The hard questions in the Lundbeck appeals: will Article 101 TFEU radically change?
The hearing in Lundbeck will take place on Thursday of next week. Readers of the blog know that this is a case that I have been following with keen interest (see for instance here). It raises some fundamental questions about the interpretation of Article 101 TFEU.
As I explained in the past, the Commission decision in Lundbeck departed in some respects from the relevant case law (and, indeed, from the authority’s own past administrative practice). The ECJ will now have to decide whether this interpretation is upheld, thereby leading to a transformation of the nature and scope of Article 101 TFEU.
These fundamental issues will become much more apparent, I believe, in the context of the appeal. Before the General Court, issues of law and fact, generally become so intertwined that it is difficult, in practice, to tell one apart from the other – or to fully appreciate the consequences of the underlying interpretation of a given provision. In this sense, the appeal is good news for the competition law community.
The areas in which Article 101 TFEU may (radically) change after Lundbeck are the following:
- The notion of competition within the meaning of Article 101 TFEU.
- The change in the understanding of the notion of potential competition, which would become a subjective concept (as opposed to the objective concept it has so far been).
- The relationship between competition law and intellectual property.
The notion of competition within the meaning of Article 101 TFEU
In Société Technique Minière, the Court declared that the notion of competition within the meaning of Article 101 TFEU must be understood as such competition that ‘would occur in the absence of the agreement in dispute’.
As subsequent judgments would consistently confirm, the analysis of restrictions does not occur in the abstract. It needs to consider whether, in the economic and legal context in which it is implemented, the agreement restricts competition that would otherwise have existed.
It is explicit in Société Technique Minière that this analysis applies both to restrictions by object and by effect (see p. 250 of the Reports, top paragraph). The Commission Guidelines on Article 101(3) TFEU are equally explicit on this point (see paragraph 18).
What are the consequences of this interpretation of the notion of competition?
The fundamental consequence of Société Technique Minière and subsequent case law is that, if an agreement does not restrict competition that would otherwise have existed, it falls outside the scope of Article 101(1) TFEU (that is, it does not restrict competition, whether by object or effect).
For instance, the Court explained in Société Technique Minière, an agreement would not restrict competition (whether by object or effect) if it is ‘really necessary’ to enter a new market. The Commission elaborates on this example in its Guidelines on vertical restraints. An agreement protecting a distributor from both active and passive sales is in principle restrictive by object. That is not the case, however, where the agreement is objectively necessary to achieve market entry. In such a case, it falls outside the scope of Article 101(1) TFEU altogether (see paragraph 61 of the Guidelines).
Lundbeck gives rise to a similar question. What if market entry is made impossible by virtue of the intellectual property regime? Can there be a restriction of competition if the only way for generic producers to enter the market is to infringe a patent? This is a central question in the case, and one that is somewhat obscured by its complex facts.
In light of Société Technique Minière and subsequent case law, the answer to these questions is clear. In such a scenario, the agreements at stake in Lundbeck would not restrict competition that would otherwise have existed. Accordingly, there would be no restriction, whether by object or effect.
The Commission, however, proposed a new interpretation of Article 101(1) TFEU in its Lundbeck decision. It suggested that an agreement can restrict competition by object even when it would entail the breach of a patent.
The ECJ, in its Lundbeck appeals, will decide whether to depart from a consistent line of case law.
The notion of potential competition as an objective (as opposed to subjective) concept
A related question in the Lundbeck appeals relates to the notion of potential competition. Is a firm a potential competitor if market entry requires breaching an intellectual property right? Until the Lundbeck case, there was little doubt that unlawful entry on the market did not count as potential competition.
E.On Ruhrgas is an eloquent example in this regard. The General Court concluded that an agreement does not infringe Article 101(1) TFEU where market entry is made impossible by virtue of the regulatory framework – in that case, the regulatory framework created a de facto monopoly. In such circumstances, it is the regulatory framework, not the agreement, which restricts competition.
The Commission also shared this view. In the 2004 version of its Guidelines on technology transfer agreement (the version published prior to Lundbeck, that is), the authority explained, in paragraph 29, that there is potential competition if market entry would have been possible ‘in the absence of the agreement and without infringing the intellectual property rights of the other party’ (emphasis added).
Lundbeck departs from these principles. How? By embracing a subjective approach to the analysis of potential competition. In Société Technique Minière, European Night Services and E.On Ruhrgas, to name a few, the notion of potential competition was an objective concept, in the sense that market entry was assessed in light of objective factors – that is, whether such entry possible and likely, given the economic and legal context.
In Lundbeck, by contrast, the economic and legal context (in other words, objective considerations) are not decisive. Subjective considerations (in other words, firms’ perceptions about the likelihood of market entry) can lead to the conclusion that two firms are potential competitors. Accordingly, potential competition may exist even when it would require infringing a presumptively valid patent.
As can be seen, Lundbeck advocates a major shift in the analysis of potential competition. Will it win the day? An analysis of the case law reveals that the ECJ clearly prefers objective approaches to the definition of legal concepts. In particular, the subjective intent of the parties is neither a sufficient nor a necessary condition to establish a restriction of competition. Similarly, the notion of aid within the meaning of Article 107(1) TFEU is an objective one.
The relationship between competition law and intellectual property
Finally, Lundbeck also marks a new relationship between competition law and intellectual property. As mentioned above, EU law does not question the existence of intellectual property rights; it only questions the way in which they are exercised.
Accordingly, EU law does not second-guess intellectual property systems, or whether the underlying ideas are worthy of protection. Thus, Articles 101 and 102 TFEU are applied without questioning the validity of the rights at stake in a case. All cases in which intellectual property is relevant, including Consten-Grundig, Coditel II or BAT (Toltecs-Dorcet) confirm this point.
By suggesting that a firm is a potential competitor even when market entry requires the infringement of a presumptively valid patent, Lundbeck alters this balance between competition law and intellectual property. According to this new balance, EU law would question the very existence of intellectual property rights (for instance, whether they are valid, or whether the underlying ideas are worthy of being protected). The consequence of the EU legal order would inevitably be significant.
[IMPORTANT: I did not want to end this post without mentioning the passing of Gil Carlos Rodriguez Iglesias. Gil Carlos led the Court of Justice at a key moment in time – Opinion 1/94 is just an example that springs to mind immediately. More importantly, he is fondly remembered by everybody who worked with him. Alfonso and I send our deepest condolences to his family and close friends. May he rest in peace.]
Chillin'Competition 
Dear Pablo.
I think this case is not about an objective and a subjective view of competition. In any event if somebody (a major player in the market) views you as a competitor one can assume that you probably are. There is nothing wrong with that and that approach is pervasive in EU competition law where the contemporaneous views of the parties are used to prove facts (an the existence of potential competition is a factual matter). This does not negate the objective nature of the competition to which the case law refers as it acts at evidentiary level only. This is standard case law and there is nothing revolutionary about it. There are precedents of analysis of potential competition under uncertain regulatory conditions (sandoz, hitachi, astrazeneca). Moreover here this is not about law in the public interest but about private rights whose exercise is at the disposal of right holders.
But I see your views on this have not changed and you continue to be fixed with the idea that there is a sort of presumption of illegality of the competitor’s conduct. And that presumption of illegality would reflect an objective view of potential competition. It is unclear why.
I would have thought that after (numerous) comments by readers on previous posts critical of your views on this subject (I cannot put the links now but I refer the readers to them, just search the posts referring to this judgment) you would have reviewed them, or at least present new arguments in their support, but I see you repeat the same old mantra of yours on this. I am afraid that your views do not become more persuasive by repeating them.
Yes, the appeals will be important, but let’s not exaggerate about how novel Lundbeck was.
Joan
18 January 2019 at 8:44 pm
I agree. And Pablo, I don’t think you have read the Lundbeck Decision. And you emphasize the importance of the legal context, with which I agree, but you don’t seem to know the legal context of the pharmaceutical industry in the EU. I have read the Decision and I suggest you do the same before making these comments, and then we can have a discussion that is more enlightening than just provoking the crowds. The Decision relies on an objective proposition and on objective elements: real and concrete possibilities, which is an analysis of the structure of the market. It identifies various possiblities of entry, investments, etc. as objective elements showing real and concrete possibilities. And Lundbeck only had process patents. Some of them had not even been granted by the time some of the agreements were concluded. I think you are ignoring reality. This is competition for the market. And very intense. The generic is trying to enter and the originator is trying to keep it out. The only weapon an originator has at this time is its patents. These are barriers to entry, but the existence of a barrier to entry does not mean that there are not folks out there who can overcome the barrier. It depends on the barrier, isn’t it? And it depends on what the potential entrant has in order to overcome the barrier. So you have to analyse that. Pharmaceutical markets in the time after compound patent expires are open to competition, in fact, generics race to be the first to enter the market, despite the risks because as rational operators they pursue profits, until more generics enter and prices go down. These are incentives to compete. And the competitive pressure exercised by them is greater – by several orders of magnitude as the CAT said- than any other constraints from other molecules. Successful generic entry completely destroys your market – overnight. Hence the incentive to collude by paying them to stay out – both parties win. There is nothing novel or revolutionary in the proposition that a patent settlement agreement in which the originator buys the generic so that it does not enter the market through payments is a restriction by object. AstraZeneca already says that delaying generic entry is an abuse, even if the dominant undertaking has patents (AZ even had a compound patent at that time) and even if you could have excluded them by enforcing your patents in court. Isn’t buying competitors by nature anticompetitive? Why would it be different for a potential competitor? or for a patent holder? I agree that this depends on showing that the generic has been bought out by payments, but assuming you have proven this. This is completely consistent with the case law on patents and competition law since Parke Davis in 1967. Patent holders do not have immunity from competition law. This was already established in the EU before the US Supreme Court said it in Actavis. Yes, we were ahead. Patent holders do not get a right under patent law to interfere with competition on the market by changing a competitor’s incentives to compete. The infringement remains Lundbeck’s allegation only as the GC said. That is a subjective view. It would be absurd to start the competition law assessment from this subjective premise that the generics were infringing as this relies on facts that have never been proven. It is just an assumption, as it has been made clear in the US while rejecting the scope of the patent test: The patent may or may not be valid and may or may not be infringed. The antitrust law assessment cannot depend on that. And the proposition in the Lunbeck decision is consistent with the US Supreme Court in Actavis. “both competition law and patent law are relevant to determine the scope of the patent monopoly”. What you suggest is not a reconcilliation but complete domination of patent law over competition law. And it is not correct that generics would have had to infringe a valid patent to enter the market. You should check the facts 🙂
Desert star
18 January 2019 at 11:58 pm
I ignore if Pablo read the decision or not. But in a more principled way, reflecting on this I think that the origin of Pablo’s views on lack of potential competition may come from a wrong understanding on the case law on exclusion of competition “by law.” Let’s leave aside for a moment that here it is not “the law” but a private right whose breach must be proven. I also leave aside the discussion of the different precedents which I think Pablo has misunderstood (I do not know if Pablo read the decision, but very few people in this world have read Sandoz, as it was only published in a summary form and it is of difficult access). For me the law is another barrier to entry in this case. We already have case law, not questioned by Pablo, which has considered that raising barriers to entry of a “legal type” (including rights) can be in certain cases an abuse, as abuse is not limited to conduct on the market (Compagnie maritime belge, astrazeneca). Such legal/regulatory environment is important in the pharma sector what explains that this type of abuse has mainly appeared here. In the same way, and the other side of the coin, evidence of potential competition is not limited to conduct on the market. I explain. Of course, the ultimate objective is to enter the market, but the generic producer may, for example, challenge in court the validity of the regulation/law (yes, even that) or right in question, in order to be able to overcome the barrier. Hence the statement by GC that such litigation is another way of “competing” so to say (or rather of overcoming barriers to competition, in view of later entry). It may also, of course (and very relevant here), enter the market, and it is up to the rightholder to prove the breach, such conduct not being illegal until determined by a court of law (not illegal per se as post appears to be suggesting). While litigation is pending, you are normally a potential competitor (otherwise you would not be a potential competitor the day before the judgment, and a potential competitor the day of the judgment). You have to take a somewhat longer time span here due to the long process of entry (astrazeneca). This also applies in a situation where a competitor would be challenging in court a law/regulation, in order to enter the market: if it is “convinced” to withdraw the case in exchange of money, potential competition which would ensue after a successful challenge of the potentially invalid law would disappear. “Legal barriers” are not impenetrable, and take different forms which should be examined carefully, if the legal barrier is challenged. Some require proof by ritghtholder (not the opposite). Of course, there may be cases where the potential for entry is so low that competition is hypothetical. Then the conduct of the dominant company would be irrational (and it would be up to the company to explain, in a shift of the evidential burden of proof, to justify its irrational conduct, as judges do take as -evidentiary- basis that informed business operators operate rationally, a principle that I hope Pablo does not want to question). And yes, I also suspect that the “balance” that Pablo proposes is quite “unbalanced” in reality.
Joan
19 January 2019 at 8:42 am
Hi Pablo, I couldn’t stop myself from commenting because the relevance of (subjective) intentions is, as you may remember, the subject of my PhD research. I don’t want to go into the merits of the argument that you make but I think you misrepresent the case law of the ECJ. While it is true that the ECJ has said that subjective (anticompetitive) intentions are not necessary to establish an anticompetitive object, that is only part of the story. Indeed, Since General Motors (2006), the ECJ has consistently stated that “even if the intention of the parties does not constitute a necessary factor in determining the restrictive character of an agreement, there is nothing to prohibit the Commission or the Community courts from taking that intention into account.” (repeated in the same words in T-Mobile (2009), GSK (2009), Allianz Hunbaria (2013), Cartes Bancaires (2014) and Dole Food (2015)). As to whether intentions are sufficient, AG Wahl in Cartes Bancaires indeed denied that this is the case (he was, like you, very negative about the relevance of subjective intentions in that opinion). In its judgment the ECJ repeated this but also pointed out that the General Court had stated in its judgment that intentions had been analysed as additional confirmation only – so arguably the argument was without object and the words of the ECJ obiter. In any event, the ECJ did not deny that subjective intentions can be relevant (on the contrary: as I said, it repeated the words from General Motors which state that the Commission and the Courts can take intentions into account).
Jan Blockx
19 January 2019 at 9:17 am
Dear all,
Thanks so much for the time and effort put into sharing these comments! These discussions are definitely the best part of blogging.
Joan is right to say that many of you have shared your views on Lundbeck. Contrary to what he appears to suggest, I have learnt a great deal from every commentator. These comments have helped me streamline my arguments, and present them more effectively. I am grateful for that.
Joan says, and he is right, that I have not changed my mind. This is not necessarily a bad thing. It is simply that, so far, I have not been persuaded by the arguments contradicting mine. And I am always open to being persuaded.
On the substance, I would say that Joan’s views reflect the new balance between intellectual property and competition law of which Lundbeck is an example. As I explain in the post, this new balance is at odds with the case law and, indeed, the traditional administrative practice of the Commission.
Not so long ago, in its 2004 Guidelines on technology transfer agreements, the Commission explained that a firm is a potential competitor if it is willing and able to enter the market without infringing the intellectual property rights of the other party (see para 29 of the Guidelines). This paragraph reflects the essence of the case law and the Commission’s past administrative practice. It is not a surprise that, at around the same time, the Commission was not even certain that the Lundbeck agreements amounted to an infringement, let alone a by object infringement (para 652 of Lundbeck).
Joan’s arguments suggest that he believes the Commission got it wrong in its 2004 Guidelines — and, by extension, that the thinking of the Commission in relation to the Lundbeck case evolved in the right direction. I respect these views. It is simply that, no matter how many times he refers to Sandoz (which I happen to discuss every year with my undergrads), these views represent a major departure from the case law.
I would not want our readers to take my word for it. I propose to all other readers a simple exercise that would prove my point. Apply Joan’s approach to Magill and IMS Health. The legal test would be very different and far less strict, wouldn’t it? And remember that the validity and/or whether the ideas were worthy of protection were at the heart of the disputes in both cases
Desert star (my favourite pseudonym so far on the blog) refers to the facts of the case. I believe they think we disagree more than we actually do. The Lundbeck agreements may very well be restrictive by object and worthy of a fine. I do not dispute this point.
I just wonder why they emphasise so much the facts of the case. My blog post is about the points of law underlying the case. More importantly, and as Desert star certainly knows, appeals before the ECJ are limited to points of law.
Jan: I greatly admire your work, but your discussion of the case law does not disprove the point I make, which is that the subjective intent of the parties is neither a sufficient nor a necessary condition to establish a restriction by object.
Pablo Ibanez Colomo
19 January 2019 at 4:11 pm
Dear Pablo.
I am not sure you have got me right. No, I do not think there has been any evolution since the 2004 guidelines. And believe me, I know what I am talking about due to my own responsibilities for both the guidelines and the evolution on IP and pharma at the time. The guidelines do not refer to the specific situation of case such as Lundbeck, where there is pending (or soon to be) litigation. They state a very general principle that potential competition must be “lawful”. Big deal. Everybody agrees with that. Such an abstract principle does not help much in this case, as explained in my second comments. You prefer the Guidelines and that’s fine, but I think you prefer your personal interpretation of the guidelines, rather than the guidelines as such.
On IMS or Magill, they are just too distant to be of relevance here, honestly. But no, I do not think the test would be less strict, and to the best of my knowledge Lundbeck is not relied upon (by the Commission, for example) to soften IMS or Magill (as interpreted later in Microsoft). They just reflect different situations where IP rights and competition law may interact. They are irrelevant here since what is at stake is whether there is potential competition in situations where there is uncertainty about one specific type of barrier to entry namely IP rights. And Lundbeck does not “second-guess” IP laws, on the contrary. If you would such trust in IP systems, one could argue that you should be happy to have the validity of the IP right (or breach thereof) be decided by courts. A patent settlement (well, as you will know some of these cases go beyond the traditional definition of a settlement, but let’s not get too specific here) is not a “right created by law”. It is a business reality. IP law does not include such right, and “affecting” it does not applying affecting IP law. It affects how certain companies may react in the context of disputes about such rights, but that’s another matter. Again, IMS or Magill were really about the IP right, which makes comparison irrelevant.
Facts are important. You stress the important of a case, and the case cannot be understood without the facts. I note that you do not necessarily disagree that Lundbeck may deserve a fine, but then it is unclear what the point of your disagreement with the GC is.
On your last paragraph, I thought your post was not about so much about the role of intention to prove “by object” restriction, but about the role and relevance of a possible infringement of the IP (that you appear to assume has been committed, or that the Commission should prove has not been committed) and whether the views of the parties at the time, documented by contemporaneous documents, is relevant to examine the existence of potential competition (what you call “subjective notion” of potential competition). E.ON or STM did not raise such issue. In E.ON the “legal barrier” was not contested, very simply. Sandoz, more to the point here, simply ignored/rejected the argument that there was no potential competition as entry of parallel imports was not possible due to legal obstacles. One can only guess what the reasons behind the ruling were (it is very short), either the judges did not believe such obstacles existed (case law on parallel imports was uncertain at the time, but there was some), or very simply they saw the conclusion of an agreement explicitly preventing such competition as evidence that such potential competition existed (it was a by object restriction), despite what Sandoz’s lawyers could later say in the context of litigation.
Joan
19 January 2019 at 5:28 pm
Dear Joan,
Once again, thanks for this exchange and for taking the time to share your views. I am sure Alfonso and our readers very much appreciate it too.
I believe this last post captures well the essence of your point. It also shows why it departs from the traditional balance between competition law and IPRs.
In essence, it looks like you do not accept the principle that patents are presumed valid (it is at least the only way I make sense of your comments). This is a principle that has sometimes when spelled out explicitly in the case law (for instance, in AstraZeneca). In any event, all the relevant case law implicitly accepts it – which goes hand in hand with the principle whereby EU law does not question the existence of intellectual property.
You appear to claim that, under regulatory uncertainty, intellectual property can only be said to prevent entry where (i) the right is exercised (before a court, since you also question that settlements are a valid expression of the exercise of intellectual property) and (ii) upheld. Hence your comment about leaving it to the judge to sort out the uncertainty.
This view of the relationship between competition law and intellectual property may win the day. As I suggest in the post, it may be embraced by the Court. But I cannot be persuaded that it is business as usual, or that Sandoz somehow enshrines it.
Importantly, the principle that patents are presumed valid is not contingent on the facts of the case. Either it is accepted, or is it not, and the legal characterisation of the facts follows.
On IMS Health: contrary to what you suggest, there was regulatory uncertainty in that case. There was so much regulatory uncertainty that the case was ultimately decided on intellectual property grounds (and against IMS Health).
Why did I bring up this case? Because, in spite of this uncertainty, the Commission decision and the Court ruling simply assumed the existence of the IPR. They did not make it conditional on the court of last resort addressing the regulatory uncertainty. And this even though, as in Magill, the right was deemed ‘weak’. But the ‘weakness’ of the right was left outside the legal test (even though it was tempting to factor this aspect in the analysis).
The opposite of what you are suggesting. Under your approach, the case would inevitably have been decided differently.
On Lundbeck: I never took issue with the outcome as such. And again, the decision and the GC ruling suggest that the agreements may very well be restrictive by object. As far as I am concerned, the controversial bits were the points of law I discuss in my post. Remove these controversial points of law, and I have the impression it is still possible to rule that the agreements were a by object infringement (and, if there was any doubt, impose a fine). Desert star’s comment effectively highlights why, in my view.
Once again, thanks a million! If only we could continue this discussion in the context of a seminar!
Pablo Ibanez Colomo
21 January 2019 at 7:32 am
Dear Pablo, we can stop here. I have the impression that your views on this case are impenetrable to “objective” analysis or discussion (even if you are in favour of “objective” notions of restriction competition and potential competition, ;-)). But it is just an impression. Besides, I know it is easier to rebut an argument by first misrepresenting it, but I hope this does not end up being the new trend in this blog.
Just four things. First, your acid-test for this case (remember, paying your competitor to stay out of the market) to be a “radical change” of the interpretation of Article 101 TFEU is some philosophical connection with Magill and IMS, cases on 102 TFEU [btw, no, I am not suggesting that IMS or Magill would have been decided differently, the reader can read what I said and did not say, and I said that Lundbeck does not affect those cases, as they are very different and the issue of potential competition was not at all decided in those cases; actually under certain strict conditions there could be an abuse which means that the ECJ took into account competition which would come from the issuing of licences, ie even if it was not possible without the licence; in IMS and Magill the courts were left to decide the underlying IP litigation thereby solving the “uncertainty” problem]. It is not because I am free to licence or not (the core of Magill/IMS) that I can put “anything” in an agreement which may indirectly relate to the IP (reserve payment to stay out of a market, for example), and claim that there cannot be any infringement of 101 since I could even refuse to licence my (presumed valid) patent. I have the impression that “invocation” of IMS or Magill here may be just reflecting a recent trend to “export” the case law where the ECJ applies the strictest standards (IMS/Magill/Bronner) to other situations in order to severely limit the scope of application of competition law. We have seen it tested, unsuccessfully, in other areas (freezers, margin squeeze, etc). Second, in Bayer/Süllhöfer (65/86, a case closer to this than IMS or Magill, and that I am sure you also explain to your students ;-)), the ECJ was not particularly “nice” to non-challenge clauses in settlements (and the settlement in that case –cross-licences- was probably, on its face, far less “damaging” than in more recent cases of the Lundbeck type), departing from the more nuanced view of the Commission in its submissions at the time. The judgment showed limited deference to the fact that the agreement was the result of a settlement of pending litigation (see paras 15 and 16). Third, even if patents are presumed valid (I am not sure I have suggested the opposite; I am not sure either that the GC would have said the opposite), the issues are (a) whether the generic producer is violating the patent (are you suggesting also that there is a “presumption of breach”?, it is still unclear to me), and (b) whether that presumption of validity would justify in IP law a payment to a competitor not to enter the market (I do not remember that being a principle in IP law, normally the users of the IP pays to the rightholder, even in IMS and Magill) and why considering that such payment may be unlawful under certain conditions in some way “weakens” IP protection (by contrast, IMS and Magill, by “obliging” the right-holder to licence touched the “core” of IP protection). Fourth, on the alleged “evolution”, I forgot to mention Lundbeck starts in 2003 (inspections, see factual part of the judgments), ie before the 2004 Guidelines.
Again, either you discuss this case or not. I am lost when you do not know if fining Lundbeck is OK or not. If you have been discussing another “abstract” case which does not fit the facts of Lundbeck, then we are wasting our time, because I am discussing this case and your post warned about the somewhat apocalyptic (yes, you did not use those words, I know) consequences of the judgment in this case, not another one in your imagination.
But as I said, I stop here. See you very soon!
Joan
21 January 2019 at 9:08 am