Lehman Brothers International (Europe)v Exotix Partners LLP

This judgment relates to a dispute as to the terms of a sale (the “Trade”) of Peruvian Government Global Depository Notes (each referred to as a “GDN”). The Trade was entered into in early 2014 by the Claimant, Lehman Brothers International (Europe) (“LBIE”) as vendor and the Defendant, Exotix Partners LLP (“Exotix”) as purchaser.

A GDN or global depository note is a debt instrument created and issued by a depository bank (in this case Citibank) which evidences ownership of a local currency-denominated debt security. Such a note emulates the terms of the underlying debt security, which comprises government-issued bonds (such as the interest rate and maturity date), in this case issued by the Republic of Peru. However, unlike the underlying bonds, each GDN provides for payment of interest and principal in US dollars and can be settled through Euroclear and Clearstream. The detailed terms of the GDNs are set out in a document prepared by Citibank called “the GDN Supplement”. The depository bank holds the underlying bonds on behalf of the GDN holders.

The Trade of GDNs which is the subject matter of these proceedings was entered into orally, but on a recorded telephone line, and there is an agreed transcript of what was actually said. It is accepted by the parties and their experts that the Trade was concluded by the traders on the telephone. However, the exchanges between the traders were somewhat informal; they were not in legal language although they were intended to have legal effect.

Ordinarily, it might be thought that, once settled, the subject matter of the Trade should not be in doubt. But in this case it is the subject of dispute. It seems plain that in settling the trade LBIE appears to have materially misunderstood what it was delivering, whilst Exotix had no expectation of receiving assets of such a value as in fact were thus delivered to and received by it. LBIE thought that its holding of GDNs amounted to ‘scraps’ having a value of some $7,000; Exotix did not expect more and paid some $7,438. In fact, LBIE’s holding had a true aggregate value of over $7 million; and it delivered all its holding to Exotix, which, after some internal hesitation, decided to sell on the entirety and pocket a windfall.

In such circumstances, the principal dispute that has arisen is as to the true meaning and effect of the bargain struck further to an oral agreement between the traders.

The dispute is further complicated because LBIE’s interpretation of the bargain struck between the traders would result in LBIE’s delivery commitment being for more or less than a whole number of GDNs; and there is an issue as to how such a commitment could be effected in circumstances where there is no express provision for fractional entitlements in respect of GDNs, and whether terms would fall to be implied to enable the Trade to be performed.

Further, and more generally, there is in such circumstances also an issue as to whether LBIE as claimant is entitled to any, and if so what, relief given that the Trade has been settled long since.

As both parties accept, it is the effect of the bargain made as expressed in the words used that is to be determined. The Court must resist any temptation to mend a bad bargain: rectification is not sought, and the contract must be given effect according to its true construction, unless impossible of performance or incapable of legal effect, in which case the matter is governed by the law of restitution.

LBIE scarcely needs introduction. It was the primary trading company within the Lehman Group of companies in the UK and Europe. It entered into administration on 15 September 2008, the same day as the ultimate holding company of the Lehman Group filed for protection under Chapter 11 of the US Bankruptcy Code. The collapse of the Lehman Group is well-known: it shook the financial world, and its reverberations are still being felt more than a decade later.

The facts relevant to the dispute took place within the context of the administration of the LBIE estate. The GDNs which were the subject of the Trade were part of a basket of miscellaneous assets in LBIE’s estate which its Joint Administrators viewed as ‘scraps’ and which they were seeking to realise by sale to third parties.

Exotix is a broker for fixed income and equity securities. Exotix states in its Defence that it does not trade securities on its own account except in certain limited circumstances, where it facilitates the execution of client orders.

Thus, Exotix does not have the regulatory permissions necessary to purchase and hold assets for its own book; thus, it could only trade acting as a broker and where it has put in place an onward sale for any assets purchased by it (a “matched principal basis”). In this way, the asset purchased is transferred back-to-back on the day of settlement and Exotix is not exposed to market risk.

However, in some cases, Exotix may, where it has not yet found a third-party buyer for a matched transaction, transfer assets purchased to an associated “warehousing” entity; and that is what it did, on 31 January 2014, in the case of the Trade. It seems clear, however, that Exotix would not knowingly have contracted to take onto its own book and, having in place no matched transaction, “warehoused” assets having so much more than the ‘scraps’ value attributed to the GDNs the subject of the Trade.

Though there was considerable dispute as to the admissibility of other evidence, there is no dispute as to the admissibility of the pre-contract factual matrix evident from the documentary record, except in relation to the documents referred to as the LBIE SignOff Pack and its attached spreadsheet mentioned in paragraphs [21] and [22] below.below.

The Trade was agreed during a telephone conversation on 31 January 2014 between

On the morning of 24 January 2014, Mr Andrew Hall of LBIE (‘’Mr Hall’’, one of Mr Radicopoulos’s juniors who thus initially set up the Trade) invited Mr Hutton of Exotix to participate in a conversation on the Bloomberg instant message platform (commonly known in the industry as “Chat”). Chat enables participation on a recorded line by a number of persons who may join and leave as they please, with each such event and the conversations between all participants being recorded.

During the course of that Chat, Mr Hall sent Mr Hutton an email setting out a list of six different securities being offered for sale by LBIE to Exotix. These are set out below, although the email did not include the “ISIN” and “Description” headings and was not in the tabular form produced below:

At this initial stage (24 January 2014), LBIE did not disclose the size of its GDN or any of its listed positions.

Following further exchanges between LBIE and Exotix regarding the securities listed above, on 29 January 2014, at 16:43, Mr Hall sent an email to Mr Hutton stating “As discussed we will require a bid on all of the following by 4.00 pm London time 31/01/14, reserve levels apply”, and then setting out the portfolio of securities available for sale in the following table (the “Portfolio”). The headings used in the table below are as set out in Mr Hall’s email.

The one material difference between this table and the initial list of securities provided by Mr Hall on 24 January 2014 (produced at paragraph [17] above) is that the securities with ISIN BRVALEDBS028 had been replaced with securities with ISIN USY68851AK32. The GDNs are in the final row of the table in paragraph 19 above.

Exotix, as a securities broker, then sought bids from third parties on the securities in the Portfolio. In doing so, Exotix’s communications seeking bids from third parties typically referred to “PERU 6.9 08/12/37 Corp – USP87324BE10 – PEN 22,955” (Footnote: ¹). It is also to be noted that Mr Hutton, both in internal emails and in seeking to elicit bids, described the securities as “scrappy lehman positions” and “small scraps”.

In order to explain how it fits into the chronology, but emphasising immediately that its admissibility in determining the issue of interpretation at the heart of this case is disputed, it is convenient here to mention that in the meantime, and prior to entering into the Trade, LBIE produced an internal “Sign-Off Pack”, which was part of LBIE’s internal authorisation process and was produced by Mr Radicopoulos’s team. The Sign-Off Pack was signed by Mr Radicopoulos (on behalf of the LBIE front office team), Mr Viegas (on behalf of the LBIE valuation team) and Mr Copley (one of LBIE’s Joint Administrators) on 29 January 2014.

The Sign-Off Pack included a spreadsheet (the “BONY Spreadsheet”) listing out a range of information including, inter alia, the value and price of the securities as understood by the Bank of New York (“BONY”). The BONY spreadsheet recorded in one column the “BONY position” as “22,955”, and in another column the value as “22,580,893.50”. Exotix contended that the reference to “22,955” must be to units, rather than notional amount. The BONY spreadsheet also included a field allowing for the identification of discrepancies between what LBIE understood the price to be and how BONY understood the price. Exotix placed considerable reliance on this document, as being not only a record made by LBIE’s custodian of its true holding of GDNs measured in units but as having been produced specifically for the Joint Administrators to approve the same and as having accordingly been intended to be a record of what was to be sold. I shall return to the BONY Spreadsheet later in this judgment; but I should note now that it is common ground that this document was intended only to be used internally within LBIE and was not shared with Exotix prior to entry into the Trade: and it is on that basis that its admissibility and that of the Sign-Off Pack as a whole is disputed by LBIE.

On the morning of 31 January 2014 (the day of the Trade), at 09:45, Mr Hutton sent an email which confirmed that there were no trading restrictions for the GDNs, so that a trade could be done for the GDNs in as small or as large a size as the parties wished.

The emails in evidence reveal that Mr Hall left work early that day but, before doing so, asked Mr Hutton to send any bids on the Portfolio to Mr Radicopoulos, his LBIE colleague, in advance of the 4.00pm deadline which had been stipulated. Mr Hall then connected Mr Radicopoulos on Chat.

At 2.30pm on 31 January 2014, Mr Hutton contacted Mr Radicopoulos by Chat to explain that he was likely to have a bid for the Venezuelan Oil Warrants (ISIN: XS0029484945) (the “Vene” (Footnote: ²)) by the 4.00pm deadline but that the “others were proving tricky”. Mr Radicopoulos explained that it would be preferable to obtain a bid for the entire Portfolio:

“we are really looking for portfolio bid otherwise we are left with all the sh*t in the end which I don’t care for”.

This Chat was followed by a recorded telephone conversation commencing at

A few minutes after that, at 16:20:19, Mr Hutton called Mr Radicopoulos back. This was the call during which the Trade was finally agreed. Following an initial conversation about the terms of the trade relating to the Vene, the (critical) conversation relating to the GDNs went as follows (again, a transcript of the call has been agreed by the parties):

The terms of the Trade were, and the case thus turns on what was meant when it was agreed, that Exotix would buy

“the 22 just on the shy of 22.. 23 thousand er sol erm at 91 and a half which is around 7,712… dollars”.

It was not in dispute between the experts called by the parties (see further below) that after traders agree a trade (such as the Trade in the present case), a summary of its terms is commonly sent by one party to the other (commonly referred to as the “ticket” or the “VCON”). It is further agreed between the experts that the ticket or VCON is meant to serve as a record of the parties’ agreement.

On the 16:20 telephone call, Mr Hutton indicated that he would be sending through a ticket, which appears to have been a reference to the VCON. In line with this, shortly after the telephone conversation on which the Trade was agreed, Mr Hutton sent Mr Radicopoulos a VCON summary of the Trade generated through Bloomberg.

I note that there is some uncertainty as to when the VCON was prepared; although it was sent to Mr Radicopoulos at 16:23 on the same day, the face of the VCON suggests it was produced at 16:17 (i.e. prior to the crucial telephone call which took place between 16:20 and 16:21 on 31 January 2014).

The VCON thus on its face records the Trade as being for: (i) the sale of “22.955 (M)” GDNs (ii) at a price of “91.500000”, being 91.5% of par or nominal value.

I should deal in passing with a pleading point in relation to the VCON which gave rise latterly to some extended debate and supplemental written submissions between the parties. LBIE’s pleaded case for trial was that although the VCON or Ticket correctly recorded (a) the agreed price and (b) the total consideration payable (US$7,707.93), its statement of (c) the quantity of GDNs to be sold as “22.955 (M)” was incorrect “if different” from the quantity of 22.955 GDNs with an assumed nominal value of S/22,955 in fact agreed to be sold. For its part, Exotix embraced the suggestion that “an objective observer believing the GDNs to have a par value of Sol 1 would…have read the Ticket as referring to 22,955 GDNs being sold for a consideration of US$7,707.93.”

At trial, however, and in opening its case, LBIE proceeded on the footing that on its true construction the VCON or Ticket correctly recorded the subject matter of the Trade as GDNs with a notional value of S/22,955; and its evidence and crossexamination of Exotix’s witnesses proceeded on that basis. When, however, LBIE sought at the end of the Trial to amend its pleading accordingly, Exotix objected on the grounds that the amendment (1) was based on a case which was “legally inconsistent and hopeless”; (2) amounted to a reversal of position amounting in effect to the impermissible withdrawal of an admission; and (3) had been proposed too late without proper excuse. I do not accept these objections. I consider the amendment to be appropriate to ensure, albeit belatedly, that LBIE’s pleading conforms with the case put forward without objection. More particularly (1) whether it is a good case is the subject matter for adjudication; (2) I would not equate the effort to conform its pleading with its case with the withdrawal of an admission; and (3) I do not accept there is any prejudice to the parties or to other court users in permitting the amendment. Any issue of costs can be dealt with after judgment.

Returning to the narrative, LBIE contended that the way the VCON was generated is of some importance. It was produced by Mr Hutton by inputting the requisite details into the Bloomberg system, and was sent to Mr Radicopoulos immediately following the call on which the Trade was done (sent at 11.23 Eastern Time, 16:23 GMT). In fact, the VCON may have been prepared immediately prior to the call (the VCON suggests it was produced at 16:17 and the call lasted between 16:20 and 16:21): but it was not suggested that anything turns on this.

When inputting the original face amount, Mr Hutton had 3 options:

For the Vene, Mr Hutton used input “X”. The reason for that was that warrants trade not with reference to a price as a percentage of their face amount, because they have no face amount, and so it was necessary to apply a price per warrant rather than a percentage price. But conversely, Mr Hutton used the “M” input for the GDNs, dividing the agreed nominal value of the GDNs to be traded (S/22,955) by 1,000 to arrive at the input 22.955 (M). He also entered the agreed price, 91.5, being 91.5% of the nominal value of the GDNs to be traded and the Bloomberg system automatically calculated the outputs shown in the VCON.

LBIE submits that these facts demonstrate that Mr Hutton (whose own experience was on the bonds desk) equated the GDNs with bonds, not warrants; and that the input must have been of the assumed face value of the GDNs times 1,000, with the price a percentage of face value and not per unit.

Exotix rejected this submission, on the basis that the premise of the argument is that the parties should be taken to have known that the GDNs were worth 1,000 and not one sol each; whereas Mr Hutton’s evidence, which was not challenged, was that his belief or assumption was that each GDN was worth one Sol. On that basis, it made no difference whether he inputted face or notional value or the number of GDNs. I discuss later the admissibility and effect of Mr Hutton’s stated understanding, and whether it is admissible and relevant whether Mr Radicopolulos shared that understanding (though I should clarify now that I accept his evidence that he was, as he saw it, trading a face or nominal amount of underlying Peruvian government debt as represented by the GDNs and never focused on or knew what number of GDNs were thus being sold).

The experts were agreed that a securities trade is settled when (a) the ownership of the securities has been transferred from seller (here LBIE) to buyer (here Exotix) and (b) the requisite consideration is paid (“Settlement”). The period between (i) the date on which the trade was agreed (here 31 January 2014) and (ii) the date on which the trade settles (here 5 February 2014) (the “Settlement Date”), is commonly referred to as the settlement period (the “Settlement Period”).

During the Settlement Period, the seller’s settlement team typically receives a trade confirmation confirming their purchase of the instrument. Using the trade confirmation, the parties’ settlement teams will give instructions which, once matched, will allow Settlement to take place.

In the present case, LBIE received a trade confirmation from Exotix on 3 February 2014 (2 days before the Settlement Date) (the “Confirmation”).

Neither Mr Hutton nor Mr Radicopoulos could recall having seen the Confirmation prior to the Settlement Date; and neither of them thought that they would have seen it. On that basis, the admissibility of the Confirmation on the issue of interpretation is disputed by LBIE, whereas Exotix contend that it is all part of the contractual process, is to be read with the VCON or Ticket, and is admissible accordingly. I return to address and determine that issue later; but for the present it seems to be an integral part of the story in any event. The Confirmation stated as follows:

“Trade Date: 31-Jan-2014

Settlement Date: 05-Feb-2014

Action: We confirm our Purchase

Quantity: USD 22,955.00

Security Description: PERU 6.9 08/12/37

ISIN: USP87324BE10

Price: 91.500000%

Total Consideration: USD 7,707.93

Euroclear Account: 12849”

On the Settlement Date, 5 February 2014, LBIE, through BONY which actually held the GDNs to LBIE’s order, delivered to Exotix 22,955 GDNs (i.e. GDNs with a notional value of S/22,955,000 (or in excess of US$7.7m). Exotix paid $7,707.93 (inclusive of accrued interest). (That consideration consisted of two components: (1) a percentage (91.5%) of the notional value of the GDNs (at least insofar as the traders understood that value), calculated at the time to be USD 7,438.14; plus (2) an amount of interest accrued since the last coupon payment (which LBIE had therefore not yet received), which was USD 269.79.)

Thus, it was that the way that the Trade was in fact settled resulted in Exotix receiving the GDNs at a price representing around 1/1000 of their market value at the time.

LBIE’s preferred interpretation of the Trade as limiting its delivery obligation to 22 GDNs and the cash value of 0.955 GDNs is therefore not what in fact happened upon settlement. Exotix attaches importance not only to this, but also to the instructions that (according to Exotix) should, in the absence of any direct evidence to the contrary, be inferred LBIE must have given BONY. In particular, Exotix contends that LBIE cannot have instructed BONY to transfer GDNs having an aggregate par value of Sol 22,955 to Exotix, as is required on LBIE’s case. Had LBIE done so, BONY, which knew that the par value of the GDNs was Sol 1,000 (since it had correctly valued them), would have understood that it was being instructed to transfer 22.955 GDNs.

Against this, however, it appears that Exotix did not, at settlement, immediately appreciate that it was being delivered a package of GDNs with a value so different

It appears that Exotix did not appreciate any discrepancy until it received on 24 February 2014 (some 3 weeks after the Settlement Date) a coupon payment on the GDNs in the sum of US$276,321.25, vastly exceeding that purchase price for the GDNs themselves.

Mr Damien Marron (“Mr Marron”, an associate in Exotix’s Dubai office, Middle

At 12:42pm that day, Mr Marron sent a further email to Standard Chartered stating:

“Can you please kindly investigate the below as we do not like to have funds in the account – of such a large sum, which we believe we are not entitled to”.

Following several email exchanges (both internally and with Standard Chartered), Exotix eventually realised that the notional value of the GDNs it had received on the Settlement Date was S/22,955,000, and not, as in Exotix’s internal records, S/22,955.

Mr Hutton then emailed Mr David Baskerville (Exotix’s Chief Operating Officer at that time), who then undertook some form of internal investigation into the matter. Around this time, in an internal Exotix email regarding the internal investigation, Mr Baskerville warned the recipients (Mr Elliott and Mr Longden)

“not to speak to ANYONE about it”.

Mr Hutton says that Mr Baskerville did not share with him the results of this internal investigation and was told that “the decision as to what to do about the GDNs was one that was above my pay grade” and that he thereafter had no further involvement in the matter.

Ultimately the decision was taken by Exotix at board level that Exotix would keep the GDNs. An excerpt of the relevant board minutes (dated 30 April 2014) notes as follows:

“…The coupon received was surprising [sic] high and initially thought to be an error but further investigation indicated that although the GDN is referenced 1:1 against the underlying bond that Bond itself has a nominal value of 1,000 compared with the normal issuance of 1. This proved back the coupon receipt and it became apparent that we had acquired the asset at undervalue due to the way in which Bloomberg have entered the standing data into their system…. If these price providers are incorrect, we have received something of a windfall should the trade stand good. Conversely, should the trade subsequently be set-aside there is a material risk to the business that will grow over time in the same way as the [omitted word(s)] claim did due to both the loss of coupon income, fx differences…. To give clarity to these risks we sought external legal opinion… and from this it is clear that [omitted words]. We therefore need to recognise the gain if we consider the market to be incorrectly valuing the asset”.

Following the Trade, Exotix received coupon payments on the GDNs of $276,312.25 (on 24 February 2014 – noted above), $278,338.49 (on 26 August 2014) and $253,421.25 (on 24 February 2015).

About a year later, on 7 April 2015, Exotix (acting by a Mr Tweedley, in the New York office) on-sold the S/22,955,000 GDNs which LBIE had delivered to it to a third party for US$7,757,921.01 (at a price of 103.581100%), thereby making a windfall of over $8.5m (when added to the coupon payments referred to in the previous paragraph).

It appears to be common ground that Exotix did not notify LBIE of the outcome of its internal investigation, nor of the windfall Exotix made on the on-sale of the GDNs. During the Trial, I was informed by Mr Bayfield QC on behalf of LBIE that LBIE first became aware of the issue in July 2015 (a few months after the on-sale) upon being notified by the FCA which was, as Mr Bayfield put it, “presumably triggered by having seen the onwards trade”. So far as I am aware, no further details relating to the FCA’s intervention have been put in evidence.

In summary, LBIE delivered GDNs with a value far more than it thought it had to sell; whereas Exotix bought and paid an amount commensurate with a ‘scrap position’ but in fact received assets of very considerable value, which it on-sold at a huge profit, which it chose not to disclose to LBIE.

What, in these odd circumstances, where the subjective expectation of the parties at the time is clear, but the objective intention apparent from their bargain is more difficult to determine, should the legal response be?

The law insists (subject to limited exceptions) on ignoring subjective evidence of the parties’ intentions; and the primary question is as to the meaning which the recorded exchanges between Mr Radicopoulos of LBIE and Mr Hutton of Exotix comprising their agreement to the Trade, as set out in paragraph [28] above, would convey to a reasonable man, having regard to all the relevant facts surrounding the transaction so far as available to the parties in the situation in which they were at the time of the Trade (and see per Lord Bingham of Cornhill in BCCI v Ali [2002] 1 AC 251).

The principal issues in that context are (1) what the parties agreed to trade (the alternatives being (a) 22,955 GDNs or (b) a number of GDNs having a face value of PEN 22,955); and (2) at what price (US$7,707.93 or (by implication, as subsequently explained) US$7,707,926.69). In other words, there are issues as to both (i) subject matter and (ii) price.

There are the following further issues, only the first two of which are in the event required to be adjudicated:

LBIE’s primary case is that, on both a subjective and objective basis, both parties only ever intended to contract in respect of ‘scraps’, and the Trade should be construed as a sale of GDNs having a value commensurate with the price paid. On LBIE’s preferred interpretation of the Trade, it inadvertently over-delivered by a factor of 1,000 and should now be entitled to restitutionary relief (alternatively damages for breach of contract or equitable compensation for breach of trust). It thus seeks to recover both the value of the coupon payment and the GDNs in so far as in aggregate exceeding the price paid and the coupon that would have been due on 22 GDNs.

LBIE’s alternative case is that even if Exotix is right that the subject-matter of the Trade was 22,955 GDNs (with a corresponding face value of PEN 22,955,000, rather than GDNs with a notional value of PEN 22,955) then the agreed price, which was stated as a percentage (91.5%) of the face value, was in fact USD 7,707,926.69 (being the product of the agreed price, i.e. 91.5% of the face value) plus a (this time larger) amount of unpaid accrued interest. On that alternative basis, LBIE claims therefore to be entitled to recover from Exotix the balance of the consideration (which LBIE has not been paid).

Exotix, on the other hand, accepts that LBIE was mistaken as to the value of the assets it wished and agreed to sell; but that this was due (to quote Exotix’s Closing Submissions) to LBIE’s own “extraordinary lack of care in selling its own assets at the wrong price”. Exotix maintains that, as matter of strict law, both delivery and the price paid were in accordance with the Trade on its true objective interpretation, and that it has received what, looking at what the words used would convey in the admissible factual context to a reasonable person, it contracted to purchase and paid what it contracted to pay.

In other words, Exotix says that on the true construction of the Trade it agreed to purchase 22,955 GDNs for US$7,707.93 (the agreed term in relation to price being the specific US$ amount, rather than the percentage of face value used (91.5%) plus accrued unpaid interest). As explained above, that amount of GDNs, and that amount of consideration, was indeed transferred on the Settlement Date.

According to Exotix, those being the agreed terms of the Trade, they should be enforced, and LBIE is not entitled to any remedy, even though in the event, LBIE has received only a fraction of the value of the 22,955 GDNs it delivered. Mr Morpuss suggested the analogy of the sale of a picture, which the seller had had for a number of years without focusing on it having been painted by Leonardo da Vinci, and thus, having overlooked its true nature and value, sold it for a fraction of its true worth. The resulting windfall to the buyer would in such circumstances in a sense be unfair and certainly a matter of regret; but it is no basis for any different interpretation of the subject matter of the contract of sale, or for reviewing the price; nor is it a basis for not enforcing the contract or restitutionary relief.

Exotix accepts and avers that it was not aware at the time of the Trade of the discrepancy between the true value of the GDNs and the price it had contracted to pay. Its evidence is that it was only after its receipt, on 24 February 2014 (some three weeks after the Settlement Date), of a substantial coupon payment on the GDNs (totalling US$276,312.25) and having queried the coupon payment with its custodian (Standard Chartered) that it investigated the matter and discovered the mistake, which it did not reveal for some time to LBIE.

Exotix also accepts that it has obtained an enormous windfall on its onward sale of the GDNs to a third party about a year later (which it did not initially disclose to LBIE); but Exotix contends that LBIE has only itself to blame for a simple mistake as to the value of the GDNs which were transferred. Exotix did not at the time appreciate the mistake either, but it contends that that is neither here nor there, and does not give rise to any restitutionary or other obligation to compensate LBIE, still less (with particular reference to the alternative way that LBIE puts its case) to pay a price it never contemplated. The bargain may have been a bad one: but it was nonetheless the bargain, and the law does not relieve from bad bargains.

In summary, Exotix cannot deny that subjectively it understood it was buying only ‘scraps’; but it seeks to have, as it were, its bond (I refer to the ‘Merchant of Venice’) and to rely for that purpose on the ‘letter of the bond’ and its strict legal rights, as it is entitled of course in this Court to do (in the absence of any principled equitable constraint). For Exotix, the crux of the case is the meaning of the letter of the bond, in this case the Trade.

The principles applicable to the construction of commercial documents have been much explored over the last two decades since Lord Hoffman’s restatement of them in his speech in Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896, often at the highest level (see BCCI v Ali [2002] 1 AC 251; Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101; Re Sigma Finance Corporation [2010] 1 All ER 571; Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900; Arnold v Britton [2015] AC 1619; and Wood v Capita Insurance Services Ltd [2017] AC 1173).

Inevitably, these decisions have spawned many summaries, at every level. I take the following from the judgment of Lord Neuberger PSC in Marley v Rawlings [2014] UKSC 2:

“When interpreting a contract, the court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party’s intentions.”

Equally inevitably, difficulties and differences in the application of these principles continue to arise, especially as to the identification of the facts to be taken as having been known or assumed by the parties at the time the contract was executed (see (iv) in Lord Neuberger’s summary in the preceding paragraph); or, as it is often referred to, in determining the relevant factual matrix and what evidence is admissible in that context.

Two particular but inter-related difficulties in the application of these principles of contractual interpretation need special consideration in this case: (1) what knowledge

In the context of (1), there has been factual dispute as to (a) whether Mr Radicopoulos knew or should be taken to have known that LBIE held 22,955 GDNs; (b) whether Exotix knew or is to be treated as having known that too; and (c) whether both parties believed that the GDNs were worth Sol 1 each and if so, with what effect. In the context of (2), a particular issue arises as to the relevance of events and documentation subsequent to conclusion of the Trade, especially in relation to the amount of GDNs in fact delivered and to documents which came into being after the Trade on 31 January 2014 but before its settlement on 5 February 2014.

I shall elaborate further on the evidence in relation to these matters, and the application of these principles to the facts of the case, later in this judgment. For the present it is enough to say that though Exotix sought to cast wider than LBIE the “evidential net” than did LBIE, which objected to various documents and other extrinsic evidence as inadmissible, both parties sought to rely on the factual matrix in which the Trade must be set for the purpose of discerning the true interpretation of its terms, and the documentary record of the Trade and its settlement.

As will already be apparent, both parties thus relied on both oral factual evidence and expert evidence, though Exotix’s depiction of this as a “mass of evidence” is perhaps hyperbolic.

Three witnesses of fact were called.

LBIE’s only witness of fact was Mr Radicopoulos, who is now employed by Attestor Capital LLP, but at the time of the Trade, was employed by LBIE with the corporate title of Managing Director. Mr Radicopoulos headed the LBIE team that dealt with the Trade, and (as previously stated) it was in the course of the telephone conversation between Mr Radicopoulos (for LBIE) and Mr Hutton (for Exotix) that the Trade between LBIE and Exotix was concluded on 31 January 2014.

Mr Radicopoulos was a candid, mostly consistent, and credible witness. But he admitted, and it was evident, that in relation to the events and matters in question, which occurred some five years ago, his recollection was imperfect and incomplete. He accepted that much of his account in his witness statement was a reconstruction from contemporaneous records. He was also, I sensed, personally embarrassed by the mistake that had arisen.

Exotix’s main witness of fact was Mr Hutton, who was at the relevant time and is still employed by Exotix, having joined the company upon graduating from university in 2008. He is now styled ‘Director, Fixed Income Sales’. He describes his role, then and now, as being primarily to “facilitate and execute trades in emerging and distressed markets”.

Mr Hutton, like Mr Radicopoulos, accepted that he had no real recollection of the Trade. That was realistic and apparent. Regrettably, however, he was not otherwise a satisfactory witness. His evidence was sometimes incoherent. I formed the impression that he was struggling to remember his lines. He became evasive on close questioning. I had no confidence that he understood the Trade. Even in Exotix’s (his employer’s) own closing submissions it was stated that “He was confused as to many of the concepts used in the markets generally”; and the language he used in establishing the Trade is pronounced in the same document to be “imprecise and in parts wrong”. (Mr Hutton also admitted under cross-examination that he had told what he presented as a ‘white lie’ to Mr Radicopoulos in a separate transaction: he tried to cover up a large profit on the onward sale of assets (bonds) that he had made on immediate onward sale after purchasing them from LBIE “to hide the embarrassment on my part that we’d made a 3.5 margin…”) I regret to say that he did not convey a sense of commercial probity in his evidence to me.

Exotix’s other witness of fact was Mr Marron. He was and remains employed by Exotix: he is now styled an ‘Equity Sales Trader’ but at the time of the Trade he was in the back-office team that processed the Trade, which was based in Dubai. He had no first-hand knowledge of the terms on which the Trade was agreed; the purpose of his evidence was stated to be to deal with the settlement of the Trade, and the settlement instructions given by Exotix to Standard Chartered.

Mr Marron was not a satisfactory witness either, but I gained the impression that, having been reminded in the course of his cross-examination of his initial reaction to the receipt of the coupon payment relating to the GDNs was that it was an error or a mix-up, and that thereafter he had been told that he must “not speak to anyone about it”, his oral evidence was destabilised by his anxiety lest he might say something which might let Exotix down in a circumstance where it had obtained a windfall it had never expected and chosen not to reveal until the proceedings.

There were, however, two linked or related features of Mr Marron’s evidence which stood out in this connection, and on which I found his evidence notably evasive. One was that he was unable or unwilling to explain what ‘nominal’ meant in a document he described as being Exotix’s “standard template”. A second was his long subsequent equivocation as to whether a settlement instruction stating “Nominal: 22,955” meant (a) the face value or (b) the number of the GDNs to be sold, having first answered that “nominal would be “the nominal value”.

In the round, however, I agree with the view expressed in Exotix’s Closing Submissions that the witness evidence is of little or no real assistance in determining the meaning and effect of the terms of the Trade.

The overall impression with which I was left was that neither Mr Radicopoulos for LBIE nor Mr Hutton and Mr Marron for Exotix had any real understanding of the subject-matter of the Trade beyond the simple fact that it was perceived by all to be a “scrap position” which Exotix would have been most unlikely to buy except at a comparatively insignificant price and as part of an overall package comprising also other securities (the ‘Vene’) which it was far more interested in. The contrast in the credibility of the two sides’ witnesses that I have identified is at heart simply a reflection of the fact that LBIE’s Mr Radicopoulos realistically admitted both his ignorance and his lack of recall, whereas Mr Hutton and Mr Marron for Exotix sought

The expert evidence concerned four main areas: (1) the standard reference points by which a trade in GDNs is typically traded and the price of any trade is fixed: and in particular, whether, if (as was common ground in the case of bonds) GDNs are typically traded by reference to nominal value, and not number, so that the price of a trade is typically fixed as a percentage of nominal or face value; (2) whether in consequence there may be a delivery obligation for a fraction of a GDN, there is any settled practice for dealing with any fractional entitlements that may thus arise; (3) how the terms of a trade are usually documented and the purpose of post-trade documentation (especially, in this case, the VCON trade ticket and the Confirmation); and (4) whether there is a settled market practice and/or usage such as to oblige Exotix, upon discovering any mistaken over-delivery of GDNs, to return any which under the terms of the Trade it was not entitled to receive.

Professor Avinash Persaud (“Prof. Persaud”), an Emeritus Professor of Gresham College and an expert in “market functioning”, was instructed and called by LBIE. Market functioning, he explained, “means the way in which buyers and sellers in financial markets find each other, the way prices are determined and trades are executed and settled”. Prof. Persaud has had a number of senior advisory and visiting posts with (amongst other bodies) the European Central Bank (2005 to 2006) and the International Monetary Fund, as well as experience as an Investment Director in a leading absolute return investor (GAM London Limited).

Prof. Persaud provided two expert reports on behalf of LBIE, his first expert report dated 11 May 2018 (“Persaud1”), and a further expert report in reply dated 19 June 2018 (“Persaud2”).

Prof. Persaud’s expertise on issues relating to “market functioning” was undisputed and obvious; his evidence was impartial and balanced. But he had never traded a GDN, and his evidence is dependent on his assumption that they should be and are equated for relevant purposes to bonds.

Mr Andrew Kasapis (“Mr Kasapis”), a Director in the Disputes and Investigations team of Duff & Phelps LLP, London (which describes itself on its website as “The global advisor that helps clients protect, restore and maximise value”) was instructed by Exotix. Mr Kasapis told me he had traded a wide range of financial products, including fixed income securities such as bonds and assets swaps and also Latin American and other emerging market bonds. He has also worked as a Market Risk Manager for two large global investment banks, managing risk on emerging market portfolios which included GDNs; and he has had experience in trading out positions (including GDNs) in the MF Global insolvency process (which he described as liquidation though MF Global is in fact in special administration). Mr Kasapis provided his expert report on behalf of Exotix dated 1 June 2018.

Mr Kasapis is plainly an experienced trader; but I was not convinced that his experience was as extensive in relation to the trading of GDNs as initially he sought to suggest, nor indeed sufficient to establish any fixed mode of trading.

Further, his central assertion that, whereas bonds trade by reference to nominal value and it is only necessary and thus standard practice when they are traded only to state that nominal value and a price as a percentage of it, by contrast GDNs are traded in units and thus it is necessary and standard practice to set out the quantity by unit as well as the notional amount and price as a percentage of it, did not match the factual position in this case that the Trade was fixed by reference to (i) the face or notional amount of the GDNs being traded and (ii) a price as percentage of that notional amount, with no express statement of the number of GDNs. It also appears to be inconsistent with Exotix’s internal email and communications with potential purchasers that bids for the GDNs alongside four different bond positions were sought by reference to their face amounts / nominal values (thus the use of the “$” or the “PEN”) and not by reference to the number of GDNs available, and that for these purposes the GDNs were equated with bonds.

The two experts signed a memorandum of agreement and disagreement on 22 July 2018 (the “Memorandum”).

There was no material disagreement between the experts as to the phases of a GDN trade process (what Mr Kasapis called “GDN trade flow”); the four key stages which they identified being (i) when the trade is agreed; (ii) when after agreement, the trade and its details are recorded (here, and usually, by a VCON trade ticket); (iii) when the trade is confirmed by the back-offices/settlement teams of the parties to the trade (in the Confirmation); and (iv) when the trade is settled by the exchange of cash for the financial instrument acquired, and delivery accordingly (Settlement), usually through a settlement agency (in this case, Euroclear). These stages are reflected in the description given above of the factual position in the present case.

The most important remaining areas of disagreement between the experts concerned

As to (1) in paragraph [98] above, Mr Kasapis stressed that (as is plainly the case) GDNs represent entitlements in respect of, but are not the same as, the underlying bonds. Each GDN is a derivative separate legal instrument created by Citibank, with separate terms and a denomination created by Citibank. The holder of a GDN has the right to surrender the GDN and receive a bond, or to be paid by Citibank (in US Dollars and following the deduction of fees) whatever is paid out to Citibank by the Government of Peru. The holder has no entitlement to be paid a fixed amount of debt by Citibank. The holder has no direct recourse against the issuer of the underlying bonds (the Government of Peru); only against the issuer of the GDNs (Citibank). Holders of GDNs may (in part) think of them economically by reference to the underlying bonds, and that may be why Bloomberg includes a par value for them (although it also includes one for Warrants, which it is common ground have no par value). However, the whole point of owning a GDN is that it is not the same as the underlying bond. Mr Kasapis contended that in such circumstances it is the fact, and unsurprising, that GDNs trade differently from bonds.

Prof. Persaud, accepted (of course) that GDNs are separate instruments and that the holder’s rights are not against the issuer of the underlying bonds but against the issuer of the GDNs (Citibank); but he adhered to his contention that the measurement of the value of a GDN, as of the value of an underlying bond, is by reference to its nominal or face value; and accordingly, the normal reference points for a GDN trade are (a) the notional value of the bond or GDN position and (b) the price as a percentage of that notional value.

As to (2) in paragraph [98] above, the disagreement between the experts reflected their disagreement on the way GDNs are traded. The problem does not arise if GDNs are traded by number rather than face or notional value. Mr Kasapis, logically and predictably, was adamant that that being so there is no market practice referable to fractional trades, just as there is not in what he suggested was the analogous market of trades in American Depositary Receipts (“ADRs”, which likewise represent but are separate from the underlying securities in question). There was no need for and no evidence of any standard practice.

Prof. Persaud, on the other hand, stuck to his line that there is a general market expectation that where a trade is entered into which gives rise to a fraction of a security requiring to be delivered, settlement will be on the basis of ‘rounding down’ the number of securities to be delivered to the nearest whole number and to “cash settle” the fraction. This market practice applies, as a default, in the absence of alternative provision being made by the parties at the time of the trade although, of course, they can agree to contract out of the convention and settle the trade on an alternative basis (for example by rounding up and the buyer paying pro rata for the fraction of the security required to make a whole).

I shall return to the dispute between the experts on these and other issues later. Before doing so, it is convenient to identify in more detail the dispute between the parties as to the admissible factual matrix and documentation.

Material agreed to be admissible and relevant

LBIE submitted that the admissible factual matrix (leaving aside the expert evidence of market practice and usage) comprised and should be confined to the following:

Additional categories of evidence which Exotix contends is admissible

Exotix identified five further categories of evidence (other than the evidence already related above, and expert evidence) as “potentially relevant and admissible”, as follows:

In line with these categories, Exotix more particularly seeks to rely on the following as admissible as part of the matrix of fact by reference to which the terms of the Trade must be construed:

LBIE’s objections as to admissibility of the above material

LBIE does not accept that all or even most of the material thus adumbrated is admissible. Stressing that, though an oral contract, there is no dispute in the present case as to the precise words which were spoken and which established the Trade, LBIE especially objects (in some instances in rather generic terms) to the admissibility of the following material on the issue of interpretation:

More particularly, LBIE contended that the following documents and matters comprised inadmissible extrinsic evidence as distinct from admissible factual matrix evidence:

My assessment of material admissible and relevant as part of the factual matrix

As to (1) in each of paragraphs [1057] and [1068] above, in my view, evidence of knowledge or information actually known to both parties prior to the Trade is plainly admissible as part of the factual matrix which should be taken (in assessing the response of a reasonable and objective observer to the words in which the parties expressed themselves) to have informed the parties in making their agreement. But I do not think there can be any doubt that the purpose of considering that as part of the admissible factual matrix is to determine the objective intention emerging from those words, and not to show what were the parties’ subjective intentions or understandings.

As to (2) in each of paragraphs [1057] and [1068] above, the question as to what knowledge or information is to be treated as being ‘reasonably available’ to the parties for the purposes of constructing the words they used remains, to my mind, a particularly difficult one. As was emphasised by Briggs LJ (as he then was) in Gladman Commercial Properties v Fisher Hargreaves Proctor [2013] EWCA Civ 1466, and also in my own decision in Challinor v Juliet Bellis & Co [2013] EWHC 347 (Ch) at [277] (cited in Lewison, The Interpretation of Contracts (6th ed) at 3.17(d)), the test of “reasonable availability” is not always easy to apply and requires restraint in its application: and all the more so given the almost unlimited information and knowledge now available through the internet.

Indeed, there is recent authority in the Court of Appeal to support the exclusion of matters “which the parties might have discovered but did not in fact discover” (see per Jackson LJ in Hamid (t/a Hamid Properties) v Francis Bradshaw Partnership [2013] EWCA Civ 470); and in Revenue and Customs v Secret Hotels2 Ltd [2014] UKSC 16; [2014] 2 All ER 685 Lord Neuberger PSC said that:

“When interpreting an agreement, the court must have regard to the words used, to the provisions of the agreement as a whole, to the surrounding circumstancesinsofar as they are known to both parties, and to commercial common sense.” [my emphasis]

However, as pointed out in Lewison on ‘The Interpretation of Contracts’ 6th ed. at [166], there is other authority, and it seems more consistent with the objective approach, to support a widening of the scope beyond what the parties actually knew. I attempted the following summary in Challinor v Bellis at [277]:

“(1)

At least where there is no direct evidence as to what the parties knew and did not know, and as a corollary of the objective approach to the interpretation of contracts, the question is what knowledge a reasonable observer would have expected and believed both contracting parties to have had and each to have assumed the other to have had, at the time of their contract;

(2)

that includes specialist or unusual knowledge which only parties entering into a contractual engagement of the sort in question might reasonably be assumed to have; and it also includes knowledge which it is to be inferred, from the nature of the actions they have in fact undertaken, that they had or must have had;

(3)

however, it does not include information that a reasonable observer would think that the parties merely might have known: that would open the gate too far to subjective or idiosyncratic speculation;

(4)

the fact that material is readily available or notorious may support an inference as to what the parties actually knew;

(5)

but (subject to (6) below) where it is demonstrated that one or more of the parties did not in fact have knowledge of the matter in question such knowledge is not to be imputed; nor is the test what reasonable diligence would or might have revealed: in either case that would be inappropriately to introduce impermissible concepts of constructive notice or a duty (actionable or otherwise) to make inquiries or

investigations;

(6)

the exception is that a reasonable person cannot be assumed to be in ignorance of clear and well known legal principles affecting or incidental the contractual engagement in question.’’

I do not think that the test of admissibility in the case of information which did not in fact “cross the line” (or in other words, was not in fact shared information) is whether there is evidence that in fact the information available to one party would have been made available to the other had the other party only asked for it. Further, in my view, the Court needs to be wary of assuming that the general availability of information is sufficient to make it “reasonably available” in the requisite sense: almost anything is available on the internet in the general sense. In my view, and subject to paragraph [114] below, that phrase envisages and requires to be made an objective judgment as to whether a reasonable man, had he known the other party to have that information, would have supposed it to be necessary in order to make sure of a proper understanding of the contract, and if so, whether he would have been likely to encounter any real difficulty in obtaining it.

A further point, reflecting paragraph (5) of my attempted summary quoted in paragraph [112] above, is that it seems to me that if a fact or circumstance is demonstrated positively not to have been known by either party, it is not to be supposed that the hypothetical reasonable observer standing in their position would

I consider that cases such as Tidal Energy Ltd v Bank of Scotland Plc [2014] EWCA Civ 1107 and Norcross v Georgallides (Estate of) [2015] EWHC 2405 (Comm) reflect just such judgments (with the Tidal Energy case illustrating the difficulty sometimes in making it, since the Court of Appeal was divided, and the Norcross case being an example of the Court accepting that parties contracting in a specialised market may find it difficult to dissuade the Court from proceeding on the footing that a reasonable man would have expected them to obtain readily available information as to the practices in that market).

As to (3) in each of paragraphs [1057] and [1068] above, I consider that a distinction must be drawn between documents which record and evidence the contract, and are exchanged between the parties for that purpose and as part of the contract process to ensure that the parties are in agreement as to all the terms, and the post-contractual internal recording of arrangements for accounting, regulatory or similar internal requirements, and which are not produced or intended for any inter-party purpose.

The question to be answered is whether the document in dispute is relied on as part of the contractual documentation establishing the Trade, intended to have contractual effect, and which falls to be interpreted, or whether that disputed documentation is not in itself contractual in effect, and in reality is post-contractual material sought to be deployed as a tool in the interpretation of the Trade.

In my view, and applying these principles, it seems to me that the contractual documentation plainly and indisputably intended to have contractual effect and which this Court must interpret are (a) the record of the words of offer and acceptance and (b) the VCON or Trade Ticket.

However, I consider, contrary to LBIE’s submission, that the Confirmation is also admissible. That is so even though I accept that it was produced, some three days after the event, by the “back office” at Exotix, and neither Mr Radicopoulos nor Mr Hutton saw it at the time or had any direct input into it. In my view, it should nevertheless be included in the admissible factual matrix, essentially because it was also one of the documents expected to be generated to record, and in its case enable settlement of, the Trade. I would have expected it to be in any contemporaneous ‘bible’ of the documentation evidencing the Trade.

I also consider that the manner of its generation in the Bloomberg system, as well as the VCON/Ticket itself, is part of the admissible factual matrix. As explained previously, the VCON on its face records that the Trade was for: (i) GDNs with a nominal value of S/22,955; (ii) for a price of 91.5% of that nominal value. The only way Mr Hutton can have created this VCON was if he entered the following two values (consistently with LBIE’s case): (i) a face amount (i.e. nominal value) of S/22,955; and (ii) a price of 91.5%. So much is clear from looking at (i) the

The potential significance of Mr Hutton’s selection of the “M” option is this. Had the Trade been for 22,955 GDNs, it would not have been appropriate: the “X” input, which allows the price to be inputted as a non-scaled price per unit rather than as a percentage of the face amount, would plainly have been the appropriate option. It was suggested by Mr Morpuss that none of this was admissible, as being evidence of Mr Hutton’s subjective intent, and that in any event Mr Hutton had insisted under crossexamination that he considered the Trade to have been for 22,955 GDNs (rather than GDNs with a notional value of S/22,955), and that his selection of the “M” option was simply a mistake, which he further suggested “lies with Bloomberg” in not clearly identifying the appropriate option and in providing for both GDNs and Warrants (and other non-bond trades) to be entered by reference to ‘Face Amount’ , even though in a case such as Warrants, it is common ground that that was a reference to quantity, not face value. Further, Mr Hutton was insistent that he assumed that each GDN had a par value of Sol 1: and on that basis it made no difference whether a party inputted quantity or face value, both being the same. Accordingly, Mr Morpuss submitted, neither the generation nor the resulting form of the Ticket should be taken to be inconsistent with a trade of 22,955 GDNs: indeed, in his written submissions in closing he went further and asserted the Ticket and its generation to be “entirely consistent with a trade of 22,955 GDNs”.

I do not accept this. Even if admissible, I do not accept Mr Hutton’s explanation of his approach. Mr Hutton used the “X” input for the ticket reflecting the purchase of the Vene (a security which, it is common ground, trades by reference to units) at materially the same time as he selected the “M” input for the GDNs. This not only demonstrates that he selected the “M” input deliberately, but it also demonstrates that he treated the GDN trade as if it were, or were akin to, a bond trade (rather than a unit trade in something like a share or warrant). In my judgmemnt, that it’s the true explanation of the inputs into the Bloomberg system and the genesis of the VCON ticket

As already indicated in paragraph [232] above, the admissibility and relevance of the LBIE Sign-Off Pack and the BONY spreadsheet attached to it showing the true extent of LBIE’s holding of GDNs was disputed.. Applying the criteria I have sought to identify previously, I do not accept Exotix’s contention that these documents should be treated (a) as part of the factual matrix and (b) as demonstrating the true subject matter of the contract, on the basis that either the document or the information it contained could and would have been made available on request with ease. I accept the evidence of Mr Radicopoulos that the Sign-Off Pack and Spreadsheet were internal documents, not intended to be or in fact ever made available to anyone

I do not see that I should suppose that a reasonable person would have sought to check whether LBIE had more to sell by seeking from LBIE its internal documents showing its aggregate holdings. The terms of the Trade provided for the sale of a

In such circumstances, I find it quite plausible that the parties did not stop to ask themselves (and each other) the par value of each GDN or precisely how many GDNs were being traded.

In that context, I should add that although Mr Hutton stated in his witness evidence and sought to maintain when cross-examined that at the time of the Trade “and for some time thereafter” his belief remained that “each GDN was referable to an underlying Peruvian bond with a face value of PEN 1” and that accordingly “each GDN was worth PEN 1”, the truth is, in my judgment, that neither Mr Hutton or Mr Radicopoulos knew or cared, nor should a reasonable observer be taken to have had reason to find out, what the par value of each GDN or the number of GDNs comprised in the Trade was: for both it was treated like a sale of bonds by reference to notional value, as in effect both expressly or impliedly accepted. I do not think it would be appropriate or legitimate to attribute to either of the contracting parties a different outlook in making the Trade.

The possibility or even fact that had Mr Radicopoulos been asked to check what the extent of LBIE’s holding was that would have been simple and would have revealed that, contrary to the shared understanding of the parties at the time of the Trade that the position was a ‘scrap position’ of little value, the holding was a substantial one worth many millions, does not seem to me to be admissible where the issue is as to what was meant by a trade of a specified quantity of GDNs thought by both parties to be a ‘scrap position’ described by reference to their assumed notional/par value at a specified price commensurate with that assumption.

I appreciate that Mr Radicopoulos’s failure to check may suggest carelessness of a surprising and concerning degree; and that it might be said, and indeed is said by Exotix, that LBIE had only itself to blame if when making the Trade Mr Radicopoulos failed to appreciate the true extent and value of LBIE’s total holding, and that any complaint about Exotix taking advantage of the situation is misplaced in consequence. But that does not seem to me to be a matter going to interpretation: the Trade was by reference to a stated subject-matter, not by reference to an unquantified holding, and evidence as to what in truth the extent of the holding was does not assist in determining what the parties should be taken to have meant by the words they used. That is especially so when it is quite clear on the uncontroverted and indisputable facts that both parties proceeded on the footing in fact that the GDNs were a scrap position with a comparatively small value.

More generally in this context, it seems to me that the Court should take care not to import notions of reasonable care and negligence into questions of contractual construction. It is a slippery slope between identifying what the actual context of a contractual engagement was, and (by contrast) what parties exercising reasonable care might reasonably have been expected to seek to make enquiries about. The proposition that the admissible factual matrix should include information “reasonably available” to the parties is not, in my view, intended to impose or connote a duty to enquire as to matters which on the basis of their shared understandings did not merit inquiry. The contractual intentions of careless parties should be honoured, and their bargains should not be corrected by reference to what they would or might have intended to do had they been less careless.

With reference to (4) in each of paragraphs [107] and [108] above as to the admissibility of post-contract events, Exotix seeks to rely on the letter before action (referred to in paragraph [110(4)] above) in which at that stage (the letter is dated 17 June 2016) LBIE’s solicitors (Linklaters) contended that the Confirmation was clearly erroneous in (amongst other things) “referring to the quantity of GDNs to be transferred as USD22,955 rather than 22,955, in circumstances where there was never any reference in the parties’ discussions to selling the GDNs by reference to a fixed US dollar amount…”

The letter, which put forward what is now LBIE’s alternative case that what had gone wrong was not the statement of the quantity but the statement of the price (which on that view should have been USD7,706,016.68) is an obvious embarrassment to LBIE.

As to (5) in paragraphs [1057] and [1068] above, Exotix also sought to rely on certain post-contract conduct of LBIE, the very fact that it delivered 22,995 GDNs, and in particular, on the instructions which it contends must have been given by LBIE to BONY (as its custodian) to effect that delivery to Exotix in return for the agreed price of US$7,707.93, and upon what it described as the “matching instructions” which Exotix itself gave to its own custodian bank, Standard Chartered Bank, Mauritius. In my view, none of this post-contract behaviour is admissible either.

Though advanced with skill and moderation, I do not accept Mr Morpuss’s submissions on behalf of Exotix that recourse to evidence of subjective intention and post-contract conduct is permissible to determine what was the subject-matter in this case. Mr Morpuss relied in particular on the decision of Chief Justice Spigelman in the New South Wales Court of Appeal decision in County Securities Pty Ltd v Challenger Group Holdings Pty Ltd [2008] NSWCA 193 in which he said, at paragraph 14:

“Even in the case of a written contract, the words identifying the subject matter being bought and sold may be susceptible to more than one meaning. This is one well established category of ambiguity, so that extrinsic evidence is admissible to identify the subject matter, even on a restrictive approach to the use of extrinsic evidence in the course of contractual interpretation.”

I do not consider that the County Securities case has any application or is of any assistance in the case in hand. That case concerned a transaction with two parts, one of which (for the transfer of certain Equity Swaps) was wholly in writing; and the other part of which (a hedge involving the acquisition of certain shares and the assumption of certain margin obligations) was not in writing, and there was no evidence of any conversation which might have established the terms: the second part of the transaction was sought to be inferred from conduct alone. The learned chief justice emphasised at paragraphs [4] and [7] that (a) the surrounding circumstances to which the court’s attention was invited had regard only to that part of the transaction that was not in writing and (b) “The issue is not one of interpretation, because there are no words to interpret. The issue is one of fact: what terms did the parties agree?” In the present case, there is a record of the Trade; the terms requiring interpretation were recorded and the question is what the words mean: it is a legal issue of interpretation.

For the avoidance of doubt, I do not think there is any viable suggestion that the parties intended the terms of the Trade to be gathered from other sources than their recorded exchanges constituting the Trade and the VCON or Trade Ticket which the parties expressly envisaged and agreed (on the 16:20 telephone call) would be produced to memorialise it, and which to enable settlement were to be confirmed in the Confirmation. Insofar as that is a question of fact, I find that there was no intention for the terms of their trade to be gathered from other sources or information. This is not, therefore, a case such as was posited by Lord Hoffmann in Carmichael and Anr v National Power PLC [1999] 1 WLR 2042 where the parties have left their agreement to be inferred from their conduct, or must be taken to have accepted that it would be.

Nor, in my view, is this a case like Savory Ltd v The World of Golf Limited [1914] 2 Ch 566, which required the Court to ascertain the subject matter of an assignment of copyright in, inter alia, “four golfing subjects”. Neville J held in that case, at pages 573 to 574:

“Then it is said that there is not a sufficient description of the subject-matter of the memorandum to be found in it, the subject being “four golfing subjects.” Now it is said that I cannot or that I ought not to admit parol evidence to identify those four golfing subjects, and that if I do not do that I cannot tell what particular golfing subjects were intended and referred to in the document itself. It appears to me that the cases that have been referred to shew clearly that I am entitled to receive evidence for the purpose of identifying the subject-matter of the contract. I think that both Shardlow v. Cotterell [(1881) 20 Ch. D. 90] and Plant v. Bourne [[1897] 2 Ch. 281] are authorities to the effect that parol evidence to identify the subject-matter of a contract is admissible. Here I think there can be no doubt upon the construction of the memorandum itself that four particular golfing subjects are referred to, and it seems to me that the difficulty in both the cases that I have referred to which was under consideration was whether the terms of the agreement indicated that no particular property was intended, but something which might be selected by one of the parties hereafter. In one case it was twentyfour acres of land in such and such a parish, and it was said there that you could not say that that was an agreement to sell any particular twenty-four acres, and it was held that you could shew by parol evidence that a certain twenty-four acres had been marked out and that they were the subject-matter of the contract between the vendor and the purchaser. I think therefore in this case I am entitled to hear evidence as to what the four golfing subjects purchased by the plaintiffs from Mr. Thomas were. In my opinion the evidence shews that those four golfing subjects included the picture “Thirteen Down.” That is the subject-matter of the present action. In my opinion, therefore, the assignment of the copyright is sufficiently shewn by the memorandum in writing signed by the proper party.”

The World of Golf case was concerned with the identification of the particular items that constituted the subject matter of the contract, which cannot be ascertained from the contract alone. The question was which particular golfing subjects comprised the subject-matter of the contract. I accept Mr Bayfield’s submission that the present case cannot be compared to cases such as The World of Golf. This is not a dispute about which particular GDNs LBIE was offering to transfer. Rather, LBIE’s Primary Case turns on how Mr Hutton’s offer to buy “22 just on the shy of 22..23 thousand er sol”, as confirmed in the VCON by reference to “22.955 (M)of PERU 6.9 08/12/37”, should be interpreted: i.e. was it a reference to GDNs of a face amount of PEN 22,955, or was it a reference to 22,955 GDNs. That is a question of construction rather than identification of subject matter, and therefore ultimately a question of law by reference to the admissible factual matrix.

I also do not consider that I should depart from the rule against the introduction of parol evidence on the basis that this is an oral agreement, as: (1) there is no dispute as to what Mr Hutton and Mr Radicopoulos actually said on the 16:20 call; and (2) it appears from the transcript of that call that the traders intended that the terms of the Trade should be recorded more precisely in the VCON. Whilst I appreciate that the words used on the 16:20 call and in the VCON may be susceptible to alternative interpretations, the way in which the law deals with that is to place greater emphasis on the admissible factual matrix, without descending into matters such as the subjective intentions of the parties, post-contractual conduct and information only available to one party, such categories of evidence being inadmissible when terms are being construed (being a question of law), as opposed to being identified (being a question of fact).

Further, even if that is wrong, I do not think this material is as telling as Exotix suggested. I accept that there is considerable forensic force in the point as made by Exotix that it “cannot be the case that LBIE instructed BONY to transfer GDNs having a face value of Sol 22,955, because BONY knew the correct par (and market) value of the GDNs, and would have understood such an instruction to mean that it should attempt to transfer 22.955 GDNs – which it did not attempt.” However (a) there was no direct evidence on, and it remained unclear, what were the terms of

In my judgment, the issue is indeed one of interpreting the words constituting, memorialising and confirming the Trade, and of discerning the objective intention of the parties from the words they used in their admissible context: it is not one of identifying what were the terms, which is a question of fact; and as to the distinction see Sea Containers Services Ltd [2012] EWHC 2547 (Ch), at [88].

The question of interpretation is as to the words used to identify the subject-matter of the Trade; the difficulty is the imprecise and colloquial way in which the parties expressed themselves in the admissible documentation:

However, subject to Exotix’s contention that the reference in the Confirmation to USD is simply incorrect and should be excised, the common thread is that both the subject-matter and the price are identified by reference to face or nominal amount stated in terms of a monetary total: the quantity being expressed, not by reference to a unit, but by reference to a monetary amount, and the price being stated as a percentage (91.5%) of the monetary amount, being face or nominal value. Further, and as a consequence, the price is stated in each case to be US$7,703.93, being the market value of the GDNs, on the agreed basis of valuation, with a face or nominal value of Sol 22,955.

The words used by Mr Hutton and the terms of the Trade then agreed do seem to me to be plain in their intent. My own impression as to their meaning was further supported by the evidence of Exotix’s own version when confronted with them. In the course of Mr Bayfield’s cross-examination of Mr Kasapis the following exchange took place:

“Q. If they’d intended to do the trade with reference to a quantity of

GDNs, a number of GDNs, then Mr Hutton wouldn’t have referred to buying the just shy of 23,000 sol at 91.5, would he?

A.

Reading what he said, then I would suppose not, but I’m not quite sure Mr Hutton clearly understood the underlying par amount of the GDNs of what he was trading at the time.” [My emphasis.]

The emphasised words represent the response to the natural meaning of the recorded terms of the Trade from an informed observer with expert knowledge of the market and (it is to be assumed) of the relevant background. Mr Bayfield, on behalf of LBIE, submitted in closing that “ultimately, the case is as simple as that”, and that the plain intent of the parties as evident from the words of the Trade and in particular the identification of what was to be sold by reference to face amount expressed in sol, and the expression and calculation of the purchase price as a percentage of that face amount without any mention of any unit quantity, demonstrates clearly and unequivocally that the Trade was GDNs with a stipulated face value of GDNs, and not a quantity of GDNs. Put another way, the parameters chosen to define the sale confirm that the subject matter was GDNs with the stipulated face amount at a price equal to a percentage of that amount; and not a number of GDNs at a price per unit.

Exotix’s case depends, then, upon extrapolating from (a) the factual matrix with the broad ambit it contends is admissible (b) the expert evidence and (c) the difficulty, if not impossibility, of a fractional trade (which is the inevitable corollary of interpreting the terms of the Trade as a sale of more/less than a whole number of GDNs) a different true intention than a reading of the words used to establish the Trade would (at least initially) suggest.

I have explained earlier why I consider that the admissible factual matrix is more confined than Exotix submitted it to be. I do not consider that there is any sufficient indication in the admissible factual matrix to warrant a different meaning for the words used than the meaning that they initially appear to convey. On the contrary, I consider that the admissible factual matrix confirms that the objective intention of the parties was to sell GDNs with a total nominal, face or par value of PEN 22,955 for a

Accordingly, in my view, on an objective interpretation the provisionally most likely available meaning of the Trade is that at the time it was agreed the parties’ agreement was for the sale and delivery of GDNs with a face value of Sol 22,955 at a price of 91.5% of par/nominal value (amounting to $7,707.93).

I must turn to consider whether the interpretation that I consider most naturally fits the words is nevertheless displaced by the other factors already mentioned, and especially (a) the problem that on that interpretation LBIE had committed to deliver less/more

Of these factors, it seems to me that (a) and the problem of a delivery obligation for a fraction of a GDN, is the most difficult. It was this which Exotix presented as its trump card. An interpretation which would result in impossibility of performance plainly requires reconsideration as to whether that interpretation can truly be correct.

It is necessary in this context, therefore, to consider whether there should be read into the agreement a provision dealing with the problem of fractions; and if not, whether an interpretation of the agreement which leaves the problem extant, so that the agreement cannot be performed, is truly the only available satisfactory interpretation, or should prompt an iterative reassessment of the conclusion which has given rise to the problem.

LBIE’s primary answer to this problem was to assert an implied term based either on (i) market practice or (ii) necessity and obviousness. In its Reply, this was pleaded as follows:

“Given that LBIE could not deliver a non-integer quantity of GDNs to Exotix, it was an implied term of the Trade, in particular (but without limitation) on the basis of market practice and/or usage and/or for reasons of business necessity, that LBIE was required to:

i.

Deliver 22 GDNs to Exotix; and ii. Pay Exotix the cash equivalent of 0.955 GDNs…”

It is, of course, plain, but nonetheless important to bear in mind, that the process of implication is at heart simply reading into the contract those terms which the parties are taken to have intended to include but failed or felt it unnecessary to make explicit. The court cannot mend nor can it alter bargains: it cannot add ad hoc terms for the sake of fairness and convenience, for that would be to change the bargain made, which is no part of the court’s function. As Lord Bingham MR put it in Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 at 482:

“The question of whether a term should be implied, and if so what, `almost inevitably arises after a crisis has been reached in the performance of the contract. So the court comes to the task of implication with the benefit of hindsight, and it is tempting for the court then to fashion a term which will reflect the merits of the situation as they then appear. Tempting, but wrong…”

This is an unusual case, where the parties have in fact apparently performed the contract; but on one interpretation such performance was not in accordance with the contract; and that is the interpretation I have been persuaded is correct. Another curious feature is that although I have reached a clear view as to the objective meaning to be given to the words used set in their admissible factual matrix, I am also clear that in reality, the parties were under a fundamental misunderstanding about the nature of GDNs and did not contemplate for one moment any difficulty in agreeing terms by reference to a notional or par value of GDNs. Any implication of terms to deal with the problem of fractions to which the Trade inexorably gives rise would be to cover a problem that in fact the parties themselves did not for a moment contemplate. Rightly or wrongly, I have found this conundrum and the tension between the objective approach required and the subjective position difficult to deal with: assessment of the “reasonable expectation of the parties” seems to me to depend critically on whether on revelation of the difficulty the parties would have sought to give effect to the contract or abandon it. I return to this conundrum later.

Leaving it on one side, for the present, and borrowing Lord Bingham’s words, the crisis is that the contract, if it bears the meaning that I consider it should in accordance with the requisite objective approach, is not capable of being performed in accordance with its express terms. The question is whether the crisis can legitimately be resolved by implying into the contract a term on the basis of establisheding custom and/or business necessity. In such a context Lord Bingham’s warning that, though sometimes beguiling, it is invariably wrong to imply a term simply to reflect the merits is of particular resonance and importance.

As apparent from its pleading (see paragraph [151] above), to justify the interpolation or implication of a term to deal with fractions, LBIE relied on a composite approach, having recourse both to what it submitted the expert evidence established to be a clear trade practice and on commercial necessity.

The Crema case is the leading modern authority as to the former; Marks & Spencer plc v BNP Paribas [2015] UKSC 72 is the leading modern restatement of the law relating to the latter. Both lines meet at the point of the whole nature and purpose of the process of interpolation or implication, which is to spell out the entirety of the contract by adding to the express terms implied terms which either because the parties are to be assumed to have wanted their contract to have commercial efficacy or because custom and usage would treat the unstated words to be part of the contract unless expressly excluded, should be read in as if they had always been intended to be included. As Aikens LJ put it in Crema v Cenkos Securities Plc [2011] 1 WLR 2066) the court’s task is:

“to see whether the proposed implication spells out what the instrument would reasonably be understood to mean.”

In the context of implication on grounds of necessity (or a related basis of ‘obviousness’) although the test is not one of “absolute necessity” (ibid.) and also, as Lord Wilberforce recognised in Liverpool City Council v Irwin [1977] AC 239, “the degree of strictness seems to vary with the current legal trend”, the test is a stringent one. As already apparent from the BNP ParibasSkyBcase quoted in paragraph [156] above, reasonableness is not sufficient (paragraph [23], and also Ali v Petroleum Company of Trinidad and Tobago [2017] UKPC 2 at [7]).

In the context of the interpolation of terms on the basis of a trade custom or usage, the term in question must be demonstrated to be “invariable, certain and notorious” (again, per Aikens LJ in Crema at [6]) such that the parties must be taken to have intended its inclusion unless they have expressly stated otherwise.

LBIE appeared to me to rely principally on trade practice; and that was certainly the principal focus of Exotix’s response in contending that no term should be implied, and that the lack of a saving term should cause the Court to reconsider the interpretation which had caused any need for it.

As explained previously, the experts were at odds as to whether there is any established trade practice or usage. Their disagreement in part reflected their conflicting views as to whether GDNs are or can be traded by reference to their nominal value (like bonds, as Prof. Persaud contended) or whether they are invariably traded by reference to their number (like shares or warrants or similar instruments, as Mr Kasapis contended). However, it seemed to me that, ultimately, the real disagreement was as to whether the usual response of parties to the ‘crisis’ of fractional entitlements (whether in the context of bonds, shares or ADRs) amounted to an implicit contractual commitment, or simply good (and usually expected) behaviour.

Neither side contended that either expert was doing other than his best to assist the court; but Mr Morpuss submitted that in those circumstances “one of them must be wrong” and that the evidence of Mr Kasapis, as someone with actual experience of trading a GDN should be preferred to that of Prof. Persaud who had no real market experience. Further, although he accepted that “whilst each expert asserts that he is right, neither has been able to produce independent evidence to support his assertion”, Mr Morpuss sought in this context to rely as “the best independent evidence that the Court has as to market practice” on (a) the BONY spreadsheet (but this time, he submitted, not for the purpose of interpretation but in illustrating market practice and supporting Mr Kasapis) and (b) “the understandings of the back offices of LBIE and Exotix as to how to settle the Trade” (again for the same purpose).

I have no reason to doubt Mr Kasapis’s evidence that his own experience when trading GDNs was that “trades in GDNs are agreed on the basis of a number of GDNs (as opposed to the underlying bond notional); and I would tend to accept his explanation that in the ordinary case this is likely to be because “traders and brokers will, generally, correctly appreciate that each GDN relates to a specific number of underlying bonds, each with a notional value, as structured by the Depositary (Issuer) Bank, in this case, Citibank”, distinguishing the GDN (ordinarily traded in units and having in strictness no ‘notional value’) from the underlying Peruvian bonds (having, and being traded by reference to, their notional value).

I would accept too that this general appreciation may have been reflected in the BONY spreadsheet, though confusion appears to me to be a sounder explanation (as Prof. Persaud suggested) of the actions of the back office. However, this was not a run of the mill case. Far from it: on the basis of my view as to the true interpretation of the words used by the traders and the wording of the VCON and Confirmation, this particular trade was for the sale and purchase of a notional value of PEN 22,955 at a price of 91.5% of that notional value, calculated as US$7,438.14 (excluding accrued interest). In this particular case, in other words, they were traded as if they were bonds, and not as if they were units or shares.

Both experts did agree that bonds are typically traded by reference to their notional values; and both also agreed that in some situations, a bond trade could throw up a fractional entitlement.

Prof. Persaud’s evidence was that in such a situation, that is, where a trade involves a fraction then the fraction will not be included in the settlement process and the whole bonds would be settled and cleared with the balance of the trade being settled in cash.

Mr Kasapis did not agree with Prof. Persaud that ADRs would often be traded in fractional amounts (though I did not take him to say it was not possible); and was rather equivocal as to any standard practice in the context of bonds, preferring to distinguish GDNs as invariably traded by whole units. He conceded that a ‘plain vanilla’ bond trade might give rise to a fractional entitlement, ‘and that it would be more usual than not for ‘rounding down’ to be agreed, even in that context, he did not concede there to be any invariable practice, suggesting that this would be agreed ad hoc (either before or after the trade) by the parties, and sometimes (especially in an illiquid market) a ‘rounding up’ solution might be agreed.

I have concluded that there is simply not enough evidential basis to establish an “invariable, certain and notorious” practice such as to satisfy the strict Crema test.

Then, the question is whether the pleaded implied term should be read into the Trade on some other basis, and in particular on the basis that a notional reasonable person in the position of the parties at the time they were contracting would have considered it both obvious and necessary to provide commercial and practical coherence to the agreement (see Marks & Spencer plc v BNP Paribas Securities Services [2016] AC 742 at [21]) and (in other words) to give the Trade business efficacy.

The determination of what is obvious and truly necessary is never easy, especially when the temptation to mend is at its strongest. In this case, the difficulty is exacerbated by the fact that the Trade, on the construction which invites and (it is said) necessitates implication of a term, plainly (in my view) proceeded on the basis of a false understanding shared between the parties as to the true nature of the GDNs (that is, the subject matter).

Thus, if recourse were to be had to the primary test usually put forward, and again reflected and confirmed in Marks & Spencer (and see [16] and [19]), of posing the hypothetical question whether the parties, on it being suggested that the desired term should be implied, would both have testily agreed that such term was so obvious as to go without saying, the likelihood in this case is that the parties would have been jolted into recognising the real problem, which is that although they were ad idem as to what they respectively intended to sell and purchase (a scrap parcel of GDNs with a nominal value of PEN 22,955 at a commensurate price fixed as 91.5% of such nominal value) they were both under a fundamental misapprehension about the nature of GDNs, assumed that the par value of each was one PEN, and thus failed to appreciate that a parcel of GDNs with a par value of PEN 22,955 was not capable of delivery since in fact each GDN has a nominal value of PEN 1,000, a parcel of GDNs with a nominal value of PEN 22,955 entails a fraction, and a fraction of a GDN cannot be delivered. Put another way, the implication of a term in this case would be to save the contract from a misunderstanding rather than an obvious omission.

That does not, however, exclude the possibility of implying a term to ensure the workability of the agreement on which, at the time, the parties were ad idem. That is the other standard basis on which a term may be implied. Both parties in this case occasionally appeared to elide the two; and as Lord Hughes stated in Nazir Ali v Petroleum Company of Trinidad and Tobago [2017] UKPC 2 (at [7]) “Usually the outcome of either approach will be the same”. But not invariably; although as Lord Hughes went on:

“The concept of necessity must not be watered down. Necessity is not established by showing that the contract would be improved by the addition. The fairness or equity of a suggested implied term is a essential but not a sufficient pre-condition for inclusion.”

The test was perhaps most helpfully explained by Lord Sumption JSC in argument in the Nazir Ali case as ‘’being that a term can only be implied if, without the term, the contract would lack commercial or practical coherence” (ibid.). Coherence obviously includes, in my view, workability: a term may be implied if it is necessary to ensure that the agreement is workable (as indeed has been recognised since The Moorcock (1889) 14 PD 64 in the Court of Appeal).

Having concluded that on true objective interpretation of the Trade the parties expressly agreed to sell and purchase GDNs by reference to a stated nominal value at a price struck as a percentage of that value, giving rise to an obligation to deliver a fraction of a GDN, the only way in which their agreement can be made to work is by implying a term for settlement of the fractional entitlement in cash. That is the implied term pleaded by LBIE in the Reply, so that LBIE was on that basis required to

Now I must acknowledge some hesitation on my part, notwithstanding the apparent logic. My hesitation has been that the truth is that the parties, if asked, would in all probability have made enquiries that revealed a shared misapprehension as to the nature of GDNs, which in turn would in all probability have resulted in them both wishing to dissolve their agreement, since both were only contemplating the sale and purchase of ‘scraps’ and Exotix could (as I understand the position) not lawfully have bought other than scraps for its own account. This is the conundrum to which I have referred previously and which has caused me concern. An alternative answer, therefore, which I turn to consider shortly, is that the parties were under such a fundamental misapprehension as to the subject matter of the Trade that though they appeared to be ad idem the consensus was reached on the basis of an assumption so fundamentally flawed as to negate not only its business efficacy and workability but the consensus apparently reached. But the law usually baulks at such an explanation and prefers to give effect to what the parties ostensibly appear to have agreed. On that basis, it seems to me that the logic of implying the term pleaded is clear: its necessity is basic because without it the contract is unworkable.

It was common ground that if LBIE succeeded in its primary submission as to the interpretation of the Trade and the implication of a term to give it business efficacy LBIE would be entitled to restitutionary relief on the ground that Exotix would on that basis have been unjustly enriched by its receipt of the overdelivery of GDNs and the coupon payments made in respect of the over-delivered amounts. No special defence was asserted by Exotix.

The appropriate remedy is monetary. LBIE did not suggest that Exotix should obtain and restore replacement GDNs; and although Mr Morpuss did at one time suggest that Exotix would have an option how to give restitution, in his closing submissions he accepted that the proper remedy was monetary. I agree that this is the correct means of restitution: and see per Lord Neuberger of Abbotsbury in Menelaou v Bank of Cyprus UK Ltd [2016] AC 176 (SC) at [81].

As to the measure of the monetary restitutionary award required to reverse Exotix’s unjust enrichment, Exotix appeared to float the possibility of a monetary payment equal to the present market value of the relevant number of over-delivered GDNs, and it appeared to contest any obligation to account for interest and/or the coupon payments it received. But at the end of his oral closing submissions, Mr Morpuss told me that the parties were “agreed in principle” what the relief would be, and anticipated no difficulty in agreeing an appropriate order.

If the parties have agreed or later agree some other measure I shall abide by that: but I would have thought that the correct measure is that Exotix should reverse its own unjust enrichment by paying to LBIE so much of the price it obtained when it on-sold the GDNs to Deutsche Bank as is attributable to the over-delivered whole number of GDNs, together with a sum equal to the aggregate of the various coupon payments it received up to the date on which Exotix sold the GDNs delivered to it to Deutsche Bank, plus interest. If the parties are not agreed, or if either wishes to contest my view (which is to that extent provisional), the matter can be dealt with at a consequential hearing after judgment.

I turn to consider the possibility I have referred to previously that, contrary to the view I have expressed, the parties were under such a misapprehension as to the fundamental basis of their agreement that they cannot be said to have made any valid and performable contract at all.

The presumption against an interpretation of a contract which requires the performance of the impossible follows naturally from the assumption that contracting parties are reasonable people who do not expect each other to do or contract to do the impossible (and see Lewison, ‘The Interpretation of Contracts’ 6th ed at [7.19]). Nevertheless, the presumption is not absolute; and it is also realistic to accept that the parties may have agreed terms without then understanding the difficulties they would cause. The Court may be bound to search for alternative interpretations; but it is not bound to force upon the parties a solution which is performable but does not accord with a more likely interpretation of their intentions.

In my judgment, the parties cannot reasonably be supposed to have agreed to sell or purchase a substantial holding worth some $7 million for $7,707, just as it is plain that Exotix cannot have intended to take on such a substantial holding and pay that £7 million. It is plain that what was in contemplation was a sale of ‘scraps’ for a scrap price of US$7,707. Price and subject matter are two sides of the same coin and the one defines the other. Even when the consequence is impossibility of performance I adhere to my provisional view as to the true interpretation of the Trade.

That raises two difficulties: it has given rise to a dispute (a) as to whether the claim as pleaded caters for the consequences, and if not whether an amendment should be permitted and (b) what indeed the consequences in law are and whether the amended plea is a good one. Exotix submitted that the answer was in the negative to each limb.

As Mr Morpuss was quick to emphasise, in his oral opening, Mr Bayfield QC (for LBIE) expressly disavowed any reliance on the law of mistake, and insisted that “LBIE’s claim is based firmly on the terms of the Trade itself”. He did, however, suggest that

“…if the contract were void for impossibility, the parties having contracted under a common misapprehension that the trade could be performed when it can’t be performed, then the consequence would be that Exotix would be unable to retain its windfall, the parties would have to be restored to their pre-trade position, which is presumably why Exotix is not taking that point.”

That last phrase is a little confusing; for if such were and are the consequences it might be though that it has always been for LBIE to take the point and not Exotix. For Exotix, Mr Morpuss in opening emphasised the point that LBIE’s primary case depended on establishing an implied term, without which (he said)

“…we have a contract on his case which is impossible to perform.”

In written closing submissions, Mr Bayfield stated that

“If the Trade would have failed because of the fraction issue, then the Trade would be void for impossibility and LBIE would in any event be entitled to restitutionary relief…

…

Specifically, in the premises the Trade will be void for common mistake, on the basis of “a fundamental assumption which renders performance of the essence of the obligation impossible”. See Great Peace Shipping Ltd v Tsavliris Salvage (International) Ltd [2003] QB 679.”

In his oral closing submissions, Mr Bayfield suggested that, although possibly not necessary since the result would be the natural consequence of a conclusion of law, LBIE would ask for permission to amend in case the Court were to think it necessary.

For Exotix, Mr Morpuss objected to the change to reliance on mistake at such a late stage, and to any amendment of LBIE’s case to plead it. He submitted that any amendment to introduce such a plea was far too late and he drew attention to the fact that the issue as to fractional trades had been raised in the Defence as originally pleaded. He suggested that had a case based on mistake been pleaded in time, there would have been other factual issues to explore: Exotix’s “requests for disclosure would have been very different”, and he would have wished to explore properly with the witnesses whether LBIE should reasonably have appreciated the mistake, which he submitted would, if demonstrated, preclude reliance on mistake.

I directed the exchange of further written submissions after the hearing to enable LBIE to formulate a proposed amendment and then for all concerned to focus on these issues. This resulted in a further revised proposed amendment, which did not expressly refer to common mistake but pleaded that the Trade would be void if the Court concluded that its performance was impossible, and four additional sets of written submissions of considerable length.

From all this it emerged (or was clarified) that:

Neither side sought a further oral hearing on these issues, both preferring to rest on their very full written submissions; but the points raised have caused me not a little difficulty. In particular, the apparent agreement between the parties that neither is averring impossibility has troubled me; and the legal foundation, different forms and varying consequences of impossibility have been much debated both in the Courts and in academic commentary for many years since Bell v Lever Brothers Ltd [1932] AC 161. Further, what I now take to be the leading recent case of the Great Peace has its curiosities and is not without its critics (especially in other common law jurisdictions), although it is of course binding at this level in this jurisdiction.

With that opening confession of my appreciation of the difficulties, I turn to address the points raised in this context, which are (as it seems to me) closely intertwined, and which in my view ultimately turn on whether the amendment sought would, if permitted, introduce a new case based on an assertion of mutual mistake previously disavowed which would require substantial factual exegesis and examination for it to be fairly adjudicated and made good.

Although I quite appreciate, and take into account, the new approach to late amendments mandated by the Court of Appeal which places emphasis on there being adequate reason and justification for the lateness of an amendment proposed in closing and only formulated after the end of the hearing, I consider that in the peculiar circumstances of this case it is important first to identify the true nature and legal basis of the case on impossibility which LBIE seeks to cover.

This is a case where the parties were (in my judgment) ad idem as to the trade but their consensus was based on a shared but incorrect assumption as to the nominal value of each GDN, such that (if no ‘saving’ term is to be implied) the Trade cannot be performed in accordance with its terms. It is not a case of supervening impossibility by reason of some unanticipated event (such as might establish frustration); nor is it a case in which performance is impossible by reason of some physical or geographical impediment; and it is not a case where the expense or onerousness of performing the outstanding contractual obligations differs from those that the parties can reasonably have contemplated at the time of their agreement. Most importantly, it is not a case where (absent an implied term) it is possible to perform the letter of the contract.

In my view, whilst recognising that there are deep waters to navigate, it is vital to distinguish cases where the mistake goes to the quality of the available or possible performance or where the subject matter simply has ceased to exist, and cases where performance in accordance with the letter of the contract always has at law been impossible. That seems to me the distinction apparent in the analysis in The Great Peace at [55].

In the one case, the essential question, which is usually at the root of any contractual dispute, is one of risk allocation: to determine where the risk of imperfect or altered performance should fall in circumstances the parties did not provide for or perhaps envisage, or in a state of affairs altered from that the parties assumed existed. In the other case, where what the contract provides for has always been impossible in law to

Since in a case of the latter character the basis of invalidity is failure of consideration in consequence of legal impossibility, as distinct from a shared but false assumption as to a “state of affairs”, it seems to me that the elements required in order to establish common mistake as set out in the Great Peace at [76] are not of relevance in a “legal impossibility” case. More particularly, it does not seem to me that, in the latter context, the Court need assess whether the mistake as to the legal quality of the subject matter is the fault of one party or the other, or whether it might have been discovered by either or both; whereas both are elements in a “state of affairs” case: ibid in the Great Peace.

Exotix treated this case as a “state of affairs” case; or at least one in which the elements identified in [76] of the Great Peace had to be established. Mr Morpuss on its behalf contended, in effect, that LBIE in such circumstances cannot pray in aid impossibility of performance to avoid the contract unless it can demonstrate not only that the impossibility is the consequence of mutual mistake but also that such mistake was based (a) not on ignorance or a failure to focus but (b) on a positive shared belief on reasonable grounds that the GDNs had a par value of one PEN. As to (a), Mr Morpuss cited the decision of Henry Carr J in Co-Operative Bank plc v Hayes Freehold Ltd [2017] EWHC 1820 (Ch), especially at [143(i)], citing Chitty on Contracts 32^nd ed at [6-001 to 6-004] for the proposition that “It is not enough if a party has not thought about the issue’’. As to (b), Mr Morpuss cited the decision of Steyn J (as he then was) in Associated Japanese Bank (International) Ltd v Credit du Nord S.A. [1989] 1 WLR 255, especially at page 268.

However, as to (a), it seems to me that the Co-Op v Hayes case was a “state of affairs” case and of a rather different nature accordingly from the present. There the mistake was as to whether a party to the Deed to be construed had the power to accept a surrender of a Superior Lease as the party asserting the common mistake (Deutsche Bank) had been advised it had, rather than a mistake as to the legal subject-matter of the contract. Further, in that case, on the facts, it was held that the cause of Deutsche Bank’s decision was not a mistake as to the contract itself but its reliance on the incorrect advice it had received, as to which it had assumed the risk (see at [137] to [138]).

Moreover, and in any event, as the passage in Chitty referred to by Henry Carr J recites (itself citing what Lord Walker of Gestingthorpe JSC said in Pitt v Holt [2013] UKSC 26 at [108] to [109]):

“A mistake encompasses two states of mind, namely an incorrect conscious belief or an incorrect tacit assumption as to a present matter of fact or law, but does not encompass mere causative ignorance but for which the claimant would not have acted as he did.”

In this case, the mistake was, as I see it, an incorrect tacit assumption shared by both parties, evident from both the statement of the subject-matter and the calculation of the price, which was fundamental and in respect of which there can be discerned no assumption of risk.

As to (b), Steyn J’s approach in determining that a “party cannot be allowed to rely on a common mistake where the mistake consists of a belief which is entertained by him without any reasonable grounds for such belief” was in the context of another “state of affairs” case, albeit that the state of affairs assumed to exist was plainly and fundamentally different than the reality. In that case, a fraudulent party had purported to sell to the Associated Japanese Bank (“the AJB”) and lease back from it four industrial machines, and the Defendant bank (“C du N”) had guaranteed the fraudulent party’s obligations under the sale and lease back. The fraudulent party was adjudged bankrupt. When AJB sued C du N on the guarantee it emerged that none of the industrial machines, which the judge held were also intended to be the “real security for the guarantee”, existed. In such a context, the judge’s view that the defendant (C du N), in seeking to avoid the guarantee, had to show that it believed the machines existed and that such belief was reasonable, is readily understandable. But that is a different case than the present, which is a “failure of consideration” case, where neither party is seeking to avoid performance or nullify the Trade, but where the Trade (on my construction of the parties’ agreement) simply cannot be performed because by reason of the definition of the subject matter it provided for the delivery of a fraction of a GDN, which is not possible.

The effect of it being legally impossible, if no saving term is to be implied, to perform the Trade in accordance with its terms, or in other words, of the impossibility of giving effect to the letter of the agreement, would in my view be to vitiate the Trade.

In such a context as this, I do not think it relevant or appropriate to enquire whether the common misconception was more the fault of one party than another. I accept, as precedent would in any event require of me, that such an enquiry is necessary in a “state of affairs” case: that is clear from the Great Peace (and see especially at [76]). But this is not, in my view, such a case: it is, by reason of the legal impossibility of performance according to the letter of the contract, a “failure of consideration” case. The Trade having been executed, the remedy is in restitution for recovery of the value of the over-delivery of GDNs (the tree) and the intermediate distribution paid in respect of the GDNs (the fruit of that tree).

That analysis and conclusion absolves me, as I see the matter, from considering at length the arguments against the proposed amendment based on (a) the modern antipathy to late amendments and (b) the prejudice to Exotix of permitting a plea

In so concluding, I have taken anxiously into account the general and fundamental rule that the judge must not “descend into the arena”, lest the trial not both be and appear to be impartial. I accept entirely that a judge may enquire and by enquiry may prompt a change of case or even a new case; but a judge should not prescribe, nor step out of the parameters prescribed by, the way the case is ultimately formulated by the parties themselves (and see Loveridge v Healey [2004] EWCA Civ 173 at [23]-[24]).

However, that, in my view, is not what would be involved in this case, notwithstanding the somewhat unsettling italicised words in paragraph [189(1)] above. In this case, impossibility of performance is the natural consequence of accepting LBIE’s submission as to the interpretation of the Trade if its principal case for an implied term is not accepted. Further, if the Court considers, as I do consider, that the true interpretation of the Trade as a whole results in it being impossible to perform, I do not see that the impossibility can in logic or fairness be resolved by judicial reticence.

In my judgment, a question has been put forward which naturally arises as a consequence of the process of contractual interpretation, and I must adjudicate upon it in accordance with my assessment of the legally correct answer. No further evidence is required: the question is a legal one on the basis of the existing record.

I accept LBIE’s submission that the amendment it proposes to introduce a new paragraph under the heading ‘Relief sought if Trade held void or unenforceable’, and to add a further consequential paragraph to the prayer for relief is unobjectionable, and, in my view, it is apposite to ensure that the Court is not trespassing beyond the pleaded case or granting relief never actually sought. On that basis I cannot see any prejudice to Exotix or any disruption which might affect other court users; and whilst I do think it regrettable that the amendment was not put forward earlier, I do not think the Court’s insistence on a more rigorous approach to amendments should be the occasion for denying itself the ability to give proper and regular effect to its conclusions on this alternative basis, in case it is found wrong in its main conclusion that the Trade is saved by the implied term pleaded.

Had I concluded that the proposed amendment did occasion further material factual enquiry, whether in terms of disclosure or cross-examination, such as to necessitate a further oral hearing, I would have refused it.

In conclusion, therefore, I consider that on an objective interpretation of the Trade set in its admissible factual matrix and having regard to the admissible documentation recording and/or implementing it, its subject-matter was a specified notional amount of GDNs stated in PENs with a price prescribed and calculated accordingly.

I would construe the Trade as subject to the implied term pleaded and find in favour of LBIE’s primary case accordingly. In that event, there is no dispute as to the appropriate relief, nor is it necessary for me to consider the fourth issue identified in paragraph [63(4)] above.

But if I am wrong that a term falls to be implied, I would consider that LBIE is entitled to restitutionary relief on the basis that without an implied term the Trade cannot be fulfilled in accordance with its terms and there would be a failure of consideration such as to make it void and unenforceable. This is not like the case of the sale of an old master. In that case, which was suggested as analogous, there is no uncertainty about the subject matter (a painting) nor any doubt as to performance of the contract. There the only problem is fairness; and that is no basis for intervention. Here the problem is that my conclusion as to the true subject-matter (and I confirm I have revisited that conclusion iteratively given the problem it has thrown up but not felt it right to change it) results, unless a term is implied, in impossibility of performance.

My provisional view is that in those circumstances LBIE should be entitled to the like relief as under its primary claim, by way of restitution. If that is contested or requires further definition, the matter can be debated further.

I should perhaps say finally that, even taking into account a surprising and regrettable lapse on the part of LBIE’s administrators in failing to ascertain that LBIE’s aggregate holding of GDNs, far from being a scrap position, was a very valuable one because the nominal value of each GDN was not Sol 1 but Sol 1,000, it seems to me that this result accords with both overall commercial good sense and commercial morality.

In my view, the commercial morality of Exotix’s position was always at best frail. Mr Marron’s reaction on receipt of such a large distribution (see paragraphs [5049] and [510] above) is a spontaneous expression of a commercial morality subsequently abandoned by Exotix’s board, whose decision to keep quiet about the whole thing and not disclose anything to LBIE seems to me to deepen the departure from ordinary norms. In that context, I consider that, even if not invariable, as between respectable parties in an established trading relationship it is likely that Professor Persaud’s evidence that

“… the routine act of trying to correct, cancel or adjust for obvious errors as soon as they are picked up by one party is a trade custom”

As to my view that my legal conclusions accord with commercial good sense as well as commercial morality, it seems to me in reality inconceivable that the parties intended, or any reasonable observer in their position would have thought they intended, to sell other than a scrap position at a nominal price calculated by reference to their assumed nominal value. Mr Marron of Exotix’s initial reaction was not only a reflection of commercial morality but an expression of the obvious commercial intention of the Trade.

In short, in the commercial world, any dispute as to the subject-matter of the sale which does not take into account the price agreed is nonsensical: the two march together and in the event of any uncertainty as to subject-matter the one helps clarify the other. At the end of an over-long judgment perhaps I should confess my view that this less complex conclusion is what really would have been the reaction of the wellinformed observer.

Lastly, I would wish to record my thanks to Counsel and their respective teams for their assistance and their patience. It seems likely that it will be necessary to have a further consequential hearing to deal with outstanding issues: but their assiduous paperwork and the clarity of their oral submissions has greatly helped me in a case which has raised many textbook questions of considerable legal difficulty, even if the legal conclusion, as well as the fair result, has ultimately seemed to me to be reasonably clear.