Hess on West Tankers
Burkhard Hess is professor of law and director of the Institute for Comparative Law, Conflict of Laws and International Business Law in Heidelberg. He has published extensively in the areas of private international law and international procedural law and is co-author of the General Report of the Study on the Application of Regulation Brussels I (“Heidelberg Report”).
1. The outcome of the ECJ’s judgment is not surprising and, from the point of view of continental procedural law, the findings are completely in line with the framework of the Brussels I Regulation. As the Italian court in Syracuse has been seised under the Regulation, it is for this court to decide on its jurisdiction (Article 5 no 3 Brussels I) and (this is only the second issue) on the scope and the validity of the arbitration clause (Article II NYC).
Despite of some heated criticism which has been brought forward against the conclusions of AG Kokott, the Court comprehensively followed her reasoning. The line of arguments developed in para. 24 of the judgment seems to be similar to the arguments of the ECJ in the Lugano Opinion: The Grand Chamber relies on the effet utile of the Regulation, its “objective of unification of the rules of conflicts of jurisdiction in civil and commercial matters and the free movement of decisions in those matters”. Mutual trust is only used as an additional argument, but much later (para. 30). In my view the judgment demonstrates that the ECJ is “defending” the proper operation of the Regulation and, finally, the priority of Community law. West Tankers is, as Lugano, a political decision.
2. However, as the AG clearly stated, the present situation under the Brussels I Regulation is not satisfactory. With all due respect, I disagree with Adrian Briggs that the issues raised by the House of Lords and the ECJ are not important. After West Tankers, the issue should be addressed in the context of the expected revision of the Brussels I Regulation. In this respect I would like to come back to the proposals of the Heidelberg Report:
The Heidelberg Report on the Application of Brussels I proposed a different mechanism for the protection of arbitration agreements. According to this proposal, a new Article 27 A shall address the situation of threatening parallel arbitral and litigious proceedings, especially when a party institutes proceedings in a domestic court of a Member State instead of enforcing the arbitration agreement. Article 27 A should read as follows: “A court of a Member State shall stay the proceedings once the defendant contests the jurisdiction of the court with respect to existence and scope of an arbitration agreement if a court of the Member State that is designated as place of arbitration in the arbitration agreement is seised for declaratory relief in respect to the existence, the validity, and/or scope of that arbitration agreement”.
This provision aims to concentrate all proceedings on the validity of the arbitration agreement in the domestic courts of the Member State where the arbitration takes place. In this respect, the Heidelberg Report proposes to insert a new Article 22 no 6 to the Brussels I Regulation. The new articles shall establish an exclusive competence for proceedings challenging the validity of the arbitration agreement. These proceedings shall exclusively take place in the Member State in which the arbitration takes place.
Article 27 A shall operate as follows: Imagine that a civil court in Member State A is called upon by a party contesting the validity of an arbitration clause providing for arbitration in Member State B. Under Article 27 A Brussels I, the civil court in Member State A shall stay its proceedings until the matter has been referred to the competent court in Member State B. The court in Member State B then decides exclusively on the validity of the arbitration clause (see Article 72 of the English Arbitration Act). In addition, the civil court of Member State A, when staying its proceedings, may set a time limit for the plaintiff (who is contesting the validity of the arbitration clause) to access the courts in Member State B where the arbitration shall take place. Still, the other party may seek redress in the courts of Member State B to get a judgment on the validity of the arbitration clause. If the plaintiff does not institute arbitral proceedings in the “designated” Member State B in a timely manner, the civil court of Member State A will dismiss its proceedings. This example illustrates the proposal’s intention to give full effect to arbitration agreements and to achieve uniform results in all EU Member States.
3. Besides, I fully agree with Horatia Muir Watt’s recent remark that the principle of mutual trust does not automatically imply the (absolute) priority of the court first seised in parallel litigation. European procedural law also provides for a (untechnical) hierarchy between the courts of different Member States (striking examples are found in Articles 11 and 20 of the Brussels II bis Regulation). To my opinion, the Brussels I Regulation should also adopt a hierarchical system giving priority to the court agreed upon in choice of court agreements and to the courts of the place of arbitration in arbitration proceedings.
I am well aware that the proposal of the Heidelberg Report to delete the arbitration exception of Article 1 (2)(d) has been criticised by many stakeholders of the “arbitration world”. However, after West Tankers/Adriatica the legal doctrine should elaborate a more balanced solution in the framework of Brussels I.
4. Finally, some authors raised the question whether the findings of the ECJ also relate to third states. I don’t believe that the Grand Chamber addressed this constellation. However, as the judgment refers to general principles of EC law (paras. 24 and 30), their application in relation to third states seems to be unlikely.
Dr Hess’s comments are, of course, interesting and valuable. Of course the issues raised by the House of Lords are important, but if one asks whether the issues raised by the *judgment* are important, the answer is, surely, not: the one answer to the one question is entirely predictable and will take nobody by surprise. Of course it leaves larger issues to be fought out elsewhere, and when it comes to those the Court will, one would suppose, apply its judgment in Hoffmann and allow all measures by way of judicial enforcemrent of awards to take place without interference from the Regulation, just as the Dutch court did in Hoffmann. A mess, no doubt, and for legislative reform (at least by a legislator who knows what he is doing), but one in which the current state of the law is not uncertain.
Fair enough, “West Tankers is […] a political decision” which seems to be bad as many important issues were overlooked by the ECJ. Moreover, the arbitration exception of Article 1 (2)(d) appears to be practically deleted by the court.
Unfortunately, the ECJ failed to grasp that the important issue, which is bound to arise again in the European context, was not to do with the anti-suit injunction, but with the question which law is governing the arbitration agreement. More particularly, the key issue before the English and Italian courts was: who were the parties to the arbitration agreement? Italian court applied Italian law and the English court applied English law. Accordingly, different results were reached.
The problems with the ECJ ruling is that the concept of preliminary issue, as adopted by the ECJ, taken together with the broad bases for jurisdiction available under the Brussels I Regulation would make easier to defeat an arbitration agreement by initiating court proceeding before one court instead of another.
The Heidelberg proposal for the protection of arbitration agreements would not suffice because it ignores the fact that the arbitral tribunal should rule on its own jurisdiction. In other words, the questions what is the scope of the arbitration agreement and who are the parties to it should be dealt with by the arbitrators. The court intervention should be limited because this is what the parties have agreed for.
Therefore, if the arbitration exception was to be deleted, then a better solution would be: if a court of the Member State at the seat of arbitration had exclusive jurisdiction in respect of the issues related to existence, validity, and/or scope of the arbitration agreement.
The ECJ judgment comes as no surprise, neither in content nor in style. The court rightly focussed on the issue whether the proceedings before the Italian court fall into the scope of the Brussels Regulation. Since this is the case, the claimant cannot be denied its right to go to the Italian court in order to question the existence of the arbitration agreement. Since Turner, this line of reasoning is well-established. What distinguished West Tankers from Turner is that the English proceedings do not come within the scope of the Brussels Regulation. Here the effet utile argument comes into play.
However, the principle of mutual trust does not add much to the analysis. Why should an English court, being concerned with proceedings that fall out of the Brussels Regulation due to art. 1(2)(d), have trust in the Italian court in ruling on the validity of the arbitration agreement? Unfortunately, the court does not provide an answer.
In this context, the argument in para 33 is beside the point. Article II(3) of the New York Convention does not “support” the ECJ’s finding. At most, it might not contradict the ruling. Since Article II(3) requires a court to refer the parties to arbitration if there is a valid arbitration agreement, one could even argue that the power to issue anti-suit injunctions in support of arbitration agreements is perfectly in line with article II(3).
Moreover, the application of article II(3) raises another problem: Whether the foreign court will actually refer the parties to arbitration depends on the law that the court applies as to the validity of the arbitration agreement. Article II (3) itself does not contain a clear-cut conflict-of-laws rule. As Article V(1)(a) of the New York Convention indicates, the determination of the applicable law is rather a matter of national law. According to many legal systems, the law of the seat of the arbitration applies (subject, of course, to a public policy exception). Yet, it is also argued that article II(3) incorporates internationally-applicable substantive contract law standards designed to ensure that idiosyncratic and discriminatory national laws do not obstruct the enforcement of international arbitration agreements. But this issue is highly controversial and far from settled even in European arbitration laws. Thus, since European procedural law does not provide any uniform rules in this respect, there is no reason for a European court to “trust” that the foreign court will eventually refer the parties to arbitration.
Yes, I agree with Burkhard Hess that the way forward now is amending the Brussels I Regulation in such a way as to provide efficient protection of choice of forum agreements in favor of courts or arbitrator in a Member State against the risk of parallel proceedings in other Member States.
One can only presume that the lameness of ECJ’s appeal to the need to ensure the effectiveness of the Regulation in order to prevent “the attainment of the objectives of unification of the rules of conflict of jurisdictions in civil and commercial matters”…is deliberate, and designed to prod the legislator into action – including action to redress the Court’s own past faux-pas(in Gasser, particularly).
I’d agree that several concrete steps are needed. First, delete the arbitration exception so as to bring court proceedings relating (that is, in support of, or concerning the validity of, the arbitration clause) to arbitration within the ambit of the Regulation (see too Pavel Simon’s address to the Legal Affairs Committee of the European Parliament on 19th January 2009); second, in cases of parallel proceedings, modify article 27 so as to exempt from the first-in-time rule cases in which the court second seized is the court chosen under a choice of court clause OR the court of the seat of the arbitration; ensure then that article 23 really means “exclusive” when it says “exclusive”(modify article 35-1 to that effect). Then let the New York Convention take over as it naturally should and consider, for instance, whether disregard for the negative effect of competence-competence by the court of a Contracting/Member State should be a cause of refusal of recognition of its judgment in other Contracting/Member States. Indeed, although one can but agree with Adrian Briggs that nothing in West Tankers will stop the arbitration from going on anyway in London, nevertheless, one cannot rule out the possibility that the Italian court will might exercise its jurisdiction on the merits (see the discussion already posted on the law applicable to the arbitration clause), so that in the (perhaps unlikely) event of an Italian judgment being swiftly obtained on the substance of the dispute, the position of the various Member States under the Regulation could become distinctly uncomfortable.
In this respect, one must rejoice, anyway, that mutual trust has taken a back seat in the ECJ’s decision, as compared with the importance given it, in support of an identical conclusion, by the AG. Mutual trust should not be made to meant that the decision of ANY court of any Member State that the breaching party happens to seize is entitled to preempt proceedings engaged everywhere else. It could mean, however, that the courts of the seat of the arbitration or the court chosen by the parties has priority to decide on the validity of the clause. Since Gasser and West Tankers, this fairly obvious point might need to be set out clearly in the Regulation.
Well, alright: here we go. The judgment just goes to show that you can expect the worst and still be disappointed. (1) It fails to refer to, never mind deal with, the Court’s previous decisions on the material scope of the Convention/Regulation and what it means for a national court to adjudicate in a matter which falls outside it. This is presumably because these cases were not even mentioned, never mind analysed, in the Advocate-General’s Opinion: what is going on, there ? But it is hard to see how they could have left much room for doubt, so maybe this is why they were ignored ? Who knows. (2) Whereas it might be seen as a ‘preliminary issue’ for a Sicilian judge to decide whether the parties’ agreement to arbitrate means that he or she should refer them to the agreed tribunal, the question of where this decision fits into the Sicilian court’s procedure is irrelevant to the distinct question of the material scope of the Regulation and what follows from being outside it. (3) Is there anybody who has any experience of commercial dispute resolution who actually thinks that this is a good outcome ? The objection has nothing whatever to do with English or other national arbitration laws, but everything to do with the fact that when the Contracting States and Member States said that arbitration was to fall outside the material scope of this instrument, that arbitration was to be allowed to get on with settling commercial disputes, it was their decision, and no-one else’s, to do that and say that. That this might be reversed at a later date is nothing to the point. How does the rule of law actually work on the Kirchberg plateau ? (4) Otherwise there is little more to this than bare assertions and non sequiturs. It is not as though the judgment and ‘reasoning’ of the Court could not be justified. It’s just that it cannot be justified by the legislative instrument which it purports to be interpreting. (5) As to priority for the courts or tribunals allegedly chosen, it is in some ways a good idea, but not an essential one: there are plenty of cases in which there is a genuine dispute as to validity and scope; and the freedom to choose, and to hold another to the choice, also connotes the freedom to not choose, or right not to constrain one’s freedom of access to a court. More thought may be needed here: after all, Professor Schlosser said (did he not?) that the idea that a party who denies that he is bound by an agreement to arbitrate has to convene a tribunal in order to have it rule that he not bound to have recourse to it after all is not one based on coherent reflections. (6) But really, does it matter ? The saving grace is that the arbitration will go ahead, just as it should, and unless there is more to the vague and imprecise nonsense in paragraphs 30 and 31 than is declared at first sight, we can all reconvene after the award. Especially if the tribunal includes in its award the costs of this ridiculous judicial frolic, as part of the loss which falls within the reference. (7) I have written this sorry judicial effort up for the May edtion of the LMCLQ. Who else, and where, please ?
Well, West Tankers is a bad political decision, but nothing more than that.
It seems to me that for arbitration in general, the consequences of West Tankers can be minimised by saying that the ECJ ruling would only be relevant in cases where there were a direct claim that a party was not bound by an arbitration agreement. (I still cannot understand how such a claim can be within the scope of the Brussels I Regulation, but I am glad for the AG, the judges and the academics, who managed to justify that.) If there were not such a direct claim, then the legal proceedings should be stayed under the NY Convention (i.e. Kompetenz-Komptenz and Separability).
In other words, if the parties to an arbitration agreement include a clause with the effect that “‘any person claiming under or through a party to an arbitration agreement’ is bound to arbitrate”, then no issue would arise at all. Would it?
Heildelberg Report: a solution of Private International Law in EU
Kompetenz-Kompetenz is useful but the exact scope of the negative effect of this principle is different in the Member States. In this case (Tankers) the arbitrator does not have the priority in Italy.
The “Heidelberg Report” seeks to provide both a coordinated solution and a territorial solution (Private International Law). The exclusive jurisdiction of the Member-State in which the arbitration takes place is entitled to decide. What decides? : “Kompetenz-Kompetenz” (Geneva-Convention /art. 1458 NCPC) or the control of the judge (§ 1032 ZPO-2) or the control of the judge and worried about saving costs (art. 9 Arbitration Act).
The proposition is a PIL (territorial) coordinated solution. The choice between an arbiter and a judge is out the scope of the Brussels I Regulation and the solution “substantial” is out. Each Member State (in which the arbitration takes place) decides according to their concept of international arbitration.
The arbitrators (London, ICC ..) don’t want PIL solutions. They are more and more preferring to extend the negative effect of Kompetenz-Kompetenz (the arbitrator has priority). But, is it possible to adapt the NY Convention (1958) whit the negative effect of the “Kompetenz-Kompetenz” (the arbitrator has priority)? For this point of view (transnational) the Heidelberg solution is territorial (national) not “neutral” and “out of date”. For this point de view the arbiter is “neutral” (transnational – delocalized) and Kompetenz-Kompetenz “the way”.
Heidelberg Report is a PIL solution (localized). Is it a choice about the concept of international arbitration (territorial, inter-national, localised v. transnational, delocalized) in the EU and in the world also? – AQE.
RE: Heidelberg Proposals
To my mind, the problem with the Heidelberg proposals seems to be that it ignores that the arbitral tribunal may rule on its own jurisdiction, including any objections with respect to the existence or validity of the arbitration agreement. (See Article 16 of the UNCITRAL Model Law; Ss 7 and 30 of the English Arbitration Act 1996).
Nevertheless, the existence, validity or scope of the arbitration agreement could be examined by national courts before an award is rendered in the context of an application for a stay of legal proceedings under Article II – NY Convention (or S 9 of the 1996 Act).
If the arbitral tribunal exceeds its jurisdiction, then an arbitral award may be set aside by the court at the seat of arbitration (see Article 34 UNCITRAL Model Law; S 67 of the 1996 Act); or its recognition and enforcement could be denied (See Article V of the NY Convention; S 103(2)). Thus, there are control mechanisms, but this is not a matter for Brussels I.
However, the Brussels I Regulation may confer exclusive jurisdiction to the Member State court at the seat of arbitration in respect of an application for a stay of legal proceedings. In this context, the conditions laid down in Article II of the NY Convention (or S. 9 of the 1996 Act) should be examined (i.e. inter alia the validity and the scope of an arbitration agreement will be dealt with by national courts). The Member State court should stay its proceedings unless there is a direct claim that the party was not bound by the arbitration agreement (e.g. no valid arbitration clause or see the facts of West Tankers).
I would like to comment briefly on the last post of Michail Danov and on the post of Anna Quinones:
1. Competence-competence of the arbitral tribunal: This problem has been addressed in the Heidelberg Report at para. 125. The competent civil court of the Member State where the arbitration takes place may stay its proceedings and refer the case to the arbitral tribunal which will decide on the question of “competence-competence”. Such a stay should not be unconditional. The civil court should set a time limit for the decision of the arbitral tribunal. To my opinion the civil court could endorse the decision of the arbitral tribunal. Accordingly, a (declaratory) judgment on the validity of an arbitration agreement could be recognised under Article 32 of the Brussels I Regulation.
The Heidelberg Report does not propose any procedure for this issue leaving it to the domestic law of the EU Member States and the NY Convention. This corresponds to the structure of the Regulation which does not address procedural law in general. However, there is no doubt that the competent courts under Article 22 No. 6 Brussels I Regulation are bound by Article II of the NYC when deciding on the validity of the arbitration clause or transferring the decision to the arbitration tribunal.
2. Delocalisation of international arbitration: You are completely right; the Heidelberg Report is based on a territorial conception of international commercial arbitration. I am well aware of the fact that this proposal contradicts those legal concepts which perceive international arbitration as delocalised and detached from national procedural laws. However, I would like to stress the fact that the underlying principle of mutual trust among EU Member States does not permit the non-recognition of a court decision of another Member State setting aside an award for a violation of EU-anti trust law. In the internal market, the uniform and consistent application of (binding) Community economic law is (absolutely) necessary. If an arbitral award has been set aside in one Member State because it contradicts mandatory Community law, it cannot be recognised as valid in another EU-Member State. This is not a question of delocalising arbitral awards, but of applying Community law in a uniform and consistent way. Accordingly, the recent decision of the French Cour de Cassation of 6/7/2008, Société SNF ./. Société Cytec Industries BV, Rev. Arb. 2008, 472, seems to be doubtful as the Belgian award had been set aside by the civil court in Brussels (I do not want to comment whether the Brussels or the French Court was right deciding on the substance of the claim).
However, it seems to be possible to distinguish intra-community situations from the relations to third states: The recognition of an award which was set aside under suspicious circumstances by the civil courts in a third state (like the famous Chromalloy award) will still be possible. The situation is different in relation to courts of other EU Member States: The principles of mutual trust and the uniform application of Community law in all Member States require the establishment of a consistent framework for the recognition (and enforcement) of arbitral awards. In this respect, the regime of the New York Convention will be complemented by a regional regime – similar to the Geneva Convention on International Commercial Arbitration.
– Finally, I would like to stress that I am expressing here only my personal opinion.
Thank you for clarifying that for me.
My concerns are with the “Kompetenz-Kompetenz” principle. You suggest to delete Article 1(2)(d) and you go further saying that new Article 22(6) aims to concentrate all proceedings challenging the validity of the arbitration agreement in the domestic courts of the Member State where the arbitration takes place. However, validity of an arbitration agreement may be challenged before the arbitral tribunals. Further, the question what is the scope of an arbitration agreement may arise as an issue before the arbitral tribunals.
Does the Heidelberg proposal mean that an arbitral tribunal must stay its proceedings and refer the matter to the court, which has exclusive competence for proceedings challenging the validity of the arbitration agreement? The example, which you have given above, adds to my concerns – you say that “the competent civil court of the Member State where the arbitration takes place may stay its proceedings and refer the case to the arbitral tribunal which will decide on the question of “competence-competence”.” Does it mean that, in your example, there is an arbitral tribunal, which is properly constituted, and the validity or scope of the arbitration agreement is still an issue for the court, which may refer it back to the arbitrators?
Why should the stay not be unconditional, provided that the conditions of Art II NYC are satisfied? If the conditions of Art II NYC are not satisfied, then why should the parties be referred to arbitration?
Further, arbitrators may render an interim award on the issue of jurisdiction. Such an award may be challenged before the court at the seat of arbitration. How would the proposed endorsement of the interim award fit into the current regime for challenges?
I am sure that the Heidelberg Proposal is not meant to undermine the “Kompetenz-Kompetenz” principle, but this should be clearly stated. For example, if you say “exclusive jurisdiction with regard to questions of existence and scope of an arbitration agreement arising in the context of an application for a stay of legal proceedings” (i.e. Art II NYC), then this will give certainty.
Dear Mr. Danov,
just a very brief clarification: The proposed Article 22 No. 6 of the Brussels I Regulation shall only address the jurisdiction of civil courts of EU Member States. It only refers to jurisdictional conflicts among civil courts of different EU Member States, it does not address the relationship between civil courts and arbitral tribunals. In this respect, the New York and the Geneva Conventions and, mainly, the domestic laws of the EU Member States shall apply. Accordingly, the Heidelberg Report leaves the question of the so-called “Kompetenz-Kompetenz” to the domestic laws of the EU Member States. However, you are certainly well aware that the issue of “Kompetenz-Kompetenz” is dealt with differently by the arbitration laws of EU Member States. For example, Germany does not recognise the “Kompetenz-Kompetenz” of the arbitral tribunal, although the word “Kompetenz-Kompetenz” is German (but not a very felicitous linguistic expression).
Dear Prof. Dr. Hess,
Many thanks for your kind response which is appreciated.
However, it would probably be better, if the Recitals of the Brussels I Regulation had reflected your clarification that despite the repeal of Article 1(2)(d) the proposed Article 22(6) refers only to jurisdictional conflicts among Member State courts. There are at least two good reasons for that. First, the “exclusive competence” of the courts in a given subject matter may sometimes affect the jurisdiction of arbitral tribunals in the same subject matter. Secondly, we all know that the relationship between the courts and the arbitral tribunals should not be a matter for Brussels I. But then, legal proceedings which had arbitration as their subject matter were supposed to be excluded from the Regulation, so some more clarity would be welcomed.
In terms of terminology, I see that you prefer to use “the place of arbitration” as a connecting factor. However, as you are aware, this could give rise to some uncertainty in cases where the seat of arbitration is in one Member State, while the place of hearings is in another Member State (see Article 16 of the LCIA Arbitration Rules 1998). It seems to me that the “seat of arbitration”, being a narrower term, might give more certainty.
Otherwise, the parties to arbitration agreements are well aware of the different arbitration laws and they take them into account when selecting the seat of arbitration.
Overall, there is no room for doubt that your proposal deserves support. Hope it goes well.
This is an interesting debate. The problem is that the Court of Justice, methodologically and historically, is a supranational court with a sort of perpetual task to safeguard and even promote the effectiveness of Community law. We owe the Court of Justice a lot as Community citizens, but one cannot deny that in the area of conflict of laws and arbitration, its teleological and effectiveness-driven methodology sounds rather one-sided and misses the broader picture. I am thinking here of cases such as Ingmar, Movil, even Eco Swiss (to an extent).
I think it is also a question of education. The legal community must convey to the Court that in a mature system such as our Community law system now is, the Court can be sufficiently relaxed to understand that “there is a world out there” and nuance its rulings.
Finally, a point always missed with regard to arbitration: It is incorrect to view arbitration only as a creation of private autonomy which from time to time may have to allow for inroads by Community law; arbitration is itself mentioned in Article 293 EC. There, the necessity of recognition and enforcement of arbitral awards is mentioned side-by-side with the necessity of recognition and enforcement of judgments. That there was no follow-up in the European context in the sense of a specific convention on arbitration following the example of the Brussels Convention is not due to the fact that arbitration was seen differently but rather to the fact that there was already a very efficient tool to ensure enforcement of arbitral awards worldwide, and thus also within the then EEC, the New York Convention. This “Treaty parameter” of arbitration must therefore not be missed. Exactly like free movement of judgments, free movement of arbitral awards within the EU furthers European integration and is extremely beneficial to the four freedoms. It should therefore be accorded the same degree of deference.