Chillin'Competition

The first instalment of this series discussed the issues of principle addressed by the Court in Generics. The judgment is equally important in what is says about pay-for-delay. It is possible to infer a fully-fledged framework for the assessment of these agreements. Crucially, this framework is wholly in line with the relevant precedents on  Article 101(1) TFEU (a question we have regularly discussed on the blog and elsewhere, for instance here).

The distinction between genuine and non-genuine intellectual property settlements

It was already possible to distinguish from the case law – in particular from BAT (Toltecs-Dorcet) – that the question of whether a settlement relates to a genuine intellectual property dispute is relevant and important in practice.

Where a supposed settlement does not relate to a genuine dispute (in the sense that the parties – or at least one of them – seek to achieve something they would not be able to achieve via the exercise of their intellectual property rights), the assessment can be relatively straightforward. Such agreements conceal a ‘naked’ market-sharing arrangement that is known to be restrictive by object.

Where the agreement relates to a genuine dispute (in the sense that the settlement replaces intellectual property litigation before the courts), on the other hand, the assessment may need to be more detailed. In this sense, Generics provide a valuable template, as I do not believe there were examples in the Court’s case law.

This difference between genuine and non-genuine disputes will be relevant for the upcoming cases on pay-for-delay. As I have written elsewhere, any moderately attentive reader of the Lundbeck decision cannot avoid the impression that the settlements in the case were a sham. You should not take my word for it: read the very representations made by the firm in the context of the proceedings and reproduced in the decision.

The status of pay-for-delay: an illustration of the Court’s default approach to ‘by object’ infringements

The Court’s approach to the assessment of restrictions by object has long been clear. An agreement amounts to a ‘by object’ infringement where it has no plausible purpose other than the restriction of competition. We also know since T-Mobile that an agreement is not caught by Article 101(1) TFEU – whether by object or effect – if it is incapable of having a restrictive impact (this is something that AG Kokott emphasised in her Opinion).

What is the added value of Generics, against this background? The judgment makes explicit some points that were implicit in the case law (it is the first time some expressions are used).

The relevant test is formulated in paras 89-90. An agreement amounts to a ‘by object’ infringement where the only ‘plausible’ explanation for it is the restriction of competition. It follows logically that such categorisation is excluded where the parties can advance an alternative explanation casting a ‘reasonable doubt’ (para 107) about the status of the practice as caught, by its very nature, by Article 101(1) TFEU.

The Court has consistently held since Cartes Bancaires that the ‘by object’ category must be interpreted restrictively. Thanks to Generics, we know what this statement means in concrete terms. It means that (leaving market integration aside) the category encompasses agreements the only plausible explanation for which is the restriction of competition.

We also know that the bar to rule out that an agreement amounts to a ‘by object’ infringement is low (‘plausibility’ and ‘reasonable doubts’ do not come across as particularly cumbersome).

Another crucial contribution of Generics relates to the legal status of a ‘reverse’ payment. In line with the case law, the Court clarifies that a payment from an originator to a generic producer does not mean, in and of itself, that the agreement is restrictive by object (paras 84-85).

As already held, the objective purpose of such payments must be considered in the relevant economic and legal context. The Court even identifies the instances in which a ‘reverse’ payment would be objectively justified and thus would fall outside the scope of Article 101(1) TFEU (para 85).

The assessment of potential competition and the role of circumstantial evidence

The assessment of potential competition in pay-for-delay cases is a fascinating puzzle. On the one hand, patents are presumed valid; on the other hand, there is not such thing as a presumption of infringement of a patent by a new entrant (such as a generic producers).

How is it possible to tell whether a generic producer has real, concrete possibilities to enter the market where there is genuine uncertainty about its ability to do so? Is it possible to rule on potential competition prior to a final court ruling on validity and/or infringement?

The Court answers in the affirmative to the latter question. Crucially, it does so without questioning the presumption of validity of intellectual property rights in EU law, and without revisiting the principle whereby EU law does not question the existence of such rights.

How? One crucial step is that ‘a patent which protects the manufacturing process of an active ingredient that is in the public domain’ is not deemed to constitute an ‘insurmountable barrier’. In this sense, the Court is careful to clarify that it is examining the status of potential competition in a very specific context (see para 51 for a description of that context).

A second crucial step in the analysis is the role given to circumstantial evidence. In a context of uncertainty, indicators such as the very fact that they conclude an agreement (para 55) or the very fact that there is a ‘reverse’ payment (para 56) can lead to the conclusion that the parties are, indeed, potential competitors.

According to the Court, the fact that the agreement relates to a genuine dispute does not rule out the finding that they are potential competitors.

What is the takeaway from the above? The key conclusion, in my view, is that potential competition exists whenever an authority can show that market entry is plausible. The threshold coincides with the threshold that applies when assessing whether the agreement restricts competition by object.

The eagle-eyed among you may remember that I wrote in my article on pay-for-delay that the threshold for potential competition is one of likelihood (that is, that market entry is ‘more likely than not’ to occur). The judgment clarifies that the bar is lower – plausibility of entry is enough.

If you ask me: what I wrote at the time made sense in my head (a potential competitor worthy of the name must exercise a constraint on existing players). Thinking about it, however, I believe it is more reasonable to align the threshold with the assessment of ‘by object’ conduct, as the Court does.

As a side note: if you were wondering whether there is a discrepancy between the judgment and my prior writings, it all boils down to this threshold.

On proof

If you read the above, you may get the impression that, after Generics, it is all about proof. Has the authority (or claimant) established, to the requisite legal standard, that the parties are competitors? Has the defendant managed to cast reasonable doubts about the characterisation of the agreement as restrictive by object?

I agree, which is why there will be a third part addressing issues of proof. Thanks in advance for your comments!

The venerable FIDE Congress will take place this year in The Hague on 20-23 May 2020 (speaking of which: you can still get the early bird rate until the end of the month here).

One of the three topics is ‘EU Competition Law and the Digital Economy‘. I am proud and grateful to have been invited to work on the report for Spain, which I have prepared with Marino García, a chief case handler at the Spanish Competition Authority (CNMC). A draft can be found here.

The UK Report was prepared by our friends Andriani Kalintiri (King’s College London) and Ryan Stones (City, University of London) and can be found here. And the general rapporteurs will be Nicolas Petit (European University Institute) and Pieter van Cleynenbreugel (University of Liege).

You may want to take a look at the Report that I prepared with Marino. It was fascinating to look systematically at all the digital economy cases examined by the CNMC. The impression one gets is that competition authorities have so far effectively addressed the challenges raised in these markets, and that usual remedies can be powerful.

I am tempted to draw your attention to two key cases. One is a merger case, Just Eat/La Nevera Roja, which deals with exclusivity and tipping in two-sided markets; and the second is an abuse case, Estudios de Mercado Industria Farmacéutica, which deals with access to data (and an old friend that goes by the name of IMS Health).

Marino and I would very much welcome your comments. We look forward to seeing many of you in The Hague!

It is safe to predict that last week’s judgment in Paroxetine will be remembered as a major landmark in EU competition law. It is unusual due to the depth and the breadth of the contributions it makes to our understanding of some fundamental principles. We have had many judgments clarifying a discrete point of law. Unlike these, Paroxetine engages with several notions at the same time (the notions of competition, restriction by a object, effects and abuse), which is uncommon in the case law, and this, in depth.

A bit over a year ago I explained why the pay-for-delay saga would test the fundamental principles (see here) underpinning Articles 101 and 102 TFEU. I do not believe it was an exaggeration: it is sufficient to take a look at the questions raised by the national court in Paroxetine. The CAT asked whether the likelihood of a patent being declared invalid should be considered when evaluating the effects of an agreement. From a system that does not question the existence of intellectual property rights, I explained, we could be moving to one treating them as probabilistic titles.

Eventually, the case law has proved to be resilient. The judgment is in line with AG Kokott’s Opinion (see here for my take on it). The notions of competition and potential competition have not changed (they are still based on objective considerations), and the relationship between competition law and intellectual property is the same it has been since Consten-Grundig. It is not because nothing has changed that the judgment is not interesting. On the contrary.

I would not do justice to such a major landmark in a single post. This one is on principles. One will follow on pay-for-delay and the final one on issues of proof, where, now that the law is clear, the attention will move.

On restrictions by object

The law has been consistent from the very early days, but we have heard claims until not so long ago (including in Budapest Bank), that the pro-competitive aspects of an agreement are not relevant under Article 101(1) TFEU. Now we can definitely say that such claims did not reflect the reality of the case law. In paragraph 103 of the judgment, the Court clarifies that the pro-competitive effects of an agreement ‘must, as elements of the context of that agreement, be duly taken into account for the purpose of its characterisation as a “restriction by object”’.

The Court is prudent to state that such pro-competitive effects must be ‘demonstrated, relevant and specifically related to the agreement concerned’. In other words, claims about the pro-competitive effects must not be merely pretextual (as they were in cases like BIDS, which concerned a plain-vanilla crisis cartel).

What is the threshold to rule out that the agreement is a ‘by object’ infringement? I had suggested in the past that the threshold is one of plausibility. Paroxetine seems to confirm this interpretation of the case law. There is an express reference to plausibility in the assessment of the reverse payment (para 89). In addition, the Court explains that the pro-competitive effects of the agreement must cast a ‘reasonable doubt’ about the object of the agreement. This position is also in line with AG Bobek’s Opinion in Budapest Bank.

On competition, potential competition and the counterfactual

Prior to the release of Paroxetine, I explained that the notion of competition in the Treaty is best understood as meaning ‘actual or potential lawful competition which would have existed in the absence of the practices under consideration’.

The judgment confirms this definition. In paragraph 40 of the judgment, the Court makes it clear that, when considering whether a firm is a potential competitor, it is necessary to consider the ‘regulatory constraints that are characteristic of the medicine sector’, thereby confirming that only lawful entry counts as competition within the meaning of Articles 101 and 102 TFEU.

Second, it is necessary to consider the counterfactual in light of the relevant legal and economic context, and more precisely the market conditions in the absence of the practice under consideration. This position puts an end to the idea that the counterfactual is not relevant at the ‘by object’ stage. As AG Kokott explained, if the market conditions would not have been affected by the agreement, the practice does not restrict competition, whether by object or by effect.

Reasonably (para 38), the Court holds that potential entry must not be a ‘purely hypothetical possibility’. By the same token, it need not be shown that entry is certain or that the potential entrant ‘will be capable, thereafter, of retaining its place’ on the market.

Third, subjective considerations may be a factor, but the notion of competition is an objective notion. In this regard, the judgment confirms the case law. I understand it as meaning that subjective considerations are relevant insofar as they shed light on the objective ability of a firm to enter the market.

On the notion of effects

There has been no shortage of posts on this blog on the notion of effect under Articles 101 and 102 TFEU. One of the points that has always been emphasised is that not every competitive disadvantage and not every limitation of a firm’s freedom of action amounts to an anticompetitive effect within the meaning of the abovementioned provisions (Post Danmark I and II, as well as MEO, insist on this point).

Crucially, Paroxetine leaves no doubt. According to the Court (para 172), in order to establish an exclusionary effect under Article 102 TFEU it is necessary to show more than just the impact of the practice on the parties involved in the transaction. The analysis must ‘[go] beyond’ such effects and consider, the market as a whole (in past cases, the Court has already had the chance to identify the relevant factors to consider in this assessment).

On the notion of abuse

The bit on whether the practice amounts to an abuse deserves to be read closely. Interestingly, it refers to the object and effect to the conduct, thereby indicating that there can indeed be abuses by object.

Second, and more importantly, the Court points out that, whether the practice has as its object or effect the restriction of competition, the application of Article 102 TFEU ‘presupposes that that conduct was capable of restricting competition and, in particular, producing the alleged exclusionary effects’ (para 154).

Whether the impact on competition needs to be shown (as in Post Danmark II) or is implicit (as in Intel), a practice is not abusive within the meaning of Article 102 TFEU if it is incapable of having restrictive effects. This clarification also means that the panoply of arguments discussed above – relating to the counterfactual, or to the fact that the impact of the practice does not go beyond the relationship between the parties – would be relevant too.

Hopefully I will address the pay-for-delay aspects next week. As ever, your comments would be most welcome.

The Ithaca Competition Summit is back! Our friend Peter Alexiadis will once again host an impressive group of top competition lawyers and economists in the island of Ithaca.

This year’s programme is about to be completed, but we can confirm that speakers will include the President of the General Court of the EU, Marc van der Woude; the President of the Portuguese Competition Authority, Margarida Matos Rosa; and Gabriella Muscolo from the Italian Competition Authority. And all but one of the former Chief Competition Economists at DG Comp (plus the current one!) have already confirmed their attendance!

I was fortunate enough to attend the first edition. The place is breathtakingly beautiful. Together with a high-level event, it is a wonderful way to end the summer.

We will be keeping you informed about the summit, and about how to register. Hopefully many of you will be able to make it!