In Part I of my post, the revised “Brussels I” Regulation (Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters) was discussed in the light of a hypothetical example from international trade and arbitration. This Part 2 strives to outline the deficiencies that the amendment may bring about against the background of the same example.
While arbitration has been excluded from Brussels I, some grassroots of arbitration concern seem to creep in. Given the natural interrelation between arbitral proceedings and state court litigation (since any award handed down by a non-state dispute resolution body is subject to state court review upon a recognition and enforcement application or motion for setting aside), there always remains the possibility to have a strain of court decisions that directly or indirectly deal with issues arising out of arbitration. The present status of the Regulation suggests it will not be applied to them. The case law of the Court of Justice of the European Union (ECJ) indicates that currently it is not easy to infer a clear cut answer. Earlier on, in its Marc Rich & Co. AG v Società Italiana Impianti PA (C-190/89) judgment ECJ sought to exclude any arbitration issues from the scope of the then effective Brussels Convention.
It would also be contrary to the principle of legal certainty, which is one of the objectives pursued by the Convention […] for the applicability of the exclusion laid down in Article 1(4) of the Convention to vary according to the existence or otherwise of a preliminary issue, which might be raised at any time by the parties. It follows that, in the case before the Court, the fact that a preliminary issue relates to the existence or validity of the arbitration agreement does not affect the exclusion from the scope of the Convention … (para 27-28)
This apparently clear position was later eroded in the Front Comor case (Allianz S.p.A. v West Tankers Inc (C-185/07)):
In that regard, the Court finds […] that, if, because of the subject-matter of the dispute, that is, the nature of the rights to be protected in proceedings, such as a claim for damages, those proceedings come within the scope of Regulation No 44/2001, a preliminary issue concerning the applicability of an arbitration agreement, including in particular its validity, also comes within its scope of application. (para 26)
ECJ introduced a test of differentiation where the principal subject matter of the proceedings determine whether the case shall fall within the Article 1(2)(d) exception. Arbitration issues that are only preliminaries to the main subject of the case would not, ECJ reasoned in Front Comor, prevent the application of the Regulation. Although this may appear as a sound judgment, it opens the door wide for disruptive divergence, especially as to qualification of the scope of litigation proceedings.
Recital 12 of the revised Brussels I purports to settle the arising controversy. It may appear that the Regulation reiterates the same old position against arbitration. However, a close look in context demonstrates that in fact it confirms to a significant extent the position introduced by Front Comor, thus moving even farther from the initial “no-arbitration” stance of the Brussels Convention as expressed in Marc Rich. The Recital explicitly draws a line of distinction between a court ruling which has an assessment of the effect of an arbitration clause as its subject matter, and a decision where this assessment forms only an incidental to the substance of the case consideration. Hence, returning to the facts of the imaginary example, if the nullity of the arbitration clause is considered as an ancillary issue wherever the enforcement of the German decision is sought, then this decision would be within the application of the Regulation rules, including with respect to enforcement matters. Such a treatment should not be a surprise since the German case may easily be regarded as a claim against a defaulting contractor where the effect of the arbitration clause in the sales contract is only a jurisdictional issue.
Let us return to the hypothetical from Part I of this post. Leaving the rules of national legislations aside and reasoning on revised Regulation level only, let us assume that the Italian and French courts admit X to enforce against Y’s assets there under the easier enforcement route set by the Regulation, applying Recital 12’s rationale and not raising any objection to German court’s treatment of the arbitration clause nullity. In Belgium the arbitral award issued by CIETAC will have to be assessed under the rules of the New York Convention. Nothing in the Regulation, Recital 12 reads, shall prejudice the applicability of the New York Convention and it can be presumed that the award would be recognised and granted enforcement. As a result, a German court would rule that there is no valid arbitration clause and French and Italian courts would grant enforcement of such ruling, while a Belgian court would approve an award based on the same (valid) arbitration clause. In effect one European Union court would rule that a claim should be decided by a state court while another European Union court – that a non-state dispute settlement mechanism has already been the proper venue for deciding the same dispute. The outcome apparently runs contrary to the whole Brussels regime underpinned by the objective of unification of conflict-of-laws rules and, beyond that, the overarching ideal of the predictability and stability of the internal market.
Furthermore, in case the Belgian court is faster, it may recognise the award before the enforcement proceedings in Italy or France are finalized. Since the lis alibi pendens principle will not be applicable, the Italian and French courts shall not be entitled to stay the proceedings until the outcome of the exequatur in Belgium. However, would then the procedure in the French or Italian court (as the Regulation is applicable to these proceedings, but not to the award enforcement in Belgium) fall within the exception in Article 34(3) – incorporating former Article 34(4)’s rationale? This provision purports to prevent issuance of contradictory court decisions in different Member States. In fact, in these circumstances there would be a decision in another Member State (Belgium) which conflicts with the outcome of the case pending before an Italian or French court. If this is applied, then the French and Italian courts may refuse enforcement. As a result of the whole saga, X has convinced a EU court (German) that its contractor Y defaulted on its contractual obligation and has obtained a favourable decision; upon its attempt to collect its contractual refund from Y (against Y’s assets in Germany but also in France and Italy), X shall fall short from obtaining enforcement there and, at the same time, Y will be allowed to lay hands upon X’s assets in Belgium asserting that X is the actual defaulting party to the sale of goods contract.
This intertwined and complex imaginary scenario may seem far-fetched. However, it is entirely possible. Although it may seem as an absurd over-stretching of the amended provisions in the Regulation, it reveals one important caveat: leaving unregulated areas outside the scope of Brussels I always poses the risk that they will pervade in its safe grounds and disrupt the effect of the Regulation. Compared to the other excepted areas such as matrimonial and inheritance matters, administrative and customs cases, etc., arbitration is a mechanism typically devised for civil and commercial disputes. Hence, the integrity of the proper settlement means is disrupted. It is not unique that one and the same type of cases may be submitted to conflicting avenues of redress only on basis of the applicable dispute resolution mechanism but the special exclusion of arbitration from the Regulation opens the possibility for radically different outcomes at the enforcement stage in cross-border scenarios.
It is true that the Regulation is careful not to undermine the New York Convention and not to intrude into the purview of its application. However, the amended Brussels I incorporates in Recital 12 the case law reasoning of ECJ into secondary EU law, and thus creates a dichotomy that may seem justified – a preliminary issue, no matter whether dealing with arbitration or not, should not dramatically alter the regime applicable to the particular case. But the factual example given, by no means impossible against the background of international trade, reveals that beneath Recital 12’s treatment of the matter there is a lack of integrity, incoherence as a matter of principle. The new version of Brussels I not only leaves the problem of identical cases running in parallel unresolved but provides even stronger support for them reaching a dramatically different outcome, as demonstrated in the example. Although, ideally, New York Convention has to take precedence by virtue of Article 71 of the Regulation, this would have effect in practice only where two colliding decisions are submitted before one and the same national authority, e.g. if X enforces the German decision in, for instance, Portugal, and Y also seeks recognition of the arbitral award there in order to lay claim upon some of X’s assets located in Portugal. Presumably the Portuguese court would have to assess the validity of the arbitration clause and if effective, would give rise to the arbitral award. However, if, as described above, the differing decision and award are pending before different EU Member States courts, then the result may be opposite.
Hence, the amendment of Brussels I has actually confirmed an approach that stands in the middle between full exclusion and full inclusion of arbitration and, naturally, this may bring, at its best, only a haphazard and inconclusive result.