Alta Trading UK Ltd v Bosworth

This appeal is the latest round in a long-running dispute as to whether the English court has jurisdiction over the claims brought by the Claimants (the Respondents in this Court) against the 1^st and 2^nd Defendants, Mr Bosworth and Mr Hurley (the Appellants in this Court). In the action the Respondents allege that Mr Bosworth and Mr Hurley, who formerly acted as the Chief Executive Officer (“CEO”) and Chief Financial Officer (“CFO”) respectively of the first three Claimants (“the Arcadia Claimants”), were guilty of a large-scale conspiracy to defraud the Arcadia Claimants of which they were stewards.

The question of jurisdiction turns on whether the claims relate to individual contracts of employment within the meaning of Art 18(1) of the Lugano Convention (or the Lugano II Convention as it is sometimes referred to). If they do, then by Art 20(1) the claims have to be brought in the state where the employee is domiciled. The Appellants are British citizens but are each domiciled in Switzerland.

The answer to that question depends on whether they were “in a relationship of subordination” to their employing company or companies. After a reference to the Court of Justice of the European Union (“CJEU”), the Supreme Court remitted the case to the Commercial Court to hear further evidence and submissions on that question.

The issue was heard by Sir Michael Burton GBE sitting as a Judge of the High Court (“the Judge”). He handed down judgment at [2020] EWHC 2757 (Comm) on 19 October 2020 (“the Judgment” or “Jmt”). He had before him a very large amount of contested evidence, but his conclusion was a firm one, namely that the Respondents had a good arguable case that the Appellants were not in a relationship of subordination to the Arcadia Claimants, in that they had a more than negligible ability to influence them (Jmt at [36]). He accordingly dismissed their challenge to the jurisdiction.

The Appellants appeal, with permission granted by the Judge himself, on the basis that he misunderstood the decisions of the CJEU and applied the wrong test to be derived from them.

I consider that he applied the right test and that there is no error in his admirable judgment. I would dismiss the appeal.

The relevant rules on jurisdiction are found in the Convention on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters between the European Community and, among others, the Swiss Confederation, signed at Lugano on 30 October 2007 (“the Lugano Convention”).

Title II of the Lugano Convention (Arts 2 to 31) deals with jurisdiction. Section 5 of Title II (Arts 18 to 21) deals with jurisdiction over individual contracts of employment. Arts 18(1) and 20(1) respectively provide as follows:

It can be seen that the wording of Art 18(1) naturally gives rise to two questions in any particular case: is there an “individual contract of employment”? and does a claim “relate to” that contract?

The facts can be taken (subject to one caveat) from the Agreed Statement of Facts and Issues that was before the Supreme Court in 2017. The Arcadia Claimants are respectively (1) the 1^st Claimant, Alta Trading UK Ltd, an English company formerly called Arcadia Petroleum Ltd (“Arcadia London”); (2) the 2^nd Claimant, Arcadia Energy (Suisse) SA, a Swiss company (“Arcadia Switzerland”); and (3) the 3^rd Claimant, Arcadia Energy Pte Ltd, a Singaporean company (“Arcadia Singapore”). They are companies in the Arcadia Group and were at the material times engaged in the business of trading physical crude oil and oil derivatives. (This Arcadia Group has nothing to do with the well-known retail group of the same name which went into administration last year). The Arcadia Group (which includes other companies than the Arcadia Claimants) is 100% owned by the 4^th Claimant, Farahead Holdings Ltd (“Farahead”), which is ultimately owned by trusts established by Mr John Fredriksen, a successful Norwegian-Cypriot businessman with extensive shipping interests. (As this indicates “the Arcadia Group” is used in the Agreed Statement of Facts and Issues to mean the Arcadia Claimants and other subsidiaries of Farahead, but not Farahead itself, and I will use it in the same sense).

Mr Bosworth was the de facto CEO of the Arcadia Group until February 2013, and Mr Hurley the de facto CFO of the Arcadia Group until September 2013. They possessed these titles although their written contracts of employment did not include provisions that they should act as group CEO and CFO.

The Respondents’ pleaded case is that they are the victims of a substantial and sustained fraud perpetrated on them by a number of individuals, including the Appellants, and corporate entities. Proceedings were commenced against six individuals and four associated companies; settlements have subsequently been concluded with some of them. The Respondents believe the Appellants to have been the principal architects of, and beneficiaries from, the alleged fraud. In a nutshell, the fraud alleged is that they and others siphoned off sums from the Arcadia Group for their own benefit by the insertion of corporate entities into buying and selling chains and the manipulation of those transactions so that almost all of the profits accumulated in the inserted corporate entities rather than in the Arcadia Group, which bore all the costs and risks of the transactions. The Respondents’ estimate is that gross profits of some US$339m were diverted out of the Arcadia Group during the period April 2007 to May 2013.

The Appellants have not yet pleaded their case but vigorously deny the allegations. In summary they say that oil trading in West Africa is fraught with reputational and regulatory risks; that Mr Fredriksen and/or Farahead knew of and consented to the insertion of ‘sleeve’ entities into the trading transactions to distance and protect the Arcadia Group from those risks; that there was no attempt to conceal this from the Respondents, who were fully aware that trading in West Africa was continuing; that Mr Fredriksen and/or Farahead knew about and authorised the continued trading; that the Respondents knew that commission payments were made to third parties so that such oil trading could continue; that there was no siphoning off, no evidence that they received any of the sums allegedly diverted, and no reason to infer fraud. They also dispute the alleged losses which have been calculated without taking account of related hedging transactions.

The Appellants were authorised by Farahead to act as CEO and CFO respectively, and invested by Farahead with, possessed, and exercised all the usual authority of a CEO and CFO. They ran the day-to-day business of the Arcadia Group. They formed part of the governing structure of the Arcadia Claimants and undertook to, and were entrusted by, them to act for and on their behalf.

They had limited formal de jure appointments as directors as follows: Mr Bosworth was a director of Arcadia Singapore from May 2010 to September 2013; Mr Hurley was a director of Arcadia London from March 2006 to July 2011, and of Arcadia Singapore from September 2011 to September 2013. Otherwise they were not formally appointed directors of the Arcadia Claimants. But the Respondents’ case is that they acted as de facto or shadow directors of each of the Arcadia Claimants.

At all material times the Appellants were also employed under contracts of employment with one or other of the companies in the Arcadia Group. Mr Bosworth was party to a contract of employment with Arcadia London between March 2006 and September 2009, with Arcadia Singapore between September 2009 and May 2012, and with a company called Arcadia Al Arabiya DMCC (a Dubai company within the Arcadia Group but not one of the Arcadia Claimants (“Arcadia Dubai”)) from March 2012 to February/March 2013. Mr Hurley was party to a contract of employment with Arcadia London between March 2006 and July 2011, with Arcadia Singapore between August 2011 and July 2012, and with Arcadia Dubai from August 2012 to September 2013.

None of these contracts of employment (which were drafted by, or under the direction of, Mr Bosworth and Mr Hurley themselves) contains any provision that they act as CEO and CFO. But notwithstanding this, they acted at all material times on behalf of all the Arcadia Group companies, and provided services and performed functions on a group-wide basis.

It is common ground that the Appellants were answerable to the ultimate shareholders in the Arcadia Group, through Farahead, who had the power to hire and fire them.

These facts were all agreed by the parties before the Supreme Court in 2017. The one caveat is that these facts did not deal in any more detail with the relationship between the Appellants on the one hand and Farahead on the other, and the extent to which they were given instructions by Farahead and its representatives.

It is necessary to give some account of the protracted and somewhat tortuous history of the proceedings, which have taken over 6 years to date and not yet resolved the question of the jurisdiction of the English court.

On 12 February 2015 the Respondents applied without notice to Teare J for, and were granted, a freezing order. The Claim Form, with Particulars of Claim attached, was issued the next day. As against Mr Bosworth and Mr Hurley these pleaded that they each owed to the Arcadia Claimants both fiduciary duties as (de jure or de facto or shadow) directors and contractual duties as employees, and advanced claims for (i) unlawful means conspiracy (relying on breaches of fiduciary and contractual duties as the unlawful means), (ii) breach of fiduciary duty, and (iii) breach of contractual duties of loyalty and fidelity.

On 9 March 2015 Mr Bosworth and Mr Hurley applied for dismissal of the claims against them for lack of jurisdiction on the grounds that such claims fell within the exclusive jurisdiction of the courts of Switzerland, an application optimistically described by their then solicitor as turning on a short point of law, namely that all the claims related to their employment by the Arcadia Group.

That application was heard by Burton J (as the Judge then was) in the Commercial Court. He gave judgment at [2015] EWHC 1030 (Comm) on 1 April 2015. By that stage the Respondents had sought to delete any reference to breach of contract in their pleading. The Judge held that the conspiracy claim was a tortious claim that did not relate to contracts of employment, and that the Respondents had a good arguable case that the breach of fiduciary duty claims also fell outside Art 18(1) of the Lugano Convention save where the relevant defendant had a contract of employment with the relevant claimant.

An appeal to this Court by Mr Bosworth and Mr Hurley was dismissed at [2016] EWCA Civ 818 on 19 August 2016. In his judgment Gross LJ (with whom Gloster and Macur LJJ agreed) agreed with the Judge that the conspiracy claims were, as a matter of substance, correctly characterised as claims in tort and did not relate to the Appellants’ contracts of employment, and that the breach of fiduciary duty claims did not relate to those individual contracts of employment either, other than for the periods for which (and in relation to the companies by which) the Appellants were employed. He also held that claims which had been added by amendment for dishonest assistance and knowing receipt fell to be treated in the same way as the conspiracy claim.

Both before Burton J and this Court in 2016 the parties and the Court proceeded on the basis that, at least as between the Appellants and the companies with which they had, as a matter of domestic law, contracts of employment, they were employees for the purposes of Arts 18 to 21 of the Lugano Convention. The argument was primarily over whether the claims related to those contracts of employment (i.e. the second of the two questions under Art 18(1) identified in paragraph 9 above).

On further appeal by the Appellants to the Supreme Court, however, the members of the Court questioned the assumption that the Appellants were employees at all for the purposes of the Lugano Convention. The Court made a reference to the CJEU and included a question addressed to that issue. Four questions were asked. Questions 1 and 3 were concerned with the correct test for whether a claim “relates to” a contract of employment. Question 4 asked whether claims by companies in the same group as a defendant’s employer also fell within Arts 18 to 21 of the Lugano Convention. Question 2 was in these terms:

“If a company and an individual enter into a ‘contract’ (within the meaning of Article 5(1) of the Convention), to what extent is it necessary for there to be a relationship of subordination between the company and the individual for that contract to constitute an ‘individual contract of employment’ for the purposes of Section 5 [of Title II of the Convention]? Can such a relationship exist where the individual is able to determine (and does determine) the terms of his contract with the company and has control and autonomy over the day-to-day operation of the company’s business and the performance of his own duties, but the shareholder(s) of the company have the power to procure the termination of the relationship?”

Advocate General Saugmandsgaard Øe delivered his Opinion on 24 January 2019. He gave his suggested answer to Question 2 first, although he went on to consider the other 3 questions. The CJEU gave judgment on 11 April 2019: Bosworth v Arcadia Petroleum Ltd (Case C-603/17) [2019] ILPr 22 (“Bosworth (CJEU)”). They also considered Question 2 first. I will have to look in detail at their reasoning and conclusions below, but in summary they said that, on the basis of the information provided by the referring court, it appeared that Mr Bosworth and Mr Hurley were not in a relationship of subordination, and that a contract concluded in circumstances such as those at issue in these proceedings did not constitute an individual contract of employment. There was therefore no need to answer the other questions.

That was, as Mr Fionn Pilbrow QC, who appeared with Mr David Heaton for the Respondents, put it, a decision expressed in what seasoned observers of the CJEU might consider unusually robust and clear terms. Nevertheless when the matter was referred back to the Supreme Court, there were extensive submissions as to the consequences, the Respondents suggesting that the jurisdiction challenge should be dismissed there and then (and indeed that the breach of fiduciary duty claims which Burton J had disallowed should be allowed back in), the Appellants that there was an issue estoppel or abuse of process in the Respondents resiling from the position that they were employees. All these submissions were rejected by the Supreme Court, but they were however persuaded by the Appellants that there was further evidence on the question of subordination which needed to be considered before the Court could reach a conclusion on whether they were in a position of subordination. The Supreme Court therefore remitted the case for further consideration of that question, their Order reading in paragraph 1 as follows:

“The appeal be allowed but only to the extent of remitting the case to the Commercial Court to hear further evidence and submissions on whether the Appellants were in “a relationship of subordination” to their employing company or companies in the sense used by the Court of Justice of the European Union in its judgment of 11 April 2019, Case C-603/17, so as to place them in an employment relationship to which Section 5, article 18(1) of the Lugano II Convention could apply.”

So it was that the case was remitted to the Judge, now Sir Michael Burton GBE, almost exactly 5½ years after he had first decided it. He handed down the Judgment on that issue on 19 October 2020.

Having set out the history at [1]-[5], the Judge dealt first with some preliminary questions at [6]-[9], and then considered the question as to the nature of the test which he had to adopt on a challenge to the jurisdiction, having regard to the well-known Supreme Court decisions in Brownlie v Four Seasons Holdings Inc [2017] UKSC 80 and Goldman Sachs International v Novo Banco SA [2018] UKSC 34. His conclusion at [13] was that the claimant must establish a good arguable case; that for this purpose the Court must decide, it if can, who has the better of the case; but that where the judge cannot decide, after conscientiously doing his or her best, who has the better of the case (due to the evidential limitations involved at the jurisdiction stage), then it is sufficient if the claimant has a plausible evidential basis. The Appellants sought to challenge that by amending their Grounds of Appeal to add a Ground 3 directed at it (based on the recent decision of the Supreme Court in Okpabi v Royal Dutch Shell plc [2021] UKSC 3 (“Okpabi”)), but permission was refused by Males LJ on 17 March 2021, and again on 31 March 2021. The correctness of the Judge’s approach is therefore not something we need to consider, but for what it is worth I do not detect in Okpabi (which Mr Richard Eschwege, who appeared for the Appellants, in fact showed us in the course of his submissions) any reason for thinking that there was anything wrong with the Judge’s pithy summary of the approach. The Judge of course has long experience as a judge of the Commercial Court and must be very familiar with the jurisprudence and practice on challenges to the jurisdiction.

At [14]-[25] the Judge considered what the CJEU required in order to establish a relationship of subordination. His conclusion at [25] was:

“I am satisfied that the issue which the CJEU resolved, though on the basis of assumed facts which I must now reconsider, is that the Defendants had a non-negligible influence over the Group companies of which they were CEO and CFO.”

At [26] he said that both sides had directed their evidence (which was extensive) to that issue and summarised their rival contentions as follows:

“The Claimants contend that the Defendants “called the shots” in relation to the companies they ran, while the Defendants claim that they were entirely subordinate to Farahead and Mr Fredriksen (and his associates).”

At [27] to [35] he summarised that evidence which he described as “a morass of papers”. He cited from certain documents and the witness statements, and then at [31] to [33] summarised two rival schedules, one prepared for the Appellants by their solicitor said to contain examples of the “directorial powers that Farahead and/or Mr Fredriksen and/or Mr Troim exercised over the Arcadia Group” (Mr Trøim being Mr Fredriksen’s then business partner), and one prepared for the Respondents by their junior counsel Mr Heaton said to contain examples “showing the absence of a relationship of subordination”, identifying which of the examples in each schedule was contested in the evidence. At [34]-[35] he considered in detail one specific area of contentious evidence concerning a restructuring in 2008/9 for tax reasons.

He then gave his conclusion at [36] in terms which it is worth citing in full:

“I am left to decide the question on the basis of this very contested and untested evidence, which I have carefully considered, both before and at the hearing and, as Lord Templeman once advised, in my room afterwards, as to whether the Defendants were in a relationship of subordination to the Arcadia companies because of the degree of control of the companies and of the Defendants by Farahead (and Mr Fredriksen). Just as the Advocate General pointed to the fact (at paragraph 41 of his Opinion) that in Holterman the Court “stated that a director having a sufficient share of the capital to influence in a ‘non-negligible’ manner the persons normally competent to give him instructions and to supervise their implementation cannot be subordinate to the company”, so the Claimants can say here that these Defendants between them, without a shareholding, had such power over the Arcadia companies that they were in a position to exercise that same influence. Whether the Defendants had the same powers as, or greater powers than, a normal CEO and CFO in such a situation, and in a case in which Farahead is in Cyprus, Mr Fredriksen in London and the Defendants are running an international group of companies, I do not need to decide. I am entirely clear however that, on the basis that the Claimants bear the onus to establish jurisdiction and my task has been to set their evidence against the rival evidence for the Defendants, and weigh it all in the context of such contemporaneous documents as are before me, I am satisfied that the Claimants have a good arguable case that there is not such a relationship, in that the Defendants had a more than negligible ability to influence the Arcadia companies. If I have, after such a difficult task, to conclude, on my assessment of the present evidence, without cross-examination, weighing the balance of the two Schedules, and taking into account, in addition to the submissions and evidence of the parties before me, in particular the Singapore Investment Memoranda and the witness statement of Ms Vaswani, that the Claimants have the better case, I do so. I am in any event satisfied, with reference to limbs (ii) and (iii) of the Brownlie test, that they have a plausible evidential basis.”

This is a concise but admirably clear conclusion, reached as I say by a very experienced Commercial Court judge, entirely in line with the exhortation by Davis LJ in Kaefer Aslamientos SA de CV v AMS Drilling Mexico SA de CV [2019] EWCA Civ 10 at [124] that judgments in such cases should so far as possible be concise.

The Judge therefore dismissed the defendants’ challenge to the jurisdiction in relation to the first three Claimants (the Arcadia Claimants); and said that the same result must also apply in respect of the fourth Claimant (Farahead) (Jmt at [37]).

By his Order dated 19 October 2020 he declared that:

“The Claimants have a good arguable case that there was no relationship of subordination in the sense used by the Court of Justice of the European Union in its judgment of 11 April 2019, Case C-603/17, between the Claimants (or any of them) and the First and Second Defendants (or either of them) and, accordingly, that neither the First Defendant nor the Second Defendant was in an employment relationship with the Claimants (or any of them) to which Section 5, article 18(1) of the Lugano II Convention could apply.”

He also granted Mr Bosworth and Mr Hurley permission to appeal in respect of two Grounds of Appeal, but on terms that there should be no challenge to the findings in [26] to [36] of the Judgment.

The two Grounds of Appeal are elaborated at some length but can be summarised as follows:

The Appellants derive the Holterman test from the decision of the CJEU in Holterman Ferho Exploitatie BV v Spies von Büllesheim (Case C-47/14) (“Holterman”) at [47]. The Appellants’ case is that this test was adopted and endorsed by the CJEU in the present case (Bosworth (CJEU) at [31]). And although Ground 2 is presented as a separate ground, it is really another argument put forward by the Appellants as to why the correct test is the Holterman test rather than the test adopted by the Judge. There is in effect therefore only one question raised by the appeal. Indeed as Mr Eschwege said, it can be reduced to the simple question: is the test for subordination the Holterman test or not?

Mr Eschwege developed his argument by reference first to a detailed analysis of Holterman. Holterman was a decision on a preliminary reference from the Hoge Raad in the Netherlands in proceedings brought by a Dutch holding company and three German subsidiaries in the Netherlands against Mr Spies von Büllesheim, a former manager who was domiciled in Germany. It raised questions on the applicability of Arts 18 to 21 of the Brussels Regulation (EC Regulation 44/2001), the wording of which is materially identical to the corresponding provisions in the Lugano Convention.

I will consider Holterman in more detail below, but I prefer to start with the decision of the CJEU in the present case (Bosworth (CJEU)). Not only is this binding on the parties for what it decides, but the Supreme Court remitted the case to the Commercial Court to consider whether the Appellants were in a relationship of subordination “in the sense used by the [CJEU] in its judgment of 11 April 2019” (paragraph 28 above). It is therefore to that judgment that we must primarily look.

In that judgment the CJEU, after a brief introduction at [1] to [2], set out the text of the relevant articles of the Lugano Convention at [3] to [5], summarised the dispute in the main proceedings at [6] to [14], and set out the questions referred at [15]. It is not necessary to refer to any of that other than to note that it is clear from [6] to [8] that they understood the difference between (i) Arcadia London, Arcadia Singapore and Arcadia Switzerland (the three Arcadia Claimants); (ii) the Arcadia Group (including, but not limited to, the three Arcadia Claimants, but not including Farahead, which they referred to as the 100% owner of the Group); and (iii) the four Claimants, being the three Arcadia Claimants and Farahead, which they referred to together as “Arcadia”. As appears below Mr Eschwege placed some reliance on this use of “Arcadia”, so I will say here that on my reading of the judgment they were simply using it as a convenient shorthand for the Claimants in the litigation.

After disposing of a preliminary procedural point they said at [20] that it was appropriate to examine Question 2 first. They proceeded to do that at [21] to [35]. We were taken through the entirety of this carefully by Mr Pilbrow, and I found this a useful exercise. Judgments of the CJEU are not written in the discursive style familiar from English judgments. They are concise and tightly worded. That sometimes makes them opaque and their meaning has to be teased out of them. I propose therefore to go through the judgment with comments as appropriate.

At [21] the CJEU restated Question 2 in their own words. At [22] they made the point that the interpretation of the corresponding provisions of the Brussels Regulation can be applied to the Lugano Convention – as already referred to, Holterman was itself a decision on the Brussels Regulation.

At [23] they said:

“23.

In order to determine whether the provisions of Section 5 of Title II (arts 18–21) of the Lugano II Convention are applicable to a situation such as that at issue in the main proceedings, it is necessary to consider whether Mr Bosworth and Mr Hurley can be regarded as having been party to an “individual contract of employment”, within the meaning of art.18(1) of that Convention, with one of the companies in the Arcadia Group, and whether they can therefore be classified as “employees”, within the meaning of art.18(2) of that Convention (see, to that effect, [Holterman] [34]).”

The reference to Holterman at [34] is to a statement in Holterman that the special rules for determining jurisdiction in Arts 18 to 21 of the Brussels Regulation only applied if Mr Spies von Büllesheim could be considered to be bound through an “individual contract of employment”. That is plain enough and unsurprising; it follows from the wording of Art 18(1).

There is one other point to notice on [23], which is that they refer to it being necessary to consider whether Mr Bosworth and Mr Hurley can be regarded as having been party to a contract of employment with “one of the companies in the Arcadia Group”. As referred to above, it is clear that they understood that the three Arcadia Claimants were companies in the Arcadia Group, but Farahead was not. Why did they pose the question in this way? The explanation I think is that there is no suggestion in the reference that Mr Bosworth or Mr Hurley ever held any position (de facto or de jure) in, or ever acted for and on behalf of, Farahead. Nor indeed had it ever been pleaded that they were employees of Farahead – the original pleading was that they were employees of the Arcadia Group; and Mr Eschwege confirmed to us in terms that Mr Bosworth and Mr Hurley never held any position in Farahead and never provided any services to Farahead. Consistently with that the schedule to the order for reference, which is drawn from, and closely follows, the Agreed Statement of Facts before the Supreme Court, refers to Mr Bosworth and Mr Hurley as having been de facto CEO and CFO of the Arcadia Group; as being responsible for the day-to-day business of the Arcadia Group; as acting for and on behalf of Arcadia London, Arcadia Switzerland and Arcadia Singapore; as having a contract of employment with one or other companies in the Arcadia Group; as being on the payroll of companies in the Arcadia Group; and as having acted on behalf of all Arcadia Group companies. In those circumstances it is not to my mind surprising that the CJEU understood the question to be whether Mr Bosworth and Mr Hurley had contracts of employment (in the EU sense) with one or more companies in the Arcadia Group.

At [24] they said:

“24.

In that regard, it should be pointed out that any such classification cannot be determined on the basis of national law ([Holterman] [36]) and that, in order to ensure that the Lugano II Convention, in particular art.18 thereof, is fully effective, the legal concepts it uses must be given an independent interpretation common to all the contracting parties (see, to that effect, judgments of 19 July 2012, Mahamdia v Algeria (C154/11) EU:C:2012:491; [2013] C.E.C. 452; [2012] I.L.Pr. 41, [42], and [Holterman] [37]).”

The principle that concepts used in EU instruments are not to be understood by reference to national law but are to be given an independent or autonomous interpretation is very well established. This paragraph contains two more references to Holterman at [36] and [37], but these say no more than that, in the specific context of a contract of employment for the purpose of Art 18(1) of the Brussels Regulation.

At [25] they said:

“25.

As regards the concept of “employee”, it must also be recalled that, as the Court has consistently held, that concept must be defined in accordance with objective criteria which distinguish the employment relationship by reference to the rights and duties of the persons concerned. The essential feature of an employment relationship is that, for a certain period of time, a person performs services for and under the direction of another person, in return for which he receives remuneration

(see, in particular, judgment of 20 September 2007, Kiiski v Tampereen Kaupunki (C-116/06) EU:C:2007:536; [2008] 1 C.M.L.R. 5; [2008]

C.E.C. 199, [25] and the case law cited).”

If one follows up the reference to Kiiski v Tampereen Kaupunki (C-116/06) (“Kiiski”) at [25] one there finds the Court saying that the concept of worker is another example of something that may not be interpreted differently according to national law but must be given a Community meaning, and describing the essential feature of an employment relationship in the terms here set out, referring to a line of cases back to 1986. The CJEU is not bound by any rigid doctrine of precedent but seeks to maintain consistency in its decisions in the interests of legal certainty, and frequently, as here, cites propositions verbatim as settled law. Kiiski was not a case about jurisdiction, and I accept Mr Pilbrow’s submission that what one can take from this paragraph is that the essence of an employment relationship – the performing of services for and under the direction of another in return for remuneration, or what the Advocate General in this case described at [AG34] as the features of “the performance of services, remuneration and subordination” – is the same across all areas of EU law.

But I also accept his further submission that although the basic features of the concept of employee or worker run across the whole of EU law, that does not mean that one can automatically transpose the working out of that concept from one area to another unthinkingly. This is well illustrated by the Advocate General in the present case. Having referred at [AG48] to cases (Danosa v LKB Līzings SIA (Case C-232/09) and Balkaya v Kiesel Abbruch- und Recycling Technik GmbH (Case C-229/14)) in which the Court had held that directors could be workers for the purposes of directives relating respectively to the rights of pregnant workers and to collective redundancies, he continued at [AG49]:

“However, the interpretation which the Court of Justice gives to a concept in one field of EU law cannot automatically be applied in a different field. As I have indicated, this is only one source of inspiration. The concept of “individual contract of employment”, within the meaning of Section 5, must be interpreted principally by reference to the scheme and objectives of the Lugano II Convention and the Brussels I Regulation and to the general principles emerging from national legal systems. The abovementioned precedents may therefore be transposed to those instruments only with caution. I would also note that, in the judgment in Holterman, the Court did not apply that case law in express terms; it merely referred to it on certain points.”

A similar point was made by Advocate General Cruz Villalón in Holterman at [AG25].

Reverting to the judgment of the CJEU, at [26] they said:

“26.

It follows that an employment relationship implies the existence of a hierarchical relationship between the worker and his employer, and that the issue whether such a relationship exists must, in each particular case, be assessed on the basis of all the factors and circumstances characterising the relationship between the parties ([Holterman] [46], and [judgment] of 20 November 2018, Sindicatul Familia Constanta v Directia Generala de Asistenta Sociala si Protectia Copilului Constanta

(C-147/17) EU:C:2018:926; [2019] 2 C.M.L.R. 8, [42]).”

Two points can be taken from this paragraph. First the focus is on the relationship between the worker and his putative employer and whether that is hierarchical. That makes logical sense. The question is whether A (a person providing services) is in an employment relationship with B (his putative employer). That, according to the CJEU depends on whether that relationship is hierarchical.

Second, the question whether a relationship exists must be assessed on the basis of all the factors and circumstances that characterise that relationship. That is indeed what is said in Holterman at [46], and repeated in the Constanţa case at [42]. There is obviously a tension between the requirement to assess the relationship on the basis of all the factors and circumstances that characterise it, and the desire that questions of jurisdiction under the Brussels Regulation and Lugano Convention should be determined with a high degree of predictability, a point I will come back to under Ground 2. For present purposes it is to be noted that the inquiry is a factual one, not one determined solely on the basis of the terms of the contract. For that one can refer to the Opinion of the Advocate General (who was again Advocate General Saugmandsgaard Øe) in BU v Markt24 GmbH (Case C-804/19) [2021] IL Pr 14 at [AG29] fn 15:

“15.

I do not mean to suggest that a contract is classified, for the purposes of art.20(1) of the Brussels Ia Regulation, solely on the basis of its terms. Indeed, where those terms do not reflect how that relationship is conducted in practice, the factual reality prevails (see [AG61] of this Opinion). Specifically, a contract held out as being an “individual contract of employment” will not be classified as such if, in actual fact, one party is not subordinated to the other, and vice versa (see, by analogy, [Bosworth (CJEU)], [34]).”

At [27] of the present case the CJEU said:

“27.

It should, moreover, be noted that, according to the wording of the provisions of Section 5 of Title II (arts 18–21) of the Lugano II Convention, the conclusion of a contract is not a condition for the application of the rules of special jurisdiction laid down in those provisions, and therefore that, as the Advocate General, in essence, indicated in [AG34]–[AG36] of his Opinion, the absence of any formal contract does not preclude the existence of an employment relationship that falls within the concept of “individual contract of employment” within the meaning of those provisions.”

At [28] the CJEU said:

“28.

However, such a relationship can be treated as an “individual contract of employment” within the meaning of the provisions of Section 5 of Title II (arts 18–21) of the Lugano II Convention only if there is a relationship of subordination between the company and the director concerned.”

This is the first reference in the CJEU judgment to “subordination”, but it was used in Question 2 itself, and extensively by the Advocate General in his Opinion. So far as one can tell from the material before us, the term comes from Holterman: see at [40] where the CJEU referred to the Report by Mr Jenard and Mr Möller on the Lugano Convention, according to which the concept of a “contract of employment” might be considered to presuppose a relationship of subordination of the employee to the employer (and see also the Opinion of the Advocate General in that case at [AG28] to [AG32]). It is however entirely consistent with the basic features of the EU concept of employment under which an employee provides services for and under the direction of his employer.

It may be noted that in [28] the CJEU gives their answer to the first part of Question 2, which was in these terms (see paragraph 26 above):

“If a company and an individual enter into a ‘contract’ (within the meaning of Article 5(1) of the Convention), to what extent is it necessary for there to be a relationship of subordination between the company and the individual for that contract to constitute an ‘individual contract of employment’ for the purposes of Section 5 [of Title II of the Convention]?”

The answer given by the CJEU at [28] is that it is necessary. The combined effect of [27] and [28] is that in order for there to be a contract of employment for the purposes of the Lugano Convention you do not need anything in the way of a formal contract of employment but you do need a relationship of subordination. Again it is to be noted that the answer in [28] is framed in terms of a relationship between a company and the director concerned.

At [29] to [30] the CJEU said:

“29.

In the present case, it should be noted that, according to the information provided by the referring court, Mr Bosworth and Mr Hurley were, respectively, chief executive officer and chief financial officer of the Arcadia Group, that they were directors of Arcadia London, Arcadia Singapore and Arcadia Switzerland, that they were each party to a contract of employment with one of those companies drafted by themselves or at their direction and that they acted at all material times on behalf of all Arcadia Group companies.

30.

It is also apparent from the order for reference that Mr Bosworth and Mr Hurley exercised control over by whom, where and on what terms they were employed.”

These matters were, as they say, all taken from the order for reference. The four matters referred to in [29] are each found in the Agreed Statement of Facts before the Supreme Court. None of them is disputed. The fifth matter, referred to in [30], is not in terms found there but it was in the schedule to the order for reference, and was taken from the judgment of Gross LJ in the judgment of this Court in 2016 at [2]. There was some debate before the Judge whether this could be reopened before him (Jmt at [9]), but on the present appeal Mr Eschwege accepted that the Judge recorded that one of the items on the Respondents’ schedule was that the Appellants moved their employment to different Arcadia Group entities as they considered expedient, including for their own benefit, and determined the terms on which they were employed (Jmt at [33(13)]), and that the Judge found that there was a good arguable case on the point. He also expressly accepted that the CJEU were right in what they said in [30], although he said that that was incomplete because it did not include any reference to what he called the instruction-giving role of Farahead.

At [31] the CJEU said:

“31.

In the circumstances, it appears that Mr Bosworth and Mr Hurley had an ability to influence Arcadia that was not negligible and that, therefore, it must be concluded that there was no relationship of subordination (see, to that effect, [Holterman] [47]), irrespective of whether or not they held part of the share capital of Arcadia.”

The interpretation of this paragraph is the key battleground for Ground 1. Before coming to Mr Eschwege’s argument, I will say what it seems to me, on a simple reading of the paragraph, the CJEU was saying. First, the reference to “In the circumstances” shows that their conclusion is built on the factors and circumstances that they have just referred to. That can only be a reference to the matters set out at [29] and [30], which, as I have said, have either always been common ground ([29]), or have now been found by the Judge to be a matter on which the Respondents have established a good arguable case ([30]). Second, the CJEU said that it did not matter whether or not the Appellants held any of the share capital. That is no doubt because in Holterman it was the fact that Mr Spies von Büllesheim had a shareholding in the Dutch holding company that raised a question whether his relationship with that company was one of subordination. What the CJEU is saying here is that that is not the only circumstance which gives rise to that question.

Third, on the face of the paragraph the CJEU says that the circumstances they have referred to establish that the Appellants had a non-negligible ability to influence “Arcadia” and that that was sufficient to demonstrate that there was no relationship of subordination. I will come back to the use of the term “Arcadia” here, which gave rise to some argument, but, leaving that point aside, the reasoning of the CJEU seems simple enough. It is that if A is in a position where he can exercise non-negligible influence over B, A is not in a relationship of subordination to B. The CJEU also evidently thought that they had enough material to reach a conclusion on that, in particular in the facts recited at [30] that the Appellants exercised control over who employed them and on what terms. That too does not seem very complex or very surprising: the CJEU’s reasoning is evidently that if you can choose your own employer and your own terms of employment, it can be concluded that you have a non-negligible influence over your employer.

Now of course the Supreme Court was persuaded by the Appellants that the CJEU did not in fact have all the evidential material necessary to reach that conclusion, and remitted the case for further evidence. But that does not change the nature of the inquiry. The question remitted by the Supreme Court, namely whether the Appellants were in “a relationship of subordination” to their employing company or companies (see paragraph 28 above) to my mind therefore required the Judge to decide, at least in relation to each of the Arcadia Claimants, whether each of the Appellants was in a position to exercise non-negligible influence over that Claimant. If so, he was not in a relationship of subordination to that Claimant, and hence did not have an employment relationship with that Claimant. (I will come back to the position of Farahead where the issue is not quite the same).

Why then does Mr Eschwege say that the Judge erred in his identification of the test? The entirety of his argument relies on the reference in Bosworth (CJEU) at [31] to “see, to that effect, [Holterman] [47]”. Mr Pilbrow, referring to the fact that the French text of the report (which he suggested was the original language of the judgment) has “voir, en ce sens”, submitted that all the CJEU meant by this crossreference was to explain what was meant by a “non-negligible” ability to influence, or in other words a semantic or textual explanation. I do not think this particular point carries the weight he sought to put on it. It is not clear to me that the French text was in fact the original – the French text itself refers to the language of the case being English (which is what one would expect in a reference from the United Kingdom) in which case my understanding is that the primary text of the judgment will be the English text – but in any event it would appear that that the French phrase “voir, en ce sens” is the standard equivalent of the English phrase “see, to that effect” and that that is the ordinary way in which the CJEU cross-refers to one of its earlier decisions: see for example Bosworth (CJEU) at [22], [23] and [24]. Those references are references to earlier decisions for propositions of law, and there is no particular reason to think that the reference in [31] is any different.

Nevertheless the question remains what the proposition of law was that the CJEU meant by the cross-reference. Holterman, as I have already referred to, was a case where proceedings were brought in the Netherlands by (i) a Dutch holding company (Holterman Ferho Exploitatie BV, “Holterman”) and (ii) three German subsidiaries against Mr Spies von Büllesheim, a German national domiciled in Germany. He was engaged as managing director of Holterman under a contract, described by the Dutch court as a “contract of employment”, and also had a contract with each of the German subsidiaries, and in fact carried out his duties in Germany. Unlike the present case therefore, he was a putative employee of the holding company Holterman. Importantly, he also had a shareholding in Holterman. The four companies brought claims for damages for improper performance of his duties as director, deceitful or reckless performance of his duties under his contract of employment, and unlawful conduct. By the time it reached the Hoge Raad (Supreme Court of the Netherlands) the claims by the subsidiaries had been held to be justiciable only in Germany and the only issue left was the claim by Holterman. One of the issues raised (it appears initially by the Commission) was whether the contract was a contract of employment for the purposes of Art 18(1) of the Brussels Regulation (see at [AG22]). The Commission’s argument was that the relationship of a managing director to his company was not one of subordination at all.

The Advocate General did not accept this (see [AG30]), nor did he accept that the mere fact that the managing director has a shareholding is sufficient to prevent a relationship of subordination (see [AG31]); it all depended on the facts (see [AG 32]). The CJEU agreed (see [45] and [46]). They continued at [47]:

“47 It is for the referring court to examine the extent to which Mr Spies von

Büllesheim, in his capacity as a shareholder in Holterman Ferho Exploitatie, was able to influence the will of that company’s administrative body of which he was the manager. In that case, it will be necessary to establish who had authority to issue him with instructions and to monitor their implementation. If it were to turn out that Mr Spies von Büllesheim’s ability to influence that body was not negligible, it would be appropriate to conclude that there was no relationship of subordination for the purposes of the court’s case law on the definition of a worker.”

Mr Eschwege drew from this the proposition that Holterman had laid down a twostage test. First you had to ask who had the authority to issue instructions and monitor their implementation, and then you had to ask if the putative employee had a non-negligible ability to influence that body. By cross-referring to this paragraph in Bosworth (CJEU) at [31] the CJEU had, he submitted, incorporated and endorsed the whole of this two-stage test. The Judge had therefore applied the wrong test (and, he continued, if he had applied this two-stage test, he would or should have found that the instruction-issuing body for the Arcadia Claimants was Farahead, and that the Appellants had no ability to influence Farahead, as the Judge recorded to be common ground (Jmt at [25])).

With all respect to Mr Eschwege, this seems to me an ingenious but overelaborate reading of Bosworth (CJEU) [31]. If the CJEU had meant to refer to and incorporate a two-stage test, one would expect them to explain how that two-stage test applied to the present case. But there is no trace of that in their reasoning. What one can see from what they say is that if A has a non-negligible ability to influence B, then A is not in a relationship of subordination to B. That is to my mind the simple proposition for which they referred to Holterman at [47] and which they proceeded to apply. How you establish in any particular case whether A does have a non-negligible ability to influence B must depend on all the facts and circumstances, and Holterman at [47] contains some guidance as to how to assess whether a managing director of a company who holds a shareholding in the company is in a position to influence the company. But the question remained whether he was in a relationship of subordination to the company which employed him. I do not think the CJEU was laying down any sort of universal test of how to assess whether there was such a relationship; it was explaining on the facts of that case the inquiries that would need to be made. There is no reason why that should be applied unthinkingly to the rather different position of a Group CEO and Group CFO who decide who their employer is and write their own contracts, where, as I have said, the CJEU evidently thought that one could conclude that they had a non-negligible ability to influence their employers.

Mr Eschwege sought to place some reliance on BU v Markt24 GmbH where the CJEU again referred to Holterman (as well as Bosworth (CJEU)). But although this was another case which concerned Art 18(1) of the Brussels Regulation, it had nothing to do with any question of subordination (BU was engaged to carry out cleaning work), and there is no reference to how one establishes whether the relationship is one of subordination. I found it of no assistance on the present question.

For these reasons, I do not accept Mr Eschwege’s Ground 1. The reference in Bosworth (CJEU) at [31] to Holterman at [47] does not in my judgment import the whole of that paragraph in the form of the so-called Holterman test or require the two

This is precisely what the Judge concluded when he said (Jmt at [25]) that the issue which the CJEU was concerned with, and which he had to reconsider, was whether the Defendants had a non-negligible influence over the Group companies of which they were CEO and CFO, that is the Arcadia Claimants (see paragraph 31 above). In my judgment he was entirely right in this conclusion. Indeed the detailed forensic scrutiny of the CJEU’s decision before us over a day and a half has only served to increase my admiration for the Judge’s Judgment which not only dealt with this point entirely correctly in my view, but also covered a large amount of other ground, legal and factual, efficiently and effectively.

Having correctly identified the task that the Supreme Court set him, he then proceeded to answer that question. Given the terms on which permission to appeal was granted, none of that is, or can be, challenged before us. Indeed Mr Eschwege accepted in terms that the Appellants had a more than negligible ability to influence the Arcadia Claimants. Subject to the separate position of Farahead, that is sufficient to justify dismissing the appeal.

It is only necessary to add some brief points on other aspects.

Mr Eschwege placed some reliance on the fact that the CJEU referred (twice) in [31] to Arcadia, which meant all four Claimants together. He suggested that this meant that focus had to be on the group as a whole, and that the question therefore was whether the Appellants had a non-negligible ability to influence the group, including Farahead, which they did not.

I do not think that this is what the CJEU meant. As I have referred to, the question posed at [23] was whether the Appellants were in an employment relationship with one or other of the companies in the Arcadia Group. This echoes the Advocate General’s Opinion at [AG38] where he said:

“Therefore, it is necessary next to determine whether the relationships that existed between the defendants in the main proceedings, in their capacity as company directors, and each of the Arcadia companies (whether or not a formal contract existed at a given moment) may be regarded as “individual contracts of employment” within the meaning of the provisions of

Section 5.” (emphasis in original)

The resolution of that question depended on whether they were in a hierarchical relationship with their putative employer [26]. That is only the case if there is a relationship of subordination between the company and the director concerned [28]. How do you assess whether a person is in a hierarchical relationship, or a relationship of subordination, with a putative employer? By having regard to all the factors and circumstances characterising the relationship between the parties [26].

The question is therefore whether Mr Bosworth and Mr Hurley were in a hierarchical relationship, or relationship of subordination, with one or more putative employers, namely the companies in the Arcadia Group for which they acted as CEO and CFO. For the purposes of the claims in these proceedings that means the three Arcadia Claimants. As Mr Pilbrow said one had in theory to ask in turn if Mr Bosworth had an employment relationship with Arcadia London, with Arcadia Switzerland or with Arcadia Singapore, and then the same with Mr Hurley.

Mr Pilbrow said that it was unclear whether the CJEU really meant Arcadia rather than the Arcadia Group, but that it did not matter. On any view the CJEU was saying that Mr Bosworth or Mr Hurley had a non-negligible ability to influence the Arcadia Claimants. It did not matter whether they really meant to say anything about the ability of Mr Bosworth and Mr Hurley to influence Farahead, as this was not in play before them, it not being suggested that they had any role in Farahead. Moreover when the matter was remitted to the Judge, it was agreed that he would only need to consider the position of Farahead separately if he concluded that the Appellants were in an employment relationship with one or more of the Arcadia Claimants.

I confess to being left in some doubt why the CJEU said that the Appellants had a non-negligible ability to influence Arcadia, and whether they really intended to include Farahead in that. It may be, although this is admittedly not quite what they say, that all they meant was that the Appellants did not have the requisite relationship of subordination with any of the four Claimants, although strictly the reason for that in the case of the Arcadia Claimants was because of the non-negligible ability to influence them, and in the case of Farahead was because it had not been put forward as a putative employer at all. But for the reasons given by Mr Pilbrow, I do not see that it matters.

The Supreme Court remitted to the Judge the question whether the Appellants were in a position of subordination to “their employing company or companies” (paragraph 28 above). That undoubtedly required the Judge to consider whether they were in a position of subordination to each of the Arcadia Claimants, which he proceeded to do. It is not clear whether the Supreme Court considered that the expression “employing company or companies” would or might include Farahead; I rather doubt it given the way the case had been pleaded and presented, but it matters not as the Judge proceeded to consider it anyway (Jmt at [37]), where he said that given his conclusions in relation to the first three Claimants, the same must apply to Farahead. Mr Eschwege had an argument that if the Arcadia Claimants were employers then Farahead would be as well for the purposes of Art 18(1) (see Jmt at [27]) but there is no suggestion in the Judgment that he was maintaining that Farahead could be an Art 18(1) employer if none of the Arcadia Claimants was, and we were referred to passages in the transcript which appear to bear out what Mr Pilbrow said, namely that it was agreed that in those circumstances it would be unnecessary to consider separately the position of Farahead.

In those circumstances I do not think that the reference by the CJEU in [31] to Arcadia as opposed to the Arcadia Group assists Mr Eschwege. It does not detract from the fact that what the Judge had to consider, and did consider, was the question whether each of the Appellants was in a relationship of subordination to, at least, each of the three Arcadia Claimants; and that he in fact considered the position of Farahead as well.

I can deal with the remainder of the judgment of the CJEU quite briefly. At [32] to [35] they said this:

“32.

The fact that Mr Bosworth and Mr Hurley were answerable to the Arcadia Group’s shareholders who, through Farahead Holdings, had the power to “hire and fire” them, is irrelevant in that regard.

33.

As the Advocate General noted in [AG46] of his Opinion, neither the general directives which a director may be given by the shareholders of the company he directs for the orientation of that company’s business nor the legal mechanisms for control by shareholders point, in themselves, to the existence of a relationship of subordination, and therefore the mere fact that the shareholders have the power to revoke a directorship is not sufficient for the conclusion to be drawn that such a relationship exists.

34.

It follows from this that a contract concluded between a company and the director of that company does not constitute, in circumstances such as those at issue in the main proceedings, an “individual contract of employment” within the meaning of Section 5 of Title II (arts 18–21) of the Lugano II Convention.

35.

Having regard to the above, the answer to the second question is that the provisions of Section 5 of Title II (arts 18–21) of the Lugano II Convention must be interpreted as meaning that a contract between a company and a natural person performing the duties of director of that company does not create a relationship of subordination between them and cannot, therefore, be treated as an “individual contract of employment”, within the meaning of those provisions, where, even if the shareholder(s) of that company have the power to procure the termination of that contract, that person is able to determine or does determine the terms of that contract and has control and autonomy over the day-to-day operation of that company’s business and the performance of his own duties.”

[32] and [33] are responding to the part of Question 2 which refers to the power of shareholders to procure termination; [34] is a conclusion expressed in definite terms that the Appellants did not have contracts of employment; and [35] is the formal answer to Question 2, reflected in their ruling or dispositif. It may be noted that both in [34] and in [35] they refer to a contract between a company and a “director of that company”, which may serve as further confirmation that they were focusing on the relationship between the Appellants and the Arcadia Claimants (of which they were (de jure or de facto or shadow) directors) and not on the relationship between the Appellants and Farahead (of which they were not).

I can deal with Ground 2 very shortly. As already explained, this is not a freestanding challenge to the Judge’s conclusion but a second reason why it is suggested that the

This really encompasses two points. The first is that the test should be easy to apply and not require a fact-intensive inquiry. The difficulty with that is that it runs straight into the jurisprudence of the CJEU that the question of subordination depends on all the factors and circumstances (Holterman at [46], Bosworth (CJEU) at [26]), coupled with the principle that this is to be determined on the basis of the actual facts, not the terms of the contract of employment (see paragraph 49 above).

The second is that it should not result in too many people losing the protection of Arts 18 to 21. I do not think we can assess how widespread the impact might be, but I doubt that this case will be a precedent for many others. Even senior managers are usually in a relationship of subordination to their employers. The Appellants had (or, to be more precise, the Respondents have established a good arguable case that they had) an unusually free hand in running the Arcadia Group as their own private fiefdom to the extent of writing their own contracts of employment. I do not think the Judge’s Judgment means that any senior manager, or even director, who is given a degree of autonomy as to how he does his job is outside the protection of Art 18(1); as the Master of the Rolls said in argument, this may appear a low bar but it is not in fact because it is quite unusual for employees, even in the case of senior managers, to be in a position of influencing the decision-making of the company that is employing them.

There is in my view therefore nothing, or at any rate not enough, in Ground 2. But in any event, Ground 2 can only at best support Mr Eschwege’s argument on Ground 1. If, as for the reasons I have given is in my view the case, Ground 1 falls to be dismissed, Ground 2 cannot get him home by itself.

I would dismiss the appeal.