Pearson & Ors (The Joint Administrators of Lean Brothers International (Europe)) v Lehman Brothers Finance SA & Ors

I must now deal with the costs of this application. I do so in a reserved judgment because the submissions which have been made give rise to a question of potentially wide application upon which there is little direct authority.

This application was, in form, an application by Administrators for directions. In paragraph 25 of my main judgment I describe the trial of the application as having taken the form of a “straight adversarial fight” between LBIE and five of its affiliates, in relation to the beneficial ownership of certain classes of securities (“Rascalled securities”) held by LBIE in house depot accounts for the books of those affiliates.

There is a sharp divergence between counsel as to how these proceedings should be categorised for the purpose of the exercise of the court’s jurisdiction as to costs. For the Administrators, Mr Milligan QC described it as, in substance, commercial litigation between rival claimants to beneficial ownership of specific property (i.e. the Rascalled securities), in relation to which costs should follow the event, as in any other commercial litigation. Counsel for the affiliates all described it as being, in substance, a joint application by the various office-holders of insolvent companies within a common group to the appropriate forum for the resolution of difficult issues affecting them all, and standing in the way of the realisation and disposal of assets to creditors, or to other persons entitled to them. It was adversarial litigation only in the sense that, for the assistance of the court, the alternative solutions to the complex common problems were advanced by counsel, including counsel for the Administrators, on an adversarial basis. The result in terms of costs was, it was submitted, either that all the parties’ costs should be borne by LBIE, or that each insolvent estate should bear its own costs. Reference was made to established Chancery authority, including Mitchell v. Gard (1863) 3 SW & TR 275 and Spiers v. English [1907] P 122, and ending with Parmalat Capital Finance Limited & ors v. Food Holdings Limited (In Liquidation) & ors [2008] UKPC 23.

There is considerable force in both those competing analyses, and it is neither necessary nor appropriate to make a stark choice between them. In favour of what I will call the commercial litigation analysis are the following factors. First, the casus belli for the application was, in a sense, the fact that LBIE invited each respondent affiliate to state whether it wished to assert a claim to beneficial ownership of the Rascalled securities, and each respondent replied that it did. Secondly, the dispute lies entirely between LBIE and a specific affiliate (or in the case of LBCCA and LBAH affiliates) in relation to each class of Rascalled securities, there being no wider class of potentially interested parties requiring representation. Thirdly, although each office-holder is, quoad the claim of his affiliate, acting in substance as a fiduciary for underlying stakeholders, there is no doubt that the office-holder is dominus litis, so as to be able to pursue or compromise claims on behalf of his insolvent estate, even though from time to time he may for that purpose have to obtain approval from creditors’ committees, or from the relevant foreign court having jurisdiction in relation to the insolvency of that affiliate (here the courts of Switzerland, Hong Kong or New York).

In favour of what I shall call the ‘joint application for directions’ analysis there are the following factors. First, the problems which led to this application arose from the carrying on by LBIE and its affiliates as associated companies of a business pursued on a Group basis, applying a common global settlement practice, Rascals processes decided on by the Group, using common form contractual documentation and a unified Group accounting system. Secondly, although the relevant insolvencies are primarily regulated by the courts of four different countries, all the foreign insolvency processes have been recognised here so that, pursuant to the Cross-Border Insolvency Regulations, all the relevant office-holders can seek the assistance of this court, as appropriate. Thirdly, as responsible office-holders, all the individuals responsible for the conduct of each affiliate’s part in the litigation may fairly be thought of as fiduciaries, in analogous positions to trustees seeking a court’s directions in relation to the administration of a trust.

A fourth factor relied upon by the affiliates, by way of analogy with the determination of issues arising in connection with a deceased’s estate, is that all the parties to this application have, to varying degrees, had their tasks made more difficult by limited access to historical documentation, and their perception of their respective cases, and those of their opponents, has therefore developed and changed over time, both before and indeed during the hearing of this application, by contrast with what it is suggested is typical of ordinary commercial litigation, where all the parties know where they stand on day one, and are therefore sufficiently informed to be able to decide whether to fight or settle. While it is true that difficulties of that type did lead to numerous changes and developments in the parties’ respective positions both before and during the hearing, I am not persuaded that this feature is unique to litigation about deceased or insolvent estates. It is by no means uncommon for ordinary commercial litigation to involve similar features, where the processes of disclosure and the search for evidence, not to mention developments which occur at trial, for example during cross-examination, give rise to a constant need for the parties to re-appraise the strengths and weaknesses of their respective cases, and to alter them by amendment.

The costs consequences of those competing analyses may be summarised as follows. First, CPR Part 44.3(2) prescribes a general rule that the unsuccessful party will be ordered to pay the costs of the successful party, which is fully applicable to commercial litigation. But even there, the ‘costs follow the event’ principle is itself only the starting point, and the court may make a different order, having regard to all relevant circumstances.

Secondly, that general rule is not in terms disapplied merely because the litigation concerns a deceased’s or insolvent’s estate. Part 44.3(3) specifically disapplies the general rule to two other types of proceedings, but not to proceedings about estates. Nonetheless, certain common features about proceedings in relation to estates have, since at least the mid-19^th century, been taken to justify a departure from the general rule. In relation to deceased’s estates, the following two features may (if present) justify a ready departure from the general rule. The first is where the issue for determination arises due to the fault of the deceased, for example where as testator he makes a confusing will which calls for construction by the court, or where because of issues about the validity of a will, the court’s task in deciding whether or not to grant probate calls for the assistance of the parties to contentious proceedings, in effect as contributors to a necessary judicial inquiry: see Mitchell v. Gard (supra), at 277-278, and Spiers v English (supra) at 123.

That these principles have survived the CPR, and been implemented after their introduction, is apparent from Kostic v. Chaplin & ors [2007] EWHC 2909 (Ch). But Henderson J noted, at paragraph 21, that there has been a change of emphasis in the extent to which those principles may justify a departure from the general rule. He gave two reasons for that change:

Kostic v. Chaplin was a contested probate case in which the testator’s extreme eccentricity had given rise to real uncertainty as to his testamentary capacity which, after full investigation, led to the conclusion that the relevant wills were invalid. Henderson J concluded that, for an initial period in the litigation, the testator had been sufficiently the author of the estate’s misfortune to justify an order for costs of the unsuccessful party out of the estate. Secondly, for a further period in the litigation (when the competing cases as to the testator’s capacity were known, but not the subject of expert evidence) he considered that no order as to costs should be made. For the final period, after the battle lines had been drawn between the experts in written reports, he ordered that costs should follow the event.

The line of authority culminating in Kostic v. Chaplin is of course applicable, if at all, only by analogy to litigation about insolvent estates. Nonetheless, in the following respects, the analogy is in my view a real one. First, it is by no means uncommon that responsibility for issues arising in the administration of an insolvent estate may fairly be laid at the door of the insolvent entity (be it individual or corporate), for example where the insolvent has kept inadequate records, or sought to interfere by preference or worse with the proper distribution of its assets to its creditors. Secondly, there is an undoubted public interest in the due administration of the assets of an insolvent’s estate in accordance with the statutory insolvency code, and parties who are joined in proceedings made necessary for that purpose should not be unduly discouraged by an unthinking recourse to the general rule where, in the end, the issue is decided against them.

As Henderson J recognised, one of the beneficial consequences of the general rule is that, by exposing unsuccessful litigants to liability for their opponent’s costs, it concentrates the litigants’ minds upon the need constantly to address the strength or otherwise of the case, the risks and benefits of advancing particular arguments, and the wisdom of searching for alternative forms of resolution of their dispute, whether by compromise or even abandonment. There is in my judgment no reason why those objectives should be any less desirable in insolvency litigation than in other forms of adversarial litigation. On the contrary, as is spelt out in Jackson LJ’s recent Report on Costs in Civil Litigation, there is an ever present risk that the costs of insolvency litigation may easily get seriously out of hand. This must be a fortiori the case if litigants are encouraged to think that they could run weak arguments at great expense to their opponents, on the basis either that costs would come out of the insolvent estate, or that each party’s costs risk would be limited to its own costs.

It follows in my judgment that although there are features of insolvency litigation which, by analogy with litigation about deceased’s estates, may justify a departure from the general rule, the court should nonetheless approach any particular case for a departure with real caution, and litigants ought to expect to have to justify such a departure by reference to the facts about their alleged predicament, rather than merely by recourse to some supposed general principle.

With those general considerations in mind, I turn to the particular facts about this litigation, so far as it affects its various parties. It is at the outset necessary to draw a line between the issues as between LBIE and LBI on the one hand, and the issues as between LBIE and the other respondent affiliates on the other hand. As its name implies, this litigation has been essentially concerned with the application of Rascals processes to securities acquired by LBIE for the account of affiliates. On analysis, there was no general application of Rascals processes as between LBIE and LBI, but merely a small number of securities affected by open stock loans which were probably not put in place by way of the application of any general Rascals policy. Furthermore, as will appear, the value of the very small number of securities to which those stock loans applied was minute as between LBIE and LBI. The contractual relationship between LBIE and LBI was governed by an agreement, the UCCBSA, regulated by New York law, to which substantial expert evidence was devoted. By contrast, securities of very substantial value were affected by routine Rascals processes as between LBIE and the other respondents, and they dealt with each other upon a broadly similar contractual and accounting basis.

Mindful of the requirements of CPR Part 44.3(7), I invited the parties at an early stage, and they agreed, to approach the matter first by a rough and ready apportionment of LBIE’s costs, as between LBI on the one hand and the other respondents on the other. After short submissions, I determined that the apportionment should be on the basis of 25% to LBI and 75% to the others. I shall therefore put LBI’s position on one side, and deal first with the incidence of costs as between LBIE and the other respondents.

At the risk of oversimplification, the course of this litigation as between those parties may be summarised as follows. LBIE began by asserting that the Rascals processes did what it said on the tin, namely that in accordance with what I have concluded was the parties’ obvious intention, they conferred absolute title to the underlying securities upon LBIE, to the exclusion of the relevant affiliates, and that title remained with LBIE both upon and following the collapse. These respondents challenged that case by technical arguments as to the ineffectiveness of the Rascals processes. LBIE responded, without abandoning its case that Rascals worked, by an ever increasing reliance upon a logically anterior case that the respondents never acquired proprietary interests in the underlying securities in the first place, so that the Rascals processes were, even if ineffective, wholly unnecessary.

In the event (subject to appeal) LBIE succeeded on its original case. It also demonstrated that, until 1995, it had in fact been unnecessary to implement the Rascals processes because, absent any decision that a particular type of security should be Rascalled, the Group’s global settlement practice was not such as to confer a proprietary interest on affiliates for which LBIE acquired and held securities in its house depot account. Nonetheless LBIE’s anterior case failed, in the sense that, once the Rascals processes were embarked upon in relation to eligible securities, those processes necessitated the conferral upon affiliates of some proprietary interest, even if only for a nanosecond, so as to make sense of the Rascals repos and stock loans thereafter applied.

In the event therefore, the long and hard fought arguments about LBIE’s anterior case proved to be unnecessary, because of the success of its original case that the Rascals processes worked. It is therefore a case in which, within the meaning of CPR 44.3(4)(b), LBIE was wholly successful in defending its claim to beneficial ownership of the Rascalled securities but, within the meaning of Part 44.3(5)(b), LBIE was by no means successful in relation to a main issue, namely the anterior issue which I have just described. As will be apparent from my main judgment, a broad brush analysis of the time and effort devoted to the Rascals issue and the anterior issue would suggest that at least half of it, and possibly a little more, was applied to the anterior issue.

There was a debate between counsel as to whether LBIE really lost the anterior issue, because of my conclusion that, but for the decision to implement Rascals processes, the global settlement practice conferred no proprietary interest on the affiliates. Nonetheless I consider that LBIE did lose the anterior issue, because of my conclusion, at paragraph 307 of the main judgment, that the mutual intention to confer a proprietary interest on affiliates (which was a prerequisite of the Rascalling of eligible securities) was not conditional upon the effectiveness of the Rascals processes. The result is that, even if the Rascals processes had for any reason been ineffective, the affiliates would nonetheless have acquired proprietary interests in the supposedly Rascalled securities.

Counsel for the respondents sought to persuade me that this was a case in which LBIE rather than the respondents was ultimately to blame for the mess which (by common consent) led to this litigation, suggesting that the Rascals processes were designed and applied by LBIE for LBIE’s purposes using LBIE’s accounting systems. Furthermore the respondents sought to rely upon difficulties of their own in obtaining access to the relevant records after the collapse, which exceeded those encountered by LBIE itself, as the basis for a submission that the respondents’ inability to focus upon their prospects of success in this litigation at any earlier date than the trial was largely the result of LBIE’s repeated changes of position, while in better command of the underlying materials than were the respondents.

I have not been persuaded by either of those submissions. Both the global settlement practice and the Rascals processes were implemented pursuant to Group policy, and even if their immediate effect was to relieve LBIE of capital adequacy and regulatory difficulties, those only arose because of aspects of the Group’s global settlement practice (such as not demanding payment in cash from affiliates) which were ultimately as much for the benefit of the affiliates as they were for LBIE. Further, LBIE’s apparently greater access to the underlying documents appears to have been the result of its readiness to pay the very large licence fee demanded for access to ITS after the collapse, by comparison with a reluctance to do so on the part of one or more, or all, of the respondents. For its part, I was told (and this was not challenged) that LBIE went to considerable effort to explain to representatives of the respondents its understanding of the Rascals process and of the Group settlement practice both before, and after, the commencement of this litigation.

More generally, I do not consider it appropriate to address costs issues upon the basis of any perception that the conduct of any of the parties was in any sense unreasonable, or such as to merit the expression of any disapprobation by the court in terms of an aspect of the costs order. In my judgment there is no basis for a conclusion that any of these parties acted unreasonably, either in the pursuit of arguments which turned out to be unsuccessful, or in the timing of the presentation of evidence, or of changes in their respective cases. This complex litigation was prepared and pursued pursuant to a relatively tight timetable in circumstances in which all the parties, to varying degrees, faced formidable obstacles both in the formulation of their own cases, and the evaluation of the merits of the dispute.

This is, therefore, not a case in which any part of these respondents’ costs should be paid out of the LBIE estate, by analogy with any perception that the mess was uniquely LBIE’s fault. Furthermore, although LBIE failed on an important and time consuming argument, and while that failure might justify disallowing part of its costs, the argument was not unreasonably advanced, and its failure did not detract from LBIE’s overall success in achieving its commercial objective, sufficient to afford any basis for an order that part of the respondents’ costs be paid by LBIE.

Nor am I persuaded that the analogy with probate litigation warrants an order that each party be left to pay the whole or even the bulk of its own costs. This was, in my judgment, litigation in which these respondents unsuccessfully advanced an adversarial case in the pursuit of a very large commercial objective, namely the obtaining of a proprietary interest in securities of enormous value. It was in truth litigation in which the expenditure in costs was by no means disproportionate to the value of the property in dispute, but that is not of itself any reason for displacing the general rule that costs should follow the event.

Furthermore, it is in my judgment nothing to the point that each respondent pursued its case through office holders acting pursuant to fiduciary duties, for the benefit of stakeholders. That is no more a reason for disapplying the general rule, than the assertion by a trustee that he is pursuing litigation for his beneficiaries. There is as I have said some force in the depiction of this application as a sort of Lehman Group insolvency application for directions, but it carries no decisive weight with me on the question of costs. There are numerous different ways in which this litigation can be portrayed. In the end, the governing consideration to my mind is that it was commercial litigation in which LBIE won, and these respondents lost.

I therefore start with a propensity to apply the general rule in LBIE’s favour, and therefore to order these respondents to pay that 75% of LBIE’s costs attributable to the litigation with them, subject to matters justifying a disallowance. In my judgment, two matters do justify some disallowance. The first is LBIE’s failure on its anterior case. Although a strict reading of CPR44.3(4) and (5) might make it appear that a party which has been wholly successful, but has failed on a particular issue should only suffer a disallowance where it was unreasonable for it to pursue that issue, I am persuaded by reference to Shirley v. Caswell [2000] Lloyd’s Rep PN 955 that the court’s general discretion as to costs is not that closely circumscribed. That was a case in which the Court of Appeal upheld a decision of Ferris J to disallow part of the successful claimants’ costs by reference to their failure on issues, which included issues which the judge had not concluded were unreasonably pursued. Mr Milligan did not suggest that Shirley v. Caswell should be understood in any different way.

Secondly, I am persuaded, by reference to the analogy with the Kostic v. Chaplin line of authority, that a significant part of LBIE’s costs was inevitably incurred in the research and preparation of a case necessitating the obtaining of the court’s directions during a period before which these respondents could reasonably be expected to form a view as to the merits of their case, to an extent beyond that inherent in ordinary commercial litigation. The natural outcome of that factor would ordinarily be to order these respondents only to bear LBIE’s costs incurred after a certain date, as indeed was ordered in Kostic v. Chaplin itself. Nonetheless there was no examination before me of the precise chronology, and none of the parties’ counsel advanced a specific date-based submission. Rather, all these respondents appeared content that I should reflect any such matter in a partial or total (but in any event proportionate) disallowance of LBIE’s costs. That is what I shall therefore do.

Taking those matters together, I propose to direct that these respondents bear 50% rather than the 75% of LBIE’s costs. That reflects a disallowance of one third of the amount attributable to the dispute as between LBIE and these respondents, as the combined result of the two particular factors which I have identified. For completeness, I direct that the costs judge should not on assessment make any further disallowance in relation to LBIE’s costs incurred in pursuing its anterior case. The costs liability of each respondent (treating LBCCA and LBAH as one for this purpose) is to be several only. In other words each respondent is to pay 16.66% of LBIE’s costs. This is because there was a separate body of securities at stake as between LBIE and each respondent, even though the issues largely overlapped.

I return to the incidence of costs as between LBIE and LBI. As I have said, this was not really part of the Rascals case at all, but rather, at least strictly, only about a very small number of securities (originally fourteen) in respect of which it appears to have been common ground that, if that was the only matter in dispute between the parties, the value of those securities would not justify a trial. In the event, LBIE abandoned its case in relation to one of them, won its case in relation to three and lost its case in relation to the other ten. By won and lost, I mean that LBIE succeeded and lost in relation to fact patterns apparently exemplified by those two groups of securities. It was not the court’s task to address issues in relation to individual securities.

Both sides made offers without prejudice save as to costs (but not Part 36 offers) to compromise the dispute, including an offer by LBIE to abandon its claims in relation to all fourteen securities in issue, and to pay LBI’s costs of the application, on 17^th September 2010. It is plain from LBI’s reply dated 27^th September 2010 that it would have accepted that offer, but only without prejudice to its continued participation in relation to what I have described (as between LBIE and the other respondents) as LBIE’s anterior case. On this rock all attempts at compromise foundered.

It was plain both from that correspondence and from Mr Brindle’s opening at trial that LBI’s real reason for wishing to participate in the Rascals litigation, while at the same time denying any real Rascals processing as between itself and LBIE, had nothing whatsoever to do with the fourteen securities subject to open stock loans, and everything to do with a perception that the outcome of LBIE’s anterior case (namely that LBI never acquired a proprietary interest in the securities in the first place) might be decisive of, or at least influential upon the outcome of a much larger dispute about non-Rascalled securities held by each for the other’s account, valued in several billions of dollars. LBIE’s response, both in correspondence and during the trial was that this application was only about Rascalled securities and that, because there had been no similar analysis of fact patterns in relation to the parties’ dealings with non-Rascalled securities, the anterior case in relation to Rascalled securities was not to stand as a decision upon issues relating to non-Rascalled securities.

In the event, as will be apparent from the main judgment, LBIE’s approach to that matter prevailed with me. I have avoided reaching any decision on LBIE’s anterior case decisive of anything other than Rascalled securities. I should add that LBI persisted in its approach notwithstanding a provisional indication from the court at the beginning of the trial that the only property in issue between LBIE and LBI was of such (relatively) modest value that a compromise, even at that late stage, might still be worth pursuing.

I can well understand the apprehension on the part of the office holder of LBI that, at a trial at which his representatives were simply absent, the court might in addressing LBIE’s anterior case as against the other respondents, stray into areas and conclusions which might have an impact upon the large dispute between LBI and LBIE in relation to non-Rascalled securities. I do not therefore conclude that LBI’s continued participation in this litigation after LBIE’s offer of 17^th September was so unreasonable as to merit special censure in costs, such as an order for indemnity costs. Nonetheless, LBI declined an offer from LBIE which, had it been accepted, would have afforded to LBI more than it in fact obtained by persisting in the litigation, since it succeeded in relation only to ten of the thirteen securities which remained in issue at the end of trial, LBIE having abandoned the fourteenth in the meantime.

One aspect of the matter as between LBIE and LBI deserves special mention. That is the introduction by both parties of substantial expert evidence about New York law. Pursuant to a direction given by the court that any party wishing to do so should identify in writing any relevant aspect of foreign law, LBI identified New York law, and advanced a case both as to the New York law interpretation of the UCCBSA, and as to the effect upon the parties’ legal relationship of the New York Uniform Commercial Code. That case was supported by expert evidence which included an inadmissible attempt to construe what turned out to be an irrelevant part of the UCCBSA. As for the New York UCC, after evidence as to its effect from LBIE’s expert, the parties eventually concurred in abandoning any reference to it. In the end, the expert evidence played a negligible part at the trial, and in the court’s deliberations. It was, from start to finish, directed to LBIE’s anterior case, rather than to the effect of the stock loans upon the ownership of the fourteen disputed securities.

In all those circumstances, I consider that LBI ought to pay all that 25% of LBIE’s total costs attributable to the issues between these two parties. While this may be an exceptional case in which, in the face of an offer of more than that to which it was entitled, the offeree’s decision to fight on may be seen as understandable in very unusual circumstances, that does not in my judgment justify a departure from the well established general rule that a party which obtains less at trial than it was offered before trial should nonetheless pay the offering party’s costs incurred thereafter. In my judgment LBI should pay 25% of LBIE’s costs incurred after 27^th September 2010. Ten days was in my view a sufficient time for acceptance. There should however be no order as to LBIE’s or LBI’s costs incurred prior to that date.

My reason for the latter conclusion is that, until then, LBIE had not made an offer in relation to the ten securities upon which LBI succeeded. While in an ordinary case that might have justified an order in favour of LBI in relation to its costs incurred until then, in the present case that consideration is neutralised by the reality that LBI’s participation, from start to finish, was to give it a platform in relation to its underlying concerns about non-Rascalled securities, which were not, at any time, the subject matter of this application. In that respect, LBIE had made its position plain in correspondence in May 2010, along the lines ultimately adopted in the main judgment.