Chillin'Competition

Part I of this post explained our agreement with the bigger picture issues discussed in DG Laitenberger’s important speech at the CRA conference last week.

This second part focuses on more specific, seemingly fine print, issues on where our assessment may diverge or, rather, incorporates some nuances that we believe are important.

Accuracy vs Administrability (and on the length of investigations) The speech identifies a debate cast as an inevitable trade-off between accuracy and administrability (a language also used here) and suggests that there are ways to reconcile both objectives. The solution, according to the speech, could be found in presumptions and rebuttals. It is interesting to note that, according to the speech, partisans of accuracy would favor time-consuming and resource-intensive investigations, which arguably “risks the untimeliness and hence the non-effectivity of enforcement”. As the DG put it, “this is antitrust to the measured Louis Armstrong tune ‘We have all the time in the world’”. Later on the speech contains another reference to “concluding [cases] within a reasonable time”. These references –or the assumption that lengthy investigations- may ultimately be untimely and ineffective have many angles and have spurred some interesting ideas. We’ll be back on this soon.

On the Notion of Restriction of Competition – By Object. The speech used two examples of presumptions-rebuttals. The first one was that of the recent ISU (International Skating Union case), in which the Commission recently found an infringement by object. According to the speech “[o]ne can say that under Article 101 TFEU, if a restriction of competition is established by the enforcing authority, the absence of a justification by efficiencies is presumed. But this can be rebutted by the business investigated”.

There is an important nuance here.  If what the DG is saying is that establishing a restriction by object implies a presumption of the absence of a procompetitive justification which can then be rebutted (as the context to the phrase in the speech suggests), then what the DG is saying is fully in line with the interpretation of the law that we developed in our piece On the Notion of Restriction of Competition (recently cited by AG Saugmandsgaard Øe; see here). If that’s the case, this is a welcome evolution or an accurate, but rare, statement of the law as we see it. But, following that same logic, finding a restriction “by effect” doesn’t presume the absence of efficiencies; on the contrary, it accepts the plausibility of the efficiencies, which the company has to substantiate. In other words, “establishing a restriction” does not imply a rebuttable presumption of lack of efficiencies; establishing a “restriction by object” (after having examined the legal and economic context), does.

Implementing Intel going forward, in theory and in practice

The speech then referred to Intel as an example of the importance of presumptions and rebuttals. In reality, however, the speech was doing more than that: it was explaining, for the first time, how the Commission interprets the Intel Judgment (for our very own comments on the Judgment, see here, here and here, and for Nicolas Petit’s most recent piece on the subject, see here too).

In DG Comp’s view, the CJ in Intel confirmed the presumption that exclusivity rebates are anticompetitive, but clarified that a dominant company can rebut the presumption by showing that the conduct was not capable of resulting in foreclosure. According to the speech, only when a company puts forward “sufficiently serious and substantiated arguments” will the Commission assess whether the conduct is liable to foreclose “as efficient competitors”.

This triggers several comments:

On the Existence of a Presumption. First, it is noteworthy that the DG has decided to emphasize the role of the presumption against fidelity rebates when applying Intel. It is a welcome development that the presumption is considered to be rebuttable “at the level of the likelihood or not of anti-competitive effects”. As Pablo has explained, it was also an inevitable development in the light of other recent case law. At the end of the day, however, whether a presumption exists or not is of relative importance; relative in relation to the standard for rebuttal…

On the Standard for Rebuttal of the Presumption. According to the speech, “obviously, the required standard for rebutting the presumption would be meaningless if the dominant firm was able to put forward general theories and abstract arguments” (…) “the dominant firm must present case-specific arguments based on concrete evidence, and this must be done during the administrative procedure” (…).

The speech explains that it is only when the dominant firm “puts forward sufficiently serious and substantiated arguments and evidence” that the Commission “must undertake an analysis showing capacity of the conduct to foreclose as-efficient competitors” The DG calls this “placing the burden of the rebuttal on the dominant firm”, and adds that “when in doubt, the analysis has to be deepened, but the procedure is a two-way street”.

The reason Intel is important, in practice, is –as explained in our first post- because of the shift in the burden of proof; it meant that it was not upon the company to demonstrate that its conduct was objectively justified (for which it bears the burden) but, once the argument is made, for the Commission to show that it is likely to foreclose competition. And case are most often won and lost on the burden of proof.

But the Commission’s intention, it seems, is to argue that the burden remains with the company, now to present “concrete evidence” that is “sufficiently serious and substantiated”. The problem here is that one must be very careful not to conflate the burden of proof and the evidential burden; the two are very different things. These difficulties also emerge when discussing the counterfactual, as previously, albeit briefly, observed here.

In my view, it should be enough for dominant companies to identify the concrete (existing or practicable) evidence that would be required to establish lack of foreclosure of as-efficient competitors. It would then be up to the Commission to gather, produce, require this evidence, or to explain why it is irrelevant or insufficient (and, in the latter case, to explain what would be sufficient). This is because sometimes the firm will not be in possession of all the info, or the info will not be at its disposal and in any event there will always be uncertainty as to what the Commission may consider to be “sufficiently serious and substantiated”. This is a shared task, and the proposed solution should combine accuracy and administrability avoiding the extremes, and respecting the respective burdens. There is no reason why this exercise could not be carried out constructively and in good faith; it is all about doing the necessary to ascertain the relevant facts.This would also be fully in line with the “proof-proximity” principle, brilliantly explained by Cristina Volpin in this paper (another young academic, Andriani Kalintri, also has done excellent work clarifying the intricacies of evidential burdens, see here.)

In other words, the Commission cannot abdicate its responsibilities. The case law has repeatedly established that “the Commission must play its part, using the means available to it, in ascertaining the relevant facts and circumstances” (e.g. CJEU in Consten/Grunding or GC in E.ON) and that it is required to have at its disposal all the relevant data that must be taken into consideration in appraising a complex situation (the Tetra Laval standard). As a matter of fact, this very speech says in a different point that “we would not do our job properly if we were not to use all available sources of evidence”.

If this doesn’t convince you, look at the very recent CJEU Judgment in Frucona Košice Judgment (State aid, which means most have not read it) according to which “the information ‘available’ to the Commission includes that which seemed relevant to the assessment to be carried out in accordance with the case-law (…) and which could have been obtained, upon request by the Commission, during the administrative procedure (71). The CJEU then observed that the Commission had “failed to obtain” (80) “all the relevant information” (81) and confirmed the annulment of the Decision. Remarkably, that was a case in which the Commission also claimed (see its arguments before the GC) –unsuccessfully- that the burden was not incumbent upon it [for my view on the details and implication of this case you can….Garrigues paywall] 😉

The Elephant in the Room- The Guidance Paper. When it comes to abuse of dominance, there are currently two parallel levels, one is the law as set by EU Courts; the other is the Commission’s enforcement criteria as set by the Commission itself. The Intel Judgment has to do with the first of these levels. The Commission’s explanations as to how it intends to implement Intel, have to do with the second.

 As I said in my first hasty comment on the case, “in its Guidance Paper on exclusionary abuses the Commission already committed to a careful assessment of likely effects (and this regardless of whether the undertaking concerned submitted evidence challenging the capability of the conduct to restrict competition). The Commission did not consider then that such an approach would make enforcement impossible nor did it consider that it contradicted earlier case law. In my view, therefore, the Commission already self-committed to applying a standard that is even stricter than that required by the Court in today’s Judgment”.

Against this background, something that at first sight may seem remarkable about the Commission’s reaction to Intel is the absence of any reference to the Commission’s own Guidance Paper or to its content, precisely at the time when the CJEU would seem more favorable to it than ever.

On the other hand, however, the speech states that “the Commission will apply the most suitable tools to assess the specific case – including, where appropriate, analysing the “as efficient competitor test” when the dominant company provides the necessary information during the administrative procedure. We are confident that this will allow us to continue to focus on the most harmful cases, while concluding them within a reasonable time”. One could read this as anticipating that cases will continue to be prioritized in the light of the AEC test following the self-commitment made in the Guidance Paper. Only time will tell.

Once again, the CRA conference proved to be the second 😉 most successful competition conference in town (congrats again to Cristina Caffarra and her team). Director General Laitenberger delivered an important speech that struck some chords with us and that, as explained below, relates to much of what has been discussed on this blog. The speech is now online.

As you will see below, we very much agree with many of the big picture messages, but have some doubts or nuances regarding the important small print, particularly on the implementation of the Intel Judgment going forward.

Part I of this post will be devoted to the big picture. Part II discusses the fine print.

The Political Background

DG Laitenberger gets the big picture and the technical competition law stuff, and the speech was a very good reflection of that. The speech run an introductory thread through a number of societal issues, including growing inequality, citizens’ doubts as to the functioning of markets, the reassessment of beliefs and certainties in the wake of the economic crisis or the role of technology. On this latter point he referred to fake news, clickbait, to “echo chambers” and increased polarization. That is the current background leading to the question, “where does this leave competition law”?

In my introductory speech at the past Chillin’Competition conference I actually referred to most of those issues, to make the point that while (really) important things happen, we competition lawyers live in our bubble and within our very own echo chamber. My point was not to connect competition law to those challenges, but rather the contrary, a bit to appease my conscience for devoting so much time to competition law when there’s so much else to be done. The DG’s diagnosis of the current background was brief, but right, and, to me, uncontroversial. The link to competition law was perhaps less evident, but still.

The Wider Competition-Related Messages

1) DG Laitenberger said that “while the economy and the society change so much, the ground rules of EU competition law remain so stable. They were designed to apply to a wide range of scenarios. They have proven capable of navigating the last 60 years. I should think that they will be able to navigate the next 60 years. This is so because the nautical charts- the fine print- which underpin the direction are continuously refined and adjusted”.

I wholeheartedly agree, with only one nuance. The refinement and adjustment cannot be radical or motivated by considerations alien to the law and that change the discipline’s direction. As we have always repeated, the stability of the law has to do with its relative isolation from small politics and by the slow judge-made distillation of common sense infused with mainstream economics in the light of experience across all countries, sectors and products.

2) The DG went on to say that “I see some polarisation in the competition community as well. My thesis today is that we can work to move away from the more extreme interpretations of the charts. We can-and should- focus on building a more common understanding”.

Again, entirely agree. We have always favored a radically centrist view of competition law, with regard to many specific issues, even connecting it to the political center in the big picture.

In that regard, I also welcomed a reference in the speech to something that is not often said and that I believe is important and true, that “ultimately, with all its refinements and adjustments, the interpretation of the antitrust remit has been more stable on this side of the Atlantic than in the U.S.”

3) The end of the speech touched on “the role of competition law vis-à-vis other societal concerns”. Echoing Commissioner Vestager’s speech at our last conference, DG Laitenberger acknowledged that competition law can’t fix it all. He nonetheless explained also that “it is possible to have big-picture concerns- about fairness, or inequality of innovation. While applying rigorous enforcement at the same time. And it is possible that rigorous competition enforcement has beneficial effects on these big picture concerns- sometimes inevitably so (…) There is no reason to be shy about the overall impact of competition law (…)”.

Once again, agreed; in fact this is the very same point I made on this editorial piece for JECLAP about fairness. As developed therein, references to the “big picture” cannot expand the reach of the competition rules, but I also believe that a more fair society is a consequence of the right (“rigorous” in the words of the DG, a very important nuance) application of the competition rules.

We have just received this video with the highlights of the conference featuring interviews with some of our attendees including a journalist whose name I can’t recall now. the President of the blog’s fan club, a speaker from last year, a speaker from this year, and a soon-to-be-fired colleague of mine.

And for a very good summary of the conference by our friends at Gecic Law, click here! 

A few days ago the European Commission opened an investigation into an alleged abrewse of dominance to restrict beer imports into Belgium (see here).

Ale those of you that thought we would barley survive without endives will now hopfully realize that this blog is lager than one single product. At yeast, we now have a hopportunity to show that we can focus on the big pitcher.

To be sure, there have been plenty of cases concerning beer before and we have written a number of posts on this drink (see e.g. here and here). In that sense, this may seem like a déjà-brew.

Indeed, alcoholic beverages are perhaps the product that has contributed the most to EU case law and to competition law in particular. Sometimes they were the subject of cases and, even when they were not, their influence clearly emanates from the content of some decisions…

But… wheat a second…. Actually, we can’t comment on this case due to a conflict: AB Inbev was the sponsor of the brewtal open bar we held at our first Chillin’Competition conference. Anything we say could therefore be regarded as an attempt to persuade the Commission to leffe the company be and drop the investigation.

Instead of providing you with our views, we will therefore provide you with some reactions from people who typically seek reactions, the members of the “Brussels competition press corps”, who have a stella reputation as competition commentators and a thorough knowledge of the relevant market. Our sources include Aoife White and Gaspard Sebag (Bloomberg), Rochelle Toplensky (FT), Lewis Crofts and Matthew Newman (MLEX) and Nicholas Hirst (Politico).

Unfortunately, after a few drinks we don’t remember who said what, so we can’t really attribute any quotes, sorry.

According to one of our sources, the decision to open the case was adopted only in light of a special report from the Chief Economist. The aim of the report was to identify the product that enjoyed the highest consumption among officials. This was part of a strategy to first adopt a decision and then lodge a follow-on action for damages suffered by the Institution, much like what happened in the elevators case (see here).

Another Brussels-based journalist reports, on the contrary, that the case originates from an informal complaint by the College of Europe alumni association (based in Place Lux) that, reportedly, is preparing a billion euro class claim.

The “Brussels Bar Association” also claims to be thd representative of the main class affected by the case. We have no confirmation of whether they represent lawyers or actual bars.

Conversations between our sources and parties connected to the case nevertheless all converge in anticipating that defence arguments will be threefold, namely (i) -“Who ever reads the small print on beer cans??”; (ii) “Competition is just one Chimay away” and (iii) “Hasn’t anyone realized that water is more expensive in this country!?”.  Economists in turn, are wondering whether one should factor in hangovers and associated lack of activity to the consumer surplus/deadweight loss analysis.

The case is also expected to shed light on several procedural issues (“if you stop drinking, can you challenge jurisdiction?”). The investigation is nonetheless expected to leffe issue unresolved and to result in consumer uncertainty (“so where do I go to buy my Christmas Kriek supply – Lille or Eindhoven?”, is a question many are asking themselves in the wake of the Commission’s press release).

If any of you has any comments on the case, feel free to comment on this post.

Cheers!

The Court Judgment in Case C-230/16 is out (available in French and in English). And, in the spirit of times, here is an online channel providing the first comment on it.

The judgment follows faithfully AG Wahl’s Opinion, which came out in July (and which we discussed on the blog). It seems to be also in line with the position consistently expressed by the Commission since the Guidelines on vertical restraints.

The vast majority of our readers will remember that the case is about the legality of a ban on the use of online marketplaces by the members of a selective distribution system. The Court has ruled the following:

• Where the conditions of Metro I are fulfilled, a restraint aimed at preserving the luxury image of a product is presumptively lawful – in other words, it falls outside the scope of Article 101(1) TFEU altogether.
• A selective distribution system that provides for an online marketplace ban is not caught by Article 101(1) TFEU where the conditions of Metro I are fulfilled (that is, where the nature of the product requires the use of the system). Such a ban does not go beyond what is necessary to preserve the luxury image of the product.
• An online marketplace ban is not a hard-core restriction within the meaning of Article 4 of the Vertical Block Exemption Regulation if it does not limit passive sales and/or the customers to which the distributor can sell the product.
• The ruling in Pierre Fabre is confined to the specific circumstances of that case, i.e. an absolute ban on online sales (paras 33-34).

More important than the ruling are the implications than one can infer from it, which can be summarised as follows:

• The Court makes it clear beyond doubt that the object of an online marketplace ban is not the restriction of passive sales and/or the customers to which distributors can sell. This is in line with the position expressed by the Commission in the context of the e-commerce sector inquiry.
• By way of consequence: the Court strongly signals that an online marketplace ban is not a restriction by object within the meaning of Article 101(1) TFEU.
• What are the implications? The first implication is that a case-by-case analysis of effects would in any event be required for non-luxury goods. Byzantine discussions about whether Asics or Mizuno shoes qualify as luxury goods seem entirely irrelevant in this regard. The ‘by object’ shortcut would not apply, irrespective of the nature of the good.
• Insofar as an online marketplace ban is not a hard-core restraint within the meaning of the Regulation, the benefit of the block exemption applies irrespective of the nature of the product. Again, the ‘luxury or not luxury’ dilemma will be largely irrelevant in practice as a result.

I have summarised my understanding of the ruling in the table below:

Other comments

Copad was the key precedent all along

The Court approaches the question in the way I would have done it. As I mentioned in July, the key precedent was Copad, an intellectual property case. In Copad, the Court ruled that a trade mark licensor can invoke its rights to prevent a licensee from selling to non-members of a selective distribution system.

That ruling is based on a key premise: if companies cannot protect their intangible property (brand image, trade mark, goodwill) when dealing with third parties, they will refrain from licensing and from selling via independent distributors. Why would firms do something that would force them to lose control of the image they want to convey?

More importantly: why would EU law penalise firms that sell or produce via third parties and favour those that produce in-house? Copad and Coty suggest that EU law is agnostic about the business model that companies choose. There is no reason why vertical integration should be favoured over licensing and/or selling via third parties. And there are many good reasons why the latter should not be treated more strictly.

The protection of intangible property is important for all producers, not only the producers of luxury goods. Asics or Mizuno shoes may not be seen as a luxury product (it is all in the eye of the beholder). This, at the end of the day, does not really matter. What matters is that conveying a particular image may also be important for these companies.

Trade mark law seeks to protect all producers, not only producers of luxury goods. There is no reason why EU competition law should be different. Coty appears to be in line with this position.

The administrability trap in EU competition law

The ‘by object’ prohibition of online marketplace bans has been defended by many in the past few months. One of the arguments that has frequently been invoked is the ease of administration of a ‘by object’ rule. If we know that a practice can be bad for competition, is it not better to lay down a prima facie prohibition so we all know where we stand?

Coty shows that the Court was not impressed with these arguments. If everything were about the easy administration of a rule, then every practice would be prohibited by object, and this is clearly not the case.

What is more, this view ignores that prima facie legality rules are also easy to administer, and are an integral element of EU competition law.

Metro I, where the Court clarified that, in some circumstances, selective distribution systems are presumptively compatible with Article 101(1) TFEU (irrespective of the market share!) is a wonderful example. We know now that the legality rule may apply even when the agreement provides for an online marketplace ban.

 CoRe, CCIA and the VUB held a great conference about competition law in digital markets last week in Brussels. Asked to focus on multi-sided markets, my claim was that the case law already provides enough guidance for us to apply the law to multi-sided business models. I posited –once again- that the real problem we have in competition law is that we are forgetting about the law.

Contrary to what many tend to assume, in the EU we have actually made quite some progress, at least on the judicial front. In fact, my bet is that when the SCOTUS rules on Amex (a case which some expect will provide us with quasi divine guidance), it will say nothing new as regards what we already have in EU case law.

Courts do get it. I witnessed that first hand when I was the guest lawyer at this special seminar organized by the European Association of Competition Law Judges. The problems arise when we don’t listen.

Building on a previous paper (“The Double Duality of Two-Sided Markets“) and on ideas voiced out in other occassions my intervention focused on 6 lessons from the case-law. I would like to believe they result from an objective reading of the case-law (even if, to be sure, I do advise several clients operating multi-sided business models).

The lessons I extract from the case law are the following:

1. One cannot look at different sides of a multi-sided business model in isolation (partly discussed here);
2. The anticompetitive value that was typically attributed to network effects needs to be nuanced (an idea partly discussed here);
3. Cross-market anticompetitive effects require, as a minimum, a showing of anticompetitive foreclosure/elimination of effective competition in the target market (a point also made by Pablo here);
4. The assessment of pro-competitive effects must consider the benefits flowing to all sides of a multi-sided system (see here);
5. Multi-sidedness considerations are part of the legal and economic context to a given practice, not a side issue to examine in isolation as a last step in the analysis (idea anticipated here);
6. The key to competition law enforcement in multi-sided markets has to do with the analysis of the counterfactual (i.e. causality/attributability) (partly discussed here).

I think I will develop these ideas in a proper article. If anyone of you has any comments on these lessons (or on others), please send them my way!

For those lucky enough to be in London next week: Judge José Luis da Cruz Vilaça, President of  Chamber at the European Court of Justice, will deliver a lecture at King’s College London on 14 December. The event is organised by the Centre of European Law and will be chaired by Professor Alison Jones.

As most of our readers know, Judge da Cruz Vilaça has presided over the chamber that has delivered some of the most important recent rulings in our field. He has acted as rapporteur in Intel and Post Danmark II.

In case you were wondering: I am going! You can sign up for the event here.

Last week the Court delivered a preliminary ruling in Case C‑547/16, Gasorba. Together with Case C‑248/16, Austria Asphalt (which came out in September) it is a very serious contender for the Most Predictable Ruling Award 2017.

In essence, the preliminary reference asked whether a national court can take action under Articles 101 and 102 TFEU where the European Commission has already accepted commitments in relation to the same practice.

In a very brief judgment (serious contender for that category too), the Court clearly ruled that it is – of course – possible for a national court to examine the compatibility of practices with EU competition rules in instances where the European Commission has closed a case with a commitment decision within the meaning of Article 9 of Regulation 1/2003.

I do not see how the Court could have come to a different conclusion. Regulation 1/2003 is crystal clear on this point – I remember discussing Recitals 13 and 22 of the Regulation, which could not be more explicit, back in 2005, when I was a teaching assistant in Bruges. I do not feel there has ever been any doubt (or expectation that the Court would see things differently).

I am sure some people will criticise this outcome – it may perhaps be argued that this outcome creates legal uncertainty, or that it reduces the incentives to offer commitments.

To which I am tempted to reply: if the outcome of the ruling is deemed undesirable, what should change is Regulation 1/2003, not a Court ruling echoing the unambiguous choice made by the legislature.

The Court, however, does not rule that commitment decisions are entirely irrelevant in proceedings at the national level. The judgment clarifies that national courts are expected to take into account the preliminary assessment of the Commission in a commitment decision, and this as an expression of the principle of sincere cooperation (or loyalty) enshrined in Article 4(3) TUE.

Because these weeks my research work is taking me to the old days, I thought immediately of the Lancôme ruling of 1980, in which the Court considered the status of the ‘comfort letters’ issued by the Commission under Regulation 17, and came to a similar solution. For younger readers: comfort letters are like commitment decisions avant la lettre (if you allow the cacophony and the bad pun).

More interesting than the ruling is the debate about the nature of commitment decisions. I have the impression that, to this day, these decisions are, for some reason, often taken as instruments stating what the law is. They are not. They are an expression of the discretion that the Commission enjoys when choosing which cases to pursue.

Because commitment decisions are about discretion, not about the interpretation of Articles 101 and 102 TFEU, they are only subject to limited review. Accordingly, the ‘preliminary assessment’ found in these decisions is, if at all, subject to the same ‘manifest error’ test that applies to the rejection of complaints.

In other words: curb your enthusiasm.