The General Court annuls the European Commission’s prohibition of the Three/O2 merger – a reset for EU merger control?

The General Court’s judgment will have significant implications regarding the circumstances in which the Commission can prohibit transactions that do not create or strengthen a “dominant” player both in the telecom sector and more widely.  Its impact is potentially comparable to that of the Airtours judgment in 2002, which opened the door to a reform of the EU Merger Regulation.  However, given the implications for merger control policy and the legal test to be applied, we expect the Commission to appeal this judgment before the Court of Justice. 

The Commision's 2016 decision

In its 2016 decision, the Commission found that by combining two of the four UK MNOs, the transaction would have reduced competition in the UK retail mobile telecommunications market, leading to higher prices and reduced choice and quality for consumers.  This was on the basis that both Three and O2 exerted an important competitive constraint in the market and, moreover, Three was an “important competitive force” – a term of art used in Commission guidelines to refer to firms that have a greater influence on the competitive process than their market shares suggest.  The Commission also concluded that Three and O2 competed closely with each other, based on evidence which included views obtained from third parties during its investigation and quantitative data on customer switching patterns.

The Commission based its conclusion that Three was an “important competitive force” on a range of evidence, including an allegedly disruptive role on the market, aggressive pricing, growth in subscriber numbers and its “gross adds” market share (ie the proportion of new customers and customers switching provider won by each firm – a metric routinely used by the Commission in assessing cases in the sector).   

The Commission also identified concerns in two other areas – first, regarding the impact of the transaction on network sharing arrangements (stemming from the fact that the merged entity would be party to both network sharing arrangements in the UK), and, second, regarding the reduced number of MNOs available to host mobile virtual network operators (MVNOs).  

In order to address the Commission’s concerns in relation to the loss of retail competition between Three and O2, CK Hutchison offered a range of remedies designed to strengthen existing MVNOs and/or encourage new entry by MVNOs. Specifically, CK Hutchison offered to give access to a share of the merged entity’s network capacity to one or two MVNOs, by divesting O2’s 50% shareholding in the Tesco Mobile MVNO joint venture and offered a wholesale agreement for a share of its network capacity to another MVNO, Virgin Mobile.  CK Hutchison also offered commitments in relation to MVNO hosting, and behavioural remedies to address the Commission’s specific concerns relating to network sharing.  However, the Commission rejected these proposals as insufficient and instead prohibited the transaction.

Breaking the wave?

The MVNO-focused remedies offered by CK Hutchison were similar to (but more extensive than) the remedy packages that had been sufficient to obtain clearance in a number of previous “four to three” MNO mergers that formed part of a push for consolidation in the telecommunications sector – beginning with Hutchison’s acquisition of Orange Austria (conditionally cleared at the end of 2012) and followed by Hutchison’s acquisition of O2 Ireland (conditionally cleared in May 2014) and Telefónica’s acquisition of E-Plus in Germany (conditionally cleared in July 2014).  While these decisions arguably served to some extent as a brake on consolidation, they suggested it was still possible to obtain clearance for a “four to three” MNO merger.

However, shortly after the German and Irish decisions, there was a change in leadership at DG COMP, with the arrival of Margrethe Vestager as Commissioner for Competition in November 2014.  That this had led to a hardening of enforcement policy was signalled by the collapse of the proposed Danish mobile joint venture between TeliaSonera and Telenor (another “four to three” case) in September 2015.  Originally notified to the Commission in April that year, the parties withdrew from the transaction after it became apparent that they could not agree acceptable remedies with the Commission.

The fact that the Commission took a harder line in the UK Three/O2 transaction than in previous cases was therefore not unexpected.  However, the outright prohibition of the transaction was still a significant blow to industry hopes for further consolidation.  In particular, the Commission’s decision was seen as evidence that in future, “four to three” MNO mergers would only be cleared on the basis of a package of divestments sufficient to lead to entry by a new fourth MNO (rather than one or more MVNOs as in previous cases).  This understanding was reinforced by the Commission’s subsequent decision to clear a joint venture combining the Italian mobile businesses of CK Hutchison and VEON (Tre and WIND), on the basis of divestments that allowed French MNO Iliad to enter the market.

The unconditional clearance of a “four to three” mobile merger in the Netherlands in 2018 (TMobile Netherlands NL’s acquisition of Tele2 NL) showed that four was not necessarily a “magic number” for the Commission as regards competition between MNOs (although that decision reflected the Commission’s assessment that Tele2 NL was what is sometimes referred to as a “flailing firm”, ie one which, while not a “failing firm” that would exit the market absent the transaction, would nevertheless decline in competitive relevance over time). Nonetheless, this background meant that CK Hutchison’s appeal was closely watched by many in the sector for its potential to shed light on the scope for future M&A activity.

The General Court's judgment

The General Court’s judgment annuls the Commission’s decision in its entirety, finding errors of law and assessment in relation to each of the three theories of harm on which the decision was based.

At the heart of the judgment is the interpretation of the legal test that the Commission needs to apply in assessing mergers. The first EU merger regulation (passed in 1989) required the Commission to assess whether a transaction would create or strengthen a “dominant position” – if not, a transaction could proceed. This did not necessarily require the creation or strengthening of a single dominant player – the Commission could also prohibit transactions that created or strengthened a position of “collective dominance”. What exactly that meant was the subject of considerable debate, and in 2002 these uncertainties led to the annulment of the Commission’s Airtours prohibition decision by the General Court’s predecessor, the Court of First Instance.

The Airtours judgment in turn contributed to the use of a modified legal test in the current merger regulation (adopted in 2004). This was designed to address a perceived “gap” in the Commission’s merger control powers, in respect of mergers that would neither create or strengthen a single dominant firm nor create a collective dominant position in the sense required by the Airtours judgment.

The test in the current merger regulation asks whether a transaction would “significantly impede effective competition […] in particular as the result of the creation or strengthening of a dominant position”. This makes clear that the Commission can prohibit a transaction in a concentrated market that nevertheless does not create or strengthen a dominant player or a collective dominant position. However, the courts had not explored the conditions for this until now.

The General Court has now clarified that two cumulative conditions must be met: first, that the transaction involves the elimination of important competitive constraints that the parties had exerted upon each other, and, second, that there will be a reduction of competitive pressure on the remaining operators.

The General Court also clarified the burden and standard of proof required in applying this test. The Commission must produce sufficient evidence to demonstrate with a strong probability the existence of significant impediments to effective competition following the transaction. This is a stricter standard than the “balance of probabilities” or “more likely than not” standard for which the Commission had argued, although less strict than one based on the impediment being “beyond all reasonable doubt”. This seems to us a point where an appeal to the Court of Justice would be required to clarify the position. 

The General Court’s judgment goes on to explain in detail why it considered that the Commission’s analysis fell short of meeting this test. A number of these criticisms of the Commission’s analysis revolve around factual aspects and evaluation of evidence, and therefore an appeal to the Court of Justice on a point of law would be more difficult here. 

In relation to competition between Three and O2 on the retail mobile market, the General Court found that the Commission had failed to establish that Three was in fact an “important competitive force”. The Commission was wrong to consider that an “important competitive force” need not “stand out” from its competitors; an approach which, the General Court noted, would effectively allow the Commission to prohibit any horizontal merger in a market with a limited number of major competitors. Moreover, the Commission’s findings were also vitiated by errors in the assessment of the underlying evidence – for example, the General Court concluded that Three’s “gross adds” market share was in fact very low in comparison to those in the previous mobile mergers described above. The General Court also held that while the Commission might have established that Three and O2 were relatively close competitors in some segments of a concentrated market, this was insufficient to meet the legal test for prohibition. Strikingly (given the discussion of four as a “magic number” for MNOs above), the General Court observed that if the Commission’s approach was correct, “any concentration resulting in a reduction from four to three operators would as a matter of principle be prohibited”.

Next, the General Court addressed the Commission’s quantitative analysis of the likely effects of the transaction on prices (so-called “upwards pricing pressure” or “UPP” analysis). The General Court confirmed that the Commission was entitled to take such analysis into account in its assessment, but found that the analysis here failed to show with sufficient probability that prices would rise “significantly”, and that the Commission had failed to address whether this would result in a significant impediment to effective competition. Of particular interest, though, the General Court faulted the Commission for failing to include cost savings resulting from the transaction in its quantitative analysis of pricing effects, ie in its analysis of its own theories of harm, where the Commission bears the burden of proof. Instead, the Commission considered cost savings only in the context of whether the transaction gave rise to efficiencies that would offset the adverse effects on competition that the transaction would otherwise have – where the parties bear the burden of proof. The General Court’s ruling touches on an important debate as to whether the Commission’s approach on this issue is unduly restrictive. The General Court has made a sweeping statement on efficiencies, saying that “any concentration will lead to efficiencies, the extent of which will also depend on external competitive pressure”. Many have read this statement as a sort of efficiency presumption which sits uncomfortably with the current framework of analysis. Again, this is a point that would require clarification through an appeal to the Court of Justice.

Turning to the Commission’s remaining theory of harm regarding the effects of the transaction on mobile network sharing arrangements, the General Court essentially found that while it subscribed to the Commission’s theory to a certain extent, the Commission had failed to show that any effects on these would significantly impede effective competition.

As to MVNO hosting, lastly, the General Court found that while Three could be seen as a credible competitor in hosting MVNOs (taking into account recent increases in its market share), the Commission had still failed to show that its combination with O2 would significantly impede effective competition at wholesale level in a context where its market share remained very low. 

Accordingly, the General Court found that none of the Commission’s theories of harm could stand and annulled its decision without needing to address CK Hutchison’s arguments regarding its proposed remedies.

What next?

The clear overall takeaway is that the General Court has made it harder for the Commission to block transactions that fall short of creating or strengthening a dominant player, even where the market is already relatively concentrated.  From a commercial perspective, the judgment will inevitably invite speculation as to whether there is now more scope for consolidation in the telecommunication sector and elsewhere.  

The Commission now has around two months to decide whether to appeal to the Court of Justice on a point of law.  As at the time of writing, Commissioner Vestager has said the Commission is “urgently analysing” the judgment and the legal issues it raises.  In our view, the Commission is highly likely to appeal, which may temper the impact of the judgment somewhat in the short term.  Any appeal will be closely watched and could potentially lead to wider revisions in the Commission’s approach to merger assessment, with parallels to the Airtours judgment discussed above.