Dec 19, 2014
CPI Europe Column edited by Anna Tzanaki (Competition Policy International) presents:
Intro by Anna Tzanaki (Competition Policy International)
The December edition of the Europe Column is dedicated to two quite important ECJ judgments delivered on the very same day and dealing with Article 101. After presenting the MasterCard ruling and its implications for efficiency claims in two-sided markets, we are very pleased to host James Killick and Jeremie Jourdan (White & Case) providing a thorough comment on the Cartes Bancaires judgment.
Admittedly, the latter judgment has much wider repercussions, going far beyond the specific context of multi-sided market settings. As the authors eloquently explain, the ECJ in Cartes Bancaires not only warned the Commission of using the “by object” restriction characterization for anticompetitive practices restrictively but also warned the General Court against any superficial analysis, rather than an in-depth judicial review. On this positive note, we wish everyone a wonderful festive season and a Happy New Year!
On 11 September 2014, the Court of Justice rendered the much awaited Cartes Bancaires judgment.2 The Court of Justice (“ECJ”) quashed the General Court (“GC”) judgment for failing to correctly apply the notion of restriction by object. The significance of the judgment is debated, some qualify it as groundbreaking, others note that it simply takes us back to the position first laid out almost 50 years ago, namely that by object restrictions must be obviously bad. In any case, the judgment has had the welcome side–effect of generating an intense debate around a notion – restriction by object –, which is surprisingly ill-defined 57 years after it was engraved in the EU Treaty but which is a key element in the EU system of competition enforcement.
Either way, the judgment is clearly a significant one given that it both clarifies the notion of restriction by object and, equally importantly, firmly requires the GC to undertake an in-depth review of Commission decisions.
Where we’re coming from: the pre-Cartes Bancaires case law
The case law setting the foundations for the notion of restriction by object echoed the simple idea that a restriction by object must be obvious and “sufficiently bad.” Back in 1966, the Court of Justice in STM explained that the clauses of an agreement that restricts competition by object must “reveal the effect on competition to be sufficiently deleterious.”3Since the early days, the Court said that to determine whether an agreement may be qualified as a restriction by object, an analysis of the content of its provisions, its objectives and the economic and legal context is required.4 The bottom line has always been that only the worst agreements, which we know are bad, are by object restrictions. To put it in plain language: by object infringements are akin to drunk driving, we know it is bad so we punish those who do it even if they have not caused harm.5
It is however fair to say that it is post-modernization that the question of the boundaries between by object and effect restrictions became more topical. One element driving this up the agenda was the feeling that effects cases required more effort by enforcers and took significantly longer than object cases. In 2008, the Court stated in Beef Industry that the distinction between an infringement by object and by effect arises from the fact that “certain forms of collusion between undertakings can be regarded, by their very nature, as being injurious to the proper functioning of normal competition.”6 This definition was endorsed in several judgments since then.7
Other than the obvious price fixing, market sharing and output restriction, restrictions on parallel trade were qualified as restrictions by object. Thus, in GSK, the Court ruled that restrictions to parallel trading in the pharmaceutical sector constituted such restrictions.8 In FA Premier League, the Court found that clauses prohibiting the resale of decoders outside a given territory amounted to a restriction of competition by object.9
Two judgments, T-Mobile (information exchange) and Pierre Fabre (selective distribution), however cast some doubts on the applicable standard. In T-Mobile, while also referring to the above statement from Beef Industry the Court stated that “for a concerted practice to be regarded as having an anti‑competitive object, it is sufficient that it has the potential to have a negative impact on competition.”19 In Pierre Fabre the Court held that “agreements constituting a selective distribution system … necessarily affect competition in the common market [and] are to be considered, in the absence of objective justification, as ‘restrictions by object’.”11 In each of these statements from the Court, what seems to be missing is the notion that a restriction by object must be limited to conduct that is “by its very nature injurious” to competition or “sufficiently deleterious.”
The more recent Allianz Hungária judgment12 (vertical agreements for insurance) arguably added to the confusion. The Court of Justice referred to both Beef Industry and T-Mobile and the need to analyze the content, objectives and context of agreements to determine whether there are restrictions by object. However, Allianz Hungária added that an agreement would also amount to a restriction of competition by object if “it is likely that, having regard to the economic context, competition on that market would be eliminated or seriously weakened following the conclusion of those agreements.”13 Some commentators criticized the judgment for requiring an effects analysis to qualify an agreement as a restriction by object, thus blurring the distinction between the two.14 However, this statement is not necessarily inconsistent with STM and Beef Industry: if it is shown, through a contextual analysis that an agreement is likely to be very bad (i.e. very harmful to competition), then one can conclude that it is sufficiently bad to be a restriction by object.
In short, the Cartes Bancaires judgment came in a context of relative confusion about the notion of restriction by object. The old cases set principles that were relatively clear, but some sentences in the more recent case law could be read as suggesting a different approach.
The facts of Cartes Bancaires
The appellant, Groupement des cartes bancaires, was an economic interest grouping created by the main French banking institutions to achieve interoperability of systems for payment and withdrawal by bank cards issued by its members. The system enabled users to make payments with bank cards to all affiliated merchants and withdrawals from ATMs belonging to any of the group members. To avoid free riding (i.e. by banks which issued lots of cards but did not do any work to ensure that merchants accepted the cards) and ensure interoperability, the group notified some new pricing measures, which included three new mechanisms, to the Commission:
- A device known as MERFA (mechanism for regulating the acquiring function), which encouraged issuers to expand their acquisition activities;
- The establishment of a three-year membership fee per active card issued; and
- A mechanism known as a “dormant member wake-up mechanism,” which consisted in a fee per CB card issued, applicable to members that were inactive or not very active before the entry into force of the measures.
The Commission had first considered the measures as a secret cartel between the banks. Following a hearing “conducted in an atmosphere of antagonism,”15 the Commission changed its approach and prohibited the measures under Article 101 TFEU as decisions by an association of undertakings, which restricted competition by object and effect. The General Court upheld the Commission decision as to the object conclusion and did not consider it necessary to examine whether the qualification of restriction by effect was well-founded.16
The search for principles in the opinion of AG Wahl
On 27 March 2014, AG Wahl delivered his opinion proposing to the Court of Justice to set aside the judgment of the General Court in its entirety.17
In its opinion, AG Wahl embarks on a useful expedition to find the roots of the notion of restriction by object and its underlining theoretical principles. He reminds us of the reasons why Article 101 TFEU provides for the possibility to find that a certain conduct restricts competition by object without analyzing its effects: first, to provide predictability and legal certainty to undertakings; second, to deter; third, to achieve procedural economy.18
AG Wahl analyzed the case law and called for a “restrictive interpretation” of, and “cautious attitude” towards, the notion of restriction by object.19 In his opinion, only conduct that (i) entails an inherent risk of a particularly serious harmful effect or (ii) in respect of which it can be concluded that the unfavorable effects on competition outweigh the pro-competitive effects can be considered as by object restriction.20 He thus advocated that: “Only conduct whose harmful nature is proven and easily identifiable, in the light of experience and economics, should therefore be regarded as a restriction of competition by object, and not agreements which, having regard to their context, have ambivalent effects on the market or which produce ancillary restrictive effects necessary for the pursuit of a main objective which does not restrict competition.”21
Also noticeable in AG Wahl’s opinion is the proposition that the context analysis can only be exculpatory for the parties.22 In other words, if the harmful nature of the agreement isn’t clear from the content or the form of the agreement, it should not be qualified as by object restriction. By contrast, an agreement that looks like a by object restriction could still be saved by a contextual analysis (e.g. taking account of the two-sided nature of the market in this case).
Beyond these general considerations, AG Wahl went on to analyze whether the measures at issue constituted a restriction by object. The General Court had considered that the anticompetitive object of the measures, apparent from the calculation formulas, was to impede competition from new entrants on the French market for the issue of cards belonging to any of the group members. However, in the eyes of AG Wahl, both the General Court and the Commission failed to demonstrate how, by their very wording, these measures restricted competition.23 The object of the measures was to require financial contribution from its members to finance the operating costs of the payment system and to avoid free riding. The fact that some members may be incentivized to change their activities or to bear additional costs does not constitute a restriction by object. This question would, if anything, come under examination of the potentially anticompetitive effects of these measures and not of their object.24 Finally, when it came to the comments made by members in preparatory documents, these were simply “not directly relevant.”25
Moving on to the assessment of the legal and economic context, AG Wahl explained that the General Court should have ascertained how the measures at issue (and unlike other measures in the same market) presented a degree of harm such that they could be considered to have an anticompetitive object.26 In doing do, the General Court failed to have regard to the double-sided nature of the market and thus to take into account the contribution of the system on both the issuing and acquiring markets.27
The Cartes Bancaires judgment: reminding us firmly of what should never have been forgotten
The ECJ set aside the judgment of the GC in its entirety. It is the first time that the ECJ sets aside a Commission decision on the definition of restriction by object.
The first important point about the judgment is that it relies on the well-established line of case law starting with STM, and including BIDS and Allianz Hungária (except, as to the latter, for the controversial §48 referred to above). It thus reconfirms the longstanding orthodoxy that a restriction by object must display a “sufficient degree of harm” to competition and include only these types of coordination between undertakings that “can be regarded by their very nature as being harmful” to competition.28 The judgment does not follow the lower standards of T-Mobile and Pierre Fabre. In fact, the Court finds that the GC erred by relying on T-Mobile and not on well-established case law.29 So the sentences in these two judgments that suggested a lower standard for restrictions by object must be considered to be rejected by the ECJ.
The second noticeable point is the ECJ’s explicit reference to the notion of experience. While this statement is made in connection to price fixing,30 it seems to be an endorsement of the relevance of experience (as suggested by AG Wahl) in determining whether a conduct should be qualified as by object restriction. It is the first time that this element features in a Court’s judgment, although it appeared in a number of AG opinions31 and in the Commission’s own guidelines.32
The third noticeable point – and perhaps the most important – is the idea, conveyed through a double negative, that the notion of restriction by object should be interpreted narrowly.33 While the idea had already surfaced in some opinions from Advocate Generals,34 it is the first time the Court states it as clearly.
In light of these guiding principles, the ECJ then assessed whether the GC was right to consider the measures at issue as a restriction by object. Similarly to AG Wahl, the ECJ held that the anticompetitive object of the measures could not be inferred from the wording of the pricing measures and that the GC did not at any point explain in what respect that wording revealed a sufficient degree of harm to competition to amount to a restriction by object.35 The GC had found that the pricing formulas, by imposing on certain members restrictions on the issuance of cards or costs not borne by other members, limited the possibility for such members to compete on prices. This, however, merely explained why the formulas were capable of restricting competition (and thus capable of falling within the scope of Article 101(1)) but the GC had not explained why that restriction of competition revealed a sufficient degree of harm in order to be characterized as a restriction by object.36 According to the ECJ, the GC “was entitled at the most to infer from this that those measures had as their object the imposition of a financial contribution on the members of the Grouping which benefit from the efforts of other members for the purposes of developing the acquisition activities of the system.” This could not be viewed as by nature anticompetitive and indeed the GC itself had found that combating free-riding in the CB system was a legitimate objective.37
The ECJ also criticized the GC judgment in several other respects:
- first, it found that the GC erred in the assessment of the context of the measures, in particular by disregarding considerations that the market at issue was two-sided;<sup38
- second, it found that while it admittedly could not be ruled out that the measures may hinder competition from new entrants and even lead to their exclusion from the system on the basis of the fees charged pursuant to those measures, such a finding fell within the examination of the effects of those measures on competition and not of their object;39
- third, it found that the GC wrongly applied the BIDS case law by analogy whereas the constituent elements of the restriction by object in that case were missing in the measures at issue;40
- finally, the ECJ confirmed that the intentions of the grouping could not in themselves lead to a characterization of the measures as by object restriction.41
The conclusion of the ECJ is a rather severe one. It found that the errors of law committed by the General Court indicated “a general failure of analysis by the General Court and therefore reveal the lack of a full and detailed examination of the arguments of the appellant and of the parties.”42 The GC is criticized for “simply reproducing on a number of occasions [..] the contents of the decision at issue,” thereby failing to review “whether the evidence used by the Commission in the decision at issue enabled it correctly to conclude that the measures at issue [..] displayed a sufficient degree of harm to competition to be regarded as having as their object a restriction of competition.”43 The ECJ (like AG Wahl) recorded the fact that the GC had failed to take any account of the “radical change” in the Commission’s position after the stormy first hearing.44
Lessons and consequences
The Cartes Bancaires judgment is clearly important. We can observe this in two distinct but related aspects.
First, on the issue of judicial review: Since the ECJ clearly stated the standard of judicial review in Chalkor and KME judgments,45 it is the first judgment in which a GC judgment is quashed for failing to meet such standard. This will likely have rippling effects on the intensity of the GC’s judicial review of Commission decisions – which chamber will want to see its judgment qualified as a “general failure of analysis”? – and consequently on the thoroughness of Commission decisions. This is a most welcome development to strengthen a system whose legitimacy today essentially rests on the intensity of the GC’s judicial review.
Second, Cartes Bancaires will likely have an impact on the Commission’s (and NCAs’) approach to cases under Article 101 TFEU. The judgment should have limited or no impact on classic restrictions of competition, such as cartels. It is long accepted (i.e. there is much experience) that price fixing, output restriction and market allocation cartels are harmful to competition and ought to be considered as by object restrictions. The bulk of NCA and Commission enforcement would therefore be unaffected.
But there will be important implications for agreements or concerted practices, which do not fall in that category. Can they still be viewed, the first time they are examined by the Commission, as restrictions by object? Or should the Commission establish, at previous cases, that such agreements or concerted practices have such anticompetitive effects that in the future they can be qualified as by object restrictions?
By referring to “experience” in Cartes Bancaires (§51), the Court seems to indicate that some past practice or experience is necessary. In short, a by object characterization must not come out of the blue.
Linked to this is the question of obviousness. In Fordham last year, DG A. Italianer said that “restrictions by object are serious – but not necessarily obvious.”46The question of obviousness is two-sided: obviousness to the enforcer and to the parties. A cartel dressed up as a legitimate agreement is still a cartel. It might not be obvious to the enforcer – indeed one would expect cartel members to disguise their practice into something else to avoid been caught – but this should not affect the legal characterization: the latter depends on the economic reality of the practice, not its appearance. What really matters is obviousness to the parties. A company engaging into a restriction by object must know that what it is doing will almost inevitably restrict competition. Indeed, it is supposed to be its very purpose. A restriction by object is judged on an ex ante basis: it is illegal no matter that it has no impact or effect. Indeed, given the difficulty of invoking Article 101(3), a by object infringement is close to per se illegality.47 Companies must be in a position to know that what they are doing is quasi-criminal on the day they sign an agreement. In short, if it ain’t obvious (to the parties), it ain’t by object. But if a serious infringement is not obvious to the enforcer because it has been disguised, it can still be an infringement by object.
The Court’s clear guidance that the notion of by object restriction must be interpreted restrictively goes in the same direction. This judgment should put an end to what has been perceived as a creeping expansion of the concept of restriction by object in the name of procedural economy. Looking backward, one may wonder whether past antitrust cases (excluding cartels) would have been decided differently, had Cartes Bancaires come out a decade before. Consider, first, commitments cases: Airlines alliances, E-Books, Siemens/Areva, Visa Debit, Visa Credit48 which were brought under the by object umbrella. Would the parties have settled in the same way if the Commission had been obliged to develop an effects analysis? (Indeed, would the Commission have indicated serious doubts in all those cases if it had been guided by Cartes Bancaires?). Next, prohibition decisions in non-cartel cases, of which there were only five in the last 5 years: LABCO, Telefonica/Portugal, Lundbeck, Fentanyl, Perindopril.<sup.49 The LABCO decision was recently confirmed. The GC considered that the evidence put forward by the Commission demonstrated that the measures in question were akin to horizontal price fixing.50 The case thus seems to be in line with Cartes Bancaires, although the GC oddly does not refer to it in its judgment. It is interesting to reflect on whether the other cases, notably the pharmaceutical cases involving settlement agreements, will pass the “restrictive interpretation” and “by nature harmful” requirements of Cartes Bancaires. Likewise, ongoing cases such as the CDS investigation,51 the container liner shipping case,52 the remaining patent settlement cases (Teva/Cephalon)53 and the benchmark manipulation cases (LIBOR, EURIBOR)54 may all be impacted in different ways by such requirements.
One final, important point: a sound competition policy must avoid false positives. Because actual effects are not considered, the risk of false positives is higher – comparatively – when the enforcer condemns practices by object. It is crucially important that an agreement or a concerted practice is not condemned by object if, at the time it is entered into, its potential effects are ambivalent – i.e. it can have negative effects, but also be neutral to competition or even enhance competition depending on future events. To refer back to the driving analogy: the act of driving to an off-license to buy a bottle of vodka cannot be prohibited ex ante. The per se violation is driving after drinking the bottle of vodka; if the bottle is not drunk, there is no violation.
Cartes Bancaires seems to close the avenues for conduct having ambivalent effects to be caught in the by object box because it reaffirms that only agreements by their nature harmful to competition can be qualified as by object restrictions. That is the only conclusion that fits with the quasi-criminal nature of EU competition law. It is also in line with the case law of the Courts dating back 50 years. That is the final conclusion of Cartes Bancaires: such ancient wisdom should not be lightly discarded.
1 James Killick and Jeremie Jourdan. The authors are, respectively, Partner and Local Partner at White & Case, Brussels. The views expressed in this article are personal to the authors and do not necessarily reflect the views of the firm or of any of its clients.
2 Case C-67/13 P, Groupement des cartes bancaires v Commission, EU:C:2014:2204.
3 Case C-56/65, Société Technique Minière (STM), EU:C:1966:38, page 249.
4 Joined Cases 96/82 to 102/82, 104/82, 105/82, 108/82 and 110/82, IAZ International Belgium, EU:C:1983:310, §25.
5 See Opinion of AG Kokott in Case C-8/08, T-Mobile, EU:C:2009:110, §47.
6 Case C-209/07, Beef Industry, EU:C:2008:643.
7 Case C‑226/11, Expedia, EU:C:2012:795, §36; Case C-32/11, Allianz Hungária, EU:C:2013:160, §35; Case C-8/08, T-Mobile, EU:C:2009:343, §29.
8 Case C-501/06 P, GlaxoSmithKline, EU:C:2009:610, §§54-67.
9 Case C-403/08, Football Association Premier League, EU:C:2011:631, §§134-146.
10 Case C-8/08, T-Mobile, EU:C:2009:343, §31.
11 Case C‑439/09, Pierre Fabre, EU:C:2011:649, §39.
12 Case C-32/11, Allianz Hungária, EU:C:2013:160.
13 Ibid, §48.
14 See G. Cosmo, “Methods for Determining whether an Agreement Restricts Competition: Comment on Allianz Hungária,” European Law Review 2013, pp.542-551; D. Harrison, “The Allianz Hungária case – The ECJ’s judgment could have ugly consequences,” Competition Law Insight, (2013) 12(6), pp.10-12; C.I. Nagy, “The Distinction between Anti-competitive Object and Effect in Allianz: The End of Coherence in Competition Analysis?,” World Competition, (2013) 36(4), pp.542-564; for a more general reference see Opinion of AG Wahl in Case C‑67/13 P, Cartes Bancaires, EU:C:2014:1958, §52.
15 Final report of the Hearing Officer in Case COMP/38.606 — Groupement des cartes bancaires, OJ L 162, 19.6.2001, p. 21. Interestingly, the report explains that the Hearing Officer recommended to Commissioner Kroes that an internal panel be held on the case, which led to the abandonment of the secret cartel theory.
16 Case T-491/07, Cartes Bancaires, EU:T:2012:633.
17 Opinion of AG Wahl in Case C‑67/13 P, Cartes Bancaires, EU:C:2014:1958.
18 Ibid, §35.
19 Ibid, §59; §74.
20 Ibid, §55.
21 Ibid, §56.
22 Ibid, §§44-45.
23 Ibid, §95.
24 Ibid, §95 and §§130-131.
25 Ibid, §110.
26 Ibid, §144.
27 Ibid, §§146-150.
28 Case C-67/13 P, Cartes Bancaires, EU:C:2014:2204, §50.
29 Ibid, §§56-57.
30 §51 of the judgment reads: “Experience shows that such behaviour [price fixing] leads to falls in production and price increases, resulting in poor allocation of resources to the detriment, in particular, of consumers.”
31 Opinion of AG Cruz Villalón in Case C-32/11, Allianz Hungária, EU:C:2012:663, §65. 32 Communication from the Commission: Guidelines on the application of Article 81(3) of the Treaty, OJ No C 101 of 27.4.2004, §21.
33 Case C-67/13 P, Cartes Bancaires, EU:C:2014:2204, §58: “the General Court erred in finding, in paragraph 124 of the judgment under appeal, and then in paragraph 146 of that judgment, that the concept of restriction of competition by ‘object’ must not be interpreted ‘restrictively’.”
34 Opinion of AG Kokott in Case C-8/08, T-Mobile, EU:C:2009:110, §44; Opinion of AG Cruz Villalón in Case C-32/11, Allianz Hungária, EU:C:2012:663, §65.
35 Case C-67/13 P, Cartes Bancaires, EU:C:2014:2204, §58 and §65.
36 Ibid, §§66-69.
37 Ibid, §75.
38 Ibid, §§76-79.
39 Ibid, §§80-81.
40 Ibid, §§83-87.
41 Ibid, §88.
42 Ibid, §89.
43 Ibid, §90.
44 Ibid, §17.
45 Case C‑386/10 P, Chalkor, EU:C:2011:815 and Case C-389/10 P, KME, EU:C:2011:816.
46 A. Italianer speech at the 40th Annual Conference on International Antitrust Law and Policy, Fordham Competition Law Institute New York, 26 September 2013: http://ec.europa.eu/competition/speeches/text/sp2013_07_en.pdf.
47 Indeed, one might argue that the increased use of by object analysis for novel infringements combined with the difficulty of invoking Article 101(3) arguably made the enforcement system unbalanced when it came to new infringements, which were too prone to being found illegal despite potential pro-competitive aspects (i.e. a problem of false positives).
48 Case COMP/39596 – Oneworld; Case COMP/39398 – Visa Debit; Case COMP/39736 – Siemens/Areva; Case COMP/39847 – E-books; Case COMP/39595 – Star; Case COMP/39398 – Visa Credit.
49 Case 39510 – LABCO/ONP; Case COMP/39839 – Telefonica/Portugal Telecom; Case COMP/39226 – Lundbeck; Case COMP/39685 – Fentanyl and Case COMP/39612 – Perindopril.
50 Case T-90/11, ONP, EU:T:2014:1049, §§321-322.
51 Case COMP/39745 – CDS – Information market.
52 Case COMP/39850 – Container Shipping.
53 Case COMP/39686 – Cephalon.
54 Case COMP/39861 – Yen Interest Rate Derivatives; Case COMP/39914 – Euro Interest Rate Derivatives.