Chillin'Competition

The provisional judgment of the Competition Appeal Tribunal in Ping is now out. Most of you will remember that the case is about the status of (outright) online sales bans under Article 101(1) TFEU.

The CAT has upheld the CMA’s decision: such bans are restrictive by object. However, it departs in some crucial respects from the authority’s analysis.

In the post I wrote back in May, I explained that the CMA had conflated the question of whether an agreement is restrictive by object with that of whether it is objectively necessary.

The CMA’s decision indeed suggested that companies can only escape the by-object qualification if they can show that there are no less restrictive means to attain the legitimate aim they seek.

The CAT concluded, in line with what I suggested, that the CMA erred in law in that respect. If the CMA’s analysis was incorrect, you may be asking why (and, more importantly, how) the decision was nevertheless upheld.

The CAT did so through a somewhat heterodox interpretation of the case law and, more controversially, by coming up with a surprising statement that questions everything we thought we knew about cartels.

Restrictions by object, pro-competitive rationales and cartels

Ping argued (para 101) in the proceedings that an agreement is restrictive by object only if it lacks a plausible pro-competitive rationale. As Alfonso and I have explained many times here, this is the most sensible way to make sense of the case law (with the exception of cases that relate to market integration, where the bar is higher, and vertical price-fixing).

Ping’s arguments are in line with AG Saugmandsgaard Øe’s opinion in F Hoffmann La Roche v AGCM (which is not cited by the CAT).

Cartes Bancaires is an example that illustrates the Court of Justice’s approach well. In the relevant economic and legal context, the agreement was understood to be a plausible means to address genuine free-riding concerns. As a result, it was found not to have an anticompetitive object.

The CAT does not read Cartes Bancaires in this way. However, it fails to explain what methodology, if any, the Court followed to ascertain the object of the agreement in that case. Interestingly, the CAT also refers to Delimitis, in which the Court’s approach to by-object infringements (paras 10-12) could not be more transparent (and more in line with Ping’s arguments).

The most interesting bit of the Ping judgment comes immediately afterwards (same paragraph). The CAT argues that Ping’s approach to identify by-object infringements (is the agreement a plausible means to attain a pro-competitive objective?) cannot be right because it contradicts the Court’s judgment in BIDS.

BIDS, the CAT acknowledges, was a cartel case (a cartel of well-meaning people, as I like to say, but a cartel nonetheless). And because it was a cartel, the Court could only conclude, as it did, that the agreement was restrictive by object (I never understood why such a straightforward case generated so much interest).

Cartels can take many different forms and can be concealed in all sorts of ways. The defining feature of cartels, their essence, is the fact that they lack a plausible pro-competitive rationale. This is what makes them stand apart from other horizontal agreements – and the reason they are the enforcement priority of competition authorities around the world.

And because cartels lack a plausible pro-competitive rationale, firms rarely ever try to come up with a justification under Article 101(3) TFEU. And when these justifications are advanced, they are irrelevant because they lack credibility. This is the – very sensible – point the Court made in BIDS: once it is established that the agreement is restrictive by object, it does not matter whether it is alleged to pursue other objectives.

For instance, pharmaceutical companies may attempt to claim that the point of their cartel is to devote the additional profits to enhance their research and development capabilities. Or book publishers may try to claim that their cartel seeks to expand the range and diversity of their titles. These arguments can be disregarded precisely because they are implausible.

Or so we thought.

The CAT points out in its judgment that it was ‘obviously plausible’ that the cartel at stake in BIDS ‘might be pro-competitive’. You read it right.

The consensus about cartels (what authorities, courts and international organisation have been saying for decades), the foundations of contemporary competition law and policy, casually questioned in a paragraph. Not only is the consensus wrong; it is ‘obviously’ wrong.

There is much debate about unilateral conduct, mergers, and vertical agreements. We thought cartels was the one area where there was no disagreement. Until Friday of last week.

What explains the CAT’s argument? Was there another way?

I can think of a (plausible) reason behind the CAT’s reasoning in Ping. On the one hand, it may have felt that Ping’s aims were legitimate. On the other, it may have felt that the outcome of the CMA’s decision was correct. Accordingly, it concluded that, even though the object of the agreement could be pro-competitive, the agreement had an anticompetitive object (I am paraphrasing).

One can come to the conclusion reached by the CAT without so many contortions, and without casually reinventing economics along the way.

The reason online sales bans are, in principle, restrictive by object has to do, I believe, with market integration in the EU. This is clear from the Guidelines on vertical restraints and the Digital Single Market Strategy. E-commerce is a powerful tool to enhance cross-border trade and integrate Member States’ economies.

One cannot be surprised, accordingly, that outright bans on online sales are treated more strictly than most other vertical restraints. It is one of these areas where it makes sense to apply the logic of Consten-Grundig. Because market integration is at stake, even a clause that is plausibly pro-competitive is deemed restrictive by object.

I have already explained that I do not have any problem with the introduction of market integration considerations in EU competition law. So long as courts and authorities are explicit about the fact that such considerations guide the outcome in the case, I fail to see why it would be controversial to rely upon consistent case law dating back to 1966.

Why the CAT’s judgment may be a problem

One could argue that the CAT’s judgment is just an anecdote to be mentioned in passing at conferences and in the classroom.

I am not so sure.

We live in times of change. It is not a secret that some sectors of the economy are trying to justify the creation of cartels in the current (disruptive) environment. And, as the Ping case shows, there may always be an astute lawyer able to persuade courts and authorities that cartels are good after all.

Creating the perception that cartels may be a plausible means to attain the objectives advanced by their members may eventually pave the way for sector-specific exemptions aimed at extracting rents from powerful suppliers and/or customers.

And (sorry to end on a gloomy note) the moment cartels start to get exemptions, (EU) competition law and policy would be over. This, I guess, is one of the battles for the years to come.