England and Wales High Court (Commercial Court) Decisions

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

		Royal Courts of Justice Strand, London, WC2A 2LL
		18/07/2014

Mr Justice Andrew Smith:

Introduction

1. This claim by Kaupthing Singer & Friedlander Limited ("KSF"), brought by their administrators, is in respect of US$65 million that the defendant, UBS AG ("UBS"), is said to have failed to pay under a foreign exchange transaction (the "FX trade") made on 1 September 2008. KSF were to buy US$100 million in exchange for £55,614,723.44, the respective payments to be made on 3 October 2008. The FX trade was made under a Master Agreement on the 1992 ISDA form made on 26 August 2004 (the "Master Agreement"), which was terminated on 9 October 2008 when UBS gave KSF notice of an "Event of Default" after the court had appointed administrators of KSF on 8 October 2008. KSF's case is that they duly made the sterling payment, UBS have paid only $35 million and the rest of the $100 million has not been paid. At the start of the trial, KSF's pleaded case was that "UBS continues to owe KSF the principal sum of $65,000,000 under the FX trade", and the relief claimed was "payment of US$65,000,000" (and interest and costs). Mr Duncan Matthews QC originally presented the claim on the basis that "KSF did not get paid and so UBS' US$65m payment obligation to KSF was never discharged. The claim is as simple as that, albeit that amid the challenges of a hugely complex administration … it took KSF's Administrators some time to piece this all together and to identify that UBS remained liable."
2. UBS pleaded in defence to the claim inter alia what I shall call the "termination defence": that, once the Master Agreement was terminated on 9 October 2008, no further payments had to be made under the Master Agreement or any transactions made under it. UBS's various other defences to KSF's pleaded claim can be gathered into these groups:

i) That they discharged their obligation with regard to the $65 million by or as a result of a payment (the "Payment") that they made on 3 October 2008 to JP Morgan Chase ("JPMC"), notwithstanding that because of a mistake on their part it was made with instructions that it be credited to the account of Kaupthing Bank hf ("Khf"), KSF's parent company: discharge defences.

ii) That they satisfied their obligation by or as a result of the close-out process under the Master Agreement: the accord and satisfaction defence.

iii) That, because of parties' understanding of their rights and obligations with regard to the FX trade and their conduct and exchanges following the Payment or in the course of the close out process (or both), KSF are precluded (whether through waiver or some form of estoppel) from enforcing the obligation to pay the $65 million: estoppel and waiver defences.

3. Although the dispute has an international flavour, not least because the Payment was made to JPMC in New York, it was agreed that I should decide all issues in accordance with English law and there was no evidence of foreign law.

The trial

4. The trial lasted only four days, but it did not run without difficulties. They arose largely from KSF's pleadings and the witness evidence that they sought to present. I must explain a pleading question concerning the termination defence in some detail at the outset (although, as I shall explain, it is not the only issue about pleadings), and for this purpose I must refer to the relevant terms of the Master Agreement. The FX trade was a part of a single agreement with the Master Agreement and other transactions under it: section 1(c) of the Master Agreement provided that "All transactions are entered into in reliance on the fact that this Master Agreement and all confirmations form a single agreement between the parties …". The payments under the FX trade were therefore to be made under section 2(a)(i) of the Master Agreement. Section 2(e) provides for payment of interest at a "Default Rate" if a party defaults in the performance of a payment obligation, the Default Rate being defined by reference to what the payee certifies would be the cost if it funded or were to fund the relevant amount, plus 1% per annum. By section 6(a) of the Master Agreement, if an "Event of Default" had occurred and was continuing with respect to one of the parties, the other party was entitled to give notice specifying an "Early Termination Date in respect of all outstanding Transactions". There is no dispute (and it is admitted by KSF on the pleadings) that UBS did give such notice and that it was effective to specify 9 October 2008 as the Early Termination Date for the Master Agreement. Section 6(c)(ii) of the Master Agreement provides that "Upon the occurrence or effective designation of an Early Termination Date, no further payments … under Section 2(a)(i) in respect of the Terminated Transactions will be required to be made …", "Terminated Transactions" being defined, as far as is relevant, as "all Transactions in effect immediately before the effectiveness of the notice designating [the] Early Termination Date". There was also no dispute that the FX trade was an "outstanding Transaction" (within the meaning of section 6(a)) and a "Terminated Transaction" (within the meaning of section 6(c)(ii)), notwithstanding that, on KSF's case, payment under it had accrued due and indeed was overdue.
5. In these circumstances, in place of the rights and obligations under Terminated Transactions including the FX trade, the parties' rights and obligations were governed by the formula in section 6(e) of the Master Agreement that the parties had chosen in the Master Agreement, in this case the so-called "Second Method and Loss", supplemented by section 6(d). So far as material, those sections provide as follows:

Section 6(d) –

"Calculations.

(i) Statement: On or as soon as reasonably practicable following the occurrence of an Early Termination Date, each party will make the calculations on its part, if any, contemplated by Section 6(c) and will provide to the other party a statement (1) showing, in reasonable detail, such calculations (including all relevant quotations and specifying any amount payable under Section 6(e)) and (2) giving details of the relevant account to which any amount payable to it is to be paid. …

(ii) Payment Date: An amount calculated as being due in respect of any Early Termination Date under Section 6(c) will be payable on the day that notice of the amount payable is effective (in the case of an Early Termination Date which is designated or occurs as a result of an Event of Default) and on the day which is two Local Business Days after the day on which notice of the amount payable is effective (in the case of an Early Termination Date which is designated as a result of a Termination Event). Such amount will be paid together with (to the extent permitted under applicable law) interest thereon (before as well as after judgment) in the Termination Currency, from (and including) the relevant Early Termination Date to (but excluding) the date such amount is paid, at the Applicable Rate. …"

Section 6(e) –

"Payments on Early Termination. If an Early Termination Date occurs, the following provisions shall apply based on the parties' election in the Schedule of a payment measure, either "Market Quotation" or "Loss", and a payment method, either the "First Method" or the "Second Method". … The amount, if any, payable in respect of an Early Termination Date and determined pursuant to this Section will be subject to any Set-off."

(i) Events of default. If the Early Termination Date results from an Event of Default:- …

(3) Second Method and Market Quotation. If the Second Method and Market Quotation apply, an amount will be payable equal to (A) the sum of the Settlement Amount (determined by the Non-defaulting Party) in respect of the Terminated Transactions and the Transaction Currency Equivalent of the Unpaid Amounts owing to the Non-defaulting Party less (B) the Termination Currency Equivalent of the Unpaid Amounts owing to the Defaulting Party. If that amount is a positive number, the Defaulting Party will pay it to the Non-defaulting Party; if it is a negative number, the Non-defaulting Party will pay the absolute value of that amount to the Defaulting Party.

(4) Second Method and Loss. If the Second Method and Loss apply, an amount will be payable equal to the Non-defaulting Party's Loss in respect of this Agreement. If that amount is a positive number, the Defaulting Party will pay it to the Non-defaulting Party; if it is a negative number, the Non-defaulting Party will pay the absolute value of that amount to the Defaulting Party".

(I have included headings in setting out the sections, but they do not affect their construction: section 9(g).)

The expression "Loss" is defined in section 14 of the Master Agreement:

"'Loss' means with respect to this Agreement or one or more Terminated Transactions, as the case may be, and a party, the Termination Currency Equivalent of an amount that party reasonably determines in good faith to be its total losses and costs (or gain, in which case expressed as a negative number) in connection with this Agreement or that Terminated Transaction or group of Terminated Transactions, as the case may be, including any loss of bargain, cost of funding or, at the election of such party but without duplication, loss or cost incurred as a result of its terminating, liquidating, obtaining or reestablishing any hedge or related trading position (or any gain resulting from any of them). Loss includes losses and costs (or gains) in respect of any payment or delivery required to have been made (assuming satisfaction of each applicable condition precedent) on or before the relevant Early Termination Date and not made, except, so as to avoid duplication, if Section 6(e)(i)(1) or (3) or 6(e)(ii)(2)(A) applies. … A party will determine its Loss as of the relevant Early Termination Date, or, if that is not reasonably practicable, as of the earliest date thereafter as is reasonably practicable. …".

6. By letter dated 20 March 2009 UBS advised KSF that they had "made [their] calculation under Section 6(e) of the Agreement", and that the "total amount payable by UBS to KSF is USD $5,199,194.05". They attached a calculation that showed that this reflected that UBS were to pay $5,956,099.00 in respect of foreign exchange transactions and a small sum in respect of equity transactions, and against that they were to receive some $800,000 in respect of fixed income, currency and commodities ("FICC") transactions. It did not bring into account anything in relation to the FX trade or the US$65 million that KSF claim was not paid under it. I shall explain KSF's response to this letter and calculation and the parties' exchanges later in my judgment, but on 6 October 2009 UBS paid KSF the $5,199,194.05.
7. When Mr Matthews opened KSF's case, I asked him about how the claim was pleaded and he accepted that UBS did not continue "to owe KSF the principal sum of $65,000,000 under the FX trade". He did not contend that the claim was that stated in his written opening, but said that "the claim is put as being an under-calculation under the close-out on the ISDA agreement". He drew my attention to paragraph 8 of the particulars of claim, which stated:

"On 9 October 2008, UBS served notice of termination of all transactions with KSF under the Master Agreement, relying upon KSF having been pleaded into administration on 8 October 2008. On 20 March 2009 UBS set out its calculation of sums payable upon early termination under Section 6(e) of the Master Agreement. UBS's calculation should have included the US$65,000,000 but failed to do so. In the premises:

a. the sum of $65,000,000 unpaid by UBS should attract interest from 3 October 2008 to judgment at the Default Rate under the Master Agreement …

b. alternatively [statutory interest at the Default Rate from 3 October 2008]

c. in the further alternative, [statutory interest awarded on some other basis]".

8. On the second day of the trial, KSF applied for permission to make these amendments to their particulars of claim:

i) to amend paragraph 7 so as to read, "In the circumstances, as at 9 October 2008 UBS continuesd to owe KSF the principal sum of $65,000,000 under the FX trade";

ii) to amend paragraph 8 so that it and a new paragraph 9 would read as follows:

"8. On 9 October 2008, UBS served notice of termination of all transactions with KSF under the Master Agreement, relying upon KSF having been placed into administration on 8 October 2008. On 20 March 2009, UBS set out its calculation of the sums payable upon early termination under Section 6(e) of the Master Agreement (a sum due to KSF of $5,199, 194. 05). UBS' calculation of its "Loss" should have included the unpaid US$65,000,000 in respect of the (now terminated) FX Trade still payable to KSF but failed to do so. In the premises:

(i) UBS' calculation was not a "reasonable determination" of its Loss for the purposes of the Master Agreement and the sum calculated and paid by UBS should have been $65,000,000 greater;

(ii) KSF is therefore entitled to a further payment by UBS of $65,000,000;

(iii) in the alternative, KSF claims damages in the sum of $65,000,000, being the shortfall in payment to KSF arising from UBS' failure to comply with its obligation to "reasonably determine" its Loss;

9. As to interest;

i) the sum of $65,000,000 …"; and

iii) to add a prayer for damages as an alternative to the claim for payment of $65,000,000.

Thus, the draft amendment acknowledges (as did Mr Matthews in presenting the application) that the original debt under the FX trade did not survive beyond 9 October 2008 (even if, contrary to UBS's contentions, it survived until then).

9. UBS resisted KSF's application to amend. Ms Sonia Tolaney QC submitted on their behalf that at the core of the amendment is an allegation that UBS's calculation is unreasonable, that that is a new allegation, and that it could not fairly be resolved at the trial. Neither party suggested that the trial should be adjourned in order to deal with the case that would be introduced by the proposed amendment. In those circumstances, as I briefly explained, I decided that the trial could fairly continue on the basis that the proposed amendments had been made, and that it should do so, and that I would determine in this judgment whether to grant the application to amend.
10. In his closing submissions, Mr Matthews submitted that the $65,000,000 might be claimed under the Master Agreement in three ways:

i) That, because UBS's calculation omitted the claim for $65 million, it was not a determination of "Loss" as defined in the Master Agreement and was "invalidated": that it was a nullity; and in those circumstances, it was said, it falls to the court to make the determination of "Loss" (the "invalidity argument").

ii) That UBS have not "reasonably determined" their loss because "as a matter of principle" (or ex hypothesi) a determination that omits a Terminated Transaction has not been made "reasonably" (the "necessarily unreasonable argument").

iii) That, even if this is not so, on the facts of this case UBS acted unreasonably in omitting the $65,000,000 when they determined their loss (the "de facto unreasonable argument").

11. Mr Matthews submitted that KSF are entitled to advance the first two arguments on their unamended pleading and that only their application to amend is required only for the third argument. I cannot accept that submission: the case that was pleaded was a simple debt claim, and did not recognise that the debt is no longer payable under section 2 of the Master Agreement. That was what was pleaded in paragraph 7 of the Particulars of Claim. The natural reading of paragraph 8, to my mind, is that it simply pleaded the claim for interest. The stark fact is, I think, that before the trial KSF did not focus on relevant provisions of the Master Agreement, although they were pleaded in the defence, admitted in the reply and reflected in the list of issues. Hence the importance of the application to amend and of the question whether I should allow the amendment, and if so whether I should allow KSF to advance all or only some of their three arguments. Having introduced the application for permission to amend, I shall come back to it at the end of my judgment.
12. Much of the important evidence is documentary, and the documents in the agreed files are admissible in evidence in accordance with the Civil Procedure Rules 32PD28.2. KSF served statements of two witnesses: Mr Patrick Brazzill, an executive director of Ernst and Young and one of KSF's administrators, and Mr James Youngs, who was employed by KSF and is now seconded to them to assist in the administration: Mr Youngs' statement was served in response to that of UBS' only witness, Ms Annette Alford, who is a Global Risk Specialist and works in (what was known at the relevant time as) their FICC Department.
13. Mr Brazzill's statement was not satisfactory, not least because it contained a great deal of argument and contentious comment on documents (a common problem with statements, despite the important guidance in 32.4.5 of the White Book). It was not only unnecessarily long, but it presented UBS with an unfair dilemma about what should be challenged in cross-examination. I was not willing for him to give evidence in chief by way of confirming the original statement. Accordingly, those acting for KSF prepared an amended version of the statement, which removed a good part of the more offensive contents, and I allowed it to stand as Mr Brazzill's evidence in chief so as not to disrupt the trial further.
14. I did not consider Mr Brazzill a satisfactory witness: it became clear that he really knew nothing about some matters still described in his statement after it was supposedly revised to omit what was simply his comment. I give just two examples:

i) He said in his witness statement of an email sent by Ms Sau Ying Man to UBS dated 17 November 2008 enquiring about the "Non-Receipt of USD 65 Mil" (to which I refer later) that he "was not specifically aware of the details of any of these enquiries at the time". He accepted in cross-examination that at the time he had no knowledge at all of the difficulty about the payment of the $65 million, let alone any "details" of enquiries about it.

ii) He gave evidence in his witness statement about internal KSF book entries which showed a credit entry in UBS' account of $65 million and a debit entry for that amount in Khf's account, saying "Importantly, … these book entries were not reviewed or authorised by the Administrators staff". When he was cross-examined, he admitted that he did not know whether or not the entries had been authorised by the Administrators' staff, and he had not even enquired of a Mr Topping, whom he described as the person at Ernst & Young who would have authorised them if anyone had.

More generally, Mr Brazzill sought in his witness statement to distance KSF from the knowledge and action of the staff who had been retained by the company when it went into administration. Thus he said, "KSF was now acting by the Administrators and we had not reached any conclusion on the point [whether Khf rather than UBS owed them the $65 million]. In so far as we, the Administrators, delegated tasks to our [Ernst & Young] colleagues, they had not reached any view either". Not only did it become apparent that Mr Brazzill had no proper basis for asserting what views had been reached by Ernst & Young "colleagues", but he accepted when cross-examined, and it was obviously the case, that retained KSF staff were acting on behalf of KSF and the Administrators after 8 October 2008. There was no real basis for distinguishing their position and that of Ernst & Young personnel: it is difficult to understand how Mr Brazzill can have thought otherwise. I do not regard Mr Brazzill as a deliberately untruthful witness and I am sure that he had no intention to do other than to assist the court, but his written evidence was not careful and his cross-examination showed the limitations of his knowledge. I cannot rely on either his oral or his written evidence on contentious points.

15. Nor was I impressed by Mr Youngs' evidence. Its purpose was to explain the roles of a Mr Jonathan Rourke and a Mr Edward Littlewood, with whom Ms Alford had dealings. In his original witness statement (until it was corrected by a supplementary statement served during the trial) Mr Youngs said that Mr Rourke was not employed by KSF but by Kaupthing Singer & Friedlander Capital Markets Limited ("KSFCM"). When Mr Matthews opened the case, it seemed to me surprising that there should be an issue about who employed Mr Rourke that required oral evidence and asked whether his contract of employment had been disclosed. It had not been, and when it was examined overnight it was found to be with KSF: Mr Youngs explained that he had looked at the contract when he made his witness statement but not noticed who the employer was. This is surprising since his statement covered that very point. It also became apparent during Mr Youngs' cross-examination that, although he gave evidence that "it was very unlikely that Mr Littlewood would have discussed the KSF trades" with Ms Alford as she had stated in her witness statement, he had not asked Mr Littlewood about this: he did not suggest that he did not know where Mr Littlewood was to be found or could not ask him for some other reason. Again, I do not consider that Mr Youngs was dishonest in his evidence, but I cannot regard it as reliable. In the end, however, I do not regard his evidence as important in any event.
16. Ms Alford was a more impressive witness. She was properly pressed in cross-examination because, while the documents do not to my mind directly contradict her evidence, they do not reflect the role that she claimed to have in closing out the Master Agreement or the account that she gave of how it was carried out. She answered the questions clearly and cogently, and was careful not to claim involvement where she had none. I therefore regard her generally as a reliable witness and accept most of her evidence, but, as I shall explain, I am unable to accept it all.

The facts

17. UBS are a Swiss bank and provider of financial services, and are based in Zurich and London. KSF (formerly called Singer & Friedlander) were a bank, and in 2005 they became a subsidiary of an Icelandic parent, Kaupthing Bank hf ("Khf"). On 8 October 2008, KSF entered administration by order of Floyd J in the Chancery Division of this court made on the application of the Financial Services Authority ("FSA"). On 9 October 2008, the Fjámálaeftirlitiö ("FME"), the Financial Supervisory Authority of Iceland, where Khf were incorporated, acted under powers conferred by emergency Icelandic legislation passed by the Congress on 6 October 2008, and took over the powers of Khf's shareholders meeting, dismissed the board of directors and appointed a Resolution Committee whose role was to prepare Khf for reorganisation or winding-up and whose authorisation was required for all payments in foreign currency. On 24 November 2008 a moratorium was granted to Khf by the Reykjavik District Court, and it is now being wound-up in Iceland. The window under Icelandic bankruptcy law for making claims in the winding-up of Khf closed at midnight on 30 December 2009.
18. There is no dispute that on or about 1 September 2008 KSF and UBS entered into the FX trade for settlement on 3 October 2008, and under it (i) UBS were to pay KSF US$100 million and (ii) KSF were to pay UBS £55,614,723.44. There is not known which individuals made the FX trade or whether it was made from UBS's London or their Zurich office. But UBS's copies of their Society for Worldwide Interbank Financial Telecommunication ("SWIFT") confirmation of the FX trade to KSF and of KSF's SWIFT confirmation to them are in evidence: they had not been disclosed during the trial, but the terms of their outwards confirmation were set out in the email sent by Ms Man on 17 November 2008 to which I have referred. After the trial, following what was described as "further targeted searches", the confirmations were found and disclosed, and the parties agreed that they be put in evidence. They give the code for KSF's account with JPMC at field 57A, which is the field for stating the account at an institution to which the payer or Operating Party requests the Beneficiary be paid. KSF's copies of these documents were not found by the administrators and have not been disclosed.
19. There is further documentary evidence of the FX trade by way of KSF's trade ticket, an internal document headed "Expanded Deal Details". It was not explained in the evidence (for example, it is obscure why it is described as "expanded"), but it records:

"Our Receiving Agent

JPMorgan Chase NY".

20. On 24 September 2008 KSF and UBS entered into a further foreign exchange transaction for settlement on 3 October 2008, under which KSF were to pay UBS the sum of Australian $44.4 million, then the equivalent of about US$35 million.
21. By the end of September 2008 there were concerns in the market about the financial stability of Icelandic banks and their subsidiaries, including Khf and KSF, and UBS were trying to reduce their exposure to them. An agreement was reached between UBS and KSF that UBS should withhold $35 million of the $100 million that they were to pay under the FX trade pending receipt from KSF of the Australian dollar payment. (Again, the details about how this agreement was reached are obscure. UBS's pleading raises an issue about whether they were entitled to withhold payment, but it has not been pursued and it is not important for present purposes.) An internal KSF e-mail of Friday 3 October 2008 sent by Mr Stephen Ong, the Head of Trading and Balance Sheet Management, reads as follows:

"I have agreed with UBS that the AUD 54m short in our UBS nostro on the back o[f] an FX payment will be covered until Tuesday by them withholding $35m of a payment they are due to send us."

Apparently the agreement was until Tuesday 7 October 2008 because 6 October was the Australian Labour Day bank holiday. The $35 million was paid by UBS to KSF on 7 or 8 October 2008.

22. UBS tried to pay the $65 million to KSF by a SWIFT transfer to JPMC on 3 October 2008. They made a mistake in entering the details of the account that JPMC should credit, and as a result JPMC were instructed to credit the funds to an account that Khf had with them, and JPMC did so. There is no dispute that this mistake was made: it is not strictly relevant to the present dispute how it occurred, but apparently it came about because, as a result of the arrangement about the $35 million, UBS's automated payment systems had to be overridden with manual instructions. The details are as follows:

i) At 3.16 pm (EDT) on 3 October 2008, Mr Andrew Newcomer of UBS's Stamford branch sent an e-mail to Ms Vesna Woehler of UBS's Zurich branch, and asked her to "please enter the manual payment of $65m" to "Kaupthing". The e-mail was sent under reference number "YEO158158605", a reference for the FX Trade.

ii) In seeking to carry out this request, the operator at UBS's Zurich branch mistakenly entered into UBS's manual payments system an instruction that the payment be made to Khf, rather than to KSF. This (erroneous) manual payment instruction was provided by UBS's Zurich branch to their Stamford branch, and the $65 million payment was accordingly made to Khf's US dollar account with JPMC at about 4pm (EDT) on 3 October 2008.

iii) The SWIFT message from JPMS to Khf in respect of the payment included this note: "REMAINDER OF PAYMENT WILL BE PAID ONCE 44.4 M AUD RECEIVED".

23. KSF and Khf both learned of the error on 3 October 2008. They did not advise UBS about it, but KSF set about correcting the position by arranging for the amount of the Payment, $65 million, to be transferred from Khf's account with JPMC to their own. On 3 October 2008 Mr Carrigan, who was "Manager–Operations" at KSF and, in the words of Mr Brazzill, "deputy head of KSF's back office", sent this internal email to others at KSF:

"Just to confirm UBS have paid the USD 65m to Iceland's account [sc Khf's account] at JPMChase, this has result in a payment for 51m not being sent to Deutsche Bank.

We spoke with the Credit Officer at Chase and they would not accept anything from Gudni [who worked in KHF's Treasury] except a swift instruction requesting the transfer. Gudni has no one in the office with swift access therefore on Monday we will backvalue the account Transfer from HF to our account….".

There is no direct evidence about what was said when KSF spoke with the JPMC Credit Officer, but I infer from the email that before it was sent KSF had informed JPMC of the error by telephone. Mr Matthews argued that the email shows only that they had asked that the funds be transferred from Khf's account to KSF, but I find it impossible to think that KSF simply asked that $65 million be transferred to them from the account of another company without explaining the problem that had arisen and the basis on which they made the request.

24. On Monday 6 October 2008, Ms Lucy Crowther, described as "Assistant Manager Settlements Team" at KSF, e-mailed Mr Einar Hjartarson of Khf, saying:

"Please note UBS paid $65m into your account on Friday. Please can you pay to our account as detailed below.

Agent – JP Morgan Chase, New York

SWIFT – CHASU33

A/C – 001 1 949245

Please ensure this is paid today and we will ensure re valuation at a later date."

Later on 6 October 2008 Mr Carrigan chased Mr Hjartarson by e-mail:

"Can you please confirm that the USD 65,000,000 which UBS paid to you in error on Friday has been release[d] to our account. Can you please send me a copy of the instruction which you have sent."

Mr Hjartarson replied:

"I can confirm that we have not send [sic] the instructions to move this amount to your account. Our Treasury don't allow that. They have told us that Hreidar Már is the only one that can tell us to move the funds…

We have send [sic] an enquiry to the sender. I hope we will get an answer tomorrow."

Mr Carrigan then sent this email to Mr Hjartarson and Mr Pablo Vergara, apparently of KSF's treasury operations:

"Have you agreed the below with Iceland? If not can you please escalate this for me, these funds were due to KSF and UBS have paid HF on Friday. We currently have a payment to make to HF of USD 21,951,000. Please advise."

Mr Hjartarson immediately responded: "Good, best to let the Treasury guys talk to each other".

25. As is reflected in Mr Carrigan's last email, on 6 October 2008 KSF were due to pay Khf $21.9 million. KSF did not pay this sum, and on 7 October 2008 Khf sent an email asking KSF to investigate and make the payment. It was forwarded to the KSF's "DL Cash Management" and "DL Treasury Middle Office" e-mail addresses, and Mr Richard Dwyer, who apparently worked in KSF's cash management function, replied on 7 October 2008, saying "They [sc Khf] didn't pay us $65m". Therefore Mr Martin Brown of KSF replied to Khf with an email headed, "KSF netting for value 6/10", which said:

"Having spoken to Cash Management there was a payment made by UBS to your account for $65,000,000 in error and it was agreed that you would return the funds yesterday. Can you please arrange to net the two payments together and pay us the difference of $43,049,000."

On 8 January 2008 KSF and Khf had entered into a so-called "netting agreement" in respect of sums in the same currency that were payable between them. In their defence UBS pleaded that under the terms of it KSF effectively netted off $21.9 million against the $65 million, and so any claim against them is reduced accordingly. They did not pursue this argument at trial and accepted that the proposed netting-off was ineffective.

26. UBS were not involved in the exchanges of 3 and 6 October 2008, and KSF did not complain to them that they had failed to discharge their obligations under the FX trade. However, on 6 October 2008 from Mr Ólafur Gunnarsson of Khf sent this email to Mr Carrigan and others:

"The reason we haven't been able to send to KSF is that we need instructions directly from UBS saying it should have gone to you guys. As soon as they will inform us we will pay to you guys [i.e., to KSF]."

27. At 2.06 pm (EDT) on 6 October 2008, Ms Delia Howser of Khf had sent a SWIFT message to JPMC, which (so far as is material) said "PLEASE URGENTLY PROVIDE FURTHER DETAILS" in relation to the transfer made by UBS to Khf. The evidence does not explain what position Ms Howser held, and it is not clear whether, when she sent this message, she was aware of the exchanges on 3 October 2008: it seems likely that she did not and was simply puzzled about the receipt. UBS were contacted over the course of 7 and 8 October 2008. The communications can be summarised as follows:

i) At 9.24 am (EDT) am on 7 October 2008, JPMC sent a SWIFT message to UBS's Stamford office, apparently in order to respond to Ms Howser's request for "further details": "KAUPISRE [i.e., Khf] IS UNABLE TO APPLY FUNDS WITH INFORMATION GIVEN, PLEASE ADVISE US OF FURTHER PAYMENT DETAILS." They sent a copy to Khf.

ii) On 7 October 2008, Ms Howser chased JPMC by e-mail for a reply to her SWIFT message.

iii) The reply from JPMC was as follows:

"I have today sent you a SWIFT acknowledgement under our case reference JPM1070406OCT08 and have contacted UBS for additional details. I will revert on receipt of their reply."

28. As I have said, when she sent the first message to JPMC Ms Howser was probably unaware of the exchanges between KSF and Khf, but on 7 October 2008, Mr Arnar Elfarsson (of Khf) e-mailed Martin Brown (of KSF):

"Oli Frimann in Treasury and Delia [Howser] are working on getting a clarification from UBS that KSF was the beneficiary of the funds.

Please if you could in any way urge UBS to send JP Morgan (our USD agents) some information about final beneficiary, that would speed the process up."

There is no evidence that KSF did contact UBS as Khf asked, but, as far as Khf is concerned, I infer that, even if (as I have found) Ms Howser did not know of KSF's interest when she first contacted JPMC about the unexplained receipt, she was by now aware of it. Ms Howser sent an e-mail to Mr Carrigan and Mr Hjartarson that she had "escalated the query with JP Morgan and [was] still waiting for their reply!" The inference of this is that, although Khf had asked for KSF's support in obtaining the information from UBS that was required for the transfer of the $65 million to KSF, the understanding between them was that it was primarily Khf who were to go about obtaining it.

29. UBS Stamford had conveyed JPMC's inquiry to UBS's Zurich office: "Please be informed that the beneficiary is unable to apply the above mentioned payment WITH THE DETAILS GIVEN. Please provide further details such as references, invoice number and/ or full address of the remitter". There followed exchanges between UBS Zurich and UBS Stamford, the details of which are not important. On 8 October 2008 UBS realised the mistake that they had made when they transmitted the money: in an exchange at 8.14am (EDT) on their internal instant messaging service between Mr Andrea-Cosimo De Blasi ("deblasan") of UBS Zurich and Ms Yolanda Acosta, who was apparently at UBS Stamford, Mr De Blasi wrote ""We paid the wrong [company]", and Ms Acosta responded "JPMorgan cannot apply". At 08.41 am (EDT) UBS Zurich sent a SWIFT message to UBS (Stamford):

"PLEASE AMEND OUR INSTRUCTIONS AS FOLLOWS:

PLEASE PAY FUNDS TO SIFRGB2L ALL OTHER DETAILS REMAINES [sic] UNCHANGED."

"SIFRGB2L" is the bank identifier for KSF. UBS Stamford sent the message on to JPMC by a SWIFT message shortly after 9.00 am (EDT) on 8 October 2008, and followed this with another message in which JPMC were asked to confirm their action.

30. JPMC then sent a SWIFT message to UBS at 9.26am (EDT) saying "WE ARE PRESENTLY CONTACTING KAUPTHING BANK FOR DEBIT AUTHORITY PER YOUR REQUEST TO AMEND YOUR PAYMENT ORDER. WE WILL ADVISE YOU FURTHER UPON REPLY." At the same time they also sent a SWIFT message to Khf, saying:

"PLEASE AUTHORIZE US TO DEBIT YOUR ACCOUNT WITH VALUE VIA AUTHENTICATED MESSAGE AS REMITTER RQSTS RTN [requests return] AS FUNDS INTENDED FOR DIFFERENT BANK.

CANCELLATION OF THE SUBJECT PAYMENT ORDER AND RETURN OF AMOUNT THEREOF IS SOUGHT AT THE REQUEST OF OUR CUSTOMER ... PLEASE CONSIDER THIS OUR INITIAL REQUEST FOR UNJUST ENRICHMENT."

When they said that the funds were intended for a different bank, they must, of course, have meant that they were intended for the account that another bank held with them: JPMC were always the intended recipient. It was submitted that they requested authority to debit Khf's account in order to return the money to UBS rather than to pay it to KSF, but I do not so read the message: they simply asked that Khf release the money to them, saying nothing of what they were proposing to do with it. At 6.17 pm on 8 October 2008, Khf's Ms Howser replied by SWIFT to JPMC:

"WE HEREBY AUTHORIZE YOU TO DEBIT OUR ACCOUNT FOR USD 65,000,000.00 THAT WAS CREDITED TO OUR ACCOUNT IN ERROR."

31. At 6.30am (EDT) on 9 October 2008, JPMC sent SWIFT messages to UBS and to Khf. Their message to Khf said: "JP DEBITING YOUR ACCOUNT … TODAY 65,000,000.00/ USD VALUE 08-OCT-08 IN SAME DAY FUNDS IN REVERSAL OF OUR CREDIT ENTRY DATED 03-OCT-08 … AS THE REMITTER REQUESTS CANCELLATION." Their message to UBS said "WE CONFIRM AMENDING THE ABOVE PAYMENT TODAY IN ACCORDANCE WITH YOUR INSTRUCTIONS." At 8.06am (EDT) UBS Stamford sent a SWIFT message to UBS Zurich: "MSG FROM BNF BK [sc. JPMC] QUOTE: WE CONFIRM AMENDING THE ABOVE PAYMENT TODAY IN ACCORDANCE WITH YOUR INSTRUCTIONS. UNQUOTE. WE CLOSE OUR FILE. WE TRUST OUR ABOVE ACTION IS IN ORDER".
32. On 7 October 2008 Mr De Blasi had received an email headed "large open USD trades", and he agreed to check the position with UBS Stamford about their account not being debited. As Mr Matthews recognised, that might be a reference to the payment of $35 million: although it is common ground on the pleadings that the payment was made at some time on 7 October 2008, the SWIFT messages indicate that it was made on 8 October 2008. There is no evidence that persuades me that UBS were aware of the mistake before 8 October 2008, and I conclude that they were not.
33. In the event JPMC never did credit KSF's account, and KSF have not received the $65 million or any part of it, but neither JPMC nor KSF nor Khf told UBS this until 2012. On 15 October 2008, a Ms Bryony Cullen of JPMC's "Nordic & Baltic Team" sought advice about the payment in an internal email:

"KAUPTHING BANK Iceland received a credit on their books for 65,000,000.00/USD … which they could not apply. Remitter advised us to amend to KAUPTHING SINGER & FRIEDLANDER LTD, London's account as funds meant for them.

Amendment has been processed and verified but held in status 45.

Please urgently advise what further action needs to be taken to clear the item."

I cannot tell from the evidence what "status 45" means. No reply to the enquiry in evidence, nor is it clear what decision was reached within JPMC about how they should deal with these funds.

34. After the $65 million had been paid into Khf's account on 3 October 2008 the balance in their account at the end of the day was $77.5 million: the evidence does not show whether or not during the course of the day the balance in the account fell below $65 million after the UBS payment had been credited to it. On 7 October 2008 the payments from Khf's account (which were more than $145 million in total) included $55 million by way of two payments to KSF. At the end of 7 October 2008 the balance in the account was $7.8 million. On 14 October 2008, the balance standing to the credit of Khf's account increased to just over $70 million: this was the first time it had risen above $65 million since 7 October 2008.
35. At 14.49 (BST) on 8 October 2008 the Chancery Division of this Court appointed four accountants from Ernst & Young (including Mr Brazzill) as joint administrators of KSF, and ordered that their affairs be managed by the administrators. The evidence does not show when UBS learned of this, but at 20.58 on 8 October 2008 internal instructions were given to "halt all business" with KSF and three other companies in the group (but not including Khf). As I have said, on 9 October 2008 UBS gave KSF notice of an Event of Default under the Master Agreement.
36. Also on 9 October 2008 the FME intervened in Khf: by SWIFT messages sent at 7.22 am (EDT), Khf wrote to UBS Zurich and JPMC, "By order of the Icelandic Financial Services Authority (FSA) we have been ordered to transfer all credit balance on our account held with you. Can you please put all pending payments out of our account on hold and send all balance to beneficiary, SISLISRE …", SISLISRE being the code for the Icelandic Central Bank.
37. The administrators of KSF faced the immediate task of managing the insolvency of a retail and commercial bank in the context of an acute liquidity crisis in the Kaupthing group, the Icelandic banking system and indeed the international financial markets generally. Mr Brazzill's evidence emphasised the scale of the work involved in the administration. The administrators' problems were aggravated because KSF's internal systems and record-keeping were incomplete, and payments could not be reconciled to underlying trades. Apparently KSF had conducted at least inter-bank trading without trade confirmations or follow-up documentation, and sometimes they did not record traders' telephone conversations. Their internal "ledger" of debts and receivables was on a system called "IBIS" that, as Mr Brazzill put it, had "limited functionality". Understandably none of the administrators became aware of the FX trade or the problems with the payment of the US$65 million in the early days after their appointment.
38. KSF had dispensed with their internal legal department, and the administrators instructed Freshfields Bruckhaus Deringer ("Freshfields") to assist them. KSF staff, together with Ernst & Young and Freshfields, sought to reconcile payments and receivables to trades. Mr Brazzill confirmed that the staff employed by the company in administration were authorised to continue with their duties, subject to "certain limitations in terms of what they could commit the company to post administration", and in particular it was within Mr Carrigan's authority to investigate the FX trade. (Mr Brazzill specifically confirmed this in relation to the position in 2009, but I infer that the position was the same in 2008 after the administrators were appointed.)
39. Mr Brazzill was not concerned with this work except at the most general level, and KSF did not adduce evidence from anyone directly involved in dealing with the FX trade and the payment of $65 million. However, the documents show that in November 2008 KSF had difficulty in understanding the payment of $35 million. On 4 November 2008 JPMC sent a SWIFT message to UBS Stamford that KSF could not "apply" the payment and requested "more details including ult[imate] beneficiary". UBS Stamford sent the request on to UBS Zurich, but they apparently received no response despite making further requests on 6, 12 and 17 November 2008. On 20 November 2008 UBS Stamford responded to JPMC (who had been chasing them for the information) that they had contacted UBS Zurich "several times without receiving any further details", and suggested that JPMC either contact them direct or return the funds to them. They concluded, "We consider the matter closed". There is no evidence that JPMC did contact UBS, and they did not return the $35 million.
40. There is no explanation as to why UBS Zurich did not respond to these inquiries. (Mr Matthews referred in this context to an internal exchange between Mr De Blasi and Ms Acosta, but, as I understand it, this was on 10 October 2008 and it is in any case quite obscure what it was about. I see no reason to connect it with this matter.) It was suggested that, had they done so, further enquiries might have led UBS to realise that the $65 million had never been received by KSF. I accept that is possible, and I revert to the point later. I do not infer, if it be suggested, that UBS deliberately avoided replying because they knew or suspected that the $65 million had not been paid, or that they have avoided proper disclosure of documents about the inquiry about the $35 million.
41. Meanwhile enquiries were being made within KSF about FX trade: I infer that they were made because of the difficulties in reconciling the receipt of the $35 million and were pursued internally when they received no information from inquiries of JPMC. The email chain within KSF was given the subject line "Ian says this one is for you from 7/10/08": apparently Mr Carrigan ("Ian") had been asked about the position and referred the matter to persons who knew of his discussions when the payment was misdirected. In reply to an email on 10 November 2008 from Ms Sau Ying Man of KSF's Treasury, Mr David Haddow, of KSF's Cash Management department, wrote:

"This is to do with a cable FX for USD 100,000,000 traded with UBS.

UBS were due to pay us USD 100m on the 3rd, but did not pay in full due to us failing to them [sic] AUD44.4m so they took USD 35m off to cover their exposure for the AUD, and paid USD65m of the 3rd, but unfortunately this was paid to HF in error, which HF should have then paid this on to us, but we are yet to receive the USD65,000,000 from HF … The USD 100,000,000 is outstanding on the IBIS side, and the USD 35,000,000 is outstanding on the CHASE side, this cannot be matched due to HF failing to pay the USD65,000,000."

I therefore infer that at this time KSF's computer system showed UBS as still owing them the $100 million that had been due under the FX trade, and apparently this remained the position until 27 October 2009: see para 93 below. On 17 November 2008, Ms Man, following a phone call to UBS's FX Operations department, sent an e-mail to the UBS's FX Investigations mailbox in Zurich headed "Non-receipt of USD 65 Mil". After providing references for the trade and identifying it, she wrote:

"I understand that you were to pay us USD100M on 03/10/08. Since we (Kaupthing Singer and Friedlander) did not pay UBS AUD 44.4M, it was agreed that USD35M would be deducted and only USD65M would be paid to us.

However, we have not received USD 65M from UBS.

It is my understanding that you have paid these funds incorrectly to Kaupthing HF (Reykjavik).

Please urgently investigate and advise.

Please also provide a copy of the SWIFT message.

FYI, here is your SWIFT confirmation for the FX: …"

Then, as I have said, she set out the confirmation of the FX trade that UBS had sent to KSF.

42. There is no evidence whether, and if so how, UBS responded to this inquiry. Neither party has disclosed any further documents about a response or any follow-up to it. It appears that Ms Man and KSF did not pursue the matter further. Mr Brazzill's evidence was that, while he knew of no "specific decision to 'park' the issue of the $65 million", it would have been "consistent with the overall approach to the administration that if an enquiry were made of a counterparty which failed to generate any useful information in response then the relevant individual at KSF and [Ernst & Young] would have been likely to move on to other things". I can understand that general approach in a busy administration, but, as Mr Brazzill's evidence confirmed and would in any case be expected, in those circumstances some record would surely have been made that the matter was still outstanding. I conclude that it is likely that Ms Man or others at KSF satisfied themselves that there was nothing outstanding in respect of the FX trade. After all, Ms Man's enquiry indicates that she knew nothing of the arrangements between KSF and Khf for the funds to be transferred between them or of UBS's amended instructions to JPMC, and the most likely inference was that, when she learned something of this history, she was satisfied that the matter had been dealt with and UBS had discharged their obligation in respect of the $100 million by the payment of $35 million and the transfer of the $65 million and the amended instruction to JPMC. This seems to me more probable than the alternative: that, simply because there was no response to the email to UBS, they abandoned the enquiry without even making a note to return to it. The amount involved was too much for KSF simply to have forgotten about it without making any record. There is some corroboration of this conclusion because an internal email at KSF (from a Mr Kirpal Gillar of KSF's Finance Department to Mr Kenneth Marsterson) dated 19 January 2009 said that, "Ian [Carrigan] is investigating an issue with an UBS $100 m fx one sided cash flow that was due to us at the time of administration as we believe it was settled partly via HF …": apparently it was the view of Mr Carrigan and others that the Payment to Khf (or a payment via Khf) had partly settled the obligation to pay $100 million.
43. Ms Tolaney went further: she submitted, although recognising that there is no direct evidence of this, that "there must have been a follow-up call or communication" between UBS and KSF in which it was established that UBS had discharged their payment obligation under the FX trade. I cannot accept that submission: Ms Man was satisfied that UBS's obligation was discharged and the FX trade "closed", but this might have been either because of some contact with UBS or as a result of further internal enquiries or discussions. The evidence does not support an inference that it was the former, and if anything it indicates the latter.
44. The documents show that in June 2009 Mr Carrigan was in communication about the $65 million with Freshfields and with Ernst & Young, in particular Ms Patricia Tay, a Manager in their Banking and Capital Markets division. By an email dated 3 June 2009 he sent Freshfields a calculation that KSF owed Khf £15,904,813.78, and the calculation brought into account the $65 million, describing it as follows: "Amount HF held which was paid to them in error by UBS". On 13 July 2009 he forwarded a copy of the email to Ms Tay. She had been engaged in June 2009 in examining what sums were owing between KSF and Khf, and for this purpose she had the emails in October 2008 in which KSF had proposed to bring the $65 million into account under the netting agreement with Khf, and she had asked for and been provided with the "tickets" underlying the trades.
45. On 27 October 2009 Mr Gillar sent an email to Mr Carrigan asking him to "pass the following entries". The entries included, "Dr 38812706 $65,000,000 Failed settlement Due from UBS paid to HF [i.e., Khf]. Cr 38266705 $65,000,000 Failed settlement Due from UBS paid settled with Hf" and "Value 3/10/08". On 9 December 2009, Mr Gillar e-mailed Mr Carrigan asking him to "post" the final ISDA close-out journals attached to the e-mail. Mr Gillar wrote that "For the record, UBS settled their liability with KSFL so once posted all accounts in [the relevant client codes] will be zero".
46. On 6 November 2008 KSF had served on Khf a notice under an ISDA Master Agreement (the agreement is not in evidence but it is clear that, so far as is relevant for present purposes, it was in standard form), and on 26 November 2009 KSF wrote to Khf with their calculation of the account between them: they calculated that £74,533,104 was due from KSF to Khf, and this calculation was made on the basis that the $65 million was due from Khf to KSF. They explained in their letter that they were using the "Loss" method because "it is not reasonably practicable to determine the amount payable in accordance with the Market Quotation", and therefore they were obliged to determine the balance of their account with Khf "reasonably" and "in good faith". On 20 November 2010, KSF wrote a further letter to Khf, again saying that the amount of £74,509,581 was due on close-out. (The change from £74,533,104 is not important.) Both letters included disclaimers in the terms of the letter of 7 April 2009 to UBS to which I shall refer shortly: see para 52.
47. On 8 December 2009, KSF submitted to the winding-up committee of Khf a claim for £753 million, which is presented as "a formal claim as a creditor". It was calculated on the basis that KSF owed Khf £74,509,581 in respect of foreign exchange transactions (which they were entitled to set off against other sums in accordance with the netting agreement), and so on the basis that KSF were entitled to claim the $65 million from Khf. The £74,509,851 was said to be payable by KSF to Khf "in accordance with the terms of the ISDA Master Agreement".
48. KSF's claim was resolved by a Settlement Agreement dated 3 May 2012, which defined the term "UBS Return Claim" as being:

"the Claim which KSF has against [Khf] in the amount of GBP 36,790,047 under claim number: 20120423-0001 for or relating to the amount of USD 65,000,000 … which was paid by UBS AG … to Kaupthing on or around 03 October 2008 and should have been paid for the account of KSF in respect of the UBS-KSF FX Trade."

Clause 5.1 provided that, "The Parties agree to exercise reasonable endeavours to negotiate and agree a settlement of the … UBS Return Claim which KSF may have against [Khf].".

49. I must go back in time to refer to the evidence about the close-out process under the Master Agreement. Ms Alford's evidence was that, as well as dealing with the FICC department's claims in the close-out processes between UBS and KSF and between UBS and Khf, she was also responsible at the relevant time for overseeing for UBS generally claims and close-out processes when UBS's counterparties under ISDA agreements defaulted on obligations under ISDA Master Agreements. She had assumed this responsibility in 2007 and has been responsible for some fifty processes, most of them being after the global financial problems of around September 2008. Her evidence was that her first communication with KSF after the termination of the Master Agreement was in November 2008. KSF approached UBS for payment of £579,000 from an account with a credit balance of some £650,000. On 24 November 2008 Ms Anastasia Aldred, a client account manager at UBS, sent an internal email asking that someone from UBS's credit department call KSF to discuss the request, and provided the name and telephone number of a Mr Jonathan Rourke. Ms Alford responded and gave instructions that funds should not be released to KSF because KSF owed UBS a similar sum: as Ms Alford explained and is reflected in the internal documents, the overall position between UBS and KSF was the more difficult to resolve because UBS was concerned that it might involve valuing the Icelandic krona. Ms Alford said that she spoke to Mr Rourke on the telephone, and he did not, as far as she recalled, mention a possible claim against UBS for the $65 million. I accept her evidence: the fact that she spoke to Mr Rourke is apparent from internal emails before ("I have put in a call already to say ... that we have a right of set off…") and after ("he was very reasonable and understood …") their conversation. I accept her evidence that, had any mention been made of a claim for $65 million, she would have remembered it and would have reported it internally. Mr Youngs' evidence was that, although Mr Rourke was employed by KSF, he was not involved with their business generally but with Kaupthing Singer Freeland Capital Markets Limited ("KSFCM"), which KSF had agreed to sell in a management buyout on 31 October 2008. I accept that evidence, and this might well explain why Mr Rourke approached UBS: an email internal to UBS that referred to his first approach said that "all activity across the account originated in their proprietary Capital Markets Desk". Nothing in Mr Youngs' evidence, however, casts any doubt on Ms Alford's account of her conversation with him.
50. In early 2009 UBS worked on the close-out calculation that was required by the Master Agreement in order to terminate it. Mr Chris Barclay, who worked closely with Ms Alford but was junior to her, collated data from different departments, and calculated that overall UBS owed KSF $5,199,194.05 (the foreign exchange department owing KSF $5,956,099 and the equities department owing them $49,871.05, but the FICC department being owed $806,776). It is clear that nobody involved in preparing the calculation thought that UBS might have an outstanding obligation in respect of the $65 million.
51. On 20 March 2009 UBS sent their calculation to KSF. On 7 April 2009 Mr Brazzill replied, asking that UBS pay them this sum "being the amount determined by UBS as the termination amount under Section 6 of the Master Agreement" and giving particulars of the account to which payment should be made. His letter continued:

"Please note that receipt and/or retention of any payment by KSF shall not constitute an acceptance by KSF or its Joint Administrators of the basis of computation on which such payment was based and accordingly the close-out statement shall remain subject to review and verification by KSF and the Joint Administrators. …

Nothing in this letter shall be deemed to constitute a waiver and KSF and its affiliates hereby reserves all other rights and remedies that it may have against you or your affiliates under any applicable law. In addition, any acceptance by KSF or its affiliates of performance from, or performance by KSF or its affiliates to you or your affiliates under any agreement between KSF or its affiliates and your or any of your affiliates or otherwise (including without limitation the rollover of, or entry into, any transactions under, or amendments, supplements or modifications to, any agreement or otherwise), or any delay in exercising any remedies KSF or its affiliates may have, shall not constitute a waiver or forbearance of any rights or remedies that those parties may have.

…

You are further advised that no oral communication from or on behalf of KSF or its affiliates by any party shall constitute any agreement, commitment, or evidence of any assurance or intention of KSF or its affiliates with respect to the subject matter hereof. Any agreement, commitment, assurance, or intention of KSF or its affiliates shall be effective only if in writing and duly executed on behalf of KSF or such affiliate".

Ms Alford did not see this letter at the time. She said that, when she saw it after KSF had made their claim, her "jaw dropped" because normal procedure, in her experience, is for the parties to review and agree a close-out payment before it is made.

52. On 7 April 2009 Freshfields also sent an email to Mr Barclay on behalf of KSF asking for details of the trades that comprised the valuation. On 21 April 2009 Mr Barclay asked Ms Alford to let him have information about the FICC trades to enable him to respond, and she supplied it. She was not asked for information about other trades, including foreign exchange transactions. On 16 May 2009 Freshfields wrote that they thought that "KSF's traders now have enough information, and the final valuations are just being reviewed by Ernst & Young". They also pressed for payment of the $5.2 million (without awaiting the outcome of the review): they attached to their email a copy of the letter of 7 April 2009, describing it as "the settlement instructions". On 21 May 2009 Mr Barclay asked whether the fact that Ernst & Young were pressing for payment meant that they had completed their review of the valuations and confirmed them, and in reply that afternoon Freshfields said that they understood that the review was not completed and the valuations had not "formally been confirmed and accepted" by Ernst & Young, but that the Administrators wished to pay an interim dividend. Meanwhile, Mr Barclay approached UBS's foreign exchange department to suggest that, since the greater part of the payment from UBS related to the business of that department, they make the payment. Mr Tom Mulroy, the business manager for the foreign exchange department, asked Mr Markus Schaufelburger, of business unit control in the foreign exchange department, to "arrange transfers and payment as recommended by [Mr Barclay]". Ms Alford said that nevertheless UBS would not have made a payment unless they believed that the close-out calculation was agreed, but that was not reflected in the responses that UBS sent to Freshfields, or in their internal emails. Ms Alford was not herself party to these exchanges with Freshfields, but some internal emails were copied to her.
53. On 22 June 2009 Mr Mulroy sent an email to Freshfields: he described himself as "trying to coordinate payment from UBS to KSF" and asked for another copy of KSF's "payment instructions". Freshfields responded by sending a copy of the letter of 7 April 2009. On 25 June 2009 Mr Mulroy sent an internal email stating that, "FX Operations is poised to make payment …", and that he wanted to check the procedure for making it. It was suggested to Ms Alford that the email was sent to her only because she was to give FICC's approval for the payment, and that she had no wider role. She responded that the approval of the legal department was required for the payment and that they would not have been given it without her agreement. In fact, however, the legal department gave their approval for the payment in response to Mr Mulroy's email in unqualified terms: "Approved for legal". It appears from UBS's internal emails that the only objection to the payment being made was raised by Mr Sam Azizo of the Credit Department, who asked whether the Finance Department had confirmed that there were no other claims that might be set off against the payment. However, Mr Mulroy also assumed, as he put it in an email to Mr Barclay, that "senior biz signoff is required" before payment was actually made, and Mr Barclay confirmed that he thought that it would be.
54. Ms Alford's evidence was that in the meanwhile she had spoken to Mr Littlewood: she was unable to remember quite when she did so except that they had conversations "in the summer of 2009". Although Ms Alford could not remember the dates more precisely, they had certainly spoken before 8 July 2009 because reference is made to them speaking in email exchanges between Mr Littlewood and Mr Dominique Thill, an Executive Director in UBS's Fixed Income sales for the Nordic Region. Ms Alford recalled, according to her witness statement, that she discussed with him "the valuation of KSF trades", as well as the valuation of Khf trades. In cross-examination she explained that Mr Littlewood had contacted her and that she had "a number of telephone conversations with him". She said that she thought that he introduced himself "as a KSF person", and that they spoke about KSF before they discussed Khf's business.
55. Mr Youngs' evidence was that Mr Littlewood, though employed by KSF, "only ever worked on Khf business and the costs of his employment were recharged to Khf": that Mr Littlewood and others engaged in winding down Khf's assets had no authority "to deal with KSF matters". The chain of email exchanges between Mr Littlewood and Mr Thill ran from 8 July 2009 to 27 July 2009, and Ms Alford was copied into some of them: it ended when Mr Thill wrote to Ms Alford on 3 August 2009 that Mr Littlewood was "still waiting for something". They were headed by reference to the close-out process between UBS and Khf, and they referred only to that process and not to the close-out between UBS and KSF.
56. By late August or early September 2009, Ms Alford understood UBS's calculation to be agreed by KSF. Her evidence was that she had a "final discussion with Mr Littlewood in late August or early September 2009 in which he agreed that I had answered all his valuation queries, and that UBS's calculations did not need to be amended. My recollection is that at the end of that call, I understood that all queries relating to the valuations of the Khf and KSF trades had been resolved". Ms Alford went on to say that she would only have authorised payment to KSF "on the basis that the KSF close-out calculations were agreed".
57. On 21 September 2009 Mr Barclay sent an internal email to Ms Alford, Mr Mulroy, Mr Azizo and others in which he said that UBS were being "chased" by KSF, recording the "sign-offs" that had been received and observing that the approval of the Finance Department was also to be sought. On 22 September 2009 Ms Alford replied for the FICC department that it was "OK to pay". Mr Barclay sent out another email on 22 September 2009: he wished to confirm that "nothing had changed ... (i.e. other debts to us have come to light etc)." There were further emails reflecting UBS's concern that the bank had no other amount to the brought into account before payment was made, and the final approval of the bank's "EC" committee (a senior business approval committee: Ms Alford could not recall what the initials stood for) was obtained for the payment.
58. On 28 September 2009 Freshfields again pressed UBS to make the payment. Mr Mulroy replied that he thought that the matter had been forgotten, but that cannot be so. A meeting had been arranged for 30 September to "review the balances with Kaupthing". It appears from the documents that the final approval for the settlement was given by UBS's Credit Control Department on 2 October 2009. On 6 October 2009 Mr Barclay asked Freshfields to confirm that the payment instructions had not changed, and Freshfields confirmed the instructions, attaching to their email a further copy of the letter from KSF of 7 April 2009. Later that day UBS paid KSF the $5,119,194.05. (I should mention that KSF's pleading refers to an internal email that Ms Alford sent on 28 September 2009, in which she said that the "actual amounts of the claims have not been agreed with the receivers yet", but that referred to the position between UBS and Khf, and KSF did not rely on it at the trial.)
59. Thus, Ms Alford's evidence and UBS's case was that she had responsibility for overseeing the close-out process; that no payment could be made under it without her approval; that she would not have given her approval had she not understood that KSF had reviewed UBS's calculation and were in agreement with it; and that, before giving her approval, she had spoken to Mr Littlewood about the calculation, and he told her that she had answered his queries about valuations and it did not need to be amended. As I have said, the documents are not directly inconsistent with Ms Alford's evidence, but they do not corroborate it on any of these points. This is surprising, although perhaps less so than first appears: many of the documents are about Mr Barclay coordinating and collecting information from different departments and about the FX department making arrangements for the payment once it was apparent that the sum due to KSF arose almost entirely from FX trades. After this work had been done, there still had to be the "senior biz sign-off". I would certainly have expected that Ms Alford's account would have been reflected at least in UBS's internal documents, if not also in the exchanges in particular with Freshfields. However, as I have said, I consider Ms Alford an honest witness and accept the general thrust of her evidence. My conclusions are these:

i) Ms Alford did oversee the close-out process: she could not have been mistaken in her evidence that this was her responsibility, and I cannot accept that she lied about that. She did not need to be involved with the details of the process, except for the exact balance for FICC, her own department. The close-out process with KSF was, as she put it, "a very simple case", and the routine work was done initially by Mr Barclay, who was junior to her, and then by Mr Mulroy of the FX department.

ii) Whether or not she actually needed in any formal sense to "sign-off" the payment (other than for the FICC department), she could have prevented it from being made and would have done so had she not been satisfied that it should be made.

iii) Ms Alford understood that KSF had reviewed UBS's calculation and were content with it. This was mainly because she knew that in May 2009 Freshfields had said that they thought that KSF had enough information about the calculation and that Ernst and Young were already reviewing them.

iv) Ms Alford and Mr Littlewood did have telephone conversations in the summer of 2009. Their focus was on the Khf valuations and the close-out process between UBS and Khf. However, Ms Alford understood that Mr Littlewood was employed by KSF (as he was) and was not made aware that he was not involved with the KSF valuations and close-out. I accept her evidence that some reference was made at the start of their conversations to the KSF valuations to the effect that they had raised no difficulties, but they were probably really little more than a pleasantry before the conversation moved on to the more difficult Khf calculations. I also accept Ms Alford's evidence that a similar reference to the KSF close-out was made in a later conversation in late August or early September 2009, but consider that it was probably in similar terms. I conclude, however, that Ms Alford must (unintentionally) be overstating the position when she spoke in her evidence of a "discussion" of the valuation of the KSF trades, and I cannot accept that Mr Littlewood would have said anything to indicate that he was giving any formal agreement or assurance that KSF accepted the calculation.

v) Had Ms Alford understood that the calculation had not been accepted by KSF, she would not have allowed the payment to be made.

vi) The delay in UBS paying KSF was not because they were waiting for assurance that the calculation had been agreed. It was apparently attributable to UBS being anxious to check that they had not overlooked anything payable to them by KSF which could be set-off against what was due to KSF and also to some extent because Mr Mulroy was unfamiliar with the procedure for a close-out of this kind.

60. As I have indicated, in 2009 and 2010 UBS were also engaged in closing-out their ISDA Master Agreement with Khf. Ms Alford was also responsible for that process. It was more complicated and took longer than with KSF, but eventually UBS concluded a global settlement agreement with Khf dated 20 December 2010. UBS's associated company, UBS Limited was also party to it. It covered both dealings under the Master Agreement and other dealings between those parties. It was signed on behalf of UBS and UBS Limited by Mr Lev Babiev, a director of both companies, but it was Ms Alford's evidence, and I accept, that this does not mean that he took the decision to enter into it on behalf of UBS. Under the agreement UBS agreed to pay Khf US$1 million and ISK 1.5 billion, and the parties gave each other various releases. As Ms Alford explained, at the time it was difficult to value Icelandic krona, but at current rates (ISK1=US$0.0088) UBS paid Khf the equivalent of something in the region of US$14.3 million. That agreement did not bring into account any claim of UBS in relation to the $65 million: they say, and I accept, that they thought that the problem had long been resolved. If UBS had thought that they had a claim against Khf for $65 million, they would not have made the close-out payment but, as Ms Alford said and was obviously the case, they would have sought to bring the claim into the account. If the claim was accepted by Khf, UBS would not have paid them anything: they could have set off some of what they owed against the $65 million, and made a claim in the liquidation of Khf in respect of the rest.
61. On 22 December 2011 Ms Alford was sent an email by Mr Haukar ?ór Hauksson of Khf, in which he said that Khf and KSF were seeking to reach a settlement which would cover the Payment of the $65 million. He proposed a "tri-party agreement" between Khf, KSF and UBS that no party would raise any claims relating to the Payment. It did not suggest that UBS might have any liability in relation to it. Ms Alford sought to investigate what the payment of $65 million might have been, but was unable to find anything about it in UBS's systems. In the event, the question of the tri-party settlement was not pursued, for reasons that are not explained by the evidence. Ms Alford referred the inquiry to UBS's legal department, and apparently further inquiries were made by Ernst and Young of that department.
62. KSF asserted a claim against UBS in respect of the US$65 million only by letter of 29 August 2012. On 2 October 2012, UBS lodged an (avowedly contingent) claim in the winding-up of Khf in order to protect their position. As I have said, the claims window in the winding-up had closed on 30 December 2009. On 23 October 2012, the winding-up committee of Khf declined to add UBS's claim to the List of Claims, inter alia on the basis that it had been lodged too late. UBS have argued in Icelandic proceedings that the claim can be made outside the claims window, but Khf maintain that it does not meet the requirement that late claims should be brought "without undue delay".

The discharge defences

63. I consider first whether UBS's payment obligation under the FX trade was discharged by the Payment of US$65 million to JPMC. "All questions relating to the payment of a sum of money in pursuance of a contract depend on the construction of the terms of the contract": Chitty on Contracts (31st Ed., 2012) vol 1, para 21-040. The Master Agreement provided (at section 2(a)(ii)) that "Payments under this Agreement will be made on the due date for value on that date in the place of the account specified in the relevant Confirmation or otherwise pursuant to this agreement, in freely transferable funds and in the manner customary for payments in the required currency". "Confirmation" is given a wide meaning in the Master Agreement ("the documents and other confirming evidence") and covers the SWIFT messages between UBS and KSF after the FX trade. Therefore UBS's payment obligation was to make payment to the place of the account at JPMC.
64. Ms Tolaney did not contend that payment was made under the contract when on 3 October 2008 UBS sent the instruction for the SWIFT transfer with the erroneous reference to Khf's account: on any view that was not a payment to KSF "in the manner customary for payments …". The first issue is whether payment was made when on 8 October 2008 UBS sent the SWIFT message amending their instructions. JPMC were made aware of the mistake that had been made on 3 October 2008. If the matter were governed by English law, their retention of the money might well give rise to a constructive trust: see Westdeutsche Bank v Islington LBC, [1996] AC 669, 715B per Lord Browne-Wilkinson. Mr Matthews submitted that there could not have been a constructive trust in this case because JPMC had notice of the mistake only from KSF (and not from UBS, the putative beneficiary, or from Khf) and therefore they could not be said to have had sufficient knowledge of UBS's interest. I cannot accept that submission: there is no suggestion that JPMC did not believe what KSF had told them, and notice from KSF would, in my judgment, have been sufficient. The question whether there was a trust might depend on whether, and if so in what circumstances, a SWIFT instruction can be revoked, about which there was no evidence. But in the end nothing, I think, turns on whether JPMC were subject to some trust of the kind that Lord Browne-Wilkinson contemplated, and I need not decide, and do not decide, whether they were.
65. I can consider how the Payment affected UBS's payment obligation on the basis that, after they had sent the amended instruction on 8 October 2008, UBS were in same position as if they had transferred the funds with correct instructions on 8 October 2008: in those circumstances, would they have made payment under and in accordance with section 2(a)(ii)? (KSF did not argue that because of the previous history by way of the initially mistaken instructions therefore payment could not be said to have been made "in the manner customary for payments…".) Mr Matthews submitted that there had been no payment to KSF, and UBS's payment obligation was not discharged, unless and until the money transferred by UBS was credited to KSF's account with JPMC, and it never was. In support of this submission he cited the Court of Appeal authority in Tenax SS Co Ltd v The Owners of the Ship "Brimnes", The "Brimnes", [1974] EWCA Civ 15 and the House of Lords authority of A/S Awilco of Oslo v Fulvia SpA di Navigazione of Cagliari, The "Chikuma", [1981] AC 314. Both those cases were about payment of hire under a charterparty on the New York Product Exchange form, clause 5 of which provides for "payment of hire … in cash". Hire had not been not paid literally in cash, and the courts had to decide how the contractual requirement should be interpreted in light of what Brandon J in The "Brimnes", [1973] 1 WLR 386, 400 called "modern commercial practice". In that case, the charterparty provided for payment "to Morgan Guaranty Trust Co of New York, 23 Wall Street, New York, for the credit of the account of [the owners] re m.v. 'Brimnes'", and in The "Chikuma" the charterparty provided for all freights to be paid to the owners' agents "care of Istituto Bancario San Paulo di Torino – Sede di Genova", which was the owners' bank. Interpreting those contracts, it was held in The "Brimnes" that it would have been good payment to make any transfer of funds to the stipulated bank for the credit of the owners' account so as to give them the "unconditional" right to the immediate use of the funds transferred, and in The "Chikuma" this view was endorsed with the clarification that an "unconditional" right to the immediate use of the funds connotes an "unfettered or unrestricted" right: see The "Chikuma" (loc cit) at p.319H per Lord Bridge. What the cases decide is this: "That when payment is made to a bank otherwise than in literally in cash, i.e. in dollar bills or other legal tender (which no one expects), there is no "payment in cash" within the meaning of clause 5 unless what the creditor receives is the equivalent of cash, or as good as cash": per Lord Bridge, loc cit at p.320B.
66. In the Master Agreement there was no requirement for payment in cash, and the question is what is required by the contract in order for payment to have been made to the stipulated bank, JPMC, in order for payment to have been made under section 2. Section 2 provided (inter alia) that payment under it by UBS to KSF should be made:

i) In "freely transferable funds";

ii) In "the manner customary for payments in the required currency"; and

iii) By payment in the place of the account at JPMC.

67. No point was taken against UBS about "transferring funds" by sending JPMC a SWIFT instruction: although there was no evidence about how payments in US dollars are (or in 2008 were) customarily made treated, the use of SWIFT was implicitly accepted as being contractually compliant. Payment was also made in the place of KSF's account with JPMC, and Mr Matthews did not suggest otherwise. The essential question, to my mind, therefore is whether payment was made "in freely transferable funds" before the account was credited with the payment. The expression is not defined in the Master Agreement (although apparently consideration has been given to clarifying what is meant: see the Commentary on the Form of Amendment to ISDA Master Agreements para 6). In my judgment, it would normally connote that the funds are to be freely transferable in the hands of the recipient and they would not normally be so until they have been credited to the account specified for receiving payment. This is, I think, consonant with the provision at section 2(b) that either party may "change its account for receiving a payment" by notice. It is also the view of Firth, Derivatives Law and Practice (loose-leaf, September 2013), who states that the requirement for payment to be in freely transferable funds "probably means no more than that the payee must have control over the money following payment": para 11.009 fn 6.
68. Is that the position in this case? As I have said, in their trade ticket KSF described JPMC as their "receiving agent". If that record reflects either an appointment to JPMC to a position of agency to receive payment or a term of the contract for the FX trade, it would be necessary to examine the authority of a "receiving agent" and the precise meaning of the expression in the context of the contract. Clearly JPMC were agents for some purpose: they would not have been entitled simply to refuse to accept money transmitted to KSF's account. That in itself might justify the description of them as agents "to receive and accept [proper] payment": see Mardorf Peach & Co Ltd v Attica Sea Carriers Corp of London, [1977] AC 850, 880 per Lord Salmon. However, although during the trial there appeared to be no question about whether the trade ticket accurately reflected the contract, in view of the (indirect) evidence of UBS's confirmation of the FX trade by way of Ms Man's email of 17 November 2008 to which I was referred later in the trial and UBS's copies of the confirmations that were disclosed after the trial, it seems to me unlikely that the contract did more than specify an account as contemplated in section 2(a)(ii), and there is no sufficient basis in the evidence that KSF gave JPMC any authority with regard to receiving the payment beyond that which they had as their bankers so specified.
69. What has troubled me more is that KSF plead in their Reply that JPMC were "specified as the receiving agent for KSF under the FX trade (so that receipt of payment by [JPMC] for the benefit of KSF would constitute valid receipt by KSF)": thus the pleading spells out the implications of JPMC being specified as "receiving agent". I think that the natural interpretation of that pleading is that, once the money was received by JPMC for KSF's benefit, the payment obligation was discharged (pro tanto). The agency role admitted in the pleading is not, as I see it, consistent with an argument that the payment obligation was not discharged unless and until there were funds that could be freely transferable by KSF, or, in the words of Firth, KSF had control over the money. In the end, however, I have decided that it would be unjust in this instance to hold KSF to this exact reading of their pleading. Had the case ultimately turned on it, I might have invited further submissions on the point, but I prefer to put my decision on other (some might say, less "technical") grounds.
70. I must therefore consider UBS's alternative submissions in support of the discharge defences;

i) That KSF treated Khf as their agent to receive the payment into their (Khf's) account, and so ratified the receipt of the funds into Khf's account as payment to them, KSF;

ii) That KSF elected to treat the payment into Khf's account as payment to them; and

iii) That a (contractually binding) agreement was made whereby KSF agreed to look to Khf and not to UBS for payment of the $65 million.

71. UBS pleaded that "KSF constituted Khf as its agent to receive the Funds [sc. the $65 million] and/or ratified KHF's receipt of the funds as agent for KSF". Clearly KSF did not make Khf their agent to receive the funds before they were sent: they were received by them only by mistake. In my judgment the doctrine of ratification does not assist UBS. It operates when a person purports to do something in the name or on behalf of another, without having actual authority to do so. If he does, then the putative principal may ratify what was done in his name or on his behalf, and so extend his rights by seeking an advantage for himself beyond any that he would have in the absence of ratification. Therefore the court more readily finds ratification than election (or an election of another kind), not least in that ratification is approval by way of a unilateral act of will (and need not be communicated to the counterparty), and may be made without the principal knowing the applicable law: see AMB Generali Holding AG v SEB Trygg Liv Holding Aktiebolag, [2005] EWCA Civ 1237 at paras 37 and 46. Ratification can be implied from "acts when you cannot logically analyse the act without imputing such approval to the party, whether his mind in fact approved or disapproved or wholly disregarded the question": Harrisons & Crossfield Ltd v London and North-Western Railway Co, [1917] 755, 758 per Rowlatt J.
72. I do not consider that Khf did any act in the name or purportedly on behalf of KSF. They "received" the funds in the sense JPMC made a credit entry in their account, but Khf had no part in that. There is no reason to think that thereafter Khf did anything other than agree that JPMC might (or should) debit their account (or reverse the credit) and enter a corresponding credit in KSF's account. UBS's pleaded case is what KSF ratified was "Khf's receipt of the Funds as agent for KSF", and that was not an act done by Khf, still less an act done by them in the name or purportedly on behalf of KSF.
73. UBS's election argument was that "KSF's immediate actions in treating Khf as indebted to it amounted to an election", which was communicated to UBS by JPMC. The doctrine of election (or of approbation and reprobation) applies where a person has two inconsistent claims and he elects to abandon one and to pursue the other: having done so, he cannot, in general, chose to return to the abandoned claim and sue on it: see Halsbury's Laws of England (5^th Ed, 2014) para 312. (The law is stated similarly but by reference to remedies, rather than claims, in Goff & Jones, The Law of Restitution (7^th Ed, 2007) para 44-022, and this statement of the law remains authoritative, although the paragraph has not, as far as I have found, been retained in the 8th edition (2011), sub nom The Law of Unjust Enrichment.) In this case, UBS's submission (as formulated in their written opening) is that "A claim against UBS and a claim against Khf in respect of the US$65 million plainly constituted alternative rights - not alternative remedies – in the hands of KSF. KSF chose to assert the right against Khf. The election was communicated to UBS by JPMC, and was also implicit in KSF's actions in not claiming the US$65 million from UBS". Thus the election is said to have been made, as I understand it, by the time of the exchanges between UBS and JPMC.
74. I reject the election contention. The evidence does not show, in my judgment, that, at the time of the communications between JPMC and UBS, KSF had made a claim against Khf that they were entitled to have Khf arrange to transfer funds to them. I infer that they had requested Khf to transfer $65 million to them, but there is no reason to suppose that they were asserting that they had a right to require Khf to do so, rather than suggesting that this was a sensible and business-like way of resolving the mistake. But even assuming that KSF were asserting such a right or claim against Khf, that claim, arising presumably from Khf having received money paid under a mistake, would not have been inconsistent with them also claiming $65 million from UBS under the payment obligation under the FX trade. Such a claim against UBS would, of course, have to be made on the basis that the payment obligation had not been met, but I do not understand why a claim against Khf should necessarily be on the basis that it had been. As I understand UBS's pleading their case is that KSF were asserting title to the $65 million credited to Khf's account, but even if they were making some proprietary claim to the funds, it would not follow that UBS had discharged their obligation to pay in freely transferable funds. That is enough to answer the election argument, but it would face other difficulties, not least that election would require an unequivocal intention to abandon any claim against UBS, and I cannot accept that KSF had evinced any such intention by (say) 9 October 2008.
75. In my judgment there is an essentially similar flaw in UBS's contention that UBS and KSF were party to a contract that, by the Payment, UBS had discharged their payment obligation or were to be treated as having done so. Their case is that arrangements were reached following the Payment whereby "in consideration of UBS confirming that the Mistaken Payment was for KSF's account (and consequently that the Funds should not be returned to UBS), KSF agreed to look to Khf (as the recipient of the Funds) rather than to UBS for payment of the sum of US$65 million in respect of the Payment Obligation and thereby to release UBS from liability to pay such sum (hence discharging US$65 million of the Payment Obligation)". Ms Tolaney submitted that the position might be analysed on the basis that KSF reached a bilateral agreement with UBS and a separate agreement with Khf, or alternatively on the basis that there was a single tripartite agreement to which UBS, KSF and Khf were all parties. The precise analysis is not important: what matters is whether there was a contractual agreement to which UBS and KSF were parties whereby KSF agreed that UBS had discharged their payment obligation under the FX trade in respect of the $65 million (in consideration of them giving the amended payment instruction to JPMC and giving up any right to recover the Payment).
76. Ms Tolaney sought support for this contention in KSF's pleading (in Further Information served on 2 April 2013) that "Khf agreed to make the payment to KSF once UBS had given its consent", and that KSF therefore have a claim in contract against Khf. (The position adopted by KSF's pleading is understandable: they are after all making a claim against Khf in the winding-up.) It is not in issue before me, therefore, that KSF and Khf made a agreement to this effect, and I shall assume for the purpose of considering this discharge argument (without deciding) that the agreement (or arrangement) had contractual effect. There was no direct communication between KSF and UBS: all UBS did was to provide amended instructions when JPMC advised that Khf could not apply the funds and, at Khf's request, sought further information. They did not, on the evidence, provide the amended instructions because they had agreed with anyone to do so, still less because they had so agreed with KSF, but because they realised that they had "paid the wrong company". In my judgment UBS and KSF did not evince an intention to enter into a contract about how the $65 million should be treated or about whether and if so how the Payment should affect UBS's payment obligation.
77. There are other answers to this version of the discharge defences. I could only conclude that arrangements about how the Payment evolved into a contract on the basis that the various exchanges between UBS, Khf and JPMC are to be taken to evince a contractual intention. But on any view the arrangements did not expressly provide that the Payment should discharge UBS's payment obligation, and so UBS's argument would have to be that the contract impliedly so provided. However, there is no need for such a term to be implied in order for the contract to make business sense (or to give it business efficacy): it makes equal (if not more) business sense for the parties to have contemplated that the obligation would be discharged if, as a result of UBS's amended instructions, KSF received freely transferable funds into their account. In those circumstances, they would, I suppose, have impliedly agreed not to complain that the payment had not been made in the customary manner. Ms Tolaney sought to meet this point by submitting that it would be inconsistent for KSF to make a claim against Khf for the $65 million and a claim against UBS under the FX trade. As I have explained, I reject that submission. In any case, any inconsistency in KSF's conduct would be irrelevant to the proper interpretation and effect of a contract between UBS and KSF unless UBS knew of it or such knowledge is to be imputed to them: there is no evidence that they did know and no basis for imputing such knowledge to them. Therefore (subject to the reservation in para 70 above) I reject the discharge defences.

The accord and satisfaction defence

78. UBS submit that, by entering into the arrangements for closing out the Master Agreement and accepting the close-out payment of $5,119,194.05, KSF agreed that the payment should settle any outstanding claim that either UBS or KSF might have in respect of the Master Agreement (that is to say, I suppose, any claim that for a loss or cost in connection with the Master Agreement or a "Terminated Transaction"), and so KSF agreed to settle any claim for the $65 million under the FX trade. There were three limbs to UBS's contention:

i) That is in the nature of a close-out arrangement of this kind and its obvious commercial purpose that the parties are settling all outstanding obligations under their Master Agreement, and the parties are to be taken to be evincing an intention to do so;

ii) That it is to be inferred that, by the time of the payment, KSF had completed their review of UBS's calculation and UBS reasonably understood that they had done so; and

iii) That Mr Littlewood agreed with Ms Alford that the close-out amount covered all their outstanding claims.

79. In view of my findings about the exchanges between Ms Alford and Mr Littlewood, in my judgment UBS have not established the factual basis for the third limb of their argument. I accept they have established the factual basis for the second limb. The mere fact that KSF were pressing for payment does not prove that KSF had completed their review of UBS's calculation: they made it clear by Freshfields' email of 21 May 2009 that Ernst & Young were looking for the amount to be paid before they had done so. However, it is apparent that the position had been further investigated by KSF before October 2009, and I infer that by the time of payment they did not have it in mind to carry out further checks on the calculation. This conclusion finds some support in remarks that, I conclude, Mr Littlewood made to Ms Alford. I also accept Ms Alford's evidence that, really for these very reasons, she supposed by the time that the payment was made that KSF had done all the checks that they were intending to make, and this was a reasonable view for her to have taken.
80. This does not mean that, by accepting the close-out payment, KSF impliedly agreed that they had no further claims against UBS under the Master Agreement. It is well-established that the test of whether a term is to be implied into a contract does not depend upon whether it would have been a reasonable provision for the parties to have made, but whether the term is necessarily implied, in order to give the contract business efficacy or from the necessary implication of the parties' words and conduct or on some other basis. I do not propose to express any view about what might be the implications of close-out arrangements under an ISDA master agreement under other circumstances. The response of KSF, through their administrators, in the letter of 7 April 2009 made clear that they did not intend to abandon any claims by receiving the payment of the close-out amount calculated by UBS. UBS submitted that the reservations in the letter were "overtaken by events", because of their exchanges with KSF or Freshfields on their behalf. In particular, it was suggested that the basis of the reservation was that UBS's calculation was to be subject to review, and UBS were led to understand well before the payment was made that the review had been conducted. I cannot accept that argument: the reservation in the letter was in much wider terms, and I would find it impossible, in view of the letter, to interpret KSF's conduct as evincing an intention to abandon the reservation in the letter. KSF went further, and submitted that this conclusion follows from the fact that by the time of the payment UBS received three further copies of the letter: on 16 May 2009, 25 June 2009 and 6 October 2009. I do not attach importance to that: as I interpret the exchanges, they were sent simply as a convenient way of confirming the details of the account to which payment should be made. But the original letter is enough to defeat the accord and satisfaction defence.

Election and waiver defences

81. Are KSF estopped (whether by convention or by acquiescence or by representation) from claiming that the payment obligation has not been discharged, and have they waived their claim under it (in the sense of KSF having relinquished rights, rather than of them having made an election between claims or remedies)? These various legal doctrines invoked by UBS all depend, in part, on them (i) proving that UBS assumed or came to understand that their payment obligation in respect of the $65 million had been discharged, and that KSF shared that assumption or understanding, or at the least did not indicate that they did not consider the obligation discharged; and (ii) showing that it would be unjust to allow KSF to assert a claim inconsistent with it.
82. The understanding of both parties is, in my judgment, reflected in their internal exchanges and records about the payment, in how they conducted the closing-out process and in the claims that they made (or did not make) in Khf's winding up. I take first the evidence about UBS. I conclude that by 8 October 2008 UBS considered that, by giving their amended instruction to JPMC, the problem arising from the mistake was resolved and that they considered the FX trade was "closed"; and JPMC's SWIFT message of 9 October 2008 corroborated UBS's understanding that the matter was concluded and the payment of $65 million had been credited by JPMC to KSF. To my mind this would be their natural understanding of the position following these exchanges, and this is also shown by an internal exchange of "chat" messages within UBS Zurich between Mr Milan Mueller and Mr Reto Neapflin: in particular, at 10.54 am Mr Neapflin enquired, "Milan, USD 100 million are closed?", and Mr Mueller replied "100 million/35 million/65 million is closed". (I attach no weight to the message from UBS Stamford that they were closing their file: that, as Mr Matthews pointed out, might well refer only to that office's involvement as a conduit between JPMC and UBS Zurich.)
83. My conclusion that UBS considered the FX trade closed is consistent with an extract from UBS's ledgers, disclosure of which was made only after the trial. (This late disclosure again was attributed by Herbert Smith Freehills ("HSF"), UBS's solicitors, to "targeted searches" carried out after the trial. It was said that the document was stored in a back office archive that had not been previously searched. On the face of it, it should have been, but I am not really in a position to comment upon or criticise UBS's disclosure processes and in any case it would be beside the point to do so.) The extract is a little difficult to interpret, but apparently ledger entries were made relating to the obligation to pay $100 million, and also about the payments of $65 million and $35 million or the arrangements to make them. It is unclear whether the entry about the $65 million was affected by the confusion between KSF and Khf: in a supplementary submission made after this late disclosure Mr Matthews said that there is no evidence that the $65 million was ever booked specifically as a sum payable to KSF. But that does not seem to me important. What matters is that there is, as I understand the document, an entry in the ledger made on 8 October 2008 that treated the payment as closed, and (perhaps less immediately relevant) there are similar entries in respect of the payments of $100 million and $35 million. I accept Ms Tolaney's submission that the natural implication is that the obligation to pay $100 million was reconciled with either the payments of $35 million and $65 million or at least the arrangements to pay them: hence all three items have the same "bundle" reference (connoting, I would infer, a single reconciliation). It appears from the record that these entries were made at 9.00 am (EDT), that is to say shortly after Mr De Blasi's exchanges with Ms Acosta and UBS (Zurich) had sent the SWIFT message to amend the payment instructions, and they were apparently initiated by or otherwise associated with Mr De Blasi. Mr Matthews observed that the ledger entries are timed before the amended instructions were passed to JPMC, and so before JPMC had even had a chance to send the requested confirmation. I cannot attach any significance to this: I infer that UBS assumed that their instructions would be implemented, notwithstanding their (unremarkable) request for confirmation from JPMC.
84. There is further evidence that the FX trade was recorded in UBS's internal records as closed. According to Ms Alford, if it had not been entered on UBS's systems that the $65 million was paid or otherwise settled, it would have shown up as a nostro break, a mismatch between the expected and the actual settlement cash-flows. When in late 2011 or early 2012 Ms Alford sought to trace the payment, it did not show up, and I infer that it did not show up in 2009 when the FX department were involved in the calculation of the close-out payment under the Master Agreement. I conclude that an entry had been made on the system that the $65 million payment obligation was discharged. Before the late disclosure, Mr Matthews suggested that the FX trade might never have been recorded on UBS's systems in the first place, but that always seemed improbable and has now been refuted.
85. In my judgment UBS continued to believe that they were under no outstanding payment obligation in respect of the FX trade despite the communications in November 2008. As I have explained, it is unclear why UBS Zurich did not respond to the enquiries made through UBS Stamford on and after 4 November 2008 about the $35 million, but those enquiries were not directly about whether the payment of the $65 million had been made. KSF argued only that had UBS investigated the position they might have learned that the $65 million had not reached KSF's account. Rightly it was not suggested that UBS were alerted to this: that would be unwarranted speculation and I would reject it.
86. What of the enquiries made by Ms Man? It was apparent from what she wrote that she knew nothing of KSF's arrangement with Khf for the Payment to be transferred or of the amended instructions. There was nothing in her enquiry that would have led UBS to doubt that the amended instructions had undone the initial mistake, and, whether or not they communicated with Ms Man or someone else at KSF in response to the inquiry, I conclude that it did not affect UBS's understanding of the position: given that Ms Man clearly had no real understanding of what had happened, there was no reasonable reason for it to do so.
87. This conclusion is confirmed by how UBS later dealt with KSF and Khf. If they had not understood that the Payment had been used to discharge their obligation to KSF, I infer that UBS would have considered that they had a claim for the money against Khf in the winding-up in Iceland and have made the claim before the window closed on 30 December 2009. I also conclude that, had UBS not understood that the FX trade was closed and their obligations under it discharged, they would have brought it into account when they calculated the close-out payment. In particular, it is clear that UBS's foreign exchange department was closely involved in the close-out process, and if they had thought or even suspected that KSF had, or might have, a claim in respect of the FX trade, they would have included it in their calculations or mentioned it when dealing with the close-out accounts. It was not suggested to Ms Alford, and there is no proper basis to think, that UBS were not seeking properly to include in the calculation all outstanding obligations between themselves and KSF.
88. I do not overlook Mr Matthews' submission that, in view of the paucity of disclosure by UBS and since UBS did not adduce evidence from anyone with contemporaneous involvement with the transaction or from anyone who dealt with KSF's enquiries about the FX trade later in 2008, I should not draw any inference in UBS's favour about their view of the FX trade or whether their obligations under it were discharged. In particular UBS did not adduce evidence from Mr De Blasi, who (as Mr Matthew submitted and I accept) was much involved for UBS with the FX trade around 8 October 2008. Mr Matthews submitted that therefore there is not sufficient evidence to conclude (i) that UBS closed the entries about the FX trade in their systems; (ii) that they formed the belief that KSF had been paid; or (iii) (a point to which I come later), if they did so, that belief or assumption was induced by KSF or something for which they are responsible.
89. In my judgment, there is no sufficient reason to think that UBS have not now made proper disclosure, and no application for specific disclosure was made. There was no evidence, and Ms Alford was not asked, about whether those contemporaneously involved with the FX trade still worked for UBS, and in any case I can well understand that witnesses would not recall what, from UBS's point of view, would have been a fairly routine trade over five years ago. In some circumstances, as Brooke LJ explained in Wisniewski v Central Manchester Health Authority, [1998] Lloyd's LR Medical 223, a court may draw adverse inferences from the absence of a witness who might be expected to give material evidence on an issue. It is then relevant to consider whether there is evidence of a convincing reason for the witness' absence: in the case of Mr De Blasi, there is no evidence about whether he still works for UBS or whether they know where he is. However, in the end I was not invited to draw adverse inferences from the fact that UBS did not call evidence from Mr De Blasi or other witnesses. Mr Matthews submitted simply that, as a result, UBS had not discharged the burdens of explaining why they closed out the trade on their systems and of showing that they regarded their payment obligation under the FX trade as discharged, but I reject that submission.
90. It is not suggested that KSF indicated to UBS that they did not consider that UBS had discharged their obligations under the FX trade (leaving aside the enquiries in November 2008, to which I have referred). I conclude that they did not do so because they, like UBS, understood that the obligation had been discharged and the FX trade was closed. In my judgment KSF's understanding of the position was that the FX transaction was closed as far as UBS were concerned, that they (KSF) were to look to Khf to have the $65 million transferred to them and that UBS owed KSF no further payment. I do not suggest that the Administrators themselves had any such understanding of the position when they were appointed: Mr Brazzill said, and I accept, that they had no knowledge of the transaction until long afterwards. What matters for present purposes is the understanding of those who were dealing with the transaction for KSF. Here Ms Tolaney pointed out that KSF called no evidence from anyone involved with the FX trade during the period between 3 and 9 October 2008, and thus there was no direct evidence as to what those at KSF understood about whether UBS still owed them $65 million; and in particular, they did not adduce evidence from Mr Carrigan, who was undoubtedly closely involved for KSF. Mr Brazzill knew that Mr Carrigan was working in London, but there had been no enquiry where: the clear inference was that he could have been traced. In view of this, to my mind, applying the approach explained in the Wisniewski decision, the fact that Mr Carrigan has not given evidence lends some support for my conclusion about KSF's understanding of the position or their assumption about it. However, my conclusion does not depend on this consideration.
91. It seems to me that KSF's understanding of the position is most clearly manifested in the fact that, having learned that the $65 million had been credited to Khf's account, they did not draw UBS's attention to the mistake. They thought, I infer, that there was nothing to concern UBS. If they had thought that UBS remained liable to them or might do so, they would surely have contacted them: to my mind that is what any reasonable bank in KSF's position would have done. The fact that they instead considered that they were entitled to the money from Khf is reflected in their request that the $65 million be brought into account under the netting arrangement of 8 January 2008. Of course, it was possible, as a matter of logic and of law, that Khf might accept that KSF were entitled to the $65 million but that UBS remained liable to KSF unless and until it was paid, but the netting-off request is consistent with KSF's internal view being that they were looking simply to Khf for the $65 million. It is clear from his exchanges with Ms Tay that this was still how Mr Carrigan saw the position in June and July 2009. I infer, in the absence of evidence to the contrary, that this is why KSF did not make the claim for $65 million in response to UBS's calculation in the close-out process. It is also why, when presenting a calculation of their account under their ISDA Master Agreement with Khf using the "loss" formula, KSF considered that they could, consistently with the ISDA Master Agreement requirement of a bona fide and reasonable determination, properly include their claim for $65 million. (I recognise that it probably should not have been included because it was not a claim in respect of their ISDA Master Agreement with Kfh or a Terminated Transaction, but that does not detract from its evidential significance about KSF's understanding of the position with regard to UBS's liability under the FX trade.) It is also consistent with the claim made by KSF in the winding up.
92. In my judgment KSF's internal records in their IBIS system also reflect their conclusion or assumption that the FX trade was closed as far as UBS were concerned. They show that on 27 October 2009 an entry was made in the account for UBS by way of a credit in the sum of $65 million with the narrative (as far as it is intelligible) "UBS paid hf". Another entry on the account for Khf showed a debit against the account to reflect that Khf, it was said, owed KSF $65 million. Mr Brazzill agreed in cross-examination that the documents showed the KSF treated UBS as having paid the $65 million and Khf as owing that sum to KSF. In his witness statement he said that this was because "KSF's internal ledgers would not have been able to record US$65 million as being due from both Khf and UBS: it would have to be one or the other. Any parallel, contingent or otherwise connected obligation could at most be reflected by entering a comment on to the system, ie; simply a text entry rather than an 'accounting' entry". I do not find this explanation convincing: it became clear on cross-examination that Mr Brazzill had only a superficial understanding of how the system operated and had not asked questions to understand it more fully. Mr Youngs understood that Mr Brazzill's description of the system applied only to foreign exchange trades, but Mr Brazzill apparently did not understand that this was the position and there is no obvious reason that only that part of the systems should have these limitations. More importantly, if the entry in respect of the $65 million on UBS's account on the IBIS system was made only because otherwise it could not record Khf's (supposed) obligation to KSF and the entry did not really represent KSF's understanding of the position, that would surely have been reflected in the narrative: it was not. In my judgment, these internal records of KSF reflect the understanding of those dealing with the FX trade for KSF that UBS's payment obligation had been discharged.
93. Do these conclusions support UBS's submission that KSF are precluded from contending that UBS did not discharge their payment obligation in respect of the $65 million? I take first the argument of estoppel by convention arising from their parties' understanding that UBS had done so. Chitty on Contract (cit sup) states at vol 1 para 3-109 "To give rise to an estoppel by convention, the mistaken assumption of the party claiming the benefit of the estoppel must … have been shared or acquiesced in by the party alleged to be estopped; and both parties must have conducted themselves on the basis of such a shared assumption: the estoppel 'requires communication to pass across the line between the parties'", this citation being from Compania Portorafti Commerciale SA v Ultramar Panama Inc (The "Captain Gregos"), [1990] 2 Lloyd's Rep 395, 405 However, this requirement does not mean that the assumption of the party seeking the benefit of the estoppel must have been created by the other party. The position is explained more fully in Halsbury, Laws of England, (5^th Ed, 2014) vol 47 at para 374: "Representation must influence the mind of the person to whom it is addressed. The representation must either produce a belief or expectation in the mind of the representee or confirm or strengthen a belief which he already holds". The Editors cite "Amalgamated Investment and Property Co Ltd (in liquidation) v Texas Commerce International Bank Ltd, [1982] QB 84, 104-105, per Robert Goff J ("there may be cases where the representee has proceeded initially on the basis of a belief derived from some other source independent of the representor, but his belief has subsequently been confirmed by the encouragement or representation of the representor"), affd [1982] QB 84, CA; and The Stolt Loyalty, [1993] 2 Lloyd's Rep 281, and affd [1995] 1 Lloyd's Rep 598 CA (where the defendant's silence encouraged an expectation in the claimant's solicitor's mind that she need take no further action in order to preserve time)".
94. UBS initially assumed that their payment obligation was discharged when on 8 October 2008 they realised that they had "paid the wrong company", but believed they had remedied it with the amended instructions. They first realised that they had paid the wrong company, as I infer from the history of events, because Khf asked, through JPMC, for further details about the sum credited to their account. That request was initially made by Ms Howser of Khf when she probably did not know that the payment was intended for KSF and was probably unaware of the exchanges about Khf remitting the receipt to KSF. KSF knew that Khf were seeking to obtain instructions to allow the remittance from UBS, and, as Khf knew, on 7 October 2008 they were relying on Khf to pursue the matter with UBS for and in the interest of KSF as well as in their own interest. However, I also conclude that, although UBS thought that the error had been put right and their payment obligation under the FX trade discharged when they sent their amended instructions, they were reinforced in that belief when JPMC replied confirming that they were amending the payment as instructed. The communications between KSF and Khf on the one hand and UBS on the other hand were conducted through JPMC, and, in my judgment, those communications and in particular the message from JPMC of 9 October 2008 satisfy the requirement of estoppel by convention that communications "cross the line". This is not a case where two parties to a contract, operating independently of each other, both formed the same (mistaken) view of their rights and obligations under it. I add that, of course, UBS would have questioned their understanding if those involved with the FX trade for KSF had communicated with them directly and said that they had not received the $65 million and it was still outstanding, but they did not do so.
95. It follows, in my judgment, UBS have established these requirements of estoppel by convention described in the passage from Chitty on Contracts that I have cited. I also consider that, if the parties did conduct their business on the basis of such a shared assumption, then KSF are not entitled to go back on it if it would be unjust or, as it is sometimes put, unconscionable for them to do so. But before coming to that, it is convenient first to consider whether UBS have also established the basis for their other arguments of estoppel and waiver. As has often been observed, the factual basis for estoppel by convention is often closely connected with that for other forms of estoppel and that is the position here. But the doctrines are distinct and each has its discrete "terrain of application": The Republic of India v India SS Co (No 2), [1998] AC 878, 914D per Lord Steyn. Lord Steyn said of estoppel by acquiescence (at p.913H/914B) that:

"The parties were agreed that the test for the existence of this kind of estoppel is to be found in the dissenting speech of Lord Wilberforce in Moorgate Mercantile Co Ltd v Twitchings, [1977] AC 890. Lord Wilberforce said, at p.903, that the question is: 'whether, having regard to the situation in which the relevant transaction occurred, as known to both parties, a reasonable man, in the position of the 'acquirer' of the property, would expect the 'owner' acting honestly and responsibly, if he claimed any title to the property, to take steps to make that claim known . . .' Making due allowance for the proprietary context in which Lord Wilberforce spoke, the observation is helpful as indicating the general principle underlying estoppel by acquiescence."

This test was examined recently by Blair J in Starbev GP Ltd v Interbrew Central European Holdings BV, [2014] EWHC 1311 (Comm), and I grateful adopt his statement of the law (loc cit at para 133):

"The question is what is meant by "acting honestly" for these purposes. It does not in my view necessarily imply that the party against whom the estoppel is raised must be guilty of actual dishonesty, in the sense of acting fraudulently. [The defendant] seemed to acknowledge this in its closing submissions, saying that what is required is "'dishonesty' in an equitable sense". It submitted that absent a relationship of good faith or partnership or something akin to a joint enterprise, the courts will not impose a duty to speak in the absence of impropriety of some description by the person who is alleged to be estopped. I would accept that way of putting it, subject to adding that the impropriety may come from the act of staying silent itself, as where a reasonable person would expect the person who is alleged to be estopped, acting honestly and responsibly, to bring the true facts to the attention of the other party known by him to be under a mistake as to their respective rights and obligations—this was how Bingham J put it in The Lutetian, [[1982] 2 Lloyd's LR 140, 157], (see also Avocet Industrial Estates LLP v Merol Ltd, 2011] EWHC 3422 (Ch) at [124], Morgan J). The "reasonable person", it is to be noted, is a reasonable person in the position of the party raising the estoppel, in this case Starbev."

96. If I am right in my conclusion that those acting for KSF understood that UBS had fulfilled the FX trade, then of course they were neither dishonest nor irresponsible in not communicating with UBS over the mistake in the Payment and its consequences. But to my mind a bank such as KSF would not be acting honestly (in the sense explained by Blair J) or responsibly if it knew about or even seriously suspected a mistake of this kind and that as a result a counterparty such as UBS had not discharged their payment obligation, but decided not to inform the counterparty of the position;. A reasonable counterparty in UBS's position would expect to be informed as a matter of honest and responsible dealing.
97. In their Reply, KSF plead that in view of the company going into administration and because the Administrators were "preoccupied with the complex process of investigating and managing the fallout from [the] combined collapse" of KSF and Khf, they could not "reasonably be taken to indicate anything about the status of UBS' Payment Obligation", and so (the implication is) KSF cannot be subject to an estoppel by acquiescence. (In their pleading, they also rely on the Administrators being under a statutory duty under "The Kaupthing Singer & Friedlander Limited Transfer of Certain Rights and Obligations Order 2008 to prioritise the orderly transfer of KSF's internet-based deposit business … to ING Direct". That point was not pursued at the trial, except by way of an explanation by Mr Brazzill about why administrators prioritised their work as they did.) This point is not supported by evidence: of course the administrators were faced with an enormous task when they were appointed. But there is no evidence that persuades me that this affected in any way the ability of Mr Carrigan or others in the operations department or "back office" to communicate with counterparties such as UBS in a normal businesslike way or inhibited them in doing so. I am not persuaded that the administration and its complexities explains or excuses KSF from contacting UBS about the Payment or the FX trade, if only to say that the payment had not reached their account and to reserve their position. KSF would have needed to adduce convincing evidence to make good that argument, and they adduced no credible evidence at all.
98. Ms Tolaney also submitted that the case can properly be seen as one of estoppel by representation, the representation being made by silence. UBS pleaded that "KSF did not contend (at any material time) that it had any outstanding claim against UBS in respect of the Mistaken Payment and/or the Payment Obligation or reserve or seek to reserve any entitlement to make such claim. By failing to do so, particularly in the context of and/or at or around the time of the close out process …, KSF ... unequivocally represented that UBS's Payment Obligation had been discharged …". As with estoppel by acquiescence, this argument would largely depend, I think, on KSF being under a duty to speak because of circumstances such as Lord Wilberforce described in Moorgate Merchantile Co Ltd v Twitchings (cit sup): see Orion Finance lt v J D Williams & Co Ltd, [1998] CLLR 36 45. I have explained why I consider that they would be under such a duty unless they considered the payment obligation to be discharged, and the argument does not, I think, require separate discussion. I only add that I would reject any suggestion that, estoppel by representation being a common law doctrine and part of the law of evidence, this analysis by-passes the need to prove that it would be unconscionable or unjust for KSF to resile from the representation. The link between the circumstances capable of giving rise to the various species of estoppel is unconscionability: see Johnson v Gore Wood & Co, 2002] 2 AC 1,41C per Lord Goff, and also Fraser v Kelly, [2012] UKPC 25 at para 17 (cited below) .
99. Nor do I propose to consider separately with the argument that UBS have an answer to the claim in the doctrine of forbearance or waiver of their breach of the payment obligation: similar questions would arise here as to whether it would be inequitable for KSF to retract their forbearance, and whether UBS's conduct was influenced by it: see Chitty on Contracts (cit sup) vol 1 paras 30-083 and 22-044, and Benjamin, Sale of Goods (8^th Ed, 2010) paras 12-037 and 12-038. Although each of these various doctrines has its own history, if there be cases in which a defence of waiver of breach or forbearance could succeed absent some form of estoppel, this is not one.
100. In their Reply KSF pleaded that they rely in answer to the defences based on estoppel and waiver on these provisions of the Master Agreement:

i) "No amendment, modification or waiver in respect of this agreement will be effective unless in writing (including a writing evidenced by a facsimile transmission) and executed by each of the parties or confirmed by an exchange or telexes or electronic messages on an electronic messaging system" (section 9(b)); and

ii) "A failure or delay in exercising any right, power or privilege in respect of this Agreement will not be presumed to operate as a waiver, and a single or partial exercise of any right, power or privilege will not be presumed to prelude any subsequent or further exercise, of that right power or privilege or the exercise of any other right, power or privilege" (section 9(f)).

Mr Matthews did not rely upon these provisions at trial. He was right not to do so, and it suffices to say that they would not cover UBS's estoppel defences.

101. I therefore come to the questions whether and how UBS's dealings were affected by their mistaken understanding and KSF's failure to communicate any claim earlier than they did; whether as a result they acted to their detriment; and whether it would be unjust or unconscionable to allow KSF now to assert a claim arising from UBS's payment obligation under the FX trade. I conclude that, because of their mistaken understanding or assumption that they had discharged their payment obligation, reinforced by KSF's failure to indicate that they had not received the Payment and did not consider the obligation discharged (or at least that they reserved their position with regard to it), UBS:

i) Closed out the Master Agreement with the payment of $5,199,194.05;

ii) Entered into the agreement with Khf and paid them some $14.3 million under it; and

iii) Compromised their position with regard to claiming against Khf in particular by making no claim until after the window for claims was closed.

Thus, UBS acted in reliance on their assumption or misunderstanding to their detriment in each of these respects, and each, in my judgment, resulted in a substantial detriment to UBS, even if it is difficult or even impossible to quantify it. As Lord Sumption explained in Kelly v Fraser, (cit sup) at para 17, "The relevance of detrimental reliance in the law of estoppel by representation is that it is generally what makes it unjust for the representor to resile from his previously stated position". In my judgment it would be unjust (for the purposes of establishing an estoppel by convention) to allow KSF to resile from the understanding of about the FX trade that they shared with UBS, or (in respect of estoppel by acquiescence) from the position in which KSF are to be taken to have acquiesced: see The Republic of India case, loc cit at pp.913D–914D. In Fraser v Kelly (loc cit) Lord Sumption went on to say that

"… the detriment need not be financially quantifiable, let alone quantified, provided that it is substantial and such as to make it unjust for the representor to resile. A common form of detriment, possibly the commonest of all, is that as a result of his reliance on the representation, the representee has lost an opportunity to protect his interests by taking some alternative course of action. It is well established that the loss of such an opportunity may be sufficient detriment if there were alternative courses available which offered a real prospect of benefit, notwithstanding that the prospect was contingent and uncertain: …"

I therefore reject the submission of Mr Matthews that, in order to establish a defence of estoppel, UBS have to demonstrate the value of the opportunities that they lost or of the detriment suffered.

102. Mr Matthews submitted, however, that UBS did not suffer detriment in relation to Khf because:

i) Khf's "immediate creditor" in respect of the US$65m was not UBS, but KSF, and UBS have only a contingent claim against Khf; and

ii) Any debts that UBS might have brought into account against the $65 million in the settlement agreement of 20 December 2010 with Khf were not owed by UBS but by UBS Limited.

I reject these arguments.

103. The first is based on the English rule of insolvency law against "double dipping": see In re Oriental Bank, (1871) LR 7 Ch App 99, 103-104 per Mellish LJ and In re Kaupthing Singer Friedlander Limited (No 2), [2011] UKSC 48 at paras 8ff per Lord Walker. The purpose of the rule is to prevent creditors of an insolvent entity being unfairly treated because multiple claims are made in respect of what is in substance same debt, and it typically applies when a surety would have a claim if he discharged a liability of the insolvent entity. KSF did not plead that they relied on the rule in respect of UBS's claim in Khf's insolvency: indeed, as I read paragraph 15 of their Reply, they recognised that UBS had a separate claim against Khf that was not contingent on discharging KSF's claim in respect of the $65 million ("it is admitted and averred … whether through UBS' own claim or (upon payment by UBS to KSF) through subrogation to KSF's rights against Khf, the economic value of UBS' rights of recovery against Khf has been preserved"). As a result, there was little material before the court about the point, and there was little argument. There was no evidence about whether Icelandic law has a similar rule against double dipping, and I am not persuaded that the English rule against double dipping would answer UBS's claim against Khf. Mr Matthews submitted that UBS's claim against Khf is "merely contingent" and will remain so unless and until they pay KSF in full. However, on the face of it UBS have a claim for money paid under a mistake, which is not contingent. It might be connected with any claim that KSF might have in that, if KSF's claim were based on a contract that Khf made before they learned of the mistake, this would provide them with a change of position defence, but that does not mean that UBS's claim against Khf is in some sense contingent or inchoate. I do not propose to decide questions of this kind: it suffices that I am satisfied that UBS have a properly arguable claim against Khf subject to the argument that it was made too late, that it was made late by UBS because they understood or assumed that their obligation to KSF had been discharged, and therefore their claim in the winding up is therefore significantly compromised.
104. I can deal with the second point more shortly: again it was not pleaded and there is little material about it and about the respective roles of UBS and UBS Limited in relation to the agreement of 20 December 2010. However, under it UBS (and not UBS Limited) agreed to make the payment to Khf, and the claim for $65 million might have been brought into account against this payment: it does not matter whether the debts that UBS were paying were their own or UBS Limited's.
105. Mr Matthews put forward another argument, conveniently labelled the "pro tanto" argument. The thrust of it was that the benefit of the estoppel should be limited to the extent of any detriment that UBS have proved, and they have not quantified what detriment they have suffered nor have they demonstrated any specific detriment. In particular, Mr Matthews sought to demonstrate when cross-examining Ms Alford that, at worst, the detriment to UBS arising from their dealings with Khf was that (i) they would not have paid Khf the equivalent of some $14.3 million, and (ii) they have prejudiced their claim in the liquidation, which in any event will not be paid in full.
106. As was reiterated by Lord Sumption in Kelly v Fraser (cit sup) the general rule is that it is not necessary to demonstrate quantifiable detriment in order to establish an estoppel, provided an estoppel must be recognised in order to prevent injustice. The court does not otherwise measure the benefit resulting from the estoppel against the detriment and seek a result that balances them. However, this rule has not been applied in some cases of money paid under a mistake: see Scottish Equitable plc v Derby, [2001] EWCA Civ 369 and National Westminster Bank plc v Somer International (UK) Ltd, [2001] EWCA Civ 970. In the Somer case (loc cit at para 60) Clarke LJ said that the exceptions to the rule that estoppels do not operate pro tanto "are or include cases where it would be unconscionable or wholly inequitable to permit the recipient of money to retain the whole of it".
107. Here another pleading point arises: Ms Tolaney made it clear in her opening that UBS contend that the pro tanto argument is not available to KSF because it has not been pleaded. The procedural history is this: in a letter dated 5 June 2014 KSF's solicitors, Quinn Emanuel, sent a draft reply pleading that would introduce the pro tanto argument: they said that amendment was unnecessary but invited UBS to consent to it. In the draft reply it was pleaded only that " … insofar as UBS is able to prove some relevant financial detriment then estoppel only operates to the extent of that detriment so that UBS has no defence in respect of the balance of the US$65 m Payment Obligation". By letter dated 12 June 2014 HSF wrote that UBS declined to consent to the proposed amendment: they said that it was too late to introduce the issue and that "The proposed amendment is in the barest of terms and does not identify what facts and matters are relied upon in support of that contention". They said that, if KSF wished to amend, they should apply at the trial. Despite Ms Tolaney's clear objection to the point being argued without being pleaded, no application to make this amendment was made.
108. It is clear that, in so far as the law recognises a pro tanto principle, it is by way of exception to the general rule: see Clarke LJ (cit sup) and Avon CC v Howlett, [1983] 1 WLR 605, in which the seeds of the principle were sown. As I see it, if a party wishes to raise an argument that the court should not apply the general rule (because, as Clarke LJ put it, it would be unconscionable or wholly equitable to do so or for some other reason), this must be pleaded. The argument will almost inevitably raise issues of fact that need to be defined. I agree with HSF's objection that the draft pleading was in too general terms to identify these issues, but in any case KSF never pursued the proposed amendment. I would reject the pro tanto argument because it has not been pleaded.
109. I would in any case have rejected it. The pro tanto principle has been recognised only in cases of a claim for unjust enrichment made on the basis of a payment under a mistake. In this context there is a tension between the operation of a change of position defence to a claim in restitution and the operation of estoppel: hence Professor Burrows has suggested in The Law of Restitution (2011) at p.557 that "the all or nothing estoppel defence is in this context inapt and should be excised". The law has not gone that far, but the arguments for an exception to the general "all or nothing" rule do not apply to this case. It would be novel to extend the scope of the exception beyond restitutionary claims, and I decline to do so.
110. Furthermore I do not consider that it would be unconscionable or "wholly inequitable" for UBS to have the full benefit of an estoppel defence. First, overall they will not be "in pocket" in the sense that they have paid away the $65 million: in a broad sense they will never profit in the way that troubled courts in the Avon case and subsequent cases where it was recognised that there might be exceptions to the general rule. Secondly, the nature of the detriment suffered by UBS is difficult, if not impossible, to quantify: this is not a case where, by departing from the general rule, the court could be satisfied that a fair balance was struck between the detriment and the benefit of the estoppel.
111. I therefore conclude that the detriment that UBS would suffer if KSF can resile from the assumption or understanding about the FX trade is such that it would be unjust to allow KSF to do so. For my part, I would regard estoppel by convention as the relevant legal principle, but other legal doctrines might well be equally applicable and others might prefer another label.

The application to amend

112. This is enough to lead me to dismiss KSF's claim, but I must come back to the application to amend the claim in order to meet the termination defence. It is clear after the decision of the Court of Appeal in Swain-Mason v Mills & Reeve, [2011] EWCA Civ 14 (if it was not before) that "a heavy onus lies on a party seeking to make a very late amendment …" (per Lloyd LJ at para 72). In Brown v Innovatorone Plc, [2011] EWHC 3221, Hamblen J summarised the position as follows (at para 14):

"As the authorities make clear, it is a question of striking a fair balance. The factors relevant to doing so cannot be exhaustively listed since much will depend on the facts of each case. However, they are likely to include:

(1) the history as regards the amendment and the explanation as to why it is being made late;

(2) the prejudice which will be caused to the applicant if the amendment is refused;

(3) the prejudice which will be caused to the resisting party if the amendment is allowed;

(4) whether the text of the amendment is satisfactory in terms of clarity and particularity.

It was because of the last point, which reflects the judgment of Lloyd LJ in the Swain-Mason case at para 82 ("[The applicants] ought to have put forward a significantly clearer, fuller and more precise pleading, one which immediately satisfied all the obligations on a party as to proper pleading, without any need to be supplemented or clarified by evidence or by further information") that I asked Mr Matthews in his closing submissions whether KSF wanted to revise in any way the draft amendment that they had put forward on the second day, and he confirmed that they did not.

113. No explanation has been given as to why KSF did not apply to amend earlier. The termination defence was pleaded by UBS, and not answered in KSF's reply. I can only conclude that its implications were not fully thought through before the trial. I have some sympathy if that is the position: sometimes the impact of arguments does become apparent only in oral exchanges during the trial process. But that is not a compelling reason to permit an amendment.
114. If I am right in my conclusion about estoppel, in one sense it might be said that KSF suffer no prejudice if I refuse their application in that their claim will in any event fail. On the other hand, KSF would suffer prejudice from me refusing permission to amend if I am wrong about estoppel and they are prevented from advancing the amended case on appeal.
115. Contrariwise, if I am right about estoppel, UBS do not need the termination defence and so they might be said to suffer no prejudice if the amendment is permitted.
116. Is the text of the amendment satisfactory? Two points about the draft might be made. First, it pleads that UBS's determination (or calculation) of their loss was not reasonable in that it did not include the "unpaid $65,000,000 in respect of the (now terminated) FX Trade", but it is not specific as to whether this is because (a) any determination that omits any obligation is not "reasonable" within the meaning of the definition of "Loss", (b) there was something about UBS's obligation under the FX trade that meant that a determination that did not bring it into account was necessarily and inherently unreasonable, or (c) UBS did not go about determining their "Loss" in a reasonable way and as a result the obligation in respect of the $65 million was omitted. Mr Matthews "fleshed out" the draft pleading in his oral submissions, but that is not how late amendments should be presented: in the nature of things oral explanations do not have the precision of a pleading.
117. Secondly, the proposed pleading criticises UBS's determination only on the basis that was not a "reasonable determination". It does not criticise it on the basis that for other reasons it is not a determination as required by the definition of "Loss": for example because it was not of a determination of UBS's "total losses" (emphasis added) or because it did not "include losses and costs (or gains) in respect of any payment or delivery required to be made …" (emphasis added).
118. In light of these considerations I must decide whether it would be just and in accordance with the overriding objective to permit the amendment. As I said at paragraph 11 above, Mr Matthews sought to advance the case in three ways. The second and third (what I have labelled the necessarily unreasonable argument and the de facto unreasonable argument) are about whether UBS made a "reasonable" determination of their "Loss", and I am unable to accept that it would be fair to UBS to open up in the course of the trial an issue about whether their determination was reasonable. That would introduce questions about the determination that UBS had not prepared to meet. Nor can I accept that on a natural interpretation of the definition of "Loss" a determination would necessarily be unreasonable if it omits a transaction (either because those making the calculation were unaware of it or because they considered it closed or for any other reason). It might be that that such an omission might be very strong evidence that the determination was not reasonable, and it might be that in practice all but conclusive evidence, but that is very different.
119. What of the other way that Mr Matthews sought to present the case in his closing submissions, the invalidity argument? As I see it, that was not an argument that UBS did not make a "reasonable determination", but an argument that it was not a determination of what the definition of "Loss" required. It is not therefore covered by the draft amendment, and was not the case that Ms Tolaney answered in her submissions.
120. I have been concerned that it would be an unduly pedantic and technical approach to pleading to refuse even a late application to amend on this basis. However, there are other broader considerations: the amended case would raise important questions about the interpretation of the ISDA Master Agreement that to my mind are far from easy. For example, on the one hand, as Mr Matthews emphasised, if a determination of "Loss" were valid notwithstanding it did not bring a transaction into account, it would create a perhaps unexpected difference between the "Market Quotation" and the "Loss" payments measures. On the other hand, it might be surprising if (as Mr Matthews submitted) the consequence of the non-defaulting party omitting a transaction from his calculation invalidates the whole determination, however complicated, careful and cogent it otherwise was and even if the omitted transaction was relatively minor. Decisions about how English law interprets the ISDA Master Agreement have major implications: see Lomas and ors v JFB Firth Rixson Inc, [2010] EWHC 3372 (Ch) at para 53 per Briggs J. They must be presented to the court properly and the court must be confident that it is properly equipped to decide them. In my judgment it would not be right to allow KSF to develop these answers to the termination defence when they would not be clearly and properly pleaded even if the amendment were permitted.

Conclusion

121. I therefore refuse permission to amend the particulars of claim, and I dismiss the claim.