BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales High Court (Commercial Court) Decisions


You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> ABN Amro Bank NV v Royal & Sun Alliance Insurance Plc & Ors [2021] EWHC 442 (Comm) (26 February 2021)
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2021/442.html
Cite as: [2021] Lloyd's Rep IR 467, [2021] EWHC 442 (Comm)

[New search] [Context] [View without highlighting] [Printable PDF version] [Help]


Neutral Citation Number: [2021] EWHC 442 (Comm)
Case No: CL-2018-000771

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

Royal Courts of Justice
Strand, London, WC2A 2LL
26/02/2021

B e f o r e :

MR JUSTICE JACOBS
____________________

Between:

 

ABN AMRO BANK N.V.

Claimant

 

 

- and -

 

 

 

(1) ROYAL & SUN ALLIANCE INSURANCE plc

(2) NAVIGATORS UNDERWRITING AGENCY LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 1221 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(3) TALBOT UNDERWRITING LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 1183 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(4) BRIT SYNDICATES LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 2987 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

 (5) HARDY (UNDERWRITING AGENCIES) LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 382 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(6) AEGIS MANAGING AGENCY LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 1225 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(7) MARKEL SYNDICATE MANAGEMENT LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 3000 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(8) ARK SYNDICATE MANAGEMENT LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 3902 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(9) THE CHANNEL MANAGING AGENCY LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 2015 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(10) ADVENT CAPITAL (HOLDINGS) LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 780 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(11) ASSICURAZIONI GENERALI S.p.A.

(12) CHARLES TAYLOR MANAGING AGENCY LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 1884 (THE STANDARD SYNDICATE) AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(13) COVERYS MANAGING AGENCY LIMITED, ON ITS OWN BEHALF AND ON BEHALF OF ALL SUBSCRIBING MEMBERS OF SYNDICATE NO. 1110 AT LLOYD'S OF LONDON FOR THE 2016 YEAR OF ACCOUNT

(14) SWISS RE LIMITED

(15) EDGE BROKERS (LONDON) LIMITED

Defendants

 

____________________

Rebecca Sabben-Clare QC, Benjamin Parker and Julia Gibbon (instructed by Reed Smith LLP) for the Claimant
Luke Parsons QC, Stewart Buckingham QC and Will Mitchell (instructed by Kennedys Law LLP) for the 1st – 14th Defendants
Siobαn Healy QC and Harry Wright (instructed by Reynolds Porter Chamberlain LLP) for the 15th Defendant
,
Hearing dates: 9-12, 16-19, 23-26, 30 November,
1-2, 8-10, 14-15 December 2020

____________________

HTML VERSION OF JUDGMENT
____________________

Crown Copyright ©

    "Covid-19 Protocol: This judgment will be handed down by the judge remotely by circulation to the parties' representatives by email and release to BAILII. The date and time for hand-down will be deemed Friday 26th February 2021"


     

    Index

    Section Para. Number
    A: Introduction
    1
    A1: The parties and the claim
    1
    A2: The issues
    12
    A3: The Trial
    19
    B: The Factual Background
    29
    Section C: The Policy terms and GUA 168
    C1: The policy subscribed by the market between 25 and 29 January 2016
    168
    C2: The policy subscribed by Mr. Beattie on 29 January 2016 , and by Prosight and Swiss Re in February 2016
    169
    C3: General Underwriters Agreement 173
    D: Construction of the policies 175
    D1: Legal principles 175
    D2: The parties' arguments 189
    D3: Discussion 206
    E: Rectification/estoppel/collateral contract
    315
    E1: Legal principles 316
    E2: The witness evidence 332
    E3: The Parties' arguments 366
    E4: Discussion
    382
    F: Non-disclosure and misrepresentation
    448
    F1: Introduction to the issues 448
    F2: The Non-Avoidance clause 463
    F3: Affirmation 485
    F4: The misrepresentations and non-disclosure relied upon: general matters
    567
    F5: Purpose/intention of the TPC 585
    F6: Non-disclosure of the NAC 634
    F7: Non-disclosure of the 2015 endorsement
    641
    F8: As expiry misrepresentation 647
    F9: Only PLOD representation to Standard
    683
    F10: The misrepresentation as the foundation for an estoppel
    692
    G: Clause 3 and lack of quality checks carried out by the Bank
    706
    G1: The issues 706
    G2: The construction of the policy 720
    G3: Was the Bank reckless in relation to the quality of the collateral?
    742
    H: Sue and Labour 764
    H1: The issues 764
    H2: Legal principles 781
    H3: The facts relating to the Bank's conduct
    792
    H4: The expert evidence 813
    H5: Is the applicable standard "good faith"?
    819
    H6: The Bank's conduct 825
    Section I: Quantum of the claim against underwriters
    854
    J: The claim against Edge 861
    J1: The issues 861
    J2: The broking of the TPC 876
    J3: Causation and Quantum Issues 941
    J4: The Bank's residual claim against Edge for irrecoverable costs
    1021
    J5: Edge's liability in respect of Clause 3 defence
    1030
    K: CONCLUSION 1034

    Mr Justice Jacobs:

    A: Introduction

    A1: The parties and the claim

  1. In these proceedings the Claimant ("the Bank") claims an indemnity of approximately £33.5 million under a policy of insurance subscribed to by the 1st – 14th Defendants as underwriters (collectively, "the underwriters"). Whilst there is no dispute that each of the underwriters did subscribe to a relevant policy of insurance, the precise identification of the relevant policy document is one of the many issues in the case.
  2. The policy was led by the 1st Defendant, Royal & Sun Alliance Insurance plc ("RSA"). It was placed by the 15th Defendant, Edge Brokers (London) Limited ("Edge"). The policy covered the period of one year commencing on 1 February 2016. The underwriters deny liability on various grounds, and the Bank's claim against Edge arises, principally, if the underwriters' main grounds of defence are successful.
  3. The dispute has arisen because the policy was placed in the marine market in London with subscribing underwriters who were specialists in insuring cargo in warehouses and in transit, and in particular the risk of physical loss and damage to that cargo. However, the policy contained an unusual clause which, as the Bank and Edge contend, was intended to and did widen the cover so as to include risks which were not dependent on physical loss and damage. The clause in question was known as the "Transaction Premium Clause" or "TPC". The Bank and Edge contend that its effect is to cover certain losses suffered by the Bank arising from the default of its customers, even if there was no physical loss and damage to the cargo.
  4. The underwriters deny that this is the effect of the clause on its true construction. They rely upon the fact, which is clear from the evidence at trial, that cover for the risk arising from the default of customers would ordinarily be placed with underwriters who specialised in providing insurance known as trade credit insurance. It is not usually provided by underwriters who provide cargo insurance. They therefore contend that the TPC should not be construed so as to provide, in effect, trade credit insurance and must be read as being only applicable where physical loss and damage is caused to the cargo.
  5. If their argument on construction fails, then they advance a case based upon conversations which took place between the leading underwriter, Mr. Brian Beattie of RSA and Mr. David Mullen who was the broker at Edge with primary responsibility for the placement of the insurance. The effect of those conversations, on the underwriters' case, is to prevent (via rectification, or estoppel or related principles) the Bank from relying on the terms of the policy on their true construction. The underwriters also contend that they are entitled to avoid the policy for non-disclosure or misrepresentation. The inevitable result of the case based on the conversations, and avoidance, is that the court has received a very large volume of evidence which would not ordinarily be admissible on an issue of construction.
  6. The claim arises out of the business of a special purpose vehicle, or SPV, of the Bank called Icestar B.V. ("Icestar"). The business of Icestar involved the provision of structured commodities finance to clients of the Bank. These clients were in the business of buying and selling commodities. The finance provided by Icestar comprised transactions known colloquially as "repo" transactions. They involved the provision by Icestar of working capital by purchasing the client's commodity for a defined period of time. At the end of that period the client was contractually obliged to buy the commodity back from Icestar. During the course of the trial, the parties and the witnesses sometimes referred to Icestar specifically, but usually simply referred to "the Bank" as being synonymous with or at least encompassing Icestar. In this judgment, I will usually refer simply to "the Bank" since it is generally not necessary to draw any distinction between the Bank and Icestar. From time to time, however, I will refer to "Icestar", either because there are some occasions on which it is appropriate to draw a distinction, but usually because a witness referred to Icestar rather than simply the Bank.
  7. The commodities giving rise to the present claim are various cocoa products such as cocoa butter, cocoa cake, cocoa liquor and cocoa powder. These products are derived from cocoa beans, and are ultimately purchased by end-users such as major chocolate manufacturers. Both cocoa beans and cocoa products are traded by commodity traders, with the volume of transactions in cocoa beans substantially exceeding the volume of transactions in products.
  8. During the second half of 2016 two of the leading players in the world cocoa market, known as Transmar and Euromar, suffered a major and ultimately terminal financial collapse. Transmar (in the US) went bankrupt and Euromar (in Germany) went into insolvency. Senior executives of both companies were convicted and imprisoned in the US for frauds committed against several banks, including the Bank. One aspect of these companies' fraudulent behaviour was that they were deliberately misstating the extent and value of their collateral.
  9. Many banks suffered significant financial loss as a result of Transmar and Euromar's fraudulent conduct and collapse, including the Bank itself. The Bank (not Icestar) had lent money to Transmar pursuant to a US$363 million syndicated revolving credit facility, sometimes described as a "borrowing base" facility and suffered losses. Icestar, which had financed the companies through a portfolio of repo deals, also suffered losses.
  10. Transmar and Euromar defaulted under the relevant repo deals with Icestar during the period August-December 2016 by not repurchasing cocoa beans and products as they were contractually obliged to do. Icestar was left having to dispose of the cargo at the best prices it could achieve. It was possible for Icestar to find buyers for the cocoa beans, which were generally of good quality, and no relevant losses were suffered by Icestar on the transactions where it held beans. However, the cocoa products proved to be of poor quality, and there was a significant shortfall between what could be recovered under sales to third parties and the amounts owed by Transmar and Euromar. This shortfall is the primary loss for which the Bank claims an indemnity under the policy.
  11. The policy was built on a foundation of conventional marine "all risks" terms including the Institute Cargo Clauses 'A'. However, it also contained a large number of detailed clauses, including extensions to the cover which went beyond ordinary physical loss and damage to the cargo. The TPC, which is at the heart of the present dispute, was a clause initially presented to Mr. Beattie in July 2015. There is a dispute as to whether Mr. Beattie agreed to amend the cover in July 2015 so as to include the TPC for the 2015/2016 policy year, and (if so) whether his agreement bound the other underwriters on risk for that year. But there is no dispute that the TPC was contained in the cover which each insurer agreed to provide in early 2016 in respect of the 2016/2017 year. The slip ultimately "scratched" (ie signed) by Mr. Beattie on behalf of the RSA in January 2016, and also by the 13th and 14th Defendants in February 2016, contained the TPC in two separate places in the policy wording. The slip scratched by the other underwriters contained the TPC once, and in a different location.
  12. A2: The issues

  13. The first issue in the case is whether, as the Bank contends, the TPC provides credit risk insurance in respect of all risks of financial default and so responds to the claim. The TPC reads as follows, with underlining to identify the words particularly relied upon by the Bank in support of their argument that the clause provides such coverage:
  14. "Underwriters note and agree that, in respect of any Transaction, it is hereby confirmed that the Insured is covered under this contract for the Transaction Premium that the Insured would otherwise have received and/or earned in the absence of a Default on the part of the Insured's client.
    'Actual Sale Price' means the sum received by the Insured upon the sale of the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    'Default' means a failure, refusal or non-exercise of an option, on the part of the Insured's client (for whatever reason) to purchase (or repurchase) the Subject Matter Insured from the Insured at the Pre-agreed Price.
    'Pre-agreed Price' mean the amount for which the Insured's client had agreed to purchase (or repurchase) the Subject Matter Insured from the Insured as specified on the relevant invoice or in the relevant transaction documents, comprising the principal together with any premium or profit element payable to the Insured.
    'Transaction' means any transaction where, following a Default on the part of the Insured's client, the Insured sells the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    'Transaction Premium' means an amount that is equal to the difference in value between the Pre-Agreed Price and the Actual Sale Price."
  15. If the underwriters' argument on construction succeeds, then none of the other defences arise, and the remaining question is the liability of Edge. The construction arguments are addressed in Section D below, and the case against Edge in Section J below. If, however, the defence on construction fails, then a number of defences advanced by the underwriters arise. Those defences, and the order in which they are considered in this judgment, are as follows.
  16. First, the underwriters rely upon statements alleged to have been made by Mr. Mullen to Mr. Beattie on various occasions. These are said to give rise to rectification of the policy, estoppel or a collateral contract: see Section E below. All underwriters seek to rely upon the case of rectification, estoppel or collateral contract arising from the discussions between Mr. Mullen and Mr. Beattie.
  17. Secondly, underwriters advance a case of non-disclosure and misrepresentation. This avoidance case has various facets which it is not necessary to describe in detail at this stage. A central theme is that Edge failed to disclose to all of the underwriters that the purpose of the TPC was to provide insurance for default in the absence of physical loss or damage. Other arguments arise from the fact that none of the underwriters, other than Mr. Beattie, recall reading the TPC. The underwriters contend that it should have been specifically drawn to their attention, particularly in the context of renewal discussions.
  18. The Bank and Edge contend that there is no substance to the case of non-disclosure. They place reliance upon the terms of a "Non Avoidance Clause" or "NAC" which was initially shown to Mr. Beattie in July 2015, and then included in the slips signed by all subscribing underwriters in January/ February 2016. They contend that this clause precludes any avoidance. In any event, they say that any avoidance is barred by affirmation as well as the intrinsic lack of merit of the defence. These issues are addressed in Section F below. That section also considers a separate case of estoppel, advanced by particular underwriters (Navigators, Ark, Advent and Standard), which is closely related to a misrepresentation defence which those underwriters advance.
  19. Third, the underwriters argue (pursuant to an amendment made in July 2020) that the Bank acted recklessly or negligently when it entered into the repo transactions, by not requiring the quality of the cargo being financed to be independently checked, and that this negligent or reckless conduct precludes any claim. This defence is grounded in the express terms of a clause requiring the Bank to do "all things reasonably practicable to prevent any claim being made under this contract". The relevant issues are addressed in Section G below.
  20. Finally, the underwriters contend that the Bank acted unreasonably after the first default: this was a default by Euromar in August 2016. They contend that the Bank should have exercised swap options, entered new hedges and sold cargo more speedily. There was therefore a failure to "sue and labour". These issues are addressed in Section H below.
  21. A3: The trial

  22. It was anticipated, at the pre-trial review held in September 2020, that the trial would take place as a "hybrid" trial with limited numbers of legal representatives in court, but with other lawyers and interested parties having remote access. In the event, following the second national "lockdown" which came into effect shortly before the trial in consequence of the Covid-19 pandemic, the majority of the trial was fully remote using Zoom technology operated by Opus 2 who were also transcribing the proceedings and using their document management platform. Very few technical difficulties were encountered and counsel were able to cross-examine factual and expert witnesses satisfactorily. At the request of the Bank and the underwriters, closing submissions took place in court, with others joining remotely.
  23. Many factual and expert witnesses gave evidence at trial.
  24. There were three factual witnesses called by the Bank: Mr. Gijs Stroink, Ms. Marieke Franssen and Ms. Pauline van de Beek. Each of them was clearly an honest witness, and the contrary was not suggested. Their evidence was consistent with the contemporaneous documents, and generally speaking (as will be apparent from later sections of this judgment) I accept their evidence.
  25. A large number of individual underwriters gave evidence for each of the Defendants. Most of them had been involved, to a greater or lesser extent, in considering or writing the risk with which I am concerned. They were (using the abbreviated names which I will use in this judgment, and using the abbreviation D1 to mean the First Defendant etc): Mr. Brian Beattie, Mr. David Vaughan, Mr. Finlay Smith and Mr. Matthew Jones (RSA – D1); Mr. James Gaiger and Mr. Michael Giles (Navigators – D2); Mr. Graham McManus (Talbot – D3); Ms. Caroline Monnickendam (Brit – D4); Mr. Matthew Pullen (Hardy – D5); Mr. Frank Chu (Aegis – D6); Mr. Richard Burnett and Mr. Graham Williams (Markel – D7); Mr. James Blewett (Ark – D8); Mr. Michael Shillabeer (Channel – D9); Mr. Gary Cooke (Advent – D10); Mr. Massimo Orsini (Generali – D11); Mr. Nick Holding (Standard Syndicate – D12); Mr. David Burns ("Prosight" Syndicate 1110 – D13); and Mr. Thomas Butterworth (Swiss Re – D14). Accordingly, evidence was given on behalf of each of the Defendants, although there were three underwriters who were actively involved in the writing of the risk (Mr. James – D3; Mr. Tobin – D4; Mr. Andrews – D11) where attempts to obtain statements were unsuccessful.
  26. The principal factual issue concerned the discussions between Mr. Beattie and Mr. Mullen, giving rise to the rectification and related arguments, and I address the evidence of those witnesses in context in Section E. The challenges to the factual evidence of the other underwriting witnesses were relatively limited, and ultimately the central issues in the case do not turn on disputed factual evidence. Their factual evidence was principally relevant, as it seemed to me, to issues concerning non-disclosure and misrepresentation, and in particular what the reaction of underwriters would have been if Edge had explained the Bank's purpose and intention in seeking the TPC. But the focus of the arguments of the Bank and Edge was not on that issue. Indeed, in its closing submissions, the Bank said (consistent with the underwriters' case) that if Edge had told cargo underwriters that the Bank wanted credit risk cover, then on the "evidence before the court, the cargo underwriters would have said "No" (save perhaps Beattie)".
  27. In these circumstances, it is not necessary to discuss the evidence of each underwriting witness. Some of the witnesses were undoubtedly more argumentative, and gave their evidence in a more tendentious manner, than others. The evidence of Mr. Beattie is considered in detail in Section E. The evidence of some other witnesses is discussed in Section F below. Generally speaking, I am inclined to accept their evidence in so far as it concerned the approach which they took to the writing of the risk, and the question of what they would have done if different information had been given. In saying this, I leave aside Mr. Beattie whose evidence is considered in detail in Section E. I also leave aside Mr. Orsini, who had no real involvement in the writing of the risk and who in my view was not in a position to assist as to what the approach of the relevant underwriter, Mr. Andrews, had been. I will also separately address, in context, the evidence of Mr. Jones of RSA: he was not an underwriter, but was a senior claims person, and his evidence is of some relevance to the issues discussed in Section E.
  28. Edge called two factual witnesses: Mr. Mullen and Mr. Lee Lockyer. Mr. Mullen faced criticism from two directions. The issues concerning the conversations with Mr. Beattie meant that Mr. Parsons, on behalf of the underwriters, sought to establish that Mr. Mullen was not a reliable witness as to what had transpired. I address those issues in Section E. The issues relating to the Bank's claim against Edge meant that Ms. Sabben-Clare sought to establish that Mr. Mullen was, from beginning to end, incompetent. I consider that both of these attacks are best considered in context. I should say, however, that I thought that Mr. Mullen sought to give honest evidence to the best of his recollection, albeit that that recollection was often hazy, and it came from a man who is now 76 years old.
  29. As far as Mr. Lockyer is concerned, he was in my view a patently honest witness who sought to give accurate evidence as to the relevant events. Indeed, the Bank placed significant reliance on his evidence as to the approach which should have been taken in January 2016 after a query had been raised by Ms. Van de Beek.
  30. Expert evidence was given by experts in a number of disciplines, which can be summarised as:
  31. i) Underwriting: Mrs. Audrey Joyce Webb for the underwriters, and Mr. Geoff Sutherland for Edge;
    ii) Broking: Mr. Aidan Meldrum for the Bank, and Mr. Nigel Russell for Edge;
    iii) Coverage available in the credit risks market: Mr. Simon Hayter for the Bank and Mr. Nick Hedley for Edge;
    iv) Issues concerning quantum, the quality of the commodities (see Section G) and sue and labour (see Section H): Mr. Gordon MacLeod for the Bank; Mr. Angus Kerr and Ms. Catherine Jago for the underwriters.
  32. Their evidence is considered in context in the relevant sections of this judgment.
  33. B: The factual background

  34. This section contains an overview of the facts, as I find them to be, which are relevant to the issues which require resolution. Where issues require more detailed fact findings and conclusions, I address those in the context of my discussion of the particular issue.
  35. The business of Icestar

  36. Icestar provided collateralised finance to clients of the Bank by way of "repo" deals, which is a form of ownership-based stock financing for commodities held in approved warehouses in numerous locations around the world. The deals concerned a wide range of physical agricultural commodities (such as coffee, cocoa, and grains) and metals. Each deal had a tenor of three months but there were often "rollovers" of the financing. At each "rollover" date the amount of financing was reconsidered and approved by Icestar and as necessary a balancing or netting payment was made by the client in Icestar's favour, thereby effecting a repayment of part of the outstanding finance. The evidence of Ms. Franssen, who was closely involved on behalf of the Bank with the conclusion of Icestar's transactions, was that it was common for clients to extend their transactions in this way.
  37. Transmar and Euromar were amongst a number of clients of Icestar to whom collateralised finance was provided. As shown by an internal Bank credit paper (headed "Loan Syndications Advice") dated December 2015, prepared at a time when the Bank was considering the renewal of a secured revolving credit facility known as the "borrowing base" facility, Transmar had been a long-standing client of the Bank for over 10 years, dating back to the time of the Bank's corporate predecessor, Fortis. The document, which recommended the Bank's continued involvement as lender to Transmar, described Transmar as one of the leading mid-sized independent cocoa merchants. The relationship was described as robust and longstanding, and the financial performance of Transmar as solid. Transmar was, at that time, a major player in the world cocoa market. Their corporate group was founded in 1980. An annex to the December credit paper showed Transmar and its affiliates to constitute a global operation, offering services spanning the entire cocoa supply chain (including sourcing beans from origin, international traders, and/or affiliates; processing of beans and products; and melting). At that time, there had been no prior defaults under its lending arrangements, or any other financial problems.
  38. At the material times Icestar was separately incorporated as Icestar B.V., but at the beginning of 2018 it was formally merged into the Bank. The Bank's title to sue in respect of Icestar's rights under the Policy has been admitted by all the Defendants.
  39. The nature of the "repo" deals

  40. Ms. Franssen described in her evidence the three separate deal structures for its financing of cocoa products, known as "Icestar 1", "Icestar 2", and "Icestar 3". The present claim arises from Icestar 2 and 3 transactions.
  41. "Icestar 1" was used for cargoes that were exchange-deliverable, such as (in the case of cocoa) cocoa beans. In such cases, Icestar bought exchange-deliverable cocoa beans from its counterparty, such as Transmar, and simultaneously granted the counterparty a call option to repurchase the beans. The client had no obligation to repurchase the beans, just a "call option" enabling it to do so as and when parcels of the cargo were needed back (for example to sell to a chocolate manufacturer, or to process). If the call option was not exercised, Icestar was able to deliver the goods to the appropriate cocoa exchange in exchange for the agreed price. It is of the nature of commodities that are exchange-deliverable that quality certificates are required as a precondition to delivering to the exchange. Accordingly Icestar always insisted on the provision of quality certification in relation to Type 1 deals.
  42. These Icestar 1 transactions were hedged with futures contracts so that Icestar could deliver under the futures contract in the event that the counterparty did not exercise the call option. As is apparent from numerous cases in the Commercial Court, and as explained in the evidence of Ms. Catherine Jago, an expert witness called by the insurers, a hedging contract enables a party to protect against the risk of an adverse market movement. By purchasing beans under Icestar 1 transactions, the Bank would be exposed to adverse downward movements which might affect the value of the beans that it was holding. This "long" position could be protected by a matching futures contract on the relevant cocoa exchange for the sale of equivalent beans. A downward movement in the value of the physical beans, causing a potential loss to the Bank, would be matched by an equivalent gain under the hedge on the exchange where the Bank would be entitled to receive a price which was above the then market price.
  43. "Icestar 2" was the mechanism used to finance cocoa products or beans that were not exchange-deliverable. It gave rise to three contracts: (1) a purchase contract whereby Icestar purchased goods from its client; (2) a call option contract in favour of the client, and (3) a swap option in favour of Icestar. The effect of the swap option was that Icestar had the contractual right to require the client to replace the product or beans with exchange-deliverable beans, or to pay their equivalent value in cash. In effect, this was a right to require repurchase. The relevant transactions between Icestar and Transmar in the present case were of the Icestar 2 variety. In due course, Transmar defaulted when Icestar did exercise its swap options in December 2016. An issue raised by insurers, in relation to "sue and labour", is whether these swap options should have been exercised at an earlier stage, in August 2016, and if so what the consequence would have been: see Section H below.
  44. It was a feature of these Icestar 2 transactions that they were also hedged with futures contracts. This was done by Icestar taking over the counterparty's own futures position through an "Exchange for Physical" or "EFP" at the moment of purchase by Icestar. That hedge would then be transferred back to the counterparty through an EFP at the same time as the repurchase.
  45. "Icestar 3" was a structure used where there was a tripartite arrangement involving a client of the Bank and another counterparty. It was this structure that was used for the transactions involving Euromar. Euromar was a corporate affiliate of Transmar, and operated a cocoa processing facility in Germany. In this structure, as far as Euromar and Transmar are concerned, the Bank purchased goods from Transmar, and simultaneously entered into a forward sale contract with Euromar. This forward contract required Euromar to repurchase at the end of the financing period for a fixed price. This repurchase price was fixed in advance, because no hedges had been entered into by Euromar. Euromar would have the benefit of a call option, and this permitted it to repurchase quantities of the commodity before the agreed forward sale date. All of the Euromar deals in the present case were Icestar 3 deals.
  46. Icestar's insurance coverage: 2008-2014

  47. Since Icestar acquired ownership of goods as a result of the transactions described above, it was natural and necessary for insurance coverage to be purchased in respect of those goods. Mr. Stroink was involved in making Icestar's insurance arrangements for many years stretching back to at least 2008. Since that time, Icestar's coverage had been placed by the 15th Defendant, Edge, or its predecessor called London Special Risks. After 2010, Mr. David Mullen was the individual responsible for handling the Bank's account. Mr. Mullen was an experienced cargo broker who, by 2016 (when he was aged 71/72), had nearly 50 years' experience as an insurance broker in the London market.
  48. It is not necessary to describe the history of these placements or the precise terms of the cover obtained during these years. Some aspects of that history and those terms were explored in the cross-examination of the witnesses. For example, Ms. Sabben-Clare on behalf of the Bank sought to demonstrate that aspects of Mr. Mullen's performance during those years were inadequate. I agree that there were inadequacies, but it is not necessary to consider them in detail. What really matters in the present case are the instructions which Mr. Mullen received in 2015 to obtain amendments to the policy then in force, and the instructions received in 2016 on renewal, the events which flowed from those instructions and any inadequacies in Mr. Mullen's work in connection therewith.
  49. The 2015 policy

  50. The slip policy which covered Icestar (as well as the Bank) was renewed by insurers in January 2015. The policy was led, as it had been for many years, by Mr. Brian Beattie on behalf of RSA. The policy was subscribed by most, but not all, of the present defendants. The cover provided by the 2015 policy included what might be described as the usual cover against physical loss or damage to goods (and referred to at trial by the acronym "PLOD"). It also included elements of coverage which did not require PLOD. The particular elements of such coverage (referred to at trial as "non-PLOD") were as follows.
  51. First, Clause 20 of Section 2 of the Policy provided coverage under a heading "Business contingency cover". This provided:
  52. "20.1 Cover against costs, expenses and losses, including consequential loss resulting from an event beyond the control of the Insured, including grading delays, which prevents or substantially delays the Insured from delivering the Subject Matter Insured to the Exchange in the related Futures Month (as defined below) for any reason other than for physical loss or damage to the Subject Matter Insured caused by a peril outside the control of the licensed warehouse keeper.
    Futures month shall mean the futures month as defined in the purchase contract by the Insured.
    …
    20.3 Any declaration under this clause 20 is to be recorded in accordance with the relevant contract terms and conditions and the Premium is to be calculated at 40% of cover rates.
    20.4 Subject to individual declarations prior to Attachment of risk and subject to no previous losses which would be collectable under this contract."
  53. It is not necessary to discuss this provision in detail. The cover under this clause would in principle extend to cases of delay in delivery to the exchange "for any reason other than physical loss or damage to the Subject Matter Insured". A typical case might be a strike or other delay at a port. However, the cover would only become available if there were "individual declarations". In fact, as the insurers pointed out, no such declarations were made, and no premium was paid.
  54. Secondly, Clause 17 of Section 2 of the 2015 policy provided cover under the heading "Fraudulent documentation". This included an indemnity, under Clause 17.3, for:
  55. "… direct financial loss suffered by the Insured, in good faith during the Period and in the ordinary course of business acting upon a "Counterfeit Document" (as defined below) … subject to a limit of USD 5,000,000, for any one event or loss during the Period."
  56. Thirdly, Section 3 of the 2015 policy was headed: "Additional Cover Confiscation and Expropriation". This type of cover is referred to by the acronym "CEND". This provided cover for
  57. "loss of and/or damage to the Subject Matter Insured directly caused by confiscation, moratorium, seizure, appropriation, expropriation, requisition, deprivation, requisition for title or use or wilful destruction by/or under the order of any government …"
    Deprivation was defined so as to include a loss of use or possession caused by the failure or refusal of a foreign government for a period of three months to permit the export of the subject matter insured from the foreign country.
  58. There was some debate at trial, in the evidence of the underwriting witnesses, as to whether these categories of coverage were, or at least were analogous to, PLOD. In my view, each of these categories of cover did extend beyond what would ordinarily be regarded as physical loss and damage to cargo. However, it is equally plain that none of these clauses provided coverage akin to coverage for losses arising from the default of the Bank's counterparties.
  59. The 2015 slip prepared by Edge, and agreed by those underwriters who subscribed to that slip, contained in its concluding pages a section headed: "Subscription Agreement Section". This section set out agreements between the subscribing underwriters in relation to contract changes, claims handling and other matters. These terms are material to the question of whether the following market (ie those subscribers who subscribed as "followers" to Mr. Beattie/ RSA's lead) were bound by any changes agreed by Mr. Beattie. The relevant provisions were as follows:
  60. SLIP LEADER: Royal Sun & Alliance [sic]
    BASIS OF AGREEMENT Subject to GUA October, 2001 with Marine Cargo Schedule 2003
    TO CONTRACT CHANGES Slip leader only to agree part two changes
    OTHER AGREEMENT
    PARTIES FOR
    CONTRACT CHANGES,
    FOR PART 2 GUA CHANGES
    ONLY
    AGREEMENT Slip leader to agree all contract changes
    PARTIES FOR
    CONTRACT CHANGES,
    FOR THEIR PROPORTION
    ONLY:

    February – 17 June 2015

  61. Subsequent to the renewal of the policy with effect from 1 February 2015, Mr. Stroink discussed the terms of the policy with the Bank's legal department and risk department in February 2015. The Bank had decided to conduct an in-depth analysis or, as Mr. Stroink described it, a "deep dive" of Icestar's set-up, products and procedures. During the course of the prior year, there had been certain discussions between the Bank and Mr. Mullen, and the latter had given advice as to the scope of the Bank's coverage. There was uncertainty within the Bank as to whether the policy met its needs, and Mr. Stroink was keen to ensure that the Bank had adequate protection against the risks to which Icestar was exposed in the course of its repo transactions.
  62. It is unnecessary to describe the internal discussions in detail. The Bank's thinking evolved over the next few months, and their needs were ultimately reflected in a draft endorsement to the policy which was given to Mr. Mullen in June 2015 and in subsequent instructions and information given to Mr. Mullen. It is therefore sufficient to describe, in outline, the following developments.
  63. On 26 February 2015, Ms. Lawar Ishak sent Mr. Mullen some proposed wording for what she wanted to include in the policy via, as she described it, a "Declaration". Ms. Ishak was an in-house lawyer working for the Bank in London, but she was not (as Mr. Mullen knew) an insurance specialist. The wording proposed by Ms. Ishak was, promptly thereafter, put by Mr. Mullen into a formal endorsement which he presented to Mr. Beattie on 26 February 2015 as Endorsement No. 3. It was scratched by Mr. Beattie on that day. It provided:
  64. "Underwriters note and agree that for the purposes of the Business Contingency Cover, it is hereby confirmed that the Insured is covered under such cover for all costs and expenses incurred by the Bank as set out in that clause, including but not limited to any market premium paid by the Insured under the relevant transaction ("market premium", being the difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents)".
  65. In contrast to the document later signed by Mr. Beattie in July 2015, this endorsement was in a standard format, containing the Unique Market Reference and clearly identifying itself as Endorsement No. 3. The endorsement also contained, at the foot of the second page, a standard "box" referred to at trial as the GUA box: GUA being the General Underwriters Agreement referred to in the "Subscription Agreement" section of the slip described above. The standard "box" is as follows:
  66. GENERAL UNDERWRITERS AGREEMENT (GUA)
    Each underwriter's proportion is several and not joint

    Slip Leader Only

    Slip Leader and Agreement Parties

    All Underwriters

     

     

     

  67. Mr. Beattie had initialled the part of the box saying "Slip Leader Only", thereby intimating that he did not consider that any of the following market needed to sign, or even see, the endorsement.
  68. On 19 March, following communications between Mr. Mullen and Ms. Ishak, Mr. Paul Power of RSA signed a further endorsement, number 4. This too had been drafted by Ms. Ishak. The language was in many respects similar to that of Endorsement No. 3:
  69. "Notwithstanding any other term to the contrary in this policy, it is hereby confirmed that the Business Contingency Policy also covers (in addition to the cover set out therein) any loss suffered by the Insured representing any market premium paid by the Insured under a transaction ("market premium") being the difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents".
  70. The documentary evidence, supported by Mr. Stroink's evidence, indicates to me that both Mr. Stroink and Ms. Ishak were at this stage under the impression, as a result of the communications which led to these endorsements as well as communications in the prior year, that Icestar was covered for losses arising purely from a shortfall in the amounts realised on a sale of the cargo; ie irrespective of whether the cargo in question had suffered physical loss or damage. The two endorsements drafted by Ms. Ishak were her attempt to capture this point. However, the Bank's risk manager, Jan Pieter Rogaar, did not altogether agree with points that Ms. Ishak was making in internal e-mails as to the nature of the coverage. He said that the policy was a "quite unclear document". He therefore proposed that "external counsel (expert on insurances)" should be consulted.
  71. Accordingly, on 24 March, Ms. Ishak sent a detailed e-mail to Norton Rose Fulbright ("NRF"), setting out the scope of the work that they wished NRF to carry out. One of the issues which NRF was asked to consider was whether:
  72. " … if there is no damage/fraud/theft, and the transaction runs with complying contracts, BUT the market premium has "vanished" o[r] "reduced", would then the insurance cover Icestar for that (transactional) loss?"
  73. On 30 April 2015, NRF issued a detailed draft advice. They expressed the view that Endorsement 3 did not cover what they described as the general "market premium" situation raised by Icestar "where the client elects or is not able to re-purchase the commodity, leaving Icestar with no option but to deliver to the exchange under a short forward hedge or liquidate on the open market at a discount to the purchase price". They were more confident that Endorsement 4 covered the general "market premium" situation, but they pointed out that there were aspects of Endorsement 4 which required clarification. These included the point that Endorsement 4 did not "expressly make the critical point that the cover provided under Endorsement 4 is not contingent on physical loss or damage to the commodity".
  74. NRF was then asked to prepare an endorsement to the policy in order to address the various issues which they had discussed in their draft letter of advice. These issues went beyond the "market premium" issue. The work on drafting the endorsement was carried out by Mr. Charles Weston-Simons. In a statement served at a time when NRF were party to the present proceedings (having been joined by Edge), Mr. Weston-Simons described the drafting of the definition of "Default" contained in the TPC. He said that whilst he did not recall his thought processes when drafting the Endorsement, he believed that:
  75. "I would have considered that the fact that a Default could be "for whatever reason" meant precisely that, and was not therefore restricted to physical loss of, or damage to, the commodity. It would have followed that, in my view, the wording encompassed not only physical loss or damage but also loss caused by non-physical damage or loss, and therefore covered what we had described in the April Advice as "the critical point".
  76. The draft endorsement was sent to Mr. Stroink and Ms. Ishak (by now married with the name Lawar Barnes) on 17 June 2015. The endorsement contained what ultimately became, with immaterial alterations, the Transaction Premium Clause. It also contained a large number of other proposed alterations to the policy. Since these, together with the clause which eventually became the TPC, were all included in the document signed by Mr. Beattie in July 2015, I will describe those proposed alterations in that context.
  77. June 2015: meeting between Mr. Mullen and the Bank and subsequent correspondence

  78. On 18 or 19 June 2015, Mr. Stroink and Ms. Barnes met with Mr. Mullen and showed him the draft endorsement that NRF had prepared. Mr. Stroink wanted the amendments recommended by NRF to be made to the policy. Mr. Stroink remembered saying that the changes were to make clear what was covered. He did not recall any specific discussion of the TPC wording, but he thought that he "probably would have said that we required cover for any loss we suffered if a client did not repurchase a cargo". I think that it is likely that Mr. Stroink did say something along those lines.
  79. Mr. Mullen's written witness statement relating to the meeting was supportive of Mr. Stroink's evidence, and indeed in some respects put the position more forcefully. He said:
  80. "I recall that there being some discussion, prompted by Gijs, to the effect that Icestar required cover against a customer defaulting on its obligations to make payment in respect of the particular commodity. What Gijs wanted was to insure against the risk of the customer failing to repurchase. In effect, the TPC was to expressly incorporate credit risk cover into the 2015 Cargo Policy. That cover was not to be contingent on physical loss or damage to the cargo".
  81. When cross-examined about this passage by Mr. Parsons on behalf of underwriters, however, Mr. Mullen agreed that the Bank had not told him, in terms, that they wanted "trade credit insurance". He couldn't remember the words "credit risk" being mentioned. He said that he came away from the meeting with instructions to incorporate certain requests into the policy, but not specifically trade credit insurance. He had the impression that Icestar's competitors already had or were looking for cover which was similar to the TPC. He said that the meeting only lasted around 30 minutes and, later in his evidence, said (unsurprisingly) that the meeting was many years ago and that he could not remember what Mr. Stroink had said. In my view, nothing turns on the question of what exactly was said at the meeting, although (if it were necessary to do so) I would prefer the evidence of Mr. Stroink.
  82. Mr. Mullen said that he was not aware that the endorsement was the product of the work carried out by external lawyers: he had no knowledge of their involvement at the material times. He did, however, understand that it had been drafted by the Bank's lawyers. His assumption was that it was Ms. Barnes who had drafted it. His evidence indicated that he thought that Ms. Barnes' drafting was often over-complicated, and he knew that she was not an insurance specialist.
  83. It is clear from the subsequent e-mail correspondence that Mr. Mullen did not study the terms of the draft endorsement at the time of the June meeting with the Bank, and it is improbable that there was any detailed discussion about it at the meeting. Ms. Barnes' email to NRF on 19 June described the meeting with Mr. Mullen as brief, and indicated that Mr. Mullen had yet to review the document. Mr. Mullen was travelling the following week, but had said that he "would take a look with his lawyer asap".
  84. On 24 June, Mr. Stroink asked Mr. Mullen to let him know the status of the "endorsement/amendments" to the wording of the policy. In his response of 25 June, Mr. Mullen said that he had:
  85. "… reviewed the document "wish list" and whilst do not feel that there is anything that is covered already [sic], the language for amendments needs to be discussed with the lead Underwriter point by point. He is on vacation until Tuesday of next week and I would rather talk to him than leave it to any deputy. also because of our regulations this endorsement will require the signatures of all Underwriters. Can you bear with me for a few days to make sure we have the required signatures from the market Underwriters? Thanks in advance."
  86. I do not consider that Mr. Mullen had reviewed the endorsement in anything more than a cursory fashion at this stage. His later e-mail of 8 July indicates that it was only around then that he did so. Had he carried out a detailed review at this stage, then he could not have said that the endorsement did not contain anything that was not already covered. It is true that his email did not actually say this: it said "do not feel that there is anything that is covered already". However, in context, it seems clear that the word "not" must have been accidentally omitted.
  87. Three other features of this e-mail should be noted. First, Mr. Mullen described the draft endorsement as a "wish list". This phrase seems to have originated with Mr. Lee Lockyer, the more junior colleague with whom Mr. Mullen worked, on 24 June. On that day, Mr. Lockyer sent Mr. Mullen an e-mail which said: "Please see the attached "wish list" of Icestar in PDF and Word format". This followed a request by Mr. Mullen to Ms. Barnes for a Word version of the document that he had been provided, because he was having difficulty with the version that he had scanned into his computer. Mr. Beattie's evidence was that the expression, "wish list" was used by Mr. Mullen when they met in July 2015. I shall return to the relevance of this when discussing the evidence as to the meeting.
  88. What is clear, however, is that Mr. Mullen understood that the Bank was seeking a contractually binding change to the terms of the existing policy. The e-mail of 25 June thus refers to the "document "wish list"…", but also to an "endorsement" which would require the signature of all underwriters. Mr. Mullen was not therefore distinguishing in his own mind between a "wish list" and a contractual endorsement, and in my view he cannot be criticised in that respect. A list of terms which a party wishes to include in a variation of a contract can easily and properly be described as a "wish list". That expression does not denote that the list of clauses requested is something other than contractual.
  89. Secondly, Mr Mullen's e-mail indicated that the document would need to be discussed with the lead Underwriter "point by point". An important question, relating to the July meeting discussed in Section E, is whether this is what actually happened. However, it is clear that Mr. Mullen anticipated that it would or at least might happen. This was a reason why Mr. Mullen did subsequently (as shown by his e-mail to the Bank on 9 July 2015) go through the document in detail, asking pertinent questions relating to many of the proposed amendments.
  90. Thirdly, Mr. Mullen indicated that "our regulations" required the endorsement to be signed by all underwriters; ie not simply the lead underwriter. The reference to regulations is obscure, and it is doubtful whether this was a specific reference to the GUA (which are not regulations which apply to Edge). It is probable that Mr. Mullen was taking the view, without any real analysis, that the proposed endorsement was one which all underwriters would be asked to sign. He was also, sensibly, seeking to lower the client's expectations as to the speed with which agreement on the endorsement could be accomplished.
  91. The documentary evidence indicates that Mr. Mullen did not at that stage move things forward. Mr. Beattie was due to return on 30 June (according to Mr. Mullen's email of 25 June), but there is no evidence of any work being carried out by Mr. Mullen until around 8 July.
  92. 8 – 25 July 2015

  93. On the morning of 8 July, at 08.26, Mr. Mullen sent an email to his colleague Lee Lockyer. The e-mail referred to "our review of the suggestions made by ABN recently". He said that:
  94. "before I think of meeting with Underwriters I wanted to run by you the areas which I think could be of concern to Underwriters"
  95. The email then addressed a number of issues. The final point was a reference to the TPC:
  96. "Page 4.Section 2. Underwriters to agree that in respect of any transaction it is confirmed that the Insured is covered under this contract for transaction premium that the insured would otherwise have received and or earned in the absence of a default on the part of the Insured's client (do not understand this – have you seen this before Lee?)
    I will be in later – let me know if you have any thoughts."
  97. At a very late stage in the case (after Ms. Sabben-Clare had delivered her closing argument, but before Mr. Parsons had started his), Edge disclosed a further document which had been created on 8 July at 12.23 by Mr. Lockyer. This document had, for reasons which it is unnecessary to describe in detail, been accidentally missed during the disclosure process. Its disclosure also led to what was ultimately an adjournment of Mr. Parsons' closing argument. Supplemental witness statements were provided both by Mr. Mullen and Mr. Lockyer which addressed this document. In the event, they were not required to attend for further cross-examination. Mr. Mullen's evidence was that the document seemed to be an aide-memoire prepared by Mr. Lockyer. But the document did not look familiar to him and he could not recall seeing it prior to December 2020, nor having discussed with Mr. Lockyer any of the matters in the document. It rang no bells at all.
  98. Mr. Lockyer's evidence was that he could see that he created the document at 12.23, but could not recall why. He thought it may have been an aide-memoire for his personal use, so that he could discuss with Mr. Mullen the questions which he had asked that morning. This was, he said, his attempt at reconstructing why he would have created the document. He could not recall discussing the document with Mr. Mullen or anyone else, nor sending it to anyone. He did not believe that he ever showed the document to anyone else.
  99. I see no reason to doubt Mr. Lockyer's evidence. It is in my view most likely that he drafted the document in order to gather his thoughts in the light of Mr. Mullen's questions that morning. The significance of the document is that it shows that Mr. Lockyer did give careful consideration to the changes which were being proposed by the Bank. His document created at 12.23 on 8 July 2015, which was headed "ABN AMRO amendments/ alterations", shows that he was working through the proposed endorsement in a methodical way, and recording his thoughts and understanding. It also does show that Mr. Lockyer did understand the effect of the TPC. Thus, his document records:
  100. Page 4. Point 2 – Effectively this clause will allow the bank to seek indemnity, without any physical loss or damage, as a result of their client defaulting on a transaction and the bank endeavouring to mitigate their financial position by tendering the subject matter insured to the exchange or third party. Any difference in price will be made whole by Underwriters.
  101. After the late disclosure of this document, Mr. Parsons indicated that he wished to consider with his clients whether or not they should seek permission to amend to plead a case of fraud. In the event, the underwriters decided not to do so. But the reason why this possibility was raised was that Mr. Lockyer's document was some evidence that at that time he, at least, clearly understood the effect of the TPC. If Mr. Mullen also understood it in the same way, then it would suggest that Edge did indeed understand the effect of the TPC, and that Mr. Mullen was not confused or (as Ms. Sabben-Clare had argued) "lost in fog". Mr. Parsons therefore wished to consider whether Mr. Mullen's failure to tell Mr. Beattie about the effect of the clause was dishonest, although in the event that argument was not pursued.
  102. In my judgment, Mr. Lockyer's document is relevant, because it does show a clear understanding on his part, at the time, of what the TPC was intended to achieve. It is inherently probable that this understanding was communicated to Mr. Mullen at the time. Mr. Mullen then raised various questions with the Bank, as described below. This included a question concerning the TPC, as to which he received an answer consistent with what Mr. Lockyer had said.
  103. On the following day, 9 July, Mr. Mullen sent Ms. Barnes a 1-page e-mail. The subject-line of the e-mail was "Endorsement No 5". The email began:
  104. "With reference to the above – and as discussed very preliminary – before we submit these to Underwriters, could we make the following observations please, and perhaps for further discussion either in London or the Netherlands."
  105. The e-mail then referred, in turn, to each of the proposed amendments to the policy. In relation to some of these, Mr. Mullen raised no queries by saying either "Fine no difficulty" or "Accepted". In relation to others, however, he raised questions which were pertinent. In particular, he asked about the non-avoidance clause which, in the proposed amendments, had three sub-paragraphs (a) (b) and (c):
  106. "Could you please explain your requirements in respect (a)-(b)-(c)".
  107. He also asked about what became the TPC. At this stage, the clause in the draft endorsement had not been given that nomenclature, but its text appeared under the following words which were underlined and in bold: "The contract change made under Endorsement 4 is deleted and replaced with the following clause". Mr. Mullen's question was:
  108. "Contract Extension under Endorsement 4. Could you please explain the relevance of the "transaction premium" contained under 1-2-3"
    The reference to "1-2-3" was because the draft clause had three paragraphs with those numbers.
  109. On the following day, 10 July, Ms. Barnes responded to Mr. Mullen's email under cover of an e-mail in which she said that she had taken "another look at our legal opinion on the insurance in order to answer your questions below". She also said that if Mr. Mullen needed any "commercial input", they would need Mr. Stroink for that and he was away, back on 20 July. Ms. Barnes then set out, in red, responses to each of the points where Mr. Mullen had raised an issue. Her responses in relation to the questions about the non-avoidance clause and (what became) the TPC were as follows:
  110. "We want to include this clause to prevent the underwriters from avoiding the policy (ie treating the policy as if it never existed) if the underwriters consider there has been a material non-disclosure prior to inception. Is it possible to include this clause?
    …
    We can discuss this over the phone. The point is that we are expressly stating that market premium loss is not contingent on physical loss or damage to the commodity, and that the additional cover under endorsement 4 is not only limited to the circumstances considered under clause 20 (business contingency)".
  111. It is clear from this e-mail that the Bank was indeed seeking coverage for a contingency such as that which ultimately occurred; ie a loss of what she called the "market premium", but which was defined in the clause as the "Transaction Premium", which occurred in the absence of physical loss or damage. Indeed, Edge's case at the trial accepted that this is what the Bank was indeed seeking to obtain, and also what Mr. Mullen was seeking to place. And although there was some suggestion in the cross-examination of Mr. Stroink that the Bank had in mind a more limited concept of "market premium" – ie more limited than is clear from the terms of the clause – I do not accept that this was so.
  112. I also consider that in the light of (i) the terms of Mr. Lockyer's aide-memoire, which he is likely to have discussed with Mr. Mullen prior to the e-mail sent to Ms. Barnes on 9 July, and (ii) Ms. Barnes' response to the question raised about the TPC, it is more likely than not that Mr. Mullen did at that time understand the intended effect of the TPC. It is fair to say that Mr. Mullen's oral evidence at trial, in cross-examination over a number of days, indicated a fair amount of woolly thinking about the TPC. For example, he sometimes used the word "delay" when it seemed, perhaps, that he was intending to refer to "default". However, I have to bear in mind that Mr. Mullen is now 76, and is seeking to recall events, and was being asked about his thought-processes, many years ago. These two contemporaneous documents, which both squarely address the effect of the TPC, provide evidence which assists Edge's case that it was understood at the time.
  113. Mr. Mullen responded on 12 July, thanking Ms. Barnes for her answers and indicating that he was out of the office until Thursday (ie 16 July), but would start drafting up a "formal endorsement based on your answers".
  114. As at 12 July, Mr. Mullen had not actually seen Mr. Beattie. The underwriters suggested that the reason why Mr. Mullen had not seen Mr. Beattie by this stage was that he realised that it would be difficult to obtain his agreement to the amendments which the Bank was seeking. I do not consider that the evidence supports this. Mr. Mullen's 8 July e-mail indicates that it was only at around this time that Mr. Mullen applied his mind to the detailed terms of the proposed endorsement. The evidence does not support a finding that this was because of some problem which Mr. Mullen recognised. The more likely explanation is that Mr. Mullen was doing other things, and had not got round to examining the endorsement in detail. Perhaps this was because he had thought that the addendum would take him some time to consider and understand (people do sometimes put off things which may be difficult), but I do not consider it necessary to speculate further as to whether this was in fact the reason.
  115. In fact, Mr. Mullen did not begin the drafting work on Thursday 16 July, but did so on the following Monday, 20 July. It is clear, both from his e-mail of 12 July and from what he subsequently did, that he did have in mind that the Bank's desired changes should take the form of a formal contract endorsement similar to the form of prior endorsements such as Endorsements 3 and 4. The underwriters' case, at least originally, was that the (alleged) non-binding nature of the document which Mr. Beattie scratched on 28 July 2015 was evidenced by the fact that this was not prepared as a formal endorsement, as it could have been. Mr. Mullen's oral evidence in cross-examination was that he had originally started to use a template for formal endorsements which was on the Edge computer system. However, he was insufficiently computer-literate to be able to incorporate the lengthy text of the Bank's proposed endorsement into the standard template. He had therefore decided that he would use an ordinary "Word" document. This evidence led to a request for disclosure, whilst Mr. Mullen was giving evidence, of the relevant computer files which might evidence this work. Disclosure provided by Edge overnight supported Mr. Mullen's account. The 'metadata' for the relevant documents showed that Mr. Mullen had worked on a proposed endorsement, using the standard template, on the morning of 20 July, but had stopped at 11.49 am and had then created a separate Word document. The metadata for the Word document shows that its last modification was on (Thursday) 23 July, although the evidence did not indicate what that modification was.
  116. On the afternoon of the day (20 July) when Mr. Mullen had worked on the endorsement, he sent a message to Ms. Barnes and Mr. Stroink:
  117. "With reference to the requested amendments to the contract wording, we apologise for the delay in replying but our lead Underwriters has been away and only returned late last week. We are in the process of discussing your suggestions and will revert very soon".
  118. Mr. Mullen's statement that he was in the process of discussing the suggestions was not accurate and was in my view, as Mr. Mullen accepted in relation to a later document, a broker "egging it". He said that he wanted a bit of "thinking time": the Bank was not his only client, but it was one of the most important clients and he "wanted to make sure that I got it right".
  119. 28 July 2015

  120. On the morning of 28 July, Mr. Mullen received a chaser from Ms. Barnes. His reply, sent at 8.33 London time, was:
  121. "Nearly finished, Lawar – Brian was away again. Will complete this week".
  122. At some point after sending this e-mail, Mr. Mullen made the relatively short walk to see Mr. Beattie in his office. By 10.29 that morning, he was back in his office with a document, signed by Mr. Beattie at the foot of each page, and with the RSA stamp alongside his signature. The document was not amended, and Mr. Beattie's signature was not qualified by words such as "for receipt only". Mr. Mullen reported to Ms. Barnes and Mr. Stroink in his e-mail sent at 10.29:
  123. "My apologies again for the time it has taken to complete the review and endorsement. Bits that needed no discussion and others that needed some explanation. However pleased to confirm all agreed by Underwriters and attached please find the initialled endorsement which we trust is suitable for your purposes. Our official endorsement will be prepared and sent to you shortly. Thanks for your patience and help in this matter".
  124. The signed document was attached to Mr. Mullen's e-mail. A copy of the signed endorsement was also placed on Mr. Beattie's files. Later that day, Mr. Mullen sent "our official endorsement" as he had indicated that he would. This document was headed: "ENDORSEMENT ATTACHING TO AND FORMING PART OF COVER NOTE NO. QM 349770". It was therefore an endorsement to Edge's Cover Note, and was signed by Mr. Mullen on the final page. The endorsement contained the text of the document signed by Mr. Beattie.
  125. It is common ground that Mr. Mullen did not take any steps to show the document signed by Mr. Beattie to any of the following market. One issue in the case is whether, even if RSA via Mr. Beattie was bound by the endorsement, the following market was bound by that document. That depends, at least principally, upon the effect of the General Underwriters Agreement or GUA. Ultimately, however, the resolution of the issue as to whether RSA or the following market were bound by the 2015 document is not determinative, or even central, to the issues in the case. No claim was ever made under the 2015/2016 policy. The claim was made under the policy which was put in place for the 2016/2017 year in January and February 2016, and that policy incorporated all the terms of the document signed by Mr. Beattie in July.
  126. I shall return to these issues, and the dispute as to what was said at the 28 July meeting and its significance, in Section E below. That meeting was the first of three occasions when Mr. Beattie's evidence was that he explained to Mr. Mullen his understanding of the TPC. In my description of the events which followed, I shall refer to the document signed by Mr. Beattie as the July endorsement, whilst recognising that there is an issue, which I will address below, as to whether it was a non-contractual "wish list" or a document containing contractually agreed terms.
  127. September 2015

  128. The Bank asked NRF to advise whether the endorsement had filled the gaps identified in their original advice. NRF provided a draft and later, in early September, a final opinion on the policy as (as the Bank and NRF understood it) amended by the endorsement agreed in July 2015. Their advice was that the endorsement should provide coverage in respect of the issues and potential gaps in cover which had been identified in their April 2015 letter. They said:
  129. 2.2 In particular, we consider that the effect of the Agreed Endorsement is that the Policy should provide cover in respect of the following matters:
    (a) "Quality deficiency" (i.e. when it comes to re-selling a particular commodity on the exchange or in the open market, it is discovered that the physical condition of the commodity is inadequate)
    ● We consider that the Policy should (in principle) respond if the physical condition of the commodity has deteriorated in the period between Icestar purchasing the commodity and re-selling it.
    ● Based on the "Proof of goods" clause and Section 2, Clause 17.1 of the Agreed Endorsement, we also believe that Icestar should (in principle) be able to recover its losses if the physical condition of the commodity was impaired when Icestar purchased it.
    (b) "Market premium" (i.e. the difference between the pre-agreed purchase or repurchase price agreed between Icestar and its client for a particular commodity, and the actual sale price if it becomes necessary to sell the commodity on an exchange or to a third party on the open market)
    ● We consider that the new Endorsement 4 should (in principle) provide cover in respect of this eventuality. (Under the Agreed Endorsement a new Endorsement 4 replaces the old Endorsement 4, which is deleted).
  130. The Bank's internal documents show, unsurprisingly, that it understood that the gaps had indeed been filled.
  131. January 2016 – the renewal and commitment of underwriters between 25 and 29 January

  132. The policy fell due for renewal at the end of January 2016. On 12 January, Mr. Stroink and his colleague Pauline van de Beek (who had joined Icestar in August 2015) met Mr. Mullen and Mr. Lockyer in Amsterdam to discuss the renewal. Ms. Barnes joined the meeting by telephone. There was no specific discussion regarding the TPC, because that had been dealt with in July. The Bank was looking to make some changes to other aspects of the policy. In particular, the Bank was seeking an increase in the contractual limits for any one individual warehouse or storage facility in respect of oil. Under the expiring policy, the limit was USD 100 million, but the Bank now wished to have a limit of USD 250 million. The Bank was seeking the removal of the USD 5 million sub-limit under the coverage (in Clause 17.3 of the expiring policy) for loss suffered in consequence of reliance on fraudulent documents.
  133. Prior to seeing any underwriters, Mr. Lockyer set to work on incorporating the amendments agreed in the prior year into new policy wording. Both Mr. Lockyer and Mr. Mullen understood, of course, that those amendments included those contained in the July 2015 endorsement signed by Mr. Beattie. On 18 January 2016, Mr. Mullen sent the Bank "amended policy wording which includes the various amendments and additions made during the last 12 months". The amended clauses were highlighted in yellow.
  134. It was Mr. Lockyer who gave consideration to where the TPC should go. He considered that it was appropriate to put it in Section 2 of the policy, headed "General Conditions", followed by words which stated that the general conditions applied to all sections of the contract. He then included the TPC as Clause 1.5 which was preceded by the heading "Definitions". I do not need to set out, at this stage, the terms of the policy, drafted by Mr. Lockyer and then sent to the Bank on 18 January: they are materially identical to the terms to which various underwriters, including Mr. Beattie, subscribed between 20 and 27 January. Those terms are set out in Section C below.
  135. Mr. Lockyer's decision to place the TPC after four clauses (1.1 to 1.4) containing definitions (of "Exchange", "Exchange Approved Storage Location", "Insured Approved Storage Location" and "Underwriters") is curious, and Mr. Lockyer's evidence did not provide a clear explanation or rationale. It is possible that this placement was because the TPC itself contained various definitions which had been highlighted in bold in the July endorsement. It is likely that Mr. Lockyer considered that he was embarking upon a somewhat mechanical exercise of transferring agreed terms into the new policy, and may have given little thought, at that stage, to the effect of the particular clauses. That this is so is confirmed by his inclusion, at Clause 1.6, of another clause (which had been included in the July endorsement) which was not, and did not contain, any definition: see Section C below.
  136. On 20 January 2016, Mr. Mullen and Mr. Lockyer went to see Mr. Beattie with the renewal slip. Except for the correction of some typographical errors (which the Bank had pointed out), and the inclusion of a "Missing Goods" clause, the renewal slip was the same as that which Mr. Mullen had sent to the Bank on 18 January. The TPC was therefore at Clause 1.5 of Section 2. Mr. Beattie scratched the slip on that day, signing it on the front page and also against the clause which increased the limits for oil (to USD 250 million), and against the "Fraudulent documentation" clause, where the USD 5 million sub-limit had been removed. Mr. Beattie's notes on the RSA file indicate that the increased limit to USD 250 million was discussed. Mr. Beattie gave an indication that his 25% line would stand; ie that he was in principle prepared to renew a 25% line. There was nothing in the renewal slip scratched by Mr. Beattie, or in Mr. Beattie's notes on the RSA file, which indicated that there was any discussion of the TPC or that any queries had been raised about it.
  137. Mr. Mullen was not at this stage looking for a binding agreement from Mr. Beattie. Mr. Beattie described this slip as a "quotation slip", meaning that it was a first run at a proposed renewal. His signature indicated that the wording of the slip was in principle acceptable to him. He did not therefore, at that stage, formally put down (or "enter") his line in accordance with the practice in the London market. However, he understood that Mr. Mullen would, having obtained Mr. Beattie's support, go to other underwriters and try to obtain their support in following Mr. Beattie's lead.
  138. Following that meeting, Mr. Mullen wrote an e-mail to his clients as follows:
  139. "Attached is the proposed new wording which incorporates the changes you mention when we were in your office last week. I hope you didn't mind but I have taken the liberty of having a full discussion with the lead Underwriter and make sure that we have no hidden surprises.
    The Underwriters have agreed to amend the "fraudulent document" wording with no sub limit. What the Underwriters have asked is a brief outline of your collateral management structure and due diligence procedures – we could discuss this next week. (Page 18 on the proposal)
    Underwriters have agreed the Missing Goods wording ( subject to your approval) on page 9.
    Underwriters will agree to increase the limit on the energy-oil- section to USD250,000,000. In doing so Underwriters will have to reserve this capacity on a product that is currently trading very sparingly. Whilst this can be achieved I will need to introduce new markets to the increase the capacity of the policy to accommodate this increase. Absolutely no problem with this but the current market premiums levels will be diluted. Just wondering, and we can discuss later, if this limit is realistic. Perhaps we can discuss at the meeting.
    Lawar, look forward to any comments on the attached."
  140. This meeting is the second occasion on which Mr. Beattie says that he "made clear" to Mr. Mullen his position as to what the TPC meant. I will discuss that evidence in Section E below.
  141. Over the following days, the renewal was broked to a number of underwriters, and they signed the quotation slip. Thus, between 20 and 22 January, the quotation slip was signed by Navigators, Talbot, Brit, Aegis, Markel and Advent. Navigators and Talbot were the lead Lloyd's syndicates on the slip. These signatures, like Mr. Beattie's, did not represent binding commitments.
  142. However, such commitments were made by RSA on 25 January (when Mr. Beattie entered RSA's 25% line) and by the 2nd – 12th Defendants between 25 January and 29 January 2016, when they entered their respective lines. Those lines were entered on a slip which contained small but, for present purposes, immaterial changes to the quotation slip signed between 20 and 22 January. Most of the underwriters who signed during that time, except for the 11th Defendant (Generali), and the 12th Defendant (Standard) had written lines on the expiring policy. The policy on which those lines were entered contained the TPC at Clause 1.5, as previously described. Edge in due course uploaded a copy of this policy onto the Xchanging system, which is a document depository used by the London market. There is no dispute, as far as the 2nd – 12th Defendants are concerned, that they became bound to the policy signed by them between 25 and 29 January. There is, however, a dispute as to whether they became bound, as a result of the events described below, by a later iteration of the policy signed by RSA.
  143. The position of the 13th and 14th Defendants (Prosight and Swiss Re) is clearer. They were not approached by Edge during the 25-29 January period, and did not sign the version of the slip in existence at that time. They only came on board in early February, after the slip had been amended so as to move the TPC from Clause 1.5 to two other locations within the policy, also as described below.
  144. Certain of the following underwriters raise specific arguments of positive misrepresentation concerning information which was given to them during the renewal broke. I will discuss those arguments, and the evidence, in the context of the discussion of non-disclosure in Section F below.
  145. For present purposes, it is sufficient to state that there is no evidence that the TPC, or indeed any of the other changes which had been introduced in the July endorsement, were discussed during the renewal meetings that took place with Mr. Lockyer and Mr. Mullen. None of the underwriters' notes of those meetings record any such discussion, and neither Mr. Mullen nor Mr. Lockyer suggested that the TPC had been brought to the attention of any of the following market by way of a discussion about it. Indeed, the only underwriter who has said that he read the clause, and then discussed it, is Mr. Beattie. He says that he did so in July 2015, again on 20 January 2016 and at a subsequent meeting on 29 January (ie after RSA had entered its line on 25 January).
  146. 27 - 29 January: the change in policy wording

  147. Accordingly, by 29 January, Edge had obtained the commitment of the 1st – 12th Defendants, over the period of 25 – 29 January, to the terms of the slip which had the TPC at Clause 1.5 as well as the other terms originally contained in the July 2015 endorsement. In fact, Edge had seen Mr. Beattie twice during that period with that version of the slip. Mr. Beattie had entered his line on 25 January, but had not initialled and dated the first and last pages of the policy. He did this on 27 January 2016.
  148. Although Edge had by 29 January obtained these contractual commitments by the 1st – 12th Defendants to renew the policy or (in the case of Generali and Standard) to subscribe to it for the first time, matters did not rest there. The events which unfolded on that day, and the next few days, had their origins in the fact that the Bank was keen to obtain NRF's advice on the terms of the proposed renewed policy. NRF were sent a marked-up version comparing the 2015 with the 2016 policy and were asked to advise and also to have a junior or trainee check that all previously agreed clauses had been incorporated. This request for advice led, in due course, to a number of proposed changes to the wording. These included the correction of typographical errors. Importantly, it also included a change in the location of the TPC within the policy. Instead of appearing at Clause 1.5, it was moved to two other locations within the policy.
  149. In order to assist in understanding the issues which arise, I shall refer to this subsequent policy, which was in due course scratched by Mr. Beattie for RSA – as well as by Prosight and Swiss Re – as the "later policy". I refer to the policy signed between 25 and 29 January as the "earlier policy".
  150. Before describing the events which followed, I will briefly indicate why the movement of the TPC may be relevant to the issues which I need to resolve. The movement of the clause has given rise to a number of issues in the case. As far as concerns the case against the insurers, these issues are principally as follows:
  151. a) The 2nd – 12th Defendants contend that the only policy which bound them was the earlier policy signed between 25 – 29 January. They did not sign the later policy, and contend that Mr. Beattie did not bind them to the later policy pursuant to the GUA or otherwise.
    b) RSA accepts that it was bound by the later policy, but contends that there was a further conversation about the TPC between Mr. Beattie and Mr. Mullen, prior to signature, which was similar in effect to the conversations alleged to have taken place in July 2015 and on 20 January 2016. This conversation is said to give rise to rectification or an estoppel or a collateral contract affecting the construction of the TPC.
    c) None of the parties contend, at least as their primary cases, that the movement of the TPC from its location at clause 1.5 in the earlier policy, to two locations in the later policy, has any effect on its true construction. Each side therefore starts from the proposition that the TPC means whatever it means, wherever it is located in the policy. Thus, both the Bank and Edge contend that the TPC has the same meaning as it had in the July endorsement, and covers the present loss, whether one is looking at the earlier policy or the later policy. And the underwriters contend that in all of those documents, the TPC has no application in the absence of physical loss or damage.
    d) Accordingly, in so far as the parties placed reliance on the positioning of the TPC within the earlier or later policy, as the case may be, this was very much by way of a secondary case. Thus, the Bank and Edge contended that the positioning of the TPC in the later policy – where it was included twice and where it appeared in its own self-contained section – dispelled any suggestion that it was simply another "basis of valuation" clause which was only applicable if there was physical loss or damage. And the underwriters contend that its positioning within the earlier policy supports their case that it only had the limited effect for which they contended, and that this did not change in the later policy.
  152. As far as the dispute between the Bank and Edge is concerned, the Bank contends that the sequence of events evidenced a considerable lack of professionalism on the part of Edge. During those days at the end of January, the importance of the TPC to the Bank, and the Bank's understanding of its effect, was made clear to Edge. The Bank contends that this should have been made clear to the underwriters. The Bank also contends that Edge's conduct resulted in an unacceptable situation where there was uncertainty as to whether the earlier or the later policy contained the relevant contract, at least as far as the 2nd – 12th Defendants are concerned. When, in due course, the claim came to be made, the parties for some considerable time were looking at the terms of the earlier policy. The later policy was not uploaded to Xchanging.
  153. Against this background, I will summarise the events which occurred without descending into unnecessary detail.
  154. On 25 January, NRF provided some advice to the Bank on the policy wording. On the following day, a conference call was arranged for 27 January: Mr. Stroink, Ms. Barnes and Ms. Van de Beek spoke to Mr. Weston-Simons of NRF. Mr. Stroink believed that it was during this call that the issue of the precise location of the TPC was discussed. His evidence was that he wanted to make sure that it was clear that, irrespective of where the clause was placed, it was an add-on and that there was coverage for client default. He did not recall exactly what was discussed, but thought that they had discussed the nature of the main insurance and the need for the cover under the TPC to be clearly seen as a separate add-on. Ms. Van de Beek could not recall what was discussed during the call, but thought it likely that the change of position would have been discussed but she did not recall the reason why.
  155. I consider that Mr. Stroink's evidence as to this call is inherently probable and consistent with the changes to the document which were then made. Later on the evening of 27 January, Ms. Van de Beek sent Mr. Weston-Simons a tracked change version of the policy. She had moved the TPC from Clause 1.5 to a new Clause 23 at the end of Section 2 (which was the General Conditions section of the policy).
  156. Mr. Weston-Simons then provided some advice in two emails sent that evening. In the second e-mail, sent at 18.46, he said:
  157. "I've been reflecting further on our call earlier and, in particular, our discussion regarding the nature of the insurance (i.e. marine cargo, storage and credit risk as a result of the Transaction Premium). Having done so, I would like to propose one further small, but potentially important, change; namely, to move the Transaction Premium clause which had appeared at clause 23 into the first / Subject Matter Insured section. The thinking behind this is that we do not want to take any chance that the Transaction Premium cover will somehow be interpreted as only applying in the marine cargo and storage contexts. Moving the clause up front (per the attached) should therefore make clear that the Transaction Premium cover applies in its own right.
    For comfort on this point, I would recommend that you also raise it with the broker in order to do two things:
    1. Confirm it is appropriate for the Transaction Premium cover to appear up front, and that it will be made clear in the course of the broking process that this cover is separate to the marine cargo and storage cover.
    2. Obtain the broker's views on whether the Transaction Premium cover requires its own "Location" / territorial limits wording (on the grounds that the existing "Location" wording only appears in relation to marine cargo and storage risks) and if specific limits for this cover need to be identified (on pages 3 and 7)".
  158. In the draft that Mr. Weston-Simons sent back, he moved the TPC into Section 1 of the policy (headed Conditions – Subject Matter Insured), placing it immediately prior to the "Basis of Valuation" and "Proof of Goods" clauses. He also, consistently with the general layout of Section 1, gave the clause a name or description (effectively a heading, although these headings were on the left hand side of the page against the clause): "Transaction Premium" which was in bold. It was therefore only at this stage that the clause was first given this nomenclature: ie the "Transaction Premium Clause" or TPC as it was referred to throughout the trial.
  159. Shortly afterwards (at 20.43), Mr. Weston-Simons explained his thinking to the partner with whom he was working, Ffion Flockhart:
  160. "The issue is where the Transaction Premium cover goes. It's essentially credit risk insurance, and therefore nothing to do with marine cargo or storage risks, and given its value I don't think we can take any chances by burying it at the back of the policy (as ABN had proposed). The point occurred to me on the way home and I wanted to get this off asap – so do please let me know if you would like to speak about it".
  161. The next morning, 28 January, Ms. Van de Beek sent Mr. Mullen a revised version of the policy "including our track and changes". This included the move of the TPC, which Mr. Weston-Simons had proposed the previous evening, from Clause 1.5 in Section 2 to Section 1 with the side marginal heading. In a Microsoft Word comment box next to the (relocated) TPC, Ms. Van de Beek commented:
  162. "David: Please confirm: is it appropriate for the Transaction Premium cover to appear up front, and is it clear that this cover is separate to the marine cargo and storage cover. Just like CEND and Business Contingent cover?"
  163. The document contained some other proposed changes to the policy wording. Many of these were typographical or minor in nature. There were, however, more extensive changes to the wording of the Missing Goods clause.
  164. Mr. Mullen forwarded the mark-up to Mr. Lockyer at 11.02, saying it contained "the changes the lawyers have suggested". He said that there was "nothing minor", by which he obviously meant nothing that was not minor (ie nothing major). He asked Mr. Lockyer to review them. Mr. Lockyer carried out a review, and he told Mr. Mullen in an e-mail sent at 12.22:
  165. "I have seen the comments made by Pauline and it appears her main concern is regarding Transaction Premium. Reading the clause it mentions that the Insured is covered by this policy for the transaction premium they would have earned if client of the insured defaults, regardless whether there has been any physical loss or damaged to the goods.
    Am I reading this correctly, and is this understanding of underwriters?"
  166. Four minutes later, at 12.26, Mr. Mullen answered:
  167. "No you are correct but they appear to want this as a separate section. I saw Brian with this and his view is "they have the coverage and they should be satisfied" Anyway nothing much to worry about.
    …
    I am proposing to make the adjustments on the copy policy and swing it."
  168. This e-mail indicates that Mr. Mullen had seen Mr. Beattie on the morning of 28 January 2016, and that there had been some discussion about Icestar's request to move the TPC into section 1. I have considered whether Mr. Mullen's statement in that e-mail, that the Bank wanted the clause as a "separate section", is correct. The Bank had proposed moving the TPC into Section 1, rather than into a separate section. Mr. Mullen may, however, have interpreted Ms. Van de Beek's question in the comment box, with its reference to CEND and Business Contingency cover, as suggesting a separate section: he seems to make a similar point in his e-mail of 28 January described below. It seems that Mr. Mullen may also possibly have considered that each part of Section 1 was a separate section: in his e-mail of 28 January, he referred to the "Basis of valuation section", whereas this would more readily be considered simply to be the Basis of Valuation clause within Section 1. Ultimately, however, I do not consider that I need to decide why Mr. Mullen said that the Bank wanted the clause as a separate section, not least because there soon came a time when it did.
  169. More importantly, however, the e-mail does suggest that Mr. Mullen thought that underwriters did have the understanding which Mr. Lockyer described: ie that the insured would be covered by the policy for the transaction premium they would have earned if a client of the insured defaults, regardless of physical loss or damage. Mr. Lockyer's question to Mr. Mullen, and Mr. Mullen's response, was the focus of submissions by the Bank in relation to Edge's liability: see Section J2 below.
  170. The e-mail also suggests that Mr. Mullen was not concerned about the proposed move of the TPC ("nothing much to worry about"), and that Mr. Beattie was irritated by the request that had been made.
  171. Mr. Beattie was not asked, on 28 January, to sign an amended slip. The last line of Mr. Mullen's email – referring to making the adjustments and swinging it – was the subject of considerable attention and submissions, as well as criticism, at the trial. The expression "swinging it" does sometimes have the connotation of misleading a person into doing something that he would not otherwise do. However, I do not think that Mr. Mullen was here indicating that he intended to mislead Mr. Beattie as and when he came back with "adjustments" on the copy policy, and in that context I bear in mind that no allegation of fraud is made against Edge in these proceedings. Rather, Mr. Mullen was indicating that he thought that he could get Mr. Beattie to agree as and when Mr. Mullen went back to him, notwithstanding Mr. Beattie's irritation. As Ms. Healy submitted, this unfortunate choice of phrase was simply a broker's shorthand for seeking to persuade Mr. Beattie to agree to move the TPC from Section 2 to Section 1.
  172. Later that afternoon, Mr. Mullen sent the following e-mail to the Bank (Ms. Van de Beek, Ms. Barnes and Mr. Stroink):
  173. "Dear Friends at Icestar.
    Thank you for your track and changes to the proposal. We very much appreciate your time on this subject. Thought it best because of the time frame to discuss with leading Underwriters and they are very comfortable with the suggested changes but make the following recommendations:
    Page 7 – sanctions exclusions: Currently the following Countries are subject to the Sanctions Exclusions clause To/from Iran ( soon to be lifted) Iraq. Syria. Libya. Myanmar, Yemen, Afghanistan, N Korea and Cuba. We can just include the current Countries that are excluded but if others are subsequently added to the "naughty boy" list then we have to add/subtract as applicable. As this is a UN requirement Underwriters would feel relaxed either way but I think the recommendation from our compliance section that to leave this clause as currently in the policy alone in case we fall foul of the regulators. However Underwriters would be happy either way. Your decision really.
    Page 9 - Transaction Premium: You are suggesting that a separate section for this similar to the CEND etc. We have deliberately sited this to the Basis of valuation section because this clause will determine the amount of recovery the Bank can obtain from the contract Underwriters. However again, we are happy to amend as you suggest.
    Other than that we preparing the cleaned up copy and would appreciate your comments on the above before sending to you for approval."
  174. This e-mail undoubtedly has some odd features. First, the statement that leading underwriters were "very comfortable with the suggested changes" is inconsistent with Mr. Mullen's earlier indication (in his e-mail to Mr. Lockyer at 12.26) that Mr. Beattie had been irritated about the proposed changes: "they have the coverage and they should be satisfied". Secondly, the opening paragraph suggests that the underwriters had made the two recommendations which were then set out. I consider that Mr. Mullen was here (not for the first or last time) expressing himself badly and without careful thought in his e-mail correspondence, since it is clear from the text of the two following paragraphs that the views being expressed were those of Edge; ie Mr. Mullen or Edge's compliance department. Thirdly, he said that Edge had "deliberately sited" the TPC to the "Basis of valuation section because this clause will determine the amount of recovery the Bank can obtain from the contract Underwriters". Whatever may be meant by "Basis of valuation section", the true position was that Edge had not deliberately sited the TPC to that location. Edge had put the TPC into Section 2, at Clause 1.5. It was the Bank, following the advice of NRF, which had put the TPC into Section 1, immediately before the Basis of Valuation clause. The insurers, as part of their argument on rectification and related issues, placed reliance on this e-mail as supporting Mr. Beattie's account of the conversations that took place with Mr. Mullen. I shall return to that issue in that context.
  175. On the morning of 29 January 2016, there was a 09.00 call between the Bank and Edge: Mr. Lockyer and Ms. Van de Beek were on that call, and it may be that Mr. Mullen and Mr. Stroink were too. During that call, Ms. Van de Beek asked for the TPC to appear in the policy twice: once in Section 1, where it already appeared, and again in Section 4. There was already an existing Section 4, which dealt with Profit Commission, Premium and related matters, and the request for the TPC to appear in Section 4 was clearly understood by Mr. Lockyer as a request for a standalone section containing the TPC. Shortly after the call, Mr. Lockyer sent Icestar a draft which acted on this request: the TPC remained in Section 1, and appeared again in a standalone Section 4.
  176. Later that morning (29 January) Mr. Mullen took the revised version of the policy, with the TPC in both places, to Mr. Beattie who scratched it. This is the third occasion on which the insurers rely, for the purposes of their rectification/estoppel/collateral contract case, upon a conversation between Mr. Beattie and Mr. Mullen as to the effect of the TPC. I shall return to that evidence in Section E below.
  177. Mr. Lockyer later emailed what he described as the "updated slip with the cosmetic amendments" to Mr. Beattie. This was the slip which Mr. Beattie had signed that morning. Mr. Lockyer also sent through the Bank's due diligence procedures. These essentially related to the warehouses which they used for storage. Mr. Beattie appears to have had this version of the slip printed, since he wrote in manuscript on it: "Tidy-up wording but this is the working slip for 2016/17 year".
  178. At 12.59, Mr. Lockyer sent the Bank the "final clean copy" of the slip, asking the Bank to let him or Mr. Mullen know if they required any further amendments. This was a copy of the slip signed by Mr. Beattie that morning, but without his signature.
  179. At 14.02, Ms. Van de Beek replied picking up five points on the wording. Three of these points were typographical in nature. A fourth was to amend the definition of "Location" in the slip to include the words "any elevator(s)" after "any storage(s)". There is no suggestion that this was a change of any significance. The final point was a request to add an additional storage facility in Milford Haven to the list of approved warehouses for oil. Mr. Lockyer replied saying that they would arrange for these amendments to be made, and then sent the Bank a clean copy incorporating the five changes made. The TPC remained in sections 1 and 4. At this point, however, Mr. Lockyer had not in fact approached Mr. Beattie to obtain agreement on the five changes, including the addition of the additional storage facility. Whilst Mr. Beattie had signed the later policy on the morning of 29 January – the principal change in that policy being the move of the TPC from Section 2 to two locations – none of the following market had been shown the later policy whose wording Mr. Beattie had agreed.
  180. The "amalgamated" document sent to the Bank on 29 January

  181. There was a great deal of reference at trial, although less so in closing submissions, to another document which Mr. Lockyer sent to the Bank on 29 January 2016. Each of the insurers considered it appropriate to adduce evidence from their underwriters commenting on this document, and Mr. Lockyer's conduct, notwithstanding that they had never seen the document at the time. In my view, this evidence was quite unnecessary and inappropriate, even prior to the introduction of new rules which relate to the content of factual witness statements. If comment on the document was to be made, this was a matter for the underwriters' expert evidence. It is convenient to refer to the document at this stage, albeit that in my view it is of no relevance to the issues which I need to resolve.
  182. The document in question was sent on 29 January 2016 by Mr. Lockyer to the Bank. Mr. Stroink had requested a "clean execution version", and Mr. Lockyer asked him: "Do you mean the final slip showing the market?". Mr. Stroink's response was: "Yes, a copy of the clean version you present to the market to sign". Mr. Lockyer then sent an amalgamated document comprising the policy with the 5 changes incorporated, plus all of the twelve subscribing underwriters' scratches. Mr. Lockyer appears to have understood that the Bank wanted to have the final wording together with details of the subscribing market. This lay behind his decision to amalgamate two documents.
  183. The document is certainly a curious one, since it is an amalgamation of (i) wording that was not finalised until 29 January, and (ii) scratches which mostly bear dates between 25 and 27 January. It is also clear from the pagination of the document and the footer that this is an amalgamation of documents. Mr. Lockyer should not have sent a document in this form, at least without explaining to the Bank how the document had been put together. Furthermore, the document incorporated the 5 amendments proposed by Ms. Van de Beek on the afternoon of 29 January, even though none of the underwriters, including Mr. Beattie, had yet seen the amendments. No doubt Mr. Lockyer was anticipating that Mr. Beattie would find the changes acceptable.
  184. In his evidence, Mr. Mullen did not seek to defend what Mr. Lockyer had done in preparing and sending this amalgamated document. I do not understand there to have been any allegation that Mr. Lockyer acted dishonestly, or that he in some way intended to deceive the Bank, or indeed that the Bank has ever (prior to trial) complained about the document. It would seem that Mr. Lockyer, in a misguided way, was seeking to give the Bank a single document which – after the slip had been through many iterations over the previous week – contained in one place the final wording and the security. It was not suggested to Mr. Lockyer in cross-examination that he had acted with a lack of integrity in relation to the preparation of this document. Mr. Lockyer, who was a witness who (as I have said) sought to give truthful evidence to the best of his recollection, was misguided in relation to this document. But this document is a distraction from the issues which I need to resolve – particularly bearing in mind that the real issues in this case involve what Mr. Mullen, who was responsible for broking this risk, said and did, rather than the conduct of Mr. Lockyer who was his assistant. There is no suggestion that the amalgamated document was the work of, or was approved by, Mr. Mullen.
  185. The Endorsement on 1 February 2016

  186. After the weekend of 30/31 January, Mr. Lockyer went to see the RSA on 1 February 2016. Mr. Beattie was not there that day, and Mr. Lockyer saw another senior individual (indeed more senior than Mr. Beattie), Mr. David Vaughan. Mr. Lockyer had prepared an endorsement to the policy on the standard form with a GUA stamp on the second page. The text of the Endorsement was:
  187. "Underwriters hereon note and agree the attached policy wording noting the minor alterations".
  188. The attachment to the endorsement was the policy which contained the 5 changes which Mr. Lockyer had made on the afternoon of 29 January, including the addition of the Milford Haven storage facility, following Ms. Van de Beek's e-mail. Save for those changes, the policy was identical to that which Mr. Beattie had signed on the morning of 29 January: ie with the TPC in two locations.
  189. Mr. Vaughan, who had had no prior involvement in this risk, scratched the endorsement and signed the attached policy, with the RSA stamp, on the first and last page. He also filled in the left hand section of the GUA box, thereby indicating that these changes had been agreed by "Slip Leader Only". But he added: List to market. This means that the market should be informed about the changes, although Mr. Vaughan was not requesting or requiring that all the underwriters should be asked to agree the changes. There was disagreement between Mr. Lockyer and Mr. Vaughan, in their evidence, as to whether Mr. Vaughan's completion of that box, including "List to market", was a consequence of advice given to him by Mr. Lockyer as to what Mr. Vaughan should do. I do not consider it necessary to make any finding on that issue. What matters is that this is what Mr. Vaughan did agree to do.
  190. Mr. Vaughan also wrote in manuscript on RSA's copy of the endorsement: "Minor alterations to slip. BB has original with amendments. DV 1/2/16".
  191. Later that day, Mr. Lockyer circulated Endorsement 1, and the attached slip signed by Mr. Vaughan, to the following underwriters under cover of an e-mail which stated:
  192. "Following the renewal of the above account, please find attached our endorsement together with our final wording incorporating some minor alterations following a review by the clients' in-house counsel i.e. capital letters replacing lower case, comma rather than semi colon etc."
  193. It is Edge's case (supported by the Bank) that this was the final version of the policy, to which all underwriters were bound. This is disputed by the insurers. RSA does not dispute (subject to the avoidance issues) that it became bound to the later policy signed by Mr. Beattie on 29 January, and then amended in 5 respects on 1 February. The 2nd – 12th Defendants, however, contend that neither the signature of Mr. Beattie on 29 January, nor that of Mr. Vaughan on 1 February, bound them to the later policy. Their only contract was the earlier policy subscribed at various points between 25 and 29 January.
  194. The following market was, however, provided with the endorsement and the attached policy on 1 February. However, those underwriters (in particular the 2nd – 12th Defendants) say that Mr. Lockyer's e-mail of 1 February was misleading (although no allegation of fraud is made). The argument depends upon what documents are being compared to each other. It may be appropriate to describe, as minor, the 5 changes made between (i) the later policy signed by Mr. Beattie on 29 January and (ii) the policy attached to Endorsement 1 scratched by Mr. Vaughan. However, it may not be appropriate to describe, as minor, the changes made when comparing (i) the earlier policy to which they subscribed between 25 and 29 January, and (ii) the policy attached to Endorsement 1 scratched by Mr. Vaughan. When these documents are compared, there was a potentially significant change, namely that the TPC had been moved from Clause 1.5 in Section 2 to the two other locations in the policy. The riposte to this argument, certainly on behalf of Edge, is that this change – although advised upon by NRF – makes no material difference to the interpretation of the cover.
  195. The subscription of Prosight (D13) and Swiss Re (D14)

  196. Following renewal of the Policy, one of the following underwriters (Talbot) wished to reduce its line from 20% to 10% because of a specific concern about the CEND cover under the Policy. On 25 February 2016, Edge therefore approached two new underwriters, David Burns of Prosight and Thomas Butterworth of Swiss Re to come on risk as new following underwriters. These underwriters each received (and only received) the final version of the policy, including the TPC in two locations and the 5 changes made on 29 January.
  197. The placement was a popular risk in the market. The written lines totalled more than 135%, with the consequence that the written lines of the following market "signed down" to a lower percentage of the risk than they had wished to write. The exception to this was RSA and Navigators, who had written on the basis that their percentage shares were "to stand": ie not to be reduced.
  198. August - December 2016 – defaults by Euromar and Transmar

  199. On 17 August 2016 Euromar defaulted under a forward sale contract with Icestar. The default related to a transaction designated as Deal 38. Mr Stroink notified this default to Edge. According to Mr. Mullen's note of the conversation with Mr. Stroink, he was told about "a provisional default position with their client, Euromar". "Legal" were currently "looking at the implications and currently the parent company Transmar are being consulted to see if a solution can be negotiated". (In fact, Transmar was not Euromar's parent company, but was an affiliate). The Bank and other financing institutions held discussions with Euromar and Transmar during August and September 2016 to see if a solution could be found. Transmar also agreed to provide financial support to Euromar.
  200. Ultimately, however, no solution could be found to Euromar's liquidity crisis and it committed defaults under the other four Euromar deals, in September and October 2016. Icestar terminated the sale contracts with Euromar on 11 October and 17 November 2016.
  201. At the end of December 2016 Transmar committed defaults under its ten extant deals with Icestar.
  202. US civil and criminal proceedings have since been pursued against the executives of both companies. In summary, a US criminal indictment was issued against the leading executives (the Johnson brothers, Peter G and Peter B, and Thomas Reich) on 2 August 2017, alleging various conspiracies to commit fraud. All pleaded guilty and were sentenced to terms of imprisonment. On 8 January 2018 a consortium of eight banks (including ABN Amro) filed a civil complaint against the executives alleging various forms of fraudulent misrepresentation, in particular relating to the extent of collateral available in respect of the syndicated borrowing facility provided by those banks. No defence was entered by either of the Johnson brothers or Mr Reich and the banks entered judgment in default in March and August 2019.
  203. The insurance claim

  204. Although Icestar was able to sell some of the relevant cargo (principally where the cargoes were beans rather than products) without incurring any loss, it nonetheless suffered substantial losses: the resale prices it was ultimately able to achieve from third parties were significantly less than the amounts contractually owing from Transmar and Euromar.
  205. Following the preliminary notification made by Icestar to Edge on 24 August 2016, Icestar started providing regular updates to Edge of the position, giving details about the attempts that it was making to avoid or reduce loss. On 15 October 2016, Ms. Franssen e-mailed Mr. Mullen explaining that Euromar had defaulted under 3 further transactions. She explained that Icestar had sent "Reservation of Rights" letters to Euromar, and in addition had sent, on the advice of German counsel, what she described as a "no demand" letter confirming that Icestar would not demand payment for the coming 13 weeks. This letter was sent so as to prevent Euromar from having to file for insolvency. She also explained in her email, under a heading "Actions undertaken to mitigate potential losses", the actions being so taken. These included cooperation with Euromar "in trying to liquidate the stock so that Icestar can be repaid". It also included using their "own network in the cocoa market to sell and liquidate the stock without involvement of Euromar". Stock details were being presented to, and quotes were being requested from, a few "good relationships". This e-mail is consistent with the evidence of Ms. Franssen described in more detail in Section H below.
  206. On 17 October 2016, Mr. Nick North of Edge, who worked on the claims side, put an entry onto the Xchanging system in the following terms:
  207. "Preliminary notification of a potential claim due to Euromar defaulting on its obligations to purchase various cocoa products.
    Would underwriters please specifically note the 'Actions undertaken to mitigate potential losses'.
    We are presently awaiting the leading underwriter's comments and will revert".
  208. The claim was dealt with for RSA by Mr. Matt Jones, who was a senior marine claims adjuster. On 19 October 2016, he made an entry on Xchanging in which he recorded that he had seen and noted the claim notification, and he asked a series of questions. The first was: "Is there any known shortage or physical damage to the cocoa in which ABN AMRO have an interest". These questions were passed on to Ms. Franssen, and her answers were sent to Mr. Jones on 28 October. The answer to the first question, concerning shortage or physical damage was: "Not to the knowledge of Icestar". Mr. Jones' next entry on Xchanging was that the responses had been noted. He asked for underwriters to be kept informed of all developments.
  209. On 7 and 12 December 2016, Ms. Franssen sent updates to Mr. Mullen in which she reported a bleak situation as far as both Euromar and Transmar were concerned. She indicated that it was reasonable to expect a bankruptcy in the near future. On 14 December, Mr. Nick North spoke to Mr. Matt Jones of RSA about the claim concerning coverage, and the conversation was reflected in an e-mail sent on that day. Mr. Jones confirmed that he was reviewing "coverage of any financial loss as a result of Euromar's default under the Slip Conditions". Mr. North said: "To us, it is very clear: this loss falls under Clause 1.5 of Section 2: General Conditions". Mr. North was therefore looking at the earlier policy scratched by the market between 25 and 29 January, rather than the later policy where the TPC had moved. Mr. Jones was asked for his confirmation regarding coverage for the purposes of a meeting with the Bank which was due to be held on the following day.
  210. Mr. North put an entry on Xchanging on 14 December 2016. This recorded that Euromar was in full insolvency proceedings and it looked as though Transmar would be too. Mr. North also spoke to Mr. Jones on 14 December. His note of that conversation indicates that Mr. Jones confirmed that there was no disputing cover. At 11.37 on 15 December, Mr. Jones e-mailed Mr. North. His email read:
  211. "It was very useful to speak with you earlier today and many apologies for not contacting you sooner.
    As we discussed I am able to confirm that cover under Section 2, Clause 1.5 of the policy would appear to attach in respect of this situation.
    Also, and again as we agreed, RSA are content that the Assured are in the best position and with the most appropriate knowledge to approach potential alternative buyers. RSA would, therefore, request that ABN obtain quotations from other entities and submit these to Underwriters for their consideration.
    I will add these comments to the ECF transaction.
    I trust this is satisfactory, but any issues please do not hesitate to contact me."
  212. Mr. Stroink's evidence is that he was told about underwriters' confirmation of cover. He said that this "confirmed what we had expected and gave us comfort that we had the backing of insurers against any potential losses".
  213. It is clear, in my view, that by this time Mr. Beattie had been consulted about the claim. At 08.35 on 15 December (ie prior to Mr. Jones' e-mail indicating that the claim was covered), Mr. Beattie sent Mr. Mullen an email stating:
  214. "Can we have a chat about ABN Amro? It appears we may have a problem over a recent loss which is effectively a Financial Guarantee not linked to any loss or damage recoverable under ICC 'A'. I have a client meeting at 10.30 but will be back in the office around 11.30".
  215. Mr. Mullen's response was to tell Mr. Beattie that ABN Amro were coming in at 11 am to discuss. He asked Mr. Beattie whether he would like to meet Mr. Stroink and Ms. Franssen after the meeting, and invited him to a lunch if he was free. Mr. Beattie said that he now had another meeting to attend and could not make lunch. But he added:
  216. "We can perhaps get together tomorrow for a chat after Nick [North] and Matt [Jones] have deliberated. If this is a financial guarantee, it is a no-no within RSA and I will be instructed to change the wording or manage an exit."
  217. There was no suggestion at this stage, by Mr. Beattie, that he had been misled into providing the relevant coverage, or that there had been an oral agreement which qualified or altered the coverage. Nor was there any indication of any dispute as to whether coverage attached in respect of the potential claim.
  218. On 15 December, Mr. Mullen e-mailed Mr. Beattie as follows:
  219. "Just looking at Nick's notes and this incident comes under the transaction premium section of the policy (page 12). This sections covers the Insured in the event of a default on the part of a Insured's client. In this case the client is Euromar – a large purchaser of commodities and they are being liquidated. ABN have established and confirmed that all the good under this purchase agreement at in situ and in good condition. They have already sold on the open market butter cake which accounts for about 10% of these transactions and the sale price was the same as the purchase price. The remaining goods should be offered to the open market and ABN have several good contacts to move this forward but before doing so want to know if Underwriters are happy for them to look for offtakers or if Underwriters would like to explore the possibilities of re-sale. In all of these movements, ABN have acted as prudently uninsured and are mitigating the position. We have invited Matt to attend and he would be welcome."
  220. On 16 December, Mr. Jones put an entry on Xchanging which in substance repeated the e-mail that he had sent on the previous day. This included the following statement:
  221. "RSA confirm that they are of the opinion that cover under Section 2, Clause 1.5 of the policy would appear to attach in respect of this situation".
  222. On 19 December, Ms. Franssen sent a summary to Mr. Mullen and Mr. North of the matters which they had discussed at the meeting on 15 December. This identified the steps that the Bank would be taking.
  223. 2017

  224. Icestar's policy fell due for renewal at the end of January 2017. RSA was not prepared to include the TPC on renewal. Mr. Stroink's evidence was that Mr. Mullen told him that although the TPC would have to come out of any renewal, the current policy would still respond to claims for default. It is clear that at this stage both Mr. Mullen and Mr. Stroink understood that underwriters would be covering the Bank's potential claim.
  225. In April 2017 an email exchange took place between Mr. Beattie and Mr. Mullen in which Mr. Beattie said that the exposure to the Bank's financial loss claim on the "aborted cocoa sale in Europe" could be as much as USD 30m. He said that this, combined with some other substantial insured losses, "will probably bring to an end our foray into bank traders' accounts. Indeed it may be better if I shut them all down prior to my final departure rather than leave the task to others". At least at this stage, there was no suggestion that the Bank's financial loss claim was not covered by the policy, or that Mr. Beattie had been misled into writing it.
  226. At around the same time, underwriters engaged investigators called Gray Page to investigate the Bank's claim. Underwriters then declined coverage on 16 August 2017. This was on the basis of their argument that the TPC was concerned only with the scope of the indemnity where there is physical loss or damage to the cargo, and that it did not operate to provide any cover for credit or financial default risks.
  227. C: The policy terms and the GUA

    C1: The policy subscribed by the market between 25 and 29 January 2016

  228. The policy wording which was scratched by Mr. Beattie on 25 and 27 January 2016, and by the 2nd – 12th Defendants on various dates between 25 and 29 January 2016, contained the following relevant terms. Some parts of the policy wording used side headings (ie headings on the left-hand side of the page, with the text of the clause on the right). Other parts used headings which were above the text of the relevant clause. Nothing turns on the location of the headings, and for convenience all headings below are shown above the relevant text.
  229. "TYPE:
    Marine Cargo and Storage Insurance.
    INSURED:
    (a) ABN AMRO Bank N.V. and or
    (b) Icestar B.V. and or
    …
    CONVEYANCE:
    Land, water and air including but not limited to steamer(s) and/or barge(s) and/or parcel post and/or road and/or rail and/or messengers and/or every conveyance or by inland waterways of any description and/or any other method of transfer approved by Underwriters ("Conveyance")
    "Carriers" include (but are not limited to) shipping companies, charterers, forwarding agents, bailees and other third parties involved in the carriage of the Subject Matter insured (as such term is defined in section 1 "Conditions"- Subject Matter Insured" below).
    INTEREST:
    All goods and or merchandise appertaining to the Insured's business for which the insured is the legal owner or for which the insured is at risk or responsible, contractually or otherwise, consisting principally but not limited to:
    (A) Hard commodities (including but not limited to aluminium, steel, copper, nickel, zinc lead and tin) ("Hard Commodities");
    (B) Soft commodities (including but not limited to coffee beans, cocoa products, butter, grain, wheat, soybeans and soya products, cotton, corn, palm oil and orange juice) ("Soft Commodities"); and
    (C) Energy commodities (including but not limited to all kinds of oil such as (but not limited to) refined or crude oil and or fuel oil, kerosene, gas oil, liquid petroleum gases and or similar oil products, including bitumen, alcohol, biodiesel, bio fuels and any green energy products, vegetable oils, consisting primarily of, but not limited to soyabean oil, palm oil and or similar products, including liquid nitrogen gas, natural gas and or any similar gas products which forms part of the Assured's activities. (Energy Commodities) and all the other interests and/or commodities traded or owned by the Insured are held and covered with or without notice at rates and on terms to be agreed by the Underwriters.
    LIMITS:
    Interest (A) and (B)
    USD 50,000,000 any one Conveyance.
    USD 100,000,000 any one individual warehouse; shed or storage facility (howsoever named) located on the same premises as any one Exchange Approved Storage Location and/or Insured Approved Storage Location. Including clients own elevators as and when declared.
    Limits deemed "or equivalent in any other currencies".
    Interest (C)
    USD 50,000,000 any one Conveyance.
    USD 250,000,000 any one individual warehouse; shed or storage facility (howsoever named) located on the same premises as any one Exchanged Approved Storage Location and/or Insured Approved Storage Location. Including clients own elevators as and when declared.
    Limits deemed "or equivalent in any other currencies".
    BASIS OF VALUATION:
    Market Value at date of claim provided that the market value shall be no less than the Market Value as used at the date of purchase. For the purposes of this clause the term "Market Value" shall mean the relevant published Exchange future price, or any other price stated in the Insured's underlying purchase contract, call option contract or any other document relevant to the purchase (as amended, supplemented, replaced or otherwise modified from time to time), plus or minus any applicable adjustments as referenced in the Insured's underlying purchase contract, call option contract, or any other document relevant to the purchase (as amended, supplemented, replaced or otherwise modified from time to time).
    …
    GENERAL CONDITIONS:
    (Applicable to all sections of this contract)
    Notwithstanding anything contained herein to the contrary, it is agreed and understood in general terms that it is the intention of this contract to protect the interest of the Insured at all times and in all circumstances.
    This contract is to protect against all risks of physical loss of or damage to the Subject Matter Insured from whatsoever cause arising.
    All other conditions are more fully detailed in each respective section and/or other parties, with the prior written consent of the Insured.
    …
    PREMIUM:
    As detailed in each section below.
    DECLARATIONS:
    Monthly declarations of amounts at risk including Seller Insurance (as such term is defined in section 1 clause 6 below). The amount declared is that recorded on the 1st day of each month or on such date as agreed between the Insured and Edge Brokers (London) Ltd.
    …
    SECTION 1: CONDITIONS – SUBJECT MATTER INSURED
    SUBJECT MATTER INSURED: THE INTEREST
    …
    LIMIT:
    Interest (A) and (B)
    USD 50,000,000 any one Conveyance
    USD 100,000,000 any one individual warehouse; shed or storage facility (howsoever named) located on the same premises as any one Exchange Approved Storage Location and/or Insured Approved Storage Location.
    …
    LOCATION:
    At and from any port(s) or place(s) in the world, via any route or Conveyance(s) including but not limited to any storage(s), interior transit and transhipment(s) incidental to the transit in any circumstances and including 30 days after arrival at first destination.
    PER:
    Conveyances – Road and or Rail and or Barge and or any other approved method of conveyances.
    SHIPMENT CLAUSE:
    All shipments are covered under this contract, whether containerized or otherwise and whether on or under deck, irrespective of bill of lading instructions. The seaworthiness of the containers is hereby admitted between the Underwriters and the Insured. The fact that the containers are found not to be seaworthy at unloading point or prior to unloading point shall not invalidate claims relevant to seaworthiness of the containers.
    STORAGE RISKS:
    The insured may declare a storage risk, irrespective if the storage is incidental to a transit declared hereunder. The limit for storage under section (A) and (B) is USD 100,000,000 in each individual warehouse; shed or storage facility (howsoever named) located on the same premises as any one Exchange Approved Storage Location. Locations are including worldwide but excluding any sanctioned Country. The limit for storage under section (C) is USD 250,000,000 in each individual warehouse; shed or storage facility (howsoever named) located on the same premises as any one Exchange Approved Storage Location and/or Insured Approved Storage Location. Locations are including worldwide but excluding any sanctioned Country.
    …
    PROOF OF GOODS:
    For the purpose of this contract, Underwriters agree that in the event of a claim arising hereunder, the relevant purchase contract of the Insured relating to the claim, together with the presentation of the applicable transportation document, storage report and or any other relevant documentation (whether or not in electronic form) as set out in the purchase contract shall be evidence of proof of the existence of the goods and their unimpaired physical condition at the time of purchase.
    …
    INSTITUTE CLAUSES AND CONDITIONS:
    Against all risks of physical loss or damage of the Subject Matter Insured from whatever cause arising including:
    Institute Radioactive Contamination Exclusion Clause CL.370
    Institute Cargo Clauses 'A' CL.382 1.1.09
    Institute Cargo Clauses (Air) CL 385. 1.1.09
    …
    B. Soft Commodities
    i: In Transit @ 0.16% per voyage inclusive of war, strikes, riots and civil commotions. Goods purchased on the high seas rated at 75% of cover rates.
    ii: In Store @ 0.13% per annum or pro-rata monthly inclusive of war, strikes, riots and civil commotions including previous preferential rated locations.
    iii: Additional premium in respect of shipments/storage of Soft Commodities to cover the risks of spontaneous combustion – 25% uplift on cover rating.
    SECTION 2: GENERAL CONDITIONS
    These general conditions apply to all sections of this contract. In the event of a conflict between these general conditions and specific clauses in this contract, the specific clause shall apply.
    1. Definitions
    [Clauses 1.1 to 1.4 contained definitions of, respectively, "Exchange", "Exchange Approved Storage Location", "Insured Approved Storage Location" and "Underwriters"]
    1.5 Underwriters note and agree that, in respect of any Transaction, it is hereby confirmed that the Insured is covered under this contract for the Transaction Premium that the Insured would otherwise have received and/or earned in the absence of a Default on the part of the Insured's client.
    "Actual Sale Price" means the sum received by the Insured upon the sale of the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    "Default" means a failure, refusal or non-exercise of an option, on the part of the Insured's client (for whatever reason) to purchase (or repurchase) the Subject Matter Insured from the Insured at the Pre-agreed Price.
    "Pre-agreed Price" mean the amount for which the Insured's client had agreed to purchase (or repurchase) the Subject Matter Insured from the Insured as specified on the relevant invoice or in the relevant transaction documents, comprising the principal together with any premium or profit element payable to the Insured.
    "Transaction" means any transaction where, following a Default on the part of the Insured's client, the Insured sells the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    "Transaction Premium" means an amount that is equal to the difference in value between the Pre-Agreed Price and the Actual Sale Price.
    1.6 Underwriters note and agree that the section relating to independent grain operators/ owner elevators will apply to all sections of the contract as and when applicable.
    Furthermore, if the owner/ operators insurance is selected then the Insured may declare under the buyers/ sellers interest sections of the policy, nevertheless the Insured is to request to be noted as a loss payee on their policy(ies).
    2. Notice of a Claim
    The Insured shall report to the Underwriters any circumstances which may give rise to a claim under this contract as soon as practicable but not more than 90 days after the responsible person becomes aware of such circumstances and shall thereafter keep the Underwriters fully informed of all developments. For the purpose of this clause any communication may be carried out by (but not limited to) telephone, email or facsimile. The responsible person will mean the head of risk and includes the title of the relevant individual.
    3. Due Diligence
    The Insured shall do (to the extend it reasonably can do) all things reasonably practicable to prevent any claim being made under this contract, providing always that following the occurrence of a peril in relation to the subject matter insured, the Insured may in its sole discretion elect an appropriate course of action, as it considers appropriate in any particular circumstance, subject to the Insured acting is good faith with the intention of minimising any ultimate potential net loss (save that the Insured shall not be required to exercise any put option following the occurrence of any such peril)
    4. Payment of claims
    Subject to other conditions in this contract in relation to payment of claims, the Insured shall provide to the Underwriters details and full documentation (if relevant) in respect of the claim, and the Underwriters will submit to their representatives such details and documents for payment of the claim within the time scale agreed between Lloyd's brokers and electronic exchanging services. Underwriters hereon will pay the Insured any recoverable amount of any loss or damage as soon as practicable but in any event no later than 30 days from receipt of the details and documents.
    ...
    17. Fraudulent documentation
    17.1 Underwriters will accept Exchange and Non Exchange warehouse receipts, warrants of any other approved Exchange Warehouse applicable to this contract and documents evidencing ownership of the subject matter insured (each a "document") as proof of the existence and unimpaired physical condition of the subject matter insured and of the Insured's interest in the subject matter insured at the time of issue of such document or at the time the subject matter insured became at the risk of the Insured if subsequently hereto.
    17.2 If as a result of a Document being stolen, lost and/or misappropriated and such document is fraudulently converted such that the Insured suffers physical loss of the related subject matter insured, or the impairment of its interests in the subject matter insured, such a loss shall be recoverable hereunder.
    17.3 The Insured is indemnified by the Underwriters for the direct financial loss suffered by the Insured including by reason of the impairment of the Insured's interest in the subject matter insured arising by reason of the Insured, either in good faith during the period and in the ordinary course of business acting or relying upon or being supplied with a "counterfeit document" of "fraudulently altered document" (as defined below). This indemnity does not include loss caused by dishonesty of the Insured's own employees.
    17.4 "Counterfeit Document" means a document that is a reproduction of an authentic document such that the Insured or its agents is deceived on the basis of the quality of the imitation so as to believe that such item is the authentic instrument.
    17.5 "Fraudulent altered document" means a document that is materially altered by any person for a fraudulent purpose.
    …
    20. Business contingency cover
    20.1 Cover against costs, expenses and losses incurred by the Bank as set out in this clause, including but not limited to any premium or profit element that the Insured would otherwise have earned but for the delayed delivery of the subject matter insured to the Exchange in the related Futures Month.
    Futures Month shall mean the futures month as defined in the purchase contract by the Insured.
    20.2 In respect of this clause 20 a notice of claim shall include written evidence that the Insured incurred any claimed costs, expenses or loss.
    20.3 Any declaration under this clause 20 is to be recorded in accordance with the relevant contract terms and conditions and the Premium is to be calculated at 40% of cover rates.
    20.4 Subject to individual declarations prior to Attachment of risk and subject to no previous losses which would be collectable under this contract.
    ...
    22. Non-Avoidance
    The Underwriters will not:
    a) Seek to avoid or repudiate this contract for non-disclosure or misrepresentation other than fraudulent non-disclosure or fraudulent misrepresentation; or
    b) Rely on, or assert any breach of warranty as grounds for the Underwriters to be discharged from any liability other than where the warranty was given fraudulently; or
    c) Seek damages for or seek to reject a claim for loss on the grounds of:
    i. Non-disclosure or misrepresentation other than fraudulent non-disclosure or fraudulent misrepresentation; or
    ii. Any breach of warranty other than where the warranty was given fraudulently.
    SECTION 3. ADDITIONAL COVER CONFISCATION AND EXPROPRIATION
    ...
    1.2 Cover for loss of and/or damage to the Subject Matter Insured directly caused by confiscation, moratorium, seizure, appropriation, expropriation, requisition, deprivation, requisition for title or use a wilful destruction by/or under the order of any government (whether civil, military or de facto and whether recognized or unrecognized) and/or public or local authority of the country or place in which the vessel(s)/craft/property hereby insured are covered by the terms of the contract whilst stored in a bonded warehouse, shed or storage facility.
    For the purposes of this clause 1.2 "deprivation" means loss of use by of possession of the subject matter insured caused by:
    (1) The failure or refusal of the foreign government for a period of three months to permit the export of the subject matter insured from the foreign country;
    (2) The Insured being prevented from exporting the subject matter insured from the foreign country for a period of three months due to its inability to obtain export licence from the appropriate authority in the foreign country.
    (3) Any loss shall be deemed to have occurred during the policy period providing that the Lead Underwriter or the Insured's representatives had notice of such permits being obtained.
    …
    SUBSCRIPTION AGREEMENT SECTION
    SLIP LEADER:
    Royal Sun & Alliance
    BASIS OF AGREEMENT TO CONTRACT CHANGES:
    Subject to GUA October, 2001 with Marine Cargo Schedule 2003
    OTHER AGREEMENT PARTIES FOR CONTRACT CHANGES, FOR PART 2 GUA CHANGES ONLY:
    Slip leader only to agree part two changes.
    AGREEMENT PARTIES FOR CONTRACT CHANGES, FOR THEIR PROPORTION ONLY:
    Slip leader to agree all contract changes."

    C2: The policy subscribed by Mr. Beattie on 29 January 2016, and by Prosight and Swiss Re in February 2016

  230. The policy wording signed by Mr. Beattie on 29 January, and then subject to the 5 amendments attached to Endorsement 1 (signed by Mr. Vaughan on 1 February) – and later signed by Prosight and Swiss Re in February – contained only one change which was potentially material to the parties' arguments. As previously discussed in Section B, this was the location of the TPC. In this later version of the policy wording, the TPC appears twice.
  231. The first location was in Section 1 of the policy wording, headed "CONDITIONS – SUBJECT MATTER INSURED" followed by "SUBJECT MATTER INSURED: THE INTEREST". It there appears immediately following the "Location", "Per", "Shipments" and "Storage Risks" clauses, and prior to the "Basis of Valuation" clause.
  232. The second location was in a separate section. In the original wording, Section 3 of the policy was the CEND cover, and Section 4 concerned profit commission, premium and other matters. In the later wording, the TPC was in a separate section (Section 4) headed "Transaction Premium", and also with a side heading "Transaction Premium". Section 4 of the earlier policy was now Section 5.
  233. One further curiosity is that the "Basis of Valuation" clause itself now appeared in two places in the policy. These were: first, in its original location, on page 3 of the policy wording, following the "Limits" provision; and secondly in Section 1 of the Policy as described above. In the earlier policy wording, it was only on page 3, following the "Limits" provision. None of the parties sought to attach any significance to the duplication of the "Basis of Valuation" clause. This duplication appears to have been an error, originating during the Bank's review of the policy with NRF.
  234. C3: General Underwriters Agreement

  235. The section of the earlier policy headed "SUBSCRIPTION AGREEMENT SECTION" was unchanged in the later policy. It was also materially identical to that contained in the 2015 policy. This section dealt with the authority of the Slip Leader, the RSA, to agree contract changes. The wording referred to the "GUA October, 2001 with Marine Cargo Schedule 2003".
  236. The relevant provisions of the General Underwriters Agreement, or GUA, were as follows:
  237. "Purpose of the GUA
    The General Underwriting Agreement is a replacement for existing Leading Underwriter Agreements, providing a form of standardisation where practical and appropriate. The purpose of the GUA is to:
    ● creates an agreement between the subscribing Underwriters on a particular contract for the management of changes
    ● clarify the extent of the delegated authority to the Slip Leader and Agreement Parties
    ● enable each class of business to define their specific requirements/needs within a common framework
    ● allow a single Slip Leader and/or Agreement Parties to agree contract alterations where empowered to do so by the GUA
    ● ensure all Underwriters are notified of alterations, where appropriate
    The GUA, as with previous Leading Underwriter Agreements, is not intended to affect the several liability of each subscribing Underwriter. As made clear throughout the GUA, each subscribing Underwriter's obligations remain several and not joint and limited to the extent of its signed subscription.
    The GUA in outline
    The GUA is an agreement between the subscribing underwriters on a particular contract relating to the level of delegated authority in respect of post placement alterations.
    The GUA structure provides a standard agreement that is referenced from the slip. If there is a difference between the GUA and the slip terms, the slip overrules the GUA. This enables, where appropriate, the terms and conditions of a Policy/Contract to be tuned to individual contract needs. The GUA is intended to be used with the new LMP Slip, but can be incorporated into any other form of slip. The slip should make clear reference to the GUA within the Subscription Agreement section under "Basis of Agreement to Contract Changes". For example "GUA October 2001 with Marine Hull Schedule May 2002".
    The structure of the GUA enables its use for any class of business with each defining its particular requirements in the Class of Business Schedules. Class of Business Schedules have been defined for Non-Marine, Marine Cargo, Marine Hull, Marine Liability, Marine Energy, Excess of Loss & Treaty Reinsurance, Political Risks, Professional Indemnity and Terrorism.
    Each Schedule is split into three Parts, defining the Underwriters whose agreement is required for each type of alteration:
    Part 1 – Slip Leader only
    Part 2 – Slip Leader plus Agreement Parties
    Part 3 – All Underwriters
    The slip should clearly identify the Slip Leader and any Agreement Parties for contract changes in the designated area (refer to the LMP Slip).
    The GUA defines administration tasks, such as distribution of certain agreed endorsements to following underwriters. This distribution can be on paper or alternatively via e-mail.
    The GUA has been designed to work within the existing endorsement process and uses the traditional endorsement document. When an endorsement is presented the GUA stamp may be applied by the Slip Leader, or alternatively the broker may wish to have it pre-printed. Two versions of the stamp have been created to support marine practices (Stamp A with listing), and the practices of the Non-Marine market (Stamp B) – refer to the examples. The Slip Leader will need to initial the appropriate box in the stamp to indicate the level of agreement required.
    2. Definitions
    2.1 The "Slip Leader" is the Underwriter identified as such on the slip.
    2..2 The "Agreement Parties" are those Underwriters identified as such on the slip. Where no such Underwriters are so identified, the Agreement Parties will all be Underwriters.
    2.3 The "Other Underwriters" are all Underwriters not identified as the Slip Leader or as an Agreement Party, other than those to whom Clause 1.3 applies.
    …
    3. Alterations
    3.1 Only Alterations set out in the applicable Schedule Part 1 may be agreed by the Slip Leader alone on behalf of the Agreement Parties and Other Underwriters, each for its own individual signed proportion severally and not jointly.
    3.2 Only Alterations set out in the applicable Schedule Part 2 may be agreed by the Slip Leader and Agreement Parties acting together on behalf of Other Underwriters, each for its own individual signed proportion severally and not jointly.
    3.3 Such Alterations shall only be agreed by Slip Leader/Agreement Party itself or by members of its staff who have been specifically designated to assume such responsibility.
    3.4 The Alterations set out in the applicable Schedule Part 3, and any Alteration that the Slip Leader and any Agreement Party so require, may be agreed only by all Underwriters, each for its own individual signed proportion severally and not jointly.
    4. Evidence of Agreement
    4.1 The Slip Leader shall incorporate the GUA Stamp (in either for A or B) in the endorsement, should it not be incorporated in or appear on the form of endorsement.

    GUA Stamp A

    General Underwriters Agreement (GUA)
    Each Underwriter's proportion is several not joint

     

    Slip Leader Only

    Box 1

    Slip Leader
    And
    Agreement Parties

    Box 2

     

    All Underwriters

    Box 3

    Notification to followers

    Yes / No
    Within______ working days

     

    GUA Stamp B

    General Underwriters Agreement (GUA)
    Each Underwriter's proportion is several not joint

     

    Slip Leader Only

    Box 1

    Slip Leader
    And
    Agreement Parties

    Box 2

     

    All Underwriters

    Box 3

    4.2 The Slip Leader (and Agreement Parties if appropriate) shall then initial in the appropriate Box the level of authorisation required.
    4.2.1 If any of the Slip Leader or Agreement Parties initials Box 3, the Alteration shall be referred to all Underwriters, each for its own individual signed proportion severally and not jointly.
    4.2.2 If the Slip Leader initials Box 2, the Alteration shall be referred to all Agreement Parties.
    4.2.3 If the Slip Leader initials Box 1 and initials and dates the endorsement in the customary place, no further agreement shall be required.
    4.2.4 Agreement to Clause 4.2.1 or Clause 4.2.2 Alterations shall be effected by each Underwriter required initialling and dating the endorsement in the customary place.
    5. Effective date of agreement
    5.1 Unless otherwise specified on the endorsement, the agreement evidenced by the Alteration shall take effect:
    5.1.1 for Clause 3.1 Alterations, on the date inserted by the Slip Leader, for the individual signed proportion of each Underwriter severally and not jointly;
    5.1.2 for Clause 3.2 Alterations, on the date when the last of the required agreements from the Slip Leader and Agreement Parties has been obtained, as inserted by that last Party, each for its own individual signed proportion severally and not jointly;
    5.1.3 for Clause 3.4 Alterations, on the date inserted by each Underwriter, so far as its proportion is concerned.
    10. Terms of the Slip
    10.1 Save as provided for in the Condition Paramount and in Clause 11
    10.1.1. where the slip or any endorsement thereto conflict with the terms of this GUA, the terms of the slip/endorsement shall prevail, provided that for the purpose of this clause, the terms of the slip/endorsement are those shown to and subscribed by each subscribing Underwriter for its own proportion.
    10.1.2 where the risk has been written as provided for in Clause 1.2, and its terms or those of any endorsement to it conflict with the terms of this GUA, the terms of the declaration, certificate or other form of contract of insurance or reinsurance or endorsement thereto shall prevail, provided that for the purpose of this clause, the terms thereof are those authorised by each subscribing Underwriter for its own proportion in the original lineslip, marine cargo cover or other contract for insurance or reinsurance.
    GUA Marine Cargo Schedule – June 2003
    This Schedule applies to insurance and facultative reinsurance and the terms insurance and insured shall include facultative reinsurance and reinsured.
    PART 1
    1.1 All Alterations that the slip specifies are to be agreed by the Slip Leader only.
    PART 2
    Alterations the Slip Leader and Agreement Parties may, if unanimous, agree on behalf of all Underwriters each for its own proportion severally and not jointly.
    2.1 All alterations that the slip specified may be agreed by the Slip Leader and Agreement Parties.
    2.2 All Alterations which do not fall within either Part 1 or Part 3.
    PART 3
    Alterations which may be agreed only by all Underwriters each for its own proportion severally and not jointly.
    3.1 All Alterations that the slip specifies may be agreed only by all Underwriters.
    3.2 All Alterations which are judged by either the Slip Leader or by any Agreement Party to be ones which ought to be agreed by all Underwriters.
    3.3 All Alterations which fall within the following list, unless the slip specifies such an Alteration may be otherwise agreed:
    3.3.1 Any Alteration which increases the monetary exposure of the Underwriters (or of any of them), whether that exposure arises in relation to the slip as a whole, or in relation to a section thereof."

    D: Construction of the policy

    D1: Legal principles

  238. It was not in dispute that the principles which govern the construction of policies of marine insurance are those applicable to commercial instruments and indeed to contracts generally: see Arnould: Law of Marine Insurance and Average 19th Edition, paragraph 3-01. These principles have been authoritatively stated in a trilogy of Supreme Court decisions in the past 10 years: Rainy Sky SA v Kookmin Bank [2011] UKSC 50; Arnold v Britton [2015] UKSC 36; Wood v Capita Insurance Services Ltd. [2017] UKSC 24. The court must ascertain what a reasonable person – ie a person who has all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract – would have understood the contracting parties to have meant by the language used. This means disregarding evidence about the subjective intention of the parties. These principles have recently been applied by the Commercial Court in The Financial Conduct Authority v Arch and others [2020] EWHC 2448. In the Supreme Court decision in that case, there was no dispute as to the relevant principles as most recently authoritatively discussed in Wood: see [2021] UKSC 1 para [47].
  239. In Rainy Sky, Lord Clarke described the exercise of construction as being essentially a "unitary exercise" in which the court must consider the language used and ascertain what a reasonable person, with the relevant background knowledge, would have understood the parties to mean. If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other. Where the parties have used unambiguous language, the court must apply it: Rainy Sky paragraphs [23] and [25].
  240. Whilst this unitary exercise of interpreting the contract requires the court to consider the commercial consequences of competing constructions, commercial common sense should not be invoked retrospectively, or to rewrite a contract in an attempt to assist an unwise party, or to penalise an astute party. This is clear from the judgment of Lord Neuberger in Arnold v Britton. He said at paragraphs [15] – [22] (omitting the principal internal citations):
  241. "[15] When interpreting a written contract, the court is concerned to identify the intention of the parties by reference to "what a reasonable person having all the background knowledge which would have been available to the parties would have understood them to be using the language in the contract to mean" … And it does so by focussing on the meaning of the relevant words, in this case clause 3(2) of each of the 25 leases, in their documentary, factual and commercial context. That meaning has to be assessed in the light of (i) the natural and ordinary meaning of the clause, (ii) any other relevant provisions of the lease, (iii) the overall purpose of the clause and the lease, (iv) the facts and circumstances known or assumed by the parties at the time that the document was executed, and (v) commercial common sense, but (vi) disregarding subjective evidence of any party's intentions.
    [16] For present purposes, I think it is important to emphasise seven factors.
    [17] First, the reliance placed in some cases on commercial common sense and surrounding circumstances should not be invoked to undervalue the importance of the language of the provision which is to be construed. The exercise of interpreting a provision involves identifying what the parties meant through the eyes of a reasonable reader, and, save perhaps in a very unusual case, that meaning is most obviously to be gleaned from the language of the provision. Unlike commercial common sense and the surrounding circumstances, the parties have control over the language they use in a contract. And, again save perhaps in a very unusual case, the parties must have been specifically focussing on the issue covered by the provision when agreeing the wording of that provision.
    [18] Secondly, when it comes to considering the centrally relevant words to be interpreted, I accept that the less clear they are, or, to put it another way, the worse their drafting, the more ready the court can properly be to depart from their natural meaning. That is simply the obverse of the sensible proposition that the clearer the natural meaning the more difficult it is to justify departing from it. However, that does not justify the court embarking on an exercise of searching for, let alone constructing, drafting infelicities in order to facilitate a departure from the natural meaning. If there is a specific error in the drafting, it may often have no relevance to the issue of interpretation which the court has to resolve.
    [19] The third point I should mention is that commercial common sense is not to be invoked retrospectively. The mere fact that a contractual arrangement, if interpreted according to its natural language, has worked out badly, or even disastrously, for one of the parties is not a reason for departing from the natural language. Commercial common sense is only relevant to the extent of how matters would or could have been perceived by the parties, or by reasonable people in the position of the parties, as at the date that the contract was made. Judicial observations such as those of Lord Reid in Wickman Machine Tools Sales Ltd v L Schuler AG [1974] AC 235 , 251 and Lord Diplock in Antaios Cia Naviera SA v Salen Rederierna AB (The Antaios) [1985] AC 191 , 201, quoted by Lord Carnwath JSC at para 110, have to be read and applied bearing that important point in mind.
    [20] Fourthly, while commercial common sense is a very important factor to take into account when interpreting a contract, a court should be very slow to reject the natural meaning of a provision as correct simply because it appears to be a very imprudent term for one of the parties to have agreed, even ignoring the benefit of wisdom of hindsight. The purpose of interpretation is to identify what the parties have agreed, not what the court thinks that they should have agreed. Experience shows that it is by no means unknown for people to enter into arrangements which are ill-advised, even ignoring the benefit of wisdom of hindsight, and it is not the function of a court when interpreting an agreement to relieve a party from the consequences of his imprudence or poor advice. Accordingly, when interpreting a contract a judge should avoid re-writing it in an attempt to assist an unwise party or to penalise an astute party.
    [21] The fifth point concerns the facts known to the parties. When interpreting a contractual provision, one can only take into account facts or circumstances which existed at the time that the contract was made, and which were known or reasonably available to both parties. Given that a contract is a bilateral, or synallagmatic, arrangement involving both parties, it cannot be right, when interpreting a contractual provision, to take into account a fact or circumstance known only to one of the parties.
    [22] Sixthly, in some cases, an event subsequently occurs which was plainly not intended or contemplated by the parties, judging from the language of their contract. In such a case, if it is clear what the parties would have intended, the court will give effect to that intention.
  242. In Wood v Capita, Lord Hodge set out the applicable principles following Rainy Sky and Arnold v Britton as follows:
  243. "[10] The court's task is to ascertain the objective meaning of the language which the parties have chosen to express their agreement. It has long been accepted that this is not a literalist exercise focused solely on a parsing of the wording of the particular clause but that the court must consider the contract as a whole and, depending on the nature, formality and quality of drafting of the contract, give more or less weight to elements of the wider context in reaching its view as to that objective meaning. In Prenn v Simmonds [1971] 1 WLR 1381, 1383H-1385D and in Reardon Smith Line Ltd v Yngvar Hansen-Tangen (trading as HE Hansen – Tangen) [1998] 1 WRL 896, 912-913 Lord Hoffmann reformulated the principles of contractual interpretation, some saw his second principle, which allowed consideration of the whole relevant factual background available to the parties at the time of the contract, as signalling a break with the past. But Lord Bingham of Cornhill in an extrajudicial writing, "A New Thing Under the Sun? The Interpretation of Contracts and the ICS decision" (2008) 12 Edin LR 374, persuasively demonstrated that the idea of the court putting itself in the shoes of the contracting parties had a long pedigree.
    [11] Lord Clarke of Stone-cum-Ebony JSC elegantly summarised the approach to construction in the Rainy Sky case [2011] 1 WLR 2900, para 21f. In the Arnold case [2015] AC 1619 all of the judgments confirmed the approach in the Rainy Sky case: Lord Neuberger of Abbotsbury PSC, paras 13-14; Lord Hodge JSC, para 76 and Lord Carnwath JSC, para 108. Interpretation is, as Lord Clarke JSC stated in the Rainy Sky case (para 21), a unitary exercise; where there are rival meanings, the court can give weight to the implications of rival constructions by reaching a view as to which construction is more consistent with business common sense. But, in striking a balance between the indications given by the language and the implications of the competing constructions the court must consider the quality of drafting of the clause (the Rainy Sky case, para 26, citing Mance LJ in Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] 2 All ER (Comm) 299, paras 13, 16); and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest: the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms.
    [12] This unitary exercise involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated: the Arnold case, para 77 citing In re Sigma Finance Corpn [2010] 1 All ER 571, para 12, per Lord Mance JSC. To my mind once one has read the language in dispute and the relevant parts of the contract that provide its context, it does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each.
    [13] Textualism and contextualism are not conflicting paradigms in a battle for exclusive occupation of the field of contractual interpretation. Rather, the lawyer and the judge, when interpreting any contract, can use them as tools to ascertain the objective meaning of the language which the parties have chosen to express their agreement. The extent to which each tool will assist the court in its task will vary according to the circumstances of the particular agreement or agreements. Some agreements may be successfully interpreted principally by textual analysis, for example because of their sophistication and complexity and because they have been negotiated and prepared with the assistance of skilled professionals. The correct interpretation of other contracts may be achieved by a greater emphasis on the factual matrix, for example because of their informality, brevity or the absence of skilled professional assistance. But negotiators of complex formal contracts may often not achieve a logical and coherent text because of, for example, the conflicting aims of the parties, failures of communication, differing drafting practices, or deadlines which require the parties to compromise in order to reach agreement. There may often therefore be provisions in a detailed professionally drawn contract which lack clarity and the lawyer or judge in interpreting such provisions may be particularly helped by considering the factual matrix and the purpose of similar provisions in contracts of the same type. The iterative process, of which Lord Mance JSC spoke in Sigma Finance Corpn [2010] 1 ALL ER 571, para 12, assists the lawyer or judge to ascertain the objective meaning of the disputed provisions."
  244. These authorities show the importance of considering the wording of a particular clause in its contractual context. The decision of Sir Ross Cranston in Engelhart CIP (US) LLC v Lloyd's Syndicate 1221 [2018] EWHC 900 (Comm) illustrates this principle in the context of a policy which has some similarities to the present, albeit without the TPC. There, as here, the policy was of a "type" described as "Marine Cargo and Storage Insurance". The claimant had suffered loss, having been defrauded into taking up documents of title for copper ingots, which did not in fact exist. The claim failed. Neither the policy as a whole, nor the specific clauses relied upon, could be construed as providing cover for what were described as "paper" losses, in contrast to physical loss and damage to actual cargo. In that case, there was no cargo physically lost or damaged, because it had never existed in the first place. The claimant's losses were economic losses due to the acceptance of fraudulent documents in the expectation that they covered physical goods.
  245. Sir Ross Cranston said (at [39] – [41]) that the authorities required one to start "from the purpose of all risks marine cargo insurance, which is to cover loss of or damage to property". He said at [40]:
  246. "Since an all risks marine cargo policy is generally construed as covering only losses flowing from physical loss and damage to goods, there must be clear words indicating a broader intention".
    He went on to describe this as a presumption:
    "Thus, the commercial context of the construction exercise is that the presumption with an all risks marine cargo policy is to insure for physical losses".
  247. He then concluded that the policy as a whole did not displace that presumption, and that neither of the specific clauses relied upon meant that paper losses or fictitious goods were included. His conclusion was that whilst the policy was broader than the standard wording of the Institute Cargo Clauses, it did not cover the loss claimed:
  248. "If the parties had intended to cover loss of this character, they would have appreciated the need for clear words before a policy covering physical losses can be read to cover as well non-physical losses".
  249. The Bank accepted, in the light of this authority, that clear words would therefore be required if the policy in the present case were to be construed to extend to the financial (ie non-physical) losses which they claimed. They argued that the language of the TPC provided language which was indeed sufficiently clear. I consider that this encapsulates the central issue of construction in this case, bearing in mind of course the need to consider the relevant language in the light of the various matters identified in paragraph [15] of Arnold v Britton.
  250. There was some debate as to whether it is appropriate to speak in terms of a "presumption". I consider that this is simply another way of expressing the idea that if a policy such as the present, or a particular clause therein, is to be construed as extending beyond physical loss and damage, there needs to be contractual language which clearly so provides. Mr. Parsons in his oral closing said that the relevant proposition to be found in Engelhart is that clear words are needed "if you're going beyond the normal risk contemplated in the marine cargo market". In my view, however, it is simpler to say that, as Sir Ross Cranston's judgment indicates, there must be clear words if the policy is to be construed as covering losses other than physical loss or damage to the goods.
  251. The insurers referred in their submissions to a number of paragraphs in MacGillivray on Insurance Law 14th Edition, where general principles of interpretation are set out. For example, MacGillivray states:
  252. "[11-006] There is a presumption that the words to be construed should be construed in their ordinary and popular sense, since the parties to the contract must be taken to have intended, as reasonable men, to use words and phrases in their commonly understood and accepted sense. This presumption can be rebutted in certain circumstances which are examined later in this chapter, but it is frequently the case that there is no reason to depart from the ordinary meaning of the words in question.
    [11-007] It is an accepted canon of construction that a commercial document, such as an insurance policy, should be construed in accordance with sound commercial principles and good business sense, so that its provisions receive a fair and sensible application. Several consequences flow from the principle. The literal meaning of words must not be permitted to prevail where it would produce an unrealistic and generally unanticipated result, as, for example, where it would unwarrantably reduce the cover which it was the purpose of the policy to afford.
    [11-008] It follows that in interpreting any clause of a policy, it is correct to bear in mind: (1) the commercial object or purpose of the contract; and (2) the purpose or function of the clause and its apparent relation to the contract as a whole. It may then become apparent that the literal meaning of the clause must yield to business sense or that an ambiguity in the wording can be resolved, or the ordinary meaning of the words used may need to be modified.
    [11-009] If a literal reading of the word leads to an absurd result or one manifestly contrary to the real intention of the parties, it should be rejected in favour of a more reasonable interpretation if that can be adopted without doing violence to the words used."
  253. I do not consider that these passages add materially to the statements of principle in the cases set out above.
  254. The parties' submissions also referred to various authorities concerning the approach to "surplusage" in commercial contracts. This is potentially relevant because the Bank and Edge contend that the insurers' construction of the TPC would mean that it was devoid of any practical effect, in the light of the "Basis of Valuation" provision. The insurers did not accept that this was so, but in any event argued that insurance policies often contain surplus words and any presumption against surplusage was weak. Thus, Arnould states at paragraph 3-37:
  255. "Policy terms may on occasion be redundant, perhaps because they have been inserted from an abundance of caution. It has been recognised that redundancy is commonly found in insurance wordings and that this should not affect the otherwise natural construction of the words used. Indeed, it has been said in a number of more recent cases that the presumption against surplusage is of little weight when it comes to construing commercial contracts generally".
  256. In Tektrol Ltd. v International Insurance Co. of Hanover Ltd. [2005] EWCA Civ 845, the court was concerned with the construction of an exclusion clause in a policy. The insurers argued that the particular word "loss" would be redundant if the insured's argument were accepted. Buxton LJ (with whom Nourse LJ agreed) said that this would be to attribute to the draftsman too precise a use of language. He referred to statements by Lord Hoffmann that "draftsmen traditionally employ linguistic overkill and try to obliterate the conceptual target by using a number of phrases expressing more or less the same idea". In Swallowfalls Ltd. v Monaco Yachting & Technologies SAM [2014] EWCA Civ 186, para [27], Longmore LJ said that arguments of surplusage were "not particularly compelling in commercial contracts".
  257. However, as HHJ Peter Coulson QC (as he then was) said in Jani King (GB) Ltd. v Pula Enterprises [2007] EWHC 2433 (QBD) at [26], a court should always think long and hard before arriving at a construction which renders otiose a part of the written agreement, although there is a good deal of modern authority to the effect that this presumption against surplusage is relatively weak. In Secretary of State for Defence v Turner Estate Solutions Ltd. [2015] EWHC 1150 (TCC) at [62], Coulson J said:
  258. "One of the canons of construction is that, in order to arrive at the true interpretation of a document, a condition must not be considered in isolation but in the context of the document as a whole. A related principle is that the court should always lean towards a construction that validates the contract, on the basis that the parties are unlikely to have intended to agree to something that was legally ineffective... And perhaps the most important canon of construction for the purposes of the present case is that, in construing a contract, all parts of it must be given effect where possible, and no part of it should be treated as inoperative or surplus. Whilst the presumption against surplusage is unlikely to be useful in interpreting a standard form of contract ... this is not of course a standard form contract but a bespoke contract carefully drafted by the parties to meet the exigencies of this particular and significant commercial arrangement".

    D2: The parties' arguments

    The Bank's argument

  259. The Bank's argument started from, and focused upon, the language of the clause. Ms. Sabben-Clare accepted that any add-on to the ordinary physical loss and damage cover provided by a marine policy would need to be clearly expressed. She submitted that the present clause was clearly expressed. As a matter of language, the TPC was clearly an insuring clause. It was clearly intended to add something of substance to the existing heads of cover. This was particularly so in the final version of the wording, where the TPC appears in two places, including in a standalone section. The Bank's primary case, however, was that the TPC meant the same wherever it was located in the policy, and whether it appeared once or twice.
  260. As a matter of construction, the Bank submitted that cover was provided for the "Transaction Premium" that would otherwise have been received or earned in the absence of "Default". What was insured was the "Transaction" and specifically the "Transaction Premium". The insured peril was "Default", and this was defined in the widest terms. There was no suggestion that the Default needed to have anything to do with loss or damage to the cargo. The coverage was tied to a Default, and that was not a concept dependent on the presence of physical loss or damage. Default was not about physical loss or damage. This was clear from, and emphasised by, the words "for any reason" which formed part of the definition of Default. It was apparent on the face of the TPC that this was a carefully drafted contract. The parties must be taken to have meant what they said: the court should not depart from the objective meaning of the words used.
  261. The contrary argument, that the TPC was concerned with the basis of valuation where there had been physical loss and damage, should be rejected. It was not supported by the language of the clause. There was, separately, a Basis of Valuation clause in the policy. That clause was sufficiently widely worded to cover losses which, on the underwriters' case, were catered for by the TPC. If, therefore, the TPC was no more than a Basis of Valuation clause, it added nothing to the clause that was already there. The court should reject the argument that a carefully-drafted clause added nothing material to the scope of cover.
  262. The TPC made no reference to physical loss and damage, and it was difficult to see how physical loss and damage could have a sensible and meaningful role to play. The clause was directed to a situation where the client had defaulted and Icestar had to sell the goods on an exchange or on the open market. If the cargo was lost, destroyed or severely damaged, such a sale on the exchange or the open market could not happen. The result would be that the TPC only applied to cases of partial loss or damage to the commodity, but not to cases of total loss or damage. It was difficult to see how commercial parties could have intended their bespoke and detailed provision to be inapplicable to the most serious heads of loss. The result would be that the more serious the loss or damage, the less the clause applied.
  263. Furthermore, the TPC framed the entitlement to an indemnity in terms of the amount that would have been received "in the absence of a Default". These words could not be explained, coherently, as simply going to a basis of valuation.
  264. There was, Ms. Sabben-Clare submitted, nothing in the context of the policy or the factual matrix which militated against this construction. The policy was not on any view a policy which only provided cover against physical loss and damage: there were "add-ons" to be found in the clauses which covered business contingency losses, CEND and fraudulent documentation. These extensions had been provided without any adjustment to the premium ratings under the policy. There was no rule of law which meant that add-ons were not permissible: it was simply an issue of construction. The broking experts had agreed that add-ons were generally available in the London market at the material times. At the time, the Bank was regarded as a good risk: the premiums payable under the policy were high, and there was a good claims record. The policy was oversubscribed with the lines of some underwriters being "signed down". The market was generally "soft" and difficult from underwriters' perspective. There were therefore good reasons why the underwriters might have been prepared to provide what was in effect credit risk cover as a free add-on. Overall, however, the factual matrix was not and could not be decisive. There were factors which pointed in different directions, and ultimately the question was: what did the parties agree here?
  265. Nor did commercial considerations mean that a different approach to construction should be taken. The Bank accepted that marine cargo insurance was normally a different class of business from credit risk insurance, and that this was an important part of the factual matrix. But this fact could not be allowed to have preclusive or dogmatic effect. Add-ons to standard PLOD cover were common in the market, and there is no reason why such an add-on could not give protection for financial default. The cover related to precisely the same cargoes that were already being declared and insured. The insurers' argument that they would thereby be exposed to billions of dollars of exposure did not assist in the task of contractual interpretation. Underwriters agreed to whatever they objectively agreed, and it is not the court's function to protect them from a commercial decision that seems unwise. In any event, the possibility of defaults of all of the Bank's clients – including in relation to Icestar 1 transactions where the real risk was upon the exchange – was highly improbable. The realistic exposure was therefore, as here, to the default of a smaller client, and it was difficult to envisage a claim which significantly exceeded the US$ 45m claimed in the present case. In any event, Arnold v Britton showed the importance of not using the concept of commercial common-sense in order to relieve a party from a contract which was a bad idea.
  266. Edge's argument

  267. Edge advanced largely the same case as the Bank on contract construction. Edge did not accept that there was any presumption that marine cargo insurance only provided cover against physical loss and damage in the absence of clear terms to the contrary, not least because the present policy provided the "add-on" covers which were not dependent on physical loss and damage. Edge emphasised the absence of any express language in the TPC which supported the underwriters' argument. Ms. Healy also emphasised the "for whatever reason" wording in the definition of Default. In its opening written submissions, Edge also identified a number of different scenarios where there might be loss and damage to the cargo. In none of these situations would the TPC, if construed simply as providing a basis of valuation, add to the existing terms. In the case of destruction of the goods, the TPC would be inapplicable in its entirety. There was nothing outlandish in construing the cover as Edge and the Bank suggested: the underwriters were already insuring the cargoes that were the subject of the repo transactions.
  268. In their written closing, Edge submitted that this is an archetypal case of a party seeking to escape from a bargain that has proved unfavourable. But for the demise of Transmar and Euromar, the clause would probably still be sitting peacefully in the Bank's policy. Arnold v Britton provided a complete answer to the underwriters' arguments as to commercial absurdity or common-sense. The construction of the policy required meaning to be given to the word "Default" which was used on a number of occasions in the TPC. If, however, there was loss and damage to the cargo, the Bank would already have cover for these events. The fact that a counterparty had committed a default subsequent to the loss and damage would add nothing. In her oral submissions, Ms. Healy submitted that the underwriters had failed to provide any satisfactory explanation as to why the word Default appeared three times in the TPC, and what it added.
  269. The underwriters' argument

  270. The underwriters submitted that nowhere does the TPC actually say in clear terms or at all that it confers cover that is independent of physical loss and damage. Therefore it did not overcome the well-established presumption that marine cargo insurance is limited to such loss. Clear words were required if the cover was to have any greater scope. At various stages in his submissions, Mr. Parsons suggested that the clarity of language required would only be fulfilled if there were words which said, expressly, that the cover was to apply in the absence of physical loss or damage. He said that, in the context of the factual matrix on which underwriters relied, only the "clearest possible words with express signposts, such as financial guarantee or trade credit, would be sufficient to displace the objective view" that cover in the nature of trade credit or financial guarantee was not being provided. The policy could easily have stated something like: "this insurance is not contingent on physical loss or damage". Overall, the relevant question is whether the clause contained clear words to extend cover beyond physical loss and damage. The answer was no.
  271. The underwriters relied upon the language of the policy as a whole. They submitted that the commercial context of the construction exercise is the presumption that it is to insure for physical risks only. Mr. Parsons referred, in opening and closing, to various provisions within the policy which were clearly referable to cover for physical loss and damage. In the present case, he submitted in opening, it was not simply a question of there being a presumption that the policy was limited to physical loss and damage. The policy contained an express provision in the opening section: "This contract is to protect against all risks of physical loss or damage to the Subject Matter Insured from whatsoever cause arising". This was therefore an express limit on the scope of cover.
  272. By the time of their oral closing submissions, however, the underwriters had modified that argument to some degree. Mr. Parsons submitted that the TPC provided cover for the transaction premium that would have been earned but for a default "following the operation of an insured peril identified elsewhere in the marine cargo and storage policy". A reasonable person in the marine cargo market would read the words as identifying the "head of loss, measure of indemnity or basis of valuation" (these expressions were interchangeable) "following a peril elsewhere". The argument therefore accepted that the TPC was not confined to recovery where there had been physical loss and damage, but would also be applicable (for example) if there was a claim arising under the CEND clause or the Business Contingency Cover. The reason for this movement in the underwriters' position was, at least as I perceived it, that if the TPC was indeed a basis of valuation clause, then it was logical to say that it was a basis of valuation for all of the risks covered under the policy, rather than simply for the risk of physical loss and damage.
  273. However, whether or not the TPC was confined to physical loss or damage, or could be extended so as to be applicable to other risks, the underwriters contended that the language could not sensibly be construed as conferring credit risk cover. In the unamended policy, it appeared at Clause 1.5 under the general heading of "Definitions". In that context, it could have meaning as a basis of valuation clause. Whilst the submission as to why the TPC was a basis of valuation clause was undeveloped in the underwriters' written opening, that case was explained to some degree in Mr. Parsons' oral opening, and ultimately articulated with clarity in his written and oral closings. In summary, the underwriters submitted that if there was physical loss and damage to the goods, and then for whatever reason the counterparty failed or refused or did not exercise the option, then the TPC would provide some additional cover. Specifically, this would provide cover for the "market premium" which had been originally identified in Endorsement No. 3 to the policy signed by Mr. Beattie in February 2015. This was defined in that Endorsement, and subsequently in Endorsement No. 4, as the "difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents". The language of the TPC enabled a recovery to be made based on the difference between what the policyholder had paid, and what the goods were worth. This would encompass this element of "market premium", but there would still need to be physical loss or damage. The TPC in this way ensured that the Bank would not be out of pocket, and could recover its entire loss, in the event of physical loss of or damage to the goods. Reasonable people in the position of the parties would have understood that the TPC was simply providing clarification in order to dispel any doubt as to whether the "market premium" was within the scope of the Basis of Valuation clause.
  274. The underwriters placed much emphasis on the factual matrix, and the need to construe the contract in that context. The key features of the factual matrix relied upon by the underwriters in opening were: all the underwriters, including Mr. Beattie, were marine cargo underwriters; whilst the policy did provide some cover for non-physical losses, these were all adjuncts to an all-risks cargo policy rather than a separate head of credit risk cover; Mr. Beattie and the other underwriters were not licensed or authorised to write credit risks, as Mr. Mullen must have known; credit risks are a separate line of cover with their own specialist insurers; the inclusion of credit risk cover would open up a whole raft of underwriting considerations, none of which were discussed at the time of placement, as well as a new raft of areas for disclosure.
  275. By the time of its closing submissions, after the extensive evidence in the case had been called, the factual matrix relied upon by the underwriters had expanded to a considerable degree, certainly in its detail if not its essentials. Reliance was placed by the underwriters on various answers Mr. Mullen had given as to how he perceived the cargo market and risk appetite of underwriters at the time. Furthermore, underwriters had asked about the Bank's due diligence procedures. They were provided with documents relating to the standard of warehouses. This was obviously relevant to ordinary storage risks. No information relevant to default risks was provided.
  276. Mr. Parsons emphasised that a cargo and storage policy does not traditionally cover credit risks, financial guarantees, the fact that the insured has entered into a bad bargain, the fact that the goods were of poor quality at the time of purchase, the risk of goods deteriorating over time, or changes in market value. One would therefore not expect underwriters to insure the risks attaching essentially to the value of the collateral which the Bank had taken from its counterparty.
  277. In their written and oral closing submissions, the underwriters placed emphasis on what they regarded as the commercially absurd results of the construction advocated by the Bank and Edge. They identified, based on Mr. Stroink's evidence as to the value of the Bank's repo transactions, a total potential exposure of around US$ 2.5 billion, and pointed to the absurdity of this cover having been provided for free. It would have been difficult to obtain this cover even from specialist underwriters in the credit risk market. Had it been available, trade credit underwriters would have ensured that terms were imposed which circumscribed their potential liability. If the Bank's argument succeeded, they would receive a "windfall"; because the only thing that the Bank had really wanted to do was to cover the "market premium" which had been the subject of Endorsements 3 and 4. If the cover was extended by underwriters in the manner suggested by the Bank, then that extension of cover would have been entirely unwitting. The RSA did not write credit risks and nor did any of the other underwriters.
  278. D3: Discussion

    Factual matrix

  279. The nature of the defences advanced in the present case – including rectification, estoppel, and avoidance because of an unfair presentation of the risk – has meant that a very large volume of evidence, both factual and expert, has been adduced which is inadmissible on issues of construction. The arguments of the parties, particularly those of the insurers, have dipped liberally into this evidence in support of their respective contentions, including as to the relevant factual matrix. For example, whilst acknowledging that subjective intention is not relevant, the insurers relied in their written closing on evidence of Mr. Beattie, Mr. Mullen and the experts, such as how they reacted when they first read the TPC, and their views as to the clarity of the wording.
  280. There is an important and salutary rule which was introduced into the Commercial Court Guide some years ago, no doubt because of the difficulty in keeping evidence bearing on the construction of a written agreement within appropriate manageable bounds. Paragraph C1.3 (h) provides:
  281. "Where proceedings involve issues of construction of a document in relation to which a party wishes to contend that there is a relevant factual matrix that party should specifically set out in its statement of case each feature of the matrix which is alleged to be of relevance. The "factual matrix" means the background knowledge which would reasonably have been available to the parties in the situation in which they found themselves at the time of the contract/ document".
  282. In the present case, the parties' respective statements of case did not comply with the Commercial Court Guide in this respect, certainly when they are compared with the evidential facts and background which were ultimately relied upon. However, no contract should be interpreted in a vacuum, and it would not be appropriate to resolve the construction of the TPC without regard to the admissible and relevant factual matrix. However, in addressing issues of construction, I put aside the views expressed by witnesses, both factual and expert, as to their understanding of the clause or its clarity. Nor is it appropriate to take into account what the Bank, or NRF, was seeking to achieve by the TPC. I therefore do not need to resolve the dispute as to whether (as the underwriters suggested) NRF misunderstood what the Bank was seeking to achieve, or the related argument that the Bank's case would involve them receiving a windfall. (I will, however, return to that matter below). Furthermore, whilst I consider that the nature of the market in which the parties were operating is relevant and admissible factual matrix evidence, I was not persuaded that Mr. Mullen's subjective views as to the risk appetite of underwriters, and other aspects of Mr. Mullen's subjective perception of the risk or the wording, is relevant and admissible. I should say, however, that having listened to Mr. Mullen's evidence over a number of days, I do not think that he subjectively appreciated that the cover that he was being asked to place was beyond the risk appetite or authority of the underwriters that he approached. Nor did he subjectively analyse the TPC as providing "trade credit" insurance, although it is clear that he did take some steps, in July 2015, to understand the risk that he was being asked to place. He did not, however, regard it as his part of his role, unless asked, to provide an explanation of the wording to Mr. Beattie, who could read the wording for himself. I do not consider that evidence, as to the way in which Mr. Mullen was or may have been thinking, was of any relevance to the construction of the TPC. Nor, of course, are the disputed discussions which are alleged to have taken place between Mr. Mullen and Mr. Beattie.
  283. Which aspects of the factual matrix are important? In my view, the most significant aspect of the factual matrix, against which the policy and the TPC falls to be construed, concerns the nature of the market in which Mr. Beattie and Mr. Mullen were operating. That market was, in broad terms, the London marine market where there were underwriters who specialised in writing insurance on cargo risks. There were other underwriters in that market who specialised in other aspects of marine business, such as hull or war risks. There is no doubt that the core business of the cargo underwriters is the writing of risks concerning physical loss and damage to cargo whilst in transit or in store. However, the underwriting experts agreed that coverages which do not require physical loss or damage were generally available in the London marine cargo market in 2015/16 as an add-on to marine cargo and storage policies.
  284. This evidence from the underwriting experts was borne out by the policy with which I am concerned. Leaving aside the disputed issues as to the interpretation of the TPC, there was ultimately no dispute in the present case that the policy did contain three separate "add-ons" which did not require physical loss or damage. These were the cover for delay under the "Business contingency cover" (or "BCC") in Clause 20 of Section 2 of the policy, the cover for "Fraudulent documentation" in Clause 17 of the same section, and the CEND cover in Section 3.
  285. In relation to the "Fraudulent documentation" cover, the 2015 policy contained wording (in Clause 17.3) which provided an indemnity for "direct financial loss" suffered by reason of the insured, in good faith and in the ordinary course of business, acting upon a counterfeit document. This wording was itself expanded in the July 2015 endorsement, and then incorporated in its expanded form into the policy subscribed by all underwriters in January/February 2016. Its effect, in broad terms, was to provide cover in circumstances where the Bank had been the victim of a fraud. The expansion of the cover in 2015/2016 reflected a widely-known problem at that time, referred to at the trial as Qingdao, where a company deployed fraudulent duplicate warehouse receipts to raise trade finance secured against the same stockpile of cargo. Multiple versions of the warehouse receipts were fraudulently created for the same parcel of goods, leaving numerous insureds suffering loss in the absence of any physical loss or damage or theft of the goods.
  286. In addition to these add-ons which were contained in the wording in the present policy, there was evidence from some underwriters as to another type of "non-PLOD" add-on offered by underwriters in the London cargo market. (As I have said, the acronyms PLOD and non-PLOD were used in the trial to denote cover requiring or not requiring Physical Loss or Damage. It is convenient sometimes to use them in this judgment). This non-PLOD add-on concerned financial risks on what were described as "project" cargo policies. Mr. Burns, formerly at Prosight (the 13th Defendant) said that he wrote consequential loss cover for such cargoes.
  287. In both July 2015 and January 2016, the evidence showed that the market was soft. An internal document of Advent (the 10th Defendant) referred to an incredibly difficult market-place. There was, clearly, overcapacity, with underwriters straining to attract and retain business. An internal RSA document from late 2014/early 2015 described the most "extreme price competition we have ever experienced". Mr. Cooke, the underwriter at Advent, described a market in which some underwriters were writing for income, slapping lines down as much as they could on anything and everything. Mr. Mullen described it as a "wee bit of a circus".
  288. Even though the market did offer add-ons, and even though the market was soft, it is right to acknowledge that there was, at the time, no precedent for marine cargo underwriters adding, to a marine cargo policy, cover which would protect the insured in respect of the contractual default of a counterparty leading to a non-physical loss on cargoes which the insured had purchased. If, therefore, the present policy so provides, then this is the first and (probably) only example of such a policy having been written in the marine market. None of the witnesses, factual or expert, had seen such an add-on before.
  289. One clause of the policy (on which attention was focused for the first time in closing arguments) did expressly contemplate a default in particular circumstances, but expressly where physical loss and damage took place. Clause 9 in Section 1 of the policy provided for "Sellers Insurance interest" at a rate of 40% of the cover rate listed in the policy's rating schedule. This would provide cover where the insured had sold goods on an FOB or CFR basis. It provided
  290. "Sellers Insurance interest clause at 40% of cover rate as listed in the Rating Schedule.
    …
    ii) claims in respect of loss of or damage to FOB Goods and/or CFR shall be payable only if and to the extent that the buyer fails to pay for such loss or damage;
    iii) Underwriters to be subrogated to Insured's right against the buyer as well as other parties in accordance with clause 5 (Subrogation) of section 2 of this contract
    iv) this contract shall not be divulged to the buyer"
  291. This clause therefore contemplated that, in a case where there was physical loss or damage, the insured's buyer might fail to pay for such loss or damage, presumably by failing to pay the full contract value. In such circumstances, the policyholder could recover, but with the underwriters being able, via rights of subrogation, to pursue the buyer. The clause is therefore some considerable distance from the TPC, as contended for by the Bank and Edge. Whilst it did refer to a buyer's failure to pay (an example of contractual default), the cover was expressly dependent on PLOD.
  292. The admissible factual matrix also indicated that there were, at Lloyd's and generally in London, underwriters who specialised in writing trade credit insurance, which is a type of financial guarantee business. The issues arising on the Bank's claim for damages against Edge meant that there was a considerable volume of expert evidence, from Mr. Hayter and Mr. Hedley, as to the approach taken by these specialist underwriters to the writing of risks. This evidence included both the rating approach and the terms which might typically be included. Whilst participants in the marine cargo market, both on the underwriting and broking side, might know at a very general level of the existence of these specialist underwriters (ie a market for trade credit risks), the evidence suggested that they would not know the detail of the risks that were being written, or the rating or other approach taken by these specialists. Nor would they readily know the materials which would typically be considered by trade credit underwriters in forming a view as to whether a particular risk was acceptable. Information about the trade credit market, including matters relating to the underwriting approach, could no doubt have been obtained by underwriters or brokers, via contacts or colleagues, if they were interested in finding anything out. To that extent, such information was reasonably available to the parties and could be said to form part of the factual matrix. Ordinarily, however, the business of marine cargo underwriters would not require them to do so, since their policies did not ordinarily include risks of that nature. I do not therefore consider that the evidence as to how trade credit insurers approached the writing of these risks is of any real assistance in deciding how the TPC is to be construed.
  293. As with any market-place, a participant might expect other participants to seek to do things which were new. As Mr. Cooke said, describing the history of his career, brokers had been trying to slip in the odd wording here and there to give them an edge in the market place. That was, he said, quite common or at least it happened from time to time, and a soft market would be an opportune time to "introduce their own slant on things".
  294. As far as the TPC is concerned, the clause was therefore a non-standard wording. Indeed, the evidence indicated that none of the underwriters in the cargo market, or indeed the expert underwriting and broking experts familiar with the marine market, had seen a clause like it. It was therefore, very clearly, bespoke wording.
  295. There was also no dispute that Mr. Beattie was told, in the July 2015 meeting, that the clause had been drafted by the Bank's lawyers – although he was not told that the lawyers concerned were NRF, because Mr. Mullen himself did not know that. None of the other underwriters was expressly told on the January 2016 renewal about the involvement of lawyers. However, the structure and drafting of the TPC, with definitions highlighted in bold text and obviously careful drafting by reference to those definitions, would have suggested the likelihood that there had been legal input into the drafting of this clause.
  296. These seem to me to be the most important aspects of the factual matrix against which the wording of the TPC is to be construed. Ultimately, as discussed below, I considered that the factual matrix served to underline the need, discussed in Engelhart, for there to be clear words before the cover could be held to extend beyond PLOD in the manner for which the Bank and Edge contended. If such clear language exists, then (as apparent from the passages quoted below from Lewison: The Interpretation of Contracts, 7th edition, paragraph 3.168), the background should not be used to create an ambiguity where none exists.
  297. Endorsements 3 and 4

  298. It is convenient to say something at this stage about Endorsements 3 and 4, upon which the insurers placed heavy reliance as a relevant part of the factual matrix; in particular in support of their argument that, in essence, the TPC was really aimed at ensuring that there was cover for the "market premium" which had been addressed by these two endorsements. In a very broad sense it can be said that these two Endorsements formed part of the factual background; because there is, as Lord Hoffmann said in BCCI v Ali [2002] 1 AC 251, no conceptual limit to what can be regarded as background; and because, as Lord Hoffmann said in Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896, the admissible matrix includes "absolutely anything which would have affected the way in which the language of the document would have been understood by a reasonable person".
  299. I do not think, however, that the interpretation of the TPC is in any way assisted by the language or interpretation of Endorsements 3 and 4. The TPC was first agreed (as I find in Section E below) in July 2015 by Mr. Beattie. The July 2015 endorsement (or "wish-list" as the underwriters described it) made significant changes to both Endorsements 3 and 4. Endorsement 3, prior to the July 2015 document, was in the following terms:
  300. "Underwriters note and agree that for the purposes of the Business Contingency Cover, it is hereby confirmed that the insured is covered under such cover for all costs and expenses incurred by the Bank as set out in that clause, including but not limited to any market premium paid by the Insured under the relevant transaction ("market premium"), being the difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents".
  301. Endorsement 3 was substantially revised in July 2015, so that it read as follows. I have reproduced all the additions and deletions. Some, but not all, of these changes were shown by underlining in the July document itself. I have shown all these changes, although the deletion of the concluding wording (after the first reference to "market premium") was not shown in the July document itself. The amended clause was preceded by the underlined words: "The contract change made under Endorsements 3 is amended and restated as follows". It provided:
  302. "Underwriters note and agree that for the purposes of the Business Contingency Cover provided under Clause 20 of this contract, it is hereby confirmed that the insured is covered under such cover for all costs and expenses and losses incurred by the Bank as set out in that clause, including but not limited to any premium or profit element that the Insured would otherwise have earned but for the delayed delivery of the Subject Matter Insured to the Exchange in the related Futures Month market premium paid by the Insured under the relevant transaction ("market premium"), being the difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents ".
  303. Endorsement 4, prior to July 2015, was as follows:
  304. "Notwithstanding any other term to the contrary in this policy, it is hereby confirmed that the Business Contingency Policy also covers (in addition to the cover set out therein) any loss suffered by the Insured representing any market premium paid by the Insured under a transaction ("market premium" being the difference between the exchange quoted price and the physical market price at purchase as specified on the relevant invoice or in the relevant transaction documents)".
  305. In the July 2015 endorsement, the TPC was introduced with the following words which were underlined
  306. "The contract change made under Endorsement 4 is deleted and replaced with the following clause".
  307. There is in my view nothing here which can reasonably assist in the construction of the TPC. Just over 5 lines of text in Endorsement 4 were replaced with 20 lines of text in the TPC, including 5 new defined terms. The TPC, as expressed in the July endorsement and ultimately in the 2016 policy, therefore used different concepts and definitions to those used in the two earlier endorsements. (This is not surprising, since NRF had subjected those endorsements to careful scrutiny and amendment). In particular, it introduced for the first time the concept of "Default".
  308. Moreover, the expression "market premium", which had appeared in both Endorsement 3 and 4, was not used in the TPC. Rather, the TPC used the concept of a "Transaction" – which was defined so as to include a requirement for a Default – as well as "Transaction Premium". The definition of the latter was very different to "market premium" as defined in the two earlier endorsements. In particular, it was not related to the exchange quoted price, but rather was defined (via the definition of "Actual Sale Price") by reference to the sum received on sale either on the applicable exchange "or to a third party on the open market".
  309. A further fundamental difference is that Endorsement 4 had referred, and referred only, to the "Business Contingency Policy": ie the BCC. That was, as the underwriters correctly argued, the clause in the policy which addressed losses consequent upon delay in the delivery of goods to an exchange. The TPC on any view went much further. Indeed, it was underwriters' contention that it applied when there was physical loss and damage to the cargo, whereas (as underwriters argued and I accept) the original Endorsement 4 on its true construction was limited to the BCC.
  310. In these circumstances the terms of the original Endorsement 4, which was deleted in its entirety and replaced by an entirely different clause (the TPC), do not provide any guide or assistance in the interpretation of the TPC. Indeed, the more obvious conclusion is that the parties, in the TPC, were not aiming at the same target as with Endorsement 4. In particular, the foundation for the argument that the parties simply had in mind the concept of "market premium", as used in the earlier endorsements, is lacking. If the parties had simply been concerned with "market premium" as previously defined, that definition could have been retained and expanded upon if required.
  311. It is also relevant that Endorsement 3 was itself amended and restated. That Endorsement was, and remained, confined to the BCC, and – expressly – to "delayed delivery of the Subject Matter Insured to the Exchange". The definition of "market premium" was again dispensed with, to be replaced by the more general words "any premium or profit element". The 5 lines of text of the revised Endorsement 3 is again to be contrasted with the 21 lines of text of the TPC, the different terminology there used, and the fact that the TPC is not limited to cases of delay.
  312. Policy context

  313. The insurers referred to many provisions within the policy which indicate that the policy was intended to cover the risk of physical loss and damage to cargo. For example, the first words of the policy, with the side-heading "Type", are: "Marine Cargo and Storage Insurance". The parties incorporated the standard Institute Cargo Clauses 'A', and these are standard physical loss and damage terms. The policy contained provisions, such as those in Section 1 with the side headings "Location", "Per", "Shipments Clause" and "Storage Risks" which would typically be seen in an ordinary cargo policy covering physical loss and damage. The policy limits (which appeared twice, in the opening provisions and then again in Section 1) were referable to "any one Conveyance" (USD 50,000,000) and "any one individual warehouse, shed or storage facility" (USD 100,000,000). The rating schedule provided for a differential between transit and storage risks. In relation to soft commodities (such as the cocoa and cocoa products with which this case is concerned) the respective rates were 0.16% per voyage, and 0.13% for goods in store.
  314. It is unsurprising that there are many contractual provisions which dealt with the situation where there is physical loss and damage to the cargo; since there is no dispute that the parties intended to cover that risk in the usual way. However, it did not seem to me that reference to these provisions was of any great assistance in resolving the issue as to the construction of the TPC. The existence of clauses in the policy relating to physical loss and damage explains and provides a firm foundation for the principle, illustrated by Engelhart, that the starting point is that the policy covers physical loss and damage to the cargo, unless there are clear words which provide wider cover. However, it does not answer the question of whether a particular clause such as the TPC is to be interpreted as providing that wider cover. That must depend upon the true construction of the relevant clause, and the clarity of the words used.
  315. This approach is borne out by the terms of the policy in the present case. It was, ultimately, common ground that there were three policy provisions which did indeed provide wider cover; ie cover in the absence of physical loss and damage. These were the Business Contingency Cover, the CEND Cover and the Fraudulent Documentation clause. The insurers argued that an analogy could be drawn between the risks covered by these clauses and physical loss and damage, and that in any event it was not unusual to see these extensions in cargo policies. I agree that it can be said that there is some analogy, certainly in relation to Business Contingency Cover and the CEND Cover, with physical loss and damage. In both cases, the insured is deprived of the physical cargo for a period of time, and in that sense might be said to suffer a physical loss at least pro tem. But even there the analogy is not exact, because the cargo itself may be entirely unharmed. Many cargoes, such as metals, can survive lengthy storage conditions, and the evidence indicated that cocoa beans (as distinct from products) may survive in store for very many years.
  316. The analogy is inapt, or at least less persuasive, in relation to Clauses 17.2 and 17.3 in the Fraudulent Documentation clause. These clauses clearly go beyond the wording of the clauses considered in Engelhart. Clause 17.2 covered not only "physical loss of the related subject matter insured", but also "impairment of its interests in the subject matter insured". This wording appears to have been added in order to cover the situation in a case such as Qingdao. Furthermore, Clause 17.3 is also not confined to physical loss, but covers "direct financial loss suffered by reason of the impairment of the Insured's interest in the subject matter insured".
  317. In his closing submissions, Mr. Parsons correctly accepted that he could not shy away from the fact that these provisions provided a form of financial loss cover, and also that they exposed the insurer to the risk of losses flowing from a dishonest counterparty of the Bank, potentially affecting goods in a number of warehouses. He said, however, that the risks covered by these clauses were part of a maritime adventure, and flowed closely from something that was already covered by the policy. This was, he submitted, very different to trade credit or financial guarantee insurance.
  318. In my view, the existence of these three clauses shows, as the Bank and Edge argued, that the policy cannot be approached on the basis that it is nothing more than a "plain vanilla" policy covering physical loss and damage. Analogies with physical loss and damage may or may not be apt. But even if such an analogy can be drawn in some respects, that does not mean that the relevant provisions do not extend beyond physical loss and damage. Ultimately, I consider that one is driven back to careful consideration of the clarity of the words which the parties have actually used in the clause alleged to provide the relevant extension of cover beyond physical loss and damage.
  319. These three clauses also answer the reliance placed by the insurers on the second paragraph (underlined below) of the "General Conditions" clause in the opening section of the policy. This paragraph and its immediate context are:
  320. "GENERAL CONDITIONS:
    (Applicable to all sections of this contract)
    Notwithstanding anything contained herein to the contrary, it is agreed and understood in general terms that it is the intention of this contract to protect the interest of the Insured at all times and in all circumstances.
    This contract is to protect against all risks of physical loss of or damage to the Subject Matter Insured from whatsoever cause arising. (emphasis supplied)
    All other conditions are more fully detailed in each respective section and/or other parties, with the prior written consent of the Insured."
  321. Whilst the underlined part of the clause does refer expressly to all risks of physical loss and damage, the immediately following wording indicates that it is necessary to have regard to all the contractual terms in order to ascertain the scope of the cover. The BCC, CEND and Fraudulent documentation cover show that the cover is not concerned exclusively with physical loss and damage, and the above wording cannot be construed (as the underwriters at one stage suggested) as an express limitation of the policy as a whole to cover for risks of physical loss or damage.
  322. Ms. Healy drew attention to the first paragraph above, and its reference to the intention "to protect the interest of the Insured at all times and in all circumstances". Rightly, she did not suggest that these general words would cover the Bank's financial losses in the absence of the TPC. This is therefore not a case where the policyholder is relying on general words in order to provide unusual cover: compare Cheshire v Thompson (1919) 24 Com Cas 114 (Bailhache J) and (1919) 24 Com. Cas. 198 (CA).
  323. However, one point which Ms. Healy made on a number of occasions was that there was no inconsistency between the coverage for which the Bank/ Edge contended, and the intention to protect the interest of the Insured as there set out. The substance of the argument was that the extended cover applied to the very same goods that were within the scope of the transit and storage cover.
  324. In considering this argument, it is sufficient to focus on the storage element of the cover. This provided coverage whilst cargoes were in store at various warehouses around the world. Whilst in those warehouses, those goods would be subject to the usual risks of physical damage, as well as to other risks such as delay or confiscation, covered or potentially covered by the various additional clauses. The effect of the TPC, on the case of the Bank/ Edge, was to provide cover for another risk, or more accurately combination of risks, to which the Bank was exposed in respect of those goods which had been purchased on the basis of repo transactions. That combination involved the risk that (i) the counterparty would default in failing to take back the goods and pay for them, and (ii) the value of the goods would fail to realise the pre-agreed price which the counterparty had agreed to pay. There was, therefore, a combination of a default risk and a risk as to the value of the collateral which affected the cargoes which were in store and which were already the subject of the ordinary cargo insurance. Ms. Healy said that there was no reason in principle why the cover on goods should not be extended to cover these additional risks. Whether or not the cover was so extended was, ultimately, a question of construction of the relevant clause. But this was not a case where the Bank was alleging that it had some general financial guarantee cover in relation to its counterparties (for example on ordinary unsecured loans that it had made), as opposed to cover which was directly related to goods which had been secured.
  325. I think that these points were correct. Ms. Healy's argument shows that there is no particular difficulty in applying the extended cover provided by the TPC (if it indeed provides such cover) to the goods which were already covered for loss and damage whilst in store. Indeed, I did not understand Mr. Parsons to submit that there was any such difficulty. His main point was not that it could not be done, but rather that it produced a commercially absurd result, particularly bearing in mind the way in which the premium had been rated (with lower rates for storage than transit) and the fact that no additional premium was charged for this additional substantial risk. The effect of construing the policy in the manner proposed by the Bank/ Edge was therefore to add considerably to the insurer's potential for loss, but without any commensurate premium. The addition of these risks also, as he correctly said, potentially resulted in more likely scenarios where goods in more than one, possibly many, warehouses could be affected. The policy limit was calculated on a per warehouse basis; ie US$ 100 million per warehouse. A fire or storm or flood might in all likelihood affect only one warehouse, or possibly more but only if they were in reasonably close geographical proximity. By contrast, a default of a counterparty, leading to an inability to realise the relevant price of the goods, could affect goods in numerous and an unlimited number of warehouses across the world.
  326. These arguments as to commercial absurdity, or commercial consequences, were much relied upon by Mr. Parsons in support of the case that the TPC should not be construed in the manner for which the Bank proposed. I will return to them after I have considered the wording of the TPC.
  327. The terms of the TPC

  328. I have previously described aspects of the factual matrix, contractual context and introduced the argument as to commercial consequences. These are all part of the unitary exercise of contract construction. In a case such as the present however, where I am concerned with a carefully drafted clause, I consider that the language used by the parties is the most important consideration in the overall unitary exercise of construction. If the language is unclear, then the clarity of language required (see Engelhart) for an extension of cover beyond physical loss and damage will, by definition, not have been achieved. If, by contrast, the language clearly provides an extension of cover beyond physical loss and damage, or (in underwriters' modified case) cover beyond the other perils covered by the policy, then it should be applied: see Rainy Sky at [23]. In those circumstances, it is unlikely that arguments based on factual matrix will result in a different conclusion being reached. In Jani-King v Pula Enterprises, HHJ Coulson said (at [10]) that the most important source of information about the agreement is the contract itself, and not the surrounding documentation, and he quoted from Buxton LJ in an earlier case:
  329. "One has to remember, when looking at issues about the factual matrix, that although reference to that matrix is not limited to cases where the words are clearly ambiguous, the first place where one expects to find the meaning of the words and the intention of the draftsman is in the words themselves. If they yield a fairly clear solution, and in my judgment these words do, then one has to pause long before concluding at that point the draftsman has used words with a meaning which do not fit in with the objective that he was seeking to attain".
  330. Although those cases preceded the trilogy of Supreme Court cases referred to above, this statement is still apposite in my view. Thus Lewison: The Interpretation of Contracts 7th edition, paragraphs 3.167 – 3.168, states:
  331. "Fourth, reliance on background must be tempered by loyalty to the contractual text. It is not permissible to construct from the background a meaning that the words of the contract will not legitimately bear.
    …
    Fifth, the background should not be used to create an ambiguity where none exists. The court must be careful to ensure that the background is used to elucidate the contract, and not to contradict it".
  332. Similarly, if the language is sufficiently clear, it is unlikely that arguments based on commercial consequences or absurdity will result in a different conclusion, essentially for the reasons given by Lord Neuberger in Arnold v Britton. To do so would result in the contract being effectively rewritten, with the consequent penalisation of the astute party and benefit to the party that was unwise.
  333. All contracts must, of course, be construed in the light of the contract terms as a whole. I have already outlined and discussed the parties' arguments on this issue. These too, for the reasons given, ultimately lead back to consideration of the clarity of language in the TPC.
  334. I therefore turn to the language of the clause itself, and for present purposes do so without reference to its location in the policy, or the related question of whether all underwriters are bound by the form of policy agreed by Mr. Beattie on 29 January (with the TPC in two locations) and subject to Endorsement 1 signed by Mr. Vaughan on 1 February. For present purposes, I therefore consider the TPC as it appeared in the unamended policy at Clause 1.5 of Section 2.
  335. For ease of cross-reference, I have added numbered square brackets to each subparagraph of the TPC:
  336. [1] Underwriters note and agree that, in respect of any Transaction, it is hereby confirmed that the Insured is covered under this contract for the Transaction Premium that the Insured would otherwise have received and/or earned in the absence of a Default on the part of the Insured's client.
    [2] "Actual Sale Price" means the sum received by the Insured upon the sale of the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    [3] "Default" means a failure, refusal or non-exercise of an option, on the part of the Insured's client (for whatever reason) to purchase (or repurchase) the Subject Matter Insured from the Insured at the Pre-agreed Price.
    [4] "Pre-agreed Price" mean the amount for which the Insured's client had agreed to purchase (or repurchase) the Subject Matter Insured from the Insured as specified on the relevant invoice or in the relevant transaction documents, comprising the principal together with any premium or profit element payable to the Insured.
    [5] "Transaction" means any transaction where, following a Default on the part of the Insured's client, the Insured sells the Subject Matter Insured to the applicable Exchange or to a third party on the open market.
    [6] "Transaction Premium" means an amount that is equal to the difference in value between the Pre-Agreed Price and the Actual Sale Price.
  337. The operative part of this wording is [1], since the remaining paragraphs [2] – [6] contain definitions which are relevant to [1] and the concepts there used. If the key definitions are incorporated (using italicised text) into the wording of [1], it reads as follows:
  338. "Underwriters note and agree that, in respect of any Transaction (i.e. any transaction where, following a Default on the part of the Insured's client, the Insured sells the Subject Matter Insured to the applicable Exchange or to a third party on the open market), it is hereby confirmed that the Insured is covered under this contract for the Transaction Premium (i.e. an amount equal to the difference in value between the Pre-Agreed Price and the Actual Sale Price) that the Insured would otherwise have received and/or earned in the absence of a Default (i.e. a failure, refusal or non-exercise of an option, on the part of the Insured's client (for whatever reason) to purchase (or repurchase the Subject Matter Insured from the Insured at the Pre-agreed Price) on the part of the Insured's client".
  339. The clause thus starts out in [1] by confirming that, in respect of any "Transaction" there is coverage for the "Transaction Premium". The Bank described this clause, and in particular the opening words ("it is hereby confirmed that the Insured is covered"), as an "insuring clause". I agree, although I was not convinced that the attachment of this label adds anything material to the Bank's argument. In my view, whether or not the clause is labelled an "insuring clause", it is indisputable that the clause confirms the existence of coverage for the matters set out in the clause. But the questions still remain: what is the extent of the coverage so provided? In particular, is it clear that the coverage so provided goes beyond cover for, or dependent upon, physical loss and damage to goods?
  340. In my view, the language is indeed clear, essentially for the reasons advanced by the Bank and Edge as summarised above. To my mind, the most important points are as follows.
  341. First, there is nothing in the express language of the clause which provides a link either to the coverage for physical loss and damage to cargo or (in relation to the case ultimately advanced by underwriters) to the other "non-PLOD" perils covered by the policy, for example the CEND and BCC coverage.
  342. Secondly, the absence of this link is emphasised by the positive point that the TPC, in particular [1], uses a series of key expressions – Transaction, Transaction Premium and Default – which are not used elsewhere in the policy. None of these words, which are critical to the coverage provided in [1], appear elsewhere. Furthermore, each of those terms is defined in terms which are unrelated to physical loss and damage to the cargo, or indeed to the other "non-PLOD" perils. If the terms of [1] are spelt out in full, the natural reading of a clause containing a series of expressions not used elsewhere, and which are defined in the clause itself, is that it is a self-contained coverage clause.
  343. The coverage so provided concerned "any Transaction". That was defined to mean any transaction where:
  344. "following a Default on the part of the Insured's client, the Insured sells the Subject Matter Insured to the applicable Exchange or to a third party on the open market".
  345. In some respects, this definition does provide a link to some other concepts contained within the policy. But it is not a link to any of the provisions concerning physical loss or damage to the cargo or to the perils covered by the non-PLOD add-ons. The links are to "Insured", "Subject Matter Insured" and "Exchange". The "Insured" was defined at the very start of the policy. "Exchange" was defined in Clause 1.1. "Subject Matter Insured" does not have a specific definition elsewhere. But it is most readily to be read as a reference back to the "Interest" referred to in the first section of the policy, ie:
  346. "All goods and or merchandise appertaining to the Insured's business for which the Insured is the legal owner or for which the Insured is at risk or responsible, contractually or otherwise, consisting principally but not limited to [(A) Hard commodities, (B) Soft commodities, (C) Energy Commodities]"
    This reference to "Subject Matter Insured" therefore supports Ms. Healy's argument that the cover provided by the TPC concerned goods, for example goods in store, which were already covered against risk of physical loss and damage.
  347. Thirdly, both [1] and the definition of "Transaction" in [5] introduce the concept, which is critical to the operation of the TPC, of "Default". Default is defined in wide terms:
  348. "Default means a failure, refusal or non-exercise of an option, on the part of the Insured's client (for whatever reason) to purchase or (repurchase) the Subject Matter Insured from the Insured at the Pre-agreed price".
  349. The word "Default" therefore appears 3 times in the TPC, although one of these times is to provide a wide definition. It is a concept which, ordinarily, is alien to a policy or clause which simply covers the risk of physical loss or damage. Ordinarily, if a person insures goods which he owns or are at his risk, and physical loss or damage occurs to those goods, a claim can be made for the value of the goods lost or for their diminished value. As the underwriters submitted in their written closing, a default is not a pre-requisite for a claim by the Bank where there is physical loss and damage, and it would be entitled to claim that loss by reference to the Basis of Valuation clause in the policy whether or not there had been a default by the Bank's counterparty. Thus, the repeated use of a term which ordinarily is alien (ie alien to a policy which does no more than to insure against the risk of physical loss or damage) is itself a further indication that the TPC was not so linked.
  350. Fourth, the underwriters' argument was primarily, or at least very strongly, based not on the language which the parties had actually used in the TPC, but rather upon the proposition that various considerations – factual matrix, the contract as a whole, and commercial consequences – should lead to the ordinary and natural meaning of the words used in the TPC being circumscribed. In my view, however, it is not appropriate, in the context of this carefully drafted clause, to read in words of limitation which are not there, and in so doing to deprive the existing language of its natural and ordinary meaning. There is nothing in the language of the clause which limits its application to default following physical loss or damage, or to default following the operation of other perils such as delay. That is not the natural reading of the operative part of the clause in [1]. The coverage in [1] was expressly for the "Transaction Premium that would otherwise have been earned or received in the absence of a Default". "Transaction Premium" was itself defined in [6] as being the difference between the Pre-Agreed Price (defined in [4]) and the Actual Sale Price (defined in [3]). The coverage provided is therefore not tied to anything other than (i) Default, and (ii) the Transaction Premium that the insured would otherwise have earned in the absence of Default.
  351. Fifth, the underwriters' argument has the effect of negating express words which are found in the definition of "Default". This is a point which the Bank and Edge made repeatedly in their submissions, and I did not think that the underwriters had an effective answer to the point. The definition of Default refers to the client's failure or refusal "(for whatever reason) to purchase or repurchase." There is therefore no suggestion that the Default need have anything to do with loss or physical damage to the cargo, or any of the other non-PLOD perils. The language indicates that Default includes the breach by counterparties (such as Transmar and Euromar in the present case), and that it is not relevant to enquire into how or why the non-performance has come about. The underwriters' argument therefore requires some event to have occurred which is then followed by a Default. If the parties' intention had been that the default must be consequential upon, or at least follow, the existence of PLOD or the operation of another non-PLOD perils elsewhere in the policy, they would not have agreed that the Default on the part of the client could be "for any reason". As Edge submitted, the underwriters' construction requires those specific words to be accorded something other than their natural meaning.
  352. Sixth, the underwriters' argument produces the result that the TPC has no application in a situation in which the cargo was destroyed (for example in a warehouse fire) or so damaged that it could not in practice be sold and delivered to an exchange or sold on the open market. It was common ground that the TPC would have no application in such circumstances; because the coverage was referable to the "Actual Sale Price" (see [2]), which was defined as the sum received upon the sale of the Subject Matter Insured to the applicable Exchange or to a third party on the open market. As Ms. Sabben-Clare said in opening, it would be a "curious beast" for a clause which provides a measure of loss or "basis of valuation" to be inapplicable in a situation of loss or damage (ie destruction or extensive damage to goods rendering them unsaleable) which could foreseeably occur and for which the parties would have expected the insurance to provide coverage.
  353. Seventh, for reasons explained in more detail below, I did not consider that the underwriters' case that the TPC was no more than a "Basis of Valuation" clause was sound or persuasive. It is not the natural meaning of the words actually used, and also deprives the TPC of any real effect.
  354. Finally, I consider that the wording of the provision concerning "Sellers Insurance interest" does, marginally, reinforce the case that there is no relevant link in the TPC to physical loss or damage. The Sellers Insurance interest provision does contemplate a situation where the Bank's buyer, on FOB or CFR terms, fails to pay for "loss of or damage" to goods. It can therefore reasonably be said that if the parties had intended that limitation to be applicable to the TPC, this would have been spelt out. I do not think, however, that too much significance should be attached to the wording of this Sellers Interest clause. None of the parties referred to this clause until closing submissions, and it is doubtful whether the draftsman would have had this clause in mind when drafting the TPC. It is therefore a marginal point, and I regard the other points to which I have referred as more significant.
  355. The TPC as a Basis of Valuation provision?

  356. In considering the construction for which the Bank and Edge argued, and the clarity of the language used, it is of course necessary to evaluate the alternative construction of the clause for which the underwriters contended. The underwriters' case was that the TPC was a clause whose purpose was to explain and clarify the amount of recoverable loss in the event that there was physical loss or damage to the cargo, or (as underwriters ultimately argued) in the event that one of the other non-PLOD perils operated. They described the clause, interchangeably, as a "Basis of Valuation" clause or a clause which concerned the measure of indemnity.
  357. On underwriters' case, the way in which the clause worked, and the reason that there were references to "Default" in the clause, was summarised by Mr. Parsons in his oral closing as follows:
  358. "… if you have something that's been covered under the policy, physical loss or delay, and then following that there is a default; i.e. the counterparty does not pay, then you are at risk of losing the sale price. Then you can mitigate that by selling it in the market or to the exchange, and, insofar as you've lost the premium or profit, then you recover under the clause
    …
    So … the reason why you see the word "default" three times … it's all focusing on the sale price, which has within it premium or profit that you wouldn't expect to get on a purchase. So normally if you had for example physical damage to goods you would get the … diminution in the market price, and the market price would either be the market price at the date of claim or the market price as identified in the purchase contract. If, however, following the physical damage your counterparty does reject the goods what you're actually at risk of is losing your premium or profit".
  359. The essence of the argument, therefore, was that the existing Basis of Valuation clause would not cover the entire field of potential loss, and that the TPC could be read as an expansion of the measure of indemnity. The argument was somewhat complex, and in order to consider its validity it is convenient to start by setting out the Basis of Valuation clause that was contained in the policy, prior (and indeed subsequent) to the introduction of the TPC. This provided:
  360. BASIS OF VALUATION:
    Market Value at date of claim provided that the market value shall be no less than the Market Value as used at the date of purchase. For the purposes of this clause the term "Market Value" shall mean the relevant published Exchange future price, or any other price stated in the Insured's underlying purchase contract, call option contract or any other document relevant to the purchase (as amended, supplemented, replaced or otherwise modified from time to time), plus or minus any applicable adjustments as referenced in the Insured's underlying purchase contract, call option contract, or any other document relevant to the purchase (as amended, supplemented, replaced or otherwise modified from time to time).
  361. The underwriters submitted that this "BOV" clause was focused on market value at the date of purchase of the goods by the Bank. Thus, it referred to Market Value at the date of claim, but provided that it was not to be less than Market Value "as used" at the date of purchase. The purpose of the BOV clause was therefore to ensure that the Bank recovered no less than the price relating to its purchase of the commodity from its counterparty. On the other hand, the measure of recovery under the TPC was focused on the "Pre-Agreed Price" which related to the sale by the Bank to its counterparty. This was defined as the amount for which the Bank would sell the commodity back to its client. There was therefore a difference in approach taken by the BOV clause (focusing on purchase by the Bank), and that taken by the TPC clause (focusing on re-purchase by the counterparty). Whilst it was agreed that the TPC had no application if the cargo was a total loss, it would have potential value in a situation where there was, for example, damage to the cargo. In that situation, the Bank would have the option of valuing the loss under either the BOV clause or the TPC clause.
  362. I did not consider this argument to be sound or persuasive, or to provide a reasonable alternative interpretation of the TPC, for a number of reasons.
  363. First, if the parties were concerned that the BOV clause was inadequate to cover the full field of market-related loss that the Bank might suffer in the event of physical loss and damage (or indeed non-PLOD perils), then the obvious course would have been to expand the BOV clause. This clause had appeared in materially identical terms in policies in prior years. If there were shortcomings, and a desire to deal with possible shortcomings in the BOV clause relating to the Bank's sales, then one would have expected the BOV clause to be changed. One would not expect the BOV clause to be expanded upon by a wholly separate clause, which makes no reference to the BOV clause itself, and uses very different terminology. Furthermore, the heading to a clause may be taken into account in construing a clause, at least as a signpost concerning the topic that is being addressed: see Lewison paragraphs 5.108 – 5.110, and The Radauti [1987] 2 Lloyd's Rep 276 (per Staughton J at first instance). Here, the parties had identified which clause addressed Basis of Valuation, and the logical conclusion is that the TPC was addressing a different issue.
  364. Secondly, there is nothing in the wording of the TPC itself which suggests that the parties were addressing any inadequacy in the BOV clause, including the suggested difference between the Bank's purchase and sale contracts. The TPC had a simple formula for calculating the Transaction Premium: the difference between the Pre-Agreed Price and the Actual Sale Price. The Pre-Agreed Price was the amount for which the Bank's counterparty had agreed to purchase or repurchase the goods: it comprised (see [4]) both the "principal together with any premium or profit element payable to the Insured". The clause was therefore focused on the total amount which the counterparty would pay, inclusive of premium and profit. It was not therefore simply concerned with the premium or profit element referable to the sale to the counterparty. That was only one part of the overall calculation.
  365. Thirdly, if the presumed intention of the parties was for the TPC to be an enhanced Basis of Valuation clause, then it would make no sense for it to have no application in circumstances where the goods were destroyed or damaged so as to be incapable of re-sale. As the Bank and Edge submitted, it would then be inapplicable in the most serious of cases.
  366. Fourth, I agree with the submissions of the Bank and Edge that the interpretation of the TPC as a Basis of Valuation clause has the effect of rendering the TPC without any practical effect. Whilst the presumption against surplusage has been said to be not particularly compelling, it would in my view be surprising to conclude that a lengthy and carefully drafted bespoke additional clause, introduced by way of amendment to a policy which already had a BOV clause, was devoid of any practical meaning and was linguistic overkill; particularly bearing in mind that the clause, when read as a whole, was not directed at the same target as the BOV clause.
  367. The reason that the TPC would be rendered without practical effect depends upon the meaning to be ascribed to the BOV clause. The underwriters' argument is based upon the proposition that the BOV clause is exclusively concerned with the Bank's purchase contracts, and that something was needed (in the event of physical loss or damage or another insured peril) in order to cover the Bank's sale contracts; ie the contracts whereby their counterparties had agreed to purchase or repurchase goods. However, this gives an unnecessarily restrictive meaning to the wide words of the BOV clause.
  368. In considering the scope of that clause, it is relevant to bear in mind the ordinary business of a bank which is financing a counterparty's stock via "repo" transactions. The evidence in the case showed that such repo transactions were a very usual way in which a bank would advance funds to its customers, with the result that it would acquire the ownership of goods for a temporary period and the customer would repurchase them later. The need for the Bank to insure the goods here arose from its ownership of goods acquired in this way, via transactions involving the purchase of goods and agreements relating to their repurchase. Whilst the insurers may not have known the precise details of the Bank's contracts with its customers, there was no suggestion that repo transactions, such as those carried out here involving a purchase and resale to a customer, were in any way unusual or outside the scope of the policy. Mr. Beattie referred in his oral evidence on Day 5 to his knowledge that the Bank was involved in repo deals, and in an email from Mr. Mullen to one of the underwriters in February 2016 (Mr. Brian James of Talbot, the 3rd Defendant), Mr. Mullen explained:
  369. "To explain how they operate, and they have just 42 people worldwide, and operate on a repo plan where they physical purchase the goods and therefore are owners, and either sell them back to the supplier after a given time, normally 6 months, or to any interested party. This is directly opposite to traders who speculate on particular interests"
  370. The BOV clause in the present case provided, as set out above, for valuation to be no less than Market Value as used at the date of purchase. The definition of Market Value encompassed:
  371. "the relevant published Exchange future price, or any other price stated in the Insured's underlying purchase contract, call option contract or any other document relevant to the purchase".
  372. This is a wide definition. It is not confined to the price stated in the purchase contract whereby the Bank acquires the goods. The clause expressly refers to the Insured's "call option contract". There is no reason to confine these words to call option contracts in favour of the Bank; ie where the Bank could call for goods. In fact, there was no evidence that the Bank entered into such contracts. Where repo transactions are concerned, one would more naturally expect that the "call options" would give the counterparty the right to call for goods, and such call options were indeed a feature of the Bank's transactions with Euromar and Transmar. This is consistent with the reference in the definition of "Default" (in the TPC) to the "non-exercise of an option, on the part of the Insured's client … to purchase (or repurchase) the Subject Matter Insured from the Insured": in other words, the client would be calling for goods from the Bank, not the other way round.
  373. Furthermore, the BOV clause then refers to "any other document relevant to the purchase". In the context of repo transactions, this would in my view plainly encompass resale contracts concluded as part and parcel of the transactions whereby the goods were acquired by the Bank.
  374. In these circumstances, the "Basis of Valuation" clause was sufficient to enable the Bank to assess Market Value by reference to prices payable by its client in what the underwriters described as the Bank's "sales" to its counterparty; ie the obligation to repurchase, or the price payable upon exercise of the call option. There was therefore no need to draft the TPC in order to cater for any lacuna in that respect. Furthermore, the absence of the suggested lacuna in the BOV clause means that the TPC is, on underwriters' construction, devoid of practical effect.
  375. The balancing exercise

  376. I therefore consider that, when the language of the TPC is analysed, the Bank's and Edge's construction of the TPC is correct, and that the wording is clear. This conclusion is based upon: (i) the natural meaning of the words actually used as described above; (ii) the absence of wording which would confine the TPC in the manner proposed by the underwriters; (iii) the inconsistency between the underwriters' proposed limitation and the words "for whatever reason" in [4]; (iv) the absence of any good reason for seeking to construe the TPC as an additional and overlapping BOV clause; and (v) the absence of any satisfactory alternative meaning (if the TPC were to be construed as a BOV clause) which would give the TPC real practical effect in the light of the BOV clause.
  377. It will be apparent from the foregoing discussion that I consider that the words used in the policy are the most significant item in the balancing exercise. This is because the TPC is a carefully drafted clause, and in such circumstances the language used is very important in striking a balance between the indications given by the language and other competing considerations: see Wood v Capita at paragraphs [11] – [12]. Nevertheless, as Lord Hodge said there, the unitary exercise of construction does involve checking each suggested interpretation against the provisions of the contract, and considering its commercial consequences.
  378. At the heart of insurers' case was an argument as to the commercial consequences of the construction for which the Bank and Edge contended. They say that the result is commercially absurd. The consequence of their argument is that the Bank received credit and financial guarantee cover in respect of every cargo that was owned by the Bank, and that it did so without paying any additional premium. They say that it makes no commercial sense for a marine cargo underwriter, with no knowledge or experience of credit risk cover, to offer cover which could, if placeable at all, only be placed in the specialist credit market at substantial additional cost. They say that neither Mr. Beattie nor the following underwriters would have done this deliberately: if the cover was provided, it was entirely unwitting. None of these underwriters was given the authority by their employers to write trade credit or financial guarantee business – although lack of authority is not relied upon as a defence in itself. They also contend that the placement did not involve any presentation of matters which would be relevant to credit risks, and they referred to the care taken, and information that would be required, by those who do specialise in the writing of such risks.
  379. They contend that the Bank would receive a windfall, which it did not expect to receive; because all that it was really seeking to do was to cover a "top slice" of potential loss, comprising the "market premium" represented by the difference that might exist between the price quoted by a recognised exchange, and the market value of the goods at a particular time.
  380. Many of these points were also deployed as aspects of the factual matrix. The insurers referred to the fact that this cover was placed by a humble marine broker (Mr. Mullen's description) in a market of marine cargo underwriters which had no history of writing trade credit insurance. Mr. Mullen knew that, ordinarily, marine cargo underwriters would not cover risks relating to the purchase of poor quality cargoes. But the effect of the Bank/Edge's construction of the policy is to place the risk of poor quality cargoes on insurers, since poor quality might lead to a default but more importantly would result in a differential between the Pre-Agreed Price and the Actual Sale Price. They say that there was no mechanism in the London marine cargo market for the assessment of trade credit risks.
  381. In my view, none of these points, whether relied upon either as part of the factual matrix, or in relation to the commercial impact of the parties' rival constructions, is such as to outweigh or decisively impact upon the construction of words which the parties have actually used.
  382. I fully recognise that cover, such as that which I consider to have been provided by the TPC, has never previously been granted by the London marine cargo market. But it is a feature of the market that cover is expanded by the addition of clauses which brokers seek on behalf of their clients. In the present case, an unusual and unique clause was presented, and agreed to by underwriters. The issue is: how to construe that clause. The construction of an unusual and unique clause, without any apparent parallel, is not materially assisted by consideration of how policies without such a clause would usually be construed.
  383. I also recognise that, certainly with the benefit of hindsight, it was unwise of underwriters to agree to this clause. It is likely to be the case, although there was no detailed evidence on this, that the effect of adding the TPC (construed in the way that I have construed it) would increase a prudent underwriter's assessment of his probable maximum loss. It did not seem to me to be right to say that the monetary exposure of underwriters was increased. The existing policy covered every cargo owned by the Bank, and therefore every existing ton of goods in each warehouse was at the risk of the underwriters. Even without the TPC, there was certainly some risk of exposure to common problems across different warehouses. For example, where goods were stored in warehouses which were geographically close, a single event such as a major explosion (such as occurred in 2020 in Beirut) or a tsunami or a hurricane could potentially impact more than one warehouse. In relation to the fraudulent documents cover, there was the potential for the actions of a dishonest counterparty to affect goods stored, or apparently stored, in different warehouses. However, the possibility of an accumulation of loss, in different warehouses, would potentially be increased by the addition of coverage which insured against the default of each counterparty. Even if only one counterparty defaulted, that could potentially affect goods in different warehouses across the world. This increase in probable maximum loss would, ordinarily, justify the payment of an additional premium, bearing in mind that the policy was rated on the basis of ordinary transit and storage rates.
  384. However, it is very important to bear in mind the approach of Lord Neuberger in Arnold v Britton. I should be slow to reject the natural meaning of a provision simply because it appears to be a very imprudent term for one of the parties to have agreed. The purpose of interpretation is to identify what the parties have agreed, not what the court thinks that they should have agreed. It seems to me that all of the arguments of the insurers, attractively presented though they were, are designed to relieve the insurers of the bargain to which they agreed.
  385. The evidence as to the way in which each party approached the terms of this contract, although canvassed in immense detail because of the various defences raised, is not relevant to contract interpretation. But it does illustrate the reason why, as Lord Neuberger explained, it is important for a judge to avoid re-writing a contract in an attempt to assist an unwise party or to penalise an astute party.
  386. There is here, on the one hand, a Bank which – together with its internal and external lawyers – paid very close consideration to the terms of its insurance policy, and sought to identify exactly what it needed to cover and what language was necessary to achieve that purpose. A careful drafting process ensued. I do not accept that, in drafting the TPC as they did, NRF misunderstood what their clients wanted, and sought to give the Bank more protection than it actually wanted. Mr. Stroink, who in my view was a reliable witness who gave his answers fairly and honestly, said that there was a process of evolution in the Bank's thinking. Whilst at an early stage they (or at least some individuals within the Bank) had been focusing on the concept of "market premium" – being the difference between the price on a recognised exchange and market price – their thinking then moved on, with the benefit of NRF's advice. This can be seen in the fact that the concept of "market premium" is not used in the TPC, and it was even removed from the revision of Endorsement 3 which concerned the BCC wording. I have no doubt that the TPC was subjectively intended by the Bank to cover the Bank for precisely the problem which ultimately occurred with Transmar and Euromar. The evidence showed that, after the TPC had been agreed in July 2015, the Bank thought that it had that coverage, and conducted its business on that basis.
  387. On the other hand, there is the position of the underwriters. The carefully drafted terms were presented initially to Mr. Beattie in July 2015, at a meeting where he was told that the wording was wanted by the Bank's lawyers. It would have been apparent that the terms had been very carefully drafted. I shall deal with the evidence as to this meeting in more detail in Section E, but it is plain that Mr. Beattie gave the wording nothing which approached the care which the Bank and its lawyers had taken. He can have read it through only once in a meeting which lasted around 15 minutes, before (as I conclude in Section E) signing the document and agreeing to the inclusion of the wording. That would have left precious little time for any discussion, particularly bearing in mind that the July 2015 endorsement contained a series of amendments and additions. As discussed in Section E, I am not satisfied that Mr. Beattie raised any questions about the TPC. But even if I were to accept his evidence, he did not receive a satisfactory or clear answer from Mr. Mullen. But he nevertheless signed the July endorsement.
  388. When the matter came back for renewal, he did not in my view give the clause any greater attention than he had done in July 2015. There is, for example, no record of any discussion with the broker about the clause or its impact. Again, even if I were to accept his evidence, there was an inconclusive discussion with Mr. Mullen about what the TPC meant, with no attempt to amend the clause or to record Mr. Beattie's alleged understanding. Instead, Mr. Beattie signed a number of slips (on 20, 25, 27 and 29 January), each of which contained the clause.
  389. As far as the following market is concerned, the evidence of the majority of the underwriters – with few exceptions, such as Mr. Gaiger of Navigators who read the policy, and Mr. Butterworth of Swiss Re who may have skim-read the policy – was that none of them noticed the clause let alone read it carefully or asked questions about it.
  390. In these circumstances, there can be little doubt that, to adopt the language of Lord Neuberger, the Bank was the astute party, and the insurers unwise. This is not relevant to how the contract is to be interpreted. But the important point is that the adverse commercial consequences, which have manifested themselves as a result of the claim, should not in my view be permitted to sway the interpretation of the policy or to depart from its ordinary language.
  391. Nor, in my view, do the commercial considerations all tell against the Bank's construction. This would be to ignore the fact that this policy was placed, and then renewed, in soft market conditions. The Bank was seen as an attractive client, and the risk was a popular one. Furthermore, again as discussed further in Section E, when the potential claim was first notified, neither the underwriter at the RSA who had written the risk (Mr. Beattie) nor its experienced claims manager (Mr. Jones) recoiled with surprise or horror when learning (as in my view they both did) that the claim arose from the default of the Bank's counterparty in circumstances where no physical loss or damage was alleged. No point was then taken that the Bank's potential claim was at odds with commercial common sense.
  392. The positioning in the policy

  393. The primary case of each party is that the TPC means what the clause means irrespective of its location. There were, of course, tactical considerations which lay behind this approach. The Bank considered that the location in the amended policy was favourable to its case, where it appeared twice including in a separate section, and therefore sought to say that the clause had the same meaning in the earlier version. By contrast, the underwriters considered that its location as Clause 1.5 in the earlier version – where it was not a separate section and appeared immediately after some definitions – was helpful in indicating the relative unimportance of the clause on its true construction; an unimportance which did not change when the same words were moved into two separate locations.
  394. I agree that the language of the clause, as discussed above (construed in the context of the relevant factual matrix and the terms of the policy as a whole) should be given the same meaning wherever it appears.
  395. However, I also consider that my conclusion, that the TPC is applicable where there is no physical loss and damage (or another non-PLOD peril covered elsewhere in the policy), is reinforced by the positioning of the TPC in the 2016 policy and the relevant factual matrix in that regard.
  396. The starting point in this context is the location of the TPC when first introduced into the cover. For reasons discussed in Section E below, the TPC first became contractually binding – at least on RSA – in July 2015. At that stage, it was not tethered to any particular part of the policy wording. Under the July 2015 endorsement, it was expressed to be a replacement of the previous endorsement which had (as discussed above) been concerned with the Business Contingency cover. However, the TPC in the July 2015 endorsement was, clearly, not confined to the BCC element of the coverage. In that regard, there is a contrast between the TPC and the revised wording, in the July endorsement, of Endorsement 3. In addition, as already described, the TPC used new definitions and new language which, in their critical respects, were not related to the risks covered elsewhere in the policy wording. In these circumstances, the TPC was, when first introduced, a standalone provision, with no relevant link to any particular section of the policy or any part of the coverage elsewhere.
  397. I have said that the July endorsement was contractually binding on RSA in July 2015. There was much argument at the hearing as to whether or not the endorsement was binding on following underwriters. I do not think that the resolution of that debate is of any real significance when considering the factual matrix to the 2016 policy. The evidence shows that none of the following market who had subscribed to the 2015 slip were told about the amendment at that time, or had their attention drawn on renewal in January 2016 to the fact that this amendment had taken place mid-year in 2015. However, all of the following market in 2016 – including the 4 insurers (the 11th – 14th Defendants) who had not subscribed to the 2015 policy – were shown and subscribed to the slip which contained the TPC. Even though none of them actually knew that the origin of the TPC was the endorsement agreed by the RSA in July, where the TPC was a standalone provision, this was information which was reasonably available to them. If they were interested in the origin of the TPC they only had to ask the broker about it – as did one underwriter, Mr. Shillabeer of Channel (the 9th Defendant) after the claim arose. Since it was information that was reasonably available, it forms part of the relevant factual matrix.
  398. In my view, however, what is more significant is where the TPC finally came to rest in the 2016 policy.
  399. As regards the RSA, and the 13th and 14th Defendants, there was ultimately no dispute that they subscribed to a policy where the TPC appeared in two locations. One of those locations was a separate Section 4. This was in my view the clearest indication that the coverage provided by the TPC was standalone coverage, not linked to the need for physical loss or damage or the operation of the perils covered by the other non-PLOD add-ons. Section 4 immediately followed Section 3, the CEND Cover. This was a self-contained add-on, not requiring physical loss and damage to the cargo. Section 4, the TPC, should in my view be construed similarly.
  400. The question remains, however, whether the other insurers (D2 – D12) who did not scratch any slip which contained the TPC in Section 4 – but only a slip with the TPC at Clause 1.5 in Section 2 – were bound by the amended slip initially scratched by Mr. Beattie on 29 January, and then the subject of Endorsement 1 scratched by Mr. Vaughan on 1 February. In view of my conclusion as to the effect of the TPC in the unamended slip (ie where it appeared at Clause 1.5), I can deal with this issue briefly. In my view, the other insurers were so bound, pursuant to the operation of the General Underwriters Agreement or GUA.
  401. The agreement of the insurers, amongst themselves, as to the scope of the RSA's authority to agree contract amendments, was contained in the "Subscription Agreement Section" of the slips subscribed by underwriters in both 2015 and 2016. For present purposes, the relevant Subscription Agreement Section is that contained in the 2016 slip which was scratched, between 25 and 29 January 2016, by the 2nd – 12th Defendant insurers. This provided, as set out in Section C above:
  402. SUBSCRIPTION AGREEMENT SECTION
    SLIP LEADER:
    Royal Sun & Alliance
    BASIS OF AGREEMENT TO CONTRACT CHANGES:
    Subject to GUA October, 2001 with Marine Cargo Schedule 2003
    OTHER AGREEMENT PARTIES FOR CONTRACT CHANGES, FOR PART 2 GUA CHANGES ONLY:
    Slip leader only to agree part two changes.
    AGREEMENT PARTIES FOR CONTRACT CHANGES, FOR THEIR PROPORTION ONLY:
    Slip leader to agree all contract changes."
  403. Contrary to the argument advanced by Edge, I did not consider that the wording "Slip leader to agree all contract changes" meant that RSA had carte blanche to agree whatever changes it wished. Those words have to be read in the context of the earlier wording which said that the "Basis of Agreement to Contract Changes" was GUA October 2001 with Marine Cargo Schedule 2003. The 2003 Marine Cargo Schedule to GUA identifies, in Part 3 thereof, those (necessarily significant) contract changes which "may be agreed only by all Underwriters each for its own proportion severally and not jointly". If a change did not fall within Part 3, then RSA could agree it. That is because of the combination of (i) the slip providing that the "Slip leader [ie RSA] only to agree part two changes", and (ii) Part 2 itself authorising the Agreement Parties (here RSA alone) to agree all alterations which did not fall within Part 3. However, if a proposed change fell within Part 3 – for example because the alteration increased the monetary exposure of the underwriters, or because the policy period was extended, or because of a change in the parties' agreement as to jurisdiction for the resolution of disputes – then the terms of Part 3 provided that each underwriter should agree for its own proportion.
  404. I do not think that the very general wording which provided for the "Slip leader to agree all contract changes" should be construed in a manner divorced from the preceding provisions of the subscription agreement section or the structure of the GUA as a whole. The consequence of Edge's argument would be (for example) that the RSA was permitted to increase the monetary exposure of the following market, and also that it was up to the RSA to decide whether or not to inform the following market that this decision had been taken. This would be a very surprising result, and I consider that the alternative interpretation, which in effect preserves Part 3 in its entirety, is preferable.
  405. This then leads to two issues which arise in relation to the amendments agreed by Mr. Vaughan on 1 February in Endorsement 1, and thereafter notified to the market by Mr. Lockyer later that day.
  406. The first issue is whether or not the GUA, which addresses "Alterations" in Clause 3, was applicable at all in circumstances where, on 1 February 2016, the renewal of the risk was still in the process of being placed. The underwriters argued that one would ordinarily expect the GUA to be applicable in the period after renewal, for example months or at least weeks after the renewal had taken place. Here, however, the position at the end of January and early February was that the brokers were still in the process of finalising the contract wording, having received a number of last-minute proposed amendments from the Bank.
  407. I do not accept this argument. The GUA does not provide a minimum lead time, starting from when the contract is bound, before its provisions become applicable. The contractual position is that the 2nd – 12th Defendants all entered their lines, on the terms of the 2016 slip with the Subscription Agreement Section attached, between 25 and 29 January 2016. They all became bound at that time. The provisions of the Subscription Agreement Section were immediately effective. Accordingly, unless the alteration agreed by Mr. Vaughan on 1 February fell within Part 3, the RSA was authorised by the following market to agree it without the need for their separate agreement to be obtained. I see nothing uncommercial in this result. Where, for example, there were typographical errors, or the need to add a further warehouse or facility, this approach would enable an amendment to be accomplished swiftly by agreement of the slip leader.
  408. The second issue is whether the relevant alteration made on 1 February 2016, which moved the TPC from Section 1.5 to the two places in the policy wording including Section 4, was an alteration which "increases the monetary exposure of the Underwriters". None of the underwriters argued, in the present context, that it did any such thing. But that was because, no doubt for the tactical reasons already identified, their primary case was that the TPC in Section 1.5 did not cover the present losses and that the movement of the TPC made no difference as to how it was to be interpreted. I have concluded that the TPC in Section 1.5 did cover the present losses. It therefore necessarily follows that the movement of the TPC, in the alteration made on 1 February, did not increase the monetary exposure of the 2nd – 12th Defendants, however the term "increase the monetary exposure" is to be construed.
  409. It follows that the 2nd – 12th Defendants were indeed all bound by the terms of the amended policy originally subscribed by Mr. Beattie on 29 January, and then the subject of Endorsement 1 on 1 February 2016.
  410. I have, however, also considered whether there was an increase in the underwriters' monetary exposure in the event that (i) I am wrong on my construction of the TPC as it appears in Section 1.5, for example because it lacks sufficient clarity but (ii) the placement of the TPC as a separate section provides the necessary clarity so that the Bank's case succeeds but only because of the way in which the TPC appeared in the amended slip.
  411. I do not think that, in such circumstances, there would be an increase in monetary exposure. Prior to the TPC, the underwriters were exposed to the potential for loss in relation to every ton of the Bank's cargo that was in store or in transit, subject to the limits of the policy. The TPC did not effect an increase in those policy limits. The effect of the TPC was that the potential for loss, within the existing policy limits, was increased. There was now coverage, for goods already at risk, in relation to additional perils. Whilst, as previously discussed, a prudent underwriter might consider that there was an increase in his probable maximum loss on his existing exposure, it does not in my view follow that his "monetary exposure" as defined by the GUA was increased.
  412. Edge drew attention to changes made to Part 3 of the GUA subsequent to the events with which I am concerned. Clause 3.3.1 was changed so as to refer to: "Any alteration which increases the limits of liability monetary exposure …". Edge submitted that this showed that "monetary exposure" was always intended to denote the limits of liability; a proposition which derived some support from the evidence of the underwriters' expert underwriting witness, Mrs. Webb. I do not consider that this is a permissible approach to construction of the GUA. At best, it provides a degree of comfort that the conclusion that I have reached in the previous paragraph is not surprising.
  413. E: Rectification/ estoppel/ collateral contract

    The issues

  414. If the insurers' arguments on contract interpretation fail, then they contend that three conversations between Mr. Beattie and Mr. Mullen prevent the Bank from relying upon the terms of the policy so interpreted. This is because the policy should be rectified so as to give effect to what was said in those conversations and the mutual intentions of Mr. Beattie on behalf of RSA and Mr. Mullen on behalf of the Bank; or that there was a collateral contract, where Mr. Beattie's signature of the relevant wordings was on the basis of an agreement as to the meaning and effect of the TPC; or because an estoppel by convention can be relied upon. The substance of the argument is that Mr. Beattie and Mr. Mullen were in agreement that the TPC was no more than a "Basis of Valuation" clause, and that it was only applicable where there was physical loss and damage to the cargo, or at least that Mr. Mullen and the Bank are precluded by estoppel from asserting otherwise.
  415. E1: Legal principles

  416. There was no significant dispute between the parties as to the applicable legal principles.
  417. Rectification

  418. The test for rectification has very recently been clarified by the decision of the Court of Appeal in FSHC Holdings v GLAS Trust [2020] Ch 365, and is summarised in the following passage at [176]:
  419. "it is necessary to show either (1) that the document fails to give effect to a prior concluded contract or (2) that, when they executed the document, the parties had a common intention in respect of a particular matter which, by mistake, the document did not accurately record. In the latter case it is necessary to show not only that each party to the contract had the same actual intention with regard to the relevant matter, but also that there was an outward expression of accord meaning that, as a result of communication between them, the parties understood each other to share that intention."
  420. Paragraphs [80] – [87] of the judgment, under the heading "Tacit Agreement", make it clear that the concept of an "outward expression of accord" does not require that the parties' common intention should be declared in express terms. The shared understanding may therefore be tacit. It may therefore include understandings that are so obvious as to go without saying, or that were reached without being spelled out in so many words. The Court of Appeal emphasised, however, that the court is concerned with what the parties actually communicated to each other, and not with identifying their presumed intention by means of an "officious bystander" test: see [87]. The concept of a tacit agreement is to be contrasted with uncommunicated intentions which happen, without the parties knowing it, to coincide. It is therefore "fundamental that contractual rights and obligations should be based on mutual assent which the parties have manifested to each other".
  421. MacGillivray on Insurance Law 14th edition, paragraph 12-002 states that:
  422. "There is a presumption that a policy which is issued by the insurer and accepted by the insured contains the complete and final contract between the parties. Consequently, the courts' equitable jurisdiction to rectify insurance policies is exercised with restraint inside certain well-established limitations, or else it would tend to destroy certainty in insurance business. When a plaintiff seeks rectification, he must establish as a fact that the parties were agreed upon the point in question, and that the policy accidentally fails to record their agreement."
  423. Ms. Healy relied upon this presumption in the context of insurance cases, and the difficulty of obtaining rectification in an insurance case. It may be that there is nothing particularly special about insurance contracts, because there is in any event a "natural presumption" that a written contract is an accurate record of what the parties agreed. In FSHC at paragraph [46], the court said:
  424. "It is necessary to show that at the time of executing the written contract the parties had a common intention (even if not amounting to a binding agreement) which, as a result of mistake on the part of both parties, the document failed accurately to record. This requires convincing proof to displace the natural presumption that the written contract is an accurate record of what the parties agreed".
  425. The presumption is, however, reinforced in the context of the London insurance market by the Contract Certainty Code of Practice published by Lloyd's, the Association of British Insurers and other industry bodies in October 2012 ("The Code"). This is not a legally binding document, but it was referred to extensively during the trial and it was common ground that it reflected best practice. The Code sets out a number of "Contract Certainty Principles" which apply at different stages. These include:
  426. "Contract Certainty Principles
    A When entering into the contract
    The insurer and broker (where applicable) must ensure that all terms are clear and unambiguous by the time the offer is made to enter into the contract or the offer is accepted. All terms must be clearly expressed, including any conditions or subjectivities.
    Explanation (A.1)
    The proposed contract is the document which contains the offer and can take many forms. Individual market protocols define these. Examples include: completed presentation templates; proposal forms; slips or other placing documents. Terms are the contractual provisions of the contract, and should be clear and unambiguous. Contract terms should comply with all relevant regulations and codes of practice.
    Guidance
    …
    All terms should either be expressed in full or unambiguously identified; for example, by specific reference to bespoke or model material.
    …
    Insurer actions (A.3)
    The insurer should check that the proposed contract clearly identifies all of the terms by the time it formally commits to the contract.
    Where there is more than one participating insurer, each insurer should satisfy itself that adequate contract checking has been completed.
    Where the contract is to provide cover that will commence prior to the contract being entered into, the insurer should ensure that:
    Broker actions (where a broker is involved) (A.4)
    The broker should provide the necessary risk and contractual information that represents the insured's demands and needs, in order to enable to agreement of all terms.
    …
    C Demonstration of performance
    The insurers and brokers (where applicable) must be able to demonstrate their achievement of principles A and B."
  427. Section C of the Code then gives examples of how Contract Certainty can be demonstrated, including verification against a checklist, sample or file audits and system or process controls.
  428. In order for rectification to be available, it must be possible to redraft the written instrument to reflect the common intention of the parties: see The Nai Genova [1984] 1 Lloyd's Rep 353, per Slade LJ at 359; and Dunlop Haywards v Erinaceous [2009] 1 Lloyd's Rep 464, per Rix LJ at [64].
  429. Collateral contract

  430. The relevant principle is summarised in Chitty on Contracts 33rd edition, paragraph 13-004:
  431. "It may be difficult to treat a statement made in the course of negotiations for a contract as a term of the contract itself, either because the statement was clearly prior to or outside the contract or because the existence of the parol evidence rule prevents its inclusion. Nevertheless, the courts are prepared in some circumstances to treat a statement intended to have contractual effect as a separate contract or warranty, collateral to the main transaction. In particular, they will do so where one party refuses to enter into the contract unless the other gives him an assurance on a certain point or unless the other promises not to enforce a term of the written agreement".

    Estoppel by convention

  432. I was referred to the decisions of the Court of Appeal in Blindley Heath Investments Ltd. v Bass [2015] EWCA Civ 1023 and of Picken J in Aras and others v National Bank of Greece SA [2018] EWHC 1389 (Comm). An estoppel by convention arises if (i) there is a relevant assumption of fact or law, either shared by both parties, or made by party B and acquiesced in by party A, and (ii) it would be unjust to allow party A to go back on that assumption.
  433. The principle is summarised in Chitty on Contracts 33rd edition, paragraph 4-108 as follows:
  434. "Estoppel by convention may arise where both parties to a transaction "act on assumed state of facts or law, the assumption being either shared by both or made by one and acquiesced in by the other". The parties are then precluded from denying the truth of that assumption, if it would be unjust or unconscionable (typically because the party claiming the benefit has been "materially influenced" by the common assumption) to allow them (or one of them) to go back on it. Such an estoppel differs from estoppel by representation and from promissory estoppel in that it does not depend on any representation or promise. It can arise by virtue of a common assumption which was not induced by the party alleged to be estopped but which was based on a mistake spontaneously made by the party relying on it and acquiesced in by the other party."
  435. This formulation reflects the summary of the relevant principles by Lord Steyn in Republic of India v India Steamship Co Ltd (No 2) [1998] AC 878 at page 913E-G:
  436. "It is settled that an estoppel by convention may arise where parties to a transaction act on an assumed state of facts or law, the assumption being either shared by them both or made by one and acquiesced in by the other. The effect of an estoppel by convention is to preclude a party from denying the assumed facts or law if it would be unjust to allow him to go back on the assumption … It is not enough that each of the two parties acts on an assumption not communicated to the other. But it was rightly accepted by counsel for both parties that a concluded agreement is not a requirement for an estoppel by convention."
  437. These principles were elaborated upon by Briggs J in HM Revenue v Benchdollar [2009] EWHC 1310 (Ch), in the context of an estoppel by convention arising out of non-contractual dealings:
  438. "… (i) It is not enough that the common assumption upon which the estoppel is based is merely understood by the parties in the same way. It must be expressly shared between them. (ii) The expression of the common assumption by the party alleged to be estopped must be such that he may properly be said to have assumed some element of responsibility for it, in the sense of conveying to the other party an understanding that he expected the other party to rely upon it. (iii) The person alleging the estoppel must in fact have relied upon the common assumption, to a sufficient extent, rather than merely upon his own independent view of the matter. (iv) That reliance must have occurred in connection with some subsequent mutual dealing between the parties. (v) Some detriment must thereby have been suffered by the person alleging the estoppel, or benefit thereby have been conferred upon the person alleged to be estopped, sufficient to make it unjust or unconscionable for the latter to assert the true legal (or factual) position."
  439. In Mitchell v Watkinson [2014] EWCA Civ 1472, the question arose as to the whether the formulation of Briggs J was applicable in contractual cases. The Court of Appeal said, at paragraph [52]:
  440. "Although there are superficial differences of formulation it seems to us that these are more apparent than real, and that in practice there is likely to be little if any material difference in the outcome whichever version of these principles is applied."
  441. In Aras, Picken J said (at paragraph [115]) that there needed to be "clarity over what comprises the common assumption (if there is a common assumption)". He also referred (at paragraph [116]) to the requirement for something to be shown to have "crossed the line" sufficient to manifest an assent to the assumption.
  442. Ultimately, the present case turns upon the facts (ie what, if anything, was said about the TPC at the various meetings relied upon by the insurers) rather than upon fine differences in the precise formulation of the principle.
  443. E2: The witness evidence

  444. I start by summarising the evidence of the two central witnesses (Mr. Beattie and Mr. Mullen) as to the relevant conversations, as well as the evidence of Mr. Lockyer who attended one of the meetings where a relevant conversation is alleged to have taken place.
  445. Mr. Beattie's evidence as to the three relevant meetings

  446. Although the first two relevant meetings were some time apart, it is convenient to summarise the relevant evidence in relation to all three meetings.
  447. In relation to the July endorsement and the 28 July 2015 meeting, Mr. Beattie's evidence in his first witness statement was that the July document or "wish list" did not represent any actual change to the wording of the policy: it was scratched for information and receipt only. Mr. Mullen told him that he had been to the Netherlands to meet the Bank's in-house legal team and risk managers, and the in-house lawyers wanted some changes to the policy wording. Mr. Mullen said that the document was a note or "wish list" of the changes that the in-house lawyers wanted to make. Mr. Beattie understood that it had been drafted by lawyers, but he was not given any explanation of why they were interested in making changes to the policy. The changes sought were extensive, and would require the approval of the following market. They were changes which he would only have been prepared to consider in the context of a renewal.
  448. Mr. Beattie said that he therefore made it clear to Mr. Mullen that he was not interested in performing a policy re-write mid-term. He also made it clear that he understood the transaction premium wording to concern the basis of valuation of goods which had been lost or damaged. He also made it clear that any such wording which might be proposed in the future must also concern only the basis on which goods subject to physical loss or damage were to be valued. Mr. Mullen did not suggest that the wording was intended to include cover for purely financial risks not consequent on physical loss or damage. Mr. Beattie told Mr. Mullen to go back to the Bank and clarify what they were hoping to achieve with the suggested changes. He scratched it for receipt and placed it on the RSA file. But Mr. Mullen did not pursue the changes at the time or afterwards. There was no attempt to develop the wish list into an endorsement.
  449. Mr. Beattie expanded upon this in his second statement. He said that after scratching the wish list for receipt, he asked Mr. Mullen to go back to the client and return with a full explanation of their requested changes. The wish list was a discussion document and not a broke of an endorsement. He asked Mr. Mullen to obtain precise detail of exactly what extension of coverage they were looking to achieve and why this was being presented mid-term.
  450. In cross-examination by Ms. Healy, Mr. Beattie said that he asked Mr. Mullen what he (Mr. Mullen) expected him (Mr. Beattie) to do with the document. He told him that he did not understand what was being requested of him. He asked whether Mr. Mullen's understanding of the TPC was that it was an "add-on to the basis of valuation", which was how Mr. Beattie interpreted it. Mr. Mullen was "from recollection non-committal but did not disagree". So he asked Mr. Mullen to go back to the client and find out "exactly what the client[s] were seeking to achieve with this three-page wish list or discussion document". He said that he read through the document. It seemed disorganised and he did not understand the reasoning behind the request or what he was being asked to do. Hence his request to Mr. Mullen to go back to the client.
  451. His evidence in response to questions from Ms. Sabben-Clare was to similar effect. He denied that he was reconstructing what he thought ought to have happened, rather than giving evidence of an actual recollection. This was, he said, "a major change which had been requested to the cover", and he recalled what happened "pretty clearly". He thought that he had asked: "what do you expect me to do with this", which implied that he was surprised by the level of changes requested. When he asked Mr. Mullen if he understood what was being asked of him, he did not receive a reply that suggested that Mr. Mullen understood any better than Mr. Beattie did. Mr. Mullen said, basically, that he was unsure. As far as the TPC itself was concerned, he said that he gave his interpretation, and then asked Mr. Mullen if he understood it in the same way, "to which I got no confirmation or denial". At that point, he told Mr. Mullen to go back to the Bank "and find out from them exactly the intention of the proposed change". When he asked Mr. Mullen the question, Mr. Beattie could not recall the exact words of the reply: "He shall we say fudged the issue". This led Mr. Beattie to ask Mr. Mullen to go back to the client "and ascertain from them exactly what they are looking for".
  452. In relation to the 20 January 2016 meeting, when Mr. Beattie signed the quotation slip, Mr. Beattie's evidence in his witness statement was that he met both Mr. Mullen and Mr. Lockyer. There was a discussion about the TPC. Mr. Beattie made it clear to Mr. Mullen that he understood this to relate only to the basis of valuation of the insured goods under the policy. It was clear to Mr. Beattie that there was not intended to be any extension of cover. On the basis that the wording related solely to a 'tied down' basis of valuation, Mr. Beattie agreed to its inclusion. He was satisfied that Mr. Mullen accepted his understanding of the meaning of the TPC wording. He did not disagree when Mr. Beattie said what it meant. Nor did he offer any other meaning.
  453. In response to questions from Ms. Healy, Mr. Beattie said that he could not recall exactly how he spotted the TPC in the quotation slip. The wish list had not been developed further, but "certain elements of it made its way into the slip" including the TPC which is when Mr. Beattie said to Mr. Mullen: "look this is what I understand this to mean as per our previous conversation and I am correct in that assumption, aren't I". At no point did Mr. Mullen point out the error of Mr. Beattie's ways.
  454. In response to questions from Ms. Sabben-Clare, Mr. Beattie said (echoing an answer he had given to Ms. Healy) that he had said that if his understanding of the TPC was incorrect, "he should find out from the assured exactly what they understand by the wording". He could not recall the exact words used but the "overall question was that I didn't understand this wording, this is my interpretation, please confirm or otherwise and if this is not – if my understanding is incorrect, please speak to our client and find out exactly what they are looking for". Mr. Mullen's response was neither to confirm nor deny that his understanding was correct.
  455. The meeting on 29 January 2016 concerned the revised policy where the TPC was moved from Clause 1.5 in Section 2 to two other places. In his witness statement, Mr. Beattie described Mr. Mullen presenting the wording on the basis that the assured wanted to make some changes to the presentation of the wording in the body of the policy. Mr. Beattie made it clear that the TPC could be moved "on the basis that it related only to the basis of valuation". Mr. Mullen did not disagree with his understanding of the clause. Mr. Lockyer's subsequent email referred to cosmetic amendments. Mr. Beattie understood that the changes were aesthetic, following a review by the Bank's own in-house legal team.
  456. In cross-examination by Ms. Healy, Mr. Beattie said that he recalled asking Mr. Mullen why he had asked for the TPC to be included twice. He also said that he made enquiries of Mr. Mullen as to the "reasoning behind this. Again [I] volunteered my interpretation of this being an add-on to the basis of valuation and still went back to asking what the assured actually were looking to achieve with these wordings and why they needed doing". Mr. Mullen told him that the move was a minor thing that the clients had asked him to do, but he could give no reason. There was an open discussion where Mr. Beattie gave his interpretation, and said that if the client was looking for something other than his interpretation, then Mr. Mullen should advise him accordingly and they would discuss.
  457. Ms. Sabben-Clare asked him whether he pressed for a response to the question of why the clause had been included twice. Mr. Beattie could not recall, but said that perhaps he did not. He was asked whether he really did let matters lie with non-committal answers given on the three occasions when the TPC was allegedly discussed (July, 20 January and 29 January), and Mr. Beattie said: "I am afraid I did".
  458. Mr. Mullen's evidence as to the three relevant meetings

  459. In relation to the July 2015 meeting, Mr. Mullen's evidence in his witness statement was that he could not recall the precise details of putting forward the July document to Mr. Beattie. He thought that he would have told him that there was "quite a lot of new stuff in here" or words to that effect. Mr. Beattie, as was his usual practice, would have read the document and would have asked any questions or made any objections if he had them, although he could not remember him doing either. Having read it, Mr. Beattie stamped, initialled and dated it on each page. This showed that he agreed the entirety of the document (to which Mr. Mullen referred in his statement as an "endorsement"). Mr. Beattie did not tell him that he required the wording to be shown to the following market.
  460. Mr. Mullen said that he did not describe the document as a wish list, and that was obviously not what the document was. In his second statement, however, Mr. Mullen said that on reflection (having reviewed an email sent to Mr. Stroink in which that phrase was used) he might have used the words "wish list", although he could not remember doing so.
  461. He said that there was no reason to waste the time of a very senior marine cargo underwriter with a wish list as opposed to a formal endorsement. Whilst he did not recall exactly what was said at the meeting, he thought that he would have described the document as an "endorsement for you to consider" or something similar. Mr. Beattie did not tell him that he was signing the endorsement for "receipt only". His experience was that if an underwriter scratches, dates and stamps every page of a contractual endorsement, he does so in order to show that he is agreeing to it. In his second statement, he said that he had never known Mr. Beattie or any other underwriter to scratch an endorsement "for information and receipt only".
  462. Mr. Mullen did not accept that Mr. Beattie made it clear that he understood the TPC to be concerned with the basis on which the insured goods which were subject to physical loss and damage were to be valued. He did not tell Mr. Beattie that he did not know why the Bank wanted the TPC. Had he been asked what the TPC was for, he would have said that Icestar wanted cover for the loss of the transaction premium which was not contingent on physical loss or damage. If Mr. Beattie had said anything at odds with this, he would have told him that he had misunderstood. On no occasion did Mr. Beattie express the view that the TPC was a basis of valuation, or that the cover it provided was contingent on physical loss or damage. Had he said this, Mr. Mullen would have corrected him since it was not his understanding. He was sure that Mr. Beattie had never said that he was only prepared to agree to the TPC on the basis that it concerned valuation only, and he was sure that he had not agreed to any such limitation on its scope. It would have been the opposite of what his client wanted. Had Mr. Beattie proposed such a limitation, he would have promptly explained to the Bank that RSA was not willing to provide the cover requested.
  463. In his second statement, Mr. Mullen largely reiterated these points, by way of response to Mr. Beattie's written statement. He denied that Mr. Beattie told him at the meeting to go back to the Bank to clarify what they were hoping to achieve by the suggested changes. Had this happened, he would not have told Ms. Barnes shortly after the meeting that the endorsement had been agreed. If Mr. Beattie had said that he was not interested in performing a policy re-write mid-term, Mr. Mullen would have sought to persuade him otherwise and, if that failed, would have explained the situation to the Bank. If Mr. Beattie had set out his understanding of the TPC in the way alleged, he would have corrected him. If he was only willing to write the TPC on that basis, he would have explained to Icestar that Mr. Beattie was unwilling to write the cover, and it is likely that a new lead underwriter would have been sought. He did not recall Mr. Beattie asking any questions regarding the endorsement.
  464. Mr. Mullen's evidence, as expressed in his witness statement as to these meetings, and what he would have done if certain things had been said, was considerably more focused, clear and coherent than his evidence under cross-examination. This was, no doubt, because of the considerable care that is usually taken in drafting witness statements with the result that they frequently become an articulation of the best points that can be made in support of a party's case. The value of cross-examination is to expose such frailties as may exist. This was graphically illustrated when Mr. Mullen was cross-examined on a passage in his witness statement which addressed the 18/19 June 2015 meeting in Amsterdam, which formed the backdrop to the broking of the July document. It was at that meeting when the Bank's draft of their desired endorsement was given to Mr. Mullen. His witness statement referred to discussion to the effect that Icestar required cover against a customer defaulting on its obligations to make payment in respect of the particular commodity, and continued:
  465. "What Gijs wanted was to insure against a customer failing to repurchase. In effect the TPC was to expressly incorporate credit risk cover into the 2015 Cargo Policy. That cover was not to be contingent on physical loss or damage to the cargo".
  466. In cross-examination, however, Mr. Mullen could not really understand the reference to "credit risk cover", and distanced himself from that passage in his witness statement. Indeed, Ms. Healy submitted that the phrase "trade credit insurance", widely used by the lawyers and experts in the case, was not a label which Mr. Mullen applied when he broked the July endorsement or the 2016 policy. He was not familiar with trade credit insurance and did not think of it in those terms. Having heard Mr. Mullen's evidence as a whole, I consider that this submission was well-founded. It does, however, call into question the accuracy of the above passage in his witness statement.
  467. None of this is to say that Mr. Mullen's evidence on the key issues should be rejected, but I do treat the clarity of his evidence in his witness statements with a considerable degree of caution. I now turn to describe his evidence in cross-examination in more detail.
  468. Mr. Mullen agreed that the meeting on 28 July 2015 had been a short one: he was in his office at 8.33 am (London time) when he sent an e-mail to Ms. Barnes, and had returned by 9.29 am (London time), having walked over to Mr. Beattie's office, a journey that would take around 5 – 10 minutes. It was, he said, a short routine meeting: he couldn't recall how long it was, but meetings with Mr. Beattie would usually last around 15 minutes. He agreed that Mr. Beattie would probably spend 10 minutes reading the document. He told Mr. Beattie that it had been drafted by the Bank's in-house lawyers. He said that sometimes Mr. Beattie asked a lot of questions, and sometimes he would ask none. On this occasion, he saw nothing untoward in the document. He did not recall Mr. Beattie asking any questions. If he had done so, Mr. Mullen would have answered them. Mr. Mullen was not "totally surprised" by the fact that Mr. Beattie did not ask questions, "because of his seniority and knowledge of the business". In response to one question, Mr. Mullen accepted that he had described the document as a wish list or note of changes that the lawyers wished to make; but later on said that he honestly couldn't remember whether he had used that term and doubted whether he had done so. He did not wish to dispute what Mr. Beattie said on that score, because "Brian is a good friend of mine". He disputed that he would have described the document as a "note for discussion".
  469. In terms of his positive recollection of the meeting, Mr. Mullen recalled being very pleased that Mr. Beattie was available to see him immediately. He sat alongside him, and was very pleased that he had read through it, seemed very satisfied and stamped the document and sent him on his way. Mr. Beattie had never said that he would only consider the changes in the context of a renewal. He did not say that he did not understand what was being requested of him with the amendments: Mr. Mullen said that he thought that Mr. Beattie "did because that's why he stamped it, Mr. Parsons". If he was not prepared to do it, he would not have stamped it. If he did not understand it, Mr. Mullen would have expected Mr. Beattie to have written something on the document, for example "for clarification".
  470. Mr. Mullen did not recall Mr. Beattie ever saying to him that the wording was a basis of valuation for goods which are lost or damaged. He did not believe that he had said it. If it had been said, Mr. Mullen said that he would then have had to take instructions from the client, since it would indicate that Brian (ie Mr Beattie) was not wholly happy with the changes to Endorsement 4. If Mr. Beattie had said that he understood it as a basis of valuation clause, Mr. Mullen could not say whether he would have agreed or disagreed with what Mr. Beattie had said. He said: "I had the wording in front of me, which was sacrosanct as far as I was concerned … I'm sure Brian didn't mention it, about the BOV [Basis of Valuation]". He said that he clearly remembered Brian not asking any questions. He was positively sure that Brian had not said that he understood the clause as a basis of valuation. He totally disagreed with the suggestion that he was sent away to find out what the clients wanted, or that the document was only signed "for receipt".
  471. In relation to the meeting on 20 January 2016, Mr. Mullen said in his witness statement that he did not remember the precise details of the discussion. He concluded from the email which he sent to the Bank on 20 January, which referred to a "full discussion" to "make sure that we have no hidden surprises", that he had gone through the proposed policy in some detail. He believed that Mr. Beattie would, as usual, have gone through the slip, asking any questions he might have had. He denied that there had been a discussion (as put forward in the insurers' response to a request for further information) where Mr. Beattie had explained the TPC to relate to the basis of valuation in the event of physical loss or damage, that he was not extending coverage, and that Mr. Mullen would have to take that interpretation back to his client, and on that basis he was prepared to agree to the inclusion of the wording. Mr Mullen said that it would have been extraordinary if he had done this, or for him to have gone along with it. It would be the complete opposite of what he clearly understood that his client wanted. If this conversation had happened, Mr. Beattie would have made sure that this was recorded in the policy.
  472. In cross-examination in relation to this renewal meeting, Mr. Mullen did not specifically recall this meeting: there were so many. But there definitely was a meeting, and he thought that Mr. Beattie had already put on his screen the endorsements showing the changes during the previous year. It was again put to Mr. Mullen that Mr. Beattie had said that he understood the TPC to be a basis of valuation clause, and that Mr. Mullen did not disagree or offer any other meaning. Mr. Mullen said that he really didn't recall Mr. Beattie saying that. He pointed out that Mr. Lockyer was also at the meeting, and he did not think that either of them would have forgotten it if it had been said. He said that the wording in the policy was very clear: it was default insurance.
  473. In relation to the 29 January 2016 meeting, Mr. Mullen's written evidence was that he walked with the final amended slip to Mr. Beattie, and told him that the client insisted upon the TPC being included in Section 4 as well as Section 1. He asked him to agree this, which he did. He again denied that there was any discussion about Mr. Beattie's interpretation of the clause, either on 20 January or 29 January.
  474. "At no time did Mr. Beattie give the interpretation of the TPC described above or say this was the only basis on which he was willing to agree to the clause. Had he done so, I would have corrected his understanding of the TPC, and if he was unwilling to agree to the TPC on that basis, I would have explained the situation to Icestar and sought their instructions."
  475. On the third day of his cross-examination he was asked about this meeting. He said that following Mr. Lockyer's question on 28 January 2016 (described in context in Section B above, and set out below in the context of Mr. Lockyer's evidence), he had discussed the TPC with Mr. Lockyer. He said that it was "very clear: it does not rely on any physical loss or damage, it's through delay. And if – and if Brian was happy with that he – and he was – he would agree it. I think Lee, after we discussed it, was also of that opinion, too". The following exchange took place under cross-examination by Mr. Parsons:
  476. "Q. Well can you recall the 29th, the meeting?
    A. I honestly can't, no.
    Q. Okay. So Mr Beattie says that he was happy to agree having the TPC moved on the basis that it only related to the basis of valuation and that's what he told you. Again, do you agree, disagree or can't recall?
    A. If it was in this document, this document you're showing me, and I imagine it was in the two places, he obviously agreed it.
    Q. Yes. But my question was: he told you that he was doing it on the basis that it was a basis of valuation clause still. Do you recall that and do you -- well, do you recall that?
    A. I don't recall him saying that, no.
    Q. He might have said that?
    A. "He might have said that?", did you say?
    Q. Yes. He might have said that to you?
    A. Again, I can't recall the conversation but he may very well have done."

    Mr. Lockyer's evidence

  477. Although Mr. Lockyer was involved in analysing the proposed endorsement in July 2015, he did not accompany Mr. Mullen to the 28 July 2015 meeting. He did, however, attend the 20 January 2016 meeting. In his statement, he agreed with Mr. Mullen's evidence that there would have been a discussion in full with Mr. Beattie on that occasion, but he did not personally recall the meeting.
  478. He went on to address the case, set out in the underwriters' pleadings, concerning the alleged discussion about the TPC as a basis of valuation clause. At that stage, it was not clear from the pleadings which particular meetings in January were being referred to. But Mr. Lockyer deduced that the relevant meeting could only have been the 20 January meeting, since Mr. Mullen had not attended on 25 or 27 January when Mr. Beattie had entered his line and then added further signatures to the slip. Mr. Lockyer said, based on his e-mail exchange with Mr. Mullen on 28 January, that he did not think that the underwriters' account could be correct. In that e-mail exchange, Mr. Lockyer had (after receiving the mark-up with Ms. Van de Beek's comments) drawn attention to what was being said about the TPC clause in the following terms:
  479. "I have seen the comments made by Pauline and it appears her main concern is regarding Transaction Premium. Reading the clause it mentions that the insured is covered by this policy for the transaction premium they would have earned if client of the insured defaults, regardless whether there has been any physical loss or damaged (sic) to the goods.
    Am I reading this correctly, and is this understanding of underwriters."
  480. Basing himself on this exchange, Mr. Lockyer reasoned in his witness statement as follows:
  481. "If the discussions alleged by underwriters had taken place during the 20 January 2016 meeting (which I attended), I would have known that Mr. Beattie understood the TPC to be a basis of valuation, as underwriters allege he did. In those circumstances, it would have made no sense for me to ask David if it was underwriters' understanding that the TPC provided cover "regardless whether there has been any physical loss or damage to the goods" as I did on 28 January 2016 at 12:21. For that reason I do not think that the underwriters' account can be correct."
  482. Mr. Lockyer was cross-examined on this exchange. It was put to him that he had not appreciated, until he saw Ms. Van de Beek's comment, that the TPC was a separate trade credit insurance or cover without physical loss or damage. He said that that was not true. He understood what the clause required, namely that in the event that the client of Icestar defaulted on its obligation or its right to purchase back those goods or default on any payment, then the clause was a stand-alone policy or clause. In his mind, it had always been a standalone clause. He said that he had read the clause before 28 January, and had thought about its meaning before then.
  483. This evidence was given prior to the late disclosure of the document which Mr. Lockyer had written on 8 July 2015 (see Section B above), when he recorded his analysis of the clauses in the proposed endorsement. In my view, the disclosure of that document confirmed that Mr. Lockyer's evidence, described in the previous paragraph, was accurate.
  484. Mr. Lockyer was asked about the 20 January 2016 meeting. He said that he did not remember the meeting, but it was clear from the correspondence that he was there. There was then the following exchange:
  485. "Q. If Mr. Mullen and Mr. Beattie discussed the TPC clause as a basis of valuation clause, that's something that you probably wouldn't even remember or remark on. That's fair, isn't it?
    A. Yes. I can't recall that being referred to as a BOV clause.
    Q But if it had been – you see Mr. Beattie says that he told Mr. Mullen that he understood the TPC related only to the basis of valuation. If he did say that, it wouldn't be too surprising if you can't now remember.
    A. Correct. I mean it was such a long time ago."

    E3: The parties' arguments

    The insurers' case

  486. The insurers rely upon the evidence of Mr. Beattie as to what transpired at these three meetings.
  487. In relation to the July 2015 meeting, the insurers contend that Mr. Mullen presented the document, which set out proposed changes, as a "wish list". Mr. Beattie read through the wish list, and made it clear to Mr. Mullen that he understood that the TPC wording concerned the basis of valuation of goods which had been lost or damaged, and that any such wording which might be proposed in the future must also concern only the basis of valuation. Mr. Mullen did not then suggest that the TPC wording was intended to extend cover for trade credit or any financial risks not consequent on physical loss or damage. The upshot of the meeting was that Mr. Beattie did not agree to the wish list. Instead, he told Mr. Mullen to go back to the Bank and clarify what they were hoping to achieve with the suggested changes. Mr. Beattie was then requested by Mr. Mullen to scratch the wish list "for receipt", and did so. Mr. Beattie therefore made it clear that he was not prepared contractually to accept any of the alterations set out on the wish list at that stage.
  488. In relation to the renewal discussions in January 2016, Mr. Beattie expressly confirmed with Mr. Mullen that the TPC was simply a basis of valuation clause. This happened on two occasions: when Mr. Beattie signed the quotation slip on 20 January, and again on 29 January when Mr. Beattie was asked to agree to changes in the slip and in particular the move of the TPC from Clause 1.5 to the two other locations in the policy.
  489. The key findings for which the underwriters contended were as follows. Both Mr. Beattie and Mr. Mullen understood that each had the subjective intention that the TPC was a "basis of valuation" extension, and did not extend to trade credit or financial guarantee. That understanding was either a binding verbal agreement between them, giving rise to a collateral contract, or the basis of a claim for rectification. Alternatively, if Mr. Mullen did subjectively intend to place cover for trade credit/ financial guarantee risks via the TPC, he nevertheless knew that underwriters were providing cover under the policy on, and only on, the assumption that the policy did not cover such risks. Accordingly, by presenting the written policy without correcting the assumption that he understood was being made by Mr. Beattie, Mr. Mullen acquiesced in it. In those circumstances, the Bank is estopped by convention from relying on the TPC as against RSA as anything other than a basis of valuation clause.
  490. The argument for a collateral contract was, in essence, that Mr. Mullen entered into a contract with Mr. Beattie such that the Bank promised to interpret the TPC in the 2016 policy as a basis of valuation clause, and Mr. Beattie promised to agree to the policy on that basis. This argument was based on the conversations in both 2015 and 2016.
  491. The argument for rectification was that both Mr. Mullen and Mr. Beattie subjectively intended that the TPC did not provide standalone credit/ financial guarantee cover, and related only to the recovery by the Bank of its profit element on the underlying transaction should the Bank have a valid claim under the policy. In other words, the TPC was a basis of valuation clause. Mr. Mullen had himself never intended to place standalone credit/ financial guarantee cover. There was an outward expression of accord on this point, operative at the time of the 2016 renewal, such that they each understood each other to share this common intention. The underwriters invited the court to accept the evidence of Mr. Beattie as to the conversations.
  492. In this context, they submitted that if there was an outward expression of accord in 2015, that was sufficient because it would still have been continuing in 2016. But they also maintained that rectification did not depend upon the court accepting the evidence of Mr. Beattie as to these conversations. They submitted that even if there was no discussion about the clause at all, there would nevertheless have been a tacit understanding between Mr. Mullen and Mr. Beattie of the other's intention. This was for a number of reasons. It was well understood between them, without words being required, that only cargo risks, not trade credit was being covered. In addition, the contractual background was a number of revisions to the BCC cover which they described as relating to the basis of valuation; so that, even in the absence of discussion, the TPC would have been understood in that vein. It was commercially absurd for either to have any different intention. These various factors would outweigh even clear words to the contrary in the TPC (although underwriters disputed that there were any such clear words).
  493. As for estoppel, the underwriters relied upon Mr. Mullen's repeated lack of protest in response to Mr. Beattie's declarations as to his understanding of the TPC as a basis of valuation clause. Mr. Mullen acquiesced in that assumption being made by Mr. Beattie, particularly given the backdrop of his duty of utmost good faith. The other requirements for an estoppel were also made out; ie it would be unjust to allow the Bank to resile from the assumption.
  494. Even if there was no discussion at all, then there would still be an estoppel. This is because Mr. Mullen (and if relevant Mr. Lockyer, who was involved in the placement in 2016) were acquiescing by their silence in an assumption which they knew was being made by Mr. Beattie and the following market. They knew that the cargo market, in which they were placing the risk, would not knowingly provide credit or financial guarantee cover. They therefore knew that underwriters were signing on the basis of an assumption that the policy did not cover trade credit/ financial guarantee, and also that it was being assumed by Mr. Beattie to be the latest clarification of the basis of valuation clause. Mr. Mullen and Mr. Lockyer therefore conducted themselves by choosing to rely solely on a written presentation of a risk that they knew, if explained, the underwriters would not provide. In such circumstances, their silence amounted to acquiescence in the assumption being made and it would again be unjust for the Bank to be permitted to depart from the assumption.
  495. It will therefore be apparent that whilst the insurers' primary case was based upon discussions between Mr. Beattie and Mr. Mullen, they also argued that a claim for rectification or a case of estoppel by convention could arise even in the absence of any relevant discussion of the TPC.
  496. As far as the following market is concerned: it was not suggested that they knew anything about the conversations between Mr. Mullen and Mr. Beattie. However, the presentation of renewal terms to each of the other underwriters (D2 to D14) involved an implied representation to them that the terms presented were those agreed by the slip leader. In those circumstances, the following market was agreeing to what RSA had agreed to do, inclusive of any collateral agreements, accords or estoppels which touched the meaning or proper interpretation of what had been agreed. In addition, the estoppel is (as explained above) not dependent upon there having been any statements made by Mr. Beattie. Alternatively, the implied representation amounted to a promise by the Bank that it would not enforce any rights against the following market which were different to those against RSA; the following market relied upon that promise; and the Bank is estopped by reason of that promise from asserting otherwise.
  497. The Bank's argument

  498. The Bank submitted that Mr. Beattie's evidence, even taken at its highest, would not meet the requirements for an estoppel by convention. This was significant because the requirements for an estoppel by convention could be satisfied more easily than those for rectification or a collateral contract. They submitted that on Mr. Beattie's version of events, no agreement was reached and there was no clear and unequivocal representation made by either of them. Mr. Mullen, even on Mr. Beattie's own account, never provided the confirmation which Mr. Beattie had sought. Mr. Beattie signed the contract anyway. This was a classic situation of risk taking: Mr. Beattie knew that a doubt existed about the effect of the TPC, and this had not been resolved when he bound RSA to the risk. In July 2015, on Mr. Beattie's evidence, there was no consensus as to the meaning or effect of the TPC. In January 2016, the position remained (on Mr. Beattie's evidence) a "fudge". In any event, there was no legal route by which other underwriters could rely on an agreement made only by RSA. There was, therefore, never any shared assumption.
  499. In her oral submissions, Ms. Sabben-Clare said that there was no clarity about the assumption relied upon; that Mr. Mullen said or did nothing which involved him assuming responsibility for any common assumption; and that Mr. Mullen did not do or say anything that could properly be regarded as assent to the assumption. They submitted that Mr. Beattie's evidence simply amounted to an express, inconclusive discussion. This was insufficient for rectification, collateral contract or estoppel.
  500. Edge's argument

  501. In relation to the position in July 2015, Edge submitted that although Mr. Mullen may not at first have understood or seen the relevance of the TPC, he did so after he had asked his colleague Mr. Lockyer to look at it and, subsequently, had received a clear explanation from Ms. Barnes in her e-mail of 10 July 2015. He then tried to prepare an endorsement on the usual "MRCE" form, but technology defeated him and he ended up preparing a Microsoft "Word" document, but with the intention that this should have the same contractual effect as an endorsement to the policy in the more usual form. He then saw Mr. Beattie on 28 July 2015, telling him that these were changes that Icestar wanted to make and that it had been drafted by Icestar's lawyers. Mr. Beattie then read it, for approximately 10 minutes, and then signed and scratched it. This was signature by way of agreement, rather than simply for information or receipt. Mr. Beattie did not say that he understood the TPC to be a basis of valuation, and that he was only willing to agree it on that basis.
  502. As far as the renewal in January 2016 was concerned, Edge submitted that Mr. Beattie read the slip again, and saw the TPC in Section 2. He did not object to the inclusion of the TPC, and did not question why it had been added to the slip. Nor did he question why the non-avoidance clause had been added into the slip. He scratched the quotation slip on 20 January 2016 to indicate his support for the renewal on those terms. He put down his line on the slip, thereby binding the RSA, on 25 January 2016 and signed it again on 27 January 2016. On 29 January 2016, after a brief discussion, he agreed to the change whereby the TPC was moved to two places in the policy wording. On none of those occasions during renewal did Mr. Beattie say or make clear to Edge that he understood the TPC to be a basis of valuation which applied in the event of physical loss or damage. In any event, even if Mr. Beattie did say this or make it clear, it was not accepted by Mr. Mullen, so that there was no agreement between them.
  503. On this factual basis, there was no case for rectification, collateral contract or estoppel.
  504. E4: Discussion

    Approach to the evidence

  505. In assessing the evidence of the factual witnesses on all issues, I adopt the approach commended by Robert Goff LJ in Armagas Ltd v Mundogas S.A. (The Ocean Frost), [1985] 1 Lloyd's Rep. 1, 57:
  506. "Speaking from my own experience, I have found it essential in cases of fraud, when considering the credibility of witnesses, always to test their veracity by reference to the objective facts proved independently of their testimony, in particular by reference to the documents in the case, and also to pay particular regard to their motives and to the overall probabilities. It is frequently very difficult to tell whether a witness is telling the truth or not; and where there is a conflict of evidence such as there was in the present case, reference to the objective facts and documents, to the witnesses' motives, and to the overall probabilities, can be of very great assistance to a Judge in ascertaining the truth."
  507. In the same case, Dunn LJ said (to similar effect):
  508. "I respectfully agree with Lord Justice Browne when he said in re F, [1976] Fam. 238 at p. 259, that in his experience it was difficult to decide from seeing and hearing witnesses whether or not they are speaking the truth at the moment. That has been my own experience as a Judge of first instance. And especially if both principal witnesses show themselves to be unreliable, it is safer for a Judge, before forming a view as to the truth of a particular fact, to look carefully at the probabilities as they emerge from the surrounding circumstances, and to consider the personal motives and interests of the witnesses. As Lord Wright said in Powell v. Streatham Manor Nursing Home sup. at p. 267:
    … Yet even where the Judge decides on conflicting evidence, it must not be forgotten that there may be cases in which his findings may be falsified, as for instance by some objective fact …
    and he referred in particular to some conclusive document or documents which constitute positive evidence refuting the oral evidence of the witnesses."
  509. The approach of Robert Goff LJ was approved by the Privy Council in Grace Shipping v Sharp & Co [1987] 1 Lloyd's Rep 207 at 215-216:
  510. "And it is not to be forgotten that, in the present case, the Judge was faced with the task of assessing the evidence of witnesses about telephone conversations which had taken place over five years before. In such a case, memories may very well be unreliable; and it is of crucial importance for the Judge to have regard to the contemporary documents and to the overall probabilities.
    That observation [ie of Robert Goff LJ] is, in their Lordships' opinion, equally apposite in a case where the evidence of the witnesses is likely to be unreliable; and it is to be remembered that in commercial cases, such as the present, there is usually a substantial body of contemporary documentary evidence."
  511. Robert Goff LJ's judgment was described as the "classic statement" in Simetra Global Assets Ltd. v Ikon Finance Ltd. [2019] EWCA Civ 1413, where Males LJ said at [48]:
  512. "In this regard I would say something about the importance of contemporary documents as a means of getting at the truth, not only of what was going on, but also as to the motivation and state of mind of those concerned. That applies to documents passing between the parties, but with even greater force to a party's internal documents including emails and instant messaging. Those tend to be the documents where a witness's guard is down and their true thoughts are plain to see. Indeed, it has become a commonplace of judgments in commercial cases where there is often extensive disclosure to emphasise the importance of the contemporary documents. Although this cannot be regarded as a rule of law, those documents are generally regarded as far more reliable than the oral evidence of witnesses, still less their demeanour while giving evidence."
  513. Robert Goff LJ's approach is also reflected in other recent authority such as Gestmin SGPS SA v Credit Suisse (UK) Ltd [2013] EWHC 3560 (Comm), at [15]-[23]. In Gestmin, Leggatt J referred at [22] to the appropriateness of basing factual findings on inferences drawn from the documentary evidence and known or probable facts. As explained by the Court of Appeal in Kogan v Martin [2019] EWCA Civ 1645, however, this does not mean that oral testimony from witnesses should be disregarded.
  514. Accordingly, my approach is to consider the objective evidence and in particular the documentary evidence, as well as the inherent probabilities, and to test the accounts of the witnesses against those matters.
  515. The July 2015 meeting

  516. The documentary evidence clearly shows that, prior to the meeting with Mr. Beattie, Mr. Mullen knew that his clients wished to have contractually agreed amendments to the 2015/2016 policy wording. For example, on 25 June, he told the Bank that the "language for amendments" needed to be discussed with the lead underwriter point by point, and that the "endorsement" would require the signature of all underwriters. On 12 July, he told Ms. Barnes that he would start drafting up "formal endorsements" based on her answers to the questions which he had asked. On 20 July 2015, he referred to the "requested amendments to the contract wording".
  517. It was on that day that he actually started work on the preparation of the physical document that he intended to show to Mr. Beattie. In his oral evidence under cross-examination, Mr. Mullen referred (for the first time) to his having attempted, unsuccessfully, to use the template for endorsements which was on the Edge system. This evidence was treated with some scepticism at the time by counsel for the insurers. However, an overnight search for documents in the light of that evidence led to the disclosure of an electronic folder of documents. This demonstrated that Mr. Mullen's recollection was correct. He had indeed tried to use the template for endorsements which was on the Edge system. I accept his evidence that he was unable to do so, essentially because of the amount of text that he was seeking to include within the endorsement. This was the reason why, ultimately, the document which Mr. Mullen prepared, and which he presented to Mr. Beattie on 28 July 2015, was an ordinary Microsoft Word document. The fact that the document was not on the MRCE form (ie the standard form for policy endorsements) does not show that Mr. Mullen was not looking to obtain agreement on varied contractual terms. He clearly was.
  518. The expression "wish list" had been used in an e-mail by his colleague, Mr. Lockyer, and it is likely that it was used by Mr. Mullen during the course of his meeting with Mr. Beattie. But this expression does not connote a lack of intention to introduce new or varied contractual terms. It can be, and was in this case, a "wish list" of contractual amendments which a party wishes to include.
  519. It is also clear that, immediately after the meeting, Mr. Mullen told the Bank that agreement had been reached. His email to the Bank that morning apologised for the time taken to "complete the review and endorsement". But he said that he was now "pleased to confirm all agreed by Underwriters and attached please find the initialled endorsement". He attached a copy of the 3-page document signed and stamped on each page by Mr. Beattie. He then referred to his intention to prepare and send "[o]ur official endorsement". On the same day, he sent a typed document which was headed: "Endorsement attaching to and forming part of Cover Note No. QM 349770".
  520. I have no doubt that Mr. Mullen did genuinely believe that he had obtained Mr. Beattie's agreement to the endorsement. The underwriters contend that Mr. Mullen was, at this stage, acting dishonestly in the way that he reported back to the Bank, since he knew that (if Mr. Beattie's evidence were accepted) Mr. Beattie had been unwilling to agree the endorsement and had sent Mr. Mullen away to ask for clarification and further information. I reject this argument. It would have been the most serious dereliction of Mr. Mullen's obligations to his client, and to his employer, to report that agreement had been reached on the terms of the signed document, and then to send out Edge's cover note, knowing that the underwriter had refused to agree terms. Even though some time had passed since he had first been requested by the Bank to obtain agreement on the amendments, and Mr. Mullen appears to have acted rather slowly, I do not consider that this would provide a reason or motive for Mr. Mullen dishonestly to mislead his clients into believing that they had cover on the amended terms when he knew that they did not. The documentary record does not contain any assurance from Mr. Mullen that he would be able to obtain the underwriters' agreement. Indeed, his email of 25 June was (as Mr. Mullen said in evidence) intended to lower expectations, by indicating that agreement by the lead underwriter was not a foregone conclusion: the amendments would need to be discussed point by point. It was also apparent, from the list of questions in the 9 July e-mail to the Bank, that there were a number of potential issues on the amendment which arose: some points were "Accepted" or "Fine no difficulty", but others required explanation. Mr. Mullen had not, therefore, boxed himself into a corner from which his only escape was to tell his client that the underwriters had agreed to the proposed wording when they had not done so. If Mr. Beattie had declined to agree to the terms of the endorsement, or had raised questions to which he required a response, Mr. Mullen would in my view have reported back to the Bank and obtained answers to the questions that were asked.
  521. Before further discussing the issues concerning the July endorsement, it is appropriate at this point to address more generally the question of Mr. Mullen's reliability as a witness. The underwriters and the Bank identified a number of instances, where it was said that Mr. Mullen had acted dishonestly, and that this should be taken into account in assessing his credibility and reliability. Both the underwriters and the Bank relied upon a sequence of e-mail exchanges with Mr. Shillabeer of Channel (D9) in July 2017. Mr. Shillabeer asked Mr. Mullen to advise when the TPC was introduced into the policy, as he could see it on the 2016 policy but not on the 2015. In a later e-mail, Mr. Shillabeer asked for copies of any endorsements from the 2015 policy since they were not on Channel's system. The e-mail exchanges culminated with Mr. Mullen explaining that his files were all with his solicitors, so that he could only:
  522. "produce our file document that was initialled by lead Underwriters and noted lists to all. These amendments were a combination of various previously agreed endorsements left of previous contracts. As soon as I get my files back I will pass it on. Thanks and sorry about bad memory".
  523. I did not consider that this sequence of e-mails contained any dishonest statements. Mr. Mullen did not have his files, and was speaking from memory. In the context of the e-mail exchange as a whole, he was not in my view suggesting to Mr. Shillabeer (as underwriters argued) that the July 2015 endorsement was itself a combination of earlier endorsements. He was in my view indicating, correctly, that the amendments to the 2016 policy (about which Mr. Shillabeer had asked in his first e-mail in the sequence) were a combination of various previously agreed endorsements. It is also important to note that no allegation of fraud was made by underwriters in relation to the placement of the July 2015 document or the 2016 policy, or indeed as to any aspects of the placement of the cover in prior years. Mr. Mullen's cross-examination involved scrutiny of his dealings with underwriters over many years and many e-mails, and the e-mail sequence with Mr. Shillabeer was the only instance where it was alleged that he had acted dishonestly towards them. Although Mr. Beattie in his evidence sought to cast doubt on Mr. Mullen's integrity generally, I was not impressed with that criticism, bearing in mind that Mr. Beattie had dealt with Mr. Mullen very extensively over many years. Other underwriters indicated that they had never had a problem with Mr. Mullen. In these circumstances, the evidence does not support a suggestion that Mr. Mullen acted with a lack of integrity towards underwriters.
  524. The Bank (and underwriters) were, however, able to point to various e-mails sent by Mr. Mullen in July 2015 where he did misreport to the Bank the work that was ongoing or which had taken place in order to obtain agreement to the terms of the endorsement. Mr. Mullen was there giving the impression that he was working on obtaining underwriters' agreement for some time, and that the process was not straightforward. The reality was that Mr. Mullen did not get down to doing the work for some time, and that when he did eventually go to Mr. Beattie there was (on the case of both Edge and RSA) only a short meeting at which Mr. Beattie signed the July 2015 document. Mr. Mullen was seeking through these e-mails to provide false excuses for his delay. Obviously, this conduct is open to criticism. But it does not lead me to the conclusion that Mr. Mullen is a witness whose word generally cannot be trusted. When I consider the documentary evidence of Mr. Mullen's dealings as whole, there was nothing to suggest that he generally acted with a lack of integrity. In his oral evidence, he sought fairly to answer all the questions that he was asked. Unsurprisingly, he had difficulty in recollecting events many years ago, and there is no doubt that some of his answers were confused. All of this in my view leads back to the need to assess Mr. Mullen's evidence in the light of the objective evidence and the inherent probabilities.
  525. I now return to the question of what transpired at the July 2015 meeting. I consider that Mr. Mullen's belief that he had obtained Mr. Beattie's agreement to the terms of the endorsement is unsurprising, because in my view Mr. Beattie did so agree. The document which he signed, both in its appearance and language, is clearly intended to be a contractually binding document. It starts with the words:
  526. "Attaching to and forming part of Policy No. QM349770L
    The following amendments are hereby noted by Underwriters hereon"
    The document then goes on expressly to use the language of agreement in relation to many of the clauses. For example the very first clause states that the "definition of the insured is agreed as follows". Clause 2 of the TPC begins: "Underwriters note and agree that, in respect of any Transaction, it is hereby confirmed that the Insured is covered under this contract …" The wording of the TPC is introduced with language which is clearly contractual:
    "The contract change made under Endorsement 4 is deleted and replaced with the following clause:" (underlining in original)
  527. Whilst it is true that the document is not headed "endorsement", and was not on the standard form, the language used by the parties indicates their intention that it was indeed an endorsement. Clause 1 of the TPC thus stated:
  528. "Unless otherwise defined in this contract, capitalised terms in this Endorsement 4 shall have the meaning ascribed to them in paragraph 3 below". (underlining supplied)
    This language, when taken together with the introductory words concerning the deletion of the existing Endorsement 4, shows clearly that the parties were contemplating the replacement of one endorsement by another.
  529. Mr. Beattie signed the document on every page, adding the RSA's stamp. This would ordinarily be taken to indicate that he was agreeing to the terms set out in the document. The signature was in ink, not pencil. There was no qualification to the signature, eg "for receipt only". If Mr. Beattie disagreed with the terms, and needed further information, I see no reason why he would have signed the document in this manner and then put a copy of the signed document on the RSA file.
  530. The 20 January 2016 quotation slip (and indeed the slip signed on 25 January 2016) is also relevant in this regard. Mr. Beattie's evidence is that he was unhappy with the terms of the July document, and asked Mr. Mullen for clarification and information. This was, on his case, the reason that he only signed the document as a record of his receipt. However, Mr. Beattie's evidence (as summarised above) indicated that he was no further forward in January, in terms of understanding the TPC, than he had been in July 2015. Yet there is no dispute that he signed the quotation slip on 20 January 2016 to indicate his agreement in principle to the terms proposed (ie not simply for "receipt"). There was no notation on the slip to indicate that there was any outstanding query in relation to the TPC, nor anything to that effect in the notes which were recorded by Mr. Beattie on the RSA file in relation to the renewal. When Mr. Beattie signed the quotation slip, he did so knowing that the quotation slip would then be taken round the market and shown to other underwriters on the basis that the slip had Mr. Beattie's support as leader. Mr. Beattie was, therefore, willing to agree those terms on 20 January, and he signed again on 25 and then 27 January. If, as Mr. Beattie's evidence suggests, he was unhappy with the proposed terms in July 2015, he would have been equally unhappy on 20, 25 and 27 January. My conclusion is that he was happy with the proposed terms on all occasions.
  531. In Edge's written closing submissions, Ms. Healy put forward "at least" six reasons why Mr. Beattie's evidence – that he refused to agree the endorsement and sent Mr. Mullen away without agreement, to ask Icestar what it wanted to achieve – was improbable. In my view, each of these reasons is powerful and collectively they are overwhelming. I will therefore identify the points raised.
  532. First, Mr. Beattie's account is irreconcilable with the e-mail which Mr. Mullen sent immediately after the meeting, at least unless Mr. Mullen was dishonest in his report to his clients. I have addressed this point already, and rejected the allegation of dishonesty. The evidence shows that Mr. Mullen valued the Bank as a client, and I regard it as most improbable that Mr. Mullen would decide to tell the Bank that it had coverage on the amended terms – and then to send Edge's formal cover note – when he knew that Mr. Beattie had declined to agree the terms. This would be the height of folly and dishonesty.
  533. Secondly, if Mr. Beattie had refused to agree the endorsement, one would have expected him to be cross or at least perplexed when Mr. Mullen returned to him at renewal in January 2016 and showed him a renewal slip which contained all the changes from the July endorsement including the TPC. If this had happened, Mr. Beattie would have repeated or followed up the alleged questions as to what the client wanted to achieve and why, in circumstances where Mr. Mullen had not come back to him in the interim.
  534. Thirdly, it made little sense for Mr. Beattie to ask Mr. Mullen what the Bank wanted to achieve by the endorsement. It was obvious that the Bank wanted amended terms. Whilst specific questions directed at particular clauses would be understandable, a general question as to the Bank's overall intention behind the endorsement is strange.
  535. Fourth, I agree that there was no need for Mr. Beattie to sign, stamp and date each page of the document in order to acknowledge receipt. Had this been the intention, it would and should have been clearly stated. The Commercial Court is familiar, particularly in the marine context, with documents which a person (such as a ship's master) signs for receipt only. The evidence of Mrs. Webb, the insurers' underwriting expert, was that if she was presented with a document which was simply a discussion document, she would sign it "for receipt only" or in pencil. It seemed to me that this would be the obvious thing to do if, as Mr. Beattie contended, he wished to go no further than acknowledging receipt of a document which was to be discussed subsequently. I did not consider that his explanation for signature was persuasive or convincing.
  536. Fifth, I agree that it is surprising that Mr. Beattie should put the document on RSA's underwriting file. If there had been no more than a preliminary discussion on 28 July 2015, the expectation would be that there would be a further discussion as and when Mr. Mullen received answers to the questions asked. Mr. Beattie would not need to put the document on the file for that purpose. Mr. Beattie saw Mr. Mullen very regularly, sometimes 4 or 5 times a week, and it would be expected that – if discussions were to resume – Mr. Mullen would bring a copy of the proposed wording with him. I was not convinced by the suggestion (Mr. Beattie said "I may be making an assumption here") that Mr. Beattie retained the document in case there was a telephone discussion, particularly since Mr. Mullen was seeing Mr. Beattie so regularly. The far more probable explanation is that the document was placed on the file because Mr. Beattie considered that it contained amended terms.
  537. Finally, Ms. Healy relied upon Mr. Mullen's firm denial that he had been sent away without agreement. In my view, for the reasons explained above, that denial is consistent with the documents and the inherent probabilities. I accept Mr. Mullen's evidence that there was agreement by Mr. Beattie to the terms proposed in the July document, and reject Mr. Beattie's evidence to the contrary.
  538. In reaching that conclusion, I do not consider that Mr. Beattie has deliberately sought to give untruthful evidence about the meeting. As frequently happens, he is a witness who has by now persuaded himself that something happened which did not; ie that the terms presented to him on 28 July 2015 were unacceptable and that he required further clarification and explanation before he could commit the RSA to what was proposed. There is in my view an obvious reason as to how this has happened, bearing in mind the observations of Leggatt J in Gestmin as to the nature of recollection. It is now very clear to Mr. Beattie, as a result of the issues ventilated in this litigation, that the July document could not easily be read and fully understood in the course of a short meeting between broker and underwriter. The TPC clause itself is detailed, and a number of witnesses commented – quite reasonably – that they needed to read it on a number of occasions in order fully to understand it. Furthermore, the TPC was only one of the clauses which were being added or amended. Amendments were being made to existing clauses. Sometimes these amendments were shown in the July document with the benefit of track changes, but in other cases the full amended clause was reproduced without the changes being tracked. The evidence indicated that, usually, Mr. Beattie is a careful and meticulous underwriter. It is also common ground that he did read through the TPC on 28 July, and I have no doubt that he did. I consider that he failed to appreciate the complexity of the document which he was asked to sign, but that he now recognises both the complexity and the potentially far-reaching nature of the terms proposed. Recognising, now, that this was a document which a prudent underwriter should not have signed without a more careful and thorough review, and questioning of the broker, he has persuaded himself that he acted as a prudent underwriter on that occasion and sent Mr. Mullen away in order to provide the explanations for which Mr. Beattie should have asked, but did not.
  539. It follows from this analysis that I do not consider that Mr. Beattie is a reliable witness in relation to a critical aspect of his evidence as to what transpired at the July 2015 meeting; ie whether his signature of the endorsement reflected his agreement to its terms. This does not, however, entirely dispose of the underwriters' case for rectification or estoppel or collateral contract. It would, at least in theory, be possible for me to reject that important aspect of his evidence, but nevertheless conclude that there was a discussion with Mr. Mullen during which Mr. Beattie expressed the view that the TPC was a "basis of valuation" clause applicable only in the event of physical loss or damage. If said, this might (depending upon the reaction of Mr. Mullen to the suggestion) provide the starting point for these arguments.
  540. However, I do not consider it realistic to reject Mr. Beattie as an unreliable witness on a critical aspect of his evidence at the meeting, but nevertheless to accept his evidence about the discussion of "basis of valuation" clause. In the light of my previous conclusions, I do not think that Mr. Beattie is a witness whose evidence as to what transpired at this meeting many years ago can be relied upon at all.
  541. There are, however, other reasons for reaching that conclusion as far as concerns the case that the parties were in agreement that the TPC was simply a "Basis of Valuation" clause. In addressing this issue, it is convenient to consider all three meetings in the round, since Mr. Beattie's evidence about the TPC discussion was in material respects the same in relation to each of the meetings.
  542. First, it seems to me to be inherently improbable that, as Mr. Beattie's evidence suggests, there would have been an inconclusive and unsatisfactory discussion about the TPC at the July 2015 meeting, but that Mr. Beattie would nevertheless have signed it on 28 July without any qualification or written suggestion indicating that further clarification was required. That conclusion is reinforced when one takes into account Mr. Beattie's evidence that a similar, inconclusive, discussion about the clause took place on three separate occasions: 28 July 2015, 20 January 2016 and 29 January 2016, with Mr. Mullen on each occasion being unable to confirm Mr. Beattie's understanding of the clause. Yet on each occasion Mr. Beattie nevertheless signed important contractual documents.
  543. Secondly, there is no contemporaneous documentation which reflects or records Mr. Beattie's understanding of the clause, or the nature of the discussion that took place. Mr. Beattie did make, in January 2016, notes on the RSA underwriting file. These were probably made on or shortly after 25 January, when Mr. Beattie entered his line, and possibly before the 29 January meeting. Had the discussion taken place in accordance with Mr. Beattie's evidence, one might have expected Mr. Beattie to record that discussion, or at least to put a question-mark against the TPC – particularly bearing in mind that there had, on Mr. Beattie's evidence, been at least two discussions with Mr. Mullen about this clause, with Mr. Mullen being unable to explain it.
  544. Thirdly, it is in my view important not to undervalue the importance of the evidence provided by the signed documentation itself. There are here a good number of signed documents, each containing the TPC, which were signed at different times. In the context of rectification, MacGillivray in the insurance context, and FHSC more generally, both refer to the presumption that the signed document is an accurate record of the parties' agreement. This is reinforced by the Code which emphasises the importance of contract certainty. In these circumstances, it is inherently improbable that the parties would have discussed the TPC on three separate occasions in the manner described by Mr. Beattie, without seeking to ensure that the issue was resolved by way of clarification to the existing wording. It would not have been difficult for Mr. Beattie to have required Mr. Mullen to add words at the start of the TPC which said: "This clause applies in case of physical loss and damage to the cargo". (If rectification were appropriate, then – notwithstanding Ms. Healy's point that the terms on which the policy is to be rectified have not been clearly identified – it seems to me to be relatively easy to do so).
  545. Fourth, it is not easy in my view to see why Mr. Beattie would have described the TPC as a "Basis of Valuation" clause, still less as a basis of valuation for physical loss and damage. The policy already contained a clause which was headed "Basis of Valuation", and it would seem strange that it was necessary to include another one. As discussed in Section D above, the TPC referred to the word "Default" in 3 places, including the definition of that clause, and this is not a concept that would ordinarily be relevant to a cargo policy.
  546. The suggestion that the TPC was still being discussed as a "Basis of Valuation" clause on 29 January 2016 is also inherently improbable in itself. That meeting took place once the TPC had (as Mr. Beattie was told) been moved to two separate places in the policy. It now had its own section, immediately after the coverage for CEND. It is in my view hard to see why a Basis of Valuation clause for physical loss and damage would require its own section, or to see how Mr. Beattie would still have felt comfortable in those circumstances that the TPC went no further than he was telling Mr. Mullen in undocumented verbal discussions.
  547. Fifth, Mr. Lockyer's question to Mr. Mullen on 28 January 2016 does not, in my view, fit at all easily with the suggestion that there had been a discussion at a meeting shortly beforehand, which Mr. Lockyer had attended, when Mr. Beattie had said that he understood the TPC to be a basis of valuation clause. Mr. Lockyer accepted in his evidence that he could not remember the discussion that took place on 20 January. But he made the point in his witness statement that it would have made no sense, if there had been a discussion on 20 January as suggested by Mr. Beattie, for Mr. Lockyer to ask whether it was underwriters' understanding that the TPC provided cover regardless of whether there had been any physical loss or damage to the goods. I think that this was a fair point.
  548. Sixth, I attach importance to the correspondence which took place once the potential claim had been advised to the RSA in August 2016. The sequence of events (described in Section B above) shows that RSA, via their senior claims adjuster Mr. Jones (who later signed the notice of declinature), was aware in October 2016 that the potential claim arose from Euromar's defaulting on their obligations, and that no physical loss or damage was involved. Mr. North's e-mail to Mr. Jones on 14 December 2016 made it clear that the claim was for financial loss as a result of Euromar's default, and that Mr. Jones had agreed to review coverage in respect of that loss. At 11.37 on 15 December, responding to Mr. North's email, Mr. Jones said that coverage would appear to attach in respect of "this situation".
  549. As I have already said (see Section B), it is clear that Mr. Beattie had been consulted about the claim by that time. His e-mail sent on 15 December 2016 to Mr. Mullen asked for a chat about ABN Amro, because "it appears that we may have a problem over a recent loss which is effectively a Financial Guarantee not linked to any loss or damage recoverable under ICC "A"". This email was sent at 08.35, and he sent a further e-mail at 08.50 in which he said: "If this is a financial guarantee, it is a no-no within RSA and I will be instructed to change the wording or manage an exit". There was, here, no reference to the discussions which are alleged to have taken place on three occasions and in which, as the insurers now contend, Mr. Beattie and Mr. Mullen were essentially ad idem that the TPC was a basis of valuation clause applicable only to physical loss or damage.
  550. Mr. Beattie's evidence, when asked about these two e-mails, was that his "initial reaction to Mr. Mullen was: no coverage here, not cargo underwriter's concern." There is, however, nothing in the documents which suggests that this was Mr. Beattie's initial reaction, or that he so expressed himself to Mr. Mullen. After receiving Mr. Beattie's emails on 15 December, Mr. Mullen then e-mailed Mr. Beattie saying that the incident came under the transaction premium section of the policy, and explaining the steps that the Bank was taking to deal with the Euromar problem. Later that morning, Mr. Jones confirmed that cover would appear to attach "in respect of this situation".
  551. Mr. Beattie suggested in evidence that he was not consulted by Mr. Jones in relation to the sending of the e-mail on 15 December which confirmed coverage. Mr. Jones gave evidence to a similar effect. I do not accept this. In my view, it is no coincidence that Mr. Beattie was sending two emails to Mr. Mullen in relation to the potential claim at the same time as Mr. Jones was being asked to confirm coverage. Nor do I accept (if this is what Mr. Beattie was saying) that Mr. Beattie told Mr. Mullen at this stage that he did not consider that there was any coverage. The documents contain no suggestion that RSA via Mr. Jones was saying that there was coverage, but that Mr. Beattie was telling Mr. Mullen that this matter was not cargo underwriters' concern.
  552. Mr. Mullen then described in his evidence a meeting with Mr. Beattie in late January 2017 when the policy came up for renewal. Mr. Beattie told Mr. Mullen that he had been told "by a higher authority" that the TPC would have to be removed from the policy on renewal, because it provided Icestar with financial guarantee cover. Again, there appears to have been no reference by Mr. Beattie to the discussions at the three meetings when the parties had, on RSA's case, effectively agreed that the TPC was only applicable to physical loss and damage.
  553. In April 2017, Mr. Beattie e-mailed Mr. Mullen as follows:
  554. "David,
    I know you are travelling, but assume you are picking up your emails. I am told today that the ABN Amro "Financial loss" on the aborted cocoa sale in Europe could be as much as $30m. This, combined with the fraudulent documentation loss on ANZ and the well documented misappropriation loss on your ex-client account will probably bring to an end our foray into bank traders' accounts. Indeed, it may be better if I shut them all down prior to my final departure rather than leave the task to others.
    This is just a heads up for discussion and we can discuss further next week.
    Regards,
    Brian."
  555. The email contains no reference to the discussions now relied upon, nor any suggestion that these discussions meant that there was no coverage for the claim.
  556. Furthermore, there was no reference to the discussions now relied upon in the notice of declinature which was sent by RSA in August 2017. This notice was sent after there had been a detailed investigation into the claim by investigators, Gray Page, and after Kennedys Law LLP (who act for all insurers in these proceedings) had been engaged.
  557. The allegation of oral discussions was first made in the underwriters' defence served in February 2019. I consider that if there were any substance to the claim that there were relevant oral discussions, the point would have surfaced long before then. In my view, these discussions are another aspect of Mr. Beattie's retrospective rationalisation of his decision-making. He has searched for an explanation as to how he came to write a policy with the TPC (because there is no case that the 2016 policy was not contractually binding), and has persuaded himself that he did so on the basis of the oral discussions which he now asserts. His suggestion that the discussions concerned the basis of valuation is in my view derived from the arguments that have developed as to how the policy is to be interpreted. The culmination of this retrospective rationalisation was an answer which Mr. Beattie gave in response to questions from Ms. Sabben-Clare, when asked what Mr. Mullen had said in response on 29 January 2016:
  558. "Again, I received no real response. Never once when I said this is my interpretation of the clause ie an add-on the basis of valuation to safeguard the profit of Icestar on a per contract basis, never once did Mr. Mullen contradict me and say, no, you have this wrong.
    In none of his statements had Mr. Beattie explained exactly how the TPC operated as a basis of valuation clause. Here, as far as I can see for the first time, he gave an explanation, and suggested that the precise impact of the clause had been discussed.
  559. For these reasons, I do not consider that Mr. Beattie's evidence either as to his thought processes concerning the TPC, or the discussions which are alleged to have taken place, can be relied upon. The evidential foundation for the case of rectification, collateral contract and estoppel therefore does not exist. I consider that the proper conclusion is, quite simply, that Mr. Beattie on behalf of RSA and Mr. Mullen on behalf of the Bank intended to contract on the terms set out in the various documents which Mr. Beattie signed. I cannot conclude, on the balance of probabilities, that there was any side-discussion which is capable of supplementing or negating or qualifying those terms. For the purposes of the rectification case, there is nothing to rebut the presumption that the signed documents represented the parties' agreement, and there is certainly nothing which amounts to convincing proof to the contrary. Nor was there any collateral contract which formed the basis on which Mr. Beattie agreed to the July 2015 endorsement or the 2016 renewal.
  560. Nor do I accept that there was any tacit understanding between Mr. Beattie and Mr. Mullen that the TPC was only applicable to physical loss or damage or that it was a basis of valuation clause in that context. In view of my findings above, the case for a tacit understanding seems to depend, impermissibly, upon uncommunicated intentions which happen, without the parties knowing it, to coincide. There is nothing in the communications between the parties upon which this tacit understanding could be based.
  561. In any event, I am not persuaded that Mr. Beattie formed any clear view as to what the TPC actually meant. In my view, he did not take sufficient time to consider it when he first agreed it in July 2015, and did not do so thereafter.
  562. As far as Mr. Mullen's subjective understanding of the TPC is concerned, I accept that, particularly in the light of some of Mr. Mullen's woolly or muddled evidence at trial, there is room for doubt as to whether he did in fact understand the TPC at the time. However, there is contemporaneous documentary evidence that Mr. Lockyer did understand it correctly and in particular that it was not confined to the situation where there was physical loss and damage: see Mr. Lockyer's 8 July 2015 aide-memoire and, later, his e-mail to Mr. Mullen on 28 January 2016 after Ms. Van de Beek had raised her question. Since Mr. Lockyer worked closely with Mr. Mullen, it is more likely than not that Mr. Mullen would have shared Mr. Lockyer's (correct) understanding.
  563. It is also important to note Mr. Mullen's response to Mr. Lockyer's question on 28 January – or strictly speaking two questions – "Am I reading this correctly, and is this understanding of underwriters". His response ("No you are correct") was to give Mr. Lockyer assurance on, as I read the e-mail, both points. He was thereby indicating that Mr. Lockyer was reading the clause correctly, and also that it was the understanding of underwriters. As further discussed in Section J below, I am not satisfied that Mr. Mullen had any firm foundation for the statement that it was the understanding of underwriters, since he had never specifically discussed the clause with any of them. But the e-mail does show that, contemporaneously, Mr. Mullen had the same understanding of the clause as Mr. Lockyer: ie that the "insured is covered by this policy for the transaction premium they would have earned if client of the insured defaults, regardless whether there has been any physical loss or damage[d] to goods".
  564. Furthermore, Mr. Mullen had on 9 July 2015 asked a direct question to Ms. Barnes about the clause, and had received a clear answer, including that: "we are expressly stating that market premium loss is not contingent on physical loss or damage to the commodity". Mr. Mullen's e-mail of 9 July 2015 asks pertinent questions, and is the e-mail of a person who wishes to understand the cover that he is being asked to place. Having received a clear answer to a pertinent question, it is in my view more likely than not that Mr. Mullen did understand what he was being told by Ms. Barnes, and indeed by Mr. Lockyer, as to the intended effect of the clause.
  565. Accordingly, I was not persuaded that Mr. Mullen failed to understand the TPC, either in July 2015 or January 2016, bearing in mind that the effect of the TPC was spelt out in writing on both occasions (by Ms. Barnes in July, and Mr. Lockyer in January). I thought that the evidence of the contemporaneous documents where the meaning of the TPC was specifically addressed was more powerful than any conclusion that could be drawn from Mr. Mullen's e-mail to the Bank sent later on 28 January (at 5.26pm) upon which the underwriters relied. In that e-mail, Mr. Mullen said:
  566. "We have deliberately sited this to the Basis of valuation section because this clause will determine the amount of recovery the Bank can obtain from the contract Underwriters"
    This sentence was inaccurate: the movement of the TPC had been made by the Bank, not Edge. The reference to a "Basis of valuation" section is also strange, since the policy contained no such section. But in any event, Mr. Mullen is not suggesting here that the TPC is only applicable where there is physical loss or damage. His statement that the clause will determine the amount of recovery the Bank can obtain from underwriters is quite general.
  567. In reaching my conclusions as to what transpired at the meetings, I do not consider that I can place any reliance on the passages in Mr. Mullen's carefully written witness statements where he gives a firm and coherent explanation as to how he would have reacted if Mr. Beattie had started discussing the TPC as a basis of valuation clause. Having heard Mr. Mullen over some days, I cannot conclude that Mr. Mullen would have articulated his response along the lines indicated in his statements. However, I do think that it is realistic, and consistent with the inherent probabilities, to conclude that if Mr. Beattie had started raising questions about the TPC at the July meeting, and that these indicated Mr. Beattie's unhappiness with the clause, Mr. Mullen would have reverted to the Bank. I therefore accept, on the balance of probabilities, Mr. Mullen's evidence that this discussion did not happen at the July meeting. I also think, as I have said, that if there had been a discussion on 20 January 2016, at the meeting which Mr. Lockyer attended, about the TPC being a basis of valuation clause applicable to physical loss or damage, then Mr. Lockyer's e-mail to Mr. Mullen on 28 January would likely have been a rather different one; since, on that premise, Ms. Van de Beek's comment box would have indicated that the Bank and Mr. Beattie were thinking differently. So I accept Mr. Mullen's evidence that if there had been a discussion along those lines, either he or Mr. Lockyer would have remembered it, at least when the issue arose on 28 January following Ms. Van de Beek's question.
  568. As far as the 29 January meeting is concerned, Mr. Mullen had no recollection of the discussion. Since he had no recollection of the discussion, and since I do not consider that I can place reliance on Mr. Beattie's evidence as to what transpired at the meetings, I do not consider that Mr. Mullen's apparent acceptance in cross-examination – that Mr. Beattie may well have said something about the TPC being a basis of valuation clause – provides a sufficient basis to conclude that this was in fact said. In any event, looking at the evidence as a whole, the contemporaneous documents and the inherent probabilities do not support the proposition that such a discussion took place at any stage.
  569. In the light of these conclusions, there is in my view no basis for an estoppel by convention. There was no relevant assumption of fact or law, either shared by both parties or made by Mr. Beattie and acquiesced in by Mr. Mullen. Nor, if it existed at all, was any common assumption expressly shared between them.
  570. Even if I had accepted Mr. Beattie's evidence as to what transpired at the meetings, I do not think that it would have supported a case of estoppel by convention (let alone rectification or collateral contract). Mr. Beattie said (in relation to the 20 January meeting) that he let matters lie with a "non-committal" answer. The effect of his evidence (if accepted) in relation to all of the meetings was that Mr. Mullen at no stage agreed that Mr. Beattie's understanding was correct: the point was raised, but without any clear or satisfactory resolution. That is not a satisfactory basis for a case based on a shared assumption: as Picken J said in Aras, there needs to be clarity over what comprises the common assumption. Nor do I think that this evidence, if accepted, would provide a satisfactory basis for a conclusion that it would be unjust to allow the Bank to rely upon the contractual terms which were in fact agreed. If Mr. Beattie took no steps to require the issue to be clarified and satisfactorily resolved, I see no injustice in saying that the contractual terms should apply.
  571. In these circumstances, it is not necessary to consider the issue of whether any of the following underwriters could rely upon these conversations between Mr. Mullen and Mr. Beattie, of which they knew nothing, in support of a case of rectification or collateral contract or estoppel in connection with each of their individual subscriptions. It suffices to say that there appeared to be formidable difficulties which confronted that argument.
  572. In relation to four of the underwriters, Mr. Parsons put forward a separate estoppel argument. This related to representations made to three underwriters (Navigators, Ark and Advent), on renewal in 2016, that the terms of the cover were "as before" or "as expiry". In relation to a fourth underwriter (Standard) reliance was placed upon a representation that the policy provided cover for loss and damage to products where finance was provided. These representations were primarily relied upon as misrepresentations which entitled those underwriters to avoid the 2016 policy. However, Mr. Parsons in his oral closing advanced an argument that these representations could found an estoppel which prevented the Bank from relying upon the TPC and also the Non Avoidance Clause. Since this argument overlaps with the arguments concerning misrepresentation in relation to these four defendants, I will address it in Section F below.
  573. Was the following market bound by the July 2015 endorsement?

  574. It is convenient at this point to address the separate question as to whether the following market was bound by the July 2015 endorsement. In view of the fact that no claim arose under the 2015/16 policy, and that the TPC was clearly included in the 2016/17 policy to which all underwriters subscribed, this question is not central to the resolution of the principal issues in the case. It does have a potential impact on the misrepresentation case advanced in particular by Navigators, Advent and Ark, but even there (as discussed in Section F below) it is not critical. It is appropriate, however, to state my conclusions on that issue.
  575. I do not accept the underwriters' argument that Mr. Beattie was not permitted, by GUA, to agree to the TPC on behalf of the following market because it involved an increase in the underwriters' monetary exposure: see paragraph 312 above.
  576. However, the underwriters put forward a separate argument as to why the July 2015 endorsement was not binding on the following market. They submitted, correctly in my view, that it did not follow from the fact that authority may have existed under GUA that it was in fact exercised on any particular occasion. Relying on Bowstead and Reynolds on Agency 21st edition, paragraph 8 – 172, they submitted that in order to determine whether an agent with authority is exercising that authority, it is necessary to consider the agent's intention. Here, they submitted that there was nothing to indicate that, if Mr. Beattie intended to agree to the terms of the July endorsement, he was exercising authority on behalf of anyone else. There is no GUA Stamp, and therefore no initial in the "Slip Leader only" box to signify an exercise of GUA authority on behalf of the followers.
  577. Edge argued in their closing submissions that the subscription agreement section of the slip empowered RSA to agree "all" contract changes on behalf of the following market. They said that over the history of the Bank's insurance, RSA had acted pursuant to that power to agree numerous changes to the cover by way of endorsement, including changes which amended the scope of cover; eg by clarifying the cover afforded under the Business Contingency Clause, by adding 5 new warehouses as approved storage locations, or covering the transit of materials between warehouses. There was only one occasion on which RSA required the agreement of all underwriters to be obtained. This was when increasing the limit for metals from US$250 million to US$300 million. Responsibility for applying the GUA stamp to an endorsement rested with RSA as lead underwriter rather than with Edge as broker. In these circumstances, it was incumbent upon Mr Beattie to make it clear if he was purporting to agree a contract change on behalf of RSA alone, rather than on behalf of all underwriters. As he did not do so, he purported to exercise authority to bind the following market.
  578. I do not accept Edge's argument. In my view, there was nothing in the events which took place at the short meeting in Mr. Beattie's office on 28 July 2015 which evidenced any intention on Mr. Beattie's part to exercise authority on behalf of the following market. Where an endorsement contains the GUA box, and where that box is then initialled by the underwriter, it will be apparent that the underwriter is intending to implement the GUA terms and hence the authority which has been conferred upon him. The way in which the underwriter is exercising that authority will be clear from the way that he fills in the GUA box. Clause 4.2 of the GUA makes this clear. It provides that the "Slip Leader (and Agreement Parties if appropriate) shall then initial in the appropriate Box the level of authorisation required". Clause 4.2.3 provides that if the Slip Leader initials Box 1 and initials and dates the endorsement in the customary place, no further agreement shall be required.
  579. The problem in the present case is that Mr. Mullen did not include the GUA Box on the July endorsement. Mr. Mullen did not suggest in his evidence that there was any discussion as to whether Mr. Beattie was being asked to agree the terms, or was in fact agreeing them, on behalf of the following market. Mr. Beattie did no more than sign and apply RSA's stamp, which was what he was being asked to do. There is nothing to which Edge (who carried the argument on the GUA issues) could point which indicated, whether objectively or subjectively, an intention on Mr. Beattie's part to sign the July endorsement on behalf of the following market.
  580. Edge's argument is not assisted by pointing to other endorsements which did contain the appropriate GUA stamp. In those cases where RSA filled in Box 1, there is no difficulty in saying that RSA was agreeing to the amendment on behalf of the followers. That is not what happened here.
  581. Nor in my view is the argument assisted by saying that it is the Slip Leader's responsibility to incorporate the GUA Stamp. Clause 4.1 of the GUA does indeed provide that he should do so, should the GUA Stamp not already "be incorporated in or appear on the form of endorsement". The significant point is that this was not done, and that there was therefore nothing which indicated an intention on the part of Mr. Beattie to contract on behalf of the following market. That intention cannot in my view be found by relying on the obligation on the Slip Leader in Clause 4.1 of the GUA in circumstances where there had been no discussion between broker and underwriter as to whether RSA was acting only on its own behalf or on behalf of the following market.
  582. Accordingly, I do not consider that the following market was in fact bound by the July 2015 endorsement.
  583. F: Non-disclosure and misrepresentation

    F1: Introduction to the issues


     

  584. The underwriters' case of non-disclosure and misrepresentation was introduced belatedly into the case. The proceedings were issued in November 2018, and the underwriters' original defence was served on 22 February 2019. No point on avoidance was then taken, although there had been a generally worded reservation of the right to take such a point in the pre-action correspondence. The avoidance case was first intimated in April 2020 with the amendments taking effect in May 2020. The Bank and Edge did not oppose the amendment, preferring to avoid a contested permission application, but making it clear (at least in Edge's case) that the case was bad in law and on the facts. The parties' arguments focused on three issues.
  585. First, the Bank and Edge relied upon the "Non-Avoidance Clause" or NAC which prevented avoidance other than for fraudulent non-disclosures or fraudulent misrepresentations. They said that the NAC provided a complete answer to the underwriters' avoidance case, since fraud was not alleged.
  586. Secondly, the Bank and Edge submitted that even if there were any substance to an avoidance case based on non-fraudulent misrepresentation or non-disclosure, affirmation provided a second complete answer. The essence of the argument was that the insurers had not relied upon, or sought to preserve, any avoidance argument at the time that they served their original defence and counterclaim, in which arguments as to the construction of the relevant policy and rectification were advanced. The insurers had also, at that stage, not tendered back the premium. The service of that defence was a positive affirmatory act, and the insurers had the requisite knowledge of the facts allegedly not disclosed or misrepresented and of their right to avoid.
  587. Thirdly, there were arguments as to whether, even if fraud was not required, there was any substance to the claim that there were material non-disclosures or misrepresentations.
  588. The misrepresentations and non-disclosures relied upon can be summarised as follows. Some of these pleas affected only some of the underwriters. The first two pleas were relied upon by all of them.
  589. (1) Purpose/intention of the TPC: All underwriters allege a failure by the Bank to disclose its purpose or intention in requesting the inclusion of the TPC. The pleaded case was that:
  590. "if, as the Bank contends, the Transaction Premium clause was intended to and/or did give rise to a separate head of cover in respect of credit risks and/or financial defaults, then … that was a material fact of which a reasonable and prudent underwriter would have desired to be informed before deciding whether, and if so on what terms, to agree [both the 2015 endorsement, if binding, and the 2016 policy]
  591. There was an element of conditionality in relation to this plea. Thus, it arose only on one or both of two premises: (i) that there was an intention on the part of the Bank for the TPC to give rise to a separate head of cover for credit/default risks or (ii) that the TPC did so provide. The underwriters did not accept that this had been the intention of the Bank. Nor did they agree that this was the effect of the TPC. However, in the event that both (or perhaps either) of these arguments were rejected, then both the RSA and the following underwriters were entitled to avoid the policy for non-disclosure of the fact that the TPC was intended to and/or did give rise to a separate head of cover in respect of credit risks and/or financial defaults. The underwriters' argument in relation to this non-disclosure possibly encompassed an argument that the presence of the TPC should itself have been disclosed, although there was no clear plea to this effect in the defence.
  592. Accordingly, the underwriters' basic case was that the purpose or intention behind the TPC – such purpose or intention being a matter of fact – was a material circumstance which should have been disclosed.
  593. (2) The non-avoidance clause. Secondly, all underwriters alleged that there was "non-disclosure of the presence of the non-avoidance clause". The underwriters' pleading on this issue is not particularly easy to follow, because it is only RSA that appears to assert that this non-disclosure gave rise to a right to avoid. However, I ultimately understood all underwriters to rely upon the non-disclosure of the presence of the clause with two consequences: it would prevent reliance on the NAC itself, and would also render the contract void.
  594. The underwriters submitted that the presence of a clause on the terms of the NAC wording proposed for renewal would be a matter of interest for any prudent underwriter. The clause would be material to the risk being written, since it restricted the nature and extent of the insured's disclosure about the risk.
  595. (3) The 2015 endorsement. The following underwriters relied upon non-disclosure of the 2015 endorsement. There was again an element of conditionality in relation to this plea: it only arose in the event that the 2015 document was contractually binding either on RSA or by way of an endorsement. In that event, however, its existence was material to the following market when subscribing the 2016 policy, in particular because of the presence of the TPC and the NAC. Even here, there was further conditionality, since (consistent with the underwriters' case) the TPC was only material if, as the Bank alleged, it was intended to or did give rise to a separate head of cover for credit risks or financial defaults.
  596. The underwriters submitted that if there had been prior agreement on the terms of the 2015 document, this would have been of great interest to an underwriter on renewal, because it would mean that the expiring terms being renewed included both credit risk cover and also the NAC.
  597. (4) "As expiring" misrepresentation. Four of the underwriters (RSA, Navigators, Ark and Advent) allege that they were told in January 2016 upon renewal that the 2016 policy was the same as the policy for the expiring year. Thus, RSA's case was that Mr. Mullen told Mr. Beattie that there was no material change to the policy terms he was presenting. Navigators allege that Mr. Mullen told the underwriter that, apart from an increase in the limit in relation to oil trading, all other terms were as before. Ark contended that Mr. Mullen said that, apart from the increase in the limit in relation to oil trading and an increase in brokerage, all terms and conditions were as expiry. A similar case was advanced by Advent. These statements were also relied upon as the foundation for an estoppel.
  598. (5) Only PLOD misrepresentation. The 12th Defendant, Standard – which had not written the policy in 2015 – alleged that it was told that the policy covered the Bank for loss and/or damage to products where finance was provided. This statement was also relied upon as the foundation for an estoppel.
  599. A feature of the three alleged non-disclosures is that each arose from, or was directly related to, the wording that was actually presented to the underwriters and which each of them scratched. The non-disclosure case therefore naturally gave rise to arguments concerning the extent to which underwriters, who are subscribing to major commercial risks, could be expected to read the wordings to which they agreed. Edge's core argument was that it was the underwriter's responsibility to read and understand the slip he is presented for agreement. If there was something that he does not understand, he may either refuse to agree it or ask the broker questions, which the broker must answer honestly. There was, therefore, no disclosure obligation in relation to the Bank's subjective purpose or intention behind seeking any contractual provision, including the TPC. If the underwriters were interested in knowing the Bank's view as to what the TPC meant, or the reason for requesting the clause, then the underwriters had the clause in front of them and could and should have asked.
  600. F2: The Non-Avoidance clause

  601. The slip policy signed by all the underwriters contained a Non-Avoidance Clause or "NAC". The clause was Clause 22 of Section 2 in the earlier policy subscribed between 25 and 29 January 2016. (The underwriters referred to this earlier policy as the "unamended" policy.) In the later policy (or the "amended" policy, in the underwriters' terminology), it was at Clause 21 of Section 2: the numbering of the clauses in Section 2 had been adjusted because, in the earlier policy, there was no Clause 10. Nothing turns on this change in numbering.
  602. The "NAC", as it was referred to at the hearing, was in the following terms:
  603. "Non-Avoidance
    The Underwriters will not:
    a) Seek to avoid or repudiate this contract for non-disclosure or misrepresentation other than fraudulent non-disclosure or fraudulent misrepresentation; or
    b) Rely on, or assert any breach of warranty as grounds for the Underwriters to be discharged from any liability other than where the warranty was given fraudulently; or
    c) Seek damages for or seek to reject a claim for loss on the grounds of:
    i. Non-disclosure or misrepresentation other than fraudulent non-disclosure or fraudulent misrepresentation; or
    ii. Any breach of warranty other than where the warranty was given fraudulently."

    The parties' arguments

  604. The underwriters did not seek to advance any case that either the Bank or Edge on their behalf had made a fraudulent non-disclosure or fraudulent misrepresentation. It is not necessary to examine the reasons for this. It is sufficient to state that the underwriters' pleadings did not clearly allege fraud, and Mr. Parsons in opening confirmed that no such case was being advanced. The possibility of an amendment to plead fraud was, as described in Section B, floated during oral closing arguments, following the late disclosure by Edge of Mr. Lockyer's 8 July aide memoire. In the end, however, no application to amend to plead fraud was made.
  605. In those circumstances, the Bank and Edge argued that the NAC provided a complete answer to the case of non-fraudulent non-disclosure/ misrepresentation that was advanced. Reliance was placed on various authorities where similar clauses had been given effect, including: Toomey v Eagle Star (No. 2) [1995] 2 Lloyd's Rep 88; HIH Casualty and General Insurance Ltd. v New Hampshire Insurance Co and others [2001] EWCA Civ 735; HIH Casualty and General Insurance Ltd. v Chase Manhattan Bank [2003] UKHL 6; Mutual Energy Ltd. v Starr Underwriting Agents Ltd. [2016] EWHC 590 (TCC) (Coulson J). Whilst non-avoidance clauses were not standard clauses in marine cargo policies, they were commonplace in project cargo policies and in any event there was no reason why they should not be given effect as stated in Arnould: Law of Marine Insurance and Average (19th Edition) paragraphs 15–110-111.
  606. They submitted that the evidence of underwriters, to the effect that they had not read or understood the clause, was beside the point. They had signed slip policies containing the clause, and it was therefore irrelevant that they did not notice the clause or reflect upon its meaning. Their signature of a contractual document meant that they must be taken to have read, understood and agreed to it: see eg L'Estrange v Graucob [1934] 2 KB 394 (CA); Springwell Navigation v JP Morgan Chase [2010] EWCA Civ 1221; Higgins & Co Lawyers Ltd. v Evans [2019] EWHC 2809 (QB) (Saini J).
  607. Ms. Healy submitted that it was not possible, as underwriters argued, to draw a line through the NAC but not the rest of the policy on the basis that it is to be treated differently from the other terms: there was no basis for treating the NAC separately from the other terms. It was not a separable provision akin to an arbitration or jurisdiction clause. In that regard, she cited the decision of the Court of Appeal in HIH v New Hampshire and the judgment of Lord Hobhouse in HIH v Chase Manhattan at page 81.
  608. Underwriters submitted that the Bank could not rely upon the NAC. In their opening submissions, they drew attention to the fact that the TPC and the NAC were introduced into the policy together. Thus, the Bank had sought to introduce cover for credit risks by means of the TPC, but without disclosing that fact to underwriters, and without giving any disclosure in relation to the credit risk that was being introduced. At the same time, the Bank/Edge sought to add the NAC, which was an unusually restrictive clause, to eliminate any non-disclosure or misrepresentation type defence, absent proof of fraud. The Bank could not rely on the NAC as an obstacle to avoidance, because it is itself "vulnerable to being avoided". It is the sort of unusual clause that must be identified in the broke. It was not expressly mentioned to Mr. Beattie or any of the following underwriters. Some of those underwriters were told that the renewal was on expiring terms, and this was a misrepresentation. Mr. Parsons explained that the way round the NAC was that the NAC was itself not disclosed, and was part of the package with the TPC. The TPC and NAC came in together, and neither of them were brought to the underwriters' attention. Since there was a non-disclosure in relation to the NAC, that clause "falls away and this allows me to have the non-disclosure in relation to the TPC".
  609. In his closing submissions, Mr. Parsons relied on the absence of any evidence that the NAC in the 2016 renewal wording had been pointed out to Mr. Beattie or any of the followers. This was not a clause that they would have been on the look-out for or expecting to see, given that it was not a clause in use in the cargo market except for project cargoes.
  610. Accordingly, there was a non-disclosure of the NAC itself. Unlike the court in Springwell, the court is here dealing with an insurance contract where the duty of utmost good faith applies. The ordinary rule that a party who signs a contract is taken to have read it, understood it and agreed to it cannot apply where, as was the case here, there was a duty to disclose the NAC. If the non-disclosure or misrepresentation relates to the clause itself, a party cannot rely on his own breach of duty. Such non-disclosure would have two consequences; it would prevent reliance on the non-disclosure clause itself, and would itself render the contract as a whole void. Mr. Parsons accepted in closing that both consequences would follow, because it was difficult to see how the NAC could be treated as severable from the contract as a whole. Mr. Parsons said that Toomey v Eagle Star was distinguishable, because there was no allegation in that case of non-disclosure or misrepresentation directed at the clause itself.
  611. Discussion

  612. I accept the submissions of the Bank and Edge on this issue. The starting point is the principle that a person who signs a document knowing that it is intended to have legal effect is generally bound by its terms, whether he has actually read them or not. As Moore-Bick LJ said in one of the cases cited in Higgins, this is an important principle of English law which underpins the whole of commercial life: any erosion of it would have serious repercussions far beyond the business community.
  613. In the present case, the insurers have expressly agreed, in writing, that they will not seek to avoid or repudiate this contract for non-disclosure or misrepresentation other than fraudulent non-disclosure or fraudulent misrepresentation. It is clear from various decisions cited to me (for example Toomey, HIH v New Hampshire, HIH v Chase Manhattan and Mutual Energy v Starr) that such clauses are effective in accordance with their terms.
  614. For example, in HIH v New Hampshire, the court was concerned with a non-avoidance clause (clause 8) which was agreed in an underlying insurance, and where there was an issue as to whether it was effectively incorporated in a reinsurance. One issue considered by the Court of Appeal, at paragraphs [126] – [138] of its judgment, was how the clause was to be construed in its original setting, and in particular whether it excluded negligent misrepresentation and non-disclosure. The court's judgment was based principally upon an analysis of the language of the clause, whose wording was very wide. The decision of the House of Lords in the later film-financing case, HIH v Chase Manhattan, was also grounded in the true construction of the relevant clause. The relevant principle is stated in Chitty on Contracts, Volume 1, paragraph 7-145 as follows:
  615. "At common law a person could not contract out of liability for fraud inducing the making of a contract with him, at least where the fraud was his own. It is, however, possible that he could do so where the fraud was that of his employees or agents and there seems no doubt that it was possible, by a provision of the contract itself, to exclude or modify the normal consequences of innocent or negligent misrepresentation."
  616. Similarly, the decision in Springwell shows that appropriately worded "no reliance" clauses will give rise to a contractual estoppel. Whether or not this is so will depend upon the language of the clause.
  617. Accordingly, given that the relevant clause was agreed by underwriters in a signed document, the only relevant question in my view is whether the clause in the present case extends, as a matter of construction, to all non-fraudulent (ie "innocent") misrepresentations or non-disclosures. The underwriters' argument posits, in effect, that there is a residual category of innocent non-disclosures or misrepresentations which can be relied upon in order to avoid the contract as a whole; ie an innocent non-disclosure affecting the NAC itself, or (in the argument originally advanced) an innocent non-disclosure affecting the NAC when taken together with the innocent non-disclosure or misrepresentation concerning the TPC. I do not accept this approach. The NAC is comprehensive. Its effect is that any non-disclosure or misrepresentation relied upon in support of avoidance must be alleged and shown to be fraudulent. That includes an allegation of non-disclosure relating to the NAC itself, whether taken on its own or in combination with the TPC.
  618. The insurers' argument that it is permissible to isolate innocent non-disclosures or misrepresentations affecting the NAC itself is in my view difficult to reconcile with the authorities in relation to such clauses.
  619. In Toomey v Eagle Star, the insurers sought to nullify the effect of (what Colman J held to be) a non-avoidance clause by relying on a misrepresentation or non-disclosure which induced the contract in which the clause was contained. They argued that if there were grounds for avoidance, the clause must go with the avoidance. If a party sought to rely upon the clause, he relies in effect on his own misrepresentation for the purposes of avoiding the impact of that misrepresentation. The insured was therefore pulling itself up by its own bootstraps; because it should not be open to a party to rely upon the clause since the very contract which includes it has gone or is avoidable by virtue of the misrepresentation. Colman J rejected this argument on the basis that it was inconsistent with prior authority. This included a (then) recent decision of the Court of Appeal (Pan Atlantic Insurance Ltd. v Pine Top Insurance Co [1993] 1 Lloyd's Rep 496) in which Steyn LJ had said that it was conceptually possible to
  620. "draft a clause which excludes the other party's right to rescind for non-disclosure, except in the case of fraud, even though the clause excluding rescission forms part of a contract which upon rescission would be rendered retrospectively null and void".
  621. Toomey v Eagle Star is an authority cited by Chitty in the context of the passage quoted above. This aspect of Colman J's decision has not been doubted in subsequent case-law. Indeed, the unsuccessful argument in Toomey was not repeated in the film finance cases, where the courts gave effect to non-avoidance clauses in the relevant contracts.
  622. In the light of these authorities, the important points are as follows.
  623. First, the cases proceed on the basis that what matters is the true construction of the relevant clauses.
  624. Secondly, it is difficult to see why, if there is a suitably worded NAC, (i) innocent non-disclosures and misrepresentations of material facts which induced the contract are insufficient to enable the insurer to rely upon the ordinary rights of avoidance, but (ii) there is an exception to this principle if there is an innocent non-disclosure or misrepresentation concerning the NAC itself. One oddity of the insurers' argument is that the contract could be avoided in its entirety for an innocent non-disclosure or misrepresentation concerning only the NAC, even if the presentation of the risk was fair in all other respects.
  625. Thirdly, the insurers' argument might be more powerful if the NAC could be regarded as a separable or severable clause akin to an arbitration agreement. If so, then it might be argued (perhaps as a matter of construction) that innocent non-disclosure or misrepresentation would invalidate the separable or collateral agreement; thereby leaving the way clear for innocent non-disclosure or misrepresentation to operate, in the usual way, as far as concerns the agreement as a whole. However, there is authority that non-avoidance clauses are not collateral or separable or akin to arbitration or jurisdiction clauses: see the judgment of Rix LJ (giving the judgment of the Court of Appeal) in HIH v New Hampshire at paragraph [182], and the judgment of Lord Hobhouse in HIH v Chase Manhattan at paragraph [98]. Furthermore, even where the separability principle applies (the classic case being an arbitration clause), it is wrong to regard the "separable" clause as a different and separate agreement from the rest of the contract: see Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb [2020] UKSC 38 at [41], [61] – [63] and [232].
  626. There is therefore no justification for regarding the NAC in the present case as subject to a special regime, whereby – contrary to the ordinary reading of the clause – innocent non-disclosure or misrepresentation is sufficient. I conclude that the effect of the NAC is that any avoidance case must be based on fraudulent non-disclosure or misrepresentation. Given that no such case is advanced, the policy (including the NAC itself) survives, and the insurers' non-disclosure and misrepresentation defences fail.
  627. F3: Affirmation

  628. The Bank and Edge contended that affirmation provided a second answer to the insurers' non-disclosure and misrepresentation arguments. Their proposition was that even if the insurers could establish the necessary ingredients for an avoidance case based on the various non-disclosures and misrepresentations on which they relied, this was a very clear case of affirmation. Since this is put forward as a decisive argument, irrespective of the merits of the non-disclosure and misrepresentation arguments, I shall address it before considering those merits.
  629. Legal Principles

  630. The legal principles relating to affirmation were not in dispute. They are summarised in Arnould paragraphs 15-92 – 15-93 by reference to various well-known authorities including, in the context of election, Motor Oil Hellas (Corinth) Refineries SA v Shipping Corp of India (The Kanchenjunga) [1990] 1 Lloyd's Rep 391. So far as relevant for present purposes (where no estoppel is relied upon by the Bank), the basic principle is that an underwriter is precluded from avoiding the policy if in the meantime he has done something to affirm it. Before there can be an affirmation, the underwriter must have full knowledge of the facts entitling him to avoid the policy. He then has a reasonable time in which to decide what course of action to take, and he cannot be said to have affirmed the policy until after such time. A party will be held to have made an election (for example to determine a contract or alternatively to affirm it) where, with knowledge of the relevant facts, he has acted in a manner which is consistent only with his having chosen one of the two alternative and inconsistent courses then open to him. His election must be communicated either by words or conduct; it will only have effect as a binding election (which is then irrevocable) if it is communicated in clear and unequivocal terms. It is generally a prerequisite to election that the party concerned is aware of the facts which have given rise to his new right. If with knowledge of those facts, he acts in a manner consistent only with treating the contract as still alive, he is taken in law to have exercised his election to affirm the contract.
  631. In Insurance Corporation of the Channel Islands v The Royal Hotel Ltd. [1998] Lloyd's Rep IR 131, Mance J analysed the applicable principles in detail. In his summary of the general principles at p 161, he stated that the party should know sufficient of the facts to know that he has the right to avoid, but it is unnecessary that he should know all aspects or incidents of the facts. The insurers drew attention to passages at page 163 where the judge considered the nature of the knowledge of material facts relevant to affirmation and the communication of an election to the other party. Mance J said that knowledge was essentially a "jury question". It was not to be equated with absolute certainty. But for practical purposes it did pre-suppose the truth of the matters known, and a firm belief in their truth, as well as sufficient justification for that belief in terms of experience, information and/or reasoning. As far as the communication is concerned: this must demonstrate objectively or unequivocally that the party affirming is making an informed choice. The communication itself or the surrounding circumstances must therefore demonstrate such knowledge to the other party.
  632. In addition to knowledge of the facts which give rise to the entitlement to avoid the policy, there is authority that knowledge of legal rights is also necessary. In Moore Large & Co. v Hermes Credit and Guarantee PLC [2003] 1 Lloyd's Rep IR, Colman J said there were authorities binding on the court that it was a requirement for a binding election that the party said to have elected must have full knowledge of the various rights among which he can elect. Arnould comments in a footnote that:
  633. "However, it can rarely if ever be suggested that the point is open to a professional underwriter when affirmation is alleged. Any underwriter must know of the existence of the right of avoidance, consequent on material misrepresentation and non-disclosure".
  634. The burden of proving both knowledge of the facts giving rise to the entitlement to avoid, and knowledge of the legal right to do so, is upon the party that seeks to establish the election: see Moore Large at [99]. In that case Colman J discussed when an inference can be drawn that a party had knowledge of his legal right, in circumstances where an affirmatory act takes place at a time when he is legally represented. He identified a potential difficulty where legal advice had been taken by the party alleged to have elected; ie that it would be impossible to prove that the lay client had chosen to give up rights of which he had knowledge without access to the legal advice which he had received. Colman J's conclusion was as follows:
  635. "[105]  In my judgment, in a case where the party said to have elected has been represented by solicitors and counsel whose conduct is relied upon as amounting to an election, it is normally to be inferred that such conduct has been specifically authorised by the client and has been the subject of legal advice. If, on the evidence before the court it is established that either the legal advisers or the client had knowledge of the facts giving rise to the right said to have been waived at the time when the affirmatory conduct took place, there must be the further inference that the party has been given legal advice as to his rights arising out of those facts. If that inference is to be displaced, there must be evidence of the advice, if any, that was given by solicitors and counsel and of the extent to which the party concerned was aware or was made aware of the right which he appears to have abandoned."
  636. Colman J's approach has been applied in a number of subsequent authorities, and indeed the underwriters did not suggest that it should not be followed. For example, in Involnert Management v Aprilgrange Ltd. [2015] EWHC 2225 (Comm) at paragraphs [15] – [161]. Leggatt J said:
  637. "[160]  The need for knowledge of the legal right, although established by authority, is difficult to justify in principle. The requirement is inconsistent both with the principle that ignorance of the law is no defence and with the principle that in the field of commerce the existence and exercise of legal rights should depend on objective manifestations of intent and not on a party's private understanding. It is also potentially extremely difficult for the other party to prove such knowledge – all the more so since any relevant legal advice which may have been received will be protected from disclosure by legal professional privilege. The unfairness of the rule is mitigated, however, by a presumption that a party which had a legal adviser at the relevant time received appropriate advice. That presumption can only be rebutted by waiving privilege and proving otherwise: see Moore Large & Co Ltd v Hermes Credit & Guarantee plc [2003] Lloyd's Rep IR 315 , 334-6, paras 92-100."
  638. In the leading case which establishes the requirement of knowledge of the legal right (Peyman v Lanjani [1985] 1 Ch 457), Stephenson LJ had similarly spoken of the need to rebut a presumption arising when a party has legal advice.
  639. Edge contended that a relevant factor in the present case is the insurers' delay in asserting an avoidance for some years, and their retention of the premium during that time. They relied upon a decision of HHJ Mackie QC in Argo Systems v Liberty Insurance [2011] EWHC 301 (Comm) where the judge said that the absence of an offer to return the premium was not of itself determinative, but was a powerful factor particularly in a case where the amount of the premium is high and there would be a reason, other than clerical inefficiency, for the insurers to retain it. The judge referred to a passage in MacGillivray which states (in the current 14th edition at paragraph 17-096):
  640. "Failure to return the premium is not per se a waiver of the right to avoid for non-disclosure. But refusal to pay a claim while not declaring avoidance and making a return of premium is evidence of an intent to affirm the contract."
  641. An insurer who has affirmed a contract with knowledge of the relevant facts may subsequently receive further information giving rise to a fresh right to avoid. In Involnert, Leggatt J said that the test for that purpose is whether the further facts subsequently discovered not to have been disclosed would make a material difference to the reasonable insurer's decision whether to affirm or to avoid the policy. He referred to the decision in Spriggs v Wessington Court School Ltd. [2005] Lloyd's Rep IR 474, para [78], where Stanley Burnton J said:
  642. "78.  It is implicit in the basis of avoidance that an insurer who has not avoided a policy despite knowledge of non-disclosure may subsequently do so if he learns of material additional non-disclosure by the insured. The requirement of affirmation that the insurer has knowledge of the facts not disclosed is to the same effect. An insurer may be reluctant to avoid for non-disclosure, for reasons connected with its appreciation of its liability or the effect of non-disclosure on its reputation or some other reason, and decide not to avoid a policy initially, but reasonably wish to do so when additional non-disclosure comes to light. But not every new fact not disclosed will entitle an insurer to avoid after affirmation. Just as it is unnecessary for affirmation that the insurer knows (to use the language of Mance J in the Icci case) "all aspects or incidents of the (undisclosed) facts", provided he "knows sufficient of the facts to know he has the right to avoid", so his learning of the immaterial aspects or incidents of the undisclosed facts cannot entitle him again to elect to avoid the policy. The new non-disclosure must make a material difference to the reasonable insurer's decision whether to affirm or to avoid the policy."

    The parties' arguments

  643. The Bank and Edge argued that, applying these principles, the underwriters had affirmed the policy. The affirmatory act, or unequivocal communication of their election, was the service of the defence in February 2019. No point on avoidance was taken in that defence. Nor was there any reservation of right to do so, either in the document itself or in the correspondence pursuant to which it was served. Instead, the case advanced was based on the terms of the policy, and it included a claim for rectification. By that time, the underwriters were fully aware of all of the matters now relied upon in support of their attempted avoidance. They had adduced no evidence which rebutted the inference or presumption that they had knowledge of their right to avoid.
  644. The underwriters argued that, on a proper analysis, there was no affirmation. There had been a full reservation of rights in their letter of declinature in August 2017, and this reservation had been maintained in the ensuing pre-action correspondence including in their response, in June 2018, to the Bank's letter of claim. That reservation had never been withdrawn. At the time that the defence was served, the underwriters only had limited documentation available to them: they did not have the parties' initial disclosure, extended disclosure or witness evidence. Against that background, the acts of underwriters relied upon in support of affirmation – by which the underwriters denied the Bank's interpretation of the policy and denied the claim – were not clear and unequivocal affirmatory acts. The necessary act was an unequivocal statement or representation that the insurers were treating the contract as binding on them, made with knowledge of the right to avoid, and with knowledge of the underlying facts giving rise to the right to avoid. The underwriters had never wavered in their denial of the claim, and their position that the policy did not cover credit risks. In the context of that argument, construction and disclosure were really two sides of the same coin, as shown by the decision in Cheshire v Thompson (1918) 24 Com Cas 114 (Bailhache J) and (1919) 24 Com Cas 198 (CA).
  645. The underwriters expanded upon these points in their written and oral closing submissions. In relation to the requirement for an affirmatory act, the underwriters relied upon the fact that the claim advanced in the Bank's Particulars of Claim – to which they were responding in their defence – was made under the later policy, ultimately contained in Endorsement No. 1 to the 2016/17 policy.
  646. The original defence put forward RSA's case that the true construction of the TPC was dependent upon PLOD, or that the same result should be reached as a result of the agreement between Mr. Beattie and Mr. Mullen, estoppel or rectification. There was in these circumstances no unequivocal representation that the RSA was treating either the later or earlier policy as binding. Rather, their position was that they were bound by the policy but only on the basis that it was restricted to PLOD and did not extend to credit risks. There was no representation that they were bound by the policy if its true scope extended to credit risks. The avoidance argument was the flip side of the other defences.
  647. The position of the 2nd – 12th Defendants was essentially the same as RSA. However, there was an additional point of importance. They said that they had subscribed only to the earlier policy (ie the policy subscribed between 25 and 29 January 2016), and were not bound by the later policy if (as alleged) it included cover for credit risks. They referred in that connection to paragraph 52 of the defence, in which the following market alleged that if the later 2016 policy did extend to cover credit risks, the RSA was not authorised to bind any of the 2nd – 12th Defendants. Accordingly, the 2nd – 12th Defendants had made it clear that the amended policy, pursuant to which the claim was made, was not binding on them if it bears the construction alleged by the Bank. The position of the 13th and 14th Defendants was essentially the same as for the 2nd – 12th Defendants, except that they accept that they are bound by the later policy.
  648. The underwriters also relied upon the fact that, in response to the initial declinature, Edge had relied upon the NAC. In those circumstances, it is unsurprising that the underwriters did not seek to avoid – they were not in a position to allege fraud. Their subsequent defence could therefore not be viewed as an unequivocal affirmation of the policy.
  649. The service of the defence has also to be seen in the context of the underwriters' reservations in their declinature. It was not necessary for there to be a reservation every time a potentially affirmatory step is taken: it must always be a question of fact whether there has been a sufficiently unequivocal act.
  650. As far as knowledge is concerned, the underwriters submitted that there had to be a firm belief in the truth of the relevant facts, with sufficient justification, and not merely something that could in theory have been pleaded. Here, there was insufficient information to give rise to a firm belief, on the part of the underwriters, that the intention of the Bank and Edge had been to obtain credit risk cover. Whilst it might be said that the underwriters could have known that the NAC had not been brought to their attention, it would have made little sense to avoid for that reason alone: the NAC and the TPC should be viewed in conjunction. Finally, as far as D2-D12 are concerned, the underwriters did not have (and indeed still do not have) a firm belief that the July 2015 document was contractually agreed. Accordingly, the argument that there was non-disclosure of the 2015 endorsement could not have been advanced.
  651. Discussion – (i) the facts

  652. I start by identifying what I regard as the important features of the factual background to the service of the underwriters' defence and counterclaim, which is the key document relied upon as constituting the affirmatory act. The factual background is relevant both to the context in which the defence was served, and also the question of whether the underwriters had knowledge of the material facts relied upon in support of the case of avoidance.
  653. On 17 October 2016, the Bank's potential claim was noted on the Xchanging system. The preliminary notification was of a potential claim "due to Euromar defaulting on its obligations to purchase various cocoa products". Mr. Jones of RSA then asked the question whether there was any known shortage or physical damage to the cocoa in which the Bank had an interest. The answer was: "not to the knowledge of Icestar". This response was also available on the Xchanging system. Mr. Jones' entry on 2 November 2016 said that the Bank's "responses to comments" were noted.
  654. In December 2016 there was the exchange (already described in Section E above) between Mr. North of Edge and Mr. Jones concerning coverage. Mr. North said, in his e-mail of 14 December 2016, that it was "very clear" that the "financial loss as a result of Euromar's default" fell under Clause 1.5 of Section 2 of the policy: ie the TPC in the earlier policy. On the following day, 15 December, Mr. Beattie had e-mailed Mr. Mullen indicating that RSA may have a problem over "a recent loss which is effectively a Financial Guarantee not linked to any loss or damage recoverable under ICC 'A'". In an e-mail later that day, Mr. Mullen said that the incident "comes under the transaction premium section of the policy. This section covers the insured in the event of a default on the part of a[n] insured's client". At 11.37, Mr. Jones confirmed by email that Section 2, Clause 1.5 of the policy "would appear to attach in respect of this situation", and a notification to similar effect was posted on Xchanging.
  655. At this stage, therefore, it was clear that the Bank and Edge were maintaining that there was coverage under the TPC for the consequences of financial default of Euromar, and RSA agreed that this appeared to be correct. The correspondence at this time did not directly address the question of whether this coverage was what had been subjectively intended by the Bank when it had proposed the TPC, or what the Bank's purpose had been when seeking agreement to the TPC. It would, however, be a very obvious conclusion that where the Bank's policy contained a bespoke and carefully drafted clause which, on a natural reading, covered financial default, that the Bank's intention and purpose would have been to have coverage for that which the clause provided. It is, however, theoretically possible, that the width of the wording and the extent of the coverage were accidental, in the sense that they went further than the Bank intended. This seems somewhat unrealistic in the present context, given the nature of the clause and the repeated reference to "Default". However, there had not at this stage been any statement on behalf of the Bank as to what its subjective intention or purpose had been. The question had not been asked, no doubt because the underwriters rightly considered that what mattered was what the words used in the policy meant, rather than what the Bank thought they meant. On that important question, both Mr. North and Mr. Jones were at that stage ad idem: it provided the coverage for which the Bank contended.
  656. There were a number of developments in the early months of 2017. Mr. Beattie declined to include the TPC in the renewed cover. It became apparent, following Transmar's default, that the likely claim was very substantial. Investigators, Gray Page, were appointed on behalf of underwriters. Mr. North was advised by Mr. Jones that the purpose of the instruction to Gray Page was "to investigate and liaise with the Insured to provide a comprehensive report on the circumstances of this matter". By June 2017, Kennedys Law LLP (a very well-known City law firm with considerable insurance expertise) had been instructed by underwriters.
  657. The important correspondence relating to affirmation begins on 16 August 2017, when a letter of declinature was sent by the RSA. It was signed by Mr. Jones. It was, however, sent on behalf of all underwriters (as paragraph 124 of underwriters' written closing submissions acknowledged). It referred to a previous meeting at the offices of Reed Smith (the Bank's solicitors), and the subsequent provision of information to Kennedys Law LLP.
  658. In that letter, RSA rejected the contention that Clause 1.5 of the General Conditions gave coverage for the loss which the Bank had suffered. There was no right to an indemnity in the absence of PLOD to the property insured. The basic argument advanced was that the terms of Clause 1.5 "were intended to extend the scope of the indemnity to which the insured would be entitled, upon the occurrence of an insured loss, but were not intended to extend the scope of cover". This is, in essence, the construction argument which I have addressed in Section D above. At this stage, these references, as to what was intended by the policy terms, were confined to the objective intention of the parties derived from their language. No reliance was placed upon the discussions between Mr. Beattie and Mr. Mullen (discussed in Section E above) subsequently relied upon in support of the rectification/ collateral contract/ estoppel case.
  659. The declinature letter also stated that the underwriters had "undertaken a review in relation to the chronology leading to the issuance of the Endorsement". Although the declinature letter referenced Clause 1.5 of the policy (ie the TPC in the earlier policy), it went on to describe the further policy wording which was scratched by Mr. Beattie on 29 January 2016, and then the signature of what purported to be "tidy-up" wording on 1 February 2016 by Mr. Vaughan. The letter referred to Mr. Lockyer's e-mail of 1 February to the market, following Mr. Vaughan's signature of the GUA box indicating that the endorsement should be listed to the market. The letter reiterated their declinature, notwithstanding the revised policy wording, stating that the policy only provided cover for PLOD to the subject-matter insured.
  660. The letter went on to refer to statements made by Mr. Mullen and Mr. Lockyer at the time of the scratch of the later policy on 29 January 2016 and the subsequent endorsement:
  661. "However, insurers specifically further rely upon the inducements made by Mr. Mullen and Mr. Lockyer in respect of their respective representations, as set out above. In those circumstances, insurers reserve their position to avoid or repudiate the policy for non-disclosure or misrepresentation based upon the above material representations
    Insurers fully reserve, and do not waive, any of their rights and defences under the policy (including the revised policy) and at law, including their right to assert additional grounds for avoidance and/or recession [sic – presumably rescission] or, alternatively, revise their coverage position after the review of any further documents and information that the assured is able to provide"
  662. At this stage, therefore, there was a comprehensive reservation of rights. It is apparent that underwriters, together with their advisers, had by this stage investigated the contractual position. They had identified that there were two policies that were in existence and potentially in issue: ie (i) the earlier policy initially subscribed between 25 and 29 January 2016 and (ii) the later policy subscribed by Mr. Beattie on 29 January and then (with some amendments) attached to Endorsement No. 1.
  663. Edge contended that, at this stage, underwriters had the necessary knowledge to avoid the policy as matters then stood. I consider that there was force in that contention in so far as the grounds for avoidance are based upon non-disclosure or misrepresentation as to the terms of the policy itself. Thus, one of the grounds for avoidance is the Bank's alleged failure to disclose the NAC. By the time of the August 2017 declinature, underwriters had the various policies in front of them, and could see clearly that they contained a NAC. Another ground of avoidance, related to the terms of the policy, is the alleged misrepresentation to Standard that the policy only covered physical loss and damage to the subject-matter insured. By the time of the August 2017 declinature, Standard could see from the terms of the policy that, on any view, it covered matters other than physical loss and damage to the cargo. The policy contained the CEND, BCC and Fraudulent documentation cover discussed in Section D above, as well as the TPC itself which, as Standard knew, the Bank alleged to apply if there was financial default unrelated to PLOD.
  664. However, other aspects of the facts, on which the avoidance case was ultimately based, were arguably not known or at least were not clearly known at this stage in August 2017. But it is not necessary to examine the state of knowledge in August 2017 in more detail, because matters then moved on.
  665. On 8 November 2017, Edge sent a detailed response (nearly 6 pages) to the notice of declinature. They said that the declinature was wrongful and could not be justified.
  666. The letter had a number of headings. The first was: "The position on coverage". This set out, over 3 pages of text, the argument as to why the policy responded. At the start of this section, Edge set out the TPC, and explained that the "preceding policy had contained identical wording by virtue of the endorsement scratched on 28 July 2015 by Mr. Beattie for RSA as leader and on behalf of the following market". Accordingly, the existence of the July 2015 endorsement was clearly identified at the start of the letter. In the concluding part of that section, Edge referred back to RSA's reaction when the claim had been presented, in particular the discussions in December 2016 leading to RSA's confirmation that coverage appeared to exist. Edge said, therefore, that their interpretation of the TPC, as creating a freestanding insured peril not dependent on physical loss or damage to the cargo, was "entirely consistent with your own understanding of the policy".
  667. The next heading was as follows:
  668. An alternative argument: Rectification (if necessary)
    As we have explained, the correct construction of the Policy is clearly in Icestar's favour. Without prejudice to that, and if it were for some reason to be found that the Transaction Premium clause did not create freestanding cover, it would necessarily follow that the Policy would need to be rectified to give effect to the parties' common intention.
    As the evidence set out above demonstrates, it was RSA's understanding that the Transaction Premium clause was a separate head of cover not dependent on physical loss or damage. Our understanding, on behalf of our client, was precisely the same. If (contrary to all that we have said above) the final wording of the Policy did not have that legal effect, this can only have been as a result of a mistake in the written expression of the agreement. The Policy would then need to be rectified to make expressly clear that the Transaction Premium clause did indeed create independent cover.
    Icestar has specifically asked us to inform you that its legal rights to seek rectification of the Policy, if that be necessary, are strictly and fully reserved.
  669. This part of the letter is significant. It went beyond arguments as to the objective meaning of the words in the policy. It related to the contractual intention of Edge, on behalf of its client, in a broader and subjective sense. It contained a clear statement that the understanding of Edge, on behalf of its client, was that the TPC was a separate head of cover not dependent on physical loss or damage. This paragraph cannot in my view be read as anything other than an assertion that the TPC was subjectively intended to provide the default cover that, on its natural reading, it did provide.
  670. The next section of the letter addressed "The circumstances of the placement" in greater detail. It emphasised that the TPC was not introduced into the policy for the first time by a last-minute endorsement made at the end of January 2016.
  671. "This, however, overlooks the fact that the identical wording appeared in the expiring terms, having been introduced by an endorsement scratched by Mr. Beattie and expressly approved by him on 28 July 2015, some six months earlier".
  672. The letter went on to explain that the same clause, identically worded, was included in the draft renewal terms for 2016/2017 which had been discussed between Mr. Mullen and Mr. Beattie on or about 20 January 2016. The course of the placement was then described. The letter then challenged the points which had been made in the declinature letter concerning misrepresentation and non-disclosure. Included amongst the points raised were:
  673. "2. Second, we do not see how it can possibly be said that there was any non-disclosure or misrepresentation: the Transaction Premium clause had been in the expiring terms, and had been expressly approved by Mr. Beattie both at the time of its introduction (28 July 2015) and again at the time of renewal. Furthermore, Mr. Beattie expressly approved the placement of the Transaction Premium clause in Section 1 and Section 4 and scratched the final slip in those terms, having reviewed it in Mr. Mullen's presence.
    …
    4. Your allegations are for all these reasons misconceived, but in any event you will have noted that the Policy contains a standard form "Non-Avoidance" clause (Section 2, Clause 21). This prevents underwriters from avoiding the Policy for non-disclosure or misrepresentation unless made fraudulently. You are not seriously suggesting that we (or Icestar) are guilty of fraud, are you?
  674. In the concluding part of the letter, Edge asserted that the declinature was opportunistic where the evidence was
  675. "entirely clear that the Transaction Premium clause was expressly agreed to by Mr. Beattie as an "add on" to traditional cargo cover, and RSA genuinely understood that the clause was a form of credit insurance not requiring any physical loss or damage to the cargo."
    This part of the letter therefore reiterated the case that it was not accidental that the language of the TPC covered the loss, arising from the default of Euromar and Transmar, which the Bank had suffered.
  676. The RSA's response on 30 November 2017 was short. They referred to underwriters' declinature in its earlier letter, and reiterated that declinature. The response did not engage with any of the points taken. It concluded with a broadly worded reservation of rights which was in materially identical terms to that contained in the declinature letter.
  677. On 21 May 2018, Reed Smith sent Pre-Action Protocol Letters of Claim to Kennedys and RPC who were acting for Edge. In the letter to Kennedys, Reed Smith aligned the Bank with the coverage analysis of Edge in its letter of 8 November 2017 as well as the rectification case:
  678. "There was a common intention that the Transaction Premium clause was a separate head of the cover not dependent on any physical loss or damage occurring to the cargo".
  679. Kennedys responded on 8 June 2018. In relation to the claim for rectification, underwriters rejected the suggestion that there was any common intention that the policy should provide cover for credit risks in the absence of physical loss or damage. They noted that Edge contended that this was their understanding, and Kennedys asked for disclosure of the brokers' file and the terms of any credit risk insurance which Edge had placed with other underwriters. Their clients' position in relation to non-disclosure and misrepresentations remained fully reserved.
  680. The present proceedings were then commenced by the Bank, against all the underwriters and Edge, on 29 November 2018. In its Particulars of Claim dated 30 November 2018, the Bank pleaded that the terms of the policy were contained in the document stamped and signed by Mr. Vaughan of RSA on 1 February 2016.
  681. The pleading contained a claim for rectification (if necessary) on the basis of mutual mistake. In that regard, Paragraph 13 asserted that:
  682. "It was at all material times the understanding and intent of Icestar, and of Edge Brokers on Icestar's behalf, that the Transaction Premium clause applied irrespective of whether any physical loss or damage occurred to the cargo. The whole purpose of the clause was to protect Icestar in the event that a client defaulted and did not repurchase cargo, causing Icestar to suffer loss when that cargo was resold; the occurrence of physical loss or damage to the cargo was irrelevant to that purpose".
  683. The Bank then set out a number of facts and matters relied upon in support of that allegation. These included the important communications between the Bank and Edge, and internally between Mr. Lockyer and Mr. Mullen, on 28 January 2016: ie when Ms. Van de Beek of the Bank raised the question of whether the cover was separate to the marine cargo and storage cover, and when Mr. Lockyer told Mr. Mullen that the Bank was concerned to have cover if the client defaults regardless of PLOD.
  684. Edge's defence was served on 21 December 2018, and was therefore available to the underwriters when they served their defence and counterclaim. Edge addressed the question of its subjective understanding of the TPC: paragraph 11.1 admitted that Mr. Mullen by no later than July 2015 understood that the Bank required insurance cover in the event that a client defaulted under a repo transaction, and that such cover was not to be dependent on physical loss or damage to the cargo. At the same time as serving its defence, Edge served an additional claim pursuant to CPR Part 20 upon NRF. The claim set out the details of material correspondence involving NRF in June 2015 and upon renewal in January 2016. The documents relating to NRF's advice were attached to the Part 20 claim.
  685. The underwriters' defence was served on 22 February 2019. By the time that it was served, the underwriters had received various documents which specifically addressed the intention of the Bank and Edge in relation to the TPC; Edge's letter dated 8 November 2017, the Bank's Particulars of Claim, Edge's Defence, and the additional claim against NRF. Some of the pleadings set out the terms of material correspondence which concerned the subjective intention of those involved. The essential case advanced was that that the TPC had been intended by Icestar, Edge and NRF to give rise to a separate head of cover in respect of the default risk, and had been drafted by NRF for that specific purpose. In addition, the materials had addressed the circumstances in which the TPC had originally been presented and signed by Mr. Beattie in the July 2015 endorsement, as well as the relocation of the TPC at the end of the renewal process in 2016.
  686. The defence is lengthy, and it is not necessary to summarise every defence advanced. There was no suggestion, however, that any point on avoidance for misrepresentation or non-disclosure was either taken or reserved. The covering e-mail contained no reservation in relation to avoidance. For present purposes, the most important points which were taken were as follows.
  687. The defence denied that the July 2015 "wish list" was binding on any of the Defendants. It relied upon the conversations between Mr. Mullen and Mr. Beattie, both in July 2015 and upon renewal, to the effect that the TPC was a basis of valuation clause.
  688. Paragraph 17 pleaded, in conventional terms, that the 1st – 14th Defendants would refer to the 2016 Policy for its full terms, meaning and effect. The 2016 Policy referred to here was a general reference to the policy presented for renewal in January 2016 (see paragraph 16 of the pleading). Later paragraphs distinguished between the "unamended" policy originally subscribed (ie the earlier policy) and the "Amended 2016 Policy" as amended by Endorsement No. 1 (ie the later policy).
  689. Paragraph 20 pleaded that when the terms were presented to the 2nd – 14th Defendants in 2016, (i) it was not suggested that Icestar was looking for cover in respect of credit risks or financial defaults and (ii) it was not even brought to their attention that the TPC had been included. None of these matters was, however, relied upon in support of a case of avoidance.
  690. Paragraph 21 relied upon the conversations on 27 and 29 January with Navigators, Ark and Advent concerning the policy being "as before" or "as expiry". It also relied upon the conversation with Standard that the policy covered the Bank for loss and/or damage to products where finance was provided. Again, however, these matters were not relied upon in support of a case of avoidance. Rather, they (and other matters) were relied upon in support of the plea in paragraph 24 that there was an agreement with the 2nd – 12th Defendants that the TPC related only to the basis on which the insured goods were to be valued. A plea of estoppel and/or estoppel by convention was also raised in paragraph 29 in consequence of these matters.
  691. Paragraph 25 pleaded that if Edge had been intending to place insurance cover in respect of credit risks or financial defaults, they ought to have approached specialist underwriters and made a fair presentation of such risks. It identified the matters that an underwriter, who was prepared to contemplate such risks, would require as part of the placing information.
  692. Paragraphs 43 – 48 set out the position of RSA. The case advanced in paragraphs 43 and 44 concerned the construction of the TPC in Section 2, clause 1.5 and then in the later 2016 policy. Paragraphs 45 – 47 advanced the claim based on the meaning of the TPC as agreed between Mr. Mullen and Mr Beattie, and the related arguments based on estoppel and collateral contract. Paragraph 45 (d) pleaded that if Mr. Mullen had said that cover for credit risks was being sought, the inclusion of the clause would have been refused. Paragraph 48 pleaded a claim for rectification, in the event that either the amended or unamended policy provided, on its true construction, for cover against credit risks.
  693. Paragraphs 49 – 57 set out the position of the 2nd – 14th Defendants. Paragraphs 49 and 50 repeated the case advanced in paragraphs 43 and 44: ie as to the true construction of the unamended and then amended policy. Paragraph 51 relied upon the agreement (ie between Mr. Beattie and Mr. Mullen) that the TPC, in both the amended and unamended policies, related only to the basis of valuation. That paragraph also indicated that the 2nd – 12th Defendants had subscribed to the unamended policy, and only the 13th and 14th Defendants had subscribed to the amended policy wording. Paragraph 52 pleaded that if the amended 2016 policy did extend to cover credit risks or financial defaults, or increased the monetary exposure of the underwriters, the RSA was not authorised to bind those Defendants to those changes. Paragraphs 54 – 57 pleaded various estoppel and related points, including in paragraph 56 a plea of rectification.
  694. Paragraphs 61 – 64 pleaded back to the Bank's claim for rectification. No admission was made as to the Bank's alleged subjective understanding of the policy as covering credit risks, but the underwriters said that such an understanding would be surprising.
  695. At the end of the pleading, the underwriters counterclaimed, if necessary, for rectification.
  696. The defence remained in that form until, in May 2020, the amendments to raise non-disclosure and misrepresentation were made.
  697. Discussion – (ii) affirmatory act

  698. As discussed above, a party's election (to affirm) must be communicated either by words or conduct. It must be communicated in clear and unequivocal terms. As Mance J said in ICCI v Royal Hotel, the communication itself or the surrounding circumstances must demonstrate objectively or unequivocally that the party affirming is making an informed choice.
  699. I consider that the service of the defence, when viewed in the context of the surrounding circumstances, does demonstrate objectively and unequivocally that the insurers were making an informed choice; ie to affirm the policy which had been concluded. The background was that there had been prior correspondence in which reservations of the right to avoid had been intimated. However, when the defence was served, there was no such reservation let alone any plea of avoidance. Instead, reliance was placed upon the terms of the policy and a positive plea was advanced, by all underwriters, that if necessary the policy should be rectified. That plea therefore sought to invoke the court's equitable jurisdiction to rectify the instrument which contained the parties' agreement. In my view, such a plea – unaccompanied by any plea of avoidance or reservation – can only be seen as a positive statement that the contract which the parties have made is binding, albeit that the document containing that contract requires rectification so as to conform with the parties' mutual expressed intentions. As Ms. Sabben-Clare submitted, a plea of rectification is necessarily affirmatory, since it positively avers that there is an enforceable agreement between the parties and seeks to have the documentary record of the contract rewritten.
  700. That overall conclusion is in my view reinforced by the background, where the possibility of avoidance had been intimated, and the reliance placed upon the terms of the policy itself. Two other matters are also relevant in this connection. First, various paragraphs in the defence, as summarised above, asserted that matters had not been discussed or disclosed, but no case was made that this led to avoidance of the policy. Rather, these matters were prayed in aid principally in support of the insurers' case on construction. Secondly, the underwriters retained the substantial premium which had been paid, with no offer (until the 2020 amendments) to return it.
  701. I did not consider that there was any force in any of the points made by the underwriters as summarised above. It is true that there had been widely-expressed reservation of rights in the pre-action correspondence, albeit no reservation between 8 June 2018 and the time when the defence was served in February 2019. Even though there had been prior reservations, the significant point in my view is that these were not repeated when the underwriters set out, in considerable detail, the case that they were actually advancing in response to the claim made in litigation, including a specific plea of rectification. The straightforward conclusion which would be drawn by a recipient of the defence is that the underwriters had decided to defend the case on the basis of the construction and rectification/ estoppel/ collateral contract arguments which featured in their defence and counterclaim. Each of those defences, and the counterclaim, is premised upon the continued existence of the contract between the parties. To adopt the words of Leggatt J in Involnert at [161], the insurers in their defence and counterclaim spoke or acted in a way which would reasonably be understood as consistent only with the insurer having made an informed choice to treat the contract as valid. The service of the defence in those terms was therefore an affirmatory act, and – as the brief decision of the Court of Appeal in Barber v Imperio Reinsurance Company UK Ltd (15 July 1993, unreported) indicates – it is incumbent on a party to reserve his position if he takes a step which would otherwise constitute affirmation, and if not he does so at the risk of being held to have affirmed.
  702. Contrary to underwriters' argument, I do not regard the construction defence as being akin to, or the functional equivalent of, a case of avoidance, nor that avoidance is simply the other side of the construction coin. They are very different defences, with different legal and evidential requirements and different consequences, leaving aside the fact that avoidance will usually require the premium to be tendered or repaid to the insured. Indeed, as Ms. Sabben-Clare submitted, even if the two defences (construction and avoidance) can be equated, in the sense that they both amount to a denial that the underwriters insured credit risks, that does not assist the underwriters. It leads to the conclusion that both points were open to the underwriters when the defence was served, but the underwriters only advanced the construction argument. It therefore reinforces the conclusion that there was an unequivocal election.
  703. Nor do I think that (as the underwriters submitted in their written submissions) the service of the defence cannot be viewed as an unequivocal affirmation of the policy, because Edge had referred to the NAC, and because it is unsurprising that the underwriters did not avoid since they were not in a position to allege fraud. In his oral submissions, Mr. Parsons said that in view of the fact that the NAC had been raised, the conclusion to be drawn from reading a defence which does not raise avoidance is that the insurers are considering themselves bound by the NAC at that stage, and not pleading fraud.
  704. Mance J said in ICCI (at 163) that the question of whether conduct amounts to an unequivocal communication of a choice to affirm requires an objective assessment of the impact of the relevant conduct on a reasonable person in the position of the other party to the contract. A reasonable person in the position of the Bank, when served with the defence in the context described above, would reasonably conclude that the insurers accepted that there was a binding policy, and had taken the view that arguments about non-disclosure or misrepresentation, previously reserved, were not being pursued. Even if (as Mr. Parsons suggested) the conclusion drawn from reading the defence is that the insurers considered themselves bound by the NAC at that stage, this does not detract from the simple point that the defence and counterclaim unequivocally represented that there was a binding contract. I do not think that the route by which the insurers reached that conclusion, and whether or not it was as a result of the NAC, affects the impact of the relevant communication on the recipient, particularly bearing in mind that the recipient could only speculate as to what that route had been. In any event, I do not think that a reasonable person would think it necessary to try to analyse why the insurers, who are professional underwriters advised by insurance specialists, had decided to take the position that, as the defence and counterclaim represented, there was a binding policy. What matters is that this was the position clearly communicated.
  705. Nor do I accept the argument, advanced by the 2nd – 12th Defendants, that there was no affirmatory act because they alleged that they were only bound by the earlier policy, rather than the later policy upon which the Bank's case in its Particulars of Claim was based. In my view, the significant point is that the 2nd – 12th Defendants' defence accepted and indeed asserted that there was a binding contract with the Bank, and sought rectification of that contract, if necessary. Whilst the pleadings indicated that there were issues as to whether the relevant contract was contained in a document signed between 25 and 29 January 2016, or a document signed on 1 February 2016, the existence of a binding agreement was not disputed. There was, therefore, no suggestion that the contract could be avoided if it was indeed contained (as the Bank contended) in the later contract. Any reasonable recipient of the defence would conclude that the 2nd – 12th Defendants accepted that there was a binding contract, and were not advancing a case of avoidance, whether that contract was contained in one document or the other. As Ms. Healy submitted, there was no dispute that there was a contract (one contract) between the Bank and each of the underwriters: the dispute was as to its terms. All underwriters, whichever wording they scratched, pleaded a case as to the true meaning of the TPC and to that extent relied upon the policy and also pleaded rectification. The fact that some underwriters did not accept that they were bound by the later policy relied upon by the Bank, but asserted that the binding contract was contained in the earlier policy subscribed between 25 and 29 January, is in my view irrelevant to the question of affirmation.
  706. Discussion – (iii) knowledge of the material facts

  707. I consider that all of the material facts relied upon by the underwriters in support of their avoidance case were known to them by the time that they received Edge's letter dated 8 November 2017, and in any event by the time that the underwriters served their defence. I shall consider each point relied upon in turn.
  708. (1) Purpose/ intention of the TPC. This ground of non-disclosure is concerned with the subjective purpose and intention of the Bank when seeking cover which included the TPC. I have already indicated that it is somewhat unrealistic to suppose that, in intimating its potential claim in 2016 in reliance on the carefully drafted TPC, the Bank was seeking to take advantage of accidental drafting of the TPC; ie drafting which enabled them to contend, contrary to their intention and purpose, that the TPC covered the financial default of Euromar in the absence of PLOD. However, the response of Edge in November 2017 made it clear that, at least from their perspective, the terms of the TPC were no accident and were indeed intended to cover the default risk irrespective of whether there was PLOD to the subject-matter insured. This was reiterated in the Particulars of Claim, where the Bank's subjective purpose and intention was clearly pleaded in the paragraphs relating to rectification, including with reference to contemporaneous documents. It was also reiterated in the defence served by Edge, and the additional claim on NRF, both of which were available to the underwriters when the defence was served.
  709. I did not think that the underwriters' argument – that they did not know whether Edge and the Bank were telling the truth as to their intention and purpose, and have always disputed it – was of any assistance. It is correct that the underwriters have, to a greater or lesser extent, put in issue the question of whether the Bank or Edge subjectively intended to cover risks of default. Ultimately, their case was that NRF went further than the Bank subjectively wanted, since the Bank was aiming at covering the "market premium" albeit (as I understood the argument) that they might have wanted cover for this element against the default risk.
  710. In my view, however, this is beside the point. If the question of the Bank's subjective intention and purpose was material for disclosure, then the underwriters had all the material which they needed, by November 2017 and certainly by February 2019, to plead that case on a sound factual footing. The underwriters could rely upon the admission, in Edge's letter, as to their subjective intention. They could also rely upon the Bank's plea of rectification, which was necessarily based upon the proposition that the Bank or its broker subjectively intended the TPC to cover the credit risk. They could also rely upon Edge's pleadings. The fact that the underwriters wished to dispute what Edge and the Bank were saying does not in my view mean that they did not have sufficient knowledge of the relevant facts. In a case where the relevant facts concern the subjective intention of the other party to the contract, and the other party has admitted and averred that subjective intention, it is difficult in my view to see what else an underwriter needs to know in order to plead an avoidance case based on subjective intention.
  711. There is no reason in principle why that avoidance case cannot be put forward on a conditional basis. The possibility of doing so, in an avoidance case, is adverted to in the Spriggs case at paragraph [77] ("On the basis of this information, RSA could have avoided any policies dating from after about 1976, if necessary on a basis qualified by reference to the truth of the Claimants' allegations"). Indeed, a conditional basis is how the avoidance case was eventually put forward. For example, paragraph 20A of the Re-Re-Amended Defence pleads:
  712. "Further or alternatively, if, as the Bank contends, the Transaction Premium clause was intended to and/or did give rise to a separate head of cover in respect of credit risks and/or financial defaults, then:
    a. that was a material fact of which a reasonable and prudent underwriter would have desired to be informed before deciding whether, and if so on what terms, to agree [the 2015 endorsement and/or the 2016 Policy] …"
  713. Similarly, paragraphs 48A and 48C plead avoidance premised upon the following:
  714. "… if, contrary to the case of the First to Fourteenth Defendants herein, the Transaction Premium clause does not have the application, meaning and effect referred to in paragraphs 43- 44 above and/or it was not specifically agreed between Mr. Beattie and Mr. Mullen that it had no such application, meaning and effect, then RSA was induced to underwrite the 2016 Policy and/or was induced to write it on more favourable terms [than] would otherwise have been the case …"
  715. Accordingly, there is no reason why the underwriters could not have said simply: "If, as the Bank and Edge assert, but the underwriters do not admit (or deny) their purpose and intention was to obtain credit risk cover, the policy is voidable and is hereby avoided for non-disclosure of that purpose and intention". That is in substance what they did say, when they eventually pleaded the point. Ms. Sabben-Clare was correct in submitting that nothing new had happened, in terms of knowledge, in order for the conditional point to be pleaded. Underwriters, apparently, have a firm belief now in their right to avoid, if these things are true now. There is no reason for them suddenly to have developed that view in May 2020 when the conditional point was pleaded.
  716. Nor do I consider that anything material, in the sense described in Spriggs and Involnert, emerged subsequent to November 2017 or February 2019. Whilst documents were disclosed which were indeed relevant to the intention and purpose of the Bank and Edge, the underwriters knew at those earlier times sufficient of the facts to know of the right to avoid which they subsequently asserted. As Mance J said in ICCI, a party does not have to know all aspects or incidents of the (undisclosed) facts. Whilst it is true that, in February 2019, underwriters had not seen the full brokers file, they had seen pleadings from both the Bank and Edge which were accompanied by the requisite statement of truth. Those pleadings quoted from documents which were relevant to the parties' subjective intentions, and some of those documents were attached to Edge's additional claim.
  717. My strong impression is that the belated appearance of the avoidance case was not the consequence of the disclosure of documents or witness statements which made a material difference to the reasonable insurer's decision whether to affirm or to avoid the policy. Indeed, the insurers have continued to dispute the case of the Bank and Edge as to their subjective intentions, notwithstanding that disclosure. Rather, and as reflected in paragraph 122 of the underwriters' closing submissions and Mr. Parsons' oral submissions, there seems to have been reconsideration of the question of whether an avoidance case can be successfully advanced in the absence of an allegation of fraud, notwithstanding the NAC.
  718. (2) Non-disclosure of the NAC. The NAC was in the policies which were reviewed by the insurers, with the assistance of their lawyers, prior to the declinature in August 2017. It was expressly referred to in Edge's response to the declinature letter. Underwriters knew at that time that they had not had their attention drawn to the NAC. If, therefore, there was a material non-disclosure, the material facts were known by August 2017 at the latest.
  719. I do not accept the underwriters' argument that it would have made little sense to avoid the policy on the basis of non-disclosure of the NAC alone, because the NAC and TPC should be viewed in conjunction. If, as underwriters maintain in these proceedings, there was non-disclosure of the NAC with the dual consequence that the policy and the NAC was avoided, then I cannot see why the underwriters could or should not have taken that point at the outset.
  720. (3) The July 2015 Endorsement. Edge's November 2017 letter set out the history of this endorsement in a number of places. Again, all of the 2nd – 14th Defendants knew that they had not been told about this endorsement at the time of placement in 2016. If, therefore, there was a material non-disclosure, the material facts were known by November 2017 at the latest.
  721. I do not accept that it is an answer that the underwriters did not, when the defence was served, have a firm belief that the July document was contractually binding, and that they still do not have a firm belief. Again, the point on the "wish list" could have been put forward on a conditional basis: ie if the "wish list" was a binding endorsement, there was non-disclosure of that fact to the following market. Indeed, this is in substance how the point was pleaded in paragraph 20B of the Re-Re-Amended Defence:
  722. "Further or alternatively, and if the wish list referred to in paragraphs 14-15 above was in fact agreed by Mr. Beattie on behalf of RSA and/or if a valid and binding endorsement to the 2015 Policy arose out of it, which is denied, then …
    b. the fact that the wish list and/or a written endorsement had been agreed was a material fact … If the wish list and/or a written endorsement had been agreed, then that would have been material …"
  723. (4) As expiry. The same conclusion follows. Edge's November 2017 description of the placement explained that terms had been introduced in July 2015 which were then included in the 2016 renewal. Accordingly, all the underwriters who have relied upon this point were aware that the 2016 renewal terms were different, because of terms originally introduced in July 2015, to the terms on which they had originally written the 2015 policy. If, therefore, the statements made to underwriters are to be understood as meaning "nothing has changed since you last saw the risk and wrote it in 2015", the material facts showing that this was incorrect were known by November 2017 at the latest.
  724. (5) Only PLOD misrepresentation. The review of the policy which preceded declinature would have revealed to Standard that the cover extended beyond PLOD in certain respects; ie CEND, BCC and Fraudulent documentation. The letter from Edge referred specifically to the BCC as being an example of a clause which covered financial losses arising without physical loss or damage. Standard was also aware that the Bank was alleging that the TPC was a clause which did not require physical loss and damage. If, therefore, a representation was made to Standard which was to be understood as meaning that there was nothing in the policy which provided cover beyond physical loss and damage to goods, the material facts showing this to be incorrect were known by November 2017 at the latest.
  725. Discussion (iv) – knowledge of the right to avoid

  726. No evidence was adduced by underwriters to rebut the inference or presumption of knowledge of the right to avoid.
  727. However, the authorities relied upon all concern policies where the potential right to avoid is not subject to a NAC which, on its face, confines that right to cases of fraud. The underwriters did not suggest that the inference or presumption was inapplicable in the present context, but nevertheless in different contexts drew attention to the potential difficulty in putting forward an avoidance case in the absence of an allegation of fraud. Thus, as already indicated, underwriters submitted (in the context of the argument as to whether there was an unequivocal act affirming the policy) that it was unsurprising that the underwriters did not seek to avoid, in light of the NAC.
  728. I consider that the inference or presumption, as to knowledge of the relevant right, remains applicable in the present context. This aspect of the case must be approached on the assumption that I am wrong in my conclusion (Section F 2 above) that avoidance requires a case of fraud; ie that the underwriters are correct in their case that an innocent non-disclosure in relation to the NAC itself has the dual effect of enabling underwriters to avoid both the policy and the NAC itself. If that is indeed their legal right, then I see no reason why I should not proceed on the basis that the presumption or inference of knowledge of this legal right applies; so that absent any evidence to the contrary, it is to be inferred or presumed that underwriters had the requisite knowledge of the right which, on their case, exists.
  729. Conclusion

  730. I therefore conclude that, even taking underwriters' avoidance case at its highest – ie assuming that the requirements for material misrepresentation or non-disclosure and inducement are satisfied in respect of each of the matters relied upon – avoidance is barred on the facts of the present case by affirmation. This is, therefore, a second reason why the case on avoidance fails.
  731. F4: The misrepresentations and non-disclosure relied upon: general matters

  732. In view of my conclusions in the preceding sections, it is not necessary to decide what the position would have been in the absence of the NAC or affirmation. I will, however, address those issues.
  733. Legal Principles

    (i) Non-disclosure

  734. I was referred to a number of authorities which identify the relevant principles (pre Insurance Act 2015) concerning non-disclosure. The most comprehensive statement in the recent authorities can be found at paragraphs [134] – [135] in the judgment of Christopher Clarke J in Garnat Trading & Shipping v Baominh Insurance. He said (omitting the citations supporting the propositions in paragraph [135]):
  735. "[134]  The starting point is s.18 of the Marine Insurance Act 1906   ("the MIA"), which provides:
    "18.  — Disclosure by assured.
    (1)  Subject to the provisions of this section, the assured must disclose to the insurer, before the contract is concluded, every material circumstance which is known to the assured, and the assured is deemed to know every circumstance which, in the ordinary course of business, ought to be known by him. If the assured fails to make such disclosure, the insurer may avoid the contract.
    (2)  Every circumstance is material which would influence the judgment of a prudent insurer in fixing the premium, or determining whether he will take the risk.
    (3)  In the absence of inquiry the following circumstances need not be disclosed, namely:—
    (a)  Any circumstance which diminishes the risk;
    (b)  Any circumstance which is known or presumed to be known to the insurer. The insurer is presumed to know matters of common notoriety or knowledge, and matters which an insurer in the ordinary course of his business, as such, ought to know;
    (c)  Any circumstance as to which information is waived by the insurer;
    (d)  Any circumstance which it is superfluous to disclose by reason of any express or implied warranty.
    (4)  Whether any particular circumstance, which is not disclosed, be material or not is, in each case, a question of fact.
    (5)  The term "circumstance" includes any communication made to, or information received by, the assured."
    [135]  Mr Ashcroft distilled a series of propositions which I am content to adopt as accurate general statements of the law. They were as follows:
    (a)  Non-disclosure is the failure to communicate a material fact within the knowledge of the assured which the insurer has not the means of knowing or is not presumed to know.
    (b)  The burden of proof in relation to any allegation that a fact or matter has not been disclosed is upon the insurer.
    (c)  In general terms, a fact or matter is material if it would have been taken into account by a hypothetical prudent insurer when assessing the risk.
    (d)  But, a minute disclosure of every material circumstance is not required. The assured complies with the duty if he discloses sufficient to call the attention of the underwriter to the relevant facts and matters in such a way that, if the latter desires further information, he can ask for it. A fair and accurate presentation of a summary of the material facts is sufficient if it would enable a prudent insurer to form a proper judgment, either on the presentation alone, or by asking questions if he was sufficiently put upon enquiry and wanted to know further details, whether to accept the proposal, and, if so, on what terms. 
    (e)  Underwriters should listen carefully to what they are being told; they cannot complain if they do not grasp the detail or the implications of it.
    (f)  In accordance with s.18(3)(b) of the MIA , in the absence of inquiry, there is no need to disclose a fact or matter that the insurer already knows, or is presumed to know; there is therefore no duty to disclose matters of common notoriety or matters that the insurers should, in the ordinary course of business, know. In the context, the test is objective. One asks what a reasonable insurer, writing the particular type or class of business concerned, would, or should, know. A reasonable underwriter is presumed to know matters which he should have known from the facts in his possession or matters which he had means of learning from the sources available to him. A reasonable underwriter is presumed to know the ordinary incidents or attributes of any peculiar or specialist risk he undertakes: every underwriter is presumed to be acquainted with the practice of the trade he insures; if he does not know, then he ought to inform himself. Because of these aspects, and absent inquiry by the insurer, only unusual elements affecting the risk have to be disclosed by the proposer.
    (g)  As regards s.18 (3) (c) of the MIA, waiver in insurance law bears a wider meaning than it does in other areas of the law. There is no need for an intentional act with full knowledge of the facts. If the facts and matters disclosed give a fair presentation of the risk, the underwriter must ask if he wishes to have more information; further, even if the initial presentation was unfair, waiver might arise if the information disclosed was such as to prompt a reasonably careful insurer to make further inquiries. In short, if the insurers receive information, which taken on its own, or in conjunction with other information known to them or presumed to be known to them, would naturally prompt a reasonably careful insurer to make further inquiries, then, if they omit to do so, they waive disclosure of the material facts and matters which such an inquiry would have revealed. A particular case in which insurers may be put upon inquiry is one where the character of the ship to be insured puts them on notice that specific preparations are or may be required before it puts to sea. Finally, an assured is entitled to assume that the insurers are waiving disclosure of matters concerning which they appear to be indifferent or disinterested.
    …
    (i)  Even where there is non-disclosure of a material fact, if this does not in fact influence the judgment of the actual underwriter, avoidance is not justified. …To justify avoidance, the non-disclosure must be a real and substantial cause affecting the decision of the insurer to enter into the contract, or to do so on the terms agreed, the insurer bearing the onus of proving inducement on the balance of probabilities. No presumption of fact applies where the underwriter is called to give evidence."
  736. The case was the subject of an unsuccessful appeal in which the principal issue, so far as concerns non-disclosure, concerned the judge's conclusion on the facts.
  737. Each side referred to authorities in which judges have described particular aspects of these basic principles in language which each considered to be helpful to the case which it advanced.
  738. The Bank and Edge referred to a passage in the decision of Hobhouse J in Iron Trades Mutual Insurance v Companhia De Seguros Imperio (1992) Lloyd's Rep. IR 213, 224 ("Imperio"):
  739. "The insurer is presumed to know his own business and to be able to form his own judgment of the risk as it is presented to him; thus the proposer is under no duty to offer the insurer advice. The duty relates to facts not opinions. The duty is essentially a duty to make a fair presentation of the risk to the insurer".
  740. They also referred to authorities in support of the proposition that matters of inference or opinion do not constitute material circumstances. In The Elena G [2001] 2 Lloyd's Rep 378, 382, Steel J said:
  741. "[25] One of the features of the present action was the need to bear in mind the distinction between the respective roles of assured and underwriter. The task of the assured is to disclose facts or circumstances material to the risk. It is the underwriter's task to appraise the risk against that fair presentation. The point is accurately and fairly summarised in MacGillivray on Insurance Law 9th Ed. paragraph 17–75:—
    "The assured is not bound to disclose what is merely a matter of inference or judgment from the facts known to the insurers. He is bound to supply the insurers with the facts but he is not bound to estimate the risk for them. If the insurers are unfamiliar with the natural inferences to be drawn from what they are told, they should ask, only counting upon the assured to disclose unusual attributes of the risk which could not ordinarily be appreciated from the facts given".
  742. This paragraph in MacGillivray appears (with the introductory heading "Assured's opinion") in materially identical terms in the current (14th) edition at paragraph 17-086. Amongst the authorities cited in support of that proposition is the classic judgment of Lord Mansfield in Carter v Boehm 3 Burr 1905:
  743. "But either party may be innocently silent, as to grounds open to both, to exercise their judgment upon."
  744. One additional sentence in the current edition, which was not quoted by Steel J is:
  745. "Neither is the insured bound to give the insurers advice on the legal consequences of the facts disclosed or on other matters"
    Two authorities are cited in support of that proposition: The Bedouin [1894] P 1, 12 and Imperio.
  746. MacGillivray goes on, in the second part of paragraph 17-086, to illustrate these principles:
  747. "Thus, when loss experience on marine business is recorded in triangulated form, the data itself is material and ought to be disclosed, but the insured's own estimates of future losses calculated from the loss statistics need not be disclosed since the insurer is in as good a position to calculate them as is the insured. In such cases the concept of utmost good faith has to be accommodated within the framework of a commercial negotiation. "As I see it", said Waller J in Pan Atlantic Insurance Co v Pine Top Insurance Co, "the negotiation is a commercial one, the broker does not have an obligation to tell the underwriter how to do his job". The same point was made by Davitt P in Kreglinger & Fernau Ltd v Irish National Insurance Co Ltd when he said that "the insured does not have to conduct the insurer's business for him".
  748. The underwriters relied on statements in cases which emphasised the width of the duty of disclosure. Thus, in Pan Atlantic v. Pine Top [1994] 2 Lloyd's Rep. 427 (HL), Lord Mustill at p 438 said:
  749. "… the vice of misrepresentation and non-disclosure is … that a breach of the duty of good faith has led the underwriter to approach the proposal on a false basis."
    And at p.443rhc, commenting on Carter v Boehm:
    "The assured is not to keep anything back which goes to the computation of the "contingent chance", for otherwise there is no "fair presentation", and the underwriter is led to approach the "risk understood to be run" on a false basis ...
    Every fact and circumstance which can possibly influence the mind of any prudent and intelligent insurer, in determining whether he will underwrite the policy at all, or at what premium he will underwrite it, is material."
  750. The underwriters also referred to other authorities in support of the width of the duty which requires disclosure of any fact or circumstance whatsoever that is material to the underwriting decision - CTI v Oceanus [1984] 1 Lloyd's Rep 476, 496-7 (Kerr LJ) and 527 (Stephenson LJ):
  751. "It follows that when ss. 17 to 20 of the Act are read together, one way of formulating the test as to the duty of disclosure and representation to cases such as the present … is simply to ask oneself: "Having regard to all the circumstances known or deemed to be known to the insured and to his broker, and ignoring those which are expressly excepted from the duty of disclosure, was the presentation in summary form to the underwriter a fair and substantially accurate presentation of the risk proposed for insurance, so that a prudent insurer could form a proper judgment - either on the presentation alone or by asking questions if he was sufficiently put on enquiry and wanted to know further details - whether or not to accept the proposal, and, if so, on what terms?"." (Kerr LJ)
    "That duty seems to require full disclosure and full disclosure seems to require disclosure of everything material to the prudent underwriter's estimate of the character and degree of the risk; and how can that be limited to what can affirmatively be found to be a circumstance which would in fact alter a hypothetical insurer's decision? Provided that there is some information which a prudent insurer would obviously want to know, or which a credible expert swears he would want to know, in considering an offer of a risk, that is a material circumstance which the greatest good faith and the rule against concealment require the assured or his agent to disclose, subject to the qualifications which the knowledge and conduct of the insurer or his agent may put upon the assured's duty." (Stephenson LJ)
  752. The underwriters accepted that there was no need to disclose circumstances which are known, or presumed to be known, to the insurer including matters which an insurer in the ordinary course of his business ought to know. This meant, as Moore-Bick J said in Glencore International v Alpina Insurance [2003] EWHC 2792 (Comm) that the "underwriter must also play his part by listening carefully to what is said to him and cannot hold the insured responsible if by failing to do so he does not grasp the full implications of what he has been told". The underwriters also accepted that (see Glencore at paragraph [41]) an insurer writing an open cover will be taken to understand the range of circumstances that may ordinarily arise in the conduct of the business in question. However, they submitted that unusual matters should be disclosed, and they referred in that context to the following underlined passages in Glencore:
  753. "[38] In Cheshire v. Thompson (1918) 20 Com. Cas. 114 …Mr. Justice Bailhache held that the risk of diversion was a particular and unusual risk not comprehended within the ordinary marine and war risks which was not covered by general wording of the kind used in that case unless it had been made clear to the underwriter what was intended. I agree with Mr. Sumption that the decision turned on the construction of the policy rather than the scope of the duty of disclosure, though it may be said to provide a further illustration of the fact that the underwriter will be presumed to have in mind only such matters as would be within the contemplation of one who is familiar with the trade in question."
    [41]… when an insurer is asked to write an open cover in favour of a commodity trader he must be taken to be aware of the whole range of circumstances that may arise in the course of carrying on a business of that kind ... the insured's duty of disclosure, which extends only to matters which are unusual in the sense that they fall outside the contemplation of the reasonable underwriter familiar with the business of oil trading, is correspondingly limited."

    (ii) Misrepresentation

  754. The principles relating to misrepresentation were not the subject of detailed submissions. The main argument concerned statements made or allegedly made to certain underwriters in 2016 that the risk was "as expiring" or "as before". The principal issue here was as follows. Were such statements accurate, on the basis that the July 2015 endorsement was an effective contractual variation binding on all subscribers to the 2015 policy, so that there was indeed no material change in terms when the varied terms were incorporated into the 2016 wording? Or were these statements misleading, at least as far as concerns the following market, because the following market had no knowledge of the 2015 endorsement, and would therefore have understood "as expiring" and similar statements to mean that the renewal terms were materially the same as to the terms written in 2015? In that context, the relevant legal principle is (as stated by Chitty on Contracts, Volume 1 paragraph 7-007):
  755. "In determining whether there has been an express representation, and to what effect, the court has to consider what a reasonable person would have understood from the words used in the context in which they were used".

    (iii) Materiality

  756. For a non-disclosure or misrepresentation to be material, it must influence the judgment of a prudent insurer in fixing the premium or determining whether he will take the risk.
  757. (iv) Inducement

  758. The underwriters referred to the principles summarised in Assicurazioni Generali SpA v. Arab Insurance [2002] EWCA Civ 1642 (CA).
  759. a) The insurer must prove on the balance of probabilities that he was induced to enter the contract by a material non-disclosure or material misrepresentation.
    b) There is no presumption of law that an insurer is induced to enter the contract by a material non-disclosure or misrepresentation, but the facts may be such that it can be inferred, even in the absence of evidence from the underwriter concerned.
    c) The insurer must show that the non-disclosure or misrepresentation was an effective cause of entering the contract, but not the sole effective cause.
  760. Since that decision, the legal position relating to inducement has been further examined and refined, with distinctions drawn between non-disclosure and misrepresentation. I refer to the more recent authority in Section F8 below.
  761. The evidence - overview

  762. There was a considerable volume of factual and expert evidence which was adduced in relation to the issues of non-disclosure and misrepresentation. Much of the focus of this evidence was whether, and the extent to which, underwriters could and should be expected to read the slips which were presented to them. The heart of the non-disclosure debate did not therefore involve a typical dispute as to materiality, where one expert says that a particular item of information would be important to a prudent underwriter and the other expert disagreed. There was really no dispute that, in a broad sense, both the TPC and the NAC, which were contained in the contracts subscribed by all underwriters, were "material" clauses. But, as Ms. Healy correctly submitted, it was difficult to see which clauses in an insurance contract are not material. Since the terms of the policy contain the basis on which the underwriters have written a risk, it follows that almost any clause would be material in the broad sense that it might influence the underwriter's decision as to whether to write the risk or not. Indeed, a non-disclosure or misrepresentation argument can scarcely be resolved without reference to the terms on which the contract has written, since a question which arises is whether (absent the non-disclosure or misrepresentation) the contract would have been written on the terms which were in fact agreed.
  763. The real debate in the present case, as described below, concerned questions such as: the effect of s. 18 (3) (b) of the 1906 Act (circumstances known or presumed known to insurers); the extent to which the non-disclosure arguments related to matters on which the underwriters should form their own view, on the basis of the information provided; and whether the arguments involved propositions rejected in the case-law summarised above (eg decisions which hold that an insured does not have to advise an insurer as to the law). On these and related issues, it was not a question of choosing between the views expressed by different experts. Their evidence was helpful, principally, in explaining the market background, and as revealing the range of different views that can be held.
  764. F5: Purpose/ intention of the TPC

    The parties' arguments

  765. The underwriters argued that an intention on the part of the assured and broker, to introduce into a marine cargo and storage policy, a clause that is intended to (and/or does) give rise to a completely separate line of insurance, for credit risk cover, and which therefore is intended to (and/or does) very considerably increase the potential monetary exposure under the policy, would plainly be of great interest to any prudent underwriter. There could therefore be no fair presentation of the risk absent the underwriters being told that the risk included credit risks as well as physical risks. This was not simply a case of requiring the disclosure of a party's subjective understanding. It concerned the very purpose of the TPC. The purpose or intention behind the TPC, which is a matter of fact, is a material circumstance within the wide definition of that term.
  766. The underwriters submitted that it was no answer to say that the underwriter should read the proffered wording and ask questions about it. There is no restriction in law on the nature or scope of matters which may constitute a material circumstance to be disclosed: this is always a question of fact. Where there is an unusual clause in a policy – unusual because of its particular wording and/or its intended effect – that is a circumstance which should be brought to an underwriter's attention. The TPC was just such an unusual clause. It would be unfair to expect any underwriter, certainly a marine cargo underwriter, to understand that its purpose was to introduce cover for credit risks.
  767. Where, therefore, there is standard policy wording, it is fair to assume that an underwriter will be aware of the nature and import of the contents of the standard wording. But an unusual clause, or a clause intended to have an unusual effect, needs to be brought to his attention so as to ensure that the underwriter can have a proper understanding of the risk. Mr. Parsons relied in that context upon the decision of Bailhache J and the Court of Appeal in Cheshire v Thompson in support of the proposition that, either as a matter of construction or as matter of the duty of disclosure, an insured should disclose an unusual risk; ie something outwith the contemplation of the underwriters.
  768. Where there is a renewal, it cannot be said that an underwriter ought to have known of the circumstance in question without it being drawn to his attention. When policies are renewed, the evidence indicated that the time available for the broker was limited. The underwriter would expect the broker to draw his attention to matters of interest. He would not painstakingly read through the terms line by line. The same applies where an underwriter comes newly onto a risk which is already established in the market.
  769. In his oral closing, Mr. Parsons said that his complaint was not so much that the underwriter should have been told the subjective intention of the brokers or the Bank, or the meaning of the clause. Rather, it was the purpose of the clause, or the risk that the underwriter was being asked to run, which should have been disclosed. The TPC was an unusual risk: not simply the wording but the risk itself. Underwriters would not, particularly upon renewal, be looking out for it. An underwriter could not, or could not necessarily, be expected to ask a question about it; because he may have assumed that it was a basis of valuation clause, or may not have focused on it in the context of a very large policy. It should be borne in mind that the practice in the marine market was for there to be relatively short 'brokes', and that the market was basically doing the same things over and over again. That was why the market, which was built upon mutual trust, relied on the brokers to draw it to their attention. The market would slow down and grind to a halt if underwriters were expected to read lengthy documents in the detail required to understand each clause. The market did not work on the basis that people would sit and read the policies that they were writing. That might happen in an ideal world, but it was not what happened in practice. Mr. Parsons made clear that the duty to disclose the purpose of a clause in the policy did not apply to every clause: it only applied to something unusual and outwith the contemplation of underwriters.
  770. The Bank contended that this (and other) aspects of the non-disclosure case amounted to a case that leading participants in the London insurance market had to be told what terms were contained in the written policy wording presented to them and what those terms meant. That contention was wrong for very many reasons and was, as the Bank said in its opening submissions, "frankly bizarre". There was no obligation to disclose subjective intentions or understandings. Matters of inference or opinion are not material circumstances for disclosure, and the insured is under no duty to offer the insurer advice. The subjective intentions of the Bank, Edge or indeed NRF, as to the meaning of the TPC are matters of opinion or inference. They are irrelevant to the true construction of the policy and not material to the risk being insured. Furthermore, those opinions arose from material which was available to both parties, namely the wording of the policy. Underwriters, in the ordinary course of business, ought to have been familiar with interpreting such material. An insurer is presumed to know its own business and to be able to form its own judgment on the risk as it is presented.
  771. In her oral closing, Ms. Sabben-Clare also relied in the present context on the authorities, considered by Saini J in Higgins (see Section F2 above), which establish that a party who signs a contractual document cannot then say that he had not read its terms. She submitted that an underwriter who chose not read the contract that he signed could not complain, whereas an underwriter who was put off the scent by a positive misrepresentation was in a different position.
  772. Edge submitted that there was no authority in which it had been held or argued that a broker is under a duty to volunteer to an underwriter his, his client's or his client's lawyers' subjective understanding or intention about the meaning and effect of a contractual term. The meaning or effect of a clause in the policy is a paradigm matter which the underwriter is presumed to know and which, in the absence of enquiry, need not be disclosed under s. 18 (3) (b) of the 1906 Act. The meaning and effect of terms in an insurance contract are matters the insurer ought to know in the ordinary course of his business. The insurer is presumed to be able to form his own judgment of the risk as presented, and the proposer is not required to offer the insurer advice. The subjective view of the broker, the client or others, is plainly a matter of opinion. In her written closing, Ms. Healy referred to the case advanced against Edge as involving the proposition that there was a "duty to nanny" and submitted that there was no such duty. (The expression was used in the context of the Bank's claim against Edge, but it was clear that Ms. Healy was also addressing the non-disclosure arguments in that context). In her oral submissions, she said that one of the oddities of the case was that Edge was being accused of not disclosing something that they handed to underwriters.
  773. Discussion

  774. The factual position can be briefly summarised. Neither the Bank nor Edge disclosed, other than via the terms of the TPC itself, what they considered to be the purpose of the clause, or their subjective intention in asking for the clause, or their views as to what the clause meant. None of the underwriters asked any questions about it. The only underwriter who says that he did so is Mr. Beattie, but for the reasons set out in Section E above I have not accepted that evidence. Even if that evidence were to be accepted, however, its effect was that although a question was asked, the broker was unable to provide a satisfactory answer and the matter was then not pursued by Mr. Beattie.
  775. The effect of the evidence of the majority of the subscribing underwriters was that they had not read through the slip policy to which they subscribed, or at least did not recall reading through it, and did not therefore see the TPC. There were exceptions to this. Mr. Gaiger of Navigators said that he had read through the policy, acknowledging that it was incumbent upon him to do so. Mr. Butterworth of Swiss Re, which was subscribing to the risk for the first time, indicated that he would at least have skim-read the policy. The facts relating to three of the subscribing underwriters is not known, because for various reasons they did not give evidence at the trial. I can therefore come to no conclusion as to whether they did or did not see the TPC. For reasons which follow, however, I do not think that – for the purposes of the non-disclosure argument – it matters whether they did so or not.
  776. I do not consider that any of the underwriters can properly allege that the TPC was not disclosed to them. It was, after all, there in the policy to which they subscribed. In the case of Mr. Beattie, he saw or had his attention drawn to it (on the basis of his own evidence) on a number of occasions: initially in July 2015; then on presentation of the renewal quotation slip on 20 January; and then again when it was the subject of discussion on 28/29 January leading to its change of position within the policy documentation. Other underwriters were shown the policy which contained the TPC on two occasions prior to making their contractual commitment in January 2016: they first signed the slip without formally entering their lines, and then subsequently saw it again when they entered their lines. Some underwriters may have seen the slip only once before subscribing. All underwriters were given a copy of the slip at around the time that it was subscribed, and many of the underwriters thereafter carried out post-placement "peer review" procedures into the risk that had been written. None of these peer reviews raised any questions about the TPC, or indeed the NAC. It is a remarkable feature of the case that, despite a large number of underwriters writing this risk, and despite a large number of peer reviews, no-one involved on the defendants' side raised any questions about the two clauses, the TPC or the NAC, which are central to the allegations of non-disclosure.
  777. In my view, whether the case is advanced as a failure to disclose the purpose of the clause, or a failure to disclose the subjective intention of the broker or the Bank, there was no non-disclosure of a material fact. I accept the submissions of the Bank and Edge in this regard, as summarised above.
  778. Section 18 (3)(b) of the Marine Insurance Act 1906 provides that there is no duty to disclose circumstances which are known or presumed to be known to the insurer. The terms of the policy that the underwriter subscribes, by scratching the slip containing the policy terms there set out, are clearly either known or presumed to be known to the insurer. The question of what a particular clause means is indeed, as Edge submitted, a paradigm matter on which the underwriter can and should form his own view. The insured is not in my view required to offer his views as to the effect or meaning of the contractual terms proposed. To do so would require the insured to estimate the risk for the underwriter. To use the colloquial language used in some of the cases, the broker does not have an obligation to tell the underwriter how to do his job, or to conduct the underwriter's business for him. Nor, as Lord Esher said in The Bedouin, is the assured bound to tell the underwriter what the law is. This necessarily extends to telling the underwriter what a particular clause means.
  779. I do not think that the argument for disclosure is improved by characterising the meaning and effect of the clause (which do not have to be disclosed) as the "purpose" of the clause or the insured's purpose in wishing to have the clause included (which, it is argued, does). I regard this as a distinction without a difference. The substance of the argument is that the insured should disclose the contractual effect of the clause, or at least the insured's view of its contractual effect. But in my view these are not material facts for disclosure. The contractual effect of the clause is a matter on which the underwriter should form his own view. The insured's subjective view of the contractual effect of the clause is irrelevant to construction of the clause which the underwriter is asked to agree, whether communicated in the course of pre-contractual negotiations or not. The insured's subjective purpose in wishing to have the clause included is similarly irrelevant to the contractual effect of the written clause which the underwriter is asked to agree and upon which he should form his own view.
  780. In any event, even if there were any validity to the point that the insured's purpose or intention should be disclosed, this is a case where (applying the principle in paragraph 135 (g) of Garnat), the insurers received information which, taken on its own or in conjunction with other information known to them or presumed to be known, would naturally prompt a reasonably careful insurer to make further inquiries. Here, underwriters were presented with what was, on the evidence, an unusual clause. It was not only unusual, but it was lengthy. It did not involve a minor tweak to standard wording, which might perhaps go unnoticed. The clause used concepts, in particular "Default" on a number of occasions, which are unfamiliar in the context of ordinary cargo insurance. It does seem to me that the language of the clause as a whole, and in particular the references to Default, was sufficient to disclose its purpose. But even if that were wrong, a reasonably careful underwriter presented with a lengthy and unfamiliar clause of this kind, would be prompted to make further enquiries if interested in the insured's purpose or intention or understanding in relation to the clause.
  781. This conclusion is in my view borne out by some of the evidence in the case. As discussed in Section E above, I consider that Mr. Beattie's evidence, as to his conversations with Mr. Mullen as to the intended meaning of the TPC, stems from his recognition that he should have paid far greater attention to the TPC and should indeed have asked questions about it. For reasons there given, I regard that evidence as unreliable. But in my view, he has transposed what he should have done into what he says that he actually did.
  782. Mrs. Joyce Webb was the insurers' underwriting expert. She was an impressive witness, with far greater underwriting experience than Mr. Sutherland who was called by Edge and in my view far more impressive than he was. Her evidence, in relation to the "wish list" was that she would have expected Mr. Beattie to have looked through it carefully at the time that it was presented. She agreed that there were all sorts of things in there which would give rise to further questions. This included the TPC. She said that her approach would have been: ""Well, what's all this about? I don't really get this," that's kind of what I would expect him to say, because I didn't get it either". She said that it took her a number of sessions to actually work out how that clause worked. She had to read it a number of times before she understood it.
  783. I do not think that it makes any difference if, as underwriters submitted (and is clearly the case), a particular clause is unusual. I asked Mrs. Webb whether it was part of an underwriter's skill set to be able to look at a document such as the "wish list" and recognise clauses which are unusual and difficult. She agreed:
  784. "Yeah, I think -- I think, to be fair, I think they should be able to -- they should be able to see. But they may not necessarily -- they might be able to see that there's something that's not normal. In which case, it's when they'd said, "Well, go away and find out" or, "Come back and talk to me about it," or ask specifically to understand "what you're trying to get at here". And I think that sort of dialogue -- and in my experience, if you're asking a broker that, he's normally done his homework before he gets there. So he knows what he needs to articulate. But if he doesn't, he'll say, "Well, actually, I'm actually not really sure, but I'll probably need to go away and find out," and he'll come back another day. You haven't got to do it on the day. There's a sense of let's try and work this out and land it and understand it. And … I do agree that underwriters need to read policies, and they need to be able to flag when something doesn't fit in with -- into their normal boxes. But at that point I think they quite rightly say, "Well, what's all this about? Tell me about it," and they want an answer." (emphasis supplied)
  785. One of the issues canvassed in the factual and expert evidence was the extent to which underwriters should read, or can be expected to read, the policies which they subscribed. Perhaps unsurprisingly, there was no consensus. However, the evidence certainly did contain support for what I would, prior to this case, have regarded as an unsurprising proposition; ie that underwriters should read the terms of the contract to which they put their names.
  786. This was certainly the evidence of Edge's underwriting and broking experts, Mr. Sutherland and Mr. Russell. There was also some support for that proposition in the underwriters' evidence and documentation. It was put to Mrs Webb that underwriters do need to read or look at the policy wording to form their judgment as to whether to underwrite it or not. She said that she was "not for one minute suggesting underwriters shouldn't or don't read policies". She said that underwriters who were completely new to a risk had an obligation to go through the slip very thoroughly in order to see "where the coverage was broader than a normal cargo and storage policy", although she said that the level of scrutiny would not be the same if the risk had been in the market for some time and the underwriter was already familiar with it. She said, however, that she would expect an underwriter already familiar with the risk, but who had not written it in the prior year, to at least go through it, looking at the main clauses and headings to see what was there.
  787. As far as renewing underwriters are concerned, she would expect them to look at the premium and claims figures, and then to look on their screens at the slip and any endorsements that attached, and then to ask: "Is it all as expiry, or are there any changes". At that point, if anything had changed, she would expect to be told about it. As she said: the practice was "to say "as expiry" and to be told yes or no". She said that underwriters might skim the policy to have a look and make sure that they remember it from last year. But if they asked the question "Is this an as-expiry slip", and were told that it was subject to some changes in limits and the fraudulent documents clause, they would stop there.
  788. Some of the underwriters were subject to guidelines which required them (eg, in the case of Advent) to consider whether the "insuring conditions" were "standard" and whether the wording was "bespoke for the interest insured". It would be impossible to do this without reading the slip. The Navigators underwriting guidelines required that at renewals, the files should have a similar level of detail as new business files and that "Information should be properly updated and re-analysed, as part of the underwriting process". Mr Gaiger of Navigators agreed that this required him to read the whole of the renewal slip he was being asked to agree. Mr. Butterworth of Swiss Re (one of the new subscribers to the risk) accepted in his evidence the proposition that it was his job and duty to his employer, when committing Swiss Re to a risk for the very first time, to read the terms of the slip carefully and make sure that they accorded with Swiss Re's rules, guidance and appetite.
  789. My conclusion from this evidence is that whilst practices may vary, good practice does require underwriters writing a risk for the first time to read the slip, even if it is lengthy. They do not necessarily need to do this under pressure of time at the box: a copy of the slip can be taken, and read later during a quiet moment. The evidence also indicates most of the insurers had teams, including those doing peer review, and it may be that the task of doing a detailed review of policy wording could be entrusted by the main underwriter to a more junior colleague. The work does not necessarily involve reading every clause in minute detail. Some clauses may be standard market clauses which are very familiar and do not require significant attention. However, non-standard clauses will obviously require more consideration.
  790. I am, however, in no doubt that policy wording, at least when it is subscribed for the first time, does need to be read by underwriters. Indeed, the 2012 Code of Practice concerning Contract Certainty requires, as its first principle:
  791. The insurer and broker (where applicable) must ensure that all terms are clear and unambiguous by the time the offer is made to enter into the contract or the offer is accepted. All terms must be clearly expressed, including any conditions or subjectivities.
    I do not see how an insurer could fulfil this aspect of the Code if it has not taken steps to read the policy wording in order to ensure that all terms are clear and unambiguous.
  792. As far as renewal is concerned, I accept that, as Mrs. Webb's evidence indicates, a short-cut can be taken. Some underwriting policies, as set out above, may require the underwriter to scrutinise the risk again, and this will involve reading the slip again. However, I do not think that an underwriter, who has subscribed on the expiring year, can be criticised for taking a short-cut: ie asking the broker whether the terms are as expiry. But if he does not do that, or does not receive an affirmative response, and does not read the policy, then I do not consider that he is in a position to complain if he does not appreciate that the terms differ from the prior year.
  793. In short, I do not accept the proposition that the insured has a duty to tell the insurer of unusual policy terms, or to explain their purpose or effect, because (whether in the marine market or otherwise) an insurer cannot reasonably be expected to read the terms of the policy that he is subscribing.
  794. Indeed, the proposition that underwriters should not read the slips which they sign would in my view come as a surprise to generations of insurance lawyers. MacGillivray paragraph 35-049 quotes Mr. John Thomas QC (a renowned insurance practitioner when a barrister, and subsequently Lord Chief Justice):
  795. "It has rightly been observed that the genius of the London market is to set out the elements of a complex transaction involving large sums of money in one short document – the slip".
  796. That statement as to the genius of the London market was made in the context of reinsurance, and it is also fair to say that the slip in the present case was a "slip policy": it is therefore longer than many slips, because the parties did not anticipate that a more detailed policy wording would be prepared subsequent to the slip. However, I do not consider that the importance of an underwriter reading the terms of the contract that he is writing diminishes as the length of the slip increases. A cargo slip which simply refers to the ICC 'A' clauses will not require detailed consideration. A slip policy, such as the present, which contains a number of bespoke clauses, will require more attention.
  797. The decision of the Court of Appeal in The Bedouin provides an illustration of the importance, in the context of non-disclosure arguments, of the information that is contained in the slip. (I was not specifically referred to this case; but it is referred to in the relevant paragraph of MacGillivray which was cited in one of the cases to which I was referred by the Bank). The Bedouin involved an allegation of non-disclosure in relation to a slip written at Lloyds. The insurer contended that he was not told, at the time of initialling the slip, that the underlying risk involved freight payable under a time charter which contained a particular "24 hours" clause; ie that freight (which would now be called hire) would cease after 24 hours. The case is authority for the proposition, stated in MacGillivray, that the insured is not bound to give the insurers advice on the legal consequences of the facts disclosed or on other matters. As Gorell Barnes J said at first instance:
  798. "… that what the underwriter complains of as having been concealed is not a fact, but a view of the law, and that can hardly be stated as a matter of concealment".
  799. For present purposes, the important point is that the court looked carefully at the terms of the slip itself in reaching the conclusion that there was no material non-disclosure, drawing conclusions from the way in which the relevant insuring clause in the slip was expressed. As Lord Esher said (at 12):
  800. "It is clear that the duty of those who effected the insurance for the assured was to lay this slip before the underwriters. They said nothing. That induces the question, does the slip conceal a material fact? In other words, is there a material fact which ought to have been disclosed in this slip and which is not? Now, what is a material fact here? It is a material fact for the underwriter to know whether he is insuring against a loss by the twenty-four hours' clause, but it does not follow from that, that is a fact which the assured is bound to tell him, for let us see what this slip does tell him". (Emphasis added)
  801. The Court of Appeal concluded, after considering the wording, that the words of the slip did inform the underwriter that he was being asked to insure chartered freight. Even if the insured's purpose behind a clause contained in a slip is a material fact which should be disclosed, the language of the TPC did inform the underwriters in the present case that the Bank's purpose was to seek cover for their counterparty's "Default".
  802. I was not referred to any prior authority in which a policy had been avoided for non-disclosure of terms which were set out in the written contract subscribed by underwriters, or non-disclosure of the effect or purpose of such terms.
  803. The closest authority was, in my view, supportive of the argument advanced by the Bank and Edge. Iron Trades Mutual v Cia de Seguros Imperio [1991] 1 Re LR 213 involved a claim under a reinsurance contract which had been subscribed by Imperio. There were various grounds of avoidance. One of the complaints concerned an increase in certain deductions from inwards premiums which the reinsured was entitled to make. The relevant account of the reinsured was being written by a company called Solar. Hobhouse J described (at 221) the increase in deductions (which came about on a renewal) as being the "result of what to an outsider was a subtle revision of the wording". The relevant change had never been drawn to the attention of Imperio, even though Solar and their principals were well aware of its significance and had deliberately introduced the change of wording with a view to increasing the deductions. The judge was left in no doubt that there was a lack of good faith on the part of the brokers: there was a deliberate decision by the brokers not to draw the change to the attention of Imperio. The judge said, however:
  804. "That is not the same as saying that there was any misrepresentation. The wording of the slip and the treaty should have been, and may have been well understood by the professional underwriters concerned at Imperio. On the evidence that I have heard the significance to the profitability of a cession on the significance of relatively small changes in the wording of the premium clause are well known in the insurance industry and are matters upon which a professional underwriter should be fully able to appreciate what is involved".
  805. Hobhouse J indicated (at 221 rhc) that although he had criticised the brokers for a lack of good faith, the consequences of that lack of good faith for the contract between the plaintiffs and Imperio remain to be considered. The judge then described the principles relating to non-disclosure, stating (224 lhc):
  806. "The insurer is presumed to know his own business and to be able to form his own judgment on the risk as it is presented to him. The duty relates to facts not opinions. The duty is essentially a duty to make a fair presentation of the risk to the insurer"
  807. He then rejected the relevant allegation of non-disclosure in the following terms (225 rhc):
  808. "As regards the deductions the actual wording of the slip as presented to Imperio should have disclosed to them that the deductions on the new treaty were to be calculated on a new basis. What was not disclosed to them was the amount of the agency commission that would now also be deducted: however this additional commission was in fact no more than they should have anticipated if they had thought about it. A reasonable underwriter sufficiently experienced to decide on behalf of a reinsurer whether to sign this slip would have appreciated that additional deductions were involved and that they would be of the order that they in fact were. He could also make up his mind whether, in the aggregate, they were going to be more than he was prepared to accept. There was evidence that deductions of this size were likely to doom the contract to loss but that was a matter for the underwriting judgment of the reinsurer: he had sufficient material upon which to make that judgment. Thus, although as I have previously indicated there are criticisms to be made of the brokers and they were certainly morally at fault, the presentation sufficed and if, notwithstanding the terms of the slip, Imperio did not fully appreciate what it meant or did not make further inquiry that was a matter for them and they cannot complain (quite apart from the affirmation point to which I have already referred)".
  809. Hobhouse J therefore placed reliance on what the actual wording of the slip "should have disclosed" to an experienced underwriter. Despite the moral fault of the brokers, the slip itself sufficiently conveyed the relevant information and if Imperio did not appreciate what it meant or did not make further enquiry that was a matter for them.
  810. In the present case, I do not consider that there was any "moral fault" on the part of the brokers akin to that in Iron Trades. The TPC was a prominent clause originally presented to Mr. Beattie in July 2015 as part of a three page document: it was not a subtle change that was in any way disguised. As far as concerns Mr. Beattie, it was then brought into focus again, at the end of January 2016, when Edge sought to move the clause from Clause 1.5 into two places in the policy. It was also a lengthy clause. I accept Mr. Parsons' point that there is some distinction, when comparing the facts of the present case to those in Imperio; namely, that a cargo underwriter would not be on the look-out for a clause such as the TPC whereas a reinsuring underwriter would be interested in the levels of deductions. Nevertheless, it seems to me that the heart of Hobhouse J's reasoning is that the relevant information was contained in the slip, which was available to be read by the reinsurer. Furthermore, the nature of the market (as described in Section D above) was that cargo underwriters could not assume that all policies placed in the market went no further than covering physical loss or damage. Add-ons were common, and in my view (for reasons already given) a professional underwriter could be expected to read the slips that he signs even if he would not be expecting to see a clause such as the TPC.
  811. By contrast, I did not consider that Cheshire v Thompson provided any substantial support for the underwriters' non-disclosure case. Indeed, as explained below, the case if anything provides some support for the Bank and Edge.
  812. The headnotes in the reports of the judgments of Bailhache J (1918) 24 Com Cas 114 and the Court of Appeal (1919) 24 Com Cas 198 are in identical terms:
  813. "When an assured has in mind a particular and unusual risk known to himself and unknown to his underwriters he does not cover that risk by general words in a policy which, taken by themselves, are as a mere matter of construction wide enough to cover that risk".
  814. This indicates, if accurate, that the case turned on the construction of the policy rather than the scope of the duty of disclosure. This indeed is how Moore-Bick J viewed Cheshire in Glencore v Alpina: see paragraph [38], and the discussion in Arnould paragraph 16-59. There was some debate before me as to whether this was the right way to view the case.
  815. In Cheshire, the plaintiffs were warehousemen in Liverpool, Birkenhead and Newport who wished to insure their profits for the handling of cargoes of nitrate carried from South America to the UK. At this time, during the war, all cargoes of nitrate were under the control of the UK Government. When the plaintiffs were informed from time to time that a cargo was destined for one of their warehouses, they would insure their profits. In addition to the ordinary marine and war risks which might prevent the nitrate arriving, there was the risk that the Government might change its mind and divert the goods elsewhere. The plaintiffs wished to obtain cover against all of these risks. The plaintiffs were in touch with a local Liverpool broker, who drafted a slip which would have covered precisely the risk that the plaintiffs wished to insure: ie to pay a total loss if the vessel "does not reach the destination named in policy through any cause arising (ie war, marine, diverted to other ports &c.)". Unfortunately, the slip that was prepared in London was drafted in more general terms: to pay a total loss "if vessel does not reach destination named in policy through any cause that may arise". Bailhache J described the words omitted from the slip as "obvious and unfortunate". There was a dispute as to what if anything had been discussed about diversion. The judge was satisfied that diversion due to a change of mind on the part of the Government, which owned the cargo, was not specifically mentioned in these discussions, and that very little on the subject was said at all.
  816. In due course, the plaintiffs made a claim for a total loss after a cargo of nitrate was diverted by the Government from Birkenhead to Savona. Bailhache J said that several points were taken for the defendants (although he did not identify all of the arguments), but he intended to decide the case on one point only. He said:
  817. "The loss in the case was clearly due to that diversion against which the plaintiff desired to be protected and to nothing else. I take it to be a rule of marine insurance law that when an assured desires to insure against an unusual risk unconnected with marine or war risks he must take one of two courses. Either he must procure the insertion in his policy of apt words descriptive of the special risk he wishes to cover, or, if he is content with wide general words, he must be prepared to show that the special risk he was minded to cover was brought to the underwriter's attention at the time when he initialled the slip or subscribed the policy in such a way that the underwriter had his mind directed to it. When an assured has in mind a particular and unusual risk known to himself and unknown to his underwriters he does not cover that risk by general words in a policy which, taken by themselves, are, as a mere matter of construction, wide enough to cover that risk".
  818. This reasoning seems to be squarely based on the construction of the policy rather than the duty of disclosure. Bailhache J identified two possible courses open to an insured who wished to cover an unusual risk. One of those courses was to include apt words descriptive of the risk, rather than to rely on general words. There is no suggestion that the policy would be voidable for non-disclosure even if apt words descriptive of the special risk were included. Bailhache J did contemplate an alternative course; ie using wide general words but informing the underwriter of the special risk that it was the intention to cover. Since this had not happened, Bailhache J did not analyse the legal basis upon which the wide general words would be sufficient, if the special risk had been identified. Nowadays, no doubt, that case would be put on the basis of arguments (rectification/ estoppel/ collateral contract) similar to those which RSA itself advanced based on the conversations between Mr. Beattie and Mr. Mullen.
  819. The significance of the decision for present purposes is, as Ms. Healy submitted, that this is a case where the language of the TPC does contain apt words descriptive of the risk which the Bank wished to cover. This is therefore not a case where the Bank is seeking to rely upon wide general words which, on a literal construction, would cover the risk. If there are apt words descriptive of the special risk, then the insured does not have to go further and spell out to the insurer what those apt words mean. By definition, the apt words are themselves sufficient. The approach of Bailhache J is therefore in my view consistent with both The Bedouin and Imperio in confirming the importance of the language which the slip actually contains.
  820. The case went to the Court of Appeal. The plaintiff's argument is briefly reported, and counsel for the insurer was not called upon. The plaintiff argued that there had been no concealment: all material facts were known to the insurer and were matters of public knowledge. The wide words "through any cause that may arise" therefore provided the cover. The Court of Appeal dismissed the appeal, the leading judgment being given by Bankes LJ. There are passages in the brief judgment which could be read as indicating that the case was decided on the basis of non-disclosure. However, in agreement with Moore-Bick J, I do not consider that the case should be read that way. At the start of his judgment, Bankes LJ said that he had come to the conclusion that Bailhache J was right. And at the conclusion of his judgment, he said that in a "passage at the end of his judgment, Bailhache J stated the law applicable to the case, and with his statement I entirely agree". This is a reference to the passage set out above, where the case was clearly decided on the basis of construction.
  821. Furthermore, the importance of the wording of the slip itself is apparent from the judgment of Bankes LJ:
  822. "The slip that was sent up to Reader [the broker in London] clearly showed that particular risk … Reader had prepared a slip from the slip sent up to him, but unfortunately it omitted words that were very material. On the slip which he presented there was nothing to indicate this particular risk".
    There is in my view nothing in the judgment that suggests that the use of apt words in the slip, identifying the particular risk, are insufficient in the absence of some additional disclosure. Indeed, it is implicit in the above passage, and clear from the court's approval of the judgment of Bailhache J., that the inclusion of apt words is sufficient.
  823. Some reliance was placed by the insurers on the decision in Glencore v Alpina itself. However, Moore-Bick J was not considering an argument which had any similarities to that advanced by the insurers in the present case, to the effect that the insured must disclose the purpose or intention behind an unusual clause. The discussion in paragraphs [33] – [41] of his judgment is directed towards the nature of the business activities of the insured. The judge concluded that an insurer who writes an open cover in favour of a commodity trader must be taken to be aware of the whole range of circumstances that may arise in the course of carrying on business of that kind. In the present case there was nothing unusual in the nature of the business carried on by the Bank. On the contrary, the financing of stock by way of repo transactions was entirely normal.
  824. Accordingly, even leaving aside the effect of the NAC and the arguments as to affirmation, I reject the underwriters' case that there was material non-disclosure concerning the purpose or intention of the TPC.
  825. In these circumstances, I do not need to consider the question of inducement. Had that issue arisen, I do not think that it would have presented any substantial difficulty for underwriters' avoidance case. The Bank's closing submission, in the context of its case against Edge, was that the evidence at trial showed that if the underwriters had been told that the Bank wanted credit risk cover, the cargo underwriters would have said "no" (save perhaps Mr. Beattie). I accept this. Edge did not substantially take issue with that proposition. In her written closing, Ms. Healy said that inducement would depend upon the nature of the non-disclosure or misrepresentation, and that "if Edge really was obliged to draw Underwriters' attention to the TPC and the NAC and say, in effect, 'are you sure you really want to agree this onerous term?', it was possible that in those circumstances the underwriters in question would have refused to agree those terms". However, in view of my conclusions in relation to the NAC, affirmation and the merits of the non-disclosure arguments, I do not need to address the issue of inducement (in the context of non-disclosure) in detail. Nor do I need to resolve whether or not inducement would be satisfied in the case of Generali, where there was no evidence from the underwriter who subscribed the risk, and to whose different position Ms. Healy drew attention in her written closing.
  826. F6: Non-disclosure of the NAC

    The parties' arguments

  827. The underwriters submitted that the presence of the NAC would be a matter of interest for any prudent underwriter. It would have the practical effect of restricting the nature and extent of the disclosure about the risk which the insurer could expect to receive. Again, it was not a clause that the insurers would be on the look-out for or expecting to see, given that it is not a clause in use in the cargo market except for project cargoes.
  828. The Bank argued that, again, the presence of the NAC in the policy wording subscribed by underwriters was a circumstance which was, under s 18 (3) known or presumed to be known to the insurer. If the underwriters did not read the terms which they were binding, that was their failing and not that of the brokers.
  829. The same essential point was made by Edge. There was no authority in which it has been held, or even argued, that an insured/ broker is under a duty to disclose to an underwriter the presence of a contractual term in a wording that is put before him. The mere act of placing the policy before the underwriter is adequate. It must then be a matter the underwriter is presumed to know about – because it is right in front of him or (at the very least) can be learned from a source (ie the slip) available to him.
  830. Discussion

  831. The factual position is that the NAC was originally included in the July 2015 endorsement. Mr. Beattie of the RSA placed his initials, and the RSA stamp, immediately adjacent to that clause. In view of my conclusion that this endorsement was contractually binding on at least RSA, I cannot see how RSA can advance a case of non-disclosure given the evidence that (i) Mr. Beattie did read through the 3 page document, and (ii) he initialled and stamped the document in that position. In his closing submissions, Mr. Parsons accepted that if (as I have concluded in Section E), the wish-list was contractually agreed in July 2015 by Mr. Beattie as part of the three page document where his stamp appeared adjacent to the NAC, then Mr. Beattie could not complain about it being in the policy on renewal in January 2016. As far as RSA was concerned, therefore, any argument concerning non-disclosure of the NAC is not available.
  832. I do not consider that the other underwriters are in any better position to allege non-disclosure of the NAC. The NAC was in the slip which all of them signed. They are therefore presumed to know about it, even if (as some of them alleged) they did not read through the slip. As Christopher Clarke J said in Garnat, a reasonable underwriter is presumed to know matters which he should have known from the facts in his possession or matters which he had the means of learning from the sources available to him.
  833. For reasons already given in Section F5, I do not agree with the proposition that an underwriter, whether subscribing for the first time or on renewal, need not read the slip and for that reason cannot be presumed to know what is in it. An underwriter who specifically asks a relevant question, such as whether the risk is the same as expiry, may be in a stronger position if he is given an affirmative answer and is not told of a clause which has just been introduced. However, that is because he may be able to rely upon a positive misrepresentation.
  834. I reach the same conclusion in so far as the underwriters advanced a case of the non-disclosure of the TPC itself. As Edge pointed out, that case was not – directly at least – pleaded by the underwriters. (In my view, it was made, indirectly, via the next non-disclosure considered in Section F7 below). However, given the presence of the TPC in the signed slips, an allegation of non-disclosure of the TPC fails for the same reasons that the claim in respect of the NAC fails.
  835. F7: Non-disclosure of the 2015 Endorsement

    The parties' arguments

  836. Underwriters argued that if the wish list had been agreed and also was binding on the following underwriters, then that would plainly have been of great interest to any prudent underwriter. It would mean that the expiring terms being renewed included credit risk cover (which none of the followers intended and/or was authorised to write) and also the NAC.
  837. The rationale for this requirement for disclosure was explained in the underwriters' written closing as follows. The TPC and NAC were each individually material in themselves. Their inclusion in an agreed endorsement was a material circumstance for the following market. For those who were followers in 2015 and who continued to follow in 2016, unless the inclusion of these terms was pointed out the underwriter would not be aware of their presence. They had not been told about the wish list when (and if) agreed in July 2015 or subsequently during the 2015 policy year, and therefore needed to be told about the inclusion of its terms including the TPC and NAC in the renewal wording. For those who were new to the risk in 2016, the inclusion of these terms was equally material and ought to have been pointed out to them.
  838. Ms. Sabben-Clare for the Bank submitted that this was not really a complaint about the existence of the prior endorsement. The only reason that the prior endorsement mattered, as could be seen from paragraph 20 of the Re-Re-Amended Defence, was because the TPC and the NAC were now included in the 2016 wording. She submitted that the mere fact that an endorsement had been agreed would not go anywhere. The real complaint was, therefore, non-disclosure of the presence of the TPC and the NAC in the 2016 renewal terms. This case failed for the same reason that the case on non-disclosure of the NAC itself failed. There was no obligation to disclose the terms of the policy. Section 18 (3) (b) applied.
  839. Edge made essentially the same point. Since the text of the July 2015 endorsement was carried through into the policy at renewal, all the underwriters had to do was read the policy to appreciate that the TPC and the NAC were there. Either the underwriters read the policy, and were content with the presence of the TPC and the NAC, or they failed to read or notice those clauses, which was not the fault of Edge and was not a non-disclosure.
  840. Discussion

  841. It is factually correct that there was no disclosure to the following market of the existence of the endorsement which Mr. Beattie had agreed in July 2015 as part of the 2015/2016 policy.
  842. However, I was not persuaded that the existence of terms in the expiring 2015 policy, under which no claim arose, is material to the underwriters' decision as to whether or not to write the 2016 policy. What matters to those underwriters invited to subscribe for the 2016 policy are the terms contained in the 2016 policy. I have already considered, in Section F 6 above, the underwriters' case that there was non-disclosure of the NAC and indeed the TPC. That case fails because there was no non-disclosure of those terms: the underwriters either knew or are presumed to know the terms which they signed and to which they agreed. Given their knowledge or presumed knowledge of the terms for the 2016/2017 year, there can in my view be no rationale for the suggestion that the 2016/7 policy can be avoided because the underwriters did not know that similar terms had been agreed by Mr. Beattie in mid-2015.
  843. F8: As expiry misrepresentation

  844. A case of misrepresentation is advanced by four underwriters: RSA, Navigators, Ark and Advent.
  845. The case of RSA can in my view be readily dismissed in view of my findings in Section E above. The representation relied upon by RSA was that on or around 20 January 2016, and during discussions in relation to the renewal, Mr. Mullen told Mr. Beattie that there was no material change to the policy terms he was presenting other than those shown to Mr. Beattie. Even if a representation in these terms was made, it was not false as far as the RSA was concerned. Mr. Beattie had agreed to the amended terms in July 2015. Those amended terms were then incorporated into the 2016 policy which Mr. Beattie signed.
  846. Furthermore, even if the July 2015 endorsement was not contractually effective, I cannot see how Mr. Beattie can allege that he was induced to write the 2016 policy on the basis of this alleged misrepresentation. This is because Mr. Beattie clearly did know that the TPC formed part of the 2016 renewal: the movement of the relevant clause was discussed with him on 28 and 29 January 2016. He would therefore have appreciated that the 2016 policy was different to the policy originally written for the 2015/2016 year in early 2015 and which (on the assumption that the July 2015 endorsement was not contractually effective) had not included any TPC at all. It would also, in my view, have been obvious to him that if the TPC from the "wish-list" had been included, that other terms from that list would have been included as well.
  847. In these circumstances, this case of misrepresentation to RSA fails. The case of the other three underwriters gives rise to factual issues as to what they were actually told by Edge, how any representations made are to be interpreted, and issues of inducement as well as estoppel considered in Section F10 below.
  848. Navigators

  849. The underwriter for the 2016/2017 policy was Mr. Gaiger, although the risk had been written in the previous year by his more junior colleague Mr. Giles. The renewal risk was broked to Mr. Gaiger by Mr. Mullen on 20 January. The meeting was brief. Mr. Gaiger scratched the quotation slip, and then made some brief notes on the file. On 25 January, having looked at Mr. Gaiger's notes, Mr. Giles put down Navigators' line on a "to be entered" basis, and Navigators' reference was entered two days later. The relevant representation was made during the broke to Mr. Gaiger on 20 January.
  850. Mr. Gaiger's evidence in his witness statement was that as a matter of practice, during a renewal broke, he would always ask whether there had been changes to the terms on expiry. If not, there was no need for him to write anything more than relatively brief notes. In relation to this renewal, his evidence was that the risk that was presented to him was on the basis that, apart from an increase in the oil limit, the terms were as expiry. This was why he wrote a note saying: "All else as before". Mr. Mullen in evidence agreed that he might have said "All as before" or "All as previously agreed".
  851. Mr. Gaiger was an honest witness. I accept his evidence that it was his usual practice to ask whether there had been any changes to the terms on expiry. He was given an affirmative response, as his note records. It may be that, as his evidence indicates, he did not make his note immediately at the box, but a little later. But that does not cast any doubt on what he was told.
  852. How is that statement to be understood, bearing in mind that the legal question is what a reasonable person would have understood from the words used in the context in which they were used? It was made to an underwriter who had no knowledge of the July 2015 endorsement, because it had never been circulated to the following market. Mr. Mullen, who made the statement, would (if he had thought about the point, which he did not) have known that Mr. Gaiger had not seen the 2015 endorsement. If, therefore, Mr. Gaiger had taken steps to familiarise himself with the risk by looking at what was on his system, he would necessarily have been working on the basis of the policy as originally written in early 2015. Mr. Gaiger's evidence was that during the renewal broke, he would have reviewed Mr. Giles' notes from the expiring year on his computer screen. Mr. Mullen would therefore have seen that Mr. Gaiger was consulting the information that was on Navigators' system. It is unsurprising and entirely natural that he would do so: Mr. Mullen gave evidence, in relation to his dealings with Mr. Beattie, that the latter looked at the prior year's documents on his screen during the broke.
  853. In those circumstances, the statement would in context have been reasonably understood to indicate that there were no material changes to the policy as it was when it was last before Navigators; ie the policy as written at the beginning of 2015. The statement as to "all else as before" would not reasonably be understood, in context, as referring to the 2015 contract as varied by an endorsement which had never been provided to the following market.
  854. I did not consider that an answer to this misrepresentation case was provided by Edge's argument that (i) the July 2015 endorsement was binding on the following market, because Mr. Beattie had agreed it on their behalf, and (ii) the statement made to Mr. Gaiger was therefore literally true. Had the endorsement been provided to Navigators, so that both parties were aware that there had been a variation, the context and the reasonable meaning of the representation in context would have been very different. But it had not been provided, and therefore the context of the representation, and what it would reasonably have been understood to mean, was as I have concluded above. I also have no doubt that Mr. Gaiger thought that, apart from the specific matters that were discussed (ie the increase in the oil limit), the terms were materially the same as written in the prior year. His understanding was reasonable and unsurprising. In any event, I have not accepted Edge's argument that the July 2015 endorsement was binding on the following market: see paragraphs 439 – 447 above.
  855. There was, therefore, a misrepresentation when Mr. Gaiger was given an affirmative answer to his question as to whether the terms were as expiry. The representation was false because there had been material amendments to the terms agreed in 2015/2016. It was not alleged that this misrepresentation was fraudulent, and for good reason in my view. Mr. Mullen was proceeding on the basis that the July 2015 endorsement was binding on all underwriters who had subscribed to the 2015/2016 policy, and that this 2016/17 policy was therefore a renewal on the expiring terms except for those matters which he specifically identified.
  856. This raises the question of whether Navigators was induced to write the policy as a result of this misrepresentation.
  857. The applicable test for inducement was not the subject of detailed submissions before me, although the underwriters did refer to the principles in Assicurazioni Generali SpA v Arab Insurance. That case, and the approach to inducement in the context of misrepresentation, was the subject of a penetrating analysis by Christopher Clarke J in Raiffeisen Zentralbank Osterreich AG v Royal Bank of Scotland [2010] EWHC 1392 (Comm) paras [153] – [194]. The approach in Raiffeisen was applied, in the insurance context, by Leggatt J in Involnert at paragraphs [210] – [217]. Involnert is treated as authoritative by MacGillivray at paragraph 16-047. In the context of misrepresentation, Leggatt J says that the critical question is what the insurer would have done in the absence of the representation made. This involves asking whether the representee would still have contracted, on the same terms, if the representation had not been made. The question of whether the representee would still have contracted, on the same terms, if the representation had not been made is not, however, the same as asking what the representee would have done if told the truth. The question of what the representee would have done if told the truth can only be relevant in so far as it bears on the critical question: ie whether the representee would still have contracted (on the same terms) if the representation had not been made.
  858. In so far as there is an enquiry into what the representee would have done if told the truth, it is important to be clear as to what the "truth" is. At paragraphs [192] – [194] of Raiffeisen, Christopher Clarke J discussed this issue and accepted the submission that:
  859. "A relevant inquiry is as to what the representee would have done if he had been given sufficient information to correct the falsity of what had been said. Any other question would not relate to the falsity of the representation but to what the representee would have done if he was given further information (of uncertain extent) beyond that necessary to ensure that there was no misrepresentation. That would involve asking what the representee would have done if he had been given a representation different to the one which he was actually given."
    Similarly, in earlier paragraphs of his judgment, Christopher Clarke J referred to the representee being told no more than is necessary to ensure that he was not told an untruth: see paragraphs [177] and [179].
  860. In its written closing, Edge submitted that even if a statement "as expiring" or similar had been made, there was no evidential basis for concluding that, in a counterfactual world where no such statement was made, those underwriters would have noticed the TPC and NAC and would therefore have refused to renew the risk with those clauses. In her oral closing, Ms. Healy drew attention to the evidence of Mr. Gaiger which indicated that he was obliged to read the policy, had read it with care, and that he was obliged to do so on a renewal as well as for a risk written for the first time.
  861. In their written closing, inducement was dealt with by the underwriters briefly and in a composite way. It was submitted that there was evidence of inducement from each of the Defendants, and in all but 3 cases the actual underwriter had given evidence. It was submitted that none of the underwriters concerned underwrote credit risks, and none of them believed that they were writing credit risks. It was therefore plausible, and likely, that they would not have agreed the TPC had it been shown and explained to them. Furthermore, non-avoidance clauses are not a feature of the cargo market, save for project cargo. It was entirely plausible and likely that none of the underwriters would have agreed to the NAC. It was also submitted that inducement was not challenged in cross-examination.
  862. The issue which I am presently considering is whether Navigators' subscription to the 2016 policy was induced by the specific misrepresentation on which they have relied. The decision in Involnert shows that it may be important not to roll up questions of inducement by misrepresentation with inducement by non-disclosure, although there may be some connection between them. The importance of focusing on the specific misrepresentation is emphasised here because I have not accepted the underwriters' case on non-disclosure.
  863. I am not satisfied on the evidence that, on the balance of probabilities, Navigators was induced by this representation. Indeed, I consider it more likely than not that Navigators would still have contracted, on the same terms, if the representation had not been made. There are three reasons for reaching these conclusions.
  864. First, it is a significant and perhaps unusual feature of the present case that it is possible to see how a very large group of underwriters approached the risk in circumstances where no "as expiry" representation was made. Leaving aside RSA, there were 13 underwriters who wrote the risk in 2016. The majority of those underwriters (10) do not allege that an "as expiry" representation was made to them. Of those underwriters, 6 were renewing the policy and some of the others, who had not subscribed on the 2015 year, said that they had some familiarity with the risk. The 2016 policy was oversubscribed. All of this would suggest that, looking at the matter generally, the policy was attractive to underwriters even in the absence of an "as expiry" representation.
  865. This is not a determinative point against Navigators on inducement. Those other underwriters did not ask the question that Mr. Gaiger did, as part of his standard practice. However, the general position does suggest that it is far from obvious that Navigators or others would not have written the policy, on the same terms, in the counterfactual world of "no representation".
  866. Secondly, I consider that Mr. Gaiger's evidence as to his approach to the renewal is significant. Mr. Gaiger was shown Navigators' underwriting manual, which said that renewal files should have a similar level of detail as new business files, and that information should be properly updated and reanalysed as part of the underwriting process. He agreed when it was put to him that this meant that, on renewal, he did not simply ask what had changed: he looked at the risk overall. He agreed that he needed to read the whole of the slip that he was being asked to subscribe. He said that he would read wordings put in front of him with care, and that he wouldn't scratch something, particularly something unfamiliar, unless he had read it with care. He said that he would have read through the slip to the end. He said that he would not necessarily have noticed at the time that the TPC was not a standard form of wording. He thought that he would have regarded it as part of the valuation of the subject-matter insured. When asked about the NAC, he said that he did read through the policy, but "obviously at the time" this did not draw his attention. He also said that if they were told that the risk was "exactly as expiry, we have to take the word of the brokers at that point and then we are happy to renew if we are happy with the results we have seen". He accepted, however, that to approach the risk in this way "maybe" would not comply with Navigators' underwriting manual.
  867. It did not seem to me that there was anything in this evidence which suggested that Mr. Gaiger would not have written the risk if the representation had not been made. Indeed, Mr. Gaiger's evidence taken as a whole was to the effect that he had read through the policy, was content with what was there, and was more than happy to renew the risk in the light of the good loss experience.
  868. Thirdly, I have reviewed in full the evidence of both Mr. Gaiger and Mr. Giles in their witness statements and oral evidence. I do not consider that their statements contain clear evidence to the effect that the risk would not have been written, on the same terms, if this particular representation had not been made. The focus of those witness statements is, in my view, very much on the non-disclosure case, and how the underwriters would have reacted if certain matters had been pointed out to them. For example, paragraph 16 of Mr. Gaiger's statement says:
  869. "If I had been broked the renewal terms for the Policy on the basis the cover was to be extended to cover financial losses or credit risks that were not contingent upon physical loss or damage to the insured cargo, then I would not have agreed to renew the Policy. I did not have the authority or licence to write that type of business".
  870. In the context of misrepresentation, however, the question is not how Mr. Gaiger would have reacted if the policy had been broked as described in paragraph 16 of his statement. The statements did not explain to me how the underwriters would have reacted if they had not been told that the risk was as before. I can see that, in theory, an underwriter might say: "if I had not been told this, I would have made sure that I reviewed the policy and asked what changes there had been, and unless satisfied with each and every change, I would not have written the policy". However, this is not what is said in the statements, and it does not emerge from a fair reading of their evidence as a whole. Furthermore, Mr. Gaiger's evidence was that he did indeed read through the policy, and that his underwriting guidelines indicated that he should do so.
  871. I did not think that there was any substance to the argument that inducement had not been challenged in cross-examination. Ms. Healy properly explored the underwriting approach of each underwriter, including Mr. Gaiger and Mr. Giles, and was entitled to make her submission that there was no evidential basis for concluding that, in a counterfactual world where no such statement was made, underwriters to whom the "no expiry" representation was made would have noticed the TPC and NAC and would therefore have refused to renew the risk with those clauses. In the case of Navigators, where Mr. Gaiger's evidence is that he did read through the policy and would have seen the two important clauses and did not raise any questions about either of them, I agree that there is no proper evidential basis for the finding on inducement which Navigators need.
  872. Ark

  873. The underwriter at Ark was Mr. James Blewett. He was a straightforward witness, and I accept his evidence that his practice was to ask if there had been any changes to the policy wording, and that he would note any material changes to the policy from the expiring year. He described himself as an ardent note taker. His note made following the renewal discussion records certain changes (increase in oil, change to the fraudulent documents sub-limit) but then says: "All rates terms and conditions as expiry". I am satisfied that this was based on what he was told either by Mr. Mullen or Mr. Lockyer. Whatever the precise words used (as to which there was some debate), I am satisfied that their substance was that the only material changes to the prior year's policy were those which had been specifically discussed, and that this is what Mr. Blewett believed.
  874. Given that the context was that Ark had not been told of or given the July 2015 endorsement, this was a misrepresentation for reasons similar to those discussed in the context of Navigators.
  875. The next question is inducement. The position of Mr. Blewett is somewhat different to Mr. Gaiger of Navigators, since Mr. Blewett did not read through the policy. I do not consider that he can be criticised in that respect, because he was a following underwriter and was told that the policy was (apart from the changes noted) as expiry.
  876. I have again looked closely at Mr. Blewett's witness statements, and reviewed his oral evidence, in order to see whether there is a satisfactory evidential foundation for the underwriters' case on the key question of whether, if the representation had not been made, Ark would not have written the policy on the terms that they did. I consider that there is. Mr. Blewett, as an ardent note-taker, was concerned to note the changes to the policy. If he had been told no more than was necessary to ensure that there was no misrepresentation, then he would have been informed that there had been a number of changes to the policy which had been agreed during the course of the previous year by the slip leader, Mr. Beattie. This would have led to a discussion as to what those changes were, and the TPC would have been identified in the context of such a discussion. Mr. Blewett's evidence was that he would have asked why the Bank wanted that clause included, and that he would not have agreed to the wording if he had been shown the terms and the reasons for their inclusion explained. I consider that Mr. Blewett's evidence, when considered as a whole, is sufficient to establish inducement.
  877. I also consider that Ark's case on inducement is supported by the fact that the policy was subject to peer review, after subscription, by two underwriters at Ark. Mr. Blewett agreed in cross-examination that it looked as though the peer review was very thorough. The review was carried out in the context of Mr. Blewett's notes, which identified the changes but which indicated that otherwise the terms were as expiry. The reviewers commented on the CEND cover and its width, and this prompted Mr. Blewett to ask some questions to Mr. Mullen. The reviewers also raised a question as to the removal of the sub-limit on the fraudulent documentation cover. No questions were raised, however, about the TPC or the NAC. Mr. Blewett believed that if the reviewers had thought that the TPC extended beyond physical loss and damage, that would have been picked up; since this would be one of the things which they would be checking. In the counterfactual world where (i) these particular changes to the prior year's policy had been identified to Mr. Blewett following his questions to the broker, and (ii) the peer review was carried out in the context of Mr. Blewett's notes which identified all the changes, I consider it likely that the peer reviewers would have picked up on the point that the TPC did extend beyond physical loss and damage. Even if Mr. Blewett had previously missed that point, it would have been brought to his attention and this would in turn have led to Mr. Blewett indicating that he was no longer happy to write the risk.
  878. Advent

  879. The underwriter for Advent was Gary Cooke. I thought that he was a good and fair-minded witness, and indeed some of his evidence as to the nature of the market in 2015/2016 was relied upon by the Bank/ Edge in support of their case.
  880. Mr. Cooke's notes record that he was told about the increase in the oil limit, and also: "As expiry but brokerage up from 22.5%". The evidence in his witness statement was that his standard practice for a renewal slip was to ask the broker about any changes to the policy wording, and then to make a note of the changes. Where he notes "as expiry", this means the renewal terms were broked as being on the same terms as the expiring policy year.
  881. Again, I accept that evidence, which was not challenged in cross-examination. Somewhat belatedly in the case, Mr. Lockyer (who broked this risk to Mr. Cooke) said that his standard practice was to say that the slip incorporated endorsements agreed during the expiring year. This evidence was not contained in Mr. Lockyer's written statements, and it is not reflected in any of the contemporaneous notes made by various underwriters of the renewal meetings. I did not consider that I could properly conclude that such a statement was made by Mr. Lockyer to Mr. Cooke.
  882. There was, again, a misrepresentation when this "as expiry" representation is considered in its context. As far as inducement is concerned, Mr. Cooke proceeded, reasonably in my view, on the basis of what he had been told as to the risk being as expiry. His evidence was that he would always ask about any changes to the policy wording, and make a note of the changes. He would therefore have noted the inclusion of the TPC, had he been told that this was one of the changes. His evidence in his witness statement was that if the TPC had been shown to him and the reasons for its inclusion explained, he would not have agreed it.
  883. In his oral evidence, Mr. Cooke said that he had probably looked through the policy the year before, when he had first written the risk. But if the broker says that it is "as expiry", you do not expect any changes: there was "a lot of trust in our business". He said, and I accept, that there was no reason for him to carry out a review of the policy on renewal, in the light of what he had been told. He said in evidence that, in his experience, there was a difference between the underwriting approach when the risk is first written, when an underwriter would "go into it in some depth", and the position on renewal where he would not do so.
  884. I thought that Mr. Cooke's position on inducement was not as strong as Mr. Blewett's. Although there was a peer review by Advent, it was not of the depth and thoroughness of Ark. Also, Mr. Cooke's witness statement did not state, as clearly as Mr. Blewett's, that he would have asked questions about the TPC. Ultimately, however, I consider that the evidence is sufficient to establish inducement in the case of Mr. Cooke and Advent. On a fair reading of his evidence as a whole, I conclude that had he been told that there had been changes to the policy since the prior year, he would have noted those changes and it is more likely than not that a discussion would then have ensued about the TPC. On the balance of probabilities, this would have led to Mr. Cooke declining to write the risk. Accordingly, Advent succeeds on the issue of inducement.
  885. F9: Only PLOD representation to Standard

  886. Standard was new to the risk in 2016. It was written by Mr. Nick Holding. His contemporaneous note includes the following: "This policy covers ABN AMRO for loss/damage to products where finance provided". Mr. Holding, based on those notes, said in his witness statement: "Thus, I believe that I was specifically told that the Policy covered the Bank for loss and damage to products where finance was provided". This statement is relied upon as a misrepresentation, on the basis that that policy was not confined to cover for loss and damage to products.
  887. If Mr. Holding had read through the slip, then it would have been apparent to him (even leaving aside the TPC) that there were add-ons to the cover, which were common in the market, such as the BCC and CEND cover, which went beyond physical loss and damage. The extent to which Mr. Holding read the slip before he signed it is unclear. Unsurprisingly, he had no real recollection of what he had read. He said that he would have read what he felt to be the important points, but this would be what it was possible to read during a broke. It was, he said, not possible to read a 30-page plus document from top to bottom and assimilate all the information in it. He said that he recalled very little of the broke but that the notes evidence clearly the points that were discussed with Mr. Mullen. He could not remember whether he had seen both Mr. Mullen and Mr. Lockyer.
  888. Mr. Mullen was shown the above note, and said that he did not recall saying anything like that, and would not have said it.
  889. Edge submitted that if this had been said, it would have been a truthful statement. The policy did cover loss and damage to products. Mr. Holding was not told that the policy only provided PLOD cover. If he had read the slip with any degree of care he would have seen that it contained add-ons. The underwriters relied upon the terms of the clear note, and submitted that this was the best evidence of what was said. They said that the clear impression given to Mr. Holding was that the policy covered physical loss and damage only, and not any other interest.
  890. The document containing Mr. Holding's notes appears in a standard form headed "Cargo – New Risk Register". Mr. Holding completed, neatly, all sections of the form. It seems likely that it was completed after the risk had been written rather than as Mr. Mullen was speaking: there is a box at the top which has a tick in the box "Written/bound". The printed form has a space for "Risk Profile/Occupancy", against which Mr. Holding wrote:
  891. "ABN AMRO through ICESTAR (investment vehicle) provides capital to commodity traders on a shared equity basis. This policy covers ABN AMRO for loss/damage to products where finance provided".
  892. Mr. Holding was clearly an honest witness. I did not, however, think that it was possible to conclude, on the balance of probabilities, that the second sentence in the above passage reflected statements made by Mr. Mullen, as distinct from Mr. Holding's own short summary of the risk after whatever review he had carried out. Mr. Holding accepted, unsurprisingly, that he remembered very little about the broke.
  893. Even if, however, the second sentence does record what Mr. Mullen told Mr. Holding, I do not accept that the statement so made would reasonably have been understood as meaning that the policy exclusively covered loss/ damage to products. Add-ons which did not require PLOD were common in the market, and this would have been known by Mr. Holding. The form completed by Mr. Holding refers to one such add-on. In the bottom section, under "Information/ Instructions/ Notes", Mr. Holding wrote: "Expropriation/Confiscation included". When asked in evidence about the CEND clause in the policy, Mr. Holding agreed that the clause could potentially provide cover without physical loss or damage to the cargo. His evidence indicated that he had some familiarity with CEND covers in the market.
  894. In these circumstances, I do not accept that there was any misrepresentation.
  895. In any event, I am not satisfied that, applying the relevant test, any misrepresentation induced the policy. If this statement had not been made, Mr. Holding would still have subscribed to the policy. He was, according to Mr. Mullen, keen to participate on the risk. All of the other underwriters in the case subscribed to this risk (and indeed it was oversubscribed) without any equivalent statement being made to them. Mr. Holding's evidence on inducement, as with the evidence of the underwriters on the "as expiry" point, is directed towards the case of non-disclosure which I have rejected. It does not in my view provide a sufficient evidential foundation for a conclusion that, if the representation had not been made, Mr. Holding would not have subscribed to this policy in the way that he did.
  896. F10: The misrepresentations as the foundation for an estoppel

  897. Paragraph 29 of the underwriters' defence pleaded as follows:
  898. "Further or alternatively, the Bank is prevented by estoppel and/or estoppel by convention from asserting:
    a. against Navigators and/or Ark and/or Advent, that the 2016 Policy was on terms (apart from the limit for oil trading and brokerage) that differed to the terms of the 2015 Policy; and/or
    b. against Standard, that the 2016 Policy provided cover beyond cover for loss and/or damage to goods; and
    and therefore from asserting that the 2016 Policy included cover in respect of credit risks and/or financial defaults.
  899. Edge's opening submissions referred to this (and other paragraphs of the underwriters' defence). Edge submitted that the relevant pleas did not contain all the necessary elements of the estoppels relied upon. Even proceeding on the basis that the alleged representations were made (which Edge denied), the plea of estoppel by representation did not set out how the relevant underwriters are said to have changed their position to their detriment in reliance on those representations, or why it would have been reasonable for them to do so. Edge also submitted that the plea of estoppel by convention did not set out how a clear and unequivocal assumed state of facts or law was shared by the parties, nor is there any basis for a finding of inequity in holding Underwriters to the bargain that was in fact struck.
  900. The case for estoppel, based upon these particular statements, was not specifically addressed in the parties' oral openings. It was then to some extent thereafter lost from sight, with the focus of the argument being upon avoidance. However, Mr. Parsons made it clear in his oral closing that he maintained the plea in paragraph 29. It was, he said, the same principle of estoppel as arose in the context of the Beattie/ Mullen discussions. He submitted that if you tell someone that the document being signed is "as expiry" or "as before", and their attention is not drawn to changes such as the TPC or the NAC, then "you would be estopped from relying on those clauses".
  901. Notwithstanding the comparative brevity with which this issue was addressed in the parties' arguments, I must resolve it as best I can on the basis of the evidence adduced at trial. It does seem to me to raise a substantial point as far as Navigators, Ark and Advent are concerned, where I have accepted that inaccurate representations were made. (The position is different with Standard, where I have rejected that case and therefore there is no potential foundation for an estoppel). The potential advantage of the argument from the perspective of these underwriters is that it potentially circumvents the difficulties in their avoidance case, and in particular the effect of the NAC and affirmation.
  902. The case advanced by underwriters was put forward as giving rise to an estoppel by convention: that was the only 'species' of estoppel discussed in their closing submissions. The relevant legal principles are set out in Section E above. An estoppel by convention can arise if (i) there is a relevant assumption of fact or law, either shared by both parties, or made by party B and acquiesced in by party A, and (ii) it would be unjust to allow party A to go back on that assumption.
  903. Here, the evidence shows that Mr. Gaiger of Navigators, Mr. Blewett of Ark and Mr. Cooke of Advent all made the assumption that the terms which they were considering were "as expiry" or "as before" in the sense discussed in Section F8 above, except for certain matters which had been specifically identified to them. That assumption was acquiesced in by Edge, and indeed was the result of positive statements made by Edge during the 2016 renewal broke.
  904. The question is therefore whether it would be unjust to allow the Bank, whose broker acquiesced in the assumption and was responsible for making the representation which induced it, to go back on it.
  905. In relation to Navigators, I consider that the answer is no: it would not be unjust. Mr. Gaiger, to his credit, did carry out a review of the policy as a whole, and read it with care. It may be that he did this because Navigators was the first Lloyd's syndicate on the slip, and was to some extent a leader itself, or because Navigators' underwriting guidelines required a reappraisal of the risk on renewal. But whatever the reason for doing so, the fact is that he did carry out the review, and in my view this should have enabled Mr. Gaiger to ask any questions about the TPC, which was a lengthy and unfamiliar clause (and which Mr. Gaiger believed that he read) or indeed about the NAC. No questions were asked, even though (see Section F5 above) the disclosure made (ie the wording of the slip policy) would naturally prompt a reasonably careful insurer to make further enquiries. In these circumstances, I do not think that it is unjust or unconscionable for the Bank to be able to rely upon the actual terms of the policy which Mr. Gaiger read through, notwithstanding the prior representation that the terms were "all else as before".
  906. It also seems to me that the question of injustice should also be considered in the light of my conclusion on inducement. I accept that the issues on inducement consequent upon a misrepresentation and estoppel by convention are not quite the same. The former depends upon whether the policy would still have been written, on the same terms, if the representation had not been made. The latter depends upon whether the requirements set out in paragraph 696 are satisfied. However, where an underwriter has sought to rely upon avoidance based on the same representation that is alleged to give rise to an estoppel, and in that context has failed to establish inducement, it would be surprising to reach the conclusion that there was sufficient injustice to give rise to an estoppel by convention.
  907. I reach a different conclusion in relation to both Ark and Advent.
  908. As far as Ark is concerned, Mr. Blewett's practice was always to ask if there had been any changes. He was (as I have said) an ardent note-taker, and his practice was to "note any material changes to the policy from the expiring year". The consequence of Mr. Blewett not being told that there had been material changes, in response to his question, was that he did not make a note of the changes and therefore did not ask any questions about them. For reasons discussed in the context of inducement, I am satisfied that if Mr. Blewett had received a different answer to his question (the minimum necessary to ensure that there was no misrepresentation), this would have led to Mr. Blewett declining the risk.
  909. In these circumstances, I consider that it would not be just to allow the Bank to go back on the assumption on which Mr. Blewett proceeded.
  910. I reach a similar conclusion in relation to Advent. I consider that the findings that I have made in the context of inducement are sufficient to mean that it would not be just to allow the Bank to go back on the assumption on which Mr. Cooke proceeded.
  911. For these reasons, in the case of Ark and Advent, I accept the plea in paragraph 29 that the Bank is estopped from asserting that the 2016 policy included cover in respect of credit risks and/or financial defaults.
  912. G: Clause 3 and the lack of quality checks carried out by the Bank


     

    G1: The issues

  913. The present issue concerns the Bank's conduct at the time when it entered into the repo transactions with Euromar and Transmar, and the impact of Clause 3 of Section 2 of the policy. That clause provides:
  914. The Insured shall do (to the extent it reasonably can do) all things reasonably practicable to prevent any claim being made under this contract, provided always that following the occurrence of a peril in relation to the subject matter insured, the Insured may in its sole discretion elect an appropriate course of action, as it considers appropriate in any particular circumstance, subject to the Insured acting in good faith with the intention of minimising any ultimate potential net loss (save that the Insured shall not be required to exercise any put option following the occurrence of any such peril). (Emphasis supplied)
  915. The underlined wording is the relevant part of the clause in the present context, because (as was common ground) the Bank's conduct in issue occurred prior to the occurrence of the insured peril.
  916. The underwriters' arguments

  917. The underwriters contend that the Bank was in breach of Clause 3 of the policy because of its failure to take any steps, when entering into the repo transactions, to establish or verify the quality or condition or market value of the products which were the subject of those transactions. They also contend in their pleadings that the Bank's failure – to have any independent system in place and/or operating to verify quality on purchase – amounted to conducting its business in a reckless manner and/or refraining to take precautions which it knew it ought to take. This case was introduced by way of amendment in July 2020. The pleaded case did not contain details of the steps that should have been taken, or the independent system that should have been in place. The case advanced at trial focused on the need for the Bank to arrange for physical inspections of the relevant goods in order to verify their quality, or at least to obtain quality certificates. The latter would then be subject to consideration by the Bank, assisted if necessary by an independent expert who could provide advice as to the quality of the products concerned.
  918. The underwriters contend that it was reasonably practicable for the Bank to verify the quality and age of the products it intended to buy as collateral, before agreeing the purchase price, by way of independent inspection and/or by reviewing any third-party quality certificates. The Bank knew that verifying quality/age was a step which would lower the likelihood of a quality deficiency risk materialising, and that it would be at risk of loss, and therefore a claim arising under the TPC, if there were a quality deficiency. Notwithstanding that, the Bank deliberately did not verify either quality or age, because it did not care whether it overpaid, wrongfully believing that the collateral was irrelevant as their counterparty would pay the pre-agreed price or presumably that underwriters would pay.
  919. The underwriters submitted that Clause 3 did not require reckless conduct. In the context of the TPC, ordinary negligence would suffice. If a higher recklessness standard was applicable, the Bank was indeed reckless in the approach that it took. In particular, it would have been reasonably practical for the Bank to obtain an independent report on quality and market value. The Bank was aware that there was a risk of a mismatch between the assumed quality and true quality, since this risk was considered in both 2014 and 2015 in the context of discussions about the Bank's insurance arrangements. Thus, knowing of the risk, the Bank deliberately omitted to verify quality at purchase. Reliance on insurance as a reason not to care about quality amounted to a complete lack of care as to whether the risk was averted or not.
  920. The underwriters submitted that the Bank's losses in the present case flowed from the breach of Clause 3. The real problem which led to the loss was not the default of the Euromar and Transmar, or at least not only that default. If the collateral had been of reasonable quality, then the Bank would not have suffered a loss at all, particularly bearing in mind that the market was rising at least up until August and September 2016. The Bank's financing did not involve its paying the full market price for the goods which were the subject of the transactions: in colloquial terms, Transmar and Euromar took a "haircut", because the finance was only a proportion of their value. Given that the market was rising, the Bank should have been in a position, if the goods were sound, to have sold them for a reasonable profit. The losses therefore all flowed from the fact that the goods were not sound. They were in many cases old and of poor quality. The Bank had therefore overpaid for them.
  921. Accordingly, the underwriters pleaded that at the time of entering into the deals extant as at 17 August 2016, Icestar suffered an effective loss of about £ 36.2m because it overpaid by that amount as against the market price for the quality of the products purchased. Had the true market price been assigned to the product given its condition at purchase, and the appropriate corresponding purchase price paid by Icestar, the later defaults of Euromar and Transmar would have resulted in it making an additional profit of about £ 2.4 million from their actual sales as against the pre-agreed price, plus some hedging gains.
  922. The Bank's arguments

  923. The Bank accepted that it had not carried out quality checks when entering into the repo transactions. Its response to underwriters' case involved arguments concerning the construction of Clause 3, the nature of the Bank's conduct, the proximate cause of the Bank's losses and a more general question as to whether, even if relevant steps had been taken, the loss would still have been suffered. In summary, the Bank submitted as follows.
  924. First, Section 2, Clause 3 is a modified form of the post event duty to "sue and labour". It does not impose obligations upon the Bank prior to the occurrence of circumstances which may give rise to a claim (ie when an insured peril has struck or is imminent). The Bank's obligation to take reasonable precautions was therefore not engaged at the time when it purchased the goods under the "repo" transactions.
  925. Secondly, the Bank's alleged "failure" to undertake quality checks was not (and cannot be dressed up as) a failure "to prevent any claim being made" on the Policy. The actual quality of the cargo had no bearing on the likelihood of an insured peril (ie a counterparty's default) occurring and acting on the cargo.
  926. Thirdly, the contention that Icestar negligently breached Clause 3 is bad in law: it is necessary for Underwriters to show that Icestar acted recklessly, in the sense that it knowingly courted a danger. Moreover the recklessness must be committed by a person or persons of such seniority that their individual recklessness will be attributed to Icestar (the insured entity) itself. In the present case, the only individual of requisite seniority was Mr. Stroink, not Ms. Franssen and certainly not anyone below their levels of seniority.
  927. Fourth, the defaults of Transmar and Euromar were on any view a proximate cause of the loss. That is sufficient to entitle the Bank to an indemnity under the policy. Underwriters' case could only succeed if the Bank's conduct was the sole proximate cause of the loss, which it plainly was not.
  928. Fifth, in any event, Underwriters' allegations of negligence and recklessness are wrong. They are based on a fundamental misunderstanding of Icestar's business. Icestar was not a commodities trader selling cargo on to third parties and so exposed to the quality of the goods. It was providing trade finance to Transmar and Euromar to enable them to finance goods for their own use. The client represented and warranted that the goods were free of defects and in any event agreed to re-purchase the goods "as is, where is and with all faults". In structuring and managing its transactions, Icestar relied upon the Bank's careful and continuing assessment of the credit risk presented by Transmar and Euromar. It believed them to be reputable clients, using these cargoes in their business activities, which included trading, processing, and manufacturing. They had a long and unblemished history of fully compliant behaviour. In all the circumstances Icestar had no need to perform quality checks and its failure to do so was not even negligent—let alone reckless.
  929. Sixth, Underwriters cannot discharge their burden of showing that Icestar would have avoided loss if it had required Transmar to provide quality certificates at the time when the repo deals were entered into. The underwriters' case in this regard is speculative, and against the balance of probabilities. Given what is now known about the modus operandi of the Transmar and Euromar fraudulent schemes, for which their principals received prison sentences in the United States, it is likely that fake quality certificates would have been provided.
  930. G2: The construction of the policy

  931. I do not accept the Bank's argument that Clause 3 is only applicable when an insured peril had struck or was imminent. The natural and ordinary meaning of the opening words is, as the underwriters submitted: "the duty is generally this, provided that in the following particular situation the duty is this". The obligation in Clause 3 therefore applied throughout the policy term.
  932. This conclusion is reinforced by the fact that, as the underwriters submitted, the Bank's construction of the opening words would render them surplus, in the light of the remaining words of the clause and the sue and labour duty which is derived from clause 16 of the ICC (A) Clauses and section 78 (4) of the Marine Insurance Act 1906. Clause 16 and s 78 (4) are discussed further in Section H below.
  933. I also agree with the underwriters that the location of Clause 3 within Section 2 does not justify a construction which confines the opening words of Clause 3, contrary to their natural and ordinary meaning, to the period when the insured peril has struck or is imminent. It is true, as the Bank pointed out, that Clause 3 comes between other clauses which deal with notice of a claim (Clause 2), Payment of claims (Clause 4), and Subrogation (Clause 5). However, as the underwriters submitted, Section 2 of the policy contains many clauses which do not relate to the claim process.
  934. Furthermore, it did not seem to me that Clause 3 could properly be seen as part of a logical sequence beginning with Clause 2. Clause 3 requires, even on the Bank's case, conduct when the peril has struck or is imminent. Clause 2 requires notification of circumstances which may give rise to a claim as soon as practicable, but no more than 90 days after knowledge of the circumstances. I think it likely that the conduct required under Clause 3, even if confined to the period when the peril has struck or is imminent, will usually occur prior to the obligation to notify circumstances under Clause 2. Indeed, if successful it will avoid the need for any notification under Clause 2. Accordingly, the steps taken under Clause 2 do not, or do not necessarily, precede those taken under Clause 3.
  935. Nor do I accept the Bank's argument that the requirement to do all things reasonably practicable to prevent any claim being made should be confined to taking reasonable precautions to prevent a default occurring. The Bank's proposition is that the clause should be construed by reference only to the insured peril consisting of the default. On this basis, the actual quality of the cargo had no bearing on the likelihood of an insured peril occurring and acting upon the cargo.
  936. I do not see why Clause 3 should be construed in this limited way. The present argument requires consideration to be given as to how Clause 3 operates in the context of a possible claim under the TPC. As discussed in Section D above, the TPC provides coverage for the "Transaction Premium that the Insured would otherwise have received and/or earned in the absence of a Default on the part of the Insured's client". The "Transaction Premium" was itself defined as an amount "equal to the difference in value between the Pre-Agreed Price and the Actual Sale Price". The TPC covers not the pre-agreed price but rather the difference between that price and the actual sale price, so that the value of the collateral is important. Thus, a claim under the TPC would arise if two risks eventuated: (i) the default risk and (ii) the risk on the collateral, namely that the goods, when sold, would not enable the Bank to receive the amount which the client had agreed to pay. If there were no default, then the value of the collateral would not matter. Equally, if the collateral were sufficient, then the default would not matter.
  937. Against that background, I consider that Clause 3 is capable of applying to steps which are reasonably practicable to ensure that there is no difference between the Pre-Agreed Price and the Actual Sale Price, thereby preventing any claim being made under the TPC. In other words, it is capable of applying to the steps which the underwriters contend should have been taken in order to ensure that the Bank did not overpay for the collateral which would, in the event of default, be sold in order to realise the Actual Sale Price.
  938. Having addressed those two preliminary arguments, I turn to what I regard as the most important construction point. The Bank submits that ordinary negligence cannot constitute a breach of Clause 3: recklessness is required. The underwriters submit that, in the context of the TPC, negligence is sufficient. Although this is the most important construction point, I do not consider that it is a difficult one.
  939. The Bank referred to the well-known line of authority concerning "reasonable precautions" clauses in the context of policies which provided cover against the consequences of the assured's negligence. Jackson J reviewed the authorities in this area of the law in The Board of Trustees of the Tate Gallery v Duffy Construction Ltd (No. 2) [2007] EWHC 912 (TCC). His conclusions at [26] were as follows:
  940. "(1) In a policy of liability or property insurance a reasonable precautions clause in the conventional form is not breached by mere negligence. Recklessness is what constitutes a breach of such a clause.
    (2) The recklessness which must [be] established is recklessness by the insured himself, as opposed to his employees.
    (3) The first two propositions are canons of construction developed by the courts, because it is improbable that the parties intend to negate a core part of the insurance cover. Nevertheless, if a reasonable precautions clause were drafted with sufficient clarity, it would be possible to achieve that harsh result".
  941. Jackson J's first proposition derives from the classic exposition of what is required for breach in the judgment of Diplock LJ in Fraser v BN Furman (Productions) Ltd [1967] 2 Lloyd's Rep 1, 12. Diplock LJ there explained, in the context of a claim against a broker concerning an employers' liability policy, what on its true construction a "due diligence" clause requires.
  942. ""Reasonable" does not mean reasonable as between the employer and the employee. It means reasonable as between the insured and the insurer having regard to the commercial purpose of the contract, which is inter alia to indemnify the insured against liability for his (the insured's) personal negligence. … What, in my view, is "reasonable" as between the insured and insurer, without being repugnant to the commercial object of the contract, is that the insured should not deliberately court a danger, the existence of which he recognises, by refraining from taking any measures to avert it… What, in my judgment, is reasonable as between the insured and the insurer, without being repugnant to the commercial purpose of the contract, is that the insured, where he does recognise a danger, should not deliberately court it by taking measures which he himself knows are inadequate to avert it. In other words, it is not enough that the employer's omission to take any particular precautions to avoid accidents should be negligent; it must be at least reckless, that is to say, made with actual recognition by the insured himself that a danger exists, not caring whether or not it is averted. The purpose of the condition is to ensure that the insured will not refrain from taking precautions which he knows ought to be taken because he is covered against loss by the policy".
  943. Jackson J's second proposition – that it is not sufficient for underwriters to show the existence of such recklessness on the part of an employee or agent of the insured, but rather of the insured itself – is derived from authorities such as Woolfall & Rimmer Ltd v Moyle (1941) 71 Lloyds LR 15 and Fraser at 12 ("It is the insured personally who must take reasonable precautions. Failure by an employee to do so, although the employer might be liable vicariously for the employee's negligence or breach of statutory duty, would not be a breach of the condition").
  944. It was (rightly) accepted by Mr. Parsons that, in the context of the policy provision of "all risks" cover against physical loss and damage to cargo, the Bank was covered against its own negligence and the consequences of such negligence. In my view it follows, as a matter of principle, that Clause 3 should be interpreted in accordance with the line of authorities discussed in Tate Gallery. Otherwise, the clause would have the effect of negating a core part of the insurance cover.
  945. Mr. Parsons submitted that Clause 3 of the policy should be construed differently in so far as it concerned the TPC. He submitted that in the unusual case of a policy where perils of the sea would cover negligence but there is a new add-on that itself does not cover negligence, then the clause is "a negligence clause for that part of the policy and it would only be recklessness if that part of the policy covered negligence".
  946. Ms. Sabben-Clare's position was that this was an impermissible approach to construction. Clause 3 could not be given different meanings depending upon which aspects of the coverage of the policy were being considered. It was therefore not possible to "parse out section 2 clause 3 and give it a different meaning in relation to the transaction premium clause".
  947. I agree with the Bank's submissions on this issue. Clause 3 must in my view have the same meaning, in particular in relation to the standard required, whichever part of the policy is being considered. Mr. Parsons did not suggest that the clause could be construed so as to mean ordinary negligence in the context of the all risks cover; no doubt because to do so would negate a core part of the cover. I do not see how the interpretation of the clause could be different in the context of the TPC.
  948. This conclusion is in my view supported by one of the authorities cited by the Bank, and referred to in Tate Gallery. In Sofi v Prudential Assurance Co Ltd [1993] 2 Lloyd's Rep 559, a claim was made for jewellery which had been stolen when the insured had left his car for a short while to go for a walk whilst at Dover en route to a holiday in France. There were two policies which were relevant, both of which contained "reasonable steps" clauses. One policy was an ordinary householder's policy, which included coverage for legal liabilities to third parties (under Section 11). The other policy was a "Travelwise Insurance" policy. A central issue in the case was whether the clauses required recklessness in order to provide a defence to the insurers, or whether an ordinary negligence standard was to be applied. In reaching the conclusion that the Fraser v Furman recklessness standard was applicable, Lloyd LJ said (at 564):
  949. "Returning to the main line of the argument, it is at once apparent that some limitation must be placed on the full width of the language of General Condition 2. This follows from the fact that the condition applies to all sections of the policy; not just section 3. If the clause were to be taken as meaning that the insured must take all reasonable care of the property insured and all reasonable care to avoid accidents, then the insurers could never be liable under section 11, for liability under section 11 pre-supposes that the insured or a member of his family is legally liable to a third party. Legal liability in the great majority of cases depends on want of reasonable care. So a wide construction of General Condition 2, requiring the insured to take all reasonable care, would be altogether repugnant to the cover apparently afforded by section 11 of the policy. Similarly, the insurers would escape all liability under sections 1 and 2 in the very ordinary case of damage to a house or its contents by fire (one of the insured perils) if the fire were caused by the negligence of the insured. That could not be right."
  950. Later in the judgment, at 565-6, Lloyd LJ said:
  951. "Secondly, Mr. Wadsworth argued that the recklessness test applies only to liability insurance such as was in issue in Fraser v. Furman, not to property insurance. In the case of liability insurance, there is nothing to insure at all unless the insured is liable to a third party. In the case of property insurance, that is not so. Therefore, in the case of property insurance, there is no need to give a restricted meaning to the words. "
    I do not accept that there is any distinction between the two types of insurance. One can see at once the difficulties which would arise if that were so, especially in the case of a composite policy such as the present. How could the condition have one meaning in relation to section 3 and a different meaning in relation to section 11? How could it have one meaning in relation to section 2 (a) to (f) and a different meaning in relation to section 2(g) which covers the insured's liability as an occupier?
    Mr. Wadsworth seeks to meet those difficulties by arguing that the clause may have the same meaning throughout the policy but impose a different standard of care in relation to the different sections. The insured must take greater care in relation to his own goods than to avoid liability to others. With respect, I find that solution to the problem quite impossible. If that is what the defendants intended, which I do not accept for one moment, then they should have made their meaning a great deal clearer.
    So I would hold that the recklessness test is, contrary to Mr. Wadsworth's argument, equally applicable whether condition 2 is included in a property insurance or in a liability insurance.
  952. Sofi shows, in my view, that a consistent meaning should be given to a "reasonable steps" clause when applied to all parts of a policy providing different aspects of coverage. The meaning cannot, in the absence of clear words, fluctuate depending upon which part of a policy is being considered. The fact that a condition is expressed to apply to all sections of the contract reinforces that conclusion. Furthermore, in deciding whether a "reasonable steps" clause is to be construed as imposing a recklessness standard, it is necessary to pay regard to the impact of the contrary argument (ie an ordinary negligence standard) on all aspects of the cover provided.
  953. Applying these principles, I do not consider that Clause 3 can be construed, when considering the TPC, as imposing a different standard to that which would be applicable to a claim under the "all risks" part of the cover. This conclusion is reinforced by the fact that the policy expressly provides, in the words at the start of Section 2: "These general conditions apply to all sections of this contract".
  954. I did not consider that any of these conclusions were affected by the decision of Tucker J in Amey Properties v Cornhill Insurance PLC [1996] Lloyds Re LR 259 to which the underwriters referred. The judge in that case was considering a clause in a motor policy which required the insured to take reasonable precautions to maintain the vehicle in an efficient and roadworthy condition. The judge decided that the test of negligence was appropriate in the context of motor policies since that was not repugnant to the commercial object of the contract. He recognised that the courts had adopted different approaches to the construction of words of such clauses depending on the nature of the policies in which they appear. The case does not in my view support the proposition that the clause in the present policy should be generally construed as imposing a negligence standard, notwithstanding that the core coverage is for all risks of property damage. Indeed, the underwriters realistically do not so contend. Nor is there any suggestion in the case that, contrary to the approach in Sofi, the clause can be construed differently in relation to different parts of the cover provided by a policy.
  955. I also considered that Ms. Sabben-Clare was correct in her additional argument that just as the all risks cover covered the negligence of the Bank, so did the TPC. The TPC applies if "Default" in failing to purchase/ repurchase arises "for whatever reason". That would include the situation where the Bank had acted negligently in relation to the circumstances in which the default arose. Since negligence is covered in that context, the underwriters' construction of Clause 3 would negate a core element of the cover that is provided under the TPC itself.
  956. For these reasons, I consider that the recklessness standard, as discussed in the case-law, is apposite in the context of Section 2, Clause 3.
  957. G3: Was the Bank reckless in relation to the quality of the collateral?

    The evidence as to the Bank's approach

  958. In describing the Bank's approach, it is to some extent necessary to distinguish between Icestar and the wider Bank; principally because the due diligence which was carried out in relation to Transmar and Euromar was, at least principally, carried out by individuals within the wider Bank rather than the specific Icestar team. The evidence as to the approach taken was given by Mr. Stroink and Ms. Franssen.
  959. Their starting point was that the Icestar deals were only conducted with clients of the Bank and their affiliates who were trusted and with whom the Bank had a relationship. Ms. Franssen acknowledged in her evidence that there was a risk that the Bank would have to deliver cocoa beans to the terminal market or sell material in a physical market if Transmar could not or did not exercise its option or were to default, but this risk was considered to be "very low because we did a lot of due diligence on the ability of the client to repay". If they were not confident of the ability of the client to repay, they would not enter into a transaction in the first place. If, therefore, there was any expectation that a worst-case scenario of a default was likely to happen, then Icestar would not have entered into the transaction at all. Mr. Stroink also referred in his evidence to the fact that Icestar 2 and 3 deals were only done with existing clients of the Bank who were reputable and where the Bank knew the strength of their balance sheet.
  960. The documentation in the hearing bundles evidenced the fact that the Bank did indeed carry out "a lot of due diligence", as Ms. Franssen said. For example, a very detailed paper was prepared in December 2015 in connection with the proposal that the Bank should participate in the continuation of Transmar's borrowing base facility. Mr. Stroink spoke to the long-standing relationship with Transmar and Euromar, which was also evidenced by the December 2015 paper.
  961. Mr. Stroink also explained that Icestar repo transactions involved reputable clients to whom the Bank was offering multiple products. Quality checks were carried out in relation to Icestar 1 transactions, where the goods were exchange deliverable and the client had no obligation to buy back. On those transactions, Icestar would require quality certificates which demonstrated that the beans met the standard required to be exchange deliverable. This was because Icestar regarded the relevant risk on those transactions as being the risk of non-performance by the exchange. Icestar therefore needed to be confident, as Mr. Stroink explained, that the commodities purchased would meet the requirements of the futures contracts.
  962. However, the position on Icestar 2 and Icestar 3 transactions was different. Here, the Bank obtained contractual representations and warranties in its favour. Its standard documentation contained a representation and warranty that the goods were free of defects. There were also binding obligations on the part of the client to repurchase the goods on an "as is, where is" basis, or to provide exchange-deliverable beans in substitution for the cargo. Mr. Stroink's evidence acknowledged the existence of a risk in relation to the quality of the goods. He denied, however, that he did not care about the quality. But he said that:
  963. "… our view was that we mitigated that risk by having the client rep to us – and these were reputable clients of the bank which we offer multiple products to and we have a banking relationship with, which we are partner to – to rep to us, next to the fact that they are obliged to repurchase on an as-is, where-is basis. So this … was our stance at that moment in time".
  964. Mr. Stroink also referred to the incentive on the client to buy back the products which were the subject of the repo transactions. The Bank had not advanced the full value of the products; so "our clients would in general always buy back because there's value for them in the commodity which you finance".
  965. Having heard their evidence, it is clear to me that neither Mr. Stroink nor Ms. Franssen thought that its clients were or might be trading sub-standard stock. They proceeded on the basis that the stock which they were financing was intended for use in their clients' ongoing business. This is not a case where there had been prior difficulties which put the Bank on notice of potential problems. Ms. Franssen acknowledged in her evidence that, with hindsight, she would "probably check it". But at the time:
  966. "… we were dealing with a well-known client that had a processing factory in Germany that was also able to sometimes reprocess cocoa products. So what would happen is they would either start from scratch with beans, or they had certain products that they would reprocess … in a different specific product itself. We had no reason to doubt the client … and that they were going to buy it back at the point we were rolling them, and that's why we kept rolling over without paying attention to the tenor".
  967. Later in her evidence, she described the steps which the Bank took in order to ensure that the collateral was held to their order by the third party warehouse. When asked about the need to make sure that collateral was worth what the client said it was worth, and about the possibility of getting quality certificates, she said:
  968. " … but you have to appreciate that commodities are stored all over the world. No commodity bank that does secured financing on commodities sends an independent quality inspector to every single stock they're financing. It's just not market practice.
    …
    But they're not traders, they're financiers and so they wouldn't always be able really look at … a certificate and assess exactly what that certificate means. They do a very detailed level of due diligence on their client, on their position in the market, what would they need the product for. For us we had pre-sold these goods on an as-is, where-is basis to a client that was at that point in time well respected and we expected to buy it from us. So no, it's also not practice for a lot of banks to get quality certificates for all the commodities that they finance on a secured basis".

    Discussion

  969. I do not consider that the Bank was reckless in its approach, and accordingly there was no breach of Clause 3. The significant points in my judgment are as follows.
  970. First, there is no material, on the evidence before me, that the Bank failed to comply with normal banking practice and procedures in relation to its approach to collateral. There was, therefore, nothing to suggest that repo financiers generally conduct quality checks, or ask for and then analyse quality certificates in relation to commodities such as the cocoa products which were being financed in the Icestar 2 and 3 transactions. Ms. Franssen's evidence, which I accept, is that repo banks do not conduct quality checks when they advance funds. She said that it was just not market practice to do so, and that the approach of financiers was to do what the bank did here, namely to conduct a very detailed level of due diligence on their clients and to obtain protection through the contractual terms agreed. She explained that it was not practical for the banks to arrange inspections of commodities owned by large companies which have stocks all over the world, and who were financing those stocks on a constant basis.
  971. Ms. Franssen was there talking about physical inspections, and this did not therefore in itself answer the insurers' argument that banks could ask their clients to provide quality certificates. However, the evidence also indicates that the Bank did not have (and one would not expect banks generally to have) in-house expertise which enabled them to draw sensible conclusions from such quality certificates as might be provided. Accordingly, the insurers needed to argue that the Bank should therefore have engaged outside experts, such as Chiodo who were eventually engaged after the defaults and the problems that emerged in 2017. There was, however, no evidence to suggest that it would be normal banking practice either to ask for quality certificates, or to engage experts to analyse them. The insurers' argument in this regard was one that developed at a late stage, and in my view owed much to hindsight. The insurers' pleaded case was set out in very general terms, without identification of the "reasonable steps" that the Bank should have taken. It is also relevant that the supposed deficiencies of the Bank were not identified by the insurers' investigators, Gray Page, and were only raised very shortly before trial.
  972. The underwriters placed much reliance on the fact that the Bank did obtain quality certificates for Icestar 1 transactions. It was therefore submitted that the same could have been done for Icestar 2 and 3 transactions. However, there were material differences between the Icestar 1 transactions and the others. Icestar 1 transactions involved goods which were exchange deliverable, and where the client had no obligation to buy back. The Bank regarded the relevant risk as being on the exchange, to which the beans would be supplied under Icestar 1 transactions if there were a default. In relation to cocoa beans, applicable standards were those laid down by the exchange, and I understood that the quality certificates on Icestar 1 transactions would enable the Bank to ascertain, without difficulty, whether or not the goods were exchange compliant. By contrast, there was, on the evidence before me, no equivalent standard in relation to the non-exchange deliverable cocoa products; ie the goods which were subject to Icestar 2 and 3 transactions. In the absence of a clear standard against which a quality certificate could be compared or judged, the simple provision of a quality certificate would not enable the Bank to draw conclusions as to the quality of the goods. This was why the underwriters were forced to argue that outside experts should have been engaged. There was no equivalent suggestion that outside experts would be required in order to assess the Icestar 1 quality certificates. In these circumstances, I did not consider that there was any significance to the fact that the Bank obtained quality certificates for Icestar 1 transactions.
  973. Against this background, I do not consider that there is any proper basis upon which I could conclude that the Bank was reckless in its approach. In Woolfall and Rimmer v Moyle (1941) 71 Lloyd's LR 15 (which was considered by Jackson J in paragraphs [17] – [20] of Tate Galleries), Goddard LJ equated recklessness (in contrast to negligence) with a person who does not run his business in the ordinary way, thereby failing to take the reasonable precautions which ordinary business people take. He illustrated the concept by giving examples of an employer providing no lights where there were stairs which could not be seen in the dark, or failing to provide fencing where it would be reasonable to expect people to fall, or leaving explosives around in a place where any employed youth could tamper with them. In the present case, however, the insurers have failed to show that the Bank's approach was unusual or anything other than the way in which ordinary business people, providing repo finance, would act. Their criticisms of the Bank cannot realistically be compared to the illustrations of recklessness provided in Woolfall.
  974. Secondly, this is not a case where the Bank changed its approach to risk management as a result of obtaining the cover provided by the TPC. Prior to obtaining the TPC, the Bank had not taken the steps to ascertain quality which, on the insurers' case, should allegedly have been taken. The Bank at that stage, when uninsured, did not see the need to guard, in the manner proposed by underwriters, against the risk of collateral on Icestar 2 and 3 transactions being inadequate. Subsequent to obtaining the TPC, the Bank carried on its business in the same way, save that there was an increase in the volume of repo transactions. There is no evidence, however, to suggest that the Bank took a more lax approach to the granting of repo facilities than it had done previously, still less that it did so in the light of the insurance that had been obtained. The effect of Mr. Stroink's evidence was that the reason for seeking the insurance was to have a second line of defence as part of its risk management strategy, not that there was an intention to lend to more risky clients.
  975. It is true that the TPC had an impact on the amount of business which the Bank was willing to do, and that the volume of transactions increased after it had been agreed in 2015. However, the evidence did not suggest to me that this was a consequence of the Bank lowering its standards or changing its approach to risk. A significant reason for the increase was that the existence of the insurance, with first-class rated insurance companies, lowered the Bank's capital requirements for its Icestar 2 and 3 transactions, so that Icestar's return on capital (sometimes called return on equity) would improve. The insurance therefore had the effect of enabling the Bank to expand its business in these respects, since more business could be conducted relative to the same amount of capital. I return to this point in Section J3 below in the context of Edge's arguments on causation.
  976. Thirdly, there is no evidence to suggest that the Bank identified the need to take the steps for which the underwriters now contend, still less that the Bank did so but decided not to take those steps because of the existence of the insurance. It is true that the Bank's internal documents, and the communications with NRF, indicate (unsurprisingly) that the Bank was aware of a risk that, in a default situation, the collateral might not realise the full amounts owed to the Bank. There is, however, nothing to suggest that anyone identified the possible need, in order to mitigate this risk, to conduct quality checks on collateral or to obtain quality certificates and have them analysed. Ms. Franssen acknowledged in her evidence that, with the benefit of hindsight in the light of the problems which occurred, it would have been better if the Bank had done more. That does not in my view come close to establishing a case of recklessness, and indeed does not even establish negligence.
  977. Fourth, this is not a case where the Bank was casual in its lending approach. As will be apparent from the summary of the evidence above, the Bank carried out due diligence in relation to the customers with which they dealt, including in relation to Transmar and Euromar. Nor is it a case where there was some prior problem which should have alerted the Bank to the need to carry out the steps for which underwriters contend. The factual position is that there had not been prior problems of the type that were ultimately experienced, and the Bank (understandably in my view) did not think that its clients would be trading sub-standard stock. As Ms. Sabben-Clare submitted in closing, the Bank had no reason to think that its clients were running anything other than a normal commercial operation and that it had good reason for holding the stocks which it had purchased.
  978. The insurers referred to a number of documents which were said to constitute "red flags", particularly in relation to the age of the products which were being financed. Ms. Franssen was asked about these documents, and she was asked questions as to whether her team or colleagues should have drawn certain conclusions. It seemed to me that this was a very long way from a case of recklessness. Moreover, recklessness would need to be established against the senior individuals of the Bank. It was common ground that Mr. Stroink would count for that purpose and I am willing to proceed on the basis that Ms. Franssen would count as well. However, Ms. Franssen had not herself reviewed the documents (principally spreadsheets) which were said to contain the red flags, and I did not see how any failure by more junior employees to spot alleged red flags would establish recklessness against the Bank.
  979. Overall, in my view, there is nothing to suggest that the Bank failed to take precautions that it knew it ought to be taking. Rather, as submitted in its closing argument, the Bank reasonably relied upon a careful client due diligence process and on the past track record of the companies with which the Bank was dealing. This is not a case of the Bank courting a danger which it knew to exist.
  980. The Bank also relied on the fact that the Bank did provide underwriters with a description of its "due diligence procedures and collateral management". This was provided under cover of an email from Mr. Stroink dated 27 January 2016, and was forwarded to underwriters. Those processes focused on the quality of the warehouses where the goods would be stored, and the steps taken to ensure that the warehouses acted on the Bank's instructions. The Bank relied upon the fact that the documentation did not indicate that any steps were taken to ascertain the quality of the cargoes, and that the underwriters did not suggest at the time that the procedures described by the Bank were inadequate. I regarded this as a marginal point. If the Bank was indeed reckless in its approach, then there would have been a breach of Clause 3 even though the underwriters did not spot it at the time.
  981. For these reasons, I do not consider that the Bank was in breach of Clause 3.
  982. It is therefore not necessary to address the other arguments advanced by the parties, such as whether any breach was causative of any loss or whether the Bank could recover in any event on the basis that the default was a proximate cause of loss.
  983. H: Sue and Labour


     

    H1: The issues

  984. This aspect of the case concerns the Bank's conduct, and alleged failure to "sue and labour" following the first default of Euromar in August 2016. Clause 16 of the Institute Cargo Clauses 'A' contained the standard sue and labour provision:
  985. MINIMISING LOSSES
    Duty of Assured
    16. It is the duty of the Assured and their employees and agents in respect of loss recoverable hereunder
    16.1 to take such measures as may be reasonable for the purpose of averting or minimising such loss, and
    16.2 to ensure that all rights against carriers, bailees or other third parties are properly preserved and exercised
    and the Insurers will, in addition to any loss recoverable hereunder, reimburse the Assured for any charges properly and reasonably incurred in pursuance of these duties.
  986. It was common ground that this duty was qualified by the specific term to which the parties agreed in Clause 3 of the General Conditions of the policy. This imposed a standard of good faith in relation to the minimisation of potential loss:
  987. Due Diligence
    The Insured shall do (to the extend it reasonably can do) all things reasonably practicable to prevent any claim being made under this contract, providing always that following the occurrence of a peril in relation to the subject matter insured, the Insured may in its sole discretion elect an appropriate course of action, as it considers appropriate in any particular circumstance, subject to the Insured acting in good faith with the intention of minimising any ultimate potential net loss (save that the Insured shall not be required to exercise any put option following the occurrence of any such peril). (Emphasis supplied)
    The underwriters' case
  988. By an amendment introduced in May 2020, the underwriters alleged that there was a failure by the Bank to "sue and labour" in four respects, which were to some extent related. The context was, on the underwriters' case, that the default by Euromar gave rise to a real risk that the extant deals across Euromar and Transmar would all default, taking into account the close relationship between the entities, and the practical reality of Euromar and Transmar being part of a family business.
  989. First, the underwriters contended that it was an obvious step for the Bank to check immediately the quality and value of the product held as collateral, and that any prudent uninsured would have done so immediately the default occurred. Had it done so, it would have quickly realised the risk of very severe loss if there were further defaults. No independent report on quality was obtained until late 2017, when a report was produced by a Dutch company called Chiodo.
  990. Secondly, the underwriters relied upon the swap options in favour of the Bank which were contained in the transactions with Transmar. The purpose of these swap options was to give an option to enable the Bank to receive exchange deliverable beans (or cash) in replacement for the cocoa products which it had financed under the Icestar 2 transactions. This would enable the Bank to de-risk its exposure to those products, and receive instead beans which were a lower risk commodity to hold.
  991. The Bank did eventually exercise these options, but not until 22 December 2016. This was far too late. A prudent uninsured would have exercised the swap options on the Transmar deals immediately, on 17 August 2016, or very shortly thereafter. This was because it would have been obvious that there was a real risk that it would need to realise its collateral, and also (from information already in its possession) that the collateral it had taken was too old, and had serious quality issues.
  992. Had the swaps been exercised then the Bank's financial position would have significantly improved. The relevant figures, on the underwriters' case, were as follows.
  993. (1) If Transmar had promptly honoured its obligation to provide cocoa beans in place of cocoa products, or the cash equivalent, and the beans so received had then been promptly sold, then the Bank would have recovered £54.2m plus US$0.9m. It would therefore have avoided any loss at all.
    (2) If Transmar had promptly honoured its obligation to provide cocoa beans, but these had been held until the end of December 2016, the Bank would have recovered £ 34.9 million less US$ 2.1 million. The lower amount, when compared to a prompt sale of the beans, reflected the fall in the market between August and December 2016.
    (3) If Transmar had failed to deliver any beans, and the Bank had then sold the cocoa products that it was holding, the Bank would have received additional revenue of physical sales of £ 16.9 million less US$3.5m. There would also have been a hedging gain of £4.6m.
  994. Thirdly, upon the default of Euromar, the Bank was exposed to market risk on the collateral which was subject to the Euromar deals: a fall in the market would result in a fall in the value of the collateral. Accordingly, the Bank should have hedged its position by entering into transactions on an exchange. In substance, this meant that the Bank's "long" position (because it held cocoa products) would be matched by a "short" position created by the hedging transaction (because it had agreed to sell cocoa beans). This would, on the underwriters' case, have avoided the exposure to market risk. Had this happened, the Bank's financial position would have improved by £3.4m. The essential reason for this is again the fall in the market after August 2016. The fall would mean that a hedging transaction, whereby the Bank sold cocoa beans at the better prices available in August, would have yielded the Bank a profit when those transactions came to be closed out.
  995. Fourth, the Bank should have taken steps immediately to liquidate the products following Euromar's and Transmar's defaults. The Bank incurred significant post-default carry costs of around £ 3 million. This off-set almost half of its revenue from the sales. Had it liquidated by a rapid salvage sale, the same sale price relative to the market price would have been achieved.
  996. There was some interrelationship between the first, second and fourth arguments. This was because, as Mr. Parsons submitted in his oral closing, if the swap options had been exercised and Transmar had failed to deliver the beans, then the Bank would know that it would be necessary to resort to its collateral. That would or should then lead to more rapid checks on quality and attempts to sell the collateral quickly.
  997. The upshot of this argument was that if, contrary to the Underwriters' primary case, the Bank's overpayment for the goods was not the proximate cause of all of the loss, then the failures by the Bank to sue and labour were the proximate cause of the Bank's losses.
  998. The Bank's argument

  999. In summary, the Bank's principal arguments were as follows.
  1000. First, the duty under clause 16 arose only after the insured peril had struck, ie once Euromar and Transmar's defaults had occurred or were a clear and obvious danger. It follows that the Transmar allegations fail because Transmar had not defaulted on 16 August 2016. The earliest that the insured peril occurred, or was imminent, in the present case was 17 August 2016 in relation to Euromar, and 29 December 2016 in relation to Transmar.
  1001. Secondly, a breach of clause 16 can only ever amount to a defence if the insured's own conduct was so unreasonable as to break the chain of causation and amount to the proximate cause of the loss. This requires extreme facts to prove. There is no question in the present case of the Bank's conduct having been so unreasonable as to have broken the chain of causation and amounted to the proximate cause of the losses, displacing Transmar's and Euromar's failure to pay as a proximate cause. On the contrary, the Bank acted reasonably. Furthermore, the Bank's conduct should be judged by the standard of ordinarily competent bankers in their shoes. The underwriters' expert evidence, from witnesses who had never worked for banks, provides no evidence which could support a finding that the Bank fell below this standard.
  1002. Thirdly, the Bank's obligation to sue and labour under the Policy was further expressly qualified by Section 2, Clause 3. Underwriters would have to establish that Icestar acted in bad faith. Quite properly, they did not seek to suggest this.
  1003. Fourth, the sue and labour allegations would only give rise to a defence if and to the extent that losses would have been avoided. The burden of proof is on the underwriters to show this. Here, no loss would have been averted if sales had been made earlier: the quality of much of the cargo was poor when it was ultimately sold, and was not significantly better in August or September 2016. Similarly, no loss would have been averted if the Transmar swap options had been exercised earlier: there is no basis for thinking that Transmar would have delivered beans or cash in or shortly after August 2016 if the swap options had been exercised.
  1004. Hence, the only point which had any potential impact on the Bank's losses was the argument that hedges should have been taken out in relation to the Euromar transactions. That claim was, in relative terms, small. At most, it was worth £ 3.43 million, although that case depended upon the Bank taking out hedges on the very day of default. If the relevant date for taking out a hedge was when the Euromar contracts were terminated, the gain on the hedge would have been £ 2.88 million. If the relevant date took into account a financial buffer which was built in to the repo contracts in order to protect the Bank from loss (essentially because the Bank did not lend based upon the full value of the collateral), the hedging gain would be £ 1.01 million.
  1005. H2: Legal principles

  1006. It was common ground that if any of the sue and labour arguments were to succeed, it was necessary for the chain of causation to be broken. Accordingly, the bare assertion that losses could have been avoided by some conduct on the part of the insured was not sufficient. A breach of the duty to sue and labour only provides a defence if the assured's conduct was such as to break the chain of causation, and supplant the original insured peril (here the "Default" leading to a loss of the Transaction Premium under the TPC) as the proximate cause of loss.
  1007. Thus, in State of Netherlands v Youell [1998] 1 Lloyd's Rep 236, 244, Phillips LJ said (when discussing s 78 (4) of the Marine Insurance Act, which provides that it is the duty of the assured and his agents to take such measures as may be reasonable for the purpose of averting or minimising the loss):
  1008. "… the duty referred to in s.78 (4) will only have significance in the rare case where breach of that duty is so significant as to be held to displace the prior insured peril as the proximate cause of the loss".
  1009. In reaching that conclusion, Phillips LJ drew upon a judgment of Colman J in National Oilwell (UK) Ltd v Davy Offshore Ltd [1993] 2 Lloyd's Rep 582, 618, where he said in relation to s 78 (4):
  1010. "It goes no further than the obvious proposition that if after the advent of an insured peril or when the advent of an insured peril was obviously imminent the assured or his agent failed to act to avert or minimise loss in circumstances where any prudent uninsured would have done so, the chain of causation between the insured peril and the loss will be broken. Clearly if the insured peril is not the proximate cause of the loss the assured cannot recover".
  1011. These and subsequent cases are discussed in Arnould paragraph 22-19, where the authors state:
  1012. "The standard to be applied in determining whether there has been a breach of the duty under s.78(4) was also discussed in The Silva. The relevant test, in that case, was whether any ordinarily competent Egyptian lawyer would have acted differently. Mutatis mutandis the same test would apply to any classes of agents, but in most cases, as we have seen the causation test will be sufficient, namely whether there has been a failure to take such obvious steps as any prudent uninsured could be expected to take, enabling it to be said there has been a breach in the chain of causation".
  1013. Both the Bank and underwriters referred to the test in The Silva [2011] 2 Lloyd's Rep 151. That case concerned the conduct of Egyptian lawyers. Burton J said (at [51]) that a breach of the sue and labour clause would depend upon whether "any ordinarily competent Egyptian lawyer would have acted differently".
  1014. The Bank also relied upon the discussion, by Gross LJ in Borealis AB v Geogas Trading SA [2011] EWHC 2789 (Comm) paras [42] – [47], of the circumstances in which the chain of causation may be broken. Although this was not an insurance case, and the issue there was whether there had been a break in the chain of causation linked to the defendant's breach of contract, it was not suggested by underwriters that Gross LJ's discussion was inapposite in the present context.
  1015. In that case, Gross LJ said that, in a breach of contract case, the conduct of the claimant must "obliterate" the wrongdoing of the defendant: the true cause of the loss must be the conduct of the claimant. He said that it was difficult to conceive that anything less than unreasonable conduct on the part of the claimant would be capable of breaking the chain of causation. Ordinarily, reckless conduct would do so, but there was no rule of law that only recklessness would suffice. The claimant's state of knowledge at the time of and following the defendant's breach is likely to be a factor of very great significance. The more the claimant has actual knowledge of the breach, or the dangerousness of the situation which has arisen and the need to take appropriate remedial measures, the greater the likelihood that the chain of causation will be broken. Conversely, the less the claimant knows, the more likely it is that only recklessness will suffice to break the chain of causation. Ultimately, the question of whether there has been a break in the chain of causation is fact sensitive, involving "a practical inquiry into the circumstances of the defendant's breach of contract and the claimant's subsequent conduct".
  1016. The approach of Gross LJ also seems to me to be consistent with that of the Supreme Court in the decision (published after the conclusion of the present trial) in Arch Insurance and others v FCA: see eg paragraph [168]:
  1017. The common-sense principles or standards to be applied in selecting the efficient cause of the loss are, however, capable of some analysis. It is not a matter of choosing a cause as proximate on the basis of an unguided gut feeling. The starting point for the inquiry is to identify, by interpreting the policy and considering the evidence, whether a peril covered by the policy had any causal involvement in the loss and, if so, whether a peril excluded or excepted from the scope of the cover also had any such involvement. The question whether the occurrence of such a peril was in either case the proximate (or "efficient") cause of the loss involves making a judgment as to whether it made the loss inevitable - if not, which could seldom if ever be said, in all conceivable circumstances - then in the ordinary course of events. For this purpose, human actions are not generally regarded as negativing causal connection, provided at least that the actions taken were not wholly unreasonable or erratic. (Emphasis added).
  1018. Whilst there is, in my view, no reason why the exposition of the general principles by Gross LJ is inapplicable in the present "sue and labour" context, it seems to me that they are ultimately reflected in the question identified by Colman J in National Oilwell: ie did the insured fail to act to avert or minimise loss in circumstances where any prudent uninsured would have done so. If so, then the chain of causation between the insured peril and the loss will be broken.
  1019. This is the question that arises ordinarily when the court is considering the application of s 78(4) of the Marine Insurance Act or clause 16 of the ICC (A) clauses, both of which impose a duty to take "such measures as may be reasonable". However, the Bank contended that the question was different in the present case in the light of Clause 3 of the General Conditions. The critical question here was, in the light of that clause, whether the Bank had acted in good faith. Unless there was a lack of good faith, the chain of causation would not be broken. The underwriters disputed this proposition. It is an important point, which I discuss below.
  1020. As far as timing of steps taken to avert or minimise loss, it was common ground that the duty to sue and labour arose when the peril had arisen or was imminent, a concept often expressed as "in the grip of a peril". In that context, underwriters referred to the decision in ICS v BTI [1984] 1 Lloyd's Rep 154. That case concerned a claim for the recovery by the insured of expenses incurred as a result of steps allegedly taken in fulfilment of the duty. Eveleigh LJ said at p 158 that s 78(4) imposes a duty to act in circumstances where a reasonable man intent on preserving his property, as opposed to claiming upon insurers, would act. It was therefore not necessary for the assured to show that a loss would "very probably" occur in order to enable costs to be recovered. Rather, the assured could recover where he could demonstrate that a prudent assured person, mindful of an obligation to prevent a loss, would incur expense of an unusual kind.
  1021. H3: The facts relating to the Bank's conduct

  1022. Ms. Franssen was the principal witness for the Bank who gave evidence as to the steps taken following Euromar's default. She was, as I have previously indicated, a reliable witness who throughout gave her evidence fairly. In fact, there was no significant challenge to the factual evidence which she gave as to the steps which the Bank had taken following Euromar's default. Although she had given some evidence about this in her first statement, the issue was addressed in greater detail in her third statement served after the underwriters had pleaded the sue and labour defence. Ms. Franssen's overall point was that she believed that she and her colleagues did everything they could to minimise the Bank's losses, based on the information that was available to them at the time and their understanding of the factual position. They had carefully considered all the options available to them, and assessed which options would best prevent or mitigate any losses. That was always her primary consideration. The evidence which Ms. Franssen gave in her witness statements was in summary as follows.
  1023. Ms. Franssen explained that following Euromar's first default in August, the Bank's priority was to check that it had good title to all the underlying commodity, in order that it could be sold and the amounts outstanding fully recouped. Independent inspectors, Control Union, were sent into the warehouses to check the quantity of stock. By September 2016, the inspectors had inspected 7 of the 22 warehouses where Euromar stock was held. The inspections showed that the quantities which should have been there were indeed in place. Subsequent inspections of other warehouses also showed that the quantities held were between 99 and 100% accurate; ie the goods were there. The Bank also obtained quality certificates for the goods and started speaking to their contacts in the market. At that stage, the Bank was assessing the entire portfolio and checking the positions held by the Bank, focusing on gathering information.
  1024. Within the Bank, the credit authority for Euromar was transferred to the Bank's Financial Restructuring and Recovery team (FR&R). Transmar was also referred to FR&R because it was a company within the same corporate group. FR&R would then closely monitor the file, together with the Icestar team originally responsible for dealing with these customers. This meant that information and decisions were reviewed by multiple people.
  1025. Following the default, the hope was that the problem was a temporary liquidity issue only. The Bank did not think that the problems would be incapable of resolution, or would lead to Euromar becoming insolvent. This was, in part, because a Chief Restructuring Officer had been recruited externally and appointed by the company. This officer was obliged to present an updated 13-week liquidity plan every week. If he did not report, or was not able to provide a viable liquidity plan, he would call for bankruptcy.
  1026. In addition, the Boston Consultancy Group ("BCG") produced a report dated 15 August 2016 which made clear that Euromar was focused on recovery, with actions being put in place in the context of a 13-week liquidity plan designed to ensure that the company remained solvent. This was a positive sign.
  1027. The report of BCG, headed "Turnaround Euromar: Validated 13-weeks rolling liquidity forecast" ran to some 52 pages, with detailed financial analysis. In the management summary, BCG described a tight liquidity situation, but that the liquidity shortage was "manageable currently – but presumably at the expense of margin potential". The company needed to "step forward to sufficient financing". A number of steps were identified under the heading: "Next steps in turnaround of Euromar".
  1028. In late August 2016, the Bank met with other lenders together with Euromar, Transmar and the Chief Restructuring Officer. The focus was on finding a solution rather than sending default notices and causing Euromar to fail.
  1029. Ms. Franssen described how within the Icestar team, and the wider Bank, discussions were taking place on an almost daily basis. She also had numerous discussions with the principals behind the companies. As a result, the Bank continued to hope that the liquidity problems within Euromar were likely to be temporary only. The Bank certainly did not think that Euromar would not be able to comply with its repurchase obligations at all, following its initial default. This was because both Transmar and Euromar were working actively with Euromar's other lenders to find a solution, and indeed Transmar furnished Euromar with formal letters of support. In the period following Euromar's default on 17 August, there was no indication that Transmar itself was in any financial difficulty, and no reason to think that it would be unable to perform its obligations or become insolvent. The fact that one company in a group has financial difficulties does not necessarily mean that the group as a whole would fail.
  1030. Accordingly, in circumstances where the Bank was actively investigating possible solutions, and had no reason to suspect that anything was wrong with Transmar, Ms. Franssen believed that it would have been "precipitate and wrong" for the Bank to have exercised all of its swap options immediately after Euromar's first default in August 2016. The Bank believed that Transmar was in a position to assist Euromar. If the swap options had been exercised, then she did not think that the company would have cooperated to help sell the Euromar goods, and this would "simply have made matters worse".
  1031. At this time, the Bank was not aware of the quality issues that subsequently became known. This came, later, as a complete surprise. Transmar was a significant trader in the cocoa market, and it supplied a number of well-known chocolate producers around the world. The Bank had been financing beans and product for years, and there had never previously been any issues with the company, and in particular no concerns about the age, quality or value of any of the products.
  1032. Ms. Franssen described how the Bank continued to work with Transmar, Euromar, other banks and potential purchasers to find a solution to the liquidity issues, and to explore other options, such as selling direct to Euromar's clients. The Bank also considered its own connections, but they also considered it prudent to work with both Transmar and Euromar to help find potential buyers. The Bank was not a commodity trader and therefore did not have the same knowledge of the market or access to buyers as their clients.
  1033. In September 2016, Mr. Peter Johnson (one of the principals behind Transmar) informed the Bank that Euromar would be able to repurchase the goods through a credit line with one of two companies (Amerra and Theobroma). Discussions also began in relation to a proposal involving the sale of cocoa products to an Indonesian company and Itochu, which was an investor in Transmar. This deal was negotiated over several months, and came close to completion, with execution versions of the contracts being prepared. But in November 2016, Itochu stopped communicating and the deal did not go ahead.
  1034. During this period, when Euromar defaulted on 5 deals with the Bank (one in August, three in September, and one in October), the Bank was discussing delivering the goods to off-takers of Euromar direct. Euromar and Transmar gave the Bank a number of client lists and existing contract details, and the Bank spoke to some of these customers and also to its own clients as regards the stock. Those clients included a number of major players in the market, and the Bank sought their assistance and advice in disposing of the cargo that had been left on their hands. The initial strategy was to sell all of the stock to a single buyer, and the Bank was hopeful that one of these companies would buy the cargo from them. But it then became clear that the only company which thought it might be able to take all of the cargo was Theobroma BV, a cocoa product specialist. The Bank had discussions with them for several months, but in the end they decided only to take a portion of the cargo. A sale of cocoa butter was made in October 2016 to Theobroma, and in January 2017 there was a sale to Theobroma of all of the Bank's cocoa beans.
  1035. One issue that was also discussed with Mr. Johnson in September 2016 was whether Transmar could buy back certain cocoa products rather than rolling those deals. This was done in order to try to reduce the Bank's overall position in a gradual way. Mr. Johnson said, however, that Transmar did not have any exchange deliverable beans, nor did it have sufficient liquidity at that point in time to pay the cash equivalent. This news was of concern to the Bank. However, the Bank took the view that if the swap options were exercised at this point, it would not have achieved anything positive. It would simply have made it more likely that Transmar would go into immediate default. It made no commercial sense to take that approach. The Bank did not think, however, that the situation at Transmar was so serious that the company would be insolvent three months later. Transmar had not committed any defaults, but in fact was continuing to service its deals with the Bank by making substantial netting payments and repurchasing goods which had been financed. It was not until December 2016 that the Bank discovered the true severity of Transmar's financial difficulties.
  1036. In those circumstances, Ms. Franssen's recommendation was that the expiring Transmar deals should be rolled, but for a shorter tenor than usual and with a higher commercial margin. The rationale was to give Transmar the opportunity to improve its liquidity, by trading through what was understood to be a temporary difficulty caused by supporting Euromar. Between 18 August and 2 December 2016, the roll of transactions resulted in £ 8.128 million and USD 6.050 million being paid to the Bank.
  1037. As far as hedging was concerned: the Bank was not a trader and did not enter into futures positions for its own account. The Bank had never done this. Instead, the Icestar 2 transactions had involved an "exchange for physical", whereby the Bank took over its clients' futures contracts whilst it financed the goods, and transferred the futures position back to its client when the client bought back. Icestar had never independently gone into the market to hedge a physical position itself. Had it done so in August 2016, it would have been taking a speculative position in a market that could have gone up rather than down.
  1038. It was in December 2016 that the Bank realised that Transmar was in serious financial difficulty. The deal with Itochu had not completed and the Bank had not been able to find other buyers for the remaining stock. At the end of December 2016, Transmar did not exercise 12 of its call options and breached its obligations when the Bank exercised its corresponding swap options. Transmar filed for Chapter 11 bankruptcy on 31 December 2016.
  1039. Ms. Franssen expanded on some of this evidence during cross-examination by Mr. Parsons, where points were put to her as to the steps that the Bank should have taken. She accepted that, following Euromar's default, there were questions in the Bank's mind as to the viability and liquidity of the whole group, and that these needed to be investigated. She referred to the meeting at the end of August with other banks. The discussion at that point was about how the group could help Euromar through its liquidity crisis, and the position appeared to be that Transmar was supporting Euromar and looking for ways to help.
  1040. She did not accept that the first thing that the Bank should have done was to audit the quality of the collateral held. She said that the first priority was to make sure that the goods were there, and to make sure that the warehouses did not act on the instructions of Transmar and Euromar. That was done. The Bank did ask Euromar for quality certificates. This was one way of gathering information. Discussions took place with some of the Bank's other clients, who knew the market much better and had a much better idea about how to start looking at liquidating a large position. The Bank were not commodity traders, and therefore did not have experience of dealing with the situation in which it found itself. They had suspicions about the quality certificates that were supplied to them by Euromar, because they bore a relatively recent date. But the Bank did not think it appropriate to carry out a full audit of the stock at that stage. They were told that if anyone was going to buy the goods, they would assess the quality themselves. There were a lot of processes that had to be thought about. The Bank was checking the legal position; obtaining warrants; checking the quantity; talking to different parties in the market as to what to do. They sent the quality certificates to some of their trusted counterparties, and none of them suggested that the first thing that needed to be done was to send in an independent quality assessor.
  1041. Ms. Franssen did not accept that the obvious thing to do was to exercise the swap options immediately. The Bank did ask, as the Euromar repo transactions expired, Transmar to repurchase the goods. But they were told that they could not do so: the support which they were providing to Euromar meant that there was a technical default under their borrowing base facility, and they did not have access to any further liquidity. The fact that there was a default by Euromar did not, however, lead automatically to the conclusion that all the swap options with Transmar should be exercised. The position was carefully considered in conjunction with the recovery and restructuring team. The Bank was still hopeful that, together with the company, a solution would be found and they would be able to repurchase.
  1042. As far as hedging was concerned, Ms. Franssen said that if a hedge had been taken out, whilst they were still talking to Euromar (who were obliged to buy on a fixed price), the Bank would be exposed if the market had gone up and this would potentially create an additional loss.
  1043. H4: The expert evidence

  1044. A considerable volume of expert evidence was served in relation to the sue and labour arguments.
  1045. The Bank's expert was Mr. Gordon MacLeod. He was an accountant by training, but had spent much of his working career within Credit Suisse. He had various roles during the 20 years that he worked there, between 1994 and 2014. He has subsequently worked as a consultant, including for HSBC and Deutsche Bank. His work extended far beyond accounting or accountancy roles. For example, he worked as a director of "Strategic Valuation Management" and later of "Strategic Risk Program" at Credit Suisse. This involved advising senior management on how to manage valuations and to some degree risk. He had considerable experience with credit, including when counterparties defaulted on credit contracts. Whilst he had not been the decision-maker on whether to enter into particular contracts, he had often been involved in investigating matters after the fact, and in advising management on what went wrong and what actions needed to be taken to remediate those things.
  1046. Mr. MacLeod was a thoughtful and careful witness. His answers in cross-examination were measured. I was impressed by his evidence. His experience of working in different roles inside a bank for 20 years meant that he was in a position to offer valuable evidence as to how reasonable bankers might be expected to react to particular situations.
  1047. The underwriters called two expert witnesses, Mr. Angus Kerr and Ms. Catherine Jago. Neither of these witnesses had any experience of working within a bank. Mr. Kerr was a very experienced commodity trader, having spent some 50 years in the soft commodities business. He had been chairman of the UK Coffee Trade Federation, and had held other senior positions in the industry. Ms. Jago had worked for over 38 years in the commercial side of the oil, commodities and shipping industry. She now works an expert witness, as well as arbitrator, in the fields of oil and commodity pricing, trading, hedging and shipping.
  1048. Given their backgrounds, I considered that Mr. MacLeod was in a position to provide more persuasive evidence than the underwriters' witnesses as to how reasonable bankers would be expected to act in the situation in which the Bank found itself. He was also, as I have said, an impressive witness who gave his evidence well, taking a restrained and analytical approach to the questions which he was asked. By contrast, it seemed to me that Mr. Kerr had formed strong views as to what the Bank should have done, and these were sometimes expressed in a somewhat uncontrolled fashion. This was not so with Ms. Jago, who gave short answers to the questions asked in a relatively brief cross-examination, and who explained the principles of hedging (both in her written reports and oral evidence) very clearly.
  1049. Ultimately, this is not a negligence case where I am required to decide between the opinions expressed by two experts who disagree as to whether a defendant fell below the requisite professional standard. It is a sue and labour case where the question is (leaving aside the issue as to whether the standard is "good faith") whether the bank failed to take steps, to act to avert or minimise loss, which any prudent uninsured would have taken; or, as Arnould puts it, failed to take such obvious steps as any prudent uninsured could be expected to take. On that issue, I have had evidence from an impressive witness who has worked within the banking industry for many years. That evidence has not been countered by any witness called by the underwriters with equivalent expertise. In those circumstances, where Mr. MacLeod had expressed views which are favourable to the Bank on the relevant question, and has provided an explanation as to why the proposed steps were not obvious or required, I consider it appropriate to accept his evidence. I have no reason to reject it.
  1050. H5: Is the applicable standard "good faith"?

  1051. In the course of his oral closing, Mr. Parsons (rightly in my view) accepted that Clause 3 of the General Conditions qualified Clause 16 of the ICC (A) Clauses, thereby in effect reducing the sue and labour duty to one of good faith. Ms. Sabben-Clare in her reply submitted that this put an end to the sue and labour argument; because there was no suggestion that the Bank had acted other than in good faith in the decisions which it had taken. She submitted that the contractual bargain was to indemnify the Bank if there was an insured event, even if the insured subsequently fails to avert or minimise loss, so long as good faith is exercised. In effect, the insured's own conduct is insured, so long as it acts in good faith.
  1052. Mr. Parsons submitted that this was not the end of the argument. The duty to sue and labour had been treated in the cases as a causation issue. That would not change if the obligation was only to act in good faith. There might not be a breach of the obligation to do so, but the causation issue would remain. As he put it: "It wouldn't be a breach, but it would still be a break in the chain of causation". He submitted that since it was a causation point, the actions of the assured can break the chain of causation. The straightforward question was: what had caused the loss? There could be actions of the assured which broke the chain of causation, even if those actions were taken in good faith and therefore did not amount to a breach of Clause 3.
  1053. In my view, Ms. Sabben Clare's submission is correct, and therefore – in the absence of any argument that the Bank failed to act in good faith – that is indeed an end of the sue and labour point. I do not consider that the causation issue can be divorced from the terms of the contract as a whole, as Mr. Parsons' argument posits. The parties have agreed that the duty to sue and labour should not be the ordinary duty to take reasonable steps, as provided for in s 78(4) and clause 16 of the ICC (A) clauses, but rather a modified and lower threshold of good faith. If there was no failure by the Bank to fulfil its obligations in that regard, then it would in my view be strange and wrong to conclude that the Bank's conduct was such as to break the chain of causation.
  1054. I was not referred to any authority concerning the relevance of contractual terms to questions of causation, but I think that it is a fairly obvious proposition that questions of causation need to be considered in the context of the terms which the parties have agreed. If authority is required, then it can be found in the judgment of Devlin J in Royal Greek Government v Minister of Transport (1950) 83 Lloyds LR 228. He emphasised that in seeking the proximate or effective cause under an indemnity provision, just as in the case of "any other contract of insurance" (at p. 238), the contract has to be construed. Speaking of contracts generally, he said (at p.237):
  1055. "'When questions of causation arise in contracts, points of construction are often involved. The contract defines the event which sets in motion the train of consequences. If in this respect the contract is misconstrued and the angle of view is, as it were, incorrectly plotted, the view will be wrong ... "
    The terms of the contract may restrict or expand the field of causation which has to be examined. There is no rigid rule that a cause to be operative must be the proximate cause. The rule is based on the intention of the parties: Reischer v. Borwick, [1894] 2 QB 548 , at p. 550. It is always subject to the contract: see Marine Insurance Act, 1906, Sect. 55 ... It is rarely that the contract expressly limits the field of causation, but it frequently does so by implication ...'
  1056. The relevance of contractual terms to causation questions can, in my view, also be seen in the Supreme Court decision in Arch. For example, at paragraph [166] of the leading judgment, the court referred to Lord Shaw in Reischer v Borwick:
  1057. "He made it clear, first of all, that the test of causation is a matter of interpretation of the policy and that "[t]he true and the overruling principle is to look at a contract as a whole and to ascertain what the parties to it really meant"."
  1058. Accordingly, I consider that Clause 3 provides a short and complete answer to the underwriters' sue and labour defence. However, for the reasons set out below, I consider that the defence would in any event fail, even if the ordinary sue and labour standard were applied. I do not think that, in relation to any of the steps on which the underwriters rely, it can be said that the Bank failed to take steps to act to avert or minimise loss which any prudent uninsured would have taken.
  1059. H6: The Bank's conduct

    Failure to exercise the Transmar swap options

  1060. The most significant point in financial terms, at least potentially, is the underwriters' case that the Bank should have exercised the Transmar swap options; ie to have called for Transmar to deliver beans or cash by way of replacement for the products which the Bank held pursuant to the repo transactions. The shape of the debate on this issue between Mr. MacLeod and Mr. Kerr, as conveniently summarised in the experts' helpful joint memorandum, was as follows.
  1061. Mr. MacLeod considered that the most likely outcome, if Icestar exercised the swap options, would be that Transmar would have defaulted. It was therefore a reasonable strategy to roll the deals, whilst negotiating with Transmar.
  1062. Mr. Kerr considered that, being confronted, Transmar would have done everything possible to perform and/or make provisions, and the situation would be made clear. A clear default would be a last resort. Mr. Kerr considered that given Transmar's strength in the market, they were capable of arranging swaps and/or alternative financing. The Bank should therefore have exercised the swaps and then started negotiations on buy backs. Transmar would have attempted to negotiate with the Bank to avoid default.
  1063. In my view, it is very far from obvious that prudent bankers in the position of the Bank would have exercised the swap options. I accept Ms. Franssen's evidence that the issue was considered, but that it was (reasonably in my view) regarded as unlikely to achieve anything. The Bank's approach, as described by Ms. Franssen, was to see if solutions could be found in order to address the liquidity problems of Euromar. This approach did have the benefit of the Bank receiving continued co-operation from Transmar in the final months of 2016, with the consequence that the equivalent of US$ 15 million of payments were received when transactions were rolled over.
  1064. It seems to me that there were considerable potential downsides of taking the course which Mr. Kerr advocated. As Mr. MacLeod explained, if the options were exercised, and then there was a default, that would produce an outcome which was now known; but it was not a desirable outcome. If a default is forced, then that usually eliminates any possibility of the company paying later, and its assets would then be sold on a fire sale basis. It was therefore appropriate to try alternative methods to recover money owed, because (as Mr. MacLeod put it) "often they lead to better outcomes". If a default was forced, then that would usually be terminal; because typically other borrowings have cross-default provisions, so that all of the company's debt would immediately go into default.
  1065. Mr. Kerr's evidence recognised the importance of continued negotiations. His point was that the Bank should have exercised the swap options, but continued to negotiate with the pressure on the negotiation having been provided by the exercise of the swap options. It seemed to me that this was a possible strategy, but that it was not without potential downsides in terms of potentially losing Transmar's co-operation, and ultimately leaving the Bank with no practical choice but to declare a default if and when Transmar failed to perform. Ms. Franssen's evidence in re-examination was that: "we knew that if we would exercise our swap option, we would put the company in default and likely lose all co-operation we were still having at that point in time".
  1066. Ultimately, I do not consider that the Bank's approach can fairly be criticised. The question of whether the Bank should (as Mr. Kerr proposed) play hard-ball, or take the more conciliatory approach which the Bank in fact took, involves a matter of judgement. I do not think that it is reasonable to criticise the approach taken by the Bank. The most that can be said, as Ms. Sabben-Clare submitted in her written closing, was that the Bank might have acted differently. But this is very different from saying that any prudent uninsured would have done so, or that this was an obvious step to take. She also made the reasonable point that if there really had been some obvious steps which the Bank had failed to take, one would have expected them to have been identified and pleaded long before they were ultimately raised in the spring of 2020; particularly bearing in mind that the underwriters had asked Gray Page to investigate the position in early 2017.
  1067. I was also unpersuaded that this argument, even if otherwise sound, led anywhere in terms of quantum. The Bank submitted that there was no basis for thinking that Transmar would actually have delivered beans or paid more money if the swap options had been exercised earlier. An enormous quantity of beans would have been required in order to enable Transmar to fulfil its obligations, if the swap options were exercised. But Ms. Franssen's evidence was that she was told by Transmar in September 2016 that it did not have any exchange deliverable beans, nor sufficient liquidity at that point in time to pay the cash equivalent for the beans. In my view, it is more likely than not that, as Mr. MacLeod said, the exercise of swap options by the Bank would simply have accelerated the events of December 2016 when Transmar defaulted and filed for Chapter 11 bankruptcy.
  1068. A separate question arose as to whether, even if no beans or cash had been provided by Transmar, the exercise of the swap options would have been beneficial; in that it would have led (following, on this hypothesis, Transmar's default in failing to provide beans or cash) to the collateral being sold more quickly and, more importantly, at higher prices. The underwriters' main point here was that there was a significant drop in the market price of cocoa beans and cocoa products after August 2016, and therefore that (on the basis of certain assumptions) an additional £ 16.9 million (less US$ 3.5m) plus a £ 4.6 million hedging gain would have followed from earlier action and sales.
  1069. Mr. Kerr's evidence in his second report, confirmed in cross-examination, was that the true quality of the stock did not materially change between the period from late 2016 through until the time it was eventually sold. It would therefore have been the same or very similar value throughout this period, and would not have had a substantially higher value at the earlier time, despite being six months younger. It was for that reason that Mr. Kerr took the prudent view that, at most, the value as compared to the market price would have increased by a small amount of around 5% (which he regarded as generous) had it been sold promptly after the first default.
  1070. In the light of this evidence, I could not see any basis to conclude that any delay in selling the cocoa products – whether or not preceded by the exercise of the Transmar swap options and a default by Transmar – led to a material or quantifiable diminution in the price ultimately achieved, notwithstanding that the market for sound cocoa products fell after August 2016. The evidence indicates that the goods were unsound when they were sold, and were not in a materially better condition in August or September 2016. As Mr. Kerr said in evidence, the goods were awful when sold in 2017 and awful if sold in late 2016 too.
  1071. As far as concerns the 5% differential: this was, as Mr. Kerr said, not a scientific calculation but was a nominal allowance based upon the fact that the actual sales were fire sales. However, as the Bank submitted and I accept, any sale in 2016 – particularly if it involved offloading very large quantities of stock at one time – would have been just as much a fire sale as those which occurred at a later time.
  1072. A subsidiary point was that an earlier sale would have been beneficial because, even if higher prices would not have been achieved, there would have been some savings in storage costs. This point has validity, at least in theory. It is, however, unclear on the evidence whether any earlier sale of the sub-standard products could in practice have been achieved. I need not, however, consider that issue further because I do not accept that there was a failure by the Bank to sue and labour in the relevant respects.
  1073. The calculation of loss in the present context also includes £ 4.5 million which, according to Ms. Jago, could have been achieved if the Bank had promptly opened a hedge on 17 August 2016, and then closed that hedge during the period 23 August to 22 September 2016 when, on the present hypothesis, the cocoa was being sold following Transmar's failure to deliver beans or cash. Since I have rejected the underlying premise – namely that the Bank should have taken the step of exercising the swap options – I will not address the question of this proposed hedge (which was not explored in the cross-examination of the experts) in detail. Suffice it to say that that argument raises, at least to some extent, issues similar to those raised in relation to the failure to take out prompt hedges for the Euromar deals. I deal with that case below.
  1074. Failure to audit the stock

  1075. The underwriters submitted that it was an obvious step for the Bank to check immediately the quality and value of the product held as collateral. Had it done so, it would have quickly realised the risk of very severe loss if there were further defaults. This was coupled with an argument that the Bank should have taken steps immediately to liquidate the products following Euromar's and Transmar's defaults.
  1076. I do not accept this argument. The case was, as with all the points raised by the underwriters, unsupported by any evidence that reasonable bankers, following a default by a customer, would usually do this or be expected to do this. Mr. MacLeod's evidence was that he had reviewed the totality of the actions taken by Icestar after the default of Euromar and Transmar, and that – based on his experience of banks seeking to dispose of distressed collateral – he believed that Icestar's actions were reasonable and well within the boundaries of the normal and ordinary responses of a lender attempting to sell collateral after a borrower's default. I accept this evidence and do not consider that there are any sound grounds on which to reject it.
  1077. It seemed to me, having heard Ms. Franssen's evidence, that the Bank was placed in a position which was not straightforward, and of which those responsible did not have prior experience. They were in the position of potentially needing to liquidate a large amount of stock which Euromar had failed to repurchase. As summarised by Ms. Franssen in re-examination, the Bank took a number of steps; sending inspectors in to check that all the goods were there; ensuring that the warehouses held the goods to their order; consulting with different law firms in Germany, Holland, Belgium, the US and the UK, to make sure that the contractual framework was in order and that they had the right to sell goods; requesting warrants for the goods; asking for quality certificates; engaging in discussions with Euromar and Transmar, the Chief Restructuring Officer, and other bankers, in relation to Euromar's liquidity issues and the position generally; speaking to their other clients in relation to the possibility of selling the goods, and accomplishing some sales in late 2016 and early 2017. Mr. Stroink supplemented this evidence by describing how he had arranged the first two sales, of cocoa butter and cocoa beans, by speaking to his contacts at Theobroma.
  1078. I was not persuaded that the Bank failed to take any obvious steps. Indeed, as I have pointed out, the alleged obviousness of the steps which should have been taken is undermined by the fact that those steps were not identified by underwriters themselves until very late in the litigation.
  1079. Nor, for reasons which I have explained, do I see that this point leads anywhere in terms of quantum. If the problems with the quality of the goods had been identified at an earlier stage, the goods would not have been materially more valuable. If the identification of these problems at an earlier stage had led to an earlier recognition that the problems with Transmar were very severe, this would at best have led to an acceleration of the events which took place in December. I do not see how it would have assisted in diminishing the Bank's losses. There may conceivably have been a saving of some storage costs, but this depends upon whether the goods could in practice have been sold earlier. The evidence did not seem to me to yield a clear answer to that question, and in any event the savings would not have been substantial in the context of the claim.
  1080. Failure to hedge

  1081. Mr. MacLeod's evidence was that upon default of a borrower, the lender's primary consideration is to recover the money owed, either directly from the borrower or from liquidation of the collateral, so as to avoid financial loss. It is also normal practice to rely on the so-called 'buffer' – ie the excess value of the collateral over the amount owed – to protect against falls in the value of the collateral and potential loss during the sale process. It was not standard practice for a lender to hedge illiquid physical collateral upon default of a borrower. In all the circumstances, he did not believe that Icestar acted unreasonably by not entering into hedges of the Euromar deals.
  1082. Again, I see no reason why this evidence should be rejected, and I accept it. For reasons already given, Mr. MacLeod was in a far better position than the underwriters' experts to give evidence as to the usual practice of bankers. Ms. Jago's comprehensive evidence relating to hedging seemed to me to establish no more than that some banks might possibly have taken out a hedge at some point after the Euromar default.
  1083. It also seemed to me that there were sensible reasons, identified in the evidence of Ms. Franssen and Mr. MacLeod, as to why banks generally, and the Bank here, would not engage in a hedging exercise.
  1084. It is relevant that the Bank did not have experience of taking out hedges. The Bank did have hedges for the Transmar deals, but this was because hedges had been transferred by Transmar to the Bank for the duration of the transaction. Accordingly, the underwriters' argument posits that the Bank should have taken the "obvious" step of entering into futures transactions of a kind which it had not previously entered; and that this should have been done when, as shown by Ms. Franssen's evidence summarised above, the Bank was focused on various issues arising from the difficult position in which it found itself. This is an unpromising starting point for the argument.
  1085. Furthermore, if hedges had been taken out for the Euromar financing, they would have been at best "imperfect" and would therefore potentially expose the Bank to further loss. The Transmar hedges transferred to the Bank operated as "perfect" hedges; because the contractual price repayable to the Bank was not fixed but varied according to the futures market price of cocoa beans (which was the subject of the hedge). This meant that if there were market movements, they would operate in the same way in relation to the price payable by Transmar and the pricing of the hedge. The Euromar transactions had never been hedged, because the price payable was a fixed repurchase price. The underwriters' argument is, therefore, that – for the first time – the Bank should take out a cocoa beans hedge in respect of the cocoa products. This would result in products being hedged by beans. However, the nature of the market was such that there could be significant changes in what was described as the "ratio" between the prices of products and beans. This ratio was not fixed: it could fluctuate, depending for example on whether there were surpluses or shortages of products. In other words, products may be more or less valuable, when compared to beans, depending upon how the market moves. This means that if cocoa products are hedged by reference to cocoa beans, the Bank is potentially exposed to loss caused by adverse changes in the ratio between these two different commodities. It is difficult to see why the Bank, as a prudent uninsured, should have exposed itself to this potential loss.
  1086. In these circumstances, I was not persuaded that any prudent uninsured in the position of the Bank would have taken out hedges, rather than concentrating on the issues on which Ms. Franssen was focused including in particular taking steps to sell the collateral provided by Euromar.
  1087. There was at least one other risk, associated with the taking out of a hedge, which would have meant that it was not prudent to do so, although the evidence did not suggest that this was a reason why the Bank did not take out a hedge in the present case. However, it may assist in explaining why, as Mr. MacLeod said, it is not standard practice for a lender to hedge illiquid physical collateral upon default of a borrower. Where a bank holds illiquid collateral (eg cocoa products when contrasted to exchange-deliverable beans), it would be important to know the actual quality of the products. The reason is that if there are problems with the quality of the product, the value of the product would not (even leaving aside the question of the ratio described above) move in line with change in the market price of beans. If the products were sound, then a rise in the value of cocoa beans might be expected to produce a rise in the value of the collateral held; although the potential movement in the ratio may mean that this rise is not linear. However, if the products are not sound, then a rise in the market for beans would make the hedge loss-making, but this would not be compensated for by any equivalent rise in the value of the products.
  1088. Mr. MacLeod was also able to identify other difficulties with the suggestion that hedges should, as Ms. Jago suggested, have been taken out promptly. The Euromar deals did not automatically come to an end on the day that Euromar failed to repay the price. The default was a breach, but not a repudiatory breach, and each forward contract sale remained alive. They were not terminated until October or November 2016. It would therefore have been wrong, or not sensible, for the Bank to have hedged on the moment of default. If Euromar had paid late, there would have been an obligation on the part of the Bank to deliver the cargo to them. But the Bank would then be left with an open futures position, and it would suffer a loss if the market price of beans rose.
  1089. Conclusion

  1090. Accordingly, even if the causation issue involves (as underwriters submit) the application of the ordinary sue and labour test, without regard to the good faith qualification in Clause 3, I do not consider that there was any failure by the Bank to take appropriate steps. Nor, except in relation to hedging and possibly storage costs, is there any basis for the conclusion that the proposed steps would have made any difference to the losses suffered by the Bank.
  1091. It is therefore unnecessary to address the Bank's argument that no duty to sue and labour arose in relation to Transmar in August/ September 2016, because there was no imminent peril.
  1092. I: Quantum of the claim against underwriters


     

  1093. In the light of my conclusions in the previous sections, the claim against underwriters succeeds in full, except in relation to Ark and Advent where it fails because of the estoppel on which those underwriters can rely.
  1094. Subject to the defences which I have considered, the quantum of the Bank's claim was admitted by underwriters.
  1095. The Bank's primary claim was for the loss of Transaction Premium as defined in the TPC, ie the difference between the Pre-Agreed Price and the Actual Sale Price for the cargo affected by the Transmar and Euromar defaults. The Bank was willing to give credit to the Underwriters for the profits it was able to make by unwinding the Transmar hedges that were left on its hands when Transmar defaulted.
  1096. In consequence the agreed amount of the Transaction Premium is:
  1097. (1) £38,919,873.09 and US$1,149,919.76;
    (2) less the amounts of £5,674,587.80 and US$4,285,772.57 (these amounts encompassing both sale proceeds and the proceeds derived from unwinding the Transmar hedge transactions).
  1098. The Bank is therefore entitled to the sterling equivalent of each underwriter's share of those sums, save in respect of Ark and Advent.
  1099. In addition, the Bank claimed the amounts of £2,521,868.04 and US$412,899.36 as sue and labour expenses, being the costs of warehousing and insuring the affected cargo in the period after the relevant Euromar and Transmar defaults. Again, these figures were not disputed as figures, and no argument was addressed as to their recoverability save for the points addressed in Sections G and H above. Each underwriter, save for Ark and Advent, is liable for its respective share of those amounts.
  1100. The Bank is also in principle entitled to claim, as against all underwriters, apart from Ark and Advent, interest on the amounts due under the policy. Any issues relating to the interest, including rate and period, can be addressed at the hearing that will follow this judgment.
  1101. J: The claim against Edge


     

    J1: The issues

  1102. The claim against all underwriters, except for Ark and Advent, has succeeded. The claim against Ark and Advent has failed because of an estoppel arising from statements made by Edge at the time when the 2016 policy was renewed, and which has the effect of precluding the Bank from relying upon the TPC in relation to those underwriters. There was no dispute, subject to issues of quantum, that Edge would indeed be liable in the event that the underwriters' defences based upon rectification/ estoppel/ collateral contract were to succeed. In relation to those two companies, the estoppel defence has succeeded and accordingly Edge is liable for the Bank's losses arising from their inability to make a successful recovery against Ark and Advent.
  1103. In respect of Ark and Advent, the quantum of that recovery against Edge should, in my view, be based on the full amount of the cover that was provided by those two underwriters, without a discount in respect of the matters discussed in Section J3 below. The reason is that the 2016/2017 slip was substantially oversubscribed. If Ark and Advent had declined to write the 7.5% share that each of them wrote, the Bank would still have had subscriptions to the cover which exceeded 100%. There is also a potential liability for any costs payable by the Bank to Ark and Advent, or any irrecoverable costs which the Bank has incurred in pursuing those underwriters. Any argument as to the extent of such liability is a matter which can addressed at the "consequentials" hearing in the light of this judgment.
  1104. As far as the majority of underwriters are concerned, where the claim has succeeded, it is nevertheless appropriate, for reasons which will become apparent, to consider the issues concerning Edge's liability to the Bank, although it is not necessary to resolve every issue that has been argued. Those issues are addressed in the remainder of this section J.
  1105. It was common ground on the pleadings that Edge owed duties of reasonable skill and care to the Bank: to procure the insurance cover required by the Bank; and to procure cover that clearly and indisputably met the Bank's requirements, and so did not expose it to an unnecessary risk of litigation.
  1106. There was also no dispute that Edge was told, and therefore understood or at least should have understood, that the Bank required to have cover against the consequence of a client defaulting under a repo transaction, and that such cover was not to be dependent on the occurrence of physical loss or damage to the cargo. Edge was explicitly told this in, for example, the e-mail sent by Lawar Barnes on 10 July 2015 in response to Mr. Mullen's questions. It was reiterated in January 2016 when Ms. Van de Beek sent a draft of the policy with the question in a comment box:
  1107. "David. Please confirm: is it appropriate for the Transaction Premium cover to appear up front, and is it clear that this cover is separate to the marine cargo and storage cover. Just like CEND and Business Contingent cover".
  1108. It was this comment which led Mr. Lockyer to write his e-mail to Mr. Mullen on 28 January 2016, saying:
  1109. "I have seen the comments made by Pauline and it appears her main concern is regarding Transaction Premium. Reading the clause it mentions that the Insured is covered by this policy for the transaction premium they would have earned if client of the insured defaults, regardless whether there has been any physical loss or damage[d] to the goods.
    Am I reading this correctly, and is this understanding of underwriters".
  1110. Mr. Mullen's response was to give a positive response to Mr. Lockyer's questions:
  1111. "No you are correct but they appear to want this as a separate section. I saw Brian with this and his view is "they have the coverage and they should be satisfied". Anyway nothing much to worry about".
  1112. Mr. Lockyer's evidence was that he was reassured by Mr. Mullen's response. He was then asked questions by Ms. Sabben-Clare as follows:
  1113. Q. If he had come back and said, "I have no idea; I have not asked the underwriters," your attitude as a competent broker would have been, "Well, we need to draw this to their attention, discuss it with them, and make sure they're happy with it, wouldn't you?"
    A. If that scenario was presented to me, then we would have looked into that further, yes.
    Q. And you'd have gone to the underwriters?
    A. Yes.
    Q. And made sure they were happy.
    A. Yes.
    Q. And made sure they clearly understood what they were meant to be covering.
    A. Yes.
    …
    Q. If Mr. Mullen had not done so before, would you agree that Mrs. Van de Beek's comments should have prompted him to double-check with underwriters that they really understood the risk the client wanted to bind?
    A. If David didn't have that conversation with the RSA, then, yes, we should have re-presented that to the RSA.
  1114. Against this background, the Bank contended that if, for some reason, the TPC was ineffective and did not cover the present claim, then Edge was in breach of its duty to procure the insurance cover required by the Bank, and indeed to procure cover that clearly and indisputably met the Bank's requirements.
  1115. I have concluded that the TPC was applicable, and accordingly it is strictly unnecessary to express a view on the question of whether the claim against Edge would have succeeded in the event that the underwriters' construction argument had been successful.
  1116. It is, however, appropriate to address that issue, not least because the issue is closely tied to an issue which does require resolution. The Bank submits that even if the TPC does (as I have held) respond to its claim, Edge is nevertheless liable for any costs which are irrecoverable from the underwriters. The basis of that claim is the second aspect of the duty described above: ie to procure cover that clearly and indisputably met the Bank's requirements, and thereby not to expose the Bank to an unnecessary risk of litigation. This duty (which was not in dispute) has been recognised in a number of cases; see FNCB Ltd v Barnet Devanney (Harrow) Ltd [1999] Lloyd's Rep IR 459; Talbot Underwriting Ltd v Nausch Hogan & Murray Inc [2005] EWHC 2359 (Comm) (Cooke J); Standard Life Assurance Ltd v Oak Dedicated Ltd [2008] EWHC 222 (Comm) (Tomlinson J). The Bank therefore contends that any irrecoverable costs of pursuing the underwriters will be recoverable from Edge for breach of the FNCB duty.
  1117. Edge contends that the Bank's claim should be rejected for a variety of reasons. In broad summary, Ms. Healy submits that even if the TPC did not provide the coverage which the Bank wanted, this cannot be laid at the door of Edge. This is essentially because the TPC had been drafted by NRF, and it was to NRF (not Edge) that the Bank was looking in order to ensure that its interests were protected. Edge did not fall below the standards to be expected of a reasonably competent broker. In the alternative scenario, where (as I have held) the TPC did provide the coverage which the Bank was seeking, Edge cannot be held responsible for the consequences of the underwriters advancing spurious arguments.
  1118. The Bank also advanced claims against Edge in the event that the underwriters' other defences (apart from sue and labour – see Section H above) succeeded. As I have said, there was no dispute (subject to quantum) that Edge would indeed be liable in the event that the underwriters' defences based upon rectification/ estoppel/ collateral contract were to succeed. In respect of two underwriters (Ark and Advent) the estoppel defence has succeeded. Similarly, there was no dispute that Edge would be liable (subject to quantum) in the event that any of the underwriters could successfully avoid for non-disclosure or misrepresentation. The Bank contended that Edge would also be liable in the event that the underwriters were successful in their defence based upon Clause 3 and the Bank's failure to take reasonable steps in relation to ascertaining the quality of the goods. Edge contended that any such claim would fail.
  1119. In addition to the issues which arose on liability, or which would (if the underwriters' defences had succeeded) have arisen on liability, there was a substantial argument relating to the quantum of any recovery by the Bank. The Bank contended that it would (if the defences had succeeded) have suffered substantial loss as a result of Edge's breaches. Had Edge performed its duties, the Bank would have sought and purchased alternative insurance cover, from specialist credit underwriters, which would have provided substantial protection against the defaults of Euromar and Transmar. If such cover had not been available for some reason, then the Bank would have taken steps in 2015 and 2016 to ensure that it was no longer exposed to those companies: in effect, the Bank would have wound down its repo transactions with those companies before any default arose. These propositions were disputed by Edge. The arguments gave rise to a debate between two well-qualified experts, with considerable experience of the credit insurance market, as to what cover would have been available at the material times.
  1120. I will start by addressing the issues relating to the way in which the TPC was broked and the Bank's case that Edge (i) is liable for any irrecoverable costs in circumstances where the cover is effective, and (ii) would be liable for very substantial damages if the cover had been ineffective. The liability issues are addressed in Section J2 below, and the general causation and quantum issues in section J3. I then consider the claim for irrecoverable costs in Section J4 below. Finally, I briefly address, in Section J5, the question of whether Edge would be liable had I concluded that the claim failed because the underwriters were entitled to rely upon Clause 3.
  1121. J2: The broking of the TPC

    The Bank's case

  1122. The Bank submitted that if the underwriters succeeded in establishing that the TPC is inapplicable, there could be no doubt that this was in consequence of negligence by Edge Brokers. On that hypothesis Edge would have failed to obtain precisely the cover that the Bank wanted and instructed them to place.
  1123. Edge's duties in relation to the broking of the risk included obligations: not to broke a clause that the broker did not understand or for a client whose insurance needs the broker did not understand; to use in-house expertise and advise the client to seek specialist advice; to take all reasonable steps to ensure that the effect of the cover obtained was clear; to communicate the client's requirements for cover clearly to the underwriter; to take reasonable care to ensure that there was a fair presentation of the risk, including asking questions of the client to elicit material information.
  1124. The Bank submitted that Edge was in breach of each of these duties. Ms. Sabben-Clare summarised the Bank's case by saying that if Edge had acted competently, then by one route or another the Bank would have received advice that credit risk cover was available from the credit risk market. It would either have received that advice on what she described as a "do not pass Go" basis, because Edge would have involved an expert credit risk broker at the outset; or the advice would have been given in consequence of a competent broke. That would have included underwriters being told what it was that the Bank wanted. Any reasonably competent broker would have explained the purpose of the TPC to underwriters, particularly after Ms. Van de Beek had raised her question on 28 January 2016 but also before.
  1125. The Bank relied upon various deficiencies in the work carried out by both Mr. Mullen and Mr. Lockyer, including the following. Mr. Mullen did not properly understand the TPC. He failed to consult specialist credit risk brokers within Edge and to advise that specialist advice be sought. He and Mr. Lockyer failed to take any adequate steps to ensure that the effect of the TPC was clear and indisputable. Mr. Mullen had not understood the clause when he had first read it. This was a new clause, not something familiar to the market. It extended cover well beyond the ordinary bounds of cargo cover, as any reasonably competent broker would have appreciated. They should therefore have explained the intended effect to underwriters. They failed to communicate to underwriters that the Bank wanted credit risk insurance. When asked the direct question by Ms. Van de Beek, as to whether it was clear that the cover under the TPC was a separate insuring clause, any reasonably competent broker would have said: "No" in response to this. The involvement of NRF and the Bank's internal lawyers was no answer to these various breaches.
  1126. The Bank relied upon the evidence of Mr. Aidan Meldrum, a cargo insurance practitioner with 44 years' experience of working on both the underwriting and broking sides of the business, and who had spent the last 25 years as a broker with an emphasis on commodities. Mr. Meldrum's view was that Edge had not acted in the manner of a reasonably competent insurance broker.
  1127. Edge therefore missed two critical opportunities to make the position on coverage clear, both (1) in July 2015, when the TPC was first introduced into the Policy, and (2) in January 2016, at the time of the renewal. On both occasions, Edge's conduct of its client's business was seriously sub-standard.
  1128. As to July 2015, an approach whereby Mr. Mullen simply handed the wording over and made no more than the most generic of remarks to Mr Beattie, was perfunctory. This was not sufficient, given the importance of the cover and its unusual nature. It was incumbent on Mr Mullen to explain to the underwriters that this was credit risk insurance, and to obtain their explicit buy-in to that broadening of the policy's scope.
  1129. As to January 2016, the striking feature of the facts here is that on 28 January 2016 (as set out above) Ms. van de Beek specifically raised with Mr. Mullen whether it was appropriate for the TPC to appear "up front" and whether "it is clear that this cover is separate to the marine cargo and storage cover". The issue raised was in substance the same as that now contested by underwriters: had it been made sufficiently clear that the TPC added a freestanding head of cover, and was not merely something that applied in the context of standard marine cargo and storage cover?
  1130. A competent broker in these circumstances would have taken up this "red flag" specifically with underwriters and obtained their express agreement that the TPC did indeed provide independent cover. But this critical opportunity was completely missed. Mr. Mullen had done no more than to provide assurance to Mr. Lockyer. But he did not then bother to pursue the point with underwriters at all. There was no substantive discussion with Mr Beattie, but merely a discussion and agreement concerning the relocation of the TPC within the policy wording. These attempts to protect his client's interests were, again, perfunctory and superficial. Had Mr. Mullen done what he should have done, the dispute in these proceedings could never have come to pass. Mr. Beattie would either have confirmed that the TPC was a freestanding head of cover, in which case underwriters' case in these proceedings would have been unsustainable. Alternatively, he would have declined to do so, in which case the Bank would have taken steps to protect itself, including by taking out a separate policy of credit risk insurance.
  1131. Edge's case

  1132. Edge submitted that the Bank was relying upon NRF to secure cover for the loss of the transaction premium in the absence of physical loss or damage. NRF had drafted the TPC and advised the Bank about where it should be positioned within the policy, and whether it provided the cover that the Bank wanted. The Bank had put itself fully into the hands of NRF in relation to the coverage which the TPC provided. Any problems with the drafting of the TPC, or its effect within the policy, were therefore the responsibility of NRF, not Edge. Any claim would lie against NRF, if anybody. Edge could not have been negligent if leading insurance lawyers drafted and approved the TPC: a reasonable marine insurance broker could not be expected to achieve a higher standard on clarity of cover than a specialist insurance law firm. In this case, the Bank was relying upon NRF to obtain for it clear cover under the TPC, not Edge. Accordingly, if the TPC did not provide the cover that the Bank wanted, this was not the responsibility of Edge.
  1133. If (as Edge contended) the TPC did provide the cover that was wanted, Edge had no liability at all, for example for any irrecoverable costs of pursuing the underwriters. The FNCB duty did not apply to "spurious" construction arguments. The underwriters' construction arguments in the present case could properly be described as spurious.
  1134. Edge contended there was no duty on the broker to highlight and explain to underwriters the Bank's subjective understanding or intention about the terms of the policy. Edge relied upon the evidence of an experienced broker, Mr. Nigel Russell, in support of that proposition. This evidence was supported by Mr. Sutherland, who gave evidence on underwriting issues. Whilst Mr. Russell's broking experience was primarily in the marine hull market, they submitted that the broking practice in that sector did not substantially differ from the marine cargo market.
  1135. Ms. Healy characterised the duty contended for as a "duty to nanny", and submitted that this was insupportable and wrong for a variety of reasons. These included the following. The insurer was presumed to know his own business, and to be able to form his own judgment of the risk as presented to him, and should therefore read the proposed policy and ask questions as appropriate. Highly experienced and well paid marine cargo underwriters should read the slip with care, and no duty should be imposed which sanctions a race to the bottom in underwriting practice. The broker's duty of care must be capable of precise definition, and the factual and expert evidence at trial had revealed different formulations of what the broker would be expected to do. There was also uncertainty as to whose understanding of a clause should be explained to the underwriter: was it the broker's, or the client's or the client's lawyer? If an explanation were proffered, the broker might inadvertently explain the cover in a way that was inconsistent with its true construction. The "duty to nanny" might place a broker in a position of conflict: by flagging an unusual clause, the underwriter might be less likely to agree it. The prospects of litigation would increase. In any controversial case, the underwriters would assert that a particular clause was not flagged and explained to them. The supposed obligation on the broker would create two divergent sets of rules on what a broker would need to disclose: one under the 1906 Act (or the subsequent 2015 Act), and another under the duty of care in tort.
  1136. In summary, therefore, it was the underwriter's responsibility to read and understand the slip that he is presented for agreement. If there is something he does not understand, he may either refuse to agree it or ask the broker questions, which the broker must answer honestly. It is not the broker's responsibility to ensure the underwriter has understood what the underwriter agrees.
  1137. Approach to the argument

  1138. I will start by considering the arguments in the context of my conclusion that the TPC does indeed respond to the Bank's claim – a conclusion based in part on my conclusion that the language of the TPC is sufficiently clear. The Bank contends that even on those premises (which is the Bank's primary case) there is a claim over against Edge for irrecoverable costs, because Edge failed to take reasonable steps to prevent the Bank from being exposed to an unnecessary risk of litigation. I start there, because if there were indeed failures by Edge to broke the risk using appropriate reasonable skill and care even on the assumption that the policy responds, then it is difficult to see how there could be any defence on liability in the event that the policy were not to respond.
  1139. The reliance argument

  1140. The starting point here is that Edge rightly accepted, in its pleadings and at the outset of its opening submissions, that Edge owed its client a duty to protect it against the unnecessary risk of litigation. The duty to arrange cover which "clearly and indisputably" meets the client's requirements was discussed by Tomlinson J in Standard Life Assurance v Oak Dedicated [2008] EWHC 222 (Comm) paras [101] – [104]. He said at [102], having referred to the prior case-law:
  1141. "This body of authority establishes that it is the duty of a broker to obtain, so far as possible, insurance coverage which clearly meets his client's requirements. Coverage is only clear in so far as it leaves no room for significant debate. The coverage will be unclear, and the broker in breach of duty, if the form thereof exposes the client insured to an unnecessary risk of litigation. Of course the risk of litigation can never be wholly avoided and the broker is not in breach of duty in consequence alone of insurers putting forward a spurious construction of the cover. The present however is not a case in which it is necessary to explore the nature of the duty at its limits".
  1142. Tomlinson J went on to focus on the particular wording which was in issue in that case, on which the insured's construction argument had failed, and to ask the question: whether a reasonably competent broker would reasonably have concluded that the policy wording was sufficiently clear to meet the insured's requirements without exposing the insured to an unnecessary risk that the insurers might argue for a more limited form of cover. He answered the question in the negative: no reasonably competent broker could reasonably have come to the view that the insured's requirements were met.
  1143. In the light of the existence of the admitted duty of the broker to arrange cover which "clearly and indisputably" meets the client's requirements, and does not expose the client to an unnecessary risk of litigation, I do not accept that issues of reliance provide any defence to the Bank's claim that Edge acted in breach of its duties. Issues of reliance are potentially important where a party makes a claim in tort based upon the Hedley Byrne v Heller principles to which Edge referred. But I am here concerned with admitted contractual duties in the context of a contractual relationship under which Edge were well paid. I do not see how alleged lack of reliance by the Bank could provide Edge with a defence to an allegation that it acted in breach of its contractual duties.
  1144. In any event, I reject the argument that the Bank was not relying upon Edge in connection with the TPC or the policy amendments which were first introduced in July 2015 and then incorporated in the renewed policy in January 2016. I accept that the Bank was looking principally to NRF for advice in connection with the drafting of the wording of the proposed amendments. I also accept that it looked to NRF for advice when questions arose concerning whether and how the 2015 amendments had been incorporated into the policy renewed in January 2016. However, I do not accept that the Bank was looking exclusively to NRF. It is clear that the Bank was looking to Edge for its professional expertise and advice as well.
  1145. The fact that the Bank was doing so is demonstrated by the important question which Ms. Van de Beek asked "David" in the comment box of the redraft sent by the Bank on 28 January 2016. She wanted Mr. Mullen's view on the issues which she raised. Both Mr. Mullen and Mr. Lockyer understood at that point, as they had done before, that the Bank was indeed looking to them to consider the policy wording, and thereby perform their ordinary contractual duties as brokers to take reasonable steps to ensure that the cover met their needs. Ms. Van de Beek's question then led to Mr. Lockyer identifying the key issue: in his 28 January 2016 e-mail, he asked Mr. Mullen "is this the understanding of underwriters".
  1146. Neither Mr. Mullen nor Mr. Lockyer was working at the time on the basis that the Bank was looking, exclusively, to external legal advisers for professional expertise in connection with the drafting of the policy. Indeed, neither of them knew of the involvement of NRF. Mr. Mullen was under the impression that the Bank's legal advice was coming from Ms. Lawar Barnes. Mr. Mullen thought that she had limited expertise in insurance matters, and he would therefore have understood that the Bank was looking to Edge for their professional views and advice.
  1147. There was also an internal exchange between Ms. Van de Beek and Ms. Barnes on 29 January, following receipt of the email from Mr. Mullen which described where the TPC had been "deliberately sited". Ms. Healy submitted that the exchange showed that Edge were being side-lined in favour of NRF. I do not agree. Ms. Van de Beek said to her colleague that it was "not clear" to her if "David thinks that Clause 23 is the best place for the Transaction premium". She was, therefore, considering the advice which Mr. Mullen had given. She was paying it due regard, but was uncertain as to what he was saying. Ms. Barnes' response was: "perhaps have a chat with him but otherwise make the amendment as per NRF's advice". This does not show Edge being side-lined, but rather that Ms. Van de Beek should speak to Mr. Mullen in order to understand his point; but if not (or presumably in the absence of any good reason given by Mr. Mullen), then the amendment of NRF should be adopted. The e-mail exchange therefore went no further than Ms. Barnes expressing the view that NRF's draft was to be adopted, at least unless Mr. Mullen could explain otherwise.
  1148. The reliance placed by the Bank on Edge is also clear from the events following the meeting in Amsterdam in June 2015, when the TPC was first proposed for introduction into the policy along with the other amendments which had been drafted. The wording was given to Mr. Mullen. Both he and Mr. Lockyer knew that they needed to consider it carefully and to understand what it was seeking to achieve. Thus, on 9 July 2015, Mr. Mullen sent a series of questions and observations on the proposed wording, indicating that there might be the need for "further discussion either in London or the Netherlands". The reason for this e-mail was, in my view, that Mr. Mullen considered that he needed to understand what was being proposed, not only because he might have to answer questions from underwriters, but also because it was possible that the Bank's answers would give rise to further discussion and possible amendment. Ms. Barnes, unsurprisingly, gave a detailed response. She did not brush off Mr. Mullen's questions on the basis that the issues raised were no concern of his, because the Bank was not looking for Edge's professional assistance in connection with what was proposed. There was, therefore, on this occasion a correct appreciation by Edge that it had obligations to understand and broke the clauses properly, notwithstanding that the endorsement, including the TPC, was a draft emanating from the client. The Bank was therefore looking to Edge for its professional advice in June and July 2015, just as it was later in January 2016.
  1149. Thus, in my judgment, Edge owed duties to the Bank and these did not reduce, still less disappear, because NRF had drafted the TPC.
  1150. Accordingly, the important question is what, in the circumstances of the present case, Edge was required to do in order to fulfil its duty to arrange cover which clearly and indisputably met the client's requirements, and did not expose the client to an unnecessary risk of litigation. This issue was the focus of the expert evidence of Mr. Meldrum for the Bank and Mr. Russell for Edge. Although Ms. Healy raised questions as to the adequacy of the Bank's pleadings, when compared to the case ultimately advanced, I did not consider that there was any substance in that point. The critical issues, concerning what Edge should have done in order to fulfil its duties to the Bank, were in my view fully ventilated in the expert evidence and I do not consider that Edge can have been taken by surprise in relation to the case advanced. Both experts therefore expressed their views as to what, in practical terms, the admitted duties required of Edge.
  1151. The expert evidence - overview

  1152. In their joint memorandum, the two experts identified what they described as a "fundamental disagreement" about the responsibilities of a reasonably competent broker.
  1153. "2.2 Mr. Russell expressed the view that a competent broker had an overriding duty to follow the instructions of the client, with the requirement to inform the underwriters of material facts and to answer honestly any questions from those underwriters. The broker did not have a duty to explain written clauses, conditions or endorsements to underwriters but, if questioned, did have a duty to answer honestly.
    Mr. Meldrum defined the role of a competent broker more widely in that the interests of an insured had to be protected and that a broker should take steps to ensure the client was properly insured. This role required the broker at times to act without the express instruction of the client, but in what the broker would deem to be in the best interests of his client".
  1154. It seemed to me that the central disagreement in the joint report, as later explored in the oral evidence, concerned the expert evidence given in response to the question: "Was Edge obliged to make clear to RSA the intended meaning and effect of the TPC and the non-avoidance clause". The joint report identified the experts' main answers to that question, and the extent to which they disagreed, as follows:
  1155. "4.1 Mr. Russell said that a broker had no obligation at any stage to explain a written endorsement or renewal slip or clauses within, but would have to answer underwriters' questions honestly. He maintained that a broker might volunteer information or explanations, but was not obliged to do so.
    4.2 Mr. Meldrum said that as the cover proposed in the TPC was not customarily placed in a cargo insurance market, Mr. Mullen should not have only advised his client accordingly, but recommended that the cover should be placed in the specialist credit insurance/ financial risk market.
    4.4 Mr. Meldrum said that Mr. Mullen should have explained the proposed coverage under Endorsement 5 to RSA and sought the agreement of the other subscribing underwriters as it was a type of cover which was not customarily expected to be included in a cargo risks policy. Further the TPC should have been explained to all the underwriters of the 2016/17 renewal.
    4.5 Mr. Meldrum said that Mr. Mullen had an over-riding responsibility to ensure that the client was properly protected by explaining the TPC to all of the underwriters, especially as Mr. Mullen would or should have been aware that the supporting market were not familiar and/or in many cases licensed to transact financial risk business.
  1156. This evidence therefore focused on the actions of Edge, and their broking approach, at essentially two stages.
  1157. First, there were issues as to how Edge should have approached the risk at the start, and whether advice should have been given that the credit risk market should be approached, with the risk then being placed in that market. (As noted above, Ms. Sabben-Clare described this as: "Do not pass Go"). The experts in their joint report returned to that issue in response to the question: should Edge have known that cover in respect of credit risks and/or financial defaults would have constituted a separate line of insurance business to marine cargo? The experts' respective views were as follows:
  1158. "4.9 Mr. Russell pointed out that Edge had previously sought terms for a separate policy in February 2014, so it already knew that there was a separate market for credit risks and/or financial defaults as did Icestar. However, its instruction from Icestar was to approach the cargo risks underwriters to seek their agreement to Endorsement 5 drafted by Norton Rose. It was under no instruction to re-approach the credit risks and/or financial defaults market. If the cargo risks underwriters declined to agree the Endorsement, it would need new instructions from Icestar.
    4.10 Mr. Meldrum said that Edge had a responsibility to the client to advise that the cargo risks market would not ordinarily provide the appropriate credit risks and/or financial defaults proposed in Endorsement 5 and consequently should have known to seek the correct level of cover in the appropriate credit risks and/or financial defaults market.
  1159. Secondly, if that did not happen, so that matters moved forward with Edge approaching the cargo market rather than the credit risk market for the relevant TPC coverage, there were issues as to how the broking of the risk should have been conducted, and in particular as to whether the intended effect of the TPC should have been explained to underwriters.
  1160. Expert evidence – advice at the outset

  1161. Mr. Meldrum's view, as expressed in his first report, was that Mr. Mullen should have recognised, from the outset, the possibility that this was not something that the existing underwriters would be prepared to, or be able to, cover. He should have recognised that if they were not willing or able to write credit risk cover, it might be necessary to approach other underwriters who did provide credit risk insurance. He should have consulted with his colleagues at Edge who specialised in that type of insurance. Mr. Meldrum could not understand why Mr. Mullen did not speak to his colleagues. This kind of liaison with colleagues was something that Mr. Meldrum had routinely done. He would have referred this business to financial risks specialists.
  1162. In cross-examination, Mr. Meldrum said that the first thing that the client should be advised would be to talk to the experts, who would be the credit risk personnel. Alternatively, if the client was insistent that cargo underwriters should be approached, then you would have a conversation with your underwriters and say: "Look, there is a credit risk. What are your feelings about it". He accepted that Mr. Mullen was not breaching his duties by going to see if cargo underwriters were willing to write the risk. He said that:
  1163. "… by all means have that discussion with the cargo underwriters. Once you've given your advice to your client, this is what you should do. If you're insistent therefore I will have that discussion with underwriters but – and at that point in the knowledge – you would obviously make full disclosure of all the facts."
  1164. In his reports, Mr. Russell said that cover for credit risks and/or financial defaults is a separate line of insurance business from marine cargo, and a broker in Edge's position should know this. He did not consider that Edge ought to have approached any underwriter apart from cargo underwriters. Edge was specifically instructed by its client to seek the agreement of the cargo underwriters to include the credit and/or financial default risks incorporated in the TPC within the cargo policy. Edge's duty was to follow its client's instructions. There was therefore no obligation to advise Icestar that the credit/ financial default risks might be placed with other underwriters.
  1165. In cross-examination, Mr. Russell did not agree with the proposition that Edge should have told the Bank to talk to the specialists, particularly since the wording of the TPC was given to Edge and they were instructed to go and see cargo underwriters. He said that that scenario might have developed if the cargo underwriters "chucks the broker out". But it was reasonable to approach cargo underwriters, even if the chances of success were small.
  1166. Expert evidence – explanation to cargo risk underwriters

  1167. The question of how, if at all, the TPC should be explained to cargo underwriters cannot be completely divorced from the question of the advice which should have been given at the outset. This is illustrated by the following question and answer in the cross-examination of Mr. Russell:
  1168. "Q. Well, Mr. Russell, if you're right that that was the proper interpretation of the instruction, going to see the wrong market made it all the more important, didn't it, for the broker to explain to the underwriter what the transaction premium clause was intended to address.
    A. Well … I think that there is obviously a duty of the broker but also the underwriter to understand what is being proposed and … one would think that an underwriter if he was being presented with documents which he did not understand, he wouldn't agree to it".
  1169. Mr. Meldrum's view was that any reasonably competent broker would have explained the intended effect of the TPC to underwriters. This evidence was not directed towards the question of whether there was a duty to do so in consequence of the duty of utmost good faith including the duty to disclose material facts. Rather, it was Mr. Meldrum's view as to what a reasonably competent broker would have done in the present circumstances, in order to discharge his core obligation to take all reasonable steps to ensure that the client's requirements are met and that the cover placed is valid and effective and fulfils the client's needs. In his oral evidence, Mr. Russell agreed that this core obligation existed. The dispute was what was required in order to discharge it.
  1170. In his report, Mr. Meldrum said that it was not sufficient for Mr. Mullen simply to hand over the 2015 endorsement, containing the TPC, to Mr. Beattie. He could and should have mentioned specifically to Mr. Beattie that the purpose of the TPC was to introduce credit risk cover. There was an onus on Mr. Mullen to explain the coverage being sought by his client. This was even more so where a new clause was being added to the policy, for which there was no precedent in past dealings with the underwriters. The TPC was not a general market clause and was not a risk that marine cargo underwriters regularly insure. Similarly, in January 2016, Mr. Meldrum's view was that Edge should have drawn underwriters' attention to the fact that the TPC was credit/financial risk cover. Even if Edge considered the endorsement wording to be clear, and that the 2015/16 policy provided the cover sought, there was no good reason not to remind underwriters of the scope of cover, to ensure that they were all happy to write the risk for another year. It was routine for a broker to seek specific confirmation from an underwriter when the scope of cover is increasing. This was "to ensure that there are no nasty surprises: it avoids doubt or uncertainty".
  1171. In his oral evidence, Mr. Meldrum said (as previously described) that if the client was insistent on seeking cover from marine cargo underwriters, the broker should "have a conversation with your underwriters and say, "Look there is a credit risk. What are your feelings about it"". The rationale for Mr. Meldrum's approach was captured in the following answers:
  1172. " …yes, quite rightly he's been instructed to get cover in place and, yes, indeed, he may well in that event discuss with the cargo underwriters but in doing so you'd make – you would need to make it abundantly clear to those cargo underwriters that this is something that you don't normally do, because they don't, and "actually I want to make it very clear to you what we're covering to avoid any doubt or any potential dispute in the future"
    Q. If Mr. Mullen presented the TPC to cargo underwriters and they said yes, we're willing to consider this and we're willing to agree covering it, he would have been complying with his instruction, wouldn't he?
    A. He would
    Q. And if that happened, there would be no reason for him to go and approach a different market who would have charged a premium for this cover because they weren't already insuring the bank?
    A. I agree, but ultimately your duty to your client means you've got the to get the best cover in place for them and also to ensure that wen the claim occurs it will be dealt with properly and paid. The issue is you can't … have a situation where there's a potential doubt or it's a grey area, and … your duty is to seek to avoid those grey areas".
  1173. He made a similar point later in his evidence:
  1174. "As far as the TPC is concerned, it's a new wording, it's a new clause. Nobody in the cargo market would be familiar with it. Therefore, your duty … would be to say, "Well, this is what it means". It's not the clearest clause, it certainly isn't to me anyway, and having incorporated that, I think it's your duty to say. "Look, this is what it's covering". Do bear in mind there is a credit element to this so you do know that and we are comfortable with that, aren't we?" Because why would you take the chance? Why not do that?
  1175. He described this approach as being open and transparent. He agreed that the slip or policy document containing the terms agreed was a "good record" of what had been agreed, but said:
  1176. " … again it is about interpretation and what was discussed at the time. A lot of clauses are created by brokers, as you will see. They are variations on various different themes and sometimes they need further explanation and for the purposes of comfort and also, again, your duty to your client, is to ensure beyond all reasonable doubt that there is cover in place. So why not just do it? I don't see why you wouldn't".
  1177. Mr. Russell's central point was that it was for the underwriters to read the wording of the slip and, if appropriate, answer questions. In his first report, he said that an underwriter was expected to read and understand an endorsement or policy wording presented to him/her for consideration, in order to decide whether to underwrite the cover on the terms presented. That was the essence of the underwriter's role. If the broker was required to provide an explanation of a clause, it could place the broker in a difficult position. It would not be practical to explain every clause. A broker would be put in the position of having to select which clauses he needed to explain. If required to volunteer an explanation, the broker might be in a position of conflict with its client, if the explanation differed from the client's understanding of the clause concerned. The broker was therefore obliged to answer an underwriter's questions honestly, but absent enquiry is not obliged to volunteer an explanation of the intended meaning or effect of any clause. In his second report, he said that it did not make any difference if a clause is not customarily included in the type of policy being broked: the broker still did not have a duty to volunteer an explanation of its meaning or effect.
  1178. He said in cross-examination that the clause itself was "enough information for the underwriter to read and understand". Ms. Sabben-Clare drew attention to Mr. Mullen's evidence, to the effect that he had not understood the TPC when he first read it, and the cross-examination continued:
  1179. "Q. And in circumstances in which he was broking the clause to participants in the same market, the cargo market, he was under an obligation to tell them what it was intended to achieve because he couldn't assume that they would understand it as credit risk, could he?
    A. Well, no … I don't agree with that. I think that … he had an instruction from the client to ask the cargo underwriters to agree. They … had an opportunity to read the endorsement and decide whether they wanted to agree it or not and [if] they were unclear to ask any questions. If you are asking me whether it would be better if Mr. Mullen was feeling [un]comfortable with the endorsement and as to what the TPC meant, then quite clearly it would have been better that he had but I don't think that he had to volunteer that information.
    Q. Wouldn't merely have been better, it's what any reasonably competent broker would have done, isn't it, to protect the interests of their client? … You agree with that?
    A. Yes."
  1180. At a later stage in his evidence, he was asked about Ms. Van de Beek's question in the comment bubble in the draft sent on 28 January, and in particular the question: "is it clear that this cover is separate to the marine cargo and storage cover". Mr. Russell agreed that the broker had to answer that question with reasonable competence. He agreed that if the position was that it was open to doubt, and not clear and unambiguous, then the only way that a reasonably competent broker could answer the question was by having a conversation with the underwriter if he had not done so already. The question would be: "Do you agree that this is credit risk cover".
  1181. Discussion

  1182. I accept Mr. Meldrum's evidence that, at the very outset in his approach to the broking of the TPC, Mr. Mullen (and therefore Edge) fell below the standard to be expected of a reasonably competent broker. The admitted duties did require Mr. Mullen to tell the Bank, at the outset, that the credit risk market was the appropriate market in which to place the cover which the Bank was seeking, and (since Mr. Mullen did not have the relevant expertise in that area) specialist brokers within Edge should become involved. Such advice would have enabled the Bank to take an informed decision as to how to proceed, and I see nothing in the evidence which suggests that (having been so advised) they would have pressed forward with an attempt to obtain the cover from cargo underwriters.
  1183. It is right, as Mr. Russell said, that the Bank and NRF had drafted the TPC and asked Edge to obtain the agreement of their existing underwriters to the changes which were contained in the July 2015 endorsement. In my view, however, the Bank was simply proceeding on the assumption that this was a possible change to which cargo underwriters might be willing to agree, and were doing so in the absence of any advice from Edge as to where such cover would normally be placed. This is not, therefore, a situation where the Bank was insistent on the cargo market being approached. If cover was to be placed which clearly and indisputably met the Bank's requirements, and did not expose it to an unnecessary risk of litigation, then in my view the starting point was for Edge to identify to its client the underwriters who provided the cover that was being requested. As Mr. Russell rightly said in his supplemental report, Edge should have known that cover in respect of credit risks and/or financial defaults would have constituted a separate line of insurance business. If, therefore, Edge was to approach underwriters who did not specialise in that line of business, and did not ordinarily write it, then in my view that should only have happened (as Mr. Meldrum said in his evidence) after appropriate advice had been given to the client and the client had taken an informed decision to pursue this course. Ordinarily, one would not expect a sensible client to take this course.
  1184. If, however, cargo underwriters were nevertheless to be approached – whether or not after Edge had given the appropriate advice discussed above – then I agree with the point which Ms. Sabben-Clare put to Mr. Russell quoted above: "going to see the wrong market made it all the more important, for the broker to explain to the underwriter what the transaction premium clause was intended to address". This question encapsulated the views expressed by Mr. Meldrum. I accept his evidence that – on the facts of the present case – the fulfilment of the relevant duty owed by Edge to its client (to arrange cover which clearly and indisputably met the client's requirements, and did not expose the client to an unnecessary risk of litigation) did require Edge to discuss with the underwriters the nature of the cover which was being sought in the TPC (ie that it was credit risk cover). This was in my view necessary in order to avoid the potential for future dispute in circumstances where: the new cover sought was of considerable importance to the client; the cover had no precedent in the marine cargo market; there did exist an established and different market in which such risks would usually be placed; the underwriters were being asked to write a risk which would materially increase the potential for losses; and where the relevant clause, the TPC, was long and tightly drafted and its full import would not necessarily be grasped by an underwriter on a first reading. I consider that there was a clear rationale for doing so, as explained by Mr. Meldrum, namely to avoid the scope for dispute later, and to ensure beyond all reasonable doubt that the cover was in place. Mr. Meldrum's rhetorical questions are in my view pertinent: why take the chance, why not just do it?
  1185. I do not reach this conclusion on the basis that the clause was unclear and ambiguous, not least because that would be contrary to the conclusion which I have reached as set out in Section D above. The clause did not, in my view, lack clarity when carefully read and understood. However, the careful drafting of a clause, in circumstances where that clause is unusual and indeed unprecedented in the market in which the cover was being placed, could not reasonably be relied upon by the broker as providing protection against the unnecessary risk of litigation. This is because the door can, and does in a case such as the present, remain open for the very arguments that the underwriters have advanced in the present case. Those arguments, as Ms. Sabben-Clare pointed out, are essentially focused not upon the language of the clause, but other matters which are potentially relevant to the construction of the contract: in particular, the factual matrix relating to the nature of the market in which the cover was placed and the existence of a specialist market for risks of the present kind, and the commercial consequences of the rival constructions for which the parties contend. This, in my view, leads back to Mr. Meldrum's rhetorical questions: why take the chance, why not just do it?
  1186. Ms. Healy argued that Edge could not have been expected to identify problems in the drafting of the clause which had escaped the attentions of NRF. In her oral closing, she said that Edge would have a good answer to the claim – even if the TPC was ineffective to provide the cover which the Bank wanted – on the basis that the Bank had not asked Edge to draft the clause. The drafting was carried out by NRF, and there was no reason to hold Edge to a higher standard than NRF. She relied on the Australian decision in Messagemate Aust P/L v National Credit Insurance (Brokers) P/L [2002] SASC 327, at paragraphs [83] – [95]. At paragraph [85], the judge said:
  1187. "As relevant to this case it is difficult to distinguish between what might be expected of a solicitor and this broker. Indeed, in some circumstances and in respect of some aspects of professional advice I could envisage that a specialist broker might be expected to be better placed than a solicitor to recognise a problem. So specialised is this business that the broker might have been expected to consult a lawyer as to the significance of differences between the various policies on offer in the market. However, insofar as the present case raises a question of construction of a document I would not expect more from the specialist broker than from a lawyer".
  1188. This passage shows, however that it is wrong to assert – as a general proposition – that a broker can have no responsibility where a particular clause has been carefully drafted by a lawyer. That is why in some cases, as the judge there recognised, the broker is better placed than a solicitor to recognise a problem. As Ms. Sabben-Clare submitted, the underwriters' points on construction were all about the market and specifically about facts known to underwriters and brokers, not the Bank. That was also the basis of the "commercial sense" case advanced by underwriters. In those regards, NRF was not a mirror to Edge. Edge were the market experts. The fact that NRF thought that the clause worked was not, as she submitted, the end of it. I agree.
  1189. It would have been, as Mr. Meldrum said, a relatively simple matter for the broker to discuss with the underwriters the intended effect of the TPC in order to avoid the scope for potential dispute. It was clearly in the interests of the Bank for this to be done, in circumstances where Edge was seeking to place credit cover in a market which was not its natural home and where, as Mr. Russell accepted in his evidence, there was a risk that credit risk insurance would fall outside the authority of the cargo underwriters who were being approached. The consequence of that simple discussion would either have been that the underwriters were comfortable with what was proposed, so that everyone was (in Mr. Meldrum's phrase) singing from the same hymn-sheet; or they were not, so that the Bank could be told that, since this cover was important to them, the credit insurance market should be approached. Mr. Russell accepted that, in the context of Ms. Van de Beek's question on 28 January, a reasonably competent broker should have a discussion with the underwriter if there was doubt as to the effect of the clause. I consider that such a conversation was appropriate even if Mr. Mullen were of the view that the clause was sufficiently clear. The clarity of the position was a matter of concern to the client: hence Ms. Van de Beek's question. Mr. Meldrum's questions – why take the chance, why not just do it – are again apposite.
  1190. I do not consider, as Ms. Healy submitted, that this is to apply the wisdom of hindsight. Here, the contemporaneous documents show that Mr. Mullen had not initially understood the clause. His question about the TPC to Mr. Lockyer on 8 July 2015, after Mr. Mullen had carried out a review of the endorsement, was: "do not understand this – have you seen this before Lee"? He then asked the Bank, on 9 July 2015, to explain the "relevance of the "transaction premium" contained under 1-2-3". Given those questions, there was clearly a risk that the underwriters and the Bank would not have the same understanding of the clause, with the client thereby being exposed (unless the position was clarified with underwriters) to an unnecessary risk of litigation. The lengthy clause was clearly of considerable importance to the Bank, as Mr. Mullen did or at least should have appreciated at the time. The need to explain the position to underwriters is, therefore, not a matter borne of hindsight. Indeed, as discussed below, Mr. Lockyer asked a pertinent question in January 2016 (" … is this the understanding of underwriters"), and so the issue was identified at the time.
  1191. Mr. Russell said, in the extract from his cross-examination quoted in paragraph 917 above, that any reasonably competent broker, who did not feel comfortable with the endorsement and what the TPC meant, would have volunteered information about it to underwriters in order to protect the interests of their client. That evidence acknowledged that it would indeed be appropriate, in certain circumstances, to discuss the intended effect of a clause with underwriters. It seemed to me that this was very close to what Mr. Meldrum was saying. Mr. Mullen may ultimately have felt comfortable with the wording of the TPC. But, as indicated above, the contemporaneous documents show that he had not initially understood it, and it had been necessary for him to ask the Bank what it was seeking to achieve. Mr. Russell said, sensibly, that quite a few people might have had to read the clause a few times before understanding it. All of this points, in my view, to the good sense of Mr. Meldrum's approach.
  1192. In so far as there was a disagreement between the two experts as to whether Edge's duty to its client required them to give an explanation to underwriters of this unusual clause, I prefer the evidence of Mr. Meldrum. His evidence was (as the Bank submitted) direct, clear, pragmatic and sensible. I thought that Mr. Russell gave his evidence fairly and conscientiously, and sought to assist the court. Nevertheless, I thought that Mr. Meldrum's approach, rather than the narrower approach which Mr. Russell put forward, better reflected what was required in order to protect the Bank from an unnecessary risk of litigation in the circumstances which I have described. As discussed in the previous paragraph, Mr. Russell's evidence acknowledged that, in certain circumstances, the underwriters should be given an explanation. In addition, Mr. Russell also accepted, when asked about Ms. Van de Beek's 28 January e-mail, that a reasonably competent broker would have had a discussion with the underwriter if he considered that the TPC was not sufficiently clear. There is, therefore, no reason in principle why a broker's obligation to its client should not require a discussion as to the effect of a clause to take place. The only question is whether it was required in the present circumstances. I consider that it was.
  1193. This conclusion is reinforced by other aspects of the evidence in this case. I consider that Mr. Lockyer's e-mail to Mr. Mullen on 28 January 2016 is indeed significant, as Ms. Sabben-Clare argued. Mr. Lockyer, having seen Ms. Van de Beek's comment, appreciated the significance of the point that she was raising, and asked about the understanding of underwriters. As his evidence (set out above) shows, if he had not received Mr. Mullen's assurance, he would have considered it appropriate to go back to the underwriters and made sure that they were happy and that they clearly understood what they were covering. This is consistent with Mr. Meldrum's view as to the approach that should have been taken.
  1194. Mr. Mullen's response to Mr. Lockyer was unsatisfactory, because there was no basis for his assurance that underwriters shared the understanding referred to. His evidence (which I have accepted – see Section E above) is that there had been no specific discussion with Mr. Beattie about the effect of the TPC. There was also no evidence of any discussion with any other underwriter. Since the issue of the underwriters' understanding had been directly raised by Mr. Lockyer as a result of reading Ms. Van de Beek's question, I consider that fulfilment of the relevant duty did require steps to be taken at that stage to check that the underwriters did indeed have the same understanding as the Bank, particularly in circumstances where Edge had not yet had a positive discussion with the underwriters on that topic. Otherwise, the Bank would be exposed to the risk of unnecessary litigation, which it was the broker's duty to take reasonable care to avoid.
  1195. Mr. Meldrum's evidence in this regard was that the email from Ms. Van de Beek on 28 January 2016 should have prompted Mr. Mullen to double check with Mr. Beattie. This was because she had raised a specific query regarding the scope of the TPC and asked for confirmation of the position. A doubt had arisen, and it was the broker's job to resolve the uncertainty and ensure there was no doubt both from the client and market perspective. It should, he said, have raised a red flag to Edge. It was therefore incumbent on Edge to take this up with underwriters: to have a specific discussion, and to make sure that the discussion was documented. He said that there is nothing unusual about that kind of situation, where a client asks for clarity and the broker is tasked to find a way to provide it. He would have expected a competent broker to specifically point out the clause to underwriters and talk through the wording and consequences. That was just part and parcel of a broker's daily work, nothing special or out of the ordinary.
  1196. I considered that evidence to be convincing, and indeed it was supported by Mr. Lockyer's evidence as to what should or would have happened if he had not received reassurance from Mr. Mullen in response to his question. I do not accept that it is an answer to the point for Edge to contend that Mr. Mullen felt that he could give a confident answer to Ms. Van de Beek's question as to whether it was clear that the cover was separate to the marine cargo and storage cover. He could not, on any view, give a confident answer to the second of Mr. Lockyer's pertinent questions; because the point had not been specifically discussed with underwriters.
  1197. I also think that some earlier correspondence provides some further support for Mr. Meldrum's evidence as to the need to discuss the TPC with underwriters, although this correspondence is not so sharply focused on the present issue as Mr. Lockyer's 28 January 2016 e-mail to Mr. Mullen and his evidence in that regard. On 25 June 2015, after the Amsterdam meeting when the proposed endorsement was handed over, Mr. Mullen told the Bank that: "the language for amendments needs to be discussed with the lead Underwriter point by point". On 20 July, Mr. Mullen reported that lead underwriters had only recently returned, and that he was "in the process of discussing your suggestions." The latter statement was incorrect, because Mr. Mullen had not started discussions. But the message continued the theme of the need for discussion with underwriters. On 28 July, he said to Ms. Barnes (when chased) that he was "nearly finished". Again, he had not in fact started to discuss matters with Mr. Beattie. Following the meeting, Mr. Mullen apologised for the time that it had taken to review, but reported (again incorrectly) that there were: "Bits that needed no discussion and others that needed some explanation". Whilst these answers were not specific to the TPC, and whilst there was an element of Mr. Mullen seeking to explain away his delay in dealing with things, they are in my view also consistent with Mr. Meldrum's evidence: the TPC was something that did need to be discussed.
  1198. It is also important to note that, as Mr. Russell accepted in his evidence, there was a risk that credit risk insurance would fall outside the authority of the cargo underwriters who were being approached. In the event, it has not been argued by underwriters that lack of authority provides a defence to the claim, although the limits of authority were referred to in the context of arguments as to the factual matrix and related matters. The significant point, however, is that if there was a risk that the writing of credit risk insurance lay outside the authority of the subscribing underwriters, there was potential scope for a future dispute on the cover if the nature of the TPC was not discussed with subscribing underwriters. The placement of cover, without any discussion with subscribing underwriters, therefore exposed the Bank to the risk of unnecessary litigation.
  1199. For these reasons, I decide that Edge failed in its duty to the Bank even on the premise that the TPC did provide the cover that was sought.
  1200. I do not accept Ms. Healy's argument that there was no relevant breach of duty because the underwriters' construction arguments can be regarded as spurious. I agree that those arguments pay little or no regard to the actual wording of the TPC. However, to the extent that they were based upon the factual matrix and context and the commercial consequences of the Bank's construction, they did in my view have sufficient strength as not to warrant being described as spurious.
  1201. Since I have decided that Edge failed in its duty even on the premise that the TPC provided the relevant coverage, I consider that the same reasoning leads inevitably to the conclusion that Edge would have failed in its duty in the event that I had concluded that the TPC did not provide such cover. Indeed, there might in that situation have been further powerful arguments as to why Edge was in breach of duty: for example, if the reason why the TPC did not provide cover was that it was insufficiently clear. However, it is not necessary to lengthen this judgment by exploring those potential arguments.
  1202. I do not consider that these conclusions result from the imposition of an unprincipled "duty to nanny". There is nothing in my reasoning or conclusions which is intended to suggest that brokers generally owe duties to their clients to explain particular clauses, including unusual clauses, to underwriters. Ultimately, the question is what was required on the facts of the present case in order to fulfil the duties which Edge rightly admitted: ie obtaining the cover that was sought, and procuring cover that clearly and indisputably met the Bank's requirements, and so did not expose it to an unnecessary risk of litigation. That question may, and does on the facts of the present case, require brokers – in order to protect the position of their clients – to give information to underwriters, or to discuss the implications of that information, even though the underwriters could not succeed on an avoidance case. That is illustrated by Ms. Van de Beek's comment, and Mr. Lockyer's response which identified the need, in the interests of the client, to check whether the underwriters had the same understanding as the Bank in relation to the effect of the TPC. The reason was to avoid problems which would potentially arise in the future, including the possibility of litigation, if the underwriters did not share the Bank's understanding of the unusual clause with which I am concerned. That is in my view a different question to that which arises in the context of the non-disclosure arguments. The latter question is significantly impacted by the Marine Insurance Act 1906, section 18, which I have considered above. The former question is not. Thus, Ms. Sabben-Clare's response to the "nanny" point emphasised the importance of identifying the right protιgι. As she said: "it's not about the duty to protect underwriters, it's about the duty to protect your own client".
  1203. In the end Ms. Healy accepted that there could be circumstances where the underwriters could not avoid the policy, but where the brokers would have been in breach of duty for failing to raise it with underwriters. But she submitted that it would require an express instruction from the client to raise the particular matter with underwriters. There was, she submitted in her oral closing, no duty to explain anything in the absence of an express instruction or an enquiry from underwriters. I disagree. I see no reason why a broker should not be obliged to raise a matter with underwriters, in order to fulfil the duties owed to the client, in circumstances where there has not been an express instruction. There are in my view no fixed rules which delineate what the broker should or should not do in order to procure cover that clearly and indisputably meet its client's requirements, and so does not expose it to an unnecessary risk of litigation. This must in my view depend upon the circumstances and the particular facts of the case.
  1204. I therefore conclude, without needing to address all of the different ways in which Ms. Sabben-Clare put the case, that Edge was in breach of duty to the Bank, and that this was the case whether or not the TPC in fact provided the coverage which the Bank was seeking.
  1205. J3: Causation and quantum issues
     

  1206. The principal issues of causation and quantum arose in connection with the Bank's claim against Edge that would have arisen if (contrary to my decision) the claim against the underwriters had completely failed. The argument largely focused on causation and quantum in the context of the situation where the TPC did not provide the coverage that the Bank was seeking, and Edge's breach of duties in that respect as discussed in Section J2 above. It was not suggested, however, that the position on causation and quantum would be any different if the cover was completely ineffective in the event that all underwriters had succeeded in their defences based on rectification/estoppel/ collateral contract or avoidance. I shall address the issues that would have arisen in the event that the claim against underwriters had failed on any of those grounds. I will then consider the issues of causation and quantum which arise in the context of my actual conclusion that the Bank's claim against the majority of the underwriters succeeds, where there is nevertheless (leaving aside the consequences of the failure of the claim against Ark and Advent) a small potential claim (Ms. Sabben-Clare described it as a 'sliver' of a claim) against Edge for irrecoverable costs. Finally, I will briefly address the possible claim, which I have not hitherto addressed, as to what the position would have been if the claim had failed, or been substantially reduced, because the Bank was in breach of Clause 3 of the policy.
  1207. The Bank's case

  1208. The Bank contended that it would have suffered practically no loss but for Edge's negligence. The Bank wanted insurance for these risks. If Edge had acted competently, the Bank would have been advised that such cover was only available from the credit risk market, under "singleton" credit policies: ie policies which provided credit risk cover in relation to individual counterparties of the Bank. In those circumstances the Bank would either: (i) have insured the transactions under such alternative credit risk cover; or (ii) if or to the extent that cover was unavailable, unwound existing transactions and not proceeded with any new ones.
  1209. There were therefore two ways in which the Bank would have avoided the losses that would be suffered in the event that the policy failed to respond. Course (i) would have avoided 90% of the loss, with the remainder self-insured, and involved paying premium. Course (ii) would have resulted in full repayment, since Transmar and Euromar were solvent at the time when, on this hypothesis, the transactions would have been unwound. Course (ii) would therefore have avoided 100% of the loss. The Bank was willing on both bases to give credit for the retention and premium under alternative credit risk cover, which meant that the court did not need to assess the relative chances of courses (i) and (ii) occurring.
  1210. In relation to course (i), the Bank provided particulars of the alternative cover that would have been obtained, in a letter from Reed Smith dated 1 May 2020. The case was:
  1211. "1. Icestar would have put in place for the year 1 February 2016 to 31 January 2017 a policy of credit risk insurance covering default by its counterparties Transmar and Euromar (alone) under their Type 2 and Type 3 financing transactions with Icestar.
    2. The policy would have been placed with underwriters operating in the London market. It would have indemnified Icestar as to 90% of the losses that it suffered by reason of the defaults committed by Transmar and Euromar during the period of the policy.
    3. The premium payable under this policy would have been calculated as a percentage of the actual financial exposure incurred to Transmar and Euromar and by reference to its actual duration. The percentage that would have been applicable to Icestar's cover would have been in the range of 1.35% to 1.65% per annum".
  1212. The Bank also contended that if (as Edge submitted) credit risk underwriters had imposed particular terms which needed to be fulfilled in order to make the cover effective, or which would have excluded particular transactions from coverage, then the Bank would have shaped its actions accordingly. It would have made sure that the underwriters' requirements were met, so that any obligations (for example to carry out quality checks) were fulfilled. If the terms of the insurance had the effect of excluding transactions with Transmar and Euromar from cover, then the Bank would have unwound those transactions so as to ensure that all transactions with those companies were indeed covered.
  1213. Edge's case

  1214. Edge denied that the Bank could or would have obtained standalone credit risk cover. The claim against Edge therefore failed as a matter of factual causation, since, but for any alleged breach by Edge, the Bank would not have been able to obtain the cover it had under the TPC. Edge contended that there was a difference between (i) credit risk cover under the TPC, which was provided as a free add-on by underwriters in the context of a large, valuable and popular risk, and (ii) standalone credit risk cover provided by specialist underwriters, without any of the attractions of the Bank's marine cargo policy.
  1215. If, however, the Bank had required cover against the risk of default by Transmar and Euromar alone, it would not have been able to obtain it; because credit risk underwriters would not have been able to provide it, and certainly not at a price which the Bank would have paid. Edge therefore submitted that if the Bank had not obtained cover under the TPC in July 2015 or January 2016, it would not have taken out credit risk insurance covering default by Transmar and Euromar alone; because it would have considered that the large cost of such insurance far outweighed the benefit, given the apparent reliability of Euromar and Transmar as counterparties.
  1216. If, however, the Bank had really been interested in obtaining this cover, credit risk underwriters would not have agreed to underwrite a policy covering default by Transmar and Euromar alone (the pleaded case), since in all the circumstances it would have appeared that the Bank was engaging in adverse selection: ie picking the least attractive and riskiest part of its portfolio to insure.
  1217. Even if credit risk underwriters had been willing to underwrite a policy covering the default of Transmar and Euromar alone, it would only have been on terms that limited their liability for the defaults which the Bank rely on in these proceedings. In particular, the cover would have been on a "risks attaching" basis; so that it did not cover transactions between the Bank and its clients which were concluded prior to the inception of the policy. There would also have been a maximum risk tenor for a given transaction of 360 days, or the useful shelf life of the goods. The policy would also have required independent quality checks of the cargo. There would also have been a retention of at least 50% rather than the 90% for which the Bank contended. A credit risk policy would also have been more expensive than the Bank contended, the rating being 2% of the value of the insured transactions or at the very least 1.85 – 1.90%.
  1218. This aspect of the dispute gave rise to issues as to what the Bank would have done if differently advised: in particular whether it would have sought and been willing to pay for credit risk insurance from specialist underwriters. It also gave rise to issues as to what cover would have been available in the credit risk market, including whether particular terms would have been imposed which would in practice have meant that the Bank was uncovered. Expert evidence on these issues was given by two underwriters experienced in that market, Mr. Simon Hayter for the Bank and Mr. Nick Hedley for Edge.
  1219. Legal framework

  1220. There was no dispute that the applicable legal framework, for the resolution of these arguments, is established by the decision in Allied Maples Group Ltd. v Simmons & Simmons [1995] 1 WLR 1602. A party's own conduct falls to be assessed on the balance of probabilities. Stuart-Smith LJ said at 1610:
  1221. "If the defendant's negligence consists of an omission, for example to provide proper equipment, given proper instructions or advice, causation depends, not upon a question of historical fact, but on the answer to the hypothetical question, what would the plaintiff have done if the equipment had been provided or the instruction or advice given.
    …
    Although the question is a hypothetical one, it is well established that the plaintiff must prove on balance of probability that he would have taken action to obtain the benefit or avoid the risk. But again, if he does establish that, there is no discount because the balance is only just tipped in his favour."
  1222. By contrast, the hypothetical acts of third parties fall to be assessed on a loss of a chance basis. The claimant in those circumstances does not have to prove, on the balance of probabilities, that the third party would have acted so as to confer the benefit or avoid the risk. The claimant can succeed provided he shows that he had a substantial chance rather than a speculative one, the evaluation of the substantial chance being a question of quantification of damages: see Allied Maples at 1611 A-C. Stuart Smith LJ concluded (at 1614 C-D):
  1223. "… the plaintiff must prove as a matter of causation that he has a real or substantial chance as opposed to a speculative one. If he succeeds in doing so, the evaluation of the chance is part of the assessment of the quantum of damage, the range lying somewhere between something that just qualifies as real or substantial on the one hand and near certainty on the other. I do not think that it is helpful to seek to lay down in percentage terms what the lower and upper ends of the bracket should be".
  1224. When it comes to the evaluation of the chance, as part of the quantification of damages, all significant factors should be taken into account. The evaluation of the substantial chance (if it exists) of obtaining the benefit in question thus forms a part of the assessment of damages: see eg Wellesley Partners v Withers LLP [2015] EWCA Civ 1146, paras [96] – [110].
  1225. How would the Bank have acted, if Edge had not been in breach of duty?

  1226. In the context of this legal framework, the Bank submitted that the starting point was to consider, on the balance of probabilities, what it would have done if Edge had acted competently. They submitted that had Edge acted competently, it would have advised the Bank to involve a credit risk specialist from the start. Alternatively, Edge would have told cargo underwriters that the Bank wanted credit risk cover, and this would (except possibly for Mr. Beattie) have led to the cargo underwriters declining to write the cover. Any competent broker would have reported the cargo underwriters' position to the Bank.
  1227. I agree that this is the correct starting point, given my conclusions as to Edge's breach of duty discussed in Section J2 above. This starting point would be equally applicable in the event that the claim had failed against all underwriters because of rectification/estoppel/collateral contract or avoidance.
  1228. The factual question which arose was, therefore, whether the Bank would in these circumstances have looked to protect itself by taking out insurance in the credit insurance market; or would not have taken on any new transactions or rolled over existing transactions if they fell outside the scope of the insurance. There was no dispute that, in the context of the legal framework set out above, it was appropriate to approach the Bank's conduct by considering, on the balance of probabilities, the steps that the Bank would have taken in order to protect itself. Ms. Healy emphasised, however, that this question had to be addressed with careful regard to the Bank's pleaded case: that it would have sought insurance cover for Euromar and Transmar alone. She submitted that the Bank's evidence was not sufficient to prove that case.
  1229. The Bank's principal witness on the question of what the Bank would have done, in different circumstances, was Mr. Stroink. He was an impressive witness, who gave his evidence carefully and honestly. In her closing submission, Ms. Healy rightly accepted that he was an honest witness.
  1230. Mr. Stroink was sometimes a little puzzled by some of the hypotheses or issues as to what would have happened in the counterfactual world that he had not had to inhabit; because he throughout believed that the Bank did have the credit risk cover for all of their counterparties. It is unsurprising and far from unusual for a witness to find that the task of answering such hypothetical questions is not straightforward, particularly when it involves hypotheses as to what third parties would have done. For example, Mr. Stroink was asked to consider how clients might have reacted to requests for them to pay for quality certificates, and how the Bank would then have reacted if they declined to do so. Nevertheless, Mr. Stroink did seek to explain what he considered would have happened in different circumstances. His evidence, as to what the Bank would have done in the counterfactual world, was in my view sensible and commercially rational. I see no reason to doubt his evidence, as summarised below.
  1231. In his witness statements, Mr. Stroink said that if he had been told in January 2016 that it was necessary for the default risk to be covered by a separate policy of credit insurance, he would have instructed Edge to obtain quotes for such a policy. He would have wanted to cover transactions where there was a client repayment risk. He would not, however, have been interested in insuring transactions where Icestar only had an exchange risk: ie the Icestar 1 transactions. He would have considered which particular clients to cover. He would not have considered it necessary to have credit risk cover in respect of clients with significant size and access to liquidity, such as Trafigura or Louis Dreyfus. However, he would have "considered" credit risk cover for clients with a smaller balance sheet. He identified 6 such clients, including Transmar and Euromar.
  1232. In relation to the cost of cover, if available in the range of 1.35% - 1.85%, or even 2%, Mr. Stroink said that he would have taken the view that the Bank should buy this additional cover. Icestar was making a margin of 3.25% on Transmar and Euromar transactions, and the cost of credit risk insurance would still leave a decent margin. The return after buying the insurance would be a good return on what would have been an investment grade risk. Whilst the Bank did not take out insurance in 2017/18, after being told at short notice that underwriters were no longer willing to include the TPC, that was because there was then no longer any need to do so. That was a consequence of all the problems that the Euromar and Transmar defaults had created. Icestar therefore changed the profile of its portfolio substantially, so that its exposure to clients with smaller balance sheets was reduced and in some cases terminated.
  1233. Mr. Stroink said that the additional cover was a very important protection. He emphasised that within the context of the Bank overall, the Icestar product was generally only an ancillary supplement. It was a product offered in addition to the Bank's wider and more general relationship with its clients.
  1234. These points were echoed by Ms. Franssen in her evidence. She referred to the healthy margin that would have been achieved even taking into account the cost of insurance, and her belief that the Bank was already exceeding its internal hurdles in terms of the return being made from Transmar across the Bank's business as a whole, regardless of any supplementary contribution that might be made through the repo transactions. The Bank's income from Transmar was going to rise from February 2016 anyway, since ABN was replacing another bank under Transmar's revolving credit facility. Any additional income from Icestar was a positive cross-selling opportunity for the Bank, even if the margin was reduced from 3.25%, which was one of the highest margins for any client.
  1235. In his third statement, Mr. Stroink said that he would have sought cover on the widest possible terms to mirror the cover that was available under the TPC. He addressed the question, following exchange of the credit risk experts' reports, as to what would have happened if the policy had contained terms which restricted cover, such as those suggested by Edge's expert: for example, terms concerning the maximum period of financing or maximum risk duration, the maximum number of rollovers, a limit on the age of the cargo financed, the time when the financing had to be advanced, or a requirement to obtain independent quality certificates. He said that he could say with certainty that Icestar would have made sure that it complied with any terms that the credit risk underwriters would have imposed. The cover was important to Icestar, and it was a part of its approach to risk and portfolio management that Icestar had already decided to pursue. Having obtained such cover, they would not have done anything to endanger the insurance by not complying with the terms that credit risk underwriters required. This meant that, from the beginning of August 2015 (when the TPC was first agreed), the Bank would have ensured that all future transactions with Transmar and Euromar complied with any relevant limitations or requirements that were imposed. Accordingly, since transactions had a maximum duration of 3 months, the Bank would only have rolled over a deal upon its expiry if doing so would still have meant that the coverage was still in place. If quality checks were required, they would have been carried out and the cost passed on to Euromar and Transmar as part of the cost of financing.
  1236. Mr. Stroink also addressed the question of what would have happened if there had been no rollover, because of the lack of cover. This would have meant that Euromar and Transmar were obliged to make repayment. He expressed the view that if that had happened at any time in the period beginning from August 2015, the Bank would have been repaid by Transmar and Euromar. He exhibited a table showing that significant financing repayments were made from August 2015 onwards. For example, repayments by Transmar and Euromar in the last months of 2015 were: £ 2.9 million (August), £ 6.3 million (September), £ 5.1 million (October), £ 458,000 (November) and £ 641,000 (December).
  1237. I did not consider that any of this evidence was materially shaken in cross-examination. Both Mr. Stroink and Ms. Franssen, in response to questions asked, expanded upon the commercial logic which lay behind the evidence given in their witness statements. Thus, Mr. Stroink explained that the review by NRF of the insurance cover in 2015 had its origins in a "deep dive" internal review which had been carried out by the Bank. This deep-dive considered Icestar's set-up, products and procedures in February 2015.
  1238. In risk management terms, Mr. Stroink explained that there were two advantages to the insurance. First, it provided a second line of defence or an additional layer of protection. Secondly, it lowered the Bank's assessment of the probability of default. The Bank could therefore, for capital allocation purposes, allocate the exposure to the insurance companies rather than the client. The insurance companies had a better credit rating than most of the Bank's clients. This would have the effect that the Bank's capital requirements for that part of the portfolio would reduce, and therefore the return on capital would be better.
  1239. Mr. Stroink described the insurance (which he thought that he had obtained) as "necessary" to risk-manage the Icestar 2 portfolio:
  1240. "As of the business review, we said: how can we decrease the risk in general within Icestar? Well we can – on the Icestar 2 portfolio, we can seek credit risk cover there. And then we decreased the risk and we increased the return".
  1241. It was clear from his evidence, as well as the contemporaneous documents such as the correspondence with Norton Rose, that the Bank was thinking about its risk management approach, and its insurance requirements, very carefully. The Bank knew what it wanted to achieve with its insurance programme, and asked NRF to look at it not only in 2015, but also at renewal in January 2016, in order to ensure that appropriate cover was in place. The Bank was therefore taking a thoughtful and thorough approach to risk management and the role of insurance in that context.
  1242. I consider that the same thoughtful and careful approach would have been taken if the Bank had received advice to the effect that the desired cover could or should not be obtained by way of an add-on to the cargo policy, but that it was advisable or necessary to seek it in the specialist credit market albeit at extra cost. There is nothing in the documents in 2015 and 2016 which suggests that the Bank was only interested in this cover if it were provided as a free add-on, or that the requirement to pay for the cover would have deterred the Bank from seeking it. The fact is that the Bank was not advised by Edge that there was any additional cost involved, and therefore the Bank had no reason to give consideration to the question of whether or not they should pay for it. Had that issue arisen, I have no doubt that the Bank would have decided to buy the cover. It made economic and commercial sense to do so, even at levels of 2%, for the reasons given by Mr. Stroink. It still allowed Icestar a healthy margin on Transmar and Euromar transactions, and would then provide an additional layer of protection as well as increasing the return on capital. Given that the Icestar business was an ancillary product to the Bank's main business, or a cross-selling opportunity, there was no reason for the Bank to carry a risk of default which could be covered by insurance at a reasonable cost which would still leave a reasonable profit. Mr. Stroink's evidence (which I accept) was that insurance was a "necessary", not simply a "nice to have" (as was put to him in cross-examination). Having reviewed his written and oral evidence, I do not accept Edge's submission that his evidence stopped short of saying that Icestar would have gone no further than merely "considering" cover for Transmar and Euromar. Nor do I accept that his evidence was vague, non-committal or fudged.
  1243. I also accept Mr. Stroink's evidence that, if advised that the insurance was not available, or that it would exclude certain transactions, the Bank would have unwound the repo transactions with Euromar and Transmar in full, or at least to the extent that they would not be protected by the insurance. Ms. Sabben-Clare made the fair point in that regard that Mr. Stroink had not really been challenged on that aspect of his evidence: cross-examination was essentially directed towards the question of whether alternative cover would have been taken out. The point is important because the potential need to obtain cover in the credit risk market would have arisen in June/ July 2015, which was a full year before the first default of Euromar. There was, therefore, a considerable amount of time to unwind transactions which only had a 90-day tenor.
  1244. I also think that Mr. Stroink's evidence as to what would have happened, absent available insurance cover, is consistent with the careful approach to risk management that the Bank took when carrying out its deep dive and then making decisions thereafter. It is also consistent with his evidence that the Bank changed the Icestar portfolio after the problems arose with Euromar and Transmar. That evidence also explains why no significance is to be attached to the fact that, when underwriters declined to include the TPC in 2017, the Bank did not seek alternative credit insurance at that stage.
  1245. I accept Mr. Stroink's evidence that, if the credit insurance had imposed particular contractual requirements (for example, if insurers required quality certificates to be obtained), the Bank would have complied. It was no part of Edge's case that the Bank was operating in an unprofessional manner. (The only suggestion to that effect was from underwriters, in the context of the Clause 3 argument which I have rejected – and where there was no evidence that the Bank had operated in a way that was different to other banks financing repo transactions). It would have made no sense for the Bank to incur the cost of buying insurance, and then fail to comply with contractual obligations.
  1246. One issue which came into particular focus in the cross-examination of Mr. Hayter, and in Ms. Healy's submissions, was Mr. Stroink's statement that he "would have considered credit risk cover for clients with smaller balance sheets. These clients would have included Transmar, Euromar, Louisiana Rice Mill, Riverland, Bluequest and Sucafina". It was clear from Mr. Stroink's evidence as a whole that he was saying that he would have taken out cover for Transmar and Euromar. For example, his evidence specifically compared the Bank's margin on those clients with the cost of insurance. The question arose, however, as to whether Mr. Stroink was saying that he would have taken out cover for all of these smaller clients, or only that he would have thought about it. In so far as he was saying that the cover would have been taken out, the question arose as to whether that was consistent with the Bank's pleaded case that it would have sought cover for Euromar and Transmar alone.
  1247. This issue was touched upon, although not explored in great detail, in the course of Mr. Stroink's cross-examination. As I have said, Mr. Stroink referred to the insurance as being a "necessary" in order to risk manage the Icestar 2 portfolio. It was suggested to him that if the cargo insurers had refused to provide the cover in July 2015 or January 2016, the Bank would not have taken out alternative credit risk insurance to cover the "Icestar 2 and 3 risks, or specifically to cover Transmar and Euromar if you'd had to pay a substantial premium for it". Mr. Stroink said that this was obviously a commercial decision: he would look at it on a case by case basis, per client. There was then the following exchange:
  1248. "Q. But you've said in your witness statement that you would look to take out insurance for all of Transmar, Euromar, Louisiana – "
    A. Yes
    Q. -- Rice Mill, Riverland, Bluequest and Sucafina: all the Icestar 2 clients.
    A. But they're not only Icestar 2 clients, we also do Icestar 1 with these clients. But the Icestar 2 exposure, I would like to – I would prefer to insure for this – for this part. And that's a hypothetical question: okay what would you have done if this would have been the case? And if I had to make a decision, I would make a decision to insure this part of the portfolio, of the Icestar 2 portfolio. (Emphasis added)
  1249. Thus, Mr. Stroink was cross-examined on the basis that he would have "looked" to take out insurance for all of the smaller Icestar 2 clients, and Mr. Stroink's answer was that he would have made a decision to insure that part of the portfolio; ie all the smaller clients. In my view, given that the Bank was concerned to manage the risk on this part of the portfolio, it is more likely than not that the Bank would indeed have sought coverage across the relevant portfolio, namely for the smaller clients including, but not confined to, Transmar and Euromar.
  1250. In her closing submissions, Ms. Healy argued that the Bank's claim on causation failed in the light of this evidence. She submitted that the Bank had failed to show that it would have sought insurance for Transmar and Euromar alone, which was the Bank's pleaded case. I reject this point. The question which I am addressing is whether the Bank would, on the balance of probabilities, have acted so as to obtain the relevant benefit or avoid the risk. The relevant benefit or risk in that context related to the taking out of insurance which would have covered the risk of losses in relation to Transmar and Euromar. In terms of causation, I consider that the Bank's case on causation succeeds whether or not the insurance sought would have covered Euromar and Transmar alone, or insurance for those companies as part of a package with some other companies. Either basis is sufficient to establish causation. I do not see why the Bank's case on causation should fail because the evidence indicates that the Bank would have sought insurance for a wider group of companies than Transmar and Euromar alone, provided that (as the evidence clearly shows) the insurance sought would have included Transmar and Euromar.
  1251. I accept that the evidence in this respect has a potential impact upon issues concerning loss of a chance, and quantification of loss. However, the evidence in the case (which I discuss in more detail below) showed that there was a real or substantial chance of obtaining insurance in relation simply to the risk of Transmar and Euromar. There was nothing in that evidence which suggested that the chance of obtaining that insurance would diminish in the event that cover was sought for the wider category of companies referred to in Mr. Stroink's first statement. Indeed, the evidence of Mr. Hedley was that this would, in principle, have been favourably received by the market. As he said in paragraph 3.8 of the joint report:
  1252. "Mr. Hedley is of the view that seeking cover for Euromar and Transmar alone would have implied adverse selection by Icestar. On the assumption that Icestar had a large client portfolio, seeking cover for two clients only would suggest that Icestar knew those clients represented an elevated credit risk and wished to lay that risk off. However, Mr. Hedley accepts that there could be legitimate grounds for selecting Transmar and Euromar only (e.g. Transmar being Icestar's largest single exposure with volumes increasing and limit relief needed) which would overturn the presumption of adverse selection. Absent such an explanation, adverse selection would be suspected".
  1253. In support of her argument that the Bank's case on causation failed (because the case advanced in the pleadings was on the basis that cover would have been sought for Transmar and Euromar alone), Ms. Healy relied upon the fact that the expert witnesses had not specifically considered what the cost of cover would have been for insurance of a portfolio of companies which included companies other than Transmar and Euromar, or indeed whether such cover would have been available.
  1254. I note, however, that there was some evidence from both experts that, unsurprisingly, the cost of cover would have been higher for a larger portfolio. This indeed is obvious: since the premium payable depended on the value of the insured transactions, as the calculations in Section K of Mr. MacLeod's first report showed. The overall premium therefore depended upon the volume of business being transacted. Mr. Hedley said that there would in effect be a discount for volume. He said that if you treated Euromar and Transmar as one customer, and added another five, the premium would not go up proportionally; but it may be three times as much. I take this evidence into account when assessing the quantification of loss.
  1255. Whilst it is true that neither expert specifically addressed the question of whether the portfolio as a whole would have been insurable, it is (as I have said) implicit in Mr. Hedley's evidence on "adverse selection" that credit underwriters would have looked favourably on insurance for the whole portfolio. The expert evidence also indicated that there were no relevant capacity constraints at the time: there was a soft market with underwriters looking for good business to write. There is nothing in Mr. Hedley's report, or in any of the expert evidence, which suggests that there would have been any difficulty in obtaining cover for the wider portfolio, and in his oral evidence Mr. Hedley gave an estimate of what the additional premium would have been. Mr. Stroink's first statement, in which he refers to the 6 companies where cover would have been "considered", was served in February 2020, some time before the expert evidence was served. If a point were to be raised, along the lines that cover would only have been potentially available for the whole portfolio of smaller companies but that there was no real or substantial chance of such cover actually being available, I would have expected such a case to have been articulated in Mr. Hedley's evidence.
  1256. In any event, I did not think that any possible doubt, as to whether all the companies in the portfolio could have been insured, was of any assistance to Edge's argument either on causation or quantification. If one or more of the other companies in the portfolio had been regarded as an unacceptable risk, then the consequence would be that the Bank's exposure to those companies would not be insured. The cover for Transmar and Euromar would, however, remain. It would also have been apparent, from the Bank's attempt to obtain coverage for the portfolio, that there was no question of adverse selection. The exclusion of cover for one or more of the other companies would, on this hypothesis, have been the consequence of an underwriting decision not to insure those companies.
  1257. I therefore do not consider that the Bank's case on causation fails by reason of any of these points, although I will take into account the additional premium cost in the context of quantum.
  1258. Edge also submitted that if the cover had required independent quality checks, the Bank would have sought to pass these costs onto its customers, and that this could or would have caused Transmar and Euromar to walk away. Icestar would not have taken out insurance for this business if that meant losing the business entirely. Mr. Stroink's evidence on this topic was that this was all very hypothetical. He did not know how the client might have reacted if asked to pay for quality certificates. But if the effect of this requirement was that the Bank needed to make a certain percentage, and the client thought that the cost of the facility was too expensive, then "we don't do a deal because then it doesn't make sense".
  1259. There was in my view nothing in this evidence which suggested that the Bank would have turned its back on the insurance which it considered necessary, in the hypothetical event that a client had decided that it was unwilling to bear the cost of providing quality certificates. In the events that (i) the cost of doing so was objected to by the client, and (ii) this meant that the repo transaction became uneconomic from the Bank's perspective, then I accept Mr. Stroink's evidence that the Bank would have declined to enter into the transaction. Ultimately, the existence of the second line of defence (the insurance) was regarded as necessary by the Bank. If the client was not willing to assist in ensuring that the terms of that insurance were complied with, by paying for quality certificates, then it would not have obtained the repo finance.
  1260. I therefore consider that the evidence shows that, on the balance of probabilities, the Bank would have acted so as to obtain the relevant benefit or avoid the relevant risk in relation to Icestar 2 and 3 transactions. It would have sought the credit insurance equivalent to that which it thought that it was getting under the TPC, or would have wound down its transactions with Euromar and Transmar if and to the extent that the insurance was unavailable. This aspect of causation is therefore established.
  1261. Actions of third parties

  1262. Allied Maples shows that that there is a second aspect of causation where the actions of third parties are in issue. The claimant must show a real or substantial chance that the third party would have acted so as to confer the relevant benefit or avoid the risk. In the present case, there are two potentially relevant third parties, or groups of third parties. In relation to what the Bank described as course (i), there are the credit risk insurers. The question is therefore whether there was a real and substantial chance that credit insurers would have agreed to write insurance which would have provided protection against the Bank's losses in relation to Transmar and Euromar. In relation to course (ii), the question is whether there was a real and substantial chance that Euromar and Transmar would, if the transactions had been unwound, have made repayments to the Bank.
  1263. Edge's submissions were directed at course (i), and did not really touch on course (ii). On the latter point, Mr. Stroink's evidence, summarised above, was that Euromar and Transmar made substantial payments after August 2015. This showed that the companies had liquidity at that time, enabling payments to be made. There was also documentary evidence, in the Boston Consulting Group report of August 2016, that Euromar's financial position was satisfactory in 2015, but had deteriorated in the first half of 2016. In these circumstances, there was clearly a real and substantial chance that the Bank would have received significant repayments if course (ii) had been followed, bearing in mind that the cover was initially sought in June/ July 2015 and that the need to approach the credit risk market would therefore have been identified at that time.
  1264. As far as course (i) is concerned, I have no doubt that there was a real and substantial chance that credit insurance would have been obtained which would have provided protection against the Bank's losses in relation to Transmar and Euromar.
  1265. The experts were agreed, in their joint report, on a number of important matters. In January 2016 there was a group of around 60 insurers who could have been approached for cover. In January 2016, the market was generally "soft". There was sufficient capacity and insurers were actively looking for business. The market in general at that time was not beginning to take a "harder" stance. Mr. Hedley confirmed that the market had been even softer in July 2015. The experts agreed that "in theory, that alternative credit risk insurance would have been available for Icestar II and III type transactions". They were also agreed that an alternative credit risk policy of 12 months could have been placed on a losses occurring or risks attaching basis. Both types of cover would have been available in the market in January 2016. It was agreed that the pricing range for a credit risk insurance policy would have been in the range of between 1.35% and 2% per annum of the actual financial exposure incurred by Icestar during the policy period. Both Mr. Hayter and Mr. Hedley would have priced the risk towards the higher end of the range, but they were aware that "there would be capacity available at prices towards or at the lower end of the stated range". They were agreed that a level of indemnity of 90% would have been available, in principle. But there were various points of disagreement which I describe in more detail below.
  1266. The issues between the experts as to the availability of cover are to be approached, as Mr. Hedley acknowledged in his evidence, in the context of the following. The market was soft, with underwriters generally looking for business. The risk to be insured fell squarely within the bounds of standard trade credit insurance. Underwriters would have welcomed the Bank as an insured: ABN Amro was an internationally well regarded corporation whose business underwriters would have sought to secure. There were also other features of the risk that made it attractive. It had a short duration: each repo transaction was 3 months, at the end of which the Bank could decide whether or not to roll over. The security structure was good, because the Bank had legal title to the goods, not merely a security interest. The countries where the risks were situated were attractive. The most recent financial statements of Transmar showed a picture of increased income and profit. As Mr. Hedley said, the financial statements would justify secured repo-type borrowing, or a secured borrowing base facility.
  1267. Against this background, I consider that there was a real and substantial chance that the Bank would have been able to obtain credit risk insurance in respect of its repo exposure to Transmar and Euromar, whether coverage was sought for those companies alone or as part of a wider package. Mr. Hayter's view was, in substance, that there would have been no difficulty in the Bank securing the coverage for Transmar and Euromar that it wanted. I consider that this evidence was more plausible, given the background summarised above, than the contrary view in so far as that was expressed by Mr. Hedley. It is supported by the fact that Mr. Hedley himself wrote a credit risk for Natixis, a financial institution, in respect of its exposure on Transmar's revolving credit facility. He did this because, as he explained in his first report, Transmar was a well-established firm which had been in business since the 1980s and had never reported a loss. It was also one of the largest players in its particular field. Their figures, as Mr. Hedley said, looked good – albeit that later on it was revealed that they were affected by fraud. Mr. Hedley also wrote a risk on Euromar for another insured in either 2014 or 2015, albeit that the policy was cancelled for non-payment of premium. He said that the credit quality of Transmar and Euromar based on the latest available financial statements would have been satisfactory.
  1268. The principal reason advanced by Edge, in their closing submissions, as to why credit risk underwriters would not have agreed to underwrite a policy for Icestar covering default by Euromar and Transmar alone, was that it would have appeared that Icestar was engaging in adverse selection. Mr. Hayter disagreed with that. He said that it was entirely normal for an insured to select a certain risk or group of risks from its portfolio to insure due to its/ their credit profile or other concerns. I accept that this is so, and I did not understand Mr. Hedley to dispute it as a general proposition. Indeed, he had himself written risks relating to a particular exposure to Transmar and Euromar as described above, and I did not understand those risks to have been part of a wider portfolio. Mr. Hedley also accepted in his report that there could be legitimate reasons for selecting an exposure to a particular counterparty for "singleton" insurance. The normality of "singleton" policies is illustrated by the fact that, as Mr. Hayter explained, there would have been a series of such policies, in respect of each insured, if the Bank had decided to insure its portfolio of smaller clients who were doing Icestar 2 and 3 transactions.
  1269. I also considered that the "adverse selection" argument had an air of unreality about it. For reasons already given, it is more likely than not that the Bank would have sought credit risk insurance for its portfolio of smaller Icestar 2 and 3 clients. There was therefore no intention adversely to select against credit risk insurers. But even if this were wrong, and a decision were taken to seek insurance for Transmar and Euromar alone, this would not in my view have led credit insurers to decline to write the risk (still less that there would have been no real or substantial chance of the risk being written). Rather, as Mr. Hedley's evidence indicated, it might have led at most to a discussion as to why it was that coverage was sought for those companies alone. There could be legitimate grounds for doing so, and I have no doubt that a legitimate and satisfactory explanation could have been given. This is particularly so when one bears in mind – as Edge emphasised in their closing submissions in the context of their argument that the Bank would not have sought credit risk insurance at all in the counterfactual world – that: the Bank had a longstanding and robust relationship with the two companies; the companies had never defaulted on any repo deal with anybody, or committed any other kind of default; and the Bank was "comfortable with the credit risk" presented by Transmar and Euromar.
  1270. Mr. Hedley said in the joint report that a legitimate reason for seeking cover for these two companies was, for example, that they represented Icestar's largest single exposure with volumes increasing and limit relief needed. It is not possible or necessary to work out exactly what explanation the Bank would have given for seeking cover on these two companies alone, if that is what they had decided to do. The Bank never had to give an explanation, because they believed that they were covered under the TPC and therefore no question arose of needing to approach the credit risk insurance market. But if there had been a conversation where they were asked to explain why cover was sought for those companies alone, the Bank would either have provided an acceptable explanation, or would simply have said that they were happy to insure more companies within their portfolio if that is what the insurers wanted. That would have put an end to any real debate about adverse selection.
  1271. The other principal argument that was advanced by Edge in this context, was that the repo deals were being repeatedly rolled over or extended. They submitted that credit risk underwriters would have been concerned that this indicated or would give rise to a concern that the counterparties could not repurchase the goods. Mr. Hayter's evidence was that there were advantages to this way in which the business was being operated. The 90 days term for the repo transactions gave the obligor the opportunity to close out the deal and on-sell the goods, and also gave the Bank the opportunity not to renew. He said that underwriters would understand that rollovers were integral to the way in which Transmar/Euromar needed to finance its stock, and that the companies would buy back when one of their clients wanted to purchase the goods. That was how stock financing worked. Credit risk underwriters would not have been concerned, particularly bearing in mind the overall record of the companies and that the stock financing deals were, as Mr. Hayter said, "very much a small part of a larger relationship that ABN had with Transmar and there is no evidence that I have seen that would suggest that credit risk underwriters would have been concerned about Transmar in January 2016".
  1272. Mr. Hayter had the advantage over Mr. Hedley of actually having underwritten risks in respect of repo transactions. His experience of working with hard and soft commodity traders and related entities was that rollovers are a common feature of their business. I accept this evidence, which is consistent with the Bank's evidence. I am willing to accept that it is possible that some credit risk underwriters might have asked questions about the rollovers. Had they done so, as Mr. Hayter said, it would have been a matter of concern if the Bank could not give a good reason. Had the conversation happened, it seems to me likely that the Bank would have given an explanation which was consistent with the fact that the Bank itself was not concerned about the rollovers; for example, that they are a well-known feature of the business, and that there had never been any difficulties with either company. I was unpersuaded that this would have led to the risk being declined, particularly bearing in mind that there was a potential market of at least 60 underwriters in a soft market. At worst, from the Bank's perspective, it might have led to terms which limited the extent of rollovers or the coverage for transactions which had been rolled over for a period which the underwriters considered to be too long. But even if that had happened, the Bank would still have obtained some insurance coverage. It would also have been in a position to shape its response to the absence or potential absence of coverage for any uninsured transactions, by closing out or not entering into transactions which would not be insured.
  1273. Accordingly, I am satisfied that there was a real and substantial chance of credit risk underwriters providing the credit risk cover that the Bank would have sought if appropriately advised.
  1274. In reaching this conclusion, I generally prefer the evidence of Mr. Hayter to Mr. Hedley where they disagree. Both witnesses were highly experienced, and I consider that both sought to give helpful evidence to the court. Both of them expressed views which they conscientiously held. However, I thought that Mr. Hayter's evidence as to the fairly ready availability of cover was more consistent with the features of the soft market, the volume of underwriters who could potentially be approached, and the overall background including the attractive nature of both the Bank and Transmar as described above. Overall, I think that underwriters would not have required a great deal of persuasion to write this cover, whereas Mr. Hedley's evidence gave the impression that they would have been searching for reasons not to do so.
  1275. Ms. Sabben-Clare also made the fair point that Mr. Hedley had jumped on a bandwagon when arguments started to emerge, in the summer of 2020, concerning the Bank's failure to carry out quality checks on the collateral. Mr. Hedley's first report did not suggest that this was something which credit risk underwriters would be interested in, and the sample policies provided to the court did not contain terms which supported Mr. Hedley's argument. I thought that Mr. Hedley's evidence in this respect was rooted in hindsight. I was more impressed by Mr. Hayter's point that the real focus of underwriters would be the risk of default, rather than the risk that the collateral would or would not prove adequate. The former was the real risk. A prudent credit risk underwriter would not expect or rely upon the collateral to provide salvation if the default risk went south.
  1276. Quantification of the chance

  1277. The Bank submitted that this was a case where the court can and should say with confidence that the victim of professional negligence would have avoided the loss but for the breach, at least to the extent of the 90% indemnity that would have been payable under alternative cover. Credit would need to be given for the premium payable, but this would have been at the lower end of the range given by the experts.
  1278. Edge submitted that if credit risk underwriters had been willing to underwrite a policy for Icestar covering default by Euromar and Transmar alone, they would have done so on terms that limited their liability for the defaults claimed in the present action. The details of that argument are further described below.
  1279. Edge submitted that the chance of getting cover satisfactory to the Bank that would have protected it against the losses on Transmar and Euromar was small: in percentage terms no more than 20%, bearing in mind that the Bank would have been unlikely to want to pay for such cover given the apparent strength of their counterparties, and that underwriters in the credit risk market would in all probability have been reluctant to cover the Bank for credit risk on Transmar and Euromar alone. Moreover, even if the Bank could have obtained cover, it would in all likelihood have been on terms which eliminated or at least substantially reduced its recovery for the sums claimed in this action.
  1280. The expert evidence addressed in some detail the question of the terms of cover which would have been available to the Bank. The principal disagreements were as follows.
  1281. Mr. Hayter considered that the level of indemnity would have been at 90% which is the standard level for trade credit insurance cover. Mr. Hedley said that he personally would probably not have written at the level of 90% cover, particularly if Transmar and Euromar had been selected out of the Icestar portfolio in isolation. There would in his view have been an indemnity level of only 50%, in effect "co-insurance".
  1282. Mr. Hayter would not have required Icestar to have checked the quality of the underlying goods prior to providing financing, and he believed that most underwriters would take the same view. Confirmation that the stocks existed would have been sufficient. Mr. Hedley said that, if underwriting the risk, he would have required Icestar to check the quality of cocoa products for which it was entering into repo transactions.
  1283. Mr. Hedley considered that the available policy would specify the maximum risk tenor for any declaration. This would comprise a maximum original financing period, and a "Maximum Extension Period". The effect of such a clause would be to impose a maximum risk tenor equal to the lower of 360 days and the period for which the relevant commodities could be expected to maintain their market value and useful life (from the date when financing was to be advanced).
  1284. Mr. Hayter disagreed. He had never seen a Maximum Extension Period clause in a single risk credit policy of the type that Icestar would have obtained. They were more often found in multi-buyer, whole turnover type policies. In any event, any competent broker would have rejected such clauses as irrelevant and unnecessary, since Icestar had an exemplary track record with Transmar and Euromar. As at January 2016, there had never been a default and there had been numerous and substantial buybacks. It had to be borne in mind that the Bank would have been considered by credit risk underwriters to be an attractive insured as at January 2016.
  1285. Mr. Hedley considered that cover would be provided on a "risks attaching" basis, thereby excluding transactions where finance had been advanced prior to policy inception. It would therefore cover new financings advanced during the policy period, and would not pick up what Mr. Hedley characterised as "stale, old, refinanced financings" that were originally disbursed before policy inception.
  1286. Mr. Hayter disagreed. Coverage was available in the market on a "losses occurring during" basis; ie to cover losses which occurred during the policy period. The suggestion that the Bank was involved in stale and old refinancings underlay many of Mr. Hedley's points. But this was, in Mr. Hayter's view, not a correct or fair characterisation of the transactions. Each rollover was a freshly agreed contractual agreement, providing short term finance. Each party could withdraw when the 90 day financing period came to an end. This was not a case where there were overdue payments which had not been made in the past. There had never in fact been any default.
  1287. Mr. Hedley considered that the policy would have contained a "stop financing" condition. This would require Icestar to stop financing new transactions when certain events occurred; for example when any transaction already on the books with the counterparty became overdue or was rescheduled beyond a pre-agreed period. This agreed period would probably be the maximum risk tenor under the policy. Mr. Hayter disagreed, and repeated his point that a rollover was not equivalent to and did not evidence an inability to repay. He thought that the majority of credit risk underwriters would not have required such a clause.
  1288. The parties expanded upon these various points in their written opening and closing submissions. Since the present issue involves the overall evaluation of the chance which the Bank lost by reason of Edge's breach, it is not in my view necessary to resolve each dispute between the experts. I am concerned with an area of uncertainty, and I must take into account all significant factors in the evaluation of the relevant chance which the Bank lost.
  1289. I do not consider that the possibility of cover only being provided on the basis of a 50% retention (effectively co-insurance), rather than the usual 10%, is a significant factor to be taken into account. Mr. Hayter said that 75-80% of all policies issued in the credit and political risks market are at 90% indemnity. This includes credit risks that are entirely unsecured and in many cases have a far worse credit and performance profile than Transmar and Euromar displayed in early 2016. Mr. Hedley's justification for a 50% retention was, as Edge made clear in its opening submissions, based upon the appearance of adverse selection, rather than his points on rollovers. For reasons already explained, I do not consider that the adverse selection argument has any substance on the facts of the present case. In my view, the evaluation of the relevant chance should not take into account the possibility that cover would be below the standard 90% indemnity level.
  1290. The remaining disagreements between Mr. Hedley and Mr. Hayter concern, in essence, issues as to whether underwriters would have imposed terms which would have had the effect of excluding certain transactions (eg because the cover was on a risks attaching basis) or required the Bank to comply with certain conditions (eg to provide quality certificates). As far as the latter point is concerned, I do not consider that the evaluation of the relevant chance should take into account the possibility that the insurers would have imposed terms, and that the Bank would then have failed to comply with them. If the Bank had taken the step of paying substantial premium for trade credit cover which imposed certain obligations, then it is not difficult to conclude that a professional organisation such as the Bank would (as Mr. Stroink said) have complied with those terms. There is no point paying premium for cover and then declining to abide by the terms of the contract.
  1291. Having considered the evidence as a whole, I have ultimately concluded that it is not appropriate to make any discount from the amounts claimed by the Bank to reflect the possibility that underwriters would have imposed the terms which were proposed by Mr. Hedley. There are two reasons for this conclusion.
  1292. First, I considered that the chances of the Bank obtaining cover, without Mr. Hedley's restrictions, were very strong. It may be that some underwriters would have taken Mr. Hedley's more circumspect approach, although there were certainly respects (for example, the late appearance of the "quality checks" argument) in which (as it seemed to me) the benefit of hindsight had an impact on Mr. Hedley's evidence. In my view, however, it is overwhelmingly likely that a substantial and sufficient number of underwriters would not have imposed the terms proposed by Mr. Hedley. I have already described important features of that market in reaching the conclusion that the cover would have been available, in particular: the soft market conditions in 2015 and early 2016 with underwriters seeking good business; the large number of underwriting players; the attraction of the Bank as an insured; the apparently good financial position of Transmar, which was a substantial and well-established trader. All of these various factors would have meant that the chances of underwriters requiring or insisting on Mr. Hedley's proposed terms were relatively small. It is also important in my view that the Bank has been able to call (in Mr. Hayter) a very experienced and competent credit risk underwriter, who was clearly of the genuine view that he personally would have written this risk and that many others would do so. This reinforces my conclusion that the chances of obtaining the relevant cover were very strong.
  1293. Secondly, an unusual feature of the present case is that the evaluation of the Bank's chance does not simply depend upon the chance of obtaining an insurance policy with satisfactory terms. That is because if special terms were imposed, in particular terms which had the effect of excluding certain transactions from cover, the Bank would still be able to protect itself. It could in those circumstances have unwound uncovered transactions or declined to enter into new transactions which would not be covered. In her opening submissions, Ms. Sabben-Clare submitted that Mr. Hedley's points on applicable restrictions are "sterile" even if they are correct. This was because if the Bank had been advised of limitations on or terms of cover, it would have closed out or not entered non-complying transactions. Thus, whether through insurance or alternative action, the Bank would have avoided the losses that it actually suffered. That case was supported by Mr. Stroink's evidence. I did not consider that Edge had any, or at least any effective, answer to this point.
  1294. If therefore the Bank's claim against all underwriters had failed (and thus, as I have concluded, Edge would have been liable to the Bank for the shares of all underwriters, and not just Ark and Advent), then I would have concluded that damages should be assessed on the basis of the full 90% that would have been covered, or the loss avoided, if the Bank had either taken out insurance cover or responded to the consequences of such cover being unavailable or restricted.
  1295. On that hypothesis it remains necessary, however, to take into account the additional premium that would have been payable for the requisite cover. The experts' range was 1.35 – 2%. Both of them would have priced the risk towards the higher end of the range, although they were aware "that there would be capacity available at prices towards or at the lower end of the stated range". Mr. Hedley explained in his oral evidence, however, that he did not think that there would be sufficient capacity to write the entire risk at the lower end of the range.
  1296. It seems to me that this issue should also be approached on a loss of a chance basis, and it should also take into account the likelihood that the Bank would have sought and obtained, and therefore paid for, cover in respect of some or all of the other smaller customers in the relevant portfolio. If one treats Euromar and Transmar as one customer, there were four such other customers (rather than the 5 to whom Mr. Hedley referred in his evidence concerning the amount of premium increase in the event of coverage of more counterparties). I consider that the appropriate allowance for these factors would be to take the relevant figures for rating at 1.6%, which Mr. MacLeod worked out to be £ 594,382 plus USD 14,681. Those figures should then be doubled so as to reflect the premium that would have been payable in respect of the other companies within the portfolio.
  1297. Accordingly, if the Bank's claim against underwriters had completely failed for the reasons discussed in this section, I would have assessed damages on the basis of 90% of the Bank's losses less the premium described in the previous paragraph.
  1298. J4: The Bank's residual claim against Edge for irrecoverable costs

  1299. The Bank's claim has, however, succeeded, except as against Ark and Advent where, for reasons already explained, the claim against Edge succeeds in respect of 100% of the liability that those underwriters would otherwise have had. I also presently consider (but reserve this for final determination subsequently) that Edge would be liable for any costs liability of the Bank towards Ark and Advent, as well as any irrecoverable costs incurred by the Bank in the action against those two underwriters.
  1300. My conclusions in Sections J2 and J3 above would be relevant in the event of a successful appeal by underwriters (apart from Ark and Advent) against this judgment. They are also relevant, however, because the Bank advances a residual, and presently unquantified, claim against Edge for any irrecoverable costs of litigation against those underwriters where the claim has succeeded, flowing principally from the FNCB duty discussed in Section J2.
  1301. The Bank submitted that I could and should not deal with this point now. It would only arise if and when an "incomplete costs order" is made.
  1302. Edge submitted that since this was a trial of all issues, I should at least address the issue of Edge's liability in principle for the different heads of costs, even if the precise quantification of any costs attributable to specific categories of the defence must await determination following judgment. Ms. Healy submitted that Edge could not be held liable for any irrecoverable costs in relation to the sue and labour dispute, in relation to which the Bank did not advance a claim that Edge would be liable if that defence had succeeded. Edge argued that the FNCB duty is concerned with whether the cover is sufficiently clear. If the cover is unclear and there is an unnecessary dispute over construction, then the broker may be liable for the costs of that dispute. But it did not follow that if underwriters took a (bad) point on construction, and then threw in every other conceivable point – giving rise to wide-ranging and expensive litigation – the broker should then pick up the whole tab for the insured's irrecoverable costs. Aside from the costs of the construction dispute, the broker cannot be liable, because those costs were not caused by a breach of the FNCB duty. Accordingly, none of the Bank's costs of litigating any issue other than the construction dispute would be recoverable from Edge.
  1303. I do not think that it is appropriate, at the present stage, to come to a final determination on the issues of causation and quantum of irrecoverable costs. It may well be that no such claim is ever made: Ms. Sabben-Clare described it as a "sliver" of a potential claim. This is because the Bank accepts that any claim for damages would need to give credit for the substantial additional premium that would have been payable, and the retention that would have been applicable, if the Bank had not obtained cover from the cargo underwriters under the TPC, but instead had obtained it in the trade credit insurance market. Also, it does seem to me to be difficult properly to address questions of causation and quantum, in the context of a possible claim for irrecoverable costs, in circumstances where no orders for costs have been made and no claim can be or has been quantified.
  1304. I consider it preferable to reserve those issues for further determination, should they arise. For present purposes, it suffices to say that I was not persuaded by Ms. Healy's argument that Edge's liability is limited to any irrecoverable costs of the construction argument, or that deductions should be made because the underwriters have advanced various unsuccessful arguments.
  1305. At present, it seems to me that the need for the present litigation, and the costs which the Bank has incurred, are a foreseeable consequence of the breaches by Edge discussed above. The construction argument was the point which RSA first took when the claim was declined, and it remained at the heart of their defence. At a comparatively late stage, it was supplemented by the other defences which I have rejected (avoidance, due diligence and sue and labour), but the construction argument remained a central point. When an insured, in consequence of a broker's breach has to litigate on a central issue as to the construction of the policy, it is far from unusual for other possible defences to be added into the mix, including defences (such as rectification and estoppel in the present case) which are closely related to the construction issue. It is therefore not at all clear why Edge should not pay for all consequences of the Bank having become embroiled in litigation, even though the underwriters' defences were ultimately not confined to issues of construction but where those issues remained of prime importance.
  1306. Even if there were to be some deduction, it is not clear how this could or would be quantified, for example whether it would involve an apportionment of any irrecoverable costs or whether it would be necessary for Edge to show more clearly the extent to which particular irrecoverable costs were said to be referable to issues other than the construction issue.
  1307. In these circumstances, I make no final decision about the issues of causation and quantification which might arise on any claim which the Bank advances for irrecoverable costs. Such decision is best made as and when a quantified claim is made, and in the light of the facts relevant to that quantified claim, when the argument will be less abstract than at present.
  1308. J5: Edge's liability in respect of the Clause 3 defence

  1309. I have not previously considered the question of whether Edge would be liable to the Bank in the event that all of the underwriters' defences failed, except for the argument concerning Clause 3 addressed in Section G above. That argument concerned the Bank's allegedly reckless or negligent conduct concerning its failure to check the age and quality of the collateral.
  1310. The Bank submitted that if the case had failed as a result of the Bank's failure to check the age and quality of the collateral, then by definition the age of the cargo and the Bank's procedures will be matters material to the risk. Edge failed to take any steps to elicit information material to the risk under the TPC from the Bank and to communicate that to underwriters. If it had done, underwriters would have been aware of the age of the cargo and the procedures followed and would either have declined the risk, in which case the Bank would have changed its procedures and/or discontinued these transactions, or would have had no avoidance defence (and would be estopped from suggesting that Icestar had failed to act with due diligence). They referred to Edge's case, through Mr. Hedley's evidence, that seeking cover in the credit market would inexorably have led to a discussion about the quality of the cargo and elicited the information that the Bank was not carrying out quality checks.
  1311. I am doubtful as to whether any liability on the part of Edge would flow, in the manner described, from its failure to elicit information material to the risk and to provide it to cargo underwriters. Those underwriters have not relied, by way of defence, upon an allegation of non-disclosure concerning the age and quality of the collateral. Their avoidance case, in essence, is that they would not have written the policy if the TPC had been pointed out to them or the Bank's intention and purpose in seeking that cover had been explained. They were not saying that they would in principle have written the policy, but would have declined to do so if they were told about the Bank's approach to collateral. Moreover, I have held that the avoidance case fails because of the NAC.
  1312. However, I consider that the Bank's case is more powerful if the relevant focus is not upon what the cargo underwriters would have done, but how matters would have proceeded if Edge had gone to credit risk insurers. (The Bank's argument, as summarised above, looks at both sets of insurers). On the basis of my earlier findings, this would have happened if the Bank had been correctly advised at the outset. For the reasons discussed in Section J3 above, an approach to the credit insurance market would, by one route or another, have resulted in the Bank being protected as to 90% of their losses. Any problems created by Clause 3 of the cargo policy would not therefore have arisen. I therefore conclude that the Bank would be entitled to recover against Edge if their claim against underwriters had failed as a result of Clause 3 of the policy.
  1313. K: Conclusion

  1314. The Bank's claim against the underwriters succeeds in full, save in relation to Ark and Advent where it fails only by reason of an estoppel arising from the manner in which the risk was broked to the underwriters of those companies.
  1315. Edge is liable for 100% of the recovery that the Bank would, but for the estoppel, have made against Ark and Advent. I also consider (subject to further submissions) that Edge's liability extends to any costs liability of the Bank to Ark and Advent, as well as any irrecoverable costs incurred by the Bank in the action against those two underwriters.

  1316.  

  1317. Edge is also liable in principle for the Bank's irrecoverable costs of the present proceedings against the underwriters in respect of whom the claim has succeeded, but issues as to the quantification of such costs (and any related issues of causation) shall be determined hereafter.


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2021/442.html