difused rights are the rights to a healthy environment, the prohibition against misleading advertising and the right to public safety. 3 Limitation of liability clause The limitation of liability clause changes the ordinary consequences of a contractual default and contractually implement between the parties only an agreed risk allocation that difers from the risk allocation that would apply should the parties remain silent in the contract. It enables the parties to make an efcient management of the project risks and reallocate them to enable a project that otherwise would not be feasible or would be feasible in a less efcient arrangement. In this sense, the non-liability clause is aligned with the main social and economic function of contracts, which is the promotion of the efcient fow of market relations. In Brazil, there is no piece of legislation that sets forth a general prohibition or imposes conditions to a limitation of liability clause. Consequently, pursuant to the freedom to contract principle,14 the parties are entitled to agree on the non-liability clause. In specifc and justifed situations, the Brazilian law forbids such type of provision, such as in consumer and transportation contracts. In consumer contracts, the natural unbalance between the consumer and the supplier can justify such legal prohibition. Most Brazilian scholars and judicial precedents recognize the validity of the nonliability clause as a legitimate manifestation of contractual freedom, but they subject such validity to the following conditions: 1 Both parties shall agree with the limitation of liability clause (bilateral consent), not being admitted unilateral declarations of a party with the purpose of limiting its own liability. 2 There must be some contractual balance among the parties, which would prevent such provisions, for example, in most adhesion contracts and employment agreements. 3 It cannot be agreed with the purpose of excluding or transferring essential obligations of the parties as otherwise it would signifcantly modify the main contractual obligation without which the contract would not have been executed. 4 It does not limit the liability in case of wilful misconduct or gross negligence under the basis that the duty to indemnify for losses arising from acts performed with wilful misconduct or gross negligence is a principle of civil liability with status of public order. 5 It does not result in violation of the public order. 14 Article 5, II of the Brazilian Federal Constitution.

209

felipe t. boechem and márcio opromolla For the purposes of this chapter, we will provide a few considerations on the fourth and ffth conditions, which trigger more debates. The majority of Brazilian scholars generally argue that the provision limiting the indemnifcation obligation in case of wilful misconduct or gross negligence (1) would correspond to a contractual exemption for the party to default its contractual obligation as if, in practice, the contractual obligation was not contracted and, consequently would be a violation of the parties’ good faith duty which is expressly recognized under the Brazilian Civil Code; and (2) could lead to a situation in which signifcant damages caused by such acts would remain unpunished, so such provision would promote undesirable social impacts and be against the public order. The frst argument is refuted by some scholars who support that the indemnifcation obligation is not the only available remedy to enforce a party to fulfl its contractual obligations. There are other means to enforce a contractual obligation, such as the possibility of the non-defaulting party seeking an injunction, claiming contractual fees, terminating the contract or not fulflling its own obligations due to such default. According to these scholars, the other remedies cannot be ignored when discussing the validity of the limitation of liability clause. Regarding the second argument, it shall be pointed out that the concept of public order is abstract and vague. Unfortunately, Brazilian scholars do not clearly defne the content of public order. Notwithstanding, two basic ideas are usually involved in the discussions of this matter. The frst idea is that the concept of public order mandatorily involves a public interest that cannot be waived. Under this perspective, the argument to support the invalidity of the provision excluding the indemnifcation obligation in case of wilful misconduct or gross negligence would be that it creates incentives for the parties to adopt negligent behaviours in the fulflment of their contractual obligations that could have material adverse impacts not only on the counterparty but also on third parties (externalities). Although one may say that this argument is reasonable, it cannot be applied to all and any situation without analysing the specifcities of each case. The second idea is that, in the context of contract interpretation, the public order shall be construed as an exception to the principle of contractual freedom which is a fundamental pillar of the theory of contracts. Following the general interpretation principles, the interpretation of an exception shall be restrictive, which means that it should avoid expanding the situations covered by such exception. It is necessary to demonstrate that the situation under analysis would be detrimental to the public interest because ultimately it is on behalf of the public interest (and not the private interests) that it will be allowed a potential judicial intervention to invalidate the non-liability clause. Expanding the concept of public order to situations in which there is no clear public interest to be protected could result in unfair decisions. More recently, the arguments for the validity of limitation of liability provisions were strengthened by Law 13,874/2019, which included a new provision in the Brazilian Civil Code to state that civil and commercial contracts shall be presumed as balanced, unless there are actual elements to the contrary, and the risk allocation defned in such contracts shall be respected and observed. 210

knock for knock under brazilian law 4 Knock-for-knock clauses in Brazil Knock-for-knock clauses are contractual provisions used worldwide, which basically defne that each party shall bear the damages caused to its people, goods, equipment, and properties regardless of whether such damages were caused by the other party. To quote a simple example, see the following language used by Petrobras in a drilling contract:15 The CHARTERER undertakes to indemnify PETROBRAS Group for any and all Damages involving: i. The CHARTERER GROUP personnel. ii. goods or facilities owned by the CHARTERER Group PETROBRAS undertakes to indemnify the CHARTER Group from any and all Damages involving: i. PETROBRAS Group personnel. ii. goods or facilities owned by PETROBRAS Group.

From the analysis of the knock-for-knock provisions, it is possible to extract some relevant considerations about them. First, the knock-for-knock provision has the nature of a limitation of liability clause under Brazilian law, as it releases the party that caused the damages from its duty to indemnify the party that sufered such damages. Please note that the general liability regime set forth by Brazilian legal system imposes on the ofender the duty to indemnify the damages caused by its acts or omissions, so the knock-for-knock clause changes the ordinary liability allocation provided by law. As previously mentioned, it shall be emphasized that, as a general principle of contractual law, knock-for-knock provisions are binding among the parties to the contract only. Second, knock-for-knock clauses in the oil and gas industry are usually structured with an initial provision covering damages to people and property as the one mentioned above and subsequent clauses covering other specifc issues like pollution. In drilling services agreements, for example, it is common to have specifc provisions regulating (1) pollution arising from the equipment above the rotatory table, which is usually allocated to the contractor; (2) pollution arising from equipment owned by the contracting party, equipment below under the rotatory table, wells and reservoir, which are usually allocated to the contracting party; and (3) big risks such as damages to wells, reservoirs and other geological formations, as well as damages arising from blowout and the use of radioactive substances, which are usually allocated to contracting party. Some Brazilian scholars argue that the principle of the human dignity, which is defned as a fundamental principal of the Brazilian Constitution, prevents the possibility of limitation of liabilities for damages in connection with the life and physical integrity of an individual and, consequently, the validity of knock-for-knock clauses could be challenged specifcally regarding the release from damages caused to people. We understand, though, that the release, operated by a knock-for-knock provision in connection with damages caused to people, should be valid and enforceable under Brazilian law. The knock-for-knock provision inserted in commercial contracts does 15 “Look before you leap: are your oil patch liability clauses enforceable? (An analysis under civil law jurisdictions with emphasis on Brazil)” (2021) Journal of World Energy Law and Business 14, 49, at 65.

211

felipe t. boechem and márcio opromolla not restrict any legal recourse that an individual may have against any of the parties to the contract as such commercial contract is binding among its parties only (does not bound such individual). For example, if “Party A” causes damage to an employee of “Party B”, such employee will be entitled to seek indemnifcation from Party A (torts liability) and, if Party A pays for such indemnifcation, Party A will be entitled to seek reimbursement from Party B. Having said that, we understand that the human dignity principle is not, as general rule, undermined by knock-for-knock provisions. Third, knock-for-knock clauses usually cover both contractual liability and torts liability. Therefore, “Party A” would contractually bear the damages to its personnel and equipment regardless of whether such damages were caused by Party A, another party to the contract or a third party. Fourth, although the knock-for-knock provision is agreed in a contract between its parties (i.e., contractor and contracting), it usually covers not only damages to equipment and personnel of such parties but also damages to the equipment and personnel of their respective group (controlling companies, controlled companies, and companies under common control). In this sense, contractor will bear all the damages not only to its own equipment and personnel but also to the equipment and personnel of other companies of the “contractor group”. Fifth, companies usually opt to contractually exclude the applicability of a knockfor-knock provision in case of wilful misconduct of the party that caused the damage. However, there is often negotiation around limiting such cases to wilful misconduct of the senior supervisory personnel (SSP) of the parties and the defnition of the SSP. In such cases, contractors tend to push for higher thresholds (if possible excluding people working in loco), while the contracting parties tend to push for lower thresholds. Finally, it is important to give some special consideration to the interconnection between knock-for-knock provisions and the insurance strategy of the parties. By agreeing to traditional knock-for-knock liability, the parties somehow recognize that they will bear the risk of their own equipment and people and, therefore, defne their risk and insurance strategy. The allocation of liability promoted by knock-for-knock provisions simplifes the distribution of risks by, for example, avoiding several discussions on who caused the damages and the extent of its liability. Such risks are then more easily identifable and channelled to one of the parties, which enables an efcient management of risk, including by means of insurance. That is why experts on this topic often refer to knock-for-knock provisions and insurance coverage as the “knock-for-knock system”. 5 Conclusion Based on the analysis above, it is possible to extract two relevant conclusions. First, Brazilian law provides a supportive backdrop to knock-for-knock clauses, even though it contains certain broad and vague concepts that may be somehow used to restrict the applicability of such provisions. As mentioned during this paper, Brazilian law supports the freedom to contract and, consequently, the allocation of risks contracted by sophisticated parties in balanced contractual position. Having said that, the Brazilian oil and gas industry, especially in ofshore exploration and 212

knock for knock under brazilian law production, tends to be a feasible feld for the development of the knock-for-knock provision. Notwithstanding, special and careful attention should always be given to the specifcities of each case. Second, although the knock-for-knock provisions have only started to be more frequently used in the Brazilian oil and gas industry in the past few years, we presume its use will increase considering (1) its benefts in terms of simplifcation of risk allocation and the possibility of developing a more rational and efcient insurance coverage, (2) Petrobras’ openness to use the clauses in its contracts and (3) more independent companies are starting to operate in Brazilian upstream industry demanding the use of these clauses.

213

CHAP TER 12

The implications of litigation post Deepwater Horizon on knock-for-knock clauses in US law Cindy Matherne Muller, Jeanne Amy, and Jennifer David

1 Introduction For many decades, operations in the oil and gas industry have been conducted under agreements with indemnity obligations that, with some variations, require a party to indemnify the counterparty for certain claims when caused by the counterparty’s negligence. This is known in the industry as “knock-for-knock” indemnity obligations, which are often agreed to with respect to personal injury and property damage claims that could be brought by each party, their respective afliates, contractors, and employees against the counterparty. This indemnity scheme is designed to create a “no-fault” system for litigation and places the risk of the loss on the party who carries frst-party insurance coverage for the respective losses. The indemnity scheme also avoids having two potential tortfeasors litigate the fault of each other to the beneft of the injured party and his attorney. For ofshore operations of the coast of the United States, the threshold issue in determining the enforceability and scope of any indemnity provision is what law will apply. The choice of law issue is not as simple as honoring the choice of the parties. Instead, a practitioner has to consider the potential application of federal law, maritime law, the Outer Continental Shelf Lands Act (OCLSA),1 or state law. The applicable law could be outcome determinative, as various states have enacted “antiindemnity” statutes, which, in varying degrees, render indemnity obligations which purport to indemnify a party against its own negligence unenforceable as a matter of public policy. These statutes vary in breadth with respect to the unenforceability of the indemnity and related insurance obligations. Care must be taken in drafting indemnity and insurance provisions to take advantage of any statutorily permissive structures to avoid unenforceability. Key to any drafting or enforcement is an understanding of what law could, or is most likely to, apply. 2 Applicable law A state or federal court in the United States will have to determine the applicable law to any contract containing indemnity provisions. Courts may consider the parties’ choice of law in the contract, but certain limitations on the parties’ freedom of contract are imposed by the OCSLA. The conficts of law issue is complicated by 1 43 U.S.C. § 1331 et seq.

214

DOI: 10.4324/9781003206798-14

the implications of litigation post deepwater horizon dual state and federal legal systems under which federal law can pre-empt state law. Moreover, a federal court could have jurisdiction either under federal question, in which a federal statute is applied, or over a controversy on the basis of diversity of citizenship of the parties, or under admiralty jurisdiction. The basis of jurisdiction could afect the confict of law analysis employed by the court. 2.1 Federal court jurisdiction Article III of the US Constitution is the basis for the formation and jurisdiction of US federal courts. Article III states: The judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority; .  .  . to all Cases of admiralty and maritime Jurisdiction; – to Controversies . . . between Citizens of diferent States.2

This jurisdiction is commonly broken into diferent categories, including federal question jurisdiction, admiralty jurisdiction, and diversity jurisdiction. 2.1.1 Federal question jurisdiction Lower federal courts have limited jurisdiction and are empowered to hear only those cases authorized by a federal statute, the US Constitution, or a US treaty.3 This is referred to as “federal question” jurisdiction. A district court (trial court) has federal question jurisdiction when an action arises under the Constitution, laws, or treatises of the United States.4 A case arises under federal law if a well-pleaded complaint establishes that either (1) federal law creates a cause of action or (2) the plaintif’s right to relief necessarily depends of the resolution of a substantial question of federal law.5 For most cases brought under federal question jurisdiction, the suit arises under the federal law that created the cause of action.6 As a general rule, state courts have concurrent jurisdiction over federal claims with the federal courts unless (1) Congress provides that the federal courts have exclusive jurisdiction or (2) there is a clear incompatibility between the state court jurisdiction and the federal interest.7 When jurisdiction is concurrent, a cause of action can be pursued in either state or federal court. Admiralty and maritime cases8 and Jones Act maritime personal injury cases9 are subject to the concurrent jurisdiction of state and federal courts. 2.1.2 Diversity jurisdiction Federal courts have diversity jurisdiction when the suit involves a controversy between parties of diverse citizenship and the amount in controversy exceeds $75,000.10 When 2 U.S. Const. art. 3, § 2, cl. 1. 3 Kokkonen v Guardian Life Ins. Co. of Am., 511 U.S. 375, 377 (1994). 4 U.S. Const. art. 3, § 2; 28 U.S.C. § 1331. 5 Empire HealthChoice Assur., Inc. v McVeigh, 547 U.S. 677, 689–690 (2006). 6 Merrell Dow Pharmaceuticals, Inc. v Thompson, 478 U.S. 804, 808 (1986). 7 Gulf Ofshore Co., Div. of Pool Co. v Mobil Oil Corp., 453 U.S. 473, 478 (1981). 8 28 U.S.C. § 1333; see Ofshore Logistics, Inc. v Tallentire, 477 U.S. 207, 222–223 (1986) (“savings to suitors” clause allows state courts to entertain in personam maritime causes of action). 9 Engel v Davenport, 271 U.S. 33, 37 (1926); see 46 U.S.C. §§ 30104, 30105. 10 28 U.S.C. § 1332(a).

215

cindy matherne muller, jeanne amy, and jennifer david jurisdiction is based on diversity, there must be complete diversity of citizenship between the parties, which requires that no plaintif can be a citizen of the same state as any defendant.11 Determining diversity of citizenship is factually based and diferent rules apply to determine the citizenship of corporations and limited liability companies. Moreover, when a court is sitting in diversity jurisdiction, the court must apply the conficts of laws rules under the law of the forum state, or the state in which the federal court is located.12 The choice of law rules of a forum state are varied, but many follow an interest analysis under which the court applies the law of the state with the most interest in the controversy. Such a determination is also highly fact dependent. 2.1.3 Admiralty jurisdiction Under 28 U.S.C. § 1333(1), US federal district courts are granted “original jurisdiction, exclusive of the courts of the United States, of any civil case of admiralty or maritime jurisdiction.” This grant includes jurisdiction “over all contracts . . . which relate to the navigation, business, or commerce of the sea.”13 A maritime contract is “one that relates to the ship in its use as such or to commerce or to navigation on navigable waters or to transportation by sea or to maritime employment.”14 Types of contracts which have been held to invoke maritime jurisdiction include, but are not limited to: contracts to furnish services, supplies, or accessories to a particular vessel; charter parties; contracts for carriage of goods and afreightment; stevedoring contracts; contracts for the carriage of passengers by water; maritime insurance contracts; salvage contracts; towage contracts; and mortgages covered by the Ship Mortgage Act of 1920,15 cargo claims subject to the Carriage of Goods by Sea Act,16 and the Harter Act of 1893.17 Maritime tort jurisdiction is based either upon statute or by law, a judicial test based upon situs and status which includes relationships to and efect upon maritime commerce. The US Supreme Court noted that the boundaries of admiralty jurisdiction over contracts – as opposed to torts or crimes – are “conceptual rather than spatial” and have always been difcult to draw.18 Contracts containing both maritime and non-maritime elements (“mixed contracts”) are maritime contracts for the purposes of federal admiralty jurisdiction where the focus of the contract is maritime regardless of the locale. In recent cases, the Supreme Court noted that the fundamental interest giving rise to maritime jurisdiction is the protection of maritime commerce.19

11 Owen Equipment & Erection Co. v Kroger, 437 U.S. 365, 374 (1978). 12 Klaxon Co. v Stentor Elec. Mfg. Co., Inc., 313 U.S. 487 (1941); Erie R.R. Co. v Tompkins, 304 U.S. 64 (1938). 13 De Lovio v Boit, 7 Fed. Cas. 418 (C.C.D. Mass. 1815) (No. 3776). 14 See J.A.R., Inc. v M/V LADY LUCILLE, 963 F.2d 96 (5th Cir. 1992). 15 Now known as the Commercial Instruments and Maritime Lien Act, 46 U.S.C. § 31301 et seq. 16 46 U.S.C.S. § 30701. 17 46 U.S.C. §§ 30701–30707. 18 Kossick v United Fruit Co., 365 U.S. 731, 735 (1961). 19 Norfolk Southern Ry. Co. v James N. Kirby, Pty Ltd., 543 U.S. 14, 25 (2004) (citing Exxon Corp. v Cent. Gulf Lines, Inc., 500 U.S. 603 (1991)).

216

the implications of litigation post deepwater horizon 2.2 Applicable law Work on the Outer Continental Shelf is performed on a wide variety of structures, including vessels, drill ships, jack-up rigs, lift boats, semi-submersible platforms, single point anchor reservoir (SPAR) and tension leg structures, foating production platforms (FPP), and foating production and storage and ofoading (FPSO). Whether these structures are considered vessels is a key component to any choiceof-law analysis, as the use of a vessel is a critical factor in determining whether a contract is maritime and whether maritime law, the OCSLA, or state law might apply. 2.2.1 Outer Continental Shelf Lands Act The OCSLA makes it clear that federal law applies ofshore: The Constitution and laws and civil and political jurisdiction of the United States are extended to the subsoil and seabed of the outer Continental Shelf and to all artifcial islands, and all installations and other devices permanently or temporarily attached to the seabed, which may be erected thereon for the purposes of exploring for, developing, or producing resources therefrom or any such installation or other device (other than a ship or vessel) for the purposes of transporting such resources, to the same extent as if the outer Continental Shelf were an area of exclusive Federal jurisdiction located within a State.20

In the Energy Policy Act of 2005, Congress amended the OCSLA to authorize the regulation of the development of ofshore energy from sources other than oil and gas. Under 30 C.F.R. § 585.100, this authority was delegated to the Bureau of Ocean Energy Management, which exercises its authority in multiple phases for leasing, siting, and permitting. Recently, in the National Defense Authorization Act for 2021, Congress amended the OCSLA to extend its application in certain instances to “nonmineral energy resources,” which removed any question that the OCSLA would apply to ofshore wind development in the United States. Congress recognized that federal law is incomplete – e.g., there is no federal law of contract, tort, liens, etc. Therefore, Congress adopted the law of the adjacent state as surrogate federal law where necessary to fll in gaps in federal law: To the extent that they are applicable and not inconsistent with this subchapter or with other Federal laws and regulations of the Secretary now in efect or hereafter adopted, the civil and criminal laws of each adjacent State, now in efect or hereafter adopted, amended, or repealed are hereby declared to be the law of the United States for that portion of the subsoil and seabed of the outer Continental Shelf, and artifcial islands and fxed structures erected thereon, which would be within the area of the State if its boundaries were extended seaward to the outer margin of the outer Continental Shelf, and the President shall determine and publish in the Federal Register such projected lines extending seaward and defning each such area. All of such applicable laws shall be administered and enforced by the appropriate ofcers and courts of the United States. State taxation laws shall not apply to the outer Continental Shelf.21

20 43 U.S.C. § 1333 (a)(1). 21 43 U.S.C. § 1333(a)(2)(A).

217

cindy matherne muller, jeanne amy, and jennifer david The Supreme Court confrmed that Congress intended to have adjacent state law apply rather than have federal courts create a new federal common law.22 In Chevron, the Supreme Court explained: Congress specifed that a comprehensive body of state law should be adopted by the federal courts in the absence of existing federal law. . . . Congress made clear provision for flling in the “gaps” in federal law; it did not intend that federal courts fll in those “gaps” themselves by creating new federal common law.23

Importantly, courts have consistently held that Congress’s choice of the substantive laws of the adjacent state as federal law is a mandatory choice of law, which applies irrespective of the parties’ choice of law in a contractual provision or the confict of laws rules of the adjacent state. In Matte v Zapata Ofshore Co.,24 the Fifth Circuit did not enforce a choice-of-law provision contained in a non-maritime contract opting for the application of general maritime law. The court held that enforcement of the parties’ choice of law would be in violation of the federal policy expressed in the OCSLA: that the law of the adjacent state be applied as surrogate federal law. Similarly, in Union Texas Petroleum Corp. v PLT Engineering, Inc., the court stated: UTP argues that even if OCSLA applies, the parties have chosen admiralty law through the choice of law provisions in their contracts. This argument can not prevail. Although Louisiana’s choice of law rules might enforce this choice of law provision, OCSLA will not. We fnd it beyond any doubt that OCSLA is itself a Congressionally mandated choice of law provision requiring that the substantive law of the adjacent state is to apply even in the presence of a choice of law in the contract to the contrary.25

Courts have rejected the argument that application of an adjacent state’s law requires applying that state’s choice of law principles. Having decided that state law should apply, it becomes necessary to determine which state’s substantive law governs. Rodrigue mandates that the law of the adjacent state, in this case Texas, applies. The Defendants have made an argument that in applying Texas law, this Court should apply Texas’ conficts of law rules for contracts. Defendants argue that Texas’ conficts rules would require that Kentucky’s substantive law should apply because the contract has little connection with Texas. However, while enticing on common sense grounds, this argument must legally fail. In Chevron Oil Company v Huson, supra, the Supreme Court held that a State’s confict of law rules have no relevance in an OCSLA case when a Federal court is applying adjacent state law as surrogate Federal law.26

Indeed, the Supreme Court in Chevron stated: It was the intent of Congress, expressed in the Senate Committee Report, in the Conference Report, and on the foor of the Senate, that state laws be “adopted” or “enacted” as federal law. Thus a federal court applying Louisiana law under § 1333 (a)(2) of the Lands Act is applying it as federal law – as the law of the federal forum. Since the federal 22 Chevron Oil Co. v Huson, 404 U.S. 97 (1971). 23 Id. at 104–105. 24 784 F.2d 628 (5th Cir. 1986). 25 Union Texas Petroleum Corp. v PLT Engineering, Inc., 895 F.2d 1043, 1050 (5th Cir. 1990) (footnote and citations omitted). 26 Walter Oil & Gas Corp. v NS Group, Inc., 867 F. Supp. 549, 553 (S.D. Tex. 1994) (footnote omitted).

218

the implications of litigation post deepwater horizon court is not, then, applying the law of another forum in the usual sense, ordinary confict of laws principles have no relevance.27

Thus, when a federal court has federal question jurisdiction under the OCSLA, it will apply applicable federal law. Where federal law is silent, however, the law of the adjacent state will apply. 2.2.2 State waters If a contract is performed on a platform in state territorial waters rather than on the Outer Continental Shelf, and the work has no maritime connection, the OCSLA will not operate as a choice-of-law trump card. Instead, when the OCSLA does not apply, a federal court sitting in diversity jurisdiction must follow the choice of law rules of the forum state.28 2.2.3 Maritime law Whether maritime law applies requires the establishment of maritime jurisdiction, an inquiry which usually depends on the nature of the claim: [I]n the context of oil and gas exploration on the Outer Continental Shelf, admiralty jurisdiction and maritime law will only apply if the case has a sufcient maritime nexus wholly apart from the situs of the relevant structure in navigable waters.29

To establish maritime jurisdiction for a tort, a party must show maritime situs and a connection to traditional maritime activity.30 If the dispute arises out of a maritime contract, i.e., the nature of the contract is maritime, then maritime law applies rather than state law. A maritime contract has been described as one “relating to a ship in its use as such, or to commerce or navigation on navigable waters, or to transportation by sea or to maritime employment.”31 2.2.3.1 Maritime contracts Although ofshore oil and gas exploration on the Outer Continental Shelf has been ongoing for more than 50 years, the question of whether a contract is maritime in the oil and gas context has been murky and often conficting. Recent jurisprudence has simplifed the test for determining a maritime contract. In 2004, the Supreme Court held that courts must examine the primary objective of a contract when determining whether the contract is maritime.32 In part based on this pronouncement, the Fifth Circuit announces a new test to determine whether an oil and gas service contract is maritime in nature. The Fifth Circuit issued a unanimous en banc decision in In re Larry Doiron, Inc.,33 which provided that to determine whether an oil and 27 Chevron, 404 U.S. at 102. 28 Roberts v Energy Dev. Corp., 104 F.3d 782 (5th Cir. 1997); Klaxon, 313 U.S. at 496 (holding that a federal court sitting in diversity must apply the confict of law rules of the state within which it sits). 29 Laredo Ofshore Constructors, Inc. v Hunt Oil Co., 754 F.2d 1223, 1230 (5th Cir. 1985). 30 Champagne v Tetra Applied Technologies, Inc., No. 05–299, 2006 U.S. Dist. LEXIS 37457, *13–14 (S.D. Tex. Feb. 6, 2006). 31 M/V Lady Lucille, 963 F.2d at 98. 32 Norfolk Southern Ry. Co. v James N. Kirby, Pty Ltd., 543 U.S. 14 (2004). 33 879 F.3d 568 (5th Cir. 2018) (en banc), cert. denied, 138 S. Ct. 2033 (2018).

219

cindy matherne muller, jeanne amy, and jennifer david gas service contract is maritime, a court must ask (1) whether the contract provides services to facilitate the drilling or production of oil and gas on navigable waters and (2) whether the parties expect that vessel will play a substantial role in the completion of the contract.34 There, the decision dealt with determining the maritime nature of contracts dealing with exploration and production, but the court stated that the test could apply to non–oil and gas activity that involved maritime commerce.35 Thereafter, the Fifth Circuit expanded the Doiron test and announced that it applies to all mixed-services contracts, not just those arising in the oil and gas context.36 In Barrios, the plaintif was injured while ofoading a generator from a crew boat to a barge. The plaintif’s employer had executed a master service contract with reciprocal indemnity provisions with the owner of a dock for repair work to be performed at the dock facility, including the installation of a concrete containment rail. To do the work, the plaintif’s employer chartered a barge, which moved up and down the river using a tugboat and a winch. The plaintif’s employer contracted with the barge owner to transport crew and equipment to the barge via a crew boat. The issue before the court in Barrios was whether the master service agreement was a maritime contract. If the master service contract was a maritime contract, then the indemnity provision would be enforceable under the general maritime law. But if the underlying contract was non-maritime in nature, then Louisiana law would invalidate the indemnity provision. The district court determined that the contract was a “land-based construction contract,” and therefore governed by Louisiana law. On appeal, the court reviewed whether the dock contract was maritime in nature. To make that determination, the Fifth Circuit grappled with the appropriate test to apply to this non–oil and gas contract. The court noted that only one district court had applied the Doiron test to a mixed-services contract.37 The Fifth Circuit found that Doiron should apply as written because it streamlines the inquiry and is consistent with the Supreme Court’s decision in Norfolk Southern Railway Co. v Kirby, 543 U.S. 14 (2004), which requires courts to look to the primary objective of a contract when determining whether the contract is maritime. Therefore, the court employed a broader test to determine whether the dock construction contract in Barrios was maritime in nature. The test enunciated by the Barrios court is as follows: “To be maritime, a contract (1) must be for services to facilitate activity on navigable waters and (2) must provide, or the parties must expect, that a vessel will play a substantial role in the completion of the contract.” Applying that test to the facts of the case, the Fifth Circuit determined that the frst prong was met because the construction of a concrete containment rail was to facilitate activities on navigable waters. The dock extended into the Mississippi River and the containment rail was designed to prevent materials from spilling into the river. On the second prong, the court determined that the parties contemplated that the substantial use of a vessel, namely a barge that would be shifted using a tugboat and winch, would be required to complete the contract. 34 35 36 37

Id. at n. 576. Id. at n. 52. Barrios v Centaur, LLC, 942 F.3d 670 (5th Cir. 2019). See Lightering LLC v Teichman Group, LLC, 328 F. Supp. 3d 625, 627–629 (S.D. Tex. 2018).

220

the implications of litigation post deepwater horizon Accordingly, the Fifth Circuit reversed the district court and remanded the case, holding that the master service contract was maritime in nature. Accordingly, under Fifth Circuit jurisprudence, “[t]o be maritime, a contract (1) must be for services to facilitate activity on navigable waters and (2) must provide, or the parties must expect, that a vessel will play a substantial role in the completion of the contract.”38 2.2.3.2 What is a vessel? The second prong of the test enunciated in Doiron, required a determination of whether equipment used ofshore for energy exploration and production or construction of any ofshore facility is a vessel. Under maritime law, a “vessel” is defned as every “water-craft or other artifcial contrivance used, or capable of being used, as a means of transportation on water.”39 When determining vessel status, the relevant inquiry is “whether the watercraft’s use ‘as a means of transportation on water’ is a practical possibility or merely a theoretical one.”40 “[A] ‘vessel’ is any watercraft practically capable of maritime transportation regardless of its primary purpose or state of transit at a particular moment.”41 “[A] watercraft is not ‘capable of being used’ for maritime transport in any meaningful sense if it has been permanently moored or otherwise rendered ‘practically incapable’ of transportation or movement.”42 Accordingly, the vessel status inquiry hinges on whether the vessel is practically capable of its maritime operations and “transport[ing] persons or things over water.”43 In Lozman, the Supreme Court held that a foating houseboat capable of being towed was not a vessel and emphasized that “[n]ot every foating structure is a vessel,” and that the statutory defnition codifed at 1 U.S.C. § 3 “applies to an ‘artifcial contrivance . . . capable of being used . . . as a means of transportation on water.’”44 The Supreme Court cited the meaning of “transportation” as involving the “conveyance (of things or persons) from one place to another” and noted that this defnition must be applied “in a practical, not theoretical, way.”45 The Supreme Court held that a structure does not fall within the scope of this statutory phrase unless a reasonable observer, looking at the foating structure’s physical characteristics and activities, would consider it capable to a practical degree of serving as a special purpose vessel.46 In Lozman, the Supreme Court held that a structure is not a vessel merely because it is “capable of foating, moving under tow, and incidentally carrying even a fair-sized item or two when they do so.”47 Instead, the structure must be “practically capable” of being used in maritime commerce as a means of transportation for persons or goods on water. The court focused on the physical features of the “vessel,” including

38 39 40 41 42 43 44 45 46 47

Barrios, 942 F.3d at 680. Stewart v Dutra Constr. Co., 543 U.S. 481, 489 (2005) (citing 1 U.S.C. § 3). Id. at 496. Id. at 497. Id. at 493; see also Riley v Alexander/Ryan Marine Servs. Co., 983 F. Supp. 2d 884, 887 (S.D. Tex. 2013). Lozman v City of Riviera Beach, 568 U.S. 115, 121 (2013). Id. at 121. Id. Id. Id. at 121.

221

cindy matherne muller, jeanne amy, and jennifer david whether it had a rudder, steering mechanism, unraked hull, electrical connection to land, portholes or regular doors and windows, and any means of self- propulsion – all things that a “reasonable observer” would conclude that it was not designed to transport persons or things over water. The “reasonable observer” test in Lozman makes any determination of what is a vessel a fact-intensive inquiry, leading to uncertainty. Post-Lozman decisions are instructive on how to apply the Supreme Court’s test for evaluating vessel status. In Martin v Fab-Con, Inc.,48 the court addressed whether a quarters barge used as a foating hotel for workmen at a job site qualifes as a “vessel.” The quarters barge did not have a rudder or steering mechanism, was incapable of self-propulsion, was spudded in place, and had remained stationary for the past several years.49 The court reasoned that a reasonable observer would not believe that the quarters barge was designed or capable to a practical degree for transporting people or cargo over water.50 Accordingly, the court found that the quarters barge was not a “vessel” for purposes of general maritime law.51 The Fifth Circuit’s post-Lozman decision in Baker v Director, Ofce of Workers’ Compensation Programs52 is also instructive as to the application of Lozman principles to ofshore drilling structures. In Baker, the plaintif was injured while building a housing module for use on a tension leg ofshore oil platform named BIG FOOT.53 The injury occurred while the BIG FOOT was berthed at a waterside marine fabrication yard in Houma, Louisiana.54 Ultimately, the Fifth Circuit concluded that BIG FOOT was not a vessel. In support of its reasoning, the Fifth Circuit noted that, like the foating house in Lozman, BIG FOOT had no means of self-propulsion, had no steering mechanism or rudder, and had an unraked bow.55 Furthermore, BIG FOOT could be moved only when under tow, and after being towed to the Outer Continental Shelf, it was not anticipated that BIG FOOT would relocate for approximately 20 years.56 Although BIG FOOT was required to carry a captain and crew when under tow, those individuals were present only to ensure BIG FOOT was safely towed to its permanent location on the OCS.57 In sum, the Fifth Circuit found that “a reasonable observer, looking to BIG FOOT’s physical characteristics and activities, would not consider it designed or capable to a practical degree for carrying people or things over water.”58 Like in Baker, most determinations of vessel status are not made at the Supreme Court. Instead, Fifth Circuit jurisprudence provides guidance on what devices used in ofshore oil and gas exploration have been found to be vessels.59

48 9 F. Supp. 3d 642, 649 (E.D. La. 2014). 49 Id. 50 Id. 51 Id. at 651. 52 834 F.3d 542 (5th Cir. 2016). 53 Id. at 544. 54 Id. 55 Id. at 547. 56 Id. 57 Id. 58 Baker, 834 F.3d at 547. 59 Single Point Anchor Reservoir (SPAR) platforms are not vessels. Fields v Pool Ofshore, 182 F.3d 353 (5th Cir. 1999); Mendez v Anadarko Pet. Corp. 466 F. App’x 316 (5th Cir. 2012); Tension Leg Platforms

222

the implications of litigation post deepwater horizon 2.2.4 Maritime law incomplete, supplemented with state law Courts have recognized the need for national uniformity in the substantive law governing maritime commerce. Where federal maritime law is incomplete and the issue does not require national uniformity, courts may apply state law.60 In Wilburn Boat v Fireman’s Fund Ins.,61 the Supreme Court held that where there is no established general maritime law on a particular issue, and the subject does not require national uniformity, admiralty courts may look to the rule of the state in which the court is sitting. There, the court held that state law will control the interpretation of a marine insurance policy only in the absence of a federal statute, judicially fashioned remedy, or need for national uniformity. As discussed below, state law typically governs the interpretation of insurance policies that are often used “hand in glove” with indemnity provisions. 2.2.5 Tension between OCSLA and maritime law In addition to the complications of whether state or federal courts could hear a controversy, and the diferent bases of federal court jurisdiction, given the nature of ofshore operations, an unavoidable tension exists between the application of the OCSLA and maritime law. Claims and contractual disputes arising out of work performed ofshore in the United States often involve determining the tension between the application of admiralty and maritime law and the OCSLA. In Union Texas Petroleum Corp. v PLT Engineering, Inc.,62 the Fifth Circuit set forth a three-part test which must be met for adjacent state law to apply rather than maritime law: (1) the controversy must arise on a “situs covered by the OCSLA” (i.e., on the subsoil, seabed, or artifcial structures permanently or temporarily attached thereto); (2) federal maritime law must not apply of its own force; and (3) the adjacent state’s law must not be inconsistent with federal law. With the recent Fifth Circuit decision in Doiron, it appears that the Doiron test will be applied to answer the second question of the Union Texas Petroleum test. If a vessel plays a substantial role in the completion of the contract, it is likely that maritime law will apply of its own force, which would prevent application of the law of the adjacent state as surrogate federal law. As should be apparent, the determination of applicable law, whether maritime law or the law of the adjacent state as surrogate federal law under the OCSLA, is a fact-intensive inquiry that is not always predictable. The applicable law matters and may well determine whether a knock-for-knock indemnity provision is enforceable.

are not vessels. Warrior Energy Servs. Corp. v ATP Titan, 551 F. App’x 749 (5th Cir. 2014); Baker v Dir., OWCP, 834 F.3d 542 (5th Cir. 2016); Mooney v W&T Ofshore, Inc., 2013 AMC 1480 (E.D. La. 2013). Semisubmersible drilling units or mobile ofshore drilling units (MODU) are vessels. Ofshore Co. v Robinson, 266 F.2d 769 (5th Cir. 1959); In re Oil Spill, Deepwater Horizon, 745 F.3d 157 (5th Cir. 2014). Drill ships are vessels, and Floating Production Storage and Ofoading Vessels (FPSOs) are vessels. BW Ofshore USA, LLC v TVT Ofshore AS, Case No. 14–1052, 2015 U.S. Dist. LEXIS 153840 (E.D. La. Nov. 13, 2015). Jack-up rigs are generally considered vessels. Texas Co. v Gianfala, 222 F.2d 382 (5th Cir.), reversed on other grounds, 350 U.S. 879 (1955). Although jack-ups have been held not to be vessels in navigation when attached to the sea bottom within OCSLA jurisdiction. Barker v Hercules Ofshore, Inc., 713 F.3d 208, 2013 AMC 946 (5th Cir. 2013); Demette v Falcon Drilling Co., 280 F.3d 492, 2002 AMC 686 (5th Cir. 2002). 60 The Tungus v Skovgaard, 358 U.S. 588, 1959 AMC 813 (1959). 61 348 U.S. 310, 1955 AMC 467 (1955). 62 895 F.2d 1043 (5th Cir. 1990).

223

cindy matherne muller, jeanne amy, and jennifer david 3 Indemnity in maritime Indemnity clauses in maritime contracts are usually enforceable in accordance with their clear and unequivocal terms. A contract to indemnify another for his own negligence imposes an extraordinary obligation.63 Long-established principles of general maritime law require that indemnifcation for an indemnitee’s own negligence be clear and unequivocally expressed.64 The clear terms will be enforced even when the negligence of the party seeking indemnity contributed to or caused the injury if the terms are clear that the indemnity was intended to apply without regard to the negligence of the indemnifed party.65 3.1 Enforceability of indemnity provisions in a maritime contract As a matter of frst impression, the court in In re Oil Spill by the Oil Rig “Deepwater Horizon” confronted, among other things, whether a contractual indemnity agreement that provides indemnity for a party even if caused by that party’s gross negligence in a drilling contract is enforceable under maritime law.66 That case arose out of the 20 April 2010 explosion and fre onboard the DEEPWATER HORIZON, which resulted in a signifcant loss of life and millions of gallons of oil that spilled into the Gulf of Mexico. Transocean, as the owner of the DEEPWATER HORIZON, and BP as the operator entered into a drilling contract that purportedly required BP to defend and indemnify Transocean for pollution emanating below the surface of the water, even in situations where Transocean is found to be strictly liable for the discharge or the pollution was caused by Transocean’s negligence or gross negligence.67 Both Transocean and BP fled cross-motions for partial summary judgment to determine whether such a provision was enforceable under maritime law.68 Transocean argued that the intent of the parties must be honored and the provision should be enforced as written, so BP’s indemnity obligation should extend to compensatory damages, punitive damages, and statutory penalties owed under the Clean Water Act (CWA).69 BP admitted that it owed indemnity for Transocean’s fault or negligence, but denied that it owed indemnity for claims arising under a strict liability theory (like that of a statutory requirement under the CWA) or for Transocean’s gross negligence.70 BP also argued that public policy invalidates the indemnity provision because a party cannot obtain indemnity for its own gross negligence or damages assessed for punitive damages or strict liability, i.e., civil penalties assessed pursuant to the CWA.71 63 Stewart v Grand Isle Shipyard, Inc., Case No. 10–4016, 2011 U.S. Dist. Lexis 148603 (E.D. La. December 23, 2011). 64 Id. 65 Theriot v Bay Drilling Corp., 783 F.2d 527 (5th Cir. 1986). 66 841 F. Supp. 2d 988, 1000 (E.D. La. 2012). 67 Id. at 992. 68 Id. The parties did not dispute that maritime law applied to the drilling contract. Id. at 994. 69 Id. at 992; 33 U.S.C. §§ 1251 et seq. 70 Id. at 992–993. 71 Id.

224

the implications of litigation post deepwater horizon The court addressed whether indemnifcation for gross negligence and punitive damages was against public policy.72 First, the court tackled the issue of gross negligence, which raised an inherent confict between “freedom of contract, which weighs in favor of enforcing the indemnity, and a reluctance to encourage grossly negligent behavior, which weighs against enforcing the indemnity.”73 On the public policy front, the court reasoned that the drilling contract allocated risk to both parties such that Transocean would have been incentivized not to act in a grossly negligence manner.74 When considering freedom of contract, the court determined that the parties had equal bargaining power in negotiating the drilling contract.75 Therefore, the court held that if Transocean is determined to be grossly negligent in causing the pollution, it could seek contractual indemnity from BP.76 Accordingly, a contractual indemnity provision that indemnifes for one’s own gross negligence is likely to be valid and enforceable as it relates to compensatory damages. The court next examined whether the contractual indemnity provision was enforceable as it relates to punitive damages.77 The court noted that the purpose of punitive damages is “to punish the defendant for egregious conduct, teaching him not to do it again, and to deter others from engaging in similar behavior.”78 Yet, “[i]n situations where the defendant’s conduct is so extreme as to merit an award of punitive damages, the cost of such must be placed upon the party responsible, and not transferred to a party innocent of any wrongdoing.”79 Based on this reasoning, the court held that Transocean could not shift the burden of punitive damages to BP, thereby rendering the indemnity provision as it relates to punitive damages void and unenforceable.80 Finally, the court considered whether BP owned indemnity for civil penalties that could be assessed under the CWA against Transocean.81 The CWA imposes a civil penalty on “[a]ny person who is the owner, operator, or person in charge of any vessel, onshore facility, or ofshore facility from which oil or a hazardous substance is discharged” where the discharge is “harmful to the public health or welfare or the environment of the United States.”82 The court reasoned that similar to the aim of punitive damages, the primary goals of the CWA “are to punish and deter future pollution.”83 In the same way that public policy prohibits indemnifcation for punitive damages, a provision that seeks to indemnify for civil penalties assessed under the CWA is void and against public policy.84

72 In re Oil Spill by the Oil Rig “Deepwater Horizon”, 841 F. Supp. 2d at 997. 73 Id. at 1000. 74 Id. at 1000–1001. 75 Id. at 1001. 76 Id. at 1003. 77 Id. 78 In re Oil Spill by the Oil Rig “Deepwater Horizon”, 841 F. Supp. 2d at 1003 (quoting Daughdrill v Ocean Drilling & Expl., 665 F. Supp. 477, 481 (E.D. La. 1987)) (internal quotation marks omitted). 79 Id. (quoting Daughdrill, 665 F. Supp. at 482) (internal quotation marks omitted). 80 Id. 81 Id. 82 33 U.S.C. § 1321(b)(3), (4), and (7). 83 In re Oil Spill by the Oil Rig “Deepwater Horizon”, 841 F. Supp. 2d at 1004. 84 Id. at 1006.

225

cindy matherne muller, jeanne amy, and jennifer david In sum, in contracts where maritime law applies and the parties are on equal footing in negotiation, an indemnity provision that includes indemnifcation for gross negligence will likely be enforceable. If the parties seek to indemnify for punitive damages or strict liability under a statutory scheme designed to punish or deter certain behavior, then it is unlikely to be enforceable. 3.2 Indemnity under the Longshore and Harbor Workers’ Compensation Act The Longshore and Harbor Workers’ Compensation Act (LHWCA)85 invalidates indemnity agreements where an employer is bound to indemnify a vessel owner for the vessel’s negligence.86 The LHWCA is a workers’ compensation scheme whereby an employer is strictly liable to an injured employee if a qualifed employee is injured on the job. The statute also provides a means for an employee to recover against a vessel owner directly for an injury “caused by the negligence of a vessel” under § 905(b) of the LHWCA. That same provision provides that “the employer shall not be liable to the vessel for such damages directly or indirectly and any agreements or warranties to the contrary shall be void.”87 The aim of the statute is to hold the negligent vessel owner liable for its own negligence instead of shifting liability via contractual indemnity to the employer, who is subject to strict liability for its injured employee. While this type of risk allocation may be rendered unenforceable under the LHWCA, courts have held that insurance protection is available for vessel owners named as additional insureds on the employer’s liability insurance policies. For example, in Voisin v O.D.E.C.O. Drilling Co., the Fifth Circuit held that an additional insured provision did not contravene § 905(b) of the LHWCA.88 There, an employee of a contractor was injured aboard an ofshore drilling vessel of the coast of Texas.89 The injured employee sued the employer-contractor and the owner of the drilling vessel, Odeco Drilling Inc. (Odeco), for vessel negligence and unseaworthiness.90 Odeco fled a third-party suit against the employer for breach of a Master Service Contract between Odeco and the employer, asserting that the employer failed to name Odeco as an additional assured as required by the agreement.91 Indeed, the Master Service Contract required that the employer name Odeco as an additional assured on its liability policies with a waiver of subrogation.92 The agreement also had a “broadly worded indemnity” provision in favor of Odeco “to the maximum extent permitted by law and to support such indemnity by liability insurance coverage to be furnished by [the employer].”93

85 86 87 88 89 90 91 92 93

33 U.S.C. § 901 et seq. 33 U.S.C. § 905(b). Id. 744 F.2d 1174 (5th Cir. 1984). Id. at 1175. Id. Id. Id. Id. at 1176 & n. 1.

226

the implications of litigation post deepwater horizon Odeco later settled with the personal injury plaintif, but maintained its claim against the employer.94 The trial court found that the indemnity and additional insured provisions under the Master Service Contract were void under the LHWCA and held that Odeco was not entitled to recover against the employer for the sums paid to the personal injury plaintif.95 On appeal, Odeco conceded that § 905(b) of the LHWCA invalidates an indemnity that requires an employer to indemnify for vessel negligence, but urged that the LHWCA does not invalidate an additional assured provision.96 The Fifth Circuit agreed with Odeco and held that § 905(b) does not render an additional assured provision unenforceable.97 The Fifth Circuit looked to the legislative history of § 905(b) of the LHWCA.98 The statute was amended in 1972 to include § 905(b), but prior to the amendment, the Supreme Court had rendered decisions that permitted vessel owners to recover against employers for vessel negligence.99 Congress enacted § 905(b) to preclude this indirect liability of employers for vessel negligence resulting from an injury to an employee.100 The court reasoned that an additional assured provision is not a similar form of indirect liability prohibited by § 905(b).101 The Fifth Circuit held that the employer breached the Master Service Contract by failing to name Odeco as an additional assured as required by the agreement.102 Section 905(c) of the LHWCA provides a remedy for longshoremen to pursue a cause of action for vessel negligence against a vessel owner for injuries sustained on the Outer Continental Shelf and for which an injured employee can recover under § 4 of OCSLA.103 The LHWCA does not preclude, however, the enforcement of “any reciprocal indemnity provision whereby the employer of a person entitled to receive benefts under this Act . . . and the vessel agree to defend and indemnify the other for cost of defense and loss or liability for damages arising out of or resulting from death or bodily injury to their employees.”104 3.3 Insurance obligations and additional assured status Generally, under federal law in the Fifth Circuit, insurance obligations take precedence over indemnity obligations. In Ogea v Lofand Bros.,105 Ogea, an employee of International Hammer, was injured working on a drilling rig operated by Lofand Brothers and located on a fxed platform owned by Phillips. Lofand fled a third-party demand

94 Voisin v O.D.E.C.O. Drilling Co., 744 F.2d at 1176. 95 Id. 96 Id. 97 Id. at 1179. 98 Id. at 1177. 99 Id. (citing and quoting Longmire v Sea Drilling Corp., 610 F.2d 1342 (5th Cir. 1980) (discussing Seas Shipping Co. v Sieracki, 328 U.S. 85 (1946) and Ryan Stevedoring Co. v Pan Atlantic Steamship Corp., 350 U.S. 124 (1956)). 100 Voisin, 744 F.2d at 1177. 101 Id. 102 Id. at 1179. 103 33 U.S.C. § 905(c). 104 Id. 105 622 F.2d 186 (5th Cir. 1980).

227

cindy matherne muller, jeanne amy, and jennifer david against Phillips seeking indemnity. Phillips asserted a counterclaim against Lofand alleging that Lofand was obligated to obtain liability insurance covering Phillips, and that its failure to do so constituted a breach of contract. The specifc indemnity provisions in the contract required Phillips to defend and indemnify Lofand for all claims for personal injury brought by Phillips and its contractors’ personnel. However, Phillips pointed to an additional assured provision in the contract, which required Lofand to obtain insurance of $500,000 under which Phillips was named as an additional assured. The Fifth Circuit concluded that the parties clearly intended that Phillips would not be held liable for any injuries incurred on its ofshore platform up to $500,000; therefore, the $500,000 additional assured coverage provided by Lofand to Phillips must be exhausted frst. However, the indemnity obligations owed by Phillips to Lofand did not come into play because the claim was settled for $60,000. Consequently, Ogea left open the possibility that if a claim in excess of $500,000 had been brought, exhausting the insurance provided to Phillips, then the indemnity obligation that Phillips owed to Lofand would be implicated. Later, in Tullier v Halliburton Geophysical Services, Inc.,106 the parties to a time charter for a vessel used in the ofshore oil and gas industry agreed to indemnify each other for job-related liability and to back up the cross-indemnity provisions with insurance. The charterer agreed to obtain CGL insurance with appropriate maritime endorsements. The vessel owner agreed to provide P&I insurance under the SP23 form with the “as owner” clause deleted. Last, the contract required that all insurance obtained by the vessel owner shall name the charterer as an additional assured and shall provide that it shall be primary insurance with respect to the charterer, irrespective of any excess or other insurance clauses. From these provisions, the Fifth Circuit concluded in Tullier that the vessel’s insurance policy was intended to specifcally cover the charterer as an additional assured with the “as owner” limitations deleted. The policy was also intended to constitute primary coverage for the charterer. The Fifth Circuit relied on Ogea and other Fifth Circuit jurisprudence for the proposition that a party who has entered into a contractual indemnity provision, but who also names the indemnitor as an additional assured under its own liability policies, must frst exhaust the liability insurance it agreed to obtain before seeking contractual indemnity. After these cases, the oil and gas and maritime industries operating within the reach of the Fifth Circuit adapted. Most contracts now include language that limit obligations to name as additional assureds and waive subrogation “to the extent of the indemnity obligations assumed by Contractor under this agreement.” Market participants in the ofshore industry understand the necessity of such limiting language and do not normally provide resistance to the addition of such language. The efectiveness of this language was tested in litigation between BP and Transocean and its underwriters related to the Macondo well blowout. Following the spill, BP sought coverage as an additional assured from Transocean for both the liabilities assumed in the drilling contract by Transocean and for BP’s sub-surface well pollution – liability that was expressly assumed by BP under the drilling contract. Transocean’s

106 81 F.3d 552, 1996 AMC 1561 (5th Cir. 1996).

228

the implications of litigation post deepwater horizon insurers fled a declaratory judgment action to limit BP’s additional assured coverage to only the obligations assumed by Transocean under the drilling contract, which expressly excluded sub-surface pollution. The Fifth Circuit certifed a question to the Texas Supreme Court on BP’s additional assured status.107 The Supreme Court of Texas answered that BP was an additional assured under Transocean’s liability policies, but only to the extent of the liabilities that Transocean assumed in the drilling contract. The court found that Transocean’s contractually assumed liability did not include the sub-surface well pollution that resulted from the oil spill. Specifcally, the Texas Supreme Court determined that the insurance policies incorporated the drilling contract by reference by including an additional assured endorsement that names other parties as additional assureds “as required by an Insured Contract.” By incorporating the drilling contract by reference, the insurance policies required that the court consult the drilling contract to determine BP’s additional assured status. Upon consulting the drilling contracts, the court held that BP’s additional assured status was inexorably linked, at least in some respect, to the extent of Transocean’s indemnity obligations. Therefore, the court held that coverage aforded to BP, as an additional assured, was limited to the obligations assumed by Transocean under the drilling contract alone, which did not include the additional pollution and environmental risk. 4 Application of state law The enforceability of a contract’s indemnity provision ultimately turns on the law governing the contract and whether a state’s anti-indemnity statute applies. Most US states have anti-indemnity statutes, so when contracting for work to be performed in a particular state, great care should be taken to understand the parameters of each state’s particular provisions.108 Pertinent to the oil and gas industry, Louisiana,109 Texas,110 New Mexico,111 and Wyoming112 each have oilfeld anti-indemnity statutes that, under certain circumstances, restrict indemnity between companies and contractors. When it comes to ofshore oilfeld disputes, however, the oilfeld anti-indemnity statutes of Texas and/or Louisiana are naturally more often implicated, considering those states border the US Gulf of Mexico. Accordingly, the body of case law interpreting the application of anti-indemnity statutes in the ofshore context largely stems from Texas, Louisiana, and the Fifth Circuit. For that reason, the oilfeld antiindemnity statutes of Texas and Louisiana are more fully summarized herein. 4.1 The Louisiana Oilfeld Indemnity Act The Louisiana Oilfeld Indemnity Act (LOIA)113 was passed to address the concerns of oilfeld contractors over the “widespread inclusion” of indemnity obligations in master 107 108 109 110 111 112 113

In re Horizon, 470 S.W.3d 452 (Tex. 2015). See Appendix A for survey of anti-indemnity statutes across the United States. The Louisiana Oilfeld Indemnity Act, La. R.S. § 9:2780. Tex. Civ. Prac. & Rem. Code § 127.003. New Mexico Oilfeld Anti-Indemnity Act (NMOAIA), N.M. Stat. Ann. § 56–7–2. Wyoming Oilfeld Anti-Indemnity Act (WOAIA), Wyo. Stat. § 30–1–131. La. R.S. § 9:2780.

229

cindy matherne muller, jeanne amy, and jennifer david service agreements, which oil companies require contractors to sign.114 The indemnity provisions exposed oilfeld contractors to worker’s compensation and tort liability, generally without an equal bargaining position with the oil companies requiring the master service agreement.115 To that end, the Louisiana legislature enacted LOIA to restrict defense and indemnity obligations in certain contracts relating to the drilling industry. Specifcally, LOIA declares null and void and against public policy any provision in a contract “pertaining to a well for oil, gas, or water, or drilling for minerals” to the extent that it provides for defense and indemnity “for damages arising out of or resulting from death or bodily injury to persons, which is caused by or results from the sole or concurrent negligence or fault (strict liability) of the indemnitee.”116 In other words, LOIA only applies to contracts pertaining to a well and for loss or claims arising out of personal injury or death, not property damage. Louisiana and Fifth Circuit courts broadly interpret LOIA when evaluating indemnity claims.117 When determining whether LOIA applies to a contract, courts apply a two-step process: (1) the contract must pertain to a well and (2) the contract must be related to exploration, development, production, or transportation of oil, gas, or water.118 With respect to the frst prong, whether the contract pertains to a well is a factintensive inquiry analyzed on a case-by-case basis.119 While the focus of the inquiry is the location of the work required under the contract, a court will examine the following factors: 1 Whether the structures or facilities to which the contract applies or with which it is associated are part of an in-feld production system; 2 The geographical location of the structure or facility relative to a well or wells; 3 The purpose or function of the facility or structure in question; 4 The owner and operator of the relevant facility or structure, and the owner and operator of the well or wells that produce the gas in question; 5 Any number of other details afecting the functional and geographic nexus between a well and the structure or facility that is the object of the agreement.120 To clarify, the services required under the contract need not be rendered on site or in connection with drilling or operation of a specifc well to fall within the purview of LOIA.121 Further, the fact that a contract also may pertain to numerous other wells or to objects other than wells is of no consequence for purposes of applying LOIA, so long as the services rendered relate to the “exploration development, production, or transportation of oil, gas, or water.”122

114 Meloy v Conoco, Inc., 504 So. 2d 833 (La. 1987). 115 Id. 116 La. R.S. § 9:2780(B). 117 Roberts v Energy Development Corp., 104 F.3d 782, 784 (5th Cir. 1997). 118 Lloyds of London v Transcontinental Gas Pipe Line Corp., 38 F.3d 193, 196 (5th Cir. 1994). 119 Id.; see also Verdin v ENSCO Ofshore Co., 104 F. Supp. 2d 682 (W.D. La. 2000), af’d, 255 F.3d 246 (5th Cir. 2001). 120 Verdin, 104 F. Supp. 2d at 690 (W.D. La. 2000) (internal citations omitted). 121 Livings v Service Truck Lines of Texas, Inc., 467 So. 2d 595, 599 (La. App. 3 Cir. 1985). 122 Id.

230

the implications of litigation post deepwater horizon In Marcel v Placid Oil Co.,123 the Fifth Circuit set forth a judicially created exception to LOIA, which provides that an indemnity provision in an otherwise qualifying contract is enforceable when the indemnitor purchases insurance for the indemnitee and the indemnitee reimburses the indemnitor for its portion of the premium. The exception is known as the Marcel exception. Stated diferently, the Marcel exception to LOIA requires that the indemnitee pay its portion of the premium for additional assured status to the indemnitor, separate and apart from the contract price. In creating the exception, the Marcel court explained that the purpose of LOIA is served by shifting the responsibility of paying for liability coverage to the indemnitee: The LOIA is aimed at preventing the shifting of the economic burden of insurance coverage or liability onto an independent contractor. If the principal pays for its own liability coverage, however, no shifting occurs. We see no need to prevent such an arrangement in order to give efect to the LOIA.124

It is worth noting that the Marcel court stressed that the exception “does not apply if any material part of the cost of insuring the indemnitee is borne by the independent contractor procuring the insurance coverage.”125 Some courts have strictly construed this requirement when considering the cost and amount of the premium paid by the indemnitee.126 In addition to its oilfeld anti-indemnity statute, Louisiana also has a separate anti-indemnity statute applicate to other certain types of contracts, including construction contracts127 and certain motor carrier transportation contracts relative to unloading, loading, transporting, or services incidental thereto (i.e., storage).128 Under

123 11 F.3d 563, 569–570 (5th Cir. 1994). 124 Id. at 569–570. 125 Id. at 570. 126 See, e.g., Amoco Prod. Co. v Lexington Ins. Co., 98–1676 (La. App. 1 Cir. 9/24/99); 745 So. 2d 676, writ denied, 99–3553 (La. 2/25/00); 755 So. 2d 253 (holding that oil company’s payment of $2,000 for $11 million in primary and excess liability coverage as an additional insured was “clearly inadequate consideration” that failed to satisfy the Marcel exception to LOIA). 127 Under La. R.S. § 9:2780.1(A)(2), a “construction contract” is defned as “any agreement for the design, construction, alteration, renovation, repair, or maintenance of a building, structure, highway, road, bridge, water line, sewer line, oil line, gas line, appurtenance, or other improvement to real property, or repair or maintenance of a highway, road, or bridge, including any moving, demolition, or excavation, except that no deed, lease, easement, license, or other instrument granting an interest in or the right to possess property will be deemed to be a construction contract even if the instrument includes the right to design, construct, alter, renovate, repair, or maintain improvements on such real property.” The statute expressly excludes contracts related to the design, construction, alteration, renovation, repair, or maintenance of “any dirt or gravel road used to access oil and gas wells and associated facilities” as well as “[o]il fow lines or gas gathering lines used in association with the transportation of production from oil and gas wells from the point that oil and gas becomes co-mingled for transportation to oil storage facilities or gas transmission lines.” 128 La. R.S. § 9:2780.1(A)(1) defnes a “motor carrier transportation contract” as “any contract, agreement, or understanding covering the transportation of property, other than agricultural products as defned in R.S. 9:3306 and timber without limitation, for compensation or hire by a motor carrier, entrance upon property by the motor carrier for the purpose of loading, unloading, or transporting property, other than agricultural products as defned in R.S. 9:3306 and timber without limitation, for compensation or hire, or a service incidental to any such activity, including but not limited to storage of property, other than agricultural products as defned in R.S. 9:3306 and timber without limitation, except the Uniform Intermodal Interchange and Facilities Access Agreement administered by the Intermodal Association of North America or other agreements providing for the interchange, use, or possession of intermodal chassis, containers, or other intermodal equipment.”

231

cindy matherne muller, jeanne amy, and jennifer david this Louisiana Anti-Indemnity Statute,129 indemnity provisions and provisions requiring an indemnitor to procure liability insurance to cover the acts of an indemnitee contained in a construction contract or certain motor carrier transportation contracts are null, void, and unenforceable as against public policy. In particular, La. Rev. Stat. § 9:2780.1 provides, in pertinent part: Notwithstanding any provision of law to the contrary and except as otherwise provided in this Section, any provision, clause, covenant, or agreement contained in, collateral to, or afecting a motor carrier transportation contract or construction contract which purports to indemnify, defend, or hold harmless, or has the efect of indemnifying, defending, or holding harmless, the indemnitee from or against any liability for loss or damage resulting from the negligence or intentional acts or omissions of the indemnitee, an agent or employee of the indemnitee, or a third party over which the indemnitor has no control is contrary to the public policy of this state and is null, void, and unenforceable. Notwithstanding any provision of law to the contrary and except as otherwise provided in this Section, any provision, clause, covenant, or agreement contained in, collateral to, or afecting a motor carrier transportation contract or construction contract which purports to require an indemnitor to procure liability insurance covering the acts or omissions or both of the indemnitee, its employees or agents, or the acts or omissions of a third party over whom the indemnitor has no control is null, void, and unenforceable. However, nothing in this Section shall be construed to prevent the indemnitee from requiring the indemnitor to provide proof of insurance for obligations covered by the contract.

To the extent ofshore activities involve the construction, design, maintenance, or repair of platforms, other ofshore structures, or oil and gas lines, Louisiana’s antiindemnity statute applicable to construction contracts may still apply. The Louisiana anti-indemnity statute for certain construction and motor carrier transportation contracts does contain a statutory exception under La. Rev. Stat. § 9:2780.1(I). The statutory exception provides that an indemnity or additional assured provision may still be enforceable where the indemnitor obtains insurance coverage only to the extent of the indemnity obligation and there is evidence that the indemnitor recover the cost of the required insurance in the contract price.130

129 La. R.S. § 9:2780.1. 130 The statutory exception under La. R.S. § 9:2780.1(I) specifcally states that: Nothing in this Section shall invalidate or prohibit the enforcement of the following: (1) Any clause in a construction contract containing the indemnitor’s promise to indemnify, defend, or hold harmless the indemnitee or an agent or employee of the indemnitee if the contract also requires the indemnitor to obtain insurance to insure the obligation to indemnify, defend, or hold harmless and there is evidence that the indemnitor recovered the cost of the required insurance in the contract price. However, the indemnitor’s liability under such clause shall be limited to the amount of the proceeds that were payable under the insurance policy or policies that the indemnitor was required to obtain. (2) Any clause in a construction contract that requires the indemnitor to procure insurance or name the indemnitee as an additional insured on the indemnitor’s policy of insurance, but only to the extent that such additional insurance coverage provides coverage for liability due to an obligation to indemnify, defend, or hold harmless authorized pursuant to Paragraph (1) of this Subsection, provided that such insurance coverage is provided only when the indemnitor is at least partially at fault or otherwise liable for damages ex delicto or quasi ex delicto.

232

the implications of litigation post deepwater horizon 4.2 Texas Oilfeld Anti-Indemnity Act Under Texas law, indemnity obligations in oilfeld contracts that indemnify a party for its own negligence are likewise generally void as against public policy. Specifcally, the Texas Oilfeld Anti-Indemnity Act (TOAIA)131 states: Except as otherwise provided by this chapter, a covenant, promise, agreement, or understanding contained in, collateral to, or afecting an agreement pertaining to a well for oil, gas, or water or to a mine for a mineral is void if it purports to indemnify a person against loss or liability for damage that: (1) is caused by or results from the sole or concurrent negligence of the indemnitee, his agent or employee, or an individual contractor directly responsible to the indemnitee; and (2) arises from: (A) personal injury or death; (B) property injury; or (C) any other loss, damage, or expense that arises from personal injury, death, or property injury.

Certain exceptions exist to the TOAIA, however, including an exception for insurancesupported indemnities.132 To clarify, Texas law provides an exception for where the parties “agree in writing that the indemnity obligation will be supported by liability insurance coverage to be furnished by the indemnitor.”133 With respect to a mutual indemnity obligation, “the indemnity obligation is limited to the extent of the coverage and dollar limits of insurance or qualifed self-insurance each party as indemnitor has agreed to obtain for the beneft of the other party as indemnitee.”134 As for a unilateral indemnity obligation, “the amount of insurance required may not exceed $500,000.”135 Under Texas law, any indemnity clause must further comply with “fair notice” requirements – i.e., the indemnifcation clause must be “express” and “sufciently conspicuous.”136 The “express” requirement requires that the intent to indemnify for a party’s negligence must be expressly stated. Practitioners have developed extensive language stating the express causes for which indemnity will apply. As for the conspicuous requirement, Texas courts look to whether the indemnity provision, on its face, “attract[s] the attention of a reasonable person when he looks at it.”137 “Language may satisfy the conspicuousness requirement by appearing in larger type, contrasting colors, or otherwise calling attention to itself.”138 Generally, indemnity provisions in Texas contracts should appear in bolded, contrasting, and/or all capitalized font in order to ensure the provision is sufciently conspicuous for purposes of meeting the “fair notice” test. 131 Tex. Civ. Prac. & Rem. Code § 127.003. 132 Tex. Civ. Prac. & Rem. Code § 127.005. 133 Id.; see also Ken Petr. Corp. v Questor Drilling Corp., 24 S.W.3d 344, 346 (Tex. 2000). 134 Tex. Civ. Prac. & Rem. Code § 127.005(b). 135 Tex. Civ. Prac. & Rem. Code § 127.005(c). 136 The fair notice requirements are not required if it can be shown that the party had actual knowledge of the indemnity provision, such as where the provision was revised by the parties during the contract negotiation. See, e.g., Cleere Drilling Co. v Dominion Expl. & Prod., Inc., 351 F.3d 642, 647 (5th Cir. 2003). 137 Storage & Processors, Inc. v Reyes, 134 S.W.3d 190, 192 (Tex. 2004) (internal quotations omitted). 138 Id.

233

cindy matherne muller, jeanne amy, and jennifer david Like Louisiana, Texas also has a Texas Construction Anti-Indemnity Act139 that renders void certain indemnity agreements in construction contracts, as well as related provisions that require procurement of additional-assured coverage.140 There are several exceptions to the Texas Construction Anti-Indemnity Act. For example, the anti-indemnity statute does not apply to a claim “for the bodily injury or death of an employee of the indemnitor, its agent, or its subcontractor of any tier.”141 The act also does not apply for certain other contracts, including, inter alia, contracts for public works for a municipality.142 5 Conclusion In the United States, the enforcement of indemnity obligations designed to indemnify a party for its own negligence is not simple. In an ofshore setting, the enforcement is complicated by the tension between application of federal and state law. A substantial amount of litigation involves the fact-intensive inquiries to determine applicable law to a particular indemnity obligation. As many agreements used in other industries ofshore take the form of master service agreements, care should be taken to ensure that indemnity and insurance provisions are drafted to maximize enforceability under the laws most likely to apply. Under any given work order, the facts could change, as could the location and thus, the law of the adjacent state. Practitioners have developed provisions to ensure enforceability under all of the likely legal regimes, through reciprocal indemnities, additional assured provisions, and premium allocations. These provisions should be revised or updated often to adapt to or take advantage of new considerations as jurisprudence continues to develop. Companies should review their existing master service agreements and examine long standing agreements to consider whether the indemnity language drafted at the time of execution remains enforceable under current jurisprudence.

139 Tex. Ins. Code § 151.102 provides: Except as provided by Section 151.103, a provision in a construction contract, or in an agreement collateral to or afecting a construction contract, is void and unenforceable as against public policy to the extent that it requires an indemnitor to indemnify, hold harmless, or defend a party, including a third party, against a claim caused by the negligence or fault, the breach or violation of a statute, ordinance, governmental regulation, standard, or rule, or the breach of contract of the indemnitee, its agent or employee, or any third party under the control or supervision of the indemnitee, other than the indemnitor or its agent, employee, or subcontractor of any tier.

140 Tex. Ins. Code § 151.104. 141 Tex. Ins. Code § 151.103. 142 Tex. Ins. Code § 151.105.

234

APPENDIX A

State

Anti-Indemnity Statute

Is an Indemnitee Permitted to Indemnify for its Sole Negligence?

Alabama

None

Alaska Arizona

Alaska Stat. § 45.45.900 Ariz. Rev. Stat. §§ 32– 1159, 34–226, 41–2586 Ark. Code § 4–56–104, § 22–9–214

Permissible if the language is clear and unequivocal. Mobil Oil Corp. v Schlumberger, 598 So. 2d 1341 (Ala. 1992). No. Prohibited by anti-indemnity statute. No. Prohibited by anti-indemnity statute. Ariz. Rev. Stat. § 32–1159. No. Prohibited by anti-indemnity statute. Ark. Code §§ 4–56–104(b), 22–9– 214(b). No. Prohibited by anti-indemnity statute. Cal. Civ. Code § 2782. No. Prohibited by anti-indemnity statute. Colo. Rev. Stat. §§ 13–50.5–102(8)(a), 13–21–111.5(6)(b). No. Prohibited by anti-indemnity statute. Conn. Gen. Stat. § 52–572k(a). No. Prohibited by anti-indemnity statute. Del. Code Ann. 6 § 2704(a). Permissible only if the agreement caps the indemnity obligation at an amount “that bears a reasonable commercial relationship to the contract.” Fla. Stat. Ann. § 725.06(1). No. Prohibited by anti-indemnity statute. Ga. Code Ann § 13–8–2(b). No. Prohibited by anti-indemnity statute. Haw. Rev. Stat. § 431:10–222. No. Prohibited by anti-indemnity statute. Idaho Code Ann. § 29–114. No. Prohibited by anti-indemnity statute. 740 ILCS § 35/1. No. Prohibited by anti-indemnity statute. Ind. Code Ann. § 26–2–5.

Arkansas California

Cal. Civ. Code § 2782

Colorado

Colo. Rev. Stat. §§ 13–50. 5–102, 13–21–111.5

Connecticut

Conn. Gen. Stat. § 52–572k

Delaware

Del. Code. Ann. 6 § 2704

Florida

Fla. Stat. Ann. § 725.06

Georgia

Ga. Code Ann. § 13–8–2(b)

Hawaii Idaho

Haw. Rev. Stat. § 431:10– 222 Idaho Code Ann. § 29–114

Illinois

740 ILCS § 35/1

Indiana

Ind. Code Ann. § 26–2–5

(Continued)

235

cindy matherne muller, jeanne amy, and jennifer david (Continued) State

Anti-Indemnity Statute

Is an Indemnitee Permitted to Indemnify for its Sole Negligence?

Iowa

Iowa Code Ann. § 537A.5

Kansas

Kan. Stat. § 16–121

Kentucky

Ky. Rev. Stat. § 371.180

Louisiana

La. R. S. §§ 38:2216(G) (public contracts), 9:2780 (oil and gas), 9:2780.1 (construction)

Maine

None

Maryland

Md. Code Ann. § 5–401

Massachusetts

Mass. Gen. Laws Ann. ch. 149 § 29C M.C.L.A. § 691.991

No. Prohibited by anti-indemnity statute. Iowa Code Ann. § 537A.5. No. Prohibited by anti-indemnity statute. Kan. Stat. § 16–121(b). No. Prohibited by anti-indemnity statute. Ky. Rev. Stat. § 371.180. Permissible if the language is clear and unequivocal. Barnett v Am. Constr. Hoist, Inc., 2011 1261 (La. App. 1 Cir. 02/10/12); 91 So. 3d 345. In oil and gas contracts, see Marcel v Placid Oil Co., 11 F.3d 563 (5th Cir. 1994). In construction contracts, see La. R. S. § 9:2780.1(I). Permissible if contract clearly and unequivocally reflects a mutual intention on behalf of the parties to indemnify for the indemnitee’s sole negligence. Emery Waterhouse Co. v Lea, 467 A.2d 986 (Me. 1983). No. Prohibited by anti-indemnity statute. Md. Code Ann. § 5–401(a)(1)–(2). No. Prohibited by anti-indemnity statute. Mass. Gen. Laws Ann. ch. 149 § 29C. No. Prohibited by anti-indemnity statute. M.C.L.A. § 691.991. No. Prohibited by anti-indemnity statute. Minn. Stat. § 337.02.

Michigan Minnesota

Missouri

Minn. Stat. §§ 337.01, 337.02 Miss. Code. Ann. § 31– 5–41 Mo. Rev. Stat. § 434.100

Montana

MCA § 28–2–211

Nebraska

Neb. Rev. Stat. § 25–21, 187

Nevada

New Jersey

N.R.S. § AB 125, § 2 (applies to residential construction contracts) N.H. Rev. Stat. § 338-A:1, 338-A:2 N.J. Stat. Ann. § 2A:40A-1

New Mexico

N.M. Stat. Ann. § 56–7–1

Mississippi

New Hampshire

236

No. Prohibited by anti-indemnity statute. Miss. Code. Ann. § 31–5–41. No. Prohibited by anti-indemnity statute. Mo. Rev. Stat. § 434.100. No. Prohibited by anti-indemnity statute. MCA § 28–2–211. No. Prohibited by anti-indemnity statute. Neb. Rev. Stat. § 25–21, 187(1). Prohibited in residential construction contracts. N.R.S. § AB 125, § 2. No. Prohibited by anti-indemnity statute. N.H. Rev. Stat. § 338-A:1, 338-A:2. No. Prohibited by anti-indemnity statute. N.J. Stat. Ann. § 2A:40A-1. No. Prohibited by anti-indemnity statute. N.M. Stat. Ann. § 56–7–1(1).

the implications of litigation post deepwater horizon State New York North Carolina North Dakota Ohio

Anti-Indemnity Statute N.Y. Gen. Oblig. Law § 5–322.1 N.C. Gen. Stat. Ann. § 22B-1 None

No. Prohibited by anti-indemnity statute. N.Y. Gen. Oblig. Law § 5–322.1(1)–(2). No. Prohibited by anti-indemnity statute. N.C. Gen. Stat. Ann. § 22B-1. Permissible if contract clearly conveys the parties’ intent. Rupp v Am. Crystal Sugar Co., 465 N.W.2d 614 (N.D. 1991). No. Prohibited by anti-indemnity statute. Ohio Rev. Code Ann. § 2305.31. No. Prohibited by anti-indemnity statute. Okla. Stat. title 15 § 221.

Oklahoma

Ohio Rev. Code Ann. § 2305.31 Okla. Stat. title 15 § 221

Oregon

Or. Rev. Stat. § 30.140

Pennsylvania Rhode Island

None R.I. Gen. Laws § 6–34–1

South Carolina

S.C. Code Ann. § 32–2–10

South Dakota

Utah

S.D. Codified Laws § 56–3–18 Tenn. Code Ann. § 62–6– 123 Tex. Ins. Code Ann. §§ 151.102, 151.103 Utah Code Ann. § 13–8–1

Vermont

None

Virginia

Va. Code Ann. § 11–4.1

Washington

Wisconsin

Wash. Rev. Code. § 4.24.115 W. Va. Code Ann. § 55–8–14 None

Wyoming

Wyo. Stat. § 30–1–131

Tennessee Texas

West Virginia

Is an Indemnitee Permitted to Indemnify for its Sole Negligence?

237

No. Prohibited by anti-indemnity statute. Or. Rev. Stat. § 30.140(1)–(2). No case law found. No. Prohibited by anti-indemnity statute. R.I. Gen. Laws § 6–34–1. No. Prohibited by anti-indemnity statute. S.C. Code Ann. § 32–2–10. No. Prohibited by anti-indemnity statute. S.D. Codified Laws § 56–3–18. No. Prohibited by anti-indemnity statute. Tenn. Code Ann. § 62–6–123. No. Prohibited by anti-indemnity statute. Tex. Ins. Code Ann. § 151.102. No. Prohibited by anti-indemnity statute. Utah Code Ann. § 13–8–1. Permissible if contractual language clearly expresses the intent to indemnify for sole negligence of indemnitee. Tateosian v Vermont, 945 A.2d 833 (Vt. 2007). No. Prohibited by anti-indemnity statute. Va. Code Ann. § 11–4.1. No. Prohibited by anti-indemnity statute. Wash. Rev. Code. § 4.24.115(1)(b). No. Prohibited by anti-indemnity statute. W. Va. Code Ann. § 55–8–14. Permissible if provision is clear and unambiguous. Gunka v Consolidated Papers, Inc., 508 N.W.2d 426 (Wis. Ct. App. 1993). Permissible if clearly stated. Union Pac. Resources Co. v Dolenc, 86 P.3d 1287 (Wyo. 2004).

CHAP TER 13

Mutual hold harmless clauses in France and francophone civil law systems Bertrand Montembault and Paul Morton

1 Introduction Although there is no general theory of risk in French law, the concept of risk occupies a central place in the French civil law tradition.1 According to its Latin etymology, the word ‘risk’ fnds its origins in the idea of “shared risk between contracting parties”,2 suggesting the idea was conceived in a contractual setting. French law has also gradually introduced the concept of risk in matters of civil liability. Although long subordinated to the existence of fault, the principles of civil liability have expanded to the point of recognising ‘no fault’ liability, the idea being that where someone’s actions create a risk for third parties, the author of those actions should be liable for the harm caused.3 The abundance of topics linked to risk highlights the importance of its place in French civil law. In addition, technological development and the associated risks it has created have given rise to new challenges that need to be refected in contractual relations. In this chapter we will focus on the analysis of the principle of contractual allocation of risk from a French civil law perspective and by extension other francophone civil law jurisdictions. More specifcally, we will look at the acceptance within such legal systems of so-called mutual hold harmless clauses as a technique for contractually allocating risk. 1.1 Francophone civil law systems French law has long served as a point of reference for legislators in other jurisdictions and has made its contribution to the creation of the international legal order. In Europe, the Napoleonic campaigns resulted in the export of a codifed legal system, as well as the institutions underpinning it. In North America, the laws of Quebec and Louisiana fnd their origins in French civil law (as well as English common law), and in Latin America many of the States created in the 19th century drew inspiration from the French legal system. In Africa, French colonisation meant that French law became an essential source of legislation at the time of independence. To this day, French law is intricately woven into the legal fabric of most francophone African jurisdictions.

1 A. Downe, La gestion des risques contractuels par le contrat: Étude du droit français, 2021, p. 29. 2 Ibid. 3 P. Le Tourneau, Droit de la responsabilité et des contrats, Dalloz Action, 12e éd., 2020, p. 83.

238

DOI: 10.4324/9781003206798-15

mutual hold harmless clauses in francophone civil law systems The analysis in this chapter will be limited to French law and other francophone civil law systems in Africa. 1.1.1 The French Civil Code of 1804 1.1.1.1 French law of obligations: contract law and civil liability Promulgated in France in 1804, the French Civil Code (or Napoleonic Code) combines all of the rules governing the legal status of persons, of property and relations between private individuals, the latter dominated by what has become known as the ‘law of obligations’. The French Civil Code divides the law of obligations into two broad categories, depending on whether the obligations arise from an agreement between the parties or by operation of the law. The detailed provisions of Title III of the Code, on contracts and contractual obligations, refected the individualistic conception of the 1804 Code, based on the autonomous will of the individual. Title V covers “commitments formed without agreement”, which are further distinguished between commitments resulting “from the sole authority of the law” (such as family relations or the protection of those designated as incapable) and those which arise as a result of a characteristic personal to the debtor. The latter includes the French notions of ‘quasi-contracts’ and tortious and quasi-tortious acts. In the context of the above, parties to a contract are free to agree in advance to limit, exclude or re-allocate their liability for failure to perform their obligations. Despite their frequent use in day-to-day contractual relations, such agreements are not subject to any dedicated regulation under French law, although they are subject to certain limitations when used in the context of certain types of contracts or sectors. 1.1.1.2 Francophone African legal systems The Civil Code of 1804 was exported to North Africa, specifcally to Algeria, Morocco, Tunisia and Egypt, where the legislation is strongly inspired by the code. It was then exported to sub-Saharan Africa; to the former territories of French West Africa, namely Benin, Burkina Faso, Côte d’Ivoire, Guinea, Mali, Mauritania, Niger and Senegal; in French Equatorial Africa, now Congo-Brazzaville, Gabon, the Central African Republic and Chad;4 as well as Cameroon and Togo, both formerly under French mandate. Finally, the Civil Code of 1804 was exported to the Great Lakes region, namely Rwanda, the Democratic Republic of Congo and Burundi (former colonies of Belgium), as well as to southern Africa where it was implanted in the Comoros, Mauritius, Madagascar and the Seychelles.5 Following independence, a number of States opted to retain a number of principles of the French Civil Code. This was the case in Tunisia and Morocco in 1906 and 1913, for instance, in the form of a Code on obligations and contracts.6

4 B. Bokolombe, L’infuence du modèle juridique français en Afrique, 2016, p. 75. 5 Ibid. 6 M. Grimaldi, “L’exportation du code civil” in Pouvoirs, 2003, p. 91.

239

bertrand montembault and paul morton Generally speaking, francophone African States retain a close link to the Civil Code of 1804, to the point where it remains in force in certain jurisdictions to this day (e.g., in Chad, the Republic of Congo and Gabon). 1.1.2 Recent developments 1.1.2.1 French law With few exceptions, the French Civil Code remained largely unchanged until ordinance no. 2016–131 of 10 February 2016 came into efect on 1 October 2016,7 bringing with it a general reform of the law of contracts and obligations. In addition to removing certain provisions that had become increasingly criticised as obsolete,8 the 2016 reform served above all to codify a number of concepts and jurisprudential principles developed to respond to the shortcomings of the 1804 code. The rules relating to contractual non-performance, which were addressed in a scattered and incomplete way in the Civil Code, have been grouped together in the same place in article 1217 et seq. of the Civil Code. This reform was also intended to be followed by a reform of the law of civil liability, a frst draft of which was published in 2017. Similarly to the 2016 reforms, the reform of the law of civil liability is intended to modernise rules originally written more than two centuries ago, as well as to codify a signifcant body of jurisprudence in a number of areas. 1.1.2.2 Francophone African legal systems As noted above, with the exception of certain jurisdictions, such as the Democratic Republic of Congo where the law of obligations is inspired by the Belgian model, most of the legislation in this area in francophone African countries is based on French civil law.9 Generally speaking, the law of contracts and more generally of obligations in these jurisdictions has changed little since independence, with very few having adopted new legislation on obligations. Notable exceptions include the Senegalese Code of Civil and Commercial Obligations, based on the amended law no. 63–62 of 10 July 1963, the Guinean Civil Code based on law no. 2019/035/AN of 4 July 2019 (replacing the 1983 Civil Code) and Malian law no. 87–31/AN-RM of 29 August 1987 establishing the general regime on obligations. Although these texts have their specifcities,10 they remain fundamentally inspired by the French civil law tradition. Despite the relatively few developments across individual jurisdictions in francophone Africa, there is the potential for a harmonisation of the law of obligations at

7 Ordinance no. 2016–131 on the reform of the law of contracts, the general regime and proof of obligations was promulgated on 10 February 2016 and entered into force on 1 October 2016. This ordinance was ratifed by the law no. 2018–287 of 20 April 2018. The ordinance modifed Book III of the French Civil Code and added a Title III relating to the sources of obligations, a Title IV relating to the general regime covering obligations and a Title IV bis on the proof of obligations. 8 For instance, the ordinance removed any reference to ‘cause’, a central concept under French contract law and condition for validity of a contract. 9 Although contract law in Cameroon notably combines elements of both French civil law and common law (B. Bokolombe, L’infuence du modèle juridique français en Afrique, 2016, p. 75). 10 These are notably in areas such as family law, where Islamic law has had an important infuence.

240

mutual hold harmless clauses in francophone civil law systems a regional level through the Organisation for the Harmonisation of Business Law in Africa (OHADA), which published a draft uniform act on contract law (in 2006) and a draft uniform act on the general law of obligations (in 2015).11 In practice, academic and jurisprudential developments in French law continue to play a persuasive role in francophone African countries. 1.2 Mutual hold harmless clauses: a contractual re-allocation of liability 1.2.1 Origins and growing prevalence in French law The remainder of this chapter will focus on the articulation of mutual hold harmless clauses with the notions of contractual and non-contractual liability12 under French law. ‘Mutual hold harmless’ or ‘knock-for-knock’ clauses, which allow the parties to a contract to give mutual undertakings to assume any losses they may sufer, irrespective of who caused the loss, have their origins in the English language, common law world, but are increasingly found in French law contracts. Such clauses are particularly common in high-risk industries, such as the petroleum industry, as well as the nuclear and space13 sectors. By avoiding the need to establish fault in the context of complex industrial operations with multiple actors, the mutual hold harmless mechanism allows participants to reduce the overall cost of operations by avoiding lengthy dispute resolution processes and multiple layers of insurance (see Gordon, Chapter 2, for more information about the general understanding of mutual indemnity, mutual indemnity and hold harmless, etc.). In addition to addressing losses sufered by the parties themselves, mutual hold harmless clauses frequently also deal with the way in which third-party liability is addressed between the contracting parties. These third parties may be related to the contracting parties (such as afliates or subcontractors) or may be ‘true’ third parties that are otherwise unconnected to the parties or the subject matter of the contract. This distinction between liability between the parties and liability towards third parties will play an important role in the interpretation of mutual hold harmless clauses under French law (as will be seen in the remainder of this chapter). As in other jurisdictions, such a contractual re-allocation of liability between parties derogates from the principles of French law on liability, which is based on the notion of fault. However, these clauses have received relatively little attention or analysis from a French law perspective. 1.2.2 A note on English terminology As noted above, mutual hold harmless clauses have their origins in contracts drafted in English and subject to the laws of common law jurisdictions (with a particularly heavy infuence from the United States and England). The specifc language and patterns of drafting used in such clauses have also developed from practices in 11 The latter was developed following the lack of progress made on the draft uniform act on the law of contracts. Finally, it is also worth noting that a Uniform Act on General Commercial Law exists already. 12 Which we will refer to here as ‘civil liability’. 13 See, for instance, A. Rakibi, “Les clauses réciproques d’abandon de recours et de garanties dans les contrats de l’industrie spatiale”, 2014.

241

bertrand montembault and paul morton diferent industries (notably the oil and gas industry). The result is a collection of operative terminology that includes, among others, ‘to hold harmless’, ‘to indemnify’, ‘to defend’, ‘to release’ and ‘to be responsible for’. While some of these terms will have specifc legal meanings in many jurisdictions, others are less well defned and appear to be included simply as a matter of industry custom. Although a number of these concepts will equate to similar or equivalent concepts under French law, this chapter will focus on the re-allocation of liability based on the principles of civil liability under French law and the construction of mutual hold harmless clauses from those principles. 2 The legal nature of mutual hold harmless clauses under French law 2.1 The legal framework for mutual hold harmless clauses 2.1.1 Principles of the law on civil liability under French law We saw above in relation to the law of obligations that a person may owe an obligation either because they agreed to such an obligation or because the law imposes it on them. Mutual hold harmless clauses fall under both regimes as they permit the parties to address both contractual liability and non-contractual (or tortious or delictual) liability of the parties. As in many jurisdictions, contractual and non-contractual liability are governed by diferent legal regimes. To begin with, while contractual liability may only be triggered by a breach of contract,14 non-contractual liability may be triggered by diferent operative events, such as an act of negligence or as a result of vicarious liability of diferent kinds.15 Reparations are also based on diferent principles. Whereas a breach of contract will entitle the other party to recover all losses foreseen or foreseeable at the date of execution of the contract, compensation in respect of non-contractual liability under French law is based on the principle of full reparation. The starting point is that contractual and non-contractual liability are non-cumulative, meaning that a creditor of a contractual obligation may not rely on the rules governing non-contractual liability.16 However, this principle is subject to certain exceptions. First, the Supreme Court (Cour de Cassation) has held that the ‘noncumulative’ principle should not prevent a party to a contract from relying on subsidiary legal grounds for a claim. In other words, where there is doubt over the extent of the contractual obligations, the claim may, although principally based on contractual grounds, also rely on non-contractual grounds.17 A claim based on non-contractual liability will also be allowed where there is no link between the loss sufered and the obligations under the contract.18 As a result, there is now a much

14 Article 1231–1 of the French Civil Code. 15 Articles 1240 to 1244 of the French Civil Code. 16 Cour de Cassation, 11 January 1989, 86–17.323. The proposed reform of the law of obligations tabled in 2016 retained this ‘non-cumulative’ principle. 17 Cass. com. 24 October 2018, no. 17–25.672. 18 Civ. 2e, 19 November 1964.

242

mutual hold harmless clauses in francophone civil law systems more nuanced position under French law in relation to potential non-contractual liability arising as between contracting parties. There are also important diferences in relation to contractual limitation or exclusion of contractual and non-contractual liability. Limitation and exclusion of liability clauses are valid in principle when applied to contractual liability (subject to certain exceptions), but this principle is more limited when applied to non-contractual liability (see section 3.1.1.1.1). 2.1.2 Contractual mechanisms for allocating risk between the parties The parties to a contract governed by French law generally have signifcant freedom to allocate between them risks arising out of a party’s civil liability. The legal mechanisms available for allocating such risks will depend on whether the risk relates to liability of the contracting parties towards each other or liability towards third parties (in each case whether tortious or contractual). Where the objective is to allocate liability of the parties towards each other, a number of options are available. First, the parties can choose to act either at the level of the obligation itself, by defning the conditions for the existence of the obligation and preventing the liability from arising in the frst place.19 Alternatively, the objective can be achieved by intervening at the level of the consequences for non-performance, by restricting the consequences of a failure to perform the obligation. Addressing the consequences of a failure to perform, the second option above, can be done either by limiting the ability of the other party to act through a waiver of recourse (clauses de non-recours), or by limiting the right of the other party to compensation. Such a limitation on the right to compensation can take several forms, either an outright exclusion of a right to damages or compensation (clause de nonresponsabilité), a limitation of that right or a pre-agreed compensation in the event of non-performance (e.g., through a penalty clause, whereby the parties agree in advance on the compensation to be paid in the event of specifc instances of non-performance20). Although conceptually distinct, the same jurisprudential principles apply to both exclusions of liability and waivers of recourse, both in relation to their validity and their efects.21 2.1.3 Contractual mechanisms for allocating third-party liability Parties to a contract may also seek to re-allocate third-party liability between them. Such third parties may either be related to the contracting parties, including for instance their subcontractors or afliates, or ‘genuine’ third parties without any connection to the parties or the contract. Although the contractual approach may vary

19 Ph. Delebecque, Fasc. 115: “RÉGIME DE LA RÉPARATION – Modalités de la réparation. – Règles particulières à la responsabilité contractuelle. – Conventions relatives à la responsabilité”, JurisClasseur Contrats, 2018, p. 3. 20 A penalty clause under French law is a clause “by which the parties determine, on a comprehensive basis in advance, the compensation due upon a breach of the contractual obligation”. See J. Carbonnier, Droit civil: les biens, les obligations, 2004, p. 2222. 21 There are, however, some distinctions. For instance, in the case of an exclusion of liability, a breach by the debtor may be remedied otherwise than through engaging the contractual liability of the debtor. The creditor may seek the forced performance or judicial termination of the contract W. Dross, Dictionnaire des clauses ordinaires et extraordinaires des contrats de droit privé, 2016, p. 542.

243

bertrand montembault and paul morton depending on the type of third parties involved, the task is more straightforward insofar as such a contractual arrangement does not afect the rights of the third-party having suferance the loss. 2.2 Legal characterisation of mutual hold harmless clauses under French law 2.2.1 Allocation of risk between the parties The general approach in French law contracts is to rely on a combination of an exclusion of liability and a waiver of recourse, as can be seen in the example below: Each party shall bear any physical losses . . . sufered by it and its afliates in connection with the performance of this contract. Accordingly, each party irrevocably undertakes not to bring any claim or to initiate any proceedings against the other party or its afliates for any reason whatsoever.22

Although exclusions of liability and waivers of recourse are well established under French law, there is no general legal defnition of, or codifed framework for, clauses allocating liability between the parties, including on a mutual basis.23 Nevertheless, such mechanisms are increasingly found in long-term, high-value contracts, notably in the oil and gas, mining and space sectors, as well as certain infrastructure contracts. Such clauses will clearly cover contractual liability between the parties. However, the intention of the parties is generally that they should extend to capture non-contractual liability as well. As a result of the contra proferentem interpretation of such clauses in many common law jurisdictions, the drafting in English mutual hold harmless clauses will frequently explicitly refer to non-contractual liability.24 The drafting in French law contracts is often less explicit, but reference to losses incurred in connection with the performance of the contract would generally be interpreted as including contractual and non-contractual liability. However, as noted above, the position on limiting noncontractual liability between contracting parties is more complex. 2.2.2 Allocation of third-party liability 2.2.2.1 Related third parties The parties will often look to expand the scope of a mutual hold harmless regime to include a wider group of subcontractors, afliates and joint venture partners linked to the parties (referred to here as members of a party’s ‘group’). Such related third parties will generally be involved in two ways. First, the beneft of the waiver of recourse granted by each of the parties may be extended to members of the parties’ groups, meaning that each of the parties undertakes not to bring claims against the members of the other party’s wider group. 22 Quoted in L. Ravillon, Espace extra-atmosphérique – aspects contractuels, 2015, p. 11. 23 Neither the 2016 reform of the law of contracts and obligations nor the proposed reform of the law on civil liability included any proposals to defne any mechanisms for the allocation of liability between parties. See N. Dowlatshahi, S. Salem, La réforme du droit des contrats, une réforme à parfaire en matière de responsabilité, 2019, p. 4. 24 For instance, “howsoever arising, whether in contract or tort, irrespective of negligence or breach of statutory duty”.

244

mutual hold harmless clauses in francophone civil law systems Second, the parties will generally seek to extend the burden of the waiver of recourse to the members of each party’s group, such that each party is protected from claims brought by members of the other party’s group as well. However, French law, like other jurisdictions, is clear that any clause purporting to limit or exclude the tortious liability of the parties towards third parties is of no efect: “articles 1382 and 1383 of the Civil Code are a matter of public policy (ordre public), from which it follows that their application may not be infringed in advance by contract”.25 Accordingly, there are two options available for extending the crosswaiver of claims. The frst option is that the parties undertake to procure that members of their group will not bring claims against the other party. Although this will have no impact on the third party having sufered the loss, the related party to the contract will nevertheless be in breach of its contractual obligation towards the other party as a result of any claim brought by a member of its group against the other party. The second option is for each party to indemnify the other for any losses sufered as a result of a claim brought by another member of the indemnifying party’s group, as in the example below: Where one or more afliates of a party pursues the other party and/or its afliates .  .  . the frst party guarantees the other party and/or its afliates in respect of the outcome of such a claim, and shall advance all necessary funds for the defence of their interests and reimburse all amounts that they may be required to pay as direct or indirect result of such a claim.26

2.2.2.2 Unrelated third parties Occasionally, mutual hold harmless clauses may also extend to other third-party liability, beyond the members of each party’s extended group. In this case, related third parties need to be distinguished from ‘genuine’ third parties that do not have any link to the parties or the subject of the contract between them. The objective with respect to this type of third-party liability is not to extend the beneft and the burden of the cross-waiver of claims, but simply to allocate a defned risk (i.e., of a third-party claim against a party) between the parties. In this case, the parties may agree to indemnify the other party for any losses arising out of such a claim (as seen above in relation to claims from related third parties). Unlike the exclusion of liability and waiver of recourse between parties, an indemnity in respect of third-party claims (whether from related or unrelated third parties) simply transfers the burden of compensating the third party from one contracting party to the other. Under French law such an undertaking is qualifed as an ‘agreement analogous to insurance’ (pacte analogue à l’assurance), which is addressed in further detail in section 3.2.

25 Voir Cass. 1e civ, 2 août 1951, JCP 1951, II, 6592. 26 Quoted in L. Ravillon, Espace extra-atmosphérique – aspects contractuels, 2015, p. 11.

245

bertrand montembault and paul morton 3 The legal regime governing mutual hold harmless clauses under French law 3.1 Exclusions of liability and waivers of recourse 3.1.1 Validity and exceptions It is well established in French law that exclusions of liability and waivers of recourse are valid in principle,27 at least insofar as these relate to contractual liability. However, French law provides for a number of conditions for the validity of clauses dealing with liability and limitations on their efect. 3.1.1.1 Exceptions linked to the law of obligations 3.1.1.1.1 LIMITING NON-CONTRACTUAL LIABILITY

Although clauses limiting or excluding liability are, on the basis of the principle of freedom of contract, generally permitted, French jurisprudence has consistently held that such limitations or exclusions are not valid when applied to non-contractual (i.e., tortious or delictual) liability,28 on the basis that tortious or delictual liability is a matter of public policy (ordre public). However, the clear distinction between contractual and non-contractual liability in the context of limitation or exclusion of liability has been brought into question on a number of fronts and the courts have taken a more nuanced position.29 The growing consensus is that clauses limiting or excluding delictual or tortious liability will be upheld to the extent that the loss sufered is not the result of the negligence (faute) of the party having caused the loss. This is also the position adopted in the proposed reforms of the French law of civil liability.30 It is worth noting that the restrictions above relate to non-contractual liability between contracting parties and therefore necessarily apply to a small range of circumstances, rather than the usual circumstances giving rise to delictual liability where the parties are otherwise unconnected contractually. Contractual provisions dealing with delictual or tortious liability towards third parties is less problematic and is dealt with in section 3.2. 3.1.1.1.2 ESSENTIAL OBLIGATIONS

In practice, an exclusion of liability could apply to any contractual obligation. However, under French law certain obligations are characterised as ‘essential’, in respect of which an exclusion of liability or waiver of recourse (for enforcement of the obligation) will not be enforceable.31 Since the reform of the law of contracts in 2016, article 1170 of the French Civil Code provides that “any clause which deprives the essential obligation of the debtor of its substance is deemed unwritten”, refecting a well-established jurisprudential position.32 27 Cass 1re civ., 19 January 1982, no. 80–15745; more generally, see G. Viney, P. Jourdain and S. Carval, Les efets de la responsabilité, 4th edition, LGDJ, 2017, nos. 185–187. 28 C. Bloch et al., Droit de la responsabilité et des contrats, 12th ed, Dalloz, para. 2335.22. 29 G. Viney, Introduction à la responsabilité. 4th ed, LGDJ, 2019, p. 433. 30 See, for instance, articles 1284 and 1286 of the draft law of 29 July 2020. 31 F. Buy, M. Lamoureux, J. Mestre, J.C. Roda, Les principales clauses des contrats d’afaires, 2019 p. 575. 32 For instance, Cass. Req. 19 janv. 1863, DP1863, 1, p. 248.

246

mutual hold harmless clauses in francophone civil law systems Such an essential obligation can be characterised as the principal undertaking in respect of which the parties specifcally contracted. The exclusion of liability in relation to such an undertaking “would amount to removing the legally binding nature of the commitment, and would thus have the efect of transforming it into a moral commitment only”.33 Although a mutual hold harmless clause under which the parties waive any recourse against each other would be subject to the requirements of article 1170 of the Civil Code, we are not aware of any jurisprudence invalidating such a clause on this basis. In any event, the ‘essential’ nature of the relevant obligations would need to be considered in the context of a particular mutual hold harmless regime. It is certainly conceivable that in many cases, the mutual hold harmless provisions would not generally extend to the performance of the core obligations of the contract (e.g., the provision of the transportation or other services), but rather to the liability with respect to, for instance, property and personnel, as a result of the performance or failure to perform the core obligations. 3.1.1.2 Specific legislative exceptions34 Clauses excluding or limiting liability are often restricted or regulated by specifc legislation that applies either to particular sector or to certain types of contracts. It is therefore important to consider the application of such legislative provisions to any contract that includes a mutual hold harmless regime. For instance, exclusions of liability are prohibited in the context of transportation of merchandise by land. Article 133–1 of the French Code of Commerce provides that the transporter remains responsible for the loss of all objects being transported, except in the case of force majeure, and any clause to the contrary is of no efect. Similarly, in relation to maritime transportation, article L. 5422–15 of the French Transportation Code provides that “any clause having as its purpose or efect, directly or indirectly, to reduce the transporter’s liability defned in the provisions of article L. 5422–12 is null and of no efect”. 3.1.2 Efects of waiver of recourse and exclusions of liability clauses under French law 3.1.2.1 Strict interpretation Similarly to the position in many common law jurisdictions, clauses related to the liability of the parties will be subject to particular scrutiny by the courts, insofar as they derogate from the ordinary law.35 More specifcally, mutual hold harmless clauses will be interpreted strictly, on the basis that they restrict the right to act.36 33 Ibid. 34 There are a number of other legislative restrictions that would cover exclusion of liability and waiver of recourse clauses but that are not relevant in the context of hold harmless clauses. These include, for instance, the restriction on abusive clauses (clauses abusives) in standard form contracts (contrats d’adhésion), as well as consumer law restrictions on clauses limiting the rights of recourse of consumers. These legislative restrictions are not further addressed in this chapter. 35 P. Le Tourneau, Droit de la responsabilité et des contrats, Dalloz Action, 12e éd., 2020, p. 1492. 36 See Cass. 3e civ., 2 mars 2010, no. 09–12.272, in relation to a general waiver of recourse in relation to insurance.

247

bertrand montembault and paul morton Furthermore, such clauses will be at higher risk of being inefective to the extent that they appear contrary to the interests of one of the parties. In other words, they will be interpreted contra proferentem, the objective of the court in such cases being “less to defne the common intention of the parties than to neutralise a clause that arouses mistrust”.37 Often this contra proferentem principle is applied in situations where the provisions of a contract are unbalanced, which should not be the case for mutual hold harmless clauses given that these are by defnition drafted on a reciprocal basis. However, a reciprocal clause will, in the event of ambiguity, be interpreted against the party seeking to enforce it. 3.1.2.2 Exceptions linked to the nature of the fault The conduct of the party seeking to enforce the clause can also have an impact on the efectiveness of a mutual hold harmless clause, notably where such conduct constitutes intentional or wilful misconduct (faute dolosive)38 or gross misconduct (faute lourde).39 Generally speaking, the French courts will refuse to apply exclusion or limitation of liability clauses in the face of faute lourde or faute dolosive by the debtor in respect of its contractual obligations. It is commonplace to expressly provide that any waiver of recourse is dis-applied in the event of dol or faute lourde. However, even in the absence of any express reference in the contract, the French courts have repeatedly refused to apply waivers of recourse in the presence of faute lourde.40 This limitation on the application of an exclusion of liability or waiver or recourse is a key diferentiating element when comparing the efect of such clauses under French law to that under other laws. Although such clauses will frequently include a carve-out in the event of wilful misconduct, under most common law governed contracts the mutual hold harmless principle is generally drafted to apply in instances of negligence falling short of wilful misconduct. Depending on the defnitions applied, the scope for achieving this result under French law is more limited. 3.1.2.3 Exceptions linked to the nature of the loss The efectiveness of a waiver of recourse or exclusion of liability under French law in the context of mutual hold harmless clauses also needs to take into account restrictions on liability clauses that relate to personal injury. Most authors contend that, on the basis of the principles in relation to the human body in article 16–1 of the French Civil Code, that any clauses dealing with liability 37 P. Le Tourneau, Droit de la responsabilité et des contrats, Dalloz Action, 12e éd., 2020, p. 1492. 38 In French law, the notion of ‘dol’ in the context of the performance of a contract is defned as a conscious or deliberate breach of one’s obligations, but without the debtor necessarily having had the intention to cause harm (Cass. civ, 2, 10 novembre 2021, no. 19–12.660; Cass. civ, 2, 28 février 2013, no. 12–12.813; Cass. civ, 3, 27 juin 2001, no. 99–21.017 99–21.284). 39 ‘Faute lourde’ is defned as a breach that “cannot result from a mere breach of a contractual obligation, even an essential one, but must be deduced from the seriousness of the debtor’s behaviour” (Cass. com, 29 juin 2010, no. 09–11841); it is “characterised by extremely serious behaviour, bordering on dol, demonstrating the inability of the obligor to perform the contractual obligation he had accepted” (Cass. ch. mixte, 22 avr. 2005, no. 03–14112). 40 Voir Cass. 1e civ., 31 mars 1987, no. 79–17.129.

248

mutual hold harmless clauses in francophone civil law systems are inefective insofar as they relate to bodily harm. Although this is not a universally accepted position,41 it is likely that any clause seeking to re-allocate the liability for personal injury will not be enforceable. Given that mutual hold harmless clauses frequently include personal injury within their scope, this is a key limitation on such clauses where they are governed by French law, as compared to similar clauses governed by other laws. 3.2 Indemnifcation of third-party losses An indemnity in respect of third-party liability typically included in mutual hold harmless clauses constitutes what is termed an ‘agreement analogous to insurance’ (pacte analogue à l’assurance), which refers to a provision pursuant to which a person not acting in a professional capacity as an insurer undertakes contractually to indemnify the counterparty in respect of the liability of the latter towards third parties in the context of the performance of their contract.42 3.2.1 Validity of clauses indemnifying third-party losses The validity of clauses indemnifying third-party losses is well established under French law.43 Such clauses can apply to both the contractual and non-contractual (tortious or delictual) civil liability of the beneftting party. Although French law generally restricts the ability to contractually restrict one’s non-contractual liability (see section 3.1.1(A)), nothing prevents a party from agreeing that another person will bear the fnancial consequences of such liability, on the basis that the person having sufered the loss is unafected by the agreement. 3.2.2 Limitations Clauses indemnifying against third-party claims, such as those found in mutual hold harmless clauses, will be subject to certain principles and limitations applicable to insurance contracts. One of the most important principles in the context of an indemnity covering third-party liability is that French law prohibits a person from insuring against their own intentional misconduct, but does allow insurance against all other forms of misconduct.44 This means that such a clause will not be enforceable in the event of the wilful misconduct (faute intentiontionelle) of the benefciary indemnity, but will be in the face of other misconduct, even gross misconduct (faute lourde) of the benefciary. The position in relation to indemnifcation for third-party liability is therefore more permissive than the position in relation to exclusions of liability and waivers of recourse, where such a clause will also fail in the face of gross misconduct (faute lourde),45 as set out in section 3.1.2.2. Accordingly, the scope for 41 W. Dross, Dictionnaire des clauses ordinaires et extraordinaires des contrats de droit privé, 2016, p. 303. 42 Ph. Delebecque, Fasc. 115: “RÉGIME DE LA RÉPARATION – Modalités de la réparation. – Règles particulières à la responsabilité contractuelle. – Conventions relatives à la responsabilité”, JurisClasseur Contrats, 2018, p. 7. 43 Civ. 3e, 10 January 1978, no. 76–11.111. 44 Y. Aubin, T. Portwood, Les clauses récriproques d’abandon de recours et de garanties contre les recours des tiers, RDAI/IBLJ, no. 6, 2001, p. 684. 45 Cass. 1e civ. 15 avril 1961, JurisData no. 1961–000198.

249

bertrand montembault and paul morton re-allocating third-party liability is broader than it is for the liability of the parties towards each other. Acknowledgement The authors wish to express their gratitude to Sumerya Dinç and Majda Hashemi for their invaluable assistance on this chapter.

250

CHAPTER14

Lessons from the application of knock-for-knock clauses under Malaysian law Wan Mohd Zulhafz Wan Zahari

1 Introduction This chapter discusses the application of knock-for-knock in countries unaccustomed with the clause. The use of legal provisions without a proper understanding of them will cause the indemnity and hold harmless clauses to be interpreted outside their meaning, potentially opening the door to abuse by a party in a dominant position. This chapter provides a case study of how these concerns might become mixed up in Malaysia. This chapter is divided into two major sections. The frst section provides an overview of Malaysian legal framework including the statutory and judicial approach on indemnity clauses in Malaysia. The second section investigates the issues and problems in Malaysia with regard to risk allocation provisions and indemnity clauses under the oilfeld service contracts. This section also presents the research fndings of the empirical studies that have been conducted in Malaysia and the analyses on indemnity clauses under oilfeld service contracts which were used and drafted by three operators in Malaysia. 1.1 General overview of the Malaysian legal system In order to fully understand the underlying legal processes and the policies from which they derive, it is crucial to describe the Malaysian legal system. This section will explain the Malaysian legal structure, including its historical basis, the Constitution and Malaysia’s fundamental laws and institutions.1 Malaysian law is settled on the English legal system, and it is a common law jurisdiction.2 This is due to Britain’s occupation of Malaya, Sabah and Sarawak from the early 19th century until the 1960s.3 Malaysian law is a hybrid of English law, which governs most aspects of daily life, and Sharia law, which governs family and personal matters for Muslims.4 Malaysian legislation is based on the Federal Constitution of Malaysia and Acts of Parliament. The British introduced the fnal version of the Civil Law Ordinance, which was frst enacted in the Straits Settlements in 1878, a

1 Arumugam Rajenthran, Malaysia: An Overview of the Legal Framework for Foreign Direct Investment (Citeseer 2002). 2 Sharifah Suhana Ahmad, Malaysian Legal System (MLJ 1999). 3 Ibid. 4 Geofrey Wilson Bartholomew, The Commercial Law of Malaysia: A Study in the Reception of English Law (MLJ 1965).

DOI: 10.4324/9781003206798-16

251

wan mohd zulhafiz wan zahari year before Malaya gained independence in 1957.5 The formal importation of the English common law and the rules of equity into the national legal system were done through the Ordinance. The Ordinance that remains until today was revised in 1972 and renamed as the Civil Law Act 1956.6 The Federal Constitution is the supreme law of the land.7 The constitution provides for a unique dual justice system: secular laws on one side (e.g., criminal and civil) and Sharia law on the other side.8 The constitution also provides the legal framework for the laws, legislation, courts and other administrative aspects of the law. Further, it defnes the government, monarch and its powers and the rights of the citizens.9 Federal laws enacted by the Parliament of Malaysia apply throughout the country.10 There are also state laws enacted by the State Legislative Assemblies which apply in the particular state.11 In respect of the judiciary, the Malaysian Court system is infuenced by and in fact, is quite similar to the English Court system which split the Subordinate and Superior Courts.12 The Penghulu Court is the lowest level of the Subordinate Courts, and each of the specifc districts is presided by a headman that is appointed by the state government. The next level in the court hierarchy is the Magistrate’s Courts which deals with minor civil and criminal cases, and at the highest level of the Subordinate Courts are the Sessions Courts. The Superior Courts are made up of the High Court, Court of Appeal and Federal Court (which is the supreme court in the land).13 The laws of Malaysia can be divided into two types of laws: written law and unwritten law.14 Written laws are laws which have been enacted or legislated by the lawmakers. In contrast, unwritten laws are laws which are not contained in any statutes and can be found in case decisions. This is also known as common law or case law. In situations where there is no law governing a particular situation, Malaysian case law may apply. If there is no Malaysian case law, English case law can be applied.15 The principle of stare decisis also applies in Malaysian law.16 Thus, any decisions by a court higher in the hierarchy will be binding upon the lower courts. 1.2 The law of contract in Malaysia In Malaysia, parties typically have freedom of contract.17 The Privy Council in Ooi Boon Leong v Citibank N.A.18 confrmed that “parties to an agreement have much 5 JN Matson, ‘The Confict of Legal Systems in the Federation of Malaya and Singapore’ (1957) 6 Int. Comp.L.Quart. 243. 6 The CLA 1956 (Revised 1972) is in fact a consolidation of the Civil Law Ordinance 1956, Sabah’s Application of Laws Ordinance 1951 and Sarawak’s Application of Law Ordinance 1949. 7 Malaysian Constitution. Art 4. 8 Malaysian Constitution. Art 121. 9 Abdul Aziz Bari, Malaysian Constitution: A Critical Introduction (Other Press 2003). 10 Malaysian Constitution. Art 74. 11 Ibid. 12 Arfah Hamzah and Ramy Bulan, An Introduction to the Malaysian Legal System (Fajar Bakti 2003). 13 Ibid. 14 Wu Min Aun, An Introduction to the Malaysian Legal System (Kuala Lumpur: Heinemann Educational Book (ASIA) Ltd 1975). 15 Subject to sections 3 and 5 of Civil Law Act, which will be discussed later. 16 Hamzah (n12). 17 Ahmad SA Alsagof, Principles of the Law of Contract in Malaysia (MLJ 1996). 18 [1984] 1 MLJ 222.

252

lessons from the application of knock-for-knock scope to negotiate and incorporate terms acceptable to them”. Currently, there are no statutory restrictions in Malaysia on contractual provisions purporting to exclude or limit the liability of one or both of the parties in relation to indemnity and liability.19 The issue of fairness and unfairness of contractual terms is seen in two aspects: procedural unfairness and substantive unfairness. Procedural fairness refers to fairness during the contract formation process.20 A contract can be regarded as procedurally unfair when it is entered by one party through an unfair means, such as coercion, undue infuence, or fraud.21 The court may fnd that such contracts are voidable under Part III of Malaysian Contracts Act 1950.22 In contrast, substantive fairness focuses on the fairness in the allocation of substantive rights and obligations under the contract itself.23 The contract is considered to be substantively unfair if the contract terms are unfair or biased or “one-sided”.24 That said, usually, “courts have been less willing to interfere in cases of substantive fairness bearing in mind the concept of freedom of contract”.25 However, it is argued that “the development of monopolistic business, in the public and private sector alike, has made it impossible for the weaker party actually to exercise the freedoms in many cases”.26 The Malaysian judicial position on the doctrines of unconscionability and inequality of bargaining power remains unclear and ambiguous.27 In Fui Lian Credit & Leasing Sdn Bhd v Kim Leong Timber Sdn Bhd,28 Datuk Cheong Siew Fai J appeared to recognise the doctrine of unconscionability, where the court, referring to the English cases of Multiservice Bookbinding Ltd v Marden29 and Hart v O’Connor,30 stated that: In order that a party may free himself from complying with an agreement he had entered into, he must show that, in the eyes of the court, it was unreasonable. A bargain cannot be unfair and unconscionable it is shown that one of the parties to it has imposed an objectionable term in a morally reprehensible manner, that is to say, in a way which afects his conscience or has procured the bargain by some unfair means.

19 Toby Hewitt, ‘An Asian Perspective on Model Oil and Gas Services Contracts’ (2010) 28 J.Energy Nat.Resour.L. 331, 333. 20 Hart v O’Connor [1985] A.C. 1000. 21 Part III of Malaysian Contracts Act 1950. 22 Maryam Rafei and Nazura Abdul Manap, ‘Legal Position of Click Wrap Agreement’, Proceedings of International Conference on Computer Communication and Management (ICCCM 2011) (2011). 23 Naemah Amin, ‘Protecting Consumers against Unfair Contract Terms in Malaysia: The Consumer Protection (Amendment) Act 2010’ (2013) 1 MLJ 1. 24 Norliza Abdul Hamid and Hariati Mansor, ‘The Legal Implications of the Consumer Protection (Amendment) Act 2010 on Contract Terms in Malaysia’, Annual Summit on Business and Entrepreneurial Studies (ASBES 2011) Proceeding (UiTM 2011). 25 Ibid. 26 John N Adams and Roger Brownsword, ‘The Ideologies of Contract’ (1987) 7 Leg.Stud. 205, 208. 27 Andrew Phang Boon Leong, Cheshire, Fifoot and Furmston’s Law of Contract, vol. 2 (Butterworths Asia, Singapore 1998). 28 [1991] 1 CLJ 522. 29 [1979] Ch. 84. 30 [1985] A.C. 1000.

253

wan mohd zulhafiz wan zahari However, the judge did not further clarify the status of Hart31 in the Malaysian context, specifcally he did not clarify whether or not the doctrine of unconscionability could be adopted as an independent ground for relief.32 Moreover, the principle of inequality of bargaining power formulated by Lord Denning in the case of Lloyds Bank v Bundy33 has also been rejected by the local courts because there is no established precedent in Malaysia.34 This could be seen in the judgment of Visu Sinnadurai J. in Polygram Records Sdn Bhd v The Search.35 In another case, Gopal Sri Ram JCA explained in his judgment at the Court of Appeal in Saad Marwi v Chan Hwan Hua & Anor36 that the doctrine of unconscionability, which is also referred to as the doctrine of inequality of bargaining power, should be explicitly recognised in Malaysian law. Gopal Sri Ram JCA stated that the time has arrived when we should recognize the wider doctrine of inequality of bargaining power. . . . What is therefore called for is a fairly fexible approach aimed at doing justice according to the particular facts of a case. . . . That brings me to the third alternative. This is to adopt the English doctrine [of unconscionability] but apply it in a broad and liberal way as in Canada.37

Unfortunately, the subsequent case at the Court of Appeal did not follow the principles in Saad Marwi. In the case of American International Assurance Co Ltd v Koh Yen Bee (f),38 on the issue of the applicability of the doctrine of unconscionability, Abdul Hamid Mohamad JCA chose not to follow Gopal Sri Ram JCA’s view. He contended that the two cases were completely contradictory and that section 14 of the Contracts Act of 1950 only recognised coercion, undue infuence, fraud, misrepresentation and mistake as elements that infuence free consent.39 It has been argued that this ruling fails to acknowledge that, in reality, it is necessary to strike a balance between the need for fair contracts and the competing interest of legal certainty.40 This situation had left Malaysian law in a state of ambiguity and uncertainty regarding its recognition and acceptance of the doctrines of unconscionability and inequality of bargaining power.41 As a result, any party who is subjected to procedurally unfairness has no recourse to seek justice and fle a cause of action in court for unconscionability and unequal bargaining power.

31 Ibid 32 Ibid. 33 [1975] QB 326. 34 Cheong May Fong, Contract Law in Malaysia (Sweet & Maxwell Asia 2010). 35 [1994] 3 MLJ 127. 36 [2001] 3 CLJ 98. 37 Ibid. 38 [2002] 4 MLJ 301. 39 Ibid., 318. 40 Cheong May Fong, ‘A Malaysian Doctrine of Inequality of Bargaining Power and Unconscionability After Saad Marwi’ [2005] MLJ 4. 41 Siti Aliza Alias and Zuhairah Arif Abdul Ghadas, ‘Inequality of Bargaining Power and the Doctrine of Unconscionability: Towards Substantive Fairness in Commercial Contracts’ (2012) 11 Austl.J.Basic&App. Sci. 331.

254

lessons from the application of knock-for-knock 1.2.1 Statutory and judicial approach on indemnity clauses in Malaysia This section demonstrates the statutory and judicial approach to indemnity clauses in Malaysia. Sections 77 and 78 of the Malaysian Contract Act 1950 govern the law relating to indemnity. However, there are no legal constraints on contractual clauses that seek to exclude or limit the liability of one or both parties with respect to indemnifcation and liability.42 The High Court in Amman Singh v Vasudevan43 held that section 77 of the Malaysian Contracts Act 1950 defnes what a contract of indemnity is,44 whereas section 78 provides the rights of the indemnity holder when he is sued by the promisor.45 In the context of section 77, the Supreme Court in South East Asia Insurance Bhd v Nasir Ibrahim46 held that a contract of indemnity is “a contract by which one party promises to save the other from loss caused to him by the conduct of the promisor himself, or by the conduct of any other person”. Such consideration provided under contract of indemnity is regarded as valuable and legally acceptable. In Macon Works & Trading Sdn Bhd v Phang Hon Chin & Anor,47 the court applied the common law meaning of consideration and followed the classic statement contained in Currie v Misa,48 where it was observed that “[a] valuable consideration may consist either in some right, interest, proft or beneft accruing to one party or some forbearance, detriment, loss or responsibility given, sufered or undertaken by the other”. Gunn Chit Tuan SCJ in delivering the judgment in South East Asia Insurance Bhd v Nasir Ibrahim49 said that the essence of consideration was that the promisee had imposed some kind of burden or detriment, for example, a legal responsibility.50 In this case, 42 Hewitt (n.19) 331; Wan Zulhafz, ‘Unfair Contract Terms Act 1977: Does It Provide a Good Model in Regulating Risk Allocation Provisions in Oilfeld Contracts in Malaysia?’ (2015) 8(1) Int.J.Trade Glob. Mark. (IJTGM) 3. 43 [1972] 1 LNS 7. 44 Section 77 of Malaysian Contract Act reads: Contract of Indemnity A contract by which one party promises to save the other from loss caused to him by the conduct of the promisor himself, or by the conduct of any other person, is called a ‘contract of indemnity’. Illustration A contracts to indemnify B against the consequences of any proceedings which C may take against B in respect of a certain sum of RM200. This is a contract of indemnity.

45 Section 78 of Malaysian Contract Act reads: Right of Indemnity holder when sued The promisee in the contract of indemnity, acting within the scope of his authority, is entitled to recover from the promisor –

(a) all damages which he may be compelled to pay in any suit in respect of any matter to which the promise to indemnify applies; (b) all costs which he may be compelled to pay in any such suit if, in bringing or defending it, he did not contravene the orders of the promisor, and acted as it would have been prudent for him to act in the absence of any contract of indemnity, or if the promisor authorized him to bring or defend the suit; and (c) all sums which he may have paid under the terms of any compromise of any such suit, if the compromise was not contrary to the orders of the promisor, and was one which it would have been prudent for the promisee to make in the absence of any contract of indemnity, or if the promisor authorized him to compromise the suit. 46 47 48 49 50

[1992] 2 MLJ 355; [1993] 1 SCR 89, SC. [1976] 2 MLJ 177, HC. (1875) LR 10 Exch 153. [1992] 2 MLJ 355; [1993] 1 SCR 89, SC. Ibid., 356.

255

wan mohd zulhafiz wan zahari the Supreme Court held that the promisee had to assume legal responsibility to its detriment by providing a third-party indemnity.51 In sum, it could be said that the law in Malaysia recognises the concept of indemnity. However, the law has not provided explanation with regards to contractual indemnity.52 An example of the use of an indemnity clause in the oilfeld service contracts in Malaysia can be seen in the case of Sabah Shell Petroleum Co Ltd & Anor v The Owners of and/or any other Persons Interested in The Ship or Vessel The “Borcos Takdir”.53 In this case, the plaintif (“the operator”) was engaged in ofshore oil and gas exploration and production activities in the ofshore areas of Sabah and Sarawak.54 The plaintif operated ofshore platform structures including underwater oil pipeline and had deployed the defendant’s vessel, the Borcos Takdir, to deliver such cargoes for food, equipment, tools and machinery to the plaintif’s ofshore platform structures. However, the vessel damaged the pipeline.55 The plaintif thus commenced the action against the defendant for the substantial loss and damage they had sufered as a consequence of the defendant’s negligence and/or breach of duty or breach.56 The defendant contended that by virtue of Clause 21 of the Contract, there was an agreement that any claim for physical loss to the property of Operator would be subject to a contractual limit of RM5,000,000.57 Clause 21 provides as follows: Clause 21 – responsibilities and indemnities (1) The Contractor shall be absolutely liable for and shall indemnify the Company (i.e., the Operator), its Associated Companies, Co-Venturers and PETRONAS from and against any and all physical loss of or damage to: (a) WORK or any part thereof, howsoever arising, and shall, at its own costs and expense, restore the WORK lost or damaged, replace or repair any unfxed MATERIALS or consumables and the like which have been lost or damaged, remove and dispose of any debris and proceed with the performance of the WORK; and (b) Property of the Company, its Associated Companies, Co-Ventures and PETRONAS caused by or contributed to by or resulting from any act, omission, fault, lack of due diligence or negligence of the Contractor, its Sub-contractors or any of their respective employees, servants and agents unless such loss or damage was caused by the sole negligence of the Company The Contractor’s liability under this Clause 21(1)(b) herein shall be limited to Malaysian Ringgit Five Million (RM5.000.000) for any one accident or series of accidents arising out of any one event. The liability in excess of such limit shall be governed by applicable law.58

51 Ibid. 52 Wan M Zulhafz and Nasarudin Abdul Rahman, ‘Unfair Risk Allocation in Oil and Gas Upstream Service Contracts in Malaysia: The Necessity for Oilfeld Anti-Indemnity Act’ (2020) Int.J.Bus.Soc. (IJBS) 21. 53 [2012] 5 MLJ 515. 54 Ibid. 55 Ibid. 56 Ibid., 516. 57 Ibid., 558. 58 Ibid.

256

lessons from the application of knock-for-knock The High Court held that the above clause is an indemnity clause that provides coverage in two tranches. The frst tranche is the amount of RM5,000,000 and is payable by the contractor regardless of whether liability is not wholly attributable to it, as long as the incident in question was not caused by the sole negligence of the operator.59 The second tranche is only available if liability has been determined and adjudicated in accordance with the law.60 In relation to the interpretation and construction of contracts, the court in Malaysia will generally apply an objective or reasonable man test in construing an indemnity clause. In Lee Kam Wah v Associated Asian Securities (Pte) Ltd,61 the plaintif agreed “to indemnify the company (the defendants) of any losses and damages that may arise on any contracts written by (him)”.62 At the foot of the document, the plaintif signed and “(agreed) to the above terms and conditions and hereby (indemnifed) the company for any losses or damages arising from any contracts written or any transactions entered into by (him)”.63 The defendants sought indemnity from the plaintif.64 The High Court held that the wording of the indemnity clauses was inelegant and implied that indemnities had to cover every conceivable loss or damage.65 Further, the court stated: It would be wrong to give such clauses such an open-ended construction without trying to ascertain what the parties really intended from the terms of the document in question and the surrounding circumstances at the time the parties entered into the agreement unless the evidence clearly established that the plaintif was negligent or had been a party to the fraudulent transaction.66

At the Supreme Court level, in Ayer Hitam Tin Dredging Malaysia Bhd v YC Chin Enterprises Sdn Bhd,67 the Court in ascertaining the meaning of “the appropriate indemnity clauses” held that the existence of an agreement depends upon the intention of the parties, who must be ad idem, which may be inferred from (1) the language used, (2) the parties’ conduct having regard to the surrounding circumstances and (3) the object of the contract.68 In another case, the Federal Court in Malaysia construed the indemnity clause as exemption and exclusion clauses. This can be seen in the case of CIMB Bank Bhd v Maybank Trustees Bhd,69 where the Federal Court agreed with the Court of Appeal’s approach and interpreted the indemnity clause under Clause 14.1 of the Trust Deed as an exemption clause where the Court of Appeal referred to the principle of exemption clause under English case of Karsales (Harrow) Ltd v Wallis,70

59 60 61 62 63 64 65 66 67 68 69 70

Ibid., 565. Ibid. [1991] 3 MLJ 286. Ibid., 287. Ibid. Ibid. Ibid. Ibid. [1994] 2 MLJ 754. Ibid., 755. [2014] MLJU 117. [1956] 2 All ER 866.

257

wan mohd zulhafiz wan zahari where Lord Denning ruled that “it is now settled that exempting clauses of this kind, no matter how widely they are expressed, only avail the party when he is carrying out his contract in its essential respects”.71 He further stated that if the party “has been guilty of a breach of those obligations in a respect which goes to the very root of the contract, he cannot rely on the exempting clauses”.72 On the other hand, the Court of Appeal held that the indemnity must be resolved by the proper construction of the exclusion clauses. The Court of Appeal had then cited few cases relating to exclusion clause such as Hotel Anika Sdn Bhd v Majlis Daerah Kluang Utara,73 Anderson v Fitzgerald,74 and Guardian Assurance Co Ltd v Condogianis.75 When there is ambiguity between the indemnifcation clause and the exemption clause, the court will apply the contra proferentem rule to settle the issue. In Syarikat Uniweld Trading v The Asia Insurance Co Ltd,76 the appellant ran a motor vehicle repair and welding shop. A fre broke out as a result of an unintentional leak in a gasoline gas pipe, resulting in the destruction of a third-party vehicle. The appellant compensated the owner of the motor vehicle and sought indemnity from the respondent. The appellant sought indemnifcation because the insurance policy expressly stated that the respondent would indemnify the appellant for any sums for which the appellant became legally obligated to pay. However, the insurance policy excluded the appellant from being held liable for accidental damage to third-party property caused by the appellant’s negligence. That exemption clause was invoked by the respondent. The High Court granted the appeal and held that, when there is an ambiguity between the indemnity clause and the exemption clause, the insurance policy should be construed in favour of the appellant under the contra proferentem rule. As a result, the insurance company (respondent) was ordered to indemnify the appellant. In Malaysia, contracting parties typically use extensive and broad phrases such as “will not be liable for any damage however caused”, “will not in any circumstances be responsible” and “arising from any cause whatsoever” in order to be exempted from liability for negligence.77 These expressions are commonly used to refer to liability for negligence.78 However, it is important to note that the party making the claim must be able to demonstrate that the damage was not caused by his negligence and that he took reasonable steps to fulfl his duty of due diligence.79 In Chin Hooi Nan v Comprehensive Auto Restoration Service Sdn Bhd,80 Siti Norma Yaakob J found that it is well established that an exemption clause, no matter how broad and general, does not relieve the respondents of the burden of proving that the

71 72 73 74 75 76 77 78 79 80

Ibid., 868. Ibid., 869. [2007] 1 MLJ 248. (1853) 4 HLC 484; (1853) 10 ER 551. (1919) 26 CLR 231. [1996] 2 MLJ 160. Alsagof (n17). Ibid. Ibid. [1995] 2 MLJ 100.

258

lessons from the application of knock-for-knock damage was not caused by their negligence and misconduct.81 They must demonstrate that they fulflled the duty and responsibility with reasonable diligence and care.82 In Sekawan Guards Sdn Bhd v Thong Guan Sdn Bhd,83 Wan Adnan J held that the appellant carry the burden of demonstrating that the exemption clause applied and that the clause protected it from liability unless the loss was caused solely by his negligence.84 To succeed, the appellant must demonstrate that other people were negligent.85 In the above case, the appellant had contracted to provide security services at the premises of the respondent.86 At the premises, a theft occurred, resulting in the loss of the respondent’s goods, which included 169 cartons of cigarettes worth RM158,237.15.87 Under the contract, the appellant’s liability was limited to RM100,000.88 The respondent fled a breach of contract lawsuit against the appellant. In addition, the respondent sued for negligence.89 The appellant relied on an exemption clause in the contract.90 However, the court found that the appellant was solely negligent and held that the exemption clause did not protect the appellant.91 In Malaysia, a party cannot use an exemption clause to protect himself from the consequences of a fundamental breach of contractual obligations.92 In Sze Hai Tong Bank Ltd v Rambler Cycle Co Ltd,93 the carrier’s agent wrongfully released the respondents’ goods to the consignee, who produced a written indemnity by the consignee’s bank (the appellant) in favour of the carrier, without the production of the bill of lading.94 The carrier’s agent was aware that this act was an illegal act.95 The respondent sued the carrier because the consignee never paid for the goods.96 The carrier admitted liability to indemnify the carrier if it was found liable.97 The carrier was argued to be immune from liability under Clause 2 of the bill of lading, which stated that the liability of the carrier shall be deemed to cease absolutely after the goods were discharged from the ship.98 The Privy Council held that the carrier was in breach of contract and was not protected by the exemption clause.99 The clause must be limited and modifed to the extent necessary to give efect to the main object and intent of the contract.100 The court ruled that the carrier should not be allowed to deliberately disregard its obligation to deliver, that is, against production of the bill 81 Ibid. 82 Ibid. 83 [1995] 1 MLJ 811. 84 Ibid., 816. 85 Ibid. 86 Ibid., 813. 87 Ibid., 814. 88 Ibid. 89 Ibid. 90 Ibid. 91 Ibid., 816. 92 Alsagof (n17) 278. 93 [1959] MLJ 200. 94 Ibid., 201. 95 Ibid. 96 Ibid. 97 Ibid., 200. 98 Ibid., 201. 99 Ibid., 200. 100 Ibid., 202.

259

wan mohd zulhafiz wan zahari of lading.101 A fundamental breach of obligations of a contract cannot be allowed to pass unnoticed under the cloak of a general exemption clause.102 It is important to note that, procedurally, a party cannot sue for a breach of contract while also claiming indemnifcation from the same party who committed the purported breach. The High Court ruled in Siong Electronic Industries (1981) Sdn Bhd v Sanyo Sales & Service Sdn Bhd,103 that a party cannot sue for a breach while also seeking indemnifcation from the party who has committed the alleged breach.104 The application of indemnity clauses was limited by the High Court.105 The court held that indemnity clauses were only applicable in situation where it saved the promisee from loss caused by third-party claims. In law, indemnity clauses denote a contract in which the promisor undertakes an original and independent obligation to indemnify the promise against third-party claims.106 In terms of evidence, the indemnitee must be able to demonstrate to the court that he sufered loss prior to fling a claim for such indemnity. In Azman bin Mahmood & Anor v SJ Securities Sdn Bhd,107 there was a question about whether a customer could be held liable on his indemnity for losses incurred as a result of a transaction where the loss claimed could not be proven.108 The majority view of the Federal Court was negative on this question.109 Based on the discussion above, it could be argued that Malaysian law applicable to indemnity clauses is too brief, unclear and incomprehensive. The Contract Act under sections 77 and 78 only lays down the defnition of contract of indemnity and provides the right to the indemnity holder to sue against the indemnitor. The courts use the principle of the “reasonable man” in construing the indemnity clause, i.e., by looking at (1) the intention of the parties, (2) the language used and (3) the surrounding circumstances when the time the parties entered into contract. It could be argued that the courts have taken the correct approach to interpret indemnity clauses. However, it could also be argued that it is inappropriate for the court to go beyond that and treat indemnity clauses in the same way as exemption and exclusion clauses. This is because, indemnity clauses are used by the parties in oilfeld service contracts to allocate risk. Conventionally, indemnity clauses distinguish between negligence and gross negligence. Treating indemnity clauses like exemption and exclusion clauses may defeat the function of indemnity clauses. For example, the indemnitee may seek indemnifcation even if the damage was solely caused by his own negligence. However, under the principle of exemption clauses from the case of Sekawan Guards Sdn Bhd v Thong

101 102 103 104 105 106 107 108 109

Ibid. Ibid., 200, 202. [1997] MLJU 162. Ibid. Ibid. See Halsbury’s Laws of England, vol. 20 (4th edn, Butterworths 1983), para. 345. [2012] MLJU 660. Ibid., 6. Ibid., 16.

260

lessons from the application of knock-for-knock Guan Sdn Bhd,110 the indemnitee was not allowed to claim for indemnity since the damage arose solely from his own negligence. 1.3 Empirical study of oilfeld service contracts in Malaysia An empirical study was conducted in Malaysia during the period of February to April 2014. Seven case studies were done and the purpose was to elicit the opinion of key players in the oil and gas industry on the issue of risk allocation. The empirical study also sought to ascertain their attitudes towards the current trends of indemnity clauses under oilfeld service contracts in Malaysia as well as their perception on how to address this problem. It is important to study actual contracting practice because, as Korobkin suggests, it is useful “to describe in-depth the contracting patterns and norms generally followed by a particular type or group of contracting party”.111 In addition, Eigen suggests that empirical are a signifcant tradition in legal scholarship because they help scholars “understand the diversity of disciplinary approaches and framings of questions about contracts raised in modern empirical explorations”.112 Eigen also argues that it is valuable to concisely articulate the interrelation between contract doctrine, theory and empirics.113 The data source for these studies was primarily semistructured interviews of ten respondents from seven organisations representing the contractor and operator. It is suggested that the low response rate could be indicative of a limited awareness of the subject matter of this research among the professionals in the oil and gas industry in Malaysia. Secondary data consisted of documentary evidence including samples of indemnity clauses of oilfeld service contracts drafted by three operators which were provided by the contractors and feld notes written by the researcher during the interviews. The present empirical study was conducted to support a previous empirical study that was done in 2008. In the previous study, a total of three case studies were conducted in diferent companies in Malaysia in respect of their experiences and perceptions about specifc issues on procurement in the oil and gas industry in general.114 The fndings indicate that there is a strong objection in the Malaysian oil and gas industry to the one-way adversarial style of the operator-contractor relationship and the way the contractor is being forced to be on the receiving end when it comes to price fuctuation, overall risks and increased workloads.115 Even though the previous study is not designed specifcally to deal with the issue of contractual risk allocation provision, there was clear assertion by the contractors in expressing their

110 [1995] 1 MLJ 811. 111 Russell Korobkin, ‘Empirical Scholarship in Contract Law: Possibilities and Pitfalls’ [2002] U.Ill.L.Rev. 1033, 1040. 112 Zev J Eigen, ‘Empirical Studies of Contract’ [2012] Ann.Rev.L.Soc.Sci. 2. 113 Ibid. 114 Mohammad Fadhil Mohammad, ‘Procurement Strategies for the Oil and Gas Industry: To Capture Changing Values and Dealing with Multi Cultural Complexity’, The Proceedings of the International Conference on Construction and Building Technology (ICCBT2008), Universiti Teknologi MARA (UiTM), Malaysia (UiTM, Malaysia 2008) 33. 115 Ibid.

261

wan mohd zulhafiz wan zahari dissatisfaction pertaining to unequal bargaining and the dominant position held by the operator in negotiating the contractual terms. There are three main questions to be addressed in this study. The frst question is that, considering the relationship between operators and contractors, to what extent do both parties exercise free will and have equality of bargaining power in negotiating the contractual terms? The second question is, are the risk allocation provisions and the indemnity clauses fair to both parties? The third question is, to what extent does the insurer provide coverage in respect of the risks which have been indemnifed by the parties? In addressing these questions, the discussion will be based on the legal theories pertaining to the concept of contractual risk allocation and indemnity clause, the theory of freedom of contract, the doctrine of inequality of bargaining power and fairness.116 These legal theories informed the construction of the interview questions. The author also used the theories as a tool for data analysis. 1.4 Research design The research method chosen for this study was semi-structured interview with oilfeld contractors and operators. Lists of predetermined open-ended questions were posed to the respondents in person. On some occasions, questions developed naturally during the course of the interview. This research method was chosen because it has some level of predetermined order but still allows fexibility in the way issues are addressed by interviewees.117 It also contains high validity as it scrutinises the ideas of the interviewees about the subject matter. Hence it encourages in-depth knowledge sharing by the respondents. Therefore, the researcher has a chance to pose complex questions and issues. Moreover, new ideas could also be discussed with the respondents, which is advantageous for the research. The questions posed during the interviews can be divided into two parts. The frst part is connected to the background of both the company and the interviewees. Meanwhile, the second part relates to the research investigation. This part deals with the process of (1) contract formation; (2) contractual negotiation; (3) preparation of standard form contracts; (4) interviewees’ experience with regards to risk allocation provisions, indemnity clauses and insurance requirements; and (5) interviewees’ views on the law and practice of risk allocation provisions and indemnity clauses in Malaysia. 1.4.1 Research process The semi-structured interviews were conducted with fve companies representing the contractor, as well as one company and one legal frm representing the operator. The respondents were (1) the heads or managers of the participating companies’ legal departments, (2) contract managers, (3) procurement managers, (4) principal technical managers, (5) head project managers of the companies and (6) a lawyer 116 Wan M Zulhafz, ‘On the Contractual Risk Allocation in Oil and Gas Projects’ (2017) Law Rev. (LR) 168. 117 K Louise Barriball and Alison While, ‘Collecting Data Using a Semi-structured Interview: A Discussion Paper’ (1994) 19 J.Adv.Nursing 328.

262

lessons from the application of knock-for-knock who used to litigate for operators in court. The contractors were selected from large, medium and small companies. The respondents were chosen due to their prominence and experience in contractual matters. The large companies are usually equipped with complete legal and contract departments, whereas the medium size company might have a legal department but not necessarily a contract department, since contractual matters are normally handled by the engineer who is also the principal technical manager. On the other hand, most of the small companies do not have a legal department or a contract department. Therefore, a procurement manager who comes from a technical background would handle contractual matters. The author experienced some difculties with getting appointments with the respondents, especially the operators, who are quite reluctant to be interviewed and not easily accessible. The respondents were given the questions together with a consent form before the interview took place to allow them to have a general idea of what was expected from them during the interview. The interviews were conducted in their ofces and the respondents were required to return the signed consent forms before the interview took place. This was done to ensure that the respondents fully understood the subject matter of the research and to confrm that they were voluntarily participating in the research. Both the recording of the interview and the notes taken during the interview have been kept confdential. All information about the respondents is made anonymous. 2 Description of the case studies This section will report the evidence found from the case study conducted amongst oil and gas companies in Malaysia. • Operator A Operator A is the largest oil company in Malaysia. It engages in locating, exploring and producing oil and gas reservoirs. It also engages in oil and gas well drilling, as well as engineering, constructing and commissioning natural gas pipelines and associated facilities. The company is based in Kuala Lumpur, Malaysia. • Operator B Operator B is a multinational company engaging in exploration activities. It entered Malaysia in 1999 and currently holds the majority interest in seven separate production-sharing contracts and three gas-holding agreements. The company is based in Kuala Lumpur, Malaysia. • Operator C Contractor C principally engages in exploration and production activities, as well as operation and maintenance of foating, production storage and an ofoading tanker facility. • Contractor A Contractor A’s involvement in the oil and gas sector focuses on pipeline services such as pre-commissioning; commissioning and de-commissioning; ofshore transportation 263

wan mohd zulhafiz wan zahari and installation; operation and maintenance; fabrication and construction; topside major maintenance and hook-up commissioning; Engineering, Procurement, Construction and Commissioning (EPCC); onshore pipeline and construction; underwater services; and ship management and catering. Contractor A has entered into a few contracts with Operators A, B and C. • Contractor B Contractor B is one of the world’s largest integrated oil and gas services and solutions providers. This company provides end-to-end solutions and services to the upstream petroleum industry and covers activities such as installation of ofshore pipelines and structures, fabrication of ofshore structures, accommodation and support vessels, drilling vessels, hook-up and commissioning, topside maintenance services, underwater services, ofshore geotechnical and geophysical services, project management, diving services, ofshore support services, infrastructure and specialised steel fabrication works. Contractor B has a number of ongoing jobs with Operators A, B and C. • Contractor C Contractor C has been active in the oil and gas industry for 35 years. It holds a business license to trade with Operator A. It is a pioneer in supplying metering and regulating skids in Malaysia for natural gas transmission and distribution. The company provides EPCC for custody transfer and non-custody transfer metering and system integration for onshore and ofshore applications. It also represents and is business partners with global leaders in the areas of control valves and instrumentation, measurement and fow management, safety, electrical and environmental work. Contractor C has signed a few contracts with Operators A, B and C. • Contractor D Contractor D provides quality and value-added products and services, especially valve-related products in the oil and gas industry. Contractor D has experience in dealing with and providing its product and services to Operators A and B. • Contractor E Contractor E engages in trading and supply of protective coating systems in the oil and gas industry. Apart from that, it also specialises in the design, manufacturing and installation of pipeline strengthening, repair and protection systems for oil and gas installations. Operator A is one of the main clients of Contractor E. • Legal Firm Z Legal Firm Z specialises in handling cases that are related to maritime, vessels and disputes in oil and gas contracts, and most of its clients are operators. The frm is located in Kuala Lumpur. Table 14.1 provides a summary for better understanding about the number of respondents involved in this study. The next section will present the fndings on how operators and contractors perceive the distribution of contractual risk during contract formation. 264

lessons from the application of knock-for-knock Table 14.1 Categories of Respondents Categories of units of analysis

Contractor

Operator 

Respondents No.

Units of analysis

No. of respondents

Designation

1

Contractor A

2

2

Contractor B

2

3

Contractor C

2

4 5 6 7

Contractor D Contractor E Operator A Legal Firm Z

1 1 1 1

Legal head + contract manager Legal manager + contract manager Legal manager + principal technical Procurement manager Procurement manager Head project Legal practitioner

10

 

Total number of respondents

2.1 Findings Two fndings of empirical study will be discussed in this section: an analysis on the perception of the contractual formation process between operators and contractors, and an analysis on the issues of risk allocation and indemnity clauses which are commonly used by the parties. The analysis of indemnity clauses will be divided into four parts: (1) liability with regard to personal injury or death of employees and loss of property of the parties; (2) third-party claims; (3) liability with regard to pollution; and (4)insurance coverage. 2.2 Perception of the contractual formation process This section presents the evidence on how the respondents perceive contractual negotiation during the contract formation process and also whether there is equal bargaining between operators and contractors.118 The case study was undertaken with the aim to primarily investigate the extent of contractual formation process that leads to the distribution of risk allocation provision in oilfeld service contracts. Competitive tendering is a standard practice for operators to obtain information on the optimal price in the contracting market.119 Tendering is part of the contract formation process. “Formation” in this context means the process by which a contract comes into being.120 Usually, the operator would invite several contractors to bid for 118 Wan M Zulhafz, ‘Perception of Contractual Risk Allocation in the Oil and Gas Contracts in Malaysia’ (2018) 11 Int.J.Trade Glob.Mark. (IJTGM) 127. 119 Kees Berends, ‘Engineering and Construction Projects for Oil and Gas Processing Facilities: Contracting, Uncertainty and the Economics of Information’ (2007) 35 Energy Pol’y. 4260; MJ Greaves, ‘Understanding and Managing Risk in Real Estate Investments’ (1983) 1 Surveyor Valuer (SISV) 21. 120 Robinson M Nigel and Lavers P Anthony, Construction Law in Singapore and Malaysia (Butterworths & Co (Asia) Pte Ltd 1988) 68.

265

wan mohd zulhafiz wan zahari the oilfeld service job in accordance with detailed technical specifcations outlined in the bid invitation121 and subsequently issue a package of documentation including a blank form tender and the instruction to tenderers. This package is called Invitation to Tender (ITT) or sometimes Invitation to Bid (ITB). Negotiation of the contractual terms is crucial before the operator accepts one of the bids and concludes the contract, as it enables the parties to identify one or more incompatibilities between them and work to fnd a mutually acceptable solution.122 Generally, standard forms of contract are not used for major oil and gas projects; in almost all cases, stand-alone bespoke contracts are used by the major oil companies which bear little resemblance to the well-established standard forms except perhaps the international FIDIC123 contract general conditions, from which a lot of bespoke construction contracts are derived.124 Most major oil companies, large independents and national oil companies have developed their own company of contract forms.125 One of the respondent representing contractor remarked: The operators or the clients will usually request a quote by issuing RFQ or sometimes, they may issue more formal request for proposals by way of RFP, ITT or ITB to the contractors. So, if the contractors wish to perform the described work, then they will respond with the price that they would charge. The acceptance of the bid by the client is considered as an “award” of the contract. These documents i.e. the RFQ, RFP, ITT and ITB contain COC (Condition of Contract) that has already been prepared by the operators. The operators, they prefer to use their own customized contract or the COC. This COC would be given to us (i.e. the contractors) as a base to initiate contractual negotiation.126

Most of the oilfeld service contracts are based on one standard model form or another and they tend to be similar in basic content with few signifcant modifcations depending on the particular commercial activities involved.127 Standard forms of contract are rarely utilised for signifcant oil and gas projects; instead, the major oil frms almost always use stand-alone customised contracts that bear little relation to the well-established standard forms.128 The operators will often keep their own database of model form of contracts and use them as a foundation in contract negotiations.129 This practice apparently saves the parties time and transaction costs and is more convenient, as they are working on the model form of contracts

121 Owen L Anderson, ‘The Anatomy of an Oil and Gas Drilling Contract’ (1989) 25 Tulsa L.J. 359. 122 David A Hensher and John Stanley, ‘Transacting under a Performance-Based Contract: The Role of Negotiation and Competitive Tendering’ [2008] Transportation Research Part A: Policy and Practice 1143. 123 FIDIC stands for Federation International Des Ingenieurs-Conseils. 124 Vincent Hooker, ‘Major Oil and Gas Projects - the Real Risks to EPC Contractors and Owners’ (2010) 26 Const.L.J. 98. 125 Cary A Moomjian, ‘Drilling Contract Historical Development and Future Trends Post-Macondo: Refections on a 35 Year Industry Career’, IADC/SPE Drilling Conference and Exhibition on 7th March 2012 in San Diego, California, USA (Society of Petroleum Engineers 2012) 9 http://www.drillingcontractor.org/wp-content/uploads/2012/04/Drilling-Contract-Historical-Development-and-Future-Trends-PostMacondo.pdf. 126 Respondent 2 from Contractor A. 127 Ibid. 128 Hooker (n126). 129 Hewitt (n19) 331.

266

lessons from the application of knock-for-knock with which they are already familiar.130 One of the respondents confrmed this practice: Operators usually prepare the contracts. Many oil and gas contracts from my experience tend to be “standard bespoke” i.e. they are “standard” with respect to the individual principal contracting entity – but on an industry basis – bespoke as they are not written by e.g. standards committees or seek industry participation in their drafting of the standard terms.131

The contracts proposed by operators are subject to qualifcation by contractors that tender for the work, resulting in negotiated contracts.132 The most important single factor in any contract negotiation is the balance of bargaining power.133 The inequality of bargaining power during contractual negotiation can create contractual unfairness between the parties. The dominant position of oil companies or the operators relative to many contractors in the oil and gas industry is such that they may assume that they can impose whatever conditions they wish.134 During the contract negotiation process, there is a high possibility of inequality in bargaining power to take place between the operators and contractors, especially when there is “imposition of will by a dominant party on a party with inferior economic bargaining power, who is being unfairly coerced into indemnifying the dominant party, stating that the result will rest on the relative bargaining power of the parties”.135 Based on the fndings of the case studies, all respondents agreed that the operators hold a dominant position during contractual negotiation process. For example, one respondent said: Of course, the operator would have a better and dominant position in negotiating the contract. The operator is the one who create the job opportunity for the contractors. As contractors, we provide them with services so as to make some proft. Sometimes, whether we like or not, we are going to agree with most of the conditions set out by the operator.136

Another respondent also confrmed this scenario: The operators are our clients as well as the project owners; of course, they have better position in negotiating the contracts.137

One of the respondents claimed that this scenario happened because of contractors’ desperation for jobs: Basically, yes, the operators have greater bargaining power over the contractors. In order to secure a job, the contractors have lesser rights to make bargaining. We can say that the contractors actually desperately need job from the operators, while the operators have

130 Ibid. 131 Respondent 5 from Contractor C. 132 Leslie Edwards, Practical Risk Management in the Construction Industry (Thomas Telford 1995). 133 Chris Thorpe, Fundamentals of Upstream Petroleum Agreements (C P Thorpe Ltd 2008) 282. 134 Peter Cameron, ‘Liability for Catastrophic Risk in the Oil and Gas Industry’ [2012] IELR 207, 207; also see ibid. 135 N Stephen Kinsella, ‘Oilfeld Indemnity and “Seperate Insurance” Provisions in the Wake of Getty Oil’ [1994] Tex.Oil Gas L.J. 29. 136 Respondent 3 from Contractor B. 137 Respondent 4 from Contractor B.

267

wan mohd zulhafiz wan zahari number of players (candidates) to do the job. For example, clause like indemnity, the contractor has to take it as it is. If the operators see there are a lot of qualifcations made by one contractor during the negotiation process, the operator might as well go for another contractor.138

Historically in the upstream industry, the party holding bargaining power has used their power to impose its preferred terms on the weaker party.139 The party with a better bargaining position, such as the operator’s, will be able to use their position to gain more favourable contract terms.140 However, the contractors might still sign an imbalanced agreement in order to secure a job even though they will be subject to unacceptable risk. In fact, some of the contractors risk double jeopardy, as any extraordinary risks will not be accepted by subcontractors and will be passed down in the usual way from the operator to the contractor but then not passed on in the subcontract.141 In this situation, the contractor is exposed to both his own contractual risk and the subcontractor’s risk.142 Generally, operators’ statutory duties pertaining to the award of contracts give theoretical power to the contracting parties in order to ensure fair, open and competitive tendering practices; however, only a brave contractor would challenge an operator with whom he aspires to work in the future.143 Here, the inferior party may well fnd that he is given no opportunity to negotiate the one-sided terms prior to the conclusion of the contract but that he should “take it or leave it”.144 This is a problem “in so-called negotiated contracts if the stronger party chooses to dictate to the other”.145 The stronger party may use his power by declining to enter any contract at all.146 2.3 Perception of risk allocation and indemnity clauses Semi-structured interviews were conducted to fnd the perception of risk allocation from the people in the industry and to discover their views on its current practice in Malaysia. Interviewees’ perceptions will be compared to samples of indemnity clauses which have been obtained from Operator A, Operator B and Operator C. In general, ranges of risk allocation clauses are commonly seen in contracts used in the oil and gas industry including indemnity and hold harmless clauses, clauses excluding liability for “consequential losses” and clauses limiting overall liability.147 The actual sharing of risk, indemnities and provisions for supporting insurances will

138 Respondent 5 from Contractor C. 139 Thorpe (n135) 282. 140 Ibid. 141 Helen Franklin, ‘Irretrievable Breakdown? A Review of Operator/Contractor Relationships in the Ofshore Oil and Gas Industry’ (2005) 23 J.E.R.L. 1, 7. 142 Ibid. 143 Ibid., p. 6. 144 Hugh Beale, ‘Unfair Contract Terms Act 1977’ [1978] Brit.J.Law Soc. 114, 115. 145 Ibid. 146 Stewart Macaulay, International Encyclopedia of Comparative Law (Mohr 1974) 18. 147 Greg Gordon, ‘Risk Allocation in Oil and Gas Contracts’ in Greg Gordon, John Paterson and Emre Usenmez (eds), Oil and Gas Law: Current Practice & Emerging Trends, vol 2nd (Dundee University Press 2011) 443.

268

lessons from the application of knock-for-knock be determined by the wording of the relevant contract documents.148 In fact, many contracts are non-standard or they are standard form contracts which are made non-standard by additional clauses.149 Non-standard contracts transfer more risk to the tenderer.150 It is important to note that subsequent to the Macondo oil spill in the Gulf of Mexico in 2010, there was an attempt by operators to depart from the established standard form contracts and shift greater risk to the contractor in the event of catastrophe.151 Post Macondo, operators are now less willing to deviate from their pro-forma contracts, and this creates a “take it or leave it” situation.152 Even before the Macondo incident, some contractors claimed that they were “required to take on onerous responsibilities without appropriate margins to absorb the consequences of the associated risks”.153 Many contractors consider such non-negotiable contracts to be problematic, primarily because they often contain onerous provisions in important areas such as allocation of risk, and this can create signifcant risk exposure.154 As a result, at the conclusion of the contracts the contractors will end up swallowing greater risk and liabilities. As one of the respondents representing contractor remarked: Usually, the risks are allocated through certain provisions such as indemnity clause, limitation clause and exclusion clause. From what I can see, these clauses are usually onesided. .  .  . Well, the thing is that, the operators would not entertain if we put so much qualifcation in the contracts. Sometimes, the operator is just going to say, “If you can’t comply and you have a lot of exceptions to the clauses, then you will be disqualifed”. As a result, we are not going to get the job. . . . We always have problem when it comes to indemnity clause. We are expected to bear most of the liabilities. The worst part is that, sometimes they (operators) expect us to be liable for something which is due to their faults . . . we have no choice but to agree with such clause. It is always be the case, either that we take or leave it.155

In distributing the risk between the operator and contractor, one of the respondents claimed that contractors were usually at the losing end and expressed their dissatisfaction at being made to indemnify operators’ negligence. This can be seen in the following remark made by one of the respondents: Supposedly, anything that are risky to us, then we need to take steps to mitigate such risks or deviate from such terms and conditions. However, most of the time, the contractor always be at the losing end, this is because in order to secure a big job, whether the contractors like it or not, the contractors have to meet the operator’s demand and must get ready to take those risks. .  .  . The problem with the indemnity clause is that, when 148 Edwards (n134). 149 Ibid. 150 Ibid. 151 Cameron (n17) 204. For further details, see ‘In Re: Oil Spill by the Oil Rig “Deepwater Horizon” in the Gulf of Mexico, on April 20, 2010: Memorandum in Support of Transocean’s Motion for Partial Summary Judgment against BP to Enforce BP’s Contractual Obligations, including BP’s Obligation to Defend, Indemnify and Hold Transocean Harmless against Pollution Claims’ (US District Court, Eastern District of Louisiana, 1 November 2011). 152 Moomjian, ‘Drilling Contract Historical Development and Future Trends Post-Macondo: Refections on a 35 Year Industry Career’ (n127). 153 Franklin (n143). 154 Ibid. 155 Respondent 1 from Contractor A.

269

wan mohd zulhafiz wan zahari the operator transfers their liabilities to us by asking us indemnify it and even though it was happened due to their negligence.156

The respondent also claimed that indemnity clauses were one-sided and that almost all of the liability in respect of indemnity clauses was placed on contractors. It could be argued that the best commercial policy would place responsibility for risk on the party best able to manage it, e.g., the party with the relevant insurance coverage.157 Insuring or contractually transferring risk to the insurer and allowing the premium to settle any charges to the other party could mitigate risk exposure and in fact, this is the most economically benefcial and practical way for risk to be dealt with.158 Insurance is used by the indemnitor as a risk cushion in a situation when he is responsible for his own employees and equipment. The insurance in fact is the underlying driver in this case rather than an ancillary tool for risk management.159 This is particularly natural for super-majors, who attempt to self-insure and minimise transaction costs.160 But this reason is not applicable to some contractors, as they cannot aford self-insurance: The indemnity clauses mostly are one sided. The contractors would be made to be liable for most of the liabilities, for example the indemnity with regards to property and equipment of the operators, the third-party liability, not to mention pollution. Usually we try to keep it and make it consistent with the insurance coverage, for example per occurrence how much we’ll be liable. Usually it is always unlimited liability and most of the time the clients refuse to negotiate on that as well.161

It is argued that the practicalities of risk allocation limited by certain basic requirements for those to whom risk is being transferred such as ability to undertake a hazardous task, willingness to take the risk, fnancial capability if the risk event occurs, continued existence and adequate fnance during the period of liability.162 It is also argued that, ideally, the responsibility for indemnifying the consequences of a risk event resulting from the activities of one of the contracting parties should rest with the party who has control over that risk.163 The operator is always in the best position to control the risk, and this was confrmed by the operator himself: As operator, it is mandatory to perform a project risk assessment and front-end study. If the risks are uncertain, big or economically not meeting the target proft margin, the project will not proceed.164

The operator also confrmed that both quantifed risk and qualitative risk, including indemnity, would be transferred to contractors. He also contended that it is the responsibility of contractors to understand and convert the risk into monitories: 156 Respondent 5, Contractor C. 157 Edwards (n134). 158 Max Abrahamson, ‘Risk Management’ (1984) 2 ICLR 241. 159 Caledonia North Sea Ltd. v British Telecommunications Plc same v Kelvin International Services Ltd. same v London Bridge Engineering Ltd. same v Norton (No. 2) Ltd (In Liquidation) same v Pickup No. 7 Ltd. same v Stena Ofshore Ltd. same v Wood Group Engineering Contractors Ltd. – [2002] 1 Lloyd’s Rep 553. 160 ‘BP Annual Report and Accounts’ (2007) 39 https://www.bp.com/content/dam/bp/business-sites/en/ global/corporate/pdfs/investors/bp-annual-report-accounts-2007.pdf 161 Respondent 3, Contractor B. 162 Edwards (n134). 163 Ibid. 164 Respondent 10 from Operator A.

270

lessons from the application of knock-for-knock As operator, all the quantifed risk will be transferred to contractor and stated in contract. The contractors are to put the price of each risk identifed in contract. It is responsibility of contractor to understand and convert the risk into monitories. . . . The risks are made clear to contractor. Contractor will put the prices for those scope specifed in the contract. Any risk is to be priced by contractor. .  .  . Uncounted quantity will use reimbursable cost plus. . . . Indemnity scope is to be taken by contractor; cost of premium for indemnity will be claim to operator.

Therefore, the contractor is inserted into a position where he has to absorb the risk and cost that risk into the price. Two respondents confrmed this scenario: Let say, there are some conditions that we could not aford to accept them in the event the operators attempt to shift greater risks to the us – now the operators go for competitive bids so our chances to be awarded is lesser if we stick to our qualifcation.165

Another respondent commented: However, we are not in the position to change the conditions, so what we normally do is, we will take note on that and accordingly advise to our technical people, “please calculate this risk into your cost”. Well, the thing is that, the operators would not entertain if we put so much qualifcation in the contracts. Sometimes, the operator is just going say, “if you can’t comply and you have a lot of exceptions to the clauses, then you will be disqualifed”. As a result, we are not going to get the job.166

However, it is difcult for contractors to calculate and set an ideal price for their services after absorbing the risk. Due to the highly competitive nature of the bidding process, the contractor faces a dilemma when setting the price. The contractor is either afraid of not getting the job if the price he sets is too high after converting the risk into monitories. Or he fears exposure to fnancial problems if the price he sets is too low. Two respondents commented on this: We need to set the price. The price should be an ideal one. Not too high as there is possibility our submission would be rejected, but not too low to the extent that it might jeopardies our proft.167

The second respondent said: Usually, in order to mitigate the risks, we have to cost in the impact into pricing. Whatever the risks involved; the cost has to be refected in the pricing. Sometimes, this would be a problem, when we were trying to cost in everything, the cost will be too high and we afraid that we are not going to get the job. But, if we neglect the risks now, then if anything happens in future, the risks would be at our own cost. So, it is real challenge for us to draw a middle line between these two.168

It is true that a contract is an equalisation of accepting between controllable and uncontrollable risks with the price deemed appropriate to undertake the work.169 However, it is said that the main reason for the increase in overall costs is due to the 165 Respondent 3 from Contractor B. 166 Respondent 1 from Contractor A. 167 Respondent 4 from Contractor B. 168 Respondent 3 from Contractor B. 169 Jur Tunay KÖKSAL, ‘FIDIC Conditions of Contract as a Model for an International Construction Contract’ (2011) 1 BMR 32.

271

wan mohd zulhafiz wan zahari usage of disclaimer clauses in allocating risk such as indemnity clauses.170 This is because, once the risk is transferred to the contractor and “the contractor has no means by which to control the occurrence or outcome of the risk, the contractor must either insure against it or add a contingency to the bid price”.171 The cost of transferring risk to the contractor through such clauses presents a few hidden costs such as “restricted bid competition, increased potential for claims and disputes and above all, more adversarial owner-contractor relationships.”172 One of the respondents, who is a practicing lawyer and used to litigate on behalf of the operator, shared her perspective on this issue: I assume when the contractors signed the contract, the contractors were fully aware of those liabilities that they will be carrying on. So, if anything happens in future, I think it is their obligation to get protection for those liabilities by way of insurance. Plus, the parties are able to practice contractual freedom. I don’t see any problem with regard to this matter, unless, there is issue with regard to fairness, but such allegation should be proven by separate cause of action and proper hearing in court.173

It is true that as a matter of contractual freedom, the parties may freely decide the terms of the contract as they wish including the risk allocation provisions.174 However, this particular respondent might not be aware of the actual situation in the industry during the contract formation process. For example, if there is inequality of bargaining power between the parties, then contractual freedom has not been exercised properly. As a result, the contract becomes one-sided and unfair risk allocation has taken place. The contractors are actually aware of the situation, but they are not in a position to change or qualify any of the terms. Nor are they given the opportunity to discuss or negotiate on the allocation of risk. One of the respondents remarked: The most we can do is to voice our dissatisfaction to the clients. Sometimes they may listen to us, unfortunately most of the time they are not. How would we mitigate? Basically, it is good to have both parties to sit down and explain and discuss about the risks involved in each project. But I suppose they are going to say that, it is them who invest money; thus, they will not accept any of our qualifcation with regard to the risks.175

There are circumstances, however – possibly owing to the self-insurance of certain risks – where one party will attempt to pass additional risks on to the other.176 This may require the contractor to indemnify the operator for the operator’s negligence 170 F Hartman, ‘Construction Dispute Resolution through an Improved Contracting Process in the Canadian Context’ (Loughborough University of Technology, UK 1993); G Jergeas and F Hartman, ‘A Contract Clause for Allocating Risk’ [1996] AACE Int. Trans.; JG Zack Jr., ‘“Risk-Sharing” – Good Concept, Bad Name’ (1996) 38 Cost Engin. (Morgantown, West Virginia) 26. 171 GF Jergeas and FT Hartman, ‘Contractors’ Protection Against Construction Claims’, Annual Meeting-American Association of Cost Engineers (AACE 1994) 8. 172 C Robert, ‘Managing Change Orders and Claims’ (1997) 2 J. Manage. Eng. 27; D Becker, ‘The Cost of General Conditions’ [1993] Am. Assoc. Cost. Eng. Trans. 7; Kwaku A Tenah, ‘The Design-Build Approach: An Overview’ (2000) 42 Cost Engin. 31. 173 Respondent 10 from Legal Firm Z. 174 Privy Council in Ooi Boon Leong v Citibank N.A. [1984] 1 MLJ 222, confrmed that “parties to an agreement have much scope to negotiate and incorporate terms acceptable to them”. 175 Respondent 10 from Legal Firm Z. 176 Franklin (n143).

272

lessons from the application of knock-for-knock to the operator’s employee and property, or vice versa. This would represent an uninsurable risk to the contractor and the adjustment severely undermines the knockfor-knock indemnity regime and may also give rise to the possibility of increased costs as the contractor attempts to insure a risk in which it does not have a true insurable interest.177 One of the respondents commented on this scenario: We always have problem when it comes to indemnity clause. We are expected to bear most of the liabilities. The worst part is that, sometimes they (operators) expect us to be liable for something which is due to their faults. Could you imagine that? But, what can we say; we have no choice but to agree with such clause? It is always be the case, either that we take or leave it. . . . The risks have to be covered by insurance. We do not aford to take the risks without any fnancial back up from the insurance company. . . . This is the problem. Sometimes, we have to absorb all risk regardless whether such risk is being covered by the insurance company.178

Moreover, from the contractor’s point of view, the risk is not one which can be passed down to subcontractors. Again, parties who seek protection from the other party’s negligence in an adjusted knock-for-knock regime similarly undermine the benefts of the concept since they, the indemnifying party, will in any event be obligated to secure additional insurance to cover the consequences of their negligence to the other party’s employees or property. Moreover, it is likely that the existence of negligence may frst have to be proved in the courts for an indemnity to operate, which defeats one of the primary objectives of the knock-for-knock regime. In the same vein, a contractor will often seek to limit its liability for third-party risks arising out of the agreement to the extent that this exceeds the limit of its thirdparty insurance.179 This is consistent with an overall desire to cap risks over which it has little or no control, including the catastrophic risks associated with ofshore exploitation; e.g., fre, explosion, blowout (underground and surface) and pollution emanating from the reservoir. Without liability caps, the contractor will have to bear what are normally uninsured risks which may well be disproportionate to the size, nature and fnancial return of the contract. However, the quid pro quo for appropriate protection by the operator is that the contractor should be prepared to insure the operator.180 It is not reasonable to expect limitation of, and/or indemnity from, catastrophic and third-party risks that the contractor’s own insurance covers. Contractors should be prepared to be open about the scope and terms of their insurance cover as well as other material information about the size and resources of the company and, where applicable, parent and group company structure. In this regard, one of the respondents commented: Again, the indemnity clauses mostly are one sided. The contractors would be made to be liable for most of the liabilities, for example the indemnity with regards to property and equipment of the operators, the third-party liability, not to mention pollution. 177 Ibid. 178 Respondent 1 from Contractor A. 179 Franklin (n143). 180 Wan M Zulhafz, ‘An Empirical Study on the Contractual Risk Allocation Provisions and Indemnity and Hold Harmless Clauses in the Oilfeld Service Contracts in Malaysia’, Paper Proceedings on ‘Second International Conference on Interdisciplinary Legal Studies (ICILS) 2015’ on 9th–10th June 2015 in Toronto, Canada (Unique Conference Canada, June 2015).

273

wan mohd zulhafiz wan zahari Usually, we try to keep it and make it consistent with the insurance coverage, for example per occurrence how much we’ll be liable. Usually it is always unlimited liability and most of the time the clients refuse to negotiate on that as well.181

In order to combat this problem, some of the respondents suggested that fairness could be achieved by legal intervention where the legislator passes a law to protect contractors. Another respondent suggested that it should be an anti-indemnity law: I think we need the government interference to come out with a regulation in order to achieve fairness.182

Another respondent commented: I think it is quite ideal to have some sort of protection by the legislator on this matter.183

A third respondent stated: I know that in US, they have oilfeld anti-indemnity law, but I am not quite sure how far does the law really efcient to address this issue. What I can confrm, we have nothing yet in Malaysia to that efect. To certain extent, I think yes, we need rules by the government to solve the problem with regards to indemnity clauses.184

One of the respondents shared his views, saying that the relevant authority should issue guidelines in order to address this problem: I think it is good if the authority come out with guidelines to monitor this problem but it has to be tightened up with our Petroleum Act.185

Therefore, based on the fndings from the semi-structured interviews with the respondents, it could be argued that operators has dominant position over contractors and in a better position during contractual negotiation in Malaysia. Due to the lesser bargaining power, contractors are given fewer opportunities to modify the terms and conditions in order to make them fairer. However, as discussed above, despite the unfair terms and conditions, contractors end up agreeing to them in order to secure present and future jobs. Moreover, contractors are exposed to the risk without covered by insurance. This situation is regarded as a serious problem because it may lead to great fnancial losses to contractors. 3 Analysis of the indemnity and hold harmless clauses of Operators A, B and C In order to assess the claims made by the contractors, the indemnity clauses in three samples of oilfeld service contracts which were drafted by three diferent operators: Operator A, Operator B and Operator C are each analysed in four diferent contexts, namely:

181 182 183 184 185

Respondent 3 from Contractor B. Respondent 5 from Contractor C. Respondent 8 from Contractor D. Respondent 3 from Contractor B. Respondent 7 from Contractor E.

274

lessons from the application of knock-for-knock 1 Liability with regard to personal injury or death of employees and loss of property of the parties; 2 Third-party claims; 3 Liability with regard to pollution; 4 Insurance coverage. These indemnity clauses are compared with two model form contracts which are produced by LOGIC186 and FIDIC.187 LOGIC is a not-for-proft subsidiary of Ofshore Energies UK (OEUK; formerly Oil and Gas UK).188 LOGIC operates as the custodian for cross-industry projects that aim to increase the efciency of working practice in the UK Continental Shelf (UKCS). The standard contracts for the UK ofshore oil and gas industry (formerly CRINE contracts) were developed by the Standard Contracts Committee and are issued by LOGIC for use within the industry between companies and their contractors.189 In South and South East Asia (India, Indonesia, Malaysia, Thailand and Vietnam), LOGIC is widely used as a starting point in drafting contracts due to its international character.190 The General Conditions of Contract for Construction are prepared for the scopes of work that are related to major fabrication, topsides installation and hook-up, signifcant topsides modifcations, construction services contracts for topsides works and also other contracting arrangements such as EPCC Contracts or Engineering, Procurement, Fabrication and Installation (EPFI) Contracts.191 The International Federation of Consulting Engineers (FIDIC) publishes international standard form contracts for works and also provides related materials such as standard pre-qualifcation forms.192 The FIDIC (Silver Book) model contract is usually used by the industry as a foundation in drafting turnkey construction contracts for lump-sum projects.193 The standard forms of contracts by LOGIC and FIDIC are the two examples that are widely used by the oil and gas industry players around the world as a foundation to draft oilfeld service contracts. These two contracts, particularly their indemnity clause provisions, will be compared to the contracts drafted by Operator A, Operator B and Operator C later in this chapter. The analysis will also make reference to the General Contracting Principles (FAIR Principles) which are provided by the International Marine Contractors Association (IMCA).194 These principles serve the long-term interests of all players in the oil and gas industry by encouraging fair contractual provisions according to the parties’

186 LOGIC Construction Contract, 2nd Edition, 2003. 187 FIDIC Conditions of Contract for EPC Turnkey Projects (Silver Book), 1999. 188 www.oeuk.co.uk/. 189 See www.logic-oil.com/. 190 Hewitt (n19) 332. 191 ‘General Conditions of Contract (including Guidance Notes) for Construction’ in Standard Contract for U.K. Ofshore Oil and Gas Industry, vol October (2nd edn, LOGIC 2003) 2. 192 See more at http://fdic.org/. 193 Huse; Axel-Volkmar Jaeger and Götz-Sebastian Hök, ‘FIDIC Contract Documents’, FIDIC-A Guide for Practitioners (Springer 2010) 125. 194 IMCA is an international trade association representing ofshore, marine and underwater engineering companies. One of its roles is to communicate and promote dialogue amongst the various companies, both clients and contractors, for the beneft of the whole industry.

275

wan mohd zulhafiz wan zahari respective risks and rewards.195 It is vital to note that the FAIR Principles are not law but merely provide guidelines in drafting oil and gas contracts, which promotes well-established industry custom and practice.196 3.1 Liability regarding personal injury or death of employees and loss of property of the parties According to the FAIR Principles, both parties should grant reciprocal, unlimited indemnities for the costs of loss of or damage to both parties’ property and personnel of whatsoever nature or the property and personnel of the parties’ group, irrespective of cause (including negligence). This principle ensures that each party to the contract has clear and easily understood liability during the project for both people and property. Further, this allocation of risk could aim to avoid increasing either party’s liability levels as a result of involvement in the project which would require additional insurance cover or the implementation of an alternative risk management regime for the project.197 It is important to note that the FAIR principles are not law. Hence they are not legally binding. The FAIR principles operate as a guideline to the parties on how to draft contracts with fair contractual terms and conditions. For this reason, the FAIR principles will be referred to in the analysis below in setting out a benchmark for what would be a fair indemnity clause. 3.1.1 Under LOGIC In the oil and gas industry, contractual indemnity clauses often take the form of a specifc type of broad form indemnity known as the “knock-for-knock” provision, where an individually party agrees to indemnify the other for injuries or damage to its personnel and property regardless of fault.198 In other words, each party agrees to take full responsibility for all bodily injury or property damage claims made by its own employees, regardless of which party may actually be responsible for the injury.199 For example, under the LOGIC Construction Contract Ed, 2 October 2003, the reciprocal indemnities between the Company and the Contractor relate to the parties included in the Company and Contractor Groups as follows: 22.1 The CONTRACTOR shall be responsible for and shall save, indemnify, defend and hold harmless the COMPANY GROUP from and against all claims, losses damages, costs (including legal cost) expenses and liabilities in respect of: (a) Loss or damage to property of the CONTRACTOR GROUP whether owned, hired, leased or otherwise provided by the CONTRACTOR GROUP arising from, relating to or in connection with the performance or non-performance of the CONTRACT; and (b) personal injury including death or disease to any person employed by the CONTRACTOR GROUP arising from, relating to or in connection with the performance or non-performance of the CONTRACT; and . . . 195 196 197 198 199

Louys (n289). Culligan and de Roo (n21). ‘IMCA General Contracting Principles (Rev.1)’ (2011), www.imca-int.com, accessed 6 January 2014. Egbochue (n166). Evans and Butler (n334).

276

lessons from the application of knock-for-knock 22.1 The COMPANY shall be responsible for and shall save, indemnify, defend and hold harmless the CONTRACTOR GROUP from and against all claims, losses damages, costs (including legal cost) expenses and liabilities in respect of: (a) loss of or damage to property of the COMPANY GROUP whether (i) owned by the COMPANY GROUP, or (ii) leased or otherwise obtained under arrangements with fnancial institution by the COMPANY GROUP which is located at the WORKSITE arising from, relating to or in connection with the performance or non-performance of the CONTRACT, but excluding the PERMANENT WORK; and . . . (b) personal injury including death or disease to any person employed by the COMPANY GROUP arising from, relating to or in connection with the performance or non-performance of the CONTRACT; and . . . 22.6 All exclusions and indemnities given under this Clause 22 (save for those under Clause 22.1(c), 22.2(c) and 22.5(b)) and Clause 25 shall apply irrespective of cause and notwithstanding the negligence or breach of duty (whether statutory or otherwise) of the indemnifed party or any other entity or party and shall apply irrespective of any claim in tort, under contract or otherwise at law.

In respect of personal injury including death or disease of the contractor’s employees, the contractor is required to indemnify the company who, in the context of this research, is the operator. On the other hand, the company will indemnify the contractor for the company’s employees with regard to personal injury including death or disease. In respect of loss or damage to property, both are required to be responsible and indemnify each other with regard to their own property. Clause 22.6 has made clear that both of the contractors and company will be liable for their own employees regardless of whether death, disease, loss and/or damage were due to the negligence of the indemnifed party. In the Caledonia North Sea Ltd v London Bridge Engineering Ltd200 case, Lord Bingham referred to the knock-for-knock indemnity provision that covered employees as a “market practice [which] has developed to take account of the peculiar features of ofshore operations”.201 The court held that the knock-for-knock indemnity provision, properly construed, entitled the operator to be indemnifed by the contractors with regards to the fatalities and injuries even where they were not liable at common law, or liable for breach of statutory obligation.202 Knock-for-knock indemnity provisions are explained in further detail in Chapter 4 of this book. 3.1.2 Under FIDIC Under the FIDIC model contract, the clause requires each party to indemnify the other Party from any claims arising out of the contractor’s execution of the works.203 This obligation extends to any third-party claims. The indemnity clause reads as follows: 17.1 The Contractor shall indemnify and hold harmless the Employer, the Employer’s Personnel, and their respective agents, against and from all claims, damages, losses and expenses (including legal fees and expenses) in respect of: (a) bodily injury, sickness, disease or death, of any person whatsoever arising out of or in the course of or by reason of the design, execution and completion 200 201 202 203

[2002] UKHL 4. Ibid., para. 7. Ibid. Axel Volkmar Jaeger and Götz-Sebastian Hök, FIDIC: A Guide for Practitioners (Springer 2009).

277

wan mohd zulhafiz wan zahari of the Works and the remedying of any defects, unless attributable to any negligence, willful act or breach of the Contract by the Employer, the Employer’s Personnel, or any of their respective agents, and . . . The Employer shall indemnify and hold harmless the Contractor, the Contractor’s Personnel, and their respective agents, against and from all claims, damages, losses and expenses (including legal fees and expenses) in respect of (1) bodily injury, sickness, disease or death, which is attributable to any negligence, willful act or breach of the Contract by the Employer, the Employer’s Personnel, or any of their respective agents, and (2) the matters for which liability may be excluded from insurance cover, as described in sub-paragraphs (d)(i), (ii) and (iii) of Sub-Clause 18.3 [Insurance Against Injury to Persons and Damage to Property].

This research interprets the word “employer” as “operator”. The contractor is required to indemnify the employer, the employer’s personnel and their agents against and from all claims, damage, etc. in respect of injury, sickness, disease or death of any person arising from the contractor’s execution of the works, including any consequences arising – from the contractor’s design (if any). However, the contractor is not required to provide indemnity in respect of any negligence or wilful act of the employer, employer’s personnel or their agents’ damage to or loss of property to the extent that such damage or loss arising out of or as a consequence of the contractor’s design, the execution of the works and the remedying of defects or is attributable to the negligence or a wilful act of the contractor, the contractor’s personnel or his agents. On the other hand, the employer is required to indemnify the contractor, the contractor’s personnel and their agents from all claims in respect of injury, sickness, disease or death attributable to the negligence of the employer, the employer’s personnel and their agents; the employer is also required to indemnify the contractor’s liability which may be excluded from insurance coverage. It is argued that while Clause 17.1 above is not knock-for-knock indemnity, it is an example of fair indemnity and hold harmless clauses which bind the contractor. This is because the clauses do not require the contractor to indemnify the employer, i.e., operator for employer’s negligence. In fact, the employer indemnifes the contractor for the injury, sickness, disease or death which is caused by his negligence. Moreover, the contractor does not bear uninsurable risk since the employer indemnifes the contractor for any matters which are not covered by insurance. 3.1.3 Under Operator A’s contract Operator A drafted the indemnity and hold harmless clauses with regard to the personal injury or death of employees and loss of property of the parties as follows: Article X – Liabilities and Indemnities For the purpose of this article, the COMPANY GROUP shall mean OPERATOR A, the COMPANY, its CO-VENTURES, its and their respective AFFILIATES, its and their respective directors, ofcers and employees. X.1 Personnel and the COMPANY GROUP The COMPANY shall be responsible for and shall protect, defend, indemnify and hold harmless the CONTRACTOR and the SUB-CONTRACTOR from and against any and all claims, liabilities, costs, damages, expenses of every kind and nature, with respect to

278

lessons from the application of knock-for-knock injury or death or damage to or loss of property of any person employed by the company group, howsoever arising. X.2 Personnel of CONTRACTOR The CONTRACTOR shall be responsible for all and shall protect, defend, indemnify and hold harmless the COMPANY GROUP from and against any and all claims, liabilities, costs, damages and expenses of every kind and nature, with respect to injury or death or damage to or loss of property of any person employed by the CONTRACTOR and/or the SUB-CONTRACTOR, howsoever arising. X.3 COMPANY’s PROPERTY The CONTRACTOR shall be liable for and shall indemnify the COMPANY GROUP against any damage to or destruction or loss of property operated and/or owned by the COMPANY arising during, and/or as a result of the performance of this CONTRACT, without regard to whether any act or omission of the COMPANY GROUP contributed to the loss. X.4 CONTRACTOR’s EQUIPMENT The CONTRACTOR shall assume the risk of, and shall be solely responsible for and in this regard shall indemnify, defend and hold the COMPANY harmless against any claims arising out of all damage to and/or loss destruction of CONTRACTOR’s EQUIPMENT lost or damaged, afects the performance of the SCOPE OF WORKS, the CONTRACTOR shall replace any lost or damaged CONTRACTOR’s EQUIPMENT at CONTRACTOR’s sole cost in the most expeditious manner possible and at CONTRACTOR’s sole expense.

Based the reading of the above Clause X.1 and X.2, it could be argued that such clause was drafted on a mutual hold harmless basis, whereby the contractor is required to indemnify and hold harmless the operator in respect of personal injury and death of the contractor’s employees. Meanwhile, the company is required to indemnify and hold harmless the contractor in respect of personal injury and death of the company’s employees. This research argues that this provision goes beyond the standard practice as set out by LOGIC and FIDIC. Under the standard practice only personal injury and death are covered, whereas the above provision also sets a condition that the contractor is required to indemnify and hold harmless the company for the “loss of property of any person employed by the contractor” and the company is required to indemnify and hold harmless the contractor for the “loss of property of any person employed by the company”. It could be argued that these extra requirements do not undermine fair practice as regards indemnity and hold harmless clauses and therefore do not cause problems for contractors. However, on further reading of Articles X.4 and X.5, one might realise that there is imbalanced risk allocation and unfair indemnity hold harmless clauses, where the contractor is required to indemnify and hold harmless the operator for the destruction or loss of property operated or owned by the operator irrespective of whether such loss was caused by the act or omission of the operator. Apart from that, the contractor is made solely responsible for and required to indemnify and hold harmless the operator for all damage to and/or loss of or destruction to property of the contractor’s equipment which occurs in the performance of the work, at the contractor’s sole cost and expense. These clauses could be unfair to the contractor. 279

wan mohd zulhafiz wan zahari 3.1.4 Under Operator B’s contract Operator B drafted the indemnity and hold harmless clauses with regard to the personal injury or death of employees and loss of property of the parties as follows: Article X – Indemnifcation X.1 CONTRACTOR agrees to indemnify, defend and save harmless COMPANY GROUP from and against any and all claims, losses and expenses including without limitation costs, demands, damages, suits, judgments, fnes, penalties, liabilities, reasonable attorneys’ fees and causes of action whatsoever nature or character, whether known or unknown, and including without limitation claims, losses and expenses for property damage, bodily injury, illness, disease, death, pollution or loss of services wages, consortium or society) in any way directly or indirectly, arising out of, or related to, the performance or subject matter of this CONTRACT or the ingress, egress, or presence on any premises (whether land, building, vehicle, platform, aircraft, vessel or otherwise) owned, operated, chartered, leased, used, controlled or hired by COMPANY GROUP or CONTRACTOR GROUP, and which are asserted by or arise in favor of CONTRACTOR GROUP and expressly including any claims, losses or expenses actually or allegedly caused by the sole, active, passive, concurrent or partial negligence (of whatever nature or character), fault or strict liability of COMPANY GROUP or any other person or the unseaworthiness, unairworthiness or defective condition of vessels, craft ore premises, whether or not preceding or during the execution of this CONTRACT.

The above indemnity and hold harmless clause is unilateral, where it is one-sided and appears to be more favorable to the operator. Under unilateral clauses, the contractor is required to indemnify and hold harmless the operator for all losses and expenses such as losses and expenses for property damage, bodily injury, illness, disease, death and pollution of not only his own employees but also the company’s employees irrespective of whether such act was caused by the fault or sole negligence of the operator. In Article X, the phrase “including without limitation . . . fnes, penalties” could be interpreted as granting punitive or exemplary damages. It could also be argued that punitive or “exemplary damages serve to punish roguish defendants to make examples of such defendants before society, to protect society’s interest in right conduct and fair play”.204 Therefore, any indemnifcation given for such punitive or exemplary damages would not be enforced as it is against public policy.205 Overall, these clauses seem to be extremely unfair to contractors because it allows the operators to dump liability and shift it to the contractor.206

204 Roy R Anderson Jr, ‘Indemnity Against Punitive Damages: An Examination of Punitive Damages, Their Purpose, Public Policy, and the Coverage Provisions of the Texas Standard Automobile Liability Insurance Policy’ (1973) 27 Sw.LJ 593. 205 Cary A Moomjian, ‘Macondo Litigation Update-How Court Decisions Will Impact Oilfeld Contracting and Insurance Practices’, IADC/SPE Drilling Conference and Exhibition on 4th–6th of March 2014 in Fort Worth, Texas, USA (Society of Petroleum Engineers 2014). 206 Wan M Zulhafz, ‘Recent Trends in Allocation of Risk Post-Macondo: The Growing Tension Between Oil and Gas Standard Forms of Contract, and Contractual Practice’ (2017) 5 Int.Energy Law Rev. (IELR) 174.

280

lessons from the application of knock-for-knock 3.1.5 Under Operator C’s contract Operator C drafted the indemnity clause regarding the personal injury or death of employees and loss of property of the parties as follows: X.0 X.1

X.2

LIABILITIES AND INDEMNITIES a) CONTRACTOR PERSONNEL CONTRACTOR shall be responsible for and shall protect, defend, indemnify and hold harmless COMPANY, its parent company, subsidiaries, AFFILIATES, consultants and their respective agents, ofcers, and employees from and against any and all claims, liabilities, costs, damages and expenses of every kind of nature, with respect to injury or death or damage to or loss of property of any person employed by CONTRACTOR arising during and/or as a result of the performance of this CONTRACT. b) COMPANY PERSONNEL COMPANY shall be responsible for and shall protect, defend, indemnify and hold harmless CONTRACTOR, its parent company, subsidiaries, AFFLIALTES, consultants and their respective agents, ofcers and employees from and against any and all claims, liabilities, costs, damages and expenses of every kind and nature with respect to injury or death of or damage to or loss of property of any person employed by COMPANY arising during and/or as a result of the performance of this CONTRACT. a) CONTRACTOR’s Equipment and Property Save as otherwise expressly provide in this CONTRACT, CONTRACTOR shall assume the risk of, and shall be solely responsible for and in this regard shall indemnify, defend and hold COMPANY, its parent company, subsidiaries, AFFILIATES, consultants and their respective agents, ofcers and employees harmless against any claims arising out of the damage to the loss, or destruction of, all CONTRACTOR and its subcontractors’ equipment and property. b) COMPANY’s Equipment and Property CONTRACTOR shall be liable for and shall indemnify COMPANY against any damage to or destruction or loss of property owned by COMPANY, its parent company, AFFILIATES, Client and other subcontractors and/or Operator B arising during, and/or as a result of the performance of this CONTRACT, without regard to whether any act or omission of COMPANY contributed to the loss.

The indemnity clause that has been drafted by Operator C is quite similar to the one drafted by Operator A. Articles X.1 (a) and (b) were based on mutual hold harmless, whereby the contractor is required to indemnify and hold harmless the operator in respect of personal injury and death of the contractor’s employee and on the other hand, the company is required to indemnify and hold harmless the contractor in respect of personal injury and death of the company’s employee. By contrast, it could be argued that there is imbalanced risk allocation and unfair indemnity and hold harmless clause in Articles X.2 (a) and (b) where the contractor is required to indemnify and hold harmless the operator and its afliates for not only the damages and loss of the contractor’s property, but also the operator’s and its afliates’ property regardless any act or omission of operator contributed to the loss. 3.2 Claims by a third party Based on the FAIR Principles, each party is best placed to assume legal liability for losses which he causes to a third party. However, in circumstances where a contractor is required to perform work on the facility, or in an area of close proximity to 281

wan mohd zulhafiz wan zahari any existing facilities, the company should reasonably be expected to indemnify the contractor group in full for all of the costs arising out of any loss or damage to such existing facilities.207 Such indemnity would extend to cover the costs of existing facilities property, third-party personnel, consequential losses and pollution efects fnancial rewards to the contractor are not increased as a result of undertaking work in an area which is in close proximity to existing facilities. Signifcant increases in risk of loss or damage occurring to such existing facilities and the contractor’s own property and personnel and additional risk in working in a congested area could therefore be fairly allocated to the company which has the ability to mitigate such third-party risk by seeking appropriate cross-indemnity agreements with the thirdparty owners prior to commencing the project or obtaining sufcient insurance cover for these signifcant risks as part of the whole project risk management.208 3.2.1 Third-party claims under LOGIC It is common for oil and gas service contracts to provide for “guilty party pays” reciprocal indemnities.209 This means that party A indemnifes party B and party B’s group against claims in respect of any death of or personal injury to third parties and loss of or damage to the property of third parties. The indemnity clause covers negligence, breach of statutory duty or breach of contract of party A or its group. On the other hand, Party B provides a reciprocal indemnity in favour of party A and party A’s group. Otherwise, contractors may give an indemnity in respect of third parties’ claims arising out of the contractor’s operations except to the extent that there is negligence, breach of statutory duty or breach of contract by the operator or any member of its group. This places the onus on the contractor to prove such negligence or breach and therefore flls a gap where it is diffcult to prove which party (if any) caused the death, personal injury, loss or damage in question. This can be seen in the LOGIC contract form: 22.1 The CONTRACTOR shall be responsible for and shall save, indemnify, defend and hold harmless the COMPANY GROUP from and against all claims, losses, damages, costs (including legal costs) expenses and liabilities in respect of: (c) subject to any other express provisions of the CONTRACT, personal injury including death or disease or loss of or damage to the property of any third party to the extent that any such injury, loss or damage is caused by the negligence or breach of duty (whether statutory or otherwise) of the CONTRACTOR GROUP. For the purposes of this Clause 22.1(c) “third party” shall mean any party which is not a member of the COMPANY GROUP or CONTRACTOR GROUP. 22.2 The COMPANY shall be responsible for and shall save, indemnify, defend and hold harmless the CONTRACTOR GROUP from and against all claims, losses, damages, costs (including legal costs) expenses and liabilities in respect of: (c) subject to any other express provisions of the CONTRACT, personal injury including death or disease or loss of or damage to the property of any third party to the extent that any such injury, loss or damage is caused by the negligence or breach of duty (whether statutory or otherwise) of the

207 IMCA General Contracting Principles (Rev. 1) (n199). 208 Ibid. 209 Chidi Egbochue, ‘Reviewing Knock for Knock Indemnities Following the Macondo Well Blowout’ (2006) 4 C.L.Int. 7.

282

lessons from the application of knock-for-knock

(d)

COMPANY GROUP. For the purposes of this Clause 22.2(c) “third party” shall mean any party which is not a member of the CONTRACTOR GROUP or COMPANY GROUP loss of or damage to such permanent third-party oil and gas production facilities and pipelines and consequential losses arising therefrom, as specifed in and defned in and in accordance with Appendix 1 to Section I – Form of Agreement where such loss or damage is arising from, relating to or in connection with the performance or non-performance of the CONTRACT. The provisions of this Clause 22.2(d) shall apply notwithstanding the provisions of Clause 22.1(c).

For the purposes of Clauses 22.1(c) and 22.2(c), the reciprocal indemnities between Company and Contractor relate to the parties included in the Company and Contractor Groups as defned in Clauses 1.2 and 1.9. The Company indemnifes Contractor Group in respect of loss/damage to permanent third-party oil and gas production facilities, and consequential losses (as defned) therefrom. A third party is defned in the LOGIC Contract as “any party who is not a member of the Company or Contractor Groups”. Consequently, certain parties, who in many cases will be present at some parts of the worksite, are third parties for the purposes of the indemnity clauses. These include in particular other contractors of the Company. It should be noted, therefore, that the Company’s other contractors (and the subcontractors of such other contractors) are not included in the “Company Group” defnition. In this regard, the LOGIC Standard Contracts Committee upholds the Industry Mutual Hold Harmless (IMHH) as the most appropriate means of dealing with the allocation of liability for injury to persons, damage to property and consequential loss between the Company’s contractors and strongly encourages all contractors to join the IMHH scheme.210 3.2.2 Third-party claims under FIDIC FIDIC drafts the indemnity clause regarding third-party claims as follows: 17. INDEMNITY 17.1 The Contractor shall indemnify and hold harmless the Employer, the Employer’s Personnel, and their respective agents, against and from all claims, damages, losses and expenses (including legal fees and expenses) in respect of: (a) bodily injury, sickness, disease or death, of any person whatsoever arising out of or in the course of or by reason of the design, execution and completion of the Works and the remedying of any defects, unless attributable to any negligence, wilful act or breach of the Contract by the Employer, the Employer’s Personnel, or any of their respective agents, and . . . The Employer shall indemnify and hold harmless the Contractor, the Contractor’s Personnel, and their respective agents, against and from all claims, damages, losses and expenses (including legal fees and expenses) in respect of (1) bodily injury, sickness, disease or death, which is attributable to any negligence, willful act or breach of the Contract by the Employer, the Employer’s Personnel, or any of their respective agents, and (2) the matters for which liability may be excluded from insurance cover, as described in sub-paragraphs (d)(i), (ii) and (iii) of Sub-Clause 18.3 [Insurance Against Injury to Persons and Damage to Property]. 210 Details of title IMHH are available on the LOGIC. See www.logic-oil.com.

283

wan mohd zulhafiz wan zahari Under the above Clause 17.1(a), the contractor is required to indemnify and hold harmless the operator, the operator’s personnel, or its agents from all claims, damages, losses and expenses in respect of bodily injury, sickness, disease or death of “any person”. The phrase “any person” includes the employees of the contractor, operator or any third party. However, the contractor would not be made liable if he is able to prove that such claims, damages and losses happened because of the negligence, wilful act or breach of the contract of the operator, the operator’s personnel, or its agents. 3.2.3 Third-party claims under Operator A’s contract Operator A drafted the indemnity clause with regard to claims by third party as follows: X.3 Third Party The CONTRACTOR shall be responsible for and shall protect, defend, indemnify and hold harmless the COMPANY GROUP from and against any and all claims, liabilities, cost, damages and expenses of every kind and nature with respect to injury, illness or death of, or damage to or loss of property of any third party (including pollution), arising during and/or as a result of the performance of this CONTRACT without regard to whether any act or omission of the COMPANY GROUP contributed to such injury, death or damage to or loss of property.

The above indemnity clause is quite straightforward. It says that the contractor is required to indemnify and hold harmless the operator, operator’s personnel and its agent with regard to the claims, liabilities, cost, damage and expenses by a third party in respect of injury, illness or death of or damage to or loss of property of any third party (including pollution). This indemnity clause appears to be one-sided. This is because the contractor would be made liable for the claims and liabilities by the third party even where such acts were caused by the negligence, wilful act or omission of the operator, the operator’s personnel, and its agents. 3.2.4 Third-party claims under Operator B’s contract Operator B drafted the indemnity clause regarding third-party claims as follows: X.1 CONTRACTOR agrees to indemnify, defend and save harmless COMPANY GROUP from and against any and all claims, losses and expenses including without limitation costs, demands, damages, suits, judgments, fnes, penalties, liabilities, reasonable attorneys’ fees and causes of action whatsoever nature or character, whether known or unknown, and including without limitation claims, losses and expenses for property damage, bodily injury, illness, disease, death, pollution or loss of services wages, consortium or society) in any way directly or indirectly, arising out of, or related to, the performance or subject matter of this CONTRACT or the ingress, egress, or presence on any premises (whether land, building, vehicle, platform, aircraft, vessel or otherwise) owned, operated, chartered, leased, used, controlled or hired by COMPANY GROUP or CONTRACTOR GROUP, and which are asserted by or arise in favor of CONTRACTOR GROUP and expressly including any claims, losses or expenses actually or allegedly caused by the sole, active, passive, concurrent or partial negligence (of whatever nature or character), fault or strict liability of COMPANY GROUP or any other person or the unseaworthiness, unairworthiness or defective condition of vessels, craft ore premises, whether or not preceding or during the execution of this CONTRACT.

The above indemnity clause is expansive and appears to be onerous for contractors. The reason for this is that it seems that the operator has shifted most of the risks 284

lessons from the application of knock-for-knock to the contractor. Even though the clause has not mentioned anything about contractor’s liability in respect of third-party liabilities with regard to property damage, bodily injury, illness, disease and death, broad phrases such as “any and all claims, losses and expenses” and “in any way directly or indirectly arising out of, or related to, the performance or subject matter of this CONTRACT” would imply that thirdparty claims, losses and expenses in relation to property damage, bodily injury, illness, disease, death and pollution are covered under the contractor’s liability. This research argues that this clause is too wide since it imposes liability to the contractor in a wide variety of situations. The contractor is to be liable irrespective of whether such damages was caused by sole, active, passive, concurrent or partial negligence of whatever nature or character, fault or strict liability of the operator or the third party itself, or even though due to which could be considered as force majeure events,211 such as unseaworthiness and unairworthiness. As mentioned previously, this kind of drafting is an example of an attempt by the operator to dump his liability by shifting risk to the contractor. This should therefore be regarded as bad drafting. 3.2.5 Third-party claims under Operator C’s contract Operator C did not specify any indemnity regarding third-party claims. 3.3 Liability regarding pollution For other areas of risk not caused by the actions of either party, standard form contracts usually share the risk between them.212 For instance, in the case of weather, the risk is often apportioned according to whether weather conditions are exceptional or otherwise; it is usually deemed uneconomic to a client for contractors to account for the risks associated with exceptional weather conditions in the tender process.213 According to the FAIR principles, each party is best placed to indemnify the other party from the efects of pollution and contamination of whatever kind emanating from the party’s respective groups’ property and facilities regardless of cause (including negligence). This principle ensures that the parties to the contract have clear, easily understood and distinct liabilities during the performance of the project. Furthermore, particularly in relation to the contractor, it ensures that liability levels do not increase as a result of participating in the project. If the liability levels increase, it would result in additional insurance or risk management being required during the project term. A key aim is to avoid duplicate insurance which leads to extra costs and often to gaps in cover.214 3.3.1 Pollution liability under LOGIC Under the LOGIC contract form, there are reciprocal indemnifcations by both parties in respect of pollution. The contractor would indemnify and hold harmless the company group against claims in respect of pollution occurring on or emanating from contractor group’s premises, property or equipment. On the other hand, the 211 212 213 214

Edward G Hinkelman, Glossary of International Trade (5th edn, World Trade Press 2009) 75. Edwards (n134). Ibid. IMCA General Contracting Principles (Rev. 1) (n628).

285

wan mohd zulhafiz wan zahari company would indemnify and hold harmless the contractor group against claims in respect of pollution emanating from the reservoir or the company group’s property. The respective indemnity clause is written as follows: 22.3 Except as provided by Clause 22.1(a), Clause 22.1(b) and Clause 22.4, the COMPANY shall save, indemnify, defend and hold harmless the CONTRACTOR GROUP from and against any claim of whatsoever nature arising from pollution emanating from the reservoir or from the property of the COMPANY GROUP arising from, relating to or in connection with the performance or non-performance of the CONTRACT. 22.4 Except as provided by Clause 22.2(a) and Clause 22.2(b), the CONTRACTOR shall save, indemnify, defend and hold harmless the COMPANY GROUP from and against any claim of whatsoever nature arising from pollution occurring on the premises of the CONTRACTOR GROUP or emanating from the property and equipment of the CONTRACTOR GROUP (including but not limited to marine vessels) arising from, relating to or in connection with the performance or non-performance of the CONTRACT. 22.5 (a) Subject to Clause 22.5(b) below, the CONTRACTOR shall be responsible for the recovery or removal and when appropriate the marking or lighting of any wreck or debris arising from or relating to the performance of the WORK or the property, equipment, vessels or any part thereof provided by the CONTRACTOR GROUP in relation to the CONTRACT, when required by law, or governmental authority, or where such wreck or debris is interfering with COMPANY operations or is a hazard to fshing or navigation and shall, except as provided for in Clause 22.2 and Clause 22.3, save, indemnify, defend and hold harmless the COMPANY GROUP in respect of all claims, liabilities, costs (including legal costs), damages or expenses arising out of such wreck or debris, whether or not the negligence or breach of duty (whether statutory or otherwise) of the COMPANY GROUP caused or contributed to such wreck or debris. 22.6 (b) Notwithstanding the provisions of Clause 22.1, where the COMPANY provides transportation for the property of the CONTRACTOR GROUP to the ofshore WORKSITE, and the COMPANY elects to, or is required by law or governmental authority to recover or remove or mark or light any wreck or debris of such property, the COMPANY shall, except as hereinafter provided, save, defend, indemnify and hold harmless the CONTRACTOR GROUP from and against any claim of whatever nature relating to the costs of such recovery, removal, marking or lighting. Provided, however, that the foregoing indemnity and hold harmless shall not apply to the extent that the recovery, removal, marking or lighting arises as a result of the negligence or breach of duty (statutory or otherwise) of the CONTRACTOR GROUP.

3.3.2 Pollution liability under FIDIC No indemnity clause under FIDIC covers pollution. However, under Clause 4.18 in the FIDIC model contract, it specifes that the contractor is required to take all reasonable steps to protect the environment both on and of the site and to limit damage and nuisance to people and property resulting from pollution, noise and any pollution otherwise resulting from his operations. According to the FIDIC model contract, contractor’s liability should only be limited to the values indicated under the employer’s requirements. The respective clause is written as follows: 4.18 Protection of the Environment The Contractor shall take all reasonable steps to protect the environment (both on and of the Site) and to limit damage and nuisance to people and property resulting from pollution, noise and other results of his operations.

286

lessons from the application of knock-for-knock The Contractor shall ensure that emissions, surface discharges and efuent from the Contractor’s activities shall not exceed the values indicated in the Employer’s Requirements, and shall not exceed the values prescribed by applicable Laws.

3.3.3 Pollution liability under Operator A’s contract Operator A drafted the indemnity clause regarding pollution as follows: X.6.1 Pollution from the COMPANY’s Wells and Facilities The CONTRACTOR shall indemnify and hold harmless the COMPANY from and against such liability resulting from or for cost incurred or payments made by the COMPANY to control or clean up the pollutant or to prevent the threat of pollution or as compensation for damage sufered by others from any pollution originated from the wells, where applicable, owned or operated by the COMPANY arising from the CONTRACTOR’s and/or SUB-CONTRACTOR’s performance of the SCOPE OF WORKS under this CONTRACT howsoever caused. X.6.2 Pollution from Equipment and/or Property and/or Vessel The CONTRACTOR shall protect, indemnify and hold harmless the COMPANY from and against all liability for pollution emanating from equipment and/or property and/or vessel owned, leased, chartered or hired by the CONTRACTOR in connection to the performance of the SCOPE OF WORKS and shall reimburse COMPANY for all control and/or cleanup costs, and/or claims related to any such pollution. The CONTRACTOR undertakes that equipment or waste in any form originating from the SCOPE OF WORKS shall not be dumped overboard.

A literal reading of the above indemnity clause suggests that it is unilateral and onerous to the contractor. This is because the contractor is solely liable and required to indemnify and hold harmless the operator for pollution whether emanating from the operator’s well and facilities, or “from equipment and/or property and/or vessel owned, leased, chartered or hired by the contractor in connection to the performance of the scope of works”. However, this is the example whereby the clause is unilateral in nature, but it is in fact fair to the contractor, and reciprocity does not necessarily apply in this case. In the above clause, the contractor indemnifes the operator for pollution at the operator’s well and facilities which is “arising from the CONTRACTOR’s and/ or SUB-CONTRACTOR’s performance of the SCOPE OF WORKS”. Thus, there is no reason for the operator to be liable for the damages, unless the operator has contributed towards such pollution. 3.3.4 Pollution liability under Operator B’s contract Operator B drafted the indemnity clause regarding pollution as follows: X.1 CONTRACTOR agrees to indemnify, defend and save harmless COMPANY GROUP from and against any and all claims, losses and expenses including without limitation costs, demands, damages, suits, judgments, fnes, penalties, liabilities, reasonable attorneys’ fees and causes of action whatsoever nature or character, whether known or unknown, and including without limitation claims, losses and expenses for . . . pollution . . . in any way directly or indirectly, arising out of, or related to, the performance or subject matter of this CONTRACT or the ingress, egress, or presence on any premises (whether land, building, vehicle, platform, aircraft, vessel or otherwise) owned, operated, chartered, leased, used, controlled or hired by COMPANY GROUP or CONTRACTOR GROUP, and which are asserted by or arise in favor of CONTRACTOR GROUP and expressly

287

wan mohd zulhafiz wan zahari including any claims, losses or expenses actually or allegedly caused by the sole, active, passive, concurrent or partial negligence (of whatever nature or character), fault or strict liability of COMPANY GROUP or any other person or the unseaworthiness, unairworthiness or defective condition of vessels, craft ore premises, whether or not preceding or during the execution of this CONTRACT.

The above indemnity clause appears to be one-sided. It was drafted as widely as possible to lessen the liability of the operator and to shift greater risk and liability to the contractor including liability in relation to pollution. Under Clause X.1, the contractor is required to indemnify and hold harmless the operator against any claims, losses and expenses for pollution whether it directly or indirectly arose from the contract irrespective of whether the act was caused by the sole, active, passive, concurrent or partial negligence, fault or strict liability of the operator. 3.3.5 Pollution liability under Operator C’s contract Operator C did not specify any indemnity regarding pollution. 3.4 Insurance coverage In Malaysia, there are no specifc types of insurance that are required by law in relation to the provision of oil and gas services.215 Even though the maintenance of certain insurance policies is often required by the contract, where the contractors will typically be obligated to carry, among other policies, contractors’ all risks insurance covering general liability to the public – which are generally dependent on the type of contracts.216 Franklin notes that the recent trend with regard to Construction All Risk is that some operators prefer to take out self-insurance rather than follow the traditional method of extending its coverage to the contractors and subcontractors in order to cut costs for the operator.217 Therefore, the operator is always on the safer side. For example, the Production Sharing Contracts usually require PETRONAS to be named as co-insured with the company in insurance policies that have been taken out by the contractor in relation to the relevant oil and gas services.218 Thus, it could be argued that without any law to monitor this obligation, the contractors are exposed to the inherent risks, as it may not be covered by the insurance industry where the risk is regarded as operator’s risk. FIDIC presupposes that risk exists and should be covered by insurance, to the extent that is possible to cover such risk.219 It is in this context that Clauses 17 (Indemnity) and (Insurance) should be read together. However, insurance cover is only available for insurable risk, which is a risk depending on fortuity, which means that the event or circumstance has to be sudden and accidental and that comprehensive data for the purposes of premium calculation exists. Obviously, some of the risk inherent to a construction contract do not depend on fortuity – it is therefore not insurable. This type of risk is also referred to as speculative risk. In principle, speculative risks

215 216 217 218 219

Hewitt (n19) 331. Ibid. Franklin (n143). Hewitt (n19) 366. Jaeger and Hök, FIDIC: A Guide for Practitioners.

288

lessons from the application of knock-for-knock are foreseeable. An experienced contractor should be able to foresee potential risks. The question is whether he should also make allowance for the event that the risk occurs. Unforeseeability means the risk is not reasonably foreseeable by an experienced contractor by the date for submission of the tender. The second paragraph of Subclause 18.1 requires a meeting between the Employer and the Contractor in order to agree the terms of insurance prior to the issue of the letter of acceptance. This is a critical point, which should not be ignored. However, it is desirable that the terms and issues of insurance are clearly identifed before the submission of the tender. Under the LOGIC contract form, the insurance required under the provisions of the insurance clause will vary depending on the scope of work for each particular contract. The exact values of insurances required by the company must be specifed in appendix 1. The clause also requires that subcontractors carry appropriate amounts of insurance as may be relevant to their work. Although not provided in the standard wording, the contractor is recommended to consider the need for a reciprocal commitment from the company to also insure.220 In this respect, the main criteria will of course be determined by the size and fnancial stability of the company in each case. Hence the insurance provisions under FIDIC and LOGIC could be regarded as models for ideal risk allocation between the parties.221 4 Conclusion Based on the empirical study conducted in Malaysia discussed in this chapter, this research argues that there is unequal bargaining power between operators and contractors in the Malaysian oil and gas industry. This inequality of bargaining power leads to imbalanced risk allocation and unfair indemnity hold harmless clauses in oilfeld service contracts. In Malaysia, the absence of a law to regulate imbalanced risk allocation and unfair indemnity and hold harmless clauses in oilfeld service contracts should be perceived as a serious problem because it leads to the problem of inequality of bargaining power resulting from the dominant position of the operators over the contractors. To date, there is no statutory restriction on contractual provisions purporting to exclude, limit or indemnify one or both of the parties in relation to liability and indemnity. Generally, contract law in Malaysia is governed by the Contract Act 1950. However, section 77 of the Contract Act 1950 merely discusses the meaning of “contract of indemnity”. Meanwhile, section 78 of the Contract Act 1950 merely talks about “right of indemnity holder when sued”. Based on the case analysis in Malaysia, the courts have not addressed this problem. Furthermore, the judicial position on the doctrine of unconscionability and inequality of bargaining power remains unclear in Malaysia since the Court of Appeal has taken inconsistent views in the cases of

220 ‘General Conditions of Contract (Including Guidance Notes) for Construction’ in Standard Contract for U.K. Ofshore Oil and Gas Industry, vol October (2nd edn, LOGIC 2003) 2. 221 Wan M Zulhafz, ‘A Comparative Analysis on the Enforceability of Knock-for-Knock Indemnities in Thailand and the United Kingdom’ (2017) 44 J.Malaysian Comp.Law (JMCL) 33.

289

wan mohd zulhafiz wan zahari Saad Marwi222 and American International Assurance Co Ltd.223 As a result of this gap in the law which results in lack of legal protection to the contractor, the operator might use his dominant position and continuously shift greater risk to the contractor. This uneven and therefore unfair risk allocation may cause signifcant fnancial setbacks to the contractor. This problem deserves attention from the Malaysian government. The problem must be resolved. If left unresolved, it presents a threat to the commercial development of the Malaysian oil and gas industry. Even though insurance provisions are sometimes provided in the contract, there are no guidelines available to govern the conduct of the parties. Additionally, the insurance requirements are not mandatory for the parties. Some operators opt to take out self-insurance rather than to buy a premium that will extend its coverage to cover the operators and subcontractors, for example in respect of Construction All Risk. This scenario can cause the contractor to assume uninsured risks, which could lead to detrimental fnancial exposure in the occasion of a catastrophic incident. The situation might get worse for contractors in the event that contractors have to assume double jeopardy contractual risk, whereby the contractors not only need to assume operators’ risk and also subcontractors’ risk. To solve this problem, it is argued that a specifc legal mechanism should be adopted in Malaysia to protect and limit the liability of the contractors under oilfeld service contracts. As discussed above, sections 3 and 5 of the Civil Law Act 1956 provide for the application of English law in Malaysia unless another provision has been or shall be written into law. The English legal principles are therefore applicable and ought to be applied to cover the gap in Malaysian law. However, the applicability of the English legal principles will be subject to the circumstances in the Malaysian oil and gas industry.

222 Saad Marwi v Chan Hwan Hua & Anor [2001] 3 CLJ 98. 223 American International Assurance Co Ltd v Koh Yen Bee (f) [2002] 4 MLJ 301.

290

INDEX

Note: Page numbers in italic indicate a figure and page numbers in bold indicate a table on the corresponding page. accident costs 65–68, 85–86; deterrence versus compensation 84–85; harm to environmental interests 82–84; illustrating the operation of knock for knock 69–71; industrial setting 69; litigation externalities 80–82; private ordering 71–73; risk interdependencies 80; social norms 71–73; social welfare 73–79 admiralty jurisdiction (United States) 216–217 Africa: Francophone African legal systems 239–240; see also France and Francophone civil law systems AI endorsements 100–102 anti-indemnity statutes (United States) 113–114 Australia: contribution to historical evolution of K4K clauses 13 back-to-back indemnity: oil and gas contracts 30–31, 32 balanced regime 120–121 BGB (Germany) 200–201 BIMCO SUPPLYTIME 2017 5–6 Brazil 204–205, 212–213; and choice-of-law 114–116; civil liability regime 205–209; contribution to historical evolution of K4K clauses 11–12; knock-for-knock clauses 211–212; limitation of liability clause 209–211 breach: contractor’s liability for 120–125; of statutory duty 40–46; validity of agreed limitations for 137–145 carve-outs: gross negligence 92–97 case law: Germany 202; see also English case law case studies (Malaysia) 261–262; analysis of 274–289; claims by a third party 281–285; description of 263–274; findings 265; insurance coverage 288–289; liability regarding personal injury or death of employees and loss of property of the parties 276–281; liability regarding pollution

285–288; perceptions of the contractual formation process 265–268; perception of risk allocation and indemnity clauses 268–273; research design 262–263 “caveat” clause (Germany) 196–197 censoring: knock-for-knock regime 145–150; provisions limiting exposure 134–135 choice-of-law 106–111; and Brazil 114–116; and United States 111–114 civil liability regime (Brazil) 205–209 co-insured 148–149 commercial contracts versus consumer contracts (Germany) 198 company groups 49–50 compensation: deterrence versus 84–85 “consequential” loss: liability for 59–62 construction contracts see fabrication and construction contracts (Norway) consumer contracts versus commercial contracts (Germany) 198 content test (Germany) 195–196 contextualism 39–40 Contract Act § 36 (Norway) 178–181 contractor 141–143; gross negligence 138–141; liability for breach 120–125 contractual formation process (Malaysia) 265–268 contractual liability (Brazil) 205–206 contra proferentem 38–39; and the problem of negligence and breach of statutory duty 40–46 control 171–172 courts: carve-outs for gross negligence 92–97; review of general terms and conditions (Germany) 194–197 damage: to the contract object 156–158; provisions limiting exposure resulting from 134–137; risks of 125–126

291

index death of employees (Malaysia) 276–281 debate: negotiation of terms demands more than debating (Germany) 196 Deepwater Horizon: and choice-of-law 111–113; contribution to historical evolution of K4K clauses 10–11; see also litigation post Deepwater Horizon default: provisions limiting exposure resulting from 134–137 deterrence: compensation versus 84–85 development of knock-for-knock clauses see knock-for-knock clauses, development of distribution of risk: normal presumptions of 35–36 diversity jurisdiction (United States) 215–216 drafting: contra proferentem and the problem of negligence and breach of statutory duty 40–46; definitional issues 49–50; “full and primary” 48–49; multi-party issues 47–48; words delimiting the circumstances in which the indemnity and hold harmless provision will take effect 46–47 efficiency: basis for liability in contractual relationships 174–175; insurance coverage 172–173; liability rules 173–174, 174 employees 50; personal injury or death of (Malaysia) 276–281 enforceability 3; United States 224–226 English case law and legal context 107, 109–111; evolution of K4K clauses in 7–10 entry into force 54–55 environmental interests: harm to 82–84 exceptions 56; French law 246–247 exclusion of liability: French law 246–248; Germany 199–202 exposure see provisions limiting exposure extension of benefits to groups 55 fabrication and construction contracts (Norway) 117, 153–156; censoring the knock-for-knock regime 145–150; contractor’s liability for breach under offshore construction contracts 120–125;damage to the contract object 156–158; do NTK 15 provisions on exclusion and limitation of liability hold good 150–152; knock-for-knock system 125–134; offshore project insurances 158–163; the problem 118–120; risk zones 156; setting aside provisions limiting exposure resulting from default or damage 134–137; validity of agreed limitations of liability for breach of contract 137–145; waivers and indemnities shall apply regardless of cause 158; waivers of liability, indemnification and waiver of subrogation 156 fairness test (Germany) 198

family zones 130–131 fault, nature of (French law) 248 fault-based liability (Brazil) 206–207 federal court jurisdiction (United States) 215–217 federal maritime law (United States) 111–113 FIDIC: personal injury or death of employees and loss of property of the parties 277–278; pollution liability 286–287; third-party claims 283–284 France and Francophone civil law systems: Francophone African legal systems 240–241; French Civil Code of 1804 239–240; French law 240; indemnification of third-party losses 249; mutual hold harmless clauses 241–249 freedom of liability, indemnity and subrogation 169–170 French Civil Code of 1804 239–240 “full and primary” 48–49 general terms and conditions (Germany): are knock-for-knock clauses general terms and conditions? 197; court review of 194–197 geographical extent 55 Germany: case law 202; contribution to historical evolution of K4K clauses 12–13; use of knock-for-knock clauses in 193–194, 202–203; validity issues 194–202 gross negligence: carve-outs for 92–97 “group concept, the” 167–169 harm to environmental interests 82–84 health and safety regulation 77–78 hold harmless clauses: and indemnity 22–24; insurance coverage (Malaysia) 288–289; interpreting 37–40; Malaysia 274–289; and mutual indemnity 24–25, 24; oil and gas context 25–34; personal injury or death of employees and loss of property of the parties (Malaysia) 276–281; pollution liability (Malaysia) 285–289; statutory control of 34–35; third-party claims (Malaysia) 281–285; words delimiting circumstances 46–47; see also mutual hold harmless clauses (France and Francophone civil law systems) implementing K4K clauses: issues in 3 indemnification 21–25; French law 249–250; indemnifiable losses (Brazil) 207–209; normal presumptions about the distribution of risk 35–36; Norway 154–155, 156; position of third parties 36–37; statutory control of indemnity and hold harmless clauses 34–35 indemnity 22–24, 98–99; anti-indemnity statutes (United States) 113–114; apply regardless of cause 158; back-to-back 30–31, 32; and

292

index general liability insurance 99–100; industry mutual hold harmless agreement (IMHH) 53–54; insurance coverage (Malaysia) 288–289; interpreting 37–40; limitations in insurance policies 103–104; Malaysia 255– 261, 268–274, 274–289; mutual 24–30, 24; oil and gas context 25–34; personal injury or death of employees and loss of property of the parties (Malaysia) 276–281; pollution liability (Malaysia) 285–288; qualified 31–34; simple 25–26; statutory control of 34–35; third-party claims (Malaysia) 281–285; United States 224–229; words delimiting circumstances 46–47; see also indemnity (Norway) indemnity (Norway) 117, 153–156; censoring the knock-for-knock regime 145–150; contractor’s liability for breach under offshore construction contracts 120–125; damage to the contract object 156–158; do NTK 15 provisions on exclusion and limitation of liability hold good 150–152; freedom of 169–170; knock-for-knock system 125–134; offshore project insurances 158–163; the problem 118–120; risk zones 156; setting aside provisions limiting exposure resulting from default or damage 134–137; validity of agreed limitations of liability for breach of contract 137–145; waivers and indemnities shall apply regardless of cause 158; waivers of liability, indemnification and waiver of subrogation 156 “individually negotiated” (Germany) 196–197 industrial setting 69 industry mutual hold harmless agreement (IMHH) 52–59 industry-negotiated model contracts (Norway) 188–189 insurance 4; co-insured 148–149; global view of 128; incorporation of indemnity limitations in insurance policies 103–104; insured perils 160–162; Malaysia 288–289; “other insurance” clauses 102–103; perfect pooling equilibrium 78–79; and regulation 127–128; United States 227–229; see also insurance (Norway); liability insurance insurance (Norway) 131–134, 153–156, 164–165; content and structure of the knock-for-knock principle 166–171; damage to the contract object 156–158; freedom of liability, indemnity and subrogation 169–170; “the group concept” 167–169; insurance law 166; insurance regulation 170–171; offshore project insurances 158–163; rationale for the knock-for-knock principle 171–175; risk zones 156; tort law 165–166; type of loss and basis for liability 166–167; validity of the

regulation 175–181; waivers and indemnities shall apply regardless of cause 158; waivers of liability, indemnification and waiver of subrogation 156 intent, exclusion of liability for (Germany) 199–202 interpretation: contra proferentem and the problem of negligence and breach of statutory duty 40–46; definitional issues 49–50; “full and primary” 48–49; indemnity and hold harmless clauses 37–40; multi-party issues 47–48; words delimiting the circumstances in which the indemnity and hold harmless provision will take effect 46–47 invalid terms (Germany) 199–202 jurisdiction see Australia; Brazil; France and Francophone civil las systems; Germany; Malaysia; Nordic countries; Norway; United States knock-for-knock clauses, development of 1–3; Australia 13; Brazil 11–12; conclusive critique 17–19; Germany 12–13; issues in using, implementing and enforcing 3; modern use 5–6; Nordic countries 13–17; offshore oil and gas sector 6–7; shipping sector 6–7; standard contracts and English case law 7–10; US jurisprudence 10–11; various uses 3–5 knock-for-knock clauses, effect of choice-of-law on 106–111; and Brazil 114–116; and United States 111–114 knock-for-knock clauses, implementing: issues in 3 knock-for-knock clauses, regulation of 87–88, 97; carve-outs for gross negligence 92–97; literature 88–89; Norway 125–128; when useful 89–92 knock-for-knock clauses, using 3–5: BIMCO SUPPLYTIME 2017 5–6; insurance 4; issues in 3; offshore contracts 4–5; P&I clubs 4; SUPPLYTIME 2005 3–4; when clauses are useful 89–92 landward areas 57 law of obligations (French law) 239; exceptions linked to 246–247 liability: Brazil 205–209;“consequential” loss 59–62; exclusion of (Germany) 199–202; French 239, 241–244, 246; Malaysia 276–281, 285–288; see also liability (Norway); liability insurance; limitation of liability liability (Norway) 164–165; censoring the knock-for-knock regime 145–150; content

293

index and structure of the knock-for-knock principle 166–171; contractor’s liability for breach under offshore construction contracts 120–125; freedom of liability, indemnity and subrogation 169–170; “the group concept” 167–169; insurance 131–134; insurance law 166; insurance regulation 170–171; the problem 118–120; provisions on 128–131; rationale for the knock-for-knock principle 171–175;and regulation 125–128; setting aside provisions limiting exposure resulting from default or damage 134–137; statutory regulation of risk 182–187; tort law 165–166; type of loss and basis for liability 166–167; unlimited no-fault 184–185; validity of the regulation 175–181 liability insurance 98–100; incorporation of indemnity limitations in insurance policies 103–104; “other insurance” clauses 102–103 licencee: channelling liability to (Norway) 186–187 limitation of liability 62–64; Brazil 209–211; Norway 124–125, 137–145, 150–152 literature 88–89 litigation externalities 80–82 litigation post Deepwater Horizon (United States) 235–237; applicable law 214–224; federal court jurisdiction 215–217; indemnity in maritime 224–229; maritime law 219–224; Outer Continental Shelf Lands Act 217–219, 223–224; state law, application of 229–234; state waters 219 LOGIC: personal injury or death of employees and loss of property of the parties 276–277; pollution liability 285–286; third-party claims 282–283 Longshore and Harbor Workers’ Compensation Act (United States) 226–227 loss: indemnifiable (Brazil) 207–209; nature of (French law) 248–249; prevention of 173–175, 174; property of the parties (Malaysia) 276–281; third-party (French law) 249–250; type of 166–167 Louisiana Oilfield Indemnity Act (United States) 113–114, 229–233 Malaysia 289–290; analysis of indemnity and hold harmless clauses 274–289; case studies 263–274; claims by a third party 281–285; empirical study of oilfield service contracts 261–262; findings 265; general overview of legal system 251–252; insurance coverage 288–289; law of contract 252–261; liability regarding personal injury or death of employees and loss of property of the

parties 276–281; liability regarding pollution 285–288; perception of the contractual formation process 265–268; perception of risk allocation and indemnity clauses 268–274; research design 262–263 maritime law, federal (United States) 111–113, 219–223; indemnity in 224–229; supplemented with state law 223; tension between OCSLA and 223–224 market practice (Germany) 198 multiple parties 50–59, 51; drafting and interpretation 47–48 mutual hold harmless clauses (France and Francophone civil law systems): and English terminology 241–242; Francophone African legal systems 240–241; French Civil Code of 1804 239–240; French law 240; indemnification of third-party losses 249; legal nature of 242–246; legal regime governing 246–250; mutual hold harmless clauses 241–249; origins and growing prevalence 241 mutual indemnity 24–30, 24 nature of the fault (French law) 248 nature of the loss (French law) 248–249 negligence: carve-outs for gross negligence 92–97; and contra proferentem 40–46 negotiation: “individually negotiated” (Germany) 196–197; negotiation of terms demands more than debating (Germany) 196 NL 5–1–2 (Norway) 176–178 no-fault liability (Norway) 184–185 non-contractual liability (French law) 246 Nordic countries: contribution to historical evolution of K4K clauses 13–15; see also Norway Norway see fabrication and construction contracts (Norway); indemnity (Norway); insurance (Norway); liability (Norway) Norwegian Petroleum Act 182–187, 192 NTK 15 150–152 obligations, law of see law of obligations (French law) obligation to notify 123–124 offshore sector 4–5, 164–165; content and structure of the knock-for-knock principle 166–171; freedom of liability, indemnity and subrogation 169–170; “the group concept” 167–169; historical encounter of the K4K clauses in 6–7; insurance law 166; insurance regulation 170–171; rationale for the knock-for-knock principle 171–175; tort law 165–166; type of loss and basis for liability 166–167; validity of the regulation 175–181;

294

index see also fabrication and construction contracts (Norway) operation of knock for knock: illustration of 69–71 operators 57 order of precedence 55 “organized financing of loss” 147–148 “other insurance” clauses 102–103 Outer Continental Shelf Lands Act (United States) 217–219; tension between maritime law and 223–224

normal presumptions about the distribution of risk 35–36; Norway 187–192; overall limitation of liability 62–64; position of third parties 36–37; statutory control of indemnity and hold harmless clauses 34–35; words delimiting the circumstances in which the indemnity and hold harmless provision will take effect 46–47

P&I clubs 4 perfect pooling equilibrium 74–79 perfect separating equilibrium 74 perils of knock for knock 79–85 personal injury or death of employees (Malaysia) 276–281 personnel 50; see also employees Piper Alpha litigation 9–10 pollution: Malaysia 285–288; Norway 184–187 private ordering 71–73 property 50; loss of the property of the parties (Malaysia) 276–281 provisions limiting exposure 134–137 qualified indemnity 31–34 “reasonable” arrangements (Norway) 149–150 reasonableness test (Germany) 197–199 recovery, basis for 162–163 regulation: insurance 170–171; liability 136–137, 182–187; see also knock-for-knock clauses, regulation of repeat interactions 76–77 research design (Malaysia) 262–263 right to defend 56 risk: of damage 125–126; global view of 128; interdependencies 80; need for contractual control of 171–172; risk insurances 155–156; “risky” terms (Germany) 198–199; risk zones 156; statutory regulation of (Norway) 182–187; see also risk allocation risk allocation 20–21; contra proferentem and the problem of negligence and breach of statutory duty 40–46; definitional issues 49–50; French law 243, 244; “full and primary” 48–49; indemnification and related concepts 21–25; indemnity and hold harmless provisions in the oil and gas context 25–34; interpreting indemnity and hold harmless clauses 37–40; known problems of drafting and interpretation 40–50; liability for “consequential” loss 59–62; Malaysia 268–274; multiple parties 47–48, 50–59, 51;

sanctions against breach 121–122 secondary obligations: breach of 123 self-regulation 75–76 shipping sector: historical encounter of the K4K clauses in 6–7 simple indemnity: oil and gas contracts 25–26 social norms 71–73 social welfare: effect of knock for knock on 73–79 standard contracts: evolution of K4K clauses in 7–10 state law (United States): Louisiana Oilfield Indemnity Act 113–114, 229–233; and maritime law incomplete 223; Texas Oilfield Anti-Indemnity Act 114, 233–234 state waters (United States) 219 statutory duty see breach of statutory duty strict liability (Brazil) 206–207 subrogation: freedom of 169–170; see also waiver of subrogation SUPPLYTIME 2005 3–4 terms and conditions see general terms and conditions; invalid terms Texas Oilfield Anti-Indemnity Act (United States) 114, 233–234 third parties 36–37; French law 243–246, 249–250; Germany 201; Malaysia 281–285; Norway 131 tort defaults 65–68, 85–86; deterrence versus compensation 84–85; harm to environmental interests 82–84; illustrating the operation of knock for knock 69–71; industrial setting 69; litigation externalities 80–82; private ordering 71–73; risk interdependencies 80; social norms 71–73; social welfare 73–79; tort law (Norway) 165–166 tort liability (Brazil) 205–206 transportation by air excluded 56–57 type of loss 166–167 UKCS: normal presumptions about the distribution of risk 35–36; position of third parties

295

index 36–37; statutory control of indemnity and hold harmless clauses 34–35 United States 235–237; applicable law 214–224; and choice-of-law 111–114; contribution to historical evolution of K4K clauses 10–11; federal court jurisdiction 215–217; indemnity in maritime 224–229; maritime law 219–224; Outer Continental Shelf Lands Act 217–219, 223–224; state law, application of 229–234; state waters 219 unlimited no-fault liability (Norway) 184–185 using knock-for-knock clauses 3–5: BIMCO SUPPLYTIME 2017 5–6; insurance 4; issues in 3; offshore contracts 4–5; P&I clubs 4; SUPPLYTIME 2005 3–4; when clauses are useful 89–92 validity: agreed limitations of liability for breach of contract 137–145; French law 246–247, 249; Germany 194–202; Norway 175–181 variation mechanism 123–124

vessel services, contracts for (Norway) 164–165; content and structure of the knockfor-knock principle 166–171; freedom of liability, indemnity and subrogation 169–170; “the group concept” 167–169; insurance law 166; insurance regulation 170–171; rationale for the knock-for-knock principle 171–175; tort law 165–166; type of loss and basis for liability 166–167; validity of the regulation 175–181 waiver of liability (Norway) 156 waiver of recourse (French law) 246–248; effects of 247–248 waiver of rights of subrogation 55–56 waiver of subrogation 104–105; maintenance of 148–150; Norway 156 waivers apply regardless of cause (Norway) 158 “willful misconduct” 143–145 words delimiting circumstances: indemnity and hold harmless 46–47

296