Property Alliance Group Ltd v The Royal Bank of Scotland Plc

This litigation arises out of interest rate swaps that the claimant, Property Alliance Group Limited (“PAG”), and the defendant, The Royal Bank of Scotland plc (“RBS”), entered into. It is PAG’s case that it is entitled to rescission of the swaps and/or damages, but in a judgment dated 21 December 2016 Asplin J (as she then was) dismissed the claims. PAG now appeals.

PAG is a property investment and development business. It operates mainly in the north-west of England and has a portfolio of industrial sites, offices and retail and leisure properties. At the relevant times, PAG’s managing director and majority shareholder was Mr David Russell, whom the Judge observed made all major decisions in relation to the company. Mr Ewan Wyse was the finance director.

RBS had become PAG’s principal source of commercial banking facilities by May 2003. It continued to supply such facilities until 2014, when, following the breakdown in the parties’ relationship, PAG refinanced with HSBC. The facilities which RBS provided to PAG were revolving and had two elements: “development facilities”, which involved relatively short-term borrowing to develop properties and were usually referenced to a margin over the Bank of England base rate, and “investment facilities”, which were used to finance income-producing property investment assets and were usually referenced to a margin over the London Inter-bank Offered Rate (or “LIBOR”).

During the relevant period, LIBOR was published on behalf of the British Bankers Association (“the BBA”) by Reuters after receiving submissions from panels of banks. The submissions were intended to represent the banks’ opinion as to:

“The rate at which an individual Contributor Panel bank could borrow funds, were it to do so by asking for and accepting inter-bank offers in reasonable market size, just prior to 11:00 [am] London time”.

The particular species of LIBOR to which the contractual arrangements at issue in this case were referenced was 3 month GBP (i.e. sterling) LIBOR.

We were taken to two agreements, dated respectively 27 April 2004 and 19 December 2006, to illustrate the basis on which RBS made funding available to PAG. Under one of them, interest was payable at 1.25% above LIBOR; the other provided for interest of 1% above LIBOR. In each case, clause 10.14 provided as follows:

“The Borrower shall ensure that an interest rate hedging instrument(s) acceptable to the Bank and at a level, for a period and for a notional amount acceptable to the Bank is entered into and maintained.”

Both agreements also contained (as clause 10.9 in one instance and as clause 10.10 in the other) a provision in these terms:

“The Borrower authorises the Bank from time to time to obtain an up to date Bank instructed and addressed professional valuation of all or any of the Charged Properties from a valuer/surveyor acceptable to the Bank and the Borrower shall meet the cost of any valuations obtained by the Bank provided that the Borrower shall not be liable for the cost of more than one valuation for each of the Charged Properties in any one calendar year other than a valuation obtained following the occurrence of an Event of Default.”

PAG and other companies associated with Mr Russell entered into some 11 transactions in derivatives between 2003 and 2014. The present proceedings arise out of four swaps (“the Swaps”) that RBS sold to PAG between 2004 and the spring of 2008.

The people with whom PAG dealt at RBS included Mr Anthony Goldrick and Mr Matthew Jones, who were successively the relationship manager, and Mr Anthony Bescoby, the individual who sold PAG the Swaps. Mr Jones took over from Mr Goldrick as relationship manager in January 2004.

PAG itself had a number of people to advise it on banking matters over the relevant years. The first of these was Ms Anne Taylor, who was taken on as a part-time consultant in October 2002 and described by Mr Russell in an email as a “banking consultant/specialist”. PAG dispensed with her services in May 2005, but in July 2007 Mr Richard Malin was engaged to work on a consultancy basis; Mr Wyse told Mr Jones in an email that Mr Malin would, among other things, “advise on new and future strategies including interest rate hedging”. Mr Malin was succeeded by Mr Jonathan Morton-Smith in early 2008, and Mr Morton-Smith was himself replaced by Mr Robin Priest in September 2009.

At times, PAG also had advice from JC Rathbone Associates Limited (“Rathbones”), a leading derivatives advisory firm. Rathbones were consulted in 2002 and were asked for their views periodically until late November 2004 when, after Rathbones had voiced concerns about the first RBS swap, Mr Wyse told them in a letter, “although we may seek to work with you in the future, we did not find it necessary to use your company to assist us”. PAG sought advice from Rathbones again from 2009.

The earliest of the Swaps (“the First Swap”) had a trade date of 6 October 2004 and a notional amount of £10 million. RBS was to have “the right to cancel the structure on 07-Oct-2009 or each quarter thereafter”, but the maturity was otherwise to be 7 October 2014. The transaction involved a “Multi Callable Libor Value Collar” summarised in these terms:

“Company [i.e. PAG] buys a 6.25%-5.25% Dual Strike Cap (Company protected at 5.25% until/unless 3 month LIBOR fixes at or above 6.25%, then company is protected at 6.25%) and sells a 5.25% Floor which is activated if 3 month LIBOR fixes at or below 3.30%”.

This meant that PAG would pay interest at no more than 6.25% however high the 3 month LIBOR rate went up and only 5.25% if the rate were between 5.25% and 6.25%. If and for so long as the 3 month LIBOR rate were between 3.30% and 5.25%, PAG would pay that rate. If, on the other hand, the 3 month LIBOR rate were to fall below 3.30%, PAG would find itself paying interest at 5.25%.

The transaction was documented in a post-transaction acknowledgment (or “PTA”) dated 7 October 2004 and a fuller confirmation dated 25 October. The PTA stated that there would be a “confirmation detailing the entire terms of our agreement relating to the transaction”, but itself contained notes that were stated to be “important”. One of these explained what PAG would pay overall if the notional amount of the swap (i.e. £10 million) were looked at together with borrowing of a corresponding sum. The note said:

“The cost to you of the overall structure is the sum of the cost of the borrowing and the net cost to you of the interest rate contract, whether this is a swap, cap, collar or any other interest rate hedging structure. This is illustrated below

You may have an interest rate swap under which you receive base rate and pay fixed. This is being used to protect interest rate risk on which you are paying base rate plus margin.

Your net pay/(receive) position under the swap is

Interest Rate Swap	Pay Receive	Fixed (Base rate)
Loan	Pay	Base rate + Margin
Net Pay		Fixed + Margin”

In other words, the net effect of the First Swap and a loan of £10 million at a margin above LIBOR would be that PAG would pay the interest rate given in the swap (say, 5.25% if the 3 month LIBOR rate were 6%) plus the margin over LIBOR (say, 1%). On the assumed facts, PAG would thus, in all, pay 6.25%.

The notes in the PTA pointed out that PAG would be exposed to interest rate risk if there were a mismatch between “the start dates of the underlying borrowing and any protection” and if there were “a difference between the value of the borrowing that is to be protected and the notional principal of your interest rate contract with us”. The notes also said:

“You [i.e. PAG] are acting for your own account, and will make an independent evaluation of the transactions described and their associated risks and seek independent financial advice if unclear about any aspect of the transaction or risks associated with it and you place no reliance on us [i.e. RBS] for advice or recommendations of any sort.”

The notes dealt, too, with what the position would be if PAG chose to bring the swap to an end before maturity. As to that, it was stated:

“If interest rate derivative contracts are closed before their maturity, breakage costs or benefits may be payable. The value of any break cost or benefit is the replacement cost of the contract and depends on factors on closeout that include the time left to maturity and current market conditions such as current and expected future interest rates. This is illustrated below.

There will be a break cost to you if the interest rates prevailing on closeout are lower than the fixed rate of the swap (that you are paying) or below the floor rate of the collar. There will be a benefit to you if prevailing interest rates are higher than the fixed rate of the swap (that you are paying) or above the cap rate of the collar.”

In the broadest terms, therefore, PAG would be liable to pay a break cost if interest rates declined but could stand to receive money if interest rates went up.

The confirmation formally recorded that RBS could terminate the structure at zero cost on 7 October 2009 and on each subsequent payment date. It also stated that each party made to the other certain representations, including these:

“(a)

Non-Reliance: It is acting for its own account, and it has made its own independent decisions to enter into this Transaction and as to whether this Transaction is appropriate or proper for it based upon its own judgement and upon advice from such advisers as it has deemed necessary. It is not relying, and has not relied, on any communication (written or oral) of the other party as investment advice or as a recommendation to enter into this transaction; it being understood that information and explanations related to the terms and conditions of this Transaction shall not be considered investment advice or a recommendation to enter into this Transaction, no communication (written or oral) received from the other party shall be deemed to be an assurance or guarantee as to the expected results of this Transaction.

(b)

Assessment and Understanding: It is capable of assessing the merits of and understanding (on its own behalf or through independent professional advice), and understands and accepts, the terms, conditions and risks of this Transaction. It is also capable of assuming, and assumes, the risks of this Transaction.”

RBS had been in contact with PAG about possible hedging arrangements for quite a number of months before the First Swap was concluded. On 5 October 2004 Mr Bescoby sent Mr Wyse a presentation entitled “Structured hedging Solution” outlining an arrangement along the lines of the First Swap. On the following day Mr Bescoby and Mr Wyse spoke several times on the telephone. During the first of these calls, which was made by Mr Bescoby, he reiterated that only RBS would be able to cancel the structure and agreed with Mr Wyse that break costs would be payable if PAG wished to break after five years. Later in the day, Mr Wyse called Mr Bescoby back to say that PAG would proceed if RBS could do the trade with a lower notional amount and at a smaller premium than hitherto proposed. The First Swap was subsequently agreed in a further conversation.

The second RBS swap (“the Second Swap”) was a “Libor Cancellable Discount Swap (Bank)” with a trade date of 25 September 2007 and a notional amount of £15 million for 4 years and then £30 million for a further six years. RBS was, however, to be entitled to cancel the swap after four years and annually thereafter. Here, PAG was to pay 5% interest and receive 3 month LIBOR. If, therefore, the swap were taken together with borrowing of £15 million at 1% above LIBOR, the net effect in the first four years would be that PAG would pay interest on its loan at a fixed rate of 6% (i.e. 5% under the swap plus the 1% margin over LIBOR on the loan). The transaction was documented in a PTA and confirmation containing the same standard notes and provisions.

The Second Swap had been preceded by a meeting attended by, among others, Mr Russell, Mr Wyse, Mr Malin, Mr Bescoby and Mr Goldrick on 18 September 2007. In the course of this meeting, which was to discuss hedging, Mr Wyse indicated that Mr Russell had a “target rate” of 5.00% in mind, and on 20 September Mr Bescoby sent Mr Wyse and Mr Malin an email headed “Structured hedging idea to achieve 5%!” Mr Wyse and Mr Bescoby discussed what was envisaged on the telephone, following which the former consulted Mr Russell and Mr Malin. On 21 September, Mr Wyse told Mr Bescoby in an email that Mr Russell’s instructions were to trade on the proposed basis when the rate dropped to 5%. On 25 September, market conditions having changed, RBS executed the trade.

The third RBS swap (“the Third Swap”) was a “Libor Collar” with a trade date of 14 January 2008 and a notional amount of £20 million. This was to last three years in the first instance, but RBS had the right to extend for a further two years. Once again, there were a PTA and a confirmation on the same lines. The PTA gave this summary of the terms:

“PAG Purchase a 5.25% Strike CAP

PAG Sells a 3.90% Strike FLOOR

If 3 month LIBOR fixes below 3.90% PAG pays the floor rate (3.90%) plus the difference between the floor strike (3.90%) and actual 3 month LIBOR fixing

3.90% + (3.90%-actual 3 month LIBOR fixing)

This is capped at 5.25%”

The final RBS swap (“the Fourth Swap”) was a “Switchable LIBOR to base rate callable swap” with a trade date of 16 April 2008 and a notional amount of £15 million. RBS could cancel after 12 months and quarterly thereafter, but subject to that the swap was to continue for five years. Unless and until the swap was cancelled, PAG was to pay 4.80% and receive 3 month LIBOR. After any cancellation, PAG would continue to receive 3 month LIBOR but pay “average base rate”. There were a PTA and a confirmation on the same lines as before.

Each of the Swaps was stated to supplement, form part of and be subject to an ISDA Master Agreement dated as of 7 October 2004 between PAG and RBS. The schedule to this agreement stated that each party would be deemed to make representations to the other to the effect of those set out in paragraph 15 above when it entered into a transaction governed by the agreement.

In 2010 RBS transferred its relationship with PAG from the bank’s management team in Manchester to its division in London known as the “Global Restructuring Group” (“GRG”). GRG instigated valuations of the properties of PAG over which RBS held security in both 2010 and 2013.

In 2011 PAG terminated the Swaps, incurring a break cost of £8.261 million. At the same time, PAG entered into a new facility agreement with RBS. Clause 21.5.1 of this provided as follows:

“The Lender [i.e. RBS] may, at any time, require the Valuer [i.e. Lambert Smith Hampton or such other valuer or surveyor as RBS might appoint] to prepare a Valuation of each Property [i.e. each of the properties over which RBS held security]. The Borrower [i.e. PAG] shall be liable to bear the cost of that valuation once in every 12 month period from the date of this Agreement or where a default is continuing.”

In 2014, having been made aware that RBS did not wish to refinance its borrowings, PAG sought funding elsewhere and secured a facility with HSBC. Repayment to RBS was completed on 25 July 2014.

RBS’s practice was to manage its risk exposure by itself selling into the market. When, therefore, it entered into each of the Swaps, it laid off its risk by hedging its open position in the market.

It was also the practice of RBS to make provision in its books for a potential default by a swap counterparty. For this purpose, it would calculate, using complex modelling, what it might come to be owed by the counterparty on a more or less worst-case scenario on the transaction being closed out. The figure would be represented in what is termed the “CLU” or, more fully, the “credit limit utilisation”.

The cost of (or benefit to be derived from) breaking a swap will be in line with its “mark-to market” value (or “MTM”) at the time. Where interest rates have moved in favour of a bank (i.e. by dropping, as happened in the present case), a swap may have a large negative MTM from the point of the counterparty. At the point, however, when each of the Swaps was entered into, its MTM (or “Day 1 MTM”) was modest: of the order of £42,000 for Swap 1, £110,000 for Swap 2, £91,500 for Swap 3 and £56,500 for Swap 4 (in favour of RBS). These figures will have reflected the fact that, at the date of the Swap, interest rates could either fall (so that RBS would be “in the money”) or rise (with the result that PAG would be “in the money”).

As is well known, interest rates tumbled following the global financial crisis of 2007-2008. Within a year of the Fourth Swap being entered into, base rate had been reduced to 0.5% and it has never since risen above this level. 3 month GBP LIBOR, to which the Swaps were tied, also went down dramatically and stayed low, with the result that the rates of interest that PAG was paying under the Swaps far exceeded what it was receiving under them. In December 2009, for example, 3 month GBP LIBOR was, as the Judge noted, no more than about 0.6%. One result of the prolonged period of unusually low interest rates was that the Swaps had a very large negative MTM from PAG’s point of view. The break cost that PAG incurred in 2011 was correspondingly substantial.

The present proceedings were issued in 2013. The relief sought included rescission of the Swaps and/or damages. PAG divided its claims into three categories: “the Swaps Claims”, which involved allegations of misrepresentation, misstatement and breach of contract on the part of RBS in connection with its proposal and sale of the Swaps to PAG; “the LIBOR Claims”, which rested on RBS’s knowledge of and participation in manipulation of LIBOR rates; and “the GRG Claims”, by which PAG complained of breaches of contract arising out of its transfer to, and subsequent management within, GRG.

Following a lengthy trial, Asplin J dismissed the proceedings in their entirety. Quite a number of the points on which PAG failed before the Judge are no longer pursued and so need not be considered further. The claims on which PAG still relies are, in broad summary, to the following effect:

With regard to the Negligent Misstatement Claim, the Judge concluded in paragraph 202 of her judgment that RBS had had no duty of care of the kind PAG contended for. She continued:

“203.

In any event, I have already found that PAG was aware of the potential for break costs which would vary according to market conditions and was also fully aware of the internal credit line necessary in relation to the Swaps. Further, it did not request any information about the extent of the MTM despite being aware of its existence. In my judgment therefore, not only was there no duty to reveal the extent of the break costs, the MTM at the outset or from time to time throughout the life of the Swaps but in any event, PAG did not enter into the Swaps as a result of the information having been withheld. I should add that to have any purpose or meaning it would have been necessary to provide information in relation to break costs on a regular basis, something which is not alleged.

204.

What of the alleged duty to provide scenario analysis as part of a duty to explain fully? It seems to me that in the light of the unchallenged evidence that such scenarios were not generally provided at the time, and the conclusions that I have reached in relation to the wider duty of care, that there is no breach in this regard. Was RBS in breach of the duty not to misstate by failing to provide such scenarios? In my judgment, it was not. The information which was provided was not inaccurate. I consider the position to be the same in relation to break costs and MTM.”

Turning to the Misrepresentation Claim, the Judge said (in paragraph 230 of the judgment):

“it seems to me that in the context to which I have referred, the reasonable representee would not have understood the references to ‘hedge’ in the way for which PAG contends. In those circumstances, including, in particular, the non-advisory relationship arising from express contractual terms, in my judgment a reasonable representee would have considered the term to be generic and would not have understood the phrases used as a representation as to the quality of the transaction upon which they could rely”.

The Judge commented (in paragraph 231) that that was “the more so because of the terms set out in the schedule to the ISDA Master Agreement and in each of the Confirmations/PTAs” and then said this in paragraph 232:

“Furthermore, and even if I am wrong about the nature of the ‘Hedge Representation’, and in addition, reference to a hedge should be construed in the narrow way put forward by Mr Virji [i.e. the expert called by PAG], I consider the evidence of Mr Wyse and Mr Russell to be fatal to their contention that they entered into the Swaps in reliance upon the Hedging Representations. In cross-examination, it was clear that neither gentleman understood a hedge in the way described by Mr Virji and relied upon for the purposes of this aspect of the claim. Accordingly, neither Mr Russell, PAG’s ultimate decision maker, nor Mr Wyse its Finance Director could have relied upon the ‘Hedge Representations’ in the manner alleged. The cancellable nature of the Swaps was also fully understood by PAG. This was revealed, for example, in the discussion between Mr Malin and Mr Bescoby on the telephone. Therefore, PAG cannot have understood the reference to a ‘hedge’ to mean as Mr Virji suggests and to have relied upon it in that way. Furthermore, it seems to me that the appetite and eagerness for derivatives shown by PAG which sought quotations on numerous occasions in relation to various structures, rates and amounts, both from RBS and other banks, also weighs against reliance upon a representation that the Swaps were ‘hedges’ in the narrow sense. In my judgment, Mr Russell’s conduct was much closer to speculation upon interest rates. The same reasoning applies in relation to the use of the word ‘protect’.”

As for the LIBOR Claims, the Judge rejected these on the basis that the alleged representations were not in fact made (paragraphs 404-413 of the judgment), that PAG did not in any case rely on them (paragraphs 417-419), that no manipulation of sterling LIBOR had been proved (paragraphs 453-475) and that neither had it been established that RBS intended PAG to rely on the alleged representations (paragraph 485).

So far as the Valuation Claim is concerned, the Judge considered that the right to call for a valuation conferred on RBS by, for example, clause 21.5.1 of the 2011 facility agreement was not fettered by any implied term (paragraphs 275-278 of the judgment) and that, even were that not so, there had been no breach of any such term (see paragraphs 302 and 308).

PAG appeals against the Judge’s decision on the Negligent Misstatement, Misrepresentation, LIBOR and Valuation Claims. The grounds of appeal involve both points of law and challenges to factual findings. We will not detail the grounds here, but will consider the issues that arise as we come to each claim.

The Negligent Misstatement Claim, the Misrepresentation Claim, the LIBOR Claims and the Valuation Claim are addressed in turn below.

During the course of the hearing of the appeal PAG’s case on negligent misstatement became refined and clearer. In summary, PAG submits that RBS was in breach of duty in tort in failing to provide PAG prior to the making of each of the swap contracts with either RBS’s internal estimation of the cost to PAG of breaking the swaps during their life on a “worst case” basis or worked examples of how much the break costs could be in different scenarios, including if interest rates moved significantly lower, that is to say against PAG (“worked break cost scenarios”). RBS’s internal estimation of a “worst case” basis was the same as RBS’s internal credit line utilisation figure, the CLU, as described above. The CLU was £2.25 million on the date the First Swap was agreed. It is to be distinguished from the MTM, which is the same as the cost or benefit of breaking the swap contract on any given day.

The CLU, the MTM and the break costs are to be distinguished from the financial consequences of the exercise of RBS’s right of cancellation. Nothing is payable either way if that is done. It is true that RBS will cancel if it is “out of the money”, and that will deprive PAG of any upside in the remaining years of the relevant swap – but that is not the complaint in these proceedings.

PAG’s case on negligent misstatement is substantially based on the judgment of Mance J in Bankers Trust International plc v PT Dharma Sakti Sejahtera [1996] CLC 518, in which the defendant (“DSS”) alleged that the claimant bank (“BTI”) was liable on various grounds, including misrepresentation and misstatement, in failing to give a full and proper explanation in connection with the decision of DSS to replace an existing swap with another. PAG relies, in particular, on the following passage (at page 533):

“I have mentioned that the existence of a duty of care does not depend upon the existence of any misrepresentation justifying rescission, and that the duty alleged by DSS extends to explaining fully and properly to DSS the operation, terms, meaning and effect of the proposed swaps and the risks and financial consequences of accepting them. The allegations go wider than those of misrepresentation and collateral undertaking. The principle on which DSS founds itself here is contained in cases such as Barclays Bank plc v Khaira [1992] 1 WLR 623, Cornish v Midland Bank plc [1985] 3 All ER 513 and Box v Midland Bank Ltd [1979] 2 Ll Rep 391. In short, a bank negotiating and contracting with another party owes in the first instance no duty to explain the nature or effect of the proposed arrangement to that other party. However, if the bank does give an explanation or tender advice, then it owes a duty to give that explanation or tender that advice fully, accurately and properly. How far that duty goes must once again depend on the precise nature of the circumstances and of the explanation or advice which is tendered. Mr Milligan accepted that BTCo and BTI did in the present case owe a duty to take reasonable care not to misstate facts in any of the relevant meetings or letters. DSS alleges that explanations and advice were tendered which went beyond the mere statement of facts, and that BTCo and BTI owed correspondingly broader duties.”

The claim failed on the facts in the Bankers Trust case but PAG relies both on that general statement of principle by Mance J and on his analysis of the application of the law to the facts.

PAG accepts that RBS was never under a general advisory duty to PAG in relation to the Swaps. PAG’s case is that RBS, by failing to disclose to PAG its estimate of the potential cost of breaking the Swaps during their life or to provide worked break cost scenarios, presented an inaccurate and incomplete explanation of each proposed swap

PAG submits that this rendered RBS liable for negligent misstatement in accordance with the classic statement of principle in Hedley Byrne & Co Ltd v Heller & Partners Ltd [1964] AC 465 as analysed and applied in the Bankers Trust case.

Very much as a second and subsidiary line of argument, Mr Tim Lord QC, who appeared for PAG with Mr Adam Cloherty and Mr Ben Woolgar, contended that, if the failure to disclose the CLU figure or to provide worked break cost scenarios falls outside the Hedley Byrne principle, it nevertheless was a breach of a duty at common law to take reasonable care when providing information to ensure that such information is both accurate and fit for the purpose for which it is provided to enable the recipient to make a decision on an informed basis. That was what counsel for the claimant in Crestsign Ltd v National Westminster Bank plc [2014] EWHC 3043, [2015] 2 All ER 133, another swaps case, called a “mezzanine” duty and has also been called an “intermediate duty”, which was described by the deputy High Court Judge in that case as less onerous than a wide duty to give advice but wider than the duty not to misstate: Crestsign at paragraphs 135-136. Again, the claimant failed on the facts in Crestsign.

In reaching her conclusion that there was no breach of duty of care by RBS, the Judge relied on the following matters in paragraphs 199 to 204 of her judgment.

In paragraph 199 she said that it seemed to her that, although PAG was not of the calibre of the counterparty in Bankers Trust, neither was it as unsophisticated as the party under consideration in Crestsign.

In paragraph 200 she said that at all material times PAG had a series of banking advisers who may not have been derivatives specialists or able to compute potential break costs or MTM but they were aware of the potential for such costs to rise and capable of considering the potential consequences of the terms being proposed and of pointing PAG in the direction of those able to calculate MTM/break costs.

In paragraph 201 she made a series of observations. She said that it was not general market practice to give information about potential break costs and the MTM at the outset at the time. She also took into account the specific warnings about break costs, mismatch and other matters which were first voiced by Rathbones in September 2002 and were mentioned in the first RBS paper provided to Mr Russell that month and repeated in the presentations of June and October 2004 entitled “Structured Hedging Solution”, the specific reference to such costs being “substantial” and the warnings in each of the PTAs. She also considered it relevant that PAG was made aware of a hedging credit line on numerous occasions. Further the CLU figure was produced for internal purposes on a worst-case scenario and could not be expected to have been revealed. She said that it was also relevant that it was not envisaged that it would be necessary to break the Swaps because it was expected that PAG’s borrowing would continue to increase and the unprecedented drop in interest rates could not have been forecast.

In paragraph 202 she said the fact that the explanations given were more extensive than in Bankers Trust militated against the existence of the duty of care rather than for it and the question of whether those explanations were misleading could best be dealt with “under the pure misstatement heading”. She said that it was also important to bear in mind that any duty to advise had been expressly excluded by the terms of the parties’ contractual arrangements and was repeated in the PTAs and that the relationship between the parties was essentially commercial.

She concluded in paragraph 202 that, taking into account foreseeability, proximity and fairness, justice and reasonableness in the context of the PAG/RBS relationship and in the light of the type of loss which it is sought to recover, there was no duty of care of the kind contended for.

In paragraph 203 the Judge held that, in any event, PAG did not enter into the Swaps as a result of the information having been withheld. In that context, she observed that PAG was aware of the potential for break costs which would vary according to market conditions; PAG was also fully aware of the internal credit line in relation to the Swaps; and PAG did not request any information about the extent of the MTM despite being aware of its existence.

In paragraph 204 the Judge addressed specifically the alleged duty to provide “scenario analysis” as part of a duty to explain fully. She said that it seemed to her that, in the light of the unchallenged evidence that such scenarios were not generally provided at the time, and the conclusions that she had reached in relation to the wider duty of care, there was no breach in that regard. She added that RBS was not in breach of the duty not to misstate by failing to provide such scenarios since the information which was provided was not inaccurate. She said she considered the position to be the same in relation to break costs and MTM.

Mr Lord submitted that it is clear that the Judge fundamentally misunderstood that PAG’s primary case is that RBS is liable for breach of the classic Hedley Byrne duty not carelessly to misstate, by giving only a partial and so misleading and inaccurate explanation. He said that she failed to appreciate that PAG’s case in that respect was correctly based on the statement of principle by Mance J in Bankers Trust at page 533 that “if [a] bank does give an explanation or tender advice, then it owes a duty to give that explanation or tender that advice fully, accurately and properly.”

It is true that the Judge did not approach her analysis on the footing that PAG’s primary case was based on breach of the Hedley Byrne common law duty not carelessly to misstate. As was pointed out by Mr Richard Handyside QC, who appeared for RBS with Mr Adam Sher and Mr Laurie Brock, that may have been due to the fact that the Particulars of Claim did not expressly allege that RBS’s explanations of the proposed swaps were inaccurate because they failed to disclose the potential size of future break costs but they did allege a general duty to ensure that information and explanations which RBS provided were full, accurate and proper, sufficient to enable PAG to understand fully the nature and effect of the transaction and the risks associated with it and to make a properly informed decision about whether to enter into the transaction. The same approach was advanced in PAG’s closing written submissions at the trial, which made no mention of Hedley Byrne.

Furthermore, the Judge appears to have thought that PAG’s complaint related, at least in part, to the failure of RBS to disclose the MTM and break costs as at the first day of each of the Swaps. It became clear during the course of the hearing of the appeal that the focus of PAG was not on that omission but rather on the failure to disclose the CLU figure or at least worked break cost scenarios if the Swap was broken in the future. The MTM mirrors the cost or benefit of breaking the swap on a given day. As is to be expected the MTM was modest at the inception of each Swap; the figures are given in paragraph 27 above. It is not suggested by PAG that the failure to disclose any of those sums was of any materiality to the decision of PAG to take up the Swaps. The complaint is about the failure to disclose the CLU figure, which was the product of complex modelling with sophisticated computer software and gave RBS a potential “worst case” cost to PAG of breaking the Swaps at any time during the lifetime of the Swaps, or at least worked break cost scenarios.

We also doubt whether the Judge was correct (in paragraph 202 of her judgment) to place weight on the fact that the explanations given by RBS to PAG were more extensive than in Bankers Trust and to say that this militated against the existence of the duty of care rather than for it.

Nevertheless, the Judge did expressly address (in paragraph 204 of her judgment) an alleged duty to provide scenario analysis; and her reasoning in paragraphs 199 to 203 was apposite in relation to PAG’s secondary case that, if necessary, there was breach of a wider duty of care than the Hedley Byrne duty not carelessly to misstate. We consider that the Judge was correct to reject both the allegation of breach by RBS of the Hedley Byrne duty and the existence of a wider duty that would have included a duty to disclose the CLU figure.

In order to understand the passage in Mance J’s judgment in Bankers Trust (at page 533) quoted above, it is necessary to place it in its jurisprudential context.

In Hedley Byrne the question was whether bankers could be held liable in tort in respect of the gratuitous provision of a negligently favourable reference for one of their customers, when they knew or ought to have known that the plaintiff would rely on their skill and judgement in furnishing the reference, and the plaintiff in fact relied upon it and in consequence suffered financial loss. The House of Lords held that, in principle, an action would lie in such circumstances in tort; but that, in the particular case, a duty of care was negatived by a disclaimer of responsibility under cover of which the reference was supplied. For present purposes, the importance of the case is that it identified that relationships might give rise to a duty to take care, not limited to contractual relationships or relationships of fiduciary duty, where there is an assumption of responsibility.

In Henderson v Merrett Syndicates Ltd [1995] 2 AC 145, Lord Goff said (at page 180C) that the principle of assumption of responsibility underlying Hedley Byrne rests upon a relationship between the parties, which may be general or specific to the particular transaction, and which may or may not be contractual in nature. Lord Goff also stated that the assumption of responsibility test is to be applied objectively (at page 181B).

Subsequent cases have established that other tests may also be used in deciding whether a defendant sued as causing pure economic loss to a claimant owed a duty of care in tort. In all, three tests have been identified. They were summarised by Lord Bingham in Customs and Excise Commissioners v Barclays Bank plc [2006] UKHL 28, [2007] 1 AC 181, at paragraph 4 as follows:

“The first is whether the defendant assumed responsibility for what he said and did vis-à-vis the claimant, or is to be treated by the law as having done so. The second is commonly known as the threefold test [in Caparo Industries plc v Dickman [1990] 2 AC 605]: whether loss to the claimant was a reasonably foreseeable consequence of what the defendant did or failed to do; whether the relationship between the parties was one of sufficient proximity; and whether in all the circumstances it is fair, just and reasonable to impose a duty of care on the defendant towards the claimant (what Kirby J in Perre v Apand Pty Ltd (1999) 198 CLR 180, para 259, succinctly labelled “policy”). Third is the incremental test, based on the observation of Brennan J in Sutherland Shire Council v Heyman (1985) 157 CLR 424, 481, approved by Lord Bridge of Harwich in Caparo Industries plc v Dickman [1990] 2 AC 605, 618, that:

“It is preferable, in my view, that the law should develop novel categories of negligence incrementally and by analogy with established categories, rather than by a massive extension of a prima facie duty of care restrained only by indefinable ‘considerations which ought to negative, or to reduce or limit the scope of the duty or the class of person to whom it is owed’.””

Lord Bingham expressed the view that the case law enumerating those three tests was not readily reconcilable and that it is preferable to take each of these tests into account when considering whether a duty arose. Lord Bingham said (at paragraph 8):

“… the outcomes (or majority outcomes) of the leading cases cited above are in every or almost every instance sensible and just, irrespective of the test applied to achieve that outcome. This is not to disparage the value of and need for a test of liability in tortious negligence, which any law of tort must propound if it is not to become a morass of single instances. But it does in my opinion concentrate attention on the detailed circumstances of the particular case and the particular relationship between the parties in the context of their legal and factual situation as a whole.”

It is clear that parts of the threefold test and the assumption of responsibility test overlap: Chandler v Cape plc [2012] EWCA Civ 525, [2012] 1 WLR 3111, at paragraph 62 (Arden LJ). The different tests usually lead to the same answer and can be used as cross-checks on each other: Playboy Club London Ltd v Banca Nazionale del Lavoro SpA [2016] EWCA Civ 457, [2016] 1 WLR 3169, at paragraph 17. They are complementary and should not be considered in isolation from each other: CGL Group Ltd v Royal Bank of Scotland plc [2017] EWCA Civ 1073, [2017] CTLC 97; cf Robinson v Chief Constable of West Yorkshire Police [2018] UKSC 4.

The Hedley Byrne common law duty of care not to misstate is, then, merely one example of a more general principle that a defendant’s assumption of responsibility may give rise to a duty of care – giving rise to pure economic loss - either in relation to a particular transaction or a continuing relationship, the existence of the duty and its extent being dependent on the particular facts.

The Hedley Byrne duty has to be seen in the context of the general principle of tort that an omission does not usually give rise to a liability in tort. All other things being equal, there is no duty to speak. If, however, a defendant does speak, they fall under a duty not to be dishonest or fraudulent in what they say: Derry v Peek (1889) 14 App Cas 337. In certain factual circumstances, such as those outlined in the speeches in Hedley Byrne, the position of the defendant in relation to the claimant, combined with the defendant’s conduct or omissions, may give rise to an assumption of responsibility and the imposition of a tortious duty. At its most basic, this is a duty not carelessly to make a misstatement. What amounts to a misstatement in this context will depend upon the factual circumstances of the relationship and identification of the matter for which the defendant has assumed responsibility. It is, therefore, an elastic duty that is factually sensitive. The duty is premised on the voluntary proffering of representations by the defendant, which may require further elucidation or the correction of misleading impressions on the claimant.

In some exceptional cases, a defendant may assume a responsibility to speak. In Cornish v Midland Bank plc [1985] 3 All ER 513, the Court of Appeal held that the explanation given by the defendant bank to the plaintiff about a second mortgage negligently misstated the position in breach of the Hedley Byrne duty. Kerr LJ expressed the view that, had it been necessary to decide the issue, he would have inclined to the view that in the circumstances of the case the bank owed a duty to the plaintiff, as the bank’s customer, to proffer to her some adequate explanation of the nature and effect of the document which she had come to sign. In a number of cases, this has been characterised as a broad distinction between a Hedley Byrne duty not to mislead and a more general advisory duty.

The passage in the judgment of Mance J in Bankers Trust, quoted above, on which PAG places such heavy reliance, must be seen in this general jurisprudential context. The starting point is, as Mance J said, that a bank negotiating and contracting with another party owes in the first instance no duty to explain the nature or effect of the proposed arrangement to that other party. The statement of Mance J that, if a bank does give an explanation or tender advice, then it owes a duty to give that explanation or tender that advice fully, accurately and properly, covers a range of possible factual situations. It is rightly qualified by his statement, which immediately follows, that “[h]ow far that duty goes must once again depend on the precise nature of the circumstances and of the explanation or advice which is tendered”. In Bankers Trust itself the factual context was that the bank put forward an explanation that entering into the proposed substitute swap would improve the risk exposure of the customer. It was against that factual background that Mance J held (at page 573H) that the duty not carelessly to misstate facts obliged the bank to present the financial implications of the proposal by a properly constructed graph and letter, which presented the downside and upside of the proposal in a balanced fashion. Those are not the facts of the present case.

The expression “mezzanine” duty or intermediate duty, first coined in Crestsign, is best avoided. It appears to reflect the notion that there is a continuous spectrum of duty, stretching from not misleading, at one end, to full advice, at the other end. Rather, concentration should be on the responsibility assumed in the particular factual context as regards the particular transaction or relationship in issue. The observation of Tim Kerr QC, siting as a deputy High Court Judge, in Crestsign (at paragraph 155) that the bank’s duty would extend to correcting any obvious misunderstandings communicated by the customer and answering any reasonable questions the customer might ask about those products in respect of which the bank had chosen to volunteer information might, depending on the particular factual context, be consistent with the standard Hedley Byrne duty not to misstate, including by omission.

Turning to the present case, it is common ground that, in explaining the terms of the proposed swaps to PAG, RBS was under a Hedley Byrne duty not to misstate. In that connection, Mr Lord took us through a detailed analysis of the transcripts of a telephone conversation between Mr Bescoby and Mr Malin (PAG’s in house banking adviser) on 27 September 2007 and a telephone conversation between Mr Bescoby and Mr Wyse on 5 November 2007. PAG draws attention to those transcripts as showing that there were specific endorsements by Mr Bescoby of the merits of the swaps under consideration in those conversations and an underplay of the disadvantages, including a complete omission to mention the potential size of any break cost to PAG in the future.

Mr Handyside countered that there was no specific allegation of misstatement in the Particulars of Claim in relation to those two telephone conversations.

The Judge did not directly address the opposing submissions of the parties on the two telephone conversations. She had no need to do so since neither of the conversations concerned the specific four Swaps in issue in these proceedings. The earlier of the conversations related to, but took place after, the Second Swap had been completed and was unrelated to the Third Swap. The later conversation related to a transaction that never took place.

In any event, PAG does not rely upon those conversations as containing any error in what was expressly and positively stated. It is not suggested by PAG that any of the formal transaction documents for the four Swaps or any communications by RBS to PAG in connection with the Swaps contained information that was inaccurate. Rather, its case is that, in relation to the entire course of dealings between RBS and PAG concerning the Swaps, RBS was in breach of the Hedley Byrne duty by failing to present a full and proper explanation since it omitted to disclose the potential size of the cost to PAG of breaking the Swaps in the future, and in particular in failing to disclose the CLU or at least to present PAG with worked break cost scenarios.

We do not agree that there was any such breach of duty. It is clear from the documentation and other evidence at the trial that PAG was made fully aware that (1) breaking any of the Swaps could carry adverse financial consequences, (2) the size of those financial consequences would depend upon interest rates at the time the Swaps were broken, and (3) the precise calculation of any amount to be paid by PAG would take into account the extent to which, if at all, the floating-rate payable by RBS under the Swaps was lower than the fixed interest payable by PAG.

A presentation document titled “Structured hedging Solution” prepared by RBS for PAG dated 5 October 2004 set out clearly the methodology for calculating break costs as follows:

“If interest rate derivative contracts are closed before their maturity, breakage costs or benefits may be payable. The value of any break cost or benefit is the replacement cost of the contract and depends on factors on closeout that include the time left to maturity and current market conditions such as current and expected future interest rates. This is illustrated below.

There will be a break cost to you if the interest rates prevailing on closeout are lower than the fixed rate of the swap (that you are paying) or below the floor rate of the collar. There will be a benefit to you if prevailing interest rates are higher than the fixed rate of the swap (that you are paying) or above the cape rate of the collar.”

There was an identical statement in Note 8 of the PTA relating to the First Swap.

In short, there was no error in the way that RBS explained the terms of the Swaps, including the circumstances in which break costs might be incurred and how they would be calculated.

Mr Lord emphasised that no reason has been given for RBS’s practice at that time of not offering information about potential future break costs. He placed weight on the fact that RBS’s policy has since changed and it would now provide worked break cost scenarios. He said that Ms Georgina Robbins, RBS’s banking expert witness, said in her oral evidence that customers had been caught out by the size of break costs. None of those matters, however, substantiates the existence of a duty on the part of RBS to disclose its estimate of the potential size of future break costs.

If RBS was under a duty to disclose the possible or probable size of future break costs at any time during the lifetime of the Swaps, that could only have arisen under one or more of the three tests for tortious liability summarised by Lord Bingham in Customs and Excise Commissioners v Barclays Bank plc. None of them are satisfied in the present case.

For the following reasons we can see no proper basis for holding, on the facts, that there was any assumption of responsibility for the disclosure by RBS of the CLU or any similar indication of the possible size of future break costs or for holding that it would be fair, just or reasonable to impose on RBS an advisory duty requiring such disclosure.

In a number of first instance decisions on swap transactions between a bank and its customer it was observed that it was not the normal practice to disclose the CLU or similar predictions and it was held that there was no breach of duty by the bank in failing to disclose them: Bankers Trust, Crestsign, Thornbridge Ltd v Barclays Bank plc [2015] EWHC 3430 (QB), Marz Ltd v Royal Bank of Scotland plc [2017] EWHC 3618 (Ch), London Executive Aviation Ltd v The Royal Bank of Scotland plc [2018] EWHC 74 (Ch). Although there is now greater disclosure by banks in relation to break costs than before, RBS still does not disclose the CLU.

The CLU is the product of the subjective view of RBS about many matters, including possible movements in interest rates in the future and the length of the outstanding term of the swaps at the time of the break, and involves a complex computer programme into which is fed a large number of different scenarios. It is an internal and subjective assessment by RBS of risk inherent in the swaps. Whether or not PAG and its advisers had the sophistication and IT facility to carry out a similar exercise, based on their own predictions of possible future movements in interest rates over the period of the Swaps, is not to the point.

Any worked break cost scenarios, intended to show what the break costs might be at any particular moment during the lifetime of the Swaps, would similarly be based on RBS’s own subjective opinion of what might happen to interest rates in the future and would not necessarily reflect RBS’s view of the degree of likelihood of the scenario actually occurring. Insofar as it is suggested that the break cost scenarios ought to have been provided merely to illustrate how the break cost methodology would work, whatever future interest rates might be, the methodology was clearly stated in the material given to PAG and it is not suggested that PAG or its advisers could not have provided worked up examples themselves.

Moreover, under the standard terms of the ISDA Master Agreement, as well as the express terms of each individual swap contract, PAG represented that it understood and accepted the risks of the transaction and was capable of assuming, and assumed, those risks. Further, in so far as the CLU reflected the suitability of the Swaps, it is relevant that PAG does not appeal the Judge’s rejection of PAG’s allegation of misrepresentation as to suitability.

Quite apart from the issue of duty of care and breach, we see no basis for interfering with the Judge’s conclusion that, in any event, PAG did not enter into the Swaps as a result of non-disclosure of the CLU or worked break cost scenarios. That was a finding of fact which, for the following reasons, she was plainly entitled to reach.

The Judge found, and, despite a challenge of PAG to the contrary, she was entitled on the evidence to find, that PAG was aware of the existence of the internal RBS credit line related to the Swaps. As explained above, PAG was fully aware of the methodology for calculating the break cost or benefit on any particular day. PAG was aware that such costs might be substantial. That was made clear in a report by Rathbones to PAG dated 16 January 2003.

It was also made clear in a document of RBS dated 6 February 2003, which addressed PAG’s future hedging requirements, and which advised that over-the-counter derivatives, including swaps:

“can provide significant benefits but may involve a variety of significant risks … In the event the market has moved a transaction you have undertaken, you may incur substantial costs if you wish to close out your position … Entering into OTC derivatives can introduce significant liquidity risk and other risk factors of a complex character.”

Yet, PAG never requested any worked break cost scenarios or information from RBS about the amount of the CLU, that is to say the amount which RBS had entered in its books in respect of the possible future break costs payable by PAG. The two telephone conversations mentioned above show that its commercial focus was on the relative advantage of the proposed fixed interest rate under the Swaps in the light of interest rates generally in the market at the time, while recognising that PAG might be tied into the Swap for an extensive period when interest rates might move against PAG by going down.

We understood from the oral submissions of Mr Lord that the only alleged misrepresentation which is in issue on the appeal is a misrepresentation in relation to each of the Swaps that it was a “hedge”. For the avoidance of any doubt, if and insofar as it is a live issue on the appeal whether or not RBS made a misrepresentation in relation to each Swap by omitting to give PAG details of the CLU, PAG fails on that issue for the same reasons as we have given for dismissing PAG’s appeal on misstatement.

On the alleged hedge misrepresentation, PAG’s case is that the Swaps did not act as a hedge against the risk of adverse movements in interest rates because they did not overall reduce the interest rate risk exposure. The Swaps plainly did act as a hedge against an increase in rates during the guaranteed period of the Swaps but PAG’s argument is that the effect of the potentially substantial cost of breaking the Swaps if interest rates dropped and of RBS’s right of cancellation is that, overall, they were not properly speaking protection against adverse movements in interest rates. Furthermore, those factors combined with the fact the Swaps lasted beyond the period of the relevant loan deprived the Swaps of the character of a hedge in the ordinary meaning of that word. In that connection, Mr Lord referred to the definition of “hedge” in the Oxford Dictionary of English (3^rd ed) as “a way of protecting oneself against financial loss or other adverse circumstances”.

Mr Lord observed that this is entirely consistent with the opinion of Mr Hanif Virji, PAG’s swaps expert, that an “interest rate hedge” is “a product which if transacted mitigates the adverse consequences of changes in interest rates” and “will reduce the risk of loss should interest rates change.” He pointed out that Mr Bescoby had himself accepted, in cross–examination, Mr Virji’s definition of a hedge.

We consider that, on the evidence and in the factual context in which the expression “hedge” was used by RBS and those acting on its behalf, the Judge was entitled to reach her conclusion (in paragraph 230 of her judgment) that the reasonable representee would not have understood that expression in the way for which PAG contends.

A critical feature of the factual context is that under each loan facility agreement RBS required PAG to enter into and maintain an interest rate hedging agreement acceptable to RBS. The purpose of such a hedging agreement was plainly to ensure that PAG would be protected against increases in interest rates which might otherwise undermine PAG’s ability to pay the interest due on its outstanding loans from RBS.

That the focus was on possible increases rather than any reduction in interest rates is reflected in PAG’s pleaded case and the evidence of its principal witnesses. Paragraph 41 (3) of the Particulars of Claim alleges that:

“the Swaps would hedge PAG against its exposure to interest rate rises under the Facilities (i.e. they would protect PAG from rises in interest rates above the relevant level in each of the Swaps).”

The witness statements of Mr Russell and Mr Wyse reflected that pleaded case. In paragraph 93 of his 3^rd witness statement Mr Russell said that, in agreeing to enter into the Swaps, he placed particular weight and reliance on representations by RBS to the effect that the Swaps would “hedge” PAG against its exposure to interest rate rises under the loan facilities, and that the interest rate products recommended by RBS would “protect” PAG against and “de-risk” the effect of rising interest rates under PAG’s borrowing.

In paragraph 186 of Mr Wyse’s 1^st witness statement he said that RBS:

“repeatedly stated that … the Swaps would provide “hedging” or “protection” or a “solution” to [PAG’s] interest rate risk under its facilities in the sense that they would protect [PAG] from rises in interest rates above the relevant levels in each of the Swaps.”

Furthermore, PAG was always fully aware that the Swaps extended beyond the period of the relevant loans. It was also fully aware of RBS’s cancellation rights under the Swaps and that, if the Swaps became commercially unattractive to RBS as a result of interest rate rises, RBS would exercise its cancellation rights. That is clear not only from the letter written by Rathbones to Mr Wyse on 26 October 2004 in relation to the First Swap but also the transcripts of the telephone conversation between Mr Bescoby and Mr Malin on 27 September 2007 and between Mr Bescoby and Mr Wyse on 5 November 2007 mentioned above. Similarly, PAG was well aware of RBS’s right to extend the Third Swap.

In those circumstances, even though Mr Bescoby accepted in cross-examination Mr Virji’s definition of a hedge, the Judge was fully justified in concluding (in paragraph 230 of her judgment) that a reasonable representee would not have understood RBS to have used the word “hedge” with that meaning in its dealings with PAG in relation to the four Swaps and so was not a misrepresentation.

In any event, the Judge also concluded (in paragraph 232 of her judgment) that PAG did not enter into the Swaps in reliance upon the hedging representations as defined by Mr Virji. That was a finding of fact based on the Judge’s view that the evidence showed that neither Mr Russell nor Mr Wyse understood a hedge in the way described by Mr Virji and relied upon by PAG. It is certainly supported by the passages in the witness statements of Mr Russell and Mr Wyse, to which we have referred above. It is impossible to say that the Judge was not entitled to reach her factual conclusion of non-reliance.

In these circumstances, it is not necessary to address PAG’s challenge to the Judge’s finding (in paragraph 231 of her judgment) that, absent fraud or recklessness, the non-reliance provisions of the ISDA Master Agreement and of the swap contracts preclude any cause of action for misrepresentation in respect of the expression “hedge”. On that issue, Mr Lord contended that, consistently with the judgment of Hamblen J in Casa di Risparmio della Repubblica di San Marino SpA v Barclays Bank Ltd [2011] EWHC 1785 (Comm), [2011] 1 CLC 701, at paragraphs 513-518, the non-reliance provisions applicable to the Swaps have no application. He conceded, however, that, on the face of it, Hamblen J’s judgment at paragraphs 525-526 is also consistent with a contractual estoppel - barring a non-fraudulent hedging misrepresentation claim arising as a result of the assumption of risk provisions in the ISDA Master Agreement and the swap contracts. He contended that Hamblen J was wrong on that point. There is, however, no respondent’s notice in this respect. We, therefore, heard no substantive argument on that issue.

In view of our conclusion about the meaning of “hedge” in the context of the facts of this case, it is also strictly unnecessary to address the challenge in RBS’s respondent’s notice to the Judge’s finding (in paragraph 223 of her judgment) that, if the expression “hedge” was to be understood in the sense for which PAG contended, RBS’s misrepresentation about the Swaps being a hedge was fraudulent. That finding followed from her conclusion that from at least the date of an internal email of 3 October 2007 from Mr Bescoby to RBS’s “North Property Team”, which was read by both Mr Jones and Mr Goldrick, all three of them knew that any representation the Swaps were hedges (in the sense for which PAG contended) was not true or at least they were reckless as to that. The Judge also relied on the evidence of Mr Russell and Mr Wyse, to which she referred in paragraph 224 of her judgment, that Mr Jones had told them on a train journey in 2010 that a credit committee member had asked Mr Jones whether his clients realised that they did not have a hedge and were not protected by the Swaps.

Although not strictly necessary, we shall address RBS’s challenge to the Judge’s finding of fraud in view of the seriousness of that finding and its implications for the reputations of RBS, Mr Bescoby, Mr Jones and Mr Goldrick.

We agree with RBS that, on a close examination of the evidence, the Judge was not entitled to make the finding of fraud. The evidence is consistent and clear that each of Mr Bescoby, Mr Jones and Mr Goldrick considered that the Swaps were hedges in that they did effectively hedge during the guaranteed or pre-cancellation period.

The internal email of 3 October 2007 set out Mr Bescoby’s critical views about an article in Property Week. The Judge (in paragraph 91 of her judgment) quoted the following passage in that email:

“In short callable structures are just one of a number of solutions which GBM can provide to clients as part of an overall risk management strategy. They are not a hedge but can satisfy a desire for lower rate funding for a pre-determined period …”

It would appear that it was that passage that led the Judge to conclude (in paragraphs 223 and 230 of her judgment) that, from at least the date of the email, Mr Bescoby, Mr Jones and Mr Goldrick knew that any representation that the Swaps were hedges was not true or at least they were reckless as to that.

The email, however, contained other passages which clearly indicate, on a fair reading of the email as a whole, that Mr Bescoby did regard derivatives like the Swaps as hedges. In the first place, Mr Bescoby began his email by describing the article as “a one-sided view of callable structures and their use within a structured hedging portfolio”. Secondly, he said that “it is typical that the guaranteed period of a callable will be set to satisfy the hedging condition (i.e. a three year condition of sanction would result in a 5 year no call 3 year trade so that GBM can only call after the 3 year period)”. Thirdly, the passage quoted by the Judge continues: “in return the user is comfortable that they may … be unhedged after the guaranteed period.”

In paragraph 82 of his 1^st witness statement Mr Bescoby said, with reference to that email:

“I note that at the end of my response I stated that such products were “not a hedge”. By this I only meant that they do not provide a guaranteed hedge against interest rate risk for the full life of the trade.”

Mr Bescoby was not challenged on that statement in cross-examination. Nor was he asked any questions in cross-examination on the passage in the email quoted by the Judge. He did say in cross-examination, with regard to the Second Swap, that it was a hedge for four of the ten years and that it was sold as a hedge to provide hedging on the loan facility. He also said that, where a transaction is only guaranteed for three months, it is not a hedge but “beyond that, as I say, for 12 months or whatever it might be, the client could assume that it was a hedge”.

Mr Jones said in cross-examination that he had reasonable grounds for saying that the First Swap was a hedge because it provided protection for the first five years. He said, with regard to the Second Swap, that it “would be referenced as a hedge and treated as a hedge for the four years of the fixed period”. He was asked whether the internal email of 3 October 2007 caused him to revise his view of any of the Swaps. He answered that it did not because the Swaps “are still providing … the no-call period, which would be the period required to sit alongside the loan facility and satisfy the credit sanction”. In relation to the Fourth Swap, his evidence in cross-examination was again that it was a hedge during the fixed term of the loan facility.

Mr Jones’s evidence with regard to the Fourth Swap was consistent with an email sent on 6 May 2008 by Mr Walter Logan of RBS’s corporate credit section to Mr Bescoby, and copied to Mr Jones, in which Mr Logan stated that “assuming you have the ability to cancel after 12 months, we will only look at this as a 12 month hedge from a risk perspective”.

Mr Goldrick was asked questions about the internal email of 3 October 2007 in cross-examination. He rejected the suggestion that the type of structured product, exemplified by the Second Swap, could not be described as a hedge. He said that he thought that what Mr Bescoby had written was “a bit ambiguous” but that the inference was that the first part of the callable was hedged. He said, with specific reference to the Second Swap, that he thought that the first part was a hedge.

It is unclear what weight, if any, the Judge was placing on Mr Jones’s comment to Mr Russell and Mr Wyse on the train journey in July 2010 that a credit committee member had asked Mr Jones whether his clients realised that they did not have a hedge and were not protected by the Swaps. In the light of the clear and consistent evidence of Mr Bescoby, Mr Jones and Mr Goldrick that they considered the Swaps were hedges, and the evidence that Mr Logan of the credit section regarded the Swaps as a hedge during the guaranteed period, and the absence of any evidence as to whether the credit committee member, to whom Mr Jones was referring, was talking about the period before or after the guaranteed period or both, we can see no reason for attributing any particular significance to the train conversation.

In the absence of evidence sufficient to justify a finding that one or more of Mr Bescoby, Mr Jones and Mr Goldrick thought that “hedge” in relation to the Swaps bore Mr Virji’s meaning and intended that references by them to “hedge” should be understood to have that meaning or were reckless as to whether such references would be understood in that way, there is no foundation for a finding of fraudulent misrepresentation against RBS.

Each of the Swaps depended on LIBOR. It is, as we have already said, defined by the BBA in the following way:

“The rate at which an individual Contributor Panel bank could borrow funds, were it to do so by asking for and accepting inter-bank offers in reasonable market size, just prior to 11.00 [a.m.] London time.”

It was at the material times set for ten currencies (including sterling, US dollars, Swiss francs and Japanese yen) and for 15 different “tenors” such as 3, 6, 9, 12 months etc. In all, therefore, there are 150 different LIBOR rates published each London business day by Thomson Reuters. The rate for the agreed swaps in the present case was 3 months sterling or, as described in the Swaps themselves, “GBP 3 months” or “GBP 3M”.

It is well-known that between January 2006 and March 2012 there was what has been called “manipulation” of LIBOR in the sense that on some and perhaps many occasions submissions were made by panel banks which did not reflect the rate at which those banks genuinely thought they could borrow funds but were rather rates which were thought to benefit the banks’ trading position and/or individual traders’ bonuses. Investigations were carried out by regulators in both England and the United States which revealed widespread LIBOR manipulation. In England the Financial Services Authority (“the FSA”) carried out such investigations including an investigation into RBS which revealed many and substantial breaches of principles 5 and 3 of the regulatory Principles for Businesses. These breaches were set out in a Final Notice (“the Notice”) of 6 February 2013 and resulted in a fine of £87.5 million which would have been £125 million if RBS had not agreed to settle at an early stage of the investigation and thus qualified for a discount.

Principle 5 required that RBS observe proper standards of market conduct. Principle 3 required RBS to take reasonable care to organise and control its affairs responsibly and effectively, with adequate risk management systems.

The FSA made specific findings of manipulation in connection with RBS’s submission of rates that formed part of the calculation of Japanese yen and Swiss franc LIBOR. United States regulators made similar findings and those manipulations are now admitted by RBS. The FSA also found that RBS inappropriately considered the impact of its LIBOR submissions on the profitability of transactions in its money market trading books as a factor when it made Japanese yen, Swiss franc and US dollar submissions. It concluded that RBS’s misconduct undermined the integrity of LIBOR.

No specific findings were made in relation to the LIBOR sterling rate. Not only did the FSA not acquit RBS of any such manipulation but RBS recognised that its original allegation (that there had been no regulatory findings of misconduct on the part of RBS in connection with GBP LIBOR) should be deleted from its pleading.

PAG says (in broad terms) that, if it had realised that LIBOR had been manipulated and was not truly set according to the expectation of the market (and, therefore, according to the BBA definition), it would never have agreed to the Swaps in the first place. The Swaps should, therefore, now be rescinded and payments made by PAG to RBS pursuant to the Swaps should be repaid by RBS.

It alleges that four representations should be implied from RBS’s conduct in proposing that PAG enter into the swap contracts. Those representations are:

“(a)

On any given date up to and including the date of each of the Swaps: LIBOR represented the interest rate as defined by the BBA, being the average rate at which an individual contributor panel bank could borrow funds by asking for and accepting interbank offers in reasonable market size just prior to 11am on that date (LIBOR Representation 1);

(b)

RBS had no reason to believe that on any given date LIBOR represented anything other than the interest rate defined by the BBA, being the average rate at which an individual contributory panel bank could borrow funds by asking for and accepting interbank offers in reasonable market size just prior to 11am on that date (“LIBOR Representation 2”);

(c)

RBS had not made false or misleading LIBOR submissions to the BBA and/or had not engaged in the practice of attempting to manipulate LIBOR such that it represented a different rate from that defined by the BBA (viz a rate measured at least in part by reference to choices made by panel banks as to the rate that would best suit them in their dealings with third parties) (“LIBOR Representation 3”);

(d)

RBS did not intend in the future and would not in the future: make false or misleading LIBOR submissions to the BBA; and/or engage in the practice of attempting to manipulate LIBOR such that it represented a different rate from that defined by the BBA (viz a rate measured at least in part by reference to choices made by panel banks as to the rate that would best suit them in their dealings with third parties) (“LIBOR Representation 4”).”

The Judge held:

The question of LIBOR representations by banks is not completely virgin territory since there have been proceedings in which defendant banks have sought to strike out or obtain summary judgment in relation to allegations that misrepresentations have been made. Two such interlocutory cases came before the Court of Appeal in litigation called Graiseley Properties Ltd v Barclays Bank plc [2013] EWCA Civ 1372. In response to a submission that, when nothing was said by a bank in connection with LIBOR, there was no obligation to disclose its own dishonesty or breach of statutory duty Longmore LJ (with whom Underhill LJ and Sir Bernard Rix agreed) said:

“27.

In the present case, however, the banks did propose the use of LIBOR and it must be arguable that, at the very least, they were representing that their own participation in the setting of the rate was an honest one. It is, to my mind, surprising that the banks do not appear to be prepared to accept that even that limited proposition is arguable.

28.

It was also submitted that doing nothing cannot amount to an implied representation. But it is (arguably) the case that the banks did not do nothing in that they proposed transactions which were to be governed by LIBOR. That is conduct just as much as a customer’s conduct in sitting down in a restaurant amounts to a representation that he is able to pay for his meal, see DPP v Ray [1974] AC 370, 379D per Lord Reid.

…

30.

The banks’ submissions boiled down to saying that they were prepared to accept that they would do nothing dishonest or manipulative during the term of the contract and that should be enough for any counterparty. I can only say that, in my view, it is arguably not enough. If the day after the contracts had been made, the banks had told their counterparties that they had been manipulating LIBOR in the past and intended to do so in the future, but would be happy to pay any loss that their borrowers could prove, the borrower would (arguably) be sufficiently horrified so as to think he would be entitled to rescind the deal. The law should strive to uphold the reasonable expectations of honest men and women. If in the end it cannot do so, that should only be after a proper trial.”

These were, of course, only interlocutory observations and it now falls to this Court to determine the extent to which, if at all, they represent the law.

As we have already mentioned the Judge called the pleaded representations “extremely complex and intricate”. This is not an unmerited description. The first alleged representation is not only intricate but appears to amount to a representation that no bank had ever made inappropriate submissions to Reuters as to the rate at which LIBOR should be set and is an impossibly wide representation; much the same criticism can be rightly made of the second representation. The third and fourth representations are also unnecessarily complex and during the argument, the Court endeavoured to reduce their complexity to something simpler. The most feasible formulation seems to us to be that RBS was representing that, at the date of the Swaps, RBS was not itself seeking to manipulate LIBOR and did not intend to do so in the future. With some justification Mr Handyside complained that no such representation had been pleaded in terms. But it seems to us that our formulation captures the essence of what has been pleaded and we will proceed on that basis.

No such representation was made expressly nor would one expect it to be. The question therefore is whether it can be implied.

Mr Handyside submitted that it would be wrong to hold that any representation should be implied since it covered ground which would normally be covered by an implied term. RBS accepted (and had indeed asserted in paragraph 10 of its opening submissions at trial) that the Court should and would imply a term into the Swaps to the effect that RBS would not dishonestly manipulate 3 month GBP LIBOR so as to cause loss to PAG. But Mr Handyside pointed out that no such term had been pleaded (though other implied terms had originally been pleaded but were not now being pursued) for the obvious reasons that PAG’s loss on the swap transactions had been caused by the financial crash of 2008 and the subsequent extremely low rates of interest which had been set as a result, rather than by any misconduct on the part of RBS.

We do not accept this submission since the law relating to misrepresentation fulfils a different function from the law relating to implied terms. The former deals with the present not the future and gives potential remedies which may be more appropriate than a claim for damages. A party to a contract containing a swap needs to be certain of the counterparty’s honesty at the beginning of the deal not just in the future but throughout its course. If a claimant has suffered no loss, that may be relevant to remedy but should not exclude a right to rely on misrepresentation if any misrepresentation has occurred.

All this does not mean that the Court should be too ready to find an implied representation. An implication can arise from words or conduct. In DPP v Ray [1974] AC 370 the House of Lords held that the offence of obtaining by deception had been committed when a customer entered a restaurant, sat down, ordered a meal and then later decided not to pay. That was because the customer impliedly represented that he could pay before he left and that representation should subsequently have been corrected. Lord Reid dissented in the result but in agreement with the majority stated that the initial conduct of the customer in sitting down and ordering a meal was a representation that he was able and willing to pay for his meal: see page 379D. To our mind this is effectively representation by conduct rather than representation by words. To like effect are In Re Shackleton (1875) LR 10 Ch App 446 and in In Re Eastgate [1905] 1 KB 465, to which Newey LJ referred counsel in the course of the hearing. In these cases it was held that a buyer, who took delivery of goods from a seller, represented that he was able and willing to pay for them. On any view it is by no means impossible that a representation can, in theory, be made by conduct alone.

The facts of those cases are, of course, a long way from a typical swaps case. Although the general principle they illustrate applies to commercial disputes there is not a great deal of authority on the material required for the making of an implied representation in the context of a banking transaction.

In Geest plc v Fyffes plc [1999] 1 All ER (Comm) 672 the defendant had promised to use reasonable endeavours to obtain a bank’s agreement to substitute itself for the claimant as a guarantor of its subsidiary’s obligations under charterparties and in any event to indemnify the claimant in respect of its liability under the guarantees it had given. When the claimant sued for breach of these obligations, the defendant alleged that the claimant had failed to disclose that the claimant had confirmed to the shipowners that its subsidiary would remain its subsidiary for the duration of the charters. In fact the subsidiary had been disposed of and was no longer the subsidiary of the claimant. Colman J held that there was no general duty of disclosure before entering into a contract of guarantee or indemnity but only a duty not to make express or implied misrepresentations. The judge said that English law had settled on four propositions, the third of which he formulated as follows (page 683b):

“Where there is no express misrepresentation, the first question to ask is whether there has been any implied misrepresentation at all and, as with any other type of contract, the essential issue is whether in all the circumstances relating to the entering into of the contract of guarantee or indemnity, including in particular (a) the nature of the contract between the beneficiary and the principal debtor, (b) the conduct of the beneficiary and (c) express representations made by him to the surety, it has been impliedly represented to the surety that there exists some state of facts different from the truth. In evaluating the effect of the beneficiary’s conduct a helpful test is whether, having regard to the beneficiary’s conduct in such circumstances, a reasonable potential surety would naturally assume that the true state of facts did not exist and that, had it existed, he would in all the circumstances necessarily have been informed of it.”

In IFE Fund S.A. v Goldman Sachs International [2007] 1 Lloyd’s Rep. 264 Toulson J said (paragraph 50):

“In determining whether there has been an express representation and to what effect, the court has to consider what a reasonable person would have understood from the words used in the context in which they were used. In determining what, if any, implied representation has been made, the court has to perform a similar task, except that it has to consider what a reasonable person would have inferred was being implicitly represented by the representor’s words and conduct in their context.”

The Court of Appeal upheld the judge’s decision that no express or implied representation had been made on the facts, see [2007] 2 Lloyd’s Rep. 449.

Christopher Clarke J cited the dicta of both Toulson J and Colman J in Raiffeisen Zentralbank Osterreich AG v Royal Bank of Scotland plc [2011] 1 Lloyd’s Rep. 123 at paragraphs 82-85 echoing Colman J’s helpful test:

“whether a reasonable representee would naturally assume that the true state of facts did not exist and that, had it existed, he would in all the circumstances necessarily have been informed of it.”

He held that on the facts, no implied representation had been made. In Casa di Risparmio della Repubblica di San Marino SpA v Barclays Bank Ltd Hamblen J also adopted the remarks of Toulson J in considering whether implied representations had been made (he held that they had not).

In UBS AG v Kommunale Wasserwerke Leipzig GmbH and Depfa Bank plc [2014] EWHC 3615 (Comm) Males J accepted Depfa’s submission that it was implicit, in what it had been told by UBS about a particular need for Depfa to become an intermediary in a transaction with KWL, in the fact that UBS had told Depfa that it had done due diligence on KWL and in the fact that the transaction was presented to Depfa at all, that UBS believed the other party to the transaction to be honest and did not have any significant doubts about its honesty. In so doing, he followed Christopher Clarke J’s endorsement of Colman J’s helpful test.

The present case appears to be the first in which Colman J’s test has been considered by the Court of Appeal. We do think it is a helpful test, in relation to the existence of an implied representation, to consider whether a reasonable representee would naturally assume that the true state of facts did not exist and that, if it did, he would necessarily have been informed of it. To that extent we would approve the dicta of Colman J in Geest plc v Fyffes plc but that is not to water down the requirement that there must be clear words or clear conduct of the representor from which the relevant representation can be implied.

In the present case there were lengthy discussions between PAG and RBS before the Swaps were concluded as set out by the Judge in the earlier part of her judgment. We have particularly in mind the facts and matters set out in paragraphs 51-54 and 82-83 of the judgment. RBS was undoubtedly proposing the swap transactions with their reference to LIBOR as transactions which PAG could and should consider as fulfilment of the obligations contained in the loan contracts. In these circumstances we are satisfied that RBS did make some representation to the effect that RBS itself was not manipulating and did not intend to manipulate LIBOR. Such a comparatively elementary representation would probably be inferred from a mere proposal of the swap transaction but we need not go as far as that on the facts of this case in the light of the lengthy previous discussions. In this sense the case is comparable to UBS v KWL in which a not dissimilar representation was implied from what Depfa had been told by UBS and the fact that the relevant transaction was put forward to Depfa by UBS. It is true that UBS had also told Depfa that it had done due diligence as KWL but we do not consider that fact to have been decisive on its own in the mind of Males J in that case.

We therefore disagree respectfully with the Judge when she held (paragraph 407), admittedly in the context of the intricate pleaded representations, that the proffering of the Swaps was not in the context of this case conduct from which any representation could be inferred.

That is by no means the end of the matter since the scope (or precise terms) of the representation have to be considered. PAG says it was a general representation about LIBOR encompassing every tenor and every currency. RBS says any representation must be confined to 3 month GBP LIBOR, as the Judge in paragraph 408 indicated she would have held, if there was any representation at all.

Mr Lord submitted that the Court should adopt the approach of Flaux J at first instance in Graiseley, when, in response to an argument that, because the series of contracts in that case were ones where the LIBOR rate was by reference to sterling LIBOR, any representation should be limited to sterling LIBOR, he said (paragraph 19):

“It seems to me that it is a wholly artificial exercise to seek effectively to divide up the various LIBOR fixings or manipulations into separate currencies. It is quite clear that there was fixing not only of sterling LIBOR but also of dollar LIBOR and of EURIBOR, and that, as I said during the course of argument, there is inevitably scope for cross-infection here. It may be that in due course, when full disclosure has been provided, it will become apparent that one or other aspect of the LIBOR fixing assumes more significance, but it seems to me to be impossible to say that the representations that were impliedly being made should be limited in the way in which [counsel] suggests. Clearly, in terms of whether there is a real prospect of success, it is fully arguable that these representations were implicitly made to the claimants before they entered the various agreements.”

Flaux J was, of course, only saying that the bank’s representations arguably applied to all LIBOR currencies in the context of an application by the claimant for permission to amend its pleadings. The full argument we have now heard in the light of the facts of this particular case persuades us that the implied representation, which we consider to have been made, cannot extend further than sterling LIBOR. The main reason for so saying is that any implied representation cannot legitimately extend further than the particular transactions allegedly induced by the representations. RBS as a panel bank made submissions in all the ten currencies in which LIBOR was calculated and the main sterling submitter for RBS was Mr Mark Thomasson. Although submitters in other currencies may have made sterling submissions when Mr Thomasson was away on holiday, he remained the substantial submitter. Mr Paul Walker was the primary United States dollar LIBOR submitter and there was evidence that Mr Thomasson made dollar submissions when he was away. But in essence the submitters were different persons.

Perhaps more importantly the swap documentation in all four Swaps with which this case is concerned referred to GBP LIBOR. That was the currency of the transactions and any implied representation, while of course relating to that currency, cannot legitimately be extended, in our judgment, to other LIBOR currencies with which the transactions were not concerned.

The admitted manipulation of Japanese yen LIBOR and Swiss franc LIBOR is, of course, deeply shocking but that is not, of itself, a reason for holding that representations made to PAG should go further than representations about the sterling LIBOR rate. If, of course, a submitter in yen or Swiss francs had also made sterling submissions, that might render false the representation about sterling LIBOR but there is no evidence that any such submission occurred. Flaux J’s concern about the risk of “cross-infection” might therefore become relevant in some cases as a matter of fact but we do not consider that that risk can extend the representation which we are prepared to imply beyond its proper sterling scope.

Mr Lord sought to rely on Mr Thomasson’s admissions, that (1) he sat with both money market and derivative traders with whom he was “embedded” in an open plan dealing room (2) potential conflicts between the functions of those traders and the LIBOR submitters had not been recognised before 2012, and (3) he had on occasion acted as a substitute submitter for US dollar LIBOR when Mr Walker was away, as supporting the need for the implied representation to be a cross-currency representation. But these are all matters of fact unknown to PAG (or other customers of the bank) at the time and cannot determine the width of the appropriate implied representation which must be determined as a matter of law depending on the expectations or assumptions of the parties at the time the representations were made.

The Judge would have been even more restrictive, if she had been prepared to contemplate any representation at all, and would have confined it to 3 month GBP LIBOR. Naturally, Mr Handyside supported the Judge in this respect. We consider, however, that that would be too narrow. The idea that RBS could contemplate manipulating some of the tenors of sterling LIBOR and not others seems to us peculiarly far-fetched and would call into question RBS’s honesty in connection with sterling across the board. We do not therefore agree with the Judge in this respect and would hold that RBS did impliedly represent that it was not manipulating and did not intend to manipulate sterling LIBOR.

The next question is therefore, whether the implied representation was false. The Judge largely accepted Mr Thomasson’s evidence and held that PAG had failed to make out its case that there had been manipulation of GBP LIBOR. That is a finding with which this Court could not normally interfere. Mr Lord, however, mounted a sustained attack on her findings with which it is necessary to deal, at least, in outline.

The Judge held that there was no evidence of manipulation by RBS of either sterling or dollar LIBOR. At trial PAG had relied on alleged instances of trader manipulation, lowballing and financial crisis manipulation all of which were rejected by the Judge but Mr Lord submits that the Judge missed the wider picture in that:

On the face of it this may appear a formidable list of criticisms of the Judge but it must be remembered that the question whether RBS had manipulated sterling LIBOR was essentially a question of fact for the Judge who heard many witnesses over many days. It would be impossible for this Court to hold, in the face of her conclusion that PAG had not made out its case on manipulation, that there was in fact such manipulation. It may also be observed that all the alleged instances of misconduct post-date the First Swap of 6 October 2004 and many of them post date the Fourth Swap of 16 April 2008. They are thus a somewhat slender foundation for alleging falsity of a representation at the time the Swaps were concluded.

Nevertheless we shall deal with the particular matters raised by Mr Lord under the above headings.

We do not accept that the Judge misunderstood the BBA definition of LIBOR. She accurately set it out and also set out the BBA guidelines in paragraphs 310-312. We consider that the Judge intended the phrase “the rate at which a bank will look to lend funds which is higher than its bid rate” to refer to “a bank other than a submitting bank”. No doubt the Judge could have expressed herself with a little more precision than she did but we do not consider that she failed to understand what she had accurately set out in the previous two paragraphs.

In the light of this, the Judge’s acceptance (paragraph 453) of RBS’s evidence, that discussions with derivatives traders about LIBOR in a general sense had taken place but were a legitimate tool for the LIBOR submitter, seems to us not only unexceptionable but entirely understandable. Mr Lord’s broad submission, that any discussion with derivatives or money market traders about LIBOR was impermissible because such discussion would be bound to reflect the bank’s funding needs, is impossibly prescriptive. It might be improper for a submitter to make a submission based on the bank’s funding needs but the Judge went on to find that there was no evidence that Mr Thomasson had done so. The Judge added:

“I also do not consider it appropriate to draw adverse inferences from the fact that the submitters were seated on the same trading floor with derivatives traders at the time and could contact those traders in numerous ways, including it would seem, by way of an unrecorded intercom. In this regard, I take account of the fact that although it is possible that improper communications were made in this way, it cannot be safe to assume it or draw some kind of adverse inference particularly in the light of the fact that despite the fact that the same means of communication was available to submitters and traders in other currencies, there is plenty of documentary evidence of their conduct.”

These observations are apt and compelling.

Failure to take account of FSA findings

In order to understand this submission it is necessary to set out some of the FSA findings. Mr Handyside made it clear that, while RBS accepted the US Department of Justice findings in relation to yen and Swiss franc LIBORs, it did not accept the findings of the FSA in relation to dollar misconduct/manipulation. The FSA anonymised its findings but it was common ground that the reference to submitters C and D were references to Mr Walker the dollar submitter and Mr Thomasson the sterling submitter and we have therefore substituted their names in the below quotations for ease of understanding:

“71.

…[Mr Walker] observed to a Broker during the financial crisis that in the absence of liquidity, “people are just setting LIBOR to suit their books” and “its just where you’ve got your fixings really”.

…

73.

With respect to USD, on 16 August 2007, [Mr Walker] directed a junior Money Market Trader to make a USD LIBOR submission on his behalf that took into account the pricing of a large forthcoming floating rate transaction that would impact the USD money market book on 17 August 2007. Specifically, [Mr Walker] told a RBS colleague, “I’ve got massive fixing in ones, so I said to [Money Market Trader A, who was making RBS’s USD LIBOR submissions on the day in question] I just want the really, really low ones, in case they do fucking cut. [Mr Walker]’s reference to “massive fixings” was a reference to a repeating USD 4 billion borrowing facility that was set to fix on 17 August 2007. On 17 August 2007, RBS’s 1 month LIBOR submission was two basis points lower than 16August 2007 and seven basis points lower than 15 August 2007. However, as many Panel Banks also reduced their 1 month LIBOR submissions on 17 August 2007, RBS’s submission ranking relative to the Panel Banks only fell slightly on that day. On 20 August 2007, the date of its next submission, RBS’s 1 month submission went back up 2 basis points (and its submission ranking moved up significantly).

74.

Also, in relation to USD, between 9 March 2010 and 18 March 2010, [Mr Thomasson] made USD submissions which took into account the pricing of large forthcoming floating rate USD transactions. A communication on 9 March 2010 illustrates the consideration that Primary Submitters, including [Mr Thomasson], gave to these transactions. Specifically, on that date, Money Market Trader B, emailed [Mr Walker] whilst [Mr Walker] was on holiday and told him of a conversation he had had with [Mr Thomasson] who was filling in for [Mr Walker] and making RBS’s USD LIBOR submissions. According to Money Market Trader B, [Mr Thomasson] told Money Market Trader B that even though Money Market Trader B wanted them higher, he [Mr Thomasson] “wanted to keep them [USD LIBORs] down because of some fixes”. [Mr Walker] replied to Money Market Trader B’s email and confirmed to him, “we do have some big fixes in London so suits for low libors”. Notably, RBS’s USD LIBOR submissions stayed low during this period when there were five large USD floating rate transactions (but they were unchanged from the rates submitted over the previous three weeks). RBS’s USD LIBOR submissions went up after the last large transaction fixed.”

Although the Judge made some reference to the FSA findings in her recording of the parties’ submissions (paragraphs 327, 348 and 436 (with regard to paragraph 73 of the Notice) and 356 (with regard to paragraph 74 of the Notice)), and also to interviews undertaken by the FSA (paragraphs 325, 328, 334 and 432) and although she appreciated that RBS disputed the findings made in paragraphs 71-74 of the Notice, it is fair to say that she ultimately made no decision on whether the FSA’s findings at paragraphs 73 to 74 were correct.

Since these findings were a hotly contested issue, we consider it unfortunate that she came to no decision about them but the question remains whether her failure to do so vitiates her overall ultimate conclusion that the implied representation which we have found to exist (that RBS was not at the date of the Swaps manipulating nor intending to manipulate sterling LIBOR) was not a false representation. We do not consider that this failure does vitiate the judgment for the following reasons:

We would add the obvious comment that this Court is not in any position, without having heard the witnesses, to form a view of its own as to whether the FSA findings were justified. All this Court could do is to order a new trial which it would only do if the Judge’s omission to deal with those findings rendered her decision impossible to uphold. For the reasons we have given we do not think her omission drives us to that unpalatable conclusion.

Mr Thomasson’s calendar entries in August and September 2009

This is a matter with which the Judge did engage. She set out Mr Thomasson’s evidence at paragraph 325, PAG’s case at paragraph 430 and rejected it at paragraph 456. The entries were themselves no more than reminders that something had to be considered on the respective dates in the diary. They read “3 mth fix low” and “LOW 6 mths”. The idea that they were advance reminders to manipulate LIBOR was possible but not inherently likely since Mr Thomasson could not decide on a particular submission until the day in question. The highest Mr Lord can put it is to say that the submissions were “consistent” with Mr Thomasson having acted in accordance with the reminders. The Judge felt that Mr Thomasson’s own explanations seven years later were speculative but found his refutation of wrongdoing persuasive. That was a decision open to her on the evidence.

Mr Lord complains that the Judge was wrong in paragraph 456 to record that PAG’s pleaded case was concerned with third party trader manipulation rather than an allegation that Mr Thomasson took account of RBS’s own money book when making GBP LIBOR submissions. This may reveal some (but not much) misapprehension by the Judge about the emphasis Mr Lord was putting on RBS’s supposed knowledge of manipulation by third parties which he continues to make (see paragraphs 156-157 below). But it is hardly material since she accurately sets out PAG’s case at paragraph 430 and concludes paragraph 456 by saying that she was not satisfied that the calendar entries were evidence of Mr Thomasson’s taking account of trader positions when making GBP LIBOR submissions, which is precisely the case which PAG was making.

Absence of senior management witnesses

No litigant is obliged to call witnesses to satisfy the curiosity or enthusiasm of his opponent. It was always open to PAG to subpoena any witness it thought would be helpful to the Court. The fact that a party who might be expected to produce witnesses does not do so may sometimes speak volumes but it is a matter for the Judge to decide whether it does so in a particular case. The critical question in the present case was whether manipulation of GBP LIBOR had taken place. The critical witness for that purpose was Mr Thomasson. If he was believed, there was nothing relevant for senior management to know; if he was not, RBS’s case collapsed anyway. The Judge did say (paragraph 461) that RBS’s decision not to call Mr Cummins (and a Mr Nielsen) in connection with the allegations of lowballing did not reflect well on RBS and repeated this in her decision on whether RBS had been fraudulent (paragraph 485). She was well aware of Mr Lord’s case (paragraph 479) but in the end was not prepared to draw an adverse inference. We do not think the Judge can be criticised.

In any event Mr Cameron’s email that the Bank of England wanted banks to play LIBOR very “straight” is hardly evidence of actual wrong doing. The exchange about “gold/bronze medal” spots is likewise not evidence of any actual wrongdoing, but is (as the Judge held in paragraph 460) far from a positive request to set LIBOR low, let alone any evidence that such request was acted upon. It must not be forgotten that October 2008 when the financial crisis struck was a fraught time for all banks and the world in general.

“Awareness” of other panel banks’ misconduct

We can only say that these allegations, based as they are on communications exchanged in November 2007 and July 2009, are scarcely compelling evidence of misconduct in relation to GBP LIBOR by RBS and still less evidence of false representations in October 2004 and September 2007, the dates of the First and Second Swaps. We are not surprised they were not given much attention by the Judge.

We cannot refrain from reiterating the warning that many judges have now given about the risk of “island-hopping”, see Fage UK Ltd v Chobani UK Ltd [2014] EWCA Civ 5, paragraph 114, and Watson Farley &Williams v Ostrovizky [2015] EWCA Civ 457, paragraph 9. Mr Lord’s attempt to focus on some parts of days of evidence and some parts of a necessarily long judgment is just the sort of exercise which appellate Courts must deprecate. Even after a seven day appeal we cannot have a small part of the Judge’s familiarity with the issues and evidence in this case. We can see no reason to reverse her on falsity which was essentially a matter of fact for her. We are certainly not satisfied she was, to adopt the phraseology of Henderson v Foxworth Investments Ltd [2014] UKSC 41, [2014] 1 WLR 2400, “plainly wrong”.

There is therefore no need to consider whether the Judge’s conclusion that fraud had not been proved is correct. If we had concluded that the implied representation was false it would be necessary to decide how the normal rule, that, for a finding of fraud, the representor must have intended to make a representation he knew to be false (see Akerhielm v De Mare [1959] AC 789, 804 per Lord Jenkins, Gross v Lewis Hillman Ltd [1970] Ch 445 per Cross LJ and Raiffeisen v RBS [2011] 1 Lloyd’s Rep. 123 paragraphs 338-340 per Christopher Clarke J) can apply to an implied representation when the implication is not present to the representor’s mind. It may be the case that an implied representation of this kind can never (or quite rarely) be fraudulent; on the other hand recent decisions about dishonesty, such as Barlow Clowes International Ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 WLR 1476 and Ivey v Genting Casinos UK Ltd [2017] UKSC 67, [2017] 3 WLR 1212, may be relevant. It is unnecessary for us to resolve that question in this case.

Nor is there any need to consider the Judge’s finding on reliance.

We would dismiss this part of the appeal.

In August 2013, RBS instructed Lambert Smith Hampton to value those properties over which PAG had granted charges in favour of RBS. The cost of the valuation was £35,000 plus VAT. According to RBS, clause 21.5.1 of the 2011 facility agreement, which is set out in paragraph 23 above, entitled it to require the valuation to be prepared and provided for PAG to bear the cost.

PAG argues otherwise. It is its case that clause 21.5.1 of the 2011 facility agreement was not unfettered but was rather subject to a “Socimer-type” implied term requiring RBS to act “reasonably, in a commercially acceptable or rational way, in good faith, for a proper purpose (i.e. the purpose for which such power or discretion was conferred), not capriciously or arbitrarily and not in a way that no reasonable lender, acting reasonably, would do” (to quote from the Particulars of Claim). On the facts, PAG maintains that the Judge ought to have permitted it to rely on passages from the witness statement of a Mr Sefton of RBS and that, had she done so, she could only sensibly have found that RBS had decided by no later than May 2013 against refinancing PAG and, hence, that the purported exercise of clause 21.5.1 in August of that year was pointless, without good or rational reason and so in breach of the implied term for which it contends.

The reference to a “Socimer-type” implied term derives from Socimer International Bank Ltd v Standard Bank London Ltd [2008] EWCA Civ 116, [2008] Bus LR 1304. In that case, the Court of Appeal considered what, if any, limitations there are on a decision-maker’s freedom of decision when a contract allocates only to one party a power to make decisions under the contract which may have an effect on both parties (see paragraph 60). After referring to a number of previous cases, Rix LJ said (in paragraph 66):

“It is plain from these authorities that a decision-maker’s discretion will be limited, as a matter of necessary implication, by concepts of honesty, good faith, and genuineness, and the need for the absence of arbitrariness, capriciousness, perversity and irrationality. The concern is that the discretion should not be abused. Reasonableness and unreasonableness are also concepts deployed in this context, but only in a sense analogous to Wednesbury unreasonableness, not in the sense in which that expression is used when speaking of the duty to take reasonable care, or when otherwise deploying entirely objective criteria: as for instance when there might be an implication of a term requiring the fixing of a reasonable price, or a reasonable time.”

One of the earlier cases, Paragon Finance plc v Nash [2001] EWCA Civ 1466, [2002] 1 WLR 685, related to a lender’s contractual power to set the interest rate applicable to a mortgage. Dyson LJ, with whom Thorpe LJ and Astill J agreed, rejected a submission that the power was completely unfettered (paragraph 30). In paragraph 36, he said this:

“I would hold that there were terms to be implied in both agreements that the rates of interest would not be set dishonestly, for an improper purpose, capriciously or arbitrarily. I have no doubt that such an implied term is necessary in order to give effect to the reasonable expectations of the parties. I am equally in no doubt that such an implied term is one of which it could be said that ‘it goes without saying’. If asked at the time of the making of the agreements whether it accepted that the discretion to fix rates of interest could be exercised dishonestly, for an improper purpose, capriciously or arbitrarily, I have no doubt that the claimant would have said ‘of course not’.”

Dyson LJ also accepted that it was an implied term that the lender would not exercise its discretion in a way that no reasonable lender, acting reasonably, would do, while noting that it would be “quite another matter to imply a term that the lender would not impose unreasonable rates” (paragraph 41).

The Paragon and Socimer cases were both mentioned in Mid Essex Hospital Services NHS Trust v Compass Group UK and Ireland Ltd [2013] EWCA Civ 200, [2003] BLR 265. After referring to Socimer and several other cases (Abu Dhabi National Tanker Co v Product Star Shipping Ltd (The “Product Star”) [1993] 1 Lloyd’s Rep. 397, Horkulak v Cantor Fitzgerald International [2004] EWCA Civ 1287, [2005] ICR 402 and JML Direct Ltd v Freestat UK Ltd [2010] EWCA Civ 34) in which there were held to be implied terms, Jackson LJ said this:

“82

In each of the above cases the implied term was intrinsic. The contract would not make sense without it. It would have been absurd in any of those cases to read the contract as permitting the party in question to exercise its discretion in an arbitrary, irrational or capricious manner….

83

An important feature of the above line of authorities is that in each case the discretion did not involve a simple decision whether or not to exercise an absolute contractual right. The discretion involved making an assessment or choosing from a range of options, taking into account the interests of both parties. In any contract under which one party is permitted to exercise such a discretion, there is an implied term. The precise formulation of that term has been variously expressed in the authorities. In essence, however, it is that the relevant party will not exercise its discretion in an arbitrary, capricious or irrational manner. Such a term is extremely difficult to exclude, although I would not say it is utterly impossible to do so….”

On the facts of the case before it, the Court of Appeal unanimously concluded that there was no implied term.

The Socimer case (but not, it seems, Mid Essex Hospital Services NHS Trust v Compass Group UK and Ireland Ltd) was also cited in Braganza v BP Shipping Ltd [2015] UKSC 17, [2015] 1 WLR 1661. In that case, Baroness Hale DPSC, with whom Lord Kerr JSC agreed, said this (in paragraph 18):

“Contractual terms in which one party to the contract is given the power to exercise a discretion, or to form an opinion as to relevant facts, are extremely common. It is not for the courts to rewrite the parties’ bargain for them, still less to substitute themselves for the contractually agreed decision-maker. Nevertheless, the party who is charged with making decisions which affect the rights of both parties to the contract has a clear conflict of interest. That conflict is heightened where there is a significant imbalance of power between the contracting parties as there often will be in an employment contract. The courts have therefore sought to ensure that such contractual powers are not abused. They have done so by implying a term as to the manner in which such powers may be exercised, a term which may vary according to the terms of the contract and the context in which the decision-making power is given.”

Baroness Hale went on to quote this passage from paragraph 37 of the judgment of Lord Sumption JSC in British Telecommunications plc v Telefónica O2 UK Ltd [2014] UKSC 42, [2014] Bus LR 765:

“As a general rule, the scope of a contractual discretion will depend on the nature of the discretion and the construction of the language conferring it. But it is well established that in the absence of very clear language to the contrary, a contractual discretion must be exercised in good faith and not arbitrarily or capriciously: Abu Dhabi National Tanker Co v Product Star Shipping Ltd (No 2) [1993] 1 Lloyd’s Rep 397, 404, per Leggatt LJ; Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (No 2) [2001] 2 All ER (Comm) 299, para 67, per Mance LJ and Paragon Finance plc v Nash [2002] 1 WLR 685, paras 39–41, per Dyson LJ. This will normally mean that it must be exercised consistently with its contractual purpose: Ludgate Insurance Co Ltd v Citibank NA [1998] Lloyd’s Rep IR 221, para 35, per Brooke LJ and Equitable Life Assurance Society v Hyman [2002] 1 AC 408, 459, per Lord Steyn, and p 461, per Lord Cooke of Thorndon.”

Mr Lord argued that the discretion conferred on RBS by clause 21.5.1 of the 2011 facility was impliedly constrained in the sorts of way described by Lord Sumption. Without such a limitation, Mr Lord said, there would have been nothing to stop RBS requiring a valuation every week or even every day. He recognised that, having regard to clause 21.5.1’s express terms, PAG could not have been called on to pay for a valuation more than once a year, but submitted that that did not provide PAG with adequate protection. Having a valuation conducted is, he suggested, burdensome and intrusive. Moreover, it could not (so he submitted) have been the intention of the parties that RBS should be able to charge PAG for a valuation for which there was no proper purpose.

The Judge took a different view. It seemed to her that RBS had “an absolute right to call for the valuation and accordingly, that the Socimer line of authorities and the necessary implication of terms in order to control the otherwise unfettered exercise of a discretion/assessment or formulation of opinion [do] not arise” (paragraph 278 of the judgment). Echoing the Judge’s conclusion, Mr Handyside argued that clause 21.5.1 of the 2011 facility did not involve any discretion, assessment or choosing from a range of options and that there was therefore no scope for any implied term to arise. Mr Handyside stressed that clause 21.5.1 had been inserted for the benefit of RBS and maintained that it was not obliged to take account of PAG’s interests when deciding whether to invoke the provision.

In our view, however, the power conferred by clause 21.5.1 of the 2011 facility was not wholly unfettered. We agree with Mr Handyside that the provision will have been inserted for the benefit of RBS, and there is, of course, no question of RBS having owed fiduciary duties. In the circumstances, it seems to us that RBS must have been free to act in its own interests and that it was under no duty to attempt to balance its interests against those of PAG. It can, however, be inferred that the parties intended the power granted by clause 21.5.1 to be exercised in pursuit of legitimate commercial aims rather than, say, to vex PAG maliciously. It appears to us, accordingly, that RBS could not commission a valuation under clause 21.5.1 for a purpose unrelated to its legitimate commercial interests or if doing so could not rationally be thought to advance them.

The next issue in this context is whether the Judge was right to decline to allow PAG to rely on the passages from the witness statement of Mr Sefton which it contended supported its case. The statement in question had been served by RBS, but it had not in the event called Mr Sefton to give evidence at the trial. PAG wished to put in as hearsay evidence these parts of the statement:

“2.

In relation to PAG, my understanding when I came to work on the file in early 2013 was that RBS was not inclined to continue financing PAG beyond the expiry of the 2011 Facility in June 2014.

…

38.

As I mentioned above, it was my understanding when I came to work on the file in early 2013 that RBS was not inclined to continue financing PAG beyond the expiry of the 2011 Facility in June 2014, largely in view of PAG’s high LTV [i.e. loan to value] ratio.”

Where a party has served a witness statement but not called its author, CPR 32.5(5) is in point. That provides:

“If a party who has served a witness statement does not—

(a)

call the witness to give evidence at trial; or

(b)

put the witness statement in as hearsay evidence, any other party may put the witness statement in as hearsay evidence.”

CPR 32.5 was considered in McPhilemy v Times Newspapers Ltd (No. 2) [2000] 1 WLR 1732, where the claimant sought to adduce statements made by a potential defence witness with a view to discrediting much of what was said. The Court of Appeal held that the trial judge had been right to refuse the application. Brooke LJ, with whom Thorpe LJ agreed, noted that, when witness statements were first introduced in 1986, there was an express rule barring a party from putting in evidence a statement served by his opponent of a witness who was not called. Brooke LJ observed (at page 1736):

“[CPR 32.5] abrogates the old rule and makes permissive what the old rule prevented. It is then a matter for the discretion of the judge whether to permit it. In my judgment, however, there is nothing in this new rule to change the basic rules of the laws of evidence which existed before the new rule was introduced by the rule-makers, and which are still in force today.”

Going on, Brooke LJ said (at page 1740):

“I know of no principle of the law of evidence by which a party may put in evidence a written statement of a witness knowing that his evidence conflicts to a substantial degree with the case he is seeking to place before the jury, on the basis that he will say straight away in the witness’s absence that the jury should disbelieve as untrue a substantial part of that evidence.”

In our view, CPR 32.5 is not applicable where a party wishes to put in only part of a witness statement. The rule itself refers to “the witness statement” being admitted, not merely some of it. Further, it makes sense that a party wanting to rely on something said in a statement should have to place all of the statement before the Court. A Court asked to attach significance to a passage from a statement should have before it the totality of what the witness said. There would otherwise, as the Judge noted in paragraph 296 of her judgment, be “real concern that cherry picking out of context would arise”. It would, moreover, be odd if a party were free to contend for the reliability of what the witness said in a particular passage while withholding the balance of the statement because he disputed it. That, in fact, would seem to have been the position in the present case had the Judge acceded to PAG’s application. PAG, we gather, was unwilling to put in Mr Sefton’s witness statement in its entirety because most of what he said was adverse to its case. It follows that, in our view, the Judge was correct to refuse PAG’s application.

It is very far from apparent, however, that the Judge would have held the valuation at issue to have been pointless, lacked good or rational reason or been commissioned for a purpose unrelated to RBS’s legitimate commercial interests or when doing so could not rationally be thought to advance them even if she had allowed PAG to put in the passages from Mr Sefton’s witness statement on which it wished to rely. PAG maintains that Mr Sefton’s evidence would have led the Judge to conclude that RBS had decided by May 2013 that it would not refinance PAG. Mr Sefton did not say unequivocally that RBS had resolved not to refinance, but rather that it was his “understanding” “when [he] came to work on the file” that RBS was “not inclined” to continue financing. It is by no means evident that that material would have led the Judge to reject the clear evidence of Mr Whatham of RBS that the decision not to refinance PAG was not taken until much later in 2013.

In short, although we differ from the Judge on whether the power conferred on RBS by clause 21.5.1 of the 2011 facility was subject to an implied limitation, there is no reason to doubt her conclusion that RBS was entitled to commission the 2013 valuation and to recover its cost from PAG. This ground of appeal therefore fails.

The appeal will be dismissed.