IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS
OF ENGLAND AND WALES
COMMERCIAL COURT
KING’S BENCH DIVISION
IN THE MATTER OF THE ARBITRATION ACT 1996
AND THE MATTER OF AN ARBITRATION CLAIM
7 Rolls Building
Fetter Lane
London,
EC4A 1NL
Before:
MRS JUSTICE COCKERILL
Between:
(1) HULLEY ENTERPRISES LIMITED (a company incorporated in Cyprus) (2) YUKOS UNIVERSAL LIMITED (a company incorporated in the Isle of Man) (3) VETERAN PETROLEUM LIMITED (a company incorporated in Cyprus) | Claimants |
- and – | |
THE RUSSIAN FEDERATION | Defendant |
Jonathan Crow CVO KC and David Peters (instructed by Stephenson Harwood LLP) for the Claimants
Khawar Qureshi KC (instructed by Dmitry Andreev) for the Defendant
Hearing dates 4,5 October 2023
APPROVED JUDGMENT
I direct that no official shorthand note shall be taken of this Judgment and that copies of this version as handed down may be treated as authentic.
This judgment was handed down remotely by the judge and circulated to the parties’ representatives by email and release to The National Archives. The date and time for hand-down is deemed to be Wednesday 1 November 2023 at 10:30am
Mrs Justice Cockerill:
Introduction
Over the course of two days I have heard the trial of two preliminary issues arising out of a longstanding and complex dispute between the Defendant (the Russian Federation: “RF”) and the Claimants, who are the former majority shareholders in OAO Yukos Oil Company (“Yukos”). The manifestation of that dispute which concerns this court is enforcement of arbitral awards of some US$50 billion plus compound interest accruing at some US$2.5 million dollars a day (“the Awards”).
These Awards resulted from determinations of a Tribunal of well known international arbitrators that (i) they had jurisdiction to hear a claim brought by the Claimants for breach of the RF’s obligations under Article 13(1) of a treaty called the Energy Charter Treaty (“the ECT”); (b) the Claimants’ allegations of breach (via unlawful expropriation via tax demands and bankruptcy proceedings) were well-founded; and (c) the RF was liable to pay the Claimants more than US$50 billion in damages.
Those preliminary issues were defined as:
Issue 1: Whether and to what extent the RF is, by reason of certain judgments of the Dutch courts, precluded from re-arguing the question of whether it has agreed in writing to submit to arbitration the disputes that are subject of the Awards;
Issue 2: Whether, if the answer to Issue 1 is that the RF is so precluded from re-arguing the relevant question, the Jurisdiction Application ought to be dismissed forthwith.
Those issues arise against a procedural background which has already been summarised at length in two judgments of this court. The parties know the facts intimately. The interested reader is referred to:
[8 – 44] of the judgment of Henshaw J reported at [2021] 1 WLR 3429 (“the Henshaw Judgment”); and
[2 – 23] and [27 – 38] of the judgment of Butcher J at [2022] EWHC 2690 (Comm) (“the Butcher Judgment”).
In essence:
On 10 November 2014, the RF commenced the Dutch Proceedings: a challenge to the Awards in the courts of the Netherlands (the arbitral seat) seeking to have the Awards set aside. In that litigation it advanced a range of challenges including ones which went to the Tribunal’s jurisdiction, and also ones relating to the conduct of the arbitration;
On 30 January 2015, the Claimants initiated these English Enforcement Proceedings seeking the recognition and enforcement of the Awards in this jurisdiction;
On 25 September 2015, the RF filed its application contesting the jurisdiction of the English Court on the basis of the RF’s state immunity pursuant inter alia to Section 1 of the State Immunity Act 1978. Its Summary of Evidence outlined the arguments (described in more detail in the Henshaw Judgment at [49]) that:
The agreement in Article 45 of the ECT did not on the true construction of that article apply to the Article 26 arbitration process (“the Article 45 argument”);
The Claimants were not entitled to invoke the provision because they were not “investors” (as defined) with an “investment” (as defined) (“the investor/investment argument”);
The claims brought lay outside the scope of the dispute resolution provisions because Article 21 of the ECT carved out taxation, and the claims were based on taxation measures (“the Article 21 argument”).
On 8 June 2016, the English Enforcement Proceedings were stayed by consent and by order of Leggatt J (“the Stay”).
Meanwhile on 20 April 2016, the Awards were set aside by the District Court of The Hague (“The Hague DC”), on the basis of the Article 45 argument.
The Hague Court of Appeal (the “Hague CoA”) allowed the Claimants’ appeal on that issue and (in accordance with Dutch procedure) it also considered de novo all of the RF’s other challenges to the Awards (including (i) Set Aside Ground 2: the Article 45 argument (ii) Set Aside Ground 3-4: the investor/investment argument (iii) Set Aside Ground 5: the Article 21 point and (iv) Set Aside Ground 1: a “fraud in the arbitration” allegation). The set aside grounds are listed more fully in the Henshaw Judgment at [44].
In the event, The Hague CoA Judgment in February 2020 rejected all of the RF’s challenges (including ruling that the RF was not entitled, in the context of a set aside application under Article 1065 of the Dutch Code of Civil Procedure , to run the fraud in the arbitration argument) and reinstated the Awards.
The RF then brought a cassation appeal in the Dutch Supreme Court (“DSC”) against some (but not all) of the conclusions in The Hague CoA Judgment. The points which were “in play” on the appeal were: (i) entitlement to run the fraud argument (ii) the Article 45 argument, (iii) investment/investor argument (iv) aspects of the Article 21 argument.
On 6 July 2020, the Claimants applied to lift the Stay following a judgment of the Hague CoA. At this point the RF’s cassation appeal to the Dutch Supreme Court (“DSC”) was pending. That application was rejected by Henshaw J by his judgment dated 14 April 2021 which continued the Stay.
On 5 November 2021, the DSC gave judgment (“the DSC Judgment”), finding that the Hague CoA’s rulings on Set Aside Grounds 2 to 7 did not result in cassation (i.e. upholding the Hague CoA), but that the Hague CoA had erred in its ruling on Set Aside Ground 1 (fraud). Accordingly, the DSC Judgment quashed the Hague CoA Interim and Final Judgments and referred the case to the Amsterdam Court of Appeal for further consideration and decision.
The net result was that apart from the procedural question as to whether or not the argument on fraud in the arbitration should have been excluded from the set-aside proceedings, all the grounds on which the Defendant sought to rely in relation to the jurisdiction of the Tribunal failed either in the Hague CoA and were not appealed (the Article 21 argument) or failed in both the Hague CoA and the DSC (investor/investment and Article 45).
Following the handing down of the DSC Judgment, the Claimants re-applied to lift the Stay. In October 2022, Butcher J acceded to that application in part, lifting the Stay “solely for the purpose and to the extent necessary for the resolution of the Defendant’s Jurisdiction Application”, and giving directions for the determination of the Preliminary Issues.
The Butcher Order granted each of the Claimants and the Defendant permission to file evidence from a single Dutch law expert on the following issue: “whether and to what extent the determinations in the Dutch Judgments are final and/or conclusive as a matter of Dutch law as between the Claimants and the Defendant”. Pursuant to the Butcher Order, the Claimants filed evidence in the form of three expert reports of Prof. Dr. Jan Willem August Biemans dated 20 December 2022, 29 August 2023 and 25 September 2023. The Defendant filed its evidence in the form of an expert report of Mr. Jacob Cornegoor dated 2 June 2023 and a rebuttal expert report dated 8 September 2023.
The Parties’ Arguments
The Claimants (in summary) urge me to answer both preliminary issues in the affirmative. They say that:
The issue which arises on the Jurisdiction Application is whether the RF has agreed to submit to arbitration the disputes that are the subject of the Awards (“the Arbitral Jurisdiction Issues”). If it has, then (a) the exception to State immunity identified in s. 9 of the State Immunity Act 1978 (“the SIA 1978”) applies; and (b) the Defendant’s assertion of State immunity in relation to the enforcement proceedings, (and hence the Jurisdiction Application) must be rejected;
A final and conclusive answer to the Arbitral Jurisdiction Issue has already been given by the curial courts pursuant to the Hague CoA Judgment and the DSC Judgment. By these judgments (collectively “the Dutch Judgments”) the Dutch courts have concluded that the Defendant did agree to arbitrate the disputes which are the subject of the Awards;
Under English law, the Dutch Judgments give rise to an issue estoppel which precludes the Defendant from re-running its failed case on the Arbitral Jurisdiction Issue;
The Defendant is therefore not permitted to re-argue the Arbitral Jurisdiction Issue before this court;
Since the Jurisdiction Application depends entirely on the Defendant being able to (a) re-argue the Arbitral Jurisdiction Issue and (b) to persuade this court to reach a different conclusion from the Dutch courts, it follows that the Jurisdiction Application must be dismissed.
The RF says the opposite. It contends that:
The Preliminary Issues require the Court to accept and conclude that the Dutch Judgments give rise to issue estoppel/res judicata, such that the English Court should not (indeed cannot) consider whether the Defendant should be afforded its immunity from jurisdiction pursuant to the SIA 1978.
The State Immunity Act 1978 is not overridden by common law doctrines. The English Court is required to consider for itself whether in fact there is a valid and binding arbitration agreement that curtails immunity pursuant to Section 9 of the State Immunity Act 1978 (“the Immunity Issue”). This evaluation must be undertaken at the very outset – and the Court can and must determine for itself the List of Issues agreed by the Parties on 19 February 2016 to determine the threshold question of jurisdiction;
A decision of a foreign Court cannot determine the Immunity Issue in the sense of res judicata/issue estoppel as a matter of English law;
The reference to Section 31 of the Civil Jurisdiction and Judgments Act 1982 by the Claimants is misconceived:
This statutory provision requires the English Court to consider whether the Defendant would not have been immune “if [the Dutch Courts] had applied rules corresponding to [Section 9 of the State Immunity Act]” – the very question the English Court is itself being asked to consider.
The fact that the Defendant is challenging the Awards in the Dutch Courts does not constitute a “submission to the jurisdiction” as required by Section 2 of the State Immunity Act 1978 – all the more so where the central basis of its presence before the Dutch courts is that it did not agree to arbitrate (and thus maintains its claim to immunity in this regard).
The Immunity Issue is itself dependent upon the determination of fundamental issues of interpretation applying Public International Law concerning a multilateral treaty (“the Treaty Issues”) and the English Court is not bound by findings made by a foreign Court on issues of treaty interpretation (let alone an arbitral tribunal) and must identify the proper interpretation of the Treaty itself.
The Treaty Issues are contentious. The Dutch Courts have themselves rendered conflicting decisions on the Treaty Issues. Henshaw J (albeit tentatively) considered that there was a “well arguable” “realistic prospect of success” underpinning the position of the RF that the Treaty had been misinterpreted by the Arbitral Tribunal and The Hague Court of Appeal;
Thus, a full de novo consideration of the Treaty Issues is required;
Further, and in any event, the Dutch Judgments are not final, irreversible or unassailable as:
The effect of the DSC judgment is to annul the Hague CoA judgment and the issues will require to be determined by the courts de novo;
The Amsterdam Court of Appeal is presently considering whether the Awards should be set aside because of fraud perpetrated by the Claimants in pursuit of the arbitration claims. It is not expected that any such decision will be reached by the Amsterdam Court of Appeal until mid-2024 (at the earliest). Therefore, the Dutch Proceedings currently remain pending, and indeed, at a crucial stage.
There are significant prospects that a CJEU reference will be made on the Treaty Issues by the Amsterdam Court of Appeal. It cannot be said they are acte clair – especially when the Dutch Courts have rendered conflicting interpretations and the Paris Cour d’Appel formulated questions for a potential reference to the CJEU in June 2017;
The Questions Posed By Issue 1
The issues between the parties on the first issue can be summarized thus:
Does the English Court have a freestanding duty under the SIA 1978 to determine whether or not to uphold state immunity of its own motion which take precedence over any final and binding determination of a foreign court which could otherwise give rise to an issue estoppel?;
Have the Dutch Judgments reached a final and conclusive determination on whether the Defendant has agreed in writing to submit to arbitration the disputes that are the subject of the Award. In particular:
Have Grounds 2,3 and 5 been finally and conclusively determined (before the Dutch Courts have reached a final and conclusive determination on Set Aside Ground 1);
Can Grounds 2,3 and 5 be finally and conclusively determined before the Dutch Courts have reached a final and conclusive determination on Set Aside Ground 1?
Does State Immunity Preclude A Finding Of Issue Estoppel?
The RF placed the well recognised duty to give effect to the immunity conferred by the SIA 1978 at the front and centre of its submissions, relying on a wealth of authority. There is however no issue as to those propositions. The question is whether they mean that issue estoppel cannot be relied on. This is an argument best taken in stages.
There is no issue that foreign judgments can result in issue estoppel: the Claimants referred me in particular to Carl Zeiss Stiftung v. Rayner & Keeler Ltd & Others [1967] AC 853 (Lords Reid at p. 818 Hodson p. 927 Upjohn p. 948 Wilberforce p 966) and to the recent Supreme Court decision of Gol Linhas Aereas SA v Matlinpatterson Global Opportunities Partners (Cayman) II LP & Others [2022] 2 Lloyd’s Rep 169. The RF did not dispute those principles.
The test is also common ground. The Claimants cite the most widely recognised formulation taken from The Good Challenger Navegante SA v. Metalexportimport SA [2004] 1 Lloyd’s Rep 67, at [50] & [72], which states that a foreign judgment may give rise to an issue estoppel if:
It is given by a foreign court of competent jurisdiction;
It is final, conclusive and on the merits – in the sense that the relevant issues cannot be relitigated in the foreign country;
There is identity of parties;
There is identity of subject matter – which means that the issue decided by the foreign court must be the same as the issue arising in the English proceedings; and
The decision on the relevant issue is one which is treated by the relevant foreign court as necessary for its decision, in the sense that it was part of the decision which it in fact reached, and was not collateral to it, or obiter.
The RF, while preferring the more recent summary of the principles by Bryan J in MAD Atelier International BV v Manès [2020] EWHC 1014 (Comm), [2020] QB 971 at 989A-C did not take issue with these elements.
The RF was also at pains to remind me of the need for caution in finding issue estoppel based on a foreign judgment which is highlighted in some of the judgments. It is important in this context to note that it is not a generalised, but an analytical caution as the judgment of Lords Hamblen and Leggatt in Gol Linhas explains:
“38. The point has been made that there may be a need for caution before finding an issue estoppel based on a foreign judgment: see Carl Zeiss at p 918 (Lord Reid) and p 967 (Lord Wilberforce); The Good Challenger, para 54(ii). The main potential reason for such caution, in the words of Lord Reid in Carl Zeiss at p 918, is that:
‘we are not familiar with modes of procedure in many foreign countries, and it may not be easy to be sure that a particular issue has been decided or that its decision was a basis of the foreign judgment and not merely collateral …’
This should not, however, be regarded as a reason to decline to treat a foreign judgment as conclusive where the domestic court is able to reach a clear view on those matters. As observed in Yukos Capital Sarl (JSC) v Rosneft Oil Co (No 2) [2011] EWHC 1461 (Comm); [2012] 1 All ER (Comm) 479, para 49:
‘… the [need] for caution … is most likely to be relevant when considering the precise identity of the issue determined, whether it was necessary for the decision and whether there has been a decision ‘on the merits’. Where differences in procedure make these issues difficult to determine then the court needs to exercise caution. However, if these matters are clear then the need for caution does not arise.’”
So far as satisfaction of the Good Challenger requirements is concerned they are largely common ground save as to “final and conclusive”, though there is an issue on identity of issue/subject matter, to which I shall return.
The faultline between the parties was principally as to how that principle interrelates with State Immunity. The RF says it does not. As I have noted, it places huge weight on the recognition of the importance of state immunity in the case law. The RF says that one starts from the principle of immunity in Section 1 SIA 1978 and the fact that Section 1(2) SIA 1978 has long been understood as imposing upon the English Court “a positive duty to give effect to the immunity conferred by the [SIA 1978]”. It notes that the English Courts have consistently underscored the importance of affording a State a full and effective opportunity to ventilate and argue state immunity as part of the Section 1(2) duty.
The RF submits that, just as the points that a State may raise as to the applicability of state immunity may not be “foreclosed” by anything that happened in the arbitrations, neither can they be so “foreclosed” by anything that has happened in proceedings brought elsewhere to challenge awards arising out those arbitrations. The English Court’s freestanding duty to inquire requires it to be satisfied on its own analysis of English law whether Section 9 SIA 1978 does or does not apply on the facts of any given case.
It is the essence of the RF’s position that if issue estoppel is found here, this court effectively fails to afford a state a proper opportunity to argue state immunity. Or, to put it another way, the duty to give effect to immunity is attenuated or abrogated by allowing the RF’s longstanding invocation of state immunity as an objection to jurisdiction to be summarily disposed of on the basis of foreign court judgments without full argument in the English Courts. It further submits that the only legitimate use of the Dutch Court judgments would have been for the Claimants to apply for recognition of those judgments under s. 31(4) of the Civil Jurisdiction and Judgments Act 1982 (“the CJJA”), which has not been done.
The Claimants’ answer to this is simple – that the RF has not understood the argument. The Claimants do not suggest the court is precluded from determining state immunity or that state immunity is overridden or sidestepped by issue estoppel. The Claimants say that what they are doing is positively inviting the Court to rule on state immunity. The Act, they say, conveys rights, but subject to exceptions, and there is nothing wrong with applying English Law (including issue estoppel) and thereby reaching the answer that there is no state immunity. They submit that the true position is that under s.1 the Defendant is prima facie entitled to immunity, but only to the extent not disapplied (e.g. because s.9 is engaged).
The Claimants therefore invite me to decide whether s. 9 applies using the full panoply of tools available to me – including issue estoppel. They submit that the Act says nothing about how the court should set about determining that issue. Nor, they say, does the CJJA limit the use which can be made of the Dutch judgments. Rather s. 31 provides a basic rule for recognition and enforcement which is satisfied here such that there is no bar to relying on that judgment for the purposes of issue estoppel.
Discussion
The starting point is that there is no clear authority on this point either way. There is (to the best of the parties’ knowledge, which can probably be presumed to be accurate given the resources involved on both sides) no case where an issue estoppel has been found against a foreign state via a foreign judgment. But there is equally no case where it has been said that such an estoppel could not arise.
The balance of authority tends to favour the Claimants, in that there are two cases which implicitly recognize the possibility of issue estoppel against a state.
The first case is Dallah v Pakistan [2011] 1 AC 763. In the Court of Appeal Rix LJ plainly alluded to the point in the following passage at [90]:
“As for the case of a successful or unsuccessful (or waived) challenge in the courts of the country of origin, that is a more controversial area. My own view is that a successful challenge is not only in itself a potential defence under the Convention or our statute but likely also to raise an issue estoppel. As for an unsuccessful challenge, that may also set up an issue estoppel.”
In the House of Lords at [23] Lord Mance noted that an issue estoppel argument based on lack of participation in the arbitration was “bound to fail” but hinted he agreed with Rix LJ in saying:
“A person who denies being party to any relevant arbitration agreement has no obligation to participate in the arbitration or to take any steps in the country of the seat of what he maintains to be an invalid arbitration leading to an invalid award against him. The party initiating the arbitration must try to enforce the award where it can. Only then and there is it incumbent on the defendant denying the existence of any valid award to resist enforcement.”
At [98] Lord Collins made a similar point:
“Consequently, in an international commercial arbitration a party which objects to the jurisdiction of the tribunal has two options. It can challenge the tribunal’s jurisdiction in the courts of the arbitral seat; and it can resist enforcement in the court before which the award is brought for recognition and enforcement. These two options are not mutually exclusive, although in some cases a determination by the court of the seat may give rise to an issue estoppel or other preclusive effect in the court in which enforcement is sought.”
The second case was that of Diag Human SE v Czech Republic [2014] EWHC 1639 (Comm), [2014] 1 CLC 750 where Eder J at [59] said this:
“However, in circumstances where a foreign court decides that an award is not ‘binding’, I see no reason in principle why that decision should not give rise to an issue estoppel between the parties provided, of course, that the other conditions referred to above apply. In particular, provided that the issue is the same and that the decision can properly be said to be ‘on the merits’, it does not seem to me that the fact that such decision was made in the context of enforcement proceedings as opposed to any other type of proceedings can of itself be material. Indeed, that is consistent with the view expressed in the leading textbook, Dicey Morris & Collins … It also seems implicit in the decision of the Court of Appeal Yukos Capital v Rosneft.”
Those authorities plainly do not decide the point or even consider it absolutely squarely, but the indications they give support the Claimants’ approach.
It is therefore worthy of note that the distinction sought to be drawn in this case by the RF between a foreign judgment in a private context being capable of giving rise to an issue estoppel and a foreign judgment against a state not being so capable, does not appear to have featured in the cases thus far. However, it seems to me that the RF is right to this extent - that the caution which is urged on judges when finding an issue estoppel in the context of a foreign judgment must be particularly in evidence when the judgment is a judgment of a foreign court against a state.
Reliance was also placed by the Claimants on an extempore judgment of my own: Zhongshan Fucheng Industrial Investments Co Ltd v Federal Republic of Nigeria [2022] EWHC 3286 (Comm) [20] (permission refusal [2023] EWCA Civ 867). That judgment is of limited assistance. The case was concerned with a very different situation. However again it does tend to support the proposition advanced for the Claimants that there is nothing in the SIA 1978 which indicates that such an estoppel is impermissible.
That is a point which also emerges from an actual consideration of the SIA 1978. There is nothing in it which disapplies procedural rules or substantive rules that would otherwise apply. There is therefore no reason interior to the Act which would make issue estoppel inapplicable to a question which arises vis a vis a state.
Finally stepping back to the point about the duty to give effect to the immunity conferred by the SIA 1978, this can only cause a problem if the effect of the issue estoppel genuinely is to preclude consideration of state immunity.
This brings us to s. 31 of the CJJA, the relevant parts of which provide as follows:
“31. Overseas judgments given against states, etc.
(1) A judgment given by a court of an overseas country against a state other than the United Kingdom or the state to which that court belongs shall be recognised and enforced in the United Kingdom if, and only if—
(a) it would be so recognised and enforced if it had not been given against a state; and
(b) that court would have had jurisdiction in the matter if it had applied rules corresponding to those applicable to such matters in the United Kingdom in accordance with sections 2 to 11 of the State Immunity Act 1978.
…
(4) Sections 12, 13 and 14(3) and (4) of the State Immunity Act 1978 (service of process and procedural privileges) shall apply to proceedings for the recognition or enforcement in the United Kingdom of a judgment given by a court of an overseas country (whether or not that judgment is within subsection (1) of this section) as they apply to other proceedings.”
This Act, which is all about enforcement of judgments, therefore sets out the tests which are applicable to recognise a foreign judgment against a state. If enforcement or formal recognition is in issue, the CJJA provides that so long as (i) the relevant enforcement test is met and (ii) the judgment is one which would not have offended against the SIA 1978 if the case were brought here recognition/enforcement should follow. In other words: you can enforce/recognise a judgment here, but only if you could have got it here. Thus in the context of recognition or enforcement the question about the duty to give effect to the immunity is captured by s. 31 (1) (b).
This argument led to a degree of confusion on the part of the RF. It first contended that the CJJA could not be invoked because the service and other requirements under s.31(4) of that act were not met.
However that is to misunderstand the nature of the Claimants’ argument. They are not attempting enforcement or recognition under the Act. What they are doing is invoking the common law doctrine of issue estoppel and the test applicable to that doctrine. However they concede that because that involves, as part of the exercise, a recognition (in substance) of a foreign judgment against a state, the s. 31 test must also be passed. And as the logical correlate they say: if we could have applied for enforcement/recognition , why should issue estoppel not be available?
That, in my judgment, makes perfect sense analytically. The concession as to s. 31 answers the requirement which I have noted above for caution as regards state immunity and foreign judgments, though the general requirement for care in the context of foreign judgments (for example as regards issue identity) remains. It also answers at least one portion of the requirement to give effect to state immunity which is stressed in the judgments to which the Claimants refer.
There is no need for the use of the procedural requirements in s. 31(4) if there is no freestanding action for recognition/enforcement. Their purpose is to ensure that the state is properly notified and served. It would be artificial to regard them as required when the recognition involved is via a common law rule in proceedings which have already been served via the appropriate process.
In response Mr Qureshi KC deployed an alternative argument, that CJJA, and in particular the procedural requirements at s. 31(4), indicates that any reliance on a foreign judgment should be via this process rather than via issue estoppel and that the Claimants should have applied for recognition/enforcement of the foreign judgments. But this is no more attractive as an argument. As I have noted, there is nothing in the SIA 1978 which ousts common law doctrines. There is equally nothing in the CJJA which purports to affect common law processes and doctrines. The Claimants do not want to recognise or enforce the Dutch judgments as an end in itself, so formal recognition or enforcement is not entirely apt. It could of course be done as a precursor to deploying the recognised judgment as part of an issue estoppel argument; but why should it be a requirement?
It follows that I conclude that there is no reason why, if the relevant hurdles are cleared, there cannot be an issue estoppel arising out of a foreign judgment against a state, just as there can be against an ordinary company or individual.
The remaining question on state immunity is therefore whether this proviso is satisfied - whether the RF would not have been immune “if [the Dutch Courts] had applied rules corresponding to [Section 9 of the State Immunity Act]”.
The Claimants say that the RF is caught by this inter alia because it has submitted to the jurisdiction in the Netherlands by challenging the awards in the Dutch courts. The RF initiated proceedings in the Netherlands. In this jurisdiction that would be squarely caught by s. 2(1) SIA 1978 (“A State is not immune as respects proceedings in respect of which it has submitted to the jurisdiction of the courts of [the United Kingdom]”). So if the Dutch courts applied s. 9 mutatis mutandis there would have been no State immunity. That logic appears perfectly sound.
It initially appeared that the RF would then rely on s. 2(4) of the State Immunity Act, on the basis that there is no such submission because the central basis of its presence before the Dutch courts is that it did not agree to arbitrate (and thus maintains its claim to immunity in this regard). S. 2(4) says:
“Subsection (3)(b) above does not apply to intervention or any step taken for the purpose only of— (a)claiming immunity; or (b)asserting an interest in property in circumstances such that the State would have been entitled to immunity if the proceedings had been brought against it.”
However (as swiftly became apparent), s. 2(4) provides no answer because it explicitly goes to s. 3(b) only. This case is a s. 3(a) case and s. 3(a) says: “A State is deemed to have submitted— (a)if it has instituted the proceedings”. The point was therefore not pursued orally and I mention it only for completeness.
I therefore conclude that the Claimants are right, and that subject to identity of issue and finality there is no bar to an issue estoppel being established against the RF. Specifically: the fact that the RF is a state does not preclude the finding of an issue estoppel, if other conditions are satisfied.
In passing in writing and orally Mr Qureshi also sought to argue that issue estoppel cannot arise in the context of public international law. No authority was cited in support of these arguments in writing, it being made on the basis that it was necessary and a logical consequence of issue estoppel being essentially a matter of domestic law. Orally the RF sought to push this argument further, contending that the courts have a freestanding obligation to interpret a treaty and cannot be bound by a foreign court’s decision. Mr Qureshi referred me to GPFGP Sarl v Republic of Poland [2018] EWHC 409 (Comm). However the need for an “independent interpretation” to which he referred me in that case does not assist. That phrase does not mean that every court must take its own view, but rather that all courts must strive to give the “independent meaning derivable from the sources mentioned in arts 31 and 32 and without taking colour from distinctive features of the legal system of any individual contracting state in principle therefore there can only be one true interpretation of a treaty.”
Further as the Claimants noted, this approach to the question ignores the jurisprudential structure of issue estoppel, which is not addressed to the courts, but rather the parties; it reflects the fact that a particular legal battle has already been fought out fully between the parties. The RF chose to dispute jurisdiction, including the construction of Article 45 ECT, in the Netherlands. It has (subject to issues of finality and identity of issue) had a determination, and cannot seek to have another one before a different court.
Identity of Issue
I will deal briefly with the identity of issue point before moving on to the three aspects of the “final and conclusive” controversy.
As to this:
The RF contended that there was a distinction between the issues here and in the Netherlands, submitting that while both proceedings are concerned with whether the RF agreed in writing to submit to arbitration the disputes that are the subject of the Awards, this question is ventilated in materially different ways. It points to the fact that the challenges in the Netherlands are directed to the existence or otherwise of a valid arbitration agreement “by reason of inter alia procedural fraud” while the English Proceedings concern the immunity of the RF under the SIA 1978;
The Claimants for their part submitted that this dispute is all about whether an issue estoppel arises in relation to the Dutch courts’ determination of the Arbitral Jurisdiction Issue and that the identity of issue is apparent inter alia from the RF’s acceptance that its objections to the Tribunal’s jurisdiction have been resolved in the Netherlands.
I conclude that the RF’s arguments lack force. Both proceedings are concerned with that central question of whether there was a valid agreement to arbitrate. If there was, there was jurisdiction and the awards are valid. This has always up until now been the RF’s position. Its solicitor Mr Goldberg swore and served witness statements saying that the Stay should be continued precisely because there was such overlap between the Dutch Proceedings and the English Proceedings. This was understood and recorded to be the case both by Henshaw J (Footnote: 1) and by Butcher J and informed the decisions which they reached.
This can perhaps best be summarised via a fairly lengthy citation from the Butcher judgment:
[34] ….All the points going to the tribunal’s jurisdiction which were raised before the Dutch SC, which involved in particular items (2), (3) and (4) in the above summary, were rejected.
[35] I consider that this is a matter of considerable significance as to whether the stay of the present proceedings should be lifted, at least to the extent of permitting the Defendant’s Jurisdiction Application to be determined. This is particularly so for the following reasons:
(1) The fact that the Defendant’s Jurisdiction Application overlapped with the jurisdictional points which were raised by the Defendant’s cassation appeal to the Dutch SC was a matter heavily relied upon by the Defendant at the hearing in front of Henshaw J last year as a reason for the stay being continued until the Dutch SC had given judgment. Thus in Mr Goldberg’s Third Witness Statement, served on behalf of the Defendant and dated 6 November 2020, there is, at paragraphs 48 – 57, an analysis of the basis for the Defendant’s jurisdiction challenge here, and of how it overlaps with the issues which were then before the Dutch SC.
(2) In Henshaw J’s judgment the extent of the overlap is recognised at paragraphs 45-50. As is said in paragraph 50, the first two of the Defendant’s grounds in its Jurisdiction Application are the same as Grounds 2 and 3 in the cassation appeal. The third ground, as Henshaw J put it, ‘overlapped’ with Ground 5 in the cassation appeal. In fact, as I have set out, when the unappealed decision of The Hague Court of Appeal in relation to Article 21 is also taken into account, it can be seen that the Defendant’s third point in its Jurisdiction Application has been the subject of a determination by the Dutch courts.
(3) An important part of Henshaw J’s reasoning as to why the stay should not be lifted was the concern that, if the English proceedings were allowed to proceed, the Defendant might be unable to present to the English court its full case on state immunity, because it could be said to be bound by the decision of The Hague Court of Appeal, notwithstanding that that decision might itself later be reversed by the Dutch SC (see para. 213(ix)). The position in relation to this has now changed.
[36] It is of course the case that there is an extant challenge to the Awards in the Dutch courts. It is clear, however, that this does not overlap with the points raised by the Defendant’s Jurisdiction Application. It has, indeed, been accepted on behalf of the Defendant that all the jurisdictional points have been resolved in the Dutch proceedings and that the remaining issue does not overlap with the jurisdictional challenge. At the directions hearing before Henshaw J on 1 April 2022, the judge asked Mr Goldberg, who was appearing on behalf of the Defendant, whether it was in dispute that the Dutch SC had ruled against the Defendant on the grounds which it had put forward as going to jurisdiction, and Mr Goldberg confirmed that it was not. In his ruling on that occasion, Henshaw J stated (at paragraph 3) that the Defendant had accepted at the hearing in 2021 that the remaining ground of challenge ‘does not go to the question of jurisdiction’ in the sense of going to whether the Defendant had agreed in writing to submit the dispute to arbitration for the purposes of s. 9 of the 1978 Act. In Mr Goldberg’s Fifth Witness Statement dated 6 May 2022, at paragraph 65, it is recognised to be a fact that the points that remain in the Dutch proceedings do not go to jurisdiction.”
The way the point is now put by the RF, with emphasis on the fraud issue, therefore attempts to row back from a point which was explicitly conceded at an earlier stage. It also (in that focus) attempts to airbrush the very considerable overlap of issues, and the complete identity of some of those issues.
I do of course bear well in mind the need for caution on the subject of identity of issue where issue estoppel is grounded in a decision of a foreign court. However this is not a case where there is uncertainty (as alluded to in Yukos v Rosneft) on the subject of whether an issue was necessary for the decision and whether there has been a decision “on the merits”.
The only question here is as to the identity of issue. Again there is no complication introduced by the procedural aspects, and I am able to be sure that there is identity of issue. In essence, the reality of the situation is that (i) the decision in the Dutch courts is about whether the Tribunal lacked jurisdiction, which is the same issue which is raised here (ii) at the more granular level the Dutch Courts’ decisions have now considered and decided the exact points which are raised by way of jurisdictional challenge here.
The only remaining aspect is that of “final and binding”.
Have The Dutch Judgments Reached Final And Conclusive Determinations?
The next question is whether the Dutch Judgments have reached final and conclusive determinations on whether the RF has agreed in writing to submit to arbitration the disputes that are the subject of the Awards.
The Butcher Judgment relied on the DSC’s determinations on Set Aside Grounds 2, 3 and 5 to lift the Stay, holding that those determinations eliminated the possibility that an adjudication of the Jurisdiction Objections in the English Enforcement Proceedings “will be based on a decision which might itself be subject to reversal in the Dutch proceedings.”
The RF however submits that, upon proper consideration, such a possibility does indeed remain for three reasons. The first is that those decisions lack res judicata effect as a matter of Dutch Law. The second is that those decisions lack res judicata effect because of the remaining live ground of fraud in the arbitration. The third concerns the possibility of a referral to the ECJ.
Expert Evidence
Expert evidence is relevant on all of these issues, and there is a clash of expert evidence, meaning I must prefer the evidence of one of the experts over another.
In deciding which evidence to prefer it was submitted by the Claimants to be a point of significance that they rely upon an expert whose expertise is manifest – he is a professor of Dutch Law, widely published on substantive and procedural law and a substitute justice of two of the Dutch Courts of Appeal (including Amsterdam). By contrast, they say, the expert evidence relied upon by the RF is not independent expert evidence, but the evidence of the RF’s own Dutch lawyer, who is arguing the case in the Netherlands. It is, they say, further afflicted by a number of problems as an expert report including points at which it appears to lack the requisite impartiality and appears to descend into the battle and one clear error in procedural law.
There is force in these submissions. Simply regarded by reference to qualifications Prof. Biemans appears better qualified to act as an expert. He has very considerable qualifications and expertise, both in the academic and judicial fora. He also has more manifest independence. While I would not put the point as high as the Claimants do and I entirely understand that (i) Mr Cornegoor was appointed to represent the RF under the Dutch Counsel Act rather than seeking the retainer in any way (ii) he was selected for his experience in relation to arbitral awards under international investment treaties and (iii) Mr Cornegoor does say that he recognises his duty to be independent, this does not entirely deal with the points made.
While it is certainly right that retained foreign lawyers are not excluded from giving expert evidence in this court (as recognised by the Commercial Court Guide) the fact of being retained to fight certain arguments cannot help but have an impact on perceived neutrality. Active involvement in litigation makes it more difficult for a lawyer to meet the neutrality required of an independent expert. Where two retained lawyers are opposed against each other in providing expert evidence (as often happens when informal expert evidence is relied upon in interlocutory disputes), both sides are on the same footing. Here that is not the case, and this adds another element to the balance in favour of Prof Biemans’ evidence.
To the extent that the decision turns upon the quality of the expert evidence the evidence of Prof Biemans therefore has the better pedigree and claim to be given the preference, if the evidence proves to be such that both experts offer arguments which can on their face be given consideration.
In the event however, for reasons which will become apparent, I conclude Prof Biemans’ evidence is, even putting these points aside, the more compelling.
Question 1: Are the Dutch Courts’ determinations Res Judicata as a matter of Dutch Law?
The RF submits that determinations of the Dutch Judgments on Set Aside Grounds 2, 3 and 5 lack res judicata effect as a matter of Dutch law. The more precise way of putting this is that the judgments are not kracht van gewijsde (i.e. “have binding force in another dispute between the same parties”) and the decisions in the judgments do not have gezag van gewijsde.
In writing the RF submitted that:
It is a requirement of Dutch law that only judicial determinations which contribute to the operational part – or “dictum” – of a judgment are capable of having res judicata effect and submit that both the experts agree that the dictum of a Dutch law judgment refers to the part containing the “final decision” of the court. The RF says that neither of the Dutch Judgments contain any currently binding dictum in that regard, in that the dictum did not mention Set Aside Grounds 2, 3 and 5 but simply quashes the Hague CoA Judgments. This means that the Amsterdam Court of Appeal (which is apparently very independent minded) can start afresh;
It is premature for the determinations by the Hague CoA on Set Aside Grounds 2, 3 and 5 to have any res judicata effect in that only decisions which are not subject to “ordinary remedies” (e.g., an appeal) are capable of having res judicata effect. It was said that the Hague CoA Judgments do not satisfy this requirement;
The “relevant operative part of the [HCA] Judgment no longer exists as a matter of [Dutch] law”, and thus the Hague Court of Appeal’s reasoning which contributed to that part of the dictum cannot, by itself, have res judicata effect on the Amsterdam Court of Appeal;
Even if the Amsterdam Court of Appeal adopts the Hague Court of Appeal’s reasoning in dismissing Set Aside Grounds 2, 3 and 5, it would not “revive” the res judicata effect of the previous Hague CoA Judgments. Instead, the Amsterdam Court of Appeal’s own reasoning would potentially be capable of having res judicata effect (provided, of course, that its reasoning contributes to the dictum of the judgment).
In the event none of these arguments were really pursued orally before me by Mr Qureshi. However I should record that to the extent that they were maintained based on the written arguments, I have had no difficulty in rejecting them.
One might perhaps start with a “sense check”. The effect of the RF’s argument is that although the challenges to all but one of the Hague CoA's rulings have been dismissed, the questions effectively required to be determined de novo. It is possible that this is correct, but it does present as a surprising proposition.
Further there was common ground between the experts which would effectively align with the Claimants’ contentions and with this sense check.
There is a route to this conclusion by virtue of DCCP, Article 424. This applies where the DSC refers a case back to a Court of Appeal. It requires the referral court to deal with the case “in a manner consistent with/with due regard to the [DSC] Judgment”. In this case, the DSC Judgment has dismissed the Defendant’s appeal against The Hague CoA Judgment in relation to all issues going to the Tribunal’s jurisdiction. The Amsterdam CoA must accordingly deal with the one outstanding issue that has been referred to it consistently with that DSC Judgment.
Second, Mr Cornegoor refers to the “Partial Effects Rule”. He describes this as:
“a rule which was developed and confirmed numerous times by the DSC, that even though the previous judgment was annulled by the DSC, those parts of the annulled judgment remain binding which were not the subject of the cassation appeal or were the subject of cassation grievances which were rejected by the DSC”
This common ground, which indicates that even leaving aside the formal res judicata equivalent in Dutch Law, the judgments are final and conclusive, was clearly pointed out in the Claimant’s skeleton and again in detailed oral argument. It was not addressed by the RF and must therefore be taken to be common ground. To the extent it was formally still in issue I unhesitatingly accept the Claimants argument here.
Thus although there might be a dispute about the technical status of the relevant judgments, the answer for all practical purposes is not contentious. The reality is that subject to (i) the question of a potential CJEU referral and (ii) the effect of the issue of fraud in the arbitration the Hague CoA Judgment on the Arbitral Jurisdiction Issue is final and conclusive under Dutch domestic law. The route by which the preclusion arises is irrelevant. The Good Challenger test is met.
To the extent that the technical point on Article 236 and “kracht van gewijsde” and “gezag van gewijsde” is relevant I would in any event prefer the Claimants’ analysis. The Claimants relied on a multi-layered analysis set out in detail in their skeleton argument, and explained further in oral argument.
These sources cover the following points which I need not rehearse in detail:
There is a substantial body of Dutch Law which proceeds on the basis that a DSC ruling does not completely annul a Court of Appeal ruling but leaves the unchallenged parts intact;
Both the DSC itself and Advocate General Vlas have used the phrase “kracht van gewijsde” to describe the effect of annulled Court of Appeal judgments whose reasoning has been partially upheld by the DSC;
The DSC’s Press Release for this case describes the effect of the DSC judgment as being that the judgment of the Hague CoA is final.
Again, none of this was challenged. But of particular interest are:
The s-Hertogenbosch Court of Appeal has recently considered the continuing effect of an (annulled) judgment of the Arnhem-Leeuwarden Court of Appeal. It concluded that:
“The Court of Appeal first of all states that it is bound as referral court to the final decisions in the annulled judgment that were not contested (or that were not successfully contested) in cassation. These have acquired [kracht van gewijsde] and therefore cannot be contested again.”
The view of Advocate General Vlas on the context of an appeal related to this case where this exact point has been taken before the DSC. He states:
“The part [of the RF’s submissions] argues, in essence, that with the Supreme Court's setting aside of the court's judgments of 25 September 2018 and 18 February 2020, the dictum of the [DC Court's] judgment of 20 April 2016 has been revived and with it the annulment of the arbitral awards pronounced therein. That argument is based on an erroneous view of the law on the partial working of the cassation appeal. It follows from paragraphs 5.1.3-5.1.19 of the Supreme Court's judgment of 5 November 2021 that the setting aside of [The Hague CoA’s] judgments was based solely on the success of the complaint that the court erred in ruling that the Russian Federation could only raise its allegations of fraud in revocation proceedings and could not base its claim for setting aside. The remaining complaints could not lead to cassation (para 7.1). The partial effect of the cassation appeal then entails that the annulment pronounced by the Supreme Court is limited to the judgment of the court of appeal that was successfully challenged in cassation. The fact that such a limitation does not follow from the dictum itself does not alter this. The other judgments of the court of appeal have become unassailable. This therefore also applies to the court of appeal's judgment that the arbitral tribunal had jurisdiction to hear and decide HVY's claims, so that [The Hague DC Court’s] judgment that no valid arbitration agreement had been concluded and the arbitral awards should be set aside for that reason cannot stand.”
This echoes relevant case law of the Dutch Courts which establish that DCCP, Article 236 applies to (a) “decisions”; (b) “concerning the legal relationship in dispute”; (c) which are contained in a judgment which is no longer “subject to ordinary legal remedies”.
Any suggestion that the Hague CoA’s determinations were not “decisions” within the meaning of Article 236, seems doomed to failure in circumstances where they were essential to both the Hague CoA Judgment rejecting the RF’s setting-aside grounds and annulling the decision of the Hague DC. There is no dispute that the Hague’s CoA’s determinations in relation to the Arbitral Jurisdiction Issue “concerned the legal relationship in dispute”.
There was some debate about whether the Hague CoA Judgment is still “subject to ordinary legal remedies”. I conclude that it is not. The legal remedy against that judgment was an appeal to the DSC. That appeal has concluded, and the DSC has upheld the Hague CoA Judgment in part, overturned it in part, and made a reference to the Amsterdam CoA which (as explained above) is consistent only with a final and conclusive finding that the Tribunal did have jurisdiction to issue the Awards. Common sense would suggest that after a Supreme Court appeal there are no “ordinary legal remedies” remaining. That is consistent with the Dutch authorities which make clear that, to the extent that it has been upheld by the DSC, the Hague CoA Judgment is “kracht van gewijsde” for the purposes of Article 236.
Question 2: Does the question of fraud in the arbitration affect finality?
The question here is whether the rulings of the Amsterdam Court of Appeal on the question of the alleged fraud in the arbitration (which has been referred back to them and remains to be determined) are going to make any difference to the jurisdictional rulings.
Here the Defendant says that “even if the English Court is of the view that Set Aside Grounds 2, 3 and 5 have been finally and conclusively determined such Set Aside Grounds cannot be considered finally and conclusively determined until a final and conclusive determination has been reached on Set Aside Ground 1 because Set Aside Ground 1 relies on facts and evidence that pertain to, and are potentially determinative of, crucial aspects of Set Aside Grounds 2, 3 and 5.”
There is, as Mr Crow KC pointed out in argument, a short route through this point. He says that it is a non-sequitur to say that “if the English Court is of the view that Set Aside Grounds 2, 3 and 5 have been finally and conclusively determined …. such Set Aside Grounds cannot be considered finally and conclusively determined until a final and conclusive determination has been reached on Set Aside Ground 1.” Alternatively, one might say that the argument was internally inconsistent or illogical. If the grounds have been finally determined, they have been finally determined. This effectively deals with the point.
If one needed to go further there is another short point: fraud in the arbitration is not a jurisdictional objection; so there is no question of the issue going straightforwardly to jurisdiction. If the points have been determined and if Ground 1 does not go to jurisdiction it seems impossible that it could be the case that the content of arguments on Ground 1 (fraud) will or may affect the other (jurisdictional) grounds.
The Claimants nonetheless addressed some detailed argument to this point, with Mr Crow outlining the nature of the arguments and what was involved in each one. Only the investor/investment argument was ultimately suggested by the RF to be potentially affected by the determination of the fraud argument – the reason being that the fraud argument is based on an allegation that the claimants withheld evidence and presented false evidence as to who controlled (as opposed to owned) them. It is clear and was tacitly accepted that the question of whether on the true construction of Article 45 the RF provisionally agreed to arbitration or whether measures which were ostensibly tax measures were in fact expropriatory have absolutely no point of connection with this issue.
As to the question of potential overlap between the fraud issue and investor/investment, the latter argument has been advanced under five headings which are summarised in the Hague CoA judgment from paragraph 5.1.3: (i) that the Claimants are shell companies owned by Russian nationals and the ECT does not provide cover in relation to disputes between a state and its nationals (ii) the Claimants were not investors because their holdings in Yukos represented a U- turn ownership structure (iii) the claimants had not made any foreign economic contribution to the RF (iv) Russian nationals behind the Claimants abused the Russian tax system so that the corporate veil should be pierced (v) the ECT does not protect the claimants because of criminal and unlawful acts on the parts of individuals behind them.
It is therefore apparent that there is potential scope for an overlap with the issue about fraud in the arbitration based on control. However the Hague CoA’s judgment did not deal with any of the facts on control. It dealt with the matter entirely by reference to questions of construction, without trespassing on controversial matters, even as regards the latter issues. Thus the grounds (iv) and (v) were dealt with as follows:
“Article 1(6) ECT provides that investment means every kind of asset that is owned or controlled by an investor. It is established that the Yukos shares are owned by HVY. There is therefore no need to establish who controls the shares. Therefore, the 'Understanding' invoked by the Russian Federation in relation to the control criterion is not relevant here. ….
In short, the Court of Appeal agrees with HVY that there is no general principle of law according to which investment treaties do not provide protection to companies wholly controlled by nationals of the host country.
Nor, in the view of the Court of Appeal, does this case support the statement of the Russian Federation that there is an international principle of law that the corporate veil should be pierced because the legal form has been abused for fraud…. In the view of the Court of Appeal, Article 1(7) ECT does not provide a basis for the application of rules of national law relating to piercing the corporate veil.
On the basis of the foregoing, the Court of Appeal is of the view that the Russian Federation has not sufficiently demonstrated that there is a generally accepted principle of law which implies that an arbitral tribunal must (always) decline jurisdiction where it concerns the making of an 'illegal' investment. As stated above, Article 1 (6) ECT does not contain a requirement of legality; it does not require that an investment must have been made in accordance with the law of the host state. Nor does the text of the ECT contain any restrictions on access to arbitration as referred to in Article 26 ECT. The Court of Appeal considers that in this case the ordinary meaning of the wording of Article 1(7) ECT prevails. As a result, the Tribunal does not lack jurisdiction if it is shown that there was 'illegal conduct' at the time of, or in making, the investment.”
Accordingly the conclusions reached were reached on the basis that the factual underpinnings were made out (i.e. on the basis that it was assumed that there was unlawfulness on the part of the Russian individuals and assuming that they did control the Claimants). Accordingly not only have the issues been determined, there is analytically no basis for those conclusions to be affected by the determination of the fraud in the arbitration point. The DSC judgment approved the Hague CoA’s approach at para 5.3.11 of its judgment.
That conclusion is reinforced by the fact that the materials which relate to fraud were in fact in play in relation to the jurisdictional debate in the sense that they were before the Hague Court of Appeal. So this is not a case where pursuing the fraud argument will open up whole new vistas of disclosure amongst which might lurk some document pertaining to the investor/investment question.
Question 3: Do determinations on Set Aside Ground 2,3 and 5 remain subject to a potential referral to the CJEU?
The RF contends that any determinations in the Dutch Judgments on Set Aside Grounds 2, 3 and 5 remain subject to a potential – and in the words of its expert “significantly probable” – referral by the Amsterdam Court of Appeal to the CJEU for rulings on the interpretation of the ECT. This turns on whether there is a principle of EU law based on the decision of the CJEU in Elchinov v. Natsionalna Zdravnoosiguritelna Kasa [2011] 1 CMLR 29 which would allow the Amsterdam CoA to take a different course to that indicated by the DSC, which has refused to make a reference to the CJEU in respect of any of the issues of interpretation of the ECT which arose before the DSC in relation to the Arbitral Jurisdiction Issue.
On 17 May 2022, the RF requested that the Amsterdam Court of Appeal “request the CJEU for a preliminary ruling on all questions of interpretation relating to the various provisions of the ECT (i.e. Article 1(6) and (7), Article 21(5), Article 26 and Article 45)” In particular, the Defendant raised eight interpretation questions (“the Interpretation Questions”).
The Defendant says that all of the Interpretation Questions relate to crucial aspects of Set Aside Grounds 2, 3 and 5. For example it says that:
Interpretation Questions 1 and 2 relate to the interpretation of Article 45 ECT, and thus have the potential to be determinative of the final outcome of Set Aside Ground 2 concerning the ECT’s provisional application to the Defendant;
Interpretation Questions 3–7 relate to the interpretation of Articles 1(6) and 1(7) ECT, and thus have the potential to be determinative of the final outcome of Set Aside Ground 3 concerning the Claimants and their investments failing to qualify as “investors” and “investments”, respectively, under the ECT; and
Interpretation Question 8 relates to the interpretation of Article 21(5) ECT, and thus has the potential to be determinative of the final outcome of Set Aside Ground 5 concerning the Tribunal having exceeded its mandate by considering the Claimants’ claims concerning “taxation measures”.
The RF says that an answer to these questions divergent from the interpretation of the ECT in the DSC Judgment could realistically cause the Amsterdam CoA to annul the awards on the ground that the dispute was not subject to arbitration.
The merits of these questions are however neither here nor there for present purposes. What matters is whether the principles set out in Elchinov indicate that there is a real possibility of a CJEU referral from the Amsterdam CoA.
Mr Cornegoor explains that the principle operates so as to:
“[. . .] preclude[] a national court which is called upon to decide a case referred back to it by a higher court hearing an appeal from being bound, in accordance with national procedural law, by legal rulings of the higher court, if it considers, having regard to the interpretation which it has sought from the [CJEU], that those rulings are inconsistent with European Union law.”
The RF’s case is that;
The Claimants cannot assert that it is impossible for the Amsterdam Court of Appeal to refer the Interpretation Questions to the CJEU;
The Amsterdam Court of Appeal is not bound by the Dutch Judgments’ determinations on Set Aside Grounds 2, 3 and 5;
Even if a rule of Dutch procedural law were to limit the scope of the Amsterdam Court of Appeal’s review of Set Aside Grounds 2, 3 and 5 any such rule is “precisely the kind of rule which is superseded by the Elchinov Exception, namely ‘a rule of national law, pursuant to which legal rulings of a higher court bind another national court’”;
An EU national court that is the highest court in its national hierarchy is required to refer questions concerning the interpretation of EU law (including the interpretation of the ECT, which forms part of EU law) to the CJEU, except where the court is “convinced that the matter is equally obvious to the courts of the other Member States and to the [CJEU]” (i.e., acte clair). As the First Cornegoor Report states, this threshold “is a high one” for relieving EU national courts of their obligation to make a referral to the CJEU;
There is a “significant probability” that such a referral will be made to the CJEU for a preliminary ruling on the Interpretation Questions because the ECT Articles that are relevant for present purposes have been subject to inconsistent interpretations by the Dutch courts themselves, as well as by the courts of France and Luxembourg.
Discussion
TFEU, Article 267, outlines the principles for a referral and states inter alia that:
“Where such a question [i.e. a question of EU Law] is raised before any court or Tribunal of a Member State, that court may, if it considers that a decision on the question is necessary to enable it to give judgment, request the Court to give a ruling thereon.”
It follows that in order for the domestic court to have any discretion to make a reference, it must therefore be satisfied that a question of EU law has been raised which that court considers must be resolved in order to enable it to give judgment. This point was not disputed by the Defendant.
Thus the discretion is only capable of arising in relation to issues of which the domestic court is properly seized and which the domestic court must resolve in order to give judgment in the proceedings before it. Like the jurisdiction to grant declaratory relief in this jurisdiction, there must be a live point in the dispute which renders it utile to undertake the exercise.
While at [32] of its Judgment, the CJEU observed that:
“European Union Law precludes a national court which is called upon to decide a case referred back to it by a higher court hearing an appeal being bound, in accordance with national procedural law, by legal rulings of the higher court, if it considers, having regard to the interpretation which it has sought from the Court, that those rulings are inconsistent with European Union Law”.
Neither Article 267, nor Elchinov nor any authority which addresses that provision, suggests that the court can or should override domestic rules which identify whether a court is properly seized of an issue at all.
This can be seen from the decision in Elchinov itself. At [31] of the CJEU’s judgment (cited by Mr Cornegoor in paragraph 34 of his report), it refers to “settled case law” to the effect that the relevant obligation is imposed upon “a national court which is called upon, within the exercise of its jurisdiction, to apply provisions of European Union law”. Similarly [24] refers to the question of whether “European Union law precludes a national court which is called upon to decide a case referred back to it by a higher court hearing an appeal from being bound…”
Prof. Biemans’ analysis on this point is careful and thorough. He refers to multiple Supreme Court decisions which indicate that the Elchinov approach is by no means adopted. Rather they disclose that under Dutch law, issues which have been finally resolved by the DSC fall “outside the legal battle” between the parties for the purposes of any subsequent debate before a lower court to which the case is referred after cassation.
Mr Cornegoor’s analysis is superficial and optimistic. He does not provide authorities which are consistent with his position, but rather asserts that “there can be no reasonable doubt that [Elchinov] applies both to Article 424 and the Partial Effects Rule”. He also does not grapple with (i) why this approach should be read as applying broadly (ii) why, if it does there is no evidence of it or (iii) why the apparently principled distinction between domestic rules which require a lower court to follow the decision of a higher court on a particular issue, and those which determine whether the lower court has jurisdiction to entertain that particular issue in the first place is not a distinction of juridical moment. His argument that the Amsterdam CoA is not bound by the determinations on Set Aside Grounds 2, 3 and 5 is in tension with his acceptance of the effect of Article 424 and the Partial Effects Rule.
Accordingly on this point also I prefer Prof Biemans’ analysis. I note that the RF’s previously instructed expert, Mr Van den Berg, although instructed to produce evidence in the context of an application to lift the stay did not suggest that there was a real prospect of a reference.
It follows that I conclude that Elchinov is therefore directed at a narrower target than the RF would suggest. It only arises where the national court has to decide something which involves interpretation and application of EU law. That is not a question at large, or to which a positive answer can be assumed. Whether and to what extent a domestic court has been called upon to address a matter of EU law (and therefore has jurisdiction to address that question) can only be determined by reference to the domestic rules allocating jurisdiction to determine issues and/or claims to a particular court. Contrary to Mr Cornegoor’s approach Elchinov does not come first and drive the conclusion. Elchinov only arises once the question as to what substantive issues are properly before the relevant domestic court are determined, and the question is asked and answered as to whether those issues involve the interpretation and application of EU law.
The correct analysis is that the Amsterdam CoA has jurisdiction to set the Awards aside, but only by reference to those arguments concerning procedural fraud (arising under DCCP, Article 1065(1)(e)) which are properly before it in light of the DSC Judgment. Having been finally resolved under Dutch law by the Dutch Judgments, the Arbitral Jurisdiction Issue is not one which is “necessary” for the Amsterdam CoA to resolve “to enable it to give judgment”. It is not, therefore, an issue in relation to which the Amsterdam CoA has any power to make a reference to the CJEU under TFEU, Article 267.
That is consistent with the purpose of the principle identified in Elchinov – namely, to ensure that domestic courts which are required to deal with issues of EU law retain the ability to make a reference to the CJEU. The approach advocated by the RF, namely to apply Elchinov regardless of domestic rules (including those identifying the issues of which a domestic court is properly seized) would be a recipe for chaos.
Conclusion
For the reasons given I conclude that the conditions required to be satisfied for the finding of an issue estoppel are met.
One final question remains. That is the question of special circumstances. There are cases in which the court will say that despite the meeting of the relevant conditions it would not be appropriate to uphold a plea of issue estoppel. While this was not a point explicitly ventilated for the RF it did appear to me that the nature of those points and the combination in which they were deployed tended to suggest that this was an argument which the RF was at least tacitly pursuing.
The point can be conveniently introduced by reference to The Good Challenger where Clarke LJ at [79] notes that “the correct approach is to apply the principles …. unless there are special circumstances such that it would be unjust to do so.”
There has been very little qualification or elaboration to this short statement of principle. The RF drew my attention to Yukos v Rosneft (No 2) [2014] QB 458 at [158] and following:
“159 The trouble with the “discretionary in special circumstances” exception is that it is so amorphous. In Arnold v National Westminster Bank plc [1991] 2 AC 93 the exercise of discretion depended on further material becoming available since the original decision. That is not the position in this case.
160 Nevertheless, if we had decided that there was an issue estoppel in this case on the basis that in truth the issue in the Dutch proceedings was the same as the issue in these English proceedings, we would be inclined to invoke the exception for ... It must ultimately be for the English court to decide whether the recognition of a foreign judgment should be withheld on the grounds that that foreign judgment is a partial and dependent judgment in favour of the state where it was pronounced. That is a question so central to the respect and comity normally due from one court to another that to accept the decision of a court of a third country on the matter would be an abdication of responsibility on the part of the English court. On matters of this kind, we should accept our own responsibilities just as we would expect courts of other countries to accept theirs.”
While the submission was made that this is an analogous case, that submission is not well founded. The Yukos case was one where the decision was about the content of public policy, and hence identity of issue was not established.
The fullest consideration of the exception is in Arnold v National Westminster Bank plc [1991] 2 AC 93, where the House of Lords was faced with a combination of newly discovered facts and an eccentric decision from which there had been no right of appeal. The exception is controversial, having been doubted in Australia (O’Toole v Charles David Pty Ltd (1991) 171 CLR 232, 258), and (as Spencer Bower and Handley Res Judicata (5th ed 2019) notes at 8.32, applied inconsistently in other jurisdictions). It has not been much used in this jurisdiction, and the fact that the words “special” and “exceptional” are used in this context indicates the soundness of Handley’s proposition that “The exception should be kept within narrow limits to avoid undermining the general rule and provoking increased litigation and uncertainty.”
In my judgment the present case is not analogous to any of the cases where the exception has been successfully invoked, and nor are the circumstances of the same order. The presence of a novel point (issue estoppel based on a foreign judgment against a State) and the involvement of a multilateral treaty are not enough. Each of those elements can be grappled with and a clear path discerned. Nor is there any real basis upon which it could be said that the recognition of an issue estoppel would work injustice (the fundamental underpinning of the exception). The continuing existence of the fraud ground does not go to jurisdiction and can (as the Claimants urged me to do) be dealt with via case management of the enforcement proceedings. The controversy to which allusion was made as to the correct construction of Article 45 is one which is live “at large” and not in this case; it therefore works no injustice. I therefore conclude that the “special circumstances” exception does not apply.
Accordingly, the answer to Issue 1 is that the RF is, by reason the judgments of the Dutch courts, precluded from re-arguing the question of whether it has agreed in writing to submit to arbitration the disputes that are subject of the Awards.
That leaves the question of whether the Jurisdiction Application should be dismissed forthwith.
As to this, one circles back to the question of State Immunity. Although the RF contended that there is a freestanding duty under the SIA to decide state immunity, as discussed above, it follows from the approach taken to issue estoppel that the present case falls within the exception to State immunity under s. 9 of the SIA 1978. Although the RF disputed the appropriateness of the determination of the jurisdiction application, it did not address the question of how it could be that if the questions were answered as I have answered them, the answer did not follow that s. 9 was engaged. The question it says must be posed is whether the RF has agreed in writing to submit to arbitration the disputes that are the subject of the arbitrations which resulted in the Awards. That is answered by the outcome of the Dutch proceedings and is a question as to which I have determined that an issue estoppel arises. That means that the RF’s assertion of immunity under s. 1 of the SIA 1978, falls to be rejected.
The answer in relation to Issue 2 is therefore “yes”: the Jurisdiction Application ought to be dismissed forthwith.