Op-Ed: “Sumal (C-882/19): Solving one problem, while creating many others?” by Miguel Sousa Ferro
By now the news has spread like tamed fire throughout the EU antitrust legal community: by its Grand Chamber judgment in Sumal (C-882/19), the Court of Justice has clarified that, as far as EU competition law is concerned, subsidiaries are (sometimes) liable for the sins of their parents. The Court continues to show itself to be the injured parties’ greatest defender.
The Sumal judgment reaffirms important clarifications from the Skanska ruling (C-724/17), namely that:
- the determination of who is liable for an infringement of Article 101 TFEU ‘is directly governed by EU Law’ (paragraph 34), and therefore it is time to stop applying national law when determining if someone is liable for an infringement of Articles 101/102 TFEU; and that
- the concept of “undertaking” is the same for public and private enforcement (paragraph 38).
Furthermore, the Sumal ruling adds two important new clarifications:
- A subsidiary may be held (jointly and severally) liable, in private enforcement actions, for the behaviour of another legal entity within the same economic unit (paragraphs 48, 50-51 and 64), but only if it is shown, not just that the subsidiary is a part of the same undertaking in the classic sense (economic, organizational and legal links), but also that there was ‘a specific link between the economic activity of that subsidiary and the subject matter of the infringement for which the parent company was held to be responsible’ (paragraph 51). The fact that it is not an addressee of the relevant Commission decision establishing the infringement cannot be used as grounds to exclude liability (paragraphs 61-63).
- A subsidiary liable under the previous clause is also bound by the binding effect of a Commission decision finding an infringement by that undertaking, even if it was not the addressee of that decision, but it must be given a chance to prove that it is not within the same undertaking (paragraphs 48, 53-59).
Requiring a link with the economic activity involved in the infringement is far shy from requiring – as AG Pitruzzella had suggested – participation of the subsidiary in the infringement, or even that the subsidiary be active on the same market. The Court of Justice’s objective was to exclude liability for subsidiaries of conglomerates which carry out completely different economic activities. Finding where to draw the line between these two types of scenarios is likely to become an issue. It seems clear that a subsidiary which sells fruit should not be held liable for a cosmetics cartel of its parent conglomerate. But what if it provides management services for supplies of the cosmetics manufacturing activity? What if it sells some cosmetics, but not the specific cosmetic in question? These problems are clearly on the horizon. In a decentralized judicial system such as the European one, one might have hoped the Court of Justice to have been a little clearer and nip some of these issues in the bud.
This judgment sets up descending and horizontal liability within the economic unit. The reasoning is applicable even if the infringement was carried out by another (non-controlling) subsidiary. And its conclusion is valid for public enforcement as well (under the homogeneity principle of Skanska), meaning that the European Commission may also use it to fine subsidiaries.
The finding on the subjective scope of the binding effects of the decisions by the Commission is, naturally, also extendable to the binding effects of National Competition Authorities’ (NCA’s) decisions under the Damages Directive. It is also an important contribution to the expected discussions in various Member States about the compatibility with fundamental rights of such binding effects. A ‘minor’ fallout from this is that every subsidiary within the economic unit must be recognized the right to appeal a Commission decision, even if it is not an addressee, and an appeal by any legal person within the economic unit must be deemed to benefit the entire economic unit.
But let us focus on the central issue. The starting point of the new clarifications is hardly new: liability is assigned to the ‘undertaking’, in the sense of economic unit, a special concept of EU competition law (paragraphs 39-43). The logical conclusion is set out in paragraph 44: ‘the concept of an ”undertaking” and, through it, that of “economic unit” automatically entail the application of joint and several liability amongst the entities of which the economic unit is made up at the time that the infringement was committed’.
This should have been the end of it. Any legal person within the economic unit could be held liable. But then the judgment (like the AG’s Opinion) took a turn. It stated that the subsidiary would only be part of the ‘economic unit’ if it carried out the same economic activity as that in question in the infringement (paragraph 52). And it justified this with a sweeping statement about the functional nature of the concept.
And this is where, I’m afraid to say, the Court of Justice may have snatched defeat from the jaws of victory. Or, at the very least, created a lot of superfluous headaches for itself (and for all courts and lawyers applying EU competition law). Because what the Court now tells us is that a single corporate group can contain several economic units or ‘undertakings’. Speaking for myself, that is complete news, and far more of a bombshell than the conclusion about subsidiaries’ liability.
The outcome in the specific case is fair and desirable. It will increase access to justice and drastically reduce costs and procedural difficulties for claimants. As a matter of policy, it may very well make sense that only subsidiaries active in the same economic activity should be held civilly liable for infringements associated to that activity. But, in its reach for good policy, the Court of Justice arguably went down a dangerous path of law. A particularly dangerous path for a legal order where one is not merely bound by the judgment’s ratio decidendi.
As was hinted at by Marcos Araujo Boyd, the Court may have sparked chaos in the case-law on the identification of an ‘undertaking’, when it introduced a functional approach allowing for several economic units (undertakings) within the same conglomerate.
We are likely to spend the next years dealing with collateral damage from this judgment, arguing whether the clarifications provided in Sumal also apply to the boundaries of the undertaking for all other purposes within competition law, such as:
- determining whether an intra-conglomerate agreement is an agreement between undertakings subject to Article 101 TFEU;
- determining the maximum fine (10% limit);
- identifying aggravating circumstances of repeat infringement;
- determining merging or acquiring undertakings under the Merger Regulation; or
- determining who is the beneficiary undertaking of State aid; etc.
Large economic groups’ lawyers are likely to seize on this judgment to try to get much more favourable outcomes for their clients on these and many other fronts.
The Court of Justice is likely to reject attempts to extend this case-law to other issues of competition law. But, at the very least, Sumal will remain as a small uncomfortable pea under the mattress of EU competition law.
Miguel Sousa Ferro is a Professor at the University of Lisbon Law School and Managing Partner of a law firm in Lisbon. His recent publications include ‘The EU Antitrust Damages Directive: Transposition in the Member States’ (Oxford University Press), and ‘Market Definition in EU Competition Law’ (Edward Elgar).