Subsidiaries as “branches” for undertakings: a new route to jurisdiction under Article 5(5) of the Brussels Regulation?

Stand alone, follow on and hybrid damages claims arising out of multijurisdictional cartels are generating some of the most novel and interesting current problems in conflicts of laws, both in relation to issues of jurisdiction and applicable law. On the jurisdictional side conventional wisdom has it that there are three main routes by which Claimants can seize English jurisdiction.

First, you can find a so-called “Anchor Defendant” that is a cartelist (and it must be an addressee cartelist if in the CAT so long as Mersen is good law) domiciled here, against which you can proceed as of right under Article 2 of the Brussels Regulation.  Then you can bring in other cartelists under Article 6 (i.e. a defendant against whom the claim is closely connected to that against the anchor defendant such that determining them together avoids the risk of irreconcilable judgments).  Where the Anchor Defendant is an addressee of the decision this tactic is unproblematic.

But of much more dubious standing, even after the recent Court of Appeal decision in Toshiba Carrier, is the treatment of a subsidiary not named in the cartel decision as an Anchor Defendant because of its use by the parent to implement the cartel in the UK.  The central perceived weakness of this argument is its artificiality, treating as it does the subsidiary as a wrongdoer in its own right, artificially imputing to the subsidiary (in effect) knowledge/conduct of the parent rather than (as is the more conventional exercise in competition law) imputing its conduct to the parent in circumstances of actual control by the parent.

The tenuous nature of such Provimi claims, which will undoubtedly lead to a CJEU reference if such claims abound (as the Court of Appeal so clearly thought appropriate in Cooper Tire) has led to the second main tactic, which is to bolster such claims against subsidiary anchor defendants by pleas of inferred knowledge.  Given the pleading latitude afforded to claimants, because of the well recognized information asymmetries, this tactic is presently in the ascendant.  Claimants generally do not really mind whether or not these equally artificial claims of inferred knowledge will stack up at trial as by then they will have served their purpose which is to seize jurisdiction, merely having to cross a Canada Trust threshold of arguability.

Then there is the third recognized tactic, evident in the Bord Na Mona case, which is used where there is not even a UK subsidiary to sue.  In such cases jurisdiction is founded on Article 5.3 on the premise that the cartel has caused damage in the United Kingdom as a result of the competition infringement.  The problem with this route to jurisdiction, at least so long as Shevill v Presss Alliance remains good law (the potential implications of eDate and Wintersteiger for Article 5.3 competition claims is probably another blog subject for another day), is that one can only recover damages for that loss suffered in the territory of the Courts so seized (i.e. the UK).  How that differs from EU wide damage in cartels that suppress parallel trade is a further interesting question.

My thesis is a simple one: there is a fourth route to jurisdiction.  That route is Article 5.5, which provides that a Defendant may be sued in a country other than its place of domicile “as regards a dispute arising out of the operations of a branch, agency or other establishment, in the courts for the place in which the branch, agency or other establishment is situated”.  The rationale for the “agency” rule is self-evident: the agency is a real extension of the principal’s area of activity that serves to establish its material presence in the host country.  That presence is then treated in conflicts policy terms as sufficient to found jurisdiction. Unsurprisingly, the CJEU’s relatively limited case-law already confirms that the term “branch, agency or other establishment” has an autonomous EU law meaning.  Whether or not an entity is in a branch or agency (“A”) for a parent or principal (“P”) is a question of substance, which has at its heart questions of whether P controls A in all material respects, whether A can take actions that bind P and/or whether P holds out A as acting for it, as the case law summarized at §2.195 of Briggs & Rees (2^nd ed) confirms.

None of this case-law as yet grapples with the problems presented by a multi-jurisdictional cartel.  But the interaction of this term in a competition law context with: (a) the economic/corporate veil-piercing concept of an “undertaking”; and (b) extant competition law on “true agency” (which can surely make supposed principals into agents, just as it makes supposed agents into principals), must clear the ground for an argument that the notionally independent subsidiary with no real (as opposed to formal) scope for determining its own commercial policy is simply the presence of the infringing “undertaking” in the jurisdiction in just the same way as a branch or agency formally so called, thus triggering precisely the same justification for proceeding against the parent (rather than the subsidiary/branch) in the place of domicile of the “branch”.  This conclusion can only be fortified by the fact that a good deal of the P/A arrangements have their origins in transfer pricing schemes designed to channel (as it happens cartelized) profits to low tax jurisdictions; and that some subsidiaries act as no more in substance than sales agents for parents, which take all material decisions on group pricing (as both the facts of Provimi v Aventis and Toshiba Carrier both strongly suggest).

In many claims this fourth route may subsume and replace the Provimi device and the (artificial) inferred claim of knowledge and the limited Article 5.3 claims; indeed it appears like a more straightforward and less artificial package for many of the arguments that moved the Judges in Provimi and Toshiba Carrier.  In other cases, this argument will provide a useful supplement to the other three arguments, creating something of a Hobson’s choice: if the subsidiary wishes to argue it had a true, independent pricing policy, then pricing convergence and timing will form the basis of a proper case of inferred knowledge (particularly if accompanied by personnel overlaps with the parent cartelist); if it wishes to downplay its scope for independent action, then the case that it is in substance a branch used by the parent to implement the cartel becomes so much the stronger.