[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Etihad Airways PJSC v Flother [2020] EWCA Civ 1707 (18 December 2020) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2020/1707.html Cite as: [2020] WLR(D) 690, [2022] QB 303, [2020] EWCA Civ 1707, [2021] 2 WLR 939

[New search] [Printable PDF version] [Buy ICLR report: [2021] 2 WLR 939] [View ICLR summary: [2020] WLR(D) 690] [Buy ICLR report: [2022] QB 303] [Help]

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND & WALES
COMMERCIAL COURT (QBD)
MR JUSTICE JACOBS
[2019] EWHC 3107 (Comm)

		Royal Courts of Justice Strand, London, WC2A 2LL
		18/12/2020

Covid-19 Protocol: This judgment was handed down remotely by circulation to the parties' representatives by email, release to BAILII and publication on the Courts and Tribunals Judiciary website. The date and time for hand-down is deemed to be 10.30 a.m. on Friday 18 December 2020

Lord Justice Henderson:

Introduction

1. This appeal raises for the first time in this court an important issue of principle concerning the scope and effect of Article 31(2) of Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast) ("Brussels Recast"). The issue, shortly stated, is whether Article 31(2), on its true interpretation as a matter of EU law, applies to an agreement conferring exclusive jurisdiction on the courts of a Member State of the EU, in circumstances where the exclusive choice of court agreement applies to proceedings initiated by one party, but not (or not necessarily) to proceedings initiated by the other party. Agreements of this type are commonly referred to as "asymmetric", although that is not a legal term of art or an expression used in Brussels Recast itself.
2. The asymmetric nature of the relevant exclusive jurisdiction agreement in the present case is apparent from its express terms. The agreement is contained in clause 33 of a Facility Agreement dated 28 April 2017, and made between Air Berlin PLC ("Air Berlin") as borrower and Etihad Airways PJSC ("Etihad", the claimant and respondent to the appeal) as lender, relating to a facility for €350 million ("the Facility Agreement"). The loan facility formed part of a package of financial support provided to Air Berlin by Etihad, which had acquired a 29.21% shareholding in Air Berlin in 2011. Despite this and other measures of financial support, however, Air Berlin encountered severe operational difficulties during the summer of 2017 which had an adverse effect on its revenues, costs and liquidity. As a result, Air Berlin applied to the Berlin court to open insolvency proceedings on 15 August 2017, and on 27 October 2017 it ceased operations. The defendant and appellant, Professor Dr Lucas Flöther, is the insolvency administrator of Air Berlin.
3. Clauses 32 and 33 of the Facility Agreement provided as follows:

"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 dispute 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").

33.1.2 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."

4. The parties therefore agreed that: (a) the Facility Agreement, and all non-contractual obligations arising from or connected with it, were to be governed by English law (clause 32); (b) the courts of England were to have "exclusive jurisdiction" to settle any dispute arising out of or in connection with the Agreement (clause 33.1.1); and (c) the courts of England were "the most appropriate and convenient courts" to settle any such disputes (clause 33.1.2). Nevertheless, clause 33.1.3, which (like the whole of clause 33) was stated to be "for the benefit of [Etihad] only", reserved to Etihad the right to take proceedings relating to a dispute "in any other courts with jurisdiction", so that "[t]o the extent allowed by law, [Etihad] may take concurrent proceedings in any number of jurisdictions." The result is that Air Berlin was bound to invoke the exclusive jurisdiction of the English courts in order to settle any dispute arising from the Facility Agreement, but Etihad was free to take proceedings in any other (i.e. non-English) courts with jurisdiction. That is the nature of the asymmetry. So far as Air Berlin was concerned, it had entered into a choice of jurisdiction agreement which compelled it to litigate any disputes in England, but the same was not true of Etihad, which on the face of it reserved an unfettered freedom to "take concurrent proceedings in any number of jurisdictions."
5. It is convenient to note at this point that asymmetric jurisdiction clauses of this general type have been in widespread use for at least twenty five years, and it is common ground that they serve a legitimate commercial purpose. As Dr Louise Merrett (Reader in Private International Commercial Law at Cambridge) observed in an article in the International & Comparative Law Quarterly in 2018 (The future enforcement of asymmetric jurisdiction agreements, I.C.L.Q. 2018, 67(1), pp 37-61 at page 40), quoting various sources, such clauses "are widely used in international financial markets". Their aim is to "ensure that creditors can always litigate in a debtor's home court, or where its assets are located", and they "also seek to reassure the creditor that it can only be sued in its preferred jurisdiction". It also appears (ibid) that the form of clause with which we are concerned is probably based on a standard form jurisdiction clause in the Loan Market Association Single Currency Term Facility Agreement.
6. Under Section 9 of Chapter II of Brussels Recast, headed "Lis pendens – related actions", the basic rule in Article 29 states that:

"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.

…

3. Where the jurisdiction of the court first seised is established, any court other than the court first seised shall decline jurisdiction in favour of that court."

7. Article 31 of Brussels Recast then provides:

"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 [which deals with submission to the jurisdiction by entry of an appearance], 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."

8. It can therefore be seen that Article 31(1) reflects the general rule in Article 29(1) by giving priority to the exclusive jurisdiction of the court first seised, but this is subject to paragraphs (2) and (3), which apply "where a court of a Member State on which an agreement as referred to in Article 25 confers exclusive jurisdiction is seised". If that condition is satisfied, Article 31(2) in effect requires a moratorium to be imposed on proceedings pending in any other court by way of a stay of the proceedings "until such time as the court seised on the basis of the agreement declares that it has no jurisdiction under the agreement". Furthermore, once the designated court has established jurisdiction in accordance with the agreement, "any court of another Member State shall decline jurisdiction in favour of that court": see Article 31(3).
9. These provisions were new in Brussels Recast, and had no predecessor in the original version of the Brussels Regulation (Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, "the Brussels I Regulation"). It is common ground that the purpose of these new provisions, in general terms, was to reverse the much-criticised decision of the Court of Justice of the European Union ("the CJEU", an abbreviation which I shall also use to refer to its precursor, the European Court of Justice) in Case C – 116/02, Erich Gasser GmbH v MISAT Srl [2005] QB 1.
10. The mischief which led to the introduction of these provisions is clearly set out in recital (22) to Brussels Recast, which it is desirable to set out in full:

"(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."

The background to the present dispute

11. In order to explain the background to the present dispute, I need to refer to one further document which formed part of the financial support package provided by Etihad to Air Berlin in April 2017. The context for the support package, very briefly, was that between 2011 and 2016 Etihad had provided a combination of equity and debt finance to Air Berlin amounting to approximately €721 million. Air Berlin had a long history of financial difficulties, and in 2016 it began work on a proposed restructuring of its business. Since Air Berlin was a UK public limited company with its registered office in London, its annual financial statements had to be prepared in accordance with the Companies Act 2006. In late 2016, KPMG, Air Berlin's external auditors, identified going concern and funding as significant issues for the forthcoming audit. With a view to ensuring that KPMG would sign off on Air Berlin's financial statements at the end of April 2017 on a going concern basis, Air Berlin's management then made various requests for financial support to Etihad. During March and April 2017, KPMG identified some further requirements which would need to be satisfied before it could sign off the financial statements on a going concern basis, and negotiations continued on the form and content of the support package.
12. Etihad was unwilling to agree to all of Air Berlin's requests, and it was ultimately agreed that, in addition to the Facility Agreement and various other contractual commitments, Etihad would provide a comfort letter ("the Comfort Letter") which, as signed on 28 April 2017 and addressed to the directors of Air Berlin, said:

"For the purpose 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 support on obtaining financing for Air Berlin."

The Comfort Letter did not contain a jurisdiction clause, nor did it cross-refer to the Facility Agreement signed on the same date. On the assumption that the Comfort Letter was intended to be governed by English law, it is clearly well arguable that Etihad did not thereby bind itself contractually to continue to provide the necessary support to Air Berlin.

13. On 24 July 2018, some nine months after Air Berlin had ceased operations, Professor Flöther in his capacity as the insolvency administrator of Air Berlin began proceedings against Etihad in the Regional Court of Berlin ("the German proceedings"). The claims made in the German proceedings relate to the Comfort Letter and advance two alternative claims against Etihad under German law. Those claims were described by Jacobs J in the judgment under appeal ([2019] EWHC 3107 (Comm)) at [3], as follows:

"(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."

The claims were set out in a lengthy document of 232 paragraphs, ending with a preliminary estimate that the amount in dispute was no less than €2 billion.

14. Some six months later, on 22 January 2019, Etihad began the present English proceedings ("the English proceedings") by the issue of a claim form under Part 8 of the Civil Procedure Rules. Professor Flöther was named as the sole defendant. The English claim sought declaratory relief to the effect that:

(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 Brussels Recast, because, on its true construction, they are within the scope of the exclusive jurisdiction clause contained in the 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 Facility Agreement;

(c) Etihad 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) Etihad 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, Etihad is not liable to Air Berlin as alleged by Air Berlin in the German proceedings.

15. It is common ground that, in substance, the English proceedings seek negative relief which mirrors the positive relief sought in the German proceedings. It is also common ground that, for the purposes of Brussels Recast, the Berlin court is the court first seised, and the English court is the court second seised.
16. On 10 April 2019, Air Berlin made an application disputing the jurisdiction of the English Commercial Court in respect of the English proceedings. The application was heard by Jacobs J on 22 and 23 October 2019. On 18 November 2019, the judge handed down his reserved judgment giving his reasons for dismissing Air Berlin's application. In his judgment, after setting out the background facts and the legal framework, the judge began by considering the parties' submissions on the scope of the jurisdiction clause in the Facility Agreement, and the applicability of Article 25 of Brussels Recast, before turning to their submissions concerning Article 31 and the question of a stay of the English proceedings. As the judge observed, at [29]:

"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."

17. The judge's discussion of the scope of the jurisdiction agreement in the Facility Agreement under English law runs from [37] to [109] of his judgment. It is enough for present purposes to quote his conclusion at [109]:

"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."

18. The judge then discussed Article 25, and the need for Etihad to establish a good arguable case that the requirement under the article for the dispute to arise "in connection with a particular legal relationship" was fulfilled. He concluded at [156] that this requirement was satisfied, both "on the basis that there is a good arguable case that the dispute concerning the Comfort Letter originates from the relationship of lender/borrower" (see [147]) and for other reasons which he discussed at [148] and following.
19. The part of the judgment with which we are now directly concerned is the judge's discussion of Article 31(2) of Brussels Recast, which runs from [157] to [220]. I will not attempt to summarise the judge's detailed reasoning at this stage, but I will of course return to it in my discussion of the rival arguments.
20. By his order dated 18 November 2019, Jacobs J granted Air Berlin permission to appeal "on the question of whether this court is obliged to stay its proceedings under Article 29 of the Brussels Recast or whether the provisions of Article 31(2) of Brussels Recast apply to the asymmetric jurisdiction clause in the Facility Agreement". He refused Air Berlin permission to appeal on any other issues.
21. Air Berlin subsequently sought permission to appeal from this court on eight grounds relating to "the scope point", concerning both the judge's application of the EU law on Article 25 and his conclusions based on English law that the dispute arising under the Comfort Letter fell within the scope of the jurisdiction clause in the Facility Agreement. This application was refused on paper by Flaux LJ on 21 January 2020. The written reasons given by Flaux LJ included these:

"3. The Comfort Letter was clearly part of the Support Package with the Facility Agreement and the judge was clearly right to conclude as he did that the dispute here has arisen from the legal relationship in connection with which the jurisdiction agreement in the Facility Agreement was concluded. The contentions in grounds 6 and 7 that he was wrong to do so are unarguable.

…

7. On any view the judge was right to conclude that there was a good arguable case that the Comfort Letter was governed by English law and that as a matter of English law it was not binding so that it could be viewed as ancillary to the Facility Agreement, from which it followed that the dispute under it was within the scope of the jurisdiction clause. The applicant has no real prospect of persuading this Court to the contrary.

8. Despite the ingenuity of the arguments in the applicant's Skeleton Argument, the applicant has no real prospect of success on grounds 2 to 9."

22. It follows that the only ground of appeal which Air Berlin has permission to pursue in this court is ground 1, which says:

"The Judge was wrong in law to conclude that the English court is not obliged to stay its proceedings under Article 29 of the Brussels Recast and erred in concluding that the provisions of Article 31(2) of Brussels Recast apply to the asymmetric clause in the Facility Agreement."

On that issue, we have had the benefit of clear and helpful written submissions on both sides, and excellent oral arguments from leading counsel (Mr David Joseph QC for Air Berlin, and Mr Robin Dicker QC for Etihad).

23. Meanwhile, on 23 January 2019, the day after the issue of the English proceedings, Etihad applied to the Berlin District Court seeking a stay of the German proceedings pursuant to Article 31(2). That application had not been determined by the date of the hearing below, but on 13 May 2020 the court issued a decision (by Presiding Judge Meder) staying the German proceedings pursuant to Article 31(2) "until the High Court of Justice, London, England, has given a final decision on its jurisdiction." The court found that Etihad's application for a stay of proceedings was well-founded, "because the conditions for such a stay are fulfilled in accordance with Articles 31(2) and 25 of the Brussels Recast Regulation." The court also rejected various arguments based on the terms of the Withdrawal Agreement between the European Union and the United Kingdom.
24. By a subsequent order made on 10 July 2020, the Berlin District Court rejected an application to reconsider its decision, and referred the matter to the Berlin appeal court for decision.
25. Since the date of the hearing before us, the Berlin appeal court has given its decision on 3 December 2020, dismissing Air Berlin's appeal against the stay of the German proceedings. We have been provided with an unagreed translation of the German judgment prepared by Etihad's solicitors, Shearman & Sterling LLP, from which it appears that the court refused a proposal by Air Berlin to refer the dispute to the CJEU for a preliminary ruling, but granted Air Berlin permission to appeal to the Federal Court of Justice (Germany's highest court for civil matters) "[i]n view of the fundamental importance of the legal issues raised".

The factual background

26. I have already given a sufficient description of the background facts which are necessary for an understanding of the legal issues which arise on the appeal. In doing so, I have gratefully drawn on the fuller description given by the judge at [6] to [25].
27. Although there was some written evidence from each side before the judge, he recorded at [25] that:

"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."

Legislation

(1) Brussels Recast

28. I have already set out the key provisions of Articles 29 and 31, as well as recital (22). It remains for me to refer to some of the other recitals upon which reliance was placed in argument, and to say a little about the general structure of the provisions relating to jurisdiction contained in Chapter II (comprised in ten Sections running from Article 4 to Article 35).
29. Recitals (14) and (15) read as follows:

"(14) A defendant not domiciled in a Member State should in general be subject to the national rules of jurisdiction applicable in the territory of the Member State of the court seised.

However, in order to ensure the protection of consumers and employees, to safeguard the jurisdiction of the courts of the Member States in situations where they have exclusive jurisdiction and to respect the autonomy of the parties, certain rules of jurisdiction in this Regulation should apply regardless of the defendant's domicile.

(15) The rules of jurisdiction should be highly predictable and founded on the principle that jurisdiction is generally based on the defendant's domicile. Jurisdiction should always be available on this ground save in a few well-defined situations in which the subject-matter of the dispute or the autonomy of the parties warrants a different connecting factor. The domicile of a legal person must be defined autonomously so as to make the common rules more transparent and avoid conflicts of jurisdiction."

30. The emphasis on "the autonomy of the parties" in recitals (14) and (15) is repeated in recital (19), which states that:

"(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."

The "exclusive grounds of jurisdiction" to which reference is made in this recital are those laid down in Article 24, and include, for example, proceedings which have as their object rights in rem relating to immovable property and proceedings concerned with the registration or validity of patents and other forms of intellectual property. None of the exclusive grounds has any bearing on the present case.

31. Recital (34) refers to the need to ensure continuity between the 1968 Brussels Convention, the Brussels I Regulation and Brussels Recast, and further states that:

"The same need for continuity applies as regards the interpretation by the [CJEU] of the 1968 Brussels Convention and of the Regulations replacing it."

32. The general structure of Chapter II of Brussels Recast may be summarised as follows. Article 4(1) states the general principle that:

"Subject to this Regulation, persons domiciled in a Member State shall, whatever their nationality, be sued in the courts of that Member State."

Article 5(1) then provides that:

"Persons domiciled in a Member State may be sued in the courts of another Member State only by virtue of the rules set out in Sections 2 to 7 of this Chapter."

33. Section 2 (Articles 7 to 9) then sets out certain exceptions under the heading "Special jurisdiction", whereby a person domiciled in a Member Sate may be sued in another Member State. Again, none of those exceptions is relevant. The same is true of the detailed provisions made in Section 3 ("Jurisdiction in matters relating to insurance"), Section 4 ("Jurisdiction over consumer contracts") and Section 5 ("Jurisdiction over individual contracts of employment"). Section 6, as I have already said, deals with "Exclusive jurisdiction" in Article 24.
34. Section 7 is headed "Prorogation of jurisdiction" and consists of two Articles, Article 25 and Article 26. To an English lawyer, the use of the term "prorogation" in this context may not be familiar, so I take the opportunity to reproduce an editorial footnote in the report of Case 22/85, Rudolf Anterist v Crédit Lyonnais, [1987] 1 C.M.L.R 333 at 343:

" "Prorogation" is a term of Scots law used to translate, accurately, the same French term. It means "the extension of the jurisdiction of a judge or court to cases which do not properly come within it." "

35. Article 25(1) is one of the key provisions in the present case, but I have not yet set it out. It provides:

"(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. The agreement conferring jurisdiction shall be either:

(a) in writing or evidenced in writing;

(b) in a form which accords with practices which the parties have established between themselves; or

(c) in international trade or commerce, in a form which accords with a usage of which the parties are or ought to have been aware and which in such trade or commerce is widely known to, and regularly observed by, parties to contracts of the type involved in the particular trade or commerce concerned."

It is also worth noting Article 25(5), which emphasises the independent nature and validity of agreements conferring jurisdiction:

"(5) An agreement conferring jurisdiction which forms part of a contract shall be treated as an agreement independent of the other terms of the contract.

The validity of the agreement conferring jurisdiction cannot be contested solely on the ground that the contract is not valid."

36. As I have already explained, it is now common ground that the agreement on jurisdiction contained in clause 33 of the Facility Agreement falls within the scope of Article 25(1), and is therefore "an agreement as referred to in Article 25" for the purposes of Article 31(2). In the context of Article 31(2), the agreement must confer "exclusive jurisdiction", but that is not a requirement of Article 25(1), because the words "Such jurisdiction shall be exclusive unless the parties have agreed otherwise" (my emphasis) make it clear that the agreement need not be one which confers exclusive jurisdiction.
37. The remaining provisions of Chapter II do not call for specific comment, apart from the provisions of Articles 29 and 31.

(2) The Hague Convention on Choice of Court Agreements

38. Unlike Brussels Recast, the Hague Convention on Choice of Courts Agreements (concluded on 30 June 2005) is not an instrument of EU law. It is an international treaty, to which the EU is itself a party. The EU's membership obviated the need for individual Member States to accede to the Convention, so the United Kingdom has not hitherto been a party in its own right. It will, however, become a party in its own right on 1 January 2021, following the UK's departure from the EU and the end of the transitional implementation period on 31 December 2020. The necessary steps have been taken to ensure that the UK will become a party with effect from 1 January 2021.
39. Since the Hague 2005 Convention (as I shall call it) features significantly in some of the arguments on the appeal, I shall refer here to some of its main provisions.
40. The scope of the Hague 2005 Convention is defined by Article 1(1), which states that:

"This Convention shall apply in international cases to exclusive choice of court agreements concluded in civil or commercial matters."

The expression "exclusive choice of court agreement" is then defined in Article 3(a), which says that it means:

"an agreement concluded by two or more parties that meets the requirements of paragraph (c) and 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."

Paragraph (c) of Article 3 requires the agreement to be concluded or documented in writing, or by certain other means of communication.

41. Article 6 is headed "Obligations of a court not chosen", and provides that:

"A court of a Contracting State other than that of the chosen court shall suspend or dismiss proceedings to which an exclusive choice of court agreement applies unless –

(a) the agreement is null and void under the law of the State of the chosen court;

(b) a party lacked the capacity to conclude the agreement under the law of the State of the court seised;

(c) giving effect to the agreement would lead to a manifest injustice or would be manifestly contrary to the public policy of the State of the court seised;

(d) for exceptional reasons beyond the control of the parties, the agreement cannot reasonably be performed; or

(e) the chosen court has decided not to hear the case."

42. Article 26 is headed "Relationship with other international instruments". Article 26(1) provides that the Convention "shall be interpreted so far as possible to be compatible with other treaties in force for Contracting States, whether concluded before or after this Convention." Article 26(6) provides that:

"This Convention shall not affect the application of the rules of a Regional Economic Integration Organisation [such as the EU] that is a Party to this Convention, whether adopted before or after this Convention –

(a) where none of the parties is resident in a Contracting State that is not a Member State of the Regional Economic Integration Organisation;

(b) as concerns the recognition or enforcement of judgments as between Member States of the Regional Economic Integration Organisation."

The effect of the latter provision, as I understand it and as Mr Joseph submitted, is that priority is afforded to the rules of Brussels Recast in any case where both parties are resident in Member States of the EU, but not where only one of the parties is so resident.

43. It is not suggested that the Hague 2005 Convention has any direct application in the present case. Etihad is a company incorporated in Abu Dhabi, United Arab Emirates, which is not a signatory to the Convention.

Air Berlin's submissions

44. In their written submissions, counsel for Air Berlin submit that the answer to the question posed by ground 1 of the grounds of appeal follows from twelve legal propositions, which not only provide the correct answer but also serve to place in context Air Berlin's criticisms of the judge's approach.
45. The twelve propositions are in substance (and omitting some points of detail) as follows:

(1) Allocation of jurisdiction for the German proceedings as between the German and English courts is a matter to be determined in accordance with the automatic and mandatory rules of Brussels Recast.

(2) Pursuant to Article 29 of Brussels Recast, 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 first seised must of its own motion stay its proceedings until such time as the court first seised is established. It is common ground that in the present case the German court was first seised.

(3) The only potentially relevant exception to this mandatory rule is that found in Article 31(2), which provides a derogation from the first seised rule, in circumstances where a court of a Member State on which an agreement as referred to in Article 25 confers exclusive jurisdiction is seised second. The fact that this is an exception to the general rule is clear from the terms of Article 31(2) and recital (22). As with all exceptions under Brussels Recast, they are to be construed narrowly. In those narrowly defined circumstances, any proceedings in a Member State court shall be stayed until the court seised on the basis of the exclusive jurisdiction agreement shall declare that it has no jurisdiction.

(4) This carve out only applies where the parties under an Article 25 agreement have conferred exclusive jurisdiction in accordance with Article 25.

(5) The nature and effect of an agreement conferring exclusive jurisdiction under Article 25 is to be interpreted autonomously under EU law, not national law. The validity or effect of such a clause under national law is not relevant. The purpose is to ensure the application of harmonised rules applied across the EU in a uniform manner to provide legal certainty.

(6) Under principles established by the CJEU, an exclusive jurisdiction agreement for the purposes of Brussels Recast is an agreement which satisfies the formal requirements of Article 25, pursuant to which the parties both confer exclusive jurisdiction on the designated court or courts of a Member State to settle any disputes which have arisen in connection with a particular relationship, and thereby exclude the heads of jurisdiction lower in the Brussels Recast hierarchy that would otherwise apply.

(7) The essential nature and effect of consensual prorogation is not a matter of contractual enforcement as between the parties, but rather is a tightly regulated procedural rule allocating jurisdiction as between courts in accordance with the framework laid down by Brussels Recast. This is reinforced by Article 25's position as a derogation from the general rules of allocation of jurisdiction in Sections 1 and 2 of Chapter II of the Regulation.

(8) The wording of Article 17 of the (original) Brussels Convention of 1968 is not reproduced in its entirety in Article 25 of Brussels Recast. The third paragraph of Article 17, which made reference to the independent possibility of a clause for the benefit of one party only, is not included in Brussels Recast, nor was it included in the Brussels I Regulation (Article 23). Instead, Article 25 and its immediate predecessor contain express provision allowing for the possibility of either exclusive or non-exclusive jurisdiction for the settlement of disputes that arise in connection with a particular relationship, according to the choice made by the parties.

(9) This settled meaning of conferral of exclusive jurisdiction is equally reflected in the definition of an exclusive choice of court agreement in the Hague 2005 Convention, to which the EU is party.

(10) Article 31(2) makes provision for a limited exception to the lis pendens rule in Article 29 and expressly applies only to clauses which confer exclusive jurisdiction, requiring all parties to bring proceedings before the designated court (or courts). The jurisdiction clause in the Facility Agreement does not do this. It is an asymmetric clause expressly designed to confer a wide range of possible jurisdictions for the benefit of Etihad. An asymmetric jurisdiction clause of this type is not an agreement conferring exclusive jurisdiction on the English Court (the court second seised) for the settlement of all disputes that may arise between Etihad and Air Berlin in connection with their particular legal relationship. Nor can it be said that the English court is "seised on the basis of" such an exclusive jurisdiction agreement as is required under Article 31(2). No court or courts are designated by the parties for the exclusive resolution of claims brought by Etihad against Air Berlin relating to the Facility Agreement. In relation to such claims, Etihad is entitled to invoke any head of jurisdiction available to it under Brussels Recast, including: (a) the place of Air Berlin's domicile, i.e. Germany (Article 4(1)); (b) the place of performance of the obligation in question (Article 7(1)(a)); and (c) non-exclusive jurisdiction in England pursuant to clause 33 of the Facility Agreement and Article 25. The English Court is seised by Etihad under Brussels Recast on the basis of its right to choose the court in which to bring proceedings.

(11) Strong support for this conclusion is found in the treatment of the same question which has arisen under the Hague 2005 Convention, given the materially similar wording of the relevant provisions and the principle that Brussels Recast and the Hague 2005 Convention are designed to operate with "maximum alignment".

(12) Finally, both Etihad and the judge placed heavy reliance on the underlying objective of Article 31(2) as leading to a conclusion that asymmetric clauses are covered by Article 31(2) and that a different solution had been reached for Brussels Recast to that concluded under the Hague 2005 Convention. Neither proposition is correct. The wording of Article 31(2) is clear and has to be applied as it stands. Further, the wording and historical development of Article 31(2) make it clear that the derogation from the first seised rule in Article 29 was designed to cover only exclusive conferral of jurisdiction, and not non-exclusive jurisdiction or asymmetric jurisdiction.

46. In the next section of their written submissions, dealing with the judge's alleged errors, counsel for Air Berlin identify the central autonomous question of EU law as being "what is meant under European law by an agreement conferring exclusive jurisdiction in connection with a particular legal relationship". Speaking for myself, I agree that this question lies at the heart of the case, and I propose to examine it by reviewing the materials placed before us by Air Berlin in support of the submission. Before doing so, however, I will deal briefly with the submission that Article 31(2) should be given a narrow construction because it constitutes an exception from the first seised rule in Article 29(1).

Should Article 31(2) be given a narrow construction as an exception to the first seised rule?

47. In support of the submission that Article 31(2) should be strictly construed, Mr Joseph referred us to Lasok and Millett, Judicial Control in the EU: procedures and principles (Richmond, 2004), at paragraph [691], where the authors say this:

"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 interest which they seek to secure; and that their scope must be determined in the light of the aims pursued by the measure containing them."

48. To similar effect, the authors say at paragraph [692]:

"Exceptions must as a general rule be strictly construed and cannot take precedence over general and unconditional rules. Exceptions to fundamental treaty provisions or other rules cannot be given a scope which would exceed their objectives."

49. Copious authority is cited in the footnotes to vouch the above general statements of principle, but to my mind they do not help in answering the prior question whether the provisions of Article 31(2), read in context, are properly to be regarded as a derogation from, or exception to, the first seised rule in Article 29(1). Since Article 29(1) begins with the words "Without prejudice to Article 31(2)", it seems to me that, if anything, Article 31(2) is expressly accorded priority over Article 29(1), and the correct approach is therefore to construe Article 31(2) in accordance with its terms and in the light of the legislative purpose reflected in recital (22). So viewed, I can see no good reason why the concept of "a court… on which an agreement… confers exclusive jurisdiction" should be construed as excluding asymmetric jurisdiction clauses, and therefore as not applying to cases where the party which invokes the jurisdiction of the court first seised is bound by an agreement on choice of jurisdiction which, so far as that party is concerned, is indeed exclusive.
50. There is nothing in the wording of Article 31(2) itself which indicates that asymmetric jurisdiction clauses fall outside its ambit, and the mischief reflected in recital (22) strongly suggests the contrary. If asymmetric agreements are not included within the ambit of Article 31(2), the path remains wide open to the kind of abusive litigation tactics which Article 31(2) was admittedly designed to counter. So far as Air Berlin is concerned, it freely entered into the choice of jurisdiction clause in the Facility Agreement which gave the English courts "exclusive jurisdiction to settle any dispute arising out of or in connection with" the Agreement. In clear breach of that agreement, Air Berlin initiated the German proceedings in Berlin. On a fair reading of Article 31(2), and subject of course to the further arguments which I have yet to consider, it seems to me quite wrong to start from a position which requires the application of Article 31(2) in such circumstances to be justified as falling within a narrow exception to a general rule to the contrary.
51. The judge dealt with this argument at [209] to [210]. After quoting from Lasok and Millett, loc. cit, he continued at [210]:

"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."

52. I respectfully agree, save that, as I have explained, I would not myself regard Article 31(2) as an exception to Article 29. On this point, I prefer the approach of Cranston J in Commerzbank AG v Liquimar Tankers Management Inc [2017] EWHC 161 (Comm), [2017] 1 WLR 3497, who began his analysis of the very similar issues which arose in that case by saying:

"62. There is no warrant, in my judgment, for giving article 29 of Brussels I 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 seised rule in the Brussels Convention and Brussels I, there is nothing in Brussels I 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."

The history of Article 25 and Air Berlin's "taxonomy" argument

53. Mr Joseph began his review of the historical European material by referring us to the history of Article 25 and relevant case law of the CJEU. He did so with the aim of establishing the thesis that a clear distinction came to be recognised in EU law between choice of jurisdiction agreements which are fully exclusive for both parties, on the one hand, and agreements which do not satisfy that test, including asymmetric agreements, on the other hand. He agreed with an observation by Newey LJ that the distinction was one of terminology, or as Mr Joseph put it "a taxonomy point", and that it did not reflect a decision in policy terms that asymmetric agreements should be treated differently from fully, or bilaterally, exclusive agreements.
54. The review began with Article 17 of the original 1968 Brussels Convention. The relevant wording of Article 17 was contained in Section 6 of the Convention under the heading "Prorogation of jurisdiction":

"Article 17

If the Parties, one or more of whom is domiciled in a Contracting State, have, by agreement in writing or by an oral agreement evidenced in writing, 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.

…

If the agreement conferring jurisdiction was concluded for the benefit of only one of the parties, that party shall retain the right to bring proceedings in any other court which has jurisdiction by virtue of this Convention."

Mr Joseph's submission was that this wording started with an unqualified concept of conferral of fully exclusive jurisdiction, and the only exception to that concept came in the third paragraph of the Article, quoted above, which permitted a carve-out of non-exclusive jurisdiction if the agreement was concluded for the benefit of one party only.

55. In this connection, Mr Joseph referred us to the decision of the CJEU in Case 24/76, Colzani v RÜWA Polstereimaschinen GmbH [1977] 1 CMLR 345, in which the Court gave the following guidance on the interpretation of Article 17 in paragraph 7 of its judgment:

"The way in which that provision is to be applied must be interpreted in the light of the effect of the conferment of jurisdiction by consent, which is to exclude both the jurisdiction determined by the general principle laid down in Article 2 and the special jurisdictions provided for in Articles 5 and 6 of the Convention. In view of the consequences that such an option may have on the position of the parties to the action, the requirements set out in Article 17 governing the validity of clauses conferring jurisdiction must be strictly construed. By making such validity subject to the existence of an "agreement" between the parties, Article 17 imposes on the court before which the matter is brought the duty of examining, first, whether the clause conferring jurisdiction upon it was in fact the subject of a consensus between the parties, which must be clearly and precisely demonstrated. The purpose of the formal requirements imposed by Article 17 is to ensure that the consensus between the parties is in fact established."

I do not find this guidance particularly helpful in the present case, because its focus seems to me to be on the need to establish the consensus between the parties, and the part played by the formal requirements of Article 17 in achieving that objective. In the present case, there is no dispute about the agreement which the parties reached, because it is clearly set out in clause 33 of the Facility Agreement.

56. The next case to which Mr Joseph took us was the Anterist case in 1986, where the CJEU gave this further guidance in paragraphs 13 to 15 of its judgment:

"13. It should be pointed out in the first place that Article 17 of the Convention… allows the parties, within the limits laid down by the second paragraph of that provision, to choose by mutual agreement a court or the courts of a Contracting State. The parties may thus confer jurisdiction on courts which would not have jurisdiction under the general or special provisions of the Convention or exclude the jurisdiction of courts which would normally have jurisdiction under those rules. According to the first paragraph of Article 17, the jurisdiction of a court or courts designated by a jurisdiction clause is exclusive, whilst the third paragraph of that article maintains the right of the party for whose benefit the clause was agreed to institute proceedings in any other court having jurisdiction under the Convention.

14. Since Article 17 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 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.

15. Clauses which expressly state the name of the party for whose benefit they were agreed and those which, whilst specifying the courts in which either party may sue the other, give one of them a wider choice of courts must be regarded as clauses whose wording shows that they were agreed for the exclusive benefit of one of the parties."

57. While Mr Joseph relies on this passage as providing support for his distinction between fully exclusive jurisdiction clauses within the first paragraph of Article 17 and other provisions for the benefit of one party only within the third paragraph, Mr Dicker on behalf of Etihad points to the importance accorded by the CJEU to the principle of party autonomy in determining the question of jurisdiction, and the need to adopt an interpretation which respects "the parties' common intention when the contract was concluded".
58. Matters rested there until the Brussels Convention was replaced by the Brussels I Regulation with effect from 1 March 2002. The introduction of the new Regulation had been preceded by the deliberations of a working party established to consider revisions of the 1968 Convention and the parallel Lugano Convention of 1988 which was concluded between the Member States of the European Community and certain Member States of the European Free Trade Association (EFTA). By April 1999, the working party had reached general agreement on a revised text for the two conventions, Brussels and Lugano. Introduction of the revised Lugano Convention was delayed, however, until 2007 because of the need to obtain a ruling from the CJEU on the question whether conclusion of the new Lugano Convention fell within the exclusive competence of the Community, or within the shared competence of the Community and the Member States.
59. In the Brussels I Regulation, the former Article 17 became Article 23. For present purposes, the important point to note is that Article 23(1) was in essentially the same form as Article 25(1) of Brussels Recast, and there was no need to replicate the third paragraph of the old Article 17 because the new text explicitly provided that "Such jurisdiction shall be exclusive unless the parties have agreed otherwise".
60. Mr Joseph took us to the Explanatory Report on the 2007 Lugano Convention by Professor Fausto Pocar, the stated purpose of which was to "clarify the meaning of the Convention and facilitate uniform application" in circumstances where the CJEU would not have jurisdiction to resolve uncertainties of interpretation that might arise in cases before national courts: see paragraph 9 of the Report. Professor Pocar had been the rapporteur of the working party to which I have already referred.
61. In paragraphs 106 and 107 of his Report, Professor Pocar discussed "The exclusive or the non-exclusive nature of the prorogation clause". He said this:

"106. The 1988 Convention lays down that a prorogation clause that meets the requirements of the Convention always confers exclusive jurisdiction on the designated court or courts. But under the laws of some of the States bound by the Convention - under English law in particular - the parties will often agree a choice of forum clause on a non-exclusive basis, leaving other courts with concurrent jurisdiction, and permitting the plaintiff to choose between several forums; and English case-law has accepted that a non-exclusive clause constitutes a valid choice of forum under the Convention. On a proposal from the United Kingdom delegation, the ad hoc working party re-examined the question of the exclusive effect of a choice of forum clause, and reached the conclusion that, since a clause conferring jurisdiction was the outcome of an agreement between the parties, there was no reason to restrict the parties' freedom by prohibiting them from agreeing in the contract between them that a non-exclusive forum should be available in addition to the forum or forums objectively available under the Convention.

A similar possibility was in fact already provided for, though within certain limits, by the 1988 Convention, Article 17(4) of which allowed a choice of forum clause to be concluded for the benefit of only one of the parties, who then retained the right to bring proceedings in any other court which had jurisdiction by virtue of the Convention, so that in that case the clause was exclusive only as far as the other party was concerned. That provision was obviously to the advantage of the stronger party in the negotiation of a contract, without producing any significant gain for international commerce. The 1988 Convention has now been amended to give general recognition to the validity of a non-exclusive choice of forum clause, and at the same time the provision in the 1988 Convention that allowed a clause to be concluded for the benefit of one party only has been deleted.

107. Article 23 does still give preference to exclusivity, saying that the agreed jurisdiction "shall be exclusive unless the parties have agreed otherwise". A choice of forum clause is therefore presumed to have exclusive effect unless a contrary intention is expressed by the parties to the contract, and not, as was initially proposed, treated as a non-exclusive clause unless the parties agree to make it exclusive."

62. Mr Joseph relied on this passage as providing an authoritative explanation of how the wording of what is now Article 25 of Brussels Recast achieved its present form. With that I can readily agree. He also submitted that the passage provides further support for his taxonomy because Professor Pocar's discussion of the English proposal shows that the working party had asymmetric jurisdiction agreements firmly in mind, and it dealt with them not by providing that they were to count as exclusive jurisdiction clauses, but rather by recognising that parties were free to enter into a non-exclusive jurisdiction agreement in the interests of party autonomy. That may be so, but I can find nothing in this passage to suggest that Professor Pocar was addressing his mind to the question whether the exclusive part of an asymmetric jurisdiction clause can properly be regarded as an exclusive agreement in relation to claims brought by the party who is bound by it and does not have the benefit of a wider choice of forum. The focus of the discussion is on the validity of the non-exclusive part of the agreement, not on the exclusive nature of the agreement for the party who has no choice of forum.
63. On this last point, Mr Dicker is in my view right to submit that some assistance can be gained from the early decision of the CJEU in Case 23/78, Nikolaus Meeth v Glacetal [1979] 1 CMLR 520. The relevant contract in that case was made between a German manufacturer of windows (Meeth) and a French supplier of glass (Glacetal). The contract was for the supply of glass by Glacetal to Meeth, and the contract provided that "if Meeth sues Glacetal the French courts alone shall have jurisdiction. If Glacetal sues Meeth the German courts alone shall have jurisdiction". In other words, each party could be sued only in the courts of its own country. Glacetal sued Meeth in Germany to obtain payment for glass which it had delivered to Meeth. Meeth wished to raise a defence of set-off relating to damage which it claimed to have suffered owing to delay or default by Glacetal in performing its obligations under the contract. The German court refused to allow this defence, on the ground that the jurisdiction clause in the contract did not permit a set-off to be claimed before the German courts. On further appeal to the Bundesgerichtshof, the latter court referred two questions of law concerning the interpretation of Article 17 to the CJEU. The first question was whether the first paragraph of Article 17 permitted a reciprocal jurisdiction clause involving the courts of two contracting States. The second question, if the agreement was valid, was whether the effect of Article 17 was to prohibit the court before which a dispute has been brought from taking into account a set-off connected with the dispute. We are not concerned with the second question, which the CJEU answered in favour of Meeth.
64. In relation to the first question, the Court said in paragraph 5 of its judgment, with reference to the wording of Article 17:

"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."

There is nothing surprising about this conclusion, but it does provide an early endorsement of the principle that the parties are free to decide which courts are to have jurisdiction to settle disputes between them.

65. Mr Dicker also relies on the views expressed by Advocate General Capotorti, who said in paragraph 2 of his Opinion:

"In my view the validity of a clause of this nature can be recognised without misgivings. Indeed, it is probable that no doubt would have arisen if the clause had been worded differently; that is, if it had prescribed that each party could be sued only before the courts of his domicile (or perhaps the State of which he is a national: in the present case the two links coincide). In those circumstances an identical intention would have been expressed in such a way as to emphasise the concern of the parties to establish a single criterion as being decisive: the domicile (or the nationality) of the defendant. Nevertheless, quite apart from the foregoing, 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 a common step but there are no grounds for considering it unlawful."

66. Drawing on that analysis, Mr Dicker submits that there is no difficulty in the present case in dividing the disputes covered by clause 33 of the Facility Agreement into two separate groups, one consisting of claims brought by Air Berlin and the other consisting of claims brought by Etihad. Claims falling within the first group are subject to the exclusive jurisdiction of the English courts, whereas claims within the second group are not.
67. Similar reasoning is also adopted in some of the academic commentaries to which we were referred. For example, Dr Merrett, in the article which I have already cited, said at pp 55-56:

"The rationale for the new rule in Article 32(1) [of Brussels Recast] is party autonomy. Recital (22) explains that the new rule was introduced "in order to enhance the effectiveness of choice-of-courts agreements and to avoid abusive litigation tactics". 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 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" they 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 other 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."

This passage was quoted by the judge at [184] of his judgment. He said that the essential point was "concisely, and in [his] view convincingly expressed" in it. I agree. The judge then went on to consider Meeth at some length, from [185] to [191], finding support in it for Etihad's case, as I have done, and as Cranston J had before him in Commerzbank at [65] to [68].

68. In summary, I remain unpersuaded that either the principles of EU law or the case-law of the CJEU require the concept of an agreement which confers exclusive jurisdiction within the meaning of Article 25(1) of Brussels Recast to be interpreted in accordance with the taxonomy propounded by Mr Joseph. On the contrary, the CJEU has laid repeated emphasis on the principle of party autonomy, and consistently with that principle I can see no difficulty in regarding a composite jurisdiction agreement such as that contained in clause 33 of the Facility Agreement as comprising (a) an exclusive agreement in relation to claims brought by Air Berlin, and (b) a separate non-exclusive agreement in relation to claims brought by Etihad. This conclusion is in my view strongly reinforced when proper weight is accorded to recital (22) and the reversal of the decision of the CJEU in Gasser. That is the aspect of the case to which I will now turn.

The reversal of the decision of the CJEU in Gasser

69. I cannot improve on the judge's concise description of the main point in issue in Gasser, and the way in which the CJEU decided it, in [194]:

"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"."

The impact of the "Italian torpedo" was exacerbated by the Court's further holding that it made no difference if the duration of the proceedings before the courts of the state first seised was excessively long: see paragraph 73 of the Court's judgment.

70. Gasser was a decision on Article 21 of the Brussels Convention, which was the ancestor of Article 29(1) of Brussels Recast. The jurisdiction agreement in issue was not an asymmetric agreement. Indeed, the existence of the alleged jurisdiction agreement in favour of the local Austrian court had not yet been determined when the reference to the CJEU was made by the Austrian appeal court.
71. It is common ground that the purpose of Article 31(2) of Brussels Recast was to reverse the decision in Gasser. That is clear from the terms of recital (22), and from other preparatory material to which we were referred. For example, in its proposal for Brussels Recast dated 16 December 2010, the European Commission identified this shortcoming in the current operation of the Brussels I Regulation:

"The efficiency of choice of court agreements needs to be improved. Currently, the Regulation obliges the court designated by the parties in a choice of court agreement to stay proceedings if another court has been seised first. This rule enables litigants acting in bad faith to delay the resolution of the dispute in the agreed forum by first seising a non-competent court. This possibility creates additional costs and delay and undermines the legal certainty and predictability of dispute resolution which choice of court agreements should bring about."

72. The Commission went on to explain that, following "extensive consultation of the interested public":

"With respect to choice of court agreements, there was a large support from stakeholders and Member States to improve the effectiveness of such agreements. Among the various ways to achieve that objective, preference was expressed for granting priority to the chosen court to decide on its jurisdiction. Such a mechanism would largely accord with the system established in the 2005 Hague Choice of Court Agreements Convention, thus ensuring a coherent approach within the Union and at international level were the Union to decide to conclude the 2005 Convention in the future."

73. Given that clear objective, the fundamental difficulty with Air Berlin's case, as it seems to me, is that on its narrow construction of Article 31(2) the job was left only half done. Although asymmetric clauses are a familiar and widely-used feature of international financing transactions, the party which had bound itself to accept the exclusive jurisdiction of a chosen court would remain free to adopt the same abusive tactics as the Italian party in Gasser, with precisely the same adverse consequences. The point was well put by the judge, when he said:

"196. 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.

197. 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."

74. As the judge went on to note at [198], Air Berlin seeks to meet this objection by suggesting that Brussels Recast represented a negotiated compromise between the Member States of the EU, not all of which take a sympathetic approach to asymmetric clauses. Like the judge, however, I find this suggestion unconvincing. As the judge said (ibid):

"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."

75. The judge's point about Article 25 is, in my view, an important one. If an agreement falls within the scope of Article 25, as it is now conceded the present agreement does, and if that agreement confers exclusive jurisdiction in relation to all claims brought by one party, I can see no difficulty in holding that the agreement also falls within Article 31(2) to the extent that it confers exclusive jurisdiction. Not only is that, to my mind, the natural reading of Article 31(2), but it is also the only way in which the mischief identified in recital (22) can be fully remedied.
76. A further point worth making in this context is that Article 25(1) itself excludes from its ambit an agreement which "is null and void as to its substantive validity" under the law of the relevant Member State. National law does therefore have a role to play in deciding whether an agreement falls within Article 25, and it remains open to a Member State which is radically opposed to asymmetric clauses on policy grounds to treat such agreements as void under its national law. Furthermore, if asymmetric jurisdiction clauses were really as unpopular across the English Channel as Mr Joseph at times appeared to suggest, one would expect to find some reflection of this in the Commission's proposal for Brussels Recast, and an explanation of why their exclusion from the ambit of Article 31(2) would not undermine "the legal certainty and predictability of dispute resolution which choice of court agreements should bring about."
77. Further support for Etihad's submissions may be found in the academic literature, a generous selection of which was included in our bundles. I have derived particular assistance from the views of Professor Richard Fentiman, not only because of his acknowledged eminence in this field, but also because his views have been expressly adopted by a Spanish court in ruling in favour of the application of Article 31(2) to an asymmetric jurisdiction clause. The question was discussed by Professor Fentiman in the second edition of his book on International Commercial Litigation (Oxford, 2015), where he said at paragraph 2.146:

"Arguably, the solution is to draw a distinction between a jurisdiction clause and the distinct agreements it may comprise. It is coherent to say that asymmetric clauses are to be classified as non-exclusive, in so far as they do not confine proceedings to a single court. However, such clauses contain separate exclusive and non-exclusive jurisdiction agreements, whereby the counterparty's agreement to sue in the designated court is exclusive, and the "beneficiary's" agreement to sue in that court is non-exclusive."

78. Returning to the same theme in a commentary included in U Magnus and P Mankowski (eds), European Commentaries on Private International Law, volume 1 (2016), at pp 751-753, Professor Fentiman said in paragraph 17:

"If X and Y agree to the jurisdiction of the English courts, for example, X alone has the right to sue in any other court of competent jurisdiction. Suppose therefore that Y launches a pre-emptive strike in Germany, and X replies by suing in England. Is Article 31(2) engaged, or does Gasser still prevent X from relying on the agreement? If such agreements are not protected by Article 31(2), there remains the potential for a party to an asymmetric agreement to disable the agreement by launching a pre-emptive strike in its preferred court. 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."

79. The Spanish case in which Professor Fentiman's views were expressly adopted was Codere SA v Perella Weinberg Partners and Others, decided in April 2016 by the Court of First Instance No 5 of Alcobendas. The dispute was between a Spanish company (Codere) and a firm based in the UK which had advised it on a restructuring (Perella). The choice of forum clause in the relevant contract was on its face framed in non-exclusive terms, but the Spanish court proceeded on the basis that, implicitly, it amounted to an exclusive English choice of jurisdiction clause in relation to proceedings brought by Codere. I can pick up the narrative from the judgment in the present case, at [165]:

"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… ."

The relevant passage in the judgment of the Spanish court is reproduced by the judge at [166].

80. As the judge records at [167], there was a sequel in this jurisdiction when the same case came before Walker J on 17 May 2016: see [2016] EWHC 1182 (Comm). The main issue debated before Walker J was whether the jurisdiction clause in issue was on its true construction exclusive as regards Codere. The judge answered that question in Codere's favour, holding that the clause was non-exclusive for both parties. However, he also briefly considered the question of what was meant by "exclusive jurisdiction" for the purposes of Article 31(2) Leading counsel for Codere, Andrew Stafford QC, did not press the argument which had been rejected by the Spanish court, and Walker J said at [18]:

"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 [a] 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."

81. The same question was then comprehensively considered, with the benefit of full argument, by Cranston J in the Commerzbank case. At [52], Cranston J accepted, rightly in my view, that "whether an asymmetric jurisdiction agreement can be characterised as conferring exclusive jurisdiction on a court of a member state within the terms of article 31(2) is a question not of English law but the autonomous interpretation of the Regulation." Cranston J's main reasons for answering this question in the affirmative were set out at [62] to [76], which were reproduced by the judge in the present case at [170]. Jacobs J rightly proceeded on the basis that, as a matter of judicial comity, he should follow the decision of another judge of first instance, unless convinced it was wrong: see [172]. Nevertheless, he clearly had no hesitation in following Cranston J's decision. As he said, at the beginning of his discussion of the rival submissions, at [183]:

"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."

The Hague 2005 Convention

82. I must now consider a separate argument upon which Mr Joseph placed considerable reliance, and which is reflected in the eleventh of his twelve propositions. The nub of the argument was that the Hague 2005 Convention was never intended to, and does not, apply to asymmetric jurisdiction clauses; that the two Conventions (Hague and Brussels Recast) operate in the same sphere, and were designed to be interpreted with the maximum alignment between them; and accordingly the concept of exclusive jurisdiction in Article 31(2) should be interpreted conformably with the Hague 2005 Convention as excluding asymmetric agreements from its scope.
83. Although this argument was persuasively advanced, I am unable to accept it. As an introduction to the subject, I will set out the reasons which Cranston J gave in the Commerzbank case for rejecting the similar submission made to him:

"71. The Hague Convention, in my view, offers no assistance in the characterisation of asymmetric jurisdiction clauses under Article 31(2) of Brussels I Recast. There is no reference to the Hague Convention in Brussels I 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 I 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 I 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 I Recast."

84. Similar conclusions were reached by Jacobs J in the present case. He discussed the question at [215] to [217], from which I cite the following extracts:

"215. 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 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":

[the definition is then set out]

…

217. Like Cranston J and Merrett [at page 58 of her article], 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."

85. I am prepared to proceed on the basis that the Hague 2005 Convention should probably be interpreted as not applying to asymmetric jurisdiction clauses, although I emphasise that it is unnecessary for us to decide that question, and I do not do so. A strong indication that this was the deliberate intention of the framers of the Convention is provided by the Explanatory Report of Professors Trevor Hartley and Masato Dogauchi, who in their discussion of asymmetric agreements said at paragraph 106:

"It was agreed by the Diplomatic Session that, in order to be covered by the Convention, the agreement must be exclusive irrespective of the party bringing the proceedings. So agreements of the kind referred to in the previous paragraph [i.e. asymmetric agreements] are not exclusive choice of court agreements for the purposes of the Convention."

86. Further support for this conclusion may be found in the Diplomatic Minutes of the Meeting of Wednesday 15 June 2005, to which we were also referred by Mr Joseph. The minutes show that a proposal by the Swiss delegate to amend the proposed definition of an "exclusive choice of court agreement" so as to make it clear that it included asymmetric jurisdiction agreement (by inserting the words "for some or all of the parties to the agreement") was debated, but found no support. The amendment was then withdrawn.
87. Despite the force of that point, and despite the general desirability (which I acknowledge) of interpreting the Hague 2005 Convention and Brussels Recast in conformity with each other where it is reasonably possible to do so, I consider that on this issue the position under Brussels Recast is clear for all the reasons which I have already given. The result may be that, in this respect, the Brussels Recast regime goes further than the Hague 2005 Convention, but that is not necessarily surprising. As Mr Dicker points out, the Hague 2005 Convention deals solely with jurisdiction agreements, whereas Brussels Recast contains a comprehensive system of allocation of jurisdiction based on various potential grounds, of which jurisdiction agreements are but one example. This means that application of Brussels Recast regularly results in the attribution of jurisdiction to the courts of more than one Member State. Furthermore, even in relation to jurisdiction agreements, the Hague 2005 Convention (on the assumption I have made) deals only with exclusive jurisdiction agreements, whereas Article 25(1) of Brussels Recast also extends, on any view, to non-exclusive jurisdiction agreements. Nor does the Hague 2005 Convention have any equivalent to the lis pendens rule in Article 29 of Brussels Recast. I agree with Mr Dicker that this is a significant point, given that Article 31(2) itself operates as a modification of the first seised rule which has no counterpart in the Hague 2005 Convention.
88. Finally, Brussels Recast nowhere refers explicitly to the Hague 2005 Convention, nor does it contain any explicit requirement that the two instruments should be interpreted consistently. There is no incompatibility in any formal sense between the two instruments, because they contain provisions which ensure that their scope is mutually exclusive: see Article 26(6) of the Hague 2005 Convention and Article 71 of Brussels Recast. So although consistency of interpretation is a desirable general objective, it cannot be elevated into a controlling principle, not least because the Hague 2005 Convention is an international agreement to be construed according to the principles of international law, and is not subject to the jurisprudence of the CJEU.

Remaining arguments

89. I hope I have now covered the main arguments advanced by Air Berlin in support of its twelve propositions, but there is one further argument which I need to consider. It forms part of the tenth proposition, and raises the question whether the English court can properly be said to be "seised on the basis of" such an exclusive jurisdiction agreement as is required under Article 31(2). The argument, in short, is that the English court is seised by Etihad under Brussels Recast on the basis of its right to choose the court in which to bring proceedings, as expressly contemplated by Article 25(1), and not on the basis of the exclusive jurisdiction agreement which required Air Berlin to take proceedings in the English courts.
90. The judge was unimpressed by this argument. He said, at [207]:

"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."

91. I respectfully agree with that analysis. I am accordingly satisfied, as were Cranston J and Jacobs J, that there is nothing in this point.

Should there be a reference to the CJEU?

92. Air Berlin invited us to make a reference to the CJEU if we had any real doubt on the critical issue of EU law which lies at the heart of the appeal. It is common ground that any reference would have to be made before the end of December 2020, so we informed the parties at the end of the hearing that, if we were minded to make a reference, we would inform them by 14 December 2020 at the latest.
93. In the event, however, I feel no real doubt about the answer to the question. For the reasons which I have given, I consider it to be acte clair that Article 31(2) of Brussels Recast applies to both the German and the English proceedings, with the consequence that this appeal must be dismissed.
94. I am reinforced in reaching this conclusion by the fact that no court considering the application of Article 31(2) to asymmetric jurisdiction agreements has yet considered it necessary to make a reference to the CJEU, although the application of the Article to such agreements has now been upheld by courts in Spain and Greece as well as by the High Court in England (in Perella Weinberg Partners v Codere by Walker J, in Commerzbank by Cranston J, and by Jacobs J in the present case) and by the Berlin appeal court in Germany (again in the present case).

Conclusion

95. I would dismiss the appeal.

Hickinbottom LJ:

96. I agree.

Newey LJ:

97. I also agree.