The Court of Justice on the Subrogation of Forum Selection Clauses
Whether a third party remains bound by a forum selection clause after acquiring a bill of lading is a matter for the court handling the case, governed by its private international law. Therefore, the ‘new’ conflict-of-laws rule in Article 25 of the Brussels I bis Regulation, whereby the court of a Member State, chosen by the parties, has jurisdiction based on the parties’ agreement ‘unless the agreement is null and void as to its substantive validity under the law of that Member State’, is not applicable. However, the principle of primacy dictates that national laws cannot contradict EU law. Therefore, Spanish special requirements for the subrogation of a jurisdiction clause are not valid.
This was the view of the CJEU in its judgment in the joined cases Maersk, C-345/22 et al, EU:C:2024:349 between Spanish insurance companies and defending maritime transport companies that objected to Spanish jurisdiction referring to jurisdiction clauses pointing out the UK.
Background
In three separate Spanish cases involving disputes between foreign maritime transport companies and Spanish insurance companies, similar issues regarding the validity of jurisdiction clauses arose.
Common to all cases was that the insurance companies had claims against the maritime companies for goods allegedly damaged during transport. As the insurance companies had not originally ordered the goods, but were successors in law of the claims, the main issue that arose in the disputes was whether the jurisdiction clauses entered in the original bills of lading were binding for the insurance companies. These clauses all designated the High Court of Justice of London. Therefore, the transport companies objected to Spanish jurisdiction meaning that the forum selection clauses should have derogative effect for Spanish courts.
To decide the issue of jurisdiction, the Spanish court, where all three cases were filed, referred questions to the CJEU on the meaning of jurisdiction clauses under the Brussels I bis Regulation.
Judgment
The CJEU began by addressing the issue of the applicability of the Brussels I bis Regulation. Although the UK has left the EU, the cases were initiated in the transition period contemplated by the Withdrawal Agreement, during which EU law continued to apply. The Court accordingly found that the Regulation was applicable.
Regarding the substantive issue on the subrogation of a jurisdiction clause, the CJEU clarified that the conflicts-of-laws rule in Article 25(1) of the Brussels I bis Regulation does not cover third-party subrogation matters. Said provision prescribes that the ‘substantive validity’ of a jurisdiction clause shall be governed by the designated country’s law, but issues related to a third party’s adherence to such clauses do not fall within its purview.
Having held that the special conflicts-of-laws rule included in Article 25 was not applicable to the issue of subrogation, the CJEU clarified that older case-law on subrogation of jurisdiction clauses still applied. Here, it must be noted that the choice-of-law rule in article 25 was a novelty when it was introduced in the Brussels I bis Regulation. The equivalent provisions in the old Brussels I Regulation and in the Brussels Convention contained no such conflict-of-laws rule. In older case-law, it has been established that the question of whether a third-party holder of a bill of lading is bound by a jurisdiction clause primarily is subject to the conflict-of-laws rules of the forum where the matter is brought. However, a harness has been put to this autonomy of the Member States so that if the governing law pointed out by the national conflict-of-laws rule does not allow for substitution, a court ‘must ascertain whether that third party has actually accepted the jurisdiction clause’ (paragraph 51 of the judgment with cited case-law).
Lastly, the CJEU dealt with how the Spanish court could fill out the gap through application of substantive law. Under Spanish law, a third-party successor of a bill of lading subrogates all rights and obligations of the original contract with one exception. A jurisdiction clause is only subrogated if it has been negotiated individually and separately with the third party. The CJEU held that such a provision would undermine Article 25 of the Brussels I bis Regulation. Such a circumvention of EU law is not compatible with the EU principle of primacy. Thus, it cannot be applied.
Comment
The Maersk judgment clarifies that third party effects are not subject to the ‘new’ conflict-of-laws rule in article 25 of the Brussels I bis Regulation. From a predictability perspective, the outcome is welcome. Through the judgment, it can be established that the choice of law rule in article 25 entails no radical change to the case law already established under article 25. Rather, the choice of law rule seems to apply only to issues of material validity that were unclear under the older regulation.
Another point worth noting is that even if Article 25 is not formally applicable, member states are not allowed to do whatever they want. This is clearly illustrated by the reasoning regarding the application of Spanish law that would not accept the subrogation of a jurisdiction clause as easily as other clauses of the contract. In my opinion, this clearly shows that the CJEU aims to do justice to the intentions of the original parties, which must have been to let a specific court adjudicate all matters.
After all, the disallowance of a jurisdiction clause could be devastating for a party that in good faith has presupposed that only a specific court will have jurisdiction. Not upholding such a cornerstone in dispute resolution could result in surprising outcomes. Ultimately, fostering predictable justice benefits all stakeholders involved in international transactions.
The Court, following the AG’s conclusions, is restrictively construing the notion of substantive validity of choice of court agreements to which the ‘new’ conflict rule applies. According to the AG, substantive validity should encompass instances of jurisdiction agreements’ invalidity because of error, misrepresentation, violence, fraud or because of the absence of authority or capacity (AG § 55). It’s not clear whether the Court follows this understanding of substantive validity. For sure it shares the AG’s view that third-party effectiveness of choice of court agreements is not an issue of substantive validity (§ 55, first sentence). It backs up then to the solution that to determine whether the choice of court clause contained in the bill of lading is binding for the third-party holder of the bill of lading, it is necessary to determine whether by purchasing the bill of lading, the third-party holder subrogated itself to the totality of the rights and obligations of one of the original parties to the contract (the shipper) under the national law applicable to the merits (§ 55, second sentence). Assuming that by “merits” the Court is referring to the assignment contract under which the shipper assigned the bill of lading to the third-party holder, it means that, to assess whether the choice of court clause circulates and binds the assignee, we have to enter into the merits and evaluate, under the law applicable to the assignment contract (Art. 14(1) Rome I) which rights and obligations of the assigned contract which includes a choice of court clause have been transferred from the assignor to the assignee. We may also have to consider whether the assigned contract can be assigned according to the law applicable to it (Art. 14(2) Rome I).
Don’t you think that this outcome is against the rule according to which a court should be able to decide whether it has jurisdiction without considering the substance of the case? And above all, isn’t this ruling against the principle of independence of choice of court agreements? If choice of court agreements “shall be treated as an agreement independent of the other terms of the contract” as required by Art. 25(5) Brussels Ibis Regulation, the fact that the rights and obligations deriving from the substantive contract have been assigned should not automatically entail that the jurisdiction clause included therein has been assigned too. In my view, both efficiency at the jurisdictional stage and respect for the independence of choice of court agreements could have been accommodated by following the approach suggested by the referring court, i.e. applying the lex fori prorogati to decide whether the third-party holder of the bill of lading was bound (only) by the choice of court clause included in the bill of lading.
Thanks for this comment Caterina. I have had to think all day to answer it. First, I think that the doctrine of separability under article 25(5) shall not be made more complex than what it is. Just like Matthias writes below in the next comment, the practical function of separating the jurisdiction clause from the rest of the contract is to avoid spill-over effects from the eventual invalidity of the main contract. I think this simple approach is well-reflected in the wording of article 25(5).
In a way, I agree with you that the court could have come to the same conclusion by characterizing the issue as regarding ‘substantive validity’ and consequently also subject to the ‘new’ conflicts-of-laws rule. If this was the first judgment on the rules, that would have been a wise solution. However, as we all know there is plenty of well-established case-law under the equivalent rules in the Brussels I Regulation and the Brussels Convention. By taking the restrictive approach that the Court did, it emphasizes, like I wrote in the post above, that the ‘new’ conflicts-of-laws rule means no radical change to well-established case-law. Therefore, I think it is good that the Court did not jettison well-established case-law, but instead confirmed that it still applies.
Thanks for the reply, Erik. I must admit that I disagree with your (and Matthias’) reading of the principle of independence of choice of court agreements. If the purpose of this principle were only to shelter the choice of court clause from the invalidity of the main contract, Art. 25(5) would be limited to its second phrase, where it specifically provides that “the validity of the agreement conferring jurisdiction cannot be contested solely on the ground that the contract is not valid”. Given that para 5 also comprises a first phrase mandating to treat the choice of court clause “as an agreement independent of the other terms of the contract”, we should read this rule as implying something more than the mere survival of the clause to the invalidity of the main contract, otherwise the first phrase would be moot. In my view, the independence of a choice of court clause incorporated in a main contract should be understood as normative independence. Since choice of court clauses are agreements with different nature, function and effects from the main contract, their vicissitudes – including third-party effectiveness – should be assessed under the law applicable to them, and not under the law applicable to the main contract. This is what happens for arbitration agreements, which are governed by the law applicable to them, which, as the case may be, could be different from the law governing the main contract (merits). Why should it be different for choice of court agreements?
As for the advantages flowing from the prosecution of the previous ECJ case law, I would object that not all that is old is good. Since the previous case law is not without inconveniences (think of the need to enter into the merits at the jurisdictional stage I mentioned before), I read the two legislative additions to Art. 25, namely the substantive validity rule and the principle of independence of choice of court agreements, as novelties which changed the status quo and justify interpretative discontinuity.
Dear Erik, A strange decision indeed! My concern is not so much about the mix of substantive law with jurisdiction – we have this in other cases as well, e.g. in Art 7(1)(a) Brussels Ibis and the Tessili case. I am also not worried about separability under Art 25(5), which really only means that the invalidity of the main contract should not spill over to the jurisdiction clause. What I find strange is rather the fact that the CJEU’ reasoning is self-contradictory. Effectively, the Court is saying (1) that the question of the effect of jurisdiction clauses is not regulated in Art 25 but must be determined under the applicable national rule and (2) that the national rule in this case is contrary to Art 25. Thus, the CJEU seems to give with one hand to national law what it takes in the same moment with the other.
Dear Matthias, thank you for your comment. I am not so sure that the reasoning is self-contradictory. Even if a certain provision is not applicable, the long arms of EU law will not let Member States do exactly what they want. The principle of primacy of EU law requires that application of national law will not undermine EU law.