BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom Competition Appeals Tribunal


You are here: BAILII >> Databases >> United Kingdom Competition Appeals Tribunal >> Hutchison 3G (UK) Ltd v Office of Communications [2005] CAT 39 (29 November 2005)
URL: http://www.bailii.org/uk/cases/CAT/2005/39.html
Cite as: [2005] CAT 39

[New search] [Printable RTF version] [Help]



     

    Hutchison 3G (UK) Limited v Office of Communications [2005] CAT 39 (29 November 2005)

    Case No: 1047/3/3/04

    IN THE COMPETITION

    APPEAL TRIBUNAL

    Victoria House,

    Bloomsbury Place,

    London WC1A 2EB

    29 November 2005

    Before:
    THE HONOURABLE MR JUSTICE MANN (CHAIRMAN)
    MR ADAM SCOTT TD
    PROFESSOR PAUL STONEMAN
    Sitting as a Tribunal in England and Wales

    BETWEEN:

    HUTCHISON 3G (UK) LIMITED

    Appellant

    and
    THE OFFICE OF COMMUNICATIONS

    Respondent

    Supported by
    BRITISH TELECOMMUNICATIONS PLC

    Intervener

    Mr. Nicholas Green QC (instructed by Freshfields Bruckhaus Deringer) appeared for the Appellant

    Mr. Peter Roth QC and Miss Kassie Smith (instructed by The Director of Legal Services (Competition), Office of Communications) appeared for the Respondent.

    Mr. Gerald Barling QC and Miss Sarah Stevens (instructed by BT Legal) appeared for the Intervener.

    JUDGMENT

    I INTRODUCTION

  1. This is an appeal by Hutchison 3G (UK) Limited ("H3G"), under section 192 of the Communications Act 2003 ("the 2003 Act"), against the determination by the Office of Communications ("OFCOM"), made under sections 48 and 79 of the 2003 Act, that H3G has significant market power ("SMP") in the market for wholesale mobile voice call termination on H3G's network and imposing certain reporting obligations. That determination is set out in paragraph 2(a) of Annex A to OFCOM's statement entitled "Wholesale Mobile Voice Call Termination", dated 1 June 2004, ("the Decision"). H3G appeals the determination in the Decision that it has SMP (and therefore the imposition of a remedy); it does not appeal any other parts of or findings in the Decision.
  2. II THE NEW REGULATORY FRAMEWORK

  3. This appeal is taking place in the context of a new regulatory framework for electronic communications networks and services which entered into force in the United Kingdom on 25 July 2003. The basis for the new framework is five new EU Communications Directives that are designed to create harmonised regulation across Europe:
  4. Directive 2002/21/EC on a common regulatory framework for electronic communications networks and services ("the Framework Directive")[1];

    Directive 2002/19/EC on access to, and interconnection of, electronic communications networks and associated facilities ("the Access Directive")[2];

    Directive 2002/20/EC on the authorisation of electronic communications networks and services ("the Authorisation Directive")[3];

    Directive 2002/22/EC on universal service and users' rights relating to electronic communications networks and services ("the Universal Service Directive") and[4];

    Directive 2002/58/EC concerning the processing of personal data and the protection of privacy in the electronic communications sector ("the Privacy Directive")[5].
  5. The first four of these Directives were adopted by the European Parliament and Council on the 7 March 2002 and came into force on the 24 April 2002. Those four Directives were implemented in the United Kingdom ("the UK") on 25 July 2003 by virtue of the Communications Act 2003.[6] These Directives are complemented by EU Decision No 676/2002/EC of the European Parliament and of the Council of 7 March 2002 on a regulatory framework for radio spectrum policy in the European Community (Radio Spectrum Decision).
  6. The new regulatory framework provides an environment in which the electronic communications sector moves towards effective competition in which state ex ante regulation will no longer be required. However, as recital 25 to the Framework Directive explains, it is envisaged that in certain circumstances there will be an ongoing need for ex ante regulation to ensure the development of competitive markets.
  7. III THE FACTUAL CONTEXT OF THE RELEVANT MARKET

  8. The relevant market in this case is wholesale voice call termination provided by H3G on its network. The market definition is not challenged in this appeal. In order to be able to consider the Decision it is necessary to understand the practical context of the market. This section sets out that background.
  9. In 2000 H3G was awarded one of the third generation ("3G") mobile telephony licences (as were other operators) enabling it to operate a mobile telephone network using the 3G technology. There were already several operators of second generation ("2G") networks. H3G is itself part of a major international group which itself once had a 2G operation (Orange). It started operating its 3G network in 2003 and the group can in no way be thought to be a small or inexperienced organisation.
  10. Understanding the relevant market in this case involves knowing how calls arrive on its network. When a caller not directly connected to a given mobile network calls one of that network's numbers the call has to be routed on to that network either directly from his/her own network (the "originating" network") or via another ("transit") operator. Such a call will either originate on a fixed line network (such as BT or Kingston) known as a "fixed to mobile" call, or on another mobile network known as a "mobile to mobile" call. The process of connecting those calls through the called mobile network is called "termination". Terminating the call through that network allows it to be passed to the relevant number. The receiving network makes a charge for that termination – a "termination charge". The UK operates a "calling party pays" ("CPP") system, which means that the entire cost of a call is born by the party who makes the call. The charge will include the charge made by the calling party's originating network, the charge made by any other transit network through which the call is passed (typically, under present arrangements, BT for non-BT originated calls) and the termination charge. Those elements are not explicitly separated out in the bill presented to the caller – he/she will merely be charged one overall amount based on retail prices. Although the information necessary to work out what the charges are is for the most part publicly available, most callers will be ignorant of the way in which the charge is made up and therefore what part comprises the termination charge.
  11. In order to be able to complete the delivery of the call on to the relevant mobile network various technical and commercial arrangements have to be put in place. Such interconnection between networks is governed partly by standard interconnect agreements and partly by additional contractual terms negotiated between the network operators. As will appear below, these agreements take place within a statutory framework which regulates, or at least is capable of providing regulation of, the terms on which that interconnection takes place. As part of the regulation of the sector, BT has been and is obliged to provide what is called "end to end connectivity", and its charges to fixed and mobile network operators ("FNOs and MNOs") are subject to price control.
  12. Over the past 7 years or so the level of termination charges made by MNOs in respect of fixed line to mobile calls, or mobile to mobile off-net calls, has attracted the attention of those concerned to protect the interest of consumers. It is not necessary for us to go into great detail of those historical concerns and their associated investigations; however, the following paragraphs set out a brief background to the appeal.
  13. In 1998 the Director General of Telecommunications referred to the Monopolies and Merger Commission (the "MMC") the matter of whether certain MNO's fixed to mobile call termination charges might be expected to operate against the public interest and, if so, whether those effects could be remedied or prevented by modifications to their licences. The MMC concluded that there was insufficient competitive constraint on termination charges and that the MNO's charges operated against the public interest. The MMC considered that the only effective means of remedying or preventing the adverse effects would be to impose a price control on termination charges. Following this, amendments were made by the Director to the MNO's respective licences to include a charge control on wholesale mobile call termination charges, to last until 2002.
  14. The Director subsequently carried out a review of the 2G (but not 3G) call termination charges of the MNOs and in 2002 a further reference was made to the Competition Commission (the "CC"). The CC concluded that the call termination charges of each of the MNOs operated against the public interest, with the adverse effects that the termination charges for 2002/3 were 30 to 40 per cent in excess of the CC's estimation of the fair charge. The CC considered that, in the absence of a charge control on those charges, the charges may be expected to operate against the public interest, with the termination charges expected to be up to double the level of the fair charge by 2005/06 if the charges were retained at their current level in real terms. The CC also considered that each operator had a monopoly of call termination on its own network and that competitive pressures at the retail level did not constrain termination charges.
  15. Three of the MNOs applied for and were unsuccessful in a judicial review of the CC's recommendations and the Director's decisions to modify their licences by imposing a charge control and to continue those controls by way of continuation notices after the implementation of the new Framework on 25 July 2003.[7]
  16. IV THE PROCEDURAL HISTORY OF THIS MATTER AND THE MAIN FINDINGS IN THE DECISION

  17. Pursuant to the Framework Directive, the European Commission established guidelines for the assessment of SMP. The Commission Guidelines on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services, OJ [2002] C165/6 (the "Commission Guidelines") were published on 11 July 2002.
  18. Pursuant to Article 15(1) of the Framework Directive, on 11 February 2003, the Commission published a Recommendation on Relevant Product and Services Markets within the electronic communications sector susceptible to ex ante regulation in accordance with Directive 2002/21/EC of the European Parliament and of the Council on a common regulatory framework for electronic communication network and services[8] (the "Commission Recommendation") and an Explanatory Memorandum to the Commission Recommendation (the "Explanatory Memorandum"). The Commission's Recommendation identified a set of markets in which ex ante regulation may be warranted. The Commission Recommendation seeks to promote harmonisation across the European Community by ensuring that the same product and service markets are subject to a market analysis in all Member States. NRAs are, however, able to regulate markets that differ from those identified in the Commission Recommendation where this is justified by national circumstances. Accordingly, NRAs are to define relevant markets appropriate to national circumstances, provided that due account is taken of the product markets listed in the Commission Recommendation.[9]
  19. Oftel also published market review guidelines[10] in August 2002 which Oftel, and subsequently OFCOM, took account of in carrying out market reviews. The Oftel Guidelines explain the criteria that Oftel intended to use to assess the existence of SMP when conducting market reviews in accordance with new EU Communications Directives.
  20. As required under the new European regulatory framework, in 2003 Oftel began a market review process to define the relevant markets for mobile wholesale voice call termination and to assess whether H3G or any of the other MNOs had SMP on any such market. This process led to the making of the Decision and comprised a series of consultations with two intermediate reviews (in May 2003 and December 2003) on which responses were invited before the Decision was made and published in May 2004. H3G participated extensively in both rounds of consultation. It is necessary to refer to this prior consultation from time to time in our deliberations; where that is necessary we take the relevant parts of the reviews in their proper context. During this consultation process, the identity of the regulator changed from the Director General of Telecommunications (or his office) to OFCOM, though the process continued as though there was no change except as to nomenclature. It is not necessary for us to distinguish between those regulators for the purposes of this appeal.
  21. The following is a short route-map through the Decision; we consider the detailed provisions in their proper context later on. At the end of the process OFCOM's conclusions on the central points were as follows. It should be noted that in the decision H3G was identified as "3".
  22. In paragraph 2.23 OFCOM expresses its conclusions on the relevant market:
  23. "2.23 Ofcom has concluded that, having taken due account of the Recommendation and the SMP Guidelines, as well as Oftel's guidelines on assessing effective competition in carrying out this review, there are six separate relevant markets as follows:
    [determinations on the other networks]
    …. wholesale voice call termination provided by '3' (such termination provided via '3''s mobile network); …"
  24. In Chapter 3 OFCOM concludes that H3G has SMP in that market. The main criteria used in that assessment are set out in paragraphs 3.2 and 3.3:
  25. "Criteria used in assessing SMP
    3.2 In its assessment of SMP in the markets for voice call termination, Ofcom has focused on single firm dominance and has relied on four of the criteria listed in the SMP Guidelines and in Oftel's Guidelines on the assessment of SMP. These criteria are:
    (a) market share;
    (b) ease of market entry;
    (c) excessive prices and profitability; and
    (d) countervailing buyer power.
    3.3 Ofcom has also considered the CPP arrangements – a key factor in shaping the competitive conditions prevailing in the wholesale mobile voice call termination markets."
    The point made at paragraph 3.3 was elaborated on in the preceding reviews, which are cross-referenced in paragraph 3.1 as containing relevant material. This is not elaborated on in this part of the Decision, but it is referred to in the preceding chapter in paragraphs 2.5 and 2.6, where it is observed that MNO's have little incentive to keep termination charges low because those charges are passed on to the caller; and in Annex B (page 173) of the May review (incorporated expressly into the Decision in paragraph 3.1) OFCOM observed that because of the CPP arrangement, and because of the lack of awareness on the part of callers of the cost of calling mobiles from other networks or from fixed lines, and lack of customer awareness of the price they were paying, customer awareness was no constraint on termination charges and therefore did not affect its provisional determination of SMP. We consider that OFCOM took that factor into account in the Decision in addition to the other factors expressly relied on.
  26. Paragraphs 3.21 and 3.22 set out the conclusions on SMP:
  27. "Ofcom's conclusion on the new entrant: '3'

    3.21 Ofcom maintains its view that '3' has SMP in the market in which it supplies wholesale mobile termination services. Ofcom considers that (i) '3''s 100 % market share in the market for wholesale voice call termination on its network; and (ii) the presence of absolute barriers to entry in that market, mean that '3' has SMP.
    3.22 In addition, Ofcom believes that purchasers of termination from '3' have insufficient buyer power to off-set '3''s market power, and thus constrain its pricing behaviour."
    The decision then goes on to deal with the question of countervailing buyer power, and finds in paragraphs 3.29 to 3.42 that neither BT nor any MNOs had such power.
  28. At paragraph 3.43 OFCOM considers excessive prices. In particular it considers H3G's averment that OFCOM failed to give proper individual consideration to the question of whether H3G can set excessive call termination prices. In paragraph 3.46 OFCOM conceded that it had not performed a detailed analysis of H3G's charges, because that would have been difficult at the time since the networks were new and capable of providing innovative services, and because H3G's charges reflected a combination of 2G and 3G termination charges. It observed, however, that that did not mean that H3G was unable to set excessive termination charges given the lack of constraints it faced, which were not sufficient to hold charges at the competitive level on a forward-looking basis. H3G's own evidence did not include information on its termination costs.
  29. At paragraph 3.51 OFCOM found that the possibility of dispute resolution would not, in practice, constrain an MNO (and therefore by implication H3G) from setting excessive termination charges. In paragraph 5.72 it concludes that:
  30. "5.72 Ofcom remains of the view that a transparency obligation including a reporting requirement is a proportionate obligation to impose on '3' at this stage, as explained in paragraphs 5.134 – 5.137 of the December consultation."
    Then, in the schedule to the Decision OFCOM imposes certain reporting obligations the details of which do not matter at this stage; the remedy itself is not challenged if the SMP determination stands.

    V SUMMARY OF THE TRIBUNAL'S FINDINGS

  31. In considering whether MNOs had SMP in the market for wholesale voice call termination on their respective mobile networks, OFCOM noted the application of CPP and relied on four criteria:[11]
  32. a) market share;
    b) ease of market entry;
    c) excessive prices and profitability; and
    d) countervailing buyer power ("CBP").
    In that regard, H3G had 100 per cent market share and there were absolute barriers to entry in the market for wholesale voice call termination on the H3G mobile network. We note that it was against the background of those strong prima facie indicators of SMP that OFCOM reached its decision that there was no effective competition in that market and, accordingly, concluded that H3G has SMP. Nonetheless, it was for OFCOM to analyse whether there was sufficient CBP in the market to negate the finding of SMP. We take the view that on that one aspect of its Decision, OFCOM did not meet the standard required of it. Notwithstanding that, on a reconsideration, it would be open to OFCOM to reach the same conclusion; it is now for OFCOM to reconsider that part of the decision relating to CBP in the light of this judgment and the information currently available.

    VI JURISDICTION OF THE TRIBUNAL

  33. Article 4 of the Framework Directive provides:
  34. "Article 4
    Right of appeal
    1. Member States shall ensure that effective mechanisms exist at national level under which any user or undertaking providing electronic communications networks and/or services who is affected by a decision of a national regulatory authority has the right of appeal against the decision to an appeal body that is independent of the parties involved. This body, which may be a court, shall have the appropriate expertise available to it to enable it to carry out its functions. Member States shall ensure that the merits of the case are duly taken into account and that there is an effective appeal mechanism. Pending the outcome of any such appeal, the decision of the national regulatory authority shall stand, unless the appeal body decides otherwise.
    2. Where the appeal body referred to in paragraph 1 is not judicial in character, written reasons for its decision shall always be given. Furthermore, in such a case, its decision shall be subject to review by a court or tribunal within the meaning of Article 234 of the Treaty."
  35. H3G's appeal is brought under section 192 of the Communications Act 2003, which implements Article 4 of the Framework Directive. The jurisdiction of the Tribunal to determine the present appeal arises under section 195 Communications Act 2003, which provides that:
  36. "195 Decisions of the Tribunal

    (1) The Tribunal shall dispose of an appeal under section 192(2) in accordance with this section.
    (2) The Tribunal shall decide the appeal on the merits and by reference to the grounds of appeal set out in the notice of appeal.
    (3) The Tribunal's decision must include a decision as to what (if any) is the appropriate action for the decision-maker to take in relation to the subject-matter of the decision under appeal.
    (4) The Tribunal shall then remit the decision under appeal to the decision-maker with such directions (if any) as the Tribunal considers appropriate for giving effect to its decision.
    (5) The Tribunal must not direct the decision-maker to take any action which he would not otherwise have power to take in relation to the decision under appeal.
    (6) It shall be the duty of the decision-maker to comply with every direction given under subsection (4).
    (7) In the case of an appeal against a decision given effect to by a restriction or condition set by regulations under section 109, the Tribunal must take only such steps for disposing of the appeal as it considers are not detrimental to good administration.
    (8) In its application to a decision of the Tribunal under this section, paragraph 1(2)(b) of Schedule 4 to the Enterprise Act 2002 (c. 40) (exclusion of commercial information from documents recording Tribunal decisions) is to have effect as if for the reference to the undertaking to which commercial information relates there were substituted a reference to any person to whom it relates.
    (9) In this section "the decision-maker" means-
    (a) OFCOM or the Secretary of State, according to who took the decision appealed against; or

    (b) in the case of an appeal against-

    (i) a direction, approval or consent given by a person other than OFCOM or the Secretary of State, or
    (ii) the modification or withdrawal by such a person of such a direction, approval or consent,

    that other person."

    VII THE TRIBUNAL'S ANALYSIS

  37. The issue arising on this appeal can be globally described as whether OFCOM were justified in coming to the conclusion that it did in relation to H3G's SMP. H3G says that it was not. Within that global issue the following issues arise:
  38. a) What was the standard of proof necessary to be achieved in order for OFCOM to be satisfied of the existence of SMP.
    b) Whether OFCOM considered all factors that it ought to have considered in reaching its conclusion.
    c) Whether the factors that it did investigate (or purport to investigate) were properly considered.
    d) Whether OFCOM was entitled to reach its decision in the light of the above matters.

    A. THE STANDARD OF PROOF

  39. H3G's case before the Tribunal was that there is a very high onus on a regulator who finds that there is SMP. Mr Green's first point, for H3G, was that once SMP was found, the regulator was obliged under the Framework Directive to impose some form of control obligation, so there was no "daylight" between the finding of SMP and the control; and that, since it was laid down in the Framework Directive that controls should only be imposed as a matter of last resort, a finding of SMP should similarly be a finding of last resort. The provision of the Framework Directive relied on by Mr Green as imposing this strict obligation was recital 27, which he said demonstrated that a designation of SMP was an administrative act of last resort:
  40. "It is essential that ex ante regulatory obligations should only be imposed where there is not effective competition, i.e. in markets where there are one or more undertakings with significant market power, and where national and Community competition law remedies are not sufficient to address the problem."
    OFCOM correctly pointed out that this passage on its face referred to the imposition of ex ante obligations, and not to a finding of SMP, but Mr Green's point was that since a finding of SMP led inevitably to some form of regulatory obligation, then one could read this passage as though it referred to a finding of SMP. This meant that a particularly serious or heavy burden was imposed in an ex ante case, requiring a "thorough economic analysis" and "mere theoretical ability" to control prices is not enough.
  41. The two exercises of finding SMP and considering what remedy to impose are two separate stages of the overall process, and Mr Green did not go so far as to conflate them. It is obviously right that they be kept separate. The factors relevant to one are not identical to the factors relating to the other. The Commission certainly takes the view that finding SMP means that a remedy must be imposed (see paragraphs 21 and 114 of the Commission's Guidelines). We do not find it necessary to express a view as to whether that is right or not, and are prepared to assume for these purposes that it is.[12] It is also right that the imposition of remedies is regarded by the legislation as a matter of considerable seriousness, and over-regulation must be prevented. However, none of that seems to us to reflect on the burden to be fulfilled in establishing SMP so as somehow to raise that burden in an ex ante regulation case, which was the thrust of Mr Green's argument at this stage. The exercise of finding SMP is what it is. On any footing it is an exercise which cannot be carried out lightly. It has serious consequences. But it remains the same exercise, and would be the same whether or not there is an obligation to impose a remedy if SMP is found. The regulator cannot be expected to carry out the exercise of determining SMP with an eye to the remedy it wishes to impose, or whether it wishes to impose one, if it finds it. All that needs to be acknowledged is that the exercise is a serious one with potentially serious consequences. That is obvious, and there is no indication at all that OFCOM ever thought otherwise. The exercise that has to be done in any given case is dependent on the facts of that case. In some cases a "thorough economic analysis" may be required; in others a differently described exercise will be required. One cannot specify more in advance, and one is not required to do so by the obvious proposition that regulation is a serious matter.
  42. We therefore reject this part of H3G's argument.
  43. Mr Green then sought to support his case in favour of a high burden of proof for establishing SMP by reference to Commission of the European Communities v. Tetra Laval BV, (Case C-12/03P), 15 February 2005. He submitted that that case established that in ex ante regulation, which required a prospective analysis of the facts, the regulator had to be satisfied to a very high degree that H3G would be able to charge, and indeed would charge, an excessive price. He did not formulate the burden in terms such as "beyond reasonable doubt", but he did urge on us the need to find with a high degree of certainty that the excessive charging would occur, and he said this was required by the jurisprudence demonstrated by Tetra Laval. He particularly relied on the following passages from the judgment of the ECJ:
  44. "42. A prospective analysis of the kind necessary in merger control must be carried out with great care since it does not entail the examination of past events – for which often many items of evidence are available which make it possible to understand the causes – or of current events, but rather a prediction of events which are more or less likely to occur in future if a decision prohibiting the planned concentration or laying down the conditions for it is not adopted.

    43. Thus, the prospective analysis consists of an examination of how a concentration might alter the factors determining the state of competition on a given market in order to establish whether it would give rise to a serious impediment to effective competition. Such an analysis makes it necessary to envisage various chains of cause and effect with a view to ascertaining which of them are the most likely.

    44. The analysis of a 'conglomerate-type' concentration is a prospective analysis in which, first, the consideration of a lengthy period of time in the future and, secondly, the leveraging necessary to give rise to a significant impediment to effective competition mean that the chains of cause and effect are dimly discernible, uncertain and difficult to establish. That being so, the quality of the evidence produced by the Commission in order to establish that it is necessary to adopt a decision declaring the concentration incompatible with the common market is particularly important, since that evidence must support the Commission's conclusion that, if such a decision were not adopted, the economic development envisaged by it would be plausible."
    Reliance was also placed on paragraph 162 of the judgment of the CFI in Tetra Laval BV v. Commission of the European Communities (Case T-5/02) [2002] ECR II-4381, which states:
    "162. It follows from the foregoing that it is necessary to examine whether the Commission based its analysis of the likelihood of leveraging from the aseptic carton markets, and of the consequences of such leveraging by the merged entity, on sufficiently convincing evidence. In the course of that examination it is necessary, in the present case, to take account only of conduct which would, at least probably, not be illegal. In addition, since the anticipated dominant position would only emerge after a certain lapse of time, by 2005 according to the Commission, its analysis of the future position must, whilst allowing for a certain margin of discretion, be particularly plausible. "
    In addition, Mr Green relied on extracts from the Opinion of Advocate General Tizzano delivered on 25 May 2004 in the Tetra Laval case, which he said suggested a high standard of proof which was required in ex ante regulation cases, and he particularly relied on paragraph 76, where the Advocate General said:
    "76. As a matter of fact, I consider that the symmetry of those requirements cannot be absolute, seeing that there is, between the cases in which the notified transactions would very probably create or strengthen a dominant position within the meaning of Article 2 and the cases in which those transactions very probably would not create or strengthen such a dominant position, a 'grey area': an area, that is to say, in which cases are to be found where it is especially difficult to foresee the effects of the notified transaction and where it is therefore impossible to arrive at a clear distinct conviction that the likelihood that a dominant position will be created or strengthened is significantly greater or less than the likelihood that such a position will not be created or strengthened. The system laid down by Regulation 4064/89 must therefore necessarily provide a yardstick for the solution of those cases which are of doubtful or difficult classification."
    This, Mr Green said, was a strong pronouncement.
  45. Tetra Laval was a case in which the Commission had ruled on a proposed merger of two companies which was a conglomerate merger. On the facts of that case it had to consider whether one of the two companies would use its dominant position in market A to produce "leverage" and improve the position of the other company so as to make that one dominant in its (different) market (market B). The facts were therefore somewhat specialised, and required more prediction of future behaviour than will be necessary in many merger contexts. The remarks of the ECJ and the Advocate General have to be read in that context. However, even allowing for that, we do not think that the remarks relied on by Mr Green have the full effect he contends for. They sound a warning about the need to carry out proper assessments, and they provide for an understandable degree of caution where what is required is an assessment as to future conduct, as opposed to an evaluation of past conduct, but they do not flag a particularly high degree of probability. So far as the Advocate General's Opinion is concerned, insofar as it might be thought to be proposing a higher test than the ECJ test, then the ECJ formulation is to be preferred. In fact it is not clear that he is doing so. In the paragraph quoted above, it seems to us that he is in substance saying that future events, if part of a chain of events said to give rise to future dominance, have to be proved to be likely on a balance of probability. That is not necessarily a "high" test. In relation to paragraph 43 of the ECJ judgment, Mr Green conceded that what was being referred to there is not a high degree of probability, but relative degrees of probability, which detracts from his strong case, and also said that what the court would do was look at the evidence carefully and evaluate whether it supported the conclusion drawn from it. That again is true, but it is not pointing to any particularly high degree of probability. The real position is that the evidence must be carefully looked at, and properly assessed.
  46. The case demonstrates (if it needs to be demonstrated) that theory and surmise is not enough. One must look to see how things operate in practice, and prove whatever has to be proved to an appropriate level of proof. It points out the need to be particularly careful in relation to that when one is considering future conduct. We do not think that at the end of the day that is particularly controversial. However, since (as will appear) OFCOM does not seek to rely on a positive prediction of future behaviour as a step towards creating dominance (which was the position in Tetra Laval) we do not need to consider just what probative levels of evidence should be required by a regulator in relation to future conduct. OFCOM's case is put differently – it relies on SMP now, not in the future. We do not think that Tetra Laval provides the assistance that Mr Green seeks to derive from it because it is directed at a different point.
  47. We note that the Irish Electronic Communications Appeals Panel ("ECAP") has taken a similar approach in an appeal brought by another Hutchison group company in that jurisdiction – Decision No: 02/05 of the Electronic Communications Appeals Panel in respect of appeal No: ECAP 2004/01. A similar submission about Tetra Laval was made to that tribunal. At paragraph 4.21 of that decision the Panel considered that it was significant that ex ante regulation was there (as here) under consideration, so that the power to raise prices was significant. At paragraphs 4.22 and 4.23 ECAP felt it relevant that in Tetra Laval the Court, at paragraph 42 of the judgment, had said:
  48. "42. A prospective analysis of the kind necessary in merger control must be carried out with great care since it does not entail the examination of past events – for which often many items of evidence are available which makes it possible to understand the causes – or of current events, but rather a prediction of events which are more or less likely to occur in the future…".
    However, ECAP went on to conclude:
    "4.23 This does not mean that because there is ex ante analysis that the Respondent [i.e. the regulator] has to meet a higher standard of proof. The standard is whether, on the balance of probabilities an undertaking has significant market power. Rather the Panel is merely asserting the common sense proposition that when one is making a finding of significant market power on the basis of a prospective analysis (as opposed to an ex post analysis) then it is necessary that this analysis be sufficiently rigorous and thorough so that a clear link can be drawn between existing circumstances and likely future behaviour. To put it another way, because the likelihood of error is greater in a prospective analysis, the prospective analysis must be proportionately more rigorous to account for this possibility."
    We respectfully agree with that approach.
  49. We therefore have to consider whether the material relied on by OFCOM justifies its conclusion that SMP exists in the relevant market. There has to be proper evidence for the conclusion; it does not have to achieve some high degree of weight.
  50. B. H3G'S ATTACK ON THE DECISION

  51. Hutchison mounted 4 principal attacks on the Decision:
  52. i. OFCOM did not carry out a sufficient analysis of prices to entitle it to come to a decision that H3G had SMP.
    ii. OFCOM failed to take account, or sufficient account, of restraints on H3G's ability to increase prices arising from regulation.

    iii. OFCOM failed to take account, or sufficient account, of the ability of customers (principally BT) to restrain pricing.

    iv. H3G's case is that as a result of these failures, and looking at the matter overall, OFCOM did not carry out the exercise required of it if it were to find SMP, and that its decision is therefore flawed.
    We will consider each of these various lines of attack in turn, but before doing so we need to set out some material which is relevant to the basis on which SMP can be determined.

    THE DETERMINATION OF SMP

  53. The starting point of a determination of SMP is the definition appearing in Article 14(2) of the Framework Directive:
  54. "An undertaking shall be deemed to have significant market power if, either individually or jointly with others, it enjoys a position equivalent to dominance, that is to say a position of economic strength affording the power to behave to an appreciable extent independently of competitors, customers and ultimately consumers."
    This definition is equivalent to the concept of dominance as defined by the case law of the Court of Justice and the Court of First Instance of the European Communities, see recital 25 of the same Directive. The words, or very similar words, appear elsewhere in the jurisprudence in that context – see e.g. Case 311/84 Michelin v The Commission [1983] ECR 3461 at paragraph 30.
  55. The Commission has published Guidelines in relation to (inter alia) such determinations – "Commission guidelines on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services." (2002)/C 165/03). The following paragraphs of those Guidelines are material:
  56. "73. In an ex-post analysis, a competition authority may be faced with a number of different examples of market behaviour each indicative of market power within the meaning of Article 82. However, in an ex-ante environment, market power is essentially measured by reference of the power of the undertaking concerned to raise prices by restricting output without incurring a significant loss of sales or revenues.
    75. As explained in the paragraphs below, a dominant position is found by reference to a number of criteria and its assessment is based, as stated above, on a forward-looking market analysis based on existing market conditions. Market shares are often used as a proxy for market power. Although a high market share alone is not sufficient to establish the possession of significant market power (dominance), it is unlikely that a firm without a significant share of the relevant market would be in a dominant position. Thus, undertakings with market shares of no more than 25 % are not likely to enjoy a (single) dominant position on the market concerned. In the Commission's decision-making practice, single dominance concerns normally arise in the case of undertakings with market shares of over 40 %, although the Commission may in some cases have concerns about dominance even with lower market shares, as dominance may occur without the existence of a large market share. According to established case-law, very large market shares — in excess of 50 % — are in themselves, save in exceptional circumstances, evidence of the existence of a dominant position. An undertaking with a large market share may be presumed to have SMP, that is, to be in a dominant position, if its market share has remained stable over time. The fact that an undertaking with a significant position on the market is gradually losing market share may well indicate that the market is becoming more competitive, but it does not preclude a finding of significant market power. On the other hand, fluctuating market shares over time may be indicative of a lack of market power in the relevant market.
    78. It is important to stress that the existence of a dominant position cannot be established on the sole basis of large market shares. As mentioned above, the existence of high market shares simply means that the operator concerned might be in a dominant position. Therefore, NRAs should undertake a thorough and overall analysis of the economic characteristics of the relevant market before coming to a conclusion as to the existence of significant market power. In that regard, the following criteria can also be used to measure the power of an undertaking to behave to an appreciable extent independently of its competitors, customers and consumers. These criteria include amongst others:
    — overall size of the undertaking,
    — control of infrastructure not easily duplicated,
    — technological advantages or superiority,
    — absence of or low countervailing buyer power,
    — easy or privileged access to capital markets/financial resources,
    — product/services diversification (e.g. bundled products or services),
    — economies of scale,
    — economies of scope,
    — vertical integration,

    — a highly developed distribution and sales network,

    — absence of potential competition,
    — barriers to expansion.

    79. A dominant position can derive from a combination of the above criteria, which taken separately may not necessarily be determinative.

    80. A finding of dominance depends on an assessment of ease of market entry…"
  57. Those Guidelines were published pursuant to the Framework Directive, and that Directive requires national regulatory authorities (such as Ofcom) to "take the utmost account of the guidelines" in carrying out a market review – see Article 16. This obligation is acknowledged by OFCOM in its own guidelines – see paragraph 1.11 et seq. It expressly professes to implement them.
  58. For the most part this appeal was conducted on the footing that those principles applied generally and, so far as applicable, to the present case. However, at one stage in his submissions Mr Green appeared to submit that the presumption of dominance from 50% market share in an ex ante case was no longer good law as a result of Tetra Laval, since that case required a full and proper and consideration of all relevant principles and imposed a heavy burden on the person asserting dominance. The Guidelines, he said, required to be revisited as a result of Tetra Laval even though the Commission had not yet got round to doing that. He went so far as to submit that the words in the definition in Article 14 of the Framework Directive "affording the power to behave to an appreciable extent [etc]" should be taken to mean "making it plausible that it will behave … [etc]".
  59. We do not accept that submission. In Tetra Laval what the Commission and the courts were concerned with was a situation where the potential dominance would arise from the potential exploitation of dominance in one market to bring about a situation of dominance in another ("leveraging", in the jargon of that case) in circumstances in which that would have required deliberate acts. The likelihood of the creation of a situation of dominance by deliberate act when none would otherwise exist had to be assessed. It is in that context that the ECJ made various remarks about the need to make a proper case, and to prove the likelihood of events properly. That is apparent from paragraph 74 of the ECJ judgment:
  60. "74. Since the view is taken in the contested decision that adoption of the conduct referred to in recital 364 in that decision is an essential step in leveraging, the Court of First Instance was right to hold that the likelihood of its adoption must be examined comprehensively, that is to say, taking account … both of the incentives to adopt such conduct and the factors liable to reduce, or even eliminate, those incentives, including the possibility that the conduct is unlawful.
    In other words, the dominance in that case did not arise from facts that could plainly be seen to exist as a result of the merger. In order for the dominance to arise one of the parties would have to take positive steps to exploit existing dominance in a different market so as to achieve a given market effect. The ECJ held that that has to be proved as a likely fact to some relevant degree (the degree does not matter for present purposes).
  61. In our view, there is nothing in any of this which requires a reconsideration of the Guidelines. The approach in the Guidelines describes a proper evidential approach to a factual inquiry, and we think that the ECJ would be very surprised to be told that its judgment was going to require the re-writing of the Guidelines in this respect. Mr Green gives the game away with his submission which changes the effect of the words in Framework Directive Art 14(2). The concepts of having a power ("affording the power") and actually doing something ("making it plausible that it will") are different concepts. The first involves a consideration of potential; the second involves an assessment of the likelihood of something happening. A man standing at the edge of a drop has the power to jump; whether it is plausible that he will jump is an entirely different question. Tetra Laval cannot be taken to have conflated those two ideas.
  62. What therefore emerges from that regime are the following points which are of importance to this appeal:
  63. a) A large market share gives rise to a presumption of dominance.
    b) However, the nature of that "presumption" must be properly understood. Normally, in English law, a presumption can be relied on by itself if there is no other evidence which goes to the point; no-one suggested that the position would be different so far as any European principles might be in play. The first sentence of paragraph 78 of the guidelines seems at first sight to detract from that principle. However, we are not satisfied that it does. What paragraph 78 provides is that a regulator is not entitled to find a large market share, rely on that as giving rise to a presumption of dominance and stop there. The regulator is obliged to go on and consider all other such factors as are relevant to a consideration of the point in the market in question. The paragraph then goes on to identify some of them. What is required by the Guidelines is a "thorough and overall analysis of the economic characteristics of the relevant market before coming to a conclusion as to the existence of significant market power." This approach is demonstrated by various authorities, including Hoffman-La-Roche [1979] ECR 461, and we do not consider that anything appearing in Tetra Laval contradicts or modifies it.
  64. We therefore approach Hutchison's appeal with those principles in mind. Hutchison's case is that the Decision was not determined on the correct basis. The review was not thorough enough, missed some material out and got some things wrong. As indicated above, the relevant commissions and omissions were by and large separately identified.
  65. PRICING AND COSTS

  66. Mr Green's submissions on pricing and costs can, we think, be reduced to the following points:
  67. a) OFCOM in substance regarded price or the power over price as the sole touchstone of its decision.
    b) In relation to the 4 2G MNOs it found an ability to charge excessive prices, and an incentive to do so. However, in the case of Hutchison it found ability, but no incentive, and did not even examine incentive. It ought to have examined the incentive question, and had it done so it would have found that there was no incentive. Absent an incentive there was no dominance.

    c) OFCOM failed to conduct an analysis of Hutchison's costs or prices and relied purely on a theoretical ability to raise prices, and therefore failed to carry out an important and necessary part of the exercise. There was therefore no proper evidence that prices were, or would become, excessive. Without this the decision is flawed.

    d) On analysis the Decision is forward-looking in terms of prices becoming excessive, and OFCOM has failed to carry out properly (or at all) the exercise of assessing when, how and to what extent they will or may become excessive. We think that on analysis this is probably a refinement or an aspect of the point at (c) or perhaps (b).
    We will take these points in turn.

    Point (a) – power over price as the sole touchstone

  68. At one level this may seem like a pointless dispute over characterisation, but close scrutiny of the way that H3G puts its case reveals that it has a significance greater than that. Part of H3Gs' complaint is that OFCOM based itself firmly on power over price, but did not carry out any investigations into pricing and costing. It is therefore necessary for us to consider it.
  69. While it is clear that OFCOM considered that power over price was a significant feature of SMP, and a manifestation or consequence of it, to say that OFCOM thought it was the sole touchstone mischaracterises OFCOM's reasoning. As appears from the narrative above, the Decision followed on from two earlier review documents. Those documents contained views, reasoning and OFCOM's response to submissions to date. There are many areas in the Decision where explicit reference back is made to the earlier reviews. One of those is the first paragraph of section 3 (dealing with SMP). In paragraph 3.1 OFCOM says:
  70. "Further details of Ofcom's reasoning for its decision on SMP, and of the analysis performed by Ofcom to arrive at the conclusions presented below, are available in Chapter 3 (paragraphs 3.9-3.45) of the December consultation and Chapter 4 and Annex B of the May consultation."
  71. One therefore has to go back to the earlier documents to find the full reasoning of OFCOM. In the May review the Director General set out Article 14 of the Framework Directive in terms indicating clearly that he intended to apply it. The review then set out various sections dealing with single market dominance (which the Director General considered to be the relevant test) and he discussed the factors that he considered relevant – market shares, ease of market entry and absence of competition, excessive prices and profitability and countervailing buyer power. There is nothing there to indicate that he used power over price, as such, as a sole or principal touchstone. It is a conventional analysis.
  72. This was followed by the December Interim Statement which again starts with Article 14(2) of the Framework Directive. Having done that it goes on to consider SMP in the light of the same four considerations as were relied on in the May paper. The excessive prices point starts with the statement that:
  73. "The ability to keep prices persistently and profitably above the competitive level is an important indicator of market power. In a competitive market, individual firms should not be able to raise prices above costs and sustain excess profits for prolonged periods of time" (paragraph 3.17).

    It then goes on to consider prices and costs for the 4 2G companies. It is true that the debate at this point involves a consideration of excessive pricing (when compared to costs) but it is only one of the factors that the Director put forward. The countervailing buyer power point was also related to pricing, because it was a consideration of the extent to which such power is capable of being exerted to prevent a price rise, but again price is part of this aspect of the consideration, not the pre-eminent point. An Annex sets out other criteria listed in the various guidelines and explains why they are said to be less relevant in this specific market.

  74. The same pattern of reasoning flows through into the Decision, where SMP is dealt with in Chapter 3. In paragraph 3.2 OFCOM lists again the four criteria referred to above, pointing out (in a footnote) the less relevant criteria as set out in the Annex to the May consultation document. In paragraphs 3.21 and 3.22 it records its conclusions shortly:
  75. "3.21 Ofcom maintains its view that '3' has SMP in the market in which it supplies wholesale mobile termination services. Ofcom considers that (i) '3''s 100% market share in the market for wholesale voice call termination on its network; and (ii) the presence of absolute barriers to entry in that market, mean that '3' has SMP.
    3.22 In addition, Ofcom believes that purchasers of termination from '3' have insufficient buyer power to off-set '3''s market power, and thus constrain its pricing behaviour."
  76. That analysis focuses on two particular aspects, neither of which expressly incorporates an explicit reference to power over prices, though it would be consistent with OFCOM's reasoning that both of them could be translated into it. The Decision then goes on to consider BT's countervailing buyer power (which we deal with below), and in that context it certainly discusses the power of BT to reject price increases. To that extent there is an explicit reference to power over prices, but it does not give the Decision the character that Mr Green seeks to place upon it. Pricing again comes into the Decision in paragraph 3.43, where OFCOM considers H3G's submission that OFCOM had failed to consider whether H3G could set excessive call termination prices. In answer to this point OFCOM concludes that:
  77. "The constraints facing "3" are … not sufficient to hold charges at the competitive level on a forward-looking basis." (Paragraph 3.46)
    At this point OFCOM is considering a submission by H3G, not putting forward its own primary reasoning. Last, in paragraph 3.50 OFCOM considers a submission by H3G that it had failed to take into account the fact that regulation would constrain pricing, pointing out that it did not consider it would.
  78. A proper analysis of this approach reveals a conventional approach to the assessment of SMP, testing it by reference to established factors appearing in the various guidelines. The positive factors relied on by OFCOM do not all explicitly refer to prices, though, as we have observed, they would (if OFCOM were right) have the potential to drive price increases or, possibly more likely in this electronic communications market, to sustain prices above levels sustainable in an effectively competitive market, but that is a more correct way of putting it than H3G's. The effect on prices was deployed as a means of countering other submissions, not as a matter of primary analysis. It is also right that the potential to increase prices, or charge excessive prices, led to the remedy imposed by OFCOM (a reporting remedy) but that deals with the effect of having found SMP, not the means of determining whether SMP exists. At the end of the day this point may not matter much (though it figured heavily in H3G's submissions), but if it is relevant then it is right to characterise OFCOM's case properly.
  79. Point (b) – the absence of a finding of incentive to raise prices

  80. Since there was something of a dispute at the hearing about what it was that the Decision found, or did not find, in relation to this issue we have to resolve that part first. Mr Green said that OFCOM found that there was no evidence of an incentive to charge excessive prices, and at more than one stage suggested that OFCOM actually found there was no such incentive. Mr Roth QC, who appeared for OFCOM, disputed that and said that OFCOM's finding in this respect was one of uncertainty – it was uncertain as to whether H3G had that incentive. It is as well to be clear as to what OFCOM did and did not find, and did and did not say, about this.
  81. The Decision does not actually say anything in terms about this. The question of incentive (in this respect) arises out of the December review. Chapter 5 of the December review deals with the question of remedies. Paragraph 5.1 states that:
  82. "The Director therefore needs to consider potential ex ante regulatory remedies to address SMP in each case [i.e. in the case of each of the MNOs, including H3G]."
    From paragraph 5.118 onwards it deals with H3G, and at paragraph 5.128 the Director
    "recognises that, whilst it has the ability, whether '3' has the incentive to set excessive charges for 2G voice call termination service is less certain. The Director accepts that so far, '3' has set charges for 2G voice call termination services in line with those of the other MNO's. This does not mean however that charges will remain so indefinitely, or that '3' lacks the ability to significantly raise the level of charges for 2G termination services."
    and in paragraph 5.130 the Director again refers to

    "uncertainty regarding '3's incentive to set excessive charges …"
  83. All this is in the context, not of considering whether H3G has SMP, but what remedy should be imposed, and the uncertainty as to incentive is one of the reasons why the Director considered that it would not be appropriate to impose an LRIC-based obligation, and that what was appropriate was a reporting obligation (which was ultimately imposed).
  84. Chapter 5 of the Decision is the chapter which contains the final decision as to remedy. It does not explicitly repeat what had previously been said about incentive. However, in paragraph 5.71 it records that:
  85. "A full explanation of Ofcom's position as regards the regulation of '3''s call termination is provided in paragraphs 5.129-5.132 of the December consultation."
    It is therefore apparent that OFCOM is still relying on the "uncertainty".
  86. What appears from this, therefore, is that OFCOM did not decide there was no evidence of incentive; and it did not decide that there was no incentive. It said something different – it was uncertain as to incentive. At the end of the day that difference may not matter much to Mr Green's argument on the point, but it is important to establish what it was that OFCOM decided or found (or did not decide or find) about this. Mr Roth complained about a mischaracterization of OFCOM's position in this respect, and we think he had some justification for doing so.
  87. With that in mind we therefore turn to Mr Green's arguments about incentive. His argument was that the mere ability to raise prices was not enough. There must also be an incentive to do so. Although he mischaracterised what OFCOM said about this, he can still make the point that he sought to make, which is based on the premise that incentive to raise prices to an excessive level needs to be found, along with ability to charge those prices, if a party is to be held to have SMP. Since OFCOM did not find that it had that incentive (which it clearly did not) it failed to establish a key ingredient of SMP (at least on the facts of this case). (For the sake of completeness we observe that at one stage Mr Green said that by "incentive" he meant an actual likelihood or probability that a company would engage in excessive pricing. Looking at the totality of his submissions we do not think he means that – we take the word to be intended to bear its normal meaning, which is a motive or reason which would incline it to do the act in question. We shall consider the point as if the word bore that meaning.)
  88. In support of his submission Mr Green pointed out that OFCOM had actually found an incentive so far as the other 4 MNO's were concerned, but the real basis of his submission was said to be the Tetra Laval decision. He pointed to various passages in the judgment of the ECJ in that case and said that those passages demonstrated that it was not only relevant, it was vital to take into account the extent to which the entity in question had an incentive to indulge in anti-competitive behaviour. In paragraphs 71 to 74 the Court held:
  89. "71 It should be observed, first of all, that paragraphs 148 to 162 of the judgment under appeal, which the Commission challenges under both its first and its second ground of appeal, form a section in which the Court of First Instance described certain specific aspects of conglomerate effects, in particular temporal aspects, and inferred from them certain general rules as to the evidence which the Commission must produce when it considers that a proposed concentration must be declared incompatible with the common market.
    "72 It was in the context of this reminder of the need for 'convincing evidence' that the Court of First Instance made reference to the obligation to examine all the relevant information.
    "73 Such an examination must be carried out in the light of the purpose of the Regulation, which is to prevent the creation or strengthening of dominant positions capable of significantly impeding effective competition in the common market or a substantial part thereof.
    "74 Since the view is taken in the contested decision that adoption of the conduct referred to in recital 364 in that decision is an essential step in leveraging, the Court of First Instance was right to hold that the likelihood of its adoption must be examined comprehensively, that is to say, taking account, as stated in paragraph 159 of the judgment under appeal, both of the incentives to adopt such conduct and the factors liable to reduce, or even eliminate, those incentives, including the possibility that the conduct is unlawful."
    What is important for Mr Green's purposes is the reference to incentives in paragraph 74. The point is taken up again in the next paragraph:
    "75 However, it would run counter to the Regulation's purpose of prevention to require the Commission, as was held in the last sentence of 159 of the judgment under appeal to examine for each proposed merger the extent to which the incentives to adopt anti-competitive conduct would be reduced or even eliminated as a result of the unlawfulness of the conduct in question, the likelihood of its detection, the action taken by the competent authorities, both at community and national level, and the financial penalties which could ensue."
    He says that it is plain from this, and from other reasoning in that decision, that incentives must be taken into account.
  90. In considering whether Mr Green is right about this our starting point is the definition in Article 14(2) of the Framework Directive, which is set out above. It does not seem to provide any basis for Mr Green's submissions. This reflects a position rather than a course of conduct. It reflects a "position of economic strength". One can logically consider whether an entity is in that position without having to consider whether it has any incentive (or indeed any intention) to use it to any purpose. Being in a position and using it, or wishing to use it, to any particular purpose are different things (though they are obviously linked – one cannot use what one has not got). The same analysis applies when one moves on through the Article – "affording the power to behave to an appreciable extent [etc]". Having a power and having the incentive to exercise it are two conceptually different things, as Mr Green himself effectively accepted when he acknowledged that a man who left a house with a knife had the ability to hurt someone with it but did not necessarily have an incentive to do so. There is nothing in the wording of the Article which suggests that one adds incentive into the equation.
  91. Nor, when correctly viewed, is there anything in Tetra Laval which assists. That case concerns a different point and a different inquiry. It was not an inquiry as to whether an entity had SMP. It was an inquiry as to whether it would use dominance that it had in one market to achieve a dominance in another (for its subsidiary). It was not testing a present position; it was assessing whether a future position was likely to be brought about by one party. The ECJ held that in the context of that inquiry all relevant evidence had to be taken into account in assessing the likelihood of that future state of affairs being brought about, and it is plain from its judgment (not surprisingly) that the incentive of the entity in question to achieve that was one of the factors that had to be considered. That is the true analysis of the case, and it does not assist Mr Green.
  92. Accordingly we reject Mr Green's submission that OFCOM left a vital consideration out of account when it did not form a view about the incentive of H3G to raise its prices to an excessive level. Such a consideration is not relevant to the assessment of SMP (though it may be relevant to the remedy to be imposed, which is on analysis the context in which OFCOM refers to its inability to find anything about incentive in the Decision and the context in which it makes findings about the other MNO's in this respect).
  93. Point (c) - the alleged failure to assess costs or to ascertain when prices would become excessive

  94. H3G's point under this head is that OFCOM did not carry out any analysis from which one could determine whether H3G's charge was excessive at the time of the Decision, or would become excessive over time.[13] It is said that OFCOM relied purely on a theoretical ability to charge excessive prices, without any reference to the actual prices charged and the costs involved. This was an error of law. It is to be contrasted with what OFCOM did in relation to the other MNO's, where it carried out an analysis which enabled it to say (in paragraph 3.7 of the Decision) that its SMP finding was:
  95. "supported by Ofcom's analysis of 2G voice call termination charges, which appear to have been substantially above a reasonable estimate of each MNO's costs for a number of years, despite both formal and informal regulation."
    In this respect Mr Green reverted to his reliance on Tetra Laval and said that that case demonstrated that it was the regulator's duty to carry out such an analysis, and it had to take place over the relevant time frame (here, the 18-24 months covered by this period of review). Without it the conclusion was speculation only, and speculation was not a proper basis for finding SMP.
  96. It is common ground that OFCOM did not seek to carry out an analysis of prices and costs such as would, or could, demonstrate that prices were or would become excessive. The Decision records (at paragraph 3.43) that H3G had complained that:
  97. "[Ofcom] has manifestly failed to give proper individual consideration to the question of whether Hutchison 3G can set excessive call termination prices …"
    By way of elaboration the Decision records (in paragraph 3.44):

    "'3' also questions whether Ofcom's analysis of the MNOs' behaviour in setting 2G voice termination charges and its conclusion that these charges have been set above a reasonable estimate of each MNO's costs for a number of years applies also to '3'. '3' argues that the December Consultation contains no analysis of its costs that could have led Ofcom to conclude that '3''s charges are excessive. In addition, '3' claims that it has provided Ofcom with evidence that demonstrates that its pricing is not excessive."
  98. OFCOM deals with those criticisms thus:
  99. "3.46 The analysis of 2G termination charges Ofcom presented in Chapter 4 of the December Consultation was limited to the charges levied by Vodafone, O2, Orange and T-Mobile. Ofcom is aware that '3''s termination charges in practice reflect a combination of its 2G and 3G termination costs, and Ofcom has not performed a detailed analysis of '3''s charges. As Ofcom has noted, 3G networks are new and capable of providing a range of innovative services, and therefore it would be difficult to assess with confidence the relevant voice call termination costs and the appropriate rate of return on capital invested. However, this does not imply that '3' is unable to set excessive termination charges, given the lack of constraints that it faces. The constraints facing '3' are similar in nature to those facing the other MNO's, and these are not sufficient to hold charges at the competitive level on a forward-looking basis.
    With regard to the evidence submitted by '3', this did not include information on '3''s termination costs and, more importantly, it only refers to ported numbers for which '3', like all the other MNOs, receives a different termination charge from its own …"
  100. What appears from this are three things. First, it is indeed true that OFCOM did not carry out any of the exercises which H3G suggested it should have carried out. Second, the reason was that it would not be an easy exercise in relation to a new product. Third, H3G itself did not provide any information on the basis of which OFCOM could have made any judgment about this. It appears from some material that H3G may have had a business plan which might have been material (at least to some extent) to the point which H3G now says should have been considered, but it was not supplied. On the basis of this material it is not clear to us that OFCOM could have done anything useful or fair in this respect.
  101. However, more fundamentally, we consider that it was not necessary for OFCOM to conduct the exercise or exercises that H3G says should have been carried out. The existence of a power to behave independently of competitors, customers and consumers may, in some cases, result in excessive prices, but that is not necessarily the case. It is perfectly possible to have SMP and not charge excessive prices either at the time the position is being tested or in the future. Excessive prices are not an inevitable manifestation of SMP. SMP was found to exist on the basis of the material referred to above. Such a finding can be perfectly proper even in the absence of excessive pricing. If excessive prices (in the sense referred to by H3G) do not have to be found to exist in order to support a finding of SMP, then it does not necessarily matter that pricing is not investigated. Accordingly, OFCOM's reasoning cannot be criticised on the footing that it did not carry out this investigation. Once again there is nothing in Tetra Laval which compels a different conclusion. The careful consideration of future conduct that that case refers to is required in cases where that future conduct has to be established in order to be able to ascertain the likelihood of future dominance, but the present is not such a case. SMP in the present case turns on the power or ability to behave independently of customers or others. That does not require that present or future pricing be investigated.
  102. Accordingly we reject H3G's criticism under this head.
  103. Point (d) – OFCOM's alleged failure to assess whether prices would become excessive over time

  104. This is another aspect of the complaint that OFCOM did not carry out a price and costs based analysis. It fails for the same reasons as apply to point (c).
  105. REGULATION, DISPUTE RESOLUTION AND COUNTERVAILING BUYER POWER

  106. The next points taken by Mr Green have overlapping backgrounds of fact and potentially some overlap of law, and can usefully be introduced together. Although we will elaborate on them below, putting the matter shortly Mr Green says that in arriving at its SMP determination OFCOM did not take proper account of the following factors:
  107. a) The effect of the possibility of regulation.
    b) The countervailing buyer power ("CBP") of BT, which was the largest purchaser of termination on H3G's network.

    c) The dispute resolution mechanism which exists under the agreement between BT and H3G in relation to the access agreement between them. That agreement has a price alteration mechanism, with the Director-General (and probably now Ofcom) having some apparent role in resolving disputes. H3G says that this means that it does not have power over price.
  108. Before turning to the background it is relevant to ascertain what the Decision found in relation to these points, because its findings curtail the extent of the factual inquiry that is necessary to deal with them.
  109. So far as the effect of regulation is concerned, the Decision did not deal with it in terms of the effect of regulation on the party whose market powers are being investigated – indeed, it was OFCOM's position that it is an irrelevant criterion. It was, however, referred to in the context of regulatory constraints on market counterparties. In paragraph 3.39 the Decision recorded:
  110. "3.39 The December consultation (paragraph 3.44) noted that there were commercial considerations which limited the countervailing buyer power of MNOs. Aside from these commercial considerations, Ofcom also considers that, in relation to whether an operator has countervailing buyer power, the threat of regulatory intervention is relevant."
    Paragraph 3.44 of the December consultation was in the following terms:
    "Against this, there are incentives on MNOs to complete their subscribers' calls to '3' customers, particularly as '3' attracts further subscribers. For example, the subscribers of an MNO might be annoyed if they could not complete calls to all mobile subscribers (including those of '3') and, if other MNOs did reach agreement with '3' it would expose that MNO to commercial disadvantage. Further, any attempt to block calls to '3's subscribers could be the subject of an investigation or dispute by the Director or Ofcom.The Director therefore maintains that, despite a level of countervailing buyer, '3' holds a position of SMP in the relevant market. The mitigating factors outlined by '3' - such as its size, its roaming agreements and the mobile number portability ('MNP') arrangements - are considered further in the chapters on the detrimental effects of SMP and in the regulatory option appraisal."

  111. It is apparent from these statements that regulation was considered only so far as it might constrain the counterparty of H3G; there is nothing to indicate that it was considered as something which was capable of operating on H3G itself so as to affect a conclusion on significant market power or dominance. That is common ground. Whether it is a correct approach is what is in issue.
  112. As to CBP, the Decision records (at paragraph 3.25) that H3G had complained that OFCOM's consideration of BTs' CBP was "cursory and inadequate". While OFCOM had relied on an obligation on the part of BT to connect, H3G contended that no such obligation existed; it complained that OFCOM had not focussed on the power of BT to delay bringing about connection and that the agreement with BT prevented changes to its charges. In relation to these matters the Decision concluded:
  113. "3.30 … While, as '3' has pointed out, there are no formal conditions in place – because they have not previously been required – the May guidance explains that BT is expected to offer end-to-end connectivity in order to meet USO requirements to provide publicly available telephone services. This weakens BT's bargaining position as it removes the threat of BT not providing connectivity if agreement over charges cannot be reached.
    3.31 It is possible that during the initial interconnection negotiations between BT and '3', '3''s urgency to launch services was a relevant factor in the relative bargaining positions of each party. However, Ofcom's analysis in this market review must be forward-looking and consider '3''s likely position in the next 18-24 months. Therefore Ofcom must also consider future negotiations between '3' and BT.
    3.32 With such a forward-looking perspective, and with delay not such a critical issue for '3', it would be difficult to argue that '3' could not set excessive charges for the termination services provided to BT … It may be that existing contractual arrangements between '3' and BT make it difficult for '3 to raise charges from their current level. However, there is no arrangement in this contract for BT to ensure that charges fall over time from their current level (in line with costs). Some evidence of this is BT's inability to enforce reduced termination payments to '3 at the time of the 15 per cent charge reduction applied to the other MNOs in July 2003."
    For those reasons the Decision found that OFCOM considered that BT was under an obligation which led to the position that it did not have CBP that offset H3G's market power in call termination.
  114. In this appeal we received a significant body of evidence as to what happened in the negotiations between the parties leading up to the Interconnection Agreement. None of this evidence was challenged by cross-examination. It is material which would go to an assessment of whether BT had CBP at the time of those negotiations. However, we do not think that we need to consider it for that purpose. The decision did not find that there was no CBP at the time of the negotiations. In essence it assumed that there was (see the opening words of paragraph 3.31), but said that it was the future that the Decision had to look to. The extent to which it is necessary to look to those negotiations is therefore limited (though they are not necessarily completely without significance, as appears below).
  115. So far as dispute resolution by the regulator is concerned, paragraph 3.51 of the Decision refers back to paragraphs 4.3 to 4.9 of the December consultation and concluded that the possibility of dispute resolution would not, in practice, constrain an MNO from setting excessive termination charges. It also refers to the point in paragraph 4.14 of the Decision where it says:
  116. "4.14 In this context, Ofcom notes it has the power to resolve the price increase dispute in question by determining that it will not prevent the increase until it has exercised its powers to set, inter alia, an SMP condition (see section 190(4) of the Act). Accordingly, Ofcom does not accept that it has made a material error of fact in rejecting dispute resolution as a constraint on the MNOs' ability to price excessively."

  117. So far as the cross-reference to the December consultation is concerned, the material occurs in a chapter which considers the effect of SMP on prices and the extent to which it would have adverse effects for consumers. It comes to the conclusion that SMP would lead to higher prices, and in doing so considers (non-exhaustively) various possible outcomes of the dispute resolution process which OFCOM participates in as regulator. The passage does not analyse the terms of the BT/H3G interconnect agreement.
  118. BT'S POSITION, THE INTERCONNECT AGREEMENT AND THE NEGOTIATION OF THAT AGREEMENT

  119. At the heart of the CBP debate is OFCOM's case that BT is obliged to produce end-to-end connectivity, that is to say that BT is under a regulatory obligation to provide inter-connections between its systems and those of other operators. It has been under such an obligation for some years. The source of the obligation has changed, and since it is now common ground that it is under that obligation we do not need to set out the various sources from time to time at this point of the judgment (though we need to do it later). Suffice it to say for these purposes that it was under such an obligation in 2001, when it was negotiating with H3G, and it is still under such an obligation now. Were it unreasonably to refuse to connect to another operator, OFCOM as regulator could intervene. BT complies with this obligation by means of standard form documentation (a "reference offer"), offered to all parties who wish to connect to its network, including a standard form interconnect agreement which governs the relationship arising out of the interconnection. This document contains annexes which are to some extent tailored to the needs of the specific transaction. The termination charge is a matter of negotiation between the parties; BT publishes a table containing the wholesale charges made by it for its services.
  120. As Mr James Westby, Interconnect Manager for H3G, points out in his evidence, all MNO's have interconnect agreements with BT. This is primarily because BT originates 70% to 80% of all fixed-line traffic in the UK. In practice, therefore, an MNO must ensure that BT routes incoming calls from its network to that of the MNO. H3G shares the view that such an interconnect agreement with BT was necessary and preliminary negotiations took place at the end of 2000. A confidentiality agreement was concluded in 14th February 2001 and negotiations between technical representatives started to take place after that.
  121. One of the things that had to be agreed was a price to be charged by H3G for BT's terminating calls on H3G's network. H3G wanted to propose an interim price for testing purposes. Prices had to be publicly disclosed, but H3G did not want to use its final price because it did not want to publicise its pricing strategy at the time. BT's view was that testing was not necessary so the parties could go straight to a final price, but if H3G wished to have an interim price then they could agree a price at that point and then vary it under the price change mechanisms in the interconnect agreement.
  122. The H3G/BT interconnect agreement was signed on 13th August 2001, but without some of its schedules containing prices for services that H3G was to provide. These were to be completed later. At three following meetings, in September and October, BT pressed H3G to come up with a termination rate. At a further meeting on 16th November H3G at last came up with the rate that they proposed. BT had various bands of retail charges attributable to fixed to mobile calls – fm1 to fm5. They covered the rates charged to BT's customers for calling a given operator (or, strictly, a band of numbers allocated to an operator). Thus, for example, fm1 was that attributable to O2 numbers, and fm4 was attributable to Orange numbers. Each of those charging bands included a sum attributable to the cost charged by the mobile network operator. At this meeting on 16th November H3G proposed charging a termination charge which was that attributable to the most expensive rate charged by BT (fm2 - a rate applicable to a particular closed network). BT did not accept that rate without more – it asked for a breakdown of the precise rates proposed to be charged according to the times of day, and in due course declined to agree the pricing proposal on the grounds that it represented too high a cost for BT's retail customers in the absence of a cost-based justification. At a subsequent meeting on 5th December BT repeated that it would reconsider its position if H3G was prepared to justify its proposed charges, and if it did not wish to justify them then H3G might like to consider adopting a rate corresponding to an existing MNO's rates. After various exchanges, on 24th December 2001 H3G proposed a rate which was the equivalent of the rates charged by One2One (now T-Mobile), which was the most expensive of the MNO rates at the time. This was ultimately accepted and became the rate charged.
  123. We have given our reasons as to why it is not necessary to examine the negotiations in depth to see to what extent BT had CBP at the time of those negotiations, but since Mr Green says that guidance can be taken from those negotiations as to how future negotiations on price are likely to go we make the following points based on our review of the evidence:
  124. a) H3G was under some time pressure because it wished to launch its product in the first half of 2002. There was an increasingly urgent need to finalise the termination rates towards the end of 2002, part of which was due to the fact that BT's Wholesale Board (which had to approve such matters) only met once a month and deadlines for presenting to the December meeting were missed. However, the problems did not arise from delays created by BT. So far as things got urgent, they arose from a failure by H3G over a couple of months to propose rates and then back up its proposals.
    b) It is said by H3G that it was vulnerable to delays created by BT, and at least to that extent BT had CBP. That may be true, but there is no evidence that BT at any time did anything which might be regarded as introducing unnecessary or tactical delays. On the evidence we have seen it dealt with appropriate expedition throughout.
    c) When H3G proposed to adopt the fm2 band, BT did not reject it out of hand, or indeed reject it in any meaningful sense. What it said was that it would require the adoption of that band to be justified by H3G. H3G chose not to justify it, and chose to suggest another band to BT.
  125. The interconnect agreement between BT and H3G was dated 13th August 2001. It was supplemented by other matters by the time H3G launched its service, but that does not matter. What is important for present purposes is clause 13. The relevant provisions of that clause are the following; in reading the provisions of clause 13 it should be understood that the charge for an Operator service includes termination charges.
  126. "2.3 A Party may terminate this Agreement by giving at any time to
    the other not less than 24 months' written notice to terminate.
    13. OPERATOR SERVICES
    13.1 For an Operator service or facility BT shall pay to the Operator
    the charges specified from time to time in the Carrier Price List.
    13.2 The Operator may from time to time by sending to such person, as BT may notify to the Operator from time to time, a notice in writing in duplicate request a variation to a charge for an Operator service or facility ("Charge Change Notice"). Such notice shall specify the proposed new charge and the date on which it is proposed that the variation is to become effective ("Charge Change Proposal"). BT shall within 4 Working Days of receipt of such notice acknowledge receipt and within a reasonable time notify the Operator in writing of acceptance or rejection of the proposed variation.
    13.3 BT may from time to time by sending such person, as the Operator may notify to BT from time to time, a notice in writing in duplicate request a variation to a charge for an Operator service or facility ("Charge Change Notice"). Such notice shall specify the proposed new charge and the date on which it is proposed that the variation is to become effective ("Charge Change Proposal"). The Operator shall within 4 Working Days of receipt of such notice acknowledge receipt and within 14 days of receipt of such notice notify BT in writing of acceptance or rejection of the proposed variation. If the Operator has not accepted the Charge Change Proposal within 14 days of receipt of such notice (or such longer period as may be agreed in writing) the proposed variation shall be deemed to have been rejected.
    13.4 If the Party receiving a Charge Change Notice accepts the Change Proposal the parties shall forthwith enter into an agreement to modify the Agreement in accordance with the Charge Change Proposal.
    13.5 If the Party receiving a Charge Change Notice rejects the Charge Change Proposal the Parties shall forthwith negotiate in good faith.
    13.6 If following rejection of a Charge Change Proposal and negotiation, the Parties agree that the Charge Change Notice requires modification, the Party who sent the Charge Change Notice may send a further Charge Change Notice.
    13.7 If following rejection of a Charge Change Proposal and negotiation the Parties fail to reach agreement within 14 days of the rejection of the Charge Change Proposal, either party may, not later than 1 month after the expiration of such 14 days period, refer the matters in dispute to the Director General.
    13.8 If the Director General upholds the Charge Change Proposal in the Charge Change Notice without modification the Charge Change Proposal shall take effect on the date specified in the Charge Change Notice and the Parties shall forthwith enter into an agreement to modify the Agreement in the accordance with this paragraph 13.8.
    13.9 If the Director General does not uphold the Charge Change Proposal in the Charge Change Notice without modification then that Charge Change Notice shall cease to be of any effect. In the event that the Director General proceeds to make an order, direction, determination or requirement following a referral pursuant to paragraph 13.7 then the Party who sent the Charge Change Notice shall send a further Charge Change Notice in accordance with the order, direction, determination or requirement of the Director General and the parties shall forthwith enter into an agreement to modify the Agreement in accordance with this paragraph 13.9."
  127. Paragraph 19 contains provisions whereby either party can seek a review of the agreement in certain events (including a material change affecting the commercial basis of the agreement) and clause 20 provides for the Director General to determine any matters arising on the review negotiations which the parties have not managed to agree for themselves.
  128. Since the finalisation of the termination charges operating between BT and H3G, H3G has not sought to increase them (or otherwise have them reviewed) or to terminate the agreement (nor has BT). Had H3G been seriously concerned at the level of termination charges that it was entitled to charge it could have adopted either of those challenges. It is not misplaced to consider a termination of the agreement, because such a termination would not necessarily have left H3G without a connection to the BT network, because BT's end to end connectivity obligation would require it to renegotiate an agreement with H3G, and the regulator would have stepped in if there had been any disagreement as to the terms of that agreement. During the course of the hearing Mr Green frequently referred to the price agreed by H3G and BT as being "embedded". This seems to us to be a misplaced word – the provisions for variation, termination and renegotiation identified above demonstrate that it was not "embedded" in the proper sense of that word (connoting a strong degree of fixedness).
  129. The reasons why H3G did not seek to renegotiate the charge, implement the provisions of clause 13 or terminate and renegotiate the termination charges in the period up to April 2004 are set out in a witness statement of Ms Clare Laurent, who was first H3G's regulatory director and thereafter a consultant fulfilling similar functions until 2nd April 2004. She explains that the possibility of terminating the agreement was not seriously contemplated because of pressure on time and resources arising from the recent launch, a desire not to imperil the relationship with BT, the difficulties of analysing termination charges so shortly after the launch of products (presumably so as to satisfy BT as to the justifiability of revised charges), signalling commercial uncertainty and the unattractiveness of potentially inviting a price review by OFCOM. She understood that attempts had been made to enlist the assistance of Oftel in 2001 and 2002 by way of an intervention with BT because it was felt that BT had forced H3G to accept an unreasonably low termination charge, but the meeting with Oftel was not productive because Oftel "did not suggest any avenues of approach which would enable resolution within a commercially acceptable timeframe." The same reasons led H3G not to seek to vary the termination charges under clause 13 of the interconnect agreement – practical implications, the damage to the relationship with BT and the commercial uncertainties involved made such a course of action unattractive at a time when H3G's priorities were to achieve a successful rollout of its services.
  130. Mr Roth, for OFCOM, did not ask to cross-examine Ms Laurent, and in terms accepted that hers was a true account of the internal thinking of H3G at the time. There was therefore no exploration of certain potential inconsistencies between her evidence and other evidence – for example, the emails and memoranda relating to the negotiations demonstrate that the possibility of fixing an initial charge for launch which would subsequently be varied via the contractual mechanism was considered by both sides as something which was a perfectly acceptable business mechanism, and there is nothing to suggest that adopting it would in any way imperil the relationship between H3G and BT, contrary to her evidence. However, in the light of the fact that neither OFCOM nor BT sought to cross-examine her evidence stands unchallenged before us, though its weight is obviously still a matter for us.
  131. Against that background, we now pass on to consider the criticisms made by H3G of the Decision in respect of related matters.
  132. REGULATION

  133. It is part of H3G's case that, in assessing whether an entity has SMP, at least in a case like this where power over price is an essential element of the regulator's decision, it is relevant, if not important, to consider the effect of regulation or possible regulation on the entity in question. One of the matters that has to be taken into account in assessing whether or not the entity can charge excessive prices is the extent to which it would be restrained from doing so by the prospects of regulatory intervention to stop it. H3G says that the relevance of regulation is acknowledged in paragraph 3.39 of the Decision, but OFCOM did not consider the point fully, because it did not consider the impact of regulation on H3G. It took it into account so far as the position of BT was concerned, but failed to do so far as H3G was concerned. H3G's own evidence showed that the possibility of regulation was a curb, because an internal memorandum prepared in the course of the negotiations with BT, in November 2001, showed that it acknowledged that one of the constraints on its future pricing was the possibility of regulation by the then regulator. This point is probably related to the point already dealt with as to the extent of the necessary inquiry as to future conduct (including a consideration of incentives and disincentives). It is also related to the point as to dispute resolution, dealt with below, but Mr Green said it also had its own individual effect.
  134. As Mr Roth pointed out, this point is an unattractive one when taken as a matter of principle. If it were a good point then one could expect it to be a frequent if not a universal answer to any attempt to impose ex ante regulation. Any entity in respect of which it was said that it could behave independently of its market counterparts would be likely to say that it could not and would not do so because if it did then it could see that it would attract regulatory intervention. It would therefore argue that it does not in fact have significant market power within the guidance given in Article 14(2) of the Framework Directive. The argument might also be extended into cases of alleged abuse and ex post regulation, where logically it might be thought to apply on the same basis (notwithstanding the different direction in which the facts might be pointing) in relation to allegations of dominance. Thus one would have the paradox: there could never be SMP where one has a vigilant regulator, so ex ante regulation would never be appropriate (or indeed necessary) despite the fact that it could turn out (on the facts) that there was dominance (and abuse) after all. That is not an attractive scenario, and it is one which we would only espouse if we were bound to by authority.
  135. We do not consider we are bound to do that. Mr Green submitted that we were, and relied on Tetra Laval as authority for the proposition that the disincentives from regulation were matters which had to be taken into account. We do not consider that that case establishes that. The CFI in that case considered what conduct it was relevant to take into account in assessing the likelihood of dominance in one market being used (by "leverage") to create dominance in another. The ECJ judgment sets out the relevant paragraphs of the CFI judgment:
  136. "56 The Commission's argument challenges paragraphs 156 to 162 of the judgment under appeal, which immediately follow paragraphs 148 to 155, which were likewise challenged by the Commission and were examined by the Court in connection with the first ground of appeal. In those paragraphs, the Court of First Instance held as follows:
    '156. In the present case, the leveraging from the aseptic carton market, as described in the contested decision, would manifest itself - in addition to the possibility of the merged entity engaging in practices such as tying sales of carton packaging equipment and consumables to sales of PET packaging equipment and forced sales (recitals 345 and 365) - firstly, by the probability of predatory pricing by the merged entity (recital 364, cited in paragraph 49 above); secondly, by price wars; and, thirdly, by the granting of loyalty rebates. Engaging in these practices would enable the merged entity to ensure, as far as possible, that its customers on the carton markets obtain from Sidel any PET equipment they may require. The contested decision finds that Tetra holds a dominant position on the aseptic carton markets, that is to say, the markets for aseptic carton packaging systems and aseptic cartons (recital 231, see paragraph 40 above), a finding which is not disputed by the applicant.
    157. It should be recalled that, according to settled case-law, where an undertaking is in a dominant position it is in consequence obliged, where appropriate, to modify its conduct so as not to impair effective competition on the market regardless of whether the Commission has adopted a decision to that effect (Case 322/81 Michelin v Commission [1983] ECR 3461, paragraph 57; Case T-51/89 Tetra Pak v Commission [1990] ECR II-309, paragraph 23; and Joined Cases T-125/97 and T-127/97 Coca-Cola v Commission [2000] ECR II-1733, paragraph 80).
    158. Moreover, in response to the questions put by the Court at the hearing, the Commission did not deny that leveraging by Tetra through the conduct described above could constitute abuse of Tetra's pre-existing dominant position in the aseptic carton markets. This could also be the case, according to the concerns expressed by the Commission in its defence, in circumstances where the merged entity refused to participate in the installation and any necessary conversion of Sidel SBM machines, to provide after-sales service or to honour the guarantees for such machines when sold by converters. However, the Commission went on to state that the fact that a type of conduct may constitute an independent infringement of Article 82 EC does not preclude that conduct from being taken into account in the Commission's assessment of all forms of leveraging made possible by a merger transaction.
    159. In this regard, it must be stated that, although the Regulation provides for the prohibition of a merger creating or strengthening a dominant position which has significant anti-competitive effects, these conditions do not require it to be demonstrated that the merged entity will, as a result of the merger, engage in abusive, and consequently unlawful, conduct. Although it cannot therefore be presumed that Community law will not be complied with by the parties to a conglomerate-type merger transaction, such a possibility cannot be excluded by the Commission when it carries out its control of mergers. Accordingly, when the Commission, in assessing the effects of such a merger, relies on foreseeable conduct which in itself is likely to constitute abuse of an existing dominant position, it is required to assess whether, despite the prohibition of such conduct, it is none the less likely that the entity resulting from the merger will act in such a manner or whether, on the contrary, the illegal nature of the conduct and/or the risk of detection will make such a strategy unlikely. While it is appropriate to take account, in its assessment, of incentives to engage in anti-competitive practices, such as those resulting in the present case for Tetra from the commercial advantages which may be foreseen on the PET equipment markets (recital 359), the Commission must also consider the extent to which those incentives would be reduced, or even eliminated, owing to the illegality of the conduct in question, the likelihood of its detection, action taken by the competent authorities, both at Community and national level, and the financial penalties which could ensue.
    160. Since the Commission did not carry out such an assessment in the contested decision, it follows that, in so far as the Commission's assessment is based on the possibility, or even the probability, that Tetra will engage in such conduct in the aseptic carton markets, its findings in this respect cannot be upheld.
    161. Moreover, the fact that the applicant offered commitments regarding its future conduct is also a factor which the Commission should have taken into account in assessing whether it was likely that the merged entity would act in a manner which could result in the creation of a dominant position on one or more of the relevant PET equipment markets. There is no indication in the contested decision that the Commission took account of the implications of those commitments when it assessed the creation of such a position in future through leveraging.
    162. It follows from the foregoing that it is necessary to examine whether the Commission based its analysis of the likelihood of leveraging from the aseptic carton markets, and of the consequences of such leveraging by the merged entity, on sufficiently convincing evidence. In the course of that examination it is necessary, in the present case, to take account only of conduct which would, at least probably, not be illegal. In addition, since the anticipated dominant position would only emerge after a certain lapse of time, by 2005 according to the Commission, its analysis of the future position must, whilst allowing for a certain margin of discretion, be particularly plausible.'"
  137. At paragraph 74 of its judgment the ECJ held that the CFI was right to take incentives and disincentives into account, but in the next paragraph held that the prospects of regulatory intervention was not something that fell to be considered in that context. Thus:
  138. "74. Since the view is taken in the contested decision that adoption of the conduct referred to recital 364 in that decision is an essential step in leveraging, the Court of First Instance was right to hold that the likelihood of its adoption must be examined comprehensively, that is to say, taking account, as stated in paragraph 159 of the judgment under appeal, both of the incentives to adopt such conduct and the factors liable to reduce, or even eliminate, those incentives, including the possibility that the conduct is unlawful.
    75. However, it would run counter to the Regulation's purpose of prevention to require the Commission, as was held in the last sentence in paragraph 159 of the judgment under appeal, to examine, for each proposed merger, the extent to which the incentives to adopt anti-competitive conduct would be reduced, or even eliminated, as a result of the unlawfulness of the conduct in question, the likelihood of its detection, the action taken by the competent authorities, both at Community and national level, and the financial penalties which could ensue.
    76. An assessment such as that required by the Court of First Instance would make it necessary to carry out an exhaustive and detailed examination of the rules of the various legal orders which might be applicable and of the enforcement policy practised in them. Moreover, if it is to be relevant, such an assessment calls for a high probability of the occurrence of the acts envisaged as capable of giving rise to objections on the ground that they are part of anti-competitive conduct.
    77. It follows that, at the stage of assessing a proposed merger, an assessment intended to establish whether an infringement of Article 82 EC is likely and to ascertain that it will be penalised in several legal orders would be too speculative and would not allow the Commission to base its assessment on all of the relevant facts with a view to establishing whether they support an economic scenario in which a development such as leveraging will occur.
    78. Consequently, the Court of First Instance erred in law in rejecting the Commission's conclusions as to the adoption by the merged entity of anti-competitive conduct capable of resulting in leveraging on the sole ground that the Commission had, when assessing the likelihood that such conduct might be adopted, failed to take account of the unlawfulness of that conduct and, consequently, of the likelihood of its detection, of action by the competent authorities, both at Community and national level, and of the financial penalties which might ensue. Nevertheless, since the judgment under appeal is also based on the failure to take account of the commitments offered by Tetra, it is necessary to continue the examination of the second ground of appeal."
  139. In other words, the ECJ was disapproving the last sentence of paragraph 159 of the CFI judgment. It was doing so on the basis that it was not right to take into account the likelihood of Tetra Laval being penalised in any one or more of the member states (or at the Community level) in ascertaining the likelihood of Tetra Laval indulging in the potentially objectionable conduct. The reason for this was that it would be "too speculative" (see paragraph 77) and would not allow the Commission to base its assessment on all the relevant facts to see if one could infer the economic scenario which would lead to leveraging.
  140. It appears that the ECJ was drawing a fine distinction here. It was only the last sentence of paragraph 159 of the CFI decision that was not upheld by the ECJ. The rest of the paragraph was apparently approved. Accordingly, to some extent it was permissible to take into account the extent to which the unlawfulness of a given course of action would be a disincentive to adopting that course. It is not clear where the boundary lies between that factor (permissible) and the actual intervention of regulators (impermissible), but we do not think that it is necessary for us to draw that boundary. It is sufficient for us to say that the decision of the ECJ does not support Mr Green's thesis that the prospect of regulation has to be taken into account, and if anything it points the other way.
  141. What that decision does not do, of course, is to address the potential paradox that we have dealt with above. It does not deal with the point as a matter of principle. That is because it was addressing a different factual scenario. As we have pointed out more than once above, the case turned on a different point – whether and to what extent a party not in a dominant position in one market would abuse a dominant position in another in order to create a dominant position in that first market. That is different from the sort of inquiry which OFCOM had to carry out in the present case. However, despite that, we do not consider that it supports Mr Green's case on whether or not to take regulation into account, and if anything it points the other way.
  142. Reliance was placed by OFCOM on a decision of the Commission in Case DE/2005/0144 – "RegTP". OFCOM submitted that the reasoning of this decision supported its case on this point. Decisions of the Commission, though not strictly binding, are matters to which we should give due deference.[14] The decision was one given as a result of a consultation by RegTP (the German telephone services regulator) on a proposed decision as to whether various German operators had SMP. The relevant procedures gave the Commission an effective veto over the proposed decision, which it exercised in that particular case. In that context the effect of regulation on the counterparties to interconnection arrangements was considered by RegTP and by the Commission.
  143. RegTP was minded to find that the incumbent operator (DTAG) had SMP in the market for call termination on its network, but that 53 other networks ("ANOs") did not. DTAG was under a regulatory obligation to interconnect with the ANOs, but in assessing whether DTAG had countervailing buyer power RegTP had left this factor out of account. It did so on what were described as a "strict Greenfield approach" and a "modified Greenfield approach". The "strict" approach was one in which it assumed that the interconnection obligation did not exist. The "modified" approach was one in which the obligation was recognised to exist but under which it was regarded as methodologically wrong to take it into account in assessing the market power of each ANO.
  144. The Commission rejected the former approach. At paragraph 22 of its decision the Commission dealt with the "strict Greenfield" approach and stated:
  145. "In economic terms, it is not appropriate to exclude regulatory obligations that exist independently of a SMP finding on the market under consideration but that can have an impact on the SMP finding on the markets under consideration. From a methodological viewpoint obligations flowing from existing regulation, other than the specific regulation imposed on the basis of SMP status in the analysed market, must be taken into consideration when assessing the ability of an undertaking to behave independently of its competitors and customers on that market." (emphasis supplied).
    The emphasised words bring out a distinction. It was appropriate to take into account the existence of the regulatory interconnection obligation on DTAG, but not the effect of regulation on the very parties whose market power was under consideration.
  146. This point is said to be developed in paragraph 23 of the decision:
  147. "The purpose of a Greenfield approach is indeed to avoid circularity in the market analysis by avoiding that, when as a result of existing regulation a market is found to be effectively competitive, which could result in withdrawing that regulation, the market may return to a situation when there is no longer effective competition. In other words any Greenfield approach must ensure that absence of SMP is only found and regulation only rolled back where markets have become sustainably competitive, and not where the absence of SMP is precisely the result of the regulation in place."
    In other words, a potentially regulated person cannot claim that it does not have SMP because regulation has procured a situation in which it no longer has it. So long as it is regulation which is bringing about competitive outcomes, the markets are not competitive independently of that regulation. It follows that the potentially regulated person cannot say that it does not have SMP because the threat of regulation means that it does not have the necessary power. That would be circular and illogical. OFCOM relied on this reasoning.
  148. Although that Decision turned on a consideration of the effect of regulation on someone other than the person who is the subject of the investigation (the equivalent of BT in the present case) we agree that the reasoning applies as OFCOM says it does. The effect of this is that the possibility of regulation being brought to bear on H3G is a factor that cannot be prayed in aid by H3G as militating against its having SMP. We reiterate that H3G's submissions would give rise to an illogical and unattractive, if not an unprincipled, position, and we consider them to be wrong. The correct position is as found in the RegTP decision, namely that regulatory obligations on a market counterparty can be taken into account, but not the potential for regulation on the party whose market position is under consideration.
  149. COUNTERVAILING BUYER POWER

  150. We have set out paragraph 78 of the Commission's Guidelines above, in which it is stressed that NRAs should undertake a full analysis of the market to measure the power of an undertaking to behave to an appreciable extent independently of competitors, customers and consumers, and that the relevant criteria include:
  151. "absence of or low countervailing buyer power"
  152. It is part of H3G's case that OFCOM did not consider this factor to a sufficient degree. It would also be its case, we believe, that if OFCOM had done so it would have found sufficient CBP to prevent SMP arising, but before us it did not seek to make a positive case to that extent and contented itself with demonstrating that OFCOM did not carry out a sufficient analysis.
  153. It is therefore necessary to identify what OFCOM determined and found in relation to CBP. The May review contains OFCOM's first views:
  154. "4.21 Countervailing buyer power exists when a particular purchaser (or group of purchasers) of a good or service is sufficiently important to its supplier to influence the price charged for that good or service. In order to constrain the price effectively, the purchaser must be able to bring some pressure to bear on the supplier to prevent a price rise by exerting a credible threat, for example not to purchase or to self-provide.
    4.22 In this case, the question of whether each MNO providing voice call termination has SMP depends on the extent to which its monopolistic position may be off-set by the buyer power of purchasers.
    4.23 BT is the major buyer of voice call termination on mobile networks (see table 4.2 below). In theory BT might credibly threaten not to purchase termination from an MNO and this would deprive that MNO of the pricing freedom that it derives from its monopoly over termination. In practice, this issue is irrelevant since BT, even if it did have buyer power, has not been able to exert it because of its obligation to complete all calls whatever the terminating network. The reasons for this obligation will be set out in the document End to End Connectivity (to be published in May 2003). This regulatory requirement curbs any buyer power that BT may have and leaves the MNOs free to set terminating charges above the competitive level."
  155. This was followed up in the December Interim Statement:
  156. "3.32 Countervailing buyer power exists when a particular purchaser (or group of purchasers) of a good or service is sufficiently important to its supplier to influence the price charged for that good or service. In order to constrain the price effectively, the purchaser must be able to bring some pressure to bear on the supplier to prevent a price rise by exerting a credible threat, for example not to purchase or to self-provide.
    3.33 In this case, the question of whether each MNO providing voice call termination has SMP depends on the extent to which its monopolistic position may be off-set by the buyer power of purchasers.
    3.34 BT is the major buyer of voice call termination on mobile networks … [figures provided]. In theory, BT might credibly threaten not to purchase termination from an MNO and this would deprive that MNO of the pricing freedom that it derives from its monopoly over termination. In practice, this issue is irrelevant since BT, even if it did have buyer power, has not been able to exert it because of its obligation to complete all calls whatever the terminating network. The reasons for this obligation were set out in Oftel's guidance document End to End Connectivity (published in May 2003). That requirement curbs any buyer power that BT may have and leaves the MNOs free to set terminating charges above the competitive level.
    3.35 In its response to the first consultation, '3' claimed that even with the existence of the end-to-end connectivity obligation conferred on BT, '3' does not have the ability to raise termination rates. Orange raised a similar concern in its response. The Director does not accept this claim. BT must ensure that its customers can call customers and services, irrespective of terminating network, i.e. it must provide end-to-end connectivity. It is therefore incorrect to assert that BT could properly exert countervailing buyer power to force an MNO to set to set lower termination rates with the threat of refusal to interconnect."
  157. OFCOM's deliberations in this respect were finalised in paragraphs 3.30 to 3.33 of the Decision:
  158. "3.30 In relation to the point about BT's countervailing buyer power, Ofcom does not believe that the existing regulatory framework would, in practice, allow BT (as an originating operator) to reject price increases by '3'. While, as '3' has pointed out, there are no formal conditions in place - because they have not previously been required - the May guidance explains that BT is expected to offer end-to-end connectivity in order to meet US0 requirements to provide publicly available telephone services. This weakens BT's bargaining position as it removes the threat of BT not providing connectivity if agreement over charges cannot be reached.
    3.31 It is possible that during the initial interconnection negotiations between BT and '3', '3''s urgency to launch services was a relevant factor in the relative bargaining positions of each party. However, Ofcom's analysis in this market review must be forward-looking and consider '3"s likely position in the next 18-24 months. Therefore, Ofcom must also consider future negotiations between '3' and BT.
    3.32 With such a forward-looking perspective, and with delay not such a critical issue for '3', it would be difficult to argue that '3' could not set excessive charges for the termination services provided to BT. With specific regard to '3''s evidence, Ofcom believes that it refers to the specific circumstances which '3' was in prior to offering services to the public. However, it does not provide a sufficient indication of how future negotiations with BT would run, given the change in '3''s circumstances (i.e. previously it required an interconnection agreement with BT to start operating, but that is no longer the case). It may be that existing contractual arrangements between '3' and BT make it difficult for '3' to raise charges from their current level. However, there is no arrangement in this contract for BT to ensure that charges fall over time from their current level (in line with costs). Some evidence of this is BTs inability to enforce reduced termination payments to '3' at the time of the 15 per cent charge reduction applied to the other MNOs in July 2003.
    3.33 Hence, for the reasons set out above Ofcom considers that BT is under an obligation which leads to a position where it does not have countervailing buyer power that off-sets '3''s market power in call termination."
  159. OFCOM's position therefore focuses on its assessment of the period after the initial negotiation. It looked back to the facts surrounding the finalisation of the Interconnect Agreement and the fixing of termination charges but made no finding as to whether CBP existed then. On a proper analysis of the Decision, it seems that OFCOM was prepared to assume that the circumstances then were such that there might have been CBP (because of the urgency that was said to have accompanied the position) although OFCOM made no finding in that respect, but then looked to the future (measured from the completion of the Interconnect Agreement) and found (in essence) that the effect of the urgency was spent when agreement on pricing was reached. At that point the effect of the end-to-end connectivity obligation weakened BT's bargaining position. The extent of that weakening is not identified, but the Decision finds that it would be difficult to argue that H3G could not set excessive termination charges, and the natural inference is that that is because of the end-to-end connectivity obligation. The following important factors emerge from this part of the Decision:
  160. i) OFCOM's reasoning does not depend on any assessment of the actual relationship between H3G and BT; rather, it depends on its overall structure as defined by the connectivity obligation and the Interconnect Agreement with no assessment of how matters would be likely to work in practice.
    ii) The connectivity obligation is probably the key factor. It is this that is said to negate (to an appropriate but otherwise undefined extent) the bargaining power that BT would otherwise have.
    iii) As an aspect of (ii), no reasoning is provided as to why it was considered difficult to resist the notion that H3G could set excessive prices in the future, save for the existence of the Interconnect Agreement (by inference).
  161. H3G says that all this is inadequate. Proper consideration should have been given to the situation; it was not appropriate to rely on the end to end connectivity obligation and say that that negated any CBP that BT had or might have had. A fuller consideration of the picture should have been obtained, including a review of what happened in the 2001 negotiation to see what conclusions could be drawn from that as to BT's bargaining power. If one were to look at what happened in those negotiations, there is material which pointed away from BT's being deprived of CBP. H3G does not invite us to take a view the other way (i.e. to find that there was CBP and therefore no SMP); it contents itself with submitting that the approach was flawed and therefore needs to be reconsidered.
  162. OFCOM maintains that BT does not have CBP, and did not have it at the time of the Decision. It relies on guidelines issued by the Commission in relation to horizontal mergers in 2004 (2—4/C 31/03) in which the Commission elaborated on what was meant by CBP.
  163. "64. The competitive pressure on a supplier is not only exercised by competitors but can also come from its customers. Even firms with very high market shares may not be in a position, post-merger, to significantly impede effective competition, in particular by acting to an appreciable extent independently of their customers, if the latter possess countervailing buyer power. Countervailing buyer power in this context should be understood as the bargaining strength that the buyer has vis-à-vis the seller in commercial negotiations due to its size, its commercial significance to the seller and its ability to switch to alternative suppliers.
    65. The Commission considers, when relevant, to what extent customers will be in a position to counter the increase in market power that a merger would otherwise be likely to create. One source of countervailing buyer power would be if a customer could credibly threaten to resort, within a reasonable timeframe, to alternative sources of supply should the supplier decide to increase prices or to otherwise deteriorate quality or the conditions of delivery. This would be the case if the buyer could immediately switch to other suppliers, credibly threaten to vertically integrate into the upstream market or to sponsor upstream expansion or entry for instance by persuading a potential entrant to enter by committing to placing large orders with this company. It is more likely that large and sophisticated customers will possess this kind of countervailing buyer power than smaller firms in a fragmented industry. A buyer may also exercise countervailing buying power by refusing to buy other products produced by the supplier or, particularly, in the case of durable goods, delaying purchases.
    66. In some cases it may be important to pay particular attention to the incentives of buyers to utilise their buyer power. For example, a downstream firm may not wish to make an investment in sponsoring new entry if the benefits of such entry in terms for lower input costs could also be reaped by its competitors."
    These passages place considerable reliance on the ability of the purchaser to find an alternative supplier, and at the hearing before us so did OFCOM. Mr Roth drew our attention to two Commission decisions, in each of which he said the ability (or lack of ability) of the buyer to find an alternative source of supply was determinative. In Alcatel/Telettra (Case No IV/MO42 [1991] OJ L122/48), the sellers (Alcatel and Telettra) were suppliers of telecommunications equipment with very large market shares. They were the main suppliers to Telefonica, a substantial Spanish telecommunications operator. The high market share of the suppliers was capable of indicating dominance, but the Commission found that Telefonica was capable of increasing supplies from their competitors, and that prevented a conclusion that the suppliers had dominance. The decision in CVC/Lenzing (Case COMP/M.2187 [2004] OJ L143/1) was said by Mr Roth to demonstrate a decision the other way – the absence of an ability to substitute sources of supply meant that there was no CBP, and he relied on paragraph 193 of the Decision. These decisions demonstrated that the ability to find alternative sources of supply was key. In the present case BT had no alternative source of supply – no-one else could provide termination on H3G's network – and since it could not walk away (because of the end to end connectivity obligation) there was no countervailing buyer power to counteract the power of H3G in relation to its termination charges. That left H3G with SMP arising out of the factors appearing above.
  164. BT supported OFCOM's case on this point.
  165. Thus OFCOM's case on CBP turns on two principal points – the obligation of BT to connect (and maintain a connection) because of the end to end connectivity obligation, and the inability of BT to have resort to a different source of supply. The Decision focuses on the former, with little reference to the latter; the argument before us focused more on the latter. Together they clearly amount to factors which are capable of negating any CBP.
  166. However, the way in which the case is put by H3G turns on whether they are, on the facts, sufficient, or whether there should have been a further consideration of the facts in order to make a more fully informed decision about CBP. In this context the first thing that we point out is the significance of CBP. The legislation does not contain a definition to which one can turn. It is part of an overall assessment of whether there is SMP. We consider that the following points have to be borne in mind:
  167. a) The underlying principle in a case like this is whether there is effective competition. SMP is a tool in determining this question – indeed, it is the central tool.
    b) Various factors are relevant in determining whether there is SMP, and one of those is CBP.
    c) For these purposes the right question is not the binary one of whether CBP exists or not. In other words, it is not enough to ask whether there is CBP, and if so to hold that there cannot be SMP. CBP is the power of counterparties to offset the powers of the party whose allegedly superior powers are under consideration, and the important question is what degree of CBP is there, and (bearing in mind all the circumstances) does it operate to a sufficient extent so as to mean that there is no SMP? CBP is not an absolute concept in terms of its strength. It is a concept which embodies a possible range of strengths. In any case where it is relevant, the relevant question is likely to be not whether there is CBP or not, but whether there is any CBP, and if so how much and what effect does it have.
  168. We turn first to OFCOM's treatment of, and reliance on, the "no alternative supplier" point. It is, of course, obviously true as a matter of fact. On the facts there is no other way in which BT can procure termination on H3G's network other than by purchasing termination from H3G itself. However, we do not accept that taken by itself it is sufficient to negate bargaining power to an extent sufficient to destroy any CBP that might otherwise have existed. Such an approach is too mechanistic. The assessment of SMP is not a mechanical one. It is one in which a number of factors have to be assessed, and they have to be assessed in their context. One of them is the buying power of the buyer. Various factors will contribute to, or detract from, the power of the buyer, and they will have various strengths depending on the market in question. Doubtless the ability of a buyer to find alternative sources will be very telling, but we do not consider that either of the two cases referred to or paragraphs 64 and 65 of the merger Guidelines require one to elevate that factor to be a touchstone in every case. In some cases it might well be conclusive. It happened to be a very important factor in the Alcatel/Telettra decision, and its absence was significant in CVC/Lenzing, but those were cases on their facts. Paragraph 64 of those Guidelines correctly identifies the exercise as being one in which the key question is bargaining strength. The three factors identified there (relative sizes, commercial significance and the ability to switch suppliers) should not be taken as being necessarily exhaustive factors.
  169. Accordingly we do not consider that the absence of an ability to switch means that nothing else need be investigated. If there should have been further consideration of factors that were not considered then this factor does not excuse its not being carried out.
  170. We turn therefore to the Decision and H3G's criticism of it. The structure of the Decision makes it plain that it is part of a developing decision-making process, having started from earlier views, allowing submissions to be made on those views and then coming to a final view. Because of the wording of the Decision it is necessary to consider that context in order to make clear what the Decision says about CBP, because there might be thought to be a certain amount of equivocation between paragraphs 3.30 and 3.32.
  171. The earlier documents clearly give the end to end connectivity obligation a central, and probably determinative, position. The May and December reviews clearly indicate that the obligation, by itself, removes any bargaining position that BT had or might have had. In its submissions H3G disputed that conclusion (and indeed it contested that there was such an obligation, but that does not matter for present purposes since it now accepts that there is such an obligation). At paragraph 4.12 of its response to the May paper it set out an explanatory memorandum from the European Commission's Recommendation on Market Definition:
  172. "Small networks will normally face some degree of buyer power that will limit greatly the associated market power … The existence of a regulatory requirement to negotiate interconnection in order to ensure end-to-end connectivity redresses this imbalance of market power. However, such a regulatory requirement would not endorse any attempt by a small network to set excessive termination charges. Consequently, there is still likely to be an imbalance of market power between large and small networks because it would be easier for a large network to initiate the step of raising call termination charges and it would be more difficult for a small network to resist a move by a large network to lower termination charges."
  173. At paragraph 2.10 of its response to the December consultation H3G asserts that it was important for the Director (who was then still in place) to make a detailed analysis of the effects of end-to-end connectivity before determining that it removed any question of CBP, and protested that that had not been done. It made submissions as to the sort of things that ought to have been considered, focusing a lot on delay but also making reference to the contractual mechanism for dispute resolution, and submitting (in paragraph 2.22) that future negotiations of termination price would not permit H3G to increase its prices to an excessive level without encountering strong resistance and countervailing power from BT.
  174. The response of OFCOM, having considered this material, is set out above. Paragraph 3.30 of the Decision is not on its face quite as unequivocal as its earlier pronouncements as to the effect of the end-to-end connectivity obligation, but we think that it seems to be intended to say the same thing. We are strengthened in this view by OFCOM's arguments before us. In its skeleton argument it contended that:
  175. "…once it is clear that BT has an obligation to provide end-to-end connectivity and therefore must do a deal with H3G, which is the only source of access to subscribers on its network, BT has no effective negotiating ploys to counter H3G's market power derived from its 100% of the market and the absolute barriers to entry to that market. BT cannot walk away if it does not like the terms on offer. If it delayed unreasonably in reaching agreement it would face regulatory intervention. The inevitable conclusion therefore is that BT does not have sufficient buyer power to off-set H3G's position on H3G's own network.
    Mr Roth emphasised this further when he said, in oral submissions:

    "BT alone, it is found, did not have any buyer power because of end-to-end connectivity"
    Thus OFCOM has maintained its position as to the effect of the end-to-end connectivity obligation.
  176. Having said that, however, the Decision demonstrates that some consideration was given to wider factual matters. Paragraphs 3.31 and 3.32 demonstrate that OFCOM recognised that it was relevant to consider what might happen in future negotiations, and it rejected submissions based on the power to create delays (or merely the existence of delays in price determination in subsequent negotiations). It was entitled to reject those submissions (and no complaint is made about that). However, there is no indication that it actually considered any wider factual picture, despite its acknowledgment that it had to consider the future, so it seems that in concluding that it would be "difficult to argue that H3G could not set excessive prices in the future" it seems that OFCOM was relying only on the existence of the end-to-end connectivity obligation.
  177. We consider that in this respect OFCOM's reasoning and Decision are flawed. We consider that there were two errors. The first is the determination or assumption that the end-to-end connectivity obligation removed any bargaining power BT might otherwise have had, with the effect that likely or possible future commercial scenarios were not considered by OFCOM. The second is an apparent misunderstanding of OFCOM's powers in relation to dispute resolution. Ultimately these factors are linked.
  178. As to the first, it is helpful to understand the obligation and its consequences. At the time of H3G's initial negotiation with BT, the then European framework included the 1997 Interconnection Directive (Directive 97/33/EC). Under Article 9 NRA's were to encourage and secure interconnection, taking into account the need to ensure satisfactory end-to-end communications for users, and were to intervene as necessary to specify conditions, terms and other things necessary to achieve that. The NRA was also to determine interconnection disputes. Article 9(6) provided that NRA's were, as a last resort, to be able to require organisations providing public telecommunications services to provide interconnect services where interconnection had not taken place, and to set terms of interconnection. That Directive was implemented by the Telecommunications (Interconnection) Regulations 1997 (SI 1997/2931). Regulation 5 provides that the Secretary of State and the Director should ensure that relevant licensees (which will have included BT) should ensure that certain conditions were met for (inter alia) interconnection to the telecommunications systems, and Regulation 6 provided:
  179. "(1) In exercising their functions conferred by or under the Act, and these Regulations, the Secretary of State and the Director shall encourage and secure adequate interconnection in the interests of all users, exercising their responsibility in a way that provides maximum economic efficiency and gives the maximum benefit to end-users, and in doing so shall have regard to the following -
    [a] the need to ensure satisfactory end-to-end communication for end-users; […]
    [f] the principles of non-discrimination (including equal access) and proportionality;
    [g] the need to maintain and develop a universal service.
    […]
    (3) In pursuit of the aims stated in paragraph (1) above the Director may intervene at any time, and shall do so on the request of either party, in order to make a direction specifying issues which must be covered in an interconnection agreement, or to make a direction that specific conditions be observed by one or more parties to such an agreement. The Director may in exceptional circumstances make a direction that changes be made to interconnection agreements already concluded where it is justified to ensure effective competition or interoperability of services for users or both.
    […]
    (6) Where there is a dispute concerning interconnection between organisations the Director shall, at the request of either party, take steps to resolve the dispute within six months of the date of the request. The direction which the Director makes to resolve the dispute shall represent a fair balance between the legitimate interests of both parties. The direction shall be notified to the parties and published in accordance with regulation 8(3). The parties concerned shall be given a full statement of the reasons on which it is based.
    […]
    (8) In exercising his duties under paragraphs (6) and (7) above, the Director shall take into account inter alia -
    [a] the interests of users; […]
    [e] the desirability of ensuring equal access arrangements; […]
    [k] the need to maintain a universal service.
    […]
    (10) Where Public Operators described in Schedule 2 have not interconnected their facilities the Director may, in accordance with the principle of proportionality and in the interests of end-users, make a direction that the Public Operators concerned shall interconnect their facilities. Any such direction shall be made only as a last resort in order to promote essential public interests. The direction may, where appropriate, set the terms of interconnection."
  180. BT was given a licence which dealt with a number of aspects, and paragraph 45.1 contained an obligation to offer to enter into an agreement with an operator such as H3G, and paragraph 45 went on as follows:
  181. "45.2 The Licensee or the Schedule 2 Public Operator may at any time request the Director to make a direction in order:
    (a) to specify issues which must be covered in an Interconnection agreement;
    (b) to lay down specific conditions to be observed by one or more parties to the agreement; or
    (c) if he thinks fit, to set time limits within which negotiations are to be completed,
    and a direction under this paragraph operates as an exercise by the Director of the power of direction conferred by regulation 6(3) or 6(4) of the Interconnection Regulations, as the case may be.
    45.3 The Licensee shall secure that the agreement or amendment referred to in paragraph 45.1 above is offered on terms and conditions which are reasonable.
    45.4 To the extent that the terms and conditions of any agreement or amendment made under paragraph 45.1 cease to be reasonable, the Licensee shall, within a reasonable period, offer to the Schedule 2 Public Operator, or agree with such Operator, as the case may be, to amend the agreement so that its terms and conditions are reasonable.
    45.5 The Licensee shall:
    (a) comply with the requirements of any directions given to the Licensee under paragraph 45.2 above or under regulation 6(3) or 6(4) of the Interconnection Regulations in relation to any negotiations or agreement to which it is or is intended to be a party;
    (b) comply with the requirements of any direction given to the Licensee under regulation 6(6) or 6(7) of the Interconnection Regulations in relation to any dispute over the terms of an agreement or amendment made under paragraph 45.1 above;
    (c) where the Director specifies conditions based on essential requirements pursuant to regulation 7(1) of the Interconnection Regulations for inclusion in an Interconnection agreement to which the Licensee is a party, forthwith secure the incorporation of those terms and conditions in such an agreement;
    (d) comply with any requirement made by the Director as a last resort under regulation 6(10) of the Interconnection Regulations to interconnect in order to protect essential public interests, and comply with any terms set by the Director for such purpose;
    (e) comply with any decision by the Director under regulation 10(2) of the Interconnection Regulations; and
    (f) comply with any facility or property sharing arrangements, or both, specified by the Director in accordance with regulation 10(3) of the Interconnection Regulations.
    45.6 So long as section 11 of the Restrictive Trade Practices Act 1976 is still in force an agreement made pursuant to this Condition shall not contain any restrictive provision, unless, before the agreement is made, the Director has consented to the inclusion of such a provision. For the purposes of this paragraph, a provision in an agreement is a restrictive provision if by virtue of the existence of such a provision (taken alone or with other provisions) the agreement is one to which the Restrictive Trade Practices Act 1976 would apply but for paragraph 1(1) of Schedule 3 to that Act.
    45.7 Paragraph 45.1 above does not apply to the extent that the Director has consented to limiting such obligation on a temporary basis and on the grounds that there are technically and commercially viable alternatives to the Interconnection requested, and that the requested Interconnection is inappropriate in relation to the resources available to meet the request.
    45.8 For the avoidance of doubt:
    (a) any question as to whether any term or condition (including a charge) is reasonable shall be decided by the Director having regard to any guidelines on the application of this Condition issued from time to time by the Director; and
    (b) in considering whether a term or condition (including a charge) is reasonable, the Director may take into account, inter alia, the effective date of the term or condition and the period during which such term or condition may already have been in effect; the Director may conclude that a reasonable charge is one which is offered or agreed, as the case may be, on terms that it take effect in agreements or amendments made under paragraph 45.1 above from the date of a complaint or the date on which the term was first offered by the Licensee or accepted by a Schedule 2 Public Operator or from any other date which is considered by the Director to be appropriate in the circumstances."
  182. The overall effect of that regime was that BT was indeed under an obligation to allow interconnection so as to provide end-to-end connectivity. However, it was not obliged to do so on terms specified by the other operator. The whole scheme obviously envisaged negotiation, and that terms and conditions would have to be agreed. In the event of a disagreement as to any terms the Director was to decide the term in question. It is quite clear that this regime extended to charges made by either party to the other.
  183. In due course OFCOM was established and took over from the Director. The Communications Act was intended to give effect to the Framework Directive, the Access Directive and the Authorisation Directive. OFCOM is given various functions, all of which are to be exercised in accordance with a principal duty set out in section 3(1):
  184. "3 General duties of OFCOM

    (1) It shall be the principal duty of OFCOM, in carrying out their functions—

    (a) to further the interests of citizens in relation to communications matters; and
    (b) to further the interests of consumers in relevant markets, where
    appropriate by promoting competition."
    There then follow other qualifications and elaborations on the duties of OFCOM. Sections 45 and following empower OFCOM to impose various conditions on network suppliers and operators. It is not necessary to set them out in detail but they include the power to impose conditions as to access to networks.
  185. Sections 185 to 191 provide for OFCOM to decide disputes. So far as material they provide as follows:
  186. "185. Reference of disputes to OFCOM
    (1) This section applies in the case of a dispute relating to the provision of network access if it is-
    (a) a dispute between different communications providers;
    (b) a dispute between a communications provider and a person who makes associated facilities available;
    (c) a dispute between different persons making such facilities available;
    (d) a dispute relating to the subject-matter of a condition set under section 74(1) between a communications provider or person who makes associated facilities available and a person who (without being such a person) is a person to whom such a condition applies; or
    (e) a dispute relating to the subject-matter of such a condition between different persons each of whom (without being a communications provider or a person who makes associated facilities available) is a person to whom such a condition applies.
    (2) This section also applies in the case of any other dispute if-
    (a) it relates to rights or obligations conferred or imposed by or under this Part or any of the enactments relating to the management of the radio spectrum that are not contained in this Part;
    (b) it is a dispute between different communications providers; and
    (c) it is not an excluded dispute.
    (3) Any one or more of the parties to the dispute may refer it to OFCOM.
    […]
    (8) For the purposes of this section—
    (a) the disputes that relate to the provision of network access include disputes as to the terms or conditions on which it is or may be provided in a particular case; and
    (b) the disputes that relate to an obligation include disputes as to the terms or conditions on which any transaction is to be entered into for the purpose of complying with that obligation.

    […]

    186. Action by OFCOM on dispute reference
    (1) This section applies where a dispute is referred to OFCOM under and in accordance with section 185.
    (2) OFCOM must decide whether or not it is appropriate for them to handle the dispute
    (3) Unless they consider-
    (a) that there are alternative means available for resolving the dispute,
    (b) that a resolution of the dispute by those means would be consistent with the Community requirements set out in section 4, and
    (c) that a prompt and satisfactory resolution of the dispute is likely if those alternative means are used for resolving it,
    their decision must be a decision that it is appropriate for them to handle the dispute.
    […]
    188. Procedure for resolving disputes
    (1) This section applies where—
    (a) OFCOM have decided under section 186(2) that it is appropriate for them to handle a dispute; or
    (b) a dispute is referred back to OFCOM under section 186(6).
    (2) OFCOM must—
    (a) consider the dispute; and
    (b) make a determination for resolving it.

    […]

    190. Resolution of referred disputes
    (1) Where OFCOM make a determination for resolving a dispute referred to them under this Chapter, their only powers are those conferred by this section.
    (2) Their main power (except in the case of a dispute relating to rights and obligations conferred or imposed by or under the enactments relating to the management of the radio spectrum) is to do one or more of the following-
    (a) to make a declaration setting out the rights and obligations of the parties to the dispute;
    (b) to give a direction fixing the terms or conditions of transactions between the parties to the dispute;
    (c) to give a direction imposing an obligation, enforceable by the parties to the dispute, to enter into a transaction between themselves on the terms and conditions fixed by OFCOM; and
    (d) for the purpose of giving effect to a determination by OFCOM of the proper amount of a charge in respect of which amounts have been paid by one of the parties of the dispute to the other, to give a direction, enforceable by the party to whom the sums are to be paid, requiring the payment of sums by way of adjustment of an underpayment or overpayment."
  187. This regime entitles OFCOM to require that telecommunications companies such as BT provide connectivity to others. In anticipation of its coming into force Oftel issued a Guidance document dated 27th May 2003 in which it set out its policy in this respect. Chapter 2 deals with the point. Paragraph 2.8 observes that the Universal Services Directive requires Member States to ensure that all reasonable requests for connection at a fixed location to the public telephone network and for access to publicly available telephone services at a fixed location are met by at least one undertaking, and paragraph 2.6 emphasises the importance of this to the consumer (who would otherwise be left to talking only to people on his or her own network). It goes on to provide:
  188. "2.10 Oftel proposes to continue the existing policy that USO providers, in meeting reasonable requests to provide access to PATS [Publicly Available Telephone Services] must ensure that their customers can call other customers and services irrespective of terminating network, that is they must provide end-to-end connectivity.
    2.11 In the USO consultation document, Oftel is proposing to designate BT and Kingston as Universal Service providers, although Kingston's designation will only apply in respect of activities in the Hull area."

    2.15 … Oftel recognises that the obligation on BT and Kingston to purchase call termination from other network providers could, in the absence of regulation, result in those other network providers exploiting their SMP and charging unreasonable prices when they sell termination services to BT and Kingston.
    2.16 For this reason, Oftel has proposed that all fixed PECNs [Public Electronic Communications Networks] should be required to provide call termination to all other PECNs on fair and reasonable terms if in receipt of a reasonable request to do so. In the event of a dispute, the Director would decide what constitutes fair and reasonable terms."

  189. What all that amounts to is a regime in which interconnection is to be expected, and compelled if necessary, but not on whatever terms a provider specifies. The regime was capable of giving rise to an end-to-end connectivity obligation, and the May Guidance made it clear that it would. But it is not the complete picture to say that BT was under an obligation to supply that connectivity. BT was under such an obligation, but not on whatever terms another network operator might propose. A complete description of the obligation involves adding that the terms would be agreed between the parties or determined by the regulator, and they would be "fair and reasonable" (paragraph 2.16 of the May Guidance).
  190. The Decision correctly identifies an end-to-end connectivity obligation but does not clearly go on to describe how it operates. Instead it assumes that the existence of the obligation removes such countervailing buyer power as BT had, or at least negates it to such an extent as leaves H3G with SMP. We consider this approach to be over-simplified and flawed. The assessment of CBP is an assessment of how the market actually operates (or is likely to operate) on the true facts, not on artificial "facts" or partial facts. If it is correct to bring the obligation into the equation (and we think it is) it must be viewed realistically and for what it is. Were it the case that the obligation were simply an obligation with no qualifications as to the terms on which interconnection was to be achieved then it would remove BT's bargaining power completely. However, that does not describe it properly. It is an obligation with some room for manoeuvre on negotiation, because the terms are to be reasonable and ultimately any dispute will be settled by someone else (the regulator). It is an obligation in those terms which has to be considered in the context of an assessment of CBP. To look just at the obligation is not to consider the true facts of the case.
  191. We consider that this approach is supported by the approach of the Commission. In its reconsideration of the German regulator's reconsidered approach to SMP (which took place after the Commission's veto of the first decision, referred to above) [15], the Commission noted that the regulator had applied a "modified Greenfield" approach to the obligation of the major fixed line operator to connect, reflecting that
  192. "… DTAG [i.e. the operator] is generally obliged to interconnect with ANOs, but that such an obligation would not require DTAG to accept unreasonable conditions for interconnection. Hence, while DTAG would have an interconnection obligation, it could still refuse unacceptably high termination rates charged by an ANO, and thereby also exercise countervailing buyer power.
    Following the Commission's criticism that the strict Greenfield approach seemed to be inappropriate in the circumstances of the German market, and that the modified Greenfield approach had been wrongly applied by [the regulator], [the regulator] undertook a new SMP assessment."
    It should be noted that what is described as the modified Greenfield approach coincides with what we think the correct approach is in relation to this particular area, and that the Commission did not think that it was wrong in principle to take that approach.
  193. This error on the part of OFCOM means that it failed to consider the full possible scope and effect of CBP. Its consideration stopped short of a full consideration. It is, of course, quite possible that it would have arrived at the same conclusion had it conducted the full exercise, but we have not been invited to determine that.
  194. There is a second error apparently underlying OFCOM's position (or at least its present position) on this point. The error relates to its perception of the limits to its powers in this area, as expressed in submissions. Part of the regulatory picture at this stage of the argument is the fact that under the statute OFCOM has (or appears to have) the power to determine the price of connection if there is a disagreement between the parties about it. As part of his argument in this appeal Mr Roth sought to argue that OFCOM did not have that power unless it had first made an SMP decision in relation to the party seeking to charge the price. This, if correct, would take the possibility of dispute resolution out of the picture, and perhaps strengthen the case for saying that BT's bargaining position was weakened to the extent that it had no sufficient CBP to stand against the apparent strength of H3G's position. Mr Roth went so far as to submit that in the absence of an SMP designation, OFCOM would have to decide the pricing dispute in favour of H3G, because to do otherwise would be to impose forbidden price control. He based his argument on the true construction of the Access Directive.
  195. We do not agree that that is the effect of the relevant provisions. We have set out above the relevant provisions of the 2003 Act. There is nothing there that supports Mr Roth's arguments. Section 190(4) refers to SMP conditions, but nothing in the wording of the Act suggests that SMP had to be found before the regulator decided a dispute over price. Mr Roth's arguments centred around Article 8 of the Access Directive. Paragraph 2 of that Article provides that:
  196. "… where an operator is designated as having significant market power on a specific market … national regulatory authorities shall impose the obligations set out in Articles 9 to 13 of this Directive as appropriate".
    Article 8(3) provides that:
    "Without prejudice to:
    the provisions of Articles 5(1), 5(2) and 6

    national regulatory authorities shall not impose the obligations set out in Articles 9 to 13 on operators that have not been designated in accordance with paragraph 2."
    Article 13 is headed "Price control and cost accounting obligations" and provides that an NRA:
    "… may, in accordance with the provisions of Article 8, impose obligations relating to cost recovery and price controls … where a market analysis indicates that a lack of effective competition means that the operator concerned might sustain prices at an excessively high level …"
    Mr Roth submitted that a ruling by OFCOM as to the price which should be charged for interconnection (in order to resolve a dispute) was price control which Article 8(3) forbad in the absence of an SMP determination.
  197. We consider this reasoning to be wrong. Under the Access Directive the NRAs have at least two sorts of powers. The first are powers to take steps to ensure end-to-end connectivity; the second are powers to intervene where SMP has been found. A power to determine a dispute as to connection is capable of falling within both, so it is certainly capable of falling within the former. If it does, the Directive makes it plain that an SMP finding is not necessary. This is apparent from the terms of Article 5. It will be noted that Article 8(3) is without prejudice to Articles 5(1), (2) and (3). Article 5(1) provides:
  198. "Article 5
    Powers and responsibilities of the national regulatory authorities with regard to access and interconnection
    1. National regulatory authorities shall, acting in pursuit of the objectives set out in Article 8 of Directive 2002/21/EC (Framework Directive), encourage and where appropriate ensure, in accordance with the provisions of this Directive, adequate access and interconnection, and interoperability of services, exercising their responsibility in a way that promotes efficiency, sustainable competition, and gives the maximum benefit to end-users. In particular, without prejudice to measures that may be taken regarding undertakings with significant market power in accordance with Article 8, national regulatory authorities shall be able to impose:
    (a) to the extent that is necessary to ensure end-to-end connectivity, obligations on undertakings that control access to end-users, including in justified cases the obligation to interconnect their networks where this is not already the case;
    …"
    A power to resolve interconnection disputes is well within this wording, and there is no basis, as a matter of construction of Article 5, for separating out disputes as to price. Indeed, it would be illogical to do so. Pricing may be at the heart of a dispute; and some disputes about connection may have aspects which are not, by themselves, directly disputes about price, but may have pricing consequences so that one cannot decide one without the other. Determinations under this jurisdiction are not price control in the sense of Article 13. The two jurisdictions exist in parallel; the fact that Article 8(3) is without prejudice to the relevant parts of Article 5 demonstrates that they each have their separate existence.
  199. Mr Roth's arguments in this respect therefore fail. The possibility of dispute resolution by OFCOM in the future is therefore part of the overall picture which has to be taken into account in assessing whether BT has a real and effective bargaining position that is sufficient to counter the factors which would otherwise point in favour of H3G having SMP.
  200. We should record that it is not apparent that this erroneous perception of OFCOM's powers was one that obviously underlay the Decision. There is a reference to its powers in Chapter 4, which deals with the detrimental effects of SMP in the markets in question. The chapter returns briefly to the question of CBP and goes on to deal with a submission made by another operator (Orange) to the effect that OFCOM could determine price disputes under section 190 of the 2003 Act. The Decision then says:
  201. "4.11 … Ofcom needs to form a view as to what is the appropriate way of exercising all of its powers under the Act in the circumstances of each case.
    4.14 In this context, Ofcom notes it has the power to resolve the price increase dispute in question by determining that it will not prevent the increase until it has exercised its powers to set, inter alia, an SMP condition (see section 190(4) of the Act). Accordingly, Ofcom does not accept that it has made a material error of fact in rejecting dispute resolution as a constraint on the MNOs' ability to price excessively."
  202. OFCOM therefore appeared to take the view that it could decide to decline to adjudicate (as a matter of discretion, or even policy) in a dispute about prices until it had made an SMP determination; it would not adjudicate until then. At the hearing before us this was strengthened to a submission that it could not (i.e. it was not permitted to) adjudicate until it had made such a decision (the point we have dealt with above). Mr Green submitted that part of OFCOM's mistaken reasoning on the CBP point stemmed from a misunderstanding as to its powers in this respect, that misunderstanding being a belief, demonstrated by paragraph 4.14, that there was no power to intervene until an SMP determination was made. While it is clear what OFCOM's present position is on the point (i.e. that it has no power to determine price absent an SMP decision) it is not clear that that is what is described in paragraph 4.14. That paragraph claims a power to take that position as a policy decision, not as something forced on it by the legislation. There was no debate before us as to whether it had that particular power to limit its intervention, so we do not express a view on it (without in any way wishing to be seen to be encouraging it).
  203. THE INTERCONNECT AGREEMENT AND DISPUTE RESOLUTION

  204. The last factor which is relied on by H3G as a curb on what would or might otherwise be dominance (or SMP) is the power and position of OFCOM under clause 13 of the Interconnect Agreement. Clause 13 is set out above. It allows for either party to seek to vary prices, so it applies to the prices charged to BT for terminating calls on its network. H3G can serve a notice, and if there is no agreement between the parties as to a new price then the matter can be referred to the Director General for a determination of the dispute. It is obviously the case that the Director General no longer exists because Oftel does not exist. The Communications Act 2003 contains no provisions which provide for OFCOM to be substituted for Oftel or the Director General where either of the latter two appeared in contracts, so in theory the arbiter provided for under the contract does not exist. However, none of the parties before us submitted that that robbed the clause of any effect, and in their differing ways they all accepted that it would be appropriate to treat the references to the Director General as if they were references to OFCOM. There may be various routes through which that conclusion can be reached, but since the end result is not disputed we will not seek to travel down any of them and merely start from the common position. To avoid confusion we will consider this point as if OFCOM were explicitly identified as the arbiter under clause 13.
  205. We specifically invited submissions on the extent (if any) to which the presence of this clause, which would allow BT to decline to initiate any price increase and have a price variation decided by a third party, would mean that there was no SMP, whether by means of providing some element going to CBP or otherwise. On a very simple view, H3G's power over price might be thought to be less because BT could always contest a price and take the matter for adjudication to someone who would obviously not set an excessive price. It seemed to us that this point needed to be argued, and (judging by their skeleton arguments) none of the parties were going to do so.
  206. Mr Green took the position that the presence of the dispute resolution procedure under clause 13 was, at the end of the day, part of, or akin to, the regulatory presence which meant that H3G could not set an excessive price and therefore had no SMP. Mr Roth's position on this clause can be shortly stated – he said one cannot contract out of a dominant position if one is otherwise in one, and he relied on authority in support of that.
  207. We agree with OFCOM that the mechanism of clause 13 does not affect the conclusions to which OFCOM might otherwise properly have come about any SMP possessed by H3G (or CBP possessed by BT). The case law does not demonstrate that this point has been considered before in the context of a contractual arbitration clause, but the effect of contractual or other restraints on pricing has. In TotalFina/Elf (Case COMP/M.1628, [2001] OJ L143/1) the Commission considered whether a contractual agreement for third party access to the market meant that there was no dominant position in a merger context. It considered that it would not, saying (at paragraph 235):
  208. "It is the creation in itself of [a dominant position] that the Merger Regulation is intended to prevent. The application of the Regulation is not affected by the argument developed by the notifying party that the clauses concerned may limit the capacity of the new entity to abuse its dominant position."
    Analytically the reason for that is probably twofold.

    a) First, the contractual position goes to the question of abuse, not whether there is SMP in the first place. This is the distinction drawn by this Tribunal in Napp Pharmaceutical Holdings Ltd and Subsidiaries v The Director General of Fair Trading [2002] CompAR 13, where a price control mechanism imposed by the NHS was held not to go to dominance for that reason – see paragraph 165. We consider the restraint in that case was analogous to the clause 13 mechanism for these purposes.
    b) The second answer lies in identifying just what the clause 13 mechanism is. It is not actually a full third party arbitral mechanism of the kind one sees in, for example, a rent review clause. The arbiter in clause 13 is the regulator. The regulator's powers are conferred and constrained by statute, and while Ofcom's are extensive they do not include the power to be a third party arbitrator. In truth clause 13 does not invoke that latter sort of status. The sort of dispute that clause 13 contemplates is a form of interconnection dispute, which OFCOM would resolve as regulator, not as a third party dispute resolver. Its intervention would therefore be as regulator, and would be a form of regulation. It therefore falls to be disregarded, as a matter of principle, just as OFCOM's general presence as a regulator with a potential effect on the conduct of the putatively regulated person falls to be disregarded, for the reasons given above. This is the same point that we have considered and dealt with above. Accordingly we do not consider that the Clause 13 mechanism for dispute resolution has any material effect on the question of whether H3G had or has SMP.
  209. The Decision therefore errs in not considering the true effect and extent of the end-to-end connectivity obligation. It does not necessarily give rise to a situation in which H3G holds all the cards in any negotiation. We cannot decide its real effect (and we were not invited to) because we do not have the information, and it is quite conceivable that a fuller consideration would lead to the same result, but we can say that the situation requires further investigation and consideration beyond that conducted by OFCOM as reflected in its Decision. We do not think that this would be pointless since there is material which is pertinent to this inquiry. For example, the following is pertinent, and may be relevant (but not conclusive).
  210. CONCLUSIONS ON CBP

  211. H3G provided some evidence to OFCOM showing how the latter part of the negotiation went in 2001 and 2002. OFCOM has come to the conclusion that that evidence went only to the negotiations at that time and "did not provide a sufficient indication of how future negotiations with BT would run, given the change in H3G's circumstances [i.e. it now had a connection as opposed to its negotiating one]". If that means that that evidence (and other evidence of the negotiation) did not provide a conclusive demonstration of how future negotiations will go, then that is plainly right. That, however, is not the point. The point is whether, and to what extent, what happened in those negotiations provides at least some useful material in assessing how future negotiations would go. For our part we do not think that it can be dismissed entirely. There were various elements which might (and we stress "might", because we are not determining the point) be taken still to point to the fact that BT had a real negotiating position. Thus:
  212. i) It will be remembered that BT did not accept the first price band proposed by H3G. It invited costs justification, and pointed out that matters had to be considered by the relevant Board. That might be said to say something about its negotiating power.
    ii) H3G put in evidence before us a document which we do not think was before OFCOM, namely an internal H3G memorandum from about November 2001 in which two employees debate a strategy for determining the termination charges that should be sought form BT. It recognises:
    "H3G's interconnect charge must be set at a level to maximise this revenue opportunity without being unacceptable to Oftel, or to our interconnect partners. A high interconnect charge will be reflected by other networks charging their customers a high retail price for calling to H3G."
    Later it says:
    "If BT reject H3G's proposed interconnect charges, and no agreement can be reached commercially, this will result in BT requesting Oftel to intervene and determine what charges should apply …"
    Those two extracts do not support any assertion that H3G had a strong hand in determining the price, and there is nothing else in the memorandum which obviously does either. While it is true that this memorandum refers to delay it might be said to provide material (albeit from H3G's side and not BT's) which would support the idea that the negotiation with BT would be a genuine one and not one in which BT held few, if any, cards. If that was true, then it may be that the same sort of factors operate now (or would have operated at the future period to which the Decision looked).
  213. There is therefore material from the prior negotiation which might be said to have some continuing relevance to what the position would be in any price negotiation between H3G and BT. It is therefore overly-simplistic and wrong to say that the end-to-end connectivity obligation determines the question of CBP. As we have said, we reach no decision on these additional matters, but we consider that the Decision, and the process underlying it, does not seem to have addressed them. Any proper consideration of CBP ought to have done so, and that failure makes the Decision flawed in this respect. It may well be that the circumstances would require a fuller investigation of BT's position (it is arguably a monopsonist), the possibility of joint dominance, and such things as its relationship with H3G and its attitude and propensity in relation to the protection of the interests of its own customers (bearing in mind the words "and ultimately consumers" in Framework Directive Art 14(2)) when considering the level of termination charges which it was inevitably going to pass on, but we say no more about it because the overall position is not (in the circumstances of this case) one which we are called on to investigate. We do not have the material to do it, and no-one has suggested we should.
  214. For the sake of completeness we observe that we have borne in mind the fact that under this head regulation is brought into account in determining CBP, whereas regulation of H3G is left out of account in looking at its side of the SMP assessment. There is nothing inconsistent in this approach. We have identified the illogicality in allowing a presumption of regulation of a putatively regulated body to operate to determine whether SMP exists. That does not apply to a consideration of CBP where one has to consider the question of a counterparty. In assessing the position of that counterparty it would be illogical not to look at the effect of regulation (and no-one suggested we should not), so OFCOM were quite correct in doing so in this case. However, as we have observed, the full factual position in this respect must be looked at – one most look at how far the regulation will actually operate in any deemed negotiations. It is in failing to do so that OFCOM erred in its Decision.
  215. There were other criticisms of the Decision in relation to CBP but in the light of the conclusion we have reached above we do not think it necessary to deal with them.
  216. THE ECAP DECISION

  217. We have referred above to the Irish ECAP decision. This decision was delivered after the hearings in this case had been finished but before we delivered judgment. We invited written submissions from the parties on that decision. We have considered, and are grateful for, those submissions. That decision was an appeal by an H3G group company from the Irish telecommunications regulator which has parallels with the Decision in this case. Each side sought to draw parallels with, and to distinguish, various aspects of that case. We have considered their submissions carefully, but at the end of the day that was a decision on its own facts and we have to make a decision on the facts of this case. In the circumstances, apart from the references made to it above, and thorough though it is, we do not gain any further assistance from it.
  218. VIII CONCLUSION

  219. We therefore conclude and find that OFCOM erred in its determination as to the existence of significant market power because it did not carry out a full assessment of the extent to which BT had countervailing buyer power. Our powers and duties include the following under section 195 of the 2003 Act:
  220. "(3) The Tribunal's decision must include a decision as to what (if any) is the appropriate action for the decision-maker to take in relation to the subject-matter of the decision under appeal."
    At present we consider that the appropriate direction would be to require OFCOM to reconsider its determination of SMP taking into account the extent to which countervailing buyer power exists in BT, and considering such other matters as are relevant as at the time of its reconsideration. However, we would wish to give the parties an opportunity to address us on the form of relief to be granted should they wish to do so.

    The Hon Mr Justice Mann Adam Scott Paul Stoneman

    Charles Dhanowa 29 November 2005

    Registrar

Note 1   OJ [2002] L108/33.    [Back]

Note 2   OJ [2002] L108/7.    [Back]

Note 3   OJ [2002] L108/21.    [Back]

Note 4   OJ [2002] L108/51.    [Back]

Note 5   OJ [2002] L201/37.    [Back]

Note 6   The Privacy Directive, which establishes users’ rights with regard to the privacy of their communications, was adopted slightly later than the other Directives and was implemented by The Privacy and Electronic Communications (EC Directive) Regulations 2003, SI 2003/2426, which came into force on 11 December 2003.The Privacy Directive does not need to be considered for the purposes of the current appeal.    [Back]

Note 7   R (on the application of T-Mobile (UK) Ltd and others) v Competition Commission and others [2003] EWHC 1555 (Admin).    [Back]

Note 8   OJ [2003] L114/45.    [Back]

Note 9   As required by section 79 of the 2003 Act.    [Back]

Note 10   Oftel’s market review guidelines: criteria for the assessment of significant market power.    [Back]

Note 11   These four criteria are taken from those listed in the Commission Guidelines and the Oftel Guidelines.    [Back]

Note 12   Though we note that H3Gs’ submission in this respect was contrary to the stance that it adopted in its response to the May Consultation, in which it suggested that “no regulation” might be an appropriate end point even if SMP were found: “1.6 Hutchison 3G therefore urges the director to reconsider his designation of SMP or to consider lesser remedies for Hutchison 3G (e.g. no regulation at all or imposition of a transparency obligation). To impose ex-ante regulation upon Hutchison 3G, as the Director proposes, would place the Director in breach of his duties and obligations to provide for appropriate regulation of telecommunications networks.”     [Back]

Note 13   During the hearing Mr Green referred to excessive prices as prices that were above the Long Run Incremental Costs (LRIC) of 3G technology, such costs including the expenditure on the acquisition of the 3G licence. An alternative definition is prices that exceed the prices that would exist in the market in the absence of SMP (in the hands of any party).     [Back]

Note 14   Crehan v Inntrepreneur Pub Co [2004] EWCA Civ 637, 148 Sol Jo LB 662.    [Back]

Note 15   Case DE/2005/0239, Decision dated 28th September 2005    [Back]


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/CAT/2005/39.html