BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales High Court (Commercial Court) Decisions


You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> Etihad Airways PJSC v Flother [2019] EWHC 3107 (Comm) (18 November 2019)
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2019/3107.html
Cite as: [2019] WLR(D) 675, [2020] 2 WLR 333, [2019] EWHC 3107 (Comm)

[New search] [Context] [View without highlighting] [Printable PDF version] [Buy ICLR report: [2020] 2 WLR 333] [View ICLR summary: [2019] WLR(D) 675] [Help]


Neutral Citation Number: [2019] EWHC 3107 (Comm)
Case No: CL-2019-000047

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

The Rolls Building,
Fetter Lane, London, EC4A 1NL
18/11/2019

B e f o r e :

MR JUSTICE JACOBS
____________________

Between:
ETIHAD AIRWAYS PJSC
Claimant/
Respondent
- and -

PROF. DR. LUCAS FLΦTHER
Defendant/
Applicant

____________________

Robin Dicker QC and Roseanna Darcy (instructed by Shearman & Sterling (London) LLP) for the Claimant
David Joseph QC, Adam Kramer and Ian Higgins (instructed by Latham & Watkins LLP) for the Defendant

Hearing dates: 22nd and 23rd October 2019.

____________________

HTML VERSION OF JUDGMENT APPROVED
____________________

Crown Copyright ©

    Mr. Justice Jacobs:

    A: Introduction and Factual Background

    The application

  1. This is an application by the Defendant ("the Insolvency Administrator"), on behalf of Air Berlin plc ("Air Berlin") for declaratory relief and a stay of proceedings commenced by the Claimant ("Etihad") on 22 January 2019. Air Berlin disputes the jurisdiction of the English court on the basis that the jurisdiction clause contained in a Facility Agreement concluded between Etihad and Air Berlin, and which provides for the jurisdiction of the English court, has no application to the claims which have given rise to the present proceedings. Those claims are made by Air Berlin in proceedings commenced by the Insolvency Administrator of Air Berlin against Etihad in the Regional Court of Berlin on 24 July 2018 (the "German proceedings").
  2. The claims made in the German proceedings relate to a letter dated 28 April 2017 from Mr James Hogan, the then President and CEO of Etihad Aviation Group PJSC, to the directors of Air Berlin (the "Comfort Letter"), which provided as follows:
  3. "For the purposes of the finalisation of the financial statements of Air Berlin plc for the year ended 31 December 2016, having had sight of your forecasts for the two years ending 31 December 2018, we confirm our intention to continue to provide the necessary support to Air Berlin to enable it to meet its financial obligations as they fall due for payment for the foreseeable future and in any event for 18 months from the date of this letter. Our commitment is evidenced by our historic support through loans and obtaining financing for Air Berlin".
  4. In the German proceedings, Air Berlin advances two alternative claims against Etihad under German Law:
  5. i) A claim for breach of the Comfort Letter on the basis that the Comfort Letter is legally binding.

    ii) Alternatively, if the Comfort Letter is not legally binding, a pre contractual claim in culpa in contrahendo, on the basis that Etihad used its negotiating power during the negotiations between the parties to avoid providing a clearly binding statement whilst, at the same time, inspiring the trust of Air Berlin that it would adhere to the commitment in the Comfort Letter.

  6. The German proceedings were commenced prior to the present English proceedings, in which Etihad seeks the following declarations:
  7. a) The claims made and declarations sought in the German Proceedings are subject to the exclusive jurisdiction of the English court within Article 25 of the Judgments Regulation, because, on its true construction, they are within the scope of the exclusive jurisdiction clause contained in the €350m Facility Agreement;
    b) The claims made and declarations sought in the German Proceedings are governed by English Law on the true construction of the governing law clause in the €350m Facility Agreement, an implied agreement between the same parties and/or the application of Rome I and/or Rome II;
    c) The Claimant is not liable for breach of the Comfort Letter, as alleged in the German Proceedings, because that letter, on its true construction, did not create a legally binding promise to provide financial support to Air Berlin;
    d) The Claimant is not liable on the basis of culpa in contrahendo, as alleged in the German Proceedings, because the facts and matters relied on in the German Proceedings do not give rise to a cause of action known to English law; and
    e) Further, and in any event, the Claimant is not liable to the Defendant as alleged by the Defendant in the German Proceedings.
  8. Air Berlin's application raises issues as to:
  9. i) the scope of the jurisdiction clause under English law;

    ii) whether the jurisdiction clause is inapplicable as a matter of EU law because the relevant dispute does not arise in connection with the "particular legal relationship" between the parties, as required by Article 25 of EU Regulation No 1215/2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters ("Brussels Recast");

    iii) whether the present proceedings should be stayed in favour of the German court which is first seised, or conversely whether Article 31 (2) of the Brussels Recast applies so that the English court is not required to stay. This question depends upon a legal issue as to whether so-called "asymmetric" jurisdiction clauses fall within Article 31 of Brussels Recast. Air Berlin seeks a preliminary reference of that issue to the Court of Justice of the European Union.

    Factual Background

  10. Etihad acquired a 2.99% stake in Air Berlin in August 2011 and, in December 2011, increased its shareholding to 29.21% by subscribing for 31.57 million new shares for a subscription price of €72.9 million pursuant to an agreement dated 19 December 2011. That agreement, which was drafted in English, was expressed to be governed by English law and contained an exclusive jurisdiction clause in favour of the English courts.
  11. Between 2011 and 2016, Etihad invested or made available further finance to the Air Berlin group, amounting to approximately a further €721 million. The relevant agreements between Etihad and Air Berlin, pursuant to which such equity or debt finance was provided, were, almost without exception, governed by English law and subject to the exclusive jurisdiction of the English courts.
  12. Air Berlin had a long history of financial difficulties and, in 2016, it began working, together with various external advisors, on a restructuring. The proposed restructuring, which became known as the "new Air Berlin strategy", had three key elements: (a) the wet lease of 40 aircraft by Air Berlin to Lufthansa in respect of which Lufthansa would pay a deposit of €220 million; (b) a joint venture between Etihad and TUI (a travel and tourism company), which was to involve Etihad purchasing Air Berlin's subsidiary airline, NIKI, for €300 million and contributing it to the joint venture with TUI; and (c) the remaining businesses of Air Berlin continuing to focus on scheduled network traffic including long-haul routes. The initial intention was that Air Berlin would fund its restructuring with the cash generated by the first two elements of this plan.
  13. Given that Air Berlin was a UK public limited company with its registered office in London, its annual financial statements needed to be prepared in accordance with the Companies Act 2006. In November 2016, KPMG, Air Berlin's external auditors, identified going concern and funding as significant issues for the forthcoming audit. Concerned to ensure that KPMG would sign off on Air Berlin's financial statements at the end of April 2017 on a going concern basis, the management of Air Berlin communicated various requests for financial support to Etihad.
  14. In November 2016, Etihad engaged Ernst & Young ("E&Y") to perform an independent review of Air Berlin's cash requirements. E&Y concluded that, in addition to the proceeds of the Lufthansa wet leases and the NIKI transaction, Air Berlin was likely to need an additional €200 million. In December 2016, Etihad indicated that, subject to formal approval, it would be willing to provide Air Berlin with an additional €350 million, thus providing a 'buffer' of €150 million over and above the amount that E&Y had identified.
  15. In February 2017, Etihad assisted Air Berlin to refinance €140 million 6% Convertible Bonds issued on 6 March 2013 in respect of which a put option was shortly due to arise on 6 March 2017 (the "Old Bonds"). Air Berlin proposed an exchange offer involving the issue of new €125 million 8.5% Guaranteed Convertible Bonds due 2019, with a put option date of 29 December 2017 (the "New Bonds"). Etihad agreed to support this proposal by accepting the exchange offer in respect of the €40 million of bonds that it held in the Old Bonds and by entering a total return swap with HSBC ("the Total Return Swap"), which had agreed to subscribe for €53.7 million of the New Bonds.
  16. During March and April 2017, KPMG identified various further requirements for it to sign-off Air Berlin's financial statements on a going concern basis and without an emphasis of matter, which Air Berlin sought to address mainly by looking to Etihad to provide additional financial support. The correspondence and discussions are described in more detail in Section B below. One important feature was that Air Berlin was now said to require additional cash accumulating to €558 million, some €208 million more than the €350 million that Etihad had already indicated that it was willing to provide. In addition, Air Berlin was found to need 'non-cash' support which amounted to a further €600 million. This related to: the put option on the New Bonds which was exercisable on 29 December 2017; the need to extend the loans granted to Air Berlin by National Bank of Abu Dhabi PJSC ("NBAD") and Abu Dhabi Commercial Bank PJSC ("ADCB"); and the need to refinance €225 million 8.25% Fixed Rate Notes due to mature in April 2018.
  17. Between 5 and 28 April 2017, Air Berlin and Etihad, together with KPMG, discussed the ingredients and terms, and exchanged drafts, of the arrangements to support Air Berlin with financial support. At the same time, Air Berlin continued to pursue arrangements with third parties, in particular the German state of North Rhine Westphalia and Lufthansa, to support its liquidity needs, the indications on which were positive. Again as further described below, Etihad had made it clear that by 26 April 2017, it would not agree to increase the amount of the loan facility from €350 million to €610 million, contrary to what they had initially discussed. In addition, Etihad informed Air Berlin that it would not agree to underwrite the refinancing of the €225 million 2018 Notes. KPMG had also identified a further item, namely the risk that Air Berlin might have to repay the €300 million in respect of the NIKI sale, in the event that the joint venture between Etihad and TUI did not go ahead.
  18. Significant shortfalls in Air Berlin's finances therefore remained to be filled before its financial statements could be signed off on a going concern basis. It was in these circumstances that, on 26 April 2017, the Comfort Letter came to be discussed and negotiated.
  19. Ultimately, between 28 and 30 April 2017, Etihad entered into a number of agreements for the purposes of providing Air Berlin with financial support. The purpose of these agreements, together with the ratification of the Total Return Swap, was to enable KPMG to sign off on Air Berlin's financial statements on a going concern basis.
  20. One of these agreements was the Facility Agreement made where Etihad agreed to advance €350 million. It is this agreement which is alleged by Etihad to give rise to the jurisdiction of the English Courts. The Facility Agreement provided:
  21. 32. GOVERNING LAW
    This Agreement and all non-contractual obligations arising from or connected with it are governed by English law".
    33. ENFORCEMENT
    33.1 JURISDICTION
    33.1.1 The courts of England have exclusive jurisdiction to settle any disputes arising out of or in connection with this Agreement (including a dispute relating to non-contractual obligations arising from or in connection with this Agreement, or a dispute regarding the existence, validity or termination of this Agreement) (a "Dispute").
    33.2.2 The Parties agree that the courts of England are the most appropriate and convenient courts to settle Disputes and accordingly no Party will argue to the contrary.
    33.1.3 This Clause 33 is for the benefit of the Lender only. As a result, the Lender shall not be prevented from taking proceedings relating to a Dispute in any other courts with jurisdiction. To the extent allowed by law, the Lender may take concurrent proceedings in any number of jurisdictions.
  22. In addition, the Facility Agreement contained the following clauses which were referred to in the course of the arguments advanced on this application:
  23. 4.2 Further conditions precedent to each Utilisation
    4.2.1 The Lender will only be obliged to comply with a Utilisation Request if on the date of the relevant Utilisation Request and the proposed Utilisation Date related thereto the Lender is satisfied that:
    4.2.1.1 no Default and/or Potential Mandatory Prepayment Event is continuing or would result from the proposed Loan;
    4.2.1.2 the Repeating Representations made by the Borrower by reference to the facts and circumstances then existing are true;
    4.2.1.3 the Borrower is not and shall not be in breach of any of the provisions of Clause 4.4; and
    4.2.1.4 the provisions of Clause 5 (Utilisation) have been satisfied.
    4.2.2 On the proposed Utilisation Date, the Lender shall have received all of the items specified in Part 2 (Conditions Precedent to each Utilisation) of 0Schedule 2 in respect of the relevant Utilisation in form and substance satisfactory to the Lender.
    Clause 16.2 Requirements as to financial statements
    16.2.1 Each set of Financial Statements delivered pursuant to Clause 16.1 (Financial statements) shall be certified by a director of the Borrower as giving a true and fair view of the Borrower's consolidated financial condition as at the end of and for the period in relation to which those financial statements were drawn up.
    16.2.2 The Borrower shall ensure that each set of financial statements delivered pursuant to Clause 16.1 (Financial statements) is prepared in conformity with IFRS, accounting bases, policies, practices and procedures and financial reference periods consistent with those applied in the preparation of the Original Financial Statements of the Borrower except the changes expressly disclosed in the financial statements.
    19. EVENTS OF DEFAULT
    Each of the events or circumstances set out in this Clause 19 is an Event of Default.
    19.6 Insolvency
    19.6.2 The value of the assets (measured at the higher of the going concern value and the liquidation value) of the Borrower is less than its liabilities, and there is no positive going-concern prognosis (keine positive Fortfόhrungsprognose), such test to be applied in accordance with Section 19 para. 2 German Insolvency Code.
    19.13 Other events of default
    19.13.2 Any event or circumstance occurs which has a Material Adverse Effect.
  24. The other documents or agreements executed by 30 April 2017 were as follows:
  25. i) The Comfort Letter, which did not contain a jurisdiction clause.

    ii) A letter agreement pursuant to which Etihad undertook, amongst other things, that it would not exercise any put option rights that it might have in respect of the New Bonds exercisable on 29 December 2017 and would continue to hold at least €40 million of such bonds until their maturity date of 6 March 2019 (the "Put Option Letter Agreement"). The Put Option Letter Agreement was expressed to be governed by English law but did not contain a jurisdiction clause. The New Bonds themselves contained an English jurisdiction clause for disputes arising in connection with the bonds.

    iii) The ratification of the Total Return Swap by Etihad in favour of HSBC dated 15 February 2017. The Total Return Swap was expressed to be subject to English law and was subject to the jurisdiction of the English Courts pursuant to an English jurisdiction clause contained in the ISDA Master Agreement.

    iv) Extensions of guarantees dated 30 April 2017, which were provided by Etihad in respect of two loans that had previously been granted by NBAD and ADCB to Air Berlin. The NBAD Guarantee Amendment was expressly governed by English law and contained an asymmetric English jurisdiction agreement in favour of NBAD. The ADCB Guarantee Amendment was expressly governed by English law and included an asymmetric DIFC jurisdiction agreement in favour of ADCB.

    v) Reimbursement deeds also dated 30 April 2017, whereby Air Berlin undertook to indemnify Etihad in respect of those guarantees (the "Reimbursement Deeds"). The Reimbursement Deeds were expressly governed by English law and subject to the jurisdiction of the English courts.

  26. The directors of Air Berlin concluded that, as a result of these agreements, there was a sufficient likelihood of the necessary support being provided to Air Berlin to permit its financial statements to be drawn up on a going-concern basis. The auditor's report, which signed off on the financial statements on a going concern basis, recorded that this was dependent in part on the Comfort Letter: "The group is reliant on a letter of support from a significant shareholder; as with such letters there remains a doubt whether this can be enforced in the event that such need arises".
  27. During May, June and July 2017, Air Berlin encountered severe operational challenges which had an adverse effect on its revenues, costs and liquidity. In summary:
  28. i) Etihad disbursed a total of a further €250 million to Air Berlin, pursuant to the Facility Agreement, following drawdown requests in May, June and July 2017.

    ii) At the end of July 2017, however, Air Berlin produced a revised liquidity forecast, which showed a need for additional financing of close to €1 billion through to 2020. On Etihad's case, this sudden requirement for very substantial additional financing had not been reflected in any of the forecasts prepared by Air Berlin earlier in the year.

    iii) On 9 August 2017, Air Berlin made a further drawdown request of €50 million to which Etihad responded by stating that it was not satisfied that the necessary pre-conditions for a draw-down were met, in particular that no Default existed under the terms of the Facility Agreement, including defaults under clause 19.6.2 (Insolvency) and clause 19.13.2 (Other events of default) and, as a result, it would not meet the draw-down request or provide further funding.

  29. On 15 August 2017, Air Berlin applied to the Berlin court to open insolvency proceedings and, on 27 October 2017, ceased operations.
  30. On 24 July 2018, the Insolvency Administrator of Air Berlin issued the German proceedings. The claim form in those proceedings contains 232 paragraphs, some of which are set out below.
  31. On 23 January 2019, Etihad applied to the Berlin District Court seeking: (i) a stay of the German proceedings pursuant to Article 31 (2) of Brussels Recast (the "Stay Application"); and (ii) to extend the deadline for service of its defence until one month after the Stay Application had been determined (an extension until 31 January 2019 having already been granted). Etihad's application for an extension was dismissed and Etihad filed its statement of defence on 31 January 2019. The Berlin District Court is expected to determine the Stay Application in the coming months.
  32. On 22 January 2019, the day before issuing the Stay Application, Etihad issued its Part 8 claim in England seeking declaratory relief, in the terms set out above, in relation to the claims made in the German Proceedings. On 10 April 2019, the Insolvency Administrator of Air Berlin issued the present application and the matter came on for a two day hearing on 22-23 October 2019 at which oral arguments were presented by Mr. Joseph QC for Air Berlin and Mr. Dicker QC for Etihad.
  33. Etihad's evidence on the application comprised two witness statements from Mr Henning zur Hausen ("Mr zur Hausen"), the General Counsel & Company Secretary of the Etihad Aviation Group. The Insolvency Administrator served one witness statement from Mr Martin Stuart Davies ("Mr Davies"), a Partner of Latham & Watkins (London) LLP, the solicitors acting on its behalf. Such factual disputes as existed were within a very narrow compass, and the factual evidence did not significantly advance matters beyond what was apparent from the contemporary documents.
  34. B: The legal framework

  35. The parties' arguments focused on two Articles in Brussels Recast, Articles 25 and 31 (which qualifies Article 29). These provide as follows:
  36. SECTION 7
    Prorogation of jurisdiction
    Article 25
    1. 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, unless the agreement is null and void as to its substantive validity under the law of that Member State. Such jurisdiction shall be exclusive unless the parties have agreed otherwise.
    …
    SECTION 9
    Lis pendens – related actions
    Article 29
    1. Without prejudice to article 31 (2), where proceedings involving the same cause of action and between the same parties are brought in the courts of different member states, any court other than the court first seised shall of its own motion stay its proceedings until such time as the jurisdiction of the court first seised is established.
    …
    Article 31
    1. Where actions come within the exclusive jurisdiction of several courts, any court other than the court first seised shall decline jurisdiction in favour of that court.
    2. Without prejudice to Article 26, where a court of a Member State on which an agreement as referred to in Article 25 confers exclusive jurisdiction is seised, any court of another Member State shall stay the proceedings until such time as the court seised on the basis of the agreement declares that it has no jurisdiction under the agreement.
    3. Where the court designated in the agreement has established jurisdiction in accordance with the agreement, any court of another Member State shall decline jurisdiction in favour of that court.
  37. During the course of the parties' submissions as to the scope of the jurisdiction clause and the applicability of Article 25, I was referred a large number of English authorities: Fiona Trust & Holding Corp v Privalov [2007] UKHL 40 ("Fiona Trust"); Satyam Computer Services Ltd v Upaid Systems Ltd [2008] EWCA Civ 487 ("Satyam"); UBS AG v HSH Nordbank [2009] EWCA Civ 585; Choil Trading SA v Addax Energy SA [2009] EWHC 2472 ("Choil"); Cinnamon European Structured Credit Master Fund v Banco Commercial Portugues SA [2009] EWHC 3381 ("Cinnamon"); Sebastian Holdings Inc v Deutsche Bank AG [2010] EWCA Civ 998 ("Sebastian Holdings"); Trust Risk Group SpA v AmTrust Europe Ltd [2015] EWCA Civ 437; Deutsche Bank AG v Petromena [2015] EWCA Civ 226 ("Petromena"); Altera Absolute v Sapinda Invest [2017] EWHC 871 (Comm) ("Altera"); Deutsche Bank AG v Comune di Savona [2018] EWCA Civ 1740 ("Comune di Savona"); BNP Paribas v Trattamento Rifiuti Metropolitani [2019] EWCA Civ 768 ("Rifiuti"); Airbus SAS v Generali Italia SpA [2019] EWCA Civ 805 ("Airbus"). I was also referred to two European cases: Case C-214/89 Powell Duffryn v Petereit [1992] ECR I-1745 ("Powell Duffryn"); Case C-352/13 Hydrogen Peroxide SA v Azko Nobel NV [2015] ECR 2015.
  38. In relation to the parties' submissions as to the applicability of Article 31 to asymmetric clauses, I was referred to the decision of Cranston J. in Commerzbank AG v Liquimar Tankers Management Inc. [2017] EWHC 161 (Comm), where the background to Article 31 is explained in detail. The parties also referred to a large volume of academic and other materials which I describe in Section E below.
  39. Both parties made their submissions as to the scope of the jurisdiction clause, including the applicability of Article 25, prior to their submissions concerning Article 31 and the question of stay. These two areas of the case were, for all relevant purposes, separate; certainly once it had been accepted, as Air Berlin accepted in its written submissions (whilst reserving its position on a possible appeal), that an asymmetric clause is a jurisdiction agreement falling within Article 25. There was therefore no material overlap between the arguments advanced on these two areas of the case. I shall follow the parties' approach by considering the scope and Article 25 issues prior to addressing the Article 31 issue.
  40. There was, however, some disagreement as to whether it was appropriate to start by considering the scope of the jurisdiction agreement under English law, or to start by considering whether there was a qualifying agreement under Article 25 of the Brussels Recast. There was, however, no dispute that both questions needed to be addressed. Indeed, Air Berlin contended in its application notice that the English court had no jurisdiction to hear "the Claim" (i.e. the claim brought by Etihad in the present proceedings) on the grounds that "the Claim did not fall within the scope of a jurisdiction agreement between the parties in favour of the English court either as a matter of English law, or as a matter of autonomous EU law".
  41. It seems to me to be logical to start by considering the scope of the jurisdiction clause relied upon. This is because if – applying the applicable (English) law of the agreement relied upon as containing the relevant jurisdiction clause – the present dispute is not within its scope, then it is unnecessary to consider the matter any further. It is therefore unsurprising that the judgments in a number of the cases cited to me, including Petromena and Altera, have considered the scope of the jurisdiction clause under its applicable (English) law before considering whether Article 25 is applicable.
  42. Standard of proof. It was, or at least appeared to be, common ground that the test to be applied in both contexts is whether Etihad could establish a good arguable case that the English court has jurisdiction, in the sense that it has the better of the argument on the material available: see Airbus at [49] – [54]. It was common ground that this test applied whether the court was considering the scope of the jurisdiction clause relied upon as a matter of English law or under Article 25 of the Brussels Recast.
  43. There was, however some suggestion by Air Berlin that a more stringent test should be applied; because of the European Union law requirement that an exclusive jurisdiction clause under Article 25 must be "clearly and precisely" demonstrated. I address this below.
  44. The relevant enquiry. The question of whether Etihad can rely upon the jurisdiction clause must be determined by reference to the claim in relation to which the proceedings have been issued: Rifiuti at paragraphs [59] – [60]. It is therefore the terms of Etihad's claim which matter: Rifiuti at paragraph [89]. Thus, in the schedule to the judgment in Rifiuti, the court analysed each of the declarations sought in the proceedings, and determined whether they were referable to the jurisdiction clause in the ISDA Master Agreement which was relied upon. A similar approach was taken in Comune di Savona: see paragraphs [24] – [29] and the Appendix to the judgment.
  45. Etihad's Claim Form, leading to the declarations set out in paragraph [4] above, raises a number of related issues. Etihad alleges, in paragraph 5 of the Claim Form, that the jurisdiction clause in the Facility Agreement is applicable to the Comfort Letter and any non-contractual claim in connection therewith. It alleges in paragraph 6 that the Comfort Letter and any contractual claim in connection therewith are subject to English law on the true construction of the governing law clause in the Facility Agreement and pursuant to the EU regulations commonly known as "Rome 1" and "Rome 2". In paragraph 7, Etihad alleges that the Comfort Letter does not amount to a legally binding promise to provide financial support.
  46. Although the critical question is whether the claims brought in the present proceedings fall within the jurisdiction clause relied upon both under English law and Article 25, those claims themselves concern and are directed towards the claims made in the German proceedings. At the heart of Etihad's claim is the proposition that the claims made by Air Berlin in Germany fall within the scope of the jurisdiction clause. In practical terms, therefore, the essential question is whether there is a good arguable case that the claim commenced by Air Berlin in Germany falls within the scope of the jurisdiction clause relied upon by Etihad: see Airbus at [81] – [82]. This is essentially the same question as asking whether there is a good arguable case that, as alleged in paragraph 5 of the Claim Form, the jurisdiction clause in the Facility Agreement is applicable to the Comfort Letter and any non-contractual claim in connection therewith.
  47. C: The scope of the jurisdiction agreement under English law

    Etihad's submissions

  48. Since the burden of establishing a good arguable case for jurisdiction is upon Etihad, I begin by summarising Etihad's submissions.
  49. Etihad submits that clause 33.1 of the Facility Agreement is a widely worded clause, and that the dispute between it and Air Berlin is within its scope. This is because, adopting the broad, purposive and commercial approach to interpreting such clauses which has been mandated by the English authorities, the dispute arises out of or in connection with that agreement.
  50. Etihad relied upon the decision of the House of Lords in Fiona Trust as providing the appropriate assumption on which to proceed; namely that the parties, as rational businessmen, are likely to have intended any disputes arising out of their relationship to be decided by the same tribunal. The Fiona Trust assumption was not limited to cases where disputes arose under a single agreement with a jurisdiction or arbitration clause. The assumption was equally applicable where the agreements were part of a package or arrangement between them. This is not a case where there are different and inconsistent jurisdiction clauses. In that situation, the Fiona Trust assumption is not applicable, and it is necessary to decide which clause applies to the parties' particular dispute.
  51. Accordingly applying the required broad, purposive and commercially minded approach, in the light of the background facts, clause 33.1 should be construed as applying to Air Berlin's claims in respect of the Comfort Letter and its "culpa in contrahendo" case which related to the negotiations leading up to the provision of the "Support Package" agreed in April 2017. The Comfort Letter and the Facility Agreement were closely linked: the parties were the same, at least on Air Berlin's case; they were concluded on the same day; they had a common purpose, namely to enable Air Berlin's financial statements to be signed off; they were commercially linked, in that the origin of the Comfort Letter was Etihad's unwillingness to provide a loan facility of €610 million, but only €350 million; they were part of a single "Support Package". It would therefore be artificial to regard the Comfort Letter and the Facility Agreement as wholly free-standing documents.
  52. There was further linkage in that it was foreseeable that the resolution of a dispute under the Facility Agreement might require the court to determine the effect of the Comfort Letter and vice versa. This is illustrated by what happened in August 2017, when Air Berlin made a further drawdown request under the Facility Agreement which Etihad refused, on the basis that Air Berlin was in default of various pre-conditions of the Facility Agreement, including those concerning insolvency. Whether or not such default existed would potentially depend on whether the Comfort Letter contained a binding contractual promise or not. The parties cannot sensibly be taken to have intended that the effect of the Comfort Letter might fall to be determined both by the English court, in connection with a dispute about the validity of a draw-down request, and also by the German court in relation to the effect of the Comfort Letter itself.
  53. Furthermore, the factual background was that financial support had been provided over a number of years on the basis of agreements which, with few exceptions, were governed by English law and contained English jurisdiction clauses. The agreements concluded in April 2017 were also connected with English law and/or English jurisdiction. If the parties had confidence in English jurisdiction for the purposes of all of those agreements, why should they not have it for the purpose of any dispute as to the Comfort Letter? By contrast, looking at what happened historically and in April 2017, the parties' agreements were not governed by German law or jurisdiction.
  54. In oral argument, Mr. Dicker QC submitted that it was relevant to pay regard to the status of the Comfort Letter from an English perspective. Whilst it would not be appropriate on the present application finally to decide issues as to the applicable law governing the Comfort Letter, or whether it contained binding obligations, that did not prevent the court from considering these questions in the context of the "good arguable case" test. He submitted that there was, at the very least, a good arguable case that the Comfort Letter was governed by English law, which was the law applicable to related agreements concluded between the parties: see F R Lurssen Werft GmbH & Co. KG v Halle [2010] EWCA Civ 587. There was therefore a good arguable case that the Comfort Letter was no more than a statement of present intention, and did not give rise to enforceable contractual obligations: see Kleinwort Benson Ltd. v Malaysia Mining Corporation Berhad [1989] 1 WLR 379. On this basis, the Comfort Letter should not be viewed as a free-standing agreement, but rather as ancillary to the Facility Agreement: Etihad was agreeing in the Facility Agreement to provide €350 million, and in the Comfort Letter was saying that "if you do need more, our present intention is to ensure that you get it". This was not, therefore, a case of a completely separate agreement. If the Comfort Letter was not a firm agreement, then it was appropriate to view it as attached or linked to something else; i.e. the Facility Agreement.
  55. Air Berlin's submissions

  56. Air Berlin submitted that the burden was on Etihad to demonstrate that it had the better of the argument as to the applicability of the jurisdiction clause relied upon. The application of this standard of proof must take into account the European Union law requirement that an exclusive jurisdiction clause under Article 25 must be "clearly and precisely" demonstrated.
  57. Air Berlin submitted that the general presumption, stated authoritatively in Fiona Trust, that the parties intended all disputes arising out of a particular relationship to be governed by a jurisdiction clause, only applied where there was a single contract with a jurisdiction clause. They argued that the presumption could not apply where there "is more than one written contract". In his oral submissions, Mr. Joseph described this as an overarching principle. He referred to the recent judgment of Hamblen LJ in Rifiuti, at [70] – [71], that a generally worded jurisdiction clause would most obviously capture claims made under that contract, not some other contract. It would be unusual for the parties to intend a jurisdiction clause in one contract to apply to disputes within the sphere of influence of another contract.
  58. Mr. Joseph also submitted, relying upon Lord Hoffmann in Fiona Trust and Longmore LJ in Petromena, that the words "arising out of or in connection with" the Facility Agreement were governed or limited by the legal relationship in that agreement.
  59. Air Berlin drew attention to the decision of Field J. in Choil in support of the proposition that even where one of the contracts does not contain an express jurisdiction clause, it will often nevertheless be intended to have its own sphere of jurisdiction. In that case, the alleged joint venture agreement had no jurisdiction agreement, and the judge decided that jurisdiction clauses in subsequent physical sales contracts were inapplicable to the joint venture agreement. More recently, in Petromena, the Court of Appeal held that jurisdiction clauses in bonds, which incorporated a Norwegian jurisdiction clause in a loan agreement, were inapplicable to disputes under an advisor/ advisee relationship which came into existence some years later and which had no applicable jurisdiction agreement.
  60. Applying these principles to the present case, the parties did not intend disputes under the Comfort Letter to fall within the jurisdiction clause in the Facility Agreement.
  61. a) The jurisdiction clause in the Facility Agreement did not refer to disputes under the Comfort Letter, nor indeed to any of the other agreements concluded in April 2017. The language of the Facility Agreement was directed entirely at the specific relationship of lender-borrower, and the specific obligations in that relationship. Neither the Facility Agreement itself, nor the jurisdiction clause within it, are drafted as a framework or umbrella agreement at the centre of a single transaction.
    b) The authors of the Facility Agreement did define a category of "Transaction Documents", but this did not encompass the Comfort Letter.
    c) The Comfort Letter is nowhere referred to in the Facility Agreement. This showed that the two agreements were not intrinsically linked and that the parties deliberately did not take up the opportunity expressly to link them.
  62. The types of dispute likely to arise under the Facility Agreement and Comfort Letter are of a quite separate character. The Facility Agreement created a relationship of lender/ borrower. The Comfort Letter does not do this. Rather, there is an overarching relationship of "patron/ protιgι". The time-frames envisaged by the documents are different: the Facility Agreement contemplated drawdown between April and July 2017, whereas the Comfort Letter looked ahead "for the foreseeable future and in any event for 18 months from the date of this letter".
  63. Air Berlin submitted that there was good reason why the parties would likely have intended the ordinary Brussels Recast jurisdiction rules, rather than English conferred or prorogued jurisdiction, to apply to Comfort Letter disputes. This was because German courts would have jurisdiction over any actual or contended insolvency of Air Berlin. If a creditor petitioned for insolvency, they would do so in Germany and the issue would then arise as to whether Air Berlin was insolvent in view of the existence of the Comfort Letter. It was improbable that the parties intended that there would be separate proceedings between Etihad and Air Berlin in England, in parallel with the German insolvency dispute. Moreover, claims for damages for breach of the Comfort Letter would relate closely to the outcome of the insolvency: the present German claim seeks damages by reference to the amounts needed to end insolvency and resume business. It made sense (as in Choil) for the ordinary jurisdictional rules under the Brussels Recast to apply to any dispute under the Comfort Letter.
  64. Air Berlin described Etihad's "support package" argument as involving a relatively arid debate. Etihad's case was "over-egged" and did not take Etihad any further.
  65. a) Etihad needed to show that the jurisdiction clause in the Facility Agreement covered claims under the Comfort Letter. There was no case that a jurisdiction agreement was to be implied into the Comfort Letter. Evidence as to past jurisdiction agreements were of little weight.
    b) There were a number of agreements reached on 28 April 2017. But these agreements created different and distinct legal relationships, with separate terms, provisions and characteristics. There were six documents which comprised the "support package". In addition to the Comfort Letter there was: the Facility Agreement; the Non-Exercise Put Option Letter Agreement; a Ratification of the Total Return Swap; and amendments to guarantees in favour of National Bank of Abu Dhabi and Abu Dhabi Commercial Bank. They were not, however, part of a single transaction, they were not inter-dependent, and did not cross-refer to each other. There were a variety of jurisdiction clauses within the "package". The Facility Agreement jurisdiction clause cannot have been intended to govern disputes under all six agreements that were reached. The various documents created separate relationships and had different purposes. The various agreements created what Mr. Joseph described as "different silos of obligation". The support package was not a unified package: it created very different legal relationships, different legal animals and different strands of activity. Each strand of support had its own arrangements.
  66. It was incorrect to suggest that the Facility Agreement was at the core of the support package. It was the Comfort Letter which created an over-arching forward-looking relationship that would lead to specific and different kinds of support transactions. This contrasted with the Facility Agreement, which was a single prior agreed transaction. Mr. Joseph emphasised that the Comfort Letter was not simply addressing Air Berlin's cash obligations and potential shortfall of cash in the future: it was also addressing all the non-cash obligations that Air Berlin might face in the period to December 2018.
  67. Furthermore, the relevant legal question is which contract or particular legal relationship is at the commercial centre of the disputes in the German claim. The Facility Agreement is not at the core or commercial centre of the German claim. Nor did the relationship of lender/ borrower encompass the other relationship such as shareholding patron/ protιgι or bondholder/ bond issuer or guarantor.
  68. Mr. Joseph also drew attention to the fact that the Facility Agreement had been approved initially in December 2016, and formally approved in February 2017, albeit that it was not finally negotiated and concluded until April 2017. By contrast, the Comfort Letter was first discussed in April 2017. Accordingly, even the "same day" argument broke down and this reinforced the fact that different aspects were being addressed in different instruments with different legal relationships and different parties.
  69. Discussion

  70. It is now clearly established that the standard of proof to be applied in determining whether the English court has jurisdiction under Article 25 of the Brussels Recast is that of good arguable case, and that the burden is on the claimant to show that it has the better of the argument on the materials available: see Airbus paragraphs [49] to [51]. It follows that there is no more stringent test arising from the European case-law which refers to the need for an exclusive jurisdiction agreement under Article 25 to be "clearly and precisely demonstrated". If the good arguable case test is properly applied, then it meets this requirement: see Bols Distilleries BV v Superior Yacht Services [2007] UKPC 45, paragraph [28]. This is not in my view affected by the fact that "good arguable case" is now understood to require the claimant to show that it has the better of the argument, rather than a "much better" argument as discussed in Bols.
  71. It is for the national court, in this case the English court, to interpret the clause conferring jurisdiction invoked before it in order to determine which disputes fall within its scope: Hydrogen Peroxide SA v Azko Nobel NV C-352/13 at [67], applying Powell Duffryn plc v Petereit C- 214/89 at [36]. The jurisdiction clause relied upon is contained in the Facility Agreement which is expressly governed by English law. Clause 32 of that agreement provides:
  72. "This Agreement and all non-contractual obligations arising from or connected with it are governed by English law".
  73. Accordingly, the question of whether, as a matter of contractual interpretation, the clause conferring jurisdiction extends to claims in respect of the Comfort Letter and the related claims advanced in the German proceedings is to be determined by reference to English law. The relevant jurisdiction clause is Clause 33.1 which is set out above.
  74. The general approach to construing jurisdiction clauses under English law is that a broad, purposive and commercially minded approach is to be followed: see, for example, BNP Paribas S.A. v Trattamento Rifiuti Metropolitan S.p.a. [2019] EWHC Civ 868 per Hamblen LJ at [68(2)].
  75. The present clause is extremely wide. It applies to "any dispute arising out of or in connection with this Agreement"; words which have been described by Males LJ in Airbus at [70] as an "all-encompassing expression". It also extends to "a dispute relating to non-contractual obligations arising from or in connection with this Agreement". Just as in Airbus, where similar words appeared, the natural meaning of the clause is that it is intended to be comprehensive.
  76. In the well-known decision in Fiona Trust, the House of Lords held that when construing an arbitration clause, the court "should start from the assumption that the parties, as rational businessmen, are likely to have intended any dispute arising out of the relationship into which they have entered or purported to enter to be decided by the same tribunal": Fiona Trust and Holding Corp v Privalov [2007] UKHL 40 per Lord Hoffmann at [13]. The same starting point applies to a jurisdiction clause. The parties' submissions disagreed, however, as to whether this starting point or assumption has any relevance to the present case, where the issue is whether the jurisdiction clause in the Facility Agreement extends to disputes arising in connection with a separate document (and which Air Berlin would characterise as a separate agreement), namely the Comfort Letter.
  77. I consider that the same starting point should apply, although (for reasons explained below) I would reach the same conclusion on the basis of the agreements and documents even if that starting point were not adopted.
  78. The general principles applicable to the construction of jurisdiction clauses, both in cases where there is a single agreement and where the parties' arrangements are set out in multiple related agreements, are summarised in the judgment of Thomas LJ in Sebastian Holdings, in particular at paragraphs [39] – [42].
  79. 39.  It is clear that in construing a jurisdiction clause, a broad and purposive construction must be followed: Donohue v Armco [2001] UKHL 64 ; Fiona Trust Holding Corporation v Privalov [2007] EWCA Civ 20 affirmed in sub nom Premium Nafta Products v Fili Shipping [2007] UKHL 40 where Lord Hoffmann observed at paragraph 7;
    "If, as appears to be generally accepted, there is no rational basis upon which businessmen would be likely to wish to have questions of the validity or enforceability of the contract decided by one tribunal and questions about its performance decided by another, one would need to find very clear language before deciding that they must have had such an intention."
    40.  The Supreme Court emphasised in Re Sigma Finance Corporation [2009] UKSC 2 the need, when looking at a complex series of agreements, to construe an agreement which was part of a series of agreements by taking into account the overall scheme of the agreements and reading sentences and phrases in the context of that overall scheme.
    41.  It is generally to be assumed on these principles that just as parties to a single agreement do not intend as rational businessmen that disputes under the same agreement be determined by different tribunals, parties to an arrangement between them set out in multiple related agreements do not generally intend a dispute to be litigated in two different tribunals.
    42.  However, where there are multiple related agreements, the task of the court in determining whether a dispute falls within the jurisdiction clauses of one or more related agreements, depends upon the intention of the parties as revealed by the agreements against these general principles: see Lawrence Collins LJ in Satyam Computer Services Ltd v Upaid Systems Ltd [2008] EWCA Civ 487 at [93], [2008] 2 All ER (Comm) 465 and the UBS case [2010] 1 All ER (Comm) 727 at [83].
  80. Etihad submitted that paragraph [42] supports the proposition that the Fiona Trust starting assumption is applicable not only to single agreement cases, but where the multiple agreements are part of a package or arrangement between them. I do not think that paragraph [42] goes quite that far: the point made in that paragraph is that parties do not intend a particular dispute to be litigated before two different tribunals.
  81. However, support for the potential applicability of the Fiona Trust assumption in the context of multiple agreements can be found in the judgment of the Court of Appeal in Am Trust Europe Ltd v Trust Risk Group. Having quoted from Fiona Trust, Beatson LJ said at paragraphs [45] – [46]:
  82. 45.  That case concerned the scope of a single arbitration clause. This case concerns an overall agreement package which contains two express choice of law and jurisdiction clauses, one of English law and jurisdiction, the other of Italian law and arbitration. Mr Samek submitted that, although the present case is not about the scope of a single arbitration clause, the Fiona Trust "one-stop"/"one jurisdiction" presumption remains a useful starting point. In principle, and subject to the qualification in the next paragraph, I agree. As Lord Collins stated in UBS AG v HSH Nordbank AG [2009] EWCA Civ 585 , reported at [2009] 2 Lloyd's Rep 272 at [84], where the agreements are all connected and part of one package, "sensible businesspeople would not have intended that a dispute of this kind would have been within the scope of two inconsistent jurisdiction agreements".
    46.  Where the overall contractual arrangements contain two or more differently expressed choices of jurisdiction and/or law in respect of different agreements, however, the position differs in that one does not approach the construction of those arrangements with a presumption. So, the 14th edition of Dicey, Morris and Collins on the Conflict of Laws stated:
    "the decision in Fiona Trust has limited application to the questions which arise where parties are bound by several contracts which contain jurisdiction agreements for different countries. There is no presumption that a jurisdiction (or arbitration) agreement in contract A, even if expressed in wide language, was intended to capture disputes in contract B; the question is entirely one of construction… (§12–094)
  83. Accordingly, the Fiona Trust starting point will potentially apply – for the reasons given by Beatson LJ – where there is an "overall agreement package", but not if the arrangements contain two or more differently expressed choices of jurisdiction and/or law in respect of different agreements. The potential importance of considering whether related contracts are part of one package is discussed by Beatson LJ at paragraph [49]. Where the contracts are not part of one package, it may be easier to conclude that the parties chose to have different jurisdictions to deal with different aspects of the relationship.
  84. The decision of Sir William Blackburne in Cinnamon illustrates the potential applicability of the Fiona Trust approach to two agreements, one of which contained a jurisdiction clause and the other (an agreement contained in a Representation Letter) which did not. The judge considered that the Representation Letter was "very closely related" to the agreement (for the purchase of certain shares) which contained the jurisdiction clause. The clause was then given a liberal construction so as to apply to a dispute arising under the Representation Letter.
  85. Similarly, the Petromena case concerned an argument that a Norwegian jurisdiction clause incorporated into certain bonds, via the terms of a loan agreement, should be applied to a subsequent advisory agreement. The latter contained no jurisdiction clause. The argument failed, with slightly different reasoning being given by the three members of the Court of Appeal. But none of the judgments indicate that there is any obstacle in principle to a jurisdiction clause in one agreement being construed under English law to apply to a dispute arising from another agreement. Indeed, at paragraph [106], Floyd LJ said (having cited Fiona Trust):
  86. "It does not follow that a claim for a breach of a separate contract can never arise in connection with the contract in which the clause is contained".

    Similarly, Ryder LJ accepted in principle (at [100]) that "a dispute arising out the advisory relationship may also be connected with the relationship arising out of the loan agreement but not on the alleged facts of this case". He went on to say [101] that the claims did not demonstrate "a sufficient connection with the loan agreement relationship."

  87. I do not consider that any different approach is indicated by the decision in Rifiuti. As Hamblen LJ said in the very first paragraph of his judgment, that case concerned (as did many of the cases referred to in argument) "apparently competing jurisdiction clauses". At paragraph [68] of his judgment, he set out a number of principles which apply where the parties' overall contractual arrangements contain two competing jurisdiction clauses. Air Berlin placed reliance on his statements in paragraphs [69] – [71] to the effect that a generally worded jurisdiction clause would most obviously capture claims under the contract containing that clause rather than "some other contract", and that even wide words "would not naturally extend to claims under a different contract". However, these passages must be seen in the context of a case where there were competing jurisdiction clauses. This is not the situation in the present case. Air Berlin does not here point to any competing jurisdiction clause in support of a contention that the parties' dispute concerning the Comfort Letter falls under that clause rather than the clause in the Facility Agreement. Rather, Air Berlin contends that there is no jurisdiction clause anywhere which is applicable to claims under the Comfort Letter, and that accordingly the normal rules of jurisdiction under the Brussels Recast apply so that Air Berlin must in principle be sued in Germany.
  88. Drawing these threads together, I consider that the position is as follows:
  89. a) There is no reason in principle why the jurisdiction clause in the Facility Agreement should not extend to disputes arising in relation to the Comfort Letter.
    b) The Fiona Trust starting assumption is potentially applicable if it can properly be said (applying the good arguable case standard) that the Comfort Letter was part of a package of agreements which contained no competing jurisdiction clause.
    c) Ultimately, the question is whether (again applying the good arguable case standard), looking at the overall scheme of the agreements, the parties' intention, as revealed by the agreements reached between them, was that a dispute under the Comfort Letter falls within the jurisdiction clause in the Facility Agreement.
    d) In ascertaining the parties' intention, it is relevant to consider the closeness of the connection between the Comfort Letter and the Facility Agreement.
  90. In considering how these principles apply in the present context, I consider that the following matters are significant.
  91. First, the jurisdiction clause in the Facility Agreement is very wide, and includes both contractual and non-contractual obligations arising out of or in connection with that agreement.
  92. Second, I consider that it is beyond serious argument (and certainly sufficient to meet the good arguable case standard) that both the Comfort Letter and the Facility Agreement were part of an overall support package which was provided by Etihad to Air Berlin in April 2017.
  93. Thus, the expression "Support package – Commercial arrangement" was used by both parties in an e-mail exchange on 3 and 4 April 2017 and this is consistent with the evidence of Mr. Henning zur Hausen (the General Counsel and Company Secretary of the Etihad Aviation Group) that the expressions "support package" or "package" were used and accepted by both parties in the course of their discussions in April 2017. An email sent by Mr. Courtelis of Air Berlin on 13 April 2017 similarly refers to the "funding package".
  94. That the description "support package" is both realistic and accurate is confirmed by the approach taken by Air Berlin in the German proceedings. Their claim repeatedly describes it in this way, often using that very phrase:
  95. [59] "In the following weeks, the parties discussed the support package including the modalities of the commitment, and repeatedly revised the documents until they – apparently – agreed on the final drafts on 19 April 2017")
    [181] "The purpose of the comfort letter, together with the other measures initiated, was to ensure the positive going concern prognosis of Air Berlin and prevent a possible over-indebtedness. For this purpose … the Defendant provided Air Berlin with the requested 'support package', part of which was the comfort letter."
    [221] "The continuation of the company depended on the granting of the support package by the Defendant (see marg. No 53 above)."
    (Paragraph 53 of the claim, referred to here, set out the emails of 3-4 April 2017 headed "Support Package – Commercial Arrangement).
  96. Third, the Comfort Letter was not only part of the support package which contained the Facility Agreement, but it was (or at least there is a good arguable case that it was) very closely connected with the Facility Agreement. The factual background to both the Comfort Letter, and the Facility Agreement, was in summary as follows.
  97. As described in Section A, the position in late March/early April 2017 was that the parties were discussing the package of financial support which Air Berlin needed in order to address "going concern" issues raised in the context of KPMG's audit. One matter addressed in the correspondence was the amount of a loan facility which Etihad had previously (in December 2016) agreed to provide, subject to formal approval, in the sum of €350 million. On 31 March, Mr. Courtelis of Air Berlin emailed Mr. Rigney, the CFO of Etihad, attaching a copy of a document headed "KPMG Going Concern requirement". This identified KPMG's requirements both for cash and for non-cash elements. The total cash element was identified as €558 million. On 3 April, Mr. Sanghavi of Etihad emailed Mr. Courtelis in an email with the subject-line "Support package – Commercial arrangement". The email asked for his comments "on the attached commercial understanding of the support package to [Air Berlin]". The email described the elements of the support package which Etihad understood to have been requested. This included a €560 million "Facility"; i.e. essentially the figure which KPMG had identified.
  98. On 5 April, Mr. Courtelis emailed Etihad indicating that an additional €50 million was required over and above the additional €208 million (which had taken the cash required from €350 million to €558 million). This was "simply because KPMG are not budging on their level of sensitivity application to the business plan". Accordingly, the request at this stage was for Etihad to provide a facility of €608 million rather than the €350 million which had previously been indicated. At this stage, the Comfort Letter had not yet been identified as a possible ingredient to the support package.
  99. The first draft of the Facility Agreement was produced in early March 2017 and it was then the subject of negotiation (although Air Berlin's evidence was that "heavy" negotiation was not required) so that most of the key terms had been agreed by 18 April 2017.
  100. Etihad's case in its skeleton argument was that by 26 April 2017 Etihad had informed Air Berlin that it would not agree to increase the amount of the loan facility from €350 million to €610 million, and that the Comfort Letter – which came to be discussed and negotiated on 26 April and over the next two days – was a response "in particular" to that refusal. I was not referred to any documentary evidence showing exactly when Etihad had refused to increase the amount of the Facility Agreement. However, there did not appear to be any substantial dispute as to Etihad's refusal to increase the loan facility, or that the Comfort Letter was – at least in significant part – a response to that problem.
  101. Thus, Mr. zur Hausen's evidence was that the Comfort Letter came to be discussed because Etihad was "unwilling to commit to a loan facility in excess of the €350m Facility Agreement that it had agreed in principle to provide or to commit to refinance the €225m Bond in the absence of other refinancing options". Similarly, in its claim in Germany, Air Berlin's case is that the Comfort Letter came about as a direct response to Etihad's refusal to increase the amount of the Facility Agreement:
  102. "[67] Over the next few days [which appears, from its context, to be a reference to the days after 20/21 April], [Etihad] and KPMG continued to discuss the financing commitment. In the course of these discussions, [Etihad] suddenly announced that it would only approve a loan for EUR 350 million instead of the EUR 610 million loan that was discussed initially. KPMG subsequently informed [Etihad] that this would require a comprehensive commitment to provide future financial support to Air Berlin in order to attain a positive going concern prognosis for Air Berlin."
  103. To the same effect in a section of the claim headed "Genesis", Air Berlin says:
  104. "[174] Since the Defendant instead of the loan initially discussed for a total of EUR 610 million only wanted to grant a loan of EUR 350 million, KPMG demanded instead a more comprehensive commitment in the comfort letter. Then [Etihad] proposed a more comprehensively formulated comfort letter on 26 April 2018"
    "[175] Air Berlin was able to continue to assume, even without an explicit mention of the bond and despite smaller changes in the text of the contract, the binding nature of the commitment. This is particularly the case, because the extension of the comfort letter dated 26 April 2018 (see marg. No 174 above) was conceived as a replacement for the reduction of the facility agreement from EUR 610 million to EUR 350 million (see marg. No. 67 above). The comfort letter could represent an adequate replacement for this without a doubt binding agreement only in the form of a binding declaration".
  105. Mr. Joseph made it clear that he was not disputing the facts and history which were set out in paragraphs [174] – [175] of the German claim. However, he emphasised that paragraph [175] was not the complete statement of the genesis. Apart from the cash that Air Berlin required, and which was to be provided by the Facility Agreement, there were other "non-cash" needs which needed to be met, and the Comfort Letter and the other agreements concluded in April 2017 were aimed at addressing the position as a whole.
  106. I accept Mr. Joseph's point that the terms of the Comfort Letter are not confined to the provision in the future (or the intention to provide in the future) of cash support, and that further non-cash support was also contemplated. However, it is clear from the statements made by Air Berlin in the German proceedings that (consistent with Mr. zur Hausen's evidence) the Facility Agreement, and Etihad's unwillingness to commit to increase the facility from €350 million to €610 million, were a very significant part of the genesis of the Comfort Letter. The commercial background is, therefore, that they were very closely connected.
  107. Fourth, I consider that, on the present material, Etihad has a good arguable case that the Comfort Letter was no more than a statement of present intention, and did not create legally binding obligations. This is relevant for a number of reasons. If there is a good arguable case that the Comfort Letter did not create contractual obligations, then (contrary to Mr. Joseph's submissions) it would not be appropriate to approach the case on the basis that it created any contractual relationship, including the relationship of patron/protιgι relied upon by Air Berlin under German law. Furthermore, if the Comfort Letter is not properly to be viewed as a separate agreement, then this provides support for Mr. Dicker's proposition that it should be viewed as ancillary or linked to something else. The most obvious agreement to which it was ancillary is the Facility Agreement. As Mr. Dicker submitted, against the commercial genesis described above, Etihad was agreeing in the Facility Agreement to provide €350 million, and in the Comfort Letter was saying that "if you do need more, our present intention is to ensure that you get it".
  108. The reason that there is a good arguable case that the Comfort Letter is non-binding is, in essence, because there is a good arguable case that English law is the law applicable to any rights and obligations created by that letter. The decision in F R Lurssen Werft GmbH & Co. KG v Halle [2010] EWCA Civ 587, and the citations within that decision at [9] – [14], indicate that an express choice of law in related transactions between the same parties may impel the court to the conclusion that a real choice of law has been made. In the present case, English law was chosen by the parties not only in the Facility Agreement but (as further discussed below) in the other agreements between Etihad and Air Berlin which formed part of the package of contracts concluded in April 2017. It was also, historically, the law chosen by those parties to govern the most significant contracts between them. Furthermore, within the Facility Agreement itself, the choice of law was expressed in very wide terms, extending to "all non-contractual obligations arising from or connected with" that agreement.
  109. Mr. Joseph submitted that I could and should not decide the issue of applicable law now, indicating that there were substantial arguments which would be raised in due course. I agree that I should not decide the issue now, and I do not do so. It suffices for present purposes that there is a good arguable case that English law is the applicable law of the Comfort Letter, and if so that the Comfort Letter does not create contractually binding obligations.
  110. Fifth, this is not a case where Air Berlin contends that the Comfort Letter, or the dispute thereunder, is more closely related to one of the other agreements (i.e. other than the Facility Agreement) which were concluded in April 2017 as part of the support package, and that therefore some other jurisdiction clause is applicable. Such an argument would have been of no real assistance, since the agreements concluded at that time, and to which Etihad and Air Berlin were party, were all governed by English law and English jurisdiction clauses.
  111. There was substantial agreement as to which contracts were concluded in April 2017. The material contracts, in my view, are those to which Air Berlin and Etihad were party, rather than those between Etihad and third parties (i.e. between Etihad and HSBC or between Etihad and the Abu Dhabi banks). Apart from the Facility Agreement and the Comfort Letter, there were three agreements to which Etihad and Air Berlin were parties.
  112. There were two "Reimbursement Deeds" dated 30 April 2017. These provided for a counter-indemnity from Air Berlin to Etihad in respect of guarantees which had been issued to the two Abu Dhabi banks. These two agreements contained English law and jurisdiction clauses which were in materially identical terms as that in the Facility Agreement.
  113. In addition, there was also the Put Option letter Agreement which was headed:
  114. "Non exercise of put options regarding the EUR 125 million 8.5% Guaranteed Convertible Bonds 2017-2019 issued by Air Berlin Finance B.V. and guaranteed by Air Berlin PLC due 6 March 2019 with a bondholder put option on 29 December 2017."

    The letter contained confirmation and undertakings by Etihad that it would not exercise any put option rights that it might have in respect of certain bonds which had been issued in February 2017, and that it would continue to hold at least EUR 40 million in bonds. There was also an undertaking to exercise certain rights under an agreement between Etihad and HSBC so as to ensure that HSBC would not exercise its put option rights under that bond issue. The effect of these undertakings was to protect Air Berlin BV and Air Berlin from having to make payment on redemption on the bonds.

  115. The Put Option Letter Agreement contained an express provision, materially identical to the governing law provisions in the other agreements, that:
  116. "This Letter and any non-contractual obligations arising out of or in connection with it shall be governed by and construed in accordance with English law".
  117. This agreement did not contain an express English jurisdiction clause. However, this is of no significance since the bonds themselves, to which this letter was ancillary, contained English jurisdiction clauses which were similar in effect to those in the other agreements.
  118. Accordingly, in so far as the support package contained agreements between Etihad and Air Berlin, such agreements were governed by English law and subject to English jurisdiction. When considering the intention of the parties, I consider it reasonable in these circumstances to pose the question identified by Lord Hope in Fiona Trust at [28]:
  119. "If the parties have confidence in their chosen jurisdiction for one purpose, why should they not have confidence in it for the other?"
  120. There is nothing in the agreements between Etihad and third parties which casts any doubt on this conclusion. The Total Return Swap between Etihad and HSBC was subject to English law and English jurisdiction. The NBAD Guarantee Amendment was also subject to English law, and contained an asymmetric English jurisdiction agreement in favour of NBAD. The ADCB Guarantee Amendment was also governed by English law, and was the only agreement in the package without an English jurisdiction clause.
  121. Sixth, I consider that it is a powerful point, in favour of Etihad's case as to the parties' intentions, that it was foreseeable that the resolution of a dispute under the Facility Agreement might require the court to determine the effect of the Comfort Letter and vice versa. This echoes another point made by Lord Hope in Fiona Trust at [28]:
  122. "… one should be slow to attribute to reasonable parties an intention that there should in any foreseeable eventuality be two sets of proceedings. If the parties have confidence in their chosen jurisdiction for one purpose, why should they not have confidence in it for the other? Why, having chosen their jurisdiction for one purpose, should they leave the question which court is to have jurisdiction for the other purpose unspoken, with all the risks that this may give rise to? For them, everything is to be gained by avoiding litigation in two different jurisdictions."
  123. Mr. Dicker illustrated the interconnection by considering what happened on 9 August 2017 when Air Berlin made a further draw-down request under the Facility Agreement. Etihad refused to allow Air Berlin to comply with that request because it was not satisfied that the relevant pre-conditions under clause 4.2 of the Facility Agreement were met, in particular that no Default existed under clause 19.6.2 (Insolvency) or clause 19.13.2 (Other events of Default). Clause 19.6.2 is concerned with whether there was a positive going concern prognosis for Air Berlin. Clause 19.13.2 is concerned with whether any event or circumstance had occurred which had a Material Adverse Effect.
  124. Whether or not there had been a Default and thus whether Etihad was obliged to honour Air Berlin's draw-down request on 9 August 2017, depended, in part, on whether the Comfort Letter contained a binding contractual promise or not. On Air Berlin's case, it was perfectly possible, for example, that Air Berlin could therefore have responded to Etihad in early August 2017 by contending that, given that the Comfort Letter contained a binding promise, there was no Default, and that it was therefore entitled to have its further draw-down request under the Facility Agreement met and that Etihad was in breach of the Facility Agreement for failing to honour that request.
  125. Another illustration of a similar point concerned the situation where a dispute arose in relation to Clause 16 of the Facility Agreement. Clause 16.2.1 required Air Berlin to provide Etihad with financial statements which were certified by a director as giving a true and fair view of Air Berlin's financial position. If statements were delivered which treated the Comfort Letter as containing contractual commitments, a dispute could well arise between Air Berlin and Etihad as to whether they were true and fair. This would require a determination of the effect of the Comfort Letter.
  126. There could also be a reverse situation, where a dispute as to the Comfort Letter gave rise to the need to consider the Facility Agreement. A situation could arise where a demand was made under the Comfort Letter, but Etihad took the point that there was no entitlement to support because Air Berlin should drawdown under the Facility Agreement. Issues could then arise as to whether or not the Facility Agreement was still available to be utilised.
  127. Mr. Dicker submitted, and I agree, that the parties cannot sensibly be taken to have intended that the effect of the Comfort Letter might fall to be determined both by the English court, in connection with a dispute about the validity of a draw-down request under the Facility Agreement, and also by the German court in relation to the effect of the Comfort Letter itself, as that could have led to mutually inconsistent decisions. Etihad and Air Berlin, as rational commercial parties, must, in these circumstances, be taken to have intended that any dispute between them arising out of the relationship into which they had entered would be decided by the same court.
  128. I consider that these illustrations of potential disputes directly between Etihad and Air Berlin are of greater significance, in determining the intention of the parties, than the possibility that the effect of the Comfort Letter would fall to be considered in the context of a claim by a creditor in insolvency proceedings in Berlin and possible defences by Air Berlin to that claim. The authorities indicate that the question is where the parties intended their disputes, rather than disputes between one or other of them and a third party to be resolved, and whether they intended their disputes to be resolved in multiple jurisdictions. These illustrations also show that the close connection between the Comfort Letter and the Facility Agreement is not confined to their historical genesis.
  129. I consider that these matters – (i) the width of the jurisdiction clause in the Facility Agreement, (ii) the fact that the Comfort Letter was part of the overall support package where all relevant agreements between Etihad and Air Berlin were governed by English law with English jurisdiction clauses, (iii) the close connection between the Comfort Letter and the Facility Agreement in terms of the genesis of the Comfort Letter, (iv) Etihad's good arguable case that the Comfort Letter did not create contractually binding obligations and was ancillary to the Facility Agreement, (v) the absence of any competing jurisdiction clause in any of the agreements within the support package, and the existence of English law and jurisdiction clauses in the relevant agreements as part of that package, and (vi) the reasonable foreseeability of disputes which required consideration of the Comfort Letter in conjunction with the Facility Agreement – all lead to the conclusion that the parties intended disputes arising in relation to the Comfort Letter to fall within the jurisdiction clause of the Facility Agreement. I reach that conclusion whether or not the Fiona Trust starting point is applied. Since I consider that it should, in the present circumstances, be applied, that reinforces the conclusion which I have reached.
  130. I did not consider that any of the arguments advanced by Mr. Joseph, in support of Air Berlin's case as to the inapplicability of the jurisdiction clause, were of any force when set against the matters which I consider important. I have to a large extent addressed those points in the course of the above discussion, but I add the following.
  131. I do not consider that the decision in Choil is of any great assistance. That case concerned an attempt to apply jurisdiction clauses in physical sale contracts to a joint venture agreement which had been concluded prior to those contracts coming into existence. Whilst it is not impossible for a jurisdiction agreement to have, on its true construction, such retrospective effect, a party seeking to rely upon a subsequently agreed jurisdiction agreement, in a separate contract, is likely to face an uphill struggle: see e.g. Satyam. One reason is that the earlier contract had an existence of its own, and hence an applicable law, prior to the conclusion of the subsequent agreements. If there was no jurisdiction agreement at the time it was concluded, then it may be difficult to conclude that it is to be found in a subsequent agreement, particularly if (as in Choil) the disputes arising under the later agreement are likely to have a very different character to disputes arising under the earlier agreement. In any event, Choil was not a case involving a package of agreements concluded at the same time, and with the features (as described above) which are present here.
  132. I do not attach significance to the absence of any cross-reference to the Comfort Letter within the Facility Agreement, or indeed the other agreements concluded in April 2017. This is explicable because the Comfort Letter arose at a late stage in the process, but more importantly because there is a good arguable case that the Comfort Letter was not intended to create binding contractual rights and obligations. There would therefore be no obvious need for any of the contractual agreements to refer to a non-binding Comfort Letter.
  133. Nor do I attach significance to the fact that the definition of "Transaction Documents" did not encompass the Comfort Letter. Again, there was no obvious reason for the Facility Agreement to refer to a non-binding Comfort Letter, and the definition of "Transaction Documents" was principally concerned with documents "executed by the Borrower".
  134. It is true that the types of dispute likely to arise under the Facility Agreement may be rather different to a dispute under the Comfort Letter. But for reasons already given, it is foreseeable that a dispute under one may require interpretation of the other. Moreover, once it is recognised that in principle a jurisdiction clause in one contract can extend to a dispute under another contract (and a fortiori a dispute under a non-binding agreement), there is no reason to require a similarity between the types of dispute that might arise both documents. Ultimately, if (as here) the documents are part of the same package, and are closely linked, then a widely worded jurisdiction clause in one contract will cover disputes under the other document even if the nature of those disputes is different.
  135. I do not accept that Etihad's "support package" argument is either arid or over-egged. For reasons already given, I consider that it is significant that the Comfort Letter was contained within a package of agreements, and that the relevant agreements between Etihad and Air Berlin were governed by English law. It is true that these agreements covered different aspects of the support provided. But that does not mean that they were not connected (and in the case of the Facility Agreement and Comfort Letter closely connected), or that they should be viewed as being in individual silos. Just as in the Cinnamon case, it is artificial here to view all the agreements (in particular the Facility Agreement) as wholly free-standing documents, operating independently of one another, and to approach the construction of one as if the other did not exist.
  136. I therefore conclude that, interpreting the jurisdiction agreement in the Facility Agreement as a matter of English law, there is a good arguable case that (i) the jurisdiction clause in the Facility Agreement is applicable to the Comfort Letter and any non-contractual claim in connection therewith, and (ii) the claim commenced by Air Berlin in Germany falls within the scope of that clause.
  137. D: Article 25 – "particular legal relationship"

    The parties' arguments

  138. It was common ground that the requirement in Article 25 for the dispute to arise "in connection with a particular legal relationship" was not determined by a conclusion that, as a matter of English law, the claims made in Germany fell within the scope of the jurisdiction agreement in the Facility Agreement. It is necessary for Etihad to demonstrate a good arguable case that this requirement of Article 25 was fulfilled.
  139. The arguments of the parties to some extent overlapped with those which I have already considered, particularly as to the strength or otherwise of the connection between the Facility Agreement and the Comfort Letter. However, the focus of the argument was rather different.
  140. On behalf of Air Berlin, Mr. Joseph submitted that the first task was to identify the "particular legal relationship". In the present case, the particular legal relationship in respect of which the parties concluded a jurisdiction agreement was that of lender/ borrower under the terms of the Facility Agreement. This agreement contained clauses referable to that relationship, for example a fixed funding amount and a limited utilisation period. There was no inextricable linkage between the Comfort Letter and the Facility Agreement.
  141. In order for Etihad to invoke Article 25, the claims in Germany would have to arise from the particular relationship of lender/ borrower. They did not do so, but rather arose under the "entirely separate, free-standing relationship generated by the Comfort Letter". This relationship had none of the features of the Facility Agreement. There was no lender/ borrower relationship, but rather an "overarching relationship of patron/protιgι under which specific instances of support would be provided"; i.e. by individual further (not in advance specified) contracts of guarantee, bond purchase, debt assumption, loan facility and so forth. The amount to be provided was not a fixed amount, but rather an amount based on need; the result being to avoid the failure of Air Berlin to meet its financial obligations. Unlike a loan, there was no provision for interest.
  142. Different disputes were likely to arise under each agreement. A dispute under the Facility Agreement would likely concern whether Etihad was obliged to advance funds. A dispute under the Comfort Letter would likely be as to the nature and extent of the legal obligation created, or what financial support Etihad was obliged to provide and the consequences of non-provision during the 18-month period.
  143. In the present case, the claims made in Germany fell under the "Comfort Letter relationship", rather than the lender/ borrower relationship. The lender/borrower relationship had "nothing to do with" the substantive claims in Germany.
  144. In his oral submissions, Mr. Joseph submitted that the resolution of this issue required a very limited exercise. It was only necessary to: (i) look at the legal relationship in the agreement which contains the jurisdiction clause; (ii) characterise the relationship; and (iii) look at the German claim and ask whether it was in connection with that relationship. The task of characterising the relationship in the relevant agreement was a simple one. It was not permissible to look at background circumstances, such as the fact that Air Berlin was looking for different types of support going forward. Nor was it permissible to look at other background facts, including: Etihad's involvement as a shareholder; the course of dealings between the parties; the timing of the relevant agreements; the fact that agreements were concluded as part of a package; and the fact that (as Air Berlin alleged in the German proceedings) the Comfort Letter was a consequence of the negotiation of the Facility Agreement. None of these matters was relevant to the characterisation of the relationship: that depended exclusively on the contract or agreement which contained the jurisdiction clause. This limited exercise was appropriate bearing in mind that Article 25 had to be applied in all of the other EU states.
  145. Applying this approach, there were two relevant relationships in the present case: lender/ borrower under the Facility Agreement and a quite different relationship of patron/ protιgι.
  146. Having identified and characterised the relevant legal relationship in connection with which the jurisdiction clause was agreed, the next stage of the exercise was to compare that legal relationship with the claim that is being made in the German proceedings. The relevant question was then whether it could really be said that the claims made in Germany were in connection with the legal relationship which has been identified. Mr. Joseph relied upon paragraph [77] of the decision of the Court of Appeal in Airbus in support of the proposition that the only exercise required was comparing the legal relationship as per the Facility Agreement to the German claim. The nature of the claim made in Germany was wholly distinct from the legal relationship of lender/ borrower in the Facility Agreement. That the claim made would or might not be a good claim under English law was nothing to the point.
  147. On behalf of Etihad, Mr. Dicker submitted that the question of whether the dispute arises from a legal relationship in connection with which the jurisdiction agreement was entered into needed to be considered on the premise that, on its true construction under English law, Clause 33.1 extends to cover the present dispute.
  148. Article 25, as interpreted in Powell Duffryn, did not require identification of the relationship constituted by the agreement which contained the jurisdiction clause. Powell Duffryn required identification of the legal relationship in connection with which the jurisdiction agreement was concluded. The purpose was to prevent a party from being surprised by the referral of a particular dispute to the specified court. The task of identifying the relationship in connection with which the agreement was concluded was a question of fact.
  149. Applying these principles to the facts of the present case, Mr. Dicker submitted that there were a number of ways in which the case could be viewed. It was permissible to look at the pre-existing position of the parties: the relationship between Etihad as shareholder providing support to the company in which it held shares. A second approach was to consider the package of agreements concluded in April 2017. This was a relationship constituted by a package of agreements and the jurisdiction clause in the Facility Agreement was entered into in connection with that package of agreements and the relationship so created. The effect of that package was that Etihad was the provider of financial support and Air Berlin was the recipient of such support, which took a number of slightly different forms. Thirdly, adopting the approach of Ryder LJ in Petromena, the facts of the present case are that the Comfort Letter arose in connection with the Facility Agreement. Finally, adopting the approach that the Facility Agreement contains a lender/ borrower relationship, the Comfort Letter referred to historic loans and the intention to provide further support. Whichever approach was adopted, it led to the same conclusion.
  150. Ultimately, the task was to identify the relationship in connection with which the jurisdiction agreement was concluded, and the reason for this was to ensure that no-one is taken by surprise.
  151. Discussions

  152. The leading case on the concept of disputes arising "in connection with a particular legal relationship" is the decision of the European Court of Justice in Powell Duffryn. The Court explained that:
  153. "This requirement aims to limit the effect of an agreement conferring jurisdiction to disputes originating from the legal relationship in connection with which the agreement was concluded. It seeks to prevent a party from being surprised by the referral to a specified court of all disputes which arise in the relationships which it has with the other party and which may originate in relationships other than that in connection with which the agreement conferring jurisdiction was concluded".
  154. I consider (in agreement with Mr. Joseph) that, applying Powell Duffryn, it is important to identify the legal relationship in connection with which the agreement conferring jurisdiction was concluded, and then to ask whether the dispute has originated in a different relationship; i.e. a relationship other than that in connection with which the agreement conferring jurisdiction was concluded. These questions should be asked bearing in mind that the purpose of the relevant words in Article 25 is to avoid a party being taken by surprise by the referral of the dispute to a contractually agreed court, because the dispute had originated in a different legal relationship.
  155. I agree with Mr. Joseph that the relevant question is not simply whether a party would be taken by surprise: this is not the legal test. However, that question serves as a very useful cross-check on what I consider to be the relevant legal questions. If it is clear that a party would not be taken by surprise by the referral of the dispute, then it is very likely indeed that the dispute has not originated in a relationship other than that in connection with which the agreement was concluded. It is therefore very likely that application of the legal test, and the answer to the question whether a party would be taken by surprise, will lead to the same result.
  156. The decision in Hydrogen Peroxide SA v Akzo Nobel NV C-352/13 provides an illustration of the distinction in practice. The issue concerned the applicability of the predecessor of Article 25 to claims for damages arising out of cartel arrangements. The Defendants sought to rely upon jurisdiction and arbitration agreements in various contracts for the sale of hydrogen peroxide. The court applied the principles in Powell Duffryn, again emphasising that the purpose of the material part of Article 25 was to avoid a party being taken by surprise.
  157. The court thus drew a distinction between two types of arbitration or jurisdiction clause. Certain clauses would not qualify under Article 25:
  158. "[69] In the light of that purpose, the referring court must, in particular, regard a clause which abstractly refers to all disputes arising from contractual relationships as not extending to a dispute relating to the tortious liability that one party allegedly incurred as a result of its participation in an unlawful cartel.
    [70] Given that the undertaking which suffered the loss could not reasonably foresee such litigation at the time that it agreed to the jurisdiction clause and that that undertaking had no knowledge of the unlawful cartel at that time, such litigation cannot be regarded as stemming from a contractual relationship. Such a clause would not therefore have validly derogated from the referring court's jurisdiction."
  159. Other clauses, however, would qualify:
  160. "[71] By contrast, where a clause refers to disputes in connection with a liability incurred as a result of an infringement of competition law and designates the courts of a Member State other than the Member State of the referring court, the latter ought to decline its own jurisdiction, even where that clause entails disregarding the special rules of jurisdiction laid down in Articles 5 and/or 6 of Regulation No 44/2001."
  161. The distinction was therefore drawn on the basis of lack of surprise and foreseeability. An abstract reference to disputes would not suffice, given the lack of knowledge of the unlawful cartel at the time. But a more specific clause, referring to competition law, would suffice. If the dispute resolution clause referred specifically to competition law, then neither party could be surprised if a dispute concerning an unlawful cartel was referred to the agreed court or tribunal, even though the existence of the cartel would not have been known by one party at the time that the contract was made.
  162. In answering the questions which I have identified, I do not accept that the court should take the narrow approach suggested by Mr. Joseph, namely: disregarding the commercial background; disregarding the timing of the relevant agreements relied upon; focusing only on the language of the claim asserted in the German proceedings and the characterisation of the relationship in those proceedings, and comparing it to the contract which contains the jurisdiction clause. Ultimately, the court has to consider, in the light of the admissible evidence as a whole, whether the dispute has originated from the legal relationship in connection with which the jurisdiction agreement was concluded. I consider that this is largely a factual question. This is the way that the question was treated by Ryder LJ in Petromena at [100] – [101]. Similarly, Longmore LJ at [85] described the existence of two relationships, with the first being "earlier and different" to the second. In so doing, he was drawing upon the background facts as described in the preceding paragraphs of his judgment, in particular paragraphs [83] – [84].
  163. Accordingly, the test requires identification, by reference to the facts of the case as a whole, of the legal relationship between the parties in connection with which the jurisdiction agreement was concluded. It then requires consideration of whether the dispute originates from that legal relationship or a different one.
  164. This wider approach is supported by the decision of Sir Michael Burton in Altera. In that case, Sir Michael Burton looked at both the origin and immediate context of the two agreements in answering the "particular legal relationship" question, including that the agreements in question were "part of the same package": see paragraphs [25] – [26]. I consider that he was right to do so.
  165. It is also important to note that the relevant question is whether the dispute has arisen from the legal relationship in connection with which the jurisdiction agreement was concluded. This is not the same as asking: is the dispute a claim which arises under the terms of contract which creates the legal relationship? At times, it seemed to me that Mr. Joseph's submission – which focused on the terms on which money was to be advanced under the loan agreement – sought to assimilate the two. But it is in my view clear that a dispute can be within a jurisdiction agreement covered by Article 25 even if it does not allege a breach of the particular contract containing the jurisdiction clause. Any other conclusion would mean that non-contractual claims fall outside the scope of an Article 25 jurisdiction agreement. This cannot be right – as illustrated by the Hydrogen Peroxide case. A recent illustration of a jurisdiction clause applying to a claim in tort is the Airbus case.
  166. Nor do I accept that Airbus is authority for the proposition that, in order to identify the relevant legal relationship, or to decide whether the dispute originates from that relationship, the court can and should only look at the way in which the claim is formulated in the proceedings (here Germany) which are alleged to have been brought in breach of the jurisdiction clause. The only authority cited in support of this proposition was paragraph 77 of the judgment of the Court of Appeal (Males LJ) in that case:
  167. "It is common ground that the question whether the appellants' claim in Italy falls within the scope of the English jurisdiction clause depends on the nature of the claim brought in Italy, not on the defences which may be or have in fact been raised by Alitalia. As Thomas LJ explained in Sebastian Holdings at para 62:
    "… the question as to whether a claim falls within the jurisdiction clause is an issue that has to be determined at the time the proceedings are issued".
  168. I do not consider that there is anything in this passage – which is principally concerned with timing and the relevance of defences raised – which indicates that the way in which a particular claim is formulated in the foreign proceedings is determinative of the issue of whether the dispute arises in relation to a particular relationship, or indeed more generally as to whether the claim falls within the scope of a jurisdiction clause. No such argument was presented in Airbus. As discussed in paragraphs [34] - [36] above, the question is whether the claim brought in the English proceedings is within the scope of the jurisdiction clause. And where that raises issues concerning a claim made in foreign proceedings, it is obviously necessary to look at the nature of the claim made in those foreign proceedings. It is clear that what is then required is for the court to consider the substance of the claim that is made. That is what Males LJ does in paragraph [81] in Airbus, when reaching the conclusion that, even though the claim in that case was made in tort, it nevertheless fell within the scope of the jurisdiction clause in the contract. Similarly, in Petromena, both Longmore LJ and Ryder LJ referred to the "thrust" of the claim, and Longmore LJ specifically disregarded an amended formulation of the claim on the basis that it was subservient to and parasitic on the main proceedings (see [89]).
  169. I consider that the legal relationship, in connection with which the agreement conferring jurisdiction was concluded, can on the facts of the present case be characterised in different ways. Mr. Joseph, looking only at the terms of the Facility Agreement, characterised the relationship as being that of lender and borrower, which are the terms used to describe Etihad and Air Berlin in the Facility Agreement itself. For reasons given below, there are other ways in which the legal relationship between the parties can be characterised or viewed, but it is convenient to start by considering the present question by reference to the borrower/ lender relationship; not least because there can be no dispute that this was one way in which the relationship could be characterised. Even if this were the only appropriate characterisation, I have no doubt that the dispute concerning the Comfort Letter can fairly (and certainly to a good arguable case standard) be said to originate from that relationship for the following reasons.
  170. (1) The genesis of the Comfort Letter was, as described above, the Facility Agreement and Etihad's unwillingness to increase the amount to be loaned under that agreement. The Comfort Letter therefore originated in the lender/ borrower relationship.
  171. (2) The Comfort Letter refers to Etihad's intention "to continue to provide the necessary support to Air Berlin to enable it to meet its financial obligations". Whilst it is true that there are different ways in which Etihad could potentially provide that necessary support, a very obvious way is by loaning further monies to Air Berlin. Against the background of Air Berlin and KPMG having indicated that Air Berlin had additional cash requirements, the parties must have had well in mind the possibility that Air Berlin would look to Etihad for the additional cash, which had been the subject of discussion but which Etihad was not at that stage willing to commit to providing in the Facility Agreement. The Comfort Letter therefore looks forward to the intention to continue the lender/ borrower relationship.
  172. (3) The Comfort Letter goes on to say that: "Our commitment is evidenced by our historic support through loans and support on obtaining financing for Air Berlin". Accordingly, the Comfort Letter expressly referenced the historic relationship of lender and borrower, as well as anticipating its intended continuation.
  173. (4) Air Berlin's claim in the German proceedings not only explains that the Facility Agreement was the genesis of the Comfort Letter, but positively relies upon the Facility Agreement in support of both aspects of its case (i.e. binding contractual commitment and culpa in contrahendo). In relation to its argument that the Comfort Letter was binding, Air Berlin relies upon the Facility Agreement: see e.g. paragraph [175] of the German claim quoted in paragraph 81 above. It also does so in relation to the culpa in contrahendo claim.
  174. Accordingly, the Facility Agreement provides an important foundation for the legal theories which are advanced in the German proceedings.
  175. (5) For reasons explained in Section C above, there is a good arguable case that the Comfort Letter was not a binding contractual commitment, but was instead a statement of intention which was ancillary to the Facility Agreement.
  176. (6) I do not consider that Air Berlin, or indeed either party, could reasonably be surprised by the referral of disputes under the Comfort Letter to the English court, bearing in mind not only its close connection to the Facility Agreement but also the English law and jurisdiction clauses in the agreements in the support package to which Etihad and Air Berlin were party. This conclusion is in my view reinforced by the terms of jurisdiction clause itself. This is potentially relevant in considering whether a party should be taken by surprise by the reference of a particular dispute to the chosen court to or tribunal: see the decision in Hydrogen Peroxide. The clause in the present case expressly refers to non-contractual obligations arising from or in connection with the Facility Agreement. Since there is a good arguable case that the Comfort Letter created only non-contractual obligations, this language is directly applicable to the Comfort Letter.
  177. In the light of these matters, it is in my view unrealistic for Air Berlin to contend that the dispute relating to the Comfort Letter originated in a relationship other than that in connection with which the agreement conferring jurisdiction was concluded.
  178. At the heart of Mr. Joseph's submission was the proposition that the relationship created by the Comfort Letter was very different to the relationship of lender and borrower. He relied on the characterisation of the relationship in the German claim; a relationship of patron and protιgι. He submitted that one simply had to compare that claim made in the German proceedings with the relationship created by the Facility Agreement, and it was obvious that they were not the same.
  179. I reject this approach. For reasons already given, Etihad has a good arguable case that the Comfort Letter and any non-contractual claim in connection therewith are subject to English law. Accordingly, the characterisation of the relationship under German law is, in my view, beside the point and is certainly not determinative. Furthermore, Etihad has a good arguable case that, applying English law, the Comfort Letter created no binding contractual rights and obligations, was no more than a statement of present intention, and was therefore a non-binding document which was ancillary to the Facility Agreement. Again, this leads to the conclusion that the different characterisation of the relationship under German law is beside the point and is certainly not determinative.
  180. Etihad's argument therefore succeeds on the basis that there is a good arguable case that the dispute concerning the Comfort Letter originates from the relationship of lender/ borrower.
  181. However, I agree that there are other reasons why Etihad has a good arguable case that the dispute originates from the legal relationship in connection with which the jurisdiction agreement was concluded. As already indicated, the substance of Mr. Dicker's submission was that, in identifying the particular legal relationship between Etihad and Air Berlin, it was permissible to look more generally at the relationship between the parties, and the agreements which the parties concluded in connection with that relationship, rather than focusing narrowly on the relationship within the Facility Agreement. He submitted, therefore, that it was permissible and appropriate to look at the "pre-existing position" between the parties; a relationship in which Etihad, qua shareholder, provided financial support to the company in which it held a significant shareholding. A second approach was to look at the relationship that was created in the package of agreements concluded in April 2017. The support package was effectively the sum of the rights and obligations contained within that package. Where there is a package of agreements, or rights and obligations, it may not be easy to ascribe a single convenient label to the legal relationship. The easiest label would be "provider of financial support" and "recipient of financial support". Whilst this relationship would not be found so described in textbooks, it was a perfectly sensible commercial description of the relationship in the present case, which was constituted by various elements of the package.
  182. I agree with both of these approaches. If a particular agreement is concluded within the context of a wider legal relationship between the parties, I consider it appropriate to look at that context in considering whether the dispute arises from the legal relationship in connection with which the agreement was concluded. This is the approach taken by the court in the Altera case. Consideration of the context is consistent with the purpose of Article 25, namely to prevent a party from being surprised by the referral of the dispute to the chosen tribunal. The contrary approach seems to me to be artificial, since it has the effect of divorcing the agreement containing the jurisdiction clause from its context. It also has the potential to lead to the inapplicability of the jurisdiction clause to the particular dispute, notwithstanding that (given the context) a party could not be taken by surprise.
  183. In my view, consideration of the wider context is fully in accordance with Powell Duffryn. The court in that case referred to the "legal relationship in connection with which the agreement was concluded". As Mr. Dicker correctly submitted, that is not necessarily the same as identifying the legal relationship contained in the contract which contains the jurisdiction clause. In some cases, such as Powell Duffryn itself, the only legal relationship will be the contract which contains the jurisdiction clause. But in other cases, depending on the facts, it may be possible to say that the jurisdiction clause was concluded in connection with a wider legal relationship.
  184. In the present case, there was a background relationship whereby Etihad was a major shareholder in Air Berlin, and as part of that relationship had provided (as is far from unusual in the context of major shareholders) financial support in a variety of forms in order to protect its investment. This was a relationship between shareholder and company, and a feature of that relationship as it developed – albeit not found in the Articles of Association of Air Berlin – was the provision of financial support, both cash and non-cash support. The Comfort Letter was provided as part of that relationship, and the present dispute originates in that relationship rather than in any separate or different relationships. Since, as a matter of factual background, the provision of financial support was an important feature of the relationship between shareholder and company, it seems to me to be appropriate to take this into account in considering the application of Article 25, and in particular in considering whether a party would be taken by surprise by the invocation of the jurisdiction clause in the Facility Agreement. In the present case, therefore, the dispute does arise in connection with the relationship between shareholder and company, and does not originate in some different relationship.
  185. The same conclusion is reached if one were only to consider the package of agreements concluded in April 2017. Where parties conclude a number of agreements as part of a package, it may not be easy to attach a convenient single label to the legal "relationship" so created. Indeed, that may also be the case where a single agreement is concluded; since it is not unusual for commercial agreements to contain a variety of different obligations to which different legal labels might attach. For example, if a package of financial support were concluded in a single agreement – whereby a shareholder agreed to provide a capital injection, loan finance, guarantees in favour of third parties, and to waive or suspend rights under other instruments or securities – it might be difficult to ascribe a textbook label to the relationship or series of relationships so created. What matters in my view, however, is not the labelling of the relationship, but whether the dispute originates from that relationship or rather from different relationships. The application of the test of surprise will usually provide an answer to that question. Here, the dispute does clearly originate from the relationship created by the package of rights and obligations which were created in April 2017, and one searches in vain for any different or more distant legal relationship.
  186. I also consider that there is nothing in Article 25 which requires that the relationship should be given a label such as borrower/ lender or shareholder/company. If, in the above example, a package of financial support is given by a shareholder in a single agreement containing a jurisdiction clause, I see nothing wrong in simply identifying the contract as containing the legal relationship. If, simultaneously, a Comfort Letter was also provided indicating an intention to provide additional support, it would not be difficult to say that a dispute about the enforceability of the Comfort Letter originates from the legal relationship in connection with which the jurisdiction agreement was concluded. The position is no different if, instead of there being a single contract containing multi-faceted obligations, there is a package of contracts concluded at the same time. In both cases, the legal relationship in which the dispute originates would in my view be the same.
  187. Some reliance was placed by Mr. Joseph on the words "as such" in paragraph [32] of Powell Duffryn:
  188. "On this point a jurisdiction clause appearing in the articles of association of a company fulfils this requirement where it covers disputes which have arisen or which may arise in connection with relationships between the company and its shareholders as such."
  189. I do not consider the words "as such" negate the conclusions which I have reached in relation to the different reasons as to why the disputes do originate from the legal relationship in connection with which the jurisdiction clause in this case was concluded. It must be recalled that these words "as such" appeared in the context of a decision about a jurisdiction clause appearing in the articles of association, and where the only relationship between the parties was that contained in those articles. Even if they are transposed so as to require the dispute to originate in the relationship of borrower and lender "as such", I think that it does for the reasons already given, particularly bearing in mind that these words do not mean that the dispute must relate to the contractual obligations contained in the Facility Agreement. Similarly, if one attempts to transpose the words "as such" into the wider legal relationships which I have identified, I do not consider that any different result is reached.
  190. For these reasons, I consider that Article 25, and the test set out in Powell Duffryn, is satisfied.
  191. E: Article 31 (2)

    The issue

  192. This issue arises because it is common ground that the German court was first seised of the proceedings which relate to Air Berlin's causes of action relating to the Comfort Letter, and that this court (i.e. the English court) was second seised. Ordinarily, Article 29 (1) of Brussels Recast would require the second seised court to stay its proceedings:
  193. "Without prejudice to article 31 (2), where proceedings involving the same cause of action and between the same parties are brought in the courts of different member states, any court other than the court first seised shall of its own motion stay its proceedings until such time as the jurisdiction of the court first seised is established".
  194. However, the opening words of Article 29 (1) ("Without prejudice to article 31 (2)") provide that the second seised court is not required to stay in the situation covered by Article 31 (2), which provides as follows:
  195. "2. Without prejudice to Article 26, where a court of a Member State on which an agreement as referred to in Article 25 confers exclusive jurisdiction is seised, any court of another Member State shall stay the proceedings until such time as the court seised on the basis of the agreement declares that it has no jurisdiction under the agreement."
  196. Article 31 (2) therefore requires a non-designated court, even if first seised, to stay its proceedings if the designated court is "seised on the basis of the agreement". This is so even if the designated court was second seised. Article 31 (2) does not expressly provide that, in circumstances where it applies, the second seised designated court can continue with the proceedings. However, in Commerzbank AG v Liquimar Tankers Management Inc [2017] EWHC 161 (Comm) at [77] – [78] ("Commerzbank"), Cranston J. rejected the argument that the second seised court could not continue with the proceedings. In his view, the argument would make a nonsense of Article 31 (2). Air Berlin did not contend that this aspect of Cranston J's decision was wrong.
  197. The important issue, therefore, is whether the jurisdiction clause in the present case is a clause which "confers exclusive jurisdiction" within the meaning of Article 31 (2). A related question is whether the English court can properly be described as being "seised on the basis of" such exclusive jurisdiction agreement within the meaning of Article 31 (2). Air Berlin says "no" to both questions, and Etihad says "yes".
  198. The essential reason why Air Berlin contends that the clause in the present case does not fall within Article 31 (2) is that is "asymmetric". This label or description has been applied to clauses which contain different provisions regarding jurisdiction, depending on whether the proceedings are initiated by one party to the agreement rather than the other. In the present case, Clause 33 starts (Clause 33.1.1) by providing that the English courts "have exclusive jurisdiction to settle any dispute arising out of or in connection with this Agreement". However, there is a qualification to that agreement in Clause 33.1.3. This qualification permits Etihad (but not Air Berlin) to take proceedings "in any other courts with jurisdiction". This means that the clause, when read as whole, does not give the English courts complete exclusivity. The ability of Etihad (but not Air Berlin) to take proceedings in jurisdictions other than England means that the exclusivity provided by the clause applies only to proceedings commenced by Air Berlin. Air Berlin contends that is sufficient to take such clauses outside the range of Article 31 (2), with the consequence that the English court must stay its proceedings pending the decision of the German courts.
  199. The case-law and the literature show that such "asymmetric" clauses have for some considerable time been a widely-used feature of international finance transactions. In her illuminating article The Future Enforcement of Asymmetric Jurisdiction Agreements" [2018] 67(1) ICLQ 37, Louise Merrett (referring to an earlier article by Professor Richard Fentiman) describes the aim of such a clause as being to ensure that creditors can always litigate in a debtor's home court, or where its assets are located. They also contribute to the readiness of banks to provide finance, and reduce the cost of such finance to debtors, by minimising the risk that a debtor's obligations will be unenforceable. Notwithstanding their common use, the question of whether such asymmetric clauses fall within Article 31 (2) has been the subject of academic comment and discussion for some time. I return to this discussion below.
  200. One issue, which has also been the subject of academic discussion, is whether or not an asymmetric clause falls within the scope of Article 25 of Brussels Recast at all. In its application notice, one basis of Air Berlin's present application was that asymmetric clauses do not fall within Article 25 (1) of Brussels Recast. If correct, this would have the consequence that neither party could use the jurisdiction agreement in the Facility Agreement, whatever its scope, as the basis for establishing the jurisdiction of the English court. There have, however, been a number of cases in England, and indeed elsewhere, where this argument has been rejected. Air Berlin did not pursue this argument and accepted that asymmetric clauses do fall to be treated under Article 25, whilst reserving their position if the matter were to be decided at an appellate level or by the CJEU. It follows that if Etihad had commenced the present proceedings prior to Air Berlin's commencement of the German proceedings, Air Berlin would not be able to dispute the jurisdiction of the English courts (even if no other basis of jurisdiction could be established), and that any subsequent proceedings commenced by Air Berlin in Germany would need to be stayed. But it was also common ground that it did not automatically follow from the fact that the jurisdiction agreement was within Article 25 that it also fell within Article 31 (2); not least because Article 25 is applicable to jurisdiction clauses which do not provide for exclusive jurisdiction, whereas Article 31 (2) uses the words "exclusive jurisdiction".
  201. The question of whether an asymmetric clause falls within Article 31 (2) has been considered in three cases to which I was referred. In each of those cases, the judge decided that it was.
  202. The first and second decisions were in related proceedings between a Spanish company (Codere) and a firm which had advised on a restructuring (Perella Weinberg Partners UK LLP). The first case, Codere SA v Perella Weinberg Partners and others, was heard in Spain by the Court of First Instance no. 5 of Alcobendas, and the decision is dated April 2016. In that case, the Spanish court was first seised of proceedings commenced by Codere in September 2015. The English court was second seised of proceedings commenced by Perella in December 2015. The Spanish court granted a stay of proceedings which had been commenced by Codere in breach of an asymmetric clause, pending determination by the English court of an application by Codere to challenge the jurisdiction of the English court. In granting the stay, the judge relied upon and applied the approach advocated by Professor Richard Fentiman in his discussion of Article 31 and asymmetric clauses in U Magnus and P Mankowski (eds), European Commentaries on Private International Law, Volume 1 (2015), pages 751-753. Professor Fentiman's conclusion was that:
  203. "Principle suggests that a finance party may rely on Article 31 (2) in such a case. Such asymmetric agreements, although non-exclusive for the benefit of the 'beneficiary' under the clause are exclusive against a counterparty. Article 31 (2) should therefore engage if a counterparty brings proceedings other than in the designated court in breach of its promise to sue only in that court."
  204. The Spanish court agreed:
  205. "This court adopts Prof. Fentiman's theory. There is no reason to avoid applying art. 31.2 to asymmetric clauses, provided that they confer exclusive jurisdiction (which is what the wording of the statute requires), even if it is only in the event that a suit is brought by one of the parties, and the conferral is non-exclusive for the other. If proceedings have been instituted where the conferral of exclusive jurisdiction may have been violated, article 25 applies, as well as art. 31.2; In fact, in this case it may make even more sense to apply the rule so that one of the courts before which complaints dealing with the same issue have been filed may rule first, and to prevent contradictory rulings, which is ultimately what is sought to be avoided. It also makes perfect sense for the "designated" court, i.e., the one that could have exclusive jurisdiction depending on the jurisdiction conferral agreement having a disputed meaning, to, in such case, settle the issue first, especially when the law governing the clause and its interpretation is its national law, as in the case at hand here."
  206. The same case then came before Walker J. on 17 May 2016: see [2016] EWHC 1182 (Comm). Prior to the hearing, Codere had raised the same argument that had been unsuccessful in Spain; namely that an asymmetrical clause could not constitute an exclusive jurisdiction clause for the purposes of Article 31.2. In paragraph [18] of his judgment, Walker J. records that counsel did not seek to press the argument:
  207. "He was plainly right not to do so. So far as article 31.2 is concerned, there is, as it seems to me, good commercial reason to focus upon the question whether party seeking to bring proceedings in a court of "another member state" has agreed that the dispute in question is to be subject to the exclusive jurisdiction of a court or the courts of another member state. Nothing in article 31.2 requires that the party relying upon the exclusive jurisdiction clause must itself be under a symmetrical obligation. In those circumstances, I did not call upon Mr Scott to respond to what was said in relation to this suggested proviso."
  208. Walker J. was therefore clearly of the view that an asymmetrical clause was an exclusive jurisdiction clause for the purposes of Article 31.2, albeit that in the end the point was not the subject of any detailed oral argument before him.
  209. Commerzbank

  210. The identical issue was, however, the subject of detailed argument, and decision, in the Commerzbank case. The borrower and guarantor, in breach of an asymmetric clause in loan and guarantee contracts, commenced proceedings in Greece. The bank subsequently commenced proceedings in England which mirrored the Greek proceedings. The defendants sought a stay of proceedings under Article 29 of Brussels Recast. This gave rise to the same issue as in the present proceedings; i.e. whether asymmetric jurisdiction clauses were agreements conferring exclusive jurisdiction on the English courts within the meaning of Article 31 (2).
  211. In a detailed judgment, which includes a valuable description of the legal background to Article 31 (2), Cranston J. held that they were. It is appropriate that I should set out the reasoning of Cranston J. in full:
  212. "62.  There is no warrant, in my judgment, for giving Article 29 of Brussels 1 Recast primacy and treating Article 31(2) as somehow an exception to it. Nor is there any warrant for giving Article 31(2) a narrow meaning. Whatever may have been the legislative history of the first-seized rule in the Brussels Convention and Brussels 1, there is nothing in Brussels 1 Recast indicating this approach. In my view, ordinary principles apply and both articles should be read together and given effect according to their language and purpose.
    63.  On its face Article 29 (1) is without prejudice to Article 31(2) , which can only mean that Article 29 (1) gives way to Article 31 (2) when the latter applies. A similar result obtains with Article 31(2) itself, which is without prejudice to Article 26: generally speaking, if a defendant enters an appearance before a court of a Member State, under Article 26 that court shall have jurisdiction even though an agreement confers exclusive jurisdiction on another court. While "subject to" is used elsewhere in the Regulation to achieve the effect that one article takes precedence over another, the terms "without prejudice to" and "subject to" are to my mind equivalent in the outcome they produce.
    64.  The natural meaning of the words in Article 31(2) - "an agreement [which] confers exclusive jurisdiction"- to my mind includes asymmetric jurisdiction clauses such as those in the various agreements in this case between the Bank and the defendants. Considered as a whole, they are agreements conferring exclusive jurisdiction on the courts of an EU member state, namely, England. That this applies in respect of a claim by the defendants alone does not detract from this effect.
    65.  Case 25/78, Nikolaus Meeth v. Glacetal Sarl [1979] CMLR 520 was decided under the first paragraph of Article 17 of the Brussels Convention (the predecessor of Article 25 in Brussels 1 Recast). The case involved a French party and a German party. There the jurisdiction clause provided that if Meeth sued Glacetal, the French court alone had jurisdiction, while if Glacetal sued Meeth, the German courts alone had jurisdiction.
    66.  The ECJ held that the first paragraph of Article 17 could not be interpreted as having no application to a clause under which two parties to a contract, domiciled in different states, could be sued only in the courts of their respective states. In effect it was an exclusive jurisdiction clause even though which court had exclusive jurisdiction turned on which party sued.
    67.  For our purposes the reasoning of the ECJ is important: an agreement such as the one it was considering, designating the courts of two states, could still be regarded under the first paragraph of Article 17 as one where a court or courts "of one Contracting State" had "exclusive jurisdiction". The court said:
    "That wording [of Article 17], which is based on the most widespread business practice, cannot, however, be interpreted as intending to exclude the right of the parties to agree on two or more courts for the purpose of settling any disputes which may arise…[I]t excludes, in relations between the parties, other optional attributions of jurisdiction, such as those detailed [elsewhere in the Convention]."
    68.  To my mind the case provides further support for the view that where a clause confers exclusive jurisdiction on the court or courts of a Member State when one party sues, the clause will still be an exclusive jurisdiction clause for the purposes of Article 31(2) even where, if the other party to the clause sues, the clause shows the parties to have agreed that jurisdiction is to be conferred differently, or allowed to engage differently.
    69.  The conclusion that an asymmetric jurisdiction clause cannot be treated as non-exclusive under Brussels 1 Recast is also consistent with the Regulation's aims. There is the aim of party autonomy in Recital 19, although that may be counterbalanced by the aims in Recital 21 of avoiding concurrent proceedings and irreconcilable judgments. But Recital 22, which is the specific background to Article 31(2) , is clear: there needs to be an exception to the general lis pendens rule to enhance the effectiveness of exclusive choice of court agreements and to avoid abusive tactics.
    70.  Thus with the asymmetric jurisdiction clauses in the present case, the defendants agreed to sue only in the courts of one EU Member State, England. Instead, they have enabled another court, the Greek court, to be seized of the matter. It would undermine the agreements of the parties, and foster abusive tactics, if the jurisdiction clauses in these agreements were to be treated not as exclusive, but as non-exclusive.
    71.  The Hague Convention, in my view, offers no assistance in the characterisation of asymmetric jurisdiction clauses under Article 31(2) of Brussels 1 Recast. There is no reference to the Hague Convention in Brussels 1 Recast, although the drafting of both occurred in tandem and Council Decision 2014/887/EU referred to ensuring coherence between the rules of the EU on the choice of court in civil and commercial matters and those of the Hague Convention.
    72.  While there is an overlap between the two instruments, however, there are important divergences. Thus there are differences between the two in the formal requirements for exclusive jurisdiction clauses, the Hague Convention in Article 3(c) requiring writing or an accessible form, Brussels 1 Recast in Article 25 allowing agreements to be established on a wider basis, through the practices of the parties or by commercial usage.
    73.  Further, there is a definition of exclusive jurisdiction clauses in Article 3(a) of the Hague Convention, whereas there is no definition in Brussels 1 Recast. The reporters record that the Diplomatic Session adopting the Hague Convention accepted that the definition in Article 3(a) did not extend to asymmetric jurisdiction clauses, something the reporters themselves do not seem to have regarded as clear.
    74.  There are good arguments in my view that the words of the definition of exclusive jurisdiction clauses in Article 3(a) of the Hague Convention cover asymmetric jurisdiction clauses. For present purposes, however, there is no need to reach a concluded view on the ambit of the definition. Even if it were to be read as excluding asymmetric jurisdiction clauses, however, that in my view is of no assistance as to the quite separate issue of their characterisation under Article 31(2) of Brussels 1 Recast.
    75.  There is nothing in my view to Liquimar's submission that the English actions are not brought in this court pursuant to any exclusive jurisdiction agreement or that the exclusive element of such a clause only governs proceedings in Greece. The fact is that the clause confers exclusive jurisdiction on the English court when the defendants sue, they have instituted proceedings elsewhere, and that is why the English actions are being pursued.
    76.  Nor do I give credence to the argument that characterising asymmetric jurisdiction clauses as conferring exclusive jurisdiction gives rise to unacceptable anomalies. Even if under an asymmetric jurisdiction clause a borrower could deprive a Bank of its right to sue in any competent court by starting proceedings in England, the designated jurisdiction, it would be even more anomalous if Liquimar is correct and by starting proceedings elsewhere than England in breach of what it agreed the borrower could cause proceedings by the Bank in the designated jurisdiction, namely England, to be stayed."
  213. Cranston J. refused permission to appeal. But an application for permission was made to the Court of Appeal, and permission was granted by Beatson LJ. His grant of permission does not contain detailed reasoning, but simply states that the defendants were "entitled to argue that the judge erred in refusing to stay Commerzbank's claims". However, the appeal does not appear to have been pursued.
  214. On this application, Air Berlin seeks to persuade me not to follow the decision of Cranston J. I approach this argument on the basis that, as a matter of judicial comity, I should follow the decision of another judge of first instance, unless I am convinced that the judgment is wrong: Police Authority for Huddersfield v Watson [1947] 1 KB 842, 848. More recently, in Willers v Joyce [2016] UKSC 44, Lord Neuberger said at [9]:
  215. "So far as the High Court is concerned, puisne judges are not technically bound by decisions of their peers, but they should generally follow a decision of a court of co-ordinate jurisdiction unless there is a powerful reason for not doing so."

    Air Berlin's arguments

  216. The arguments advanced by Air Berlin are materially the same as those which failed to persuade Cranston J. There has, on both sides, been some citation of authority, including academic commentary, which was not before Cranston J. But the arguments remain essentially the same.
  217. Air Berlin argues that the Commerzbank decision is wrong and unsupportable and that the reasoning is flawed. At the very lowest, the question is one which would benefit from guidance from the CJEU. They submitted as follows.
  218. First, the Judge at [62] failed to adopt a restrictive construction of Article 31(2) because he failed to characterise it as an exception. Indeed he did the opposite, adopting a broad construction. From then, his decision was built upon error.
  219. Second, the Judge was wrong to reject at [71] – [74] as irrelevant the clear authority that materially identical wording (to those in Article 31(2) Brussels Recast) in the Hague 2005 Convention excluded asymmetric clauses.
  220. Third, the learned Judge failed to identify that Article 31(2) requires a jurisdiction clause that is exclusive with respect to all disputes that arise in connection with a particular legal relationship. This led him to find that a clause could be exclusive even though it "applies in respect of a claim by the defendants alone" [64], which is directly contrary to the words of Article 25 incorporated into Article 31(2) (and then qualified by the word 'exclusive').
  221. Fourth, the learned Judge misapplied Meeth. That case is authority that "exclusivity" requires the total disapplication of the other grounds for jurisdiction. The Judge misunderstood the passage he himself quoted from Meeth [67] and failed to cite the important CJEU decision in Case C-24/76 Estasis Salotti di Colzani Aimo e Gianmario Colzani v RUWA Postereimaschinend GmbH [1976] ECR 1831 which is followed in Meeth. His conclusion in relation to asymmetric clauses [68] runs counter to the reasoning in Meeth, rather than taking support from it.
  222. Fifth, the learned Judge ignored the clear pointer that the Article 31(2) wording gives that asymmetric clauses are excluded. It provides for proceedings in the court first seised to be stayed until 'the court seised on the basis of the agreement' has declined jurisdiction. In the case of an asymmetric clause such as the present, the court second seised (England) is not seised on the basis of any jurisdiction agreement (which is silent as to where Etihad may bring proceedings) and there is no court seised on the basis of the agreement. Article 31(2) is drafted to give jurisdiction to a court seised under an agreement. It is not drafted to give jurisdiction to any court seised other than in breach of an agreement (although it could have been drafted in this way).
  223. Sixth, the judge ignored the anomalies that could arise if asymmetric clauses were included within the scope of Article 31 (2). Air Berlin argued that if Article 31(2) were to apply to an asymmetric clause, such a clause could be used against the benefiting party, contrary to its obvious intention. For example, if the clause in the Facility Agreement applied, but Etihad had sued Air Berlin first in Germany (rather than the other way around, as occurred), that would be entirely proper (the Clause not mandating jurisdiction for proceedings brought by Etihad) and Germany would be the court first seised. However, if Air Berlin then sued Etihad in England (its only available jurisdiction under the Clause), England would be the court second seised. Nevertheless, on Etihad's argument, Article 31(2) would require the German court to stay the proceedings pending determination of the jurisdiction issues by the English Court 'seised on the basis of the agreement', even though Etihad's own claim brought first in Germany was not contrary to a choice of court agreement. If Article 31(2) were intended to apply to asymmetric clauses, it would need to be drafted differently to deal with the complexities that arise.
  224. Finally, Cranston J's conclusion that treating asymmetric clauses in this way comports with the aims of Brussels Recast (see [69]-[70]) was wrong. This not only misstates the nature of an agreement for exclusive jurisdiction but does not deal with the important treatment of this very subject in the Heidelberg Report. This report, produced in September 2007, had been commissioned by the European Commission. Paragraph 892 stated that the new contemplated lis pendens rules in Article 31 (2) would not apply to asymmetric jurisdiction clauses.
  225. Etihad, by contrast, invited me to follow the decision and reasoning in Commerzbank. Mr. Dicker's oral submissions expanded, where appropriate, on the reasons given by Cranston J.
  226. Discussion

  227. Even if this case were not covered by pre-existing authority, I would have no difficulty in saying that Article 31 (2) applies in the present case. It is important, in my view, to approach the issues not by applying labels but by identifying, at the start, the agreement which the parties have reached. This requires understanding the particular obligations upon which the parties have agreed. There can be no doubt that the effect of Clause 33.1 is that Air Berlin has agreed upon the exclusive jurisdiction of the English courts in respect of any proceedings which they commence against Etihad. Such an agreement, viewed as a matter of English law (which is the governing law of the Facility Agreement) carries with it an obligation not to sue in other jurisdictions. This is clearly a case where (see BNP Paribas SA v Anchorage Capital Europe LLP [2013] EWHC 3073 para [88]) the commencement and pursuit of the German proceedings are things which Air Berlin has promised not to do.
  228. Once these contractual obligations have been identified, there is in my view no difficulty in saying that the relevant agreement is an agreement for exclusive jurisdiction within Article 31 (2). This is because, as Professor Fentiman puts it, the agreement is "exclusive against a counterparty". Equally, as the Spanish court correctly understood, there is a conferral of exclusive jurisdiction on the English court in the event that proceedings are brought by Air Berlin. This remains a conferral of exclusive jurisdiction in respect of disputes in respect of which Air Berlin wishes to commence proceedings, even if there is no similar conferral for disputes in respect of which Etihad wishes to commence proceedings. The essential point is concisely, and in my view, convincingly expressed by Merrett at 55-56 as follows. Having referred to the rationale for the new rule in Article 31 (2) of Brussels Recast, and Recital (22) thereof – to which I return below – her analysis is straightforward:
  229. "In an asymmetric agreement, the borrower has promised not to sue anywhere other than the chosen jurisdiction. The question of whether the other party did or did not agree to do the same does not arise when the bank is seeking to enforce the agreement and should be irrelevant. Thus, the point is not so much that "considered as a whole" [asymmetric agreements] are agreements conferring exclusive jurisdiction, as the judge put it in Commerzbank. Rather, each obligation can be considered on its own; the clause includes a promise by the borrower not to sue in any jurisdiction and that promise is capable of being protected by Article 31(2). Each different obligation necessarily falls to be considered separately and the fact that the bank is not under a similar obligation is neither here nor there."
  230. I consider that this approach, namely to identify the relevant obligation, is correct in principle and consistent with the decision of European Court of Justice in Case 23/78, Nikolaus Meeth v. Glacetal Sarl [1979] 1 CMLR 520. The relevant clause was in a contract between a French supplier (Glacetal) and a German buyer (Meeth). It provided for the French court to have exclusive jurisdiction if Meeth sued Glacetal, but the German court had exclusive jurisdiction if Glacetal sued Meeth. The clause in that case was reciprocal, but could be described as asymmetric; in the sense that the obligations of each party in relation to jurisdiction differed. Glacetal began proceedings, in accordance with the clause, against Meeth in Germany for non-payment of certain deliveries. Meeth wished to counterclaim and exercise a set-off for late delivery. The German court decided that Meeth could not do this, because the French courts had exclusive jurisdiction over claims against Glacetal. On a reference to the European Court of Justice, the question was whether the agreement came within Article 17 of the Brussels Convention (the predecessor of Article 25): "Does the first paragraph of Article 17 of the Convention permit an agreement under which the two parties to a contract for sale, who are domiciled in different States, can be sued only in the courts of their respective states". The court was therefore concerned with the validity of the clause for Article 17 purposes. The possible invalidity arose from the fact that the clause conferred jurisdiction on the courts of two contracting states, whereas Article 17 referred to an agreement to confer jurisdiction on the courts of "a contracting State". The case was therefore not concerned with whether a particular clause can or cannot be described as "exclusive". At this stage, of course, there was no equivalent to Article 31 (2) of Brussels Recast. Nor was the court concerned to provide a definition of what constitutes an exclusive jurisdiction clause.
  231. The Advocate General, in his opinion, considered that the validity of the clause could be recognised "without misgivings". He said:
  232. "… it can be recognised that the parties to a contract may stipulate that the courts of two States shall have jurisdiction to settle disputes arising from that contract, provided that each jurisdiction is restricted to a specified class of dispute. In short, there is nothing to preclude the parties, instead of treating all disputes which could arise from their contract as a whole, from dividing them into two or more groups in accordance with criteria which they are free to establish and prescribing the courts of a different State for each group. This is not common but there are no grounds for considering it unlawful."
  233. The Advocate General then noted that Article 17 of the Brussels Convention did not "prescribe limits" to the right to choose a particular court "in that it does not require that there be any material link between the contract and the State to which the court or tribunal specified by the clause pertains". The parties' choice did not infringe any of the other requirements of the Brussels Convention. He went on to say:
  234. "Accordingly, it is merely superfluous to point out that the choice expressed in the clause is fully in accordance with the general system of the Convention: the courts specified are those of the defendant's country of domicile, as is provided in the first paragraph of Article 2. The only effect of the clause is thus that the jurisdiction based on domicile is rendered exclusive."
  235. This opinion is instructive in that it expressly recognised, consistently with the analysis set out above, the ability of the parties to divide up their disputes into "two or more groups", even if they arose from the same contract. In the present case too, this is what the parties have done: disputes in respect of which Air Berlin wished to commence proceedings are exclusively for the English court, but disputes in respect of which Etihad wished to commence proceedings are permissible in a wider group of courts. The opinion also acknowledges party autonomy, recognising that there were no limits to the parties' choice. Furthermore, the fact that the particular choice in Meeth was consistent with the general system of the Brussels Convention was "superfluous": in other words, the parties could validly agree on the jurisdiction of a court which would not otherwise have had jurisdiction under the Convention. The fact that the clause was reciprocal (and therefore could loosely be described as "symmetrical") played no part in the analysis.
  236. The decision of the European Court of Justice, whilst more briefly expressed, is in my view to the same effect. The court's reasoning is contained in paragraph [5] of its judgment:
  237. "[5] According to the first paragraph of Article 17 'if the parties… have agreed that a court or the courts of a Contracting 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 exclusive jurisdiction'. With regard to an agreement conferring reciprocal jurisdiction in the form in which it appears in the contract whose implementation forms the subject-matter of the dispute, the interpretation of that provision gives rise to difficulty because of the fact that Article 17, as it is worded, refers to the choice by the parties to the contract of a single court of the courts of a single State. That wording, which is based on the most widespread business practice, cannot, however, be interpreted as intending to exclude the right of the parties to agree on two or more courts for the purpose of settling any disputes which may arise. This interpretation is justified on the ground that Article 17 is based on a recognition of the independent will of the parties to a contract in deciding which courts are to have jurisdiction to settle disputes falling within the scope of the Convention, other than those which are expressly excluded pursuant to the second paragraph of Article 17. This applies particularly where the parties have by such an agreement reciprocally conferred jurisdiction on the courts specified in the general rule laid down by Article 2 of the Convention. Although such an agreement coincides with the scope of Article 2 it is nevertheless effective in that it excludes, in relations between the parties, other optional attributions of jurisdiction, such as those detailed in Article 5 and 6 of the Convention."
  238. This passage therefore recognises the importance of party autonomy ("the independent will of the parties"). It also affirms the ability of the parties to agree upon two or more courts for the purposes of deciding disputes under a particular contract. It is inherent in a choice of two or more courts that, as the Advocate General said, the parties must have divided up their disputes into groups. It is therefore clear from this case that (contrary to a theme in Mr. Joseph's submission) it is not necessary that all disputes from a particular legal relationship should be submitted to the same court. There also is nothing in the judgment which requires that the parties' choice should be reciprocal, let alone symmetrical: the court simply noting that where there is a reciprocal choice of the courts of the parties' domicile, that was "particularly" a case where the independent will of the parties should be allowed to prevail.
  239. Contrary to a submission made by Air Berlin, I do not consider that the last sentence of paragraph [5] means that a jurisdiction clause can only be "effective", or indeed exclusive, if all other optional attributes of jurisdiction are excluded. Mr. Joseph relied heavily upon this passage as showing that the present clause was not exclusive, because Etihad would be able to start proceedings against Air Berlin in any court with jurisdiction under any of the provisions of Brussels Recast. I consider that this argument misses the key point. Party autonomy means that the parties are able to divide up their disputes into separate groups. In the present case, the material "group" is disputes in respect of which Air Berlin wishes to commence proceedings. In relation to that group, the parties have agreed upon exclusivity and have indeed excluded the other jurisdictional bases in Brussels Recast upon which Air Berlin might otherwise have been able to establish jurisdiction against Etihad.
  240. In paragraphs [65] – [68], Cranston J. derived support for his conclusions from the decision in Meeth. I consider that he was right to do so.
  241. Against this background, I turn to another important matter upon which Cranston J. relied. In paragraphs [69] – [70], Cranston J. refers to the background to and aims of Brussels Recast as recorded particularly in Recital (22). The material recitals to which Cranston J refers are as follows:
  242. "(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.
    (21) In the interests of the harmonious administration of justice it is necessary to minimise the possibility of concurrent proceedings and to ensure that irreconcilable judgments will not be given in different Member States. There should be a clear and effective mechanism for resolving cases of lis pendens and related actions, and for obviating problems flowing from national differences as to the determination of the time when a case is regarded as pending. For the purposes of this Regulation, that time should be defined autonomously.
    (22)  However, in order to enhance the effectiveness of exclusive choice-of-court agreements and to avoid abusive litigation tactics, it is necessary to provide for an exception to the general lis pendens rule in order to deal satisfactorily with a particular situation in which concurrent proceedings may arise. This is the situation where a court not designated in an exclusive choice-of-court agreement has been seised of proceedings and the designated court is seised subsequently of proceedings involving the same cause of action and between the same parties. In such a case, the court first seised should be required to stay its proceedings as soon as the designated court has been seised and until such time as the latter court declares that it has no jurisdiction under the exclusive choice-of-court agreement. This is to ensure that, in such a situation, the designated court has priority to decide on the validity of the agreement and on the extent to which the agreement applies to the dispute pending before it. The designated court should be able to proceed irrespective of whether the non-designated court has already decided on the stay of proceedings. This exception should not cover situations where the parties have entered into conflicting exclusive choice-of-court agreements or where a court designated in an exclusive choice-of-court agreement has been seised first. In such cases, the general lis pendens rule of this Regulation should apply."
  243. The background to Recital (22), and Article 31 (2), is described in paragraphs [26] – [28] of Cranston J's judgment. Their origin is the much-criticised decision of the ECJ in Erich Gasser GmbH v MISAT Srl (Case C-116/02) [2005] QB 1. The Austrian court was the designated court in an exclusive jurisdiction agreement. Nevertheless, proceedings were first taken in Italy by the Italian party. The subsequent proceedings commenced in the designated court by the Austrian party meant that the Austrian court was second seised. The European Court of Justice held that the designated (Austrian) court had to await the decision of the non-designated (Italian) court as to whether the latter had jurisdiction. If the Italian court decided that it had jurisdiction, the court second seised would have to decline jurisdiction in its favour. The outcome is sometimes referred to as the "Italian torpedo".
  244. It was common ground before me that Article 31 (2) of Brussels Recast sought to reverse the decision in Erich Gasser. This is reflected in the references in Recital (22) to the need to enhance the effectiveness of exclusive choice-of-court agreements, to avoid abusive tactics, and to ensure that the designated court has priority to decide on the validity of the agreement and the extent to which it applies to the dispute pending before it.
  245. The effect of Air Berlin's arguments, however, is that this reversal of Erich Gasser, and the aims set out in Recital (22) are only partially achieved; so that the Italian torpedo remains fully effective in the context of very widely-used asymmetric clauses. I consider that there can be no logical justification for this difference in approach. In the present case, Air Berlin entered into an agreement that proceedings that it commenced would be brought exclusively in England; it had no option to bring proceedings elsewhere, and Air Berlin agreed not to bring such proceedings elsewhere. It is accepted that such clauses are effective under Article 25 of Brussels Recast. The clauses are therefore entitled to be enforced like any other jurisdiction clause.
  246. Given that the parties had agreed on the exclusivity of the English courts for proceedings brought by Air Berlin, Air Berlin's argument leads to the conclusion that, in relation to a jurisdiction clause of this type, party autonomy should not be respected; that the effectiveness of choice of court agreements should not be enhanced; and that the chosen court does not have priority to decide on the validity of the agreement and the extent to which the agreement applies to the pending dispute. It is in my view not possible to see why these conclusions should follow simply because the clause leaves open the possibility that, in relation to another "group" of disputes, namely those where Etihad may wish to begin proceedings, the parties agreed that Etihad was not confined to the jurisdiction of the English courts.
  247. Air Berlin offered an explanation as to what might lie behind this illogicality. It was suggested that the Brussels Recast was a negotiated compromise between all EU states, many of which take a different approach than England to asymmetric clauses. However, I agree with Mr. Dicker that this is not a satisfactory or convincing explanation. There is nothing in the Brussels Recast itself which indicates, let alone makes clear, that asymmetric clauses are being treated differently to other exclusive jurisdiction clauses, with the consequence that the parties' agreement on exclusivity for particular disputes should fall outside the aims identified in Recital (22). Moreover, once it is accepted that asymmetric clauses are within Article 25 as a matter of EU law – so that any national laws which may cast doubt on the validity of such clauses are not relevant – it makes little sense to say that the effectiveness of such clauses should be decided in the first instance by a non-designated court, if first seised. In the case of a symmetrical English jurisdiction clause, it was common ground that Gasser had been reversed, so that the English court would decide upon the validity and scope of the clause. If the aims of the Brussels Recast are to be respected, the same should apply to an asymmetrical clause. This approach is in my view reflected in the reasoning of the Spanish court as set out above, more briefly in the decision of Walker J, and again in the decision of Cranston J in Commerzbank.
  248. I was referred to a large number of academic commentaries on the application of Article 31 (2) to asymmetrical clauses. A number of writers, addressing the issue prior to the decision in Commerzbank, endorsed the view that, in accordance both with principle and the aims of the Brussels Recast, namely, to enhance party autonomy and avoid abusive litigation tactics, Article 31(2) should not be confined to symmetrical exclusive jurisdiction clauses: see I Bergson, The Death of the Torpedo Action? (2015) 11(1) J Priv Int L 1-30 and R Fentiman (2015) in the article quoted by the Spanish court.
  249. I was also provided with a number of more recent academic commentaries referring to Commerzbank, in none of which is Cranston J's decision on this point said, or even suggested, to be wrong. In at least four commentaries, Commerzbank is cited, without detailed discussion or any hint of criticism, as uncontroversial authority for the proposition that an asymmetric jurisdiction clause is an exclusive jurisdiction clause for the purposes of Article 31(2), see: Dicey, Morris & Collins on The Conflict of Laws, 15th edn (Sweet & Maxwell, 2018) Fifth Cumulative Supplement at 12R-098 – 12-101 and 12-157 – 12-163; The White Book at 6JR.35; Arnould's Law of Marine Insurance & Average, 19th edn (Sweet & Maxwell, 2018) at 5-06; and P Wood, Conflict of Laws and International Finance, 2nd edn (Sweet & Maxwell 2019) at 21-014 – 21-016.
  250. The most detailed discussion of the issue is by Merrett, who considers the decision in Commerzbank to be consistent both with principle and the policies underlying Article 31(2): L Merrett, The Future Enforcement of Asymmetric Jurisdiction Agreements (2018) 67(1) ICLQ 37.
  251. In its written submissions, Air Berlin relied on two academic commentaries, both published before the decision in Commerzbank, which it said directly supported the proposition that Article 31(2) is limited to jurisdiction agreements that are exclusive as regards both parties: A Dickinson and E Lein, The Brussels I Recast (OUP, 2015) and M Ahmed, The Legal Regulation and Enforcement of Asymmetric Jurisdiction Agreements in the European Union [2017] EBLR 403. However, I considered that both commentaries were somewhat tentative on the issue.
  252. i) Dickinson and Lein stated at 11.54:

    "It is debatable whether [Article 31(2)] can be applied, in terms or by analogy, to other [non-exclusive] jurisdiction clauses, such as clauses for the benefit of only one of the parties. Though there may be reasons to accept this from a policy perspective, the wording of the provision limits its application to those agreements which confer exclusive jurisdiction simpliciter." (emphasis supplied).

    ii) On page 421, Ahmed considered that the wording of Article 31(2) "may have the effect that an asymmetric jurisdiction agreement, of the type frequently encountered in cross border finance transactions, is not caught by Article 31(2)", commenting that if this conclusion was correct, "there remains the potential for a party to an asymmetric jurisdiction agreement to disable the agreement by launching a pre-emptive strike in its preferred court".

  253. The overall picture therefore is that there is a significant body of academic writing, as well as reasoned decisions of the Spanish and English court (Cranston J) and a brief decision of Walker J., which considers that Etihad's argument in this case is consistent with principle and the aims of Brussels Recast. I am far from persuaded that these conclusions are wrong, and there are in my view powerful reasons for thinking that they are right.
  254. I would of course have hesitated before reaching this conclusion if it would involve reaching an interpretation of the language used in Brussels Recast which was unconvincing or tortuous. I do not consider that this is the case here. Cranston J. considered at paragraph [64] that the natural meaning of the words "an agreement [which] confers … exclusive jurisdiction" includes asymmetric clauses. His reasoning involves considering the clause "as a whole", and his conclusion that exclusive jurisdiction "applies in respect of a claim by the defendants alone". Mr. Dicker acknowledged that (as Merrett points out) it is not really a question of considering the clause "as a whole", at least in the sense of taking a global view of the clause. However, the conclusion that there is exclusivity in respect of a claim by the defendant alone, and that this qualifies as an agreement which "confers exclusive jurisdiction" is consistent with the analysis of Merrett which I have set out above and which I consider to be convincing.
  255. Since, in my view, Cranston J's decision is supported both by the aims of Article 31 (2) and by the ordinary meaning of an "agreement which confers exclusive jurisdiction" – as well as by other judicial decisions and a body of academic writing – I do not think that there is any basis on which I should decline to follow it. This means that I can deal (below) relatively briefly with Mr. Joseph's other arguments. Generally speaking, I considered that whilst some of these arguments were points that could fairly be made, there were reasonable answers to each of them, and that there was nothing that was so powerful or convincing as to lead me to a view contrary to that which I have reached on the basis of the above discussion.
  256. Mr. Joseph identified a supposed textual difficulty; namely that, in an asymmetric clause, the second (designated) court is not seised "on the basis of" the jurisdiction agreement. He described this as a point with a "tiny bit of subtlety". The argument is that it is only in relation to a symmetrical jurisdiction clause that the English court would be seised on the basis of an exclusive jurisdiction clause. In the present case, however, Etihad has commenced proceedings on the basis of a clause that allows them to bring proceedings wherever they wish. Although they have seised the English court, this is not on the basis that it is exclusive as regards them.
  257. I did not consider that there was any force in this argument. It was rejected by Cranston J. in paragraph [75] of his judgment, and in my view, he was right to do so. Etihad has seised the English court for the simple reason that the agreement provides for the exclusivity of the English court in relation to proceedings commenced by Air Berlin, and Etihad has brought the present proceedings on the basis of that agreement. There is no difficulty in saying that, in those circumstances, the English court is "seised on the basis of the [exclusive jurisdiction] agreement". Etihad is able to rely, and does rely, on Clause 33.1.1 and the agreement of the parties that the courts of England have exclusive jurisdiction.
  258. Furthermore, I also consider that Etihad does not in the present proceedings rely, and does not need to rely, upon any basis other than Clause 33.1.1 to establish this court's jurisdiction over Air Berlin. Etihad has not sought to invoke Clause 33.1.3, which would potentially enable the English court to exercise jurisdiction if it could be established on some different basis. In those circumstances, the English court is indeed seised in the present case on the basis of the English jurisdiction clause.
  259. Mr. Joseph criticised the judge's failure to take a restrictive approach to Article 31 (2), on the basis that it was an exception to the primary rule in Article 29 that the court second seised should stay its proceedings. There is indeed some authority, albeit not in the present context of the relationship between Article 29 and Article 31 (2), that exceptions within EU legislation should be strictly construed. Lasok and Millett, Judicial Control in the EU: procedures and principles, state at [691]:
  260. "Derogations from and exceptions to the treaty or other legislation must be strictly construed. It has also been said that they cannot be given a meaning that goes beyond what they expressly provide; that they cannot be interpreted in such a way as to extend their effects beyond what is necessary to safeguard the interests which they seek to secure; and that their scope must be determined in the light of the aims pursued by the measure containing them".
  261. In the present case, I accept that Article 31 (2) can be viewed as an exception to Article 29. It is clear that the lis pendens rule in Article 29 is subject to and qualified by the provisions of Article 31 (2). However, both of these provisions must be viewed in the context of Recital (22) of Brussels Recast, and the "aims pursued by the measure containing them". The interpretation advanced by Etihad gives effect to those aims. Air Berlin's interpretation does not. It is in my view inappropriate to give a restrictive approach to Article 31 (2) which produces a result contrary to the aims of Brussels Recast.
  262. Mr. Joseph submitted that Cranston J's decision failed to take into account the jurisprudence of the European Court of Justice in a trilogy of cases, Salotti, Meeth and Case 22/85 Anterist v Credit Lyonnais [1987] 1 CMLR 333. He submitted, in particular in reliance on Meeth, that these cases established that the essence of an exclusive jurisdiction clause was both to confer exclusive jurisdiction on a particular court, but also to exclude the other optional basis for jurisdiction that might otherwise exist. This asymmetric clause did not both confer and exclude, because Clause 33.1.3 expressly permitted Etihad to take proceedings in other courts with jurisdiction. The vital features of both conferring and excluding were therefore absent.
  263. I have already dealt in detail with the decision in Meeth, and the reasons why this supports the conclusion that Article 31 (2) is applicable. I do not consider that the other two cases are of any real assistance. The decision in Salotti pre-dated Meeth and is referred to in the opinion of the Advocate General in Meeth. The court in Salotti was not considering asymmetric clauses, or anything close to them. Rather, the issue concerned the requirements for proving whether the parties had agreed upon a jurisdiction clause in circumstances where, for example, the clause relied upon was printed on the back of the contract.
  264. Nor did I consider the decision in Anterist to be of any assistance. In one sense, the case could be said to be concerned with an asymmetric clause. The parties' agreement clearly provided for a particular French court to have exclusive jurisdiction, namely the court within whose jurisdiction the branch of Credit Lyonnais was situated. However, notwithstanding the clause, the bank brought proceedings in Germany, which was the domicile of Mr. Anterist. The bank submitted that since the jurisdiction agreement was for its benefit, it should be permitted to disregard it and establish jurisdiction against Mr. Anterist on another basis. The bank was therefore in a sense arguing that the clause was asymmetrical, because it permitted them to sue Mr. Anterist elsewhere. The court rejected this argument, because the clause did not make sufficiently clear that the bank was permitted to sue elsewhere:
  265. "[14] Since Article 14 of the Convention embodies the principle of the parties' autonomy to determine the court or courts with jurisdiction, the third paragraph of that provision must be interpreted in such a way as to respect the parties' common intention when the contract was concluded. The common intention was to confer an advantage on one of the parties must therefore be clear from the terms of the jurisdiction clause or from all the evidence to be found therein or from the circumstances in which the contract was concluded."
  266. In my view, the case provides some marginal support for Etihad's argument. This is because it clearly contemplates that a clearly worded asymmetric provision will be effective, and paragraph 14 of the judgment refers to Article 17 of the Brussels Convention (the predecessor of Article 25 of Brussels Recast) as embodying the principle of the parties' autonomy. However, the court was not considering the issue of Article 31 (2) which arises in the present case.
  267. Air Berlin repeated substantially the same arguments concerning the Hague Convention which had failed to persuade Cranston J.: see paragraphs [71] – [74] of his judgment in Commerzbank. It is true, of course, that the some of the materials which preceded the Hague Convention indicate that asymmetric clauses were not to be equated with symmetric clauses for the purposes of that Convention. However, the language of the Hague Convention itself does not make that clear. Unlike the Brussels Recast, it contains a definition of the expression "exclusive choice of court agreement":
  268. "… means an agreement concluded by two or more parties that … designates, for the purpose of deciding disputes which have arisen or may arise in connection with a particular legal relationship, the courts of one Contracting State or one or more specific courts of one Contracting State to the exclusion of the jurisdiction of any other courts."
  269. Some academic commentaries proceed on the basis that this definition in the Hague Convention does not apply to an asymmetric clause. The question was also touched upon by Bryan J. in Clearlake Shipping Pte Limited v Xiang Da Marine Pte Ltd. [2019] EWHC 536 (Comm) at paragraphs [62] – [64], but he did not need to reach a concluded view on the issue. Cranston J. was of the view that there are good arguments that the definition does apply to asymmetric clauses. The issue is discussed by Merrett at page 58 of her article. Her view is that the position is not entirely clear, but that the question should again be addressed by considering the effect of a particular obligation on a particular party. On this basis, she concludes that there is nothing in the structure or rationale of the Convention to mean that "if the claim is made against a borrower who has agreed to be sued in a particular jurisdiction and only that jurisdiction that the rules should not engage".
  270. Like Cranston J. and Merrett, I consider that there are good arguments that the rules in the Hague Convention are engaged by an asymmetric clause. But in any event, I am concerned here with the rules in Brussels Recast which are differently worded and also have the important Recital (22). I have come to a clear view, based on the wording of Brussels Recast, its aims and background, as well as the decision in Meeth and the three prior cases on Article 31 (2), that Article 31 (2) of Brussels Recast does apply to asymmetric clauses. I am far from convinced that even if a different result might arguably be reached under the Hague Convention, that this should dictate the answer under Brussels Recast. Indeed, there is a powerful case for saying that the conclusions reached in relation to Brussels Recast should assist in dictating the answer under the Hague Convention.
  271. Separately, Mr. Joseph was able to refer to the 2007 Heidelberg report as indicating that, at that early stage of considering proposals for what ultimately became Brussels Recast, the view was taken that the change to the lis pendens rules would not apply to asymmetric clauses. As Mr. Dicker pointed out, however, these proposals were at an early stage, and they contemplated a rather different scheme – including adherence to a standard form of jurisdiction agreement – to that which was in due course adopted. Ultimately, the question for me is whether the view expressed in this report was carried through to the Brussels Recast and in particular Article 31 (2). For the reasons given, I do not consider that it was.
  272. Finally, there is the question of the "anomaly" identified by Mr. Joseph. The situation described in this anomaly had some complexity, and there was argument as to whether the anomaly actually existed. Even if it does, I agree with Cranston J. that a far greater anomaly would arise if Air Berlin is correct; i.e. that it is entitled to start proceedings in breach of its agreement, and then require the designated court to stay its proceedings pending the determination of the jurisdiction dispute by the non-designated court.
  273. I do not consider, in the exercise of my discretion, that it would be appropriate for me to refer this issue to the CJEU. I have reached a clear view on this issue, and it is consistent with the views reached by three other judges as well as a body of academic writers. I do not have any real doubt as to the answer to the issue, and I do not consider it necessary to refer in order to give judgment.
  274. Conclusion

  275. For the above reasons, the application by the Defendant is dismissed.


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWHC/Comm/2019/3107.html