Posts Tagged Competition
Fly lal: Locus delicti commissi for anticompetitive agreements. And application of Article 7(5)’s extension to branch domicile.
AB flyLAL — Lithuanian Airlines (‘flyLAL’) operated flights from Vilnius airport in Lithuania until it was put into liquidation. According to flyLAL, its demise was caused by predatory (that is, below cost) pricing by the Latvian airline Air Baltic Corporation A/S (‘Air Baltic’). That predatory pricing was, it is alleged, part of an anticompetitive strategy agreed between Air Baltic and the operator of Starptautiskā lidosta Rīga (Riga international airport in Latvia, ‘Riga Airport’). Thus, Riga Airport and Air Baltic agreed to drastically reduce the prices paid by Air Baltic for services at Riga airport. The savings were then used by Air Baltic to finance the predatory pricing that drove flyLAL out of the market in Vilnius, Lithuania.
Can Air Baltic and Riga Airport for damages before the courts in Vilnius? The national court and parties refer to three alleged infringements of competition law: (i) abuse of dominance consisting in the system of reductions implemented by Riga Airport; (ii) an anticompetitive agreement between Riga Airport and Air Baltic; and (iii) abuse of dominance in the form of predatory pricing by Air Baltic. Those infringements, it is argued, were interrelated, forming part of a strategy to oust flyLAL from the market in Vilnius and move passengers to Riga airport to the benefit of both Riga Airport and Air Baltic.
There is a lot in the Opinion – among others because as the AG points out, the referring court’s description of the alleged infringement of competition law is not entirely clear. Bobek therefore sets out a set of variables. The Court itself is bound not to distinguish among quite so many. Of note are the AG’s suggestions
- that locus damni here is not place of financial loss, rather the place within the markets affected by the competition law infringement where the claimant alleges loss of sales.That suggestion in my view is helpful for neither the Regulation’s aim of predictability, nor the protection of those damaged by infringement of competition law (the latter not however a stated aim of the Regulation). Put differently: damage located in a Mozaik fashion assists the tortfeasor. The Advocate General reaches this conclusion after a thorough revisit of the initial Bier judgment (and Capotorti AG’s Opinion in same), ditto Marinari and Dumez France. Yet the continuing need to conceptualise the Court’s Bier rule illustrates again in my view the mistake made in that original judgment, to introduce a forum damni despite the utter lack of textual support for same.
- for locus delicti commissi with full jurisdiction, the AG distinguishes between Article 101 TFEU (as regards the alleged anticompetitive agreement between Air Baltic and Riga Airport, the place of the event giving rise to the harm (that is, the loss of sales by flyLAL), is the place of the conclusion of the agreement) and 102 TFEU (alleged predatory pricing by Air Baltic, the place of the event giving rise to the harm is the place where the predatory prices were offered and applied). With respect to Article 101 TFEU, Bobek AG suggests this is identical to the Court’s judgment in CDC . I am not too sure but I am biased. As I noted above, in my view the Court should steer clear of an application of Article 7(2) which allows those infringing competition law to forum shop by manipulating the place of decision-making. In CDC the Court held that ‘the identification, in the jurisdiction of the court seised of the matter, of a specific event during which either that cartel was definitively concluded or one agreement in particular was made which was the sole causal event giving rise to the loss allegedly inflicted on a buyer’ cannot be ruled out. That implies that in other cases the identification of such singular event can be ruled out and that many places may be consider locus delicti commissi.
- finally with respect to (now) Article 7(5), the activities of a branch. The AG does not specify what must be meant by a ‘branch’ – for the national court has already concluded there is such branch. The Advocate General here is perhaps unusually deferential to the factual finding. Whether there is a sufficient nexus between the activities of the branch and the dispute, in the case of tort-based claims requires the branch participate in at least some of the actions constituting the tort (at 137). Offering the fixed prices or otherwise having been instrumental in concluding contracts for services at those prices suffices. In such cases, the branch has again participated in the commission of an act that constitutes a necessary precondition for the abuse (at 142).
A lengthy opinion. And it all started with the fairly straightforward facts of Bier…
(Handbook of) EU private international law), 2nd ed. 2016, Chapter 2, Heading 18.104.22.168
In one of my many ponderings on research I would like to do but might never get an opportunity do (hence my repeated sharing of potential PhD topics) I came across an excellent post by Daniel Jowell QC on the application of EU competition law in the UK courts post-Brexit.
The usual disclaimer of course applies (let’s wait and see what happens in the future Treaty between the UK and the EU) yet one important consideration has wider appeal: how does one apply the classic conflicts suggestion that courts do not apply foreign public law, or if they do, do so with great caution?: both out of comity with the foreign State; and to protect one’s own ordre public.
Competition law is often seen as being of quasi-public nature. Daniel justifiably suggests that post Rome II (in which competition law is assigned a specific (if complicated) lex causae), the UK will revert to its standard rules which increase the possibility that UK courts might refuse to apply foreign competition law, including the EU’s, on public policy grounds.
One to remember.
(Handbook of) EU private international law, 2nd ed. 2016, Chapter 4, Heading 4.6.2.
Heavily loaded. Applicable law in follow-up competition cases: watch the Dutch Supreme court in Air Cargo.
Update March 2018. Quentin’s blog has a link to the SC decision refusing to take the case, considering it was academic given that an appeal against the decision of the European Commission is still pending before the EU courts. It has therefore not irreversibly been decided whether the eleven air carriers had violated European competition law. Most probably the case will be back, one imagines.
Quentin Declève alerted me to the Air Cargo damages compensation case currently making its way through the Dutch courts. (I have previously reported on jurisdictional issues re such cases; searching the tag ‘damages’ should help the reader).
I have difficulty locating the actual judgment addressing the issue in this post: namely applicable law in follow-up competition cases. I have however located one or two previous judgments addressing the damages claims assignment issue in same. This web of litigation seems to be particularly knotty and any help by Dutch or other readers would be appreciated.
At issue is whether Rome II applies to the facts ratione temporis; if it does, how Article 6 should be applied, in particular: locus delicti commissi, locus damni and ‘affected markets”; and if it does not, how the previous Dutch residual connecting factor ought to apply.
A case of great relevance to competition law and fair trading cases.
(Handbook of) EU private international law, 2nd ed. 2016, Chapter 4, Heading 4.6.2.
Towards the end of July, the Court at Amsterdam applied the recent CJEU judgment in CDC, on the application of (now) Article 8’s rule on anchor defendants. The case also involved CDC – busy bees on the competition enforcement front, this time pursuing inter alia Kemira, a Finnish company, using Akzo Nobel NV, domiciled in The Netherlands, as anchor defendants.
The court referred in extenso to the CJEU’s CDC case, noting inter alia that it is not up to CDC to show that the suit was not just introduced to remove Kemira from the Finnish judge: that Kemira suggests that introduction of the suit in The Netherlands is not very logical given the absence of factual links to that Member State, does not suffice. The court also adopted the CJEU’s finding on choice of court and liability in tort. In the absence of specific proviso in a standard contractual choice of court, the application of such choice of court to extracontactual liability [such as here, for infringement of competition law] cannot be assumed.
Finally, at 2.18, the Court also referred to argument made by Kemira that Finish and Swedish law ought to apply to the interpretation (not: the validity) of the choice of court agreement. That would have been an interesting discussion. However in light of the court’s earlier judgment on the irrelevance of the court of choice, the court did not entertain that issue.
Don’t leave the store without asking. Joinders, and the Aldi principle applied in Otkritie. On the shopping list for the EU?
Postscript 21 November 2017: For an application in Hong Kong see Far Wealth Ltd v Lo Ki Mou, reported here: proceedings dismissed as an abuse of process because the plaintiffs could have protected their position by way of a counterclaim in prior proceedings commenced against them by the defendants.
A posting out off the box here, so bear with me. Neither Brussels I nor the Recast include many requirements with respect to (now) Article 8(1)’s rule on joinders. A case against a defendant, not domiciled in the court’s jurisdiction, may be joined with that against a defendant who is so domiciled, if the cases are ‘so closely connected that it is expedient to hear and determine them together in order to avoid the risk of irreconcilable judgments’. There is of course CJEU case-law on what ‘so closely connected’ means however that is outside the remit of current posting.
As I reported recently, the CJEU has introduced a limited window of abuse of process viz Article 8(1), in CDC. The Court’s overall approach to Article 8(1) is not to take into account the subjective intentions of plaintiff, who often identify a suitable anchor defendant even if is not the intended target of their action. The Court does make exception for one particular occasion, namely if it is found that, at the time the proceedings were instituted, the applicant and that defendant had colluded to artificially fulfil, or prolong the fulfilment of, (now) Article 8’s applicability.
What if at the time the proceedings were instituted, applicant artificially ignores the fulfilment of, (now) Article 8’s applicability?
The Aldi rule of the courts of England and Wales, and its recent application in Otkritie, made me ponder whether there is merit in suggesting that the CJEU should interpret Article 8(1) to include an obligation, rather than a mere possibility, to join closely connected cases. I haven’t gotten much further than pondering, for there are undoubtedly important complications.
First, a quick look at the Aldi rule, in which the Court of Appeal considered application of the Johnson v Gore Wood principles on abuse of process of the (then) House of Lords, to an attempt to strike out a claim for abuse of process on the basis that the claim could and should have been brought in previous litigation. Aldi concerned complex commercial litigation, as does Otkritie. The result of Aldi is that plaintiffs need to consult with the court in case management, to ensure that related claims are brough in one go. Evidently, the courts need to walk a fine rope for the starting point must be that plaintiffs have wide discretion in deciding where and when to bring a claim: that would seem inherent in Article 6 ECHR’s right to a fair trial.
In Otkritie [the case nota bene does not involve the Brussels Regulation], Knowles J strikes the right balance in holding that the Aldi requirement of discussing with the court had been breached (and would have cost implications for Otkritie in current proceedings) but that otherwise this breach did not amount to abuse of process.
Now, transporting this to the EU level: to what degree could /should Article 8 include a duty to join closely related proceedings? Should such duty be imposed only on plaintiff or also on the court, proprio motu? A crazy thought perhaps for the time being, but certainly worthwhile pondering for future conflicts entertainment.
Anchor defendants in follow-up competition law cases. The ECJ confirms AG’s view on joinders. Sticks to Article 5(3 /7(1). Locus damni for purely economic loss = registered office.
In Case C-352/13 CDC, in which the ECJ held last week, at issue is among others the use of Article 6(1) of the Brussels I-Regulation (8(1) in the recast) when the claim against the anchor defendant has been settled before the trial is well and truly underway.
I reviewed JÄÄSKINEN AG’s opinion here. The ECJ’s overall approach to Article 6 is not to take into account the subjective intentions of plaintiff, who often identify a suitable anchor defendant even if is not the intended target of their action. Like its AG, the Court does make exception for one particular occasion, namely if it is found that, at the time the proceedings were instituted, the applicant and that defendant had colluded to artificially fulfil, or prolong the fulfilment of, Article 6’s applicability. I had expressed reservation vis-a-vis this suggestion, obviously in vain. In cases such as these, where tort is already clearly established (via the European Commission’s cartel finding), the intention of ECJ and AG seem noble. Collusion to defraud is disciplined by the non-applicability of Article 6. However this arguably serves the interests of the parties guilty of the other type of collusion involved: that of defrauding not procedural predictability, but rather consumers’ interest.
Next, the referring court enquired about the application of Article 5(3)’s special jurisdictional rule in the event of infringement of competition law, where that infringement concerns a complex horizontal agreement, spread over a long period of time, and with varying impact in various markets. The AG had suggested dropping application of Article 5(3) (now 7(1)) altogether, both with respect to locus delicti commissi and locus damni. Here the Court disagreed. Difficult as it may be, it is not to be excluded that locus delicti commissi can be established. At 50: one cannot rule out ‘the identification, in the jurisdiction of the court seised of the matter, of a specific event during which either that cartel was definitively concluded or one agreement in particular was made which was the sole causal event giving rise to the loss allegedly inflicted on a buyer.’
For locus damni, the Court again has no sympathy for either mozaik effect of Article 5(3), or indeed the often great difficulties in establishing locus damni, flagged by the AG. At 52: ‘As for loss consisting in additional costs incurred because of artificially high prices, such as the price of the hydrogen peroxide supplied by the cartel at issue in the main proceedings, that place is identifiable only for each alleged victim taken individually and is located, in general, at that victim’s registered office.‘
Registered office as the locus damni for purely economic loss, lest my memory fails me, has not been as such confirmed by the ECJ before. It is also currently pending in Universal. The Court is in my view a bit radical when it comes to justifying registered office as the Erfolgfort: at 53: ‘That place fully guarantees the efficacious conduct of potential proceedings, given that the assessment of a claim for damages for loss allegedly inflicted upon a specific undertaking as a result of an unlawful cartel, as already found by the Commission in a binding decision, essentially depends on factors specifically relating to the situation of that undertaking. In those circumstances, the courts in whose jurisdiction that undertaking has its registered office are manifestly best suited to adjudicate such a claim.‘
Finally, on the issue of choice of court in the agreements between the victims of the cartel, and those guilty of the cartel, the Court follows the AG’s lead. Such clauses are not generally applicable to liability in tort (the clause would have to refer verbatim to tortious liability). Neither do they in principle bind third parties, lest of course there be subrogration (Refcomp). (The referring national court has given very little detail on the clauses at issue and hence the ECJ notes that it could not reply to all questions referred).
In the end, it is the finding with respect to economic loss for which the judgment may be most remembered.
The use of anchor defendants in follow-up competition law cases. JÄÄSKINEN AG in CDC questions i.a. arbitration clauses in competition cases.
A particularly sticky point in competition cases, are follow-up suits for damages. I have already reported on (private international law aspects of) the issue of the piercing of the corporate veil, and on the use of a related undertaking as an anchor. [I report more extensively on competition law and conflicts in Jacques Steenbergen’s liber amicorum here. I hope to translate it into English some time soon].
In Case C-352/13 CDC (Cartel Damage Claims, in effect private anti-trust enforcement), at issue is among others the use of Article 6(1) of the Brussels I-Regulation when the claim against the anchor defendant has been settled before the trial is well and truly underway.
JÄÄSKINEN AG [whose Opinion at the time of writing was not available in English; indeed the absence of English translation of quite a few important Opinions is becoming a bit of a pattern. (That’s an observation. not an accusation)] suggests in his Opinion that only the time of service of the suit is relevant to assess the criteria of Article 6(1). This suggestion in my view finds support in the ECJ’s overall approach to Article 6: the subjective intentions of plaintiff, who often identify a suitable anchor defendant even if is not the intended target of their action, does not feature in the application criteria of Article 6. While this may lead to abuse of procedural power, establishing malicious intent is all but impossible. All but impossible: but not totally excluded. For that reason the AG does suggest that if one can prove that plaintiff and anchor defendant (in the case at issue: Evonik Degussa) had secretly agreed to settle, prior to the introduction of the suit, such collusion should be punished by non-applicability of Article 6(1), for in that case the conditions of Article 6 arguably are no longer met.
I am not sure the ECJ should follow the latter suggestion, particularly not in cases such as the one at issue, where defendants have been found to have acted illegally under EU competition law. (Misdemeanor or indeed criminal act therefore has already been established). In a way it would be an application of nemo auditur propriam turpitudinem allegans not to reward those who infringe EU competition law in the way the AG suggests. (This may be different in the event of as yet unsubstantiated claims of tort, in which case one may argue the defendant should not routinely have to defend the claims in a court other than the one identified by Article 2).
Next, the referring court enquired about the application of Article 5(3)’s special jurisdictional rule in the event of infringement of competition law, where that infringement concerns a complex horizontal agreement, spread over a long period of time, and with varying impact in various markets. One can probably not at all establish a locus delicti commissi for the tort as a whole: for such behaviour often takes shape in a variety of meetings, electronic correspondence et al. For locus damni, too, the picture would be one of a complex patchwork. Predictability and manageability of the ensuing suits would be impossible to establish in some coherent way, thus endangering some of the very foundations of the Brussels regime. In conclusion therefore the AG suggests not to apply Article 5(3) at all to current scenario, and to stick with application of Article 2, often then in conjunction with Article 6.
Although the last word on Article 6 needs to be said by the national court who alone is the judge of the risk of irreconcilable judgments, clearly in the AG’s mind there is a strong likelihood of such risk in the event of follow-up damages in the case of a cartel which has been found to be illegal by the European Commission and where all members to it have acted within one and the same intent (again, as established by the EC). Article 6(3) b Rome II [not applicable in the case but the AG suggests it would not hurt looking ahead] hints at such scenario where many defendants are sued in one and the same court.
Finally, the Court is asked to give input on the issue of choice of court, and arbitration clauses, in the agreements between the victims of the cartel, and those guilty of the cartel: do such clauses have any impact on the legal position of CDC, who has acquired the rights to seek damages for the cartel infringement? The AG suggests, in line with most national case-law (see more on this in my Steenbergen chapter, linked above), that such clauses cannot include follow-up damages for cartel infringement: for the latter is arguably not within the legitimate contractual expectations. This would be different for such clauses concluded after the tort has been committed: for Article 23 of the Regulation allows parties to agree on a different forum than those identified in the special jurisdictional rules. The AG finds additional support for this argument in the overall objectives of the very recent Directive 2014/104, the damages Directive. He takes the opportunity to argue that in the case of arbitration clauses, these may hinder the effet utile of Article 101 TFEU, just as choice of court clauses might, unless parties are shown beyond doubt to have consented to the clause, and provided the tribunal or court at issue, is under an obligation to apply EU competition law as matter of public policy. (Whether that is the case is subject to national law).
(It is quite likely that the Court itself will not review the last question for as the AG indicates, the referring national court has given very little detail on the clauses at issue).
This case could turn out to have quite a wide relevance for a large part of commercial practice. Or not: that depends on how far the ECJ itself will decide to entertain it.