Jump to Content Jump to Main Navigation

6 Enforceability of asymmetric jurisdiction clauses before EU Member State courts

From: Asymmetric Jurisdiction Clauses

Brooke Marshall

From: Oxford Legal Research Library (http://olrl.ouplaw.com). (c) Oxford University Press, 2023. All Rights Reserved.date: 27 February 2024

Subject(s):
Validity and effect — Validity of contract — Jurisdictional agreement — Jurisdiction

(p. 195) Enforceability of asymmetric jurisdiction clauses before EU Member State courts

I.  Introduction

6.01  This chapter examines whether asymmetric jurisdiction agreements are enforceable under the Recast and 2007 Lugano Convention and, to the extent permitted by those instruments, prone to be constrained by the national laws of the European Union (EU) Member States on abuse of rights. A body of decisions, applying Regulation 44/2001 and the 2007 Lugano Convention,1 from France’s highest Court has precipitated the fear that they are not. That case law reveals that the Cour de cassation was not concerned in limine with whether asymmetric agreements are ‘agreements conferring jurisdiction’2 under EU law. The key issue, though laconically expressed, was whether the asymmetric agreements in those cases were compatible with the ‘requirements’3 for jurisdiction agreements to be enforceable under EU law. In particular, with the requirement that the content of a jurisdiction clause be sufficiently precise to allow a court to determine and, arguably, the parties to predict which court has jurisdiction. Put another way, the relevant question is not whether asymmetric agreements are, as (p. 196) a matter of EU law, admissible (licit) or ‘valid’ tout court.4 Instead, it is whether asymmetric agreements, in their various forms, satisfy the discrete EU law requirements for enforceability.5

6.02  By reference to those requirements, this chapter argues, in Part I, that most types of asymmetric clauses6 will, in principle, be enforceable under the Recast and 2007 Lugano Convention, though several discrete questions ought to be referred to the Court of Justice of the European Union (CJEU) to confirm that conclusion.7 As has been seen,8 the Recast will generally govern the enforceability of an asymmetric agreement which confers jurisdiction on the courts of an EU Member State. The 2007 Lugano Convention will govern the enforceability of an asymmetric agreement that confers jurisdiction on a European Free Trade Association (EFTA) Lugano State court, provided at least one party is domiciled in an EU or EFTA Lugano State.9 For proceedings initiated on or before 31 December 2020 that are ongoing, English courts are to be treated as though they are EU Member State courts under the Recast and 2007 Lugano Convention.10 For proceedings initiated after that date, English courts are to be treated as Third State courts, unless proceedings initiated before that date in the same or a related matter are already pending in an English or EU Member State court.11 It should be noted that the obligations of English courts to have regard to CJEU case law in comporting themselves (p. 197) as EU Member State courts in these transitional cases are, from the UK’s perspective, tempered.12

6.03  Overlaying the Recast and 2007 Lugano Convention is the EU prohibition on abuse of law, explored in Part II. There, it is argued that the EU prohibition can control only some rights arising out of asymmetric jurisdiction clauses. This suggests that national abuse-of-rights doctrines may play a marginal role in the rare circumstances where this is needed, redressing what may be a secondary concern of the Cour de cassation.

II.  Enforceability requirements under the Recast and 2007 Lugano Convention

6.04  This Part analyses those requirements under the Recast and 2007 Lugano Convention most prone to affect asymmetric jurisdiction agreements specifically.13 Analysis begins in Part II(A) with the CJEU’s requirement—of which the French Cour de cassation has repeatedly found asymmetric agreements to fall foul—that the courts, contemplated by a jurisdiction agreement, must be identifiable. It highlights an uncertainty, as a matter of EU law, as to whether those courts must be identifiable from the point of view of both parties as well as a court seised. It explores how different types of asymmetric clauses14 cohere with that requirement, underscoring a misapprehension on the part of the French court. Part II(B) then considers the requirement of form and evidence of consensus, which, in most cases, will be satisfied. The chapter goes on to analyse a requirement specific to the Recast, namely, that a jurisdiction clause be ‘substantively (p. 198) valid’ under its governing law, in Part II(C).15 It explores the application of that rule to asymmetric jurisdiction clauses and its implications for their enforceability. It argues that some doctrines of national law, which may apply to this question, carry the potential to redress a further concern reflected in the French case law. That is the legitimate concern that, in some cases, the asymmetric structure of these clauses reflects a gross asymmetry in the parties’ bargaining power,16 such that a normative basis for their enforcement is lacking.17

A.  ‘Precision of … content’

6.05  The first enforceability requirement which an asymmetric jurisdiction clause must satisfy is that the courts contemplated by the agreement must be objectively identifiable.18 It is uncertain, however, whether this requirement is to be assessed exclusively from the perspective of a court that is seised or also from the perspective of both parties.

1.  From the perspective of a court seised

6.06  The CJEU implied a requirement into the Brussels Convention and Regulation 44/2001 that the wording of a jurisdiction clause must specify, with sufficient precision, the objective factors that would enable a court that is seised to ascertain whether it has jurisdiction.19 In other words, whether a court has jurisdiction must be determinable.20 The CJEU has ruled that jurisdiction agreements conferring jurisdiction on the courts ‘in the country where the carrier has [its] principal place of business’21 and on ‘the Paris Courts’22 both satisfy that requirement.

6.07  Those cases show that the CJEU is concerned not only with the ability of a court, on which the clause confers jurisdiction, to determine whether it has jurisdiction in light of the clause, but also with the ability of other Member State courts seised to determine whether they have jurisdiction in spite of it. (In neither of those cases was the referring court the court on which the clause conferred jurisdiction.23) This suggests that the requirement of determinability attaches to the jurisdiction clause as a whole, not merely to the part of the clause which confers jurisdiction on one or more Member (p. 199) State courts.24 In other words, any court whose jurisdiction is prorogated or preserved by the clause25 must be identifiable for any Member State court which is seised. There is nothing to suggest that this requirement is not similarly implied into the Recast or the 2007 Lugano Convention. Under the Recast, however, it may now be for a Member State court, on which the clause apparently confers exclusive jurisdiction to assess, if it is seised after and concurrently with another Member State court.26

6.08  By reference to the CJEU’s rulings,27 a narrowly drafted Rothschild clause will be sufficiently precise if it allows the non-option holder to be sued in the named court or other identifiable court. For example, ‘jurisdiction is attributed exclusively to the courts of Passau/Germany. The supplier is, however, also entitled to bring an action against the reseller before the competent court at the place of the latter’s registered office.’28 The Civil Chamber of the French Cour de cassation upheld this clause, reasoning that ‘the courts which could be seised were precisely identified, such that the clause satisfied the objective of predictability which the Recast pursues’.29 Consistently with that reasoning, a Rothschild clause which gives the option holder the ‘right to refer to other competent courts consistent with the rules of legal procedure’ would also be sufficiently precise.30

6.09  In ICH v Crédit Suisse, however, the Civil Chamber of the Cour de cassation held that the optional aspect of a standard Rothschild clause31 ‘did not specify the objective factors on which this alternative jurisdiction was based’ and remitted the case.32 The clause provided:

The borrower acknowledges that the exclusive forum for all proceedings is Zurich or the place of the bank’s branch where the relationship between the parties was formed. The bank nonetheless reserves the right to commence proceedings against the borrower before any other court with jurisdiction.33

(p. 200) 6.10  When this case came before the Cour de cassation for the second time, the Court maintained its position.34 It is difficult to see why the criterion ‘any court with jurisdiction’ is not an objective factor or why it is insufficiently precise, assessed only from the perspective of a court seised: any court which would otherwise regard itself as competent has jurisdiction.35 Although the optional limb does not confer jurisdiction on those courts, it provides a criterion allowing any court which is seised to determine whether it is one of the courts before which the parties contemplated the option holder could bring proceedings. When a non-option holder, domiciled in a Third State, is being sued, this aspect of the clause would mean all EU Member State and EFTA Lugano State courts, which have jurisdiction according to their national—including exorbitant—rules of jurisdiction,36 and all other courts in the world, which have jurisdiction under their own rules of private international law.37 When an EU-domiciled non-option holder is being sued, ‘any court with jurisdiction’ must mean any court outside of the EU with jurisdiction under its own rules of private international law or any EU Member State or EFTA Lugano State court that has general or special jurisdiction under the Recast or 2007 Lugano Convention.38

6.11  Unilateral non-exclusive jurisdiction agreements equally satisfy the requirement of precision from the perspective of the court seised. For instance, (Codere) ‘agrees for the benefit of (Perella) that the courts of England will have non-exclusive jurisdiction to settle any dispute which may arise in connection with this engagement’.39 These clauses have the same effect as (symmetric) non-exclusive jurisdiction agreements under the Recast and 2007 Lugano Convention.40 The relevant objective factor under unilateral non-exclusive and symmetric non-exclusive jurisdiction agreements alike is the descriptor ‘non-exclusive’ or a form of words which makes clear to a court seised that it is not prevented from exercising the jurisdiction it otherwise has, even though the nominated court has jurisdiction.

6.12  More questionable are asymmetric clauses which contain subjective rather than objective factors.41 Depending on how they are interpreted, this may include broad (p. 201) variants of Rothschild clauses, the optional aspects of which require the non-option holder to defend proceedings in any court of the option holder’s choice ‘anywhere in the world’.42 At least from the perspective of French courts and the French doctrine, such clauses are null under the 2007 Lugano Convention and Recast.43 The criteria for determining which court the option holder can seise in each of these clauses are in the option holder’s mind’s eye. That said, a court examining its competence with reference to a Rothschild clause of this kind could simply ask itself whether it is ‘somewhere in the world’ and, given that the answer would be ‘Yes’, could decide that it has jurisdiction.44

6.13  Unilateral indeterminate jurisdiction clauses, be they narrow45 or broad,46 also contain subjective factors. These require the non-option holder to bring or defend proceedings in a court to be chosen by the option holder after a dispute has arisen. Again, a court examining its competence with reference to a unilateral indeterminate clause could simply ask itself whether it is a court of ‘F and N respectively, I or the buyer’s place of business’ or whether it is a court ‘in England or elsewhere’ and, if the answer is ‘Yes’, decide that it has jurisdiction.

2.  From the perspective of the parties

6.14  What the foregoing analysis suggests is that, from the perspective of the Cour de cassation, a jurisdiction agreement governed by the Recast or 2007 Lugano Convention must be predictable both from the perspective of a court seised and the parties.47 In other (p. 202) words, the jurisdiction clause itself must specify with sufficient precision the objective factors that would enable a court seised to ascertain whether it has jurisdiction and that would enable both parties to foresee where they will be required to bring or defend proceedings. Without having explicitly said so, it appears that this is how the Cour de cassation has understood the CJEU’s ruling in Coreck.48

6.15  The cases in which the Cour de cassation has struck down Rothschild clauses (namely, the Rothschild case, ICH v Crédit Suisse, ICH v Crédit Suisse II, Dexia49) and the cases in which it has upheld them (Diemme,50 Ebizcuss51) are all consistent with that view. It is telling that in all these cases52 the Court tethered the requirement of objective factors to the aim of legal certainty for litigants, realized through predictable jurisdictional rules.53 It is also clearly expressed in a 2014 decision of the Cour d’appel d’Aix-en-Provence concerning a narrow Rothschild clause which provided for the option holder to sue in Antwerp or ‘in any other jurisdiction in which the “merchant” has assets and/or a domicile’.54 The Court reasoned that ‘the choice given to [the option holder] is clearly limited to defined jurisdictions known to its contracting partner … and constitutes an element of legal certainty’.55 In the case, however, of a standard Rothschild clause which ‘effectively incorporate[s]’ the default rules of jurisdiction in the optional limb,56 the words ‘any competent court’ will be insufficiently precise—what the Cour de cassation appears to require is that the clause specify expressly that the court’s competence is ‘based on procedural rules’57 or ‘an ascertainable rule of domestic or international law’58 or ‘a rule of jurisdiction applicable in a Member State’59.60

(p. 203) 6.16  As a matter of EU law, whether it is a requirement that both parties be able to ascertain from the objective factors on the face of the clause where they must bring or defend proceedings is not an acte clair. On one view, perhaps the CJEU has not said that the parties must be able to ascertain where they can sue and be sued because party autonomy is the prevailing legal value.61 If the non-option holder agrees to jurisdictional uncertainty by placing unilateral control in the hands of its counterparty,62 article 25 requires that it be enforced. The CJEU’s statement that ‘the concept of a “jurisdiction clause” must be interpreted as … giving full effect to the principle of freedom of choice on which [a]rticle 25(1) of Regulation No 1215/2012 is based’63 supports that view.

6.17  On another view, it would be surprising if it were not a requirement that the parties must be able to foresee where they can sue and be sued: the CJEU has also said that which ‘underpins’ the predecessor to article 25 of the Recast is ‘the objective of foreseeability’.64 As Advocate General Jääskinen observed, a ‘teleological interpretation’ of that provision ‘seeks to ensure the foreseeability of jurisdiction … and therefore the legal certainty of the parties to a dispute, in particular where clauses conferring jurisdiction are concerned’.65 Those observations suggest that the objective of legal certainty (in the sense of foreseeability or predictability)66 necessitates that a ‘well-informed [would-be] defendant’67 be able to foresee, at the time it agrees to it, from a jurisdiction agreement where it can be sued,68 just as much as it necessitates that a prospective defendant be able ‘reasonably to foresee’ where it may be sued under the default rules.69 It follows that if that objective is to be taken seriously, the only reasonable answer to the question of whether a jurisdiction clause must specify with sufficient precision the objective factors that would enable both parties to foresee where they must bring or defend proceedings is ‘Yes’. What, then, is the upshot for asymmetric clauses?

6.18  It seems that broad variants of Rothschild clauses, if they are interpreted as conferring jurisdiction on ‘any court’ which the option holder may choose, will be imprecise and, therefore, unenforceable.70

(p. 204) 6.19  Conversely, a narrow Rothschild clause providing for the option holder to bring proceedings in ‘Passau’ or the place of non-option holder’s ‘registered office’ is sufficiently precise.71 The identity of those places is within the non-option holder’s control and any change to the identity of those places between the time of contracting and the time of a dispute is equally within the control of the non-option holder.

6.20  Those elements are more foreseeable for the non-option holder than a standard Rothschild clause which allows it to be sued in ‘any competent court’—whether a court considers itself to be competent is not within the non-option holder’s control.72 But can it be said that a standard Rothschild clause, allowing a non-option holder to be sued before ‘any competent court’, is not reasonably foreseeable from the perspective of a well-informed defendant? To the extent that such a clause simply incorporates the rules of special and general jurisdiction, that clause will, for an EU-domiciled non-option holder, allow it to foresee where it can be sued; concluding otherwise would mean that the default rules themselves do not allow for that.73 In other words, impugning the clause would mean impugning the rules, which it seems most improbable the CJEU would do. That is especially so since impugning the clause means that those rules apply anyway.74 Granted, the rules of special and general jurisdiction only apply to EU-domiciled defendants.75 It might, therefore, be thought that where the defendant is domiciled in a Third State, such a clause is insufficiently certain, in so far as it exposes the non-option holder to the national, including exorbitant, rules of jurisdiction of the Member States.76 Again, though, that uncertainty is first and foremost a product of the EU law instruments, as opposed to the clause, which show little concern for the plight of non-EU defendants.77

6.21  A possible problem for standard Rothschild clauses, both where non-EU and EU-domiciled non-option holders are concerned, is that the phrase ‘any competent court’ necessarily also encompasses any court competent in a Third State. By reference to that phrase, courts of many Third States would likely assert jurisdiction.78 The paradigm in which they would be less inclined to assert jurisdiction is if the clause were to confer exclusive jurisdiction on a Member State, which a Rothschild clause does not do, so far as proceedings against the non-option holder are concerned.79 The CJEU could consider it to be against the interests of EU-domiciled defendants for a Rothschild clause to extend to the courts of Third States and thus consider the clause to be insufficiently (p. 205) imprecise on that basis. But neither the CJEU nor a Member State court has control over whether Third States would assert jurisdiction over a non-option holder, just as neither has control over whether a Third State court would assert jurisdiction over a defendant under a non-exclusive jurisdiction clause or over a defendant in the absence of any jurisdiction clause at all.

6.22  Since it is uncertain what the CJEU may say about this state of affairs, it might be thought prudent for drafters to specify that ‘any competent court’ is restricted to those within the EU, if that is their intention. Most cases in which the optional part of the clause has been exercised, however, often involve Third States,80 suggesting that it is not the parties’ intention to restrict the option to the EU and that such a change would diminish the practical utility of these clauses.

3.  Conclusion

6.23  In sum, all types of asymmetric clauses are, as a matter of EU law, compatible with the requirement of precision of content, examined purely from the perspective of a court seised.

6.24  If, however, EU law also requires that the parties be able to foresee from a jurisdiction clause where they must defend proceedings, then broad variants of Rothschild clauses81 and unilateral indeterminate jurisdiction clauses82 which contain ‘subjective factors’ will fail. Whether that is a requirement is a question which must be put to the CJEU. The position as concerns standard Rothschild clauses is more complex. French courts will probably consider a standard Rothschild clause, providing for the option holder to seise ‘any competent court’, to be insufficiently precise from the parties’ perspectives.83 If that clause is interpreted as being limited to EU Member State or EFTA Lugano State courts, then the French position can only be right to the extent that the underlying rules of the Recast are themselves insufficiently precise. It is improbable that the CJEU will say that they are.

B.  Form and evidence of factual consensus

6.25  The Recast and 2007 Lugano Convention both require a jurisdiction agreement84 to satisfy two distinct but related elements to be (p. 206) enforceable:85 satisfaction of a form requirement86 and consensus in fact.87 As to the first, a jurisdiction clause must satisfy one of three alternative form requirements in articles 25(1)/23(1). The clause must be in writing or evidenced in writing (sub-clause (a))88 or in a form that accords with established practice, either between the parties in the past (sub-clause (b)) or within their industry (sub-clause (c)).89 ‘Any communication by electronic means which provides a durable record of the agreement shall be equivalent to “writing”.’90 Asymmetric jurisdiction clauses are always in writing, so this first element of formal validity is generally unproblematic.

6.26  Where the clause is in writing or evidenced in writing, to be formally valid, the clause must satisfy the second element.91 It must be ‘in fact the subject of consensus between the parties, which must be clearly and precisely demonstrated’.92 This means that independently of the requirement that the clause be in writing or evidenced in writing, the parties’ consent to that clause must also be in writing.93 It is the duty of the court first (p. 207) seised to determine whether ‘consensus between the parties is in fact established’.94 This is said to mean actual consensus.95 It will generally be established where the jurisdiction clause is contained in a contract which both parties have signed,96 even if that contract is entirely drafted by one party and has not been negotiated.97 Consensus will also be established where the jurisdiction clause is contained in the standard contract terms of one of the parties and the contract, signed by both parties, expressly refers to and incorporates those terms.98 This is said to ensure that the jurisdiction clause does not go unnoticed by the ‘weaker party’ when it is contained in the other party’s standard contract terms and incorporated by reference without its knowledge,99 though it is doubtful whether it achieves this aim.100 ‘Weaker party’ here is not confined only to those structurally weaker parties which the Recast and 2007 Lugano Convention specifically protect (namely, most consumers, employees, and some insureds)101 but to all parties asked to adhere to the other party’s standard contract terms.102 Consensus will not be established where the parties have concluded an oral contract with no reference to a party’s standard terms and where an invoice subsequently refers to those terms, which themselves contain a jurisdiction clause.103

6.27  Consensus is concerned with both parties’ consent, which together constitutes consensus,104 not merely with the consent of the party that seeks to contest the (p. 208) jurisdiction of the court whose jurisdiction is based on the agreement. The German Bundesgerichtshof has held that this requires both parties to have signed the contract, if the party seeking to rely on the jurisdiction clause has drafted the contract,105 but that it requires only the party contesting the jurisdiction clause to have signed the contract if the contract has been negotiated by both parties.106 The party contesting the jurisdiction clause will generally be the claimant in proceedings brought before a non-nominated court and the defendant in proceedings before the nominated court.107 This may explain what is otherwise an inconsistent use of the terms ‘consent’ and ‘consensus’ in the jurisprudence of the CJEU.

6.28  In Salotti,108 the CJEU ruled that a jurisdiction clause in the supplier’s standard contract terms, printed on the back of the contract drawn up by the supplier which did not refer to those terms, was not an agreement under the Brussels Convention. There was no ‘guarantee’ that the buyer ‘really consented to the clause’,109 even though the rest of the standard contract terms had been incorporated into the contract.110 The Court ruled that consensus would have been made out had the contract referred to the standard terms which contained the jurisdiction clause rather than referring to a letter of offer which itself referred to the standard terms containing the jurisdiction clause.111 Although apparently unnoticed in the literature and case law,112 the jurisdiction clause was a narrow variant of a Rothschild clause, which provided for both parties to bring proceedings in Cologne but gave the seller the option ‘at all times … to elect (p. 209) to commence proceedings in the buyer’s place of establishment’.113 The Court did not refer to the asymmetric nature of the clause, presumably because the seller had brought proceedings in Cologne under the mutual, anchor limb of the clause, which was governed by article 17(1) of the Brussels Convention. Nor did it refer to article 17(4) of the Brussels Convention,114 which would have applied had the seller brought proceedings pursuant to its option.

6.29  In sum, most asymmetric agreements will satisfy the Recast and 2007 Lugano Convention’s form requirement. Most asymmetric agreements will also satisfy the threshold of consensus in fact. This merely requires signature of the contract, which either contains the jurisdiction clause or incorporates the standard terms of one party that themselves contain the jurisdiction clause.

C.  Substantive validity

6.30  The Recast introduces a conflict-of-laws rule relevant to enforceability of a jurisdiction agreement. It determines the national law that governs the substantive validity of that agreement. Regulation 44/2001 and the 2007 Lugano Convention115 contain no choice-of-law rule. They simply state that a choice-of-court agreement is valid if it satisfies one of the alternative form requirements and is the subject of consensus in fact between the parties, as per section B.116 On one view, this means that those instruments isolate the substantive validity of jurisdiction agreements from any conflict-of-laws analysis, such that the application of national law is excluded.117 As the English Court of Appeal has reasoned, there does not appear to be:

any decision of the ECJ/CJEU where, for the purposes of Article 17 [of the Brussels Convention]/23 [of Regulation 44/2001], it has accepted the proposition that, in some circumstances, the court considering whether that article applies must resort to the (p. 210) applicable law of the substantive contract concerned to see whether the jurisdiction clause in it is valid.118

6.31  On another view, the silence of those instruments means that national choice-of-law rules can be implied to designate the law applicable to the substantive validity of a jurisdiction clause.119 According to that view, validity has always been subject to a conflict-of-laws analysis, and the only change which the Recast introduces is to unify the applicable choice-of-law rule. A final view is that substantive validity is under the 2007 Lugano Convention to be assessed against an autonomous definition of what constitutes an ‘agreement’.120 To the extent that this continues to be a point of uncertainty under the 2007 Lugano Convention, it is a question to be put to the CJEU.

6.32  Whatever the case was under Regulation 44/2001 and is under the 2007 Lugano Convention, the Recast differs. It introduces a choice-of-law rule subjecting the substantive validity of a jurisdiction clause to the law of the designated court. The Recast also introduces a substantive rule, which provides that a jurisdiction agreement contained in or associated with121 a contract is legally independent of that contract.122 It codifies CJEU case law,123 meaning that the same principle applies under the 2007 Lugano Convention. One consequence of that rule is that matters which impugn the contract’s substantive validity will not necessarily affect the jurisdiction agreement,124 although they may,125 and vice versa.

6.33  The analysis which follows investigates the requirement of substantive validity under the Recast and the operation of the choice-of-law rule to which it is subject. It first explores how the substantive validity rule works in the context of exclusive and non-exclusive jurisdiction agreements, the former being the paradigm for which the rule was designed (Part II(C)(1)). It then examines the rule’s application to asymmetric jurisdiction agreements (Part II(C)(2)).

(p. 211) 1.  Exclusive and non-exclusive jurisdiction clauses

6.34  The Recast provides that the agreed court or courts of a Member State shall have jurisdiction ‘unless the agreement is null and void as to its substantive validity under the law of that Member State’,126 including that Member State’s own ‘conflict-of-laws rules’.127 This choice-of-law rule is inspired by the 2005 Hague Choice of Court Convention,128 which requires a court designated in an exclusive choice-of-court agreement to exercise jurisdiction unless ‘the agreement is null and void under the law of that State’,129 including that state’s ‘choice-of-law rules’.130 This section begins by exploring the issue of characterization of substantive validity under the Recast; a question, in part, arguably determined by the juridical nature of that agreement (subsection a)). It then examines the choice-of-law rule and the extent to which the law that it designates may be affected by imperative norms (subsection b)).

a)  Characterization: What issues are covered by ‘substantive validity’?

6.35  Three uncertainties attend the process of characterization and the substantive validity rule. The first is as to which law is applicable to the interpretation of the phrase ‘null and void as to its substantive validity’. It is uncertain whether the meaning of that phrase and the scope of the issues it covers is a question to be resolved by the law of the court seised, by reference to an autonomous interpretation or by reference to the law applicable to substantive validity itself.131 Though the CJEU has not been asked to rule on the question, an autonomous approach would ensure that issues of substantive validity, governed by national law, do not overlap with other aspects of the Recast which are governed by an autonomous approach.132

6.36  The second issue is a question of interpretation of the rule itself. Is ‘null and void as to its substantive validity’ (entachée de nullité quant au fond, materiell nichtig) restricted to voidness (Anfechtung, nullité absolue) or does it also encompass voidability (nullité (p. 212) relative,133 Anfechtbarkeit134) ? Most authors favour the view that aspects of the applicable law, which render a jurisdiction agreement voidable, are also encompassed by the rule.135 Fentiman, however, argues that ‘[e]ven an autonomous interpretation of the phrase “null and void” could hardly embrace … cases’ ‘where a vitiating factor … renders an agreement voidable, not void ab initio’.136 Although all official language versions of the Recast would need to be consulted to reach a conclusive view,137 excluding voidability would mean that some issues which are clearly intended to be covered by the rule, such as duress, may go unregulated if, for instance, German law were the applicable law.138 Less clear is whether ‘null and void’ also extends to ‘unenforceability’ or ‘ineffectiveness’139 (‘réputation non écrite’140 inopposabilité, inefficacité, Unwirksamkeit).

6.37  The third, and most vexed, uncertainty which arises is as to which kinds of issues can be characterized as issues of substantive validity.141 There is a consensus in the literature that the provision encompasses issues of defective party consent142 and that it may encompass ‘capacity and authority to conclude a jurisdiction agreement’,143 although capacity is excluded from the Recast itself.144 It follows that there is equally a consensus that relevant doctrines of national law, which would apply as part of the applicable law, at least include common mistake, fraud, threat, misrepresentation,145 and (p. 213) duress146.147 Beyond that, there is a grey area, though there are signs that the CJEU may favour a broad interpretation.148 The following four subsections will precis that grey area before positing in the fifth an alternative reason as to why the relevance of other specific doctrines or provisions is less clear.

aa)  All aspects of validity, including issues of form?

6.38  There seems to be no support for the view, and sensibly so, that ‘substantive validity’ encompasses issues of form. The fact that article 25 of the Recast itself regulates formal requirements149 means that additional requirements of the applicable law as to form must be excluded.150 But, again, less clear is which law ought to characterize a requirement as formal or substantive.151

bb)  All aspects of validity, except those as to form?

6.39  Several authors appear to envisage as relevant to a jurisdiction clause’s validity any aspect of substance.152 That view is admittedly open on the wording of the rule; it does not qualify the words ‘substantive validity’ except through the descriptor ‘null and void’. It is also consistent with the way ‘substantive validity’ has been understood in the private law literature.153

cc)  All aspects of validity except those as to form or consent in fact?

6.40  Several authors154 have suggested that apart from formal requirements which are exclusively regulated by the requirements of article 25, the substantive validity rule ‘reverse[s] the [CJEU’s] case-law on the autonomous determination of whether there is in fact (p. 214) consent’.155 This means that instead of subjecting the question of whether the parties in fact consented to a jurisdiction agreement to the autonomous test discussed in Part II(B), that question is now determined by national law. Magnus, Mankowski, and Advocate General Szpunar in Hőszig156 have argued against this interpretation on the basis that the word ‘unless’ suggests that the ‘validity’ of a jurisdiction agreement is presumed:157 ‘the agreement must … already be valid according to a certain law before the “unless” can come into play’.158 An agreement that is the subject of consensus in fact, following the approach autonomously developed by the CJEU, will be presumed valid, and that presumption will only be refuted if the applicable law ‘invalidates’ the agreement.

6.41  Advocate General Szpunar also refers, in this connection, to the goal of aligning the Recast with the 2005 Hague Choice of Court Convention.159 With respect, the Convention does not assist this point. While it also uses the word ‘unless’,160 which could suggest that a jurisdiction agreement governed by that instrument is presumed to be valid under some law other than the law of the chosen court, it is unclear which law that would be. There is no autonomous approach to consensus of a jurisdiction agreement governed by the 2005 Hague Choice of Court Convention which would presume an agreement to be valid.161 And there is no clarity as to whether an agreement governed by the Convention must exist by reference to ‘any normal standards’162 or to the national law of the court seised before its validity is assessed,163 or whether this issue is (p. 215) subsumed entirely by the Convention’s provisions on formal and substantive validity, public policy,164 and capacity.165

6.42  Whatever the position under the 2005 Hague Choice of Court Convention, the autonomous approach to consensus set out in Part II(B) ought to continue to apply under the Recast, despite the substantive validity rule;166 a view which is consistent with a recent decision of the CJEU. In Ryanair, the Court observed that it is for the court to ascertain first whether a jurisdiction agreement satisfies the requirements of form and consensus in fact under the Recast before turning to the separate issue of substantive validity.167

dd)  All aspects of validity except those as to form, consent in fact, or public policy?

6.43  A reduction in the scope of the issues covered by the substantive validity rule raises the question of whether public policy and mandatory laws should equally be excluded. Fentiman has argued that ‘public policy’ of the applicable law should be excluded both on the basis of the CJEU’s ruling in Sanicentral GmbH v Collin,168 concerning the Brussels Convention, and because public policy ‘typically affects not the existence but the enforceability of an agreement’.169 Garcimartín similarly argues that public policy should be excluded170 because including it would reverse existing CJEU case law and because it is an issue of ‘contractual enforceability exogenous to the consent or capacity of the parties’.171 Mankowski has argued that to characterize issues regulated by the lois de police of the applicable law as issues of substantive validity would, inter alia, ‘curtail party autonomy … dramatically’.172 Magnus, by contrast, has argued that substantive invalidity should encompass those ‘grounds of invalidity’ contained in harmonized European contract law instruments, such as the Draft Common Frame of (p. 216) Reference,173 which includes infringement of both ‘fundamental principles’ and ‘mandatory rules’.174

ee)  A question of the juridical nature of the clause?

6.44  As will be clear from the foregoing, the scope of the substantive validity rule is unclear. One little-noticed175 reason for the lack of clarity as to the kinds of issues which can be characterized as issues of substantive validity is that the juridical nature of jurisdiction agreements governed by the Recast176 is uncertain. If the question ‘Which kinds of issues does the substantive validity rule cover?’ is to be resolved autonomously, then the juridical nature of an article 25 agreement provides the answer. Even if that is a question to be resolved by the law of the court seised or the law applicable to the clause’s substantive validity, both of which are methodologically doubtful,177 arguably the juridical nature of the agreement, whether adjudged by article 25 or by one of those bodies of national law,178 is relevant.

6.45  If jurisdiction agreements are contractual, substantive doctrines of contract law will determine their substantive validity. If, instead, they resemble non-obligatory agreements, many of these doctrines are likely to be excluded. Take the requirement of consideration under English law, which has been raised as a particular concern by German writers with the substantive validity rule.179 If a jurisdiction agreement is contractual and English law is the applicable law180 to its substantive validity, it is difficult to see why consideration would not in principle be required,181 though the need for that (p. 217) requirement would be overcome where the jurisdiction agreement is solemnized by being executed and witnessed as a deed. Whether it would suffice that the jurisdiction clause is part of an exchange of value in the underlying contract or whether there would need to be an exchange of value within the jurisdiction clause itself, considering the principle of severability, is a more difficult question.182

6.46  One should pause to note that it remains possible, even though English courts are no longer Member State courts, for English law to apply to the substantive validity of some jurisdiction clauses governed by the Recast. This could occur, for instance, if the jurisdiction clause designates German courts but there is an express choice-of-law clause selecting English law.183 Under German private international law rules, whether the parties reached a materially valid agreement as to jurisdiction184 is generally governed by the lex causae applicable to the contract in which the jurisdiction agreement is contained.185 In this example, that would be English law.

6.47  As the previous example illustrates, the nature of a jurisdiction agreement governed by the Recast has practical ramifications for its enforcement. An agreement on jurisdiction is an independent concept under the Recast and 2007 Lugano Convention.186 It is not a reference to national concepts of jurisdiction by agreement.187 That much the CJEU has said. But the CJEU has not clarified what the nature of a jurisdiction agreement as an independent concept is. Until the Recast’s introduction of the substantive validity rule, there was little cause for a Member State court to seek clarification.188

(p. 218) 6.48  It is uncertain whether a jurisdiction agreement is, for the purposes of article 25, a procedural dispositional agreement (Verfügung),189 or a promissory (obligational) contract, or neither. The CJEU’s consistent references to the procedural function and purpose of a jurisdiction agreement190 tend to suggest that they are not obligational. Equally, there are indications to suggest that they are not dispositional either: the CJEU, in relation to a jurisdiction agreement governed by the Brussels Convention, reasoned that ‘[b]y its nature … such a choice has no legal effect for so long as no judicial proceedings have been commenced and only becomes of consequence at the date when judicial proceedings are set in motion’.191 This statement may be explicable by its context: it was made to justify a ruling that a jurisdiction agreement is to be assessed by the legal rules in force at the time proceedings pursuant to it are commenced. But it may be a broader statement of principle, detracting from the view that jurisdiction agreements have a dispositional character, because a fundamental characteristic of a dispositional contract is that it has a direct effect on the default rules at the time of its conclusion.192 The fact that the Recast treats a jurisdiction agreement as severable from the contract within which it is contained could suggest that the agreement is contractual. Subjecting a jurisdiction agreement to the law of the Member State of the court designated in the agreement, however, tells us little: it might be thought that this points away from a jurisdiction agreement as a ‘contractual creature’,193 but English choice-of-law rules similarly subject the validity of arbitration agreements to the law of the seat194 and there is no suggestion that, by doing so, arbitration agreements cease to be contracts. Definitively determining the nature of a jurisdiction agreement under the Recast and 2007 Lugano Convention is not an undertaking which can or need be completed here, though, as the foregoing observations suggest, it is likely that the CJEU, if asked, would not conclude that they are promissory. Returning to the present enquiry, if it is right that the juridical nature of a jurisdiction agreement as contractual or otherwise informs the characterization of the kinds of issues affecting the substantive validity of that agreement under the Recast, then precisely what the juridical nature of that agreement is, as a matter of autonomous EU law, is a question for the CJEU.

(p. 219) b)  Which law applies to substantive validity?

6.49  The Recast introduces a choice-of-law rule which includes national conflict-of-laws rules (subsection aa)). This raises the question whether the law, which those designate, can be overridden or displaced by the overriding mandatory rules and ordre public international of the chosen court, when the chosen court’s conflict-of-laws rules designate a law other than the law of the chosen court (subsection bb)). The relevance of the overriding mandatory rules and ordre public international of the forum is also an open question (subsection cc)).

aa)  Law applicable according to the choice-of-law rule

6.50  The Recast provides that the substantive validity of a jurisdiction agreement shall be governed by the law of the Member State of the court designated in the agreement.195 At least according to a recital in the Recast, and as suggested by article 25 in its French language version,196 ‘law’ includes that Member State’s conflict-of-laws rules.197 The CJEU, however, has not yet referred to renvoi in relevant case law,198 so a preliminary question on the effect of this recital can be expected, even if it is not strictly necessary.

6.51  The conflict-of-laws rules, contemplated by the recital, are the national conflict-of-laws rules of each Member State199—there is no uniform EU conflict-of-laws rule for the law governing a jurisdiction agreement, this matter having been carved out of the Rome I Regulation.200 In France, the conflict-of-laws rule applicable to the validity of a jurisdiction agreements is unsettled.201 It may be the law chosen by the parties,202 the law of a court excluded by the clause,203 the law of the chosen court,204 or the law of the court seised.205 In Germany, as we have seen, the relevant conflict-of-laws rule is that the law governing the contract, in which the jurisdiction clause is contained, will govern the jurisdiction agreement.206 The Recast’s recital refers, in terms, to the law of the Member (p. 220) State designated in the agreement, including its ‘conflict-of-laws rules’. Some authors have suggested that it might not be limited to ‘conflict-of-laws rules’ and, instead, include all the private international law rules of that Member State.207

bb)  Overriding mandatory rules and ordre public international of the chosen court (assuming the law of the chosen court is, under its choice-of-law rules, not the applicable law and another court is seised)?

6.52  If the law of the Member State designated in the agreement includes all the private international law rules of that Member State, a further question arises (assuming that it is plausible that issues of public policy and mandatory laws can be characterized as issues of substantive validity).208 Can the law which those rules designate be overridden by the overriding mandatory rules or displaced by the ordre public international of the Member State designated in the agreement?209 For example, if the jurisdiction agreement designates exclusively the courts of France and French conflict-of-laws rules lead to Luxembourgish law as the applicable law, should the Luxembourgish court, if seised, nonetheless apply French overriding mandatory rules or give effect to France’s ordre public international? On the one hand, coherence with the Rome I Regulation would militate against this because it would have the result of the court of one Member State applying the overriding mandatory rules of another; a result which the Rome I Regulation, and its interpretation by the CJEU,210 greatly circumscribe.211 On the other hand, article 9(3) of the Rome I Regulation does allow a Member State court to give effect to the overriding mandatory rules of another country, the laws of which render performance unlawful in the place where the contract is to be performed. If a jurisdiction agreement designating French courts under the Recast is a contract,212 then its performance must necessarily take place in France, where French overriding mandatory rules would render it unlawful.

cc)  Overriding mandatory rules and ordre public international of the forum?

6.53  In a traditional conflict-of-laws analysis, one would ask whether application of the applicable law is overridden by any overriding mandatory rules of the forum or displaced by the ordre public international of the forum.213 Several commentators suggest, or appear to suggest, that this step should not apply under the Recast. Basedow argues that ‘courts outside the country of the selected forum will not be able to enforce their own (p. 221) imperative norms as against an otherwise valid forum selection clause under Article 25’.214 Mankowski, like Winkler,215 also rejects the application of the overriding mandatory rules of the forum and recourse to its ordre public international. He appears to do so because the issues with which they deal are not issues which should be characterized as issues of substantive validity, consistent with his argument regarding public policy and mandatory rules of the applicable law.216 Fentiman leaves open the question of whether the public policy of the forum is relevant, ‘notwithstanding [his view] that public policy is outside the scope of the substantial validity test’.217

6.54  The uncertainty arises because of CJEU jurisprudence concerning the Brussels Convention. Relevant CJEU jurisprudence tends to suggest218 that imperative norms relating to the effects of the agreement cannot impugn a jurisdiction agreement conferring jurisdiction on a Member State court. In other words, it is immaterial that enforcement of the clause will mean that the forum’s overriding mandatory rules will not be applied or that the forum’s ordre public international cannot be respected.219

6.55  Relevant CJEU jurisprudence also says something about imperative norms relating to the jurisdiction agreement itself.220 In Sanicentral, the CJEU was concerned with an imperative provision of French law which rendered void clauses derogating the jurisdiction of French courts in employment matters. It ruled that the provision did not apply where the agreement, designating the jurisdiction of German courts, satisfied the (p. 222) requirements of article 17 of the Brussels Convention.221 It remains to be seen whether this jurisprudence concerning the Brussels Convention is still applicable under the Recast, even though the texts of those instruments are, with respect to substantive validity, obviously different.222

2.  Asymmetric jurisdiction clauses

6.56  Having considered how the substantive validity rule operates in the context of simple jurisdiction agreements, this section will turn to its operation in cases involving complex ones. Subsection (a) will examine how a rule which is designed to test the validity of a jurisdiction agreement, by reference to the law of the chosen court, can apply to those asymmetric agreements which confer, or appear to confer, jurisdiction on more than one court. Subsection (b) explores the kinds of doctrines and provisions of national law that may render an asymmetric jurisdiction clause substantively invalid. That analysis will demonstrate that while the substantive validity rule will not lead to the systematic setting aside of asymmetric jurisdiction clauses, it may have that effect in certain cases.

a)  Law applicable according to the law of which court?

6.57  As has already been seen, the Recast subjects the substantive validity of the agreement to the law of the Member State court on which the clause confers jurisdiction, or, if its conflict-of-laws rules so provide, the law of another state.223 As Briggs has remarked, ‘[h]ow [the rule] is supposed to work if the clause designates two courts with jurisdiction … is unexplained probably because it is inexplicable’.224 Because the rule presupposes that all jurisdiction agreements confer jurisdiction on only one court,225 the starting point is to determine the court or courts on which different types of asymmetric clauses confer jurisdiction.226 For unilateral non-exclusive jurisdiction clauses,227 the rule is unproblematic:228 the only court on which the clause confers jurisdiction is the (p. 223) court which is named in the clause. How the rule is supposed to apply to different variants of Rothschild clauses and unilateral indeterminate jurisdiction clauses is more complex.229

aa)  Standard Rothschild clauses

6.58  As argued in Chapter 5, the most cogent characterization of a standard Rothschild clause is that it prorogates the jurisdiction of the anchor court, and only that court, for both parties; it preserves but does not prorogate the jurisdiction of other courts for one of them.230 That characterization is consistent with the view that a standard Rothschild clause has exclusive effect for proceedings brought by the non-option holder and non-exclusive effect for proceedings brought by the option holder.231 This means that the only court on which a standard Rothschild clause confers jurisdiction is the anchor court.

6.59  If that is right, application of the substantive validity rule, when applied to standard Rothschild clauses, is less problematic than appearances first suggest. If the anchor court is seised by either party, it will apply the law designated by its conflict of laws rules to the substantive validity of the clause. If, instead, another court is seised by either party, it will apply the law designated by the anchor court’s conflict-of-laws rules to the clause’s substantive validity.232 Whether that other court is seised by the option holder, according to the agreement, or by the non-option holder, contrary to it, is immaterial for the purposes of the rule.233 The following example is illustrative. Assume there is a standard Rothschild clause in the following terms:

The Tribunal de Commerce de Paris has exclusive jurisdiction to settle any dispute arising out of or in connection with this Agreement … no Finance Party shall be prevented from taking proceedings relating to a Dispute in any other courts with jurisdiction … [and] in any number of jurisdictions.234

(p. 224) Assume that the courts of Luxembourg are courts ‘with jurisdiction’ because the contract is to be performed there.235 If the non-option holder or option holder (a Finance Party) brings proceedings before the Tribunal de Commerce as the anchor court, the substantive validity of this clause will be assessed by the law designated by French conflict-of-laws rules. Those rules may designate French law as the law of the chosen court or the law of the court seised.236 Instead, those rules may designate Luxembourgish law, as the law of the court excluded by the clause.237 The position would not change if the non-option holder, contrary to the terms of the clause, brings proceedings in Luxembourg or the option holder, under its option, brings proceedings in Luxembourg.238 In either case, the Luxembourgish court should apply French conflict-of-laws rules, leading either to Luxembourgish law or French law as the law governing the substantive validity of the clause.

6.60  There is, however, a potential caveat to these propositions. They hold only to the extent that overriding mandatory rules and the ordre public international do not enter the fray. If the conflict-of-laws rules of the anchor court lead not to the law of the anchor court but to the law of another country, a question arises as to whether the applicable law may be displaced by the overriding mandatory rules and ordre public international of either the anchor court or the court seised. Returning to the example, that question would arise in three scenarios, first, if a French court is seised and French conflict-of-laws rules designate Luxembourgish law. In that situation, Luxembourgish law may be displaced by the overriding mandatory rules and ordre public international of France as the forum. Second, if instead, a court in Luxembourg is seised and French conflict-of-laws rules designate French law, there is a possibility that a Luxembourgish court could apply the overriding mandatory rules and ordre public international of Luxembourg, as the forum, to displace French law. Third, if a court in Luxembourg is seised and French conflict-of-laws rules instead lead to Luxembourgish law, the question of whether the Luxembourgish court should apply the overriding mandatory rules and ordre public international of France as the ‘chosen’ (anchor) court may arise. As seen above, the relevance of overriding mandatory rules and the ordre public international to the substantive validity rule is not an acte clair,239 so the extent to which this caveat will matter is uncertain.

bb)  Broad and narrow Rothschild clauses and broad and narrow unilateral indeterminate jurisdiction clauses

6.61  Chapter 5 demonstrated that unilateral indeterminate jurisdiction clauses confer jurisdiction on more than one court.240 For instance, ‘Place of jurisdiction shall be, at our option, F and N respectively, I or the buyer’s place of business’.241 Or the non-option holder ‘accepts … any court—in England or elsewherethat (p. 225) the [option holder] may designate’.242 The substantive validity rule for these clauses is therefore clearly problematic. The substantive validity rule will also be problematic for broad variants of Rothschild clauses if they are construed as conferring jurisdiction on any court which the option holder may choose. For instance:

… each of the parties irrevocably: (a) agrees for [C’s] benefit that the courts of [France] shall have jurisdiction to settle any suit, action or other proceedings relating to this Agreement (‘Proceedings’) and irrevocably submits to the jurisdiction of such courts (provided that this shall not prevent us from bringing an action in the courts of any other jurisdiction).243

Further, the substantive validity rule will be problematic for narrow variants of Rothschild clauses if they are interpreted as conferring jurisdiction on the nominated court in the anchor and other court or courts identifiable from the clause.244

6.62  There seems to be five conceivable approaches to this rule which would allow it to be applied to broad and narrow Rothschild clauses and unilateral indeterminate jurisdiction clauses. These are that the clause must be:

  1. 1.  valid by reference to the law designated by the conflict-of-laws rules of all courts on which the clause confers jurisdiction;

  2. 2.  valid by reference to the law designated by the conflict-of-laws rules of the court that is seised, provided it is a court on which the clause confers jurisdiction;245

  3. 3.  valid by reference to the law designated by the conflict-of-laws rules of any one of the courts on which the clause confers jurisdiction, irrespective of which court is seised;

  4. 4.  invalid under the law designated by the conflict-of-laws rules of all the courts on which the clause confers jurisdiction if the court seised is not one of those courts, for the court seised to accept jurisdiction; or

  5. 5.  valid, in the case of a broad Rothschild clause, under the law designated by the conflict-of-laws rules of the law of the anchor court and, in the case of a broad variant of unilateral indeterminate clause, by reference to the law of the named court246 and not to the law of the other courts on which the clause confers jurisdiction. This approach is, however, unworkable for narrow variants of unilateral (p. 226) indeterminate jurisdiction clauses247 and narrow variants of Rothschild clauses in which multiple courts are named.

6.63  Take again the broad Rothschild clause introduced above:

… each of the parties irrevocably: (a) agrees for [C’s] benefit that the courts of [France] shall have jurisdiction to settle any suit, action or other proceedings relating to this Agreement (‘Proceedings’) and irrevocably submits to the jurisdiction of such courts (provided that this shall not prevent us from bringing an action in the courts of any other jurisdiction).248

Assume that the law expressly chosen to govern the container contract, in which this jurisdiction clause is contained, is French law. Assume that the non-option holder has moveable assets in Germany so that German courts are those on which the clause confers jurisdiction if the option holder chooses to bring proceedings there. Assume, too, that the courts of Luxembourg would otherwise have special jurisdiction under the Recast over an action against the option holder, were that jurisdiction not excluded by the jurisdiction clause. Finally, assume that a French court has been seised. Based on these assumptions, Table 6.1 shows the implications of applying the above approaches to the Recast’s substantive validity rule to that clause.

(p. 227)

Table 6.1.  The implications of various approaches to the substantive validity rule for a broad variant of Rothschild clause

Approach

Implications for the clause

1. Valid by reference to the law designated by the conflict-of-laws rules of all courts on which the clause confers jurisdiction.

1. Valid by reference to Luxembourgish law (which, under French conflict-of-laws rules, may be the law governing the substantive validity of the clause because it is the law of a court excluded by the clause)a or French law (which, by virtue of French conflict of laws rules, may instead be the law governing the substantive validity of the clause because it is the law of one of the chosen courts),b and valid by reference to the law designated by German conflict-of-laws rules (which would, because of the choice-of-law clause in the container contract, be French law).

2. Valid by reference to the law designated by the conflict-of-laws rules of the court that is seised, provided it is a court on which the clause confers jurisdiction

2. Valid by reference to French law (which, also by virtue of French conflict-of-laws rules, may be the law governing the substantive validity of the clause because it is the law of one of the chosen courts and the law of the court seised).

3. Valid by reference to the law designated by the conflict-of-laws rules of any one of the courts on which the clause confers jurisdiction, irrespective of which court is seised.

3. Valid by reference to Luxembourgish law or French law.

4. Invalid under the law designated by the conflict of laws rules of all the courts on which the clause confers jurisdiction, if the court seised is not one of those courts, for the court seised to accept jurisdiction.

4. Not applicable on these facts because the clause confers jurisdiction on French courts.

5. Valid by reference to the law designated by the conflict-of-laws rules of the law of the anchor court.

5. Valid under either French law or Luxembourgish law.

a See para 6.51.

b As explained at para 6.51, the French conflict-of-laws rule applicable to jurisdiction agreements is unsettled.

6.64  As this illustration demonstrates, the Recast’s substantive validity rule, applied to a broad Rothschild clause, is complex. Realistically though, it will only lead to a choice between several rather than myriad competing laws. Granted, that complexity would be amplified in this illustration if the parties had not expressly chosen French law to govern the container contract or if the rules that a German court would apply in the absence of an express choice of law would lead not to French or Luxembourgish law but to German law. But it does not make the rule impossible to apply.

6.65  The best and simplest solution from among the approaches examined is the second approach: assess the substantive validity of the clause by reference to the law designated by the private international law rules of the court that is seised, provided it is a court on which the clause confers jurisdiction. This approach is equally workable in the case of narrow variants of unilateral indeterminate jurisdiction clauses and narrow variants of Rothschild clauses, in which multiple courts are named. It also has the advantage over the other approaches of preventing the outcome of a court being seised and forced to exercise jurisdiction conferred by a jurisdiction clause which, under its law, would be invalid. Finally, this approach is also consistent with the CJEU’s view that a jurisdiction agreement does not have any procedural effect until proceedings have been commenced.249

6.66  Admittedly, this solution does not solve the problem of which law should be applied by a court seised on which the clause does not confer jurisdiction. But that is a problem (p. 228) which will not arise whenever one of the designated courts is also seised: article 31(2) of the Recast requires the non-designated court to stay its proceedings so as to let the designated court to rule, inter alia, on the substantive validity of the clause.250 The third and fourth approaches also present a solution to that problem, though they effectively impose a favor validitis principle which, albeit in different terms, was rejected during the drafting of the Recast.251

6.67  An alternative solution would be to allow the parties to choose the substantive law governing the substantive validity of their jurisdiction agreement. This could be achieved either by amending the Recast252 or by Member States developing their national conflict-of-laws rules where they currently do not allow for this.253 This would mean that, regardless of which court’s conflict of laws rules are applied to determine the law applicable to the substantive validity of the clause, the result will be the same: the law applicable to that issue will be the law chosen by the parties to govern it.

6.68  It might be thought that allowing parties to choose the law applicable to an asymmetric jurisdiction agreement further increases the likelihood of opportunism by the option holder. Given that the option holder is generally the party dictating the contract terms,254 one would expect it to choose a law which has a particularly generous approach to substantive validity, meaning that the question of whether the non-option holder’s consent in fact to the clause has been vitiated will be determined by the option holder’s preferred law. The same result would follow if the Recast’s rule were that the law applicable to the substantive validity of the jurisdiction clause is the law expressly chosen in the container contract.255 In both cases, it could be said that the option holder ‘unilaterally determines the level of protection of the other party’.256 But because the (p. 229) parties’ choice of law would be subject, at least, to the ordre public international or overriding mandatory rules of the forum, allowing parties to choose the law applicable to the substantive validity of an asymmetric jurisdiction clause may be somewhat more innocuous than first appears.257

b)  Doctrines of the applicable law potentially relevant to substantive validity

6.69  Once the Member State whose law will apply to the substantive validity of an asymmetric clause has been identified, the next question that arises is which doctrines of that law will be relevant. Depending on the juridical nature of a jurisdiction clause governed by the Recast, any number of doctrines or types of provisions of national law may affect the substantive validity of asymmetric jurisdiction clauses.258 Of potential relevance for asymmetric agreements is that agreements, or at least contracts, which are grossly imbalanced, or in which one party takes an excessive advantage, are invalid under the laws of multiple Member States.259 In several Member States, such as France, disparity is considered to be an issue of defective consent.260 Regardless of their juridical nature, it is equally plausible that asymmetric clauses will be invalidated on public policy grounds.261 This subsection explores three specific doctrines/types of provisions which may be relevant: the sanctioning under French law of clauses which subject a trading partner to a significant imbalance (subsection aa)) or which are subject to a potestative condition (subsection bb)), and, under both French and German law, the review of standard contract terms (subsection cc)).

aa)  Subjection of a trading partner to a significant imbalance (L. 442-1 I–2 of the Code de commerce)

6.70  Where French law is relevant to the substantive validity of the jurisdiction clause,262 one basis on which it may potentially be rendered null is article L. 442-1 I–2 of the Code de commerce.263 This provision applies especially to (p. 230) distribution contracts,264 in which asymmetric jurisdiction clauses can be found.265 It provides that a party which ‘[s]ubject[s] or seek[s] to subject the other party to obligations that create a significant imbalance in the rights and obligations of the parties’ ‘shall be held liable and obliged to make good the damage caused’.266 A clause which infringes this provision may be declared null.267 L. 442-1 has an ordre public nature.268 According to the Rome I Regulation definition, L. 442-1 I–2 is also an overriding mandatory provision (disposition de loi de police) under French law,269 regarded as ‘crucial for safeguarding [France’s] social and economic organisation’.270 It is unclear whether a French court would apply L. 442-1 I–2 to a jurisdiction clause, which French law does not characterize as contractual,271 although this does not appear to be an obstacle from the perspective of the French doctrine.272 In any event, there would be no obstacle to the application of L. 442-1 I–2 if the CJEU were to rule that jurisdiction agreements under article 25 are, as a matter of autonomous EU law, contractual in nature.273

6.71  French courts may be influenced by CJEU case law in considering whether L. 442-1 I–2 applies to jurisdiction clauses. As is well known, the CJEU has held that a non-negotiated jurisdiction agreement will be invalid in a standard-form consumer contract, ‘in a case internal to a Member State’ if, contrary to the EU Unfair Contract Terms Directive, it creates a significant imbalance between the parties.274 The CJEU and the Belgian Cour de cassation have since, however, held that this case law also applies in cross-border cases to jurisdiction agreements governed by article 25 of the Recast: the substantive validity rule refers to the law of the designated Member State, which includes the legislation implementing the Directive.275 Although these cases all involved consumers and the Directive does not apply to commercial contracts, French courts may be influenced by that case law in interpreting article L. 442-1 I–2 of the Code de (p. 231) commerce, which employs similar language to the Directive.276 If so, where French law is relevant to the substantive validity of an asymmetric clause governed by article 25 of the Recast, a Rothschild clause which designates the courts of the option holder’s domicile and provides for the option holder to seise any competent court may, depending on the circumstances of the case,277 be null. The same conclusion would apply to a broad variant of Rothschild clause, which allows the option holder to seise any court of its choice, or a unilateral indeterminate jurisdiction clause, which requires the non-option holder to bring or defend proceedings in a court to be chosen by the option holder after a dispute arises.

bb)  Clauses subject to a potestative condition

6.72  Where French law is the law applicable to the substantive validity of a Rothschild clause,278 it might be thought that a second basis on which it could be invalidated is article 1304-2 of the French Code civil. The following analysis suggests that this is improbable. Usunier suggests that the substantive validity rule would allow for the invalidation of an asymmetric jurisdiction clause subject to a potestative condition.279 That is one interpretation of the reference to the clause’s ‘potestative character’280 made by the Civil Chamber of the Cour de cassation in the Rothschild case281 and by its Commercial Chamber in Diemme.282 In the former decision, the Court held that the fact that the clause bound only the non-option holder meant that ‘it was contrary to the objective and finality of a prorogation clause, provided by article 23 of [Regulation 44/2001]’.283 In the latter, the Court held that the fact that the clause bound only the non-option holder was irrelevant, given the principle of party autonomy.284 Both cases were decided under Regulation 44/2001, which, on one view, prevented the application of national law to the substantive validity of the clause.285

6.73  Article 1304-2 of the revised French Code civil286 provides that ‘an obligation subject to a condition, the fulfilment of which depends solely on the will of the obligor, is null. Nullity on this ground cannot be invoked if the obligation has been performed with (p. 232) knowledge of the circumstances.’287 Several French commentators have argued that what is now article 1304-2 of the Code civil does not apply to jurisdiction agreements because they do not give rise to rights and obligations.288 Whether this is so for jurisdiction agreements governed by the Recast is uncertain, given that the CJEU is yet to define their exact nature.289

6.74  Let us assume, for argument’s sake, that jurisdiction agreements do contain obligations,290 such that article 1304-2 would apply to a Rothschild clause whose substantive validity is governed by French law. Take a standard Rothschild clause in the following terms:

The Tribunal de Commerce de Paris has exclusive jurisdiction to settle any dispute arising out of or in connection with this Agreement … no Finance Party shall be prevented from taking proceedings relating to a Dispute in any other courts with jurisdiction … [and] in any number of jurisdictions.291

6.75  Article 1304-2, applied to that clause, would mean that the obligation to sue in Paris is subject to a condition, the fulfilment of which depends solely and exclusively on the choice of the option holder.292 That obligation is null unless the option holder does actually sue in Paris while aware of the fact that it was not obliged to do so. The clause effectively provides, ‘I will sue you in Paris, if I choose to.’293 ‘[T]he condition is merely (p. 233) a state of mind of the obligor, and the promise if enforced inflicts no further duty upon … [it] other than that of not wishing to [sue in France].’294 On the face of it, such a clause would appear to be subject to a purely potestative condition, such that it would be null unless the option holder sues in France.

6.76  On closer analysis, however, the clause set out above is not subject to any condition at all.295 The option holder’s obligation is potestative (‘I will sue you in Paris, if I choose to’), but it is not subject to any condition, potestative or otherwise. The point is a fine one, but there is a distinction between the obligation itself being potestative and the obligation being subject to a condition that is potestative. An example of an obligation subject to a potestative condition would be: ‘I will sell you an apartment at Bondi Beach, if I choose to buy one.’296 My obligation (to sell you an apartment at Bondi) is subject to a condition (my buying an apartment at Bondi) and the fulfilment of that condition is subject to nothing more than my decision on whether or not I choose to buy one. If, however, I do choose to buy an apartment at Bondi, then I am obliged to sell it to you, the condition rendering the clause null having been fulfilled.297

6.77  The preferable view is that an asymmetric jurisdiction clause may be sanctioned under French law if it contains a ‘potestative right’298 that is misused.299 That view is more consistent with the language of ‘potestative character’ used by the Cour de cassation in connection with the significant discretion the clause afforded to the option holder.300 A contractual option or discretionary contractual power is a potestative right.301 A right or option with its source in the general law may also be a potestative right.302 Fenouillet explains that:

The particularity of a potestative right resides in its effect, whether it be the content (modifying an existing legal situation), the source (unilateral intention) or the nature (uncertainty) of the right: it confers on the right holder the power to modify a legal situation, with the result that a right holder’s unilateral will has consequences for the (p. 234) counterparty without the counterparty knowing in advance what the right holder will ultimately do.303

Unlike an obligation subject to a purely potestative condition which, if article 1304-2 applies to a Rothschild clause, would result in the nullity of the entire clause,304 a potestative right is not per se invalid.305 This suggests that it is not relevant to the substantive validity of a jurisdiction clause under the Recast. Instead, the exercise of a potestative right may, depending on the circumstances of the case, amount to an abuse of rights under French law, as Part III will show.306

cc)  Review of standard contract terms

6.78  A third basis on which an asymmetric clause may be impugned under the substantive validity rule is by reference to the standard contract terms control test under French and German law.307

6.79  Article 1171 of the revised French Code civil308 may be relevant if the asymmetric clause is contained in a standard-form contract.309 Several authors have argued that this provision is d’ordre public310 and applicable to commercial contracts.311 Article 1171, which has no equivalent in the former Code civil, further entrenches into French law the concept of ‘déséquilibre significatif’. This concept may reflect a more fundamental principle of France’s ordre public which is part of the ordre public international.312 It provides that:

in a contract of adhesion, any clause which creates a significant imbalance between the rights and obligations of the parties to the contract shall be deemed to be unwritten.313 (p. 235) In assessing a significant imbalance neither the main subject-matter of the contract nor the adequacy of price vis à vis performance is relevant.314

6.80  If ‘null and void’ in article 25(1) of the Recast also means ‘ineffective’,315 it is arguable that § 307 of the German Bürgerliches Gesetzbuch is also relevant. It will apply to asymmetric jurisdiction clauses, contained in standard-form contracts,316 between commercial parties,317 whose substantive validity is governed by German law.318 § 307 is considered to be a simple mandatory provision of German law.319 It aims to correct ‘a partial market failure’ caused by the rational disinterest of an adhering party in the content of a standard form.320 It also aims to compensate the adhering party for restricting its freedom to choose.321 § 307 provides that a standard contract term is ineffective if, contrary to the requirement of good faith, it unreasonably disadvantages the proferee.322

6.81  An unreasonable disadvantage is assumed to exist if a provision is not compatible with the essential principles of the default statutory provisions from which it deviates.323 These principles apply where the parties have agreed to derogate from or extend the (p. 236) statutory default rules.324 The essential principles of the Recast325 are: that the defendant be able to predict where, other than at its domicile, it will be called upon to defend proceedings326 and that ‘creating a situation in which a number of courts have jurisdiction in respect of one and the same contract’ is to be avoided.327 In this context, Rothschild clauses which allow the option holder to sue in any court of its choice may give rise to an assumption of an unreasonable disadvantage; Rothschild clauses which allow the option holder to sue in any court with jurisdiction (and which therefore incorporate the statutory default rules) will not.

3.  Conclusion

6.82  Although it is yet fully to be tested, the substantive validity rule which the Recast introduces changes the way jurisdiction agreements designating EU Member State courts are to be assessed. Difficulties in identifying which law applies to asymmetric clauses by reference to this rule exist, but they are not insurmountable. The doctrines of national law which may apply via the substantive validity rule, including standard form contract review, could go some way towards addressing a key normative problem to which asymmetric jurisdiction clauses give rise, namely, the problem of the non-option holder not having genuinely consented to the clause.328 But the applicability of a number of these doctrines may depend on the nature of a jurisdiction clause and the way substantive validity should be characterized, both of which the CJEU is yet to clarify.

III.  Prohibition of abuse of rights under EU and national law

6.83  An asymmetric jurisdiction clause may be constrained if the option holder, or indeed the non-option holder, invokes the clause abusively or for an improper purpose. Several scholars have argued that the decision of the Cour de cassation in the Rothschild case implicitly acknowledged that the clause was abusive, although the option holder neither exercised (nor exercised abusively) its option in the circumstances of the case.329 With (p. 237) a view to controlling the abusive exercise of rights under jurisdiction clauses generally, some scholars330 argue that an abuse mechanism does or ought to apply to all jurisdiction clauses governed by the Recast or 2007 Lugano Convention.331 As Basedow puts it, Member State courts applying EU law should assess ‘not the validity of the clause, but the exercise of the rights flowing from it. This exercise may be abusive in the particular circumstances of the case.’332 Wais has argued that an abuse mechanism should apply to Rothschild clauses specifically, controlling either the clause as a whole or the exercise of the option.333

6.84  This Part will argue that while an EU abuse test controlling the exercise of any rights of either party under an asymmetric clause may be desirable, the EU ‘prohibition of abuse of rights’, in its current form, is capable of controlling only some of them. The reason for that, as will be further elaborated below, is because the EU prohibition targets the misuse of a rule, not the misuse of a right conferred or preserved by agreement. Unless or until the CJEU further develops its autonomous test, CJEU jurisprudence suggests that Member State courts can apply mechanisms, substantive or procedural, arising out of the law of the forum to control an abuse, at least before their own courts.334 They may do so to the extent that such mechanisms do not undermine the effectiveness of the Recast or 2007 Lugano Convention.335 These national law doctrines are distinct from those which Member State courts may apply to asymmetric clauses under the Recast via the substantive validity rule examined above.336 As has been seen, those doctrines only apply to the extent that they result in the nullity, ineffectiveness, or setting aside of the entire clause.337 Moreover unlike abuse doctrines, some of those doctrines applicable to substantive validity may only apply if jurisdiction agreements governed by those instruments are obligational ‘contracts’.338

6.85  This Part proceeds as follows: Part III(A) introduces several abuse-control mechanisms which exist under French and German law. Part III(B) explains the principle of (p. 238) effectiveness. Why the EU ‘prohibition of abuse of rights’ has a more limited application to asymmetric jurisdiction clauses than might be thought is demonstrated in Part III(C). Finally, Part III(D) applies national abuse doctrines to several scenarios in which asymmetric jurisdiction clauses might be misused.

A.  Introduction to abuse mechanisms under national law

6.86  Abuse of rights is broadly a doctrine which prevents ‘an abusive exercise of an otherwise lawful right’.339 Abuse of rights finds reflection in the French concept of abus de droit, the German concept of Missbrauch, and in international soft law instruments and European restatements on contract law.340

6.87  Abus de droit is a doctrine which controls the improper ‘exercise of the right held by an individual … for example, a right to sue, a right not to contract, [or] a right to terminate a contract for non-performance under an express termination clause’.341 Relevantly for asymmetric jurisdiction clauses, it applies under French law to unilateral powers (or potestative rights or options)342 in contractual clauses,343 presumably via article 1104 of the revised Code civil,344 although only to the extent that constraining the power does not render it devoid of substance.345 It also applies to procedural rights, including the (p. 239) right to bring and defend proceedings.346 A similar idea is reflected in the German concept of Missbrauch (misuse): § 242 of the German Bürgerliches Gesetzbuch—the general provision on good faith—encompasses Missbrauch.347 Relevantly for asymmetric clauses, that provision applies both to procedural rights348 and to unilateral contractual powers.349 German case law makes clear that Bundesgerichtshof sees a role for § 242 in cases governed by the Recast or 2007 Lugano Convention.350

B.  The principle of effectiveness

6.88  In Kefalas v Dimosio (Greek State), the CJEU held that Member States may apply their own doctrines of national law, with their own yardsticks,351 to control an abuse of a right. This is subject to the proviso that the application of these doctrines is not incompatible with the objectives of the ‘uniform application’ of the provision of EU law in question.352 Similarly, in Shevill and others v Presse Alliance, the CJEU ruled that ‘as regards procedural rules, reference must be made to the national rules applicable by the national court, provided that the application of those rules does not impair the effectiveness of the [Brussels] Convention’.353 The caveat, referred to by the Court in each of those cases, is the principle of effectiveness. It commands that national law must not make ‘the application of European Union law impossible or excessively difficult’, (p. 240) ‘jeopardise the achievement of the objective pursued by [the EU instrument]’,354 or be ‘incompatible with the system of the [EU instrument]’.355

6.89  The principle of effectiveness has been employed by the CJEU in the context of the Recast’s predecessors to prevent the application of a procedural rule of English law designed to control an abuse356 before another Member State court. As has been seen,357 the Court ruled that anti-suit injunctions impermissibly affect a Member State court’s power to assess its own jurisdiction, which is inherent to the principle of mutual trust. Because the enhancement of mutual trust is, among other things,358 an objective359 of the Recast, each court should have the same authority to apply the Recast’s rules of jurisdiction.360

C.  Existing mechanisms at an EU level

1.  EU ‘prohibition of abuse of rights’

6.90  The CJEU has, in multiple cases, held that there is a ‘general principle of EU law on the prohibition of abuse of rights’.361 That statement might lead one to think that the CJEU has established a principle of primary (constitutional) EU law362 which could control the exercise of an option under an asymmetric clause in a similar way to abus de droit or Missbrauch. But that view is mistaken. The general principle targets the misuse of a rule, not the misuse of a right conferred or preserved by agreement. Applied to a Rothschild clause, for instance, this means that the EU prohibition of abuse of rights will only apply to control either party’s reliance on the anchor part of the clause; it will not control the option holder’s ability to rely on the option.363 Better described as abuse of law, the general principle articulated by the Court resembles fraude à la loi, fraude à la competence, or Gesetzesumgehung.364 Those terms denote rule appropriation or avoidance; (p. 241) a deliberate attempt to avoid or divert the application of a rule that would otherwise apply to ensure the application of another.365 In Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas, the CJEU articulated a test for abuse of EU law.366 The CJEU has subsequently applied the test to cases involving private parties. In Nils-Johannes Kratzer v R + V Allgemeine Versicherung AG, the Court stated that the test requires, first, that ‘objective circumstances’ show that, ‘despite formal observance of the conditions laid down by EU rules, the purpose of those rules has not been achieved’;367 and second, that ‘objective factors’ show that ‘the essential aim of the transactions concerned’ from the subjective perspective of the party concerned ‘is to obtain an undue advantage’.368 In establishing the second element, a court may take particular account of any artificiality involved in the transaction.369

6.91  ‘Artificiality’ has been described as a ‘touchstone’ of the CJEU’s test.370 It is an element371 which one does not find in national doctrines on abuse of rights (including abus de droit or Missbrauch),372 reinforcing the point that the general principle of EU law is concerned with abuse of law.373 That view is further supported by the CJEU’s decision in Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Akzo Nobel NV.374 It establishes that ‘the invocation of the jurisdictional rules in the Recast Regulation [can] amount to an abuse of EU law’.375 Advocate General Jääskinen, in Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Evonik Degussa GmbH (Cartel Damage Claims),376 reframed the referring court’s question regarding article 6(1) of Regulation 44/2001 as one of abuse of rights (abus de droit). That provision, now contained in article 8(1) of the Recast, allows for multiple defendants to be sued in the place of domicile of one of them. Its purpose is to avoid the risk of irreconcilable judgments that would arise out of the same situation of fact and law if the defendants were to be sued separately.377 The (p. 242) referring court in that case had asked whether it was significant that the claims against the anchor defendant (from whose domicile the court had derived jurisdiction over the other defendants) had been withdrawn by the applicant.378 Advocate General Jääskinen opined that it would be for the national court to establish whether the postponement of an out-of-court settlement between the applicant and the anchor defendant until after the applicant had instituted proceedings was an abuse of rights, in that its ‘sole purpose’ was to deprive ‘one of the other defendants of the jurisdiction of the courts of the Member State in which that other defendant is domiciled’.379

6.92  The CJEU did not adopt this language, and instead used the language of circumvention. The Court reasoned that:

the court seised of the case can find that the rule of jurisdiction laid down in that provision [art 6(1) of Regulation 44/2001] has potentially been circumvented only where there is firm evidence to support the conclusion that the applicant artificially fulfilled, or prolonged the fulfilment of, that provision’s applicability.380

The fact that Advocate General Jääskinen and the CJEU framed the same problem as abuse of rights and abuse of law, respectively, indicates that where a right derives directly from a rule, it will often, though perhaps not always, be possible to frame its misuse in terms of abuse of law. This allows for it to be sanctioned via the EU prohibition, provided the requisite element of artificiality is satisfied. But where a right derives from an agreement and only indirectly from a rule,381 reframing an abuse of rights as an abuse of law will not be possible. The option holder’s rights, under the optional limb of a standard Rothschild clause,382 do not derive directly from a rule; rather, they derive from an agreement, indirectly from the default rules of special and general jurisdiction preserved by that agreement,383 and indirectly from a rule giving effect to the agreement.384 That being so, the general principle of EU law on the prohibition of abuse, in its current form, is not relevant to the optional part of the clause.385 Conversely, reliance by the option holder on the anchor part of the clause does arguably derive directly from a rule of EU law which gives the clause non-exclusive effect in circumstances where the option holder sues.386 If that is right, reliance by the non-option holder on the anchor part of the clause also derives directly from a rule of EU law that gives the clause (p. 243) exclusive effect when the non-option holder sues.387 Put simply, the EU prohibition may be instructive where the non-option holder or option holder argues that being required to defend proceedings in the anchor court is abusive; but it would not be instructive where the non-option holder argues that proceedings brought by the option holder pursuant its option are abusive.

2.  Dionisios Diamantis v Elliniko Dimosio (Greek State)

6.93  Apart from the EU prohibition of abuse of law just discussed, several authors have interpreted the CJEU’s ruling in Dionisios Diamantis v Elliniko Dimosio (Greek State)388 as establishing a disproportionality criterion for abuse of rights at a then European Community, now European Union level.389 In that case, the referring Greek Court asked whether the factual circumstances of the case were sufficient to warrant a finding that a shareholder’s invocation of its rights under a European Community directive, regulating changes in capital,390 contravened article 281 of the Greek Civil Code on abuse of rights. That article, as summarized by the CJEU, provided that ‘the exercise of a right is prohibited where it manifestly exceeds the bounds of good faith, morality or the social or economic purpose of that right’.391 One could be tempted to derive from paragraph 33 of the Court’s reasoning the articulation of a self-styled test by the CJEU.392 However, in light of the case to which it refers in that paragraph,393 and when viewed in context of the rest of the judgment,394 the better view is that the Court is rather giving guidance to the Greek Court in assessing whether application of article 281 of the Greek Civil Code, in the circumstances, would impermissibly ‘detract from the full effect and uniform application of Community law in the Member States’. In essence, that guidance was that application of a national doctrine on abuse of rights will not detract from the full effect and application of the Second Directive if:

(p. 244)

of the remedies available for a situation that has arisen in breach of [a] provision [of European Community law], [a shareholder] has chosen a remedy that will cause such serious damage to the legitimate interests of others that it appears manifestly disproportionate. Such a determination would not alter the scope of that provision and would not compromise its objectives.395

In other words, the CJEU was helping a Member State court to establish whether it could apply its own doctrine of abuse of rights without undermining the effectiveness of the Second Directive in the circumstances; it was not seeking to craft, out of Greek law, an autonomous abuse of rights doctrine.

D.  Application of national abuse mechanisms to asymmetric clauses by reference to the principle of effectiveness

6.94  Discerning to what extent abus du droit and Missbrauch can apply to asymmetric jurisdiction clauses governed by the Recast or 2007 Lugano Convention is challenging. Under each of these doctrines, the yardstick for impropriety in the use of an asymmetric clause is different. For abus de droit, the yardstick would be the option holder’s intention or the objective purpose of the option. French authors refer to the ‘more or less subjective’ requirement of an ‘intention to harm’,396 of ‘vexatious and unreasonable behaviour’397 on the part of the right-holder by reference to good faith or an exercise of the right that is abnormal or contrary to its social purpose.398 In the particular context of contractual powers, relevant at least to broad variants of Rothschild clauses,399 the test is considered to be an exercise of the power that is not consistent with the objective of the power to which the subject of the power has consented.400 For Missbrauch, the yardstick would be the objective purpose of the option. It focuses on the result—an abnormal or excessive exercise of the right, largely regardless of the right holder’s intention.401

6.95  Although the French and German abuse doctrines arrive at equitable results ‘in different ways’402 and, within each of these doctrines, the exact ‘contours’ of abuse are not precisely (p. 245) defined,403 several scenarios, involving asymmetric clauses, can be envisaged in which either of these doctrines might engage. The following analysis predicts how several of the scenarios—referred to in the literature or case law—would likely be treated under French and German abuse doctrines and the extent to which that treatment is consistent with the EU law principle of effectiveness.

1.  Exercise of the option purely for procedural advantages

6.96  Several authors have argued that unilateral forum shopping will be abusive where the claimant chooses a forum principally for some reason other than receiving a judgment on the merits disposing of its claim.404 As Chapter 3 showed, asymmetric clauses may be seen as a guise for unilateral forum shopping by the option holder. Applied to standard variants of Rothschild clauses, this would mean that an option holder which chooses a forum for the attractiveness, absolute or relative,405 of its procedural or substantive law would be acting abusively. If this behaviour were sufficient to constitute an abus du droit under French law,406 or Missbrauch under German law, the principle of effectiveness would nonetheless prevent these doctrines from applying when the Recast or 2007 Lugano Convention governs the jurisdiction clause. Sanctioning unilateral forum shopping would be incompatible with the full effect of those instruments because that is something which they permit.407 Any negative regulatory consequences produced by such clauses, in so far as they encourage unilateral competition within the EU,408 cannot be remedied by asserting ex post that it is abusive to use them.

(p. 246) 2.  Requiring the non-option holder to bring or defend proceedings in a remote location

6.97  Several authors have also suggested that an asymmetric clause might be abused by an option holder if its effect is to require the non-option holder to defend or bring proceedings in a remote location.409 The CJEU has made clear, in commercial cases, that there is no need for proximity in the rules of special jurisdiction above and beyond that provided by the rules themselves.410 It is therefore clear that an option holder insisting that a court that has or would otherwise have special jurisdiction over proceedings brought against the non-optional holder pursuant to the optional limb of a standard Rothschild clause cannot be considered abusive. To consider those proceedings as abusive would infringe the principle of effectiveness, even if that forum is ‘remote’, whether geographically or in terms of its connections with the parties.

6.98  If, instead, the complaint is that the court before which the non-option holder must bring proceedings under a Rothschild clause is remote, it is difficult to see how the option holder’s reliance on the clause could be abusive. Those difficulties were reasonably foreseeable at the time the non-option holder agreed to it because the identity of that court is named in the anchor limb of the clause. Requiring the non-option holder to sue in that court could therefore not be said to be an abnormal or excessive exercise of the option holder’s right. Remoteness also seems difficult to assess: it is not clear, for example, why a relying on a Rothschild clause requiring, in the anchor limb, a Finnish non-option holder to sue a Swedish option holder in Romania is any more abusive than relying on a clause requiring, in the anchor limb, a Finnish non-option holder to sue a Bulgarian option holder in Romania.411 Arguably, it is just as inconvenient for the Swede to defend proceedings in Romania as it is for the Finn to sue there, and the fact that the Swede has an option not to sue in Romania does not change the fact that it has to defend proceedings there.

6.99  Even if relying on a Rothschild clause which requires a Finnish non-option holder to sue a Swedish option holder in Romania could, when viewed in conjunction with other factors (such as the quantum of the claim), be considered abusive, the Finn would, in many cases, nonetheless need to go before a Romanian court to argue that being required to bring proceedings there is abusive. The Finn could not go before a Swedish court to make that argument because EU courts can only control abuses before their own courts.412 Mailhé has made a compelling argument that the CJEU jurisprudence (p. 247) supports that view only in situations in which another Member State court has already been seised.413 The Finnish non-option holder could go before a Swedish court to make that argument only until such a time as the Swedish option holder seises the Romanian anchor court in an effort to have the Finn bring its proceedings there. As has been seen, article 31(2) of the Recast requires a Member State court to stay its proceedings until the exclusively designated anchor court declares that it has no jurisdiction under the agreement.414 Presumably, in accordance with that provision, the contention that the Swede’s reliance on that agreement is abusive in the circumstances is a matter properly for the anchor court.415 That is an obstacle to regulating this problem (if it is a problem) via abuse doctrines.416

6.100  The situation may well be different in situations involving Third States. Assume, for instance, that a German option holder relies on a unilateral indeterminate clause in order to force a French non-option holder to bring proceedings against it in Australia. In circumstances in which neither party has any connection to Australia, the German party would clearly be acting abusively.417

6.101  A similar principle would apply if, on exercise of its option under a Rothschild clause, the German option holder sued the French non-option holder in Australia.418 The non-option holder could seise a competent court under the Recast419 or the anchor court, alleging that having to defend the suit in Australia is an abuse. To make good that claim, the non-option holder would need to show that being required to defend proceedings on the other side of the world is an exercise of the option not consistent with the objective of that option or is an abnormal or excessive exercise of the option holder’s right.420 This scenario, however, seems unlikely to occur in practice: no rational German party would choose to incur the costs and hassle of bringing proceedings in a country where there is no tangible benefit in doing so. In other words, the German party would only bring proceedings in Australia if the non-option holder had significant assets there, in which case, it could hardly be said that the option holder is exercising its option inconsistently with the objective of that option or in an abnormal or excessive way. That said, (p. 248) there may be cases, although rare in practice, where an option holder brings tactical proceedings in a remote Third State location where the non-option holder has no assets, for instance, simply to vex the non-option holder. In these kinds of strategic cases, national abuse-of-rights doctrines could usefully apply to prevent the abuse.

3.  Misuse of concurrent proceedings provisions

6.102  Previous chapters have shown that many standard Rothschild clauses are accompanied by a clause expressly providing for option holders to bring proceedings against the non-option holder in two courts concurrently or in circumstances in which the non-option holder has already brought proceedings in the anchor court. As explained in Chapter 5, in cases internal to the EU and EFTA, concurrent proceedings provisions will generally have no effect.421 Their main field of application is in cases involving an EU Member State/EFTA Lugano State court and a Third State court. It was seen earlier that abus de droit concerns itself with the option holder’s motivations in the use of its right or the social purpose of that right, and Missbrauch concerns itself with the outcome of the use of that right in a certain way. It follows that where an option holder uses a concurrent proceedings provision to see how favourably trial progresses in one set of proceedings in comparison to the other or simply to vex the non-option holder, French and German doctrines would likely consider this to be an abuse. For instance, if an option holder, with such an objective, first brings proceedings in Singapore, under the anchor limb of a Rothschild clause, and then in France, pursuant to the optional limb, it is likely that a French court would consider the bringing of both sets of proceedings to be an abuse.

E.  Conclusion

6.103  Abuse doctrines arising out of the law of the forum can be applied under the Recast or 2007 Lugano Convention to the extent that they are compatible with the principle of effectiveness. Concrete scenarios involving asymmetric clauses in which they will engage are limited. The EU prohibition of abuse of rights, for its part, has limited application so far as asymmetric clauses are concerned. It will control proceedings brought by either party under the anchor part of a Rothschild clause but not proceedings brought by the option holder pursuant to its option.422 Wais has argued that contours of an EU abuse test could indirectly be article 6(1) of the European Convention on Human Rights and article 47 of the Charter of Fundamental Rights.423 As will be explained in Chapter 8, the protection, secured by these articles, already has indirect horizontal effect on EU courts. It does so, however, only to the extent that that protection has not been waived or restricted by the parties via an asymmetric jurisdiction clause itself. Wais’ proposal is a fruitful one because an abuse test framed by these rights would operate regardless of whether those rights have been notionally waived or restricted.

(p. 249) IV.  Conclusion

6.104  Examination of asymmetric jurisdiction clauses, by reference to the enforceability requirements established by the Recast and 2007 Lugano Convention, reveals some uncertainties. To resolve them, the following preliminary questions should be put to the CJEU:

  1. 1.  Must a jurisdiction clause, governed by the Recast or 2007 Lugano Convention, only specify with sufficient precision the objective factors that would enable a court seised to determine whether it has jurisdiction?424 Or must it equally specify with sufficient precision the objective factors that would enable the parties to foresee where they must bring or defend proceedings?

  2. 2.  Which law applies to the characterization of the issues encompassed by the words ‘null and void as to its substantive validity’ in article 25 of the Recast?

  3. 3.  Must the application of the dispositive law, applicable to a jurisdiction clause’s substantive validity, result in the clause being ‘null and void’? Or is it sufficient that this law results in the clause being ineffective or voidable?

  4. 4.  What is the juridical nature of ‘a “jurisdiction clause”, which must be interpreted as an independent concept of EU law’?425

  5. 5.  Can the law applicable to a clause’s substantive validity under art 25 of the Recast, which is selected by the private international law rules of the Member State designated in the clause, be overridden by the overriding mandatory rules or displaced by the ordre public international of either the designated court or the forum?

  6. 6.  Which law, if any, applies to the substantive validity of a jurisdiction clause, governed by article 23 of the 2007 Lugano Convention?

6.105  If the answer to the first question is that the parties must also be able to foresee where they must bring and defend proceedings, the fate of broad variants of Rothschild clauses and unilateral indeterminate jurisdiction clauses—but not standard Rothschild clauses—is uncertain.426 For reasons of consistency, the question of which issues the substantive validity rule should cover ought to be determined autonomously and doctrines of national law that result in an ineffective or voidable clause should be covered. Whether some of those doctrines will bite may turn on the juridical nature of a jurisdiction clause under EU law being contractual (ie, obligational), rather than dispositional, or something else entirely—as Briggs has remarked in a related context, ‘[e]ach view is perfectly tenable; their combination is not’.427 Abuse of rights doctrines will, for the enforceability of asymmetric clauses in cases involving only EU Member States, add little sting.(p. 250)

Footnotes:

1  And the Recast’s predecessor, Regulation 44/2001.

2  An ‘agreement conferring jurisdiction’ is an independent concept under EU law; not ‘a simple reference to the national law of one or other of the States concerned’: Case C–222/15 Hőszig kft v Alstom Power Thermal Services [2016] ECLI:EU:C:2016:525, para 29 (Regulation 44/2001).

3  See, generally, Case C–64/17 Saey Home & Garden NV/SA v Lusavouga-Máquinas e Acessórios Industriais SA [2018] ECLI:EU:C:2018:173, para 23.

4  Hélène Gaudemet-Tallon and Marie-Élodie Ancel, Compétence et exécution des jugements en Europe (6th edn, LGDJ 2018) para 166; Case C–595/17 Apple Sales International and others v MJA [2018] ECLI:EU:C:2018:541, Opinion of AG Wahl (opining at para 37 that ‘the fact that the jurisdiction clause … is asymmetrical … cannot in itself be a relevant factor in the assessment of … that clause in the light of the requirements laid down in Article 23 of Regulation No 44/2001’ and contrasting that provision in fn 14 with art 25 of the Recast, which refers the substantive validity of a jurisdiction agreement to national law (emphasis added)); Alex Mills and Uglješa Grušić, ‘Jurisdiction under the Brussels/Lugano System’ in Paul Torremans and James J Fawcett (eds), Cheshire, North & Fawcett: Private International Law (15th edn, OUP 2017) 187, 235, fn 379; Trevor C Hartley, Civil Jurisdiction and Judgments in Europe: The Brussels I Regulation, the Lugano Convention, and the Hague Choice of Court Convention (OUP 2017) para 13.135; cf Carmen Otero García-Castrillón, ‘Legal Certainty and Predictability in the EUPILLAR Project’s Regulations: An Assessment’ in Paul Beaumont and others (eds), Cross-Border Litigation in Europe (Hart Publishing 2017) 583, 596 (stating that it is unclear ‘whether asymmetry is a question of substantive validity or a question of what constitutes a choice of court agreement falling within Article 25 of the Recast’) (emphasis added); LIC Telecommunications Sarl v VTB Capital plc [2019] EWHC 1747 (Comm) [254].

5  As explained earlier, ‘enforceability’ refers to the question of whether a clause is invalid, void, vitiated, or otherwise liable to be constrained, set aside, or réputée non écrite (‘deemed unwritten’) for want of compliance with those requirements.

6  Set out in paras 1.37–1.54.

7  These questions are listed in para 6.105.

8  See para 5.02.

9  By reference to the typology of asymmetric clauses in paras 1.37–1.54, this means that the Recast will govern the enforceability of a standard, narrow, or broad Rothschild clause of which the anchor limb designates an EU Member State’s courts; of a unilateral non-exclusive clause which designates an EU Member State’s courts; and of an indeterminate unilateral jurisdiction clause, if the court which the option holder chooses to designate at the time of a dispute is that of an EU Member State. The 2007 Lugano Convention will govern in the same fashion where an EFTA Lugano State’s courts are designated and at least one party is domiciled in an EU Member State or EFTA Lugano State. A possible exception is where the agreement confers jurisdiction on an EU Member State and the defendant to the proceedings is domiciled in an EFTA Lugano State. In that case, it is unsettled whether an EU Member State court should apply the Recast or the 2007 Lugano Convention. National private international law rules of the EU Member States will govern the enforceability of asymmetric clauses that confer jurisdiction on Third State courts: Gulf International Bank BSC v Aldwood [2019] EWHC 1666 (QB), [2020] 1 All ER (Comm) 334 [125] referring to Case C–387/98 Coreck Maritime GmbH v Handelsveem BV [2000] ECR I–9362, para 19 (Brussels Convention) (although using the term ‘validity’).

10  See para 5.02.

11  ibid.

12  See Andrew Dickinson, ‘Realignment of the Planets—Brexit and European Private International Law’ (2021) 41 Praxis des internationalen Privat—und Verfahrensrechts (IPRax) 213, 214.

13  Other requirements, applicable to jurisdiction agreements generally, will not be discussed. On the requirement, under both instruments, that the transaction have a cross-border element, see, generally, Ulrich Magnus, ‘Introduction’ in Ulrich Magnus and Peter Mankowski (eds), European Commentaries on Private International Law, vol 1 (Otto Schmidt 2016) Brussels Ibis Regulation, art 25, paras 23–26; Francisco Garcimartín, ‘Prorogation of Jurisdiction’ in Andrew Dickinson and Eva Lein (eds), The Brussels I Regulation Recast (OUP 2015) 277 paras 9.30–9.31; Horatia Muir Watt, ‘Le principe d’autonomie entre libéralisme et néolibéralisme’ in Marc Fallon, Paul Lagard, and Sylvaine Peruzzetto (eds), La matière civile et commerciale, socle du code européen de droit international privé (Dalloz 2009) 77, 78. On the requirement, under both instruments, that there be ‘a particular legal relationship’ between the parties which also affects the scope of disputes to which the agreement may apply, see, generally, Adrian Briggs, Agreements on Jurisdiction and Choice of Law (OUP 2008) paras 7.88–7.92; Gaudemet-Tallon and Ancel (n 4) para 156. On the effects of the jurisdiction agreement on third parties, see Case C–214/89 Powell Duffryn plc v Wolfgang Petereit [1992] ECR I–1769; Case C–543/10 Refcomp SpA v Axa Corporate Solutions Assurance SA [2013] ECLI:EU:C:2013:62; Magnus, ‘Introduction’ (n 13) art 25, paras 160–62; Richard Fentiman, International Commercial Litigation (2nd edn, OUP 2015) paras 2.159–2.179. On the 2007 Lugano Convention requirement, abolished in the Recast, that at least one party must be domiciled in an EU Member State or EFTA Lugano State in order for the Convention to govern the jurisdiction agreement, see Fausto Pocar, Explanatory Report to the Convention on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters [2009] OJ 319/1, paras 104–05. It is unsettled whether domicile should be assessed at the time a dispute arises or at the time the agreement is concluded: see Andreas Bucher, in Andreas Bucher (ed), Commentaire Romand : Loi sur le droit international privé/Convention de Lugano (1st edn, Helbing Lichtenhahn 2011) art 23, paras 3–5. Though that uncertainty may be less material than first appears: Adrian Briggs, Civil Jurisdiction and Judgments (7th edn, Informa Law 2021) para 12.10.

14  A typology of asymmetric clauses can be found in paras 1.37–1.54.

15  In this respect, the Recast differs materially from art 23 of Regulation 44/2001 which preceded it, and the corresponding provision of the 2007 Lugano Convention, both of which contain no choice-of-law rule applicable to an agreement’s substantive validity.

16  See François Mailhé, ‘France’ in Mary Keyes (ed), Optional Choice of Court Agreements in Private International Law (Springer 2020) 197, 211–12.

17  See para 3.111.

18  Hőszig (n 2) para 43.

19  Coreck (n 9) ruling para 1 (Brussels Convention); referred to with approval in Hőszig (n 2) para 43 (Regulation 44/2001).

20  See Jean-Baptiste Racine, ‘Les clauses d’élection de for asymétriques’ in Marie-Élodie Ancel and others (eds), Le droit à l’épreuve des siècles et des frontières : Mélanges en l’honneur du Professeur Bertrand Ancel (IPROLEX 2018) 1323, 1339.

21  Coreck (n 9) para 5 (Brussels Convention).

22  Hőszig (n 2) para 45 (Regulation 44/2001).

23  See, generally, para 5.07.

24  But see Mailhé, ‘France’ (n 16) 210.

26  The point is unclear; see para 5.97–5.98.

27  Coreck (n 9) para 5 (Brussels Convention); Hőszig (n 2) para 45 (Regulation 44/2001).

28  Which was the clause in (Cour d’appel de Toulouse, 26 June 2017) No 17/00315 (author’s translation) and in issue on appeal in Aquitaine caravanes v Knaus Tabbert GmbH (Cour de cassation, First Civil Chamber, 11 July 2019) No 18-11.456, Non publié au bulletin [Not publicized by the Court]. Narrow variants of Rothschild clauses are explained further in paras 1.46–1.47.

29  Aquitaine (n 28) (author’s translation).

30  This was the wording of the optional limb of the clause in issue in Diemme v Chambon (Cour de cassation, Commercial Chamber, 11 May 2017) No 15-18.758, Non publié au bulletin [Not publicized by the Court] (author’s translation). The Court below had impugned the clause on the basis that it was insufficiently precise; a decision which the Commercial Chamber of the Cour de cassation overturned. Although, in doing so, the Court did not refer expressly to the requirement of precision of content, Ancel and Marion state that it was a central element of the Rapporteur’s analysis which led to the Court’s decision: Marie-Élodie Ancel and Léa Marion, ‘Clause attributive de juridiction (ou clause d’élection de for)’ (2018) 145 Journal du droit international 1166, 1172. For further reasons given by the Court, see para 6.72.

31  See further the typology in paras 1.40–1.42.

32  ICH v Crédit Suisse (Cour de cassation, First Civil Chamber, 25 March 2015) No 13-27.264, Publié au bulletin [Publicized by the Court] (author’s translation). See Brooke Adele Marshall, ‘Imbalanced Jurisdiction Clauses under the Lugano Convention’ [2016] Zeitschrift für Europäisches Privatrecht 515.

33  ICH v Crédit Suisse (n 32).

34  Crédit Suisse v ICH II (Cour de cassation, First Civil Chamber, 7 February 2018) No 16-24.497, Non publié au bulletin [Not publicized by the Court] (hereafter ‘ICH v Crédit Suisse II’); Ancel and Marion (n 30) 1166.

35  See, eg, Hannes Wais, ‘Einseitige Gerichtsstandsvereinbarungen und die Schranken der Parteiautonomie’ (2017) 81 Rabels Zeitschrift für ausländisches und internationales Privatrecht 815, 830. The same could be said of a Rothschild clause which provides for the option holder ‘to bring proceedings in any other court which would normally have competence with respect to the Guarantee and the Guarantor’, as the Cour d’appel de Luxembourg has held: CA Luxembourg (Comm) 7 December 2016, Pasicrisie luxembourgeoise 2017, 195 (note), cited in Ancel and Marion (n 30) 1179 (author’s translation).

36  As explained in para 2.07, exorbitant grounds of jurisdiction are those which found the jurisdiction of a court, generally based on tenuous connections with the parties or their dispute.

37  See Recast, art 6 and the discussion in para 5.64. As explained there, this argument turns on art 6(1) being subject to art 25, and the parties’ art 25 agreement referring back to the national rules of private international law of the Member States contemplated by art 6 in the case of an agreement with non-exclusive effect for the option holder. See also 2007 Lugano Convention, arts 4, 23.

38  Recast, art 5; 2007 Lugano Convention, art 3.

39  Perella Weinberg Partners UK LLP v Codere SA [2016] EWHC 1182 (Comm) [33].

41  See Marie-Élodie Ancel, Léa Marion, and Laurence Wynaendts, ‘Réflexions sur les clauses de juridiction asymétriques : A propos de Cass 1re Civ., 26 septembre 2012’ (2013) 148 Banque et Droit 3, 5.

42  Broad variants of Rothschild clauses are described in paras 1.43–1.45. English courts have interpreted these as referring to any court of competent jurisdiction, despite their broad language. If EU Member State courts adopt that interpretation, then the clause will contain objective rather than subjective factors. If they do not, then such a clause indeed contains subjective factors and will fail under EU law; cf the likely position under common law principles of a clause containing subjective factors: see paras 7.16–7.19. See also MJA (mandataire judiciaire de la société Ebizcuss.com) v Apple Sales International (Cour de cassation, First Civil Chamber, 7 October 2015) No 14-16.898, Publié au bulletin [Publicized by the Court]; François Mailhé, ‘Licéité d’une clause attributive de compétence asymétrique’ (19 October 2015) 43 La Semaine Juridique—Édition Générale 1123 (note). In that case, the Cour de cassation held that a Rothschild clause providing for the option holder to sue ‘in the courts having jurisdiction in the place where Reseller has its seat or in any jurisdiction where harm to Apple is occurring’ allows for the identification of the jurisdictions before which litigation would be brought. Whether the italicized criterion in the clause refers to identifiable courts is doubtful. The formulation of this criterion seems to contemplate proceedings, including concurrent proceedings, in any jurisdiction where harm to Apple is occurring anywhere in the world. The Court’s reasoning may, nonetheless, be justified on the basis that the reference in the second part of the clause to ‘any jurisdiction where harm to Apple is occurring’ may be analogized to the default rule under the now art 7(2) of the Recast for matters brought in tort or relating to delict or quasi-delict. See, generally, Case C–12/15 Universal Music International Holding BV v Michael Tétreault Schilling [2016] ECLI:EU:C:2016:449, paras 27, 34–35, 40. If such a criterion is sufficiently narrow to allocate jurisdiction under the default rules, such a criterion must be sufficiently precise when incorporated into an asymmetric jurisdiction clause. See also Ancel and Marion (n 30) 1174.

43  Gaudemet-Tallon and Ancel (n 4) 188, para 140.

44  Wais (n 35) 830 (arguing that a clause which allows a non-option holder to be sued before ‘any court’ is consistent with this requirement because the agreed courts are those of every state. For completeness, it should be noted that Wais understands ‘any court’ to be any Member State court).

45  See, eg, ‘Place of jurisdiction shall be, at our option, F and N respectively, I or the buyer’s place of business’: OLG Brandenburg, 26 June 2012, 6 U 3/11, BeckRS 2012, 15696 (author’s translation). (Obviously, ‘F’, ‘N’, and ‘I’ have been anonymized.)

46  See, eg, under which the non-option holder ‘accepts … any court—in England or elsewherethat the [option holder] may designate’, which was the clause in issue in ‘Cour d’appel de Paris, 5 July 1989, No 89-4373’ (1990) 117 Journal du droit international 152 (note) (author’s translation).

47  See also Édouard Treppoz, ‘L’imprévisibilité du juge élu’ in Malik Laazouzi (ed), Les clauses attributives de compétence internationale : de la prévisibilité au désordre (Éditions Panthéon-Assas 2021) 91, 91–94, 99–105.

48  Coreck (n 9); François Mailhé, ‘Les clauses attributives de compétence asymétriques dans les relations d’affaires’ (2018) 1 Revue de droit international d’Assas 422, 429.

49  Madame X v Banque Privée Edmond de Rothschild (Cour de cassation, First Civil Chamber, 26 September 2012) No 11-26.022, Publié au bulletin [Publicized by the Court] (hereafter the ‘Rothschild case’); ICH v Crédit Suisse (n 32); ICH v Crédit Suisse II (n 34); Société Saint Joseph v Dexia banque internationale (Cour de cassation, First Civil Chamber, 3 October 2018, No 17-21.309) Publié au bulletin [Publicized by the Court] (author’s translation); François Mailhé, ‘Une asymétrie peut en cacher une autre’ (2018) 37 La Semaine Juridique—Édition Générale 2233 (note) (underscoring the weight placed by the Cour de cassation in Dexia on recital 11 of Regulation 44/2001). For the notes d’arrêt accompanying each of these see para 1.01.

50  See para 6.08. But see Treppoz (n 47) 92, 99.

51  Diemme (n 30); Ebizcuss (n 42).

52  Ancel and Marion (n 30) 1172. See also Racine (n 20) 1323, 1332–33; Mary Keyes and Brooke Adele Marshall, ‘Jurisdiction Agreements: Exclusive, Optional and Asymmetrical’ (2015) 11 Journal of Private International Law 345, 372.

53  Case C–452/12 Nipponkoa Insurance Co (Europe) Ltd v Inter-Zuid Transport BV [2013] ECLI:EU:C:2013:858, para 36; Case C–157/13 Nickel & Goeldner Spedition GmbH v ‘Kintra’ UAB [2014] ECLI:EU:C:2014:2145, paras 38, 41; Recast, recital 15.

54  (Cour d’appel d’Aix-en-Provence, Second Civil Chamber, 17 July 2014) No 12.05792 (Regulation 44/2001) (author’s translation); note Philippe Delebecque, ‘Clauses de compétence : valorisation constante et extension du cercle des bénéficiaires’ [2015] Le Droit maritime français 1015.

55  (Cour d’appel d’Aix-en-Provence, Second Civil Chamber, 17 July 2014) (n 54) (emphasis added); note Delebecque (n 54).

56  Louise Merrett, ‘The Future Enforcement of Asymmetric Jurisdiction Agreements’ (2018) 67 International and Comparative Law Quarterly 37, 54. See further paras 5.22–5.23.

57  Diemme (n 30).

58  ICH v Crédit Suisse II (n 34).

59  Dexia (n 49).

60  Ancel and Marion (n 30) 1173, 1176.

61  See Recast, recital 19. See, generally, Peter Mankowski, ‘The Role of Party Autonomy in the Allocation of Jurisdiction in Contractual Matters’ in Franco Ferrari and Francesca Ragno (eds), Cross-Border Litigation in Europe: The Brussels I Recast Regulation as a Panacea ? (Casa Editrice Dott. Antonio Milani 2016) 97, 103, 115–16.

62  Mankowski, ‘The Role of Party Autonomy’ (n 61) 118. But see Pippa Rogerson, ‘Plus ça change ? Article 5(1) of the Regulation on Jurisdiction and the Recognition and Enforcement of Judgments’ (2000) 3 Cambridge Yearbook of European Legal Studies 383, 396–97.

63  Case C–519/19 Ryanair DAC v DelayFix [2020] ECLI:EU:C:2020:933, para 38 (Recast).

64  Apple Sales (n 4) para 34 (Regulation 44/2001).

65  Refcomp (n 13) Opinion of AG Jääskinen, paras 44, 48 (Regulation 44/2001) (references omitted).

66  See Andrew Dickinson, ‘Background and Introduction to the Regulation’ in Dickinson and Lein (n 13) 1, para 1.64.

67  See Case C–440/97 GIE Groupe Concorde v Master of the vessel ‘Suhadiwarno Panjan’ [1999] ECR I-6307, paras 23–24 (Brussels Convention).

68  See the Rothschild case (n 49) (Regulation 44/2001); ICH v Crédit Suisse (n 32) (2007 Lugano Convention); LG Ulm, 31 July 2014, 4 O 66/13, unreported, 13–14 (a case involving a consumer) (2007 Lugano Convention); Carine Brière, ‘Note’ (2013) 140 Journal du droit international 175, 178.

69  See Case C–337/17 Feniks sp. z o.o. v Azteca Products & Services SL [2018] ECLI:EU:C:2018:805, para 34.

70  See para 5.24.

71  See para 6.08.

72  Though, of course, some of the grounds on which the default rules of jurisdiction are based rely on a factual connection that is established by the non-option holder’s choice, eg, domicile. Default rules are discussed generally in paras 2.04–2.08.

73  See Ancel and Marion (n 30) 1174.

74  See, eg, Mailhé, ‘Une asymétrie’ (n 49) 2237.

75  Recast, art 5(1).

76  ibid art 6.

77  That said, to the extent that the clause could be seen as a further gateway to those rules, giving effect to it under EU law may be incompatible with the Charter of Fundamental Rights of the EU; see para 8.14.

78  Whether the courts of those Third States will exercise jurisdiction may turn on other factors, such as whether the jurisdiction clause is accompanied by a concurrent proceedings provision: see, generally, paras 7.86–7.119.

79  See para 5.62.

81  ie, those which are construed as conferring jurisdiction on whichever court where the option holder chooses to bring proceedings: see para 6.18.

82  See, eg, under which the non-option holder ‘accepts … any court—in England or elsewherethat the [option holder] may designate’, which was the clause in issue in Cour d’appel de Paris, 5 July 1989 (n 46) (author’s translation).

83  For a clause example, see para 6.09.

84  Again, the principle of continuity, explained in para 1.36, means that the Recast and the 2007 Lugano Convention are to be interpreted identically to the Brussels Convention and the Regulation 44/2001 in respect of this requirement. The form requirements were relaxed in successive versions of the Brussels Convention. For an overview, see Carlos Alberto Arrue Montenegro, L’autonomie de la volonté dans le conflit de juridictions (LGDJ 2011) paras 84–86. The form requirements in the consolidated version of the Brussels Convention are the same as those in the Recast and the 2007 Lugano Convention, with the sole difference that the Recast and the 2007 Lugano Convention clarify, in a separate paragraph, that electronic communications are equivalent to writing. See Diego P Fernández Arroyo, ‘The Limits to the Parties’ Free Choice of Jurisdiction: Formal and Territorial Requirements for the Efficacy of Choice-of-Court Agreements’ in Georges Affaki and Horacio Grigera Naón (eds), Jurisdictional Choices in Times of Trouble (International Chamber of Commerce 2015) 71, 72–73.

85  Calyon v Wytwornia Sprzetu Komunikacynego PZL Swidnik SA [2009] EWHC 1914 (Comm) [8], [13].

86  See Paul Beaumont and Burcu Yüksel, ‘The Validity of Choice of Court Agreements under the Brussels I Regulation and the Hague Choice of Court Agreements Convention’ in Katharina Boele-Woelki and others (eds), Convergence and Divergence in Private International Law—Liber Amicorum Kurt Siehr (Eleven International Publishing 2010) 564, 565–71.

87  See Saey (n 3) para 25 (Recast); Case C–436/16 Georgios Leventis v Malcon Navigation Co Ltd [2017] ECLI:EU:C:2017:497, para 34 (Regulation 44/2001); Case 71/83 Partenreederei ms Tilly Russ v NV Haven & Vervoerbedrijf Nova [1984] ECR 2418, para 15 (Brussels Convention). See also Briggs, Agreements (n 13) paras 7.10, 7.12, 7.17, 7.22, 7.25. On the issue of third-party consent to a jurisdiction agreement, not considered here, see the sources cited in n 13.

88  Recast, art 25(1)(a); 2007 Lugano Convention, art 23(1)(a). ‘Evidenced in writing’ refers to an oral agreement on jurisdiction that the parties have recorded in writing: Partenreederei (n 87) para 15; see, eg, BGH NJW 2019, 1300, paras 28–31.

89  Recast, art 25(1)(b)(c); 2007 Lugano Convention, art 23(1)(b)(c). Art 25(1)(c) of the Recast and art 23(1)(c) of the 2007 Lugano Convention provide:

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 [emphasis added].

90  Recast, art 25(2); 2007 Lugano Convention, art 23(2). See, generally, Case 322/14 Jaouad El Majdoub v CarsOnTheWeb.Deutschland GmbH [2015] ECLI:EU:C:2015:334 (Regulation 44/2001).

91  Case 24/76 Estasis Salotti di Colzani Aimo e Gianmario Colzani snc v Rüwa Polstereimaschinen GmbH [1976] ECR 1831, para 9; BNP Paribas SA v Anchorage Capital Europe LLP [2013] EWHC 3073 (Comm) [44]; Mills and Grušić (n 4) 239.

92  See Case C–366/13 Profit Investment SIM SpA v Stefano Ossi [2016] ECLI:EU:C:2016:282, para 27 (Regulation 44/2001). See also Case C–159/97 Trasporti Castelletti Spedizioni Internazionali SpA v Hugo Trumpy SpA [1999] ECR I–1636, para 48.

93  Briggs, Agreements (n 13) para 7.22; Mastermelt Ltd v Siegfried Evionnaz SA [2020] EWHC 927 (Comm) [87]. Where the jurisdiction agreement is not in writing or evidenced in writing (as per sub-clause (a)), and is instead reflective of a practice which the parties had used in the past (as per sub-clause (b)) or an industry practice of which the parties ought to have been aware (as per sub-clause (c)), the second element of actual consensus in fact is, oxymoronically, presumed to exist: see Case C–106/95 Mainschiffahrts-Genossenschaft eG (MSG) v Les Gravières Rhénanes SARL [1997] ECR I–911, para 19 and (Cour de cassation, First Civil Chamber, 20 November 2019) No 18-21.854, Non publié au bulletin [Not publicized by the Court]; Benjamin Remy, ‘À quelles conditions le paiement d’une facture renvoyant à une clause d’élection de for contenue dans des conditions générales vaut-il acceptation de celle-ci ?’ (2020) Revue critique de droit international privé 317 (note). See especially the text following fn 13 in Remy’s note.

94  Saey (n 3) para 25.

95  Case 221/84 F Berghoefer GmbH & Co KG v ASA SA [1985] ECR 2704, para 13. See also Refcomp (n 13) Opinion of AG Jääskinen, para 41; Aspen Underwriting Ltd v Credit Europe Bank NV (The Atlantik Confidence) [2020] UKSC 11 [24].

96  Fentiman, International Commercial Litigation (n 13) para 2.87. Magnus, ‘Introduction’ (n 13) art 25, para 95 (on electronic signatures).

97  For examples, see Magnus, ‘Introduction’ (n 13) art 25, para 99.

98  See, eg, Hőszig (n 2) paras 39–41; Saey (n 3) para 27. See also para 6.28.

99  Hőszig (n 2) para 36 (Regulation 44/2001).

101  2007 Lugano Convention, arts 13, 17, 21; Recast, arts 15, 17, and 19.

102  In none of the cases in which the CJEU stated that one of the aims of art 23(1) of Regulation 44/2001 and of art 17 of the Brussels Convention was to protect the weaker party was a consumer, employee, or insured involved. See Hőszig (n 2) paras 10–15, 36 (manufacturer and buyer; jurisdiction clause contained in manufacturer’s standard terms and conditions); Trasporti (n 92) paras 2, 19 (commercial consignee and agent for the ship and carrier; jurisdiction clause contained in the bill of lading issued by the original shipper); Mainschiffahrts-Genossenschaft (n 93) paras 2, 6, 17 (ship owner and charterer; pre-printed jurisdiction clause contained in a commercial letter of confirmation sent by the ship owner).

103  Saey (n 3); Louise Merrett, ‘Orally Agreed Jurisdiction Agreements under the Brussels I Regulation Recast’ (2018) 77 Cambridge Law Journal 472, 473–74.

104  See Case 150/80 Elefanten Schuh GmbH v Pierre Jacqmain [1981] ECR 1671, para 24:

formal requirements were inserted out of the concern … to cancel out the effects of clauses in contracts which might go unread … if they were not agreed to by the party against whom they operate. For those reasons, jurisdiction clauses should be taken into consideration only if they are the subject of a written agreement, and that implies the consent of all the parties [emphasis added].

Note, though, that there is an inconsistent use of the terms ‘consent’ and ‘consensus’ in the jurisprudence of the CJEU; eg, in both CarsOnTheWeb (n 90) paras 26–30, and Refcomp (n 13) paras 26–28 the Court deduced the requirement of ‘consensus’ from the wording of art 23(1) and Recital 11 of Regulation 44/2001 and from CJEU case law on art 17 of the Brussels Convention. From that requirement, the Court inferred that an aim of art 23 of Regulation 44/2001 is, like art 17 of the Brussels Convention, ensuring consent. See also Profit Investment (n 92) para 27. The CJEU’s use of the terms in earlier case law is equally inconsistent. In Case 201/82 Gerling Konzern Speziale Kreditversicherungs-AG and others v Amministrazione del Tesoro dello Stato [1984] ECR 2504, the Court referred exclusively to consent. In Salotti (n 91) and Case 784/79 Porta-Leasing GmbH v Prestige International SA [1980] ECR 1517, the Court referred exclusively to consensus.

105  Either as its standard form contract or as memorializing the terms of the party’s oral agreement.

106  cf BGH, 22 February 2001, IX ZR 19/00, NJW 2001, 1731, 1732 (1988 Lugano Convention) (holding that a jurisdiction clause in a bank’s standard terms signed by the surety, but not by the bank, could not be relied upon by the bank) with BGH, 25 January 2017, VIII ZR 257/15, Beck RS 2017, 102068 (2007 Lugano Convention) (holding that a contract which had been negotiated by both parties but signed only by the party seeking to contest the jurisdiction clause was enforceable against that party, if the other party commenced performance soon after the contract was signed) and BGH, 26 April 2018, VII ZR 139/17, BeckRS 2018, 9534 (Recast) (holding that a contract drafted by only one party and orally accepted by the other party was not enforceable against that party); Peter Mankowski, ‘BGH: Anforderungen an die Behauptung eines Handelsbrauchs’ (2018) Lindenmaier-Möhring—Kommentierte BGH-Rechtsprechung 406706. See also IMS SA v Capital Oil and Gas Industries Ltd [2016] EWHC 1956 (Comm) [43]–[52] (holding that the consent of the party seeking to contest jurisdiction based on the clause, evidenced by that parties’ signature on a deed, did not establish consensus-in-fact in circumstances where the party seeking to rely on the jurisdiction clause had not signed it).

107  See, eg, IMS SA (n 106) [43]–[52]. cf Adrian Briggs, ‘Private International Law’ in Andrew S Burrows (ed), English Private Law (3rd edn, OUP 2013) 1527, para 20.51 (stating that ‘there is no question whether the claimant agreed, as he [or she] has issued the writ’).

108  Salotti (n 91). For French decisions applying Salotti, see Gaudemet-Tallon and Ancel (n 4) 193–94, para 146, fn 74.

109  Salotti (n 91) para 9. In Case 25/76 Galeries Segoura SPRL v Société Rahim Bonakdarian [1976] ECR 1851, the CJEU reached the same conclusion but in circumstances where the jurisdiction clause was in the seller’s standard terms, which the seller gave in a confirmation document to the buyer, to which the buyer did not respond, after the oral contract of sale had been concluded and upon completion of the seller’s obligation under the contract: paras 2–3. The jurisdiction agreement was not the subject of the oral contract between the parties: para 10.

110  Briggs, Agreements (n 13) para 7.44.

111  Salotti (n 91) paras 9, 12.

112  cf Etihad Airways PJSC v Flöther [2019] EWHC 3107 (Comm) [212] (where Jacobs J remarked that ‘[t]he court in Salotti was not considering asymmetric clauses, or anything close to them’).

113  Salotti (n 91) 1833. Narrow variants of Rothschild clauses are described in paras 1.46–1.47.

114  Article 17(4) provided: ‘If an 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.’ As explained in para 1.27, in the consolidated version of the Brussels Convention, it appears as sub-para 5. For a detailed discussion of this provision, see paras 1.25–1.36.

115  The 2007 Lugano Convention has not been amended to reflect the choice-of-law rule introduced by the Recast.

116  See para 6.25. See also Jonathan Harris, ‘Agreements on Jurisdiction and Choice of Law: Where Next?’ [2009] Lloyd’s Maritime and Commercial Law Quarterly 537, 539.

117  Joint Stock Co Aeroflot—Russian Airlines v Berezovsky [2013] EWCA Civ 784 [51]–[65]; Provimi Ltd v Roche Products Ltd [2003] EWHC 961, [2003] 2 All ER (Comm) 683 [59], [82]; Louise Merrett, ‘The Enforcement of Jurisdiction Agreements within the Brussels Regime’ (2006) 55 International and Comparative Law Quarterly 315, passim; Fentiman, International Commercial Litigation (n 13) para 2.113; Jérôme Barbet, ‘Invalidité en droit européen d’une clause optionnelle unilatérale offrant la faculté à une seule partie d’agir devant les juridictions de deux États différents, ainsi que devant « tout autre tribunal competent »’ [2013] Cahiers de l’arbitrage 443, 447–48; Bernhard Berger, ‘Art 23, Titel II: Zuständigkeit’ in Christian Oetiker and Thomas Weibel (eds), Basler Kommentar: Lugano-Übereinkommen (2nd edn, Helbing Lichtenhahn Verlag 2016) para 23.

118  Joint Stock Co Aeroflot (n 117) [63] (Aikens LJ, with whom Mann J and Laws LJ agreed) (the law of the substantive contract generally being, from the perspective of English conflict-of-laws rules, the law governing the jurisdiction agreement).

119  Fentiman, International Commercial Litigation (n 13) para 2.113; Thalia Kruger, Civil Jurisdiction Rules of the EU and Their Impact on Third States (OUP 2008) paras 4.27–4.32; Jan Kropholler and Jan von Hein, Europäisches Zivilprozessrecht (Verlag Recht und Wirtschaft GmbH 2011) para 28.

120  Briggs, Civil Jurisdiction (n 13) para 12.20.

121  See Adrian Briggs, ‘Choice of Forum and Submission to Jurisdiction’ in Jürgen Basedow and others (eds), Encyclopedia of Private International Law (Edward Elgar Publishing 2017) 304, 308.

122  Article 25(5), first sentence. See Deutsche Bank AG v Asia Pacific Broadband Wireless Communications Inc [2008] EWCA Civ 1091 [24] (a case involving a Rothschild clause governed by Regulation 44/2001). The doctrine of severability or separability, which this rule reflects, is discussed in para 7.26.

123  Case C–269/95 Benincasa v Dentalkit Srl [1997] ECR I–3767, paras 24–25, 29. See, generally, Adrian Briggs, ‘The Subtle Variety of Jurisdiction Agreements’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 364, 380–81.

124  Recast, art 25(5), sentence 2.

125  See, eg, if a party’s signature was forged or if a party alleges that a person who purportedly signed on its behalf had no authority: Premium Nafta Products Ltd v Fili Shipping Co Ltd [2007] UKHL 40 [17] (obiter) (providing examples of matters which might impeach both a contract and an arbitration agreement contained within it).

126  Article 25(1).

127  Recast, recital 20.

128  Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (Recast)’ COM (2010) 748 final, para 3.1.3.

129  Article 5(1). The rule has a counterpart in art 6(a) of the 2005 Hague Choice of Court Convention, which allows a non-designated court to exercise jurisdiction if the agreement is null and void under the law of the chosen court. This was, in turn, inspired by art II(3) of the New York Convention on arbitration agreements. Art II(3), however, does not require a court seised in respect of a matter of which the parties have made an arbitration agreement to refer to the substantive law or conflict-of-laws rules that the arbitral tribunal would apply. See, generally, Laurence Usunier, ‘Le droit applicable à la validité des clauses attributives de jurisdiction en vertu de l’article 25 § 1 du Règlement Bruxelles I bis’ in Affaki and Naón (n 84) 158, 167.

130  Masato Dogauchi and Trevor C Hartley, ‘Convention of 30 June 2005 on Choice of Court Agreements: Explanatory Report’ in Permanent Bureau of the Conference, Proceedings of the Twentieth Session, Tome III (Intersentia 2010) para 125 (hereafter ‘Hartley/Dogauchi Report on the 2005 Hague Choice of Court Convention (Final)’).

131  See Fentiman, International Commercial Litigation (n 13) paras 2.101–2.102; Stéphanie Francq, ‘Les clauses d’élection de for dans le nouveau règlement Bruxelles Ibis’ in Emmanuel Guinchard (ed), Le nouveau règlement Bruxelles Ibis (Bruylant 2014) 107, 134–35. See also Ronald A Brand and Paul Herrup, The 2005 Hague Convention on Choice of Court Agreements: Commentary and Documents (CUP 2008) 81 (who appear to assume, in the context of the similar rule under the 2005 Hague Choice of Court Convention, that the applicable law itself applies to characterize an issue as falling within or outside of the ‘ “null and void” category’).

132  See Francq (n 131) 137.

133  ibid 135.

134  Magnus, ‘Introduction’ (n 13) art 25, para 81c.

135  Christian Heinze, ‘Choice of Court Agreements, ‘Coordination of Proceedings and Provisional Measures in the Reform of the Brussels I Regulation’ (2011) 75 Rabels Zeitschrift 581, 585; Garcimartín (n 13) para 9.70. But see Peter Mankowski in Thomas Rauscher (ed), EuZPR/EuIPR, Band I: Europäisches Zivilprozess—und Kollisionsrecht (5th edn, Otto Schmidt 2021) art 25 Brüssel Ia-VO, para 60.

136  Fentiman, International Commercial Litigation (n 13) para 2.102 (Fentiman refers to ‘a vitiating factor in the law of the forum’ but it appears from para 2.95 that he meant ‘a vitiating factor in the law of the chosen court’). See also Quim Forner-Delaygua, ‘Changes to Jurisdiction Based on Exclusive Jurisdiction Agreements under the Brussels I Regulation Recast’ (2015) 11 Journal of Private International Law 379, 396 (suggesting that the Spanish-language version indicates that only aspects of the applicable law which result in the agreement being null and void are relevant).

137  Case C–296/95 The Queen v Commissioners of Customs and Excise, ex parte EMU Tabac and others [1998] ECR I–1605, para 36. See, generally, Case C–283/81 CILFIT v Ministero della Sanità [1982] ECR 3415, para 18 (referring to the importance of comparing different language versions of an EU instrument to interpret it).

138  See, eg, duress must be raised by one of the parties to avoid an agreement under § 123(1) BGB. But see also § 142 BGB discussed in Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 60.

139  See Brand and Herrup (n 131) 80 (suggesting, in relation to the 2005 Hague Choice of Court Convention, that it might).

140  Mailhé, ‘Les clauses attributives’ (n 48) 432. This concept is explained at n 315.

141  See Mankowski, ‘The Role of Party Autonomy’ (n 61) 104 (observing that the term gives rise to problematic ‘issues of characterisation’); Fentiman, International Commercial Litigation (n 13) para 2.101; Magnus, ‘Introduction’ (n 13) art 25, paras 81c, 81g.

142  See Heinze (n 135) 585; Ulrich Magnus, ‘Sonderkollisionsnorm für das Statut von Gerichtsstands—und Schiedsgerichtsvereinbarungen’ [2016] Praxis des Internationalen Privat—und Verfahrensrechts 521, 525; Garcimartín (n 13) para 9.69; Fentiman, International Commercial Litigation (n 13) para 2.101. See also Hartley/Dogauchi Report on the 2005 Hague Choice of Court Convention (Final) (n 130) para 126 (suggesting that these issues are intended to be encompassed by the equivalent terms under the 2005 Hague Choice of Court Convention).

143  Fentiman, International Commercial Litigation (n 13) para 2.85 (suggesting that, alternatively, this matter would be governed by the national law of the forum). See also Forner-Delaygua (n 136) 394. But see Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 68 (arguing that in order for the substantive validity rule in the Recast to be consistent with the 2005 Hague Choice of Court Convention (see para 6.41), legal capacity must be excluded from the substantive validity rule).

144  Article 1(2)(a); see Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 67.

145  Which ‘encompasses cases both of error (negligent and innocent misrepresentation) and dolus (fraudulent misrepresentation)’ in continental legal systems: Sebastian Lohsse, ‘Validity’ in Nils Jansen and Reinhard Zimmermann (eds), Commentaries on European Contract Laws (OUP 2018) 649, 650 (‘Introduction before Art 4:101’).

146  Which ‘more or less resembles what in continental laws is dealt with under the concept of violence’: ibid 650, para 2 or ‘threat’: ibid 695, para 1 (‘Art 4:108: Threats’).

147  See Heinze (n 135) 585; Fentiman, International Commercial Litigation (n 13) para 2.101; Garcimartín (n 13) para 9.69; Magnus, ‘Sonderkollisionsnorm’ (n 142) 525. For those authors who took the view that national law could be implied into Regulation 44/2001 (and can be implied into the 2007 Lugano Convention) to regulate defective consent or lack of capacity (see para 6.31) mistake, fraud, duress, threat, and misrepresentation were/are equally considered to be relevant: see Magnus, ‘Introduction’ (n 13) art 23, para 83; Tena Ratković and Dora Zgrabljićrotar, ‘Choice of Court Agreements under the Brussels I Regulation (Recast)’ (2013) 9 Journal of Private International Law 245, 253–54. See also Francq (n 131) 128–30; Garcimartín (n 13) para 9.60; Jonathan Hill and Adeline Chong, International Commercial Disputes: Commercial Conflict of Laws in English Courts (4th edn, Hart Publishing 2010) paras 5.3.38–5.3.40.

148  Ryanair (n 63) para 50.

149  See para 6.25.

150  Usunier (n 129) 162; Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 57.

151  Francq (n 131) 134–35. See also Cyril Nourrissat, ‘L’avenir des clauses attributives de juridiction d’après le Règlement « Bruxelles I bis »’ in Mélanges en l’honneur du Professeur Bernard Audit : Les relations privées internationales (LGDJ 2014) 567, 575.

152  See, eg, Vincent Heuzé, Pierre Mayer, and Benjamin Remy, Droit international privé (12th edn, LGDJ 2019) 255, para 364.

153  Ulrich Drobnig, ‘Substantive Validity’ (1992) 40 American Journal of Comparative Law 635, 637–38 (arguing that in cross-border contracts, four types of invalidity are involved: ‘causes of invalidity which compete with remedies for non-performance’, ‘causes of invalidity rooted in the general law of contract where there is no competition with remedies for non-performance’, ‘causes of invalidity in the general area of civil law outside contract law’, and ‘causes of invalidity located in the area of public policy as well as in the public law regulation of business and business transactions by mandatory rules’).

154  Jan von Hein, ‘Die Neufassung der Europäischen Gerichtsstands und Vollstreckungsverordnung (EuGVVO)’ (2013) 59 Recht der Internationalen Wirtschaft 97, 105 (referring inter alia to ‘Gebundenheit an die Erklärung’ (‘lack of consent’)); Forner-Delaygua (n 136) 393.

155  cf Hőszig (n 2) Opinion of AG Szpunar, para 47; Magnus, ‘Sonderkollisionsnorm’ (n 142) 524 (both arguing against this suggestion).

156  Hőszig (n 2) Opinion of AG Szpunar, para 47, fns 34–36 (adopting Magnus’ argument).

157  See Robert Freitag, ‘Halbseitig ausschließliche Gerichtsstandsvereinbarungen unter der Brüssel I-VO’ in Peter Mankowski and Wolfgang Wurmnest (eds), Festschrift für Ulrich Magnus zum 70. Geburtstag (Sellier 2014) 419, 428.

158  Magnus, ‘Introduction’ (n 13) art 25, para 79a. See also Magnus, ‘Sonderkollisionsnorm’ (n 142) 524; Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 56.

159  Hőszig (n 2) Opinion of AG Szpunar, para 49.

160  Article 5(1) of the 2005 Hague Choice of Court Convention provides: ‘The court or courts of a Contracting State designated in an exclusive choice of court agreement shall have jurisdiction to decide a dispute to which the agreement applies, unless the agreement is null and void under the law of that State.’

161  Masato Dogauchi and Trevor C Hartley, ‘Preliminary Draft Convention on Exclusive Choice of Court Agreements: Explanatory Report’ (2004) HCCH Preliminary Document No 25 of March 2004, para 42:

The application of autonomous rules may have been correct in the context of the Brussels Convention, but it is not correct with regard to the Hague Convention, under which the law of the State in question must decide whether there is consent: the explicit references in various articles to State law clearly indicate this.

Masato Dogauchi and Trevor C Hartley, ‘Preliminary Draft Convention on Exclusive Choice of Court Agreements: Explanatory Report’ (Preliminary Document No 26 of December 2004) in Permanent Bureau of the Conference, Proceedings of the Twentieth Session, Tome III (Intersentia 2010) 167, 183 para 62; Hartley/Dogauchi Report on the 2005 Hague Choice of Court Convention (Final) (n 130) para 94. See also ‘Minutes of the Second Commission: Minutes No 1: Meeting of Tuesday 14 June 2005 (morning)’ in Permanent Bureau of the Conference, Proceedings of the Twentieth Session, Tome III (Intersentia 2010) 566, 572–73, 575; cf Peter Nygh and Fausto Pocar, ‘Preliminary Draft Convention on Jurisdiction and Foreign Judgments in Civil and Commercial Matters: Adopted by the Special Commission’ (2000) HCCH Preliminary Document No 11 of August 2000, 42–43.

162  Hartley/Dogauchi Report on the 2005 Hague Choice of Court Convention (Final) (n 130) paras 94–96 (against which ‘the basic factual requirements of consent’ are to be assessed).

163  Brand and Herrup (n 131) 40–41, 41 fn 1, 79–80; William J Woodward Jr, ‘Saving the Hague Choice of Court Convention’ (2008) 29 University of Pennsylvania Journal of International Law 657, 674; Ronald A Brand, ‘The Evolving Private International Law/Substantive Law Overlap in the European Union’ in Mankowski and Wurmnest (n 157) 371, 377. Australian case law on the International Arbitration Act 1974 (Cth), which implements the New York Convention, on which the 2005 Hague Choice of Court Convention is based, suggests that the question of formation (existence) of the agreement is separate from its validity and will be governed by the law of the forum: Trina Solar (US) Inc v Jasmin Solar Pty Ltd (2017) FCR 1 [129].

164  Paul Beaumont, ‘Hague Choice of Court Agreements Convention 2005: Background, Negotiations, Analysis and Current Status’ (2009) 5 Journal of Private International Law 125, 137–40 (favouring this approach).

165  If a non-nominated court has been seised, in addition to the nominated court, capacity will be assessed by reference both to the law of the court seised (2005 Hague Choice of Court Convention, art 6(b)) and the law of the chosen court via the null and void exception by both the chosen court under art 5(1) and the non-nominated court under art 6(a). If a party has capacity under the law of the non-nominated court but no capacity under the law of the chosen court in the eyes of the non-nominated court, the non-nominated court need not suspend or dismiss its proceedings: Hartley/Dogauchi Report on the 2005 Hague Choice of Court Convention (Final) (n 130) para 150. If the nominated court arrives at a different conclusion as to the capacity of that party applying its own law, the result will be parallel proceedings.

166  See Mankowski, ‘The Role of Party Autonomy’ (n 61) 104.

167  Ryanair (n 63) paras 41, 48–61. See also Saey (n 3) paras 23–29.

168  Case C–25/79 Sanicentral GmbH v Collin [1979] ECR 3423. This case, which concerned an imperative norm of the law of the forum, will be considered at para 6.55.

170  It is uncertain whether Garcimartín (n 13) is concerned with public policy of the applicable law, public policy of the forum, public policy of the chosen court (assuming that the law of the chosen court is, by virtue of its conflict-of-laws rules, not the applicable law), or all three.

171  ibid para 9.69 (making the same point in relation to illegality). Again, it is not clear whether Garcimartín is concerned with all public policy (ie, rules of the ordre public and ordre public international) or one of the two.

172  Mankowski, ‘The Role of Party Autonomy’ (n 61) 105, 106 (referring at 106 to ‘the lois de police of the lex fori prorogati’).

173  Study Group on a European Civil Code and Research Group on EC Private Law (Acquis Group), Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference (DCFR) (Sellier 2009) 214–15, arts II 7:301–7:302

174  Magnus, ‘Sonderkollisionsnorm’ (n 142) 525. See, generally, Gilles Cuniberti, ‘Article 1.4’ in Stefan Vogenauer (ed), Commentary on the UNIDROIT Principles of International Commercial Contracts (PICC) (2nd edn, OUP 2015) 167, para 2. See also Burkhard Hess, ‘Report on the Application of the Regulation Brussels I in the Member States’ in Burkhard Hess, Thomas Pfeiffer, and Peter Schlosser (eds), The Brussels I Regulation 44/2001: Application and Enforcement in the EU (Beck, Hart Publishing, Nomos 2008) 17, para 378.

175  But see Adrian Briggs, ‘The Brussels I bis Regulation Appears on the Horizon’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 157, 161 (‘The rules of substantive law to which reference may be made are neither identified as contractual nor confirmed as being not necessarily contractual’).

176  And the 2007 Lugano Convention.

177  Although characterization is a matter for the lex fori, which is the law of the court seised, the Recast clearly replaces the lex fori. It would be very peculiar, however, if characterization were to be conducted by reference to the applicable law itself.

178  The juridical nature of jurisdiction agreements from a German and French perspective was discussed in paras 2.21–2.23.

179  Magnus, ‘Sonderkollisionsnorm’ (n 142) 524; Ulrich Magnus, ‘Gerichtsstandsvereinbarungen unter der reformierten EuGVO’ in Normann Witzleb and others (eds), Festschrift für Dieter Martiny zum 70. Geburtstag (Mohr Siebeck 2014) 785, 792; von Hein (n 154) 105 fn 159. See also Monica Herranz Ballesteros, ‘The Regime of Party Autonomy in the Brussels I Recast: The Solutions Adopted for Agreements on Jurisdiction’ (2014) 10 Journal of Private International Law 291, 302.

180  Because consideration does not have an overriding mandatory nature and it is not part of the ordre public international, it would only be required if English law were the applicable law. The requirement of consideration is a simple mandatory rule under English law. See, generally, Mindy Chen-Wishart, ‘Regulating Unfair Terms’ in Louise Gullifer and Stefan Vogenauer (eds), English and European Perspectives on Contract and Commercial Law: Essays in Honour of Hugh Beale (Hart 2014) 105, 119.

181  But see Adrian Briggs, ‘One-Sided Jurisdiction Clauses: French Folly and Russian Menace’ [2013] Lloyd’s Maritime and Commercial Law Quarterly 137, 141; Briggs, Agreements (n 13) paras 7.11–7.12 (although not referring to consideration specifically, calling for a distinction under Regulation 44/2001 between a doctrine of national law which ‘ascribes binding force’ to an agreement and the need ‘to ascertain whether there really was, in substance, an agreement’).

183  See, eg, Alfred Walchshöfer, ‘BGH, Urt. v. 3.12.1973, AZ II ZR 91/72—OLG Hamburg’ (1975) 88 Zeitschrift für Zivilprozess 318 (note) (a case involving a Rothschild clause designating the courts of Bangkok in the anchor limb in a bill of lading expressly subject to German law); OLG Bamberg, 31 October 2018, 8 U 73/18, paras 4, 40–41 (a case involving a consumer and a standard Rothschild clause, designating the courts of Luxembourg in the anchor limb in a loan agreement expressly subject to German law).

184  BGH, 29 February 1968, VII ZR 102/65, NJW 1968, 1233.

185  See BGH, 18 March 1997, XI ZR 34/96, NJW 1997, 2885, 2886, para 2(a); Haimo Schack, Internationales Zivilverfahrensrecht (8th edn, Beck 2021) para 557; Magnus, ‘Sonderkollisionsnorm’ (n 142) 531. See also BGH, 25 March 2015, VIII ZR 125/14, NJW 2015, 2584 [49].

186  Ryanair (n 63) para 38 (Recast); Refcomp (n 13) para 21 (Regulation 44/2001); Hőszig (n 2) para 29 (Regulation 44/2001).

187  Refcomp (n 13) para 21 (Regulation 44/2001).

188  Whether an art 25 jurisdiction agreement may also constitute a contract, under the national law of an EU Member State, so that its breach may sound in an award of damages is a related but different issue. That issue—a jurisdiction agreement’s contractual effects—gives rise to two questions, as explained in para 2.19: (a) whether, according to the national law of a Member State, damages are available for breach of a jurisdiction agreement and in what circumstances; and (b) whether the application of that law in the circumstances of the case is compatible with the EU law principles of effectiveness and mutual trust. A decision of the German Bundesgerichtshof concerning an exclusive jurisdiction agreement nominating German courts, governed by the Recast, supports that view. The Bundesgerichtshof relevantly held that (a) the effect of any contractual obligations that may be created by a jurisdiction agreement is not governed by the Recast: BGH, 17 October 2019, III ZR 42/19, NJW 2020, 399 [27]–[29]; (b) the principal of mutual trust is not an obstacle to an award of damages in an extra-EU case: at [30]. That a jurisdiction agreement can have a Janusian nature for the issue of relief does not mean, however, that it can for the issue of substantive validity. In other words, the issue of whether a jurisdiction clause can have both procedural effects under art 25 of the Recast and contractual effects under national law is different from the one being asked in this section. The one asked here is whether the jurisdiction agreement has a contractual or procedural nature within art 25 of the Recast itself, which, in turn, influences the kinds of doctrines relevant to its substantive validity. It is improbable that a jurisdiction agreement can lead a double life within the confines of that article.

189  See Rainer Hausmann, ‘Invalidity of Unilateral Jurisdiction Clauses under Article 23 Brussels I Regulation?’ (2013) 2 European Legal Forum 37, 42 (describing jurisdiction agreements under art 23 of Regulation 44/2001 as having a ‘purely procedural nature’).

190  Benincasa (n 123) para 25 (Brussels Convention); Hőszig (n 2) para 31 (Regulation 44/2001). See also Mario Giuliano and Paul Lagarde, Report on the Convention on the law applicable to contractual obligations [1980] OJ C 282 1, art 1, para 5 (explaining that jurisdiction agreements were excluded from the 1980 Rome Convention because ‘the matter lies within the sphere of procedure and forms part of the administration of justice’).

191  Sanicentral (n 168) para 6.

192  Christoph Reithmann in Christoph Reithmann and Dieter Martiny (eds), Internationales Vertragsrecht (8th edn, Otto Schmidt 2015) para 5.252. See also Jennifer Antomo, Schadensersatz wegen der Verletzung einer internationalen Gerichtsstandsvereinbarung ? (Mohr Siebeck 2017) 443–48 (who makes a compelling case, especially at 444, that a jurisdiction agreement under German law cannot, as conventionally thought, be a Verfügung because it is an agreement which modifies the default jurisdictional rules at some future point in time. It does not have an immediate effect on those rules at the time of its conclusion).

193  Briggs, Civil Jurisdiction (n 13) para 12.20.

194  Where the parties have not chosen a law to govern it: Enka Insaat ve Sanayi AS v OOO Insurance Co Chubb [2020] UKSC 38, [2020] 1 WLR 4117, [170] para (iv).

195  Recast, art 25(1). Assessment of the general merits of this rule is outside the scope of the present analysis; cf Briggs, Private International Law in English Courts (OUP 2014) para 4.206 with Francq (n 131) 132–33 (observing that, at least from the perspective of harmony of solutions, the rule is preferable to the inconsistency that prevailed under Regulation 44/2001 in so far as the substantive validity of a jurisdiction clause nominating France will always be governed by French conflict-of-laws rules).

196  Dominique Bureau and Horatia Muir Watt, Droit international privé, vol 2 (4th edn, Presses universitaires de France 2017) para 869 (observing that art 25(1) of the Recast refers to ‘droit’, not to ‘loi’, which might be translated as ‘system of law’ (droit) and ‘internal law’ (loi)).

197  Recast, recital 20.

198  Ryanair (n 63) paras 49–50.

199  Gilles Cuniberti, ‘La clause attributive de jurisdiction’ in Marie-Élodie Ancel and others (eds), Le banquier luxembourgeois et le droit international privé (LGDJ 2017) 67, 71; cf Weco Projects ApS v Loro Piana [2020] EWHC 2150 (Comm) [98] (suggesting that the relevant renvoi is to the rules of the Recast itself).

200  Article 1(2)(e); Barclays Bank plc v Ente Nazionale di Previdenza [2015] EWHC 2857 (Comm), [2015] 2 Lloyd’s Rep 527, [137].

201  Gaudemet-Tallon and Ancel (n 4) para 160; Usunier, ‘Le droit applicable’ (n 129) 164; Jean-Michel Jacquet, Philippe Delebecque, and Laurence Usunier, Droit du commerce international (4th edn, Dalloz 2021) para 969.

202  loi d’autonomie—it is not entirely clear whether Gaudemet-Tallon is referring to the law chosen by the parties to govern the jurisdiction agreement specifically or the law chosen by the parties to govern the container contract: Gaudemet-Tallon and Ancel (n 4) para 160.

203  ibid para 152; Vincent Heuzé, ‘Faut-il confondre les clauses d’élection de for avec les conventions d’arbitrage dans les rapports internationaux ?’ in Vincent Heuzé, Rémy Libchaber and Pascal de Vareilles-Sommières (eds), Mélanges en l’honneur du Professeur Pierre Mayer (LGDJ 2015) 295, 302.

204  Gaudemet-Tallon and Ancel (n 4) para 160.

205  ibid.

206  See para 6.46.

207  Francq (n 131) 131 (suggesting that the rule may be a ‘règle de conflit de système’); Bureau and Muir Watt (n 196) vol 1, para 222-2 (firmer, stating that it is a règle de conflit de système). See also Hartley (n 4) paras 13.138–13.143; Pietro Franzina, ‘The Substantive Validity of Forum Selection Agreements under the Brussels Ibis Regulation’ in Peter Mankowski (ed), Research Handbook on the Brussels Ibis Regulation (Edward Elgar 2020) 95, 104 (both arguing that double renvoi or the foreign court theory should be applied). But see Cuniberti, ‘La clause attributive de jurisdiction’ (n 199) 75 (stating that at least those aspects of the private international law rules of a Member State which condition the acceptance of a renvoi by the court seised must be excluded).

208  As seen in para 6.43, a number of authors have argued that they should not. It is not necessary for my argument to take any view on this, and I do not do so. Suffice it to say, until the CJEU rules on the question, it appears to be an open one.

209  Francq (n 131) 131–32. See also Mankowski, ‘The Role of Party Autonomy’ (n 61) 106.

210  Case C–135/15 Republik Griechenland v Grigorios Nikiforidis [2016] ECLI:EU:C:2016:774, para 49.

211  Francq (n 131) 132.

212  See para 6.48.

213  This is also the approach which prevails under English rules of private international law and in other common law countries; see para 7.84.

214  See Jürgen Basedow, ‘Exclusive Choice-of-Court Agreements as a Derogation from Imperative Norms’ in Patrik Lindskoug and others (eds), Essays in Honour of Michael Bogdan (Juristförlaget 2013) 15, 19, 20 (emphasis added).

215  Matteo M Winkler, ‘Overriding Mandatory Provisions and Choice of Court Agreements’ in Mankowski, Research Handbook (n 207) 347, 358–60.

216  Mankowski, ‘The Role of Party Autonomy’ (n 61) 106; see para 6.43.

217  Fentiman, International Commercial Litigation (n 13) para 2.101 (last sentence), para 2.107 (last sentence). The quote is at para 2.101; see para 6.43.

218  See Trasporti (n 92), in which the CJEU ruled, at para 51, that the substantive rules of liability for the carrier applicable before the English courts nominated in the clause (and, implicitly, the fact that the choice of the English courts excluded the substantive rules of liability for the carrier applicable before the Italian courts) were not relevant to the validity of the jurisdiction clause under art 17 of the Brussels Convention. The character, imperative or otherwise, of those Italian rules of liability was not discussed.

219  Member State courts have also enforced jurisdiction agreements conferring jurisdiction on Member State courts governed by Regulation 44/2001, both where it was expected that the designated court would apply the overriding mandatory rules of the forum, presumably by exercising the discretion in art 9(3) of the Rome I Regulation, eg (Cour de cassation, First Civil Chamber, 18 January 2017) No 15-26.105, Publié au bulletin [Publicized by the Court] and where it must have been expected that the designated court would not apply the overriding mandatory rules of the forum. A decision of the French Cour de cassation, involving an asymmetric jurisdiction clause, is illustrative of the latter: Diemme (n 30). In that case, the Court enforced the Rothschild clause in issue, which required the non-option holder to bring proceedings in Italy, even though the Italian court could not, by virtue of the Rome I Regulation, apply French overriding mandatory rules on abusive practices to the dispute. French courts have taken a similar approach in cases involving jurisdiction agreements designating Third State courts, enforcing those agreements in circumstances where the Third State court would be unlikely to apply the imperative norms of the forum; cf Monster Cable Products inc v Audio Marketing Services (Cour de cassation, First Civil Chamber, 22 October 2008) No 07-15.823, Publié au bulletin [Publicized by the Court] (2009) 98 Revue critique de droit international privé 69 (note) with BGH, 5 September 2012, VII ZR 25/12, BB 2012, 3103; discussed in Basedow, ‘Exclusive Choice-of-Court Agreements’ (n 214) 15, 16–17 and Matthias Weller, ‘Choice of Court Agreements under Brussels Ia and under the Hague Convention: Coherences and Clashes’ (2017) 13 Journal of Private International Law 91, 104.

220  cf Marshall, ‘Imbalanced Jurisdiction’ (n 32) 524 (where I argued, erroneously, that this case involved an imperative norm relating to the results of the jurisdiction agreement).

221  Sanicentral (n 168). See Briggs, Civil Jurisdiction (n 13) para 12.19. As mentioned in para 1.16, the Recast and 2007 Lugano Convention both contain specific provisions protecting employees.

222  A more complex question is whether the decision in Sanicentral (n 168) applies to jurisdiction agreements governed by the 2007 Lugano Convention, although it does not contain the substantive validity rule introduced by the Recast. It may now be arguable that it is an implicit requirement of art 23 of the 2007 Lugano Convention that a jurisdiction agreement itself—as distinct from its effects—does not infringe the ordre public international or overriding mandatory rules of the forum. Indeed, Weller does not see Sanicentral (n 168) as fatal to his argument that the relevance of imperative norms going to the validity of a jurisdiction agreement itself may be implicit in Regulation 44/2001, though he does not refer to that case: ‘Choice of Court Agreements’ (n 219) 107–09.

223  Recast, art 25(1), recital 20. But see para 6.51, where the possibility was raised that ‘conflict-of-laws rules’ should be understood as all the private international law rules of that Member State.

224  Briggs, Private International Law in English Courts (n 195) para 4.206, fn 282 (referring to Case 23/78 Nikolaus Meeth v Glacetal Sarl [1978] ECR 2134 (Brussels Convention), which involved a non-uniquely exclusive jurisdiction clause). See also Ancel and Marion (n 30) 1181.

225  Racine (n 20) 1338.

226  As seen in para 5.07, the term ‘confer jurisdiction’ is synonymous with ‘prorogate’ or ‘designate’.

227  For a clause example, see para 6.11. As explained in paras 5.78–5.80, the effect of such a clause is to require one party to defend proceedings in a nominated court only if no other competent court is first seised by either party.

228  In the same way that it is unproblematic for symmetric non-exclusive clauses; cf Winkler (n 215) 355 (arguing that the rule ‘is useless when the parties have opted for a non[-]exclusive forum’).

229  See the typology of asymmetric clauses in paras 1.37–1.54.

231  See para 5.62.

232  This view finds marginal support in LIC v VTB Capital (n 4), a case concerning proceedings brought by the non-option holder against the option holder in England, contrary to a standard Rothschild clause designating the courts of Luxembourg in the anchor limb: [245]–[246]. Moulder J observed in obiter at [249] that the non-option holders ‘did not pursue the argument that the clause is null and void as a matter of Luxembourg law’, suggesting that if they had pursued such an argument, which would go to the substantive validity of the clause, it would have fallen to be determined by Luxembourg law as the law of the anchor court named in the clause. It must be said, however, that his Honour did not refer to the relevance of Luxembourgish conflict-of-laws rules to that inquiry.

233  cf Hausmann (n 189) 42 (suggesting that this will only be true where the non-option holder is bringing proceedings; where the option holder is bringing proceedings, the clause’s substantive validity will be determined by the law of whichever court the non-option holder seises) with Garcimartín (n 13) para 9.67 (citing Hausmann in support of the former proposition but not making reference to how the substantive validity of the clause is to be measured where the option holder is bringing proceedings), with Merrett, ‘Future Enforcement’ (n 56) 52 (stating that the substantive validity of the clause will be measured by the law of the anchor court where the option holder is bringing proceedings without making reference to how it will be measured where the non-option holder is bringing proceedings), and with Fabienne Jault-Seseke, ‘Validité de la clause attributive de jurisdiction : un revirement attendu tempéré par de nouvelles exigences’ (2015) 191 Recueil Dalloz 2620, 2622 (suggesting that the substantive validity of a Rothschild clause could be determined by the law of the court which the option holder seises, although acknowledging that this approach would allow for the validity of the clause to wax and wane with the ‘procedural initiatives of the parties’ or rather, I would add, of one party).

234  Loan Market Association, ‘Multicurrency Term and Revolving Facilities Agreement’ (French law version LMA.MTF.09, 14 June 2016) cl 38 (including its footnote) (reprinted with permission).

235  Recast, arts 7(1), 25.

236  As explained at para 6.51, the French conflict-of-laws rule applicable to jurisdiction agreements is unsettled.

237  See para 6.51.

238  Assuming the defendant in either scenario is domiciled in an EU Member State.

240  See para 5.81.

241  OLG Brandenburg, 26 June 2012 (n 45).

242  Cour d’appel de Paris, 5 July 1989 (n 46) (author’s translation).

243  Adapted from the clause in issue in Barclays Bank v Ente (n 200) [30], which referred to English courts rather than French courts in the anchor limb.

244  As explained in paras 5.76–5.77, narrow Rothschild clauses need to be so interpreted in order for them to have the effects the parties intend them to have under the Recast. But, as also queried there, in light of the substantive validity rule that the Recast introduces, it is not certain that a jurisdiction agreement can confer jurisdiction on more than one court under that instrument.

245  See Garcimartín (n 13) paras 9.65–9.66; Briggs, Civil Jurisdiction (n 13) para 2.146; Bureau and Muir Watt (n 196) vol 2, para 869, fn 3 (all making this suggestion in relation to a jurisdiction clause which names the courts of several jurisdictions).

246  ie, by reference to English law in the case of a clause providing that: ‘the purchaser accepts … any court—in England or elsewherethat the seller may designate for the resolution of all disputes arising from the present contract’: Cour d’appel de Paris, 5 July 1989 (n 46).

247  See, eg, the clause providing that: ‘Place of jurisdiction shall be, at our option, F and N respectively, I or the buyer’s place of business’: OLG Brandenburg, 26 June 2012 (n 45).

248  Adapted from the clause in issue in Barclays Bank v Ente (n 200) [30], which referred to English courts rather than French courts in the anchor limb.

249  Sanicentral (n 168) para 6 discussed at para 6.48. See also Pierre Gothot and Dominique Holleaux, La Convention de Bruxelles du 27 septembre 1968 (Jupiter 1985) 96 (expressing the view that unless and until litigation is initiated, the jurisdiction clause does not exist, such that no law applies to validate or nullify it).

250  See para 5.97.

251  The French and Danish delegations, as noted in Council Document 9474/11 ADD 14 of 16 June 2011 at 10, put forward a favor validitis principle during the negotiations of the instrument but this was not taken up:

1. An agreement conferring jurisdiction shall be valid as to substance if it is so considered by: (a) the law of the Member State of the court or courts designated by the agreement, or (b) the law chosen by the parties to govern that agreement, or (c) where such choice has not been made, the law applicable to the contract in which the agreement is included, or (d) in all other cases, the law applicable to the particular legal relationship from which the dispute between the parties has arisen. 2. The law designated by points (b) to (d) of the previous paragraph shall apply even where such law is not that of a Member State.

252  See Heinze (n 135) 585–86 (proposing in relation to jurisdiction agreements generally a tiered approach to the determination of substantive validity: apply the law chosen by the parties to govern the jurisdiction agreement; in the absence of such a choice, apply the law chosen by the parties to govern the contract in which the clause appears; and in the absence of such a choice, apply the substantive law of the chosen court); Usunier, ‘Le droit applicable’ (n 129) 165 (advocating applying the law chosen by the parties or, in the absence of such a choice, the law governing the contract in which the clause appears unless the law of the chosen court is more closely connected to the clause, in which case, that law should govern the substantive validity of the jurisdiction clause. It is difficult to see how the last aspect of Usunier’s proposal would work in practice; presumably, the law of the chosen court would systematically have closer connections to a clause designating its own courts than would the law applicable to the contract in which it is contained (assuming that law is different)).

253  See Magnus, ‘Sonderkollisionsnorm’ (n 142) 529.

254  See paras 2.74 and 2.100.

255  See, generally, Symeon C Symeonides, ‘What Law Governs Forum Selection Clauses’ (2017) 78 Louisiana Law Review 1119, 1154–55.

256  See, generally, Christian Kohler, L’autonomie de la volonté en droit international privé : un principe universel entre libéralisme et étatisme (Brill 2013) 106, 109 (raising this concern in the context of a choice-of-law clause being determined by the putatively chosen law).

257  A further alternative, which Weller has proposed, is to amend the Recast such that the substantive validity of a jurisdiction agreement is determined by a harmonized European contract law instrument, such as the Draft Common Frame of Reference: ‘Choice of Court Agreements’ (n 219) 101; Study Group on a European Civil Code and Research Group on EC Private Law (Acquis Group) (n 173) 230.

259  See, generally, Lohsse (n 145) 703, para 3 (‘Art 4:109: Excessive Benefit or Unfair Advantage’) (referring to German, Greek, Italian, Portuguese, and Swiss law), cf International Institute for the Unification of Private Law (UNIDROIT), Unidroit Principles of International Commercial Contracts (UNIDROIT 2016) 137, art 3.2.7, which gives the disadvantaged party the possibility of avoiding the contract. Ulrich Magnus argues that this issue, described in the Draft Common Frame of Reference as ‘Unfair exploitation’ should be covered by the substantive validity rule: ‘Sonderkollisionsnorm’ (n 142) 525.

260  Lohsse (n 145) 703, para 3 (‘Art 4:109: Excessive Benefit or Unfair Advantage’).

261  See Etihad Airways PJSC v Flöther [2020] EWCA Civ 1707 [76] (Henderson LJ):

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 … 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.

262  Whether as the applicable law or where its overriding mandatory rules are entitled to apply.

263  See Cyril Nourrissat, ‘L’incidence du déséquilibre des forces entre parties non réputées faibles’ in Laazouzi (n 47) 107, 110, 111, 112, 119 (concluding that the invocation of L. 442-1 I–2 is unlikely to be successful in many cases). See, generally, Emilie Gicquidaud, ‘Le contrat à l’épreuve du déséquilibre significatif’ (2014) 67 Revue trimestrielle de droit commercial 267; Philippe Stoffel-Munck, ‘The Revolution in Unfair Terms’ in Simon Whittaker and John Cartwright (eds), The Code Napoléon Rewritten (Hart Publishing 2017) 145, 152–56.

264  Though the French Cour d’appel has clarified that L. 442-6 (the equivalent predecessor to L. 442-1) is not confined to distribution contracts: ‘Cour d’appel de Paris, 21 June 2017, Ministre de l’economie, de l’industrie et du numérique v Expedia inc, No 15-18784, 27’ (2018) 194 Recueil Dalloz 966, 968 (note).

265  See, eg, Diemme (n 30). See further para 2.102.

266  Author’s translation (emphasis added).

267  L. 442-4 I, para 2; Nourrissat, ‘L’incidence’ (n 263) 111–12.

268  Bertrand Fages, ‘Obligations et contrats spéciaux’ (2011) 64 Revue trimestrielle de droit civil 120, 122 (discussing a decision of the Conseil constitutionnel (13 January 2011, No 2010-85) on L. 442-6 (the equivalent predecessor to L. 442-1) to this effect).

269 Ministre de l’economie’ (n 264) 27, confirmed in (Cour de cassation, Commercial Chamber, 8 July 2020, No 17-31.536) Publié au bulletin [Publicized by the Court].

270  Rome I Regulation, art 9(1) (which contains the definition of overriding mandatory provisions applied by the court).

271  Pascal Ancel, ‘Force obligatoire et contenu obligationnel du contrat’ (1999) 52 Revue trimestrielle de droit civil 775, 785 fn 73; Pierre Bourel, Pascal de Vareilles-Sommieres, and Yvon Loussouarn, Droit international privé (10th edn, Dalloz 2013) para 716.

272  Nourrissat, ‘L’incidence’ (n 263) does not raise this point.

274  Joined Cases C–240/98 and 244/98 Océano Grupo Editorial and Salvat Editores [2000] ECR I–4941, paras 21–24; Case C–137/08 VB Pénzügyi Lízing Zrt v Ferenc Schneider [2010] ECR I–10847, paras 53–56. See also Case C–243/08 Pannon GSM Zrt. v Erzsébet Sustikné Győrfi [2009] ECR I–4713, paras 36–44; Marshall, ‘Imbalanced Jurisdiction’ (n 32) 524.

275  Ryanair (n 63) paras 49–61; Happy Flights, SPRL v Ryanair (Cour de Cassation de Belgique, 8 February 2019) No C.18.0354.N. In the light of the CJEU’s ruling, the decision in Weco (n 199) [97]–[98] must now be considered wrong.

276  EU Unfair Contract Terms Directive, art 3: ‘A contractual term … shall be regarded as unfair if … it causes a significant imbalance in the parties’ rights and obligations arising under the contract, to the detriment of the consumer’ (emphasis added).

277  In a commercial case, it is suggested that something more than the remoteness of the optional court for the non-option holder would be needed for the clause to be unfair: see para 6.98.

278  This provision would not be relevant if French law were not the applicable law because it is improbable that it is an overriding mandatory rule. See, generally, Barbet (n 117) 447.

279  Usunier, ‘Le droit applicable’ (n 129) 160.

280  cf Caroline Verbruggen, ‘La compatibilité d’une clause (unilatérale) optionnelle de juridiction avec l’article 23 du Règlement Bruxelles I’ (2013) 119 Revue de Droit Commercial Belge 443, 450; Keyes and Marshall (n 52) 370 fn 133 (suggesting that the use of the word potestative in the Rothschild case by the Cour de cassation may have been descriptive).

281  The Rothschild case (n 49); Usunier, ‘Le droit applicable’ (n 129) 160.

282  Diemme (n 30).

283  The Rothschild case (n 49) (author’s translation).

284  Diemme (n 30).

285  See para 6.30.

286  Those aspects of the Code civil relating to the law of contract were reformed by Ordonnance n° 2016-131 du 10 février 2016 portant réforme du droit des contrats, du régime général et de la preuve des obligations which entered into force on 1 October 2016.

287  Author’s translation (emphasis added). ‘Est nulle l’obligation contractée sous une condition dont la réalisation dépend de la seule volonté du débiteur. Cette nullité ne peut être invoquée lorsque l’obligation a été exécutée en connaissance de cause.’ The latter sentence did not appear in the former version of the Code civil. An alternative translation can be found in John Cartwright, Bénédicte Fauvarque-Cosson, and Simon Whittaker (trs), ‘The Law of Contract, the General Regime of Obligations, and Proof of Obligations: The New Provisions of the Code civil Created by Ordonnance n° 2016-131 of 10 February 2016 Translated into English’ (2017) Ministry of Justice, French Republic, 32 <http://www.textes.justice.gouv.fr/art_pix/THE-LAW-OF-CONTRACT-2-5-16.pdf> accessed 22 December 2021 (‘An obligation undertaken subject to a condition whose satisfaction depends on the will of the debtor alone is a nullity. Nullity on this ground cannot be invoked where the obligation has been performed while aware of the position’). Art 1304-2 replaces arts 1170 and 1174 of the former Code civil which provided: ‘A potestative condition is one which makes the execution of the agreement depend upon an event that one or the other of the contracting parties has the power to bring about or to prevent’ and ‘Any obligation is null when it has been contracted subject to a potestative condition on the part of the party who binds [it]self’ (Legifrance translation).

288  Pascal Ancel and Gilles Cuniberti, ‘Cour de cassation française (1ère ch. civ.), 26 septembre 2012, Observations’ (2013) 25 Journal des tribunaux Luxembourg 7, 10; Dominique Bureau, ‘Clause attributive de juridiction potestative et pluralité de défendeurs dans des actions fondées sur des loi différentes, Cour de cassation (Civ 1re)—26 septembre 2012, D 2012 2876’ (2013) 102 Revue critique de droit international privé 256, para 6. The nature of jurisdiction agreements from a French perspective was explored in para 2.23.

289  See para 6.48.

290  If jurisdiction agreements do contain obligations, other provisions of the revised French Code civil may be relevant. See Merrett, ‘Future Enforcement’ (n 56) 53 (referring to the possible relevance of art 1169 concerning onerous contracts). Article 1169 provides that ‘[a]n onerous contract is a nullity where, at the moment of its formation, what is agreed in return for the benefit of the person undertaking an obligation is illusory or derisory’: translation of Cartwright, Fauvarque-Cosson, and Whittaker (n 287).

291  This is the clause contained in the Loan Market Association’s Multicurrency Term and Revolving Facilities Agreement, though the optional limb of the clause is contained in a footnote: (n 234) cl 38.

292  ie, it is ‘purely potestative’. See, generally, Bertrand Fages, Droit des obligations (10th edn, LGDJ 2020–21) para 149; Wood Brown, ‘Potestative Conditions and Illusory Promises’ (1930) 5 Tulane Law Review 396.

293  This example is based on Brown’s example of a contract subject to a potestative condition giving rise to nullity under art 1174 of the former Code civil: ‘I will pay you one hundred dollars if I want to’: (n 292) 399. I think readers will agree that an apartment at Bondi Beach is the better option: see para 6.76. See also Ancel and Cuniberti (n 288) 10 (‘je m’engage à saisir la juridiction du Luxembourg, sous condition que je le veuille bien’).

294  Brown (n 292) 399. It could also be described as a state of mind of the obligor’s advisors.

295  See also Matthias Lehmann and Alexander Grimm, ‘Zulässigkeit asymmetrischer Gerichtsstandsvereinbarungen nach Artikel 23 Brüssel I-VO’ [2013] Zeitschrift für Europäisches Privatrecht 890, 895.

296  See Thibault Douville, ‘Art 1304–2’ in Thibault Douville (ed), La réforme du droit des contrats (Gualino 2016) 265, 267.

297  ibid.

298  David Martel, ‘A la découverte de la clause attributive de juridiction potestative’ (2012) 188 Recueil Dalloz 2876, 2879. See also Racine (n 20) 1334–35.

299  See Ancel and Cuniberti (n 288) 10.

300  The Rothschild case (n 49).

301  ‘[Un] droit d’option … est un droit potestatif’ ; ‘des droits potestatifs [sont] de[s] pouvoirs’ [A right to an option is a potestative right and potestative rights are powers] (author’s translation): lbrahim Najjar, Le droit d’option : Contribution à l’étude du droit potestatif et de l’acte unilatéral (LGDJ 1967) 103–04, 108–09. But see Dominique Fenouillet, ‘La notion de prérogative : instrument de défense contre le solidarisme ou technique d’appréhension de l’unilatéralisme’ [2011] Revue des contrats 644, 668, 672–73 (suggesting that the elision between a potestative right and a power is disputed).

302  Though this is contested. Fenouillet (n 301) 669 argues that ‘even if the law gives a potestative right to a contracting party, a potestative right remains contractual and is worthy of control, in that it is dangerous because it subjects the contractual situation resulting from the parties’ autonomy to the whim of one of them’ (author’s translation) (emphasis added).

303  Fenouillet (n 301) 668 (author’s translation).

304  Strictly, the nullity relates only to the obligation subject to the potestative condition (in our example, that would be the obligation of the option holder to sue in Paris): see Elisabeth Omes, ‘Cour de cassation française (Civ, 1re), 26 septembre 2012’ (2013) 51 Bulletin droit et banque 27, 32 (note). But the French doctrine and case law support the view that, as a consequence of that nullity, the correlative obligation of the counterparty (in our example, the obligation of the non-option holder to sue in Paris) disappears: see Douville, ‘Art 1304–2’ (n 296) 266; ‘Cour de cassation, Third Civil Chamber, 7 June 1983 No 82-10.281’ [1983] III Bulletin civil no 132 (note) (in which the court annulled the whole contract in which the obligation subject to the potestative condition was contained). See, generally, Brown (n 292) 403–04.

305  Ancel and Cuniberti (n 288) 10.

307  As already noted at para 6.46, there will still be cases in which English law is the law governing the substantive validity of an asymmetric clause under the Recast, despite the fact that English courts are no longer EU Member State courts. Where that is the case, it will be relevant to consider the Unfair Contract Terms Act 1977 (UK), discussed in paras 7.48–7.57.

308  See, generally, Stoffel-Munck, ‘The Revolution’ (n 263) 149–65.

309  See Nourrissat, ‘L’incidence’ (n 263) 110. See, generally, Fenouillet (n 301) 677 fn 175 (suggesting that one possible mechanism to control potestative rights—which, as demonstrated at para 6.77, asymmetric clauses often contain—is by restricting the freedom to include potestative rights in contracts of adhesion). Art 1110 defines a contract of adhesion as ‘one in which the general conditions, not the subject of negotiation, were formulated in advance by one of the parties’ (author’s translation).

310  François Ancel, Bénédicte Fauvarque-Cosson, and Juliette Gest, Aux sources de la réforme du droit des contrats (Dalloz 2017) 112, 230; Cécile Pérès, ‘Règles impératives et supplétives dans le nouveau droit des contrats’ (2016) 16 La Semaine Juridique—Édition Générale 770, 772. Again, whether this provision will eventually be found to have an overriding mandatory character (loi de police) appears yet to be tested: see Ancel, Fauvarque-Cosson, and Gest (n 310) 230–31.

311  Hein Kötz, European Contract Law (2nd edn, OUP 2017) 138–39; Ancel, Fauvarque-Cosson, and Gest (n 310) 171–72; cf Anne-Sophie Chone-Grimaldi in Douville, La réforme (n 296) 139, 140.

312  Marshall, ‘Imbalanced Jurisdiction’ (n 32) 524.

313  The phrase ‘réputée non écrite’, literally translated as ‘deemed to be unwritten’, means very little in English. The intention behind the provision is to require a court to invalidate the clause: Ancel, Fauvarque-Cosson, and Gest (n 310) 173. The Cour de cassation has interpreted the phrase ‘réputée non écrite’ in another context to allow for a partial invalidation of the clause, although this approach has been criticized: Bertrand Fages and Hugo Barbier, ‘Obligations et contrats spéciaux’ (2013) 66 Revue trimestrielle de droit civil 832, 836–37.

314  Author’s translation. ‘Dans un contrat d’adhésion, toute clause qui crée un déséquilibre significatif entre les droits et obligations des parties au contrat est réputée non écrite. L’appréciation du déséquilibre significatif ne porte ni sur l’objet principal du contrat ni sur l’adéquation du prix à la prestation.’

315  See para 6.36.

316  A standard form contract is defined in § 305(1) BGB. A contract is considered to be a standard form contract when it is pre-formulated, drafted with the intention of being used on at least three occasions, presented by one party to the other, and where no individual negotiation takes place: Jürgen Basedow in Franz Jürgen Säcker and others (eds), Münchener Kommentar zum Bürgerlichen Gesetzbuch, vol 2 (8th edn, Beck 2019) § 305, paras 17–19.

317  See § 310(1) BGB; BGH, 4 July 2017, XI ZR 562/15 and XI ZR 233/16, NJW 2017, 2986.

318  Wais (n 35) 847. cf Kilian Bälz and Philipp Stompfe, ‘Asymmetrische Streitbeilegungsklauseln in internationalen Wirtschaftsverträgen’ (2017) 15 Zeitschrift für Schiedsverfahren 157, 162, fn 37 (who exclude the application of a standard form control test under German law, based on literature that relates to art 23 of Regulation 44/2001. Art 23 did not expressly refer the substantive validity of a jurisdiction clause to an applicable national law.)

319  Wolfgang Wurmnest in Säcker and others (n 316) § 307, [266]: ‘Kein Indiz für die Unangemessenheit einer Gerichtsstandsklausel ist es ferner im Allgemeinen, dass durch die Vereinbarung eines ausländischen Gerichtsstandes die Anwendung der AGB-Vorschriften oder anderen zwingenden deutschen Rechts auf den Vertrag praktisch oder rechtlich ausgeschlossen wird.’

320  Basedow (n 316) § 305 [5]; Wurmnest in Säcker and others (n 31) § 307, [41]; Wais (n 35) 846.

321  Thomas Pfeiffer in Manfred Wolf and others (eds), AGB-Recht Kommentar (7th edn, Beck 2020) ‘Einleitung’ [15]; Wais (n 35) 846. See, generally, Dan Wielsch, ‘Global Law’s Toolbox: How Standards Form Contracts’ in Horst Eidenmüller (ed), Regulatory Competition in Contract Law and Dispute Resolution (Beck, Hart Publishing, Nomos 2013) 71, 80.

322  § 307(1) BGB.

323  § 307(2) sub-para 1 BGB. In several decisions, the German Bundesgerichtshof has applied § 307(1) sentence 1 and § 307(2) sub-para 1 to hold administrative charges clauses to be ineffective in commercial loan agreements: BGH, 4 July 2017 (n 317). Clauses levying administrative fees and charges separately from interest charges were ineffective under § 307 BGB because they unreasonably disadvantaged the commercial borrower in comparison to the creditor. The Court held that the protective aim of § 307 BGB (namely, to limit the use of one-sided power to shape the contract terms) equally applied to informed and experienced commercial parties. The fact that a commercial party might be better able to estimate the total cost of finance resulting from various fee components does not substantiate the reasonableness of the clause. (Even if there was not necessarily a partial market failure, the Court’s policy decision behind these cases may well have been market driven: the comparability of loan prices is naturally facilitated if price is purely reflected in interest fees.) The Bundesgerichtshof had previously struck down similar clauses in cases involving consumer borrowers.

324  § 307(3) BGB.

325  Thomas Pfeiffer has persuasively dismissed the argument that the parties’ ability to agree on jurisdiction is itself a statutory default rule and thus a jurisdiction clause merely mirrors the default rule: ‘Halbseitig fakultative Gerichtsstandsvereinbarungen in stillschweigend vereinbarten AGB?’ [1998] Praxis des Internationalen Privat—und Verfahrensrechts 17, 23 (discussing § 9 Gesetz zur Regelung des Rechts der Allgemeinen Geschäftsbedingungen [Standard Terms Act], the predecessor to § 307 BGB). While party autonomy is undoubtedly an essential principle of the Recast (see recital 19), a jurisdiction clause under the Recast, by its very nature, is an extension of and/or derogation from the default rules of general and special jurisdiction: see paras 2.14–2.17. To argue that any and every jurisdiction agreement fulfils the essential principle of party autonomy would be to justify the enforcement of any unreasonably disadvantageous clause: ibid 23; also in support of Pfeiffer’s point, see Wais (n 35) 848–49.

326  Recast, recitals 15–16; GIE (n 67) paras 23–24.

327  Case C–256/00 Besix SA v Wasserreinigungsbau Alfred Kretzschmar GmbH & Co KG (WABAG) [2002] ECR I–1699, para 27.

328  See Chapter 3.

329  Étienne Cornut, ‘Clause attributive de juridiction potestative’ (2012) 86 La Semaine Juridique 1065, 1065; Hélène Gaudemet-Tallon, ‘L’abus de droit en droit international privé’ in Louis d’Avout and others (eds), Mélanges en l’honneur du Professeur Bernard Audit : Les relations privées internationales (LGDJ 2014) 383, 388; Ulrich Magnus, ‘Article 25, Brussels Ibis Regulation’ in Magnus and Mankowski, European Commentaries (n 13) para 73, fn 242 (tentatively); the Rothschild case (n 49).

330  Basedow, ‘Exclusive Choice-of-Court Agreements’ (n 214) 22; Magnus, ‘Article 25’ (n 329) para 73 and sources cited in fn 242. Magnus appears to argue that art 25 as a matter of current law implicitly contains such a test, although acknowledges that this is not the majority view; cf Eckart Gottschalk and Steffen Breßler, ‘Missbrauchskontrolle und Gerichtsstandsvereinbarungen im europäischen Zivilprozessrecht’ [2007] Zeitschrift für Europäisches Privatrecht 56, 68–70, 72, 80; Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 100; Arnaud Nuyts, ‘Forum Shopping et abus du Forum Shopping dans l’espace judiciaire européen’ in Philippe Gérard and others (eds), Mélanges John Kirkpatrick (Bruylant 2004) 745, 785.

331  German case law makes clear that, as a matter of current law, there is no implied misuse test arising out of EU law itself: OLG Hamburg, 14 April 2004, 13 U 76 03, para 2(c), NJW 2004, 3126 (Regulation 44/2001).

332  Basedow, ‘Exclusive Choice-of-Court Agreements’ (n 214) 22; cf Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO, para 99.

333  Wais (n 35) 833–34. The author appears to favour a test targeting the misuse of the option rather than the clause as a whole: see 835. Rothschild clauses are described in paras 1.38–1.47.

334  See further para 6.99.

335  See Mankowski, EuZPR/EuIPR (n 135) art 25 Brüssel Ia-VO para 102; see paras 6.88–6.89.

337  That is not to say, of course, that the issues which these doctrines address are unrelated.

339  Anna di Robilant, ‘Abuse of Rights: The Continental Drug and the Common Law’ (2010) 61 Hastings Law Journal 687, 688 (providing a broad description of the concept across legal systems). See also Jürgen Basedow, ‘Das Verbot von Rechtsmissbrauch und Gesetzesumgehung im europäischen Privatrecht’ in Michael P Stathopulos, Timētikos tomos Mich P Stathopoulou (Ekdoseis Ant N Sakkoula 2010) 159, 161 (arguing that the lawful existence of the right is what distinguishes abuse of rights from other doctrines).

340  See Jan Peter Schmidt, ‘Art 1:201: Good Faith and Fair Dealing’ in Jansen and Zimmermann (n 145) 101, paras 54, 91 (observing at para 91 that ‘the prohibition of “abuse of rights” [ … ] has been the only uncontested field of application of the general duty of good faith’ among these instruments). See, generally, Research Group on the Existing EC Private Law, Principles of the Existing EC Contract Law (Acquis Principles): Contract I (Sellier 2007) 261, 264; Axel Metzger, ‘Abuse of Law in EU Private Law: A (Re-)Construction from Fragments’ in Rita de la Feria and Stefan Vogenauer (eds), Prohibition of Abuse of Law (Hart Publishing 2011) 235, 250; Study Group on a European Civil Code and Research Group on EC Private Law (Acquis Group) (n 173) 230, vol III, 1:103; UNIDROIT (n 259) 19, art 1.7; Ole Lando and Hugh Beale (eds), Principles of European Contract Law (Kluwer Law International 2000) 6, art 1:201.

341  Simon Whittaker, ‘Comments on “Abuse of Law” in European Private Law’ in de la Feria and Vogenauer (n 340) 253, 258. Several authors have argued that it is preferable to regard an impermissible exercise of a right as a case where a person no longer has a right or acts outside the scope of the right that it has, rather than as an abuse of the right. James Gordley traces this scope argument back to Marcel Planiol in the early 1900s: ‘The Abuse of Rights in the Civil Law Tradition’ in de la Feria and Vogenauer (n 340) 33, 40–46; Schmidt (n 340) para 92.

342  As explained earlier, according to lbrahim Najjar, an option is a potestative right and a potestative right is a power: (n 301) 103–04, 108–09; cf Fenouillet (n 301) 672–73.

343  See, generally, Philippe Stoffel-Munck, L’abus dans le contrat : Essai d’une théorie (LGDJ 2000) paras 745–54; Fenouillet (n 303) 669.

344  Abuse of rights was formerly governed by art 1134 of the Code civil, which provided that ‘Agreements lawfully entered into have the force of law for those who have made them … They must be performed in good faith’ (Legifrance translation). See, eg (Cour de cassation, Commercial Chamber, 10 July 2007) No 06-14.768, Publié au bulletin [Publicized by the Court]. Art 1104 of the revised Code civil provides: ‘Agreements must be negotiated, formed and performed in good faith. This provision is a provision of the ordre public’ (author’s translation). See, generally, Gilles Raoul-Cormeil in Douville, La réforme (n 296) 33, 35.

345  See Alain Bénabent, Droit des obligations (19th edn, LGDJ 2021) para 303; Bertrand Fages, ‘Mais la bonne foi ne fait pas échec au bénéfice d’une garantie de passif’ (2007) 60 Revue trimestrielle de droit civil 773; cf Fenouillet (n 301) 659–62, 665.

346  Angelo Dondi, ‘Abuse of Procedural Right: Report for Italy and France’ in Michele Taruffo (ed), Abuse of Procedural Rights: Comparative Standards of Procedural Fairness (Kluwer Law International 1999) 109, 117. Article 32-1 of the Code de procédure civile [French Code of Civil Procedure] provides that ‘The one who acts in justice in a dilatory or abusive way may be condemned to a civil fine’ (Legifrance translation). Article 30 makes clear that ‘acting in justice’ relates to both parties: ‘the action’ is the right of the claimant to bring proceedings and the right of the defendant to respond to them.

347  See, generally, Matthias Klöpfer, Missbrauch im Europäischen Zivilverfahrensrecht (Mohr Siebeck 2016) 45.

348  Heinz-Peter Mansel, Jauernig Kommentar zum Bürgerlichen Gesetzbuch (18th edn, Beck 2021) § 242, [38] (stating that ‘The exercise of a right is inadmissible if the rights-holder is not pursuing their objective, permanent self-interest but rather is exercising that right as a pretext for obtaining an unfair advantage beyond the scope of the contract or alien to the law. This also applies to the exercise of procedural powers’ (author’s translation)); Jan Lüttringhaus, ‘Der Missbrauch des Gerichtsstandes im Zivilprozess’ (2014) 127 Zeitschrift für Zivilprozess 29, 30. See also Gottschalk and Breßler (n 330) 59 (in relation to jurisdiction agreements specifically).

349  § 242 applies to the exercise of ‘Gestaltungsrechte’ that are not governed by more specific provisions; unilateral contractual powers that give the right to one party to change or terminate the contractual relation: Lorenz Kähler, Beck Online Großkommentar (Beck 2021) § 242, [1514].

350  Jan D Lüttringhaus, ‘Treu und Glauben und Rechtsmissbrauch im Internationalen Zivilvefahrensrecht (zu OLG Karlsruhe, 20.12.2018—4 U 31/18)’ [2020] Praxis des Internationalen Privat—und Verfahrensrechts 433, 437 (note) referring to BGH 16 October 2008, III ZR 253/07, NJW 2009, 148, 150; OLG Karlsruhe 20 December 2018, 4 U 31/18, NJW 2019, 1086.

351  Basedow, ‘Das Verbot’ (n 339) 181.

352  Case C–367/96 Kefalas v Dimosio [1998] ECR I–2843, paras 22, 29 (quote is at para 22). Whether this case law is still good law is disputed; cf Raluca Nicoleta Ionescu, ‘L’abus du droit de l’union européene’ in Catherine Kessedjian (ed), Autonomie en droit européen (Éditions Panthéon-Assas 2013) 207, 207, 219 and Laurence Usunier, ‘Le règlement Bruxelles I bis et la théorie de l’abus de droit’ in Guinchard (n 131) 449, 466–67 (who both appear to suggest that this case law is still relevant notwithstanding the CJEU’s later development of the EU prohibition on abuse of law, discussed in paras 6.90–6.92, which would cover the situation which occurred in that case) with Stefan Vogenauer, ‘The Prohibition of Abuse of Law: An Emerging General Principle of EU Law’ in de la Feria and Vogenauer (n 340) 521 (arguing at 563 that ‘national anti-abuse doctrines apply only to the extent that they conform with the abuse test’ of the CJEU and, at 556, that the circumstances which occurred in Kefalas v Dimosio could equally be framed as an abuse of law).

353  Case C–68/93 Shevill and others v Presse Alliance [1995] ECR I–415, para 36; Case C–365/88 Hagen v Zeehaghe [1990] ECR I–1845, para 20.

354  Case 413/12 Asociación de Consumidores Independientes de Castilla y León v Anuntis Segundamano España SL [2013] ECLI:EU:C:2013:800, paras 34, 52 (the quote is at para 52); Andrew Dickinson, ‘Background and Introduction to the Regulation’ (n 66) para 1.113.

355  Case C–159/02 Turner v Grovit [2004] ECR I–3578, para 37 (Brussels Convention).

356  Dickinson, ‘Background and Introduction to the Regulation’ (n 66) para 1.113.

358  Matthias Weller, ‘Mutual Trust in Search of the Future of European Union Private International Law’ (2015) 11 Journal of Private International Law 64; Eduardo Gill-Pedro and Xavier Groussot, ‘The Duty of Mutual Trust in EU Law and the Duty to Secure Human Rights: Can the EU’s Accession to the ECHR Ease the Tension?’ (2017) 35 Nordic Journal of Human Rights 258, 265–69.

359  See Nipponkoa (n 53) para 36 (Regulation 44/2001).

360  Turner (n 355) paras 25–27.

361  See, eg, Case C–50/14 Consorzio Artigiano Servizio Taxi e Autonoleggio v Azienda Sanitaria Locale di Ciriè, Chivasso e Ivrea [2016] ECLI:EU:C:2016:56, para 65; Case C–113/13 Azienda Sanitaria Locale n 5 ‘Spezzino’ v San Lorenzo Soc Coop Sociale [2014] ECLI:EU:C:2014:2440, para 62; Joined Cases C–131/13, 163/13 and 164/13 Staatssecretaris van Financiën v Schoenimport ‘Italmoda’ Mariano Previti vof [2014] ECLI:EU:C:2014:2455, para 53. See also Case C–321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I–5818, para 38 and its discussion in Annekatrien Lenaerts, ‘The General Principle of the Prohibition of Abuse of Rights’ (2010) 6 European Review of Private Law 1121, 1138.

362  On the effect of general principles of EU law, see Anthony Arnull, ‘What Is a General Principle of EU Law?’ in de la Feria and Vogenauer (n 340) 7, 12.

363  See para 6.92.

364  Vogenauer, ‘The Prohibition of Abuse of Law’ (n 352) 524–25, 555, 557. See also Basedow, ‘Das Verbot’ (n 339) 160–61.

365  See Nuyts (n 330) 763; Whittaker (n 341) 258; Ionescu (n 352) 222. Gilles Cuniberti defines it as ‘the use of one rule with the sole object of ousting the application of another’: ‘The Discreet Influence of Abuse of Law in International Civil Procedure’ in de la Feria and Vogenauer (n 340) 279, 284. See also Vedanta Resources plc v Lungowe [2019] UKSC 20, [2020] AC 1045 [36]: ‘the abuse of law doctrine is limited to the collusive invocation of one EU principle so as improperly to subvert another’ (Lord Briggs).

366  Case C–110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I–11595, paras 52–53; see Lenaerts (n 361) 1133–35.

367  Case C–423/15 Nils-Johannes Kratzer v R + V Allgemeine Versicherung AG [2016] ECLI:EU:C:2016:604, para 39 (emphasis added).

368  ibid para 40.

369  ibid para 41.

370  Vogenauer, ‘The Prohibition of Abuse of Law’ (n 352) 535.

371  Although one might read Nils-Johannes Kratzer (n 367) para 4 as suggesting that artificiality is only an indicator of the intention to obtain an undue advantage rather than a necessary condition for it, on the facts of this case there was artificiality. Moreover, in Emsland-Stärke (n 366) para 53, in which the CJEU established the test, artificiality was a necessary condition: ‘It requires … a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it.’

372  See Gaudemet-Tallon (n 329) 384–85.

373  cf Klöpfer (n 347) 174 (arguing that the EU prohibition encompasses both abuse of rights (Missbrauch) and abuse of law (Gesetzesumgehung and fraude à la loi)).

374  Case C–352/13 Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Akzo Nobel NV [2015] ECLI:EU:C:2015:335. The same applies to the 2007 Lugano Convention.

375  Dominic Liswaniso Lungowe v Vedanta Resources plc, Konkola Copper Mines plc [2017] EWCA Civ 1528 [38].

376  Case C–352/13 Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Evonik Degussa GmbH [2014] ECLI:EU:C:2014:2443, Opinion of AG Jääskinen.

377  See CDC v Akzo (n 374) para 20.

378  ‘Is it significant in this regard if the action against the defendant domiciled in the same State as the court is withdrawn after having been served on all the defendants, before the expiry of the periods prescribed by the court for lodging a defence and before the start of the first hearing?’: ibid para 14.

379  CDC v Evonik (n 376) Opinion of AG Jääskinen, para 90.

380  CDC v Akzo (n 374) para 29.

381  See Caroline Cohen, ‘Les normes permissives en droit international privé. Étude critique’ (Thèse en droit, Université Panthéon-Assas (Paris II) 2015) para 175 (discussing asymmetric jurisdiction clauses).

382  For a clause example, see para 6.09.

383  Recast, arts 4, 7, 8, or 9 (assuming the non-option holder is EU-domiciled); see para 5.33.

384  Recast, art 25.

385  Even if one were to adopt the interpretation of asymmetric clauses proposed in para 5.149 (ie, as exclusive agreements which one party can renounce in their entirety and thus rely on the default rules of jurisdiction to bring proceedings), the right to renounce the agreement would still derive directly from an agreement and only indirectly from the rule in art 25 of the Recast allowing for the agreement.

386  As to non-exclusive effect, see paras 5.46–5.52.

387  As to exclusive effect, see paras 5.36–5.45.

388  Case C–373/97 Dionisios Diamantis v Elliniko Dimosio (Greek State) [2000] ECR I–1705.

389  Gaudemet-Tallon (n 329) 385; Bureau and Muir Watt (n 196) vol 2, para 869; Étienne Cornut, ‘Forum shopping et abus du choix de for en droit international privé’ (2007) 134 Journal du droit international 27, 41; Nuyts (n 330) 780; Usunier, ‘Le règlement’ (n 352) 474.

390  Second Council Directive 77/91/EEC of 13 December 1976 on coordination of safeguards which, for the protection of the interests of members and others, are required by Member States of companies within the meaning of the second paragraph of Article 58 of the Treaty, in respect of the formation of public limited liability companies and the maintenance and alteration of their capital, with a view to making such safeguards equivalent [1977] OJ L 26, 1.

391  Dionisios (n 388) para 13.

392  Dionisios (n 388) at para 33 states that:

Community law cannot be relied on for abusive or fraudulent ends (see Kefalas and Others, cited above, paragraph 20, and the case-law cited there). That would be the case if a shareholder, in reliance on Article 25(1) of the Second Directive, brought an action for the purpose of deriving, to the detriment of the company, an improper advantage, manifestly contrary to the objective of that provision (Kefalas and Others, cited above, paragraph 28).

393  As seen in paras 6.88–6.89, Kefalas v Dimosio (n 352) is authority for the proposition that a Member State can apply national law to prevent an abuse of what is now EU law, provided that it is consistent with the principle of effectiveness.

394  See particularly paras 34, 35, and 38.

395  Dionisios (n 388) para 43.

396  Sophie Le Gac-Pech, La proportionnalité en droit privé des contrats (LGDJ 2000) para 1119 (author’s translation). See also Ancel, Fauvarque-Cosson, and Gest (n 310) 113.

397  Ionescu (n 352) 210.

398  Le Gac-Pech (n 396) para 1119. See also Ionescu (n 352) 209 (explaining that by referring to the abnormal exercise of the right, objective elements clearly temper the subjective test); Laura Melusine Baudenbacher, Vom gemeineuropäischen zum europäischen Rechtsmissbrauchsverbot (Nomos Beck 2016) 33.

399  For a clause example, see para 6.12, n 42.

400  Stoffel-Munck, L’abus (n 343) paras 746–47 (‘it seems improbable that the objective is for the subject of the power to submit itself to the arbitrary discretion of the power holder’) (author’s translation). See also Fenouillet (n 301) 653.

401  Baudenbacher (n 398) 37 (‘intention can only be a guiding not decisive element’) (author’s translation). See also Ionescu (n 352) 210.

402  Schmidt (n 340) para 51.

403  See Philipp Eichenhofer, Rechtsmissbrauch (Mohr Siebeck 2021) 241 (author’s translation); Joseph Voyame, Bertil Cottier, and Bolívar Rocha, ‘Abuse of Rights in Comparative Law’ in Council of Europe, Abuse of Rights and Equivalent Concepts: The Principle and Its Present Day Application: Proceedings of the Nineteenth Colloquy on European Law (Publishing and Documentation Service 1990) 23, 23–24; Robert Krieps, ‘General Report’ in Council of Europe, Abuse of Rights and Equivalent Concepts (n 403) 166, 166–67.

404  Nuyts (n 330) 779–80. See also Usunier, ‘Le règlement’ (n 352) 449; Lüttringhaus, ‘Der Missbrauch’ (n 348) 45, 58; Klöpfer (n 347) 238–40. Étienne Cornut traces the utility of abuse of rights in combatting unilateral forum shopping back to Louis Josserand, the French father of abuse of rights, although in a domestic context: ‘the possibility sometimes accorded to a claimant to choose between several competent courts may become an opportunity for abuse if the choice is exercised in a way that is particularly prejudicial for the defendant and does not benefit or advantage the claimant’ (author’s translation): ‘Forum shopping’ (n 389) 35. See, eg, Louis Josserand, De l’esprit des droits et de leur relativité. Théorie dite de l’abus des droits (2nd edn 1939, Dalloz 2006) para 51.

405  See Basedow, ‘Exclusive Choice-of-Court Agreements’ (n 214) 22–23.

406  Dominique Bureau and Horatia Muir Watt argue that something more than the fact that the substantive law of the forum chosen is unfavourable to the defendant would need to be shown: (n 196) vol 1, para 224.

407  See Case C–98/06 Freeport plc v Olle Arnoldsson [2007] ECR I–8319, Opinion of AG Mengozzi, 8334 fn 27 (‘[c]hosing[sic] a forum according to the advantages which may arise from the substantive (and even procedural) law applied there … is undoubtedly permitted’); CDC v Evonik (n 376) Opinion of AG Jääskinen, para 89 (‘it is open [to the claimant] to exercise that option in the manner [it] considers most suitable and advantageous’). See also Matthias Lehmann and others, ‘Special Jurisdiction’ in Dickinson and Lein (n 13) 131, para 4.02 (‘a legally allowed kind of forum shopping’). See also Bureau and Muir Watt (n 196) vol 1, para 224.

408  Arguably, the effects of unilateral forum shopping in driving negative unilateral regulatory competition, and therefore the risk of a race to the bottom (discussed in paras 3.56–3.62), are the same as unilateral forum shopping on a global scale; it is just that within the EU the pool is smaller. Further consideration is outside the scope of this work.

409  Wais (n 35) 836. See also Basedow, ‘Exclusive Choice-of-Court Agreements’ (n 214) 22 (suggesting, in the context of exclusive jurisdiction clauses, ‘the remoteness of the chosen court’ as one consideration to be taken into account).

410  Case C–288/92 Custom Made Commercial Ltd v Stawa Metallbau GmbH [1994] ECR I–2913, paras 18–21. The Court in that case was interpreting the Brussels Convention but the same principle applies under the Recast (Bureau and Muir Watt (n 196) vol 1, para 224) and 2007 Lugano Convention.

411  cf Wais (n 35) 836.

412  The majority interpretation of the relevant CJEU jurisprudence suggests that the principle of mutual trust prevents Member States courts from controlling abuses before other Member State courts: See Case C–185/07 Allianz SpA v West Tankers Inc [2009] ECR I–663, paras 28–30; Turner (n 355) para 31. See Adrian Briggs, ‘The Rejection of Abuse in International Civil Procedure’ in de la Feria and Vogenauer (n 340) 261, 272–74; Cuniberti, ‘The Discreet Influence’ (n 365) 287–88.

413  François Mailhé, L’ organisation de la concurrence internationale des juridictions : La compétence face à la mondialisation économique (Economica 2016) paras 659–661.

414  Article 31(2) of the Recast requires a Member State court to stay its proceedings until the exclusively designated court declares that it has no jurisdiction under the agreement. On the application of art 31(2) to asymmetric clauses, see paras 5.87–5.147.

415  See, generally, paras 5.93–5.103.

416  cf Wais (n 35) 839.

417  This type of situation appears to be that to which Basedow is alluding: ‘Exclusive Choice-of-Court Agreements’ (n 214) 22.

418  This type of situation appears to be that which Wood is contemplating: Philip R Wood, Conflict of Laws and International Finance (3rd edn, Sweet & Maxwell 2019) para 21-010 (although Wood does not use the term ‘abuse’).

419  Or the 2007 Lugano Convention, as applicable.

420  See para 6.94.

421  See para 5.64.

422  For a clause example, see para 6.09.

423  Wais (n 35) 834.

424  Coreck (n 9) ruling para 1 (1968 Brussels Convention); Hőszig (n 2) para 43 (Regulation 44/2001).

425  Ryanair (n 63) para 38 (Recast).

426  See paras 6.23–6.24. A typology of asymmetric clauses is provided in paras 1.37–1.54.