JSC Commercial Bank Privatbank v Kolomoisky & Ors

This is the judgment of the Court to which each member of the Court has contributed.

PJSC Commercial Bank Privatbank (the Bank) appeals against the order of Fancourt J dated 4 December 2018 whereby he declared that the court had no jurisdiction to try the claim brought by the Bank in these proceedings against the first, second and sixth to eighth defendants, stayed the proceedings against the third to fifth defendants and set aside service of the claim form and particulars of claim on the sixth to eighth defendants, and (subject to appeal) discharged a worldwide freezing order earlier granted against all the defendants restraining each of them from disposing of assets with a value of up to US$2.6 billion.

The appeal raises a number of issues. The first is whether article 6(1) of the Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters 2007 (the Lugano Convention) is subject to a requirement that a claim brought against a defendant in the courts of that defendant’s domicile must not be brought for the sole object of joining a defendant domiciled in another Convention state and, if so, whether the present proceedings were brought against the third to fifth defendants with the sole object of joining the first and second defendants. The second is whether, assuming the answer to one of the questions under the first issue is in the negative, the court has jurisdiction to stay proceedings brought against defendants in accordance with the Lugano Convention and the Recast Brussels Regulation on grounds of lis alibi pendens in favour of proceedings in a non-Convention state (a third state) and, if so, whether the judge was wrong to hold that, if it had arisen, he would have exercised his discretion to stay the proceedings. The third is whether the judge was wrong to stay the proceedings against the sixth to eighth defendants on grounds of forum non conveniens and to set aside service of the proceedings on them. This largely turns on the first and second issues.

The judge decided all these issues against the Bank and gave permission to appeal on those issues. We refer to them respectively as Grounds 1 to 3.

The Bank seeks permission to appeal on three further issues. They arise only if the judge was wrong on all or some of Grounds 1 to 3. The further issues concern (i) the quantum of the worldwide freezing order, (ii) whether there was material nondisclosure of relevant matters to the court on the initial, without notice, application for the worldwide freezing order and, if so, whether the judge was wrong to hold that he would on this ground have discharged the order and refused to re-grant it and (iii) whether the judge failed to give sufficient reasons for his decisions on certain points. The judge refused permission to appeal on these issues, which we refer to respectively as Grounds 4 to 6. On the Bank’s renewed application to this Court for permission to appeal, Patten LJ adjourned consideration of the application to the hearing of the appeal of Grounds 1 to 3. Lord Pannick QC, on behalf of the Bank, told us that the points arising under Ground 6 were sufficiently dealt with under Grounds 4 and 5 and

We are agreed that the Bank’s appeal should be allowed. We are also agreed on all the issues raised by the appeal, save in one respect. On the issue of principle arising under Ground 1 (see paragraph [3] above), the majority (David Richards and Flaux LJJ) are agreed that article 6(1) is not subject to a sole object condition, for the reasons set out in this judgment. Newey LJ takes a different view, which he sets out in a separate judgment. This difference has no effect on the outcome of the appeal, because we are all agreed that the judge’s finding that the present proceedings were brought with the sole object of joining the first and second defendants to these English proceedings cannot stand.

The Bank is incorporated under the laws of Ukraine and has its head office and principal operations in Ukraine. It was founded in 1992 by, among others, the first and second defendants, Igor Kolomoisky and Gennadiy Bogolyubov. They became, if they were not already, the majority shareholders, with direct and indirect holdings varying between 80% and almost 100% between 2006 and 2016. They sat on the Bank’s supervisory board until 2016 and it is the Bank’s case, which for present purposes is not denied, that as regards the key decisions relevant to these proceedings they controlled the Bank. The Bank grew to become one of Ukraine’s largest banks. By 2016, it had 30 regional offices and 2,445 high street branches in Ukraine, and a branch in Cyprus. It provided services to more than 20 million customers, just under half the population of Ukraine.

The Bank was nationalised in December 2016, following a declaration that it was insolvent by the National Bank of Ukraine. Mr Kolomoisky and Mr Bogolyubov, and the other members of the supervisory board, were dismissed and replaced by a new management team.

It is common ground that, for the purposes of the Lugano Convention, Mr Kolomoisky and Mr Bogolyubov are, and were at the date of the commencement of these proceedings, domiciledin Switzerland.

The third to fifth defendants are companies incorporated in England (the English Defendants). The sixth to eight defendants are companies incorporated in the British Virgin Islands (the BVI Defendants). All the defendants accept, for the purposes of the applications before the judge and this appeal, that there is a good arguable case that the English and BVI Defendants were at all material times owned and/or controlled by Mr Kolomoisky and Mr Bogolyubov.

On 19 December 2017, at the conclusion of a full day’s hearing and after a day’s prereading, Nugee J granted a worldwide freezing order (the WFO) against the defendants for up to US$2.6 billion on a without notice application. The application was supported by a very substantial affidavit (Lewis (1)) sworn by Richard Lewis, a partner in the Bank’s solicitors.

The claim form was issued on 21 December 2017and subsequently served, together with particulars of claim, on the English Defendants and the BVI Defendants.

The WFO was continued, with amendments not material to this appeal, without prejudice to the right of the defendants to apply to the court to challenge the court’s jurisdiction and to vary or discharge the WFO. The defendants issued such applications, which were heard by Fancourt J over five days in July 2018 with subsequent written submissions and some further evidence, leading to the order under appeal.

It will be necessary to look in detail at the claim as formulated by the Bank in its particulars of claim and in the evidence filed on the applications below.

In general terms, the Bank alleges that Mr Kolomoisky and Mr Bogolyubov orchestrated the fraudulent misappropriation of over US$1.9 billion from the Bank. The precise amount is alleged to be US$1,911,877,385, but for convenience we will refer to it as US$1.9 billion. The Bank believes that the total amount misappropriated by or at the behest of Mr Kolomoisky and Mr Bogolyubov is likely to run to many billions of US dollars, but the claim in the present proceedings is confined to US$1.9 billion.

The Bank alleges that this misappropriation was achieved through loans by the Bank (the Relevant Loans) to some 46companies (the Borrowers), all incorporated in Ukraine and controlled by Mr Kolomoisky and Mr Bogolyubov. The Relevant Loans were made over a period of 17 months between April 2013 and August 2014 in either US dollars or Ukrainian Hryvnias. For convenience, we will refer to all payments in US dollars. The terms of each loan provided that it was advanced for the purpose of financing the Borrower’s “current activities” and that it would be secured by a “Pledge Agreement”.

The Borrowers entered into supply agreements with some 35companies (the Suppliers), all of them incorporatedoutside Ukraine and most of them in offshore jurisdictions. The Bank alleges that the Suppliers were controlled by Mr Kolomoisky and Mr Bogolyubov and that the supply agreements were bogus. They were for the supply of wholly unrealistic quantities of commodities and industrial equipment and were never intended to be performed. Their terms were uncommercial, and in particular provided for the pre-payment of the entire purchase price before the time for delivery of the commodities or equipment in question.

Until May 2014, the pre-payments were re-cycled between the Borrowers and the Suppliers. Funds were lent by the Bank to the Borrowers in Ukraine. The Borrowers transferred the funds to accounts with the Bank’s branch in Cyprus and then to accounts of the Suppliers with the same branch. In order to comply with Ukraine’s exchange control regulations, the funds were re-transferred to the credit of the Borrower in Ukraine before being used again for the apparent pre-payment of sums due under further supply agreements.

The Bank alleges that pre-payments totalling US$1.9 billion made under supply agreements dated between May and August 2014 (the Relevant Supply Agreements)

The rights to receive goods under the supply agreements were pledged as security for the Bank’s loans to the Borrowers. It is the Bank’s case that, by virtue of their uncommercial terms and because in truth they were sham agreements, they were without value as security. Further purported supply agreements bearing dates between December 2013 and October 2015, which provided for payment to be made after the delivery of the goods under the agreements, were apparently made by at least 36 of the Borrowers (Loan File Supply Agreements). Apart from the payment terms, the Bank alleges that in all other respects these agreements were as uncommercial as the other supply agreements and that they too were shams. It alleges that the only purpose of the Loan File Supply Agreements was to provide security which would have a more plausible appearance, in order to mislead auditors and regulators.

The defendants, including Mr Kolomoisky and Mr Bogolyubov, accept, for the purposes of this appeal, that there is a good arguable case that the Bank lost approximately US$515 million through these transactions and that they were orchestrated by Mr Kolomoisky and Mr Bogolyubov, using the Borrowers and Suppliers in the manner generally alleged by the Bank. Mr Kolomoisky and Mr Bogolyubov have not themselves to date proffered any explanation for the transactions in question or sought to explain their commercial rationale, if any.

The judge observed in his judgment at [25] that there was no difficulty with the Bank proving a good arguable case of a fraudulent scheme. The evidence was “strongly indicative of an elaborate fraud perpetrated by someone, allied to an attempt to conceal from any auditor or regulator the existence of bad debts on the Bank’s books, and money laundering on a vast scale”. The Borrowers had no commercial track record or any substantial assets. The documentary evidence clearly demonstrated that the supply agreements were shams, and “were used as a deceptive basis on which to justify very large sums of money flowing out of the Bank”. The artificial complexity of the re-cycling of funds was “itself indicative of a fraudulent scheme”. At [104], the judge noted that Mr Kolomoisky and Mr Bogolyubov had admitted “a good arguable case of fraud on an epic scale”.

As will become apparent, the applications of the defendants and their submissions before the judge and before us are based to a significant extent on a close analysis of the Bank’s claim made in these proceedings. This is particularly true in relation to the issues of the quantum of the freezing order and of non-disclosure, but it is relevant also to the application of the “sole object” test, if one exists as a matter of law under article 6 of the Lugano Convention.

It is therefore convenient at this stage to give some indication of the competing positions of the parties on the nature of the case advanced by the Bank in these proceedings.

This centres on the role of the English and BVI Defendants. The defendants submit that the Bank’s case hinges on the funds advanced by the Bank under the Relevant Loans having been paid to those Defendants without any subsequent repayment to the

However, they submit that the amount arguably derived from the Relevant Loans was not US$1.9 billion but was, at most, US$514million. Moreover, evidence which they adduced, and which the Bank accepts, shows that all amounts received by the English and BVI Defendants were immediately on receipt paid on to other companies with accounts at the Cyprus branch of the Bank.

The defendants submit that the Bank’s pleaded case, and the case that it put forward to the court on the without noticeapplication for the freezing order, is restricted to a claim based on the English and BVI Defendants having received a net amount of US$1.9 billion derived specifically from the Relevant Loans. In this way, the Bank presented the English and BVI Defendants as central to its claim, because they had knowingly received the very funds said to have been misappropriated by means of the Relevant Loans.

The Bank asserts that this is indeed part of its case. However, there is a fundamental difference between the parties as to how it is to be determined whether the English and BVI Defendants received funds derived from the Relevant Loans. The Bank relies on its pleaded case of Ukrainian law which, it submits, does not require the sort of tracing exercise on which the defendants’ case is based.

The Bank further asserts that its claim against the English and BVI Defendants is also based on the Ukrainian equivalent of the tort of conspiracy. It alleges that those defendants assisted in the misappropriation of US$1.9 billion, through their participation in the Relevant Supply Agreements and associated sham transactions, and that it can establish their liability for that amount without any need to show that they received funds derived from the Relevant Loans themselves.

We look at this issue in detail later in this judgment but at this stage we do no more than state our conclusions. We consider that the Bank’s pleaded case includes a claim based on assistance as well as a claim based on receipt of funds derived from the Relevant Loans. We are satisfied that, on the without noticeapplication before Nugee J for the WFO, the claim was advanced to a significant extent, but not exclusively, on the basis that the English and BVI Defendants had themselves received funds derived from the Relevant Loans. The claim was put forward on both bases before Fancourt J, although it may be that the submissions focused more on the receipt-based claim. Ground 1

Article 6 of the Lugano Convention provides:

“A person domiciled in a State bound by this Convention may also be sued:

1.

where he is one of a number of defendants, in the courts for the place where any one of them is domiciled, provided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments from separate proceedings;

2.

as a third party in an action on a warranty or guarantee, or in any other third party proceedings, in the court seised of the original proceedings, unless these were instituted solely with the object of removing him from the jurisdiction of the court which would be competent in his case;

3.

on a counter-claim arising from the same contract or facts on which the original claim was based, in the court in which the original claim is pending;

4.

in matters relating to a contract, if the action may be combined with an action against the same defendant in matters relating to rights in rem in immovable property, in the court of the State bound by the Convention in which the property is situated.”

It is common ground that Mr Kolomoisky and Mr Bogolyubov were at the commencement of the present proceedings domiciled in Switzerland and that the English Defendants are, by virtue of having their registered offices in England, domiciled in England (see article 60 of the Lugano Convention).

The issue raised before the judge and on this appeal is whether, in addition to the express qualification in article 6(1) that the claims against the defendants are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings, article 6(1) is subject to a further qualification that it may not be invoked if the proceedings in question are instituted with the sole object of removing a defendant from the jurisdiction of his domicile or from another appropriate jurisdiction (the sole object test).

Not only is this not an express qualification to article 6(1), but it is an express qualification to article 6(2). As will be seen, this is a relevant but not of itself a determinative factor.

The Lugano Convention has in this, as in many other respects, closely followed the EU instruments on jurisdiction in civil and commercial matters. The link is made clear in the recitals to the Lugano Convention. The original Lugano Convention of 16 September 1988 extended the application of the rules of the original EU instrument, the Brussels Convention of 27 September 1968 (the Brussels Convention), to certain members of the European Free Trade Association. The Brussels Convention was replaced by Council Regulation (EC) No 44/2001 of 22 December 2000 (Brussels 1) and its provisions were largely reproduced in the current Lugano Convention.

For present purposes, the purpose of the EU instruments and both versions of the Lugano Convention has been to allocate jurisdiction in civil and commercial matters among the member states of the EU and among the parties to the Lugano Convention, with the overarching principle that a party is to be sued in the state of that party’s domicile, subject to express exceptions.

Article 2 of the Lugano Convention states the primary position, giving priority to a defendant’s jurisdiction of domicile:

“1.

Subject to the provisions of this Convention, persons domiciled in a State bound by this Convention shall, whatever their nationality, be sued in the courts of that State.

2.

Persons who are not nationals of the State bound by this Convention in which they are domiciled shall be governed by the rules of jurisdiction applicable to nationals of that State.”

Articles 2 of the Brussels Convention and of Brussels 1 are in materially the same terms, as is article 4 of the Recast Brussels Regulation.

The existence of exceptions, and their limited effect, is made clear by article 3(1) of the Lugano Convention (and, in materially the same terms by articles 3 of the Brussels Convention and of Brussels 1 and by article 5 of the Recast Brussels Regulation):

“Persons domiciled in a State bound by this Convention may be sued in the courts of another State bound by this Convention only by virtue of the rules set out in sections 2 and 7 of this Title.”

The only exception relevant to this appeal is that contained in article 6 of the Lugano Convention, quoted above. It appears in Section 2, headed “Special jurisdiction”, with equivalent provisions in article 6 of Brussels 1 and article 8 of the Recast Brussels Regulation.

Article 6(1) of Brussels 1 made a significant amendment to its equivalent in the Brussels Convention, article 6, which had provided:

“A person domiciled in a Contracting State may also be sued:

1.

where he is one of a number of defendants, in the courts for the place where any one of them is domiciled;

2.

as a third party in an action on a warranty or guarantee or in any other third party proceedings, in the court seised of the original proceedings, unless these were instituted solely with the object of removing him from the jurisdiction of the court which would be competent in his case;

3.

on a counterclaim arising from the same contract or facts on which the original claim was based, in the court in which the original claim is pending.”

The significant change was to introduce into article 6(1) the qualification of a close connection between the claims brought against the defendants. The same change was made when the Lugano Convention replaced the original Lugano Convention in 2007. The background to this change, and the reasons for it, are significant for the purposes of this appeal.

There can be no dispute that the issue of whether a sole object test is applicable to article 6(1) involves some conflicts of principle. On the one hand, the primacy of allocating jurisdiction to the state of domicile of the defendant leads to a strict interpretation of the express exceptions. Given that primacy, the commencement of proceedings in one jurisdiction for the sole object of removing a defendant from the jurisdiction of his domicile can be seen as contrary to the scheme of the Lugano Convention and the Regulations. On the other hand, it is a primary aim of the Lugano Convention and the EU instruments to promote certainty and predictability in the allocation of jurisdiction, which is achieved by the express terms of article 6(1) but could be put at risk by the imposition of a sole object test.

It is common ground that there is no decision either of the Court of Justice of the European Union (or, as it was previously called, the European Court of Justice) (the CJEU) or of any UK domestic court which provides an answer binding on us.

It is necessary to trace the development of the CJEU case law on the application of article 6(1). None of the cases concerned the Lugano Convention, but this is not significant because courts applying and interpreting the Convention must “pay due account to the principles laid down by any relevant decisions” on the EU instruments: Article 1 of Protocol 2 to the Lugano Convention.

The unqualified terms of article 6(1) of the Brussels Convention were a cause of concern, which led to a largely unanimous view among academic lawyers and national courts that it was necessary for there to be a connection between actions brought against different defendants: see the opinion of Advocate General Darmon in Kalfelis v Bankhaus Schröder (case 189/87) [1988]ECR 5565, [1989] ECC 407 (Kalfelis) at p.409. This was the position taken in the Jenard Report on the Brussels Convention published in 1979 in its comments on article 6(1):

“In order for this rule to be applicable there must be a connection between the claims made against each of the defendants, as for example in the case of joint debtors. It follows that action cannot be brought solely with the object of ousting the jurisdiction of the courts of the State in which the defendant is domiciled.”

The existence and extent of any limitation arose for decision by the CJEU in Kalfelis.

In its judgment, the Court noted at [8] the principle laid down by the Convention that the courts of the state in which the defendant is domiciled have jurisdiction and that article 6(1) constituted an exception to this principle, so that it followed “that such an exception must be treated in such a way that it cannot call into question the very existence of the principle”. The Court continued:

“[9] That might be the case if a plaintiff were free to bring an action against several defendants for the sole purpose of removing one of them from the courts of the State where he is domiciled. As pointed out by the report of the committee of experts who drew up the text of the Convention, any such possibility must be excluded. For this purpose it is necessary for a connection to exist between the actions against each defendant.

[10]

It appears that, to ensure so far as possible the equality and uniformity of the rights and obligations which flow from the Convention for Contracting States and persons concerned, it is necessary to determine the nature of the connection independently.

[11]

On this point it should be noted that the abovementioned report by the committee of experts expressly justifies Article 6(1) by the concern to avoid the delivery of decisions in Contracting States which would be incompatible as between themselves. That is, moreover, a concern which has been embodied in the Convention itself in Article 22, which governs cases where related actions are brought in the courts of different Contracting States.

[12]

Therefore the rule laid down by Article 6(1) applies where actions against different defendants are connected at a time when they are commenced, that is to say, when it is expedient to hear and determine them together to avoid judgments which might be irreconcilable if the actions were determined separately. It is for the national court to ascertain whether this condition is satisfied in each particular case.”

In his opinion, the Advocate General referred to the unanimous view of academic lawyers and national case law that it was necessary for there to be a connection between the actions. He said that the reason for this requirement was “to uphold the principle laid down by the rule actor sequitur forum rei so as to ‘prevent [article 6(1)]

The Advocate General further stated:

“It would be difficult to apply a subjective criterion which would necessitate asking whether the plaintiff did or did not intend to remove one of the defendants from the court which would normally have jurisdiction.

Whatever the circumstances, it must be possible to deduce which court has jurisdiction from objective rules. Legal certainty would hardly be compatible with an examination of the plaintiff’s intentions, which would be both difficult and uncertain.”

He concluded that the right approach was to adopt a requirement of a close connection to avoid the risk of irreconcilable judgments resulting from separate proceedings, being the test expressly applied by article 22(3) of the Convention, adding that “[t]he prevention of irreconcilable judgments is the ratio legis of both Article 6(1) and Article 22(3)”.

To similar effect was the decision of the CJEU in Réunion Européenne SA v Spliethoff’s Bevrachtingskantoor BV (Case C-51/97) [2000] QB 690 (Réunion). At [16], the Court noted that it was settled case law that the general principle was that a defendant’s state of domicile determines the court with jurisdiction over him and that it is only by way of derogation from that principle that the Brussels Convention provided for cases “which are exhaustively listed” in which a defendant may be sued in the courts of another contracting state. At [46] the Court observed that the Convention pursued “the objective of legal certainty” and then said:

“47.

In any event, the exception provided for in article 6(1) of the Convention, derogating from the principle that the courts of the state in which the defendant is domiciled are to have jurisdiction, must be construed in such a way that there is no possibility of the very existence of that principle being called in question, in particular by allowing the plaintiff to make a claim against a number of defendants with the sole purpose of ousting the jurisdiction of the courts of the state where one of those defendants is domiciled: Kalfelis v. Bankhaus Schröder, Münchmeyer, Hengst & Co. (Case 189/87)[1988] E.C.R. 5565, 5583, paras. 8 and 9.

48.

Accordingly, after pointing out that the purpose of article 6(1) of the Convention, and of article 22, is to ensure that judgments which are incompatible with each other are not given in the contracting states, the court held in Kalfelis that, for article 6(1) of the Convention to apply, there must exist between the various actions brought by the same plaintiff against different defendants a connection of such a kind that it is expedient to determine the actions altogether in order to avoid the risk of irreconcilable judgments resulting from separate proceedings.”

Although article 6(1) of Brussels 1 was drafted in a manner that reflected and gave effect to the decisions in Kalfelis and Réunion, they are cases that have continued to be cited by the CJEU. Mr Mark Howard QC, on behalf of all the defendants on this issue, submitted that they demonstrate the existence of a general principle to which article 6(1) is subject, that it cannot be relied on for the sole object of subjecting a defendant domiciled in one member or contracting state (a foreign defendant) to the jurisdiction of the courts of another member or contracting state. In other words, a foreign defendant may not be named as a defendant in proceedings in another state where the proceedings against a defendant domiciled in that state (an anchor defendant) are commenced with that sole object.

Lord Pannick QC submitted on behalf of the Bank that the decisions demonstrate that the vice posed by proceedings commenced with that sole object was met by the requirement laid down by the Court that there must exist a connection between the various claims against the defendants of such a kind that it was expedient to determine the claims together in order to avoid the risk of irreconcilable judgments resulting from separate proceedings. Lord Pannick drew attention to the passage in the Jenard Report quoted above where it is said to follow from the requirement for a connection that proceedings cannot be brought for the sole object of ousting the jurisdiction of the courts of a defendant’s domicile. He relied also on the view of the Advocate General in Kalfelis that “it must be possible to deduce which court has jurisdiction from objective rules. Legal certainty would hardly be compatible with an examination of the plaintiff’s intentions, which would be both difficult and uncertain”. The Advocate General may have had in mind a subjective criterion when making this remark, whereas, as discussed below, all parties are agreed that any such criterion, if it exists, is to be judged objectively. Nonetheless, Lord Pannick submitted, it would introduce a lack of legal certainty, requiring an examination of the circumstances in which the claims were brought going beyond the test of a sufficient connection.

As mentioned above, the wording of article 6(1) of Brussels 1, carried forward unaltered to the Recast Brussels Regulation, reflected the decisions in Kalfelis and Réunion, by requiring “the claims to be so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings”.

It is significant to note that in developing the draft that was to become Brussels 1 and, in due course, the Lugano Convention of 2007, the EU Commission proposed that article 6(1) should read as follows:

“A person habitually resident in a Contracting State may also be sued in another Contracting State:

1.

Where he is one of a number of defendants, in the courts for the place where any one of them is habitually resident, unless the action has been brought solely in order to cause the co-defendants to appear in a court other than their own court.”

This was not adopted and the Pocar Report on the Lugano Convention, published in 2009, explains at paragraphs 69-70 thatthe ad hoc working party of experts composed of representatives of EU and EFTA states, established by the Council of the EU in December 1997 to examine amendments to the Brussels and original Lugano Conventions, considered it advisable to codify the case-law that required a connection between claims but:

“did not believe it necessary to codify the other principle stated in the Jenard report, according to which jurisdiction is justified only if the claim does not have the exclusive purpose of removing one of the defendants from their proper court. It felt that the close relation that must exist between claims, together with the requirement that the court before which the matter was brought be the court of the domicile of one of the defendants, was sufficient to avoid the misuse of the rule (fn 8); this was not the case with an action on a warranty or guarantee or other third party proceedings regulated by Article 6(2), where the principle was expressly referred to in order to prevent a third party from being sued in an unsuitable court.”

The apparent clarity of this explanation is muddied somewhat by Professor Pocar’s footnote 8 which states: “This consideration is not meant to imply that Article 6(1) may be interpreted in such a way that it would allow a plaintiff to bring an action against a plurality of defendants in the court competent for one of them with the sole purpose of removing the other defendants from their proper court”, followed by reference to two subsequent decisions of the CJEU (Reisch Montage and Freeport) to which we later refer. Given the references, this would appear to be Professor Pocar’s own view, rather than a statement of the position of the ad hoc committee.

In any event, the member states refused to include in article 6(1) the sole object limitation that appears in article 6(2), taking the view that the general condition that the claims be connected was more objective: see the judgment of the CJEU in Freeport plc v Arnoldsson (Case C-98/06) [2008] QB 634 (Freeport) at [51].

The first case before the CJEUon article 6(1) of Brussels 1 was Reisch Montage AG v Kiesel Baumaschinen Handels GmbH (Case C-103/05) [2007] I.L.Pr. 10 (Reisch Montage). The claimant brought an action to recover a debt in Austria against the debtor, an individual who was domiciled in Austria, and the guarantor, a company which was domiciled in Germany. At the commencement of the action, the debtor had already been made bankrupt in Austria, and under Austrian law he could not be made a defendant in such proceedings while he remained bankrupt. The referring Austrian court made clear in its judgment (reported at [2005] I.L.Pr. 44) that it had not been established, and it could not without more be assumed, that the claimant knew that the debtor was bankrupt when it commenced the action.

The question referred to the CJEU was whether a claimant could rely on article 6(1) “where the claim against the person domiciled in the forum state is already inadmissible by the time the claim is brought, because bankruptcy proceedings have been commenced against him, which, under national law, results in a procedural bar”.

The answer given by the CJEU was that article 6(1) “must be interpreted as meaning that, in a situation such as that in the main proceedings, that provision may be relied on…even when that action is regarded under a national provision as inadmissible from the time it is brought in relation to the first defendant”.

In its judgment, the Court referred to the general principle stated in article 2, that the courts of the state where a defendant is domiciled are to have jurisdiction over him, and the need to interpret strictly the exceptions which are exhaustively listed in Brussels 1: see [22]-[23]. At [24], the Court said that the exceptions must be interpreted “having regard for the principle of legal certainty, which is one of the objectives” of Brussels 1, as it was of the Brussels Convention. At [25], it stated that the principle of legal certainty required the exceptions to “be interpreted in such a way as to enable a normally well-informed defendant reasonably to foresee before which courts, other than those of the state in which he is domiciled, he may be sued”.

The decision of the Court on the issue before it appears to have turned on the inapplicability of domestic rules to the operation of article 6(1): see [26]-[31], the conclusion being stated at [31] in these terms:

“In those circumstances, Art. 6(1) of Regulation 44/2001 may be relied on in the context of an action brought in a Member State against a defendant domiciled in that state and a codefendant domiciled in another Member State even when that action is regarded under a national provision as inadmissible from the time it is brought in relation to the first defendant.”

The significance of the Court’s decision in Reisch Montage lies in what it said at [32]:

“However, the special rule on jurisdiction provided for in Art. 6(1) of Regulation 44/2001 cannot be interpreted in such a way as to allow a plaintiff to make a claim against a number of defendants for the sole purpose of removing one of them from the jurisdiction of the courts of the Member State in which that defendant is domiciled (see, in relation to the Brussels Convention [Kalfelis] at [8] & [9] and [Réunion] at [47]). However, this does not seem to be the case in the main proceedings.”

While not entirely clear, it seems likely that in the last sentence of [32], the Court was referring to the fact, as stated by the Austrian court, that it had not been established and could not be assumed that the claimant knew of the Austrian debtor’s bankruptcy and hence that the action was inadmissible against him under Austrian bankruptcy law. It may be for this reason that its answer, quoted above, is qualified by the words “in a situation such as that in the main proceedings”.

This paragraph naturally features large in Mr Howard’s submissions, and it was a significant part of the judge’s reasons for his decision that article 6(1) is subject to a sole object test. It is striking that the Court repeated this principle, notwithstanding that article 6(1) of Brussels 1 had been drafted to include the close connection test but not the sole object test. However, Lord Pannick pointed out that the authorities cited

Mr Howard referred us to the Opinion of Advocate General Colomer, who said at [AG28]:

“There are two reasons for the connection requirement. On the one hand, it reduces the risk of diverging judgments; on the other hand, it avoids the unwarranted removal of one of the defendants from the courts of the state where he is domiciled.”

Mr Howard also drew attention to the footnote to this paragraph where the Advocate General states:

“Although this second reason is mentioned in article 6(2)….it is not mentioned in para (1); but it can be deduced from the spirit and purpose of the provision, as a corollary to the connecting link (Jenard Report) or in an autonomous manner

(Droz…considers that it is due to an involuntary omission rather than a voluntary silence).”

The suggestion that the absence in article 6(1) of any reference to the removal of a defendant from the courts of his state of domicile was “an involuntary omission rather than a voluntary silence” is inherently improbable and cannot stand with the drafting history described by Professor Pocar and by the CJEU in Freeport, to which we have referred. It should be noted that the Court did not adopt the answer and analysis proposed by the Advocate General.

The significance of paragraph [32] of the judgment in Reisch Montage needs to be assessed in the light of later developments.

The next case before the CJEU was Freeport. The claimant commenced proceedings in Sweden against a Swedish company and its English parent company. The English company challenged the jurisdiction of the Swedish court on the grounds, among others, that the claim against the Swedish company had been brought with the sole object of suing the English company in Sweden. The relevant question posed by the Swedish court to the CJEU was:

“is it a precondition for jurisdiction under article 6(1), in addition to the conditions expressly laid down therein, that the action against a defendant before the courts of the state where he is domiciled was not brought solely in order to have a claim against another defendant heard by a court other than that which would otherwise have had jurisdiction to hear the case?”

The Court’s answer was:

“Article 6(1) of Regulation No 44/2001 applies where claims brought against different defendants are connected when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings, without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled.”

In his Opinion, Advocate General Mengozzi said at [47] that in essence the Swedish court was asking whether article 6(1) “applies only provided it is established that the action against a defendant domiciled in the member state of the court seised has not been brought solely with the object of removing another defendant from the jurisdiction of the court which could be competent in the case”. He observed at [48] that the question raised “the sensitive issue of the limits on the fraudulent or wrongful use of the bases for jurisdiction which Regulation 44/2001 lays down”. At [53], he said that “the system of rules also establishes certain mechanisms which make it possible to curtail the opportunities for using it in a fraudulent or wrongful manner”. He noted the express limitation of close connection in article 6(1) and the express limitation in article 6(2) as regards sole object and said at [57] that the Swedish court was asking whether the limitation in article 6(2) also applied to article 6(1), even though it did not specifically provide for it.

The EU Commission proposed that this question should be answered in the negative, taking the view that article 6(1) must be interpreted as meaning that if the claims are sufficiently connected, there can be no questioning of the objectives the claimant is pursuing. The Commission submitted that this was supported by the Court’s judgment in Kalfelis.

The Advocate General rejected the Commission’s approach. First, he interpreted the decisions in Kalfelis and Réunion as establishing a presumption that there was neither fraud nor abuse if the specific connection required by article 6(1) existed. He further regarded Reisch Montage as showing that the presumption could be rebutted if the circumstances made it possible to establish the fraudulent or wrongful use of the close connection under article 6(1). Further, the Commission’s submission, while respecting the express requirement of article 6(1) “does not preclude the possibility of the claimant using the basis for jurisdiction under article 6(1) with the sole object of ousting the jurisdiction of the court for the place of domicile of one of the defendants and, consequently, does not eliminate the risk of fraud or abuse”. He then gave the example of proceedings being brought against a fictitious co-defendant, and expressed the view that the claimant in Reisch Montage would not have been entitled to rely on article 6(1) if it had been established that it was acting in bad faith. In his opinion, the rules in Brussels 1 were limited by “fraud relating to the jurisdiction of the courts” and fraud of that nature occurred if the rules are applied “as a result of manipulation on the part of the claimant which is designed to oust and has the effect of ousting the jurisdiction of the courts of a particular member state”.

At [63] the Advocate General considered the question whether it is possible to identify a general prohibition on the abuse of the right to choose the court as more delicate. His view was that the express limitation in article 6(2) applied also to cases within article 6(1) and that the Court should answer the Swedish court’s question accordingly.

We have set out at some length the Advocate General’s reasoning because it clearly expounds the case for the application of a sole object test to article 6(1).

The Court did not accept the recommendation of the Advocate General as to the answer to be given. In giving its reasons, the Court stated:

“52 It should be recalled that, after mentioning the possibility that a plaintiff could bring a claim against a number of defendants with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants was domiciled, the court ruled, in Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst & Co. (Case 189/87) [1988] E.C.R. 5565,that it was necessary, in order to exclude such a possibility, for there to be a connection between the claims brought against each of the defendants. It held that the rule laid down in article 6(1) of the Brussels Convention applies where claims brought against different defendants are connected when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings.

53.

Thus, that requirement of a connection did not derive from the wording of article 6(1) of the Brussels Convention but was inferred from that provision by the Court of Justice in order to prevent the exception to the principle that jurisdiction is vested in the courts of the state of the defendant’s domicile laid down in article 6(1) from calling into question the very existence of that principle: the Kalfelis case, para 8. That requirement, subsequently confirmed by the judgment in Réunion Européenne SA v Spliethoff’s Bevrachtingskantoor BV (CaseC51/97)[2000] QB 690, para 48, was expressly enshrined in the drafting of article 6(1) of Regulation No 44/2001, the successor to the Brussels Convention : Roche Nederland BV v Primus (Case C-539/03) [2006] ECR I-6535, para 21.

54.

In those circumstances, the answer to the question referred must be that article 6(1) of Regulation No 44/2001 applies where claims brought against different defendants are connected when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings, without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled.”

This appears to us to provide a clear, negative answer to the question of whether article 6(1) is subject to an implicit sole object test. However, in the present case the judge held at [90] that, while not wholly clear, it appeared to be a decision that a claimant need only prove the close connection and expediency explicitly referred to in article 6(1) and need not also disprove that the claim was brought with the sole object

We are unable to accept this reading of the decision. In civil law systems, the notion of a burden of proof on issues of the court’s jurisdiction does not sit easily with the duty of the court to satisfy itself as to its jurisdiction: see the Advocate General’s Opinion in Cartel Damage (infra) at [86]. Further, and importantly, the English defendant in the Swedish proceedings positively asserted that the claimant had brought the action against the Swedish defendant with the sole object of suing the English defendant in the Swedish court, albeit it is fair to note that this challenge was based on its primary submission that the claims lacked the necessary close connection: see the Advocate General’s Opinion at [19] and the Court’s judgment at [20]. There is no suggestion in the Opinion or the judgment that the case was concerned with burdens of proof or rebuttable presumptions.

However, the decision in Freeport is not the last word on this issue, and it is now established that, at least in certain circumstances, abuse of the right to commence proceedings against an anchor defendant will not confer jurisdiction against a foreign defendant. That is the effect of the decision of the CJEU in Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Akzo Nobel NV (Case C-352/13) [2015] QB 906 (Cartel Damage).

Before coming to Cartel Damage, there are two further decisions of the CJEU to note. The first is Painer v Standard Verlags GmbH (Case C-145/10) [2012] ECDR 6 (Painer). This was a claim for breach of copyright brought in Austria against various newspaper and magazine publishers domiciled in Austria and Germany. Most of the questions referred to the CJEU concerned article 5 of Brussels 1 and are not material for present purposes, but one question concerned the application of article 6(1) to claims for copyright infringement identical in substance but based on differing national legal grounds. In the course of its judgment, the Court repeated, almost verbatim, paragraph [32] of Reisch Montage, with the same references to Kalfelis and Réunion. Virtually the same paragraph with the same references, to which was added

In Cartel Damage, claims were made for damages and disclosure in respect of alleged collusive agreements made in different parts of Europe and causing loss to various business undertakings. The claims were made in proceedings in Germany against six companies domiciled in various EU member states, one of which was domiciled in Germany. Subsequently, the claimant settled the claim against the German company. The remaining defendants challenged the jurisdiction of the German court, claiming that the claimant and the German defendant had deliberately delayed the conclusion of the settlement until proceedings had been commenced, with the sole object of securing jurisdiction for the German court.

One of the questions referred to the CJEU was concerned with the application of article 6(1) and asked, in part, “Is it significant…if the action against the defendant domiciled in the same state as the court is withdrawn after having been served on all

The answer given by the Court was that article 6(1) applied “even where the applicant has withdrawn its action against the sole co-defendant domiciled in the same state as the court seised, unless it is found that, at the time the proceedings were instituted, the applicant and that defendant had colluded to artificially fulfil, or prolong the fulfilment of, that provision’s applicability”.

In considering the effect of a deliberate delay to a settlement with the anchor defendant until after proceedings had been issued and served, Advocate General Jääskinen stated in his Opinion at [84]:

“In accordance with the court’s consistent case law, “the rule [on jurisdiction laid down in article 6(1) of the Brussels I Regulation] cannot be interpreted in such a way as to allow the plaintiff to make a claim against a number of defendants for the sole purpose of removing one of them from the jurisdiction of the courts of the member state in which the defendant is domiciled” (my emphasis): the Reisch Montage case, para 32 and Painer’s case, para 78, and the case law cited. This restriction concerning a potential ousting of jurisdiction of the court seised, which was supported in the Jenard Report is perfectly consistent with the requirement that derogations from the jurisdiction in principle of the courts for the place of the defendant’s domicile, laid down in article 2 of the Brussels I Regulations, should be interpreted restrictively.”

Referring to the allegation made by the defendants in Cartel Damage that the settlement with the anchor defendant had been deliberately delayed, the Advocate General said at [88]:

“Provided that the deceitful tactics alleged, disputed in this case by the parties concerned, are not just probable but confirmed, which it will be for the national court to establish, such an abuse of rights, which is designed to deprive one or more of the defendants of the general jurisdiction as a rule of the court for the place where they are domiciled, ought in my view to be penalised by refusal to apply article 6(1) of the Brussels I

Regulation in those circumstances, given that the criteria relating to connecting factors were not truly satisfied at the time the action was brought. The advantage of its being the court for the domicile of the “anchor defendant” that examines and tries the claims against several defendants at the same time, in accordance with that provision, disappeared as soon as a binding transaction put an end, vis-à-vis that defendant, to the legal obligation that the applicant could have relied upon against it before that court. Apart from these specific cases, there is, on the other hand, no need in my view to examine and penalise an abuse of rights in a legal context of this nature.”

The closing words of that paragraph indicate that the Advocate General contemplated a narrow range of circumstances as amounting to an abuse of article 6(1), a point emphasised by Lord Pannick. In the context of the facts alleged in Cartel Damage, the Advocate General referred to “deceitful tactics” at [88] and knowing concealment at [90]. See also his suggested answer to the relevant question at [133(b)].

The Court set out the applicable approach in general terms in the following terms:

“25 It must therefore be considered that determining separately actions for damages against several undertakings domiciled in different member states which, contrary to EU competition law, participated in a single and continuous cartel may lead to irreconcilable judgments within the meaning of article 6(1) of Regulation No 44/2001.

26

That said, it remains to be considered to what extent the applicant’s withdrawal of its action against the sole codefendant domiciled in the same member state as the court seised is capable of rendering the rule of jurisdiction provided for in article 6(1) of Regulation No 44/2001 inapplicable.

27

According to settled case law, that rule cannot be interpreted as allowing an applicant to make a claim against a number of defendants for the sole purpose of removing one of them from the jurisdiction of the courts of the member state in which that defendant is domiciled: the Reisch Montage case [2006] ECR I6827, para 32, and in Painer’s case [2011] ECR I-12533, para 78.

28

The court has nevertheless stated that, where claims brought against various defendants are connected within the meaning of article 6(1) of Regulation No 44/2001 when the proceedings are instituted, the rule of jurisdiction laid down in that provision is applicable without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled: Freeport plc v Arnoldsson [2008] QB 634, para 54.

29

It follows that where, when proceedings are instituted, claims are connected within the meaning of article 6(1) of Regulation No 44/2001, the court seised of the case can find that the rule of jurisdiction laid down in that provision has potentially been circumvented only where there is firm evidence to support the conclusion that the applicant artificially fulfilled, or prolonged the fulfilment of, that provision’s applicability.”

In our view, these are critical paragraphs. Paragraphs [27] and [28] set out conflicting principles, the thesis and the antithesis. The synthesis is stated at paragraph [29]. It accepts that there is a role in the application of article 6(1) for a principle of abuse of

Other examples of an artificial fulfilment might be the hypothetical case posited by the Advocate General in Freeport of proceedings being commenced against a fictitious anchor defendant and a claimant in a case such as Reisch Montage commencing proceedings against the anchor defendant knowing that such proceedings were inadmissible against him.

We were referred to three domestic decisions, two of the Court of Appeal and one of the Supreme Court.

The decisions of the Court of Appeal are of limited assistance. JSC Aeroflot v Berezovsky [2013] EWCA Civ 784, [2013] 2 CLC 206, concerned, so far as relevant, a challenge to the English court’s jurisdiction by a company incorporated and domiciled in Luxembourg which was named as a co- defendant in reliance on article 6(1) of Brussels 1. The challenge was on the basis that there was no arguable claim against that defendant and that therefore it could not be expedient to hear and determine the claim to avoid the risk of irreconcilable judgments. Aikens LJ, with whom Laws LJ and Mann J agreed, relied on the CJEU’s decision in Freeport for the proposition that “the national court should not concern itself with the question of whether the claim against the non-resident defendant was brought in those proceedings with the sole object of ousting the jurisdiction of the court of the Member State where that defendant is domiciled. The ECJ therefore specifically rejected the proposition that the national court should consider and decide whether other possible motives for bringing that defendant into the proceedings should be taken into account” (emphasis added). The emphasised words show that the court was looking at the application of a sole object test to the foreign defendant, not the anchor defendant.

The other decision of this court is Sabbagh v Khoury [2017] EWCA Civ 1120. The claim in that case was made against several defendants, one of whom was domiciled in England. Most of the others were domiciled either in EU member states or in Switzerland, and they were joined as defendants in reliance on article 6(1) of Brussels 1 or, as appropriate, of the Lugano Convention. Those defendants challenged the English court’s jurisdiction on the grounds that reliance could not be placed on article 6(1) where there was no sustainable claim against the anchor defendant. This court was divided on the issue whether the merits of the claim were a factor relevant to the application of article 6(1). The majority (Patten and Beatson LJJ) held that they were, while Gloster LJ disagreed. The case does not assist us because it was accepted by the claimant that article 6(1) was subject to a sole object test, the very issue that we have to decide: see [169(i)]. Further, the views of all three members of the court were obiter, as they were for other purposes satisfied that the claim against the anchor defendant was sustainable: see [1].

In Lungowe v Vedanta Resources plc [2019] UKSC 20, [2019] 2 WLR 1051

At [29], Lord Briggs referred to the primary rule laid down in article 4 and the need to construe narrowly the express exceptions to it. He said: “If, therefore, the Recast Brussels Regulation also contains (as it probably does) an implied exception from the otherwise automatic and mandatory effect of article 4, based upon abuse of EU law, then that is also an exception which is to be narrowly construed”.

Commenting on article 8(1) of the Recast Brussels Regulation (article 6(1) of Brussels 1), Lord Briggs said at [31]:

“Since article 8.1 is itself to be restrictively interpreted because it derogates from the primary rule of jurisdiction in article 4, it might be thought that the Court of Justice would liberally apply an abuse of law principle where it perceived that article 8 was being misused as a means of circumventing article 4. None the less the cases show that abuse of EU law has been restrictively interpreted, even in this context.”

Having referred to Freeport and Cartel Damage andto the Opinion of the Advocate General in the latter case where he stated the sole object test, Lord Briggs said at [34]:

“In its judgment, the Court of Justice [in Cartel Damage] expressly affirmed that opinion in para 27, adding at para 33 that in the context of cartel cases nothing short of collusion between the claimant and the anchor defendant would be sufficient to engage the abuse of law principle.”

Lord Briggs continued:

“35 Those decisions of the Court of Justice show that, even before the Freeport case [2008] QB 634, there was an established line of authority which limited the use of the abuse of EU law principle as a means of circumventing article 6 (now article 8) to cases where the ability to sue a defendant otherwise than in the member state of its domicile was the sole purpose of the joinder of the anchor defendant. Even though there appears to be no authority directly upon abuse of EU law in relation to article 4 itself (or its predecessors), the need to construe any express or implied derogation from article 4 restrictively would appear to make the position a fortiori in relation to article 4, as indeed the judge himself held.

36 But the matter does not stop there. Such jurisprudence as there is about abuse of EU law in relation to jurisdiction suggests that the abuse of law doctrine is limited to the collusive invocation of one EU principle so as improperly to subvert another.”

Applying the CJEU authorities, including those on forum non conveniens to the case in hand, Lord Briggs said at [40] that “leaving aside those caseswhere the claimant has no genuine intention to seek a remedy against theanchor defendant, the fact that article 4 fetters and paralyses the English forum conveniens jurisprudence in this way in a necessary or proper party case cannot itself be said to be an abuse of EU law…” (emphasis added). Mr Howard stressed the emphasised words.

In the light of Cartel Damage, Lord Pannick accepts that a principle of abuse of law is applicable to article 6(1), but that it is restricted to cases of fraud and collusion. Mr Howard argues for the wider proposition that proceedings brought against an anchor defendant for the sole object of removing a foreign defendant from the courts of his state of domicile is an impermissible abuse of article 6(1).

Entirely reasonably, Mr Howard, like the judge, relies heavily on the general statements of the sole object test in Reisch Montage at [32] and in Painer and in Solvay, as well as similar statements in the opinions of the Advocate General in some of the cases. Like the judge, he also interprets the CJEU’s decision in Freeport not as rejecting the sole object test but as stating that the court is not concerned to investigate that issue unless it clearly appears on firm evidence that it applies on the facts of the specific case before the court. He submits that the existence of a general sole object test was affirmed by the CJEU in Cartel Damage at [27]. This general principle is consistent with the basic rule of all the Conventions and Regulations that, subject to express exceptions which are to be restrictively interpreted, a defendant is to be sued only in the courts of the state of his domicile.

A principal aspect of Mr Howard’s submissions was that the fundamental premise of article 6(1) was that, but for its application, there would be separate proceedings which might lead to irreconcilable judgments. In the absence of separate proceedings, there would be no such risk and no ground for the application of article 6(1). The claimant must have a genuine intention to sue the anchor defendant separately, before article 6(1) can be invoked. Accordingly, the Bank could not rely on article 6(1) to join Mr Kolomoisky and Mr Bogolyubov as co-defendants to the English proceedings unless it would in any event sue the English Defendants alone.

There is a significant difficulty with this submission. It is not difficult to think of circumstances in which a claimant would wish to sue two defendants together but would not sue either of them separately. Take the example of a conspiracy claim against two persons, each domiciled in a different state. If Mr Howard’s submission were correct, neither could be sued at all unless the claimant were prepared to sue each of them separately in their states of domicile. Ultimately, Mr Howard accepted that it was difficult to sustain this approach to article 6(1) and that it was not an essential ingredient of the sole object test that the claimant had a genuine intention to sue the anchor defendants on their own.

As regards the test of artificial fulfilment of the express condition of connected claims under article 6(1) set out in Cartel Damage, Mr Howard submitted that it was simply another way of expressing the sole object test. They were two sides of the same coin. Artificial fulfilment was not limited to collusive conduct of the type contemplated by the CJEU in Cartel Damage. It encompassed the situation where a claimant commenced an action in the court of one state against an anchor defendant for the sole purpose of subjecting a foreign defendant to the jurisdiction of that court. The fact that the claim against the anchor defendant had merit would not affect the application of the sole object test.

While we see the force in these submissions, we have concluded that they do not represent the correct position as regards article 6(1) of Brussels 1 and of the Lugano Convention and article 8(1) of the Recast Brussels Regulation. An artificial fulfilment of the express condition of a close connection will not permit reliance on article 6(1). However, if the question is asked, is a claimant with a sustainable claim against an anchor defendant, which it intends to pursue to judgment in proceedings to which a foreign defendant is joined as a co-defendant, entitled to rely on article 6(1) even though the claimant’s sole object in issuing the proceedings against the anchor defendant is to sue the foreign defendant in the same proceedings, we consider that the question should be answered affirmatively.

We reach this conclusion for a number of connected reasons.

First, the terms of article 6 and the history of its drafting do not support the existence of the general sole object test for which the defendants contend. There is a clear contrast between articles 6(1) and 6(2), with only the latter containing a sole object condition. It is fair to ask why it is not also included in article 6(1) if it was intended to apply. As appears from Freeport at [51], the member states refused to include it, on the grounds that the close connection condition was more objective.

Second, the close connection condition was included in article 6(1) of Brussels 1 in response to the CJEU’s decisions in Kalfelis and Réunion. While expounding the need to avoid a defendant being removed from the courts of the state of his domicile, the CJEU proceeded on the basis that the point was met by a close connection condition.

Third, as the Advocate General warned in Kalfelis, an additional requirement based on the intention of the claimant would introduce an undesirable degree of uncertainty into a question, the allocation of jurisdiction among member states, that required certainty and predictability. This is so, even if the claimant’s intention is to be assessed on an objective, rather than subjective, basis. The Recast Brussels Regulation contains recital (16) which emphasises the need for certainty:

“In addition to the defendant’s domicile, there should be alternative grounds of jurisdiction based on a close connection between the court and the action or in order to facilitate the sound administration of justice. The existence of a close connection should ensure legal certainty and avoid the possibility of the defendant being sued in a court of a Member State which he could not reasonably have foreseen.”

The recital contains no reference to a sole object test. The object of the claimant is irrelevant to the factor which is mentioned, that a defendant should reasonably be able to foresee the courts in which he might be sued. That presents no difficulty in the present case, in which it is alleged that Mr Kolomoisky and Mr Bogolyubov used English companies for their fraudulent scheme. They could well foresee that they might be sued in the English courts.

Fourth, we take the view that the CJEU’s decision in Freeport is a statement of principle that article 6(1) is not subject to an unexpressed sole object test. We are not persuaded by the defendants’ submission that, as a result of Freeport, the only difference between articles 6(1) and 6(2) is that, under article 6(1), the object of the claimant in suing the anchor defendant does not necessarily have to be investigated.

Fifth, we regard the CJEU’s decision in Cartel Damage as rejecting a general sole object test but subjecting reliance on article 6(1) to the principle of abuse of law in cases of artificial fulfilment of the close connection condition. In general, all rights under EU law are subject to this principle and there is no reason to exclude article 6(1). It is noteworthy that the example given by the Court is a collusive arrangement between the claimant and the anchor defendant to conceal a settlement of the claim until the proceedings have been issued and served on the foreign defendants. As earlier mentioned, other examples might be naming a fictitious person as the anchor defendant (Freeport) and commencing proceedings against an anchor defendant knowing that it was an inadmissible claim (Reisch Montage).

Sixth, this is consistent with the analysis of Lord Briggs, giving the judgment of the Supreme Court in Vedanta. Although strictly obiter, his analysis deserves, of course, great respect.

Seventh, while the statements in Reisch Montage at [32], and in Painer and Solvay, may be read as unqualified statements of the applicability of a general sole object test, we accept Lord Pannick’s analysis that, as shown by the references to Kalfelis and Réunion, they are no more than restatements of what those cases decided, that the vice in using article 6(1) to remove a foreign defendant from the courts of the state of his domicile was met by a close connection condition.

Accordingly, we conclude in the present case that the Bank, which has a sustainable claim against the English Defendants and which intends to pursue the claim to judgment against those Defendants in combination with its claims against Mr Kolomoisky and Mr Bogolyubov, is entitled to rely on article 6(1) even if its sole object in commencing the proceedings against the English Defendants is to be able also to sue those individuals in the same proceedings.

We proceed to consider this issue on the basis, contrary to what we have decided above, that article 6(1) is subject to a sole object test.

Having held that article 6(1) was subject to a sole object test, the judge held that it applied to the claim brought against the English Defendants and it was therefore an abuse of article 6(1) to join Mr Kolomoisky and Mr Bogolyubov to these proceedings. On that basis, he declared that the English court had no jurisdiction to try the claim against Mr Kolomoisky and Mr Bogolyubov and set aside the WFO against them.

At [94] the judge listed the facts he considered to be relevant to an assessment of the Bank’s purpose in issuing proceedings against the English Defendants, as follows:

“(1)

There is a good arguable claim against them in tort for loss of about US$515m.

(2)

The claim against the English Defendants based on unjust enrichment is hopeless.

(3)

The English Defendants have no substantial assets, as the Bank was aware.

(4)

The English Defendants have no particular significance as corporate entities involved in the fraudulent scheme, other than the fact that they are English companies.

(5)

The English companies appear to have been set up to act as agents for offshore companies and are run by Cypriot corporate service providers with accounts in Cyprus, which facts were at all times available to the Bank.

(6)

The Particulars of Claim present a misleading picture of the English Defendants being central to the fraudulent scheme and as being the intended recipients of US$1.8bn.

(7)

The First and Second Defendants are known to be extremely wealthy businessmen with substantial assets worldwide under their direct or indirect control.

(8)

The claim against the First and Second Defendants is valued at US$5.5bn according to the Bank, but the Bank has chosen to bring instead a claim for US$1.91bn of money that passed through the English and BVI Defendants' bank accounts.

(9)

As compared with a trial in Ukraine or Switzerland, proceedings in England and Wales confer substantial procedural advantages on the Bank, including (a) the ability to obtain worldwide freezing orders extending to assets only indirectly controlled by the First and Second Defendants, (b) an extensive disclosure regime and (c) cross-examination of witnesses at trial.”

The judge said at [95] that these “facts themselves give rise to a strong inference that the English Defendants are only being sued in order to be able to bring a claim” against Mr Kolomoisky and Mr Bogolyubov. He asked: “What other reason could there be for bringing a claim against limited companies that, on any fair analysis of

Having considered and rejected justifications advanced by the Bank at [96]-[97], the judge stated his conclusion and reasons:

“98.

After considering carefully what the Bank has submitted and the likelihood of the Bank suing the English Defendants if it could not have sued the First and Second Defendants, I have no doubt that the Bank has sued the English Defendants in order to establish jurisdiction to sue the First and Second Defendants in London under Article 6 of the Lugano

Convention. I conclude that it is also the sole reason that this particular claim has been brought against the English Defendants. My reasons, in brief summary, are the following.

99.

First, no direct evidence has been given as to why in fact the Bank brought a claim against the English Defendants in December 2017, and none of the arguments that the Bank puts forward now as justification are at all persuasive. The judgment in the Cartel Damages Claim makes clear that the relevant time is, unsurprisingly, the time at which the action was brought, not any later time. Some of the justifications on which the Bank relies could not have existed as reasons in December 2017 because the Bank was unaware of them.

100.

Second, the claim that was brought has been artificially constructed in order to seek to enable the Bank to satisfy Article 6. The claim has been limited to US$1.91bn so that the claims against the First and Second Defendants and the English Defendants appear to be broadly co-extensive and their importance as defendants equivalent. The role of the English Defendants in the fraudulent scheme has been presented to make it appear that the English Defendants were central players in the fraudulent scheme and/or the recipients of the US$1.8bn, such that the Bank would naturally wish to pursue them. A restitutionary claim against the English Defendants, used to support this conclusion, is unsustainable.

101.

Third, the Bank has presented its claim – in the particulars of claim and in its evidence upon the without notice application – so as to omit highly relevant facts. A true presentation of the facts would have revealed that the role of the English Defendants was no more than incidental to the working of the fraudulent scheme.

102.

Fourth, there is no real attraction or benefit to the Bank in suing the English Defendants; the real defendants to the claim are the First and Second Defendants.

103.

Fifth, the Bank has admitted that bringing a claim against the First and Second Defendants in London gives it significant procedural advantages, many of which the Bank has already enjoyed as a result of the worldwide freezing order and the Defendants' compliance with associated disclosure obligations. That was a real benefit to the Bank, as compared with the lesser attraction of litigating in Switzerland or Ukraine, and is self-evidently the reason for the proceedings in London.

104.

I do not shut my eyes to the fact that the Defendants admit that the Bank has a good arguable claim against them for at least hundreds of millions of dollars. On the basis of the evidence that I have read, the proceedings would be very complicated and expensive. That is another reason why, in my judgment, the Bank would not have brought proceedings against the English Defendants on their own. It is, of course, unattractive for the First and Second Defendants to admit a good arguable case of fraud on an epic scale against them and yet seek to prevent this court from investigating the matter. That is particularly so where the First Defendant is on record as previously having stated to the press in Ukraine that the English court can be expected to get to the truth of the matter. However, the consequence of my conclusion is not that the First and Second Defendants will escape justice but that they are entitled under the terms of the Lugano Convention to have any claim brought against them in Switzerland, where they are domiciled. Alternatively, the Bank could bring its proceedings against them in its own State, Ukraine, which is not a party to the Lugano Convention. That is where the entire factual subject matter of the claim naturally resides and where the First Defendant has himself issued proceedings relating to the subject-matter of the alleged fraud. What the Bank is not permitted to do is forum shop without regard to international conventions on jurisdiction.”

Before considering the Bank’s challenge to the judge’s assessment and the parties’ submissions, there are two preliminary points.

First, the Bank accepts, and has never concealed, that a substantial reason for bringing the proceedings in England lies in the advantages that it perceives in the English courts as a venue for major commercial litigation. The statistics for actions commenced in the Business and Property Courts of England and Wales suggest that it is not alone in taking that view. More importantly, the authorities establish, as a matter of European as well as English law, that it is entirely legitimate for a claimant to have regard to this consideration in its choice of jurisdiction: see, for example, the Advocate General’s Opinion in Cartel Damage at [89]. Having said that, if the only reason for commencing proceedings against a defendant in England were to obtain those advantages in a claim against defendants domiciled in another Convention state, that would offend against a sole object test.

Second, it was common ground in this court between the Bank and all the defendants that the objects of a claimant in commencing proceedings in a particular jurisdiction were, for the purposes of a sole object test, to be determined objectively, not subjectively. In other words, the court will assess the claimant’s object or objects by reference to all the established facts and relevant considerations, not on the basis of direct or indirect evidence of the actual motives of the claimant.

A subjective approach would have the major disadvantage of increasing uncertainty and cost at the initial stage of determining jurisdiction. If a claimant gave evidence that it had an object other than (or in addition to) suing the non-domiciled defendant in the jurisdiction in question and such evidence was not self-evidently wrong, it is difficult to see how the issue could be resolved without the very mini-trial that both European and domestic law deprecate. It could also lead to perverse results. A claimant who, objectively, had legitimate reasons for commencing proceedings in England would be prohibited from joining a non-domiciled defendant if, subjectively, its only object was to sue that defendant in England, while a claimant who genuinely but unreasonably believed that it also had some other object could nonetheless proceed.

An objective approach to the ascertainment of purpose or intention is well-established in other areas of European law: see, for example, Gemeente Borsele v Staatssecretaris van Financiën (Case C-520/14) [2016] STC 1570, as regards determining whether property is exploited “for the purposes of obtaining income therefrom on a continuing basis” within article 2(1) of the Principal VAT Directive (2006/112/EC).

It follows from this common ground that the first reason given by the judge, at [99], cannot stand. The judge there relied on the absence of any direct evidence “as to why in fact the Bank brought a claim against the English Defendants in December 2017”, but that could only be relevant to the Bank’s subjective assessment of its own purpose or object in bringing the proceedings.

The judge’s second reason, given at [100], is that the claim has been “artificially constructed in order to seek to enable the Bank to satisfy article 6”. By limiting the claim to US$1.9 billion, the claims against Mr Kolomoisky and Mr Bogolyubov and against the English Defendants “appear to be broadly co-extensive and their importance as defendants equivalent”. The case has been presented “to make it appear that the English Defendants were central players in the fraudulent scheme and/or the recipients of the US$1.8bn, such that the Bank would naturally wish to pursue them”.

Closely allied to the judge’s second reason is his third reason at [101], that the Bank presented its claim both in the particulars of claim and in Lewis (1)so as to omit highly relevant facts, whereas a “true presentation of the facts” would have revealed that the English Defendants’ role “was no more than incidental to the working of the fraudulent scheme”.

These are very important and serious assessments by the judge which we address in detail under Grounds 4 and 5 below. For the reasons given there, we do not consider that they are assessments that can stand.In any case, the fact that the Bank may have framed its claim in such a way as to allow it to take advantage of article 6(1) of the Lugano Convention need not be objectionable. It is consistent with the Bank having a wish to sue Mr Kolomoisky and Mr Bogolyubov in this jurisdiction either as its sole motivation or as merely one purpose, however important. Had the Bank’s claim been “artificially constructed” in the sense of “bogus”, that might have been important. As it is, however, the judge himself considered the Bank to have a good arguable claim against even the English Defendants for some US$515 million.

The judge made a further point at the end of [100] where he said that a restitutionary claim against the English Defendants was “unsustainable”. At [94(2)], he judged it to be “hopeless”. It would appear from [79] that he regarded this as a very significant factor in his decision. All the defendants accept that this was not argued before the judge and that it is a conclusion which was not open to him.

It appears from [60] that the judge based this view on the fact, as it appears to be, that the missing pre-payments passed through each of the English and BVI Defendants in the course of a single day, so that colloquially they were conduits. Accordingly, the judge reasoned, these Defendants “were not enriched by acquiring US$1.91 billion or any money or rights” and they did not benefit from the Bank’s loss. The unchallenged expert evidence was that an unjust enrichment claim could lie under Ukrainian law in these circumstances. The Bank’s expert witness, Oleh Beketov, stated in his report of 6 July 2018 at paragraph 18 that “As a general proposition, where B has been unjustly enriched at the expense of A, A may pursue a cause of action under Article 1212 of the Civil Code against B irrespective of whether B has subsequently dissipated its enrichment”. There is nothing surprising in this statement to an English lawyer nor in its application to the receipt of funds which are dissipated very shortly, even momentarily, after their receipt.

It may well be that the basis of the judge’s view lies in what he said at [96(i)] that the “money never became the English Defendants’ money” because “the English defendants only received the prepayment very briefly, under a pre-arranged scheme in which they were immediately to pass the funds on or back to the borrowers”. With respect to the judge, neither the brief possession of the money nor the existence of a pre-arranged scheme would necessarily have the result that the money did not become the property of the English Defendants while they held it. It is, of course, possible that they agreed to receive the money on terms that imposed a trust, but the full circumstances would need to be investigated at trial before reaching that conclusion. Given that the money was supposedly received as a pre-payment for the supply of commodities or equipment and the only alternative is that the money was paid as part of a fraudulent scheme, the prospects of establishing a trust of the monies appear to be a good deal less than overwhelming. This would be the position in English law and there is no evidence to suggest Ukrainian law would treat the funds as not being the property of the English Defendants while they held them.

The fourth reason, that there was no real attraction or benefit to the Bank in suing the English Defendants, is a ground to which we return below.

The fifth and final reason, at [103], is that the significant procedural advantages to bringing a claim against Mr Kolomoisky and Mr Bogolyubov in England were “selfevidently the reason for the proceedings in London”. As already mentioned, the Bank readily accepts that those advantages are a substantial reason for commencing these proceedings against the English Defendants but that does not constitute an abuse of article 6 unless it is the sole reason. If the judge was right in his assessment that there

At [96] the judge addressed the reasons advanced by the Bank for its submission that its sole object in suing the English Defendants in England was not to sue Mr Kolomoisky and Mr Bogolyubov here rather than in Switzerland. The reasons, in headline terms, were: possible claims against the directors of the English Defendants, possible claims against their shadow directors, possible claims under the agency agreements which the defendants had produced in evidence but whose authenticity was denied by the Bank and disclosure by the English Defendants.

The judge attached no weight to any of these reasons. In our view, he was right not to do so as regards the second and third reasons. There is no evidence of shadow directors other than Mr Kolomoisky and Mr Bogolyubov. The reasons for bringing the proceedings against the English Defendants are to be assessed as the date of their commencement (see Cartel Damages at [31]) and the existence of the purported agency agreements was not known to the Bank. Moreover, the Bank denies their authenticity. The possibility of claims under the agency agreements cannot provide a plausible reason for the proceedings.

At [97] the judge said that all these reasons had been advanced by the Bank after the event to seek to establish that its sole purpose was not to sue Mr Kolomoisky and Mr Bogolyubov in England, and that they gave the appearance of having been developed retrospectively in an attempt to justify the claim. We do not accept that this is a good or a fair point to make. The judge himself accepted in the same paragraph that the Bank’s skeleton argument for the hearing before Nugee J asserted that the English and BVI Defendants might have valuable rights and important disclosure to give.

We have reached the conclusion that the assessment by the judge of the Bank’s reasons for bringing these proceedings is based on too many false grounds for it to stand. It falls to us to make the assessment: no-one has suggested that these proceedings should be prolonged or delayed by remitting the issue to be determined at first instance.

In submitting that the Bank’s sole purpose was to sue Mr Kolomoisky and Mr Bogolyubov in England, Mr Howard on behalf of the defendants placed great emphasis on four matters: (i) the “false” centrality given to the English Defendants by the Bank in its particulars of claim and in the case presented to Nugee J in December 2017, used as “a pretext” for choosing to sue them here; (ii) the emphasis given to the allegation that the English Defendants had received US$1.8 billion of the Bank’s funds, thus giving the appearance of a “follow the money” claim; (iii) the “bogus” reliance on the possibility of claims against the directors; and (iv) the lack of substance in disclosure as a reason for suing them here.

We examine the first and second of these matters later in this judgment when considering Grounds 4 and 5. For the reasons given there, we reject the defendants’ submissions.

It does not follow that the Bank’s sole object was not to sue Mr Kolomoisky and Mr Bogolyubov in England, rather than Switzerland or Ukraine. It is necessary to examine the remaining two matters.

Before doing so, it is right to bear in mind that, as already set out, the defendants accept that the Bank has a good arguable case against the English Defendants for US$515 million in respect of their part in a fraud “on an epic scale”, to use the judge’s phrase. It might be thought entirely natural that a claimant in these circumstances would wish to obtain judgment against those defendants in the jurisdiction of their domicile. Of course, it may be found on an objective assessment that this provides no reason for commencing proceedings against them, but it does suggest that caution is required before arriving at this essentially counter-intuitive conclusion.

The two remaining issues considered by the judge are the possibility of claims against the directors of the English Defendants and disclosure. In his submissions, Lord Pannick concentrated on the second of these issues. He had good reason for doing so. If judgment were obtained against those defendants, there would almost certainly be a strong basis for claims for breach of fiduciary duty against the directors. If they were knowing participants in the fraudulent scheme, Mr Howard is right to say that the Bank would have direct claims in tort against them. However, claims for breach of fiduciary duty would not require proof of dishonesty and would therefore be more easily established.

However, there is no evidence to suggest that it would make commercial sense to bring any such claims against the directors. All the directors appear to be based in Cyprus. Michalakis Tsitsekkos is a lawyer, with his own law firm in Cyprus, and is one of two directors of the third defendant and the sole director of the fourth defendant. Two other individuals are respectively the other director of the third defendant and the sole director of the fifth defendant. There is no evidence that any of them have substantial assets, such as to make it worth contemplating proceedings against them. It might be said that this is an issue to be investigated when the time comes to cause those companies to commence such proceedings but, viewed objectively, it is impossible to consider that this possibility is likely to have been a reason for suing the English Defendants in England or indeed elsewhere.

The focus is therefore on the benefits that the Bank might realistically expect to obtain from disclosure.

The judge dealt with the Bank’s submission that the English Defendants would have important disclosure to give at [96(iv)] where he said:

“iv)

The English Defendants will have important disclosure to give. The Bank points in particular to documents likely to show the circumstances of drafting the supply agreements and the loan file supply agreements and negotiations with the borrowers; the negotiation of the agency agreements; documents that – according to the terms of the agency agreements – would have to be produced in accordance with them relating to the actions of the English Defendants, and documents relating to the beneficial ownership and administration of the English Defendants. If the English Defendants were substantial entities and central to the fraud, as the Bank contended, then there might well be a reasonable expectation that some or all of the identified documents exist. However, the reality is that the English Defendants are mere creatures and conduits, and it is highly implausible that the English Defendants negotiated any of the documents in question or that the agency agreements have been operated in accordance with their terms. The Bank's skeleton argument asserts that "none of the [English and BVI Defendants] had any website, offices, staff, warehouses, workforce or any other public presence". If the Bank were entitled to sue the First and Second Defendants, it is understandable that it might also choose to sue the English and BVI Defendants in the hope that something additional, or inconsistent with the First and Second Defendants' cases, might emerge on disclosure. But that is not the question: the question is whether there was any real purpose in suing the English Defendants other than to join the First and Second Defendants to the claim.”

Lord Pannick criticised the reasoning in this passage in the judgment. The judge said in the third sentence that if the English Defendants were substantial entities and central to the fraud “there might well be a reasonable expectation that some or all of the identified documents exist”. He then observed that it was highly implausible that those defendants negotiated any of the documents in question, but he appears to accept in the penultimate sentence that the English Defendants may have documents to disclose. Moreover, if, as he accepted in that sentence, the Bank would have a good reason for suing the English Defendants if it were also suing Mr Kolomoisky and Mr Bogolyubov, his conclusion that nonetheless the Bank had no reason for suing them in these proceedings beyond subjecting Mr Kolomoisky and Mr Bogolyubov to the jurisdiction of the English courts becomes difficult to sustain. We think there is force in these criticisms.

But, as Lord Pannick submitted, there is a more fundamental point to make about this part of the judgment. When the Bank commenced these proceedings, it had evidence which established a good arguable case that the English Defendants were controlled by Mr Kolomoisky and Mr Bogolyubov who had procured them to participate in a fraudulent scheme, but the Bank needs to prove its case. When it commenced the proceedings, it was not in a position to assume that, without more, it would succeed in establishing its case of a fraudulent scheme or that the English Defendants were the creatures that the Bank alleges them to be. In this connection, Lord Pannick drew attention to the fact that they commenced their own proceedings for defamation in Ukraine in October 2017, alleging damage to their business reputation, almost two months before the commencement of these proceedings. Of course, that may well be yet further evidence of the control exercised over them by Mr Kolomoisky and Mr Bogolyubov, but it creates more uncertainty about their role.

Nobody, including the judge, discounts the possibility that the English Defendants may have relevant documents to disclose. Mr Howard, speaking on this issue for all the defendants, made it clear that he was unable to say that they had no documents. There may be real benefit to the Bank in obtaining the disclosure of documents. Moreover, even if they have no documents to disclose, that very fact will assist the Bank to prove its case. If the English Defendants had played a genuinely commercial role in the transactions, one would expect that they would have relevant documents to disclose.

Mr Howard’s principal response to this was to ask why the Bank had not also sued all the companies alleged to have participated in the fraudulent scheme, numbering in total some 200 companies. This submission ignores that the English and BVI Defendants were the only Suppliers who received pre-payments that were not repaid. It was through them that the missing US$1.9 billion was finally dissipated and it has yet to be recovered. We do not find it surprising that in those circumstances the Bank should choose to concentrate its fire on those companies. It is fair to ask why the Bank did not also sue the sixcompanies to which it appears the English and BVI Defendants paid the pre-payments received by them, but their absence as defendants cannot drive us to conclude that the Bank had no reason for suing those Defendants other than to sue Mr Kolomoisky and Mr Bogolyubov in England.

In order to engage the sole object rule, the desire to sue Mr Kolomoisky and Mr Bogolyubov in England rather than Switzerland must be the sole object of the Bank in proceeding against the English Defendants in England. If the Bank had an additional reason to sue them here, albeit not as significant as bringing the non-domiciled defendants to the English court, the rule is not engaged. We are satisfied that, on an objective assessment, the ability to obtain disclosure from the English Defendants provided a real reason for bringing these proceedings against them.

For these reasons, we have concluded that the judge’s assessment cannot stand and that the correct assessment is that suing Mr Kolomoisky and Mr Bogolyubov in England was not the sole object of the Bank in commencing these proceedings.

The judge concluded that even if, contrary to his determination in relation to Mr Kolomoisky and Mr Bogolyubov, the English court had jurisdiction over them, he would have granted Mr Kolomoisky and Mr Bogolyubov a stay of the English proceedings on the grounds that there were pending proceedings in Ukraine. Those proceedings are defamation proceedings brought by Mr Kolomoisky against a Ukrainian journalist and the Bank in respect of a magazine article in which it was alleged that Mr Kolomoisky had siphoned off substantial funds from the Bank through the fraudulent scheme which is the subject of the Bank’s claim in the present proceedings. Thus, as the judge found (and the Bank does not challenge), there was a substantial overlap between the factual allegations in the two sets of proceedings. The English and BVI Defendants were joined as third parties to the defamation proceedings and the English Defendants commenced their own defamation proceedings in Ukraine. Mr Bogolyubov was also joined as a third party to the defamation proceedings but has not commenced his own defamation proceedings.

Whilst recognising that, since Ukraine is not an EU or Lugano Convention state, the provisions of article 28 of the Lugano Convention in relation to pending proceedings in another Convention state did not apply directly, the judge held that the article could be applied “reflexively” or by analogy, applying the analysis of Andrew Smith J in Ferrexpo AG v Gilson Investments Ltd [2012] EWHC 721 (Comm), [2012] 2 Lloyd’s Rep 588 (Ferrexpo). He stayed the proceedings against the English Defendants pursuant to article 34(1) of the Recast Brussels Regulation on the basis that there were “related proceedings” in Ukraine.

Having concluded that the court had no jurisdiction to try the claim against Mr Kolomoisky and Mr Bogolyubov and that the claim against the English Defendants should be stayed pending the final determination of the defamation proceedings in Ukraine, the judge also stayed the proceedings against the BVI Defendants on the grounds that England was not the convenient forum for the determination of the claim against them.

By Ground 2, the Bank seeks to challenge the judge’s conclusion on lis alibi pendens for four related reasons: (A) the judge erred in the exercise of his discretion in granting a stay of the claims against Mr Kolomoisky and Mr Bogolyubov and the English Defendants; (B) the judge erred in law in finding that the preconditions for the grant of a stay under article 28 of the Lugano Convention and/or article 34(1) of the Recast Brussels Regulation were fulfilled by the proceedings in Ukraine; (C) the judge erred in law in granting a stay of the Bank’s claims against Mr Kolomoisky and Mr Bogolyubov by applying article 28 of the Lugano Convention by analogy and/or reflexively and (D) the judge erred in law in finding that the requirements of “relatedness” in article 28 and article 34(1)(a) were fulfilled in circumstances where the Bank’s tort claims could not in fact be heard and determined together with the defamation proceedings in Ukraine.

Ground 3 contends that the judge was wrong to decline jurisdiction against the BVI Defendants on the grounds of forum non conveniens. It can be dealt with shortly as it essentially follows on from the first two grounds, in the sense that if the Bank is right that the English court has jurisdiction over the claims against Mr Kolomoisky and Mr Bogolyubov and that the judge erred in granting a stay against them and the English Defendants, it necessarily follows that the BVI Defendants are necessary or proper parties to the claims in the English proceedings against the other defendants, as the judge himself recognised at [166] and [170] of his judgment. On the other hand, if the judge was correct in granting a stay against Mr Kolomoisky and Mr Bogolyubov and against the English Defendants, he was also correct to stay the proceedings against the BVI Defendants.

Before dealing with these Grounds in more detail, we propose to set out the current status of the proceedings in Ukraine. In October 2018, between the hearing before the judge and the handing down of his judgment, the Ukrainian court dismissed Mr

Thus, whilst the judge was correct that there was no contested hearing on the merits, to the extent that [131] and [136] of his judgment suggest that the dismissal of the claims in Ukraine took no account of the merits (or lack of merits) of the claims, that fails to take account of the fact that (as that passage from its judgment shows) part of the reasoning for dismissing the claims was the adverse view formed by the Ukrainian court of the merits of the claim and its assessment that Mr Kolomoisky had no real interest in pursuing the claim. We were unimpressed by the submissions by Ms Sonia Tolaney QC (who argued this part of the appeal on behalf of the defendants) that the judge had not been incorrect in reaching that conclusion. Whether it was because of how the proceedings in Ukraine were presented to him or otherwise, the judge was wrong to conclude, as he clearly did, that the Ukrainian court had dismissed the claim on purely procedural grounds without any consideration of the merits.

The judge referred at [136] to the fact that Mr Kolomoisky and the English Defendants intended to appeal against the dismissal of the claim. There was indeed an appeal to the Court of Appeal in Ukraine where Mr Kolomoisky relied upon article 6 of the European Convention on Human Rights to contend that, because his claim had been struck out, he had not had a fair trial. Given the basis on which the court at first instance had struck out the claim, this argument was given short shrift and the Court of Appeal dismissed the appeal. In doing so, the Court agreed with the lower court that “the claim is clearly frivolous and meritless”. Both Mr Kolomoisky and the English Defendants (whose defamation proceedings were also struck out) launched cassation appeals to the Supreme Court of Ukraine from the decisions of the Court of Appeal.

The Supreme Court of Ukraine gave judgments on 7 August 2019 allowing the appeals of the English Defendants and on 11 September 2019 allowing the appeal of Mr Kolomoisky. In relation to Mr Kolomoisky (with whose claim in Ukraine we are primarily concerned) the reasons for allowing the appeal were in summary: (i) it had not been established that the applications made by him for the judge at first instance to recuse herself were an abuse of procedural rights; (ii) even if they were any abuse of procedural rights, the sanction was to strike out the applications not the whole claim; (iii) article 257 of the Ukrainian Civil Procedure Code sets out exhaustively the grounds for dismissing a claim without consideration, none of which applied here, so that the court of first instance was wrong to dismiss the claim without consideration; (iv) the conclusion of the court of first instance that the claim was artificial was based on assumptions by the court and without giving proper reasons; (v) the approach of the court of first instance violated Mr Kolomoisky’s rights under article 6(1) of the European Convention on Human Rights to a fair hearing. Accordingly the Supreme Court set aside the decisions of the court of first instance and the Court of Appeal and remitted the case to the court of first instance for fresh consideration.

The proceedings in Ukraine are clearly pending and were pending within the meaning of the Lugano Convention even before the appeal succeeded, as the judge found at [137] of his judgment. The Bank has abandoned any attempt to argue the contrary on this appeal. Ground 2(B) is thus in effect no longer pursued. The Bank also no longer pursues the other aspect of that Ground, that the judge erred in law in finding that

In relation to Ground 2, we will consider first the issue whether the judge was correct to apply article 28 of the Lugano Convention reflexively or by analogy in deciding that the court had jurisdiction to grant a stay of the present proceedings against Mr Kolomoisky and Mr Bogolyubov. In his oral submissions on this Ground, Lord Pannick addressed this issue first.

Article 28 provides as follows:

“Article 28

1.

Where related actions are pending in the courts of different States bound by this Convention, any court other than the court first seised may stay its proceedings.

2.

Where these actions are pending at first instance, any court other than the court first seised may also, on the application of one of the parties, decline jurisdiction if the court first seised has jurisdiction over the actions in question and its law permits the consolidation thereof.

3.

For the purposes of this Article, actions are deemed to be related where they are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings.”

It is common ground that since Ukraine is not an EU or Lugano Convention state, the provisions of article 28 of the Lugano Convention and article 34 of the Recast Brussels Regulation (in comparable terms) which concern pending proceedings in another Convention state, do not apply directly. The question which arises is whether article 28 should be applied reflexively or by analogy.

Lord Pannick submitted, by reference to a number of what he contended were general principles, that the provisions of the Lugano Convention and the equivalent EU Convention and Regulations should not be given reflexive effect. He submitted first that the jurisdiction provisions of the Lugano Convention (like the equivalent provisions in the Brussels Convention, in Brussels 1 and in the Recast Brussels Regulation) provide an exclusive code, which meant that it was not open to the court to create an additional exception to the default jurisdiction position under article 2, which is essentially what applying article 28 by analogy would be doing. He relied upon [16] of the judgment of the CJEU in Réunion Européenne to which we have already referred at [50] above. He also submitted that article 34(4) of Brussels 1 and the Lugano Convention, which expressly provide for recognition of a judgment in proceedings in a third state in certain circumstances, demonstrate that the drafters of the Conventions, when they wanted to address the position in third states, were quite capable of doing so, but did not do so in the context of either article 28 of the Lugano Convention or article 28 of Brussels 1 (in materially identical terms).

Lord Pannick submitted that it was significant that article 28 of Brussels 1 had been amended, since article 34 of the Recast Brussels Regulation (for which, see [182] below) did now provide for lis alibi pendens to be taken into account in relation to pending proceedings in a third, non-member state, whereas no equivalent amendment had been made to the Lugano Convention. The Standing Committee had been asked to look at whether the Lugano Convention should be amended in the light of the Recast Brussels Regulation, but made no recommendation as to any amendment.

In support of the Bank’s case that the court should not go beyond the terms of article 28 and seek to apply it by analogy to a case such as the present, Lord Pannick relied upon the principle of interpretation of an international treaty stated by Lord Bingham of Cornhill in R (European Roma Rights Centre) v Immigration Officer at Prague Airport [2004] UKHL 55, [2005] 2 AC 1 at [18]-[19], specifically the passage at [18]:

“However generous and purposive its approach to interpretation, the court's task remains one of interpreting the written document to which the contracting states have committed themselves. It must interpret what they have agreed. It has no warrant to give effect to what they might, or in an ideal world would, have agreed. This would violate the rule, also expressed in article 31(1) of the Vienna Convention, that a treaty should be interpreted in accordance with the ordinary meaning to be given to the terms of the treaty in their context.”

He submitted that article 28 meant what it said. It is confined to proceedings taking place in another Convention state and the court should not interpret it otherwise. The court had no jurisdiction to grant a stay in relation to proceedings other than those before another Convention state.

Lord Pannick also placed considerable reliance upon the decision of the CJEU in Owusu v Jackson (Case C-281/02) [2005] QB 801, that a national court with jurisdiction over a defendant under article 2 by virtue of his domicile in that state cannot decline jurisdiction on the grounds that the court of another state is the appropriate forum. At [37]-[38] and [41] of its judgment, in passages upon which Lord Pannick particularly relied, the Court said:

“37.

It must be observed, first, that Article 2 of the Brussels Convention is mandatory in nature and that, according to its terms, there can be no derogation from the principle it lays down except in the cases expressly provided for by the Convention [citations omitted]. It is common ground that no exception on the basis of the forum non conveniens doctrine was provided for by the authors of the Convention, although the question was discussed when the Convention of 9 October 1978 on the Accession of Denmark, Ireland and the United Kingdom was drawn up, as is apparent from the report on that Convention by Professor Schlosser.

38.

Respect for the principle of legal certainty, which is one of the objectives of the Brussels Convention [citations omitted] would not be fully guaranteed if the court having jurisdiction under the Convention had to be allowed to apply the forum non conveniens doctrine.

41.

Application of the forum non conveniens doctrine, which allows the court seised a wide discretion as regards the question whether a foreign court would be a more appropriate forum for the trial of an action, is liable to undermine the predictability of the rules of jurisdiction laid down by the Brussels Convention, in particular that of Article 2, and consequently to undermine the principle of legal certainty, which is the basis of the Convention_.”

The question which the CJEU was addressing concerned the application of the doctrine of forum non conveniens and the CJEU declined to answer a second question referred by the Court of Appeal, as to what the position would have been if proceedings had been pending in the third state, on the grounds that it was hypothetical, as no such proceedings were pending in the third state, there Jamaica. Nevertheless, Lord Pannick relied upon the fact that Advocate-General Léger had addressed that second question in his Opinion at [253]-[256], concluding that the Court was not entitled to decline jurisdiction under article 2 save in the special cases provided for by the provisions of the Brussels Convention which were the equivalent of articles 27 and 28 of the Lugano Convention, which did not arise in that case, because even if parallel proceedings were commenced, they would be in Jamaica, a non-contracting state. Lord Pannick submitted that Lewison LJ had been incorrect in Mittal v Mittal [2013] EWCA Civ 1255, [2014] Fam 102, when he had said at [37] that both the CJEU and the Advocate General had declined to answer the second question referred. Lord Pannick submitted that the Advocate General had addressed and advised on the second question.

Subsequent first-instance decisions have taken different views as to whether, even though the CJEU did not deal with the second question, the effect of the decision of the CJEU in Owusu v Jackson is that the English court should not decline jurisdiction under article 2 by applying the provisions of the Conventions concerned with proceedings in another Member or Convention state (here articles 27 and 28 in the Lugano Convention) by analogy. Lord Pannick relied upon the decision of Barling J in Catalyst Investment Group Ltd v Lewinsohn [2009] EWHC 1964 (Ch), [2010] Ch 218. In that case, the defendants applied to stay English proceedings on the grounds that the United States District Court in Utah was first seised of the dispute and the English court should apply article 27 of Brussels 1 (in materially identical terms to article 27 of the Lugano Convention) reflexively by analogy. The argument, as appears from [68] of the judgment, was not that the pending proceedings in Utah were a freestanding basis for a stay, but that the existence of those parallel proceedings meant that applying article 27 reflexively, the Court could apply conventional forum conveniens principles and stay its proceedings.

Barling J rejected that argument, saying at [71] that the submission that a reflexive use of article 27 so as to import a forum conveniens discretion was consistent with the aims of Brussels 1 was difficult to maintain in the light of the reasoning of the CJEU and the Advocate General in Owusu. He pointed out that if the argument were correct, the defendants in Owusu could have bypassed the effect of that decision by the simple expedient of commencing proceedings in Jamaica at any time.

Lord Pannick also relied upon what Barling J said at [99]:

“…it is not open to me to interpret and apply Article 27 reflexively so as to enable me to exercise a discretion to stay proceedings which have been properly founded on Article 2, on the grounds that the same dispute is pending between the same parties in the Utah courts and that the latter and not this Court is the natural and appropriate forum. Such an interpretation would introduce the wide forum conveniens discretion by the back door, contrary to the ruling of the ECJ in Owusu. In my view the submission fails whether this Court is the first or second seised.”

In giving the leading judgment in this Court in Lucasfilm v Ainsworth [2009] EWCA Civ 1328, [2010] Ch 503, Jacob LJ was critical, at least implicitly, of Barling J’s reasoning, saying at [134]:

“We do not have to decide whether that [decision] was correct, though we note that, if he is right, there is this oddity: that there is a clear lis pendens rule, with associated court first seized rule, for parallel cases within the EU but none for parallel cases where one is running within an EU Member State and one without. What Barling J did not decide was that Art. 2 conferred extra-EU subject matter jurisdiction generally.”

In JKN v JCN [2010] EWHC 843 (Fam) Lucy Theis QC sitting as a Deputy High Court Judge (as she then was) carried out a careful analysis of the authorities and the academic writings, some of which had been critical of the decision in Catalyst. She noted what Jacob LJ had said at [134] of Lucasfilm and declined to follow the decision of Barling J in Catalyst. At [149] of her judgment, she concluded that it was neither necessary nor desirable to extend the Owusu principle to cases where there are parallel proceedings in a non-Member state. She gave as her principal reasons for that conclusion:

“(a)

The risk of irreconcilable judgments which undermine two important objectives of the Brussels scheme namely: avoiding irreconcilable judgments between Member States and ensuring recognition of judgments between Member States.

(b)

It would lead to an undesirable lacuna, as there will be no mechanism in place for resolving this situation with the consequence of both proceedings continuing with the consequent increased uncertainty and cost.

(c)

The supporting rationale by Jacob LJ in Lucasfilm

'… the EU could not legislate for third countries' [111]:

'The Regulation is not setting up the courts of the Member States as some kind of non-exclusive world tribunals for wrongs done outside the EU by persons who happen to be domiciled within the EU.' [129]

[She then also cited [134] of Jacob LJ’s judgment which we have already quoted]

(d)

The reasoning that underpins Owusu is not incompatible with retaining the discretionary power where there are parallel proceedings in a non-Member State. It does not undermine certainty for the defendant (as he will be bringing the proceedings in the non-Member State); the claimant (although not mentioned in Article 2) will have knowledge of the proceedings in the non-Member State and it is likely to be in his interests to have one set of proceedings rather than two (the latter would happen if the Owusu doctrine was extended); there would be less risk of irreconcilable judgments given in Member States which are not recognised in another Member State; Coreck (which was decided 4 years before Owusu) permits judicial discretion in circumstances where there is no provision for it in Brussels 1.”

That reasoning was followed and approved by Andrew Smith J in Ferrexpo. That was a case where the defendants sought to rely by analogy on articles 22 and 28 of Brussels 1 in relation to proceedings in Ukraine where the dispute affected the constitution of a Ukrainian company. If the company had been incorporated in an EU Member state or a Lugano Convention state, article 22(2) of Brussels 1 and the Lugano Convention would have provided that the courts of that state had exclusive jurisdiction. Andrew Smith J held that article 22 should be applied reflexively by analogy, holding that the reasoning of Barling J (with which he did not agree) had no application to cases of the reflexive application of article 22 (see [137] of the judgment of Andrew Smith J).

In Ferrexpo the defendants also argued that the court should give reflexive effect to articles 27 and 28 of Brussels 1. Andrew Smith J concluded that those provisions should be given reflexive effect. He said at [163] that, because the CJEU in Owusu had not decided the second question referred to it, it did not decide whether it was consistent with the Brussels Convention for the court to stay proceedings on the grounds of lis alibi pendens in the court of a non-contracting state. He referred to the decision of Barling J in Catalyst and the contrary decision of Lucy Theis QC in JKN v JCN. He continued at [165] and [166]:

“165 The "general rule" is that where there are conflicting decisions of courts of co-ordinate jurisdiction, the later decision is to be preferred, if it is reached after full consideration of the earlier decision: Minister of Pensions v Higham [1948] 2 KB 153, page 155. There is good reason for the rule and I should depart from it only if convinced that Miss Theis was wrong: Colchester Estates v Carlton Plc [1986] Ch 80, page 85. I am not. On the contrary, I agree with both her decision and her reasoning. As Mr Malek submitted, many of the reasons for giving article 22 reflexive effect apply to article 28, and this conclusion reflects the opinion expressed in textbooks such as Dicey, Morris and Collins (loc cit at para 12-022) and Briggs and Rees (loc cit at para 2.260).

166 The connection between the principle of forum non conveniens and the principle of lis alibi pendens is undeniable, and was recognised by Hobhouse J in S & W Berisford Plc v New Hampshire Co [1990] 2 All ER 321 at page 331j and Potter J [in] Arkwright Mutual Insurance Co v Bryanston Insurance Co Ltd [1990] 2 All ER 335 at page 347 (as well as Barling J in the Catalyst case). Both doctrines are aspects of the law's general policy expressed by Bingham LJ as being "to favour litigation of issues only once, in the most appropriate forum": Du Pont v Agnew [1987] 2 Lloyd's Rep 585, page 589. The fact that there is lis alibi pendens might influence the court's decision on an application to stay on the grounds of forum non conveniens, as Barling J observed in Catalyst at para 109. But I see no difficulty in giving effect to the ECJ's injunction that because of article 2 a defendant cannot dispute that his domicile is an appropriate forum (and so not contend that it is forum non conveniens), but be protected from multiplicity of proceedings. Mr Smouha accepted that, notwithstanding Owusu, a defendant is entitled to have stayed proceedings against him that are abusive, and therefore would be entitled to a stay if a claimant has oppressively started proceedings both in another jurisdiction and here, but he distinguished the position where a defendant to English proceedings has brought the foreign proceedings, albeit before the proceedings were brought here. I cannot accept that the court's powers to protect a party from multiplicity of proceedings are subject to this limitation, or that, whereas Owusu allows protection in the former circumstances, it

prevents it in the latter.”

Lord Pannick’s overall submission was that Barling J had been correct and that the later decisions of Lucy Theis QC and Andrew Smith J were wrong, since the Lugano Convention provided an exclusive code as to jurisdiction and the only exceptions to article 2 jurisdiction were those expressly provided for by the Convention, as [37] of Owusu confirmed. In those circumstances, a reflexive application of article 28 was not permissible. He would, if necessary, go so far as to submit that reflexive effect could not be given either to article 22 of the Lugano Convention (specific cases of exclusive jurisdiction) or article 23 (exclusive jurisdiction clauses) though he contended that it was not necessary to go that far for his argument in relation to article 28 to be correct.

He submitted that there was no binding decision of this Court which supported the analysis of Andrew Smith J in Ferrexpo. In Huawei Technologies Co Ltd v Conversant Wireless Licencing S.A.R.L. [2019] EWCA Civ 38, at [29] of his judgment Floyd LJ, giving the leading judgment, stated that the conclusion of Andrew

The only other decision of the Court of Appeal which was of any relevance to this issue was Mittal v Mittal [2013] EWCA Civ 1255, [2014] Fam 102. Lord Pannick submitted that that case concerned the provisions of Brussels II in relation to family law and Lewison LJ, delivering the main judgment, was careful not to be drawn into the wider question as to the effect of the CJEU judgment in Owusu in relation to Brussels 1. As Lewison LJ said at [41] of his judgment:

“It is not necessary for us to be drawn into a wider debate (which Ms Theis also considered) on the extent to which Owusu v Jackson applies to the Judgments Regulation; and anything I might say on that topic would be simply obiter.”

So far as the general principles for which Lord Pannick contended are concerned, there is no doubt that the various Conventions (including the Lugano Convention) provide an exclusive code as regards jurisdiction for the matters which they cover, namely nationals of and proceedings before Member and Convention states. However the Conventions, including the Lugano Convention, do not purport to cover proceedings in third states and nothing in the language of the Conventions precludes the application of their provisions by analogy. Were the position otherwise, the default jurisdiction provision in article 2 would have the very extra-territorial effect in relation to non EU or Lugano Convention states which Jacob LJ deprecated in Lucasfilm at [134] of his judgment, cited at [170] above.

Nor do we consider that the application of the Lugano Convention by analogy somehow subverts the principles enunciated by Lord Bingham in the Roma Rights case. The application of articles 22 and 23 or 27 and 28 of the Lugano Convention by analogy to cases involving potential or actual proceedings in third states is not an interpretation of the Convention which involves an impermissible extension of its scope, but a recognition that the same principles which underlie those articles should be applicable in the case of proceedings pending in a third state. This approach does not subvert the Convention but, on the contrary, is in line with its purposes, to achieve certainty in relation to jurisdiction and to avoid the risk of inconsistent judgments: see Briggs: Civil Jurisdiction and Judgments 6^th edition (2015) para 2.05; JKN at [149] and Ferrexpo at [126]-[127] and [165]-[166].

Lord Pannick’s submission that the application of articles of the Conventions by analogy is impermissible because each of the relevant Conventions is an exclusive code proves too much. As he was constrained to recognise, the logical consequence of the submission is that even article 22 (the various cases of exclusive jurisdiction) and article 23 (exclusive jurisdiction clauses) could not have reflexive effect. However, that those articles should be given reflexive effect in an appropriate case has been recognised, albeit obiter, by this Court. Thus, in Masri v Consolidated Contractors International [2008] EWCA Civ 303, [2008] 2 Lloyd’s Rep 128 at [125], having noted that whether or not the particular provision of article 22 had reflexive effect did not arise in that case, Lawrence Collins LJ said:

“The problem has long been recognised: see Dicey, paras 12021-12-022; 23-026-027. The way in which it arises can be illustrated by two examples. First, Article 22(1) of the Brussels I Regulation gives exclusive jurisdiction, in the case of proceedings which have as their objects rights in rem in immovable property, to the courts of the Regulation State where the property is situate. What if the defendant is domiciled in England and is sued in England, and the land is in a non-Regulation State, such as Canada? Second, Article 23 provides that if the parties, one or more of whom are domiciled in a Regulation State have agreed that the courts of a Regulation State are to have jurisdiction to settle any disputes which may arise between them. What if they have designated the courts of a State which is not a Regulation State, such as the courts of New York and an action is brought in England in breach of the jurisdiction agreement? In such cases it would be odd if the Brussels I Regulation did not permit the English court to stay its proceedings.”

We have already referred at [175] above to the passage from the judgment of Floyd LJ in Huawei where the analysis of Andrew Smith J as to why article 22 should have reflexive effect was approved.

We do not consider that there is any significance in the fact that article 34 of the Recast Brussels Regulation now specifically addresses pending proceedings in a third state, whereas the Lugano Convention continues not to do so. Not only is there nothing to indicate why the amendment to what is now article 34 of the Recast Brussels Regulation was made (as Lord Pannick accepted), but it simply does not follow that, prior to the amendment, there could not have been a reflexive application of what was then article 28 of Brussels 1 or article 28 of the Lugano Convention to proceedings in a third state. As Mr Daniel Jowell QC for Mr Bogolyubov (who dealt with the reflexive effect of article 28 on behalf of all the defendants) submitted, amendments to the wording of European Conventions are often clarifications of matters previously addressed in the case law. We would also not read anything into the fact that the Standing Committee did not make any recommendations as to amendment of the Lugano Convention, as there is no material available from which we could deduce what the rationale was for its position.

In our judgment, the analysis of Andrew Smith J in Ferrexpo as to the reflexive effect of article 28 is correct. Contrary to what Barling J appears to have thought in Catalyst, we do not consider that giving reflexive effect to the article would lead to uncertainty or be inconsistent with the purpose of the various European Conventions. On the contrary, giving reflexive effect to the article in relation to pending proceedings in a third state will avoid the risk of inconsistent judgments. Accordingly, we consider that the judge in the present case was correct in his conclusion at [125] of his judgment that article 28 should be given reflexive effect.

We propose to consider Ground 2(D) next. This concerns a threshold question to the application of both article 28 of the Lugano Convention and article 34 of the Recast Brussels Regulation, whether the English proceedings and the Ukrainian defamation proceedings are “related” as defined by article 28(3): “actions are deemed to be related where they are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings.” Article 34(1) of the Recast Brussels Regulation expresses the same concept in a slightly different way:

“1.

Where jurisdiction is based on Article 4 [domicile] or Articles 7, 8 or 9 and an action is pending before a court of a third State at the time when a court in a Member State is seised of an action which is related to the action in the court of the third State, the court of the Member State may stay the proceedings if:

a)

It is expedient to hear and determine the related actions together to avoid the risk of irreconcilable judgments resulting from separate proceedings;

b)

It is expected that the court of the third State will give a judgment capable of recognition and, where applicable, of enforcement in that Member State; and

c)

The court of the Member State is satisfied that a stay is necessary for the proper administration of justice.”

The Bank argues that the actions are not “related” in the sense that it is expedient to hear and determine them together, because consolidation of the Bank’s claim with Mr Kolomoisky’s claim in the defamation proceedings would not be possible. It is submitted that unless the two actions can be consolidated and actually heard together, it is not “expedient” to hear and determine them together. In other words, the Bank submits that expediency in this context means practicability. The judge, following the decision of Eder J in Nomura International Plc v Banca Monte Dei Paschi Di Siena SpA [2013] EWHC 3187 (Comm), [2014] 1 WLR 1584, rejected this argument, concluding that the fact that actions cannot in practice be heard together does not mean that it is not “expedient” to hear and determine them together. The Bank submits that this conclusion is wrong and urges the court to conclude that expediency is to be equated with practicability.

As with the issue of the reflexive effect of articles in the European Conventions, this is an issue on which there are conflicting decisions at first instance. In Cardosa de Pina v MS “Birka” Beutler Sciffahrts KG [1994] I. L. Pr. 694, the English court was asked to stay a negligence action against his employer by a seaman who had been injured on board a ship, in circumstances where he had previously commenced an action in Germany against the Compensation Board for Seamen. The evidence was that a claim against the employer could not be brought in the German proceedings, because the liability of the employer for non-international accidents had been replaced by an action against the Compensation Board (see [15] of the judgment of Richard Buxton QC, sitting as a Deputy High Court Judge). The judge concluded, albeit with some hesitation, that the two actions were not “related”. At [16] he said:

“The words in the Article must be given proper weight. They envisage actions as being related to each other which could, in the circumstances in which they are brought, be tried together. Otherwise it cannot be “expedient” to try them together. It simply seems to me to be the case that these two actions cannot be so tried.”

Centro Internationale Handelsbank AG v Morgan Grenfell Trade Finance Limited [1997] CLC 870 was a case where, having commenced proceedings in Italy against an export credit insurer, the benefit of whose policy had been sold to the plaintiff bank by the defendant under a forfaiting transaction, the plaintiff bank later commenced proceedings in England against the defendant for breach of the forfaiting contract and in restitution. The plaintiff sought a stay of the English proceedings on the basis that the actions were related actions under article 22 of the Brussels Convention. The defendant resisted the grant of a stay on a number of grounds, including that it was only if the two actions could be combined in one jurisdiction that it could be said to be “expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings.”

Although he refused to grant a stay for other reasons, Rix J rejected that argument. He disagreed with the decision in Cardosa di Pina and concluded that it was not right that the actions were only related if they could both be brought in Italy and consolidated. At p.889B he said:

“…the test of whether ‘it is expedient to hear and determine them together’ …may perhaps be glossed as whether the two actions should be heard and determined together (to avoid the risk of irreconcilable judgments); but in my judgment there is a real difference between asking whether the actions can be brought together and asking whether they should be brought together.”

The Bank relied upon Haji-Ioannou v Frangos [1998] CLC 61, where the defendant sought a stay under article 22 of the Brussels Convention of what was in essence a breach of trust claim against him in England on the grounds that the plaintiff had previously instigated criminal proceedings in Greece where the public prosecutor recommended that no charges be brought against the defendant, a recommendation which was accepted by the Greek criminal court, upheld on appeal. At the time of the application to the English court, an appeal by the plaintiff to the Greek Supreme Court was pending. The judge, Neuberger J, concluded that article 22 did not apply and the English proceedings and the Greek criminal proceedings were not “related actions”:

“not least because the substantive relief sought in the instant proceedings, which is essentially of an equitable and tracing nature, could not apparently be pursued under the Greek civil law, and could therefore not be the subject of a claim in the Greek criminal proceedings…Obviously, the Greek criminal proceedings could not be heard in this country. Accordingly, I do not see how the Greek criminal proceedings and the instant proceedings could be ‘heard and determined together’” (p.72BC).

It does not appear that either Cardoso de Pina or Centro Internationale was cited to the court in that case and, perhaps for that reason, the analysis of the judge does not focus, as at least Rix J had, on the meaning of the word “expedient” in the article.

The Bank also relied upon the decision of Cooke J in JP Morgan v Primacom AG [2005] EWHC 508 (Comm), [2005] 2 Lloyd’s Rep 665. There, two sets of proceedings had been commenced in Germany, notwithstanding an exclusive English jurisdiction clause in the English law contract. Three separate sets of English proceedings were then commenced. The defendants applied to stay those proceedings. The judge concluded that the declaratory proceedings were based on the same cause of action as the German proceedings and ordered a compulsory stay under Brussels 1. He then went to consider the position under article 28 of Brussels 1 but clearly this part of his decision was obiter. At [57], he said:

“If I had not found that the Declaratory proceedings involved the same cause of action as the Mainz proceedings, I would have found that the two actions were connected but not that it was expedient to hear and determine them together to avoid the risk of irreconcilable judgements resulting from separate proceedings. I have in mind the comments of Lord Saville in Sarrio SA v Kuwait Investment Authority[1999] AC 32at page 40, but here the two actions are not capable of being heard and determined together in any real sense of the word because the Declaratory Proceedings will proceed on the basis of English law whilst the German actions proceed on the basis of the validity of the SSFA as a matter of English law but then seek to apply German public policy considerations.”

All these earlier authorities were considered carefully by Eder J in Nomura. Having done so, he concluded at [57]:

“Against this somewhat chequered and uncertain background and in the absence of clear authority to the contrary, it seems to me that the answer to Mr Handyside's threshold point ultimately lies in the wording of Article 28(3). As I have already said, the focus of that wording is in my view what in principle is expedient which I read in the sense of genuinely desirable, not what is "capable" or "possible". For that simple reason, I am unable to accept Mr Handyside's threshold point. Thus it is my conclusion that the existence of an exclusive jurisdiction clause in favour of the court second seised does not of itself mean that proceedings commenced in that court may not be "related" to proceedings in another court for the purposes of Article 28(3).”

We agree with Ms Tolaney QC that the approach of Rix J in Centro Internationale and of Eder J in Nomura is to be preferred to that of Mr Buxton QC, Neuberger J and Cooke J. Both Rix J and Eder J correctly focus on the language of article 28/article

The word “expedient” is more akin to “desirable”, as Rix J put it, that the actions

“should” be heard together, than to “practicable” or “possible”, that the actions “can” be heard together. We also consider that there is force in Ms Tolaney’s point that, if what had been intended was that actions would only be “related” if they could be consolidated in one jurisdiction, then the Convention would have made express reference to the requirement of consolidation, as was the case in article 30(2) of the Recast Brussels Regulation.

Accordingly, on this threshold issue, we consider that the judge was right to conclude that the actions were related, even if they could not be consolidated, so that the judge did have jurisdiction to grant a stay in the present case. However, the fact that the actions could not be consolidated was relevant to the exercise of discretion, the final issue on Ground 2 to which we now turn.

Having reached the conclusion that the English proceedings and the defamation proceedings in Ukraine were “related” proceedings or actions, the judge turned to consider the exercise of discretion as to whether to grant a stay. He concluded at [150] to [153] that a judgment of the Ukraine court would be recognised in England, a conclusion which the Bank no longer challenges. He then went on to consider whether a stay was necessary for the proper administration of justice. He considered that there was a substantial degree of relatedness between the Bank’s claim and the defamation claim in Ukraine, although the causes of action were different, another conclusion which the Bank does not challenge. He then concluded that there was a clear risk of irreconcilable judgments on the issues as to whether there was a fraudulent scheme, who set it up and operated it, how it worked and its purpose, who benefited from it and how much money was unlawfully removed (see [144] and [155]).

He considered that, if any appeal in Ukraine were successful (as it ultimately has been), it might still be more likely that a final determination in Ukraine would be obtained before the trial of the Bank’s claim in England, but as a result of the appeal in Ukraine, there was no longer a significant difference in timescale and he recognised that as a result of the dismissal of the defamation claim there might never be a judgment on the merits (see [157]). On the issue of proximity the issues raised in the two sets of proceedings were almost exclusively concerned with events in Ukraine, the majority of witnesses would be Ukrainian and Ukrainian law would apply, whereas, in contrast, the issues have no connection with England and the existence of the English Defendants was of no materiality. The proximity of the claim to Ukraine pointed strongly in favour of a stay (see [158]).

The judge noted at [159] that the Bank nevertheless argued that a stay would be contrary to the proper administration of justice. It contended that the proceedings cried out for determination by a truly independent tribunal, although it did not contend that the Ukrainian court would be unable to resolve the issues or that the Bank could not obtain justice in Ukraine.

The judge referred at [160] to Lord Clarke’s observation (in The Alexandros T [2013] UKSC 70, [2014] 1 Lloyd’s Rep 223 at [27]) that the aim of the Lugano Convention and Recast Brussels Regulation is to avoid parallel proceedings and conflicting decisions. The question of whether a stay should be granted should be considered in that light. The Judge said that it might be said by the Bank that the current proceedings would take so long to come to trial that there was no real risk of mutually irreconcilable decisions because any decision in the Ukrainian proceedings would be

The judge considered at [161] that the argument against a stay would have greater weight, if the stay granted under article 34 was a once and for all decision, but it was not, because under article 34(2), the current proceedings could be continued at any time when it was appropriate to do so, and therefore potential prejudice to the Bank in granting a stay was limited. If the appeal in Ukraine was dismissed or Mr Kolomoisky did not properly pursue the claim to judgment, the grounds for continuing a stay were likely to fall away. Accordingly, the judge concluded at [162] that, if he had concluded that the Court had jurisdiction to try the claim against Mr Kolomoisky and Mr Bogolyubov, he would have granted a stay of the claim under the inherent jurisdiction or under CPR 3.1(2)(f) by analogy with article 28 of the Lugano Convention. A stay of the claim against the English Defendants would be granted under article 34 of the Recast Brussels Regulation. In each case the stay would not be final, but only while the defamation proceedings in Ukraine were pending and subject to review, depending on future developments in Ukraine.

The principal argument advanced by Lord Pannick on this part of the Bank’s appeal was that the exercise of discretion by the judge to grant a stay was plainly wrong on the facts. The judge had failed to stand back and ask himself whether, in all the circumstances, it was appropriate to stay an English fraud claim, involving what he found was fraud and money laundering on an “epic scale”, where he found the Bank had a good arguable claim for US$515 million, in favour of Ukrainian defamation claims. The answer was “obviously not”.

In relation to the dismissal of the Ukrainian proceedings at the time of the hearing before the judge, Lord Pannick submitted that the judge proceeded on the fundamentally erroneous basis that the Ukrainian court had dismissed the claim only on a procedural ground, without any consideration of the merits, whereas in fact the court had examined the case file and concluded that the claim was frivolous and fabricated and that Mr Kolomoisky had no genuine intention of pursuing the claim to judgment. It was plainly wrong to grant a stay in favour of Ukrainian proceedings which had been dismissed on that basis (as well as on procedural grounds), but the judge had not taken into account at all the fact that the Ukrainian court did consider the claim was unmeritorious.

The Bank also submitted that the judge had erred in the exercise of his discretion in failing to recognise that, even though the fact that the Bank’s claim and the defamation claim could not be consolidated did not preclude their being related actions or proceedings for the purposes of article 28 of the Lugano Convention and article 34 of the Recast Brussels Regulation, the fact that consolidation was not possible was an important factor militating against the grant of a stay, when it came to the exercise of discretion as to whether to do so. This was recognised by both Rix J in Centro Internationale and Eder J in Nomura.

In Centro Internationale at p.889G-H, Rix J stated:

“Where…an action is merely stayed on the court’s own motion, the court need not at that time be satisfied that both actions can be brought together in the court first seised; and the court may subsequently be persuaded, by evidence that the two actions cannot in fact be brought together in the court first seised, to revoke its stay.”

Likewise in Nomura at [45], Eder J said:

“…Article 28(3) would appear to be focussing on the question of principle i.e. whether the actions are so closely connected that it would be expedient to hear and determine them together to avoid the risk of irreconcilable judgments from separate proceedings. In particular, it does not seem to me necessarily to follow from the fact that it may not be capable of hearing the actions together in the court first seised because, as Mr Handyside would submit, of the existence of the jurisdiction clauses (in particular the exclusive jurisdiction clause in the

ISDA Master Agreement governing the Asset Swap

Transactions) that it would not, in principle, be expedient to do so. In my view, the focus of the definition (if that is what it is) in Article 28(3) is what in principle is "expedient" in the sense of genuinely desirable not what is "capable" or "possible". Of course, the fact that it may not be capable or possible to hear the actions together in the court first seised may provide an unanswerable or at least compelling argument in favour of the court second seised exercising the discretion under Article 28(1) to refuse a stay.”

The Bank submitted that, in failing to take account of this “compelling argument” against the grant of a stay, the judge had erred in the exercise of his discretion in another respect. It was submitted that this Court should exercise the discretion afresh and decline to grant a stay.

The defendants challenged the Bank’s argument that the judge had exercised his discretion in a fundamentally erroneous manner because he proceeded on the basis that the Ukrainian defamation proceedings had been dismissed on procedural grounds only, without any consideration of the merits. Ms Tolaney QC argued this part of the appeal on behalf of the defendants. As we have already noted at [155] above she sought to argue that the judge’s conclusions at [131] and [136] of his judgment about the basis upon which the claim was dismissed were not incorrect. We do not accept that submission. It is quite clear that the judge did proceed on the incorrect basis that the Ukrainian court had not considered the merits of the claim, whereas in fact it had, determining that the claim was frivolous and fabricated.

Next Ms Tolaney submitted that, provided that proceedings were still pending, notwithstanding that they had been dismissed, because an appeal was possible, the exercise of discretion as to whether to grant a stay should not be influenced by the reason for which the proceedings had been dismissed. Accordingly, she submitted that, even if the judge had erroneously assumed that the Ukrainian court had not considered the merits of the claim, he had still exercised his discretion correctly under

She also submitted that the present case was to be contrasted with the decision of this Court in Easy Rent a Car Ltd v Easygroup Ltd [2019] EWCA Civ 477, [2019] 1 WLR 4630, upon which Lord Pannick had relied. In that case the proceedings in Cyprus had been dismissed and the appellants were appealing that dismissal. However there was evidence before this Court which, as David Richards LJ said at [70] of his judgment, made the position “fundamentally different” from the situation which existed at the time of the hearing before HHJ Hacon at first instance, namely that it would be three to four years before the appeal in Cyprus was decided. David Richards LJ continued at [71]:

“In all the circumstances of this case, the fact that there will be no Cypriot proceedings unless the appellants succeed in their appeal in Cyprus, and the fact of the very lengthy delay in any progress in the Cypriot proceedings even if the appeal is allowed, are overwhelming factors against the grant of a stay of the English proceedings under article 30.”

Ms Tolaney submitted that in contrast with that case, here the appeal to the Supreme Court of Ukraine was imminent (as to which she proved correct) and there was no question of a very lengthy delay in the progress of the proceedings in Ukraine.

In relation to the question whether a stay should have been refused because consolidation of the Bank’s claim with the defamation claim would not be possible in Ukraine, Ms Tolaney submitted that neither Rix J nor Eder J was laying down an immutable principle that, if consolidation were not possible, a stay should be refused. It all depended on the circumstances. The judge had adopted the correct approach to this issue at [145], where he said at the end of the paragraph:

“If there were shown to be a real possibility of consolidation of the related proceedings, that would be an additional, strong reason for exercising the discretion in favour of a stay, but it is

not a condition of such exercise.”

Attractively though Ms Tolaney’s submissions were presented, we cannot accept them. For whatever reason, the judge did not appreciate that the court of first instance in Ukraine had considered the merits of the claim and had concluded that it should be dismissed not only on the grounds of Mr Kolomoisky’s procedural abuse of process, but on the grounds that the claim is frivolous and fabricated and that Mr Kolomoisky had no genuine intention of pursuing the proceedings to judgment. Accordingly, the judge exercised his discretion to grant a stay on a fundamentally erroneous basis, as Lord Pannick submitted. Had the judge appreciated that the court in Ukraine considered the claim unmeritorious on those grounds, in exercising his discretion correctly he should have refused to grant a stay. This conclusion is unaffected by the fact that, subsequent to his judgment, the Supreme Court of Ukraine has allowed Mr Kolomoisky’s appeal against the striking out of his claim.

We consider that the judge also erred in his exercise of discretion in relation to the fact that the two sets of proceedings could not be consolidated and heard together in Ukraine. The expert evidence of Ukrainian law from Mr Beketov was that the Pechersky District Court, the court of first instance before which the defamation claims are proceeding, does not have jurisdiction to hear the Bank’s claim, which would have to be brought before the Ukrainian commercial court. That evidence was challenged by Mr Kolomoisky’s expert who referred to certain cases where Ukranian civil courts had apparently heard claims for which they did not in fact have jurisdiction. We were not persuaded by any of that evidence and consider that the better view is that expressed by Mr Beketov that the claims could not be consolidated for the reasons he gives.

Whilst Ms Tolaney is no doubt correct that neither Rix J nor Eder J was laying down a rule of law, what [45] of Eder J’s judgment demonstrates is that, absent some strong countervailing factor, the fact that proceedings cannot be consolidated and heard together will be a compelling reason for refusing a stay. The problem here is that the judge seems to have considered the exercise of discretion from the wrong end of the telescope: he concluded that the availability of consolidation would be a strong reason to grant a stay, but its unavailability would not in itself be a reason not to grant a stay. He thus erroneously failed to consider that, as Eder J had held, unavailability of consolidation will usually be a compelling reason to refuse a stay. There was certainly no strong countervailing factor in this case pointing in favour of a stay.

In our judgment, although the appeal of Mr Kolomoisky in Ukraine has been allowed and the matter remitted to the court of first instance, so that this court should proceed on the basis that the proceedings in Ukraine will continue and be pursued to judgment, the unavailability in the Ukrainian court of consolidation of the Bank’s current claim with Mr Kolomoisky’s defamation claim remains a compelling reason for refusing to grant a stay. In particular, the fact that the Bank’s claim would have to be brought before the Ukrainian commercial court rather than before the Pechersky District Court in which the defamation proceedings are being heard means that if a stay were granted, the risk of inconsistent findings in these different courts would remain. Furthermore, we accept Lord Pannick’s overall submission that, standing back in this case, it would be entirely inappropriate to stay an English fraud claim in favour of Ukrainian defamation claims, in circumstances where the fraud claim involves what the judge found was fraud and money laundering on an “epic scale” and where, as we have concluded, the Bank has a good arguable case to recover the pleaded sum of US$1.9 billion. We consider that for those reasons, in exercising the relevant discretion afresh, this court should refuse to grant a stay.

Accordingly, we consider that Ground 2(A) of the appeal succeeds and overall the appeal on this Ground should be allowed. The judge’s decision that he would have granted a stay in favour of Mr Kolomoisky and Mr Bogolyubov by analogy with article 28 of the Lugano Convention and his decision to grant a stay in favour of the English Defendants under article 34 of the Recast Brussels Regulation cannot stand and should be set aside to enable the claim to proceed in England against Mr Kolomoisky and Mr Bogolyubov and the English Defendants.

Given our conclusion that Ground 2 succeeds and that the English claim against Mr Kolomoisky and Mr Bogolyubov and the English Defendants should be allowed to proceed, it inevitably follows that the BVI Defendants are necessary or proper parties to that claim and that the judge was wrong to conclude that the proceedings against the BVI Defendants should be set aside or stayed. This inevitable conclusion is essentially accepted by Ms Tolaney on behalf of the BVI Defendants. Accordingly, Ground 3 also succeeds.

As mentioned earlier, the judge considered that there was “no difficulty with the Bank’s proving a good arguable case of a fraudulent scheme” ([25]). He concluded, however, that the Bank’s “realistically arguable claim” in these proceedings is only US$514,965,295 rather than the US$1.9 billion put forward in the amended particulars of claim. As the judge saw things, the Bank’s claim is “limited to funds under Relevant Loans being misappropriated by being paid to and not repaid by the English and BVI Defendants” ([56]). That meant that, “unless the Bank has a good arguable case that it was Relevant Loan monies that were paid to the English and BVI Defendants as the Unreturned Prepayments, its case cannot be proved as a matter of fact” ([56]).

There was “a good arguable case that at most about US$515m of Relevant Loans passed through [the English Defendants’ and BVI Defendants’] accounts in such a way as to be capable of causing loss to the Bank in connection with those loans”

In the circumstances, the judge considered that the Bank had not established a good arguable case for tortious loss in excess of US$514,965,295 ([58]). As regards the Bank’s alternative claim for unjust enrichment, the judge said this in [60]:

“The unjust enrichment claim … appears hopeless when it is considered that, contrary to the appearance given by the Particulars of Claim, the English and BVI Defendants were no more than conduits for the US$1.91 billion to pass through en route for immediate return to the borrower or transfer to other offshore corporate entities (and then in many cases immediate return to the borrower and the Bank). The English and BVI Defendants were not enriched by acquiring US$1.91 billion or any money or rights. Para 79 of Mr Beketov’s first expert report [i.e. evidence as to Ukrainian law] emphasises that a Defendant sued for unjust enrichment has to have benefited from the Bank’s loss. It is clear, when the evidence about the nature of the scheme and what happened to the Relevant Loan money is properly analysed, that the English and BVI Defendants did not do so.”

The Bank now challenges the judge’s views. Its contentions can be conveniently discussed under the following headings:

The Bank’s most fundamental contention is that the judge was mistaken in thinking that its case depended on showing that the Unreturned Pre-payments were made to the English and BVI Defendants from funds drawn down under the Relevant Loans. Mr Andrew Hunter QC, who argued this part of the case for the Bank, explained that, while the Bank alleges that the English and BVI Defendants did receive misappropriated funds, this is not a necessary part of the Bank’s case. In contrast, Mr Howard, who presented this part of the case for the defendants, endorsed the judge’s analysis.

Mr Howard took us to various paragraphs of the particulars of claim to support his submission that the Bank’s claims against all the defendants are (as the judge thought) based on receipt by the English and BVI Defendants. The passages to which we were referred included these:

Mr Howard also addressed us on [54] of the particulars of claim, which reads:

“The Defendant Suppliers procured and/or assisted in the misappropriation of $1,911,877,385 from the Bank and procured and/or assisted in the subsequent concealment of the same. For the reasons particularised below, the actions, decisions and/or omissions of the Defendant Suppliers were unlawful because they deprived the Bank of its monies in a manner not provided for by the Constitution or other laws of Ukraine (contrary to Article 3(1)(2) of the Civil Code) and acted in a manner that violated the Bank’s rights and with the intention to injure the Bank (contrary to Article 13 of the Civil Code).

Mr Howard submitted that it is not only sub-paragraphs (b) and (c) of [54] which relate to receipt. The assistance in misappropriation alleged in sub-paragraph (a) can only, Mr Howard said, be in having received and paid away the funds.

For his part, Mr Hunter maintained that the Bank’s primary case against the English and BVI Defendants is that they are liable because they assisted Mr Kolomoisky and Mr Bogolyubov in their fraudulent scheme with “unity of intent”. One of the matters on which it relies in support of that case is the allegation that the English and BVI Defendants are to be seen as having received proceeds of the fraud, but, Mr Hunter said, the Bank’s case would stand even if none of the Relevant Loans had made their way to the English and BVI Defendants. While the Bank alleges receipt by the English and BVI Defendants, that represents no more than a particular of the key allegation of assistance. The Bank, Mr Hunter stressed, has never advanced any proprietary claim, and its tortious claim is for compensation for the amount drawn down under the Relevant Loans which was not repaid. The particulars of claim, Mr Hunter argued, include allegations that the English and BVI Defendants received proceeds of the fraud, but the Bank’s case does not depend on that having been the case.

In our view, Mr Hunter’s submissions are borne out by the terms of the particulars of claim. Thus:

Mr Howard suggested that the judge’s interpretation of the Bank’s case is consistent with the evidence and submissions on which the Bank relied at the without notice hearing before Nugee J on 19 December 2017. As Mr Howard pointed out, Lewis (1), which was the main evidence before Nugee J, contained various references to the English and BVI Defendants having received proceeds of Relevant Loans. Thus, Mr Lewis said, for example, that “approximately US$1.9 billion of the Bank’s money was paid out to the Defendant Suppliers” ([27]), that funds lent to the Borrowers “were paid out to the Defendant Suppliers” ([24]) and “ultimately transferred to the Suppliers” ([250]), and that the pre-payments to the English and BVI Defendants “emanated from lending advanced to [the Borrowers] by the Bank” ([49]) and were made “using funds that the Bank advanced to the Borrowers under certain loan facilities (i.e. the Relevant Loans)” ([101]). In a similar vein, the Bank’s skeleton argument for the hearing before Nugee J spoke of the English and BVI Defendants having “received the Bank’s monies, albeit via the Borrowers” ([24]) and of the English Defendants having “received and transferred the Bank’s monies (purportedly pursuant to the Prepayment Supply Agreements)”.

On the other hand, the Bank’s skeleton argument did not frame its claim against the English and BVI Defendants exclusively by reference to the English and BVI Defendants’ alleged receipt of “the Bank’s monies”. The Bank was said to have tortious claims against the various defendants on the basis that they had “caused harm to the Bank by their unlawful decisions, actions and/or omissions, contrary to Article 1166 of the Ukraine Civil Code” ([5]), with “those who cause harm by collective actions” being “jointly and severally liable for the same” ([19.3]). More specifically, the English and BVI Defendants were “closely involved in the Misappropriation” because they “entered into the Relevant Supply Agreements”, “neither delivered the goods that they said they would deliver or returned the prepayments they had received” and “entered into the Loan File Supply Agreements for the apparent purpose of disguising the Misappropriation”, with the result that each was likely to be “jointly and severally liable for the entire sum misappropriated from the Bank” ([21]). At the hearing before Nugee J, Mr Stephen Smith QC, who was then appearing for the Bank, agreed with Nugee J that the Bank’s case in tort or delict was that “the suppliers are parties to a conspiracy, effectively”, going on to say:

“I mean, Ukrainian law doesn’t use the word ‘conspiracy’ but that’s effectively what it is. They are all jointly and severally liable for inflicting harm on the bank.”

In the circumstances, the case that the Bank advanced to Nugee J was not wholly based on the English and BVI Defendants having received proceeds of the Relevant Loans. In any case, we do not think that the scope of the Bank’s pleaded case can be controlled by what was said on the Bank’s behalf at that stage. Moreover, materials put before Fancourt J by the Bank asserted that its case against the English and BVI Defendants did not depend on establishing that they had received misappropriated funds. Mr Beketov, the Bank’s expert on Ukrainian law, explained in a report dated 6 July 2018 that “[i]t is not necessary for the Bank’s claim in tort … that any of the

Defendants received any part of the very same monies (in the proprietary sense) transferred from the Bank to the Borrowers under the Relevant Loans, or indeed that any of the Defendants received any benefit from the Relevant Loans at all” ([10]) and that, “for the purposes of a claim in tort under Article 1166 of the Civil Code, it is sufficient that the Defendants’ unlawful acts and/or omissions have caused harm to the Bank” ([11]). Mr Beketov went on in [12]:

“Even if the defendant Suppliers received prepayments under the Supply Agreements that, from a tracing standpoint, consisted of monies that were not extracted from the Bank via the Relevant Loans, this would not change my prior analysis. It would still remain the case that (i) each Defendant Supplier participated in the misappropriation of the Bank’s monies via the Relevant Loans, and (ii) the Defendant Supplier’s failure to perform the Supply Agreement after receiving prepayment thereunder, by either providing the relevant goods to the Borrower or returning the prepayment, deprived the Bank of the value of its collateral under the pledge agreements and the ability to pursue the Borrower for return of the loaned funds, thereby assisting and facilitating the misappropriation of its property.”

In keeping with Mr Beketov’s report, the Bank’s written submissions for the hearing said that the Bank alleged that the English and BVI Defendants “assisted in the fraud perpetrated by [Mr Kolomoisky and Mr Bogolyubov] … and are jointly and severally liable for the losses that have been caused by it” ([94]), adding (in [95]):

“In addition, and though it is not necessary for the Bank to show as much, the Bank can also demonstrate that the drawdowns of the Relevant Loans and the transfers to [the English and BVI Defendants] are transactionally linked.”

Similarly, in the course of the hearing Mr Smith QC, appearing for the Bank, told the judge that it was “not a necessary element of the claim in tort” that the money that the English and BVI Defendants received was the same as that which was lent to the Borrowers under the Relevant Loans, observing that “Ukrainian law doesn’t require, for the purposes of a tort claim, the very same money has to pass from the borrowers to the suppliers” and that there is no need to undertake a “payment tracing analysis”.

In short, we agree with Mr Hunter that the judge was mistaken in thinking that the Bank’s case depended on showing that the Unreturned Pre-payments were made to the English and BVI Defendants from funds drawn down under the Relevant Loans.

As already mentioned, the judge considered that the Bank could not have a strong prima facie to recover more than US$1,484,463,028 since the defendants had “sufficiently shown that, of the total funds drawn down under the Relevant Loans, US$1,010,956,101 came back to the Bank and discharged Relevant Loans of Ukrainian borrowers” (see [215] above). The Bank disagrees.

An important strand in the Bank’s contentions on this point is the proposition that, as a matter of Ukrainian law, a purported repayment of a Relevant Loan can be disregarded if the repayment was made with funds derived from another fraudulent loan. In this respect, the Bank relies on evidence from Mr Beketov, who expressed the opinion in his report of 6 July 2018 that, “where monies purportedly used to repay an unlawful loan are themselves referable to other, unlawful lending, as a matter of Ukrainian law the repayment would not be considered to operate to reduce the Bank’s loss” ([15]).

Mr Hunter gave a hypothetical example. Suppose, he said, that there is a fraudulent scheme under which borrowers A and B each draw down US$100 million and borrower A purports to repay his loan using the money that borrower B has drawn down. As a matter of Ukrainian law, Mr Hunter said, the lender would be entitled to ignore the purported repayment of the loan to borrower A. Putting matters slightly differently, Mr Hunter argued that the overall scheme would have left the lender with a loss of US$100 million, and it would be open to it to attribute that to the loan to borrower A.

Mr Hunter illustrated the practical significance of the issue to the present case by reference to the Bank’s dealings with AEF LLC (“AEF”), one of the Borrowers. The Lafferty Speadsheets record that AEF drew down sums totalling US$45 million both on 7 October 2013 and on 27 December 2013. Also on 27 December AEF made a pre-payment of US$45 million to Kalten Trade SA (“Kalten”), apparently in respect of contract ST-01/77. That same day, according to the Lafferty Spreadsheets, Kalten returned to AEF a US$45 million pre-payment it had received on 7 October in respect of contract ST-01/65 and AEF made loan repayments to the Bank. The Bank, Mr Hunter submitted, is entitled to treat the repayments as relating to the drawdowns of 7 October rather than those of 27 December. There is at least a good arguable case to that effect, Mr Hunter said.

Mr Howard countered that, looking at matters in a common sense way, AEF can be seen to have repaid the US$45 million that it had borrowed that very day rather than any earlier drawdown. Come trial, the defendants may succeed on the point. It nonetheless seems to us that the Bank has a good arguable case in this respect. The Bank’s contentions derive support from Mr Beketov’s expert evidence.

We consider that, contrary to the judge’s view, the Bank’s claim cannot at this stage be reduced on the basis that the Relevant Loans have been repaid to an extent greater than the Bank’s US$1.9 billion figure assumes.

The judge considered that it had been “plausibly shown” that large amounts of the net drawdowns under the Relevant Loans did not reach the English and BVI Defendants and that the Bank’s claim fell to be scaled back correspondingly. The significance of this in the present context is much reduced by our conclusion on the pleading point, but we should nevertheless address the rival contentions that the parties put forward on this aspect of the appeal.

Mr Hunter placed in the forefront of his submissions the fact that the Bank’s case reflects contemporary documents. One of these is a spreadsheet (“the Gurieva Spreadsheet”) found on the laptop of Ms Tetiana Gurieva. According to evidence adduced by the Bank, she was a trusted confidante of Mr Kolomoisky and Mr Bogolyubov who headed a “secret” unit within the Bank referred to as “BOK” which dealt with lending to companies ultimately owned by Mr Kolomoisky and Mr Bogolyubov. The spreadsheet had been deleted from her laptop, but was recovered.

The Gurieva Spreadsheet has a number of columns. The first lists the Borrowers, the second gives “loan as of today (19/11/2014)” in respect of each Borrower, the next four appear to contain details of outstanding pre-payments to the various English and BVI Defendants (and no one else), and the final four seem to provide information about antecedent transactions. With regard to AEF, for example, the Gurieva Spreadsheet records a loan of “931,471,735” (presumably, Ukrainian Hryvnias), a pre-payment to the fourth defendant of US$45 million in respect of contract PT-13/06 on 26 June 2014 and (seemingly) return by Kalten of a pre-payment that AEF had previously made to it in respect of contract ST-01/11. The total of the pre-payments outstanding from the English and BVI Defendants is specified as US$1,911,877,385, the figure used in the particulars of claim.

Mr Hunter also placed reliance on judgments which the Borrowers obtained against the English and BVI Defendants in Ukrainian proceedings in the latter part of 2014. To take AEF again, it claimed (and was granted judgment against the fourth defendant) for pre-payments totalling US$45 million which it had made pursuant to contract PT-13/06 on 26 and 27 June 2014. The judgment further stated that AEF and the Bank (which was also a party to the proceedings) had “entered into loan agreements No.4A13578D of 20.09.13 for 34,000,000.00 USD; No.4A13579D of 24.09.13 for 33,000,000.00 USD; and No.4A13580I of 24.09.13 for 460,000,000.00 UAH [i.e. Ukrainian hryvnia]”. Such judgments, Mr Hunter suggested, confirm that pre-payments outstanding from the English and BVI Defendants are appropriately to be linked with loans made to Borrowers much earlier in time, notwithstanding intervening transactions.

Mr Hunter explained that the Bank’s approach is to seek to identify what transactions in substance resulted from fraudulent drawdowns and to determine whether those transactions resulted in a loss to the Bank. On the Bank’s approach, Mr Hunter said, what is relevant is whether there was a transactional and therefore causal link between a fraudulent drawdown and an Unreturned Pre-payment such that the latter should be regarded as resulting from the fraudulent drawdown for the purpose of assessing the Bank’s loss. Looking at matters in this way, it does not matter, Mr Hunter argued, whether a pre-payment was made with the very money that was the subject of the fraudulent drawdown.

Mr Hunter sought support in a passage from one of Mr Kolomoisky’s skeleton arguments in which this is said:

“If it was not the same money drawn under the Relevant Loans as was used to make the Unreturned Prepayments, then it is unclear how the Unreturned Prepayments could nevertheless ‘be regarded as derived from’ the Relevant Loans. The Relevant Loans may have caused the Unreturned Prepayments, but not the other way round – and thus it is not the Unreturned

Prepayments which cause the Bank’s loss.”

Mr Hunter focused on the words, “The Relevant Loans may have caused the Unreturned Prepayments”, which, he said, confirm that the Bank has a good arguable case on the causation it alleges.

Mr Howard decried the approach espoused by the Bank. It ignored the reality, he suggested. By way of example, he referred to the transactions mentioned in [232] above and, specifically, to those of 27 December 2013. Once, he argued, Kalten had repaid US$45 million to AEF and AEF had in turn made repayments to the Bank, there was no money left. There can therefore, so it was said, be no question of the US$45 million which AEF drew down on 27 December having reached the English and BVI Defendants even if that indebtedness remained outstanding. Regardless of whether the US$45 million paid to the Bank on 27 December is deemed to have related to that day’s drawdowns or those of 7 October, the simple fact is that the proceeds of the 27 December drawdowns cannot have been received by the English and BVI Defendants, Mr Howard submitted.

Mr Hunter responded that Kalten had received two pre-payments from AEF of US$45 million each (on respectively 7 October and 27 December) so that it owed AEF US$45 million even after it had returned £45 million on 27 December. That it may not have had US$45 million (or any money) in its bank account after it had paid the US$45 million to AEF is, Mr Hunter suggested, neither here nor there. Kalten remained a debtor to the tune of US$45 million, and the Bank can properly link that indebtedness to payments totalling US$45 million which Kalten made to AEF on 2627 June 2014, which were immediately followed by Unreturned Pre-payments in the same sums to the fourth defendant.

A further problem with Mr Howard’s approach relates to the derivation of the Lafferty Spreadsheets. The judge referred to their having been prepared “with the benefit of work done by expert analysts” ([38] of the judgment). Mr Lafferty himself explained in a witness statement that “Mr Kolomoisky personally and a team put together by him” had been in contact with “former employees and clients of the Bank” who had provided them with “the Bank’s transactional data, relevant information and explanations” but who were unwilling to be identified and that the transactional data had been analysed by teams “engaged by Mr Kolomoisky” including “former employees of the Bank, who were aware of the transaction patterns … and the companies involved in the Scheme” (see [17]-[20] of Mr Lafferty’s statement of 9 March 2018). There is thus reason to believe that the exercise was undertaken by individuals who were themselves involved in the alleged fraud.

Mr Howard countered that the Lafferty Spreadsheets were a convenient way of presenting information in bank statements to which the Bank itself had access and that, as the judge noted in [70], Mr Lewis’s second and third witness statements did not challenge the factual analysis of the drawdowns and pre-payments presented by Mr Lafferty. It can also be observed, however, that neither Mr Kolomoisky nor anyone else who played a part in the preparation of the Lafferty Spreadsheets has attempted to explain in evidence the myriad of transactions having no evident commercial logic that features in the Lafferty Spreadsheets. In the circumstances, the judge ought, we think, to have approached the Lafferty Spreadsheets with caution.

The courts have often warned against conducting mini-trials at an interlocutory stage. In Swain v Hillman [2001] 1 All ER 91, Lord Woolf MR explained (at 95) that “the

In our view, the judge was wrong to attach the importance he did to the Lafferty Schedules and to reject the Bank’s analysis in favour of the defendants’ approach to tracing. The relevant facts are not “so plain that they can be readily and summarily established” and the issues “should be more properly reserved for the trial itself”. If Mr Howard’s contentions have plausibility, as the judge thought, so too do Mr Hunter’s. The validity and significance of Mr Hunter’s approach need to be assessed, not at this interim stage, but at trial, with full evidence and a proper understanding, informed by expert evidence, of what matters for the purposes of the Ukrainian-law claims that the Bank is bringing. It is neither appropriate nor even possible to say now which of the rival arguments are right. The Bank’s case on linking may or may not prove to be well-founded at trial, but it cannot be dismissed at this stage.

As mentioned earlier, the judge characterised the Bank’s alternative claim in unjust enrichment as “hopeless”. We have already said that we disagree. It seems to us that the Bank has a good arguable case in unjust enrichment.

We grant permission to appeal on Ground 4 of the Grounds of Appeal and consider that this Ground of Appeal has been made out.

The judge concluded that there had been serious breaches of the Bank’s duty of disclosure in respect of its without notice application to Nugee J on 19 December 2017. The judge considered, moreover, that the breaches “were deliberate at least in the sense that the Bank deliberately presented its case in a particular way and did not draw attention to material that would cast real doubt on the validity of its claim” ([83]

The relevant legal principles were summarised in these terms by Ralph Gibson LJ in Brink’s Mat Ltd v Elcombe [1988] 1 WLR 1350 (at 1356-1357):

“In considering whether there has been relevant non-disclosure and what consequence the court should attach to any failure to comply with the duty to make full and frank disclosure, the principles relevant to the issues in these appeals appear to me to include the following. (1) The duty of the applicant is to make ‘a full and fair disclosure of all the material facts:’ see Rex v. Kensington Income Tax Commissioners, Ex parte Princess Edmond de Polignac [1917] 1 K.B. 486, 514, per Scrutton L.J.

(2)

The material facts are those which it is material for the judge to know in dealing with the application as made: materiality is to be decided by the court and not by the assessment of the applicant or his legal advisers: see Rex v. Kensington Income Tax Commissioners, per Lord CozensHardy M.R., at p. 504, citing Dalglish v. Jarvie (1850) 2 Mac. & G. 231, 238, and Browne-Wilkinson J. in Thermax Ltd. v. Schott Industrial Glass Ltd. [1981] F.S.R. 289, 295.

(3)

The applicant must make proper inquiries before making the application: see Bank Mellat v. Nikpour [1985] F.S.R. 87. The duty of disclosure therefore applies not only to material facts known to the applicant but also to any additional facts which he would have known if he had made such inquiries.

(4)

The extent of the inquiries which will be held to be proper, and therefore necessary, must depend on all the circumstances of the case including (a) the nature of the case which the applicant is making when he makes the application; and (b) the order for which application is made and the probable effect of the order on the defendant: see, for example, the examination by Scott J. of the possible effect of an Anton Piller order in Columbia Picture Industries Inc. v. Robinson [1987] Ch 38; and (c) the degree of legitimate urgency and the time available for the making of inquiries: see per Slade L.J. in Bank Mellat v. Nikpour [1985] F.S.R. 87, 92–93.

(5)

If material non-disclosure is established the court will be ‘astute to ensure that a plaintiff who obtains [an ex parte injunction] without full disclosure … is deprived of any advantage he may have derived by that breach of duty:’ see per Donaldson L.J. in Bank Mellat v. Nikpour, at p. 91, citing Warrington L.J. in the Kensington Income Tax Commissioners’ case [1917] 1 K.B. 486, 509.

(6)

Whether the fact not disclosed is of sufficient materiality to justify or require immediate discharge of the order without examination of the merits depends on the importance of the fact to the issues which were to be decided by the judge on the application. The answer to the question whether the nondisclosure was innocent, in the sense that the fact was not known to the applicant or that its relevance was not perceived, is an important consideration but not decisive by reason of the duty on the applicant to make all proper inquiries and to give careful consideration to the case being presented.

(7)

Finally, it ‘is not for every omission that the injunction will be automatically discharged. A locus poenitentiae may sometimes be afforded:’ per Lord Denning M.R. in Bank Mellat v. Nikpour [1985] F.S.R. 87, 90. The court has a discretion, notwithstanding proof of material non-disclosure which justifies or requires the immediate discharge of the ex parte order, nevertheless to continue the order, or to make a new order on terms.

‘when the whole of the facts, including that of the original non-disclosure, are before [the court, it] may well grant … a second injunction if the original non-disclosure was innocent and if an injunction could properly be granted even had the facts been disclosed:’ per Glidewell L.J. in Lloyds Bowmaker Ltd. v. Britannia Arrow Holdings Plc., ante, pp.

1343H–1344A.”

In the same case, Slade LJ said (at 1359):

“The principle is, I think, a thoroughly healthy one. It serves the important purposes of encouraging persons who are making ex parte applications to the court diligently to observe their duty to make full disclosure of all material facts and to deter them from any failure to observe this duty, whether through deliberate lack of candour or innocent lack of due care.

Nevertheless, the nature of the principle, as I see it, is essentially penal and in its application the practical realities of any case before the court cannot be overlooked. By their very nature, ex parte applications usually necessitate the giving and taking of instructions and the preparation of the requisite drafts in some haste. Particularly, in heavy commercial cases, the borderline between material facts and non-material facts may be a somewhat uncertain one. While in no way discounting the heavy duty of candour and care which falls on persons making ex parte applications, I do not think the application of the principle should be carried to extreme lengths.”

More recently, Males J (as he then was) drew attention to the following points in National Bank Trust v Yurov [2016] EWHC 1913 (Comm) (at [18]):

“a.

A fact is material if it is one which the judge would need (or wish) to take into account when deciding whether to make the freezing order.

b.

Failure to disclose a material fact will sometimes require immediate discharge of the order. This is likely to be the court’s starting point, at least when the failure is substantial or deliberate.

c.

Nevertheless the court has a discretion to continue the injunction (or to impose a fresh injunction) despite a failure of disclosure; although it has been said that this discretion should be exercised sparingly, the overriding consideration will always be the interests of justice.

d.

In considering where the interests of justice lie, it is necessary to take account of all the circumstances of the case including (without attempting an exhaustive list) (i) the importance of the fact not disclosed to the issues which the judge making the freezing order had to decide; (ii) the need to encourage proper compliance with the need for full and frank disclosure and to deter non-compliance; (iii) whether or to what extent the failure to disclose was culpable; and (iv) the injustice to a claimant which may occur if an order is discharged leaving a defendant free to dissipate assets, although a strong case on the merits will never be a good excuse for a failure to disclose material facts.

e.

The interests of justice may sometimes require that a freezing order be continued, but that a failure of disclosure be marked in some other way, for example by a suitable order as to costs.” 253.Males J further said this (at [84]):

“In this context a failure may be regarded as ‘innocent’ if the fact in question was not known to the applicant or its relevance was not perceived. That was the sense in which the word was used by Ralph Gibson LJ in Brink’s Mat at 1357D and the judgments of Balcombe LJ at 1358G and 1360H were to the same effect. Moreover in Behbehani v Salem [1989] 1 WLR 723 Woolf LJ expressly rejected at 728F-G a submission that a failure could not be regarded as innocent if the fact in question was not recognised as material but ought to have been, while

Nourse LJ observed at 736F that ‘in the Brink’s Mat case all three members of the court defined an innocent non-disclosure as one where there was no intention to omit or withhold information which was thought to be material’. This formulation would rightly include as culpable blind eye knowledge, that is to say a decision not to investigate for fear of discovering facts which would have to be disclosed, but that is not this case. I am satisfied that all three failures in this case were innocent in the sense described.”

In broad terms, the judge’s criticisms of the Bank have two themes. One is that the Bank failed to disclose the extent of repayments to Borrowers and the Bank. The other is that the Bank exaggerated the role of the English and BVI Defendants.

With regard to the first of these, the judge said this in [76]:

“In my judgment, there were material facts on which the Defendants would have relied and of which the Bank was aware that should have been but were not explained to the Judge [i.e. Nugee J]. These were, in particular: that a very substantial proportion of the sums drawn down under the Relevant Loans were repaid to the Bank by the very same borrower, or by another of the forty-six Ukrainian borrowers, prior to the relevant prepayments to the English and BVI Defendants, and that in a substantial number of cases monies prepaid to the English and BVI Defendants between May and September 2014 were themselves, either directly or indirectly, repaid to the borrower and to the Bank, often on the same day. The Bank must have been aware of this because it would inevitably have sought to identify where the money went. There is evidence that it did carry out an exercise of following the money in the case of AEF LLC. These facts would have been immediately apparent from a first study of the bank statements of the borrowers and the English and BVI Defendants (and their alleged principals), all of which held accounts with the Bank. If the Bank did not actually know those facts then it undoubtedly should have done because it should have carried out investigations before applying for the extensive relief that it did.”

In paragraph [79], the judge said:

“A more complete factual picture would have led the Judge to question whether the quantum of any arguable claim of the Bank (on the basis pleaded) was not considerably overstated, for all the reasons that have led me to conclude that there is no good arguable case (on the basis pleaded) in excess of about US$515 million.”

The judge was approaching matters, of course, on the basis that the amount for which the Bank had a good arguable case did not exceed US$515 million, in part because Relevant Loans had been repaid. We have taken a different view, concluding that the Bank has a good arguable case to recover the full US$1.9 billion given in the particulars of claim. That, though, is by no means a complete answer to the judge’s concerns. The fact that an applicant may have a good arguable case does not excuse him from disclosing matters that might be thought adverse to him.

Any complaint that the Bank did not explain to Nugee J that “in a substantial number of cases monies prepaid to the English and BVI Defendants between May and September 2014 were themselves, either directly or indirectly, repaid to the borrower and to the Bank, often on the same day” can nonetheless be disposed of shortly. Mr

Turning to whether the Bank was at fault in failing to disclose matters indicating that “a very substantial proportion of the sums drawn down under the Relevant Loans were repaid to the Bank by the very same borrower, or by another of the forty-six Ukrainian borrowers, prior to the relevant prepayments to the English and BVI Defendants”, the Bank specifically highlighted the possibility of “Potential defences” including that the defendants’ conduct “caused no harm because the Relevant Loans were repaid” both in [25.1] of its skeleton argument for the hearing before Nugee J and in [339]-[340] of Lewis (1)on which the Bank was relying. It is fair to say that the Bank did not refer to the specific repayment argument which the defendants have developed by reference to the Lafferty Spreadsheets, but we do not think it was incumbent on it to do so. The case that the Bank was advancing on repayment accorded with contemporary documents (the Gurieva Spreadsheet and the judgments obtained by the Borrowers in Ukraine). Moreover, we do not think that the defendants’ repayment defence would have been “immediately apparent from a first study of the bank statements of the borrowers and the English and BVI Defendants (and their alleged principals)”. The Lafferty Spreadsheets, which the defendants use to support their case, will have taken weeks to prepare even with input from people who were familiar with the relevant transactions, having, it seems, been involved in them. In any event, it cannot, in our view, be inferred that failure to disclose matters relating to the repayments that the defendants allege was deliberate.

There remains to be considered the question of whether the Bank exaggerated the role of the English and BVI Defendants. In this connection, the judge said in [77] that the

For the reasons we have given earlier in this judgment, we do not agree with the judge that the unjust enrichment claim is hopeless. The main ingredients of the judge’s criticisms, however, were to the effect that the Bank failed to disclose (a) the fact that the alleged fraud had extended over a number of years and involved a multitude of companies, (b) the fact that the English and BVI Defendants were not in a significantly different position from other participants in the scheme which were not being joined as defendants and (c) what had happened to the pre-payments that the English and BVI Defendants had received.

Taking (a) first, the evidence before Nugee J consisted principally of Lewis (1). This affidavit, which Nugee J was asked to read during his day’s pre-reading, explained that loan facilities granted to “46 Ukrainian companies” (i.e. the Borrowers) had been used to make payments to “35 companies … incorporated in jurisdictions including England, the BVI and Cyprus” (i.e. the Suppliers) ([93]), that the Bank’s transactional data suggested that “this system of prepayment and return of prepayment that characterises the early part of the Lending Scheme may have been implemented between certain of the Borrowers and Suppliers as early as 2008” ([97]) and that, “during a period of almost six years, using loans advanced by the Bank, tens to hundreds of millions of US Dollars at a time were moved repeatedly between the accounts of the Borrowers and the Suppliers”, with “just under US$13.2 billion [being] transferred from the accounts of the Borrowers to the accounts of 35 Suppliers during this period, and just over US$11.2 billion in total [being] transferred to the same Borrowers’ accounts” ([270]). Echoing that evidence, Mr Smith referred during the without notice hearing to the existence of 35 suppliers to which payments had been made, adding that “until 28 May 2014 no payments appeared to have been retained by any of the 35 … original suppliers”. In the circumstances, it seems to us that the Bank sufficiently revealed the fact that the alleged fraud had lasted for some years and involved Suppliers and other companies which were not being sued.

Moving on to (b), the English and BVI Defendants could, in our view, legitimately be seen as in a different position from other Suppliers and, indeed, other companies which had played a part in the allegedly fraudulent scheme. The Bank’s approach was consistent with the picture painted by the Gurieva Spreadsheet and the Ukrainian judgments which the Borrowers had obtained. Those identified the English and BVI Defendants as having received pre-payments which had not been returned and, moreover, portrayed the English and BVI Defendants as the only Suppliers in that position. All other Suppliers appear to have repaid pre-payments in full within about 90 days. On top of that, it was the English Defendants, and not other Suppliers, which entered into the Loan File Supply Agreements.

With regard to (c), there can be no question of the Bank leading Nugee J to believe, or of Nugee J in fact thinking, that the English and BVI Defendants still had the Unreturned Pre-payments. Mr Lewis referred in Lewis (1) to the modest net assets apparent from the English Defendants’ filed accounts, and Nugee J mentioned this feature during the hearing before going on to observe to Mr Smith, “They [i.e. the

As for what had become of the money paid to the English and BVI Defendants, Mr Lewis said this, albeit only in a footnote to Lewis (1):

“As is explained below, the Bank has attempted to discover what happened with the funds paid to the Defendant Suppliers, and has some information in that regard as the money was paid to them in PrivatBank Cyprus accounts. It appears, however, that the money was split up and paid on quickly to a significant number of further companies in such a way that it has not been possible to discover where it now resides, but it is anticipated that one or more of the Defendants will know, and that the grant of freezing and disclosure orders – before there is time to dissipate the money any further – will provide the Bank with the best chance of securing the funds.”

The judge had a particular concern that it was not explained to Nugee J that the English and BVI Defendants were mere “conduits” through whose accounts money “passed fleetingly”. The footnote to Lewis (1) quoted in the previous paragraph answers that criticism to an extent, since Mr Lewis referred to the money being “split up and paid on quickly to a significant number of further companies”. In a broadly similar vein, the Bank’s skeleton argument for the hearing before Nugee J said that “the funds passed through the Defendant Suppliers” ([22]) and Nugee J spoke of money having “ended up going through, even if it wasn’t still in, English companies”.

On balance, we consider that the Bank should have gone further than it did and spelled out more clearly the fact that pre-payments passed through the English and

Since we have arrived at substantially different conclusions from Fancourt J on the non-disclosure allegations, we must decide for ourselves whether the WFO granted by Nugee J should be discharged and, if so, whether injunctive relief should be regranted. We have concluded that it is not appropriate to discharge Nugee J’s order. While we take the view, as we have said, that the Bank should have spelled out more clearly the fact that pre-payments passed through the English and BVI Defendants’ hands only “fleetingly” before being passed on to other entities, we do not think that the failure was deliberate or that it can have affected Nugee J’s order. It does not, as it seems to us, warrant the discharge of a freezing order made in support of a claim for US$1.9 billion in respect of which there is a good arguable case.

Accordingly, we shall grant permission to appeal on Ground 5 of the Grounds of Appeal and consider that this Ground has also been made out.

We therefore allow the appeal.

I share the views expressed in the main judgment, to which I have myself contributed, subject only to one point. That point, which does not affect the way in which the appeal should be disposed of, relates to the issue discussed in [31]-[111] above: whether article 6(1) of the Lugano Convention is subject to a “sole object” test. David Richards and Flaux LJJ consider that a claimant with a sustainable claim against an anchor defendant, which it intends to pursue to judgment in proceedings to which a foreign defendant is joined as a co-defendant, is entitled to rely on article 6(1) even though the claimant’s sole object in issuing the proceedings against the anchor defendant is to sue the foreign defendant in the same proceedings. I respectfully disagree.

Writing in 2015, Professor Adrian Briggs referred to the law in this area as a “sorry mess” (see Civil Jurisdiction and Judgments, 6^th edition, at [2.231]). That still seems a fair description, notwithstanding the CJEU’s subsequent decision in Cartel Damage.

Professor Briggs observed that in Freeport the CJEU had “ruled, clearly and precisely, that even if the action against the Swedish defendant had been brought solely to remove the English co-defendant from the court which would otherwise be competent in a claim against him, this did not make what is now Article 8(1) [of the Recast Brussels Regulation] inapplicable”, the “requirement for connexity, first expressed in Kalfelis, then taken up into the legislated terms of the Article itself”, being “considered sufficient to prevent any misuse of this rule of special jurisdiction” (Civil Jurisdiction and Judgments, at [2.231]. As, however, Professor Briggs also noted, the CJEU did not say in Freeport that Reisch Montage had been wrongly decided and, despite Freeport, it stated in both Painer and Solvay that article 6(1) of Brussels I “cannot … be applied so as to allow an applicant to make a claim against a number of defendants with the sole object of ousting the jurisdiction of the courts of the State where one of those defendants is domiciled” (see the judgment in Painer at [78]). In the circumstances, Professor Briggs said that it seemed that “this non-textual requirement must now be taken to be part of the law” (Civil Jurisdiction and Judgments, at [2.231]). That, it appears, was also Professor Pocar’s view when he wrote the Pocar Report in 2009 (see [56] above).

Cartel Damage confirms that the provisos to article 6(1) of the Lugano Convention and article 8(1) of the Recast Brussels Regulation do not state the limitations on those articles comprehensively. Each article says that a person who is one of a number of defendants may be sued in the courts for the place where any of them is domiciled “provided the claims are so closely connected that it is expedient to hear and determined them together to avoid the risk of irreconcilable judgments resulting from separate proceedings”. It is clear from Cartel Damage, however, that there are circumstances in which a claimant cannot rely on article 6(1) or article 8(1) (as the case may be) despite the “closely connected” requirement being met. What is at issue is not whether the articles are subject to an implied exception at all but the extent of that exception.

As is pointed out at [86] of the main judgment, paragraphs [25] to [29] of the judgment in Cartel Damage are critical. The CJEU there concluded that a court could reject a claim brought on the strength of what is now article 8(1) of the Recast

Mr Howard, however, argued that in Cartel Damage the CJEU did no more than restate the “sole object” test. According to Mr Howard, a claimant may be said to have “artificially fulfilled” the requirements of article 6(1) or article 8(1) where his sole object in suing the anchor defendant was to allow him to claim to have met those requirements.

On balance, I agree that that is the preferable view. In [27] of its judgment, the CJEU recorded, citing Reisch Montage and Painer, that “settled case law” was to the effect that what is now article 8(1) of the Recast Brussels Regulation “cannot be interpreted as allowing an applicant to make a claim against a number of defendants for the sole purpose of removing one of them from the jurisdiction of the courts of the member state in which that defendant is domiciled”; it did not voice any criticism of the “sole purpose” test established by the “settled case law” or, more specifically, of Reisch Montage. In the following paragraph of its judgment, the CJEU took Freeport as authority for the proposition that, if claims are “closely connected” in the way laid down in article 6(1) and article 8(1), the article will apply “without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled”. The point, as I see it, was that there was no need “to establish separately” that the claimant did not have “the sole object of ousting the jurisdiction of the courts of the member state where one of the defendants is domiciled”, not that, were that “sole object” to be evident, it would not matter. In the circumstances, the conclusion in [29] that circumvention can be found “only where there is firm evidence to support the conclusion that the applicant artificially fulfilled, or prolonged the fulfilment of, that provision’s applicability” is, I think, to be read as referring to the need for “firm evidence” rather than marking a rejection of a “sole object” test. The Court was not, to my mind, introducing a new, and narrower, “artificially fulfilled, or prolonged the fulfilment of”, requirement, but rather saying that a court before which proceedings had been brought in pursuance of article 6(1)/8(1) need not concern itself with the claimant’s object in the absence of strong evidence on the matter. Reisch Montage and Freeport were to be reconciled in that way, not by departing from the “settled case law” indicating (as the CJEU saw things) that article 6(1)/8(1) would not apply if it was apparent that the claimant had the “sole purpose of removing [a defendant] from the jurisdiction of the courts of the member state in which that defendant is domiciled”. The significance of Freeport, as interpreted in Cartel Damage, is thus that the court in which a claim has been brought under article 6(1)/8(1) need not investigate the claimant’s object routinely.The CJEU may perhaps have used the language of “artificial” fulfilment instead of “sole object” in [29] of its judgment because that was felt to fit the facts of the case, which involved allegations of collusion. I do not believe that the CJEU was intending to hold that a claimant could proceed under article 6(1)/8(1) even if there was “firm evidence” that he was suing the anchor defendant for the “sole purpose of removing [another defendant] from the jurisdiction of the courts of the member state in which that defendant is domiciled”

As explained in the main judgment, in Vedanta Lord Briggs noted at [34] that in Cartel Damage the CJEU “add[ed] … that in the context of cartel cases nothing short of collusion between the claimant and the anchor defendant would be sufficient to engage the abuse of law principle” and said at [36] that “[s]uch jurisprudence as there is about abuse of EU law in relation to jurisdiction suggests that the abuse of law doctrine is limited to the collusive invocation of one EU principle so as improperly to subvert another”. He also, however, expressly “[left] aside those cases where the claimant has no genuine intention to seek a remedy against the anchor defendant” when making his observations at [40] and his (tentative) remarks were in any event obiter. While, therefore, analysis by Lord Briggs is plainly deserving of great respect, Vedanta does not cause me to alter the conclusions I arrived at in the previous paragraph.

In all the circumstances, my own view is that the fact that a claimant’s sole object in issuing proceedings against an anchor defendant was to sue the foreign co-defendant in the same proceedings would be enough to render article 6(1) of the Lugano Convention inapplicable. Professor Briggs may well be right, however, that the “sole object” test will in practice turn out to set a “standard which is hard to attain” (see Civil Jurisdiction and Judgments, at [2.231]).