The General Court’s judgment in CK Telecoms[1] annulling the European Commission’s (“Commission”) 2016 prohibition decision of the UK mobile merger raised quite a bit of interest beyond the Brussels competition bar, particularly with European telecom companies, who have long argued that the Commission has taken a far too strict stance on in-country market consolidation.
Mobile markets are blueprints for oligopolistic markets and four-to-three mobile mergers are the poster child of the so-called “gap cases”, i.e., horizontal mergers where the merged entity is below the dominance threshold, but where changes to the market structure brought about by the merger could negatively impact competitive dynamics. In the past ten years, the Commission has reviewed seven mergers involving a reduction from four mobile operators to three. Each case involved a detailed and lengthy “Phase II” investigation. Under Commissioner Vestager, the Commission blocked four-to-three mobile mergers in Denmark (2015)[2] and in the UK (2016)[3] and cleared a four-to-three mobile merger in Italy (2016)[4] only after the parties offered to divest assets to allow the entry of a new mobile operator in the market (making the Italian case effectively a four-to-four mobile merger). In 2018, in a somewhat surprising twist, the Commission unconditionally cleared a four-to-three merger in the Netherlands after a contentious Phase II review.[5] The Dutch case energised some telecom executives, but its somewhat surprising outcome was largely due to the very specific circumstances of the case. Then the General Court delivered a blow to the framework developed by the Commission to assess mobile mergers in CK Telecoms.
CK Telecoms makes it significantly more difficult for the Commission to challenge four-to-three mobile mergers (as well as other transactions in oligopolistic markets). The judgment criticises the Commission for making numerous errors in law in determining that the transaction results in “unilateral effects”. With limited exceptions, the Commission’s concerns in mobile mergers are chiefly based on the so-called “unilateral” or “non-coordinated” effects theory of harm, rather than on the “coordinated effects” theory of harm, which requires showing that the transaction would reduce competition in the market by facilitating collusion (or “coordinated effects”). CK Telecoms is the first case in which the methodology relied on by the Commission in the assessment of mobile mergers and, more broadly, the “unilateral effects” theory of harm, was subject to court review. The results are shattering for the Commission. This relates in particular to the two key aspects of the Commission’s analysis and their relevance to assessing the transaction: closeness of competition between the parties, and whether the transaction would eliminate “an important competitive force”. The judgment rejects some of the key assertions made by the Commission in that regard and sets a significantly higher standard in relation to both considerations. It also raises the standard of proof that the Commission needs to meet to show its case (and not just in cases involving a theory of harm based on unilateral effects).
During the initial wave of excitement, some commentators compared CK Telecoms to the Airtours case,[6] which re-shaped EU merger control enforcement nearly two decades ago. Whereas the magnitude of the Commission’s defeat in CK Telecoms is comparable to Airtours, CK Telecoms does not appear to have provoked similar soul-searching at DG Competition. On the contrary, some of the senior officials criticised the judgment[7] and indicated that the Commission does not intend to introduce any radical changes in its assessment of unilateral effects. The Commission insists that its approach based on its Horizontal Merger Guidelines is “economically sound” and challenges the General Court’s reasoning as flawed from an economic perspective, and its conclusions as limiting the Commission’s ability to conduct an effects-based analysis.[8] Not surprisingly, the Commission appealed the judgment, challenging all of its key aspects.[9]
To be sure, the outcome of the appeal is far from clear but unless and until CK Telecoms is overturned, it is binding on the Commission and must be taken into account in its decision-making process. Moreover, the unconditional clearance in the Dutch case confirms that the Commission may entertain unconditional clearance in some four-to-three mobile mergers. Notably, there are some parallels between the Dutch decision and the Court’s reasoning in CK Telecoms. It would thus appear that the Commission is likely to take stock of at least some of the criticism made in the judgment, in particular in relation to the “toolbox” employed by the Commission to analyse four-to-three mobile mergers. This note analyses the Commission’s practice in that regard and focuses on the likely impact of CK Telecoms on the Commission’s assessment of mobile mergers.
Four-to-three mobile mergers: The UK case and the Commission’s toolbox
The UK transaction involved the acquisition of Telefonica UK’s “O2” by Hutchison 3G UK’s Three, combining the respectively second- and fourth-largest operators. The merger would have reduced the number of Mobile Network Operators (MNOs) on the UK market from four to three. The merged entity would have become the new leader on the retail market (with a share of more than 40%) and its various sub-segments, including, in particular, the key private post-paid segment. As in other four-to-three mergers, the main theory of harm was based on the non-coordinated effects in the mobile retail market. The Commission also asserted that a situation where the merged entity would participate in two different network sharing agreements could negatively impact its network sharing partners, i.e., Vodafone and EE.[10] Finally, the Commission alleged that the transaction would lead to non-coordinated effects on the wholesale mobile market. The Commission blocked the merger following failed remedies negotiations. Hutchison appealed the Commission’s decision, which led to the CK Telecoms judgment.
The UK case followed a spate of four-to-three mobile mergers[11] in which the Commission developed a toolbox to assess whether such transactions would lead to a significant impediment to effective competition (or “SIEC”)[12] based on the unilateral effects theory of harm.[13] With the single exception of the Dutch case, the Commission found that each four-to-three mobile merger it reviewed in the last decade would have resulted in a SIEC.
While both the EU Merger Regulation (“EUMR”)[14] and the Horizontal Merger Guidelines[15] make it clear that a SIEC generally results from the creation or strengthening of a dominant position, dominance is not a necessary condition for a SIEC. As noted in recital 25 of the EUMR, a SIEC may arise in oligopolistic markets, in the absence of a likelihood of coordination, in transactions that involve “the elimination of important competitive constraints that the merging parties had exerted upon each other, as well as a reduction of competitive pressure on the remaining competitors.”
The Commission’s Horizontal Merger Guidelines list the factors that are relevant to the assessment of “unilateral effects”, including market shares of the merging parties, closeness of competition, ease of switching, the ability of competitors to increase supply in case of a price increase and whether one of the parties is an “important competitive force”.[16]
Two of the factors listed in the Horizontal Merger Guidelines have been pivotal in the Commission’s assessment of four-to-three mobile mergers. The first one is the “closeness of competition”, i.e., the degree of substitutability between the products offered by the parties. The second concept is the concept of an “important competitive force.” Both are analysed in turn below.
Close, closer, closest
With respect to closeness, the Horizontal Merger Guidelines note “[t]he higher the degree of substitutability between the merging firms’ products, the more likely it is that the merging firms will raise prices significantly.”[17] Thus, closeness of competition may make the rivalry between the parties an important source of competition on the market.[18] The logic behind this assertion (as explained by the leading Commission economists at the time) is as follows: “if the merging parties sell very close substitutes, they impose on each other a significant competitive constraint. Pre-merger, if a firm raises prices customers may simply switch to its rival. However, following a merger, customers may have no other close substitutes to turn to, and the merged entity could then raise prices significantly, irrespective of whether it becomes the market leader.”[19]
Broadly, in mobile mergers, the Commission attached significant weight to the switching data and, specifically, the data from number portability databases that track telephone numbers in a country and reflect customer churn and migration between the different operators. There are various ways of looking at that data: the most intuitive one is looking at which operator is the first choice when a customer switches away from the merging parties. If that first choice is the other merging party, that party is second best to the other merging operator and the closest substitute. But the Commission asserted that it is not required that the majority of customers choose one of the parties as their first best option, to consider the other merging party as the second best option.[20] In other words, according to the Commission, the fact that the parties were “close” was by itself sufficient to show that the merging parties exerted “important competitive constraints” upon each other and therefore the first prong of the test referred to in Recital 25 of the Merger Regulation was met. Since in a mobile market all MNOs are “close”, this meant that any four-to-three transaction met that requirement.
In the UK case, the Commission also found that Three and O2 competed closely, both against each other and against the other MNOs.[21] Notably, the evidence in the Commission’s file did not suggest that the parties are each other’s closest competitors. The Commission, however, asserted that it is not required to show that the merging parties’ products are each other’s closest substitutes for competition concerns to be raised. In other words, the fact that other MNOs were closer competitors to, respectively, Three and O2, than Three and O2 were close to each other, did not preclude the finding of a SIEC.[22] Rather, the relative closeness of Three and O2 was viewed as a factor contributing to the finding of a SIEC.
The concept of an “important competitive force”
The second factor driving the Commission’s analysis in all mobile mergers was whether the transaction would eliminate an “important competitive force”.[23] In line with paragraph 37 of the Horizontal Merger Guidelines, there may be a situation where a firm has “more of an influence on the competitive process than [its] market shares or similar measures would suggest.” According to the Guidelines, “a merger involving such a firm may change the competitive dynamics in a significant, anticompetitive way, in particular when the market is already concentrated”.
The Commission asserted that an “important competitive force” did not have to have “a disproportionately high impact on the competitive dynamics” or to be a maverick.[24] According to the Commission, a market player is “an important competitive force where it contributes, substantially and consistently, to the competitive process on the market, based on parameters such as price, quality, choice and innovation.”[25] In practice, in assessing whether a company is an “important competitive force”, the Commission looked at variety of quantitative and qualitative factors, including gains in market share, the share of gross adds, capacity to innovate and steer market trends, aggressive pricing, as well as perceptions of customers and competitors expressed in market testing. The threshold required to show that a party was an “important competitive force” was rather low. [26] In fact, in pretty much all of the four-to-three mobile mergers reviewed by the Commission in the past decade, one of the merging parties was considered an “important competitive force”. In the German case, the Commission went as far as to assert that both parties were important competitive forces.[27]
In the UK case, the Commission found that Three, as the relatively recent entrant into the UK market and the smallest (but growing) MNO, was an “important competitive force”, or, in any event, exerted “an important competitive constraint” on the mobile retail market.[28] This conclusion was based on the finding that (i) Three’s market share and subscriber base had been constantly growing; (ii) following its market entry, Three introduced a “series of innovative and disruptive product introductions”, such as abolishing roaming charges in a number of countries and offering 4G at no extra cost to customers. Internal documents were also used in support of the argument that Three was a maverick in the UK market.[29] The Commission further analysed Three’s tariffs and concluded that, via its tariffs, Three consistently exerted an important competitive constraint on the market.
The Commission asserted that it was not required to show that a firm “stands out from its competitors in terms of its impact on competition” in order to find that that firm constitutes an “important competitive force.” Rather, in the Commission’s view, a SIEC could be established on the basis of the decline in competitive pressure resulting from the loss of a company having more of an influence on the market than its market share would suggest.[30]
The concept of an “important competitive force” as interpreted by the Commission was also key to the assessment of the effects of mobile mergers on the wholesale market. In several cases, the Commission found a SIEC on the wholesale market despite the fact that one or even both merging parties had virtually no presence in that market, on the basis that one of the parties was “an important competitive force”.[31] In particular, in the UK case, the Commission found that the transaction would have reduced the number of MNOs willing to host mobile “virtual” operators (“MVNOs”) on their networks from four to three. The Commission asserted that Three played an important role on the wholesale market despite its very small market share (less than 5%).[32] The Commission relied on Three’s internal documents as evidence that Three actively bid for wholesale contracts, that it had a strategic interest in hosting non-MNOs, and that its importance as a competitor in the wholesale market was increasing. On this basis, the Commission concluded that Three was “an important competitive force which has more of an influence on the competitive process than its market share would suggest“[33] and therefore the competitive conditions on the UK wholesale market would “materially deteriorate” post-transaction.[34]
The quantitative analysis
Much to the chagrin of the financial departments of the merging operators, which have often spent weeks responding to the Commission’s requests for information, in all mobile mergers, the Commission used economic models to predict the likely price effects of the transaction. Each decision issued by the Commission in merger cases has included an economic annex setting out the Commission’s upward pricing pressure (“UPP”) and indicative price rise (“IPR”) analyses. In a Phase II investigation, these models were typically developed to make a more sophisticated merger simulation, taking into account the reaction of competitors.
All these models by their very design would always show that the transaction would generate a price increase. While in principle, it is possible to account for efficiencies in a merger simulation, the Commission largely refused to do so in mobile mergers. In the Dutch and Italian cases, for example, the Commission rejected the models proposed by the parties that accounted for efficiencies, considering them to be unreliable.[35] That said, in the Dutch case, the Commission’s own model reflected some variable cost efficiencies claimed by the parties.[36]
In the UK case, the Chief Economist’s Team ran the same types of analyses as in other mobile mergers (from the simpler indicative price rises (“IPR”) analyses to the calibrated merger simulation) and on that basis concluded that the transaction would result in significant price effects. The Commission’s first model (based on diversion ratios at the network level) predicted price increases for Three and O2 of respectively 12.9% and 9% in the overall private segment. The resulting predicted segment-wide price effects were 10.3% in the pre-paid segment, 4.8% in the post-paid private segment and 5.7% on the overall private segment.[37]
In line with common criticism, the parties argued that the economic models used by the Commission would predict price increases in every horizontal merger and were inappropriate to estimate price effects of the proposed merger. The parties submitted that such models could only be used as a first screen to identify cases that require closer investigation, rather than as evidence to support a self-standing theory of harm. The Parties further contrasted the Commission’s approach with the approach of the U.S. agencies, which place less evidentiary weight on such analyses. The Parties further submitted that the use of such analyses effectively reverses the burden of proof to the merging parties and leads to arbitrary discrimination against mergers in markets where the required switching data is available.[38] Finally, given that Commission’s models would always predict a price increase, the parties criticised the Commission for a failure to state a critical threshold of materiality, below which a price increase would not be viewed as evidence of potential anti-competitive effects of the transaction.
The Commission rejected these arguments, stating that its quantitative analyses were based on a “rigorous and standard economic framework” and were not used in isolation from other qualitative evidence. The Commission acknowledged that its models would always predict some price increase, but took the view that the magnitude of the price increases could be used as one of the relevant elements for the Commission’s overall assessment of the case.[39]
Four is the magic number (despite limited exceptions)
All in all, while the Commission has stated on various occasions that there was no “magic number” of MNOs,[40] the substantive test applied by the Commission to assess potential “unilateral effects” in mobile mergers implies that the Commission would effectively find issues in any four-to-three mobile merger. While the Commission unconditionally cleared a four-to-three merger in the Netherlands,[41] it took pains to stress that that was largely due to the very specific circumstances of the case, which the Commission acknowledged only after a contentious Phase II review, including a Statement of Objections and an Oral Hearing.
The reasons for that unconditional approval were twofold. First, the Commission recognised that Tele2 NL’s role as a competitive force in the Dutch market would most likely decrease in the future. While the Commission did not consider (nor the parties argued) that Tele2 NL was a failing firm, the Commission acknowledged that Tele2 NL’s ability to compete would likely deteriorate (due to the specific circumstances relating to the network sharing agreement between T-Mobile NL and Tele2 NL) and considered it relevant in its assessment of the counterfactual. Second, the Commission recognised that the potential anti-competitive effects would be limited due to the relatively small combined market share of the merging parties (the merged entity remained the third largest player in the mobile retail market with a share of approx. 25%) and the limited increment (around 5%) brought about by the transaction. Further, the concerns in this case were limited to the mobile retail market and, specifically, its post-paid private segment. The business segment was not the focus of the Commission’s assessment (it was not an “affected market”, as the parties’ combined share in that segment did not exceed 20%).[42] While the Commission investigated potential concerns in the wholesale market for access and call origination on mobile networks, it ultimately concluded that the transaction did not raise issues in that regard.[43]
As in other mobile mergers, closeness and the concept of an “important competitive force” were key in the Commission’s analysis of the Dutch case. Interestingly, while the Commission at first suggested that the parties were close competitors, it ultimately acknowledged that they were not “particularly close (or the closest) competitors to each other” and there was “a significant degree of closeness of competition among the four MNOs in the market.”[44] The Commission also found that it was unclear whether Tele2 NL’s would continue to be a close competitor to TMNL in the future given Tele2 NL’s uncertain network situation, which was likely to further “increase of the gap in network performances (capacity and quality) between TMNL and Tele2 NL”.[45] The fact that the parties were not “particularly” close contributed to the Commission’s finding that the Dutch transaction would not result in a SIEC.
The Commission also found that Tele2 NL could not be considered an important competitive force given: (i) its limited market share; (ii) its competitive behaviour and performance; (iii) its network limitations; and (iv) the fact that Tele2 NL’s competitive strength would likely deteriorate.[46]
Last but not least, the Commission’s quantitative analysis indicated that the price effects of the transaction would be “moderate” (below 5% in the post-paid private and overall private segment) even without taking into account the weakened position of Tele2 NL in the future.[47] On that basis, the Commission concluded that the transaction would not lead to a SIEC and cleared it unconditionally.
The remedies
While the theories of harm and the assessment of four-to-three mobile mergers has been consistent in the past decade, there has been one important policy change in relation to the remedies required to obtain clearance. Under Commissioner Almunia, the Commission approved three four-to-three telecommunications deals in Austria, Ireland and Germany.[48] Each of the three cases was eventually cleared on a remedy package, chiefly aimed at supporting the entry and expansion of MVNOs.
In the Austrian case, the main remedy was wholesale access to 30% of the network capacity for up to 16 MVNOs based on a reference offer. Hutchison also agreed to offer spectrum, sites and a national roaming agreement to a new entrant, contingent on a new market entrant acquiring spectrum in the next spectrum auction. That second part of the remedy never materialised because the spectrum auction did not result in market entry, which the Austrian telecoms regulator blamed on the lack of effectiveness of the remedies imposed by the Commission. In Ireland and in Germany, the commitments also included wholesale access to up to 30% of the network capacity to an MVNO. To ensure swift market entry, the remedy taker had to be approved before closing. It also had to commit to purchasing minimum capacity usage at fixed prices (to incentivise quick market expansion). In both cases, the remedy package included structural elements: a commitment to offer spectrum, sites, retail outlets and a national roaming agreement, which the remedy taker could opt to purchase if it were to become an MNO. However, the remedy takers have shown little interest in becoming MNOs.[49]
Under Commissioner Vestager, the Commission grew increasingly weary both about effectiveness of MVNO remedies and about the ability of MVNOs to compete with MNOs.[50] The change came swiftly: a year after approving four-to-three mobile mergers in Ireland and Germany based on MVNO remedies, the Commission refused to clear a similar transaction in Denmark,[51] even though the parties offered remedies going beyond those offered in the cases cleared under Commissioner Almunia.
A similar scenario played out in the UK case. To alleviate the Commission’s concerns, Hutchison proposed an innovative package of structural and behavioural remedies (again, going beyond the Almunia-era remedies). This included an offer to divest O2’s stake in the Tesco Mobile joint venture and to offer a wholesale agreement for a share of its network capacity to Tesco Mobile, plus to grant a perpetual fractional network interest of up to [10-20]% of capacity on the merged entity‘s network to one or two MVNOs. Hutchison also offered behavioural commitments to remedy the Commission’s concerns relating to the network sharing agreements between the UK MNOs. On both occasions, Commissioner Vestager made public statements that the Commission could not approve the merger because the parties did not offer structural remedies and, specifically, an MNO remedy.[52] Shortly after blocking the UK merger, the Commission cleared a four-to-three mobile merger in Italy, but only on the condition that the parties divest assets needed to create a MNO.[53] Based on this remedy, Iliad entered the market with aggressive price plans in 2018 (roughly two years after the merger was cleared). By 2020, Iliad had 6.3 million subscribers in Italy, corresponding to 8% market share.[54]
As mentioned above, these cases were followed by an unconditional clearance in the Dutch case, which could create the impression that if the Commission were to find that a mobile merger results in a SIEC, an MNO remedy would be required.
The General Court’s judgment
The General Court delivered its judgment in CK Telecoms four years after the Commission delivered its deadly blow to the UK merger. The Court criticises the Commission’s approach for the lack of any limiting principles for finding a “significant” impediment and makes clear that it is not sufficient for the merging parties to be “close” and “important” competitors to find a SIEC. The Court notes that one cannot assume “unilateral effects” or a SIEC merely on the basis that the market is characterised by a high degree of concentration and the transaction would further increase market concentration levels by reducing the number of competitors from four to three.[55] In other words, there could be no presumption or a lower standard of proof to show that a four-to-three merger will lead to a SIEC. No magic number means no magic number.
The standard of proof
Many of the Court’s findings are relevant beyond mobile mergers and unilateral effects cases. This relates in particular to the passages of judgment relating to the standard of proof required to show a SIEC. It is also one of the key aspects of the judgment challenged by the Commission on appeal.[56]
Referencing General Electric[57], Bertelsmann and Sony Corporation of America v Impala [58], and Airtours[59], the General Court sets out an analytical framework requiring a two-step analysis. In the first step, the Commission must evaluate which “economic outcome attributable to the concentration” is most likely to ensue.[60] In that regard, the Court notes that the “scenarios and theories of harm [considered by the Commission] must appear sufficiently realistic and plausible, and cannot therefore solely be conceivable from a theoretical point of view, in the light of an analysis of all the relevant factors.”[61] In the second step, the Commission must assess whether that economic outcome is likely to lead to a SIEC. [62] In that regard, siding with the more restrictive view as to the standard of proof that the Commission is required to meet to block a merger,[63] the Court found that “the Commission is required to produce sufficient evidence to demonstrate with a strong probability the existence of significant impediments following the concentration.“[64] According to the General Court, that standard of proof is “stricter than that under which a significant impediment to effective competition is ‘more likely than not’, on the basis of a ‘balance of probabilities’, as the Commission maintains.” While, as the General Court stressed, the Commission is not obliged to prove its case “beyond all reasonable doubt”,[65] the standard of proof sets a relatively high bar for the Commission to challenge a merger.
Novel theories of harm (such as the theory of harm developed by the Commission in relation to the network sharing agreement) may be subject to a higher standard of proof. In that regard, the Court reasoned: “the more prospective the analysis is and the chains of cause and effect dimly discernible, uncertain and difficult to establish, the more demanding the EU judicature must be in terms of the specific examination of the evidence produced by the Commission.“[66] This has had an impact on the Court’s assessment of the Commission’s allegations relating to the fact that the merged entity would have participated in two different network sharing agreements.
The requirements to show a SIEC in a unilateral effects case
In relation to the non-coordinated effects theory of harm, the Court starts its reasoning by noting that the legislative process for and the wording of the EUMR in its current form indicate that the non-coordinated effects must be “equivalent” to the creation or strengthening of a dominant position in order to justify an intervention.[67]
Based on recital 25 of the EUMR, the Court infers that non-coordinated effects could be established if the transaction leads to “(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’”.[68] Moreover, the Court makes clear that a SIEC can be established only if both conditions are present: “that the mere effect of reducing competitive pressure on the remaining competitors is not, in principle, sufficient in itself to demonstrate a significant impediment to effective competition in the context of a theory of harm based on non-coordinated effects”.[69] The Court then applies this test to raise the bar for the Commission to establish a SIEC on the basis that one of the parties is an “important competitive force”, that the parties are “close” competitors, and on the basis of quantitative analysis, i.e., the key tools used by the Commission to analyse unilateral effects in all four-to-three mobile mergers. This is addressed in turn below.
The crime of closeness
The General Court firmly rejected the Commission’s assertion that a SIEC could be inferred just from the fact that the parties to the transaction are “close” competitors, in particular in a market characterised by a low level of product differentiation (like mobile telephony), where all competitors are close to some degree.
The Court acknowledges that the concept of “closeness” is relevant for the assessment of the unilateral effects of the merger[70] and that, under certain circumstances, in markets comprising products with less-close substitutes, closeness may be even more important to the assessment of the transaction than market shares.[71] However, in markets characterised by a low degree of product differentiation, where effectively all competitors are relatively close, closeness alone is insufficient to establish that the transaction would lead to the elimination of important competitive constraints between the Parties (and therefore would satisfy the first prong of the test spelled out in recital 25 of the EUMR).[72] The Court went on to stress if that were the test to show a SIEC, “any concentration resulting in a reduction from four to three operators as a matter of principle be prohibited.”[73]
The judgment clarifies further that in homogenous markets (such as mobile markets), closeness is relevant if it leads to the elimination of an important competitive constraint that the parties exercise upon each other, which in turn results in the parties being “particularly close”.[74] The Court found that the Commission had not shown to the requisite standard of proof that this condition was met. In particular, the evidence adduced by the Commission had only shown that Three and O2 were “relatively close” (and, indeed, the Commission asserted in its decision that all MNOs were “close”). In particular, the diversion ratios indicated that customers switching away from Three were more likely to switch to other operators than to O2; similarly, O2’s customers were more likely to choose other operators than Three.[75] Thus, while the Court does not specify what it means to be “particularly close”, it clearly raises the threshold from the Commission to show how closeness might contribute to a SIEC in a mobile merger cases. Further, as the Dutch case illustrates, the Commission appears to be open to an argument that relative closeness should not be seen as an aggravating factor where all operators are close.
“Important competitor” vs. “Important competitive force”
The General Court found that the EC had not established to the requisite standard that Three was an “important competitive force” and that the finding that Three exerted “important competitive constraint” (or was an “important competitor”) was not in itself sufficient to establish that the transaction would lead to a SIEC.
The concept of an “important competitive force” is defined rather vaguely in the Horizontal Merger Guidelines, as a market player that has “more of an influence on the competitive process than their market shares or similar measures would suggest.”[76] The Guidelines give two examples of a situation where one of the market players could be considered an “important competitive force”: (i) “a recent entrant that is expected to exert significant competitive pressure in the future” and (ii) an important innovator with promising pipeline products (in markets where innovation is important).[77] As explained above, the Commission adopted a rather broad interpretation of an “important competitive force” both in the UK case and in other mobile mergers. On appeal, the Commission argued that a company does not need to be a maverick or “stand-out” to be considered “an important competitive force”, which one could view as a relatively exceptional situation. Rather the test should be whether the company contributes “substantially and consistently, to the competitive process on the market based on parameters such as price, quality, choice and innovation” – a significantly lower threshold.[78]
The General Court rejected that interpretation, asserting that it effectively amounted to confusing the concepts of “significant impediment to effective competition” (Article 2(3) of the EU Merger Regulation), “elimination of an important competitive constraint” (recital 25 of the EU Merger Regulation), and “elimination of an important competitive force” (paragraph 37 of the EC’s Horizontal Merger Guidelines).[79] In the Court’s view, such an approach effectively results in a situation where “any elimination of an important competitive force would amount to the elimination of an important competitive constraint which, in turn, would justify a finding of a significant impediment to effective competition.”[80] Instead, in the Court’s view, the finding of a SIEC could be made only if the transaction leads to the elimination of an “important competitive force”, understood as a competitor that “stands out” from its competitors in terms of its impact on competition.[81] The Court further found 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 a significant impediment to effective competition”.[82]
As mentioned above, in all mobile mergers and in the UK case in particular, the Commission has attached particularly strong importance to the parties’ share of gross adds and subscriber growth. The Court found in that regard that the evidence adduced by the Commission in relation to Three’s growth (and in particular the fact that Three’s share of gross adds exceeded its market share) was insufficient to establish that Three was an “important competitive force”. Although Three grew in the market, its growth (in the range of 1% per year) was not particularly fast. Further, the Court found that the Commission’s analysis of Three’s record as an innovator in the market had been anecdotal. Notably, the Court reasoned that the disruptive role that Three may have historically played on the market was insufficient to show that it would continue to do so in the future. In doing so, the Court implies that the conclusion that one of the merging parties is “an important competitive force” cannot be based solely on historical data. The Commission must conduct appropriate forward-looking analysis of the counterfactual, i.e., the situation on the market in the absence of the transaction. As mentioned above, such a forward-looking analysis in the Dutch case suggested that the ability of one of the merging parties will deteriorate in the future. This factor strongly contributed to the unconditional clearance.
The ruling does not spell out the specific conditions needed to establish whether a merging party “stands out” from its competitors. Neither does it suggest that it is necessary to show that one of the parties is an important competitive force for the transaction to lead to a SIEC. However, the Court rejected the assertion that the mere fact that one or both parties significantly contribute to the competitive process is sufficient to establish that the transaction would lead to “a reduction of competitive pressure on the remaining competitors”. The Court also examined in detail the specific factors taken into account by the Commission to show that Three was an “important competitive force” and the related evidence adduced by the Commission against the test it established. What the judgment makes clear in that regard is that being a credible competitor, having a higher share of gross adds than market share, or having strengthened one’s market position are in themselves insufficient grounds to conclude that the party is an important competitive force or that its removal would result in a SIEC. All this points to the conclusion that, in line with CK Telecoms, the notion of an “important competitive force” is reserved for an extraordinary competitor that one could expect to drive the competitive process in the future.
The higher standard needed to establish that a merging party is an “important competitive force” also meant that the Commission did not discharge the burden of proof in relation to its allegations relating to the impact of the transaction on the wholesale market. As indicated below, Three’s share in the wholesale market was minimal, but the Commission asserted that Three was “an important competitive force” because Three was actively bidding for new wholesale contracts and had a gross add share that was “disproportionately high” in comparison with its market share.[83] The Court dismissed that finding (in particular, because the Commission did not demonstrate why Three’s share of gross adds was of such importance) and ruled that the mere fact that Three was an “important competitor” in the wholesale market alone was insufficient to establish a SIEC.[84]
That said, CK Telecoms should not be read as indicating that the Commission must prove that one or both merging parties are important competitive forces to establish that the transaction reduces competitive pressure on the remaining competitors. In other words, elimination of an “important competitive force” is only one possible way of establishing a SIEC based on a unilateral effects theory of harm. It seems plausible that the Commission could instead focus, for example, on how the ability and the incentives of other MNOs to compete would change as a result of the transaction or that it would assert that the reduction of competitive pressure on the remaining competitors is proved by the results of its merger simulation.
Price effects must be “significant” and the analysis must reflect efficiencies
CK Telecoms goes on to closely scrutinise the economic models the Commission used to assess the effects of the transaction, and the Commission’s methodology in that regard. The judgment echoes some of the criticism in relation to the quantitative models used by the Commission to assess mobile mergers. The Court makes clear in particular that the mere fact that the quantitative analysis shows a price increase is not in itself sufficient to show a SIEC.[85] The Court notes that the IPR tool “may be useful for screening purposes”, but cannot “be regarded as credible forecast of price increases” and therefore, of a SIEC.[86] The Court accepted, however,, that the more sophisticated model (i.e., a merger simulation, which takes into account the likely reaction of competitors to price increases and which the Commission carried out in Phase II investigation in the UK case) could be suitable for this purpose.[87]
This should not be a particularly controversial finding, given that the Commission itself acknowledged that its IPR analysis should not be regarded as an exact quantification of the price effects resulting from the transaction, but only an “indication of likelihood” of such increases, and therefore was not regarded as decisive evidence by the Commission.[88] There was, however, no certainty as to whether any level of price increase (which the model was bound to show) would be treated as an indication of a SIEC or whether a negligible price increase would not be treated as such. The ruling makes clear that, in order to establish a SIEC, the Commission must demonstrate with “a sufficient degree of probability” that the negative price effects resulting from the transaction would be “significant”. The Court noted that, given that the Commission’s model would always result in some price increase in the absence of efficiencies, the magnitude of such increases is relevant and the Commission’s quantitative analysis must show that prices would rise “significantly” to have probative value to establish potential unilateral effects resulting from the merger.[89] There is no guidance in the judgment as to what level of price effects should be viewed as such, but presumably low single-digit increases, such as those calculated by the Commission in the Dutch case (cleared unconditionally),[90] would be viewed as “insignificant”.
Finally, and this is one of the more controversial aspects of the judgment, the General Court found that “standard efficiencies” must be reflected in the Commission’s model. The Court distinguishes the notion of “standard efficiencies” from the efficiencies referred to in Section VII of the Horizontal Merger Guidelines that the parties must demonstrate if they want to claim an efficiency defence. These “standard efficiencies” should be taken into the account in the Commission’s competitive assessment and, specifically, in its quantitative analysis of the transaction’s price effects. According to the General Court, the Commission confused these two types of efficiencies, to the detriment of the merging parties.[91] The Court reasons that since “any concentration will lead to efficiencies”, such standard efficiencies should be “a component of a quantitative model designed to establish whether a concentration is capable of producing such restrictive effects”.[92]
This is an important development, given that the Commission has traditionally relied on efficiencies being an aspect of a merger for which the parties bear the full burden of proof. The Commission typically dismissed most of the efficiencies claimed by the parties to a transaction as not proven. In mobile mergers, most of the efficiencies related to merging the networks were deemed as not “merger specific”, as the Commission took the view that the parties could achieve such efficiencies by entering into network sharing agreements. The Commission has also traditionally dismissed all fixed-cost savings, on the ground that they would not be passed on to consumers.
In principle, the fact that the Commission would have to take into account more efficiencies in its quantitative analysis means that it would be more difficult for it to establish “substantive” price effects. That said, it should not be assumed that the Commission would account for any and all efficiency claims made by the parties in its quantitative analysis. Some form of filtering will likely be required. And since the judgment does not clarify what should be viewed as “standard efficiencies”, there is some uncertainty as to how it will be implemented and whether it will make a significant difference in the assessment of mobile mergers.
Outlook
CK Telecoms delivered a blow to the framework developed by the Commission to assess mobile mergers. The Court’s interpretation of the notion of “important competitive force” significantly raises the threshold for the Commission to challenge the case on the basis that one of the merging parties is an aggressive competitor. It also makes it impossible for the Commission to challenge a mobile merger solely on the ground that the parties are close competitors. The judgment’s criteria to establish “unilateral effects” thus truly implements the principle that there is no “magic number” of MNOs.
That said, the Commission remains deeply suspicious of four-to-three mobile mergers and it will continue to scrutinise such transactions very closely. In its appeal to the Court of Justice, the Commission contested all key aspects of the General Court’s reasoning.[93] Comments from high-ranking Commission officials make clear that the Commission will not radically change the way it analyses mobile mergers. It will still apply the Horizontal Merger Guidelines and look at factors similar to those it examined in the past. Until the Court of Justice decides on the Commission’s appeal, we should expect no revolution in the Commission’s approach to mobile mergers. What is more, CK Telecoms does not address remedies, the key topic for merging parties in potentially contentious merger cases. In that regard, as discussed above, the Commission appears to have a strong preference for structural remedies that would allow the entry of a new MNO. That strong preference will very likely persist: there is no indication that CK Telecoms will mean a return to MVNO-style remedies. This means that if the Commission asserts that there are concerns and a remedy has to be found to address such concerns, more likely than not a structural remedy will be required.
Given the judgment’s criticism and the high standard of proof that it sets for the Commission, it is likely that the Commission would conduct an even more detailed assessment of mobile merger cases. In practice, this means longer pre-notification periods and more information requests for both the parties and other market players.
On a positive note, the Commission could be expected to engage more closely with the evidence submitted by the parties (both in relation to competitive dynamics and theories of harm and in relation to the efficiencies) and focus on a more forward-looking analysis. All in all, while telecommunications companies should not expect an easy ride in Brussels, the unconditional clearance in the Dutch case and CK Telecoms indicate that the Commission should be more open to clearing four-to-three mergers where there are specific market circumstances that limit the potential harm and where there is a strong pro-competitive rationale.
Katarzyna Czapracka represented Telia in Case COMP/M.7419 and Deutsche Telecom in Case COMP/M.8792. The views expressed in this post are entirely those of the author and cannot be attributed to White & Case LLP or its clients. Warm thanks to Thilo Klein, Tilmann Kuhn, Thilo Wienke and Mark Powell for their comments on an earlier version of this post.
[1] Case T-399/16 – CK Telecoms UK Investments v Commission, ECLI:EU:T:2020:217 (“CK Telecoms”), annulling the Commission decision in Hutchison 3G UK/Telefonica UK, Case COMP/M.7612 (“the UK Decision”).
[2] TeliaSonera/Telenor/JV, Case COMP/M.7419, abandoned by the parties after failed remedies negotiations. See the statement of Commissioner Vestager available at: https://ec.europa.eu/commission/presscorner/detail/en/STATEMENT_15_5627.
[3] The UK Decision.
[4] Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016.
[5] T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018.
[6] Case T-342/99, Airtours plc v Commission, ECLI identifier: ECLI:EU:T:2002:146. For comments on the broader implications of the judgment see e.g., Tilman Kuhn, Stefan Thomas, The More Economic Judicature: How the General Court has Recalibrated the Merger Gauge, Competition Policy International, 7 June 2020.
[7] For example, Guillaume Loriot (Director – Markets and cases II: Information, Communication and Media, DG COMP) said that CK Telecoms was “wrong in law” and that the General Court’s interpretation of the concepts of “important competitive force” and “closeness of competition” contradicted the Horizontal Merger Guidelines. Mr Loriot also suggested that the judgment would not “change a lot” in terms of how the Commission analyses mobile mergers. See the transcript of the Concurrences seminar “The Future of EU and UK Merger Control”, 17 September 2020, available at: https://www.concurrences.com/en/conferences/the-future-of-eu-and-uk-merger-control-enforcement; European Commission’s Chief Competition Economist, Pierre Régibeau, criticised the judgment for creating a presumption that mergers “generally” create efficiencies, Interactive Brussels Conference: The Bigger Picture, GCR, 22 September 2020. Tommaso Valletti, who was the Chief Competition Economist when the Commission adopted the UK Decision, criticised the Court for committing several “conceptual and factual mistakes” in his LinkedIn post.
[8] See the remarks of Guillaume Loriot cited in FN 9 above.
[9] Pending appeal case before the Court of Justice C-376/20 P. The summary of the Commission’s appeal has been published in OJ C390, 16.11.2020, p. 20.
[10] Three and EE had combined their networks in the Mobile Broadband Network Limited (“MBNL”) joint venture. O2 and Vodafone had combined their networks to create “Beacon”. The Commission’s theory was that given its unique market position, the merged entity could de-prioritise one of the shared networks without harming its own services to the same extent as those of the relevant network sharing partner.
[11] Hutchison 3G Austria/Orange Austria, Case COMP/M.6497, Commission decision of 12 December 2012; Hutchison 3G UK/Telefónica Ireland, Case COMP/M.6992, Commission decision of 28 May 2014; Telefónica Deutschland/E-Plus, Case COMP/M.7018, Commission decision of 2 July 2014; TeliaSonera/Telenor/JV, Case COMP/M.7419, abandoned by the parties after requests by the EC to create a fourth mobile network operator to remedy competition concerns; Hutchison 3G UK/Telefonica UK, Case COMP/M.7612, Commission decision of 11 May 2016; Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016 and T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018.
[12] When merger control was first introduced at the EU level in the 1989 Merger Regulation, the substantive test under which mergers were assessed was whether they would “create or strengthen a dominant position as a result of which effective competition would be significantly impeded”. In other words, the creation or strengthening of market power or dominance was at the heart of the assessment. This focus on dominance was seen as a potential shortfall, an enforcement “gap” with regard to non-collusive mergers in oligopolistic markets. Thus, the currently binding 2004 EU Merger Regulation (“EUMR”), as part of the broader efforts to reform EU competition rules and introduce more “effects-based” enforcement, introduced instead the Significant Impediment of Effective Competition (“SIEC”) test, shifting the focus of the test away from dominance. For the debate on the topic, see e.g., J Fingleton and D Nolan, “Mind the Gap: Reforming the EU Merger Regulation”, Mercato, Concorrenza, Regole, 29 May 2003 and LH Röller and M De La Mano, “The Impact of the New Substantive Test in European Merger Control”, European Competition Journal, April 2006, p. 9.
[13] With the exception of the Danish and Italian cases, the Commission did not find sufficient evidence to show that the mobile deals it reviewed would also lead to coordinated effects.
[14] Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (the “EUMR”), OJ L 24, 29.1.2004, p. 1.
[15] Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, OJ C 31, 5.2.2004, p. 5 (“Horizontal Merger Guidelines”).
[16] Horizontal Merger Guidelines, at paragraphs 27-38.
[17] Horizontal Merger Guidelines, at paragraph 28.
[18] Id.
[19] LH Röller and M De La Mano, “The Impact of the New Substantive Test in European Merger Control”, European Competition Journal, April 2006, pp. 14-15.
[20] See, e.g., Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016, at paragraph 783 and T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018, at paragraph 352.
[21] UK Decision,, at paragraph 438.
[22] UK Decision, at paragraph 324.
[23] Horizontal Merger Guidelines, at paragraph 37.
[24] See, e.g., Telefónica Deutschland/E-Plus, Case COMP/M.7018, Commission decision of 2 July 2014, para. 121.
[25] Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016, at paragraph 432.
[26] For a summary of the Commission’s view on mobile mergers see, e.g., L Manigrassi, E Ocello, V Staykova, “Recent Developments in Telecoms Mergers”, Competition Merger Brief 3/2016; B. Buehler, T. Büttner, E. Ocello and M. Piergiovanni, “Europe – The Final Countdown: Five, four, three … no, wait!”, 2017, Competition Law & Policy Debate, 3(4), p. 18.)
[27] Telefónica Deutschland/E-Plus, Case COMP/M.7018, para. 460. By contrast, in the Dutch case, which the Commission cleared unconditionally in 2018, the Commission found that the target company was not an “important competitive force.” T-Mobile NL/Tele 2 NL, Case COMP/M.8792, para. 443.
[28] UK Decision, at paragraph 681.
[29] See, e.g., ibid., at paragraphs 499-521 and 640-650.
[30] UK Decision, at paragraphs 636 (retail market) and 1875 and 1920 (wholesale market).
[31] See, e.g., Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016 and Hutchison 3G UK/Telefonica UK, Case COMP/M.7612, Commission decision of 11 May 2016.
[32] According to the submissions of the notifying party, Three hosted only some minor wholesale customers and struggled to attract serious interest from existing large and new ambitious wholesale customers. See, e.g., ibid., at paragraphs 888-895, 1840-1844.
[33] Ibid., at paragraph 1920.
[34] Ibid., at paragraph 2301.
[35] See T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018, at paragraph 816.
[36] Specifically, these related to the savings related to the elimination of the national roaming agreement between TMNL and Tele2 NL. Tele2 NL only had a limited network and had to rely on roaming on TMNL’s network to achieve nationwide coverage, which would no longer be necessary post-merger.
[37] Ibid., at paragraph 1209.
[38] Ibid., ANNEX A: THE COMMISSION’S QUANTITATIVE ANALYSIS, at paragraphs 242-244.
[39] Ibid., ANNEX A: THE COMMISSION’S QUANTITATIVE ANALYSIS, at paragraphs 245-257.
[40] See, e.g., S Vande Walle and J Wambach, “No magic number to dial – The Commission’s review of mobile telecoms mergers”, Competition Merger Brief 1/2014, p. 14, and “Statement by Commissioner Vestager on competition decision to prohibit Hutchison’s proposed acquisition of Telefónica UK”, 11 May 2016, accessible at: https://ec.europa.eu/commission/presscorner/detail/en/STATEMENT_16_1713.
[41] T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018.
[42] Ibid., at paragraph 806.
[43] Ibid, at paragraphs 724 and 775-781. These price effects were lower than those in other mobile mergers.
[44] T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018, at paragraph 720.
[45] Ibid, at paragraph 721. See also section 1.4 below.
[46] Ibid., at paragraph 443.
[47] Ibid, at paragraphs 816-817 and 823.
[48] Hutchison 3G Austria/Orange Austria, Case COMP/M.6497, Commission decision of 12 December 2012; Hutchison 3G UK/Telefónica Ireland, Case COMP/M.6992, Commission decision of 28 May 2014; Telefónica Deutschland/E-Plus, Case COMP/M.7018, Commission decision of 2 July 2014.
[49] Notably, 1&1 Drillisch, the remedy taker in Telefónica Deutschland/E-Plus, Case COMP/M.7018, Commission decision of 2 July 2014, acquired 5G spectrum in 2019.
[50] The Commission has repeatedly stressed its preference for structural remedies and became increasingly sceptical of the effectiveness of MVNO remedies to resolve perceived competition concerns, in particular because it took the view that they raised issues at the implementation stage. See, e.g., L Manigrassi, E Ocello, V Staykova, “Recent Developments in Telecoms Mergers”, Competition Merger Brief 3/2016.
[51] TeliaSonera/Telenor/JV, Case COMP/M.7419, abandoned by the parties.
[52] See Statement by Commissioner Vestager on competition decision to prohibit Hutchison’s proposed acquisition of Telefónica UK, 11 May 2016, and Statement by Commissioner Vestager on announcement by Telenor and TeliaSonera to withdraw from proposed merger, 11 September 2015.
[53] Hutchison 3G Italy/Wind/JV, Case COMP/M.7758, Commission decision of 1 September 2016. For an overview of the Commission Decisions in four-to-three mobile merger cases, see K Czapracka and M Powell, “The shifting sands of EU merger control – un, deux, trois, piano!,” Competition Law & Policy Debate, 3(4), November 2017, pp. 41-53.
[54] See https://www.iliad.fr/finances/2020/CP_030920.pdf.
[55] CK Telecoms, paragraph 434.
[56] See the summary of the appeal brought by the Commission in Case C-376/20 P, published in OJ C390, 16.11.2020, p. 20.
[57] General Electric, op. cit., para. 464.
[58] Bertelsmann and Sony Corporation of America v Impala, Case C-413/06 P, EU:C:2008:392, para. 51.
[59] Airtours, op. cit., para. 59.
[60] CK Telecoms, para. 113.
[61] Ibid., paras. 117.
[62] Ibid., para. 115.
[63] The issue was controversial in other merger cases reviewed by the EU Courts and previously not explicitly addressed in judgments. In Commission v Tetra Laval, AG Tizzano opined in that the EC may only intervene if “on the basis of solid elements gathered in the course of a thorough and painstaking investigation, and having recourse to its technical knowledge, the Commission is persuaded that the notified transaction would very probably lead to the creation or strengthening of such a dominant position.” (opinion of Advocate General Tizzano in Commission v Tetra Laval, C-12/03 P, EU:C:2004:318, para. 74). By contrast, in AG Kokott’s opinion in Bertelsmann and Sony Corporation of America v Impala, it was considered “inappropriate to set the bar higher [than ‘more likely than not’] in the context of merger control and, for example, to require that the market development envisaged by the Commission should be ‘very probable’ or ‘particularly likely’ in order to be accepted by the Court. Such a higher standard of probability would seriously weaken the Commission in carrying out its competition policy functions.” (Opinion of Advocate General Kokott in Bertelsmann and Sony Corporation of America v Impala, C-413/06 P, EU:C:2007:790, para. 210).
[64] CK Telecoms, paragraph 118.
[65] Ibid.
[66] Ibid., para. 332.
[67] CK Telecoms, paragraphs 85 et seqq.
[68] CK Telecoms, paragraph 96.
[69] Ibid., para. 97.
[70] CK Telecoms, at paragraph 241.
[71] CK Telecoms, at paragraph 239.
[72] CK Telecoms, at paragraph 242.
[73] CK Telecoms, at paragraph 249.
[74] CK Telecoms, at paragraph 247.
[75] CK Telecoms, at paragraphs 244 and 248.
[76] Horizontal Merger Guidelines, paragraph 37.
[77] Ibid., paragraphs 37-38.
[78] CK Telecoms, paragraph 167.
[79] CK Telecoms, paragraph 173.
[80] Ibid.
[81] CK Telecoms, at paragraph 174.
[82] CK Telecoms, paragraph 173.
[83] UK Decision, paragraphs 1868 to 1920.
[84] CK Telecoms, paragraph 446.
[85] CK Telecoms, at paragraph 268.
[86] CK Telecoms, at paragraph 256.
[87] CK Telecoms, at paragraph 258.
[88] CK Telecoms, at paragraphs 267-268.
[89] CK Telecoms, at paragraph 282.
[90] The calibrated merger simulation conducted in the Dutch case showed price increases in the range of 0-5% at the provider level and 5-10% at the network level; see T-Mobile NL/Tele 2 NL, Case COMP/M.8792, Commission decision of 27 November 2018, at paragraph 829.
[91] CK Telecoms, at paragraph 279.
[92] Ibid.
[93] The summary of the Commission’s appeal has been published in OJ C390, 16.11.2020, p. 20.
________________________
To make sure you do not miss out on regular updates from the Kluwer Competition Law Blog, please subscribe here.
