The judicial review of a merger decision by the European Commission is a rare phenomenon. Judgments that rule against the Commission are rarer still. In an administrative system of merger control, it should perhaps not be surprising that only a small fraction of decisions become the subject of judicial appeals. By the same measure, an appeal that results in a comprehensive annulment of a Commission decision is by definition an extraordinary event.
On May 28, 2020, the General Court of the European Union delivered its decision in CK Telecoms UK Investments Ltd v. Commission, annulling the Commission’s 2016 decision to prohibit the proposed merger of Hutchison 3G UK (Three) and Telefónica UK (O2), two of the UK’s cellular network operators. The case involved the assessment of the competitive effects of a transaction in an oligopolistic market. In its judgment, the Court found wanting the Commission’s interpretation of the standards to be applied in assessing such transactions, as well as the quantitative analysis it used as evidence to support the prohibition of the transaction.
If the Court’s findings in Hutchison survive the Commission’s pending appeal to the Court of Justice of the European Union, the case will dictate at the very least a re-assessment by the Commission of some of the key substantive assumptions upon which it relies when examining mergers in oligopolistic markets. The question of whether a merger gives rise to a significant impediment of effective competition (SIEC) has been the central plank of the Commission’s merger reviews since the SIEC test was introduced in the EU Merger Regulation in 2004. The Court’s decision in Hutchison is the first from the Court on the question of whether—in the context of an oligopolistic market—the creation of a non-dominant firm as a result of a merger could constitute an SIEC. The Court’s judgment in Hutchison suggests that the approach the Commission has developed over the years, in the absence of case law, requires something of an overhaul—with potentially significant implications for future transactions in oligopolistic markets.
The combination of Three with O2, proposed in 2015, would have represented a four-to-three transaction, with two of the four mobile network operators (MNOs) active in the UK combining to form the largest MNO in the UK market, holding approximately 30–40 percent of the market. A combination of this magnitude in an oligopolistic market in any industry sector would be expected to attract close interest from regulators. That the merger would be in the telecom industry—an industry characterized by dense oligopolies throughout the European Union—meant that the proposed merger of Three and O2 was inevitably bound to attract a particularly intense degree of scrutiny.
At that time, the Commission had reviewed several telecom mergers in the preceding years and on several occasions had concluded that the combination of non-dominant players in four-to-three merger situations would likely give rise to SIECs. The Commission issued conditional clearance decisions in several four-to-three mergers of MNOs, subject to substantial commitments, such as the divestment of physical infrastructure and radio frequency spectrum allocation, or undertakings requiring the merged entities to take steps that would enable the entry of a fourth MNO in the market. The prevailing assumption was that the Commission was becoming increasingly skeptical of four-to-three mergers in the telecom sector. When Hutchison notified the Commission of the Three/O2 merger in 2015, the challenging regulatory landscape was made more fraught still by public opposition to the proposed combination by both the Competition and Markets Authority (the UK national antitrust authority) and Ofcom (the UK telecom regulator). Against this background, the Commission’s decision in May 2016 to prohibit the merger was perhaps not an outright surprise.
Ultimately, the Commission developed and relied on three separate theories of harm in reaching the conclusion that the combination of Three and O2 would likely give rise to an SIEC in the UK market, and one that could not be adequately neutralized through structural remedies proposed by the merging parties. The Commission found that the merger would result in non-coordinated anticompetitive effects (1) in the retail market for mobile communications services; (2) related to network-sharing arrangements; and (3) in the wholesale market for mobile communications services. In its judgment annulling the Commission’s decision, the Court found that the Commission had erred with respect to each of those theories of harm.
The Court’s findings with respect to the first theory of harm—non-coordinated effects in the retail market—were particularly apposite to the overarching question of what constitutes an SIEC in the context of a merger in an oligopolistic market. In its prohibition decision, the Commission had concluded that the merger would likely significantly weaken competition in the UK mobile telecom retail market. In reaching that conclusion, it had found that (1) the merger would lead to the elimination of Hutchison’s Three as an “important competitive force,” (2) Three and O2 were significantly close competitors, and (3) the merger would lead to the loss of other important competitive constraints in the market—as evidenced by an upward pricing pressure (UPP) analysis conducted by the Commission.
The Commission’s Horizontal Merger Guidelines recount the relevance of a merger’s eliminating an important competitive force in assessing the likelihood of an SIEC. The Guidelines state that “[some] firms have more of an influence on the competitive process than their market shares . . . would suggest. A merger involving such a firm may change the competitive dynamics in a significant, anti-competitive way, in particular when the market is already concentrated.” The Court took issue with the Commission’s interpretation of what constitutes an important competitive force in the context of a concentrated market: the Commission had concluded in its 2016 Decision that a competitor did not need to stand out in particular from other competitors to constitute such a force, and that the mere removal of a competitor that was exerting some form of competitive constraint could, in an oligopolistic market, qualify as the elimination of an important competitive force.
In finding fault with the Commission’s approach, the Court observed that the Commission, in its defense to the appeal, had “conceded that an important competitive force must have more of an influence on competition than its market share would suggest, compete in a particularly aggressive way and force other players to follow that conduct.” The Court added, however, that the Commission had gone too far by taking the position that “the mere decline in the competitive pressure which would result, in particular, from the loss of an undertaking having more of an influence on competition than its market share would suggest is sufficient, in itself, to prove [an SIEC].” For example, under the Court’s reasoning, a competitor with a 10 percent market share is not necessarily an important competitive force just because its influence is equivalent to that of another competitor with slightly higher share. Rather, the competitor would need to play a significant role in the market, enabling it to exert disproportionately strong constraints on other competitors, compared to the power that its market share would suggest.
Aside from finding that the Commission had interpreted the concept of an important competitive force too broadly, the Court concluded that the Commission had erred by confusing the concept of an important competitive force in its own Guidelines with the legal concepts included in the Merger Regulation itself. The concept of an SIEC—the actual legal criterion in Article 2(3) of the Merger Regulation—and that of an important competitive constraint, contained in recital 25 of the Merger Regulation, constitute the statutory legal concepts.
The Court found that permitting the Commission to persist in a confusion of these three separate concepts would allow the Commission to “omit to analyse the possible elimination of the important competitive constraints that the merging parties exert upon each other in favour of a theory of harm based solely on a reduction of competitive pressure on the remaining competitors.” The Court added that the SIEC criterion set out in the Merger Regulation must be interpreted as permitting the Commission to prohibit mergers on oligopolistic markets where the effect on competitive conditions is equivalent to the creation or strengthening of an individual or collective dominant position. This would occur where the merged entity has “the power to enable it to determine, by itself, the parameters of competition and, in particular, to become a price maker instead of remaining a price taker.”
In this respect, the central thrust of the Court’s findings seems aimed at avoiding a situation where reviews of transactions in oligopolistic markets, carried out under the Merger Regulation, move too readily toward findings of SIECs simply because there is some reduction in competitive pressure—as opposed to the elimination of a competitor that truly represents an important competitive constraint.
Assessments of the closeness of competition between merging parties is a familiar feature of merger reviews by antitrust authorities in many jurisdictions, including the European Union. Analyzing the intensity of competition between two merging companies can be a useful barometer of how important a competitive constraint one company represents for the other. The concept of “closeness of competition” does not appear in the Merger Regulation itself. Rather it is captured in the Commission’s Guidelines, which state that “the fact that rivalry between the [merging] parties has been an important source of competition on the market may be a central factor in the [Commission’s] analysis.”
In its 2016 Decision, the Commission pointed to various factors it interpreted as indicative of closeness of competition between Three and O2, including the results of a survey of subscriber switching data designed to show the diversion ratio of subscribers between the two MNOs. Analyses of closeness of competition have been scrutinized in previous cases before the Court, and have been found to form a valid component of merger reviews by the Commission. Nonetheless, in Hutchison—and in something of a theme in this judgment—the Court reiterated the fact that the concept of a “close competitor” is limited to the Commission’s own Guidelines as opposed to forming part of the Merger Regulation itself (incidentally, casting in a rather different light the Commission’s references in its 2016 Decision to closeness of competition as a “legal test”). The Court considered that the Commission had wrongly focused on the general closeness of competition among all four MNOs active in the UK, as opposed to scrutinizing the intensity of competition between Three and O2 specifically. Aside from observing weaknesses in the survey data used by the Commission (in particular the way in which diversion ratios between networks were calculated by the Commission using a limited sample size, and the fact that the results of that survey did not match up with the results of the fuller quantitative analysis set out in an annex to the 2016 Decision), the Court concluded that the Commission’s analysis did not indicate that Three and O2 were “particularly close” MNOs, “even if, on such a market, all operators are, by definition, close to a greater or lesser extent.”
Complex economic assessments have also formed part of the Commission’s merger reviews for many years, with important Court of Justice judgments, such as Airtours plc v. Commission, highlighting the need for Commission merger decisions to be sufficiently grounded in concrete evidence. In approving conditionally other MNO mergers prior to the 2016 Decision, the Commission had conducted UPP analyses to predict the probable increase in average prices that would occur as a result of the merger. Having carried out a UPP analysis in its assessment of Three’s proposed acquisition of O2, the Commission had concluded that the merger would likely lead to a 7.3 percent increase in average prices for the private segment of the UK mobile telecom market. In examining the Commission’s UPP analysis, the Court recalled that its role in this case was not only to verify that the Commission had relied on evidence that was “factually accurate, reliable and consistent, but also [to] ascertain whether the evidence contains all the information which must be taken into account in order to assess a complex [merger] situation—and whether it is capable of supporting the conclusions drawn from it.”
The Court found that the Commission’s UPP analysis had relied on a limited number of inputs (diversion ratios and margins), and was insufficient to support a conclusion that the elimination of whatever competitive constraint existed between the parties would likely lead to significant increases in prices. The Court recognized that the Commission itself had never suggested that the UPP analysis results were absolutely conclusive, and indeed the Court pointed to the fact that the Commission had taken a “somewhat prudent approach” to the probative value of its quantitative analysis. Ultimately, the Court concluded that the results of the analysis could not be considered to be sufficient evidence that the elimination of the competitive constraint that the parties exerted on each other would result in an SIEC.
The Court cross-referenced in its assessment of the UPP analysis one of its broader conclusions in this case, and what may represent one of the most consequential findings of the entire judgment: in complex cases where the Commission is attempting to establish an SIEC from a body of evidence and indicia based on several theories of harm, “the Commission is required to produce sufficient evidence to demonstrate with a strong probability the existence of significant impediments following the [merger].” Through this finding, the Court established the applicable standard of proof at a level higher than the Commission’s “balance of probabilities” approach in its 2016 Decision, but below the level of “beyond all reasonable doubt.” While underlining that it does not believe that the standard of proof applicable in non-coordinated effects cases is different from that applicable in coordinated effects cases, the Court does imply the existence of a continuum, with relatively greater importance attached to the quality of evidence where an analysis is more uncertain or difficult in the circumstances. The Court concluded that:
[T]he more a theory of harm advanced in support of [an SIEC] put forward with regard to a concentration is complex or uncertain, or stems from a cause-and-effect relationship which is difficult to establish, the more demanding the Courts of the European Union must be as regards the specific examination of the evidence submitted by the Commission in this respect.
As a final strike against the Commission’s quantitative analysis in this case, the Court found that the Commission confused two distinct types of “efficiencies” analyses when appraising the transaction. The Commission’s Guidelines provide that the Commission should take into account any asserted efficiencies arising from a transaction and determine whether those efficiencies would likely outweigh whatever anticompetitive effects are otherwise likely to arise as a result of the merger. The Court faulted the Commission for confusing that weighing of efficiencies against the merger’s anticompetitive effects with the initial consideration of “standard” efficiencies, such as the elimination of duplicate production and distribution structures, and more generally through the rationalization and integration of production and distribution processes, as a component of a quantitative model designed to establish whether the merger is capable of producing anticompetitive effects.
The Hutchison Court concluded that the Commission had wrongly determined that it did not need to take such standard efficiencies into account when conducting its quantitative analysis, rather assuming that its separate assessment of efficiencies claims under the terms of its Guidelines would suffice. The Court explained that those standard efficiencies, as “merely a component of [the] quantitative model,” should have been taken into account by the Commission as part of its economic analysis of whether the merger was likely to restrict competition regardless of and before any separate weighing of specific efficiencies claims by the parties as contemplated in the Guidelines. In effect, the Court underlined the importance of the Commission maintaining an assessment of “standard” efficiencies in any dedicated quantitative analysis since, even if standard efficiencies are not sufficient to outweigh anticompetitive effects of a merger, they may nonetheless be relevant for other aspects of the SIEC analysis.
The Commission’s decision with respect to the second and third theories of harm (non-coordinated effects arising from network sharing, and non-coordinated effects in the wholesale market) was also faulted by the Court. Regarding network-sharing agreements, the Commission had concluded that the transaction would raise serious concerns: future investment in UK telecom infrastructure that was managed via two network-sharing agreements (one between Three and EE and the other between O2 and Vodafone) would result inevitably in a reduction in incentives for the other parties to those agreements to invest in future network roll-outs. The Court rejected the Commission’s findings in this respect, finding that there was no solid basis for the Commission’s claim that a disruption to operators’ alignment of interests in the network-sharing agreements would likely harm competition.
As regards non-coordinated effects on the wholesale market—with MNOs such as Three hosting services for mobile virtual network operators (MVNOs)—the Commission found that Three was an important competitive force, the loss of which would likely lead to anticompetitive, non-coordinated effects in a highly concentrated market. The Court was dismissive of the Commission’s assessment. It essentially found that, while Three was a credible competitor in supplying MVNOs and other providers in the wholesale market, and influenced competition even when it did not win bids, that was not sufficient to classify Three as an “important competitive force.” Moreover, the Commission had failed to establish that Three and O2 exerted important competitive constraints upon each other.
Hutchison can reasonably be considered a significant defeat for the Commission, heralding a potential recalibration of the Commission’s current approach to determining what constitutes an SIEC and the standard of proof to establish an SIEC in oligopolistic markets. The SIEC test was originally introduced in the Merger Regulation in 2004 as a means of dealing with a perceived gap in EU merger enforcement: the previous EU merger regulation had provided for a substantive test based solely on the creation or strengthening of an individual or collective position of dominance. This test failed to give the Commission a basis for intervention in mergers that did not meet this dominance test but could still lead to a substantial lessening of competition in oligopolistic markets.
The Court’s judgment in Hutchison represents the first full assessment by the Court of the way in which the Commission has been interpreting the SIEC test in mergers in oligopolistic markets and the standard of proof that applies. In issuing this judgment, the Court has clarified that there must be a “strong probability” of harm as a result of a transaction, and the Commission is not entitled to automatically label all operators in a concentrated market as “important.” Rather, the player that is being removed from the market as a result of the merger must “stand out” as posing a significant competitive constraint (not just having been a competitive force) in the market. The standard of proof that the Commission must satisfy to demonstrate that a player is sufficiently important has, arguably, been raised substantially.
The judgment in all likelihood makes it more difficult for the Commission to prohibit mergers—or to extract significant remedies from merging parties—in oligopolistic markets in the absence of clearer evidence of harm to competition as a result of the transaction. It effectively curtails the Commission’s discretion in these types of cases, and makes the very notion of what constitutes “clear evidence of harm” a more fraught matter than it seemed previously. Among a crowded field of negative findings for the Commission in this judgment, the Court’s conclusions about the applicable standard of proof are conspicuous. Prior case law held that where the Commission puts forward an “inherently complex” theory of harm in its review of a merger, “such complexity does not, of itself, have an impact on the standard of proof which is required [to establish the existence of an SIEC].” In other words, the Commission’s decision should be based on identification of the most likely outcome of a merger. However, the Court’s findings in Hutchison dictate that—insofar as determining the existence of an SIEC in oligopolistic markets is concerned—the Commission must effectively demonstrate a strong probability of an SIEC.
Although the Court’s decision does not mean that the original Three/O2 merger is automatically revived, these findings matter because the Commission can be expected to adapt its merger enforcement practice in response to the Court’s decision. Commission case teams handling what are already complex merger investigations in oligopolistic markets face the prospect of having to meet this new, higher threshold when trying to prove that a merger is likely to lead to the removal of an important competitive constraint in a market. Initial reaction to the judgment from Commission officials reflects acute concern: aside from the Commission’s having appealed the judgment, in remarks that are simultaneously forceful and unsurprising, a senior Commission official is reported as having described the Court’s findings as “contrary to the very spirit” of the Merger Regulation. A Commission spokesperson is also quoted as having stated rather more prosaically that, in appealing the Court’s decision, the Commission “respectfully considers that on these issues the General Court made a number of errors of law and is asking the European Court of Justice to set aside the General Court’s judgment in that respect.” Further detail as to the exact grounds that the Commission will have set out in its application to appeal will become available when the Commission’s application is published in the Official Journal of the European Union—something that should happen within the next few months. Outside the Commission perimeter—but not too distant—Tommaso Valletti, Chief Competition Economist at the Commission from 2016–2019, commented that the judgment was “pretty ugly and full of bad consequences… extreme and indefensible.”
It is not surprising that the Commission has appealed the Court’s decision, given the potentially significant implications of the judgment. The Commission has suffered a number of pushbacks from the courts in recent years with respect to certain aspects of the procedural conduct of its merger reviews, including UPS/TNT (prohibition decision annulled for violation of parties’ procedural right to be informed fully of the evidentiary basis on which the Commission based its decision); Lufthansa (failure to properly assess parties’ application for a waiver of the remedies given by the parties at the time of the merger to secure conditional clearance); and Liberty Global/Ziggo (clearance decision annulled for failure to give reasons for not analyzing certain vertical foreclosure theories).
It would be an overstatement to portray this recent run of successful challenges as indicative of a trend toward greater judicial intervention in the EU merger review process—or indeed of a more fundamental problem in the conduct of EU merger reviews. Nonetheless, when set against this backdrop, Hutchison stands out as a decision in the mold of a triumvirate of judgments from the early 2000s—Airtours, Tetra Laval, and Schneider—each of which had a significant impact on the Commission’s conduct of merger reviews. The impact of these judgments was particularly pronounced because they found errors in the Commission’s substantive analysis of merger reviews, and in its interpretation of the relevant statutory standards—arguably a category into which Hutchison now falls.
Each of those decisions came in the wake of a debate that had been ongoing at the time, provoked by the transatlantic divergence in approach that manifested itself in the Commission’s prohibition of GE/Honeywell—a merger cleared by the U.S. Department of Justice under the U.S. substantial lessening of competition standard but prohibited under the European Union’s old dominance-based standard contained in the 1989 Regulation. The ensuing debate and identification of an “enforcement gap” in the European Union gave rise to the shift in the new Merger Regulation from a structural approach to a more effects-based approach, with a corresponding update to the concept of an SIEC. While the changes introduced in 2004 in the new Merger Regulation were closely related to the enforcement gap debate, to some extent they also reflected the substance of the Court’s findings in the Airtours-Tetra Laval-Schneider trio of judgments.
The outcome of the Commission’s appeal against the Court’s judgment will determine whether Hutchison may join the small constellation of judgments that have shaped the Commission’s substantive review standards over the years. However, until the Court of Justice appeal process reaches its conclusion, the Court’s findings in Hutchison will stand. This in itself guarantees at the least an intriguing period during which the Commission’s conduct of merger reviews will be watched carefully for any signs of change in reaction to the Court’s findings.
