Relaxing whilst doing Competition Law is not an Oxymoron

Two important Opinions: AG Kokott on Post Danmark II (C-23/14) and AG Wahl on AC Treuhand (C-194/14 P)

with 7 comments

A few hours ago two important competition law-related Opinions were made public by the European Court of Justice. Both are remarkable and, interestingly, I would even dare to say –and note that this is not a criticism- that the direction of each of them could have been expected in the light of the track record of their respective authors, Mr. Wahl and Ms. Kokott. Similarly, I also have the sense that the Court might finally be inclined to follow only one of the two Opinions, I let you guess which.

That said, both Opinions raise most interesting questions regarding two very different issues, one of them novel (can a cartel “facilitator” be sanctioned under 101 TFEU despite not being a party to the agreement?) and one of them fairly old but always hot (what are the criteria to assess loyalty rebates by a dominant firm?; Is it mandatory to follow the “as-efficient competitor test”; Is there an appreciability threshold for such conduct to fall under 102 TFEU?).

Let’s look at them one at a time. I have more extensively summarized Kokotts Opinion (because I know you’re too lazy to read Opinions in full), but have included all key messages in bold for those lazy enough to not read even the blog posts 😉

AG Kokott on Post Danmark II (more on the fight for the soul of EU Competition Law)

The legal treatment to be applied to rebates on the part of dominant companies remains one of the most contentious issues in contemporary EU competition law and is in many ways the main battleground on the –legal or economic- soul of EU competition law. Discussions about it have abounded in recent times (see here for a summary), and have also occupied our attention (for Pablo’s views see here, here or here, and for my own views click here).

Kokott sends a clear message right from the start (para. 4), noting that the case comes at a particularly controversial time when many push for a more “economic approach”, and recommends that “in is replies, the signal effect of which is likely to be extended well beyond the present case (she refers to Intel in a footnote) the Court should not allow itself to be influenced so much by current thinking or ephemeral trends, but should have regard rather to the legal foundations on which the prohibition of abuse of a dominant position rests in EU Law”.

In her Opinion the AG first goes on to identify the general criteria that should be taken into account in order to assess rebates, referring first to the special responsibility of the dominant company (para. 24), underlining that the “quantitative” or “loyalty” labels are irrelevant, and that what is decisive is the possibility that they may lead to an exclusionary effect which is not economically justified (para. 29). AG Kokott then insists but that there is not a closed list of factors to be considered given that each rebate might have its peculiarities, but identifies some particular criteria, namely (a) the “criteria and rules governing the grant of the rebate” (paras. 36-41: referring to loyalty building/suction effects, which depend inter alia on retroactivity, the volume and time-span of the rebate, as well as intent -the latter is referred to as a “strong additional indication” as opposed to a “mandatory precondition”-; she also adds that the charging of “negative prices” should not be a precondition either (para. 41); and (b) the conditions of competition in the market and the position of the dominant company (see paras. 42-50, very closely linked to the facts of the case). She sums all this up in a “interim conclusion” (para. 55) stating that a rebate scheme operated by a dominant company will be abusive “where an overall assessment of all the circumstances of the individual case shows that the rebates are capable of producing an economically unjustified exclusionary effect, it being important to take into account in that regard, in particular, the criteria and rules governing the grant of the rebate, the conditions of competition prevailing on the relevant market and the position of the dominant undertaking on that market”. Nothing groundbreaking or too controversial here.

From para. 57 onwards she refers to the Commission’s Guidance on exclusionary abuses and its endorsement of the “as efficient competitor test” that the Institution imposed upon itself. She notes that such an administrative practice “is not, of course, binding on the national competition authorities and Courts”. Importantly, she says that “although the national authorities themselves are not precluded from following the Commission’s example and using the AEC test, they are none the less, from a legal point of view, bound only by the requirements arising from Article [102]” and that “[i]t is for the Court to define what those requirements are”. This isn’t groundbreaking at all either, but some might consider it controversial.

It is at this point that the most relevant stuff comes. In para. 61 the Opinion observes that Article 102 does not support the inference of any legal obligation requiring the use of the AEC test. It then observes that in previous cases (Telia Sonera or Post Danmark I), the ECJ has validated this test but not as an “absolute requirement” for all price-related cases. The AG remarks, first, that the said-case law is specifically concerned with other pricing practices that are by their nature closely related to the cost structure of undertakings and also, second, that the wording used by the Court in those cases made it clear that anticompetitive exclusion is not only that which affects equally efficient competitors (62-63).

With regard to rebates in particular the Opinion refers to the ECJ’s Judgment in Tomra (para. 92 later mentions that Tomra was rendered “at about the same time” as Post Danmark I) to support the contention that a cost-price assessment is not mandatory. Although she contemplates the possibility of establishing this requirement, she expresses “skepticism” towards any reorientation of the law (65) given that (i) “the added value of expensive economic analyses is not always apparent and can lead to the disproportionate use of resources” [economist will love this..] (66); (ii) “it is wrong to suppose that the issue of price-based exclusionary effects can be managed simply and in such a way as to ensure legal certainty by applying some form of mathematical formula based on nothing more than [business data] not uncommonly open to different interpretations” (67); and (iii) “the finding of an abuse requires taking into account all the relevant circumstances of the individual case in question and must not be confined to an examination of price and cost components alone” (68).

In para. 69 the Opinion explains that taking into account all circumstances + considering whether there is any objective justification for the rebate “adequately ensures that the legal requirements (…) do not disregard economic realities”.

Paras. 71 to 75 then develop some further objections to the AEC test, notably regarding the fact that when a dominant company is present the structure of the market often rules out the presence of equally efficient competitors (due e.g. to barriers to entry, economies of scale or network effects) which implies that “the competitive pressure exerted by less efficient undertakings must not be underestimated.

In the light of the above, Kokkot’s recommendation for the Court in para. 75 is to respond that Article 102 does not require the abusive nature of rebates to be established pursuant to an AEC test, but that national authorities and Courts are at liberty to avail themselves of a price/cost analysis unless, on account of the circumstances, it would be impossible for another undertaking to be as efficient as the dominant one.

Finally, the Opinion addresses the question about how “likely and serious” the exclusionary effect must be in order for Art. 102 to apply.  With regard to likelihood, it states that “hypothetical effects” are not enough because the rebate must be capable “not only in the abstract but also in practice of making it difficult or impossible for the dominant undertaking’s competitors to gain access to the market”; in its view, the provision is triggered in the face of “likely” effects, not of “very likely” or ·particularly likely” or “beyond reasonable doubt”. At most, the Opinion explains, the degree of likelihood may have a bearing on sanctions. With regard to seriousness (appreciability) she first observes that the doubts of the Danish Court may have to do with a deficient translation of the Judgment in Post Danmark I from French to Danish (the latter version referred to appreciable effects/elimination effects instead of exclusionary effects). In her view, likely exclusionary effects are enough, there not being a need to qualify it those effects as serious or appreciable; the Opinion then cites Tomra for support, and adds that a de minimis threshold doesn’t seem necessary given that there will already be a an assessment of all relevant circumstances and also given the fact that Art. 102 extends only to conduct that is likely to affect trade between Member States [I personally don’t think that this latter argument is valid, for the effect on competition and on trade between Member States are two different things assessed pursuant to different criteria; this, in my view, is quite clear in the case law on 101]

This very last section of the Opinion is what I find less satisfactory (many people will probably take issue with the previous stuff too) because it leaves a question unaddressed (in its defense, one that was not posed directly in the case, and one that I think is at the root of most major current substantive discussions: what is really anticompetitive exclusion/foreclosure? when is it enough to warrant intervention? is it about making life more difficult to competitors –and how much more?- or about their elimination –and to what extent-?) I’m not sure that the argument that “we will know after considering all circumstances” is enough. In practice the issue if often solved by prosecutorial discretion (the EC at least has chosen well its cases) but, query, is that the appropriate solution? Perhaps the question is not so relevant for loyalty rebates since –according to Michelin and Intel –the only two Judgments that, unless I’m wrong, contain the expression- they are considered restrictive “by object” (pending the objective justification assessment), but it is the key question to every other practices assessed under 102.

As for the rest of the Opinion, I think there is nothing new; it fits within the line of the established and controverted case-law on the issue that we have extensively discussed here. I suspect that (i) people with strong views on either sides will regard this Opinion as a lost opportunity for very different reasons; (ii) the ECJ is likely to endorse this view; and (iii) I also suspect AG Wahl might take a different view when he writes his Opinion in Intel. And speaking of AG Wahl:


AG Wahl on AC-Treuhand (or what is a restriction of competition?)

The second Opinion rendered today concerns a novel issue which AG Wahl proposes to address by returning to the fundamental –and unclear– concept of restriction of competition.

The case concerns an appeal against the General Court Judgment endorsing the Decision which –for the first time- sanctioned a company for its role as a “cartel facilitator” despite not being a player in the affected markets. In essence, the company’s role consisted in arranging and participating in meetings, gathering and circulating data, moderating tensions and fostering commitments in exchange for a remuneration.

The ground of appeal that is dealt with in the Decision raised two interesting questions, namely: (i) does Article 101 encompass this sort of conduct?; and (ii) subsidiarily, could the company be sanctioned in a manner compliant with the principle of legality considering that there was no previous case-law establishing that such conduct fell within the scope of Article 101?

In the view of AG Wahl, “in order to identify a restriction of competition it must be shown, following the pertinent economic analysis, [intermission, note the difference in the language compared to the previous commented Opinion] that the company at issue has renounced, totally or partially, by its conduct, to exert a pressure characteristic of effective competition on the rest of the operators in the market or markets affected to the prejudice of economic efficiency and consumer welfare” (para. 1, later paraphrased at various key paragraphs of the Opinion, notably 47, 50, 51, 62 and 69). In the light of this notion of restriction, and considering that AC Treuhand did not exert any competitive pressure on the other participants in the cartel prior to the agreement, it never ceased exerting any such pressure and therefore, according to AG Wahl, cannot be held directly responsible for the cartel. Consequently, he recommends the ECJ to annul the General Court’s Judgment.

I see the point, but at the same time I have doubts: didn’t the company participate in an agreement that had as its object the restriction of competition? Also, it is true that a wide interpretation of Art 101 to capture facilitators could potentially extend even to lawyers not doing their job properly; at the same time, however, organizing cartels should probably not be a legitimate business.
Btw, the  notion of restriction used here –despite the reference to economic analysis- seems close to that often criticized as ordoliberal; I’m not saying this pejoratively, I’m simply observing it.

According to the Opinion – which in para. 71 is quite blunt- if the Court were to endorse the view of the General Court and of the Commission, it would “profoundly disturb” the method of identification of anticompetitive conduct by disconnecting the conduct and the economic restriction in such a way that the definition of the relevant market and the identification of anticompetitive constraints therein would become completely superfluous. Again, I see AG Wahl’s major point, but I’m not really persuaded by the latter part of this particular argument, for market definition is already deemed superfluous when it comes to cartels…

The Opinion then goes on to consider the theoretical question of whether the company could be sanctioned as an “accomplice” (paras. 77-83). It notes that whereas this would seem convincing at first sight, the charges were not framed in that sense and, moreover, the concept of “accomplice” belongs to criminal law and is alien to administrative law, so resorting to it in a case like this would not make sense (para. 82).

In AG Wahl’s view, it is exclusively for the legislator to foresee a sanction for accomplices under EU Law. After stating this, at the very end of the Opinion, he sends a clear message that I, for one, am likely to quote in the future-: it is necessary to underline that the will of the Institutions of safeguarding the effectiveness of their policies must be conciliated with legality and legal certainty. As pointed out by an author (a footnote clarifies that the “author” is Pierre Pescatore) the effectiveness -effet utile- doctrine cannot lead the Court of Justice to interpret Treaty provisions so as to extend to the maximum the competences of the Institutions, but must permit to interpret the pertinent rules in the light of their objective and goal”. (Note that an English version of the Opinion is not yet available; this is my own translation).

The second question raised by the applicants was, in my view, equally interesting, but was not addressed in the Opinion (although I predict that it may be more relevant to the eventual Judgment…). Could the company be sanctioned for acting as a facilitator when the law was unclear –there was no precedent- as to whether it violated Art. 101? In practice the Commission has sometimes decided not to sanction a company resorting to this reasoning but it has done so on its own motion (see here). However, is there a legal obligation for the lege to be clear for the poena to be imposed? This is a question –or rather a problem- that, in reality, concerns not only this issue but the whole of competition law (with the exception of cartels, or at least of how the term “cartel” was traditionally understood). I will recall the answer that the General Court gave to an argument that also concerned the principle of legallity in Case T-167/08, Microsoft (compliance):

  1. “(…)the use of imprecise legal concepts within a provision does not prevent liability being established as against a person who contravenes it. As the Commission points out, if it were otherwise, an infringement of Article 101 or 102 TFEU – which are themselves drawn up using imprecise legal concepts, such as distortion of competition or ‘abuse’ of a dominant position – could not give rise to a fine without the prior adoption of a decision establishing the infringement“. 

I guess that says a lot about our discipline…

Written by Alfonso Lamadrid

21 May 2015 at 6:53 pm

Posted in Case-Law, Uncategorized

7 Responses

Subscribe to comments with RSS.

  1. Dear Alfonso, AG Wahl is shaking the grounds of yet another well-established habit of the public enforcement agencies in EU – ”for market definition is already deemed superfluous when it comes to cartels…”.
    This is precisely where the Opinion is ”revolutionnary”. It says that there is no competition infringement in the absence of competition, which means the parties must be active in the same relevant market, in the same time. Pretty obvious but not to the enforcement agencies, which tend to take the loose requirement – ”the relevant market does not have to be identified exactly” – as the absence of such a requirement.
    It happens in Romania and I guess it does elswhere in EU: for instance we have a decision which sanctions companies for participating in a single infringement, although the Romanian Competition Council acknowledges that the undertakings were not competing in the same market (but each was competing with another one, in distinct relevant markets).


    22 May 2015 at 8:24 am

  2. Just read both opinions.
    With all due respect, I do not find either of them persuasive. Too little economic thinking in the first, too much in the second.
    Kokott’s opinion is basically a lot of … ‘the same old stuff’. It could have been written in the 80ies. I am not sure it reflects current/modern economic and legal thinking. It is very much orthodoxy covered with new clothes.
    Wahl’s opinion is equally poor. As you noticed, it is extremely brief and blunt, but lacks any real legal explanation for his theory. He describes what the world should, in his personal view, be. But forgets to deal with the very text of the Treaty provision at stake. There may be some ordoliberal economic vision behind his views, but there is certainly no robust legal analysis to support it.
    Two missed opportunities. Pity.

    Jozsef (ELR)

    22 May 2015 at 8:41 am

  3. […] nog blijken. Op zichzelf voert AG Wahl valide argumenten aan, echter die kunnen worden weerlegd. In de praktijk wordt daarom zeer betwijfeld of het Hof van Justitie van de EU deze Opinie zal volgen. In het […]

  4. Jozsef: you are right it cannot reflect modern econ and legal thinking, which is said at §4 to be “Zeitgeist”.
    I have a less dire reading of the Opinion of Ms Kokkott though, because it feels it cuts halfway through more econ and existing, more formal approaches. My beef goes at the inconsistency/flaw planted at the beginning and end of Opinion. If AEC test is to be prohibited (§74) when structure makes that no rival can be equally efficient (which can be called the returns to scale hypothesis) how then can it be that returns to scale can constitute objective justification (§29 amongst other)? How can this defense work in practice? You need price cost analysis at defense stage when burden of proof is on domco, but there is no price cost analysis at offense stage when burden of proof is on agency? In other words, domco must argue efficiencies in the dark? And is it entirely convincing conceptually? I mean this comes close to saying that efficiency plays as a quasi-offense (strict test, no AEC) and can be a defense at the same time.

    Nicolas Petit

    22 May 2015 at 7:53 pm

  5. I leave the “Zeitgeist Opinion” for another occasion… With regard to AG Wahl’s Opinion, it may, indeed, seem somewhat surprising that the AG is proposing an analysis that may not be very friendly to the principle of effectiveness of the prohibition of Article 101 TFEU. However, in reality, the AG is simply introducing a question of competence, legality and proper scope of a Treaty provision. There is no doubt that you may have accomplices that help cartels and other anti-competitive agreements and structures. There is also no doubt that there will be legal consequences of civil and criminal law nature for such accomplices. Their contracts will be invalid, they may also be liable to damages, there may be criminal liability, etc. Of course, such civil and criminal sanctions in national law will have to take duly into account the principle of effectiveness of EU law. However, it is quite a different issue if the scope of application of Article 101 TFEU, as interpreted by the 50-year old case law, can catch such behaviour, a behaviour outside a “market”, by a non-market participant, which does not compete himself, as the other undertakings do (the parties to the illegal behaviour)! Should we be extending the scope of application of the Treaty to such undertakings and make their conduct the subject of also public / administrative (or indeed quasi-criminal) sanctions at supranational level? This is the question for the Court of Justice. It will be interesting to see what its response will be.

    Makis Komninos

    23 May 2015 at 8:14 pm

  6. I very much agree with Nicolas on Kokott’s conclusions and with Josef regarding Wahl’s conclusions. I have in cery high esteem both members but this time they performed poorly

    Pang Pang

    24 May 2015 at 6:17 pm

  7. […] that earlier this year we commented on AG Wahl’s Opinion in AC-Treuhand (C-194/14 P) (see here) and anticipated that, in spite of its thought provoking reasoning, it was likely not to be […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: