PJSC NATIONAL BANK TRUST & Anor v BORIS MINTS & Ors

There is no dispute between the parties that an issue estoppel can arise between the parties to an arbitration award, which can then be deployed if there is an attempt to relitigate that issue between those parties in court proceedings. That has been clear since at least the decision of the Court of Appeal in Fidelitas Shipping Co Ltd v V/O Exportchleb [1966] 1 QB 630, and possibly since Doe on the demise of Davy v Haddon (1783) 3 Doug KB 310. However, as a matter of English law, issue estoppels which arise from court judgments bind not only the parties, but also their “privies”. There is a significant body of English case law which has developed since the decision in Gleeson v Wippell & Co Ltd [1977] 1 WLR 510, 514-515 which proceeds on the basis that the concept of privy extends not only to the parties’ successors in title in respect of the litigated right which has given rise to the issue estoppel, but also a wider class of persons where “having due regard to the subject matter of the dispute, there [is] a sufficient degree of identification between the two to make it just to hold that the decision to which one was a party should be binding in proceedings to which the other is party”. I will refer to that wider class of potential privies beyond successors in title as Gleeson privies.

As I explain at section C2 below, Mr Rabinowitz QC and his team argue that, even when considering the position of court judgments, there is in fact no class of Gleeson privies, that Gleeson (to the extent that it did formulate such a category of privies) is wrong and that none of the authorities which have proceeded on the basis of the Gleeson test are binding on me. Logically, that ought to be my starting point, but (as Mr Rabinowitz QC accepted) that is a very big topic with very significant implications. For that reason, I intend to begin with Mr Rabinowitz QC’s entrée rather than his plat principal. That is the argument that, whatever the position may be for judgments, arbitration awards cannot bind Gleeson privies, with privies to arbitrating parties being limited to the successors in title to the parties or the relevant rights.

I should immediately acknowledge, at the abstract level, the attractions of an argument which would bring the doctrine of privity for the purposes of issue estoppel in arbitrations into line with the doctrine of contractual privity, in what is ultimately a consensual process of dispute resolution. However, it is a conclusion which might well have significant consequences, and its merits require testing. It is an issue which has received limited attention in judicial decisions to date. Among the 150 authorities cited for the hearing, I was referred to only three decisions on the subject:

In addition, while writing this judgment, I became aware of the decision of the Court of Appeal in Vale SA & ors v Steinmetz & ors [2021] EWCA Civ 1087, in which Males LJ also assumed, at [31], that the privy extension to issue estoppel could apply to an arbitration award:

In the absence of any clear authority on the issue, it is necessary to approach it from first principles.

Mr Rabinowitz QC’s first argument is that the category of persons bound by an arbitration award falling within Part 1 of the Arbitration Act 1996 is defined (and delineated) by s.58(1), which provides:

S.82(2) of the Act defines “a party to an arbitration agreement” (rather than to the award) as including “any person claiming under or through a party to the agreement” (the reason for reversing the “under or through” wording in s.58 not being clear), which definition is carried through into, for example, s.9 dealing with the stay of court proceedings. However, the issue of who is a party to the arbitration agreement (and therefore able to take the benefit or be made subject to the burden of the contractual promise to arbitrate when the court is asked to enforce that promise through a stay) is not, ex facie, the same question as who is bound by an award once made. The former is a question which is essentially contractual in nature, the latter one capable of engaging wider public interests of finality.

As Mr Rabinowitz QC submitted, both of these provisions can be traced back to the Arbitration Act 1889. The forebear of s.58 was the provision to be implied (as a term) into any submission to arbitrate that “the award to be made by the arbitrators or umpire shall be final and binding on the parties and the persons claiming under them respectively” (Schedule 1). S.82(2) can be traced back to s.4 dealing with the power to stay court proceedings commenced by, or against, a party to the submission or persons claiming “through or under” such a party. As I have noted, Mr Rabinowitz QC’s submission effectively treats s.58 (and its predecessors) not only as identifying persons who are bound by an arbitration award, but as circumscribing the categories of those who may be bound. There are statutory contexts in which that would be a very formidable argument. Before determining whether this is one of them, it is convenient to consider Mr Rabinowitz QC’s second argument.

By reference to National Ability SA v Tinna Oils and Chemicals Ltd (The Amazon Reefer) [2009] EWCA Civ 1330, [14], he submits that an arbitration agreement is enforceable “because of the implied contractual promise to pay an arbitration award” (an observation made when dismissing an argument that enforcement of an award under s.26 of the Arbitration Act 1950 was not subject to a limitation period, whereas enforcement by an action of the award would have been). Mr Rabinowitz QC then contends that the rule that an arbitration award gives rise to an issue estoppel also rests on the implied promise the parties have made in their arbitration agreement to perform the award. For that proposition, he is able to point to the not inconsiderable authority of Lord Hobhouse in the Privy Council in Associated Electric and Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] UKPC 11. The immediate issue in that case was whether the arbitral obligation of confidentiality prevented a party to one arbitration from relying on the award to found a defence of issue estoppel in another. Lord Hobhouse noted at [9] that:

Lord Hobhouse went on to hold that the preclusive effect of an arbitration award by way of an issue estoppel between the arbitrating parties reflected that implied term, and the confidentiality obligation could no more prevent reliance on the award to enforce that right than it could prevent reliance on the award for the purposes of other forms of enforcement ([15]). Given that the normative force of the award derives in the vast majority of cases from the parties’ agreement to arbitrate, it can certainly be said the preclusive effect of the award is ultimately derived from the contract to arbitrate, or from a statutory prohibition on denying such a contract (for an exception see [25(iii)] below). I do not accept, however that it follows from the AEGIS decision that the capacity for an award to generate an issue estoppel in subsequent proceedings is limited to the parties to the arbitration agreement pursuant to which the award was rendered.

That analysis assumes that the preclusive effect of a prior determination in a subsequent dispute is a legal incident of rights arising from the original determination, rather than the result of a rule of law applicable by the second tribunal as to the legal effect of that original determination. However, the doctrine of issue estoppel appears to me to depend on a rule of law of the “receiving” tribunal rather than the rights adjudicated on by the “transmitting” tribunal:

It follows, therefore, that while I accept (for reasons which I explain at [27] below) that the contractual source of the tribunal’s jurisdiction is a highly relevant factor when considering whether the findings in an award bind a non-party to the arbitration, I do not accept that the existence and ambit of any issue estoppel arising from an award is essentially a question of contract, dependent on which parties were entitled, and bound, by the scope of any implied promise to perform the award.

I am pleased to have reached this conclusion, because of a number of difficulties which would follow from Mr Rabinowitz QC’s analysis:

For these reasons, I reject Mr Rabinowitz QC’s submission that the effect of the AEGIS case and ss.58(1) and 82(2) of the Arbitration Act 1996 (and their predecessor provisions) is to confine the scope of issue estoppel arising from an arbitration award to contractual privies (i.e., effectively to successors in title to or those engaged in the derivative exercise of the relevant contractual right). That does not mean, however, that the contractual source of an arbitral tribunal’s substantive jurisdiction is irrelevant to the application of the doctrine of issue estoppel by the receiving court. Far from it. I accept that it is one of a number of reasons why any attempt to establish the preclusive effect of an award against anyone except the parties or their contractual privies will be an extremely challenging task.

In particular, I am satisfied that the following features of the arbitration process require a more restrictive approach to giving an award a preclusive effect in the context of Gleeson privies (just as they have been held to require a more restrictive approach when determining whether it would be an abuse of process for a non-party to advance a case which is inconsistent with a prior arbitration award in court proceedings: see [81] below):

Perhaps encouraged by the view expressed by the Honourable KR Handley in Spencer Bower & Handley: Res Judicata (5^th), [9.42] that “the requirements for privity in interest were settled until the decision of Megarry VC in Gleeson v Wippell”, Mr Rabinowitz QC argued that the concept of privity is limited to “a successor in title, i.e. a person who claims through or under a party to the prior proceedings such as an executor, trustee in bankruptcy or assignee”, and that the test which Megarry VC has been treated as formulating in Gleeson was “wrong in principle and contrary to authority”. Although accepting that “the Gleeson test has been endorsed (often obiter) in a number of first instance and some Court of Appeal cases”, it was suggested that these decisions were not binding on me because the criticisms made by Mr Rabinowitz QC of the Gleeson decision had not been raised.

It is important to note how extensively Gleeson has been cited with approval since 1977. I shall not attempt a comprehensive list, but relevant decisions include Lord Bingham’s (obiter) approval of the test, whose application had been conceded, in Johnson v Gore Wood [2002] 1 AC 1, [31]; the Court of Appeal in House of Spring Gardens v Waite [1990] 1 QB 241, Resolution Chemicals Ltd v H Lundbeck A/S [2013] EWCA Civ 924 and Ward v Savill [2021] EWC Civ 1783; and numerous first instance decisions. The case has also been mentioned without disapproval in many more Court of Appeal decisions. Against that background, I hope Mr Rabinowitz QC will not think me unduly pusillanimous for not embarking on a detailed analysis of an argument which I am satisfied is not realistically open before a first instance judge.

I confine myself to three observations:

These three paragraphs do scant justice to the detailed argument I received on this issue. However if the English law on the preclusive effect of prior determinations is to be extensively restructured in the manner contended for, a two-day hearing before a puisne judge is not the appropriate occasion to begin that process.

In Gleeson, 515, Megarry VC formulated the test for a privy in interest as follows:

Megarry VC described the concept of privity as “protean”, and his “test” is criticised in Spencer Bower & Handley as “circular”. It is fair to say that the “test” is essentially conclusory, and that it falls to be applied in circumstances in which there are a wide variety of combinations of factors which might lead to a conclusion of privity, or be insufficient to support it, in different cases. To that extent, it is a multi-factorial rather than rule-based principle. This limits the extent of the guidance which can be obtained from cases considering particular applications of the test. Without in any way purporting to identify all relevant factors (which I suspect would be an impossible task, as well as a pointless one when it is the particular combination of factors which matters), the authorities to which I was referred provided a number of “signposts” which I have found of particular assistance in this case:

It has been suggested that a relevant factor is whether the party sought to be bound could have been compulsorily joined to the original proceedings (which is to say that it is suggested that if it could not have been so joined, that is likely to tell against a finding of privity): Standard Chartered Bank, [145] (and see also the Court of Appeal, [31]) and Mad Atelier [64], [70]. I do not accept the suggestion that these cases establish that the ability to join the purported privy to the original proceedings is a necessary condition for the application of the doctrine (not least because it would render the application of the (domestic) doctrine of issue estoppel subject, in foreign judgment cases, to the vagaries of the relevant foreign procedural code). The rationale for this factor is not entirely clear – it may reflect the fact that too ready a recognition of the doctrine of privity in interest in such circumstances could circumvent the territorial or other limits of the original court’s jurisdiction. However, where it is open to the claimant to join the proposed privy to the original proceedings, but it does not do so, the claimant’s failure to remove any doubt as to the effect of a decision on that party at a time when the outcome of the dispute was not yet known might also be thought relevant (and I note that Sales J was of this view in Seven Arts, [65]). If the proposed privy sought to join the original proceedings, but that joinder was resisted by the successful party, that should also tell against a finding of privity (see by analogy the position so far as abuse of process is concerned in Bragg v Oceanus Mutual Underwriting Association (Bermuda) Ltd [1982] 2 Lloyd’s Rep 132, 137 and Michael Wilson & Partners Ltd v Sinclair [2017] 1 WLR 2646, [90]).

Mr Pillow QC placed some reliance, having regard to the decision in House of Spring Gardens v Waite [1990] 1 QB 241, on the allegation that the LCIA Claimants and the Respondents were alleged to be joint tortfeasors. By way of explanation, in House of Spring Gardens judgments had been entered in Ireland against three defendants as joint tortfeasors. Two of the defendants (D1 and D2) had applied to set the judgment aside on the ground that it had been procured by fraud and failed. The third (D3) had not joined in that application, but later sought to resist enforcement of the judgment in England on the same grounds. It was held that he was precluded by the doctrines of abuse of process and issue estoppel from doing so. The significance of the joint and several liability in that case was that the application to set aside brought by D1 and D2 necessarily engaged the interests of D3, because, if it had succeeded, the judgment would have been set aside against all three defendants (p.254G). That apart, I am not persuaded that there is any special rule which makes proceedings by one alleged joint tortfeasor seeking to challenge a liability already established against another joint tortfeasor more susceptible to a Gleeson-estoppel or a finding of abuse of process (and I note Mr Justice Arnold was of a similar view in Resolution Chemicals Ltd v H Lundbeck A/S [2013] EWHC 739 (Pat), [122]).

Finally, I have already referred at Section C1 above to the even greater caution required in finding that a non-party is bound by the terms of an award as a privy.

Without presuming that anyone other than the parties to this litigation will have any interest in any part of this judgment, the section which now follows, addressing the particular facts of this case, will undoubtedly only be of interest to the parties. For that reason, and in order to reduce the length of that part of the judgment which considers and applies the relevant legal principles, I address the particular matters relied upon by the Banks as establishing privity in this case in an annex.

By way of a summary of my conclusions:

On my findings, there can be no question of there being a summary determination in the Banks’ favour that any of the Respondents are privies of the LCIA Claimants, not least because I have been unable summarily to determine all of the underlying factual issues on which the Banks rely in their favour, or sufficient of them to make a summary determination a realistic prospect.

The issue which arises in these circumstances is whether I should simply give permission to amend, or go further and refuse permission to advance the issue estoppel plea. The former course would require a merits hearing dealing with the “live” factual issues relied upon in the context of the issue estoppel plea, together with a repetition of the extensive legal arguments I have heard. That is not a particularly attractive prospect. In any event, I do not believe the dispute in this case turns on the finer points of the parties’ competing factual cases, so much as a fundamental difference of view as to the proper approach having regard to the dispute’s broader outlines (cf [3] above).

Having regard to those parts of the Banks’ case which are either clear or arguable, but also:

In particular, many of those acts – control or the funding of a litigating company, taking decisions on its behalf or giving evidence for it – are common actions on the part of directors or controllers of corporate bodies, requiring the particular caution identified at [33(iii)] above. Having regard to the cumulative effect of the issues I have referred to in this and the preceding paragraphs, the Respondents’ desire that an LCIA arbitration tribunal pronounce on the status of the Otkritie Replacement Transactions before a Russian court did so, because of the benefits which it was perceived that could bring in the context of the alleged “campaign against the O1 Group and the Mints family”, does not change the position. This is particularly the case when it cannot realistically be claimed (i) that the LCIA Arbitrations were being conducted on the basis that they would be legally determinative so far as any liability of the individual Respondents was concerned; or (ii) that that was the Banks’ own position.

In view of the conclusion I have just reached, I intend to deal with this issue briefly.

In this context, there are three relevant requirements.

The first is that the determination of that issue must be necessary for the decision: Spencer Bower & Handley, [8.23]. This is sometimes explained in positive terms (the issue must be “fundamental”, “essential” or an “ultimate” issue) and sometimes in negative terms (it must not be “collateral” or merely “an evidentiary fact”). I accept that the test I should apply is whether the issue was “an essential step in the reasoning” of the first tribunal (Spencer Bower & Handley, [8.01]) or whether “the determination [is] so fundamental that the decision cannot stand without it” (ibid, [8.24]).

The second is that the determination of that “ultimate” issue must be clear. That requirement is even more important when the original determination is said to have been made by an arbitration award. Consistent with the distinct roles of the court and the arbitral tribunal, and the policy expressed in s.1(c) of the Arbitration Act 1996, as Gross J noted in Norsk Hydro ASA v State Property Fund of Ukraine [2002] EWHC 2120 (Comm), [17]-[18] the court will not second-guess the intentions of the arbitration tribunal or “stray into the arena of the substantive reasoning and intentions of the arbitration tribunal” where the relevant issue is not crystal clear on the face of the award.

The third is that for an issue estoppel to arise, the issue must be the same in both sets of proceedings. This proposition is sometimes qualified by the adjective “substantial” (Spencer Bower & Handley, [8.05] and Butcher J in Carpatsky Petroleum Corp v PJSC Ukrnafta [2020] Bus LR 1284, [122]). I accept some such qualification is appropriate, both to reflect the fact that the doctrine will sometimes require applying determinations under one system of law to claims under another, and to avoid an over-technical application of a doctrine which gives effect to an important public policy.

The proposed pleading of issue estoppel in this case is advanced by quoting extensive passages from the LCIA Award in the Amended Particulars of Claim in compendious form against all three Respondents. With the benefit of hindsight, this gave rise to a number of difficulties:

There are clearly a number of matters pleaded which could not on any view constitute “ultimate issues”. By way of some (amongst many) examples:

In the course of the hearing, at my request, Mr Pillow QC produced (on very short notice) a list of the ingredients of the causes of action raised by the Banks in the LCIA Arbitrations, to allow this issue to be explored in terms more appropriate to the identification of ultimate issues. Given the sheer volume of material and points which had to be addressed during the two-day hearing, there was only limited consideration of the implications of this analysis for the issue estoppel argument. For present purposes, I should note that I formed a necessarily preliminary view that findings on the following issues in the LCIA Award might arguably be said to concern substantially the same issue as an issue arising in this case:

Had I reached the conclusion that the Respondents (or any of them) were arguably privies, before giving permission to amend I would have required the Banks to serve a further document which set out clearly, for each Respondent, which particular paragraphs of their Defence were said to be precluded by which “fundamental” or “ultimate” issues which had been determined by the LCIA Award. I would then have heard further argument on the question of “identity of issue” by reference to that formulation before reaching a conclusion on the application for permission to amend. For that purpose, it might have been necessary to engage with some of the Russian law evidence which featured only fleetingly at the hearing.

However, there are three further issues I should mention.

First, I would not have accepted Mr Rabinowitz QC’s argument that a finding that an LCIA Claimant had knowledge because (i) a particular Respondent had that knowledge and (ii) the knowledge of the particular Respondent was attributable to that LCIA Claimant did not embrace the same issue as whether that same Respondent had that same knowledge. On Mr Rabinowitz QC’s approach, a finding against employer A that it was vicariously liable for the negligence of employee B would determine the issue of B’s negligence on the basis of the “employee’s tort” theory of vicarious liability, but not if the “employer’s tort” theory had prevailed. Such an approach could only commend itself to the most committed legal conceptualist.

Second, I would not have accepted Mr Pillow QC’s submission that because the LCIA Claimants had originally sought a negative declaration as to the status of the Okritie Replacement Transactions in wide terms, that meant that the Tribunal was necessarily determining “any realistic ingredient of any realistically arguable claim”. The practical position in the LCIA Arbitrations is that the Banks’ counterclaim served to define the issues to be determined, and the scope of the determinations made. In any event, I was referred to no passage in the LCIA Award in which the Tribunal was said to have made any determination which went wider than the specific allegations brought in the Banks’ counterclaim.

Third, it is necessary to say something more about the position of D1. In short, but highly effective, submissions, Mr Edey QC submitted that the LCIA Award did not arguably contain any finding of the requisite clarity that D1 had held the knowledge or dishonest state of mind alleged against him in these proceedings. I have been persuaded that he is right:

For these reasons, I would have refused permission to plead an issue estoppel case against D1 in any event.

I will also deal with this issue briefly.

It was common ground that even if the ingredients of an issue estoppel are otherwise established, the court may nonetheless refuse to give effect to the estoppel in “special circumstances”. In Arnold v National Westminster Bank plc (No 1) [1991] 2 AC 93, 109, Lord Keith explained the position as follows:

I accept the observation in Spencer Bower & Handley, [8.32] that “the exception should be kept within narrow limits to avoid undermining the general rule and provoking increased litigation and uncertainty”. I also accept that there is no closed list of special circumstances: R (East Hertfordshire District Council) v First Secretary of State [2007] EWHC 834 (Admin), [24].

One issue which does not appear to have been considered in the authorities – at least in those to which I was referred – is whether there is any scope for arguing that the “special circumstances” exception should be wider in the case of Gleeson privies than for the original parties or their successors in title. I am not attracted by that argument, because the premise of the estoppel binding a privy is that it is just and fair to treat them in the same way as a party. In those circumstances, it would be inconsistent then to apply a wider “special circumstances” exception on the rationale that it would not be fair to treat the non-party in the same way as a party.

Given my conclusion in [60] above, I have approached this issue by considering what the position would be if the LCIA Claimants had sought to re-fight ultimate issues determined in the LCIA Arbitration. So approached, I was not persuaded that it was arguable that the matters relied on here were sufficient to engage the “special circumstances” exception.

The “special circumstances” relied upon by the Respondents fall into three categories.

First, it is said that there will or may be evidence before the court at the trial which was not available in the LCIA Arbitrations:

If the question is asked whether any of these matters could have constituted “special circumstances” so as to permit the LCIA Claimants to re-fight their liability, the answer would surely have been no. The different evidential regimes between court and arbitral process, in circumstances in which no one suggests that the s.68 jurisdiction is engaged, cannot amount to “special circumstances” sufficient to deprive an award of whatever preclusive effect it would otherwise have, nor could the fact that the LCIA Claimants had instructed an unsatisfactory expert (the Banks’ failure to do so being something which, in the ordinary course, should have been a positive benefit for the LCIA Claimants). The submission made on behalf of D2 and D3 that the issues on disclosure “comprised matters which might be acceptable in arbitration but not in these proceedings” is wholly out of kilter with the policy of English arbitration law. As the Departmental Advisory Committee on Arbitration Law observed in its Report on the Arbitration Bill (February 1996) when discussing what would become s.68 of the Act at [280]:

I would also note, although I have not relied on this factor, that Mr Dankevich (who was very much at the forefront of the Respondents’ submissions on this issue) was in Israel during the LCIA Arbitration. I was shown nothing to suggest that the court’s assistance could not have been sought to obtain his evidence through letters rogatory if anyone had thought it sufficiently worth their while (or been willing to take the risk), utilising the court’s powers under s.44(2)(a) of the Arbitration Act 1996 (see Dame Sara Cockerill, The Law and Practice of Compelled Evidence in Civil Proceedings, [5.41]).

Second, it is said that the court will hear evidence of a potential commercial justification for the Otkritie Replacement Transactions from Bank Otkritie’s perspective, namely that they avoided the need to provision for certain loans. That point did feature, albeit late, in the LCIA Arbitrations. No explanation has been given for why it could not have been advanced at an earlier stage if it was believed to have merit and could be reconciled with the case theory then being advanced. On the hypothesis on which the present argument is proceeding (that the Respondents are the LCIA Claimants’ privies), the failure to run a particular argument sufficiently early in the LCIA Arbitrations could not begin to amount to a special circumstance. No doubt anyone who has ever lost a case can identify in retrospect alternative arguments which might have been more viable, or which offer at least a prospect of a way through the wreckage, but that is no basis for depriving the original decision of the preclusive effect it would otherwise have.

The final point raised by the Respondents is the risk of inconsistent findings on the issue of the dishonesty of the Otkritie Replacement Transactions and those involved in them as between any persons estopped by the LCIA Award, and those such as D4 to D7 (and on my findings, D1) who are not. It was suggested that the risk was particularly acute here because there is nothing which would prevent the estopped individuals from giving evidence on behalf of other defendants. The overall effect, it was said, would be to create an “affront to justice”.

The Respondents relied in this regard on Tugushev v Orlov [2021] EWHC 926 (Comm). In that case, there was a dispute between three parties (T, O and R) as to whether there was a three-way joint venture between them in relation to a fishing boat, or a two-way joint venture involving O and R. There had been unfair prejudice proceedings involving O and R in Hong Kong (conducted on the basis that O and R held 50% each in the joint venture company). T's putative interest was noted, and R asked the judge to provide for it, but no mechanism for doing so was identified. The judge made an order requiring R to buy out O’s 50% stake, observing that he was not persuaded that it would not be possible satisfactorily to address T’s position if the existence of a three-way agreement was ultimately established. In English proceedings commenced by T to assert his rights under a three-way joint venture, R counterclaimed against O on the basis that if the three-way agreement alleged by T was established, he would have paid O too much when buying out his stake, to be met with the response that R was estopped by the Hong Kong proceedings from raising that argument. Sir Nigel Teare concluded that the “special circumstances” exception was engaged because:

In Tugushev, the injustice identified by Sir Nigel Teare arose because a stranger to the original proceedings raised a claim, and the claim which was sought to be precluded was contingent on that stranger’s claim being established. The case did not involve, therefore, a unilateral attempt by one of the original parties or their privies to re-open the earlier determination. It was also a case in which the judge in Hong Kong appeared to have contemplated that it would remain possible to work out the financial consequences if T did have a one third interest. Those features are not present in this case which (on the current hypothesis) would involve the same parties or their privies litigating the same point for their own purposes, not contingently in response to a third-party claim.

While I do not pretend I would have been able to contemplate the scenario outlined in [67] with any degree of sanguinity, I am not persuaded that the risk of inconsistent findings between parties who are estopped, and parties who are not, in respect of the same transaction or issue of itself gives rise to special circumstances. This risk arises inevitably whenever some parties to an overall dispute are subject to arbitration agreements and others are not (as Sir Nigel Teare noted in Tugushev, [54]). It can also arise when the same transaction features in successive High Court actions with overlapping but not identical parties and common witnesses. Current and former commercial practitioners of a certain vintage will recall a vivid example of this in the Orion v Sphere Drake litigation, in which the effect of a three-way arrangement was determined in Orion’s favour in proceedings between Orion and Sphere Drake ([1990] 1 Lloyd’s Rep 465; [1992] 1 Lloyd’ Rep 239), in which the relevant witness from the third party, Baloise, gave evidence, and then determined to contrary effect in subsequent proceedings between Orion and Baloise after hearing from some of the same witnesses (Sphere Drake Insurance Plc v Basler Gesellschaft [1998] 1 Lloyd’s Rep (Insurance and Reinsurance) 35). While Sphere Drake eventually succeeded in setting aside the original judgment on the basis that it had been procured by fraud ([2001] Lloyd’s Rep IR 1), such that the estoppel fell away, it was never suggested that the risk of inconsistent findings alone was sufficient to free Sphere Drake from the effect of the earlier judgment. I cannot think that if Sphere Drake had been joined to the second action, this would have circumvented the need to apply to set aside the judgment, nor that if Sphere Drake and Baloise had been jointly sued for failing to pay amounts due under the pool arrangements, it would have been open to Sphere Drake to renew the argument that these had been finally settled at the famous 23 April 1975 meeting simply because the court would be hearing argument from Baloise to that effect, relying on the same witnesses.

There is no dispute that the court has an inherent power to prevent the abuse of its procedures by actions which, although not involving an express breach of the rules, would give rise to manifest unfairness to another party or bring the administration of justice into disrepute: Hunter v Chief Constable of the West Midlands Police [1982] AC 529, 536 (Lord Diplock). Such an abuse may arise from an attempt to relitigate an issue determined in other proceedings, even though no issue estoppel has arisen. However, it is likely to be a rare case in which this is so. As Lord Lowry noted in Shaw v Sloan [1982] NI 393, 397:

It has been said that in order to determine whether proceedings which seek to re-litigate such an issue are abusive, it is necessary to “engage in a close ‘merits based’ analysis of the facts” (Michael Wilson & Partners Ltd v Sinclair [2017] 1 WLR 2646, [48]).

There are decisions in which findings of abuse have been made in respect of attempts to relitigate issues raised in prior litigation and where the identity of parties necessary for an issue estoppel is lacking. Mr Pillow QC placed particular reliance on two such cases.

The first is Secretary of State for Business, Innovation & Skills v Potiwal [2012] EWHC 3723, (Ch). In that case, the VAT tribunal in proceedings brought by HMRC against a company, Red 12, found that Red 12 through its sole director Mr Potiwal had engaged in fraudulent MTIC transactions. Mr Potiwal was the sole source of instructions for Red 12 and its sole witness at that hearing, at which he was subjected to a lengthy cross-examination. The Secretary of State then sought to rely on those same findings in director’s disqualification proceedings brought against Mr Potiwal. The parties to both proceedings were different, and while Briggs J found Mr Potiwal to be a privy of Red 12, he was not persuaded that the Secretary of State was a privy of HMRC. Given that both HMRC and the Secretary of State were (to adopt language used in another context) emanations of the UK state, that might be thought to be a “near miss” in privity terms (assuming the correctness of the decision so far as Red 12 and Mr Potiwal are concerned), and, to that extent, the case might be said to fall within the “spirit” of the doctrine of issue estoppel (adopting a phrase used by Lord Hoffmann in Arthur J Hall & Co v Simon [2002] 1 AC 615, 701 when describing the role of abuse of process in re-litigation cases).

Briggs J found that Mr Potiwal’s attempt to require the Secretary of State to prove his knowledge of and involvement in the fraudulent transactions an abuse of process, a rare example of a finding of abuse against a defendant seeking to raise arguments determined against a closely-related party in prior proceedings (see further [76]-[77] below). Briggs J laid particular emphasis on the fact that the tax payer was meeting the costs of both sets of proceedings ([27]), and the lack of evidence that Mr Potiwal could meet any order requiring him to reimburse those costs. The case undoubtedly represents the “high watermark” of abuse findings in cases concerned with the re-litigation of issues where there is no identity of parties. The dilution of the rule in Hollington v Hewthorn suggested by cases decided since Potiwal might provide an alternative means of addressing the vice in that case (see JSC BTA Bank v Ablyazov [2016] EWHC (Comm) 3071, [24]).

The second case is Tinkler v Ferguson [2021] EWCA Civ 18, in which it was held to be abusive for a claimant, who had lost litigation in the Commercial Court against a company said to be vicariously liable for breaches of fiduciary duty by other directors in issuing a press release referring to him, to sue those directors for malicious falsehood in relation to the contents of the same press release in separate proceedings. In that case, there was no identity of parties, but there was something close to it because the litigation in the Commercial Court concerned the vicarious liability of the defendants for the alleged wrongs of the directors who were defendants in the malicious falsehood action in respect of substantially the same alleged wrongs. To that extent, it might also be said that the case fell within the “spirit” of the doctrine of issue estoppel, and I note that Peter Jackson LJ described the case as one which was “not formally between the same parties or their privies” (emphasis added) (at [62]), although he also noted that the absence of identity of parties was a “powerful factor” against any finding of abuse ([66]).

I accept the Respondents’ submission that it is likely to be easier to establish an abuse where the claimant seeks to re-litigate a claim which has failed against another defendant than where a defendant seeks to run a defence which failed in proceedings brought by the same claimant against a different defendant. The sentiments to that effect expressed by Ward LJ in Conlon v Simms [2008] 1 WLR 484 and Blair J in OMV Petrom SA v Glencore International AG [2014] 2 Lloyd’s Rep 308, [30] have the same intuitive appeal which animated the words of a traditional French song: “ce chien est très méchant; quand on l’attaque, il se défend”. As those decisions show, the intuitive appeal of that argument is likely to be stronger still when an attempt is made to rely on a prior determination to preclude the defendant from defending allegations of dishonesty, although Potiwal would suggest that even these considerations have a finite reach.

The Banks referred me to a passage in Zuckerman on Civil Procedure, 4^th edition, [26.145], questioning the utility of the distinction between the position of claimants and defendants in the abuse context. While I agree that the formal character of a party is unlikely to be determinative (e.g., where proceedings are initiated by a party seeking a negative declaration as to the other’s claim, with a counterclaim following), I accept that there is a difference to be drawn between those who positively invoke the court’s processes in order to secure substantive relief from the other party, and those whose involvement is essentially defensive in response to the latter’s claims. The actions of the former more readily lend themselves to the characterisation of “vexing” an opponent than the latter. In any event, I note the relevance of the distinction is also supported by the Court of Appeal in Michael Wilson, [93] and Bragg, p.137.

Finally, it may, in appropriate circumstances, be an abuse of process to seek to relitigate in court an issue which had already been determined in an arbitration award, even where there is no “identity of parties” for issue estoppel purposes: Hamblen J so held in Arts and Antiques Ltd v Richards [2014] Lloyd’s Rep IR 219, [20], as did Reyes J in Hong Kong in Parakou Shipping Pte Ltd v Jinhui Shipping and Transportation Ltd [2010] HKCFI 817 (Reyes J emphasising the need for caution when applying this doctrine to a prior arbitral determination ([173])). The Court of Appeal in Michael Wilson agreed ([67]). On this basis, there may be an asymmetry between the effect of a prior court decision on arbitral proceedings, and the effect of a prior arbitral award on court proceedings: see [25(v)].

Arts and Antiques involved an attempt by an unsuccessful arbitral claimant (and hence a party to the arbitration agreement) to re-litigate a dispute as to the terms of an insurance contract in court proceedings against the insurer and broker, having failed on the same arguments in an arbitration against the insurer. Hamblen J relied on another case (Taylor Walton (A Firm) v David Eric Laing [2997] EWCA Civ 1146) in which a party had lost litigation against its contracting party as to what the terms of the contract were, and then sought to bring a claim against its solicitors in subsequent proceedings premised on its (failed) case as to the terms of the contract. They were both cases, therefore, in which the foundation of a second action against an agent was a particular contractual state of affairs which had been held not to exist in litigation between the contracting parties.

The caution which I have found is required before finding that a non-party will be bound by an arbitration award as a privy of an arbitrating party (see [27] above) also applies when an attempt is made to argue that a prior determination in an arbitration award makes it an abuse of process for the non-party to seek to raise a particular issue in subsequent litigation. Indeed, if anything, the fact that the non-party does not meet the requirements for Gleeson privity would suggest that this should be an even more challenging argument. It is not surprising, therefore, that Simon LJ in Michael Wilson, [68] suggested it would be “perhaps a very rare case, where court proceedings against a non-party to an arbitration can be said to be an abuse of process”.

The factors at [41]-[42] above which led me to conclude that it is not arguable that the Respondents are privies of the LCIA Claimants in this case themselves provide compelling grounds for concluding that there is no abuse of process in the Respondents raising issues in their defences which are inconsistent with the findings in the LCIA Award. This is not a case in which I have found the issue estoppel case failed on some technical requirement, such that it can be said that preventing the Respondents from advancing their defences in a manner which was inconsistent with the LCIA Award falls within the “spirit” of an issue estoppel.

The Banks have failed to identify some special feature of this case which arguably requires the court to conclude that there would be an abuse of process here, whereas the Respondents are able to point to further strong reasons why there would not:

For all these reasons, I have concluded that the abuse of process amendment is not realistically arguable either.

For these reasons, the Banks’ application for permission to amend is refused, and the further applications do not arise.

The parties are asked to agree proposals for the court’s approval as to what consequential issues arise for determination, and how they should be resolved.

It will be apparent that the (important) issues and (high quality) arguments involved in this application required a great deal more time than the two days set aside to do them justice. Although the court sat extended hours, and the pace of submissions was such that the two days’ argument generated 208 and 223 pages of transcript respectively, there were a significant number of authorities or other documents for which the court was simply given references “for your Lordship’s note” to read in its own time. My experiences in this regard bear some similarities to those described by His Honour Judge Pelling QC in Libyan Investment Authority v Credit Suisse International [2021] EWHC 2684 (Comm), [139]-[140]. It goes without saying that this is not a satisfactory way in which to conduct heavy interim applications in this court (as the recent Practice Notice from the Judge in the Commercial Court dated 29 March 2022 makes clear).