Brussels 1 Recast was introduced to give greater efficacy to exclusive jurisdiction clauses but since its enactment in 2015 practitioners and academics have sought confirmation that asymmetric jurisdiction clauses, common in many contracts including especially banking documentation, qualify as valid exclusive jurisdiction clauses for the purposes of Brussels 1 Recast. The Commercial Court handed down judgment on 3 February 2017 in Commerzbank Aktiengesellschaft v Pauline Shipping and Liquimar Tankers [2017] EWHC 161, in which Poonam Melwani QC of Quadrant Chambers instructed by Stephenson Harwood LLP on behalf of the Claimant Bank successfully argued that asymmetric clauses are indeed valid exclusive jurisdiction clauses for the purposes of Brussels 1 Recast so that the usual lis alibi pendens rules do not apply and potentially abusive litigation tactics cannot prevail.


A typical asymmetric jurisdiction clause requires one party to bring proceedings in a specified jurisdiction only but allows the other party to sue in that jurisdiction and also in any other Court of competent jurisdiction. Such clauses, providing for an exclusive jurisdiction if proceedings are brought by A but for non-exclusive jurisdiction if proceedings are brought by B, are ubiquitous in the banking world and have repeatedly been upheld by the English Courts as valid, requiring the restricted party A to sue in the specified jurisdiction alone.

However, doubt was cast on the validity of such clauses by Mme X v Societe Banque Prive Edmond de Rothschild, [2012] where the French Cour de cassation held that such an asymmetric clause was invalid for the purposes of the Brussels 1 Regulation (the precursor to Brussels 1 Recast). That decision was subject to trenchant academic criticism, described by Briggs as “French folly”, but has been effectively followed in other EU jurisdictions such as Bulgaria and Poland.

Further, even if such clauses are valid, do they qualify as exclusive jurisdiction clauses for the purposes of Brussels 1 Recast? That is a matter of real significance given the welcome changes that Brussels 1 Recast introduced, but in relation to exclusive jurisdiction clauses only.

Prior to Brussels 1 Recast, and as a result also of the much criticised decision of the ECJ in Erich Gasser GmbH v MISAT Srl [2005] QB 1, a party could commence proceedings in State A (usually merely for a declaration of nonliability), notwithstanding that the parties had agreed to the exclusive jurisdiction of State B. The subsequent commencement of proceedings by the wronged party in the agreed exclusive jurisdiction of State B was to no avail. That court was “second seized” and the lis alibi pendens provisions of the Brussels Regulation meant that State B had to simply stay its proceedings and await the decision of State A that it had no jurisdiction. If, for example, the first seized state was Italy then any such decision was likely to take years and the “Italian torpedo” as it became known, had struck again.

The strong representations to the ECJ in the Gasser case, including on behalf of the United Kingdom Government, that this should not be allowed and that the Brussels Regulation should be construed so as to require the lis alibi pendens regime to yield to exclusive jurisdiction clauses, fell on deaf ears and much of the legal world was in uproar, with the Court of Appeal decision in Continental Bank NA v Aeakos Compania Naviera SA [1994] QB 588 being effectively overruled.

Fast forward a few years and amendments were proposed specifically to deal with the problem. Thus Brussels 1 Recast came into force on 10 January 2015 expressly explaining in Recital 22 that an exception to the general lis pendens rule had been introduced at Article 31 in order to enhance the effectiveness of exclusive choice of court agreements and avoid abusive litigation tactics. So far, so good. But is an asymmetric jurisdiction clause an exclusive jurisdiction clause for these purposes?

The Facts

The facts of this Commerzbank case mirrored those in the Continental Bank case. An asymmetric jurisdiction clause was agreed between a Bank and Borrower/Guarantor, with the Borrower/Guarantor being restricted to sue in England only. Notwithstanding that, the Borrower/Guarantor commenced proceedings in Greece claiming a declaration of non-liability and claiming damages against the bank for allegedly acting contrary to business morality pursuant to Article 919 of the Greek Civil Code.

Commerzbank subsequently commenced proceedings in England, relying on the new provisions of Brussels 1 Recast, and the Borrower/Guarantors sought to set aside/stay those proceedings on the basis that England was second seized and that the asymmetric jurisdiction clauses did not trigger any different result.

The Potential Problems

Brussels 1 Recast, whilst referring to exclusive jurisdiction and exclusive choice-of-court agreements, contains no definition of these phrases. Further, there is nothing in the wording of the Recast or in the Recitals about asymmetric jurisdiction clauses, notwithstanding their long standing and wide-spread use. The Defendants argued that an asymmetric clause could not be said to be exclusive – it had elements of exclusivity but also of nonexclusivity – and that the wording of Article 31 of Brussels 1 Recast was inapt to encompass asymmetric jurisdiction clauses.

Moreover, the Convention of 30 June 2005 on Choice of Court Agreements developed by the Hague Conference on Private International Law (“the Hague Convention”) had been concluded in 2005 but had not yet been enacted when the Recast was being considered. The European Commission Proposal of 2012 regarding the Recast explained that another aim of the Recast was to bring its wording closer to that employed in the Hague Convention and so as to facilitate the possible conclusion of the Hague Convention by the EU. And the Hague Convention, having laid unused for a decade, then did come into force on 1 October 2015. However, asymmetric jurisdiction clauses are apparently not exclusive choice of court agreements for the purposes of the Hague Convention. The explanatory report to the Hague Convention specifically says so. So is an asymmetric jurisdiction clause exclusive for the purposes of Brussels 1 Recast, but not exclusive for the purposes of the Hague Convention?

The Decision

In a robust and clear decision, Mr Justice Cranston has held that asymmetric jurisdiction clauses are valid clauses under Brussels 1 Recast, notwithstanding the Mme X case, which he held was based on the French concept of potestative, and not any autonomous concept in EU law. He accepted the Bank’s argument that the original Brussels Convention 1968 specifically recognised asymmetric jurisdiction clauses (see Article 17 penultimate line) and held that, “it would need a strong indication that Brussels 1 Recast somehow renders what is a regular feature of financial documentation in the EU ineffective.”

And on the meatier point as to exclusivity, he held that an asymmetric jurisdiction clause did qualify as an exclusive jurisdiction clause for the purposes of Brussels 1 Recast. He agreed with the Defendants that whether an asymmetric jurisdiction clause can be characterised as conferring exclusive jurisdiction within the terms of Article 31(2) was not a question of the governing English Law but was a question of autonomous interpretation of the Recast. Thus, English authorities were not binding.

However, as a matter of autonomous interpretation, such clauses are exclusive. The Judge held that such clauses confer exclusive jurisdiction on the Courts of an EU member state, in this case England, and the fact that this applies in respect of a claim brought by one party alone does not detract from that analysis. The Judge found support for the analysis in the case of Nikolaus Meeth v Glacetal Sarl cited by the Bank. There the French party was bound to sue in Germany alone and the German party was bound to sue in France alone. The EJC held that such a clause was valid and provided for the exclusive jurisdiction of Germany in proceedings brought by the French party and the exclusive jurisdiction of France in proceedings brought by the German party. Thus, as the judge held, an exclusive jurisdiction clause did not require the designation of the courts of one state only and it was legitimate to consider exclusivity from the perspective of the parties separately.

Further, the policy and aims of Brussels 1 Recast supported such an analysis. Party autonomy is a specific aim outlined in Recital 19 and the parties here had provided for the Defendants to sue in England alone. Of even more significance was the specific background to the new Article 31 and the aims set out in Recital 22; namely that 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. For asymmetric clauses to be treated as non-exclusive would be to undermine the agreements of the parties and foster abusive tactics.

As for the Hague Convention, the Judge agreed with the Bank that it did not detract from the proper analysis of Brussels 1 Recast. Firstly, whilst the two instruments overlap, there are in any event divergences between them - such as the formal requirements for exclusive jurisdiction clauses. Secondly, there is a definition of an exclusive jurisdiction clause in the Hague Convention where none appears in Brussels 1 Recast. The reporters of the Hague Convention had recorded that the Diplomatic Session adopting the Hague Convention had decided that the definition did not extend to asymmetric jurisdiction clauses but the earlier reports and discussions cited to the Court had shown that that was not something the reporters had themselves regarded as clear. The Judge’s provisional view was that on its face the definition in the Hague Convention did indeed cover asymmetric jurisdiction. However, there was no need to decide that definitively. Whatever the position under the Hague Convention, characterisation under Article 31 of Brussels 1 Recast was a different issue.

Further, it was irrelevant that the English actions were not brought pursuant to any exclusive jurisdiction agreement or that the exclusive element of the clause governs only proceedings brought by the Defendants. The clause confers exclusive jurisdiction on the English Court when the Defendants sue; but they instituted proceedings elsewhere, and that is why the English actions were being pursued.

The judgment goes on to discuss and dismiss other points raised by the Defendants, including as to alleged anomalies and an argument that whilst Article 31 enables the English Court to decide whether it has jurisdiction, the English Court is nonetheless then required to stay its proceedings until the foreign Court makes a ruling. There is also useful discussion of the principles applicable to a stay application under Article 30.

One practical point should be borne in mind. It has always been the case that the Bank’s freedom to sue in any court of competent jurisdiction other than the designated court might be restricted by the Borrower. If proceedings were commenced first by the Borrower in the court designated as exclusive for him, then the Bank could not challenge that jurisdiction since it had agreed to it. (Lornamead Acquisitions v Kaupthing Bank [2011] EWHC 2611 at 112.) It may be that the position now is that this result ensues even if the Bank sues in a competent jurisdiction and the Borrower then commences proceedings in the designated exclusive court second. However, as the judge said, it would be even more anomalous if, by starting proceedings other than in the designated court, the borrower could cause proceedings by the Bank in the designated jurisdiction, to be stayed.