Chillin'Competition

Searching for an answer. A few days ago I asked Pablo in public (following some private teasing) whether there is any Article 102 TFEU decision adopted by the Commission that he liked. He tells me he replied with a blog post last week (see here). Perhaps I read it too quickly, because I don’t see an answer 🙂  In any event, the fact that he did not identify any case with which he agrees probably means that there is no such thing.

[Intermission: the fact that we are good friends enables us to discuss things in a way which would be much harder to do with other people. This was also the case with Nico back in the day. This, by the way, confirms that having another brilliant academic with views not always coincidental with mine was a great decision for this blog].

This failure to choose is interesting because, in my view at least, the Commission is quite (perhaps too) selective when it comes to picking abuse of dominance cases (that is unless they are predestined to go through the commitment route; it’s those that I personally like the least, not the Article 7 ones, which tend to be quite solid).

Law in abstract and Law in casu. The reason Pablo doesn’t reply citing specific cases is because he says he “does not see Commission decisions that way” (I can see how people don’t have a list of “best” and “worst” decisions, but if anyone had one, it would have been Pablo…). His point is that he doesn’t really care about cases, nor about who wins or loses, but about “the way in which the law is shaped and evolves over time”.

This is commendable for an academic, but I’m not sure I agree with the implications. In competition law it is cases that shape the law and that drive its evolution, so one can perfectly assess cases in the light of their contribution to the state of the law. If what Pablo means is that he doesn’t care what party wins or loses, I can testify that he truly doesn’t. If what he means is that cases should not be driven by policy but by the law, then we fully agree. However, I don’t see why all of this could mean that there cannot be cases that he likes or dislikes.

The approach of a practitioner is not, or should not be, so different. I only care about the party who wins when the case involves a client of mine. I also take an interest, but one that has nothing to do with the law, when a friend is involved (for disclosure purposes: I have good friends involved in Intel and Post Danmark II). As for the rest of the cases, I have enough with understanding the case-law and how it can relate to my clients’ issues, and I am not so concerned about contributing to the evolution of the law in a particular direction (partly because I don’t know what side I’ll be on in the future, and partly because it would be pretentious on my part: I would rather leave that to those whose jobs is to study cases in the depth they deserve, like the parties to every case, the judge, the clerks, or the academics who may want to contribute to the debate).

In sum, I’m not in the business of trying to influence the evolution of the law (except when paid to do it), and this is what explains that I haven’t written about the ongoing debates on exclusivity rebates, Intel and Post Danmark II, that occupy Pablo and others at a time when these important specific cases are pending.

Pablo’s whole post is about returning a question to me (never mind that mine wasn’t answered!), and to compel me to spend part of a Sunday morning typing instead of doing better things break my silence; the question is:

Do I believe exclusive dealing should be prohibited absent an objective justification or whether, instead, Article 102 TFEU enforcement should follow the principles set out by the Commission in the Guidance and by the Court in Delimitis?

On the key assumptions underpinning the debate. Pablo’s post notes that his (brilliantly written) paper on Intel , everything,  exclusive dealing and loyalty rebates focused not on who won or didn’t, but on the “key assumption underpinning 35 years of case law”.

Discussing key assumptions makes a lot of sense, so let’s start from there:

-Unless I’m wrong, the key assumption underpinning 35 years of case law is that in markets characterized by the presence of a dominant company (not the case in Delimitis, mentioned by Pablo), the use of exclusivity inducing arrangements can be deemed prima facie restrictive of competition. I’m familiar with the case law in this regard and actually think that this is sort of intuitive, for exclusivity almost by definition raises barriers to entry and deprives rivals of scale (as Pablo has very well explained in other domains –see here-, EU Courts have been able to implicitly incorporate sound economic insights to their case law). Many of the most reputed competition economists do not seem to question this. I won’t bother to conduct research on this point for a blog post, but I happen to have read this Carl Shapiro piece yesterday for a case in which I’m working, and it is quite clear.

– And unless I’m wrong, the key assumption underpinning the critique to that case-law is that “the lessons of experience and economic analysis” (this is the formulation in vogue, also used in Pablo’s post, tailored to evocate the Cartes Bancaires Judgment and draw a parallel) undoubtedly show that exclusivity arrangements are more often than not procompetitive, also when carried out by a dominant firm. Leaving aside the fact that experience and economics do not always go hand in hand, there is this widespread assumption that according to economic “science”, it is absolutely beyond discussion that the law here is a mess.

Leaving the theoretical economic literature aside (basically because I don’t know much about it; query: is there so much conving research on the advantages of exclusivity when carried out by a dominant firm?), I know from my personal experience with companies that exclusivity inducing rebates may –in certain cases- be perfectly justified by reasons other than anticompetitive motive. But I frankly do not know whether these outweigh, in the abstract or in general, the anticompetitive perils associated to these practices when carried out by a dominant player.

Those who have heard my presentation on two-sided markets or that will read my forthcoming Competition Law Journal article on the subject will realize that I’m all for taking into account economic lessons -when they are well established- for the application of the law. I’m simply not fully persuaded that economic research so clearly shows that the current state of affairs in the EU is so manifestly wrong.

Since I am asked, in my view the current state of the law strikes what seems to be a reasonable balance, at least in theory.  It may, like almost anything, be debatable, but I fail to see it as the epitome of absurdity:

-In the field of Article 101, the Commission’s soft law as well as the case-law (mainly Delimitis, cited by Pablo) explicitly acknowledge the mixed effects that exclusivity agreements may have, and subject them to a balancing test in which the burden of proof rests on the accusing party. No one seems to complain about this.

-In the field of 102, the case-law takes into account that the degree of competition is already lessened by the presence of a company that is, by definition, able to behave independently of competitors and customers (in a way, conducting a strict foreclosure/effects assessment from this starting point risks incurring a variant of the cellophane fallacy, which is what the Court said, in a way, in the heavily criticized para. 245 of Michelin II) and strikes a different balance. This is explicitly explained in para. 89 of the Intel Judgment.

In the 102 domain, precisely because competition is considered to be reduced, restrictive effects are presumed in a first stage, BUT there always remains an open door to show that the practice is not abusive because parties can always show that the arrangement is objectively justified.

This reflects a double assumption that exclusivity always makes life more difficult for the competitors of the dominant company (as explained in para.150 of Intel, the analysis favored by the Court is one of difficulty, not the one of impossibility linked to the as-efficient competitor test set out in the Guidance; this is a crucial point that many overlook and that has to do with how we define foreclosure) and that sometimes exclusivity is part of a procompetitive strategy.

As I hinted in the first comment to Pablo’s post on Intel, this is key. If this escape door were not here, I would also take issue with the case-law. But it is, as unequivocally stated in paras. 94 and 173 of the Intel Judgment. In my view, this explicitly acknowledged the economic lesson that in some cases these practices may be procompetitive and hence should not be prohibited. Why is this, legally speaking, a problem? To the extent the presumption can effectively be rebutted, I see no problem to it.

Now, a different debate is whether an “objective justification” defense is a mere chimera or not, and there, I do agree that it should be a real possible defense, not just some nice wording.

In an ideal world,  and like I have said more generally with respect to 101, presumptions at the level of establishing the restrictive effect of a given practice should not be so important as they are. Firstly, because if something is so obviously restrictive to be deemed restrictive, then it should not be so difficult to show effects (as, by the way, both the Commission and the General Court were able to do in a few hundred pages in Intel; this, on the other hand, is probably a very good example of why shortcuts may make sense). Secondly, and conversely, because if a practice is so obviously pro-competitive, and if defences (like 101(3) and the objective justification notion) were effectively available, then there would be no obstacle for the practice to be redeemed this way.

On the old debate of form and effects. I have in the past set out very clearly my views on the interface between competition law and competition economics (see here), so I won’t repeat myself. Let me just add that when it comes to exclusivity inducing rebates, even people not at all suspect of “ordoliberalism” [one day we should try to clarify here what this means], like Commissioner Josh Wright, are of the view that cost assessments (like the one advocated for in the Guidance Paper) might not be well suited for loyalty discounts, because their essence lies not in price but on exclusivity (see his speech “Simple but Wrong or Complex but More Accurate? The Case for an Exclusive Dealing-Based Approach to Evaluating Loyalty Discounts”. On this point, see also the excellent writings of one of our Friday Slotters, Einer Elhauge (see e.g. pages 463-464 of this great one).

“Rules and standards need to be crafted to ensure that they are accurate and administrable” is a phrase that appears at the end of Pablo’s post, and that leads me to one final comment. The current situation is, in my view, the one that is easiest to be administered, and, importantly, the one that requires less work from the lawyers, and particularly from the economists advising the dominant company. If we were to apply the Guidance paper test to all these cases, we would need to deploy hordes of public officials, and countless hours of lawyers and economic consultants (as if something good comes out of that mix…)

Under the current situation, on the contrary, companies know that except for one red line, they can design their rebate schemes the way they wish. Most of the objectives of exclusivity inducing practices can be achieved through other, perfectly legitimate and less risky, means. With less intense but more refined and creative legal and economic advice companies could continue competing intensely on the merits whatever the rule on loyalty-inducing rebates.