Socimer International Bank Ltd v Standard Bank London Ltd

This is large-scale litigation between two banks which, before the failure of one of them, had been trading together in the securities of emerging markets. When, shortly before the failing bank entered into liquidation, it was put into default, it owed its counter-party bank some US $24.5 million in so-called “Unpaid Amounts” in respect of a portfolio of forward sales of securities which it had bought. That was on 20 February 1998, the so-called termination date. Under the operative agreement between the banks, the “Standard Terms for Forward Sale Transactions” dated 8 November 1996 (the “agreement”), the creditor bank, the seller, had to “liquidate or retain” that portfolio (the “Designated Assets”) to satisfy the amount due to it. For these purposes it had to value the portfolio on the termination date. The critical sentence of the agreement, of which much more will have to be set out below, found in its clause 14(a)(bb), is this:

The seller bank, Standard Bank London Limited (“Standard”), did not in fact carry out a clause 14(a)(bb) valuation. Instead, it sold what it could over the ensuing months and (and in the case of one security) almost years, and credited the proceeds to the buyer bank, Socimer International Bank Limited (“Socimer”), in dribs and drabs. Until sale of any security, Socimer received no credit for it. In the meantime, Socimer had entered into liquidation on 3 March 1998, about two weeks after the contractual termination date. It was not until November 2002 that the liquidator’s London solicitors, Messes Allen & Overy, wrote to Standard to complain that no clause 14(a)(bb) valuation had been carried out. That complaint led to the issue of these proceedings by Socimer’s liquidator on 9 April 2003, in which, ignoring the amounts for which Standard had actually sold the portfolio and asserting other valuations as at the termination date, then alleged to be 3 March 1998, Socimer claimed a surplus of some $13.8 million as due to it. Standard, however, defended these proceedings on the basis that it had done what it ought to have done and was itself still owed money in the liquidation.

The issue of construction thrown up by this claim and defence, namely whether Standard should have carried out an immediate valuation and credited the resultant amount to Socimer, as Socimer maintained, or was obliged to credit Socimer only with what it had obtained from sales in its discretion, as Standard maintained, was decided as a preliminary issue in the litigation, by the judgment of Mr Justice Cooke dated 11 May 2005. He held in favour of Socimer’s submissions on the construction of clause 14(a)(bb). He concluded [2004] EWHC 1041 (Comm) at para 34 that –

For the purpose of that decision, Cooke J had accepted the submissions of Mr Richard Millett QC on behalf of Socimer that “Standard was given a wide discretion in relation to valuation” at para 16. Thus the judge expressed that discretion as follows:

Assisted as they were in this way by Cooke J, and despite a mediation, the parties were nevertheless unable to settle for themselves the resultant question of valuation of the portfolio. Indeed, they remained over $14 million apart when default interest was taken into account. Accordingly, the valuation issues were addressed at a witness trial in June and July 2005 before Mrs Justice Gloster. Before her, Socimer now submitted that Standard no longer had a “wide discretion” in valuing, but was bound (to take reasonable care) to find the true market value. That was put as a matter of contractual implication, or alternatively as a matter of equity by analogy with the duties of a mortgagee with a power of sale. It was submitted on behalf of Standard, on the other hand, that the answer depended not on what a reasonable, objective, valuation should have produced, but on the valuation which Standard itself would have performed, if it had been alive at the time to the true meaning of its contract. Indeed, it was also said that Socimer’s “should have” case (as contrasted with Standard’s “would have” case) was not open to it. Gloster J, however, agreed with Socimer’s submissions. For good measure she rejected the evidence of Standard’s factual witnesses on valuation, and, founding herself in the main on Socimer’s expert witnesses, produced a valuation greatly in excess of Standard’s case. Most of that difference, however, was made up on two particular securities, a corporate bond known as Socma DRs, where the judge’s valuation was $3,000,000 as against Standard’s valuation of $500,000; and a species of Brazilian sovereign debt known as Brazilian TDA-Es, where the valuation was $8,500,000, as against Standard’s valuation of $3,500,000. That created a difference of $7,500,000 (plus consequential interest) on those two assets alone.

Gloster J also found in favour of another, major, aspect of Socimer’s case, which was premised on the assertion that even if Standard’s valuations had been otherwise accepted, but Standard had also credited to the “Unpaid Amounts” various cross-credits owed to Socimer deriving from other transactions between the parties, amounting to some $4,000,000, then Standard would have found itself with surplus assets in its hands and would have returned to Socimer the Socma DRs as having been fully paid for. On that basis, referred to at trial as Socimer’s Socma Primary Case, the judge agreed with Socimer that the value of the Socma DRs in Socimer’s hands was not merely $3,000,000, but their amortised face value namely $8,424,966. Accordingly, the judge ended up giving judgment for Socimerin the sum of $11,686,295.43, plus interest (at US Prime) in the sum of $3,528,936.59, in the total sum of $15,215,232.02.

Standard now appeals, with the leave of Thomas LJ. Of a number of issues on appeal, as to which there has even been some dispute as to whether or not permission to appeal has been granted, we have heard only two, or perhaps three, as we shall explain. The two issues which we have heard debated in full, for which it is common ground permission to appeal has been granted, relate to the two main matters of decision referred to above, namely:

whether Standard’s valuation obligation was to carry out a reasonable, objective, valuation (Socimer’s “should have” case) or only the honest but otherwise subjective valuation which Standard would have carried out if it had been aware of its contractual obligation to value (Standard’s “would have” case); and

The importance of issue 1 is that the judge only made findings of valuation on the basis of an objective valuation. The importance of issue 2 is that it is common ground that it is only if it is answered as the judge answered it that Socimer is able to advance its Socma Primary Case. Issue 1 may be described, loosely, as the “implied term” issue, since Socimer accepts that the objective valuation for which it contends, so far successfully, needs to be introduced by implication. Issue 2 may be described as the “Unpaid Amounts” issue, since its purpose is to define the total of the Unpaid Amounts to which Standard’s valuation needs tobe applied.

The third issue which the parties have addressed is a procedural issue inherent in issue 1, and that is whether it was open to Socimer to deploy its implied term, as it has done successfully, in circumstances where Cooke J had construed the agreement, as Standard contends, as giving to Standard a discretion as to the valuation which it was obliged to perform. If that issue were decided in Standard’s favour, it would mean that the judge deployed the wrong valuation test in any event. A further related procedural issue, which has also been debated before us, is whether the judge was entitled to reject the valuation evidence of Standard’s witnesses in circumstances where Socimer deliberately chose not to challenge that evidence. That issue may or may not matter if issue 1 was open to the judge and she was otherwise right in the answer she gave to it. If, however, that issue was not open to her, or she gave the wrong answer to it, the answer to the related procedural issue might guide the consequences for the parties: as between this court now accepting the evidence which Standard gave as to its “would have” valuation on the one hand, or there having to be a new trial as to the “would have” valuation on the other hand. It is also possible that this related procedural issue could by itself lead to a conclusion that the judge erred in her answer on issue 1. I shall call these procedural issues respectively issue 3(1) or the “open to Socimer” issue, and issue 3(2) or the “unchallenged evidence” issue.

The relevant terms of the agreement, expressed to be between Standard as “Seller” and Socimer as “Buyer”, are as follows –

It will be observed that: (i) the agreement essentially concerns transactions defined as a sale of assets on a forward basis (clause 1) under which, pending a forward settlement date during which the buyer receives credit on payment of a finance charge which is incorporated in the price of the asset sold, the asset remains in the entire legal and beneficial ownership of the seller until full settlement (clause 3(c)); (ii) a part payment or “Downpayment” is required on purchase, which is non-refundable (clause 2(a)) and the balance of the transaction price is the “Unpaid Amount” payable on the forward settlement date (clause 3(a)); (iii) the critical valuation clause is contained within clause 14 as part of the “Events of Default” and in particular is to be found in clause 14(a)(bb); (iv) thus the whole valuation process is a consequence of the buyer’s default; (v) although the seller is entitled to sell Designated Assets “in its sole and absolute discretion…at such time, in such manner and at such price as it deems reasonable and appropriate”, the valuation obligation covers both “liquidated or retained” Designated Assets and is expressed in the simple terms that “The value [of such assets]… shall be determined on the date of termination by the Seller”; (vi) it is possible to contrast the expression “The value” in that valuation sentence with the expression “Market Value” which is defined in clause 1 (Definitions) as “the value…determined by the Seller in its sole and absolute discretion for assets of the same description and type”; (vii) the Market Value definition is utilised in clause 6 (Additional Downpayments; Subsequent Additional Downpayments) by reference to “Mark-to-Market Loss” which is itself defined in clause 1 by reference to Market Value, and for these purposes the concept of Additional Downpayment, a form of (further) margin requirement, is spoken of as something “allocated by the Seller for the purposes of subsequent settlement in its absolute and sole discretion” (clause 6(b)); (viii) clause 14(a)(bb) also provides that “following the satisfaction of the Seller’s claims”, any remaining Designated Assets shall be “sold” (which it is common ground means in this context transferred) to the buyer, and any remaining proceeds from the sale of Designated Assets shall be paid to the buyer: ie the buyer recovers either in specie or in cash the balance of Designated Assets remaining after full satisfaction of Standard’s claims; (ix) clause 14(a)(bb) also provides that where the valuation sentence applies but there is insufficient to satisfy Standard’s claims in full so that there is a deficiency, then Standard shall “prepare a certificate specifying the amount of the deficiency” and such certificate “shall be conclusive and binding on the Buyer, in the absence of any manifest error”; (x) set-off provisions are contained in clause 8 (Payments) and clause 9 (Call Account and Set-Off), sub-clause (c) of which gives to Standard a general discretion (“may at any time…after the Buyer commits any event of default”) to apply any credit balance on any account maintained with Standard in the name of the buyer in or towards satisfaction of “any moneys due and owing to the Seller”.

Cooke J’s judgment is dated 11 May 2004, [2004] EWHC 1041 (Comm). It was concerned with “the trial of a preliminary issue concerning the true construction” of the agreement (Cooke J, para 1). The preliminary issue is not in terms defined in his judgment, but the order for directions pursuant to which the trial before him was held, made by Langley J on 14 November 2003, provided as follows:

Paragraph 11 of Socimer’s particulars of claim asserted in relevant part that “the market value of any Designated Asset for the purposes of [Standard’s] continuing obligations to Socimer under clause 14 was to be determined as at the date of termination…”. Paragraph 6.3 of Standard’s defence, on the other hand, referred to “Standard Bank being entitled to the relevant assets and having an absolute discretion following default”; and paragraph 6.5 alleged that Standard “would have complete flexibility to liquidate Socimer’s portfolio of assets at its absolute discretion in the event of default” and “would have the right to hold those assets for such time as it might consider to be appropriate in the circumstances before they were sold”. In response, in paragraph 34(1) of its reply, Socimer pleaded that “while Standard bank had an absolute discretion, under clause 14(a)(bb) of the Agreement, as to how, when and at what price it sold the Designated Assets (whether by way of liquidation or by way of retention), Standard Bank had no discretion at all following such termination as to the valuation of the Designated Assets for the purposes of the state of the account between itself and Socimer, which had to be valued on (or as at) the date of termination”, so as to provide certainty going forward as to the state of the account between the parties, which otherwise would be liable to fluctuate almost daily while Standard played the market in the Designated Assets at Socimer’s expense, since default interest would continue to run until Standard sold off portfolio items at a time and price of its own choosing (paragraph 36 of Socimer’s reply).

Although this is not a complete statement of the “issue” tried by Cooke J or of the parties’ refined submissions before him, it is nevertheless plain that there arose for decision two matters at least: (1) whether the state of the accounts between the parties, so far as it depended on Socimer’s outstanding forward sale portfolio, was to be settled once and for all on or as at the date of termination, as Socimer contended, or as and when Standard sold items from the portfolio on or at any time following the date of termination, as Standard contended; and (2) how, if any valuation had to be performed as at the date of termination, it was to be carried out: in Standard’s absolute discretion, as Standard asserted for all its dealings with the Designated Assets following default, or with “no discretion at all” as Socimer appeared to assert.

It may be true that the first of these two issues was the primary issue debated (reflecting the circumstance that Standard had never in fact carried out a valuation on or as at the termination date), but it was still tied up inherently with the issue of discretion and how any valuation was to be performed: particularly as Socimer accepted that Standard had, in the express terms of clause 14(a)(bb), an absolute discretion as to how it sold the Designated Assets and yet asserted “no discretion at all…as to the valuation”. It is possible, however, that Socimer had simply meant to say that Standard had no discretion at all but to carry out a valuation on or as at the termination date, while leaving it open how such a valuation was to be performed - other than by pleading in the particulars of claim that it was to be at “market value”. That, however, in the light of the agreement definition of market value at any rate when expressed as “Market Value”, might be said to maintain the ambiguity.

To the extent that this second aspect of the issues of construction remained in doubt on the pleadings, it was nevertheless elucidated during submissions before Cooke J, because Mr Richard Millett QC then accepted that Standard retained an absolute discretion over how the valuation he contended for was carried out by Standard. Indeed, he used the existence of this discretion as one of his arguments for insisting on a termination date valuation: for, as he submitted, the combination of the obligation on Standard to value as at a date certain, the termination date, and of the discretion in Standard as to how to perform that valuation, created a balancing of the interests of the parties which reflected the context and needs of their agreement as a whole.

Thus the following exchange occurred between Mr Millett and Cooke J right at the end of the former’s closing submissions (at Day 2.161):

This exchange developed out of a discussion between Mr Millett and the judge about the way in which a valuation on or as at a certain day would work in practice. For instance, what would happen if there was no market in a given security on that day? Or if a price on a screen did not reflect the realities? There had been evidence from Mr Ian Beckman (an independent expert called by Standard) that both were possibilities and that if there are no buyers in the market and it is necessary to value on a given day, a price as low as zero will need to be applied. These possibilities emphasised that valuation can be difficult and subjective. Mr Millett said (Day 2.117/8):

Cooke J found the following facts proved by the agreement or its matrix (para 7). Liquid assets could be valued on the basis of screen prices and conversations with market participants. For more illiquid assets, however, where there were no readily available markets and no screen prices, “Standard might have to use its own commercial judgment”. The value placed on them by Standard was not necessarily that which would be obtained on an actual sale, because of the difficulty of evaluation of such assets and the volatility of the market. Smaller debts from less developed economies are highly illiquid, are traded infrequently, and are as known as “exotics”. Volatility is a hall-mark of emerging markets as a whole. Downpayments were designed to mitigate the risk that the seller would default on his obligations. On an illiquid asset they could range up to 50%. Where illiquid assets are concerned, terms of trade invariably favour the seller in the event of the buyer’s default because of the difficulty of realising the debt. Mr Beckman’s evidence was that the only definitive measure of value was the price for which assets could actually be sold, as distinct from prices to be seen on screens or databases or calls to market professionals: if there was no real price, it would be necessary to value the asset at zero.

Cooke J set out the parties’ submissions. It is sufficient to record this (at paras 16/17) –

The balance which Mr Millett had accepted and emphasised in his submissions between the obligation to value on a day certain and the discretion to perform that valuation is a recurrent theme in Cooke J’s discussion and analysis. For instance:

Since it is common ground that the judgment of Cooke J, which was not taken to appeal by either party, is binding on them both, it is necessary to set out the judge’s conclusions and his reasoning. Gloster J set out paras 27/33 in full. I would begin at para 26:

Thus Cooke J concluded as follows:

Much of that analysis was primarily aimed at Standard’s particular case, now water under the bridge, that it was only as and when it sold an asset that any sum needed to be brought into account in favour of Socimer. For present purposes, it is perhaps useful to recapitulate certain aspects of the judge’s analysis which are particularly pertinent to the submissions we have heard on this appeal: namely (i) that the leading consideration under sub-clause (bb) is Standard’s obligation to value all Designated Assets on or at any rate as at the termination date and in any event as soon as practicable; (ii) that the whole of that value, both that which represented assets immediately liquidated and that which represented assets retained for Standard’s own account, had to be credited to Socimer; and (iii) that any surplus value had to be transferred to Socimer in specie or in cash, again as soon as practicable after the termination date. It follows that all market risk from the point of valuation onwards is transferred formally from the buyer, Socimer, to the seller, Standard.

It may be said that this contractual provision is remarkably favourable to the buyer, seeing that these consequences (a) arise on the buyer’s default; (b) when the counter-party risk is therefore at its greatest; (c) produce the result that, for all assets which cannot be immediately sold, ie the more illiquid assets, the seller is compelled to take them on his own book at his own risk, so as to produce as it were “the proceeds of their sale” even when they have not been in fact sold, and even though he has not selected them for his own investment at all; (d) and are consequences assumed to have been selected by the seller as his option under sub-clause (bb), in preference to the position under sub-clause (aa), on the basis that the assets in question have been falling in value since the forward sale was made and/or are subject to a falling market. However, the judge’s analysis, consistently with Mr Millett’s own submissions about Standard’s “wide discretion” (see paras 16/17), balance these consequences with the other side of the coin, namely that the valuation exercise “lies entirely in [Standard’s] hands” subject to good faith (at para 32), otherwise expressed as being “left entirely to Standard’s own discretion” (at para 19); that where there is no market on the valuation date the value may turn out to be zero (at para 32); and that in the case of any deficiency following the valuation process, the size of it will be conclusively determined in the absence of manifest error by Standard’s certificate (para 31).

One other aspect of this analysis may here be noted. If the clause 14(a)(aa) option is chosen, the seller keeps both downpayments and assets, but gives up his right to any additional downpayments or to be paid the amounts outstanding, viz the Unpaid Amounts. If the clause 14(a)(bb) option is chosen, and the assets cannot be immediately sold, the seller has to credit to the buyer the “value” of the assets retained and thus, even though he is in principle entitled to keep the downpayments because they are part of the price due and are in any event non-returnable nevertheless, were the valuation to make no allowance for the risks inherent in the asset, then the upshot would be that the seller is required to disgorge in the form of “value” the downpayments which are designed to reflect those inherent risks.

It is against this background that following the judgement of Cooke J the parties prepared for the outstanding issues between them. On the basis that the issues of construction had been determined at the trial of the preliminary issue, those issues were limited to those of valuation.

It emerged, however, that a considerable issue of construction continued to divide the parties: in essence whether the valuation which Standard ought to have carried out but had failed to put into operation, because they had misconstrued or overlooked their agreement, should now be determined by the court at the figures which Standard “would have” used in an assessment within their own discretion, had they performed their contract, or at the objectively ascertained figures which Standard “should have” used on a proper and careful market analysis.

That issue led to two of the issues which have now been debated on this appeal: one, the issue of construction, as to whether the determination of value is that of Standard (which might be termed a “subjective” valuation), or whether it is that of the court itself (applying entirely objective criteria); the second, as to whether that issue was in any event properly and fairly open at the second stage of the trial process.

To understand those issues as they were developed before Gloster J and this court it is necessary to trace the procedural conduct of this litigation from the conclusion of the trial before Cooke J to the conclusion of the trial before Gloster J.

In his order at the conclusion of his trial, Cooke J gave directions for the future conduct of the litigation. That order provided for disclosure, the exchange of witness statements and also for experts to prepare reports “confined to expert evidence on the valuation of the Designated Assets…on or as at the date of termination”. It also provided for mediation.

On 2 September 2004 Standard served its witness statements. These included statements from Mr Jeffrey Clifford and Mr David Feld, who had been in 1998 and continued to be employed by Standard. Both gave evidence as to how Standard “would have” approached the valuation exercise required by the judgment of Cooke J. Therefore from at least that time it would have been obvious to Socimer and its legal advisers that that was how Standard approached the valuation trial. Messrs Clifford and Feld gave evidence that in the first place the valuation would have been delegated to Mr Clifford. Mr Feld, as the more senior man, agreed with Mr Clifford’s view that Standard’s senior management would have been involved at the review stage, and said that the ultimate responsibility and decision would have been his and that of other members of Standard’s senior management. Mr Feld explained that, despite the passing of six years from the default, it was important to avoid any element of hindsight. Standard would have been gravely concerned about Socimer’s default and imminent insolvency and the great financial risks arising from the obligation to value and assume the risk of extremely illiquid assets on a same day basis. Senior management would have had as its primary objective a desire to fulfil (what had been found to be) Standard’s obligations in a way which best protected Standard’s interests.

On 25 November 2004 Socimer served the expert report of Professor Luis Rosenberg, a distinguished financial consultant whose firm advises investors and investment banks on Brazilian markets. He had in the past been directly involved in fund management and at one time as a Brazilian government official. His evidence was concerned with the valuation of the Brazilian TDA-Es. He gave detailed evidence in support of his valuation of that security. He addressed the evidence of Messrs Clifford and Feld first at paras 6.26/7 of his report, where he said –

Professor Rosenberg subsequently gave a detailed critique of Mr Clifford’s evidence, stating in passing that it was a “good academic lecture on all possible factors that could eventually be considered in order to diminish the value of a financial asset” (at para 7.11), but concluding that “In summary, based on the above mentioned conceptual and quantitative analysis, I consider that the “internal valuation” proposed by Mr Clifford was biased towards undervaluing the assets and should not be used as a valuation for the TDA-E portfolio in question” (at para 7.31). His “Conclusions” (at para 8) included those that the TDA-Es were of “medium liquidity” being neither illiquid nor highly liquid, that the “task of selling the portfolio could be completed in a matter of two to three weeks”, and that the availability of Brazilian Government bonds with currency indexation clauses “provided a sanctuary to the proceeds of such sale, even if bureaucratic obstacles would prevent automatic remittance abroad”. Professor Rosenberg’s report was clearly aimed at an objective valuation of the security in question: he did not appear to dispute the “would have” aspect of Standard’s evidence, but he disagreed with the subjective approach to that he plainly considered to be an objective appraisal of value. He appears to have accepted that the large portfolio of TDA-Es could not have been sold, or at any rate all sold, on the termination date. He allowed two to three weeks for their sale.

Standard’s expert on the Brazilian TDA-Es was Mr Ricardo Quintero, whose report was served on 2 December 2004. His expertise was gained during his employment in banks in Brazil over a period of 15 years. At the beginning of his report he described his instructions as follows:

I refer to Mr Quintero’s instructions because it was Mr Millett’s submission on the appeal that the fact that experts were used at all for the purposes of the valuation trial showed that it had always been the mutual contemplation of the parties that the valuation required by the agreement was an objective one. In my judgment, Mr Quintero’s instructions show the very opposite. At most, one aspect only of Mr Quintero’s expertise was invoked to provide his own valuation, and in context that was, as Mr Stephen Auld QC on behalf of Standard submitted, nothing more than a “reality check”.

In his executive summary, Mr Quintero described TDA-Es as “illiquid securities with various risks attached”, gave a personal valuation “based upon unlimited time within which to value” of between $5.4 and $5.7m, agreed with “the methodology and approach taken by [Standard] in accordance with the Judgment of the Court in May 2004, namely at or immediately following an event of default under the Agreement”, agreed with the valuation of Mr Clifford in the circumstances prevailing (and the still more conservative valuation of Mr Feld at an executive level) and disagreed with Professor Rosenberg’s approach in a number of respects, including that of using the benefit of hindsight.

On 2 December 2004 the report of Mr Beckman was also served by Standard. Mr Beckman had already given evidence before Cooke J. He was a managing director of WestLB AG and had until December 2002 been head of its department specialising in emerging markets. His evidence was particularly directed to “Valuation Approach upon Default” and the Socma DRs. As for the former, he described what in his experience was an appropriate process, and concluded: “Since the Court has interpreted Clause 14 to mean that Standard bank is effectively at risk when it comes to sell designated assets after accounting for their notional value to the defaulting customer, a conservative approach to valuation is not only prudent but also entirely within the bank’s rights.” As for the Socma DRs, he described these as totally illiquid, with an imbedded credit risk in Socimer as debtor, and concluded that a bid would not have been found so that a valuation at or very close to zero was appropriate. I should explain that Socimer was itself an intermediate debtor under the securitised issue of these corporate bonds by Socma (not itself an associate of Socimer), even if there was for some time uncertainty in the markets as to whether Socimer was a fiduciary rather than a debtor.

Mr Beckman’s letter of instruction was also exhibited. This contained inter alia the following request:

On 20 April 2005 Standard served a re-amended defence, in response to Socimer’s substantially re-amended particulars of claim served on 21 March 2005 inter alia introducing its Socma Primary Case about the Socma DRs. Among the amendments made by Standard were passages specifically relying on citations from the judgment of Cooke J. In particular, at section 14 of the re-amended defence, in reliance on the judgment of Cooke J Standard pleaded that “the Court is seeking to reconstruct the contractual valuation as it would have occurred” as of 20 February 1998, and that for this purpose the valuation process and the value itself, subject obviously to good faith, was entirely within the discretion of Standard. It then proceeded to plead the essence of Standard’s case on its “would have” valuation which had been previously set out in the statements of Messrs Clifford and Feld, to which reference was made.

In its re-amended reply served on 6 May 2005, Socimer pleaded (at para 132) that “the matters there alleged are irrelevant to the question of valuation, which is a matter of expert opinion and not a matter of what Standard Bank would or would not have done had it not breached the Agreement”.

Although Mr Millett was to submit to Gloster J that Standard’s “would have” case was the result of a late amendment for which permission should not be given, and was wrong and irrelevant and should be rejected at the outset, the fact is that it had always been inherent in Cooke J’s judgment that the valuation which had to be carried out at trial was the contractual valuation which Standard ought to have carried out on or as at the termination date but had failed to carry out; and that such a valuation was one which lay in Standard’s hands and for which it had a wide discretion. On the contrary, it was Socimer’s case, namely that the contractual valuation which had to be reconstructed was an objective one based entirely on independent expert evidence, which was new to the litigation, being inconsistent with what had been common ground before Cooke J and had been incorporated in his judgement.

As will appear below, Mr Millett failed before Gloster J in his procedural objection to shut out Standard’s case and its witnesses’ evidence on it as a matter of principle, although the ultimate answer Gloster J arrived at was in practice not far removed from his original submission. This was because she also accepted Mr Millett’s new case, developed only at the conclusion of the trial and never pleaded, that the valuation obligation contained in the agreement was subject to an implied term requiring an objective valuation to be carried out with reasonable care; and also because she was persuaded that, on such a case, Messrs Clifford’s and Feld’s evidence, although not challenged in cross-examination, was worthless. It is necessary therefore to show how the argument at trial developed procedurally.

Battle was first joined at a pre-trial review held before Andrew Smith J on 27 May 2005. Socimer there sought to shut out much of Standard’s evidence, both factual and expert, on various grounds. We are interested in the attack on the evidence of Messrs Clifford and Feld, which was put on the basis that they were factual witnesses who could not give expert valuation evidence. Andrew Smith J ordered that such issues should be left over for the trial judge.

Socimer’s opening submissions in writing were served on 3 June 2005. They contained passages dealing with the status of Messrs Clifford’s and Feld’s evidence, an issue on which it was said that a ruling would be needed at the start of the trial. The objection remained that they were purporting to give expert valuation evidence (for which permission had not been given) under the guise of being witnesses of fact. But that rather technical objection was premised on the more substantive submission that it was only independent expert evidence that could be relevant to the valuation exercise on which the trial was engaged. At that stage this submission was put in this way (at paras 20/23 of Mr Millett’s written submissions):

Now this may be shrewd advocacy, but its analytical basis, viewed through what is now known about this appeal, is suspect. Mr Millett accepts that the basis of Cooke J’s judgment is that Standard “can apply its subjective views” in exercise of a “broad discretion” in good faith to its assessment of value, but at the same time asserts the need for an entirely objective appraisal of value by the Court based on only independent expert evidence. Other than a plea, which this passage is in part, that Standard’s factual evidence of what it would have done is worthless as being partisan and therefore, as is hinted, possibly lacking in good faith or motivated by hostility or other irrationality, there is no foundation at all for the submission that the subjective assessment required by Cooke J should be supplanted by an entirely objective one. The answer to this advocacy could well be: “Cross-examine and we will see how we go. The court will not assume the worst of witnesses before they are heard.” The court might also have asked, to pick up the words of Cooke J to which Mr Millett had there alluded: “Do you have a case of bad faith or challenge on any other basis?”

In Standard’s written opening submissions in response, Mr Auld sought to interpret Socimer’s approach under the heading of “Reasonableness” (at para 47). He deprecated any attempt to “try and assert some implication of “reasonableness”; and he stoutly defended his witnesses against any charge of unreasonableness.

The trial before Gloster J began on 8 June 2005 and these themes were addressed by Mr Millett on the first morning. He submitted that where Designated Assets are retained, an “objective standard” was necessary because of the potential conflict of interest between the parties. If Cooke J had not imported a reasonableness qualification into the process, “then I invite your Ladyship to do so…We say it is not a would have case, it is a should have case” (at Day 1.35/37). Before the close of that day Gloster J raised the pending application to exclude the evidence of Messrs Clifford and Feld. She expressed a reluctance to rule anything out at the beginning, and a preference to hear the evidence de bene esse. Mr Millett, however, said he wanted the issue decided out front so that it would be clear that he did not have to cross-examine or put his expert case in cross-examination to them. However, he was then instructed by Socimer to accept the judge’s suggestion, and Gloster J said: “You will have to prepare your cross-examination”, to which Mr Millett replied: “As far as I am concerned, it gives me some practice for the real experts” (Day 1.162/166). At that stage, therefore, the decision in principle was going to be delayed for judgment, and Mr Millett was going to cross-examine.

On Day 2, Mr Auld made his opening oral submissions on behalf of Standard (see at Day 2.120/125). He contrasted Standard’s would have valuation limited by good faith with Socimer’s should have valuation based on reasonableness. His approach was that it was a matter of construction. The judge then raised the question whether it was not one of those cases where an obligation to do something involved “an implied duty to do so carefully”. Mr Auld said, accurately in my judgment, “That is not being put against us at this stage”. The judge next raised with Mr Auld another way in which Socimer’s position might be protected, namely by way of analogy with the position in equity as between mortgagor and mortgagee. It would seem therefore that both the concepts of an implied term and of the position in equity between mortgagor and mortgagee originated from the judge. As it was, however, neither concept had been pleaded by Socimer.

When Messrs Clifford and Feld came to give evidence, Mr Millett did not cross-examine them on the evidence they had given as to how Standard would have valued the Designated Assets. Despite his earlier exchanges with the judge on such cross-examination, Mr Millett’s decision was quite deliberate. This was remarked on by Mr Auld in his closing oral submissions, when there was the following exchange with an apparently unsympathetic judge (at Day 10.55/56):

In that exchange Mr Auld was dealing with the “broad point of fairness” that in the absence of cross-examination counsel cannot simply assert a challenge of dishonesty or capriciousness. A little later, however, he made the additional submission that in any event Socimer had no pleaded basis for an alternative challenge on the ground of unreasonableness or breach of a duty of care. I will return to that point.

Later that day Gloster J raised with Mr Millett in his closing speech the issue that he had not cross-examined Messrs Clifford and Feld on their evidence of valuation. Mr Millett then made it plain that this had been deliberate tactics. The following exchange occurred (Day 10.138/141):

In this important exchange, Mr Millett makes the following points: (1) that he had chosen not to cross-examine the witnesses deliberately, in reliance on his legal point (which of course the judge had still not resolved) that their “would have” evidence was irrelevant and inadmissible; (2) that he did not expect that any cross-examination would have made any difference to their evidence; (3) that despite those matters, he would invite the judge to regard their evidence with scepticism as being “self-serving” and indeed to reject it as “incredible”; (4) while at the same time acknowledging that he could not otherwise dispute it. In my judgment, those are unsatisfactory submissions. Of course, if the evidence was irrelevant and inadmissible, there was no problem. The judge, however, had asked her question on the basis that she decided otherwise. The other points were a polite way of asking the judge to reject the unchallenged evidence on the basis that it was not given in good faith. The only other theoretical possibility was that the evidence was given in good faith but was the result of the witnesses honestly but wrongly persuading themselves that an incredible account was a true account of how they would themselves have behaved. In the circumstances, however, that was not a feasible alternative, and was inconsistent with the idea of their evidence being self-serving. It was also inconsistent with the submission that no amount of cross-examination would have altered the evidence. Moreover, there was no suggestion that what was in issue was that entirely normal experience in the courts of witnesses having honestly different views about the extent of their duties or the performance of them.

I revert to the separate point as to whether a case of unreasonableness or negligence could be advanced, in the absence of a pleading, on some such basis as an implied term. This was the subject-matter of increasing debate beyond the confines of the trial itself.

The final day of evidence had been on 24 June 2005. There was then a pause while the parties put their closing submissions into writing. During that pause Mr Auld’s written submissions were served on 4 July 2005. He dealt briefly with “reasonableness” at para 40 of his document. He asked the judge to be extremely cautious, since “There is no basis for implying contractual terms to this effect (if this is what is alleged)”.

Mr Millett’s written submissions were exchanged on the same day. They opened with new and extensive submissions of construction as to the content and basis of the valuation obligation. He now submitted that there were two possible legal sources of a duty which he defined as one to perform the valuation exercise “not only in good faith but with reasonable care, by reference to objective criteria existing at the time”. The first legal source was “An implied term to perform the valuation with reasonable care”. The second was “An obligation in equity not to damage the residual interests of the mortgagor in the mortgaged property”. In both respects he made use at this point of suggestions floated by the judge during Mr Auld’s opening oral submissions.

On 7 July 2005 the parties returned to court for a final day of closing oral submissions. I have already cited passages from them in relation to the issue of cross-examination. As for the implied term now put forward, Mr Auld observed that nothing of that kind had ever been pleaded against Standard so as to go beyond the matters to which Cooke J had alluded, such as good faith, capriciousness and the like. He further observed that if such a case had been made, Standard would have wanted to adduce further evidence, such as that of market practice. Mr Millett responded to say that it was a pure point of law, although he accepted that an implied term had never been pleaded. During his final oral submissions Mr Millett elaborated slightly in an exchange with the judge (at Day 10.98/99) on the difference between “unreasonable” in the sense of negligent and in the sense of “Wednesbury unreasonable”. The judge suggested that the former went beyond the latter. Mr Millett suggested that in this case it amounted to the same thing, and that “whether you call it an obligation of care is probably just playing with words”.

On 17 July 2005 Socimer sent to the judge a “Post-Trial Note” containing extensive further legal submissions concerning the mortgagor/mortgagee analogy. On 22 July 2005 Standard responded with further written submissions of its own. On 14 October 2005 Socimer had the last word with its “Response Submissions” which developed still further its implied term and equity bases for a duty of reasonable care. That document repeated the oral submission that Socimer perceives “little if any difference between” the duty to act with reasonable care and the duty not to behave Wednesbury unreasonably.

The judge was then left in peace to write her judgment. Before I turn to that judgment, it will I think be convenient if I refer to some authority discussed in her judgment about which there has been little controversy on this appeal. That authority is concerned with the implications implicit in the contractual allocation to one party of a power to make decisions.

When a contract allocates only to one party a power to make decisions under the contract which may have an effect on both parties, at least two questions arise. One is, what if any are the limitations on the decision-maker’s freedom of decision? The other is, what is to happen if the contractual power was not in fact exercised at the time when the relevant party was obliged to make a decision?

The answer to the first question is illustrated by cases such as the following. In Abu Dhabi National Tanker Co v. Product Star Shipping Ltd (The “Product Star”) (No 2) [1993] 1 Lloyd’s Rep 397 the charterparty contained a clause which gave charterers the right to alter the destination of the cargo in circumstances where the contractual port of loading or discharge was blockaded owing to war and “the loading or discharging of cargo at any such port be considered by the Master or the owner in his or their discretion dangerous”. The trial judge, upheld by the court of appeal, held that the owners’ purported decision under this clause was wholly unwarranted, and that in fact they did not consider it dangerous to proceed to the contractual loading port. Leggatt LJ (with whom Balcombe and Mann LJJ agreed) said this about the content of the owners’ power (at 404):

Ludgate Insurance Company Ltd v. Citibank NA [1998] Lloyd’s Rep IR 221 concerned an agreement by which the London Market Letter of Credit Scheme was operated by Citibank. In certain circumstances the agreement gave to the bank the rights “to retain in the account(s) such additional margin as it considers appropriate in all the circumstances” and “to allocate the drawing(s)…in such manner as the Bank considers appropriate in its sole discretion”. Waller J and this court held that Citibank had exercised its decision-making rights in accordance with the purposes for which they were granted. Brooke LJ (with whom Mummery and Russell LJJ agreed) said this (at 239/240):

Gan Insurance Co Ltd v. Tai Ping Insurance Co Ltd (No 2) [2001] EWCA Civ 1047, [2001] 2 All ER (Comm) 299 concerned the claims co-operation clause in a facultative reinsurance policy. The clause required the prior approval of the reinsurers for any settlement or compromise of an underlying loss. The issue was raised whether it was to be implied that reinsurers could not withhold such approval unless they had reasonable grounds for doing so. This court, in a judgment given by Mance LJ (in which Latham LJ and Sir Christopher Staughton shared) held that no such implication was to be made. Mance LJ said –

Paragon Finance plc v. Nash [2001] EWCA Civ 1466, [2002] 1 WLR 685 concerned a variable interest clause in a mortgage agreement. The issue was whether the discretion given to the mortgagee to vary the interest rate was subject to an implied term that it was bound to exercise the discretion “fairly as between both parties to the contract, and not arbitrarily, capriciously or unreasonably”. Dyson LJ (with whom Thorpe LJ and Astill J agreed) accepted a limited implication in which “unreasonably” was understood in a sense analogous to the Wednesbury sense: Associated Provincial Picture Houses Ltd v. Wednesbury Corporation[1948] 1 KB 223. That was the sense in which Leggatt LJ had used the expression in The Product Star: see at paras 37/38. Dyson LJ concluded:

The second issue referred to above, is what is to happen if the contractual discretion ought to have been exercised but has not. That is answered in this court by Cantor Fitzgerald International v. Horkulak[2004] EWCA Civ 1287 (14 October 2004). There an employee who had been wrongly dismissed sought compensation to include a discretionary bonus which he might otherwise have been awarded. This court held that the court’s task in such a case is to put itself in the shoes of the decision-maker. Potter LJ (giving the judgment of the court, which also included Carnwath LJ and Bodey J) said:

It is plain from these authorities that a decision-maker’s discretion will be limited, as a matter of necessary implication, by concepts of honesty, good faith, and genuineness, and the need for the absence of arbitrariness, capriciousness, perversity and irrationality. The concern is that the discretion should not be abused. Reasonableness and unreasonableness are also concepts deployed in this context, but only in a sense analogous to Wednesbury unreasonableness, not in the sense in which that expression is used when speaking of the duty to take reasonable care, or when otherwise deploying entirely objective criteria: as for instance when there might be an implication of a term requiring the fixing of a reasonable price, or a reasonable time. In the latter class of case, the concept of reasonableness is intended to be entirely mutual and thus guided by objective criteria. Gloster J was therefore, in my judgment, right to put to Mr Millett in the passage cited at para 57 above the question whether a distinction should be made between the duty to take reasonable care and the duty not to be unreasonable in a Wednesbury sense; and Mr Millett was in my judgment wrong to submit that it made no difference which test you deployed. Lord Justice Laws in the course of argument put the matter accurately, if I may respectfully agree, when he said that pursuant to the Wednesbury rationality test, the decision remains that of the decision-maker, whereas on entirely objective criteria of reasonableness the decision maker becomes the court itself. A similar distinction was highlighted by Potter LJ in para 51 of his judgment in Cantor Fitzgerald. For the sake of convenience and clarity I will therefore use the expression “rationality” instead of Wednesbury-type reasonableness, and confine “reasonableness” to the situation where the arbiter on entirely objective criteria is the court itself.

I do not think that the body of authority cited above is in dispute. Rather, what Mr Millett submits is that it is always open in any particular contract to argue for a more extensive implied limitation. His reasons for saying why an implication requiring an objective assessment by the court is to be made in this case, if it is open to Socimer, will still need to be considered below. It is, however, against the background of this authority that the argument will proceed.

In his judgment Cooke J cited no authorities and I do not know what if any authorities were cited to him; possibly none. Nevertheless, when he spoke (at para 32) of the valuation exercise lying “entirely in [Standard’s] hands” subject to the proviso that its assessment “is in good faith and is not challengeable on any other basis”, he was referring either through knowledge or sound intuition to the body of authority which I have cited above. He was not intending to suggest an implied term changing the whole basis of the assessment.

I now turn to the judgment of Gloster J.

We are concerned only with those parts of Gloster J’s judgment which relate to the three issues defined in the Introduction above. For reasons which will become clearer, it is not easy to state precisely the terms of her findings or holdings. It will suffice for the present to say, however, that on those issues she held in favour of Mr Millett’s submissions, as follows: on issue 1, the “implied term” issue, that the agreement was subject to an implied term requiring Standard to carry out its valuation reasonably and with reasonable care, so as to arrive at what objectively can be said to be a proper value of the Designated Assets, as supported by the analogy with the duties imposed by equity on a mortgagee in possession; on issue 2, the “Unpaid Amounts” issue, that whether or not Standard was obliged to set off the $4,000,000 of other credits against the Unpaid Amounts as of the valuation date, it would have done so, thus opening the door to a further finding that, even if Standard’s own valuations had been accepted, there would have been a surplus of assets which would have led to the transfer to Socimer of its purchase of Socma DRs; on issue 3(1), the “open to Socimer” issue, that there was nothing to prevent Socimer adopting its new construction of the valuation obligation; and on issue 3(2), the “unchallenged evidence” issue, that although the factual evidence of Standard’s witnesses as to their “would have” valuation was admissible, it was worthless and could fairly be rejected, even though it had not been challenged in cross-examination.

It is necessary, however, to go into greater detail.

As for issue 1, the judge’s expression of her decision was not uniform throughout her judgment. She considered the parties’ submissions and stated her conclusions on this issue at paras 31/45 of her judgment. Mr Millett’s submission was that Standard was “bound to perform the valuation exercise not only in good faith but with reasonable care, by reference to objective criteria existing at the time” (see para 56 above). The valuation sentence was silent as to machinery and did not confer on Standard the broad discretion elsewhere referred to in the agreement. The rationale was the need to protect Socimer from the abuse of Standard’s self-interest. The two possible legal sources were (a) an implied term, whose necessity was proved by the rationale just referred to, and (b) an obligation analogous to that in equity “not to damage the residual interests of the mortgagor in the mortgaged property”. Mr Auld, however, submitted that where a party has a contractual discretion, the law makes no implication that such discretion should be exercised reasonably with regard to the interests of the other party. There was in any event no justification of necessity for the implied term in question. As for the mortgagor/mortgagee analogy, it had no application to the agreement and would have the effect of completely rewriting it.

The judge concluded that “Standard was obliged to act honestly and reasonably and to arrive at a value which properly reflected the actual value of the Designated Assets as at the termination date”, albeit it might also be entitled to adopt “an extremely conservative approach” to the question of valuation (para 36). However, when the judge came to expand on her reasons, it appears that there might be some confusion or conflation in her mind between the concepts of reasonableness and rationality, for she referred to the cases from The Product Star to Cantor Fitzgerald as demonstrating that the courts are not slow to imply a term that a party must exercise its contractual discretion “not only honestly and in good faith, but also reasonably and not capriciously” (at para 38). I have made the point above that those cases use “reasonably” in that context to refer to a Wednesbury-type rationality. Moreover, she went on to say this (at para 40):

Up to the words “in a reasonable manner”, that formulation would have had Mr Auld’s agreement; but the words thereafter, reflecting similar words in para 36, are plainly liable to go beyond the concepts of good faith and rationality. This extension of the concepts of good faith and rationality is again demonstrated by the judge’s acceptance of the analogy of the duties of a mortgagee in equity. The judge listed those duties (at para 42(a) to (h)) by reference to the relevant jurisprudence reviewed by Lawrence Collins J in Meftah v. Lloyds Bank plc (No 2) [2001] 2 All ER (Comm) 741 as including: a duty of care, a duty to obtain market value, and a duty fairly and properly to expose the property in the market, (albeit the mortgagee will not be adjudged in default unless he is plainly on the wrong side of the line). She said that there was every reason, in the need to value and to return any surplus to Socimer, “for the implication of a duty closely analogous to that of a mortgagee” (at para 43). She also relied (at para 44) on a similar restatement of the mortgagee’s duties in equity taken from para 19 of Lightman J’s judgment in this court in Silven Properties Ltd v. Royal Bank of Scotland [2003] EWCA Civ 1409, [2004] 1 BCLC 359. Gloster J then ended this passage of her judgment by restating her implied term in the following terms:

Therefore, despite some uncertainty engendered by the judge’s acceptance that Standard had a discretion as to its assessment and by her conflation of good faith and reasonableness, it appears that on balance she was accepting in full or at least in general both Mr Millett’s approach as a matter of law and his submission that, subject at any rate to the possibility of taking a conservative approach, what was required on objective market-based criteria was the true market value. Only such a value could also be described as fair, reasonable or proper. Standard not only had an obligation to use reasonable care to value at such a value, but had to arrive at it: the judge sometimes stressed the former aspect and sometimes the latter.

These conclusions of the judge have, however, to be balanced with what she said about the issue between the parties as to whether Standard’s factual “would have” case was relevant and admissible. She discussed this issue even before considering the matters of law just reviewed above, at a passage in her judgment headed “Objective/subjective approach to valuation” (at paras 22/28). On this issue, on balance she favoured Mr Auld’s approach. She said (at para 25):

Of course, the “would have” valuation to be expected is that which Standard would have performed if carrying out its obligations. In other words, Standard could not perform a valuation in bad faith, to refer to the obligation of good faith about which the parties are in common accord. However, that is not the “should have” valuation which Mr Millett was contrasting with Mr Auld’s “would have” obligation. Mr Millett’s “should have” valuation is his obligation to arrive, objectively, at the true market value, based on his implied term. Mr Auld’s “would have” obligation was limited to that valuation which Standard was entitled to make in the exercise of its own discretion, in good faith and rationally, but otherwise consulting its own interests and therefore in that sense aiming at a subjectively valid as distinct from objectively true value. Thus, when Gloster J went on later in her judgment to prefer Socimer’s submissions as to the implied content of the valuation sentence, she was removing with one hand what she had granted to Standard with the other.

Gloster J returned to these problems when she came to assess the evidence before her on valuation and to state her decisions about it. It was in the particular context of the valuation of the Brazilian TDA-Es that she made the following observations, which in my judgment encapsulate and point up the inconsistency which she had allowed to develop in her previous conclusions. She said (at para 94):

Had this been all that the judge had said on these matters, then, like Cicero, Mr Auld would have been able to disdain the adversarial swords of Mr Millett’s advocacy. (Footnote: ¹) However, on this basis what she had said about the need to arrive at an objective market value would have been beside the point. The concern is therefore that ultimately in valuing the Designated Assets she reversed her stated position in this passage, by the combination of maintaining her implied term analysis requiring an objective assessment of the true market value, and of rejecting Standard’s (albeit unchallenged) evidence.

That concern is supported by the following considerations. First, this passage at para 94 comes after the judge had already valued the Socma DRs not only on Socimer’s primary but also on its secondary case. The judge therefore would presumably at that earlier point have had her implied term basis of valuation predominately in mind. Secondly, in valuing the Socma-DRs on Socimer’s primary case (ie on the basis that they should have been transferred over to Socimer) the judge said in terms that in such circumstances the value to be placed on them “must be the value of the Socma DRs in Socimer’s hands, not an objective market based value” (emphasis added, at para 73), where the contrast is with the objective market value which would be applied on Socimer’s secondary case. Thirdly, in dealing with that secondary case (at paras 83/86) the judge rejected the evidence of all the witnesses she had heard (Socimer’s case was that the Socma DRs were worth some $7.5 million, Standard’s that it would have valued them at $500,000), and selected a figure of about $3,000,000. It is true that she did so on the basis that “Standard would have been likely, acting cautiously as it was entitled to do, to have ascribed a value of approximately US$ 3 million” to them: but she gave no reason for that figure other than going on immediately to say that “I am supported in this conclusion by the actual price of US $2 million that was achieved subsequently when Standard sold the instruments on 12 January 2000”, that is to say nearly two years after the termination date (at para 86). It is not clear how that figure supports the judge’s figure, but I am left with the conclusion that the judge has made her own assessment of an objective market value and ascribed that to Standard. In effect, having made her own assessment, she has ascribed that to Standard as the valuation it would have made in accordance with its valuation obligations.

Fourthly, turning to the valuation of the Brazilian TDA-Es, I observe that the judge essentially adopted the evidence of Professor Rosenberg, and rejected that of all Standard’s witnesses, both factual and expert (although she gave some detailed consideration to Mr Quintero’s evidence as to his own hindsight figure of some $5.5 million). Professor Rosenberg’s figure was between $8.9 and $9.4 million (para 114(i)), and he was giving evidence based on Socimer’s “should have” objective market value approach. The judge adopted that figure, while making a discount to $8.5 million out of consideration for Standard’s entitlement to a cautious and conservative approach. She commented (at para 114(iii)):

What is going on here? It is hard to say. Mr Millett did not address us on the role of para 94, apart from referring to it as one of a number of places where the judge indicated that she did not simply ignore the evidence of Standard’s witnesses. It remained his overall submission nevertheless that “the whole trial proceeded on the basis that the numbers were properly the subject of expert evidence”, that “The search was for an objective true value” and that “the judge was faced with really a range of numbers from which she could choose” (12 December 2007 at 133/134). Doing the best I can, I would provisionally interpret the judge as saying that because Standard’s evidence was unacceptable by reference to its obligations of good faith and reasonableness, she was left only with expert evidence as to what a proper valuation should have been, on which subject she preferred the evidence of Professor Rosenberg. That is why she expressed her conclusion at para 114(iii), quoted above, by saying that the figure of $8.5 million was one that Standard “could legitimately have put forward in compliance with its obligations of good faith and reasonableness”. In other words a lower figure could not be put forward on that legitimate basis. Subject to the ambiguity inherent in the word “reasonableness” discussed above, that is a return to the judge’s implied term. This would be consistent with my understanding expressed above of the judge’s conclusion on Socimer’s secondary case as to the value of the Socma DRs. In other words, the judge has made her own assessment of an objective market value and then ascribed that to Standard, subject to a discount for cautiousness, on the basis that that figure is the figure it would have arrived at if complying with its (implied) obligations. Moreover, the reference by the judge to the obligation of good faith suggests that she also found Standards’s valuation to be lacking in this respect as well.

Because these matters, on my provisional view, are linked with the judge’s attitude to the legitimacy of rejecting Standard’s factual evidence, I proceed next to consider in further detail the judge’s conclusions as to issue 3(2), the “unchallenged evidence” issue. The judge did not refer at all to Mr Auld’s submissions that it would be unfair to reject the evidence of Messrs Clifford and Feld in the absence of any challenge to it by way of cross-examination. Nevertheless, the judge did reject their evidence on valuation. She reflected Mr Millett’s submission throughout the trial that their evidence should be discounted on the ground that it was partisan and self-serving. Thus she said of Mr Clifford that “I had to approach his evidence on the basis that there was a real risk that it might be self-serving and partisan” (at para 106); and she went on immediately to say of Mr Feld that “Likewise I did not find Mr Feld to be a persuasive witness. His evidence also gave the impression at times of being self-serving, partisan and argumentative” (at para 107). Those were general comments. Under the immediately following heading of “The evidence relating to the valuation of the TDA-Es” she said “I did not find the evidence given by the Standard witnesses as to the valuation figure which they said they would have ascribed to the TDA-Es persuasive” (at para 108). The reason she gave was that neither of them had done “a retention valuation of the sort contemplated by the Agreement”, ie a clause 14(a)(bb) valuation, before. “Thus it was not a situation where retention valuations of Designated Assets following default were routinely carried out by Standard.” That was so, for clause 14(a)(bb) had never been interpreted by Standard in the way Cooke J had interpreted it. However, the judge did not say that they were not experienced bankers in the business of trading in and valuing the securities in emerging markets: as at any rate on their evidence they purported to be. The judge also gave other reasons, such as lack of support from Mr Quintero, for rejecting their evidence. We have not heard argument about those reasons in themselves. Therefore, I assume for present purposes that those reasons are sound, as far as they go. Clearly, Professor Rosenberg was an impressive witness.

The judge was of course entitled to form a view about the evidence before her, as Mr Millett has correctly submitted to us. Even so, there are three things which concern me in this regard. First, did she reject evidence which had not been fairly and properly challenged by Socimer in the absence of cross-examination? Secondly, did she reject their evidence on a basis on which Standard had never been challenged at all, quite apart from the absence of cross-examination? Thirdly, did she reject their evidence because she was expecting too much of it by imposing an “objective true market” test on Standard’s obligation?

I ask my first question because Mr Millett accepted both before the judge and again before this court that he had deliberately decided not to cross-examine Messrs Clifford and Feld on their evidence that they would have valued the assets in the way that they said they would have. I ask the second question, because Mr Millett acknowledged before us that he did not challenge the honesty or good faith of their evidence and had never done so. He was quite specific about that. I ask the third question because of those parts of the judgment which suggest that Gloster J was considering the evidence through the prism of an obligation (to take care) to arrive at an objective true market value.

I observe that in this connection: (1) The judge gave no reason for rejecting Messrs Clifford’s and Feld’s “would have” valuation evidence regarding the valuation of the Socma DRs. In this respect, as I have remarked above, she did not accept the evidence of either party’s expert as to this Argentinian corporate bond over which Socimer’s impending insolvency threw its own pall (Socimer’s relevant expert, Mr Tether, valued this asset at about $7.5 million, Standard’s expert, Mr Doyhambehere, said there was no market at all and valued it at nil). However, other than saying that she had “heard all the evidence”, she gave no reason for rejecting Messrs Clifford’s and Feld’s valuation at $500,000 and for concluding that Standard “would have” ascribed a value of “approximately US$ 3 million”. (2) In connection with the Brazilian TDA-Es, one of the judge’s reasons for rejecting the evidence of Messrs Clifford and Feld was expressed as follows (at para 109):

I therefore conclude that the provisional view expressed above, namely that the judge had it in mind that the evidence of Messrs Clifford and Feld was unacceptable because it was not consistent with Standard’s obligations of good faith and reasonableness, at any rate in the light of its implied term obligation, is the correct way to understand her judgment. Thus I am also concerned that one of the reasons which has led Gloster J to this strong conclusion is that she is also judging Standard’s evidence by a test (the implied term test) for which its evidence was never prepared.

Issue 3(1), the “open to Socimer” issue is not unconnected with issue 3(2), because, if Socimer’s implied term basis for criticising Standard’s valuation evidence was not properly available to Socimer, then that would be an alternative reason for being concerned if it be the case that the judge’s criticism of Standard’s evidence was premised at least to some degree on that test of its valuation obligation. On this issue 3(1), the judge rejected Mr Auld’s submission that a timeous pleading of an implied term would have affected the course of the litigation. She said (at para 37):

It is true that the implication is essentially one of law, being first and foremost a matter of construction. It may also be the case that the judge may have been right to have been sceptical of Mr Auld’s expressed concern, had such an implied term been pleaded, to have brought forward further (unidentified) evidence of market practice bearing on the question of implication (ibid). However, there was a good deal of factual and expert evidence about the matrix of the contract before Cooke J, and there must be a real possibility that the nature of that evidence would have been different had Standard known that Socimer was contending for an implied term that both limited the broad valuation discretion which had been common ground before Cooke J and required the search for an objective true market value. Moreover, there is a still further concern, in the light of Gloster J’s judgment, about the application of such a new implied term to the facts of the case. There is no point in bringing a new implied term to bear unless there is also a case that Standard would have been in breach of it by valuing the Designated Assets in the way in which they said they would have valued them. However, such a case had not been pleaded. If it had been, Socimer’s witnesses would presumably have wished to address it in their evidence, beginning with their witness statements. It is fanciful to suppose that there was not further evidence which Standard would have wished to deploy, had Socimer pleaded its implied term and particulars of breach of it.

This is again linked to the absence of challenge by way of cross-examination to Standard’s factual witnesses. If, on the basis of the new implied term, Standard’s obligation was to take reasonable care in its valuation and/or to arrive at an objective true market value, what is the function of such a term unless Standard is given proper notice of how it is said that Standard’s “would have” valuation failed to meet Standard’s contractual obligations, and its witnesses are also given a fair opportunity in cross-examination to meet and answer any challenge which is being made? And if no challenge is made to Standard’s factual witnesses, what is the point of the new implied term? And yet, seeing that the implied term was never pleaded, was new to the trial and indeed had not been clearly formulated until final speeches or even beyond final speeches, I find it difficult to discern a proper case being made of how it is said that Standard breached that implied term.

It is therefore possible to see that at the root of all these concerns is the appearance that Gloster J was judging Standard’s witnesses against a test which they had never come to court to deal with. They prepared their evidence of a hypothetical “would have” valuation on the basis of the valuation exercise as interpreted by Cooke J. But their evidence appears to have been rejected on the basis that it did not match up to a different interpretation of their obligations.

Issue 2, the “Unpaid Amounts” issue is self-contained. I shall therefore deal with the judge’s analysis of it below, together with the parties’ submissions on it at this appeal, and my own conclusions about it.

I turn now therefore to the resolution of each of the three issues we have heard argued.

The essence of the conflicting submissions has already been stated above. In brief, on behalf of Standard Mr Auld submits that, even if the case in favour of an implied term was open to Socimer, it should have been rejected by the judge. An implied term was not necessary. It ran against the vein of the agreement as a whole, which plainly gave the determination of value to Standard, in the exercise of its subjective judgment and subject to a wide discretion. Even if that was not expressly recognised in the valuation sentence itself, it was reflected in the definition of “Market Value”, and in the sentence immediately preceding the valuation sentence, and in the provision for a conclusive certificate from Standard on the amount of any deficiency, which was binding subject only to manifest error. The context was the forward sale and purchase of assets in volatile and illiquid emerging markets. The valuation provision was triggered by the buyer’s default. The effect of the “liquidated or retained” option was to require Standard to retain for its own account and risk any and all assets which it was unable immediately to liquidate on the valuation date itself (subject to any choice not immediately to liquidate a liquid asset). Even though the definition of “Market Value” did not apply in terms to the “value” referred to in the valuation sentence, it would be nonsense to give to that latter word the meaning of “market value” in an essentially different sense from that in which “Market Value” was deployed in the agreement. The hindsight imposition of an objective market value for an assessment which had to be done in an emergency on the termination date itself, in respect of Standard’s own property, which it was forced in breach of contract and against its will to retain, would be unfair and uncommercial. All of this was underlined by the terms in which the preliminary issue had been argued before and decided by Cooke J.

Moreover, no insight or support was to be derived from the cases about the position in equity of a mortgagee. The Designated Assets did not belong to the buyer, there was no equity of redemption, and the context of the agreement between two sophisticated trading partners in volatile markets was wholly different.

On behalf of Socimer, however, Mr Millett submitted to the contrary. He acknowledged that the valuation sentence was silent on the mechanism of valuation and that an implication therefore had to be made if his case was to succeed, but he submitted that it ought to succeed, for his term was necessary, in the traditional sense, for the sake of business efficacy. It would have been the common accord of the parties if raised as an issue by the “officious bystander”. What was his term? It was that which Gloster J referred to at para 36 of her judgment, viz that “Standard was obliged to act honestly and reasonably and to arrive at a value which properly reflected the actual value of the Designated Assets at the termination date” (paras 35/36 of his skeleton argument on appeal). He also glossed this as an obligation “to perform the valuation exercise not only in good faith but with reasonable care, by reference to objective criteria existing at the time” (ibid at para 36(1)). The rationale of the implication was that, although Standard could be trusted in its own interest to get the best price it could when selling a Designated Asset, however when valuing a retained asset it was in conflict with the interests of its buyer because it would wish to ascribe as low a value as possible so as to maximise a potential profit on later sale. “That is the abuse from which Socimer needs protection by the imposition of a standard of reasonable care in the valuation process” (ibid at para 36(1)(c)). An implied term to ward against such abuse was all the more necessary in the light of Standard’s ability to certify its deficiency for itself. The only way in which to ensure that Socimer is protected from the harshness of the binding nature of the valuation is to provide that it must be carried out by reference to objective criteria against which it can be assessed. A duty of good faith was insufficient, for it would not protect from honest but unreasonable or unfair valuations, such as the suggestion (already deployed before Cooke J) that where there was no buyer available on the termination date itself, the asset could be valued at zero. On the other hand, Standard did not need the protection of its own subjective assessment of value, for it already had sufficient protection in the other provisions of the agreement, such as the right to obtain downpayments on sale and additional downpayments as the market might decline or become more volatile.

Moreover, the fact that Standard was obliged to transfer to Socimer assets or cash which represented a surplus in Standard’s hands both confirmed the need for an objective assessment of value and demonstrated the applicability of the mortgagee analogy.

Finally, Mr Millett submitted that there was nothing inconsistent in the judgment of Cooke J. The primary issue debated before him had been the different question of whether there was any need for a valuation at termination date at all, and when he concluded that the valuation was to be done “in good faith and not challengeable on any other basis” he was allowing for the implication now relied upon.

In connection with these submissions it is necessary to refer to two additional authorities. The first is Lion Nathan Ltd v. CC Bottlers Ltd [1996] 2 BCLC 371 (PC), which Gloster J cited and Mr Millett relied on in his skeleton, but not in his oral submissions. There a share sale agreement contained a warranty that a profit forecast upon which the buyer had relied “has been calculated in good faith, and on a proper basis…and [is] achievable based on current trends and performance”. The trial judge held that the seller had warranted that reasonable care had been taken in the preparation of the forecast and that the seller was in breach of warranty because it had deduced its forecast from the starting point of the sale price, rather than the other way round. Thus the actual forecast had been prepared on a wholly wrong basis. The Privy Council held that it was necessary to approach the question objectively and ask what a reasonable forecast would have been. This would have permitted a range of forecasts all of which would have been reasonable. In those circumstances it was submitted on behalf of the seller that it should be assumed to have adopted a forecast which was the highest within the legitimate range. The Privy Council held however that in the absence of any evidence as to how the seller would actually have forecast its results on the correct methodology, the court could not choose any particular variation within that range and could make no assumption in favour of the seller. Lord Hoffmann said (at 380a):

As for that figure, Lord Hoffmann went on to reason that, in the absence of any evidence to displace the prima facie assumption that the most likely forecast would have accurately predicted the actual result, it was the latter which should be adopted as representative of the former. Damages were therefore awarded on that basis. Lord Hoffmann concluded (at 380i/381b):

Gloster J referred to Lion Nathan in two passages of her judgment: at para 27, in a part of her judgment under the heading “Objective/subjective approach to valuation” where she was considering whether Standard was entitled to tender evidence about how it would have valued the Designated Assets; and at para 41, where in the context of deciding the implied term issue, she turned aside from the question of construction to comment that Standard could not say that it had honestly valued an asset at nil (in the absence of a market on the termination date) if “the evidence showed that the proper value of such assets that would have been attributed to such assets by the hypothetical reasonable financial institution in its position was far greater”.

In his skeleton argument Mr Millett referred to Lion Nathan in support of the judge’s and his implied term (at para 36(9)).

In my respectful judgment, Lion Nathan throws little if any light on our present problems other than supporting, but less directly than Cantor Fitzgerald, the judge’s view that it was in order for Standard to tender factual evidence about how it would have carried out a contractual valuation. However, it throws no light whatsoever on the question of the construction of our agreement. In Lion Nathan the issue of construction had been decided by the trial judge, that the warranted forecast required the taking of reasonable care. The only issue of construction in the Privy Council arose out of the fact that the New Zealand court of appeal had differed from the trial judge in saying that the warranty that the forecast was “achievable” was a strict warranty and not merely one requiring the taking of reasonable care. There is no similarity between the forecast (looking into the future) of a warranty under a share sale agreement and clause 14(a)(bb) in our agreement; nor is there any similarity between the language of that warranty and our valuation sentence. In the present case, the issue is what the valuation sentence means or implies: and beyond the fact that a warranty which on one view is strict may be more generously construed as only requiring the taking of reasonable care (as well as honesty) Lion Nathan teaches nothing at all as to construction. As it is, I suspect that the citation of Lion Nathan is the reason why there is some uncertainty as to whether Socimer’s and the judge’s implied term is one only to take reasonable care to arrive at the objective true market value, or is one actually to arrive at it.

The only other lesson of Lion Nathan goes to the application of a forecast warranty requiring good faith and reasonable care (not the issue with which I am presently engaged). Lord Hoffmann said that the requirement of good faith in a term requiring reasonable care did not permit the seller simply to pick the highest possible forecast within a range of reasonable forecasts on the basis that that would then support the highest possible price for the sale of the shares (at 379h). In this case, however, whatever the judge may have decided, Mr Millett has made it clear to us that he has had and has no challenge to Standard’s honesty and good faith. Lord Hoffmann also said that in the absence of evidence to the contrary it should be assumed that the reasonably careful profit forecast and the actual result would be the same. That again has nothing to do with this case – save that if we had been concerned with the valuation of some wholly liquid security (a paradigm example of which would be (say) BP shares on an ordinary trading day) then it might be said that a valuation of such an asset which had been sold on the valuation (termination) day should, in the absence of evidence of anything untoward, be taken to be the sale price. However, we are not concerned with that scenario either. Neither the Brazilian TDA-Es nor the Socma DRs are suggested by Socimer as having been sufficiently liquid to have been liquidated on 20 February 1998.

The other authority which I need to mention at this point is the only authority cited in oral submissions by either party on the question of any implication (going beyond those of good faith and rationality etc discussed above). That is Phillips Electronique Grand Public SA v. British Sky Broadcasting Limited [1995] EMLR 472 (CA). Although not reported other than in a highly specialised series of reports, it is a useful and authoritative modern restatement of the relevant principles upon which terms may be implied and of the rationale of doing or not doing so. The judgment is of Sir Thomas Bingham MR and is a judgment of the court also constituted of Stuart-Smith and Leggatt LJJ. Sir Thomas Bingham said (at 480/482):

The judge made no mention of such doctrine or of any cases which discuss it. Implications of good faith and rationality, and of lack of arbitrariness or perversity, are standard, for they represent the very essence of business (and other) relationships. Once one goes beyond them, however, the matter becomes much more uncertain.

It is against the background of these submissions and this doctrine, that I approach the current issue on this appeal.

In my judgment, the implied term is not necessary or sufficiently certain and I would reject it. I would seek to put the matter in the following way.

Mr Millett accepts that an implied term is needed, for there is no express provision. In this connection he does not simply rely on the word “value” as containing the answer, but rather as disguising it.

There is uncertainty as to the term required. In essence, the term contended for is the finding, on solely objective criteria, of the true market value. However, sometimes the term is expressed as the obligation to take reasonable care to arrive at the true market value. These are not the same terms. A true market value, arrived at solely on objective criteria, at a certain moment, is likely to lie within a very narrow range indeed for the most liquid of assets and a somewhat broader but essentially identifiable range for an asset which, even if not entirely liquid, remains marketable. Of course there will always be a spread between a bid price (what it would cost a buyer to buy) and an offer price (what a seller would obtain to sell). That is not the range I am talking about, because I assume that the relevant price in this context, where Standard already owns the asset and retention is a substitute for liquidation, is an offer price. Even so, there will be a range: both because the market fluctuates during the day and because once one leaves behind the most liquid of securities, market making and pricing becomes increasingly more of an art than a science, or at best an artistic science. That said, if nevertheless the term is to find the true market price, then although experts may differ somewhat about it, it should always be possible, at any rate if there is a market, to obtain a reasonably narrow range of answers. Moreover, the best evidence will be that of the most expert and independent witnesses. If, however, the term is to take reasonable care to find the market price, then, where at any rate there is factual evidence of how the valuer would have valued, there will always potentially be a larger range of values within which the valuer cannot be said to be at fault. Therefore, the objective value arrived at by expert evidence and the hypothetical value proved by witnesses of fact who were seeking and are assumed to have carried out the valuation with reasonable care will not necessarily be the same. Moreover, a breach of the obligation to carry out a valuation with reasonable care will not be proved by the mere fact that a preferred expert arrives at a different answer from that of the bank’s hypothetical valuation, or even that he adopted a different methodology. The question will be whether it can be shown that the valuer in question was negligent. Negligence would have to be proved in the ordinary way.

Not only is there a difference between the various formulations of the implied term adopted by Socimer, but there is of course a difference between the implied terms adopted by Standard and Socimer. Standard says that the valuation in question is not that of an expert at all: it is one to be performed by Standard, and one to be performed by it within its discretion and according to its own subjective criteria. There are standard limits on that discretion and subjectivity, which are common ground: described by concepts such as good faith, honesty, rationality, arbitrariness, perversity, capriciousness. It might perhaps in the abstract be said to be strange to think of the value of securities in terms of discretion or subjective criteria: but that is not in fact said by Socimer. Socimer (and both Cooke J and Gloster J) accept such concepts. It is entirely clear from the contractual definition of “Market Value” that it is common ground between the parties that it is possible to talk in such terms. One might debate exactly what such concepts mean, although that debate has not been canvassed before us. I suppose that the ideas of discretion and subjective criteria reflect both the context that the securities traded under the agreement are or may be more or less volatile and illiquid (see for instance clause 16 and the acknowledgments by the buyer) and the consideration that Standard, as the party who is entitled to protect itself against market risk imposed on it by its counter-party’s default, is entitled to act in its own interests in measuring the risk: and that in such circumstances there may be much room to differ about value.

Thus in the specific context of a default and a forced retention of Designated Assets, Standard is compelled by its buyer’s default to retain what it never sought, save to the extent that it can immediately liquidate the assets on the termination date. The question whether it can sensibly in the interests of either party liquidate on the termination date is part of the complex uncertainties of this emergency situation. If it decides not to liquidate, it is forced to retain. If in that context it has to value the assets, why should it not be entitled to value them at a value which reflects the value of such assets to itself? It may dislike the risk they pose, in terms of the nature of the particular asset, its currency and/or nationality and so on. The decisions have to be taken very quickly, namely “on the date of termination” (see further below). Once the asset is not immediately sold, the risk of retention is entirely transferred to Standard. In theory and sometimes in practice anything may happen the next day, or within the time in which a sale might become possible. The difficulty multiplies if the asset is relatively or entirely illiquid. Then there is no market price by which the value can be set on the relevant day. Who knows at what price the asset can be sold when a buyer appears? In such circumstances, Standard is entitled, it may be said, to consult its own interests, subject of course to the requirements of good faith and rationality. Those factors include both subjective and objective elements, but the essence of that construction is that the decision remains that of Standard, not of the market or the court, and that in coming to its assessment, subject to the limitations of good faith and rationality, it is entitled primarily to consult its own interests.

There are still further possible variations of what might in theory be meant by objective criteria of reasonableness: such as that the value has to be “reasonable” in a mutual sense, as when a “reasonable price” is what implication requires to be found. In such circumstances, what is reasonable will depend on the assessment of something that is in both parties’ interests. The parties are in a good position in most cases to work that out for themselves, as long as goodwill continues to exist between them. If they cannot agree, the valuation will effectively be that of an independent valuer, and in the absence of any such person, of the court or arbitrator involved in resolving the parties’ dispute, again probably with the help of experts. However, I do not think that this fits the context of this case. The Designated Assets do not belong to Socimer, but to Standard. Standard does not want them, but is forced, subject to immediate liquidation, to retain them for itself, together with the ongoing risk. Socimer, although not the seller, is in the position of a forced seller, and Standard, although not the buyer, is in the position of a forced buyer. And what has forced this situation on both of them is Socimer’s default.

The valuation has to be performed “on the date of termination”. Cooke J contemplated that this might not be possible, for instance in a case where the default which marks the termination date occurs shortly before midnight (23.59 hours): to which might perhaps be added any period after the relevant markets have closed. Only in this sense might the valuation have to be conducted after the termination date, on a subsequent day. Where, as here, the termination date is a Friday, it may not be possible to value until the next business day. As a matter of fact, Friday 20 February 1998 was the eve of the Mardi Gras weekend, which created further difficulties. But even if in practice the valuation has to be performed later than the valuation date, it has to be performed as soon as practicable, and by reference to valuation date value: in other words “as of” the termination date. Mr Millett sought to submit that the valuation could be performed at almost any time, as long as it was “as of” the valuation date: but that is not what the agreement says. Cooke J was quite clear about this: see para 29 of his judgment cited above.

It follows that where there is no buyer on or as at the valuation date, there is an additional difficulty. What is the value to be put on an asset for which there is no buyer? Or for which there is no market? Or for which there is a volatile or illiquid market? Or a market which can only absorb a small amount of stock before the price is affected? The last quoted deal before the termination date may be of little assistance in such a case. The next quoted deal will be of little assistance, because it looks beyond the relevant day. In such circumstances, it might be rational to value the asset at or close to zero: but of course it does not necessarily follow. I do not think that there was any real difference about that, either before Cooke J or Gloster J. What is plain, however, is that where it is not possible to liquidate an asset on the termination and valuation date it is unreasonable that there should be any risk at all on Standard. It makes the implication of an objective hindsight valuation an unnecessary and unreasonable imposition.

Mr Millett relies on the fact that any surplus cash or assets must be returned to Socimer to submit that an implication is necessary for the avoidance of abuse. In my judgment, the requirements of good faith and rationality are a sufficient protection. The danger to be guarded against, he says, is abuse caused by self-interest. That is precisely what implicit good faith deals with. Commercial contracts assume such good faith, which is why express language requiring it is so rare. What Mr Millett’s implication is dealing with, however, is not abusive self-interest, but a (possibly negligent) failure to get the valuation exactly right because of the unconscious and insidious danger of self-interest. I agree there is such a danger: but the argument assumes that Standard is in the position of a neutral valuer, rather than a bank forced by its customer’s default to protect its own position in potentially highly volatile and illiquid markets.

The agreement was a carefully worked out contract between sophisticated parties. An implication in such circumstances is, as Sir Thomas Bingham said, “an altogether more ambitious undertaking”. There is no standard term (other than that of good faith and rationality) which applies in such circumstances. Cooke J referred to the fact that Standard’s terms differed from those of a standard set of terms known as the PSA/ISMA type agreements which made provision for a “Default Market Value” and which did not give to the seller the wide discretion which at that time it was common ground that Standard’s terms gave to the seller in a default situation.

There was disagreement before Gloster J as to the extent to which the Brazilian TDA-ES and the Socma DRs were more or less liquid or illiquid investments. Whatever be the answer to that dispute, there was no dispute that the Standard agreement was designed to cater for assets among which were those illiquid to the extent of being called “exotics”. In such circumstances, it may be that in the case of entirely or essentially liquid assets there would be little difficulty or difference in valuation, whatever the standard of value might be. However, the agreement had to cater for assets of all kinds, even the most exotic. Here, Standard could sensibly require the protection of a standard of value which left it with a wide discretion.

Mr Millett relied on the fact that, as was common ground, “value” in the valuation sentence was not the same expression as “Market Value” and that it was only the latter which was defined expressly in terms of “the value…determined by the Seller in its sole and absolute discretion”. However, I do not think that is of any particular significance. Of course, if the valuation sentence had referred to “Market Value”, the test would have been set out expressly. As it is, as Mr Millett accepts, the reference to “value” is silent about the value to be found and how it is to be found. Nevertheless, the valuation sentence does say that it was to be “determined…by the Seller”. In context, and with the certificate of deficiency to be settled by Standard alone, unchallengeable save for manifest error, Cooke J was entitled to say that the express discretion left to Standard in terms of “Market Value” was “a crucial factor when approaching the question of construction of clause 14” (para 19 of his judgment). I agree with Mr Auld’s submission that it would make no sense to give “value” the meaning of “market value” but with an essentially different meaning from that of “Market Value” as defined. The fact is that there was nothing in the agreement which supported the implication contended for before Gloster J, and much to make it unlikely. There was no appeal by Socimer from the judgment of Cooke J. I agree with him that the whole context of the valuation sentence was to leave the assessment “entirely in [Standard’s] hands” (at para 32 of his judgment). In those circumstances, the implication contended for becomes an impossible one. I do not agree with Mr Millett’s submission that Cooke J’s language “provided that the assessment is in good faith and is not challengeable on any other basis” (ibid) was leaving room for an entirely objective evaluation. That would have been entirely inconsistent with the judge’s general approach, and to what was common ground before him.

In this connection it will be recalled that one of the pleaded issues of construction at the time of the preliminary issue went to the nature or absence of Standard’s discretion. By the time of the oral hearing before Cooke J, however, that issue had gone: see paras 12/16 above. Socimer accepted that the logic of the situation might well be “Tough on the buyer”.

At the end of the day, Mr Millett asks us to imply a term in circumstances where his own proposed implication takes two different forms, and where it is at least possible to take different, equally rational views on an implication either restricting an assessment in Standard’s discretion to the traditional concepts of good faith, rationality and the like on the one hand, or an implication imposing entirely independent and objective “true market price” criteria on the other. This alone demonstrates that the parties would not have been likely to rejoin “of course” to the officious bystander; and that an implication such as that contended for by Socimer is neither necessary nor sufficiently certain..

In these circumstances, it is difficult to see how the analogy of the position of a mortgagee in equity could throw any light on the situation. In any event, I would respectfully agree with what Lord Justice Lloyd has to say about that. In my judgment, the analogy breaks down. Standard’s position is governed by its commercial contract, not by the law of equity. This is the world of sophisticated investors, not that of consumer protection. These merchants in the securities of emerging markets have made an agreement which speaks of the need for a spot valuation, not of the more leisurely process of taking reasonable precautions, such as properly exposing the mortgaged property for sale, designed to get the true market price by correct process. Meanwhile, the assets involved are those of Standard, not of Socimer: and the underlying background is that where the buyer defaults, he loses both the right to complete his purchase and his downpayment.

Of course, under the clause 14(a)(bb) option any surplus is the buyer’s, not his seller’s: but that tells one nothing in itself. The question remains how the termination date valuation, which may or may not throw up such a surplus, is to be made. If the clause 14(a)(aa) option had been exercised, then the buyer is excused completion, the full market risk of the Designated Assets rests with Standard, but Standard retains the downpayments to cushion it against market loss, past or future. If, however, the clause 14(a)(bb) option operates, then the buyer not Standard gets the benefit of the downpayments, and subject to that credit the buyer remains liable for the Unpaid Amounts: in such circumstances one would prima facie expect the buyer not Standard to remain liable for the market risk. What happens, however, is that a valuation is made on the termination date to allocate the market risk between the past and the future. In such circumstances that allocation needs to be made in order to protect Standard’s interests, not the buyer’s. Of course that allocation must be made honestly and rationally, but subject to that there is no reason whatever to think that Standard should be left with any risk of what the markets will do. Otherwise, in a manner quite contrary to the whole of the rest of the agreement, the full risk from the termination date forwards would bear unfairly and entirely on Standard, who is simply reacting in an emergency to its buyer’s default. Just as in other circumstances Standard is entitled to determine and exact, in its own protection, the downpayments, so, under the clause 14(a)(bb) option, where exceptionally the downpayments remain credited to the buyer, Standard continues to need the protection of the equivalent discretion. See para 23 above.

It follows that Gloster J erred in my judgment in construing the valuation sentence as requiring an objective inquiry into the true market value of the Designated Assets, or as imposing a duty of reasonable care upon Standard. It must also follow that if she found Standard’s evidence, as to how it would have carried out such a valuation, to have fallen short of such obligations, then she was judging that evidence by the wrong test and her conclusions could not survive on appeal. My attempt to analyse her conclusions in the section above, headed The judgment of Gloster J, suggests that it is not entirely easy to determine on what basis she rejected Standard’s evidence. However, it seems difficult to suppose that she went to so much care to determine the legal dispute between the parties as to the correct test to apply, unless she proposed to use her view as to that correct test in resolving the ultimate issues at trial. Moreover, it is reasonably plain that she judged Standard’s evidence against the background of Socimer’s experts whose evidence was directed to an ascertainment of the true market value on objective criteria. The only concession to that approach was that she was prepared to allow, perhaps somewhat inconsistently, a subjective element of conservatism and caution in reaching her final valuation results. At the same time, however, she appears to have also concluded that Messrs Clifford and Feld had either not been honest in their evidence or had spoken to a dishonest valuation. It is also possible that when she said that their valuations had been unreasonable, she was intending to say that they were irrational as well as, or rather than merely, careless. However, those conclusions, as to honesty and perhaps rationality, were themselves influenced by her adoption of the requirements of Socimer’s implied term. There must therefore be real uncertainty as to whether the judge’s findings as to honesty or rationality could retain any validity (even on the assumption that those findings could survive the absence of challenge to Standard’s witnesses) once the implied term has been shown to have been wrongly adopted. Certainly, Mr Millett did not submit that the judge’s judgment could survive this appeal, were it shown that she had been wrong to adopt Socimer’s implied term, on the ground that she had in any event rejected all Standard’s evidence because of bad faith or irrationality.

I therefore conclude that on this first issue Standard succeeds, and that its success must at least earn Standard a new trial in relation to the termination date valuation. I will consider further the ramifications of the question of bad faith and irrationality below.

The importance of this issue is that first, it is common ground that it arises even if Standard succeeds on the first issue that all the Designated Assets are to be valued as Standard says that it would have valued them; and secondly, that it is only if this issue is answered in Socimer’s favour that its Socma Primary Case arises. The judge found in favour of that primary case, in the sum of $8,424,966, which constitutes the major part of her judgment award in financial terms.

In question was the status of four credits owed by Standard to Socimer, totalling some $4 million (the “other credits”). The essential question was whether these other credits went in immediate and prior reduction of the Unpaid Amounts as of the termination date, so as to reduce the Unpaid Amounts from the otherwise agreed total of some $24.5 million to a revised net figure of only some $20.4 million. If so, then Socimer argued that, even on Standard’s valuations, a surplus became due to Socimer which ought to and would have resulted in Standard’s return to Socimer of the Socma DRs. The judge agreed, and found that in Socimer’s hands the Socma DRs were to be valued at the nominal face value figure of some $8.4 million.

Gloster J reasoned the matter as follows. She said that regardless of whether or not Standard was obliged to set off the other credits to Socimer, it was entitled to do so under clauses 8 and 9(c). She adverted to the fact that Standard had in fact set off the other credits in August/September 1998, as shown by Standard’s subsequent letters to the liquidator of Socimer about the state of the parties’ accounts. There is no dispute about that: those letters referred to various sales of Designated Assets: there was at all such material times a deficiency in respect of the Unpaid Amounts, even after such sales had been credited to Socimer, and the other credits applied by way of set off (ie the other credits not deriving from sales of Designated Assets) merely reduced the deficiency, but never eliminated it.

The most significant of the four other credits was called the “Spot Trade Balances”, and these were worth some $3 million of the total $4 million concerned. On this appeal, the parties were content to argue this issue solely by reference to these Balances. These Balances were not in fact due and owing to Socimer until 25 February 1998, five days after the termination date at which the Designated Assets had to be valued. They were not in fact credited to Socimer’s account with Standard until the relevant settlement day of 25 February, as the judge found. The judge nevertheless also found, as a matter of fact, that Standard would have credited Socimer by way of set off with the value of the Spot Trade Balances as of 20 February 1998, before they were actually due. She said (at para 48):

After dealing with further arguments no longer relevant for present purposes, the judge stated her conclusion in respect of the Unpaid Amounts in these terms:

Other matters with regard to these Balances were argued at trial. However, for present purposes this Issue 2 on this appeal was confined, at any rate for the present, to the short point whether such a finding was justified on the evidence on which Socimer continued to rely, namely Mr Feld’s evidence and the later correspondence with the liquidator. Socimer no longer contended that Standard was obliged to set off these other credits. It simply supported the judge’s decision on the basis that she was right to find that Standard would have set off the Spot Trade Balances as at the termination date, even before they were due. One additional problem about the terms of para 57 however, is that, somewhat inconsistently with what she had said earlier, she appears to have concluded that the set-offs depended in one form or another on Standard’s obligations. That, as I have said, is no longer maintained by Socimer.

Indeed, Socimer’s argument before the judge and on appeal involved the additional factor that Standard would have set off the other credits as at the termination date even before conducting its valuation of the Designated Assets. This was a necessary part of the argument: because, as I have observed, the Socma Primary Case depended on the premise that the Unpaid Amounts due were not in total some $24.5 million, which was prima facie common ground, but that figure less the other credits following their set off, namely a “net” figure of only some $20.4 million. It was only if the relevant Unpaid Amounts totalled that lower figure that Socimer was in a position to contend that Standard was sufficiently in surplus after the retention of all assets other than the Socma DRs to require their transfer to Socimer.

The judge nowhere dealt in terms with this additional point, possibly, but I am not sure, because it was common ground. What she did deal with was Socimer’s argument that, on the premise of the lower net figure for the Unpaid Amounts, it would have been the Socma DRs rather than any one or more of the other Designated Assets which would have been transferred to Socimer as surplus to Standard’s retention requirements. It has not been necessary so far to hear this aspect of Standard’s appeal. In brief, however, the judge found (at para 72) that had Standard appreciated in full its clause 14(a)(bb) obligations it would have returned the Socma DRs to Socimer: because it was otherwise fully recouped for what it was owed, because the Socma DRs were the most difficult to value and therefore the asset most likely to give rise to dispute, because the value of the Socma DRs was peculiarly subject to the Socimer insolvency risk (owing to the prospect that Socimer was not merely the fiduciary of the asset but the debtor in respect of it), and because of its own purported valuation of the asset at nil (in fact $500,000). The judge appears merely to have assumed that in this context the Unpaid Amounts would have been reduced by the set-off of the other credits before Standard would have undertaken its valuation exercise.

On this appeal, Mr Millett has abandoned, if it was relevant at trial, any contention that Standard was obliged to set off the Spot Trade Balances. What he continues to say is that Standard would have set off these Balances (and the three other credits) before turning to its valuation exercise. He submits that the judge’s findings on this point are in Socimer’s favour and irrevocably control this issue. On Standard’s side, Mr Auld submits that there was nothing in the evidence to lead to the judge’s conclusion, and that she misunderstood what Mr Feld said. He therefore submits that, on the basis that the Balances were not due to Socimer until 25 February 1998, the judge’s findings on this issue were impossible.

This is a short point. The basic premise, as acknowledged by the judge herself, is that the Balances were not in fact due to Socimer until the relevant settlement day of 25 February. They were in fact credited to Socimer’s account on that day, not before, and that was correct accounting. That was of course in the absence of any clause 14(a)(bb) valuation. Why should the hypothetical exercise have made any difference? To the extent that the judge relied on the correspondence in August/September 1998, there was nothing to support her conclusion there. Socimer was always in a deficiency and Standard was entitled to set off Socimer’s credits against its greater liabilities. There was nothing whatsoever in that conduct to suggest that Standard could or would have set off the Balances against the Unpaid Amounts if it had been carrying out the valuation sentence exercise on and as at 20 February 1998.

What was critical to the judge’s thinking, however, was Mr Feld’s evidence at trial. The relevant passage of evidence was this (at Day 4.13/16):

What is being said here? The settlement day of the Spot Trade Balances is 25 February. The valuation date is 20 February. This is a five day period (referred to by the judge at one point as the “three-day period”). Mr Feld is saying that if, after the valuation on 20 February, a deficit had been left, then Standard would have applied the Spot Trade Balances credit to that deficit, ie would have exercised its right of set-off. The judge then asked Mr Feld about how the five day time value of money would have been dealt with, and was told “We would not have discounted for the time value, I do not think.” In other words, a credit to Socimer of (say) $100 as of 25 February is not worth as much as a credit to Socimer of $100 as of 20 February. So, strictly speaking, the credit as of 25 February should have been discounted by the time value of 5 days, alternatively 5 days default interest would be added to the deficit between 20 and 25 February. But Mr Feld said that he thought that the credit, being only five days later than the established deficit, would have been squared off dollar for dollar, without discounting the later credit: ie Socimer would have been given an allowance for a five day gap before the credit became due, as if the credit had become due and had been set off on 20 February rather than on 25 February. There is nothing whatsoever in that evidence to suggest that Standard would have credited Socimer with the value of the Spot Trade Balances before the valuation exercise. On the contrary, Mr Feld’s evidence is that that would have happened after the valuation exercise, if a deficit had been left. Of course, the set-off could not have occurred the other way around, because the credit did not occur until five days later.

I fear the judge is simply mistaken about this. In these circumstances, if Standard’s valuations had left Socimer with a deficit of (say) $4 million, then the other credits could have been set off so as to reach balance. If the deficit had been (say) $1 million, then Standard could have set off $1 million, and would have had to return $3 million to Socimer. What Standard could not do, even if it had wanted to, was to say: We would much rather keep the $3 million cash, rather than $3 million of Designated Assets. For the balance on the valuation exercise would have already been struck.

In my judgment, therefore, the judge should have determined this issue in Standard’s favour. Socimer’s Socma Primary Case could not arise.

In the light of my conclusion on issue 1, it does not matter whether Socimer’s implied term case was open to it or not, and therefore I will deal with this issue relatively briefly.

In my judgment, the implied term case was not open to Socimer. I do not say that the matter was not open because of res judicata (as a result of the judgment of Cooke J) or abuse of process, for Mr Auld said that he did not rely on such doctrines, and I see that in phased litigation it may sometimes be difficult to steer an entirely clear line in these respects. However, Standard’s evidence was prepared on the basis that the valuation exercise which Cooke J’s judgment had identified was one in which the determination of value was for itself and was one in which it had a wide discretion. However, the argument it faced during the hearing before Gloster J was that the valuation exercise in question was a different one, being ultimately for the court to decide based on objective criteria. That was unfair, and was likely to lead to error. If Standard had known that its valuation was to be attacked on the basis of lack of reasonable care or failure to achieve the true market value, then its evidence would have been likely to have been more complex than it was. As it was, Standard suffered all the difficulties of having prepared its evidence for what, by the time of judgment, had become a mistaken test, and of not having prepared its evidence for the objective test which the judge applied. In such circumstances, it may not be surprising that Standard’s witnesses were found wanting.

Finally, in my judgment the difficulties caused by the late introduction of the implied term test were exacerbated, as it turned out, by Socimer’s tactical decision not to cross-examine Messrs Clifford and Feld. I say that it was a tactical decision in the light of the material set out above, which shows that Socimer wished to avoid that evidence being admitted, had then recognised, when the judge decided that it should be heard de bene esse, that it would have to be cross-examined, and had finally chosen not to challenge it by cross-examination. In the light of the valuation exercise as interpreted by Cooke J, Standard’s evidence could have been challenged on the basis that (a) it was not credible (for whatever reason), or (b) was dishonestly false, or (c) while in itself credible, was premised on a bad faith approach to valuation, or (d) was premised on an irrational (or arbitrary, capricious or perverse) approach to valuation. On the basis of the implied term finally adopted by the judge, that evidence might have been challenged on the further bases that, however credible and honest the evidence in itself and however honest the witnesses’ approach to valuation, nevertheless (e) their evidence was irrelevant because it spoke to the wrong test, and/or (f) flawed because it demonstrated an objectively careless approach to valuation, and/or (g) flawed, even in the absence of carelessness, simply because it failed to identify by objective criteria the true market value of the Designated Assets. However, when Mr Millett decided not to cross-examine Messrs Clifford and Feld at all on their evidence as to the valuation which they would have conducted, he put himself in the position, in my judgment, where he was unable, as a matter of simple fairness, to make any of those submissions, save for (e), a matter essentially of law (albeit one on which it is not impossible that further matrix evidence might have been available to Standard, just as matrix evidence had been deployed before Cooke J), and, perhaps, (g).

As it was, Mr Millett, by an exercise of shrewd advocacy, was able, even in the absence of any challenge, to address the judge on the basis that Messrs Clifford’s and Feld’s evidence should be disregarded or discounted on the ground that it was incredible, self-serving and partisan: and in essence he succeeded in those submissions so far as the valuation exercise was concerned, as I have sought to show above. And yet, before this court, Mr Millett expressly accepted that he had no case of bad faith to make against Standard’s witnesses. In these circumstances, his submission to us that the judge was simply entitled to make up her own mind as to the value of Standard’s evidence did not meet the point. Of course the judge was entitled to make up her own mind about the evidence before her: but she was not entitled, at any rate not without the most explicit warnings and possible need for an adjournment, to find that evidence which had not been challenged by any cross-examination was false evidence, or evidence which spoke to a valuation exercise which was flawed by bad faith, irrationality or negligence. It is simply unfair for Standard’s witnesses to be found wanting in such respects, unless the relevant challenges had been put to them and they had been afforded the opportunity (and I mean a proper opportunity, for which the ground had been sufficiently prepared in the pre-trial process) to answer the accusations made against them.

This is not one of those cases where a minor issue arises as to whether some detailed point had been sufficiently covered in cross-examination. As has often been said, it is not necessary to cover every point. This case, however, is remarkable. There was, quite deliberately, no cross-examination at all of Messrs Clifford and Feld on their relevant evidence, but the forensic submission was nevertheless made in effect that such evidence should be totally rejected on the ground that “They would say that, wouldn’t they?”

Of course, the judge would have been entitled, as she was invited, subject to issue 3(1) above, to say that the Standard evidence was simply beside the point. However, she did not so rule, and if she had, she would, in my respectful judgment, have been in error.

In conclusion, in my judgment Standard has succeeded on all the issues which have been so far argued before us. It therefore seems to me that the judge’s decision cannot stand. However, quite what the full consequences are is a matter on which we told counsel at the end of the hearing that we would wish to hear further submissions.

I agree, and wish to add only some comments on the analogy which the judge drew with the duties incumbent on mortgagees.

If parties enter into a transaction which is a mortgage, then the law imposes certain obligations on the mortgagee, and confers certain rights on the mortgagor, which go back to the intervention of equity in the early development of mortgages. Although a mortgage is a contractual transaction, the imposition of such duties has nothing to do with the implication of terms in a contract under the general law of contracts: see Yorkshire Bank v Hall[1999] 1 W.L.R. 1713, at 1728D. Whether these duties are imposed on a given party depends only on whether, on the true analysis of the transaction, it is or is not a mortgage.

Other consequences may follow if the transaction is a mortgage, for example obligations to register the security if it has been created by a company, failing which it may be void against a liquidator or other creditors.

It is therefore important to draw a clear distinction between a transaction which is a mortgage, on the one hand, and one which, however similar it may be to a mortgage in economic or commercial effect, is not a mortgage as a matter of true legal analysis.

In some circumstances it may not be easy to determine whether an agreement does or does not create a mortgage: see, for example, Welsh Development Agency v Export Finance Co[1992] BCC 270. In the present case, however, it was common ground, and the judge rightly accepted, that the agreement did not involve lending, nor security for a debt, and therefore it did not amount to, or involve the creation of, a mortgage.

At paragraph 42 of her judgment, the judge summarised the duties imposed on a mortgagee in respect of the exercise of powers to realise the security. At paragraph 43 she said that

Mr Auld had submitted that such duties did not apply because they arose only from the relationship of mortgagor and mortgagee. She rejected that argument, and said at paragraph 43 that, given the rights of Standard, there was “every reason for the implication of a duty closely analogous to that of a mortgagee”. At paragraph 45 she said that

That passage suggests that she was proceeding on the basis of the implication of a term into the contract, but by analogy with the terms imposed by law in relation to a different type of transaction, to which this agreement had economic similarities. It seems to me, with respect to her, that she was led by that similarity into drawing, and applying, an analogy with mortgage law, while overlooking, on the one hand, the need to justify the implication on the basis of conventional contract law and, on the other hand, the fact that, in relation to a mortgage, the duties by reference to which she drew the analogy do not derive, and cannot be derived, from such a process of implication, but are imposed as a matter of general law, which does not apply in the present case because the transaction is not a mortgage.

It seems to me that the duties to which a mortgagee is subject are no guide at all on the question whether it is legitimate to imply into the contract a term under which Standard would be subject to such a duty such as the judge found.

I therefore disagree with the judge’s proposition in her paragraph 43 that “similar principles must apply, with the necessary adjustments, to the valuation powers of … a person in an analogous position to a mortgagee, such as Standard, in circumstances such as the present”, and with her conclusion on the point expressed at her paragraph 45.

Like Rix LJ, I consider that the requirements of necessity and otherwise which are laid down by the law of contract for the implication of terms cannot be satisfied in the present case in respect of the term which the judge held ought to be implied into the contract. For those reasons, and for all the reasons given by Rix LJ, I would allow this appeal.

I agree with both judgments.