Non-binding opinion finds that General Court erred in applying heightened standard of proof to cases involving unilateral effects in oligopolistic markets.

The Case and the Opinion

On October 20, Advocate General Kokott issued her opinion[1] on the European Commission’s appeal to the Court of Justice (“ECJ”) of the General Court’s landmark May 2020 judgment overturning the Commission’s prohibition of the Three/O2 UK mobile telecommunications merger.[2]  The opinion advises the ECJ to uphold the Commission’s appeal on all main grounds and refer the case back to the General Court for reconsideration.

The case represents the first opportunity for the ECJ to rule on the concept of “significant impediment to effective competition” (“SIEC”) as it relates to non-coordinated (unilateral) effects in an oligopolistic market – precisely the sort of “gap case” that the EU Merger Regulation’s 2004 change from the dominance test to the SIEC test was intended to capture.

Standard of Proof

The Commission’s first and central ground of appeal is that the General Court committed an error of law by applying a stricter standard of proof than recognised by the ECJ.  The General Court held that the Commission is required to produce sufficient evidence to demonstrate with a “strong probability” that a concentration will give rise to harm, expressly noting that this standard lies between a balance of probabilities test and a requirement of proof beyond reasonable doubt.[3]  This appeared to reopen the previously settled question of the appropriate standard of proof, which the Court of Justice established in Bertelsmann and Sony – a case in which Advocate General Kokott herself submitted an influential opinion – as being a balance of probabilities, where the Commission is required to predict the outcome that is “most likely to ensue.”[4]

The Advocate General’s new opinion dissects the General Court’s ruling, finding that the Court went wrong chiefly by confusing the appropriate standard of proof with the strength and quality of evidence that must be adduced in order to reach a conclusion on the likelihood of a particular outcome.  According to the Advocate General, Bertelsmann and Sony and other case law establish that the standard of proof – which is symmetrical in respect of decisions to approve or prohibit transactions and the same as applied to all types of concentrations and theories of harm – is the balance of probabilities.

As to the quality of evidence that should be adduced to meet that standard, in Tetra Laval[5] and other cases the ECJ has held that the more complex or uncertain the theory of harm identified by the Commission, the more persuasive the evidence should be.  The Advocate General finds that the General Court erred, in essence, by inferring that a heightened evidentiary standard in certain circumstances means that those cases are subject to a stricter standard of proof.

The Concept of SIEC and Closeness of Competition

The Commission’s second main ground of appeal is that the General Court committed an error of law by misinterpreting the concept of an SIEC.  The General Court cited EUMR Recital 25 in holding that, in the absence of finding that a concentration will create or strengthen a dominant position, to prove an SIEC the Commission must establish that the concentration would involve: (i) the elimination of important competitive constraints that the merging parties had exerted upon each other; and (ii) a reduction of competitive pressure on the remaining competitors.

The Advocate General accepts the Commission’s argument that this approach is too restrictive and would prevent the Commission from taking account of all relevant circumstances affecting competition in an oligopolistic market – including in particular a reduction in competition by competitors following a concentration.  This limitation is not justified by any convincing consideration and would be contrary to the basic objective of Article 2 EUMR, which is to effectively control concentrations that are liable significantly to impede effective competition.  Accordingly, so the Advocate General opined, the concept of an SIEC is more flexible than what the General Court set out.

The Commission also challenged the General Court’s application of “closeness of competition” in its assessment of the extent to which the merging parties exercised competitive constraints on each other.  In its prohibition decision, the Commission had established that Three and O2 were close – but not “particularly” close – competitors on the UK mobile telecommunications market.  The General Court observed that in such an oligopolistic market, all competitors are likely to be relatively close in some segments, which means that for the Commission to establish that a concentration will remove an important competitive constraint, the Commission must show that the merging parties are particularly close competitors.  Were this not the case, any 4-to-3 transaction would in principle be prohibited.

The Advocate General observes that closeness of competition is only one of the factors that the Commission considered in respect of its theory that the concentration would have eliminated an important competitive constraint.  The Commission’s Horizontal Merger Guidelines recognise that closeness of competition is a matter of degree, and neither the Guidelines nor the EUMR require that the merging parties be particularly close competitors in order to establish the elimination of important competitive constraints, or an SIEC as such.  This purported requirement is ultimately a reflection of the General Court’s excessive standard of proof, including in respect of the elements that may be assessed in order to determine whether they were capable of demonstrating the existence of an SIEC.  Closeness of competition, as measured for example by diversion ratios, is one such element, but the merging parties need not be particularly close competitors for this factor to be relevant.


The Advocate General’s position on the standard of proof issue is not surprising.  We previously noted that the General Court’s raising the standard of proof from the balance of probabilities test appeared vulnerable, and we think the Court of Justice will likely follow the Advocate General on this point.  However, the practical significance of this for future transactions should not be over-stated.  The more important point, which is not under appeal, is that the more tenuous a theory of harm being advanced, the stronger and more convincing must be the evidence adduced in support of a conclusion that harm to competition is the most likely outcome.

Similarly, we think the Court of Justice will agree with the Advocate General’s position on closeness of competition.  Depending on the circumstances, merging parties need not necessarily be particularly close competitors in order for a concentration to lead to competitive harm.

Even if the Court of Justice accepts the Advocate General’s recommendations and the case is sent back to the General Court for reconsideration, the directions from the ECJ may not have much effect beyond the individual case.  In its June 2022 Thyssenkrupp v. Commission judgment, the General Court considered another appeal of a Commission prohibition decision based on unilateral effects in an oligopolistic market.  The General Court upheld the prohibition, and in so doing made no reference to the heightened standard of proof it had set out in Three/O2 and had no difficulty identifying Tata Steel as an important competitive constraint.  In this light, we expect the state of the law to revert to its previously settled position.

[1]           European Commission v. CK Telecoms UK Investments Ltd, Case C-376/20 P, Opinion of Advocate General Kokott, October 20, 2022.

[2]              Case COMP/M.7612 Hutchinson 3G UK / Telefónica UK, Commission decision of September 29, 2016.

[3]              CK Telecoms UK Investments Ltd v. Commission (“Three/O2”), Case T-399/16 EU:T:2020:217, para. 118.

[4]              Bertelsmann and Sony v. Impala (“Bertelsmann and Sony”), Case C-413/06 P EU:C:2008:392, para. 52.

[5]              Commission v. Tetra Laval, Case C-12/03 P EU:C:2005:87, para. 44.