CONTRACTUAL jurisdiction agreements take several forms. They can be exclusive, permitting one country's courts alone to determine disputes between the parties. Or they can be non-exclusive, conferring jurisdiction on one forum, but allowing each party to sue the other in any other competent court. But where one party is commercially dominant, as in financing transactions such as loans, bonds or swaps, a third formula is common. Often the parties agree that each may sue the other in one country, but that the party providing finance may also sue the debtor in any other court of competent jurisdiction. Such unilaterally non-exclusive clauses are ubiquitous in the financial markets. They ensure that creditors can always litigate in a debtor's home court, or where its assets are located. They also contribute to the readiness of banks to provide finance, and reduce the cost of such finance to debtors, by minimising the risk that a debtor's obligations will be unenforceable.
Such agreements are valid in English law: Continental Bank NA v Aeakos SA [1994] 1 W.L.R. 588 (CA). Indeed, despite their asymmetric, optional character it is difficult to conceive how in English law their validity could be impugned, or what policy might justify doing so, given that they are invariably clearly drafted, and affect only well-advised commercial parties. In a decision of considerable importance the highest court in France has, however, taken a different view: Cass civ, 1ère, 26.9.2012, Ms X v Banque Privée Edmond de Rothschild, No 11-26.022.
In Rothschild, Ms X, a French national, brought proceedings in Paris against Banque Rothschild, a Luxembourg bank with which she had an account. The contract between them provided that the Luxembourg courts should have exclusive jurisdiction over any claims brought by Ms X, but permitted the bank to sue either in Luxembourg or in “the courts of the client's domicile or any other competent court”. Relying on the first limb of the clause, the bank argued that the French courts had no jurisdiction. Like the courts below, however, the Cour de cassation held that Ms X was free to sue in France. The jurisdiction agreement, purporting to confine her to suing in Luxembourg, was ineffective. Deploying a principle familiar in French contract law, it held that the clause was potestative, in so far as it merely granted an option to the bank to sue in Luxembourg. As such it was contrary to the objectives of Article 23 of EU Regulation 44/2001 (O.J. [2001] L 12/1), which regulates jurisdiction agreements in favour of EU national courts.
It is unclear why the court considered that such agreements are incompatible with Article 23's objectives. Possibly the mischief lies in their uncertainty, since the dominant party might choose to sue anywhere. This reflects the familiar principle that jurisdiction under the Regulation must be certain. But the court's explicit reference to the agreement's potestative character implies that such agreements may also conflict with Article 23's purposes because they impose no obligation on the dominant party.
Four aspects of the decision are notable. First, although drawing on the French law concept of potestativité (expressed in 1174 Code civil), the court did not apply French law directly. So the decision contributes nothing to the familiar controversy about the relevance of national law in determining the validity of Article 23 agreements. Secondly, although Ms X, being Rothschild's customer, was a consumer, nothing apparently turned on this fact. Indeed, the reasoning in principle extends to any transaction involving a unilateral jurisdiction agreement. Thirdly, being a decision on Article 23, as distinct from a case under French law, it is relevant (if not binding) in proceedings in any Member State concerning Article 23 – even before an English court. Fourthly, Rothschild concerned whether Ms X could litigate other than in Luxembourg, as stipulated in the agreement, not whether the bank could properly sue elsewhere. Her case concerned not the effectiveness of the bank's option, but the validity of the clause in its entirety.
Rothschild flies in the face of market practice and has caused consternation amongst practitioners. For many reasons the decision is as perplexing as it is controversial.
First, a natural reading of Article 23 would seem to demand the opposite conclusion to that reached by the court. Article 23 expressly provides that agreements within its scope are exclusive unless otherwise agreed, thereby recognising that non-exclusive jurisdiction agreements are compatible with the Regulation. But what was the agreement in Rothschild but a form of non-exclusive jurisdiction agreement?
Secondly, the decision is inconsistent with the assumptions underlying Article 23. Its predecessor, Article 17 of the 1968 Brussels Convention, expressly provided that an agreement concluded “for the benefit” of one party gave that party the option of suing in any court having jurisdiction under the Convention. Those words were omitted from Article 23, not because such agreements are objectionable, but because (it was generally assumed) the endorsement of non-exclusive jurisdiction in Article 23 made explicit reference to unilateral agreements unnecessary.
Thirdly, all non-exclusive jurisdiction agreements inevitably assume that proceedings may be brought in any competent court. Does not Article 23 therefore contemplate precisely the uncertainty as to which court will ultimately hear a case to which the Cour de cassation might have objected?
Fourthly, Article 23 aside, the Regulation clearly contemplates the uncertainty (if any) consequent upon allowing claimants a choice of venue. Manifestly, for example, a claimant may rely on subject-matter jurisdiction under Article 5 as an alternative to personal jurisdiction under Article 2.
Fifthly, the argument that jurisdictional choice generates uncertainty is misconceived. Once such a choice is exercised, and the claimant's preferred court is seised, it is clear that no other court may entertain proceedings, a result required by Article 27.
Sixthly, if the optional character of such agreements is the difficulty, it is hard to conceive how that alone could render them incompatible with Article 23. The Regulation itself contemplates that claimants may have a choice of venue, as when a choice exists between proceeding under Article 2 or Article 5. Nor is there anything in Article 23 to suggest that the asymmetric, optional character of such agreements is objectionable. Indeed, Article 23's language clearly suggests that such concerns about form are irrelevant. Article 23 expressly demands that jurisdiction agreement be in writing, or otherwise in a form evident to the parties, but says nothing more about their permitted form. Any wider objections to the form of jurisdiction agreements belong to the law of contract, not the law of jurisdiction, and are irrelevant to whether a given clause is compatible with Article 23.
Finally, given the strength of the arguments for upholding unilateral jurisdiction agreements – many of them advanced by the bank – it hard to see how the Cour de cassation could have concluded that the matter was acte clair. So why was the issue in Rothschild not referred to the CJEU?
Rothschild concerns only agreements subject to Article 23, leaving unaffected agreements to sue in a Member State neither party to which is EU domiciled. But its implications for Article 23 agreements are alarming. The decision destabilises a term ubiquitous in commercial transactions. Whenever proceedings now occur in an EU state concerning such agreements (even in English courts) Rothschild will inevitably be relied upon to disable them, if only as persuasive authority. Even if any argument based on Rothschild might fail outside France, the effect is to generate uncertainty and expense, and to make a future reference to Luxembourg likely. Again, if such agreements are invalid entirely, as Rothschild holds, neither party could rely on the agreement to found jurisdiction even in the primary court identified in the clause. Suppose that the parties to a loan agreement confer jurisdiction on the English courts, but permit the lender to sue alternatively in any other competent court. The borrower could now object even to proceedings brought by the lender in England. Moreover, the decision threatens a flurry of torpedo actions, in which counterparties, anxious to avoid a unilaterally exclusive jurisdiction agreement in favour of one court, launch pre-emptive proceedings elsewhere on Rothschild grounds. Debtors, contractually bound to litigate in England under English law, often seek to escape the pro-creditor approach of the English courts when applying English law by seeking a declaration of non-liability in countries offering them greater protection. Such pre-emptive strikes are doomed to failure if the parties have agreed to the exclusive jurisdiction of the English courts. But they become feasible if the debtor can impugn the validity of such agreements, as Rothschild now permits.
The decision in Rothschild is as insecure in principle as it is damaging in practice. Doubtless its effects can be avoided in future by redrafting the standard terms and conditions common in commercial transactions. But the effectiveness of jurisdiction clauses in numberless financial contracts is now jeopardised. And any decision which undermines a creditor's ability to enforce its rights threatens to increase the price of finance, so prejudicing those debtors whose interests the Cour de cassation apparently sought to protect.
