Putting Limits on Judicial Creativity: AG Kokott’s Opinion in Commission v CK Telecoms

Keywords: SIEC – non-collusive oligopoly gap – standard of proof – Advocate General – UPP test – horizontal mergers.

On 20 October 2022, Advocate General (AG) Kokott delivered her Opinion in Commission v CK Telecoms.[1] This constitutes a major recent development in the area of merger control.[2] The case provided her with a unique opportunity to shed light on the concept of Significant Impediment to Effective Competition (SIEC) in the context of a concentration that gave rise to non-coordinated effects without creating or strengthening a dominant position. Moreover, the AG encourages the Court to settle long-standing debates on the matter of the relevant test for the standard of proof and the scope of judicial review in merger control. Siding with the European Commission (EC or the Commission), the AG proposed setting the decision[3] of the General Court (GC) aside in its entirety and referring the matter back to that court. The case at hand has become a somewhat violent point of contention for two opposing visions of antitrust, one encapsulated in the GC decision and the other in the AG’s opinion. While the GC’s general policy seems to consist in curtailing the Commission’s discretion, as the Commission would otherwise “systematically prohibit all horizontal concentrations in an oligopolistic market”,[4] the AG warns against “underestimating from the outset the competitive forces present within an already concentrated oligopolistic market”.[5] The case comes at a time when both the U.S. and the EU are witnessing a steady rise in industry concentration,[6] as cries for equipping competition agencies with more power to effectively halt such trends grow louder. While some commentators welcome greater recourse to structural presumptions in an effort to reinvigorate antitrust,[7] others warn against adopting a standard of proof that is too lax which would result in over-enforcement. As the European Court of Justice (ECJ) finds itself at the crossroads, it remains open whether the AG’s opinion will tilt the balance in the Commission’s favour.

The case concerned the legality of a four-to-three merger between British mobile telephony operator CK Hutchinson Holdings Ltd, which sought to acquire sole control over Telefónica Europe via its indirect subsidiary Hutchinson 3G UK Investments. It is the first case in which the ECJ has the chance to evaluate the question of when an impediment to competition is “significant” if caused by unilateral effects without dominance. Prior to the adoption of the revised EU Merger Regulation (EUMR),[8] the concept of “dominance” was the linchpin of the merger control regime. As a result, transactions in oligopolistic markets that gave rise to unilateral effects effectively slipped through the cracks. In order to close what became known as the “non-collusive oligopoly gap”, the legislator introduced the SIEC test (in Regulation 139/2004) as an alternative to the dominance test. It remained unclear, however, when exactly such effects would be grave enough to warrant a prohibition decision. In some of its earlier decisions, the Commission specifically focused on whether the merger would eliminate a maverick firm.[9] Absent any distinct feature of the target in the present case, the enforcer argued that the notion of “important competitive force” mentioned in paragraph 37 of the Horizontal Merger Guidelines (the Guidelines)[10] should be interpreted broadly, and that it therefore should not be limited to mergers resulting in the elimination of a maverick. This in turn would mean that the mere reduction of competitive pressure on the remaining competitors would suffice to establish a significant impact on the relevant market.[11]

The GC, however, held the Commission to a higher standard, putting forward a novel test which considers firstly i), whether the merger led to the removal of an “important competitive constraint”, and secondly ii), whether it was likely that the remaining players would continue competing aggressively post-merger. Arguing that the notion of being an “important competitive constraint” as mentioned in Recital 25 of the EUMR is more stringent than the notion of being an “important competitive force”, which is the phrase used in the Guidelines, the GC insisted that the target company must stand out from its competitors.[12] In conjunction with this, the GC advanced an equally narrow understanding of the concept of “close competitors”, requiring the merging parties not only to compete closely, but also to be particularly close competitors.[13] When it saw all of its findings quashed, the Commission brought a direct action against the GC’s judgment, relying on six grounds of appeal which the AG addressed separately and at length. The rigour with which the AG addressed each legal issue demonstrates her willingness to assist the ECJ in as many aspects as possible.[14] Using unmistakable words that describe the GC’s analysis as “unduly conflated or even confused”,[15] she reiterated the arguments she made previously in Impala II.[16]

The first issue relates to the allegation that the GC applied an excessively strict standard of review. In principle, the competencies are clearly delineated: while questions of law and fact are subject to full judicial scrutiny, the doctrine of marginal review restricts the EU Courts to evaluating whether the Commission committed any manifest errors if a case involved either policy considerations or complex assessments of an economic or technical nature.[17] Traditionally, the Court followed a careful approach, treating economics as the province of the Commission.[18] The advent of the GC as a first-instance adjudicator in competition cases inevitably changed the institutional equilibrium.[19] Its greater willingness to engage with economic matters gradually moved the EU Courts toward closer scrutiny, and culminated in the annulment of the Commission’s decisions in Airtours, Tetra Laval and Schneider Electric in 2002.[20] The reformulated test under this new generation of cases requires EC decisions to be factually accurate, reliable, consistent and complete,[21] and allowed the Courts to carry out an in-depth probe into the Commission’s evaluations.

In the judgment under appeal, the GC went even further, omitting any reference to the manifest error test and letting mere “errors of assessment” suffice.[22] To rein in further moves to limit the Commission’s discretion, AG Kokott declares “that the review [...] of a Commission decision relating to concentrations”[23] is subject to marginal review. This infelicitous choice of words has caused some commentators to view this as a suggestion to “make limited review the default in merger control”.[24] A closer reading of the passage reveals, however, that the AG was referring to the interpretation of art. 2 EUMR specifically, and therefore to the SIEC test, which has been expressly qualified as a complex economic assessment.[25] Arguably, merger cases are inherently complex and automatically entail a certain margin of discretion; this raises the concern that the Commission could block any transaction based on form-based presumptions. According to critics, the Commission’s increasingly interventionist approach to assessing four-to-three mergers in the mobile telecoms sector is difficult to reconcile with the “more economics-based approach”.[26] However, the AG Opinion is merely a restatement of the law as it stands,[27] leaving it up to the Court to make the final call. In the same vein, the AG is not swayed by the argument that the Commission alone was empowered to define complex economic concepts found in the Guidelines.[28] She emphasizes that only the Courts have the competence to interpret and define legal concepts, even of a complex economic nature, if derived from EU primary or secondary legislation.[29]

Furthermore, the AG reproaches the GC for setting the bar unreasonably high by requiring the Commission to show with “strong probability” that the concentration would lead to a SIEC. This was incompatible with the legislative history, as it would realign the conditions for establishing non-coordinated effects with those for establishing dominance.[30] Emphasizing the “perfectly symmetrical” nature of the standard of proof and the fact that the concept of SIEC is a unitary concept of law,[31] the AG concludes that it would be implausible to apply different standards of proof depending on the type of merger or its complexity.[32] This argumentation echoes her Opinion in Impala II.[33] While the standard of proof itself is invariable, only the type of evidence or its quantity and quality required for the Commission to discharge its burden of proof can change.[34]

AG Kokott reiterates that only a standard based on the “balance of probabilities” is in line with the Commission’s discretion,[35] which she deems more pronounced in merger cases whose prognostic nature renders it “[im]possible to provide ‘objective’ proof of a forecast or for it to be free of uncertainties and doubts”.[36] She accuses the GC of creating a false impression by citing the test suggested by AG Tizzano in Tetra Laval, according to which the Commission must prove “very probable” anticompetitive effects,[37] without mentioning that the ECJ has not assented to this test.[38] Although the ECJ has yet to “expressly identify the relevant test […] as being the ‘balance of probabilities test’”, the AG insists that it has applied it in practice, “requir[ing] the Commission to provide evidence of the ‘most likely’ outcome or ‘plausibility’ of its prospective analysis”.[39] She contends that a stricter standard would usurp the Commission’s ability to act as an effective merger control authority.[40]

The second group of issues focuses on AG Kokott’s criticism of the GC for adopting “both a formalistic and a reductionist reading”[41] of art. 2(3) of the EUMR by limiting the SIEC to two cumulative and exhaustive conditions. Apart from that, she probes the narrow interpretation of these substantive criteria. This applies to the GC’s definition of the concept of “important competitive constraint” as requiring the target to stand out from among all competitors by “competing particularly aggressively in terms of prices” and “forc[ing] the other players on the market to align with its prices or that its pricing policy [is] capable of significantly altering the competitive dynamics on the market”.[42] Likewise, she disagrees with the condition that the parties must be “particularly close competitors”.[43] As the GC acknowledged, the concept of “closeness” does not appear in the EUMR, but only in the Guidelines. Nevertheless, it inaccurately stated “that the applicability of Article 2(3) [EUMR], read in the light of recital 25 of that regulation, requires” that the transaction, if allowed, would eliminate the direct pressure the merging operators exerted on each other.[44] Thus, the GC was wrong to infer from this that the parties must be each other’s closest competitors.[45]

The last issue concerns the GC’s argumentation with regard to the quantitative assessment of evidence as carried out by the EC. In particular, it relates to the upward pricing pressure test (the UPP test), which is used in mergers in differentiated product markets to evaluate the likelihood of a price increase that would result from the transaction if it proceeds.[46] The first point of contention was that the Commission had issued a prohibition decision despite the fact that the expected price increase in the present case was lesser than that in previous decisions[47] where the Commission had granted clearance. AG Kokott dismisses this argument, stressing not only that had the GC failed to provide any further explanation in this regard, but also that “[it] was not entitled to base that conclusion on a comparison with other cases”.[48] Given that the Commission had approved those prior mergers subject to commitments, the AG concluded that the decisions could not provide guidance for the case at hand. In all three cases, the Commission had reached its decision only after an in-depth Phase II investigation and a careful examination of the remedies offered by the merging parties. Out of a total number of 21 “gap mergers” in the mobile telecom sector issued between 2006 and 2019,[49] Hutchinson 3G UK/Telefónica UK is the only case in which the Commission adopted a prohibition decision given the very specific features of the mobile telecoms market in the UK. Against this background, it could be suggested that the cases were indeed materially different.

But perhaps a comparison between these cases was unwarranted in the first place. The GC’s reference to these earlier decisions raises the broader question of the precedential value of Commission decisions. Notably, the GC contradicts its own earlier findings in General Electric, where it held that “an applicant is not entitled to call the Commission’s findings into question on the ground that they differ from those made previously in a different case […] even where the markets at issue in the two cases are similar, or even identical”.[50] Therefore, a comparison between the cases was arguably misplaced. By distinguishing the present decision from previous ones, however, the AG implicitly adopts a broader understanding of what constitutes precedent in the jurisprudence of the ECJ.

The GC then submits that the EC did not review all relevant evidence, as it did not include an assessment of efficiencies in its UPP test in a way the Guidelines allegedly prescribe. In a jolt of creativity, the GC draws a distinction between two different types of efficiencies, namely those “specific to each concentration” (also referred to as “default” or “standard” efficiencies) and those detailed in section VII of the Guidelines.[51] It views the former as part of a “quantitative model designed to establish whether a concentration is capable of producing such restrictive effects”, and “therefore an evidential matter relating to the existence of restrictive effects which arises prior to the overall competitive appraisal as provided for in […] the Guidelines”.[52] In other words, the GC treats “standard” efficiencies as substantive criteria pertaining to the Tatbestandsebene, whereas the second category only becomes relevant at the justification level. Citing academic commentary,[53] AG Kokott remarks that this “rather innovative approach”[54] cannot be retrieved from the EUMR, nor from the Commission Regulation implementing it,[55] nor even from the Guidelines. Moreover, the GC’s suggestion would lead to peculiar results given that the burden of proof is on the incumbent firm to assert any mitigating effects.[56] The AG concedes that the idea of an “efficiency credit” is reminiscent of the developments in the area of art. 101 TFEU. This jurisprudence recognizes that, for the determination of whether an agreement restricts competition by its object, it is possible to take into account procompetitive effects as part of the economic context.[57] However, the EU Courts have expressly denied incorporating the application of a “rule of reason” in EU law, which would have reversed the burden of proof.[58]

Conversely, the Commission claimed that the GC had misconstrued the scope of its review by confining itself to assessing the concepts of “important competitive force” and “closeness of competition”, while the GC itself treated these criteria as part of a set of non-exhaustive factors. Once again stressing the symmetry of the standard of proof, AG Kokott concludes that the GC erroneously placed too much weight on those two factors.[59] Finding fault with the GC’s approach that deemed invalidating the first theory of harm to be sufficient, she recalls the GC’s obligation to “not only establish whether the evidence relied on is factually accurate, reliable and consistent” but also to review the probative nature of the entirety of the relevant factors and evidence.[60]

The AG’s Opinion is an invitation for the ECJ to consolidate its case law and fend off ill-reasoned attempts at legal innovation. Except for some ambiguities, there can be few objections to the AG’s reasoning. Contradictions in the GC’s analysis become even more apparent when contrasted with the Opinion, which pays due regard to established case law and offers sound justifications for each of the propositions made. True to the saying that “you can’t have your cake and eat it, too”, the AG reminds the GC “to carry out its own overall analysis”[61] before criticizing the Commission for lack of rigour when exercising its discretion.