All EU member states are party to the New York Convention, the worldwide treaty underpinning arbitration.

Article II of the New York Convention imposes an obligation on contracting states to recognise an arbitration agreement. Articles III to V impose obligations on contracting states to recognise and enforce arbitral awards, save in limited circumstances. Under Article IV, a party need produce only the award and the arbitration agreement to obtain recognition and enforcement. Under Article V.1(a), a contracting state can refuse to recognise and enforce an award if the arbitration agreement is not valid under the law to which the parties have subjected it, or failing any indication thereon, under the law of the country where the award was made.

The Brussels Regulation (44/2001/EC) allocates jurisdiction amongst the courts of EU member states in civil and commercial matters and provides for mutual recognition of judgments, subject to limited exceptions.

Article 1(2) provides:

Article 1(2)(d) is known as the "arbitration exclusion". It mirrors an equivalent provision in the Brussels Convention (the Brussels Regulation's predecessor), and the Lugano Convention (the parallel convention with Iceland, Norway and Switzerland).

Article 73 of the Brussels Regulation requires the European Commission to present a report on its application and proposals for adaptations, no later than five years after its entry into force. Somewhat belatedly, the Commission presented its report (COM(2009) 174 final) and a green paper (COM(2009) 175 final) proposing reforms on 21 April 2009.

In connection with this review, Professors Burkhard Hess, Thomas Pfeiffer and Peter Schlosser were commissioned to prepare a study into the workings of the Brussels Regulation (Study JLS/C4/2005/03, Report on the Application of Regulation Brussels I in the Member States, Final Version 2007) (the Heidelberg study).

The green paper proposes the following reforms to the Brussels Regulation, based largely on suggestions made in the Heidelberg study:

The green paper makes additional proposals, which were not put forward in the Heidelberg study. These include to:

As part of their research, the authors of the Heidelberg study canvassed opinions on the interface between arbitration and the Brussels Regulation. The overwhelming response was that the Brussels Regulation should not be extended to arbitration (paragraphs 108-114, Heidelberg Study).

Why, then, has the Commission proposed reform? At a recent roundtable discussing the proposals, Karen Vanderkerckhove of the Commission explained its concerns. Within the "European area of justice" the European authorities "have not tolerated uncertainty as to which courts have jurisdiction to act, we have not tolerated parallel proceedings for dispute resolution and we have not tolerated uncertainty when conflicting decisions exist" (D Samuels, The policy backdrop, Global Arbitration Review, Special edition: roundtable on the EU green paper, 5 August 2009).

The capacity for parallel proceedings and inconsistent judgments under the existing regime was highlighted in February 2009 in the West Tankers case. The English courts considered that proceedings brought by Allianz in the Italian courts were in breach of an agreement providing for arbitration in London. However, the ECJ held that it would be incompatible with the Brussels Regulation for the English courts to issue an anti-suit injunction, which would prevent the Italian court deciding for itself whether it should decline jurisdiction on the basis of Article II of the New York Convention.

The Commission's solution is to centralise decision-making in the courts of one member state: the courts of the seat of arbitration. If the proposed new Article 27A had been in force, the Italian court in West Tankers would have had to stay its proceedings as soon as an application was made to the English courts for a declaration that the arbitration agreement applied to the dispute. Moreover, the deletion of the arbitration exclusion would mean that the English court's judgment would be entitled to recognition in Italy (and other member states) under the Regulation. The proposals to give the courts of the seat exclusive jurisdiction in proceedings in support of arbitration and to certify the enforceability of awards would further reinforce this centralisation of decision-making.

The Commission invited responses to its green paper by 30 June 2009. These responses are available on the Commission's website.

Some respondents have expressed support for some or all of the Commission's proposals (see, for example, the responses from the Council of Bars and Law Societies of Europe and the Law Society of England and Wales). However, many arbitration practitioners have criticised the Commission's proposals as likely to undermine the operation of the New York Convention and make EU member states less attractive as seats of arbitration (see, for example, the responses of the International Bar Association's Arbitration Committee, the Association for International Arbitration, Brussels, the Comité Français de l'Arbitrage and the Chamber of National and International Arbitration of Milan).

The ball is now back in the Commission's court. It is currently considering the responses to the consultation exercise and is expected to produce revised proposals later this year. The following options are open to the Commission. It can:

These options are discussed in turn below.

The first option available to the Commission is to preserve the status quo. This is the preferred approach of a number of distinguished respondents to the consultation who are concerned that the proposed reforms would harm arbitration in Europe (see, for example, the responses of the Comité Français de l'Arbitrage and of Professor Emmanuel Gaillard).

However, the ECJ's West Tankers decision already threatens serious harm to arbitration in the EU. Much of the publicity around that decision focused on anti-suit injunctions. However, the ECJ's reasoning has potentially far-reaching consequences going beyond this.

The ECJ had previously held that court proceedings are excluded from the Brussels Regulation if their subject matter is arbitration (Marc Rich & Co AG v Società Italiana Impianti PA (Case C-190/89)). In West Tankers, the ECJ developed this reasoning. The English proceedings seeking a declaration upholding the arbitration agreement (and an anti-suit injunction) fell outside the Regulation. On the other hand, the Italian proceedings were brought on the merits of the dispute and fell within the scope of the Regulation. Moreover, "a preliminary issue concerning the applicability of an arbitration agreement, including in particular its validity, also comes within its scope of application" (West Tankers, at paragraph 26).

This produces what the Bar Council describes in its response to the green paper as an "unacceptable asymmetry" (Bar Council response at paragraph 7.3). Indeed, West Tankers produces perverse results:

This is likely to lead to further costly litigation as parties and courts seek to mitigate the adverse effects of West Tankers. This has already started. In one case, the English High Court invoked public policy to refuse recognition of a judgment of a Spanish court which denied that a dispute was subject to an arbitration agreement (National Navigation Co v Endesa Generacion SA [2009] EWHC 196 (Comm)). It is questionable whether the ECJ would uphold that decision. At the very least, there will have to be further references to the ECJ to settle the matter.

There are good reasons, therefore, to conclude that the status quo is not sustainable.

The next alternative is to take action to undo or mitigate the effects of the West Tankers decision. In its response to the consultation, this was the approach favoured by Allen & Overy LLP: to amend the Brussels Regulation to permit a court to refuse enforcement of a judgment that was irreconcilable with an arbitration agreement or award which would be recognised in that member state. This could be achieved by a limited amendment consistent with retaining the arbitration exclusion (Allen & Overy's response at paragraphs 105-107).

The Bar Council goes further, advocating amendments to undo not only the effects, but also the reasoning of West Tankers. It proposes clarifying the scope of the arbitration exclusion to make clear that a judgment obtained in breach of an arbitration agreement and a judgment holding that an arbitration agreement is ineffective would fall outside the scope of the Brussels Regulation altogether (Bar Council response at paragraph 7.4).

These proposals leave open the possibility of parallel proceedings and inconsistent decisions. However, the perverse incentives of West Tankers are eliminated, and no member state's court would be forced to choose between its obligations under the Brussels Regulation and under the New York Convention. A party faced with a torpedo action brought by its opponent would be free to choose whether or not to enter an appearance to contest the court's jurisdiction, or to ignore the proceedings and pursue the arbitration. If its opponent had assets in jurisdictions other than that where the torpedo action was commenced, it might well decide to ignore it.

The third possibility is that the Commission will proceed with the approach advocated in the green paper. That approach has at least one advantage when compared to the status quo following West Tankers. If any court is going to decide the existence, validity and scope of an arbitration agreement, it is submitted that it is preferable that the courts of the seat do so. It is better to give priority to the courts of the seat, which can be chosen by the parties in their arbitration agreement, than to a court which just happens to have been seised with the merits of the dispute by one party, particularly if such court has been seised as a deliberate torpedo tactic. (For a contrary view, see the response of the Swiss Bundesamt für Justiz, which approves of the West Tankers decision.)

However, the Commission's proposals are seriously flawed. At the heart of these flaws is the fact that the Brussels Regulation is simply not an appropriate instrument for harmonising arbitration law. Its object is to allocate jurisdiction between courts; it seeks to achieve consistency through procedural mechanisms: lis pendens rules and mutual recognition of judgments, and it looks inward towards intra-EU conflicts of jurisdiction without considering relationships with other countries. This gives rise to a number of issues:

There are further reasons to criticise the details of the proposals:

A further important consideration is that, if the Brussels Regulation is amended so as to encompass arbitration-related court proceedings, the EU will gain external competence over (at least some) arbitration matters. It will be for the EU, not the member states, to agree further international instruments, and member states will be bound by the EU's common position at international bodies such as UNCITRAL. Given the differences between member states' approaches to arbitration law, finding such common positions is likely to be extremely difficult. More generally, including arbitration within the Brussels Regulation may lead to pressure to harmonise further aspects of arbitration law, thus reducing the opportunity for innovation. The risk is that any common position will in fact be the lowest common denominator, to the detriment of the users of arbitration and of member states that wish to attract them.

The final option open to the Commission is to come forward with a modified version of its proposals. Some of the possible modifications will be evident from the discussion above; for example, it could exclude provisional measures and evidence-gathering from the proposed exclusive competence of the seat. Other modifications are less obvious: the proposed default rule for the seat of arbitration has attracted much criticism but few alternative suggestions.

The key element of the reform package, however, is the allocation of jurisdiction to decide the existence, validity and scope of an arbitration agreement. Professor Hess has indicated that it was never the intention of the authors of the Heidelberg study to upset the rules on kompetenz-kompetenz in the member states when they proposed the new Article 27A (See D Samuels, Segment on "anti-torpedo torpedo" proves bittersweet for its inventor, Global Arbitration Review, Special edition: roundtable on the EU green paper, 5 August 2009). The proposed Article could be redrafted to provide for a stay of court proceedings whenever the existence, validity and/or scope of an arbitration agreement is either before the arbitral tribunal or before the courts of the seat. This ought to preserve kompetenz-kompetenz. A number of respondents to the green paper have made suggestions along these lines (see, for example, the responses of the IBA Arbitration Committee at paragraph 27, Allen & Overy LLP at paragraphs 118-120, Herbert Smith LLP at paragraph 7(a)).

Such modifications will not entirely eliminate the difficulties with the Commission's proposals. Only by allowing a court to refuse to recognise a judgment that is irreconcilable with an arbitration agreement that it considers valid can the clash of obligations between the Brussels Regulation and the New York Convention be avoided. However, such a provision, largely defeats the object of bringing such judgments within the scope of the Regulation.