Merinson v Yukos International UK BV & Ors

Regulation (EU) No. 1215/2012 of the European Parliament and of the Council (“Brussels Recast”) and its predecessors introduced jurisdiction rules differing markedly from those hitherto prevailing at common law. In place of flexibility and judicial discretion (including the doctrine of forum non conveniens), fixed rules were introduced, prioritising certainty and predictability, in accordance with the philosophy underpinning Brussels Recast. It is those fixed rules which have given rise to the present dispute. Under common law there would have been little doubt that the present Action, by one route or another, would have been litigated in the Netherlands. But the position may be different under Brussels Recast, as indeed Mr Richard Salter QC, sitting as a Deputy Judge of the High Court held, in his clear and comprehensive judgment dated 27 February 2018 (“the judgment”) [2018] EWHC 335 (Comm); [2018] QB 1113.

This appeal from that judgment involves the contention by a former employee that proceedings brought by the former employer should not proceed before this Court - the Court of the Member State where the former employee was domiciled at the time the Claim Form was personally served on him.

The Appellant (“Mr Merinson”) was employed by various entities within the Respondent group of companies (“Yukos”) from March 2002. For most of the relevant period, Mr Merinson was based in the Netherlands. In the events which happened, disputes arose between the parties, negotiations ensued and various proceedings were commenced before the Courts of the Netherlands (“the Dutch Proceedings”). Thereafter, as set out in the judgment (at [5]):

The Dutch Court Settlement, both in its original form and as subsequently amended, was widely drawn and included a clause conferring (or purporting to confer) exclusive jurisdiction on the Courts of the Netherlands.

On 3 May 2017, Yukos issued the Claim Form in this action, advancing two groups of claims:

The Claim Form was personally served on Mr Merinson on 4 May 2017. As already foreshadowed, Mr Merinson was on that date domiciled in England (although, shortly thereafter, he and his family moved to Germany).

By way of the present application, Mr Merinson seeks a declaration that the English Court has no jurisdiction to try the claims brought against him by Yukos and an order that the Claim Form be set aside. The essence of Mr Merinson’s submission is that all of the Yukos claims against him fall within the scope of the Settlement Agreement incorporated in the Dutch Court Settlement. Given the exclusive jurisdiction clause contained in the Dutch Court Settlement, it is the Courts of the Netherlands on which that clause would, if valid, confer exclusive jurisdiction under Art. 25 of Brussels Recast. For the purposes of this jurisdiction dispute only, Yukos is content that the matter should proceed on the basis that all of its claims in the action do fall within the scope of the Dutch Court Settlement.

I gratefully adopt (with only minor adaptations) the Judge’s summary (at [18]) of the principal Issues before him:

The Judge answered Issue I “yes”; Issue II “no”; and Issue III “no”. It followed (at [85]) that the English Court, “as the court of the Member State in which Mr Merinson was domiciled at the date this action was commenced”, had jurisdiction in relation to all the claims made in the present action. The Judge accordingly dismissed Mr Merinson’s application for a declaration to the contrary.

From that decision, Mr Merinson appeals to this Court. The principal issues are, again, Issues I – III, as introduced above, with only some limited qualification (below).

There is some common ground on the appeal, as there was before the Judge. First, Mr Merinson accepts that the Damages Claims fall to be characterised as “matters relating to individual contracts of employment”. The dispute under Issue I accordingly relates only to the Annulment Claims. Secondly, for present purposes only, Yukos remains content that the matter should proceed on the basis that all of its claims in the Action do fall within the scope of the Dutch Court Settlement.

Before turning to the principal Issues, it is convenient to set out the terms of the Dutch Court Settlement, as amended by an Addendum dated 18 November 2016 (“the Addendum”) and on about 12 December 2017 (“the December 2017 amendment”).

The Dutch Court Settlement, as amended by the Addendum, provided as follows:

“To end their dispute, the Parties agree as follows:

…..

2.

The employment contract and any other legal relationships (save for this agreement) between the parties ended on 1 October 2016.

3.

Merinson has no obligation to repay the salaries and other emoluments received by him to date and Yukos has no obligation to pay any salaries and other emoluments to Merinson other than the salaries and other emoluments paid to Merinson to date.

…..

5.

Yukos International will pay to Merinson a sum of 200,000 Euros net.

6.

This sum should have been deposited in the bank account of

Merinson…with reference ‘severance compensation’, no later than by 5 October 2016.

7.

Merinson declares that to this day he has complied with the confidentiality clause as included in article 19 of the employment contract in relation to all information concerning Yukos and will continue to do so, now and in the future. On violation of this confidentiality clause is set a penalty of 360,000 Euros immediately due and payable to Yukos

International.

……

Merinson will not render his co-operation to, work for or render services to any party involved in any legal action concerning Yukos or persons affiliated with it; if Merinson should be requested to appear in any legal action versus Yukos, he will inform Yukos of this immediately.

8.

The moment that payments can lawfully be made from the assets of the Dutch branch to former shareholders of Yukos Oil, Yukos will pay to Merinson….a second net sum of 200,000 Euros…

9.

At present, Yukos has no proof of any violation of Merinson of the above obligation [i.e., under cl. 7], based on the current information.

10.

Parties give mutual and final discharge for all that they have claimed in these legal actions and all they might yet claim in the context of the legal relationship that used to exist between them, notwithstanding the right to claim specific performance of this settlement agreement.

11.

The Parties intend that the disputes released here be construed as broadly as possible. This release extends to Yukos and any current or former Yukos Entity officer, director, employee, consultant, agent and attorney, whether or not acting in his/her representative, individual or any other capacity.

……

13.

The Parties will keep the contents of this settlement agreement confidential.

14.

This agreement shall in all respects be interpreted, enforced and governed by the laws of the Netherlands. Any disputes regarding or relating to any aspect of this agreement formation, meaning, performance or breach, including any claim for breach of the confidentiality provision, shall be submitted to the courts of the Netherlands.

15.

Parties each bear their own procedural costs and agreed to have this current action deleted.”

As noted by the Judge (at [9]), the copy of the Dutch Court Settlement produced to the parties on 28 September 2016 did not bear the words “In naam van de Koning” (“in the name of the King”). It was common ground between the parties that those words were required for a Dutch Court Settlement to be formally enforceable under Dutch law. Accordingly, by the December 2017 amendment, a copy of the Dutch Court Settlement bearing those stamped words was procured from the Dutch Court.

For completeness and a curiosity, though nothing turns on it, the copy of the Dutch Court Settlement bearing the stamped wording did not include the amended wording of cll. 2 and 3 – which I have set out above; in the copy with the stamped wording, cll. 2 and 3 appeared in their previous, unamended form. For whatever reason, the Addendum had been in English and was signed by the parties. It was not incorporated into any proces-verbaal of the Dutch Court.

(A) The Brussels Recast Framework: So far as concerns Issues I and II (deferring consideration of Issue III until later), the context of the present dispute is found in the protective policy adopted by Brussels Recast to the presumptively weaker party in the area of (inter alia) employment contracts, qualifying the autonomy otherwise enjoyed by contracting parties.

The Recitals provide as follows:

“(18)

In relation to insurance, consumer and employment contracts, the weaker party should be protected by rules of jurisdiction more favourable to his interests than the general rules.

(19)

The autonomy of the parties to a contract, other than an insurance, consumer or employment contract, where only limited autonomy to determine the courts having jurisdiction is allowed, should be respected subject to the exclusive grounds of jurisdiction laid down in this Regulation.”

Chapter II, Section 5 of Brussels Recast deals with “Jurisdiction over individual contracts of employment”. Art. 20.1 provides that “In matters relating to individual contracts of employment…” jurisdiction shall be determined by “this Section.”. Art.

contains the protective rule of jurisdiction designed to favour employees:

“An employer may bring proceedings only in the courts of the Member State in which the employee is domiciled.”

Art. 23(1) provides that the provisions of Section 5 may be departed from only by an agreement:

“which is entered into after the dispute has arisen…”

Chapter II, Section 7 deals with the recognition of exclusive jurisdiction agreements, subject (inter alia) to the provisions of Art. 23. Art. 25.1 is in these terms:

“If the parties, regardless of their domicile, have agreed that a court or the courts of a Member State are to have jurisdiction to settle any disputes which have arisen or which may arise in connection with a particular legal relationship, that court or those courts shall have jurisdiction….Such jurisdiction shall be exclusive unless the parties have agreed otherwise….”

However, Art. 25.4 provides the qualification, of the first importance for present purposes, that agreements conferring jurisdiction “…shall have no legal force if they are contrary to…” (amongst others) Art. 23.

(B) The judgment: After reviewing the framework of Brussels Recast, together with the relevant authorities, the Judge expressed his conclusion succinctly (at [25]):

“…this action as a whole, including the Annulment Claims, is in reality and substance a matter relating to Mr Merinson’s individual contract of employment for the purposes of Section 5. There is plainly a highly material nexus between the Annulment Claims and that contract.”

The Judge’s reasons followed, at [26.1] – [26.4]. Thus:

(C) The rival cases: For Mr Merinson, Mr Jowell QC submitted that, overall, the judgment had produced a “counter-intuitive” outcome; the Settlement Agreement had been widely drafted and these claims cried out to be heard in the Netherlands. On Issue I, the Annulment Claims were to be considered separately from the Damages Claims; they had a distinct object and once separately considered could be seen to relate to the Settlement Agreement, not to Mr Merinson’s contract of employment. Moreover, the Dutch Court Settlement had a special status; at all events, there was now a new relationship with its own bespoke legal regime. So far as the significance of the Damages and Annulment Claims was concerned, as indicated by its pleaded case, Yukos had been compensated by Julius Baer in respect of the sums (allegedly) received by Mr Merinson from that financial institution. What remained of the Damages Claims as such was not that large. Mr Jowell invited us to test the approach of the Judge by asking:

“…whether a Dutch court would have adopted the same approach if the Annulment Claims had been commenced in the Netherlands….”

He submitted that it could not credibly be suggested that a Dutch Court would have been obliged to decline jurisdiction over the Annulment Claims on the basis that they related to a contract of employment.

The decision of this Court in Aspen Underwriting Ltd v Credit Europe Bank NV [2018] EWCA Civ 2590; [2019] 1 Lloyd’s Rep 221 (to which I shall come, below) was distinguishable. First, the settlement there did not terminate the contract of insurance; it was simply a mechanism for performance of the policy. By contrast, here, the Settlement Agreement confirmed that the employment was at an end. There had been a clean break and there was now a different contractual relationship on a different footing. Secondly, in Aspen, a close consideration of the insurance policy had been essential. Here, breach of the contract of employment was neither central nor dispositive. The question of kickbacks was not coterminous with the issues in the Annulment Claims. The contract of employment was no more than a matter of history.

For Yukos, Mr Willan submitted that the Settlement Agreement was indivisibly connected with Mr Merinson’s contract of employment. Indeed, Yukos could have issued a Claim Form advancing only the Damages Claims, leaving Mr Merinson to raise the Settlement Agreement by way of defence; by contrast, there was no benefit to Yukos in setting aside the Settlement Agreement as a goal in itself. A dispute as to the validity of the Settlement Agreement would affect termination of the contract of employment or reinstatement thereunder. Properly considered, there was a material nexus between the Annulment Claims and the contract of employment:

“The very basis of the attack on the Settlement Agreement is that it was induced by, inter alia, an error or deceit arising from the concealment of Mr Merinson’s misconduct as an employee. The Annulment Claims will necessarily require an investigation into Mr Merinson’s duties and conduct as an employee….”

Even considered on its own, the Annulment Claims were closely related to the contract of employment and the Settlement Agreement was akin to a severance agreement; could it be said, for example, that an employee on the verge of dismissal was outside the protective policy of Section 5 of Brussels Recast? The Annulment Claims depended indispensably on Mr Merinson’s conduct as an employee; Aspen could not properly be distinguished. The status of the Settlement Agreement (as a Dutch Court Settlement), whatever the correct analysis and its importance under Issue III, had no bearing on the present Issue. The Judge’s decision on Issue I was wellfounded and should be upheld.

(D) Discussion: I am persuaded that this action, including the Annulment Claims, is, in reality and substance, a matter related to Mr Merinson’s individual contract of employment. I agree with the Judge that there is a highly material nexus between the Annulment Claims - whether considered together with the Damages Claims or separately from them – and that contract of employment. My reasons follow.

(1) The decision in Aspen: It is convenient to begin with the decision of this Court in Aspen, which post-dated the judgment of the Judge and, as to which, there was considerable argument before us. In that case there was a dispute as to jurisdiction; Underwriters brought proceedings in this jurisdiction; the Bank challenged jurisdiction, contending that it could be sued only in the place of its domicile, the Netherlands. A settlement agreement was interposed between Underwriters’ claim and the underlying policy of insurance. One of the issues was whether Underwriters’ claim was a “…matter relating to insurance” within the protective policy Chapter II, Section 3 of Brussels Recast. The decision is thus of relevance to the present dispute, both as to the test to be applied and the approach to a settlement agreement.

The facts were that, following the loss of the vessel Atlantik Confidence (“the vessel”), Underwriters entered into a settlement agreement with shipowners (and managers, “owners”). Pursuant to the settlement agreement, the proceeds of the insurance were paid to brokers and were thereafter received by the Bank as mortgagee and sole loss payee under the policy. Subsequently, Underwriters sought (inter alia) the avoidance and/or rescission of the settlement agreement together with repayment of the sums paid under the settlement agreement, or damages. The crucial (if not the only) underlying question, essentially binary in nature, was whether the vessel was lost by reason of a peril insured against under the policy or whether the loss arose by reason of wilful misconduct on the part of owners: Aspen, at [78]. For present purposes and as foreshadowed, the relevant jurisdictional issue was whether Underwriters’ claim came within Section 3 of Brussels Recast as a “…matter relating to insurance” or whether the settlement agreement constituted a “firewall” or broke the chain between the policy and Underwriters’ claim (Aspen, at [62] and [70])?

With regard to the test adopted, its nature appears from the discussion in Aspen, at [72] and following - where the Court was guided by authority, including Bosworth v Arcadia Petroleum [2016] EWCA Civ 818, together with the authorities cited in that judgment, notably for present purposes, Alfa Laval Tumba v Separator Spares

On the facts in Aspen and applying the test as just outlined, I held (at [78]), giving the lead judgment of the Court, that, as a matter of reality and substance, there was “the most material nexus” between Underwriters’ claims and the policy, notwithstanding the interposition of the settlement agreement. Accordingly, Underwriters’ claims came squarely within the heading “matters relating to insurance”.

For completeness, I repeat the caveat expressed in Aspen (at [72]) as to the ultimate fate of Arcadia, given the reference by the Supreme Court in that case of various questions to the CJEU for a Preliminary Ruling: Official Journal of the European Union, 18 December 2017.

(2) The relevant test: I did not understand there to be any or significant dispute as to the relevant test - in shorthand, whether in reality or substance there is a material nexus between the Annulment Claims and Mr Merinson’s contract of employment. That was the test adopted by the Judge (at [20] – [21]), following consideration of Brussels Recast and various passages drawn from authority, which need not be repeated here. It was also the test adopted by this Court in Aspen and it is the test I apply here. The focus of this ground of appeal, however, was on the application of the test, where the parties were very much in dispute. To that battleground, I next turn.

(3) The application of the test: First, while Aspen does not bind this Court to decide Issue I in favour of Yukos, there are telling similarities between the crucial question there and the basis for the Annulment Claims here, so that the reasoning there lends support to the Judge’s decision here. Though the question of kickbacks was not coterminous with the issues raised by the Annulment Claims, as it seems to me, the basis for the Annulment Claims turns on Mr Merinson’s alleged breach/es of his contract of employment. If that is right, then the contract of employment is much more than a mere matter of “history or pathology”: Aspen, at [69]; to the contrary, it is necessary to consider the terms of the contract of employment to resolve the matter in dispute. Accordingly, the nexus between the Annulment Claims and the individual contract of employment is highly material, notwithstanding the interposition of the Settlement Agreement. To such extent, this case bears a significant resemblance to Aspen. Though there are points of difference (for instance, that the Settlement Agreement here confirmed the termination of the contract of employment, whereas there was no need for any similar provision in the settlement agreement in Aspen), I do not think that any such differences reduce the reliance to be placed on the reasoning in Aspen and its relevance here.

Secondly, it will be noted that I reach that conclusion on the assumption most favourable to Mr Merinson – namely, by considering the Annulment Claims on a stand-alone basis, separately from the Damages Claims. This approach provides a

Thirdly, at first blush, I acknowledge a degree of attraction in Mr Jowell’s rhetorical question or challenge, namely, whether it could credibly be suggested that had Yukos brought the Annulment Claims in the Netherlands, a Dutch Court would have been obliged to decline jurisdiction because they related to Mr Merinson’s contract of employment. Closer scrutiny, however, reveals that there is, with respect, less force in this point than might first be thought. On that factual hypothesis, the key lies simply in Mr Merinson’s attitude to the proceedings. Had Mr Merinson “submitted” to the Dutch jurisdiction, then of course no question would have arisen of the Dutch Court declining jurisdiction; but the reason would have been that no jurisdictional point was taken by Mr Merinson, rather than any more complex analysis. If, on the other hand, Mr Merinson had contested Dutch jurisdiction, then the Dutch Court would have been obliged to decline jurisdiction in favour of the court in the Member State of his domicile. That is a necessary consequence of the protective policy of Section 5 of Brussels Recast. It does not at all follow that all claims relating to or arising from the Settlement Agreement could not be pursued in the Dutch Court; at the very least, insofar as claims within the scope of the Settlement Agreement with its exclusive Dutch jurisdiction clause came within Art. 23(1) of Brussels Recast, the Dutch Court would have been entitled and obliged to take jurisdiction.

Fourthly, I see force in Mr Willan’s submission that the Settlement Agreement was, in significant respects, akin to a severance agreement. Having regard to the protective policy of Section 5 of Brussels Recast, there is a powerful case that (in principle) disputes as to severance agreements (ex hypothesi arising at or about the time when the employee is leaving his employment) should be brought in the court of the Member State where the employee is domiciled.

Fifthly, whatever the significance of the Annulment Claims “impeaching a Dutch juridical act” (even assuming in Mr Merinson’s favour, without in any way deciding, that they do so), that is of no relevance to Issue I – whatever the relevance may be under Issue III.

Sixthly, looking at the matter in the round, any oddity with regard to Issue I does not arise from the analysis of the nexus between the Annulment Claims and Mr Merinson’s contract of employment, taking full account of the Settlement Agreement with its Dutch exclusive jurisdiction clause. Instead, the oddity is that it is the individual employee - and presumptive beneficiary of the protective policy of Section 5 of Brussels Recast – who does not wish the dispute to proceed in the court of the Member State of his domicile at the relevant time. For better or worse, however, Brussels Recast does not contain relevant flexibility in that regard.

For the reasons given, I would dismiss the appeal under Issue I.

(A) Introduction and test: This Issue calls for consideration of the balance struck by

As the Judge observed (at [31]), the 1968 Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters (“the Brussels Convention”) did not originally include special jurisdictional rules relating to contracts of employment. It followed that the Jenard Report [1979] OJ C/59/1 on the Brussels Convention (“the Jenard Report”) could not and did not comment on any predecessor to Art. 23(1).

The Jenard Report did, however, say this as to Art. 12 of the Brussels Convention, the predecessor to Art. 15(1) of Brussels Recast, dealing with matters relating to insurance:

“Article 12 relates to agreements conferring jurisdiction. Agreements concluded before a dispute arises will have no legal force if they are contrary to the rules of jurisdiction laid down in the Convention.

The purpose of this article is to prevent the parties from limiting the choice offered by this Convention to the policyholder, and to prevent the insurer from avoiding the restrictions imposed under Article 11.

A number of exceptions are, however, permitted. After a dispute has arisen, that is to say ‘as soon as the parties disagree on a specific point and legal proceedings are imminent or contemplated’, the parties completely regain their freedom.”

As it seems to me, there is no good reason why this test (“the Jenard test”) should not be equally applicable to the provisions of Art. 23(1) of Brussels Recast, dealing with contracts of employment; the policy considerations as to matters relating to insurance and individual contracts of employment are materially similar in this regard. Accordingly and without reading it as a statute, the Jenard test for establishing when a dispute has arisen under Art. 23(1) contains two – cumulative - limbs:

In Civil Jurisdiction and Judgments (6^th ed., “Briggs”), Prof. Briggs has commented (at para. 2.95, fn. 593) that the Jenard test “…may be thought to be too vague, or if not, too late in the day”. Emphasising the difficulty under the protective policy Sections of Brussels Recast of identifying the precise point at which a dispute has arisen, Prof. Briggs himself goes no further than saying (at para. 2.95) that it must be:

“…somewhere between the occurrence of the facts which will give rise to the claim and the service of process. But it is not possible, at this stage, to be more precise about it.”

For their part, Layton and Mercer in European Civil Practice (2^nd ed., “Layton and Mercer”), at para. 16.060 refer to a range of different views on this point and express the following view:

“It is suggested that the test proposed by Jenard should be modified slightly, so that proceedings should be both imminent and contemplated. This is because the practical need for an agreement on jurisdiction ought not to arise until the dispute becomes one which will apparently need litigation to resolve it. A jurisdiction agreement entered into at any earlier stage would be a precautionary measure only, and would be contrary to the same policy objections as support the existence of …[the now Art. 23(1)]…in the first place.”

Against this background, the Judge’s summary (at [35]) is both succinct and welcome:

“To summarise these commentaries it seems that, in the authoritative view of Mr Jenard, a dispute will have

‘arisen’….only if two conditions are satisfied: (a) the parties must have disagreed on a specific point; and (b) legal proceedings in relation to that disagreement must be imminent or contemplated. Prof. Briggs considers that would put the relevant point too late: Layton and Mercer that it would put it slightly too early.”

For my own part, as will be seen and for reasons to which I shall come, I favour the Jenard test.

(B) The facts: Though the parties are very much in dispute as to the inferences and conclusions to be drawn from them, I do not discern any or any significant or strenuous dispute as to the primary facts found by the Judge. I therefore take these essentially from the judgment and venture the summary which follows.

First, from Mr Merinson’s perspective the claims brought in this action are precisely those he wished to avoid by entering into the Settlement Agreement. He knew that Yukos had taken an aggressive approach towards a former colleague, a Mr Feldman – alleging a fraudulent scheme and deposing Mr Feldman in US proceedings (inter alia) as to whether he (Mr Merinson) had been involved. According to Mr Peters, Mr Merinson’s Dutch lawyer, he had explained to the Dutch Court that a crucial part of any settlement, so far as Mr Merinson was concerned “..was that it should release Mr Merinson from any future claims of any nature which might otherwise be made against him…”. The upshot (as found by the Judge, at [57]) was that “the claims made in the present action were the very sort of claims which Mr Merinson wished to prevent being raised”.

Secondly, a deposition of Mr Feldman took place in New York on 21 June 2016 and thus some three months before the conclusion of the Settlement Agreement. The transcript includes the following exchange:

“Yukos Attorney: During your tenure at Yukos, did you ever receive any kickbacks from banks or financial institutions that held Yukos funds?

Feldman: No.

Yukos Attorney: What about Dmitri Merinson?

Feldman: I don’t know.

Yukos Attorney: Do you have any reason to think Dmitri Merinson did receive some form of introductory fee or kickback?

Feldman: I don’t know.

Yukos Attorney: What about Falcon Bank? Did either of you, to your knowledge, ever receive any sort of fee or award from Falcon Bank?

Feldman: I can’t speak for him. I haven’t. ”

Mr Merinson’s evidence was that he was aware, prior to entering into the Settlement Agreement, that these questions had been asked of Mr Feldman. As these questions had been put on behalf of Yukos and though there is no express finding by the Judge, the justifiable inference must be that Yukos had sufficient suspicion of Mr Merinson to justify asking them; it would be surprising if they were no more than an unauthorised frolic on the part of the Yukos Attorney. The significance of Falcon Bank is that the Amended Particulars of Claim in the present action allege in terms that Mr Merinson received illicit payments from that Bank.

Thirdly, the Yukos evidence before the Judge, from Mr Godfrey (the director) and Mr O’Sullivan (the Yukos solicitor) was that the dispute the subject of the present proceedings had not arisen and was not in contemplation at the time the Settlement was entered into. The circumstances giving rise to the current claim only came to the notice of Yukos in about March 2017. It was even said (by Mr Godfrey) that before the Settlement Agreement “there was no suggestion that the defendant [Mr Merinson] had been wrongfully taking payment from the banks”. Whatever the state of Mr Godfrey’s personal knowledge, that statement is belied by the question/s put on behalf of Yukos in the New York deposition of Mr Feldman and, therefore, at the least, requires qualification. That said, it is also right to note Mr Godfrey’s further comment, that had Yukos “been aware of the facts and matters which are the subject

Fourthly, against this background, the Judge found (at [59]):

“It may well be that both Mr Merinson and the Yukos Group were (at least separately) aware of at least some of the circumstances which give rise to the present claims. However, they had not joined issue or communicated with each other about those specific circumstances.”

Fifthly, as recorded in the judgment (at [14]) and as already foreshadowed, Yukos is content that the matter should be decided on the basis that all of the claims in this action fall within the scope of the Settlement Agreement.

(C) The analysis in the judgment: The Judge reminded himself (at [55]) that the fact that it was Mr Merinson who did not want the action to be brought in this Court was neither here nor there and further reminded himself (at [56]) of the policy underlying Section 5 of Brussels Recast. The fact that the application of the policy “may work harshly” on the facts of an individual case was not a matter he could legitimately take into account.

In the Judge’s view (at [57]):

“The relevant purpose of these sections seems to me to be to protect employees from bargaining away their right to be sued in their home jurisdiction (which is presumed to be the most favourable to them), except where they are positively choosing a specific jurisdiction for the resolution of a specific existing dispute. Only at that point, where a specific dispute is already in existence, will the employee know the implications of his or her decision as to the jurisdiction in which that dispute is to be resolved. It seems to me that there is force in the observation of Layton and Mercer that a jurisdiction agreement entered into at any earlier stage would be a precautionary measure only, and would be contrary to the protective policy which underlies these provisions.”

The Judge went on to say (at [58]) that it was not uncommon for settlement agreements, such as the Settlement Agreement here, “to seek to compromise potential claims other than the immediate dispute which has led to the settlement”. However (ibid), settlement agreements did not fall to be treated as a “special category” for the purposes of Art. 23(1). Jurisdiction clauses in settlement agreements were no different in principle from jurisdiction clauses in employment contracts themselves.

“To the extent that they seek to regulate jurisdiction in relation to particular disputes which have not yet actually arisen, they contravene the protective policy of Section 5.”

Moreover (at [59]), the concept of a dispute within Art. 23(1) required the existence

“of some particular matter on which the parties have actually joined issue”. However, as already underlined (above) that had not happened here. Thus:

“Parties are not ‘in dispute’ merely because one or both of them is aware of circumstances which could potentially give rise to a claim. A dispute will not have arisen for these purposes unless and until the subject-matter of the claim has been communicated by one party to the other.”

With “some hesitation” (at [60] – [61]), the Judge parted company with Layton and Mercer and adopted the Jenard test. It was sufficient if the parties were “positively choosing” the forum in which their specific existing dispute may eventually have to be resolved “even if they are merely choosing it as a last resort after other methods – negotiation, mediation etc – have first been tried and failed”. On the facts of the present case, the Jenard test was not satisfied at the date of the Settlement Agreement.

Accordingly, the Judge answered Issue II in the negative. It followed that the jurisdiction provisions of the Settlement Agreement were ineffective to override the otherwise mandatory jurisdiction provisions of Section 5.

(D) The rival cases: For Mr Merinson, Mr Jowell submitted that the Judge had taken a wrong turning by imposing an unnaturally constrained interpretation on the wording of Art. 23(1). The Jenard test was not of enormous assistance and, in any event, the Judge erred (at [59]) in requiring “joinder of issue”; that was an unwarranted gloss on the test. Further, the Judge had lost sight of the need for a purposive approach and had, wrongly, sought to apply the Jenard test mechanistically; the importance of predictability was recognised – but that did not mean predictably foolish. Still further, the Jenard test did not address the particular circumstances of this case; namely where:

“(i)

the parties entered into the jurisdiction agreement in the context of a wide settlement of a large number of actual and potential claims and counter-claims between them and a complete breakdown of the employment relationship, and (ii) the disadvantaged party specifically desired to settle claims such as that now brought, and to subject any subsequent proceedings about such claims to the exclusive jurisdiction of a

particular court (in this case, the courts of the Netherlands).”

Pressed by the Court as to his formulation for satisfying Art. 23(1) of Brussels Recast, Mr Jowell produced the following:

“(1)

The facts giving rise to the issue between the parties have taken place prior to the agreement.

(2)

The issue which is the subject of the claim has been articulated by the presumptively stronger party and has been communicated (directly or indirectly) to the weaker party so that it was in their contemplation prior to the agreement.

(3)

The weaker party has chosen to include a claim relating to that issue within the ambit of the agreement.

Any requirement for a ‘disagreement’ can be fulfilled where the parties are in a general dispute that affects the entirety of the relevant relationship and where all wrongdoing is denied by the weaker party.”

Mr Jowell emphasised the facts of the present case. The deposition transcript revealed that the present disputes must have been in the contemplation of Yukos prior to the conclusion of the Settlement Agreement; it was implausible that Yukos would have had a “wild shot in the dark”; moreover, the specific reference to Falcon Bank was telling. Here, the subject-matter of the claim had been communicated to Mr Merinson, even if not directly by Yukos; he thus knew the implications when he agreed to the Settlement Agreement with its Dutch jurisdiction clause.

For Yukos, Mr Willan underlined that Section 5, including Art. 23(1) was to be given an autonomous European meaning. Art. 23(1) meant what it said; it referred to “the” dispute, i.e., the same dispute forming the subject-matter of the proceedings in which the jurisdiction clause is invoked. A jurisdiction agreement would only be valid if the parties were actually in dispute as to a specific subject-matter at the time the agreement was concluded; the concept of a “dispute” required the existence of some particular matter on which the parties have actually joined issue. It was furthermore necessary that the parties had communicated as to that dispute though no particular formalities were required. The purpose of Section 5 required a live actual dispute at the time of entry into the jurisdiction agreement, which, qualitatively, would be akin to submission to the jurisdiction by the weaker party (the employee) for the dispute in question. The Jenard Report was not simply another commentary. There was a difference between potential disputes which the parties wanted to settle and disputes which had actually arisen. Mr Willan underlined the importance of certainty and predictability in this area; the matter could not depend on what Mr Merinson had heard from Mr Feldman; a communication from a third party did not equate to a dispute having arisen. Yukos would have to guess as to Mr Merinson’s knowledge. As Mr Willan expressed it:

“…a dispute cannot have arisen before the subject-matter of the claim has been communicated by one party to the other. That is plainly necessary to achieve the objectives of certainty and predictability: the validity of a jurisdiction clause cannot possibly depend, as Mr Merinson submits, on an inquiry as to whether an employee, at the relevant time, had second-hand knowledge that a question had been asked by a related party to a third party about the employee’s conduct.”

Here, at the time of entry into the Settlement Agreement, there was no dispute between Yukos and Mr Merinson as to kickbacks; there was suspicion but the claims comprising the action were not in mind at the time.

(E) Discussion: I am driven to conclude that Yukos has the better of the argument on Issue II and that the Judge was correct to answer this Issue as he did.

(1) The test: There is no authority on this Issue and, as already indicated, I prefer the Jenard test to such other formulations as have been suggested. Indeed, given the wellknown status of the Jenard Report, I would be reluctant to depart from the Jenard test unless an alternative was plainly to be preferred. As always with Brussels Recast (and its predecessors), a purposive approach is to be adopted; it is, therefore, necessary throughout to keep in mind the protective policy underlying Section 5, including Art. 23(1). As it seems to me, the Jenard test suffices to protect the presumptively weaker party (the employee).

Thus Limb 1 (as I have termed it) requires the parties to have disagreed on a specific point. Necessarily, this limb assists in distinguishing an actual dispute from a potential dispute and, in so doing, marking the divide between a jurisdiction agreement entered after “the” dispute has arisen and a jurisdiction agreement entered by way of a precautionary measure only. The inescapable connotation of disagreement on a specific point is that the parties have “joined issue” on that specific point. I am unable therefore to accept Mr Jowell’s submission that the Judge had added an unwarranted gloss to the Jenard test. It further seems to me that Mr Willan was plainly correct to say that a dispute cannot have arisen before there has been communication as to its subject-matter by one party to the other. Given the context of these protective provisions and the premium placed on certainty and predictability, the outcome cannot sensibly turn on what one party (the presumptively weaker party) may have gleaned from a third party (at all events, a third party not an authorised agent of the “presumptively stronger party”).

I am bound to say that the difficulties presented by Mr Jowell’s formulation (set out above) for satisfying Art. 23(1) are themselves revealing. Thus para. (1) is selfevident but takes the matter no further. For the reasons already given, para. (2) of the formulation can only be acceptable if the communication is made by or, perhaps, by an agent suitably authorised on behalf of, the presumptively stronger party. A hit or miss approach to knowledge acquired by the employee from a third party cannot suffice. While there can be no objection to para. (3) of itself, it hinges on paras. (1) and (2). The unnumbered final paragraph, with respect, fails to grapple with the need to focus on a specific dispute and thus to distinguish between an actual and a potential dispute.

As to Limb 2 of the Jenard test, where litigation in relation to the dispute in question is either imminent or contemplated, it seems to me that the presumptively weaker party is adequately protected. I am not, with respect, attracted to the Layton and Mercer suggestion that litigation must be both imminent and contemplated. I do not think that such further postponement of the relevant date is either warranted or necessary. I also think that there is force in the Judge’s observations at [60] – [61] (set out at [59] above), which might well fall foul of the Layton and Mercer suggestion, were that suggestion well founded.

While, with respect, I understand the unease expressed in Briggs as to the timing of the Jenard test (that it might be too late), absent a cogent suggestion of some earlier cut-off point, I cannot see a good reason for departing from the Jenard test.

(2) Applying the test: On the facts of the present case, it seems plain to me that the Jenard test was not satisfied.

First, the key is Limb 1. It is common ground or realistically indisputable that there was no direct communication between Yukos and Mr Merinson in advance of the Settlement Agreement as to any question of his receipt of kickbacks from financial institutions. That Yukos may well have harboured suspicion in this regard (sufficient to prompt questioning of Mr Feldman in his New York Deposition) and that Mr Merinson heard of this questioning from Mr Feldman, is neither here nor there. In the absence of any such direct communication between Yukos and Mr Merinson, there could be no actual dispute between Yukos and Mr Merinson going to the particular subject-matter of the present action. In my judgment, this is an insuperable hurdle for Mr Merinson’s case on Issue II and fatal to it.

Secondly and if necessary to go further, the Settlement Agreement was, advisedly, certainly from Mr Merinson’s perspective, drafted in very wide terms. As has been seen, for the purposes of these proceedings, Yukos accepts that all the claims in the action fall within its scope. Again, at first blush, this consideration might be thought to assist Mr Merinson but, upon analysis, it does not go nearly far enough. The reason is that settlement agreements, especially if widely drafted, frequently encompass potential as well as actual disputes. Potential disputes will not, however, satisfy Art. 23(1). Moreover, with regard to Art. 23(1), settlement agreements do not comprise some special category, so that it remains necessary to distinguish actual from potential disputes. In terms of the Settlement Agreement, the subject-matter of this action comprised no more than a potential dispute. The Settlement Agreement neither amounted to a communication between the parties of the specific actual dispute (the subject-matter of this action) nor did it indicate that litigation was imminent or contemplated in respect of the present dispute. This difficulty too, Mr Merinson is unable to surmount.

Thirdly and for my part, I readily accept that there would be no prejudice to Mr Merinson were this action to proceed in the Netherlands rather than here – and, had I the discretionary power to ensure that it was heard in the Netherlands, I would exercise it accordingly. But, under Brussels Recast, I do not have such a power and the perhaps curious outcome in an individual case is of no moment. In another case, it could readily be that the former employer seeks to litigate an additional claim in the jurisdiction of the settlement agreement and the erstwhile employee (and presumptively weaker party) would be fighting tooth and nail for the litigation to proceed in the Court of the Member State of his domicile. Brussels Recast is concerned with such broader issues of policy and principle and it is those which dictate the outcome here.

For these reasons, I would uphold the decision of the Judge and dismiss the appeal on

If it is of any consolation, I underline the limited nature of the decision to which I have come on Issues I and II. That decision does not mean that Mr Merinson may not ultimately prevail in his reliance on the Settlement Agreement as barring the claims now pursued against him. The decision means no more than that the Annulment Claims can be considered by the English Court. Like the Dutch Court, however, the English Court must apply Dutch law when considering those claims. The English Court is vastly experienced in deciding questions of foreign law, so that the outcome ought to be unaffected.

It remains to deal with Issue III.

(A) The nature of the Dutch Court Settlement: As the Judge observed (at [63]), this Issue requires consideration of whether the nature of the Dutch Court Settlement means that the English Court is in any event precluded from entertaining the Annulment Claims.

The Jenard Report (cited by the Judge at [64]) explains that the Brussels Convention included a provision as to court settlements on account of the German and Dutch legal systems.

“…Under German and Netherlands law, settlements approved by a court in the course of proceedings are enforceable without further formality.

….

The Convention…makes court settlements subject to the same rules as authentic instruments, since both are contractual in nature. Enforcement can therefore be refused only if it is contrary to public policy in the State in which it is sought…” England has no equivalent of enforceable (i.e., “authentic”) instruments.

A “court settlement” is defined in Art. 2(b) of Brussels Recast, as follows:

“…a settlement which has been approved by a court of a Member State or concluded before a court of a Member State in the course of proceedings;”

An “authentic instrument” is defined in Art. 2(c) of Brussels Recast as:

“…a document which has been formally drawn up or registered as an authentic instrument in the Member State of origin…”

It is next necessary to go to Chapter IV of Brussels Recast, headed “Authentic Instruments and Court Settlements”. Arts. 58 and 59 are in these terms:

An authentic instrument which is enforceable in the member State of origin shall be enforceable in the other Member States without any declaration of enforceability being required. Enforcement of the authentic instrument may be refused only if such enforcement is manifestly contrary to public policy (ordre public) in the Member State addressed.

In Solo v Boch (Case C-414/92) [1994] I.L. Pr. 457, the CJEU said this as to Court settlements and authentic instruments:

“17.

….to be classified as a ‘judgment’ within the meaning of the Convention, the act must be that of a court belonging to a Contracting State and ruling on its own authority on points in dispute between the parties.

18.

However, this condition is not fulfilled in the case of a settlement, even if it is reached before a judge of a Contracting State and puts an end to a dispute. Court settlements are essentially contractual in nature, in the sense that their terms depend primarily on the parties’ intentions, as the expert report explains.

….

24.

…the enforcement of an authentic instrument in a Contracting State other than the one where it was drawn up and is enforceable may be refused only if enforcement is contrary to public policy in the state in which enforcement is sought.”

In short summary, therefore, a court settlement is contractual in nature and not a judgment; however, if enforceable in the Member State of origin, it is and is intended to be enforceable in other Member States, subject only to refusal if enforcement would be contrary to the public policy of the Member State where enforcement is sought.

(B) The judgment: Before the Judge, Mr Jowell submitted under this Issue that the English Court had no jurisdiction to entertain the Annulment Claims on two (overlapping) grounds. First, that the Dutch Court Settlement was a juridical act of the Dutch Court. Secondly, the Dutch Court Settlement was a “court settlement” for the purposes of Chapter IV of Brussels Recast, which the English Court was obliged to recognise and enforce.

After considering expert evidence of Dutch Law, the Judge (at [78]) distilled the matter into the following propositions:

“78.1

The Dutch Court Settlement is a ‘court settlement’ within the definition in Article 2 of the Recast Judgments Regulation. It is therefore enforceable in the United Kingdom under Chapter IV of the Recast Judgments Regulation.

78.2

The only ground on which this court could refuse enforcement of the Dutch Court Settlement would be if such enforcement were manifestly contrary to public policy: see Articles 58 and 59.

78.3

Nevertheless, the Dutch Court Settlement does not have the status of a judgment…It remains essentially contractual in nature….

78.4

The applicable law of the Settlement Agreement is Dutch law. Under Dutch law, it can be impugned on the same basis as any other contract, and is not subject to any special regime in that regard….

78.5

Any court of a Member State which, under the Recast Judgments Regulation, would have jurisdiction to set aside the Settlement Agreement as a contract still has that jurisdiction, even though the Settlement Agreement has been incorporated into the Dutch Court Settlement. The Dutch courts do not have exclusive jurisdiction in that regard to any greater extent than they would have in the case of any other contract…..

78.6

The judgment of such a court setting aside the Settlement Agreement would be entitled to recognition and enforcement in the other Member States, including the Netherlands, notwithstanding the Dutch Court Settlement….”

The Judge appreciated (at [79] – [81]) that these conclusions gave rise to a tension between enforcement on the one hand and proceedings to set aside a Court Settlement on the other. The provisional solution lay in case-managing the “…enforcement application and the set-aside action, so that they are dealt with together, the result of the action determining the enforcement application”. No such complexity arose here, in that no enforcement application was before the Court.

Against this background, the Judge’s conclusions (at [78], set out above) disposed of Mr Jowell’s grounds:

“82.1

If, as I have held, the Dutch Court Settlement remains an agreement which can be impugned on the same basis as any other contract, despite being incorporated in the proces-verbaal of the Dutch court, then there is no reason intrinsic to the Dutch Court Settlement to prevent the English court exercising jurisdiction in relation to it. That reasoning is fatal to Mr Jowell’s first argument.

82.2

As to his second argument, based specifically on Chapter IV, it confuses enforceability with jurisdiction. If, as I have held, the English court has jurisdiction under Chapter II of the Recast Judgments Regulation to set aside the Settlement Agreement as a contract, it does not lose that jurisdiction merely because the Settlement Agreement is enforceable in the United Kingdom as part of the Dutch Court Settlement….”

(C) The rival cases: For Mr Merinson, Mr Jowell effectively repeated the submissions he had advanced before the Judge. The judgment had not taken account of the special features of a court settlement, the clear terms of Art. 52 of Brussels Recast and the decision in Solo (esp., at [24]). Comity pointed to only the “home court” (here the

For Yukos, Mr Willan submitted that there was no special rule of jurisdiction for court settlements and no principled reason why one should exist. A court settlement did not involve a juridical act. Once the true contractual nature of a court settlement was understood, it could be impugned on the same basis as any other contract and in accordance with the same jurisdiction rules otherwise applicable under Brussels Recast. Enforceability was not to be confused with jurisdiction. Were Mr Merinson to take steps to enforce the Dutch Court Settlement, by reason of Art. 52, the English Court would not be entitled to review its substance:

“But that does not mean that the English court is precluded from considering whether the Settlement Agreement can be impugned in the context of a separate and distinct claim to annul the Settlement Agreement, still less that article 52 excludes the English court’s jurisdiction under Chapter II entirely.”

In the alternative, Mr Willan submitted that the Court’s jurisdiction fell to be determined at the date the claim was issued; at that time the Dutch Court Settlement was not enforceable in the Netherlands because it lacked the wording “In naam van de Koning”; accordingly, any considerations otherwise relating to the Dutch Court Settlement were simply inapplicable.

(D) Discussion: My conclusions can be very briefly expressed. With respect to Mr Jowell’s submissions, I do not think there is anything in this point. I agree with the Judge, essentially for the reasons he gave.

First, it is clear both generally (the Jenard Report and the CJEU decision in Solo, esp., at [17] – [18]) and from the evidence of Dutch law (summarised by the Judge), that the Dutch Court Settlement is contractual in nature.

Secondly, on the evidence of Dutch law, the Settlement Agreement (governed by Dutch law) can be impugned on the same basis as any other contract.

Thirdly, provided the English Court would otherwise have jurisdiction to entertain the Annulment Claims it does not lose that jurisdiction simply because the Settlement Agreement has been incorporated in the Dutch Court Settlement.

Fourthly, the Judge (rightly, in my view) concluded that the submissions advanced on behalf of Mr Merinson confused enforceability and jurisdiction. Reconciliation between the demands of enforcement of the Dutch Court Settlement (if enforcement is sought) and consideration of the Annulment Claims can be achieved by way of case management, along the lines outlined by the Judge. On any view, the provisions of Arts. 52, 58 and 59 of Brussels Recast and the observations in Solo at [24] do not preclude the English Court’s jurisdiction to entertain the Annulment Claims.

It follows that I would accede to Mr Willan’s primary submission. It is therefore unnecessary to express any view as to his alternative submission and I do not do so.

For the reasons given, I would dismiss the appeal on Issue III. It follows that I would dismiss the appeal as a whole.

I agree.

I also agree.