ASYMMETRIC jurisdiction clauses, which confine one litigant to a single court while handing the other a passport to multiple forums, have long kept scholars sparring and courts divided. French courts have condemned such clauses; English judges, by contrast, view them as a commercially rational allocation of risk if the benefitting party is limited to courts of competent jurisdiction. Until recently, the Court of Justice of the European Union (“CJEU” or “Court”) had not considered whether this asymmetry is compatible with the jurisdictional architecture of the Brussels Ia Regulation (“BIR”).

In Società Italiana Lastre SpA v Agora SARL, Case C‑537/23, EU:C:2025:120 (“Lastre”), the CJEU confronted the issue head–on. Although the practical impact of the ruling may be limited in the UK – especially post–Brexit – it nevertheless supplies valuable comparative insight and raises several core questions of private international law.

The dispute itself was straightforward. Società Italiana Lastre SpA (“SIL”) agreed to supply materials to Agora SARL (“Agora”) for a construction project in France commissioned by two natural persons (the “Property Owners”). The contract contained a jurisdiction clause designating the court of Brescia (Italy) for any disputes, but it allowed SIL – unilaterally – to bring proceedings in “any other competent court in Italy or elsewhere” (emphasis added). When defects emerged, the property owners initiated proceedings in the Rennes Regional Court (France) against both Agora and SIL. Agora then sought to recover from SIL under a guarantee claim before the same court. SIL objected, relying on the jurisdiction clause in favour of the Brescia court. Both the trial court and the Rennes Court of Appeal dismissed SIL’s objection, holding the clause vague and unfairly one–sided, and therefore contrary to Article 25 BIR. On further appeal, the Cour de Cassation asked the CJEU (1) whether asymmetry and alleged imprecision fall under the BIR’s autonomous standards or the “substantive validity” carve–out in Article 25(1) (explained below); (2) if the former, are such clauses valid; and (3) if the latter, how to determine the governing law when several forums are named.

The CJEU endorsed an autonomous approach and accepted in principle the validity of such clauses, clarifying three issues. First, the concept of the agreement being “null and void as to its substantive validity” in Article 25(1) (the “substantive validity carve-out”) – much like the same concept in the 2005 Hague Convention on Choice of Court Agreements – refers narrowly to the agreement’s nullity in respect of general causes of nullity of a contract under the national law of the court designated by that agreement (e.g. fraud, mistake, duress, lack of capacity) (at [36]–[37]). Therefore, asymmetry and imprecision are not questions of substantive validity for the purposes of Article 25(1). Instead, they fall under the autonomous regime of the BIR (at [51]–[53]), which sets out the formal and substantive requirements that such clauses must satisfy “in addition” to the national law conditions under the substantive validity carve-out (at [35]–[36]).

Second, those autonomous standards require that a jurisdiction clause clearly identifies, at least, the objective factors for choosing the competent court (at [45]). The CJEU held that asymmetry of this kind complies with the requirement that the chosen court is identifiable with precision as Article 25(1) does not require the parties necessarily to designate the courts of a single and the same Member State (at [55]). In addition, the Court referred to the older provision of Article 17 of the Brussels Convention which allowed agreements to the benefit of one of the parties, albeit this provision had not found its way into the BIR (at [52]).

Third, an asymmetric clause is not inherently unlawful under EU law, provided it “permits a party to bring proceedings before the courts of different Member States or States that are parties to the Lugano II Convention” (at [58]). A clause, however, permitting proceedings before non-EU or non-Lugano Convention States was found to undermine legal certainty and foreseeability, and therefore “it would be contrary to the BIR” emphasis added (at [60]).

Three issues merit comment. First, while the Court did not specifically label the issue in these terms, the point of departure is characterisation of the various issues given their cross-border nature. As Merrett notes (“Future Enforcement of Asymmetric Jurisdiction Agreements” (2018) 67 I.C.L.Q. 37), the contractual validity of an agreement and the legal effects it produces are distinct enquiries. Asymmetric features in jurisdiction clauses can be examined at various levels: first, what kind of obligations are undertaken by the parties; second, whether the asymmetry in the obligations undertaken is an issue affecting the actual choice of a court; third, whether it is an issue affecting the substantive validity of the agreement at a contractual level; or finally, whether, even if it is considered valid, what legal effect it has.

The answer to each of these questions in the context of the EU jurisdiction regime is given either directly in a uniform manner by the BIR itself or by the conflict of laws rule included therein, which is itself common for all Member States. In the field of private international law, harmonised rules dealing with the issues directly in the substance – in translation from the French term (règles de droit international privé matériel), substantive private international law rules – have methodological priority to conflict of laws rules which aim at choosing among competing domestic-law rules. Such rules are also different to harmonised conflict of laws rules in a multi-lateral or multi-state context, such as the one established by the Rome I and Rome II Regulations.

Although rules on jurisdiction are distinguished from rules on choice of law, a similar approach is followed where a multi–state regime creates harmonised jurisdictional rules, including substantive private international law rules, such as the 2005 Hague Convention on Choice of Court Agreements and the BIR or the 1958 New York Convention. All these regimes include (1) harmonised rules for certain aspects of dispute resolution agreements, such as form; and (2) conflict of laws rules for other aspects, such as the substantive validity of these agreements.

It was in this context that the CJEU in Lastre had to examine the questions of asymmetry and alleged imprecision of a choice of court agreement. As discussed above, the Court held that these issues fall within the scope of the harmonised autonomous rules established by the BIR and not within the substantive validity carve-out. The rationale of the Court is clear: to avoid diverging approaches between Member States, the asymmetric features of a jurisdiction clause must be assessed against uniform standards. This, however, arguably puts the cart before the horse, as the question of the imbalance of power and the effect this has on any contract is an issue affecting the consent of the parties – an issue squarely within the null–and–void category. The opposite conclusion would not mean that the asymmetry of the clauses could not be assessed at all. The issue would be assessed by all Member States’ courts under the same national law due to the conflict of laws rule in Article 25(1).

Second, the CJEU referred to the special rules in relation to contracts with “weaker parties” which provide for special validity conditions directly under the BIR. These provisions and the protection by way of expanding the options available for the protected parties are triggered whenever there is an imbalance between the parties (at [49]). The imbalance, however, created by asymmetric clauses stems from the unequal obligations created by the parties themselves, not from the parties’ attributes or power difference. It would be clearer if the Court explained that, despite the different source of the imbalance, the effect, namely one party having more options than the other, is similar and this shows that the BIR is not opposed to such asymmetries. Furthermore, the Court’s judgment does not clearly explain how the existence of these provisions supports the conclusion that every asymmetric jurisdiction clause must be tested against autonomous standards and not against national law standards under the substantive validity carve out.

Third, the Court’s conclusion (at [60]) that a jurisdiction agreement contravenes the BIR if, and to the extent that, it designates courts from states outside the EU or the Lugano II Convention (“Third States”) was unnecessary. The CJEU and courts of EU Member States are only concerned with determining their own jurisdiction under EU rules. They have no authority regarding the jurisdictional validity of courts in Third States. As long as the court seized one of an EU Member State and competent according to either the BIR’s jurisdictional provisions or its own jurisdictional rules (see art. 6 of the BIR), that should suffice. Questions as to the existence of parallel proceedings would be dealt with under the provisions of the Regulation itself depending on whether proceedings are commenced in a court of a Member State or Third States.

In conclusion, the Court’s recipe – party autonomy on a short, Brussels‑supplied leash – blends commercial pragmatism with systemic order. Whether one regards the judgment as a triumph of certainty or a missed opportunity to address substantive inequality, it undeniably resets the terms of the debate across Europe and provides a roadmap for the next generation of jurisdiction clause litigation.