Hutchison 3G UK Limited v Office of Communications [2008] CAT 11 (20 May 2008)

Neutral citation [2008] CAT 11

IN THE COMPETITION 1083/3/3/07

APPEAL TRIBUNAL

Victoria House

Bloomsbury Place 20 May 2008

London WC1A 2EB

Before:

VIVIEN ROSE

(Chairman)

PROFESSOR ANDREW BAIN OBE

ADAM SCOTT TD

Sitting as a Tribunal in England and Wales

BETWEEN:

HUTCHISON 3G UK LIMITED

Appellant

-v-OFFICE OF COMMUNICATIONS

Respondent

supported by

O2 (UK) LIMITED

T-MOBILE (UK) LIMITED

VODAFONE LIMITED

ORANGE PERSONAL COMMUNICATIONS SERVICES LIMITED

BRITISH TELECOMMUNICATIONS PLC

Interveners

Heard at Victoria House from 24 January to 5 February 2008

JUDGMENT ON NON PRICE CONTROL MATTERS (Non-Confidential Version)

APPEARANCES

Miss Dinah Rose QC and Mr Brian Kennelly (instructed by Baker & McKenzie) appeared on behalf of Hutchison 3G (UK) Limited.

Mr Peter Roth QC, Mr Josh Holmes and Mr Ben Lask (instructed by the Office of Communications) appeared for the Respondent.

Mr David Anderson QC and Miss Sarah Lee (instructed by BT Legal) appeared on behalf of British Telecommunications plc.

Miss Kelyn Bacon (instructed by SJ Berwin) appeared on behalf of O2 (UK) Limited.

Miss Marie Demetriou (instructed by Field Fisher Waterhouse) appeared on behalf of Orange Personal Communications Services Limited.

Mr Jon Turner QC and Mr Meredith Pickford (instructed by Miss Robyn Durie, Regulatory Counsel, T-Mobile) appeared on behalf of T-Mobile (UK) Limited.

Miss Elizabeth McKnight (Partner, of Herbert Smith) appeared on behalf of Vodafone Limited.

Note: Excisions in this judgment (marked “[… ][X]”) relate to commercially confidential information: Schedule 4, paragraph 1 to the Enterprise Act 2002.

I. INTRODUCTION

1. This appeal is brought by the Appellant (“H3G”) under section 192 of the Communications Act 2003 (“the 2003 Act”). H3G is a mobile network operator which challenges certain aspects of two decisions adopted by the Respondent (“OFCOM”) on 27 March 2007.

2. H3G’s appeal concerns the prices that mobile network operators charge for mobile call termination (“MCT”). Mobile call termination is the process of connecting a voice call from the caller’s network to the recipient’s network. Consumers expect to be able to make calls from their fixed line or mobile phone to any other retail customer irrespective of the service provider (fixed or mobile) to which the receiving party subscribes. Network operators enter into contractual arrangements with each other for the provision of access to each other’s networks. Under those arrangements the terminating network operator makes a charge for each call terminated on its network, known as a mobile call termination charge.

3. The charge for mobile call termination is expressed in pence per minute or “ppm”. Usually the mobile network operators (“MNOs”) set different prices for terminating day-time, evening and weekend minutes. There are tens of billions of minutes terminated on the networks of the MNOs each year so that changes of a fraction of a penny in the rates make a difference of many millions of pounds in the income and expenditure of these companies.

4. In the United Kingdom there are two main forms of mobile network commonly known as ‘2G’ and ‘3G’. Second Generation or 2G networks were originally designed to support mobile voice calls and text messaging services using a radio transmission technology known as Global System for Mobile Communications (“GSM”). 2G networks were subsequently enhanced to support low speed mobile data services such as mobile internet access and picture and multimedia messaging services. Third Generation or 3G networks are aimed at supporting higher speed call services (for video telephony) and higher speed mobile data

services for faster internet access and multimedia messaging. The radio technology for 3G is different from that used within 2G but many of the services delivered over the technologies are similar. A key difference is that 2G networks cannot offer the higher speed data services now possible on 3G networks.

5. In 2000 the Government held an auction for licences to operate 3G spectrum. At that time there were four main MNOs in the mobile market using 2G technology: the firms now called O2 (UK) Limited (“O2”), T-Mobile (UK) Limited (“T-Mobile”), Vodafone Limited (“Vodafone”) and Orange Personal Communications Services Limited (“Orange”). To ensure that there was sufficient competition to encourage the roll out and adoption of 3G technology, the Government designed the auction so that one licence was reserved for a new entrant. The new entrant who acquired the fifth licence was H3G. There are three main spectrum bands used by the five MNOs each of whom has a separate block of spectrum within these bands. The sums paid by the MNOs for these licences were considerable. The MNOs differed in the amount of spectrum they were allocated but they all paid more than £4 billion for their allocation, with one of them paying almost £6 billion.

6. The four MNOs operating in the UK who used to operate only 2G networks now operate both 2G and 3G networks. They are all substantial companies belonging to groups which operate across Europe. They have intervened in these proceedings because they are directly affected by any challenge to the rate which OFCOM has determined they can charge for mobile call termination on their networks. These four are commonly referred to collectively as the “2G/3G MNOs”. H3G, which entered the market as the fifth licensee of the 3G spectrum, operates a 3G network only but it has always had roaming arrangements in place so that in areas of the country which are not covered by H3G’s 3G network, its customers can interconnect using a 2G network.

7. The dates on which the MNOs began to offer their 2G and 3G services are as follows:


Operator	2G service	3G service
Vodafone	December 1991	November 2004¹
Orange	April 1994	March 2004
O2	July 1994	February 2005²
T-Mobile	September 1993	October 2005³
H3G	n/a	March 2003

8. It is important to bear in mind that so far as the retail market for mobile phone services is concerned, the MNOs are not the only service providers. There are also the mobile virtual network operators (“MVNOs”) who compete with the five MNOs for retail mobile phone customers. The MVNOs are companies which do not operate their own mobile network infrastructure but contract with one of the MNOs to use its network. From the perspective of the customer they appear very much like MNOs, billing customers directly for the mobile services provided to them. An example of such an MVNO is Virgin Mobile, which offers a range of retail tariffs and which bills its customers for mobile services but provides those services using the T-Mobile network. There are thus many more companies participating in the retail mobile market than there are networks.

9. In addition to the MNOs and MVNOs there are the fixed network operators or FNOs, the largest of which is British Telecommunications plc (“BT”). Fixed network subscribers also need to be able to interconnect with subscribers to mobile networks and BT has in general paid the same mobile call termination charges to the MNOs as the MNOs pay to each other. More than 15 billion fixed-to-mobile call minutes are originated every year of which BT’s share is about 50 per cent. The FNOs also charge each other, and the MNOs, for terminating calls on their fixed networks. The charges that BT can impose for termination are fixed by OFCOM at a level such that, we were told, the average charge is about 0.4 ppm.

¹ Vodafone carried out commercial trials of its 3G voice services before the November 2004 launch date and launched its 3G data card services in April 2004.

² O2 began to offer 3G data services for business customers in September 2004 and offered 3G voice and data services for post pay customers from February 2005.

³ T-Mobile began using its 3G spectrum in 2004 but the first 3G specific service was launched in October 2005.

10. The United Kingdom operates a “calling party pays” (“CPP”) system which means that the entire cost of the call is paid for by the calling party. Mobile call termination charges are paid in the first instance by the originating network operator to the terminating operator and thus form an element of the costs that determine the charge collected by the originating operator from its retail subscriber customer. CPP is contrasted with RPP (receiving party pays) although the term RPP has two senses. It can refer to retail RPP where the retail recipient of the call pays a charge for each call received but it can also refer to an arrangement whereby the cost of terminating the call simply lies with the receiving network, leaving that network to cover that cost in whatever way it considers appropriate other than by a direct charge to the originating network.

11. BT is important not only as the major FNO in the UK but also because it provides transit services to other fixed and mobile operators. BT directly interconnects with approximately 180 communications providers in the UK and is under a regulatory obligation as regards certain parts of its transit business – for example, charges it can impose for transit are regulated. Many operators therefore rely on BT to terminate their calls on other networks under BT’s interconnection agreement with that network rather than having to negotiate their own agreement with each of the 180 communications providers. In such a case BT pays the MCT charge imposed by the terminating network and charges the transiting operator that MCT charge plus the transit fee and an additional circuit charge for conveyance. The terminating MNOs are not able to identify in respect of calls coming from BT whether the call comes from a BT subscriber or whether the call originates with a subscriber of another operator who is using BT’s transit services to route the call.

12. At the time when H3G was negotiating its first interconnection agreement with BT prior to the launch of H3G’s service in the United Kingdom, BT charged its subscribers a range of retail price bands for calls from fixed-to-mobile telephones with each price band relating to calls to a particular mobile operator. There were distinct retail charges per minute for calls made during a weekday, during the evening and night, and during a weekend. The retail charges reflected a margin over the mobile call termination charge paid to the relevant mobile operator.

Since then, BT has increased the variety of fixed-to-mobile bands but has consolidated its retail charges so that calls to each of the four 2G/3G MNOs are now charged at rates that do not discriminate between the four of them. BT has, however, kept a different retail rates band for calls from its network to H3G.

II. REGULATORY BACKGROUND (i) The EU Legislation

13. Regulation of electronic communications across Europe is now based on the European Common Regulatory Framework (“CRF”) which was promulgated in April 2002 and had to be implemented by the Member States by July 2003. This superseded earlier EU regulatory instruments. The CRF comprises (amongst other instruments) Directive 2002/21/EC on the common regulatory framework for electronic communications networks and services [2002] OJ L108/33 (“the Framework Directive”) and four other directives referred to in the Framework Directive as the Specific Directives. The most relevant Specific Directive as regards this appeal is Directive 2002/19/EC on access to, and interconnection of, electronic communications networks and associated facilities [2002] OJ L108/7 (“the Access Directive”).

14. Under the Framework Directive, the Member States must designate a national regulatory authority (“NRA”) to carry out the regulatory tasks set out in the CRF. Such NRAs must be independent of the government of the Member State and must exercise their powers impartially and transparently. Article 8 of the Framework Directive sets out the policy objectives and regulatory principles of which the NRAs are required to take the utmost account in carrying out their tasks under the Framework Directive and the Specific Directives. These objectives include the following in article 8(2) -

“The national regulatory authorities shall promote competition in the provision of electronic communications networks, electronic communications services and associated facilities and services by inter alia:

(a) ensuring that users, including disabled users, derive maximum benefit in terms of choice, price, and quality;

(b) ensuring that there is no distortion or restriction of competition in the electronic communications sector;

(d) encouraging efficient use and ensuring the effective management of radio frequencies and numbering resources.”

15. Among the specific tasks conferred on the NRAs is an obligation to carry out an analysis of relevant markets in the telecoms sector. In identifying such markets, the NRA is required to take the utmost account of recommendations and guidelines published by the EC Commission as to what product and service markets should be analysed. Once the NRA has identified the relevant markets in its own territory it must determine whether each of those markets is “effectively competitive”. Where an NRA determines that a relevant market is not “effectively competitive” it must identify undertakings with “significant market power” (“SMP”) on that market and must then impose on such undertakings appropriate specific regulatory obligations or maintain or amend such obligations where they already exist. The Access Directive also provides for the NRA to conduct market analysis into the markets identified in the Commission’s Recommendation. Article 7(3) of the Access Directive provides that Members States shall ensure that, as soon as possible after the entry into force of the Directive, and periodically thereafter, NRAs undertake a market analysis to determine whether to maintain, amend or withdraw any obligations that are in place at the date that the CRF comes into effect. Further article 8 of the Access Directive provides that where an operator is designated as having SMP on a specific market as a result of the NRA’s market analysis, the NRA must impose obligations of the kind set out in articles 9 to 13 of the Access Directive. Those obligations, commonly referred to as the “SMP conditions”, include, in article 13, the setting of price controls:

“13. Price control and cost accounting obligations

1. A national regulatory authority may, in accordance with the provisions of Article 8, impose obligations relating to cost recovery and price controls, including obligations for cost orientation of prices and obligations concerning cost accounting systems, for the provision of specific types of interconnection and/or access, in situations where a market analysis indicates that a lack of effective competition means that the operator concerned might sustain prices at an

excessively high level, or apply a price squeeze, to the detriment of end-users. National regulatory authorities shall take into account the investment made by the operator and allow him a reasonable rate of return on adequate capital employed, taking into account the risks involved.

2. National regulatory authorities shall ensure that any cost recovery mechanism or pricing methodology that is mandated serves to promote efficiency and sustainable competition and maximise consumer benefits. In this regard national regulatory authorities may also take account of prices available in comparable competitive markets.”

16. In February 2003 the EC Commission published, pursuant to its obligation under article 15 of the Framework Directive, its Recommendation on relevant product and service markets within the electronic communications sector susceptible to ex ante regulation (“the Recommendation on Market Definition”). Market 16 in the Annex to the Recommendation defined “voice call termination on individual mobile networks” as one of the markets which the NRA ought to analyse to see if it is effectively competitive within that NRA’s territory.

17. The United Kingdom’s NRA is OFCOM. The decisions with which this appeal is concerned relate to OFCOM’s analysis of Market 16 as defined in the 2003 Recommendation. That Recommendation on Market Definition has since been replaced by an updated Recommendation published in December 2007. The new Recommendation still identifies voice call termination on individual mobile networks as a market which NRAs should analyse but this is now Market 7 rather than Market 16.

18. Where an NRA intends to impose ex ante regulation on an undertaking with significant market power in a case which would affect trade between Member States, it must consult with the EC Commission and the NRAs in the other Member States. This obligation is imposed by article 7 of the Framework Directive and is usually referred to as “Article 7 consultation”. The Commission and the other NRAs are entitled to comment on the draft measure and the reasons for it within a period of one month and the NRA must take the “utmost account” of such comments. Where the measure proposed by the NRA is a measure which decides whether or not an undertaking has SMP in a relevant market in a case which affects trade between Member States, the Commission may indicate to the NRA that it has serious doubts about whether the measure is compatible

with Community law and the objectives under the CRF. Ultimately the Commission may issue a decision requiring the NRA to withdraw such a draft measure. This power is commonly referred to as the Commission’s veto. The Commission’s power of veto extends only to certain proposed decisions by the NRA including the decision to designate an undertaking as having or not having SMP. At present the Commission does not have a veto over the kinds of ex ante regulation that the NRA can impose on an undertaking with SMP, although the Commission may comment to the NRA on the suitability of the proposed regulatory measures on which it is consulted.

(ii) Implementation of the CRF in the United Kingdom

19. The CRF was implemented in the United Kingdom by the 2003 Act. Section 45 of the 2003 Act empowers OFCOM to set conditions of various kinds, including SMP conditions. Section 47 provides that OFCOM must not set a condition under section 45 unless the condition is non-discriminatory, proportionate and transparent. We consider the test which OFCOM must apply before setting a price control condition in more detail later in this judgment.

20. Section 79 of the 2003 Act sets out what OFCOM must do when carrying out its market analysis and before it makes a determination that an undertaking has SMP, requiring it, for example to take due account of all applicable guidelines and recommendations issued by the European Commission. Section 87 provides that where OFCOM has made a determination that a person has SMP it shall set such SMP conditions authorised by this section as they consider it appropriate to apply to that person. Section 87 goes on to list the kinds of SMP conditions that OFCOM can impose, including price controls which are described in subsection (9) –

“87 (9) The SMP conditions authorised by this section also include (subject to

section 88) conditions imposing on the dominant provider-(a) such price controls as OFCOM may direct in relation to matters connected with the provision of network access to the relevant network, or with the availability of the relevant facilities;

(b) such rules as they may make in relation to those matters about the recovery of costs and cost orientation;

(d) obligations to adjust prices in accordance with such directions given by OFCOM as they may consider appropriate.

(10) The SMP conditions authorised by subsection (9) include conditions requiring the application of presumptions in the fixing and determination of costs and charges for the purposes of the price controls, rules and obligations imposed by virtue of that subsection.”

21. Section 88 of the 2003 Act then sets important preconditions which must be satisfied before OFCOM sets a price control authorised by section 87(9):

“88 Conditions about network access pricing etc.

(1) OFCOM are not to set an SMP condition falling within section 87(9) except where-(a) it appears to them from the market analysis carried out for the

purpose of setting that condition that there is a relevant risk of adverse effects arising from price distortion; and

(b) it also appears to them that the setting of the condition is appropriate for the purposes of-(i) promoting efficiency; (ii) promoting sustainable competition; and

(iii) conferring the greatest possible benefits on the end-users of public electronic communications services.

(2) In setting an SMP condition falling within section 87(9) OFCOM must take account of the extent of the investment in the matters to which the condition relates of the person to whom it is to apply.

(3) For the purposes of this section there is a relevant risk of adverse affects arising from price distortion if the dominant provider might-(a) so fix and maintain some or all of his prices at an excessively high

level, or

(b) so impose a price squeeze,

as to have adverse consequences for end-users of public electronic communications services.

(4) In considering the matters mentioned in subsection (1)(b) OFCOM may-

(a) have regard to the prices at which services are available in comparable competitive markets;

(b) determine what they consider to represent efficiency by using such cost accounting methods as they think fit.

…”

(iii) Regulation of the market for mobile call termination charges in the United Kingdom

22. The CRF and the 2003 Act superseded the pre-existing regulatory regime in the telecoms sector which had been implemented in the United Kingdom by the Telecommunications Act 1984. In 1999 the former Monopolies and Mergers Commission concluded that the mobile call termination charges of two of the MNOs might be expected to operate against the public interest and recommended the imposition of price controls on termination charges. The former Director General of Telecommunications amended the two MNOs’ licences to include charge controls. In 2003 charge controls were imposed in respect of the mobile call termination charges of all four 2G/3G MNOs. Following the coming into force of the 2003 Act and the publication by the Commission of its Recommendation on Market Definition, OFCOM conducted its analysis of Market 16. OFCOM’s determinations were set out in its statement dated 1 June 2004 on Wholesale Mobile Call Termination (“the 2004 Statement”). Broadly, the 2004 Statement concluded -

(a) that there were separate relevant services markets for mobile call termination on each of the MNOs’ networks, regardless of whether termination took place on the 2G or 3G network;

(b) that all MNOs had 100 per cent share of the market for termination of calls on their own network and there were absolute barriers to entry which precluded the possibility of any other undertaking providing mobile call termination services on those markets;

(c) that a price control should be imposed on the price of mobile call termination charges of the 2G/3G MNOs using the 2G spectrum;

(d) that there should be no price control in respect of termination using 3G spectrum – at the time this affected only H3G since the other MNOs had not yet launched their 3G services;

(e) the regulatory obligation imposed on H3G should not include a price control in respect of either termination on its 3G spectrum or of termination via its roaming arrangements on 2G spectrum.

23. The price control in the 2004 Statement was set to apply until 31 March 2006. H3G appealed against the 2004 Statement on grounds that OFCOM had erred in finding that it had SMP. The Tribunal’s judgment delivered in November 2005 in Hutchison 3G (UK) Limited v Office of Communications [2005] CAT 39 (“H3G (1)”) found that OFCOM had erred in its analysis of market power and remitted the case back to OFCOM. We will need to consider in some detail exactly what was decided by the Tribunal in that judgment. The Tribunal in H3G (1) made an order requiring OFCOM to reconsider whether H3G has SMP taking into account the extent to which BT has countervailing buyer power and any other matters relevant at the time of OFCOM’s reconsideration. On 13 September 2006, OFCOM published a consultation paper setting out its proposed reassessment of H3G’s market power and on 27 March 2007, OFCOM published its “Assessment of whether H3G holds a position of SMP in the market for wholesale mobile voice call termination on its network Statement” (“the Reassessment Statement”) confirming its earlier conclusion that H3G had SMP during the period covered by the 2004 Statement.

24. Meanwhile, on 7 June 2005 OFCOM published a consultation document proposing a one year extension of the price control set in the 2004 Statement for a further year, until 31 March 2007. On the same day, in parallel with that proposal, OFCOM published a Preliminary Consultation to initiate consideration of the issues which would need to be addressed during the next review of Market 16 for the period after the price control set in the 2004 Statement expired. Towards the end of 2005 OFCOM issued a statement extending the price control in the 2004 Statement for a further year up to 31 March 2007, making it clear that the extension was not intended to limit in any way the range of conclusions

that might be drawn from the consultation that had commenced for the review of the market for the period thereafter.

25. On 30 March 2006 OFCOM published a more detailed consultation document Wholesale mobile voice call termination (“the March 2006 Consultation”) in respect of Market 16 after March 2007. That consultation set out OFCOM’s initial view that there are separate markets for mobile call termination supplied by each of the five MNOs, and that the prima facie evidence indicated that each of these mobile operators had SMP in the market in which it supplies call termination. The March 2006 Consultation analysed the detriments which could arise from the exercise of SMP in these markets and explored a number of regulatory options for addressing those detriments. On 13 September 2006, having considered responses to the March 2006 Consultation, OFCOM published a third consultation Mobile call termination – Proposals for consultation (the “September 2006 Consultation”). In the September 2006 Consultation, OFCOM again set out its view that there are separate markets for mobile call termination supplied by each of the five MNOs and each of these mobile operators has SMP in the market in which they supply MCT. The September 2006 consultation also described the detriments which are likely to arise from the exercise of that SMP, and the remedies which OFCOM proposed should be imposed. These remedies included price controls to apply to each of the five MNOs for four years to 31 March 2011, obligations to meet reasonable demand for call termination on fair and reasonable terms, prohibitions of undue discrimination and obligations concerning transparency of charges and contract terms.

26. Having considered responses to the September 2006 Consultation, OFCOM published the final statement on 27 March 2007 (“the 2007 Statement”) setting out its conclusions that -

• There are separate markets for the provision of wholesale mobile voice call termination in the UK to other Communications Providers by each of Vodafone, O2, Orange, T-Mobile and H3G;

• Each of the five MNOs has SMP in the market for termination of voice calls on its network(s);

• Charge controls should be imposed on the supply of MCT by each of the five MNOs, and those controls should apply to all voice calls whether terminated on 2G or 3G networks;

• The charge control should apply for 4 years from 1 April 2007;

• Average charges of Vodafone, O2, Orange and T-Mobile should be reduced to 5.1 ppm (2006/7 prices) by the final year of the charge control period (1 April 2010 to 31 March 2011). The reduction should be implemented in 4 equal (percentage) steps across the four years starting from the 2G capped rate which applied during 2006/07 pursuant to the application of the price control in the extended 2004 Statement;

• Average charges of H3G should be reduced to 5.9 ppm (2006/7 prices) by the final year of the charge control (1 April 2010 to 31 March 2011). This level reflected cost differences between H3G and the 2G/3G MNOs. The change was to be implemented by an initial reduction to 8.5ppm (2006/7 prices) followed by three reductions each of equal (percentage) changes across the next three years (i.e. from April 2008 to March 2011);

• Further conditions were imposed requiring provision of voice call termination on fair and reasonable terms and conditions (including contract terms), prohibiting undue discrimination, and requiring charge transparency.

(iv) OFCOM’s dispute resolution function

27. The issues raised in this appeal also turn in part on the proper interpretation of a different set of powers that the CRF requires Member State to confer on the NRA, namely the power to resolve disputes between undertakings. There are two separate powers. The first is in article 20 of the Framework Directive which provides -

“1. In the event of a dispute arising in connection with obligations arising under this Directive or the Specific Directives between undertakings providing electronic communications networks or services in a Member State, the national regulatory authority concerned shall, at the request of either party, and without prejudice to the provisions of paragraph 2, issue a binding decision to resolve the dispute in the shortest possible time frame and in any case within four months except in exceptional circumstances. The Member State concerned shall require that all parties cooperate fully with the national regulatory authority.

2. Member States may make provision for national regulatory authorities to decline to resolve a dispute through binding decision where other mechanisms, including mediation, exist and would better contribute to resolution of the dispute in a timely manner in accordance with the provisions of Article 8. …

3. In resolving a dispute, the national regulatory authority shall take decisions aimed at achieving the objectives set out in Article 8. Any obligations imposed on an undertaking by the national regulatory authority in resolving a dispute shall respect the provisions of this Directive or the Specific Directives”.

28. The second power is in article 5(4) of the Access Directive:

“With regard to access and interconnection, Member States shall ensure that the national regulatory authority is empowered to intervene at its own initiative where justified or, in the absence of agreement between undertakings, at the request of either of the parties involved, in order to secure the policy objectives of Article 8 of [the Framework Directive], in accordance with the provisions of this Directive and the procedures referred to in Articles 6 and 7, 20 and 21 of [the Framework Directive]”.

29. Article 20 covers all disputes arising in connection with obligations under the Directives without distinguishing between disputes relating to the provision of network access and other disputes. Article 5(4) covers disputes “with regard to access and interconnection” whether they arise in relation to a regulatory obligation or not. There is therefore an overlap between them in that a dispute which is “with regard to access and interconnection” and which also arises in connection with a regulatory obligation will fall within both provisions.

30. These powers so far as they relate to dispute resolution were implemented in the United Kingdom by section 185 of the 2003 Act. OFCOM’s task in resolving a dispute is set out in section 188 of the 2003 Act; OFCOM must consider the dispute and make a determination resolving it, using whatever procedure OFCOM considers appropriate. OFCOM must make its determination as soon as practicable and, except in exceptional circumstances, within four months of the date on which it accepts jurisdiction over the dispute.

31. Sections 185 to 190 implement the dispute resolution powers in both article 20 of the Framework Directive and article 5(4) of the Access Directive. The power to intervene on its own initiative which is also required by article 5(4) is implemented by section 105 of the 2003 Act. Section 105 applies where it appears to OFCOM that a “network access question” has arisen and needs to be determined and where it considers that, for the purpose of determining that question, it would be appropriate for OFCOM to exercise certain of its powers to set, modify or revoke conditions imposed on communications providers. A “network access question” is defined as “a question relating to network access or the terms or conditions on which it is or may be provided in a particular case”.

III. PROCEDURAL BACKGROUND

(i) The non price control matters in H3G’s appeal

32. H3G’s appeal covers three main areas:

(a) the findings of SMP in each of the Reassessment Statement and the 2007 Statement (see paragraphs [44] to [141] below);

(b) OFCOM’s decision to impose a price control in the form imposed by the 2007 Statement (see paragraphs [142] to [298] below);

33. Although the appeal therefore challenges two separate decisions by OFCOM, it has been treated as a single appeal. BT has also lodged an appeal against the 2007 Statement, arguing that price control imposed in the 2007 Statement sets rates which are too high. These two appeals are the first occasion on which the Tribunal has followed the procedure set out in section 193 of the 2003 Act. That procedure requires the Tribunal to identify whether an appeal raises any “specified price control matters” as defined. The price control matters to which the procedure applies have been specified in rule 3 of the Competition Appeal Tribunal (Amendment and Communications Act Appeals) Rules 2004 (SI 2004

No. 2068) (“the 2004 Rules”). If an appeal does raise specified price control matters, then those matters are to be referred by the Tribunal to the Competition Commission for its determination. Matters raised by the appeal which are not price control matters are to be decided by the Tribunal. Once the Competition Commission has notified the Tribunal of its determination of the price control matters referred to it, the Tribunal must decide the appeal on the merits and, in relation to the price control matters, must decide those matters in accordance with the determination of the Competition Commission, unless the Tribunal decides, applying the principles applicable on an application for judicial review, that the Competition Commission’s determination would fall to be set aside on such an application.

34. The issues as to whether H3G has SMP in the relevant market for the purposes of the Reassessment Statement and the 2007 Statement, and the issue as to whether a price control should have been imposed on it in the 2007 Statement are not price control matters and hence fall to be determined by the Tribunal. We refer hereafter to the first group of issues as the “SMP Issue” and to the second group of issues as the “Appropriate Remedy Issue”. The issues in the H3G appeal and in the BT appeal as to the level of the price control imposed by the 2007 Statement are specified price control matters which must be determined by the Competition Commission.

35. BT and the four 2G/3G MNOs have intervened in the H3G appeal. Broadly the stance taken by them in relation to the non price control matters which the Tribunal must determine is as follows:

(a) Orange intervenes in relation to the SMP issue only to the extent that the issues raised by H3G go to the proper interpretation of OFCOM’s dispute resolution powers under section 185 of the 2003 Act. Orange broadly supports OFCOM’s interpretation of its dispute resolution powers. In relation to the Appropriate Remedy Issue, Orange also supports the arguments raised by OFCOM in its Defence;

(b) O2 limited its intervention, as regards the non price control matters, to supporting OFCOM’s Defence on the Appropriate Remedy Issue;

(c) Vodafone adopts OFCOM’s submissions in their entirety and urges the Tribunal to reject all aspects of H3G’s appeal on the non price control matters;

(d) T-Mobile largely supports and adopts OFCOM’s submissions in relation to the SMP issue, save (i) in respect of certain submissions relating to the appropriate approach to dispute resolution and (ii) as regards OFCOM’s focus on the end-to-end connectivity obligation in the determination of the disputes referred to OFCOM under section 185 of the 2003 Act. As regards the Appropriate Remedy Issue, T-Mobile largely supports and adopts OFCOM’s arguments;

(e) BT intervened in support of OFCOM on both the SMP Issue and the Appropriate Remedy Issue. However BT disagrees with OFCOM’s interpretation of its dispute resolution powers and supports instead OFCOM’s second line of defence in relation to the SMP findings (described below).

(ii) The Termination Rate Disputes

36. OFCOM’s dispute resolution powers under section 185 of the 2003 Act have been described earlier. Between 21 December 2006 and 19 March 2007 (that is before the publication of the 2007 Statement), a number of disputes between BT and the MNOs over the level of MCT charges for calls from BT’s network to the mobile networks were referred to OFCOM for determination. BT’s complaint was, broadly speaking, that the 2G/3G MNOs wanted to charge BT a “blended rate” for calls incorporating an additional charge in respect of calls which were terminated on the 3G network of the 2G/3G MNOs. The rate that those MNOs could charge in respect of termination on the 2G spectrum was set by the price control imposed in the 2004 Statement; but that Statement had left 3G termination unregulated.

37. In addition to these disputes, there was also a dispute between BT and H3G as to the price which H3G should charge for calls to its subscribers (whether terminated on H3G’s own 3G network or on the 2G network of its roaming partner) and disputes between H3G and Orange and between H3G and O2 as to the rate of Orange’s and O2’s MCT charges payable by H3G.

38. On 7 July 2007⁴ OFCOM issued determinations of the disputes concerning BT in its Statement Determinations to resolve mobile call termination rate disputes between T-Mobile and BT, O2 and BT, Hutchison 3G and BT and BT and each of Hutchison 3G, Orange and Vodafone (“The BT Dispute Determinations”) and on 10 August 2007 OFCOM issued its determination of the disputes concerning H3G and Orange/O2 Determinations to resolve mobile call termination rate disputes between Hutchison 3G and each of O2 and Orange (“the H3G Dispute Determinations”). Broadly, OFCOM found that the 2G/3G MNOs were entitled to charge a blended rate and upheld the level of charges which all five MNOs sought to apply.

39. Four appeals were lodged against the BT Dispute Determinations, by BT, by T-Mobile, by H3G and by a number of FNOs. Further, H3G appealed against the H3G Dispute Determinations. These appeals are referred to as the Termination Rate Dispute Appeals. The rates fixed in the BT and H3G Dispute Determinations were to a large extent overtaken by the 2007 Statement which fixed the rates for the period 1 April 2007 to 31 March 2011. Those Determinations were still relevant for the period up to 31 March 2007 and the appeals were pursued in respect of that period. As will be seen, some of the issues raised in the Termination Rate Dispute Appeals overlap with the issues raised by the SMP Issue in the current H3G appeal. This is because a key element of OFCOM’s reasoning in deciding that H3G has SMP in the market for call termination on its network concerned the test that OFCOM would apply in relation to any dispute about H3G’s charges referred to it by BT under section 185 of the 2003 Act. By an order dated 31 October 2007 in the Termination Rate Dispute Appeals and an order dated 20 November 2007 in H3G’s Appeal

⁴ The BT Dispute Determinations were reissued on 19 July 2007 with some minor typographical errors corrected.

against the Reassessment Statement and 2007 Statement, the Tribunal ordered that the overlapping issues in H3G’s Appeal and the Termination Rate Dispute Appeals be heard at a combined hearing in January and February 2008. This hearing took place between 24 January and 5 February 2008.

IV. PRELIMINARY REMARKS

40. This is the judgment of the Tribunal on the non price control matters arising from H3G’s appeal against the Reassessment Statement and the 2007 Statement. The Tribunal is handing down at the same time a separate judgment on the core issues in the Termination Rate Dispute appeals.

41. A number of the parties to the H3G appeal lodged statements from witnesses of fact relating to the non price control matters. None of the parties asked to cross examine the witnesses of fact, so this evidence has been treated as uncontested in so far as it sets out primary facts. H3G, OFCOM and T-Mobile also served expert evidence from economists relating to certain aspects of the case, as discussed below. All the parties emphasised at the hearing that they relied not only on the points made in their oral submissions but on the points raised in their pleadings, witness statements and their skeleton arguments. The Tribunal has carefully considered all the written material submitted by the parties as well as the oral argument, in arriving at the conclusions set out in this judgment.

42. In drafting this judgment, the Tribunal has also had regard to the introductory remarks in the judgment of the Court of Appeal in Argos & Littlewoods [2006] EWCA (Civ) 1318:

“5. Complicated as appeals of this kind to the Tribunal are often likely to be, it may be that the Tribunal will, over time, find it possible to deal with such appeals in judgments which are not so long as those under appeal in the present cases. …The reasoning of Griffiths LJ (as he then was) in Eagil Trust Co Ltd v. Pigott-Brown [1985] 3 All ER 119 at 122, endorsed by the Court of Appeal as being of general application in English v. Emery Reimbold & Strick Ltd [2002] EWCA Civ 605, [2002] 1 W LR 2409, applies to such a judgment of the Tribunal as to any other. Griffiths LJ said:

"a judge should give his reasons in sufficient detail to show the Court of Appeal the principles on which he has acted and the reasons that have led him to his decision. They need not be elaborate. I cannot stress too strongly that

there is no duty on a judge, in giving his reasons, to deal with every argument presented by counsel in support of his case. It is sufficient if what he says shows the parties, and if need be, the Court of Appeal the basis on which he has acted … (see Sachs LJ in Knight v. Clifton [1971] Ch 700 at 721)."

6. The same applies to findings of fact, so that the Tribunal may not need to make a finding on every disputed factual issue. Nor is it always necessary for the Tribunal to set out each party's submissions in detail before explaining its reasons for deciding the case. We therefore express the hope that, in future, it will be possible for the Tribunal to express its findings of fact and its reasoning in more succinct form. Its efforts to do so will have the support of this court, provided always that the essential tasks identified by Griffiths LJ have been fulfilled.”

43. According to Schedule 4 to the Enterprise Act 2002 the Tribunal must have regard to the need for excluding from its decision, so far as practicable, commercial information the disclosure of which would or might, in the Tribunal’s opinion significantly harm the legitimate business interests of the undertakings to whom the information relates. This is subject to the requirement that the Tribunal also has regard to the extent to which any such disclosure “is necessary for the purpose of explaining the reasons for the decision”: see paragraph 1(3) of Schedule 4. At an early stage of these appeals, a confidentiality ring was set up by the Tribunal to ensure that information that the parties considered confidential was kept within the circle of the parties’ legal advisers and external consultants. In preparing this decision, the Tribunal has had regard to the fact that some aspects of its decision are only comprehensible if the relevant figures are set out and that much of the information initially regarded as confidential by the parties is now out of date. The Tribunal has therefore only redacted figures which relate to the period after 1 April 2007 (when the price control which is the subject of these appeals) came into force.

V. THE FINDINGS OF SIGNIFICANT MARKET POWER (i) Background

44. The first main part of H3G’s appeal is its assertion that OFCOM erred in finding that H3G had SMP over the period covered by the Reassessment Statement as extended (that is between 2004 and 2007) and by the 2007 Statement (that is between 2007 and 2011). The relevant provisions of the Framework Directive which refer to the concept of SMP have already been described in broad terms.

Article 16(3) of the Framework Directive provides that OFCOM must determine, on the basis of its market analysis, whether a relevant market for mobile call termination is “effectively competitive”. Article 16(4) of the Framework Directive goes on to provide –

“Where a national regulatory authority determines that a relevant market is not effectively competitive, it shall identify undertakings with significant market power on that market …”

45. Recitals (25), (27) and (28) of the Framework Directive explain the background to the concept of significant market power as used in the CRF. Recital (27) states that it is essential that ex ante regulatory obligations should only be imposed where there is not effective competition, that is 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. Article 14(2) of the Framework Directive defines SMP in the following terms:

“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 it the power to behave to an appreciable extent independently of competitors, customers and ultimately consumers”.

46. This definition is reflected in section 78 of the 2003 Act which provides that a person shall be taken to have significant market power in relation to a market if he enjoys a position which amounts to or is equivalent to dominance of the market. Section 78 also provides that references to dominance of a market must be construed in accordance with any applicable provisions of Article 14 of the Framework Directive.

47. H3G does not dispute OFCOM’s definition of the relevant market as the market for wholesale mobile voice call termination provided to the other communications providers by H3G in the United Kingdom. It also does not dispute OFCOM’s finding that H3G has 100 per cent market share in that relevant market and that there are absolute barriers to entry which prevent the development of any competitor to H3G in that market. What H3G disputes is OFCOM’s conclusion that BT did not have sufficient countervailing buyer

power to negate what would otherwise be H3G’s SMP. H3G’s case in a nutshell is that on a proper analysis of the legal and factual position, H3G is unable to act independently of its competitors, customers and ultimately its consumers, because of the countervailing buyer power (“CBP”) exercised by BT.

48. The debate focuses on the buyer power of BT, rather than any of the other communications providers who interconnect with H3G because BT is the largest purchaser of mobile call termination. The charge that BT agrees with each MNO acts as a ceiling to the charge that other operators would be willing to accept from the MNO. It also acts as a floor because the terminating MNO would be unwilling to agree a more favourable rate with another MNO. OFCOM therefore treats BT’s charge as effectively setting the charge for all other agreements between suppliers and purchasers of mobile call termination. Even if this was not the case, the fact that BT is by far the largest purchaser of mobile call termination means that if it were found not to have a level of CBP sufficient to negate any prima facie finding of SMP, it could reasonably be assumed that neither would any other purchaser of mobile call termination.

49. The importance of assessing CBP when considering whether an undertaking has SMP is recognised in the guidance issued by the EC Commission. First there is the Guidance published by the EC Commission in 2002 on market analysis and the assessment of significant market power under the Common regulatory framework [2002] OJ C165/6 where the Commission states (paragraph 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. …”

50. Similarly, the EC Commission notes on page 20 of its Explanatory Memorandum to its Recommendation on Market Definition that a market definition of call termination on individual networks:

“… does not automatically mean that every network operator has significant market power; this depends on the degree of any countervailing buyer power and other factors potentially limiting that market power. 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 (as required by the regulatory framework) 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.”

51. The Tribunal was also referred to Guidance issued by the Director General of Communications in August 2002 on criteria for the assessment of SMP and to a Working Paper on the SMP concept in the new regulatory framework produced in October 2004 by the European Regulators Group (“ERG”). The ERG is a group of NRAs established by the European Commission to encourage cooperation and coordination in order to promote the development of the internal market for electronic communications: see recital (36) of the Framework Directive. Both these documents point to an assessment of CBP being an important factor when assessing the existence of SMP.

52. H3G’s appeal against the 2004 Statement also centred on the existence and extent of any CBP on the part of BT. As we have already mentioned, that appeal was successful and the Tribunal in its judgment H3G (1) remitted the 2004 Statement to OFCOM for further consideration, giving rise to the Reassessment Statement. The reasoning used by OFCOM in rejecting H3G’s argument on CBP was essentially the same in both the Reassessment Statement and the 2007 Statement. H3G’s submissions, at least by the time of the hearing, also did not distinguish between the two Statements. We therefore set out our conclusions on the challenge to the finding of SMP in Statements together.

(ii) OFCOM’s reasoning on the existence of CBP

53. OFCOM began its assessment of CBP by referring to the fact that CBP is “not a binary issue” – it is not an absolute concept in terms of its strength. It is a concept which embodies a possible range of strengths so that 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. Having referred to guidance issued by the Office of Fair Trading and the ERG,

OFCOM identified the economic factors which are relevant to a finding of CBP and considered how they applied to BT. OFCOM found a number of factors present in the market which would point towards BT having CBP:

(i) BT is a well informed and price sensitive purchaser of mobile call termination and is increasingly price sensitive as H3G’s subscriber base grows.

(ii) BT is an important customer for H3G – it is generally accepted that no network operator can survive in the market without an interconnection agreement with BT.

However, OFCOM also found that there were factors that pointed against BT having CBP:

(iii) Whereas BT may have been in a position to exercise CBP at the time of the initial negotiations on MCT rates before the launch of H3G’s retail service because it could have delayed or threatened to delay interconnection and hence disrupt the launch of H3G’s service, it was not in a position to engage in such conduct during either of the relevant periods, that is between 2004 and 2007 or between 2007 and 2011 because there was a binding contract in place.

(iv) There was no “reciprocity of trade”: that is to say that because BT’s own call termination rates are fixed by OFCOM, BT cannot try to bring the price at which it sells termination on the BT network to H3G into the negotiation over the price at which it buys termination on the H3G network.

(v) There are no alternative sources of supply to which BT could turn if H3G tried to impose too high a price.

(vi) BT does not have the option either to refuse to buy mobile call termination from H3G or to delay purchasing mobile call termination in

order to put pressure on H3G to lower its price. This is partly due to commercial constraints which prevent BT from acting in such a manner and partly due to the fact that, as explained further below, BT is subject to a regulatory obligation to interconnect with other network operators.

(vii) That regulatory obligation imposed on BT (generally referred to as BT’s end-to-end connectivity obligation) entitles BT to refer to OFCOM disputes with H3G over the reasonableness of the prices H3G charges for MCT. But the existence of this dispute resolution procedure does not, OFCOM found, mean that BT has sufficient CBP.

54. OFCOM also analysed other aspects of the market and what they indicated about the existence of CBP. It looked at the recent negotiations which had taken place between the different parties and the attempts by the parties to change the level of the mobile call termination rates. OFCOM concluded (paragraph 5.75 of the 2007 Statement) that the behaviour of purchasers and suppliers of MCT, in respect of proposals to increase or decrease MCT charges, had been strongly conditioned by the existence or threat of regulation in these markets, and by the expectation that these markets may be subject to further regulation from April 2007. OFCOM reasoned that no inferences could therefore be drawn about the parties’ relative bargaining power from observing their behaviour on the market at such a time.

55. As well as the end-to-end connectivity obligation, there were other regulatory constraints imposed on BT which, OFCOM found, may weaken BT’s CBP. BT is required to allow Carrier Pre Selection (CPS) and Indirect Access (IA) which enable competing retail service providers to provide calls to customers using the BT network. OFCOM noted (paragraph 5.91) that while the purpose of these conditions is to promote competition in a range of downstream markets, they also have a specific impact on the retail market for calls to mobiles. The ability of consumers to switch to alternative CPS or IA based providers of such calls may weaken BT’s ability to threaten to cease purchasing wholesale MCT.

56. There were also more general commercial constraints on BT’s ability to refuse to buy interconnection from MNOs. OFCOM noted that there would potentially be a significant commercial imperative for all originating operators, including BT, to provide their subscribers with the opportunity to call each of the mobile networks. OFCOM commented (paragraph 5.142 of the 2007 Statement, footnotes omitted):

“Although it might be thought that this may not be the case where a new entrant MNO (with few, if any customers) wishes to sell call termination to a large incumbent network, Ofcom notes that the evidence suggests that BT, the largest purchaser of MCT, regarded the entry of H3G in 2001 as an opportunity for incremental income from its retail customers rather than a potential threat to its own access and origination revenue. In this case BT therefore judged that it had a commercial incentive to purchase call termination services from H3G.”

57. In light of all these factors OFCOM’s view was that BT does not have sufficient CBP to constrain the MNOs’ ability to behave to an appreciable extent independently of competitors, customers and ultimately consumers, such that MNOs are unable to sustain charges appreciably above the competitive level.

58. As regards the Reassessment Statement, OFCOM concluded that nothing had occurred in the market over the period between the publication of the 2004 Statement and the adoption of the Reassessment Statement to alter its earlier assessment and hence OFCOM concluded that BT did not have CBP sufficient to negate H3G’s market power. It further concluded that H3G had SMP in the relevant market but that it was not appropriate to impose a price control on H3G’s mobile call termination. In the 2007 Statement OFCOM largely repeated the reasoning on which it relied in the Reassessment Statement.

59. As regards the different factors for and against BT’s CBP, H3G adopts and relies on points (i) and (ii). H3G’s appeal relates to that part of OFCOM’s findings which decided that BT did not have power to refuse or delay purchase of mobile call termination from H3G because of BT’s regulatory obligation to interconnect. H3G accepts that OFCOM has correctly proceeded on the basis that the crucial question is the extent to which BT’s CBP is constrained by its obligation to interconnect with other operators. However, H3G argues that OFCOM has misconstrued the obligation to which BT was subject. Key to this appeal

therefore are the proper interpretation of the regulatory constraints imposed on BT and the way that OFCOM would be expected to resolve a dispute about the reasonableness of MCT charges if BT referred such a dispute to it.

60. There has been some confusion caused by the use of different terms to describe different levels of pricing. We therefore set out at the start how the Tribunal is using these terms in this judgment, whilst recognising that these may not bear the same meanings as those used by OFCOM or the parties in this case.

(a) An abusive price means a price which is so high as to amount to an abuse by the undertaking of its dominant position;

(b) An excessively high price means a price which would satisfy the test set out in section 88 of the 2003 Act that OFCOM must apply when considering whether it has power to include a price control in the SMP conditions it decides to set under section 87 of that Act;

(c) A price appreciably above the competitive level means a price which may not be cost based and is in fact significantly above a price set at a competitive level;

(d) A price set at the competitive level means a price which normally reflects the costs incurred by an efficient operator. (The Tribunal does not thereby prejudge in any way the approach to be adopted by the Competition Commission in determining the price control matters in this appeal.)

These categories are not mutually exclusive – for example an abusive price will also clearly be an excessively high price and a price which is appreciably above the competitive level may, or may not, be so high as also to be abusive or excessively high.

61. Before considering OFCOM’s reasoning and H3G’s challenge to it in detail, it is necessary to describe BT’s obligation to interconnect with other communications providers and how BT contracts with the MNOs for interconnection.

(iii) BT’s end-to-end connectivity obligation

62. “End-to-end connectivity” describes the process of enabling retail customers to make calls to other customers on the same network or on other providers’ networks. Competing communications providers need to be able to interconnect with other networks so that their subscribers can call any person who has a mobile or fixed phone irrespective of the network used by the person being called. The Common Regulatory Framework recognises the importance of interconnection and article 5(1) of the Access Directive provides that -

“National regulatory authorities shall, acting in pursuit of the regulatory objectives set out in article 8 of [the 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”.

63. Article 5 further provides that NRAs must be empowered to impose obligations on undertakings to the extent that this is necessary to ensure end-to-end connectivity and, as we have seen earlier, article 5(4) provides for OFCOM to take action either on its own initiative or when asked to resolve a dispute between the parties.

64. Before September 2006, OFCOM had not imposed an explicit obligation on BT aimed at ensuring end-to-end connectivity. However, it appears that the industry in general and BT in particular acted on the basis that BT was bound, whether formally or informally, to provide interconnection because of BT’s obligations as a universal service provider (in accordance with Guidance issued by the former Director General of Telecommunications on “End-to-end connectivity” dated 27 May 2003) and before that, because of a condition in BT’s licence under the old regulatory regime. On 13 September 2006 OFCOM imposed a condition on BT under section 74(1) of the 2003 Act requiring it to provide “end-to-end connectivity”, that is to say, a condition which obliged BT to purchase wholesale

MCT services on reasonable terms from any MNO requesting it to do so (“the E2E Statement”). The end-to-end connectivity obligation is an “access-related condition” for the purposes of section 73(2) of the 2003 Act and that subsection provides that OFCOM may impose such conditions as appears to it appropriate for the purpose of securing efficiency on the part of communications providers, sustainable competition between them and the greatest possible benefit for the end-users.

65. OFCOM explained in the E2E Statement why it decided to impose an end-to-end connectivity obligation only on BT and not on all the MNOs. OFCOM noted that once a communications provider has secured an agreement to send calls to BT’s network, they are in a position to send calls to all other networks (thereby securing end-to-end connectivity for their subscribers) because of BT’s position as a transit provider. This meant that imposing an obligation on all providers was neither appropriate nor proportionate.

66. OFCOM further explained its decision as regards the terms and conditions under which BT would be obliged to contract with a public electronic communications network (PECN):

“3.32 Ofcom is also proposing that BT is not obliged to purchase wholesale narrowband call termination services at any price, but to do so where requested by a PECN and where the terms and conditions offered by that PECN are reasonable. Whether a particular term or condition (including charge) is reasonable will depend on the particular circumstances relating to any decision not to purchase in the context of the need to ensure end-to-end connectivity and may lie within a broader range of outcomes than that which might be considered in the circumstances of SMP. In particular, as Ofcom has to ensure that any charges it imposes are proportionate, it is unlikely to set charges at a level set in the context of addressing a finding of SMP.”

67. None of the parties has argued in this appeal that the Tribunal should treat the period before 13 September 2006 differently from the period after that date for the purposes of these appeals. We have therefore proceeded on the basis that BT was at all material times subject to an end-to-end connectivity obligation but that there was at all times a proviso (“the E2E Proviso”) to that obligation to the effect that if the terms proposed by an actual or proposed counterparty did not

seem reasonable, after negotiations in good faith, BT was entitled to refer the matter to OFCOM for resolution.

(iv) BT’s Standard Interconnection Agreement and dispute resolution

68. When BT enters into an interconnection agreement with another operator it does so on the terms of a Standard Interconnection Agreement or “SIA”. This SIA is a substantial document which sets out a wide range of services provided by BT to the counterparty and by the counterparty to BT. The SIA is entered into for an indefinite term and can be terminated only on 24 months’ notice. Clause 12 deals with the provision of services by BT to the MNO and clause 13 deals with the provision of services by the MNO to BT. Both clauses stipulate that the charges payable by the recipient of the services are the charges specified from time to time in a document known as the Carrier Price List. Both clauses also contain provision for the variation of those charges though these are not the same in both clauses. Clause 13 sets out the mechanism whereby the parties can seek to vary the price charged for the services that the MNO provides to BT by sending to each other a Charge Change Notice proposing a new charge. If the parties cannot agree on whether the proposed charge should come into effect, then either party may refer the matters in dispute to OFCOM. If OFCOM upholds the charge proposed in the Charge Change Notice the change takes effect on the date specified in the Charge Change Notice and the parties must enter into an agreement to modify the SIA accordingly. If OFCOM does not uphold the proposed change then that Charge Change Notice ceases to be of any effect. The parties refer to a Charge Change Notice served under clause 13 as an “Operator Charge Change Notice” or “OCCN” to distinguish it from a notice concerning a proposed change in BT’s prices served under clause 12 of the SIA. It is common ground that when a dispute is referred to OFCOM under clause 13 of the SIA, its jurisdiction to determine the dispute is the jurisdiction now set out in section 185 of the 2003 Act.

(v) OFCOM’s reasoning concerning the end-to-end connectivity obligation

69. OFCOM considered the effect of BT’s end-to-end connectivity obligation in the context of assessing H3G’s market power because it analysed whether BT could

counter that SMP by refusing to purchase or delaying the purchase of mobile call termination from H3G. OFCOM noted that BT is constrained in its ability to refuse or delay purchase by the end-to-end connectivity obligation. But because of the E2E Proviso if BT and an MNO were unable to agree upon terms for the supply of MCT, either party could refer the dispute to OFCOM to resolve. Therefore, the extent to which any CBP that BT would otherwise possess is negated by its end-to-end connectivity obligation could depend in part on the expectation of the parties to a dispute as to how OFCOM would set about resolving that dispute.

70. OFCOM then stated as follows (paragraph 5.152 of the 2007 Statement):

“In resolving a dispute relating to the application of BT’s end-to-end connectivity obligation, Ofcom would consider each dispute on its merits, in the light of the specific facts and circumstances and the arguments put to it by the parties to the dispute, including the reasonableness of any resolution on both parties.”

71. OFCOM considered that a number of alternative approaches would be open to it in dealing with a dispute. But it would consider the question of what is “reasonable” for the purposes of the E2E Proviso by reference to the purpose underlying BT’s end-to-end connectivity obligation. OFCOM identified this purpose as to remove the risk of a potential market failure from BT refusing to buy call termination. OFCOM recognised that while in principle BT’s customers value calling customers of smaller networks and customers of smaller networks may value receiving such calls, there may be some circumstances in which BT would not have an incentive to provide interconnection for such calls. On this basis, OFCOM continued (paragraph 5.154):

“… a reasonable charge for BT to purchase MCT with a view to ensuring end-to-end connectivity may be at a price appreciably above the competitive level. As such, if a charge appreciably above the competitive level were in dispute, Ofcom considers it unlikely that it would insist on a strictly cost based charge (such as used in deriving cost benchmarks ... to set efficient regulated charges in [an SMP condition] charge control) ie a charge that was not appreciably above the competitive level.”

72. In OFCOM’s view neither party in a negotiation over MCT charges would assume that OFCOM when resolving such a dispute would impose a charge for

MCT that was not appreciably above the competitive level. OFCOM concluded (at paragraph 5.161) that:

“… a purchaser and supplier of mobile call termination, properly apprised as to Ofcom’s approach to dispute resolution, would therefore negotiate on the basis that if a charge appreciably above the competitive level were in dispute, Ofcom would be unlikely to impose a charge for MCT in the context of such a dispute that was not appreciably above the competitive level.”

73. Given this expectation, the E2E Proviso did not qualify BT’s obligation to buy interconnection from H3G to a degree which meant that BT was in a position to exercise countervailing buyer power against H3G. Therefore OFCOM’s dispute resolution function in the context of BT’s end-to-end connectivity obligation did not have the effect of negating H3G’s market power.

74. OFCOM went on to consider other aspects of the dispute resolution function which also contributed to its conclusion that that function did not rule out SMP on the part of H3G. OFCOM considered that certain limitations of the dispute resolution process mean that it should not be seen as a substitute for the appropriate regulatory processes for addressing the question of market power as set out in articles 15 and 16 of the Framework Directive. In addition, dispute resolution is of limited assistance in curbing pricing appreciably above the competitive level as it is aimed at resolving a dispute between two (or more) parties and regulators can only act in the context of that dispute - which may not address the “regulatory” issue i.e. general pricing issues. Moreover, it is only a mechanism that OFCOM can rely on when asked to do so by one or more of the parties in dispute. It is therefore not necessarily the case that: an MNO would bring a dispute; or, another provider would refuse to purchase interconnection at a charge appreciably above the competitive level.

(vi) The Grounds of Appeal and OFCOM’s Defence

75. H3G relied on three grounds of appeal in its challenge to the findings of SMP:

(a) first H3G argued that BT’s ability to refer disputes to OFCOM itself precluded SMP on the part of H3G;

(b) in the alternative, H3G argued that the likely outcome of a dispute referred to OFCOM would, if OFCOM interpreted its powers properly, inform the negotiations and act as a constraint on H3G;

(c) finally H3G argued that if OFCOM is correct in interpreting the end-to-end connectivity obligation in a manner which in effect requires BT to buy MCT at high prices, then that is a flaw in the drafting of the end-to-end connectivity obligation and OFCOM cannot cure that fault by imposing price control regulation on H3G.

76. During the course of the proceedings, the first two grounds tended to merge into one. The extent to which the ability to refer a dispute will act as a constraint on the negotiating positions of the parties depends on their expectation as to how OFCOM would resolve that dispute. We do not therefore deal separately with the “ability to refer” point and the “expectation of the likely outcome” point.

77. H3G argued that the basic error that OFCOM has made throughout its consideration of this matter is in relation to its interpretation of the E2E Proviso which limits BT’s obligation to interconnect to where interconnection is offered on reasonable terms and conditions. Going back to the wording of the E2E Statement, H3G identifies the first appearance of OFCOM’s error in the statement made by OFCOM that when it is considering whether terms presented to BT are reasonable, it is unlikely to set charges at the level it would set in the context of addressing a finding of SMP. OFCOM, H3G says, jumps from saying that there is an obligation on BT to pay for interconnection at a “reasonable” price to saying that they would not impose the same price that they would if considering a “reasonable” price for the purpose of setting an SMP price control condition. This is wrong, H3G say, because where a dispute concerns the reasonableness of proposed price under BT’s end-to-end connectivity obligation, OFCOM should not be imposing a price at all but simply be declaring the maximum price that BT can reasonably be expected to pay.

78. The thrust of H3G’s case on how OFCOM should approach resolving a dispute between BT and H3G over H3G’s mobile call termination charges was as follows:

(a) H3G accepts that the existence of the end-to-end connectivity obligation on BT would be an important consideration in any such dispute;

(b) However, OFCOM erred in treating the end-to-end connectivity obligation as having only the purpose of securing interconnection between BT and the other networks – according to section 73(2) of the 2003 Act, the purpose is to ensure interconnection in a way which promotes competition and accords with OFCOM’s wider statutory obligations under the Common Regulatory Framework and the 2003 Act;

(c) given OFCOM’s wider duties under the 2003 Act it would be ultra vires for OFCOM to determine a dispute by setting a price at a level which constitutes an excessively high level likely to have adverse consequences for end-users so as to trigger OFCOM’s power to set a price control under section 88 of the 2003 Act.

79. The wider duties to which H3G refers in this part of its argument are those contained in article 8 of the Framework Directive as implemented by sections 3 and 4 of the 2003 Act. The objectives in article 8 so far as they relate to the NRAs’ duty to promote competition have already been described (see paragraph [14] above). Section 3 of the 2003 Act sets out the two principal duties of OFCOM in carrying out its functions, namely to further the interests of citizens in relation to communications matters and to further the interests of consumers in relevant markets “where appropriate by promoting competition”. The other subsections of section 3 set out a wide range of matters to which OFCOM must have regard, where relevant, in performing its functions. These include securing the optimal use of the spectrum (section 3(2)(a)); the desirability of encouraging investment and innovation in relevant markets (section 3(4)(d)); and the interests of consumers in respect of choice, price, quality of service and value for money (section 3(5)).

80. Section 4 of the 2003 Act sets out six Community requirements in accordance with which OFCOM must act when carrying out various of its functions, including its dispute resolution function under section 185. The relevant Community requirements for the purposes of these appeals are:

(a) the first Community requirement which includes the requirement to promote competition in relation to the provision of electronic communications networks and services (section 4(3)(a));

(b) the fourth Community requirement which is a requirement to take account of the desirability of OFCOM carrying out its functions in a manner which, so far as practicable, does not favour one form of network or service or one means of providing such a network or service over another (often referred to as the requirement of technological neutrality) (section 4(6)); and

(c) the fifth Community requirement which is to encourage the provision of network access and service interoperability for the purpose of securing efficiency and sustainable competition and the maximum benefit for customers of communications providers.

81. OFCOM’s main line of Defence relied on the reasoning set out in the Reassessment and 2007 Statements namely that it was legally possible that the outcome of the exercise of its function under section 185 in a dispute between H3G and BT could be a price set at a level which was appreciably above the competitive level. This meant that the dispute resolution powers were not a sufficient constraint on H3G’s exercise of market power to negate a finding of SMP.

82. In its Defence OFCOM elaborated on what this would mean:

“ … When determining a dispute as to the price for [call termination], Ofcom will be concerned principally with reasonableness as regards BT, the party that is subject to the obligation in question. Ofcom will accordingly seek to ensure that the price asked of BT for termination is not so high that it would be unreasonable for BT to be expected to purchase at that price; and conversely that BT does not

insist on a price that is so low that it cannot reasonably expect termination to be supplied at that price (i.e. in effect a constructive refusal by BT to purchase).”

83. OFCOM argues that compliance with its other regulatory obligations does not mean that its only option in determining a dispute between H3G and BT would be to impose a price at the competitive level and thus a consideration of H3G’s costs. On the contrary, such an approach would amount, in OFCOM’s view to using the end-to-end connectivity obligation as a regulatory device for imposing a price control on an MNO, circumventing the thorough SMP regime.

84. Whether, when referring to a price “appreciably above the competitive level” in its reasoning, OFCOM meant a price which was or which could include an “excessively high price” or an “abusive price” in the sense that the Tribunal has described those terms was never entirely clear. H3G interpreted it in that way and submitted that OFCOM had erred by holding that it could, in the exercise of its dispute resolution powers require BT to connect with other networks even at prices which were excessive and uncompetitive – even so excessive as to have an adverse effect on consumers. The Tribunal understands OFCOM’s reasoning in the Statements and in its submissions in these appeals as indicating that OFCOM was not accepting any particular constraint on the price which it might consider appropriate to set in the context of dispute resolution.

85. OFCOM put forward a second line of defence. If, contrary to its primary position, the dispute resolution powers did require OFCOM to resolve a dispute in a manner which ensured that the price was not appreciably above the competitive level, then it would still not constrain H3G’s SMP. This is because such dispute resolution powers would amount to regulation of H3G and hence fall to be disregarded when assessing H3G’s market power. Such powers must be disregarded, OFCOM argues, to avoid the circularity which arises from an argument that an undertaking cannot be regarded as having the ability to act independently of customers and consumers if there is a mechanism in place to allow the regulator to prevent the undertaking from attempting to do so. OFCOM argued that H3G’s position was contrary to the scheme of the Common Regulatory Framework because it would in effect make it impossible for the regulatory authority to find SMP in relation to wholesale markets where the

participants had the right under the Directives to refer a dispute about terms and conditions to the regulator.

86. Vodafone supported OFCOM’s conclusions as regards the effect of its dispute resolution powers on BT’s CBP whilst appearing not to go quite as far as OFCOM went – or at least as far as H3G argues that OFCOM went. Vodafone argued that OFCOM might well approve charges which may exceed the most efficient level but which are nonetheless within the bounds of reasonableness such that BT could pay them without suffering any significant detriment itself and without giving rise to any significant detriment to end users. There is no reason to expect that OFCOM would resolve a dispute by holding that BT’s obligation to purchase applies only if H3G offers the most efficient cost-reflective price for such services. This means, according to Vodafone, that OFCOM’s exercise of its dispute resolution powers “would be unlikely to be sufficient to prevent H3G from setting its mobile call termination charges at a level which appreciably exceeds the most efficient level for such services”. On Vodafone’s argument, it is not necessary to conclude that OFCOM could lawfully set an abusive price or an excessively high price under its dispute resolution powers in order to negate SMP. It is only necessary to acknowledge that it could set a price above the level that would be the most efficient cost related price.

(viii) Can OFCOM determine a dispute pursuant to section 185 by setting a price appreciably above the competitive level?

87. The first issue between the parties was therefore whether OFCOM was right to assert that the parties could not expect, when negotiating about a proposed change of price, that OFCOM would determine a dispute between them in a way that ensured that BT did not have to pay a price which was appreciably above the competitive level. If OFCOM was wrong about that and BT’s and H3G’s expectations should be that BT can always avoid being forced to pay such a price by referring the dispute to OFCOM then, say H3G, BT has sufficient countervailing buyer power to counteract any significant market power arising from H3G’s monopoly on call termination on its network.

88. This question as to the parties’ expectations about OFCOM’s approach to dispute resolution had become a live issue by the time the Reassessment Statement and the 2007 Statement were issued in March 2007. This was because the disputes over MCT charges between BT and the MNOs and between H3G and two of the 2G/3G MNOs had by that time been referred to OFCOM under section 185 of the 2003 Act (see paragraph [37] above). The reasoning set out by OFCOM in its decisions determining those disputes closely reflected how OFCOM had said, in the 2007 Statement, that it would approach dispute resolution in such circumstances. Thus H3G’s allegations in the current appeal that OFCOM erred in law in stating how it would resolve any disputes translate, in the context of H3G’s appeal against the BT and H3G Dispute Determinations, to allegations that OFCOM erred in law in the way it in fact determined those disputes. The Tribunal’s judgment on the core issues in the Termination Rate Dispute appeals sets out in full the Tribunal’s reasoning as to why, in the Tribunal’s judgment, OFCOM erred in its exercise of those powers.

89. For the purposes of the current judgment, the Tribunal’s conclusions can be stated as follows. OFCOM is correct in asserting that it has a range of options available to it when exercising its dispute resolution powers in relation to disputes concerning MCT rates. OFCOM must, however, approach dispute resolution having regard to all its statutory obligations and not focus unduly on the existence of other regulatory constraints imposed on one or other of the parties to the dispute, such as BT’s end-to-end connectivity obligation. In resolving a dispute, there will be a range of possible prices which it would be intra vires for OFCOM to set. But it could never be appropriate for OFCOM to set a price which was an abusively high price. If OFCOM’s reasoning in the two Statements was intended to indicate that it did have power to set an abusively high price when resolving a dispute, then that would, the Tribunal finds, constitute an error of law.

90. There might, however, be cases in which it would be appropriate for OFCOM to resolve a dispute by setting a price which is not at a competitive level and which is even appreciably above that competitive level. In any dispute, there will be two prices being put forward, presumably a higher one being urged on OFCOM

by the supplier and a lower one being urged by the purchaser. It may be that in fact both these prices are prices above the competitive level and even appreciably above that level, without being abusive. This might happen for example where the prevailing price before the service of the OCCN was a price set at a time when it was a cost reflective price but where, by the time of the OCCN, costs have reduced so that the prevailing price is no longer cost reflective. The purchaser may still be content to pay the prevailing price but wish to resist a proposed increase in the supplier’s OCCN. If OFCOM resolved the dispute in the purchaser’s favour and upheld its rejection of the OCCN it would in effect be setting a price which was appreciably above the competitive level. Provided that the price is not so high that OFCOM should exercise its powers to intervene on its own initiative, it might be reasonable for OFCOM to resolve a dispute in that way. This is the case whether or not the dispute is referred in the context of one party being subject to an obligation to agree to “reasonable” terms – an obligation which can arise through a regulatory or a contractual condition.

91. This leaves the question of whether it would be ultra vires for OFCOM to resolve a dispute under section 185 by setting a price at an excessively high level within the meaning of section 88(3) of the 2003 Act. The Tribunal notes that this terminology is taken from article 13 of the Access Directive which refers to a situation where market analysis indicates that a lack of effective competition means that the operator concerned might sustain prices at an excessively high level, or apply a price squeeze, to the detriment of end-users. It is not at all clear whether the concept of “an excessively high price having adverse consequences for end-users” is intended to be the same as or different from the concept of “an abusively high price”. It is noteworthy that whereas the Common Regulatory Framework expressly provides that the concept of significant market power is intended to be the same as the concept of dominance under Article 82 EC (see article 14(2) of the Framework Directive) it does not say whether an abusively high price and a price sustained at an excessively high level are meant to mean the same thing or not.

92. In any event, there is nothing in the Common Regulatory Framework to suggest that the NRA is required, when exercising its dispute resolution powers, to ask

itself the rather convoluted question whether, if the price which is being proposed in the context of a dispute were instead being considered as the price which there was a risk that an undertaking with SMP might charge, that risk would be such that the power to set a price control SMP Condition was triggered. The test under article 13 of the Access Directive and under section 88(3) of the 2003 Act forms no part of the dispute resolution process. Rather it is one step in a complex series of tests to be applied in the course of a market review, balancing a wide range of different objectives, when the NRA is deciding which of a range of remedies it is appropriate to impose following a finding of SMP. We doubt that it is useful to ask whether OFCOM is precluded from setting a price which would, in fact, be “excessively high” whether or not OFCOM turned its mind to that question. In so far as the price would also be abusive we have held that OFCOM cannot set that price. In so far as the price would be something less than an abusively high price but still a price appreciably above the competitive level then whether it is appropriate to resolve the dispute at that price will depend on the circumstances of the particular case just as whether it would be appropriate to set a price control under section 88 in order to avoid the risk of that price prevailing would depend on all the circumstances of a different particular case.

93. Further, in the Tribunal’s judgment OFCOM did not need to decide whether it had the power to set an excessively high price when resolving a dispute in order to decide whether BT’s countervailing bargaining power was sufficient to negate H3G’s SMP. The Tribunal agrees with the submissions made by Vodafone in this regard. The expectations that the parties to a negotiation could properly have about the way in which OFCOM would resolve the dispute are not such as to give BT CBP sufficient to negate H3G’s market power.

94. Therefore the Tribunal finds that even if OFCOM did make the error of law alleged by H3G, the error was not such as to vitiate the finding that H3G had SMP in the Reassessment Statement and the 2007 Statement. OFCOM was right to reject H3G’s primary argument because the existence of the dispute resolution powers does not constrain H3G’s market power so as to undermine the finding of SMP.

(ix) OFCOM’s second line of defence

95. The Tribunal has heard extensive argument on OFCOM’s second line of defence which can be summarised as follow. OFCOM’s dispute resolution function under section 185 of the 2003 Act is a regulatory function. The determination of a dispute between H3G and BT amounts to regulation of both those parties, not just of BT. In so far as the dispute resolution powers therefore constrain H3G from setting an excessively high price, they fall to be disregarded as a matter of law when assessing whether H3G has SMP.

96. This is because of the application of what is called the “modified greenfield approach”. The modified greenfield approach is a way of distinguishing between those regulatory constraints existing in a market which ought to be taken into account when assessing market power under the Framework Directive and those which should be ignored. The application of the modified greenfield approach in a situation where the regulator is considering whether SMP exists has been approved both by the European Commission in its monitoring of the decisions of NRAs under the CRF and by this Tribunal in the H3G (1) judgment. In the H3G (1) judgment, discussed in more detail in paragraphs [108] et seq. below, the Tribunal held that under the modified greenfield approach, the potential power of the regulator to impose SMP conditions on H3G must be disregarded when considering whether H3G has SMP. Conversely the regulatory obligation imposed on BT to provide end-to-end connectivity did not fall to be disregarded but should be taken into account when considering the scope of BT’s countervailing buyer power and hence the strength of H3G’s market power.

97. OFCOM therefore argues that whatever the Tribunal decides as regards the correct test applied by OFCOM when exercising its dispute resolution powers, it would be contrary to the scheme of the CRF for the potential exercise by OFCOM or any NRA of its dispute resolution function to be treated as precluding SMP on the part of a network operator. H3G’s reliance on dispute resolution in this manner would effectively make it impossible for an NRA to find that a wholesaler has SMP in so far as that wholesaler is operating in a

market in respect of which the counterparty has a right to refer disputes to the regulator.

98. This line of argument was strongly supported by BT. BT’s position on the SMP Issue was that OFCOM’s power to regulate the MCT rates set by H3G through the dispute resolution process cannot be taken into account by OFCOM when assessing SMP. Thus, all H3G’s arguments about the proper test to be applied under the dispute resolution powers are irrelevant. BT asserts that in fact OFCOM’s understanding of its dispute resolution powers was wrong and that it is not entitled to set H3G’s mobile call termination rate at a level appreciably above the competitive level. This is consistent with the stance taken by BT in its challenge to the BT Dispute Determinations. T-Mobile also argued that although OFCOM’s approach to its dispute resolution powers was incorrect, none of OFCOM’s errors leads to H3G’s conclusion that the end-to-end connectivity obligation properly applied removes H3G’s or any other MNO’s SMP.

99. H3G’s counter-argument is that the function of dispute resolution in this case does not regulate the seller, H3G but only the buyer, BT. H3G accepts that OFCOM was right to regard its task in any dispute between BT and H3G as being a dispute about what are “reasonable” terms and conditions for the purposes of BT’s end-to-end connectivity obligation. Since BT is the only operator subject to an end-to-end connectivity obligation, OFCOM’s implementation of the E2E Proviso in the context of a dispute resolution amounts to regulation of the price at which BT buys mobile call termination – it is not regulation of the price at which H3G sells mobile call termination. OFCOM’s dispute resolution powers do not, therefore, fall to be disregarded in assessing H3G’s SMP under the modified greenfield approach.

100. In the alternative, H3G argues that if it is wrong and dispute resolution does amount to regulation of H3G as well as of BT, that regulation does not fall to be disregarded as a matter of law. According to H3G, the modified greenfield approach only precludes the regulator from taking account of the potential for SMP regulation in assessing whether an undertaking has SMP. It does not

require the regulator to ignore all potential regulation which might be brought to bear on the putative dominant undertaking.

101. H3G rejects the suggestion that its argument “emasculates” the SMP provisions in the CRF. The finding of SMP is very fact sensitive and if H3G is right that the potential for dispute resolution under BT’s end-to-end connectivity obligation prevents H3G from having SMP, that is because of the particular circumstances of H3G in this market. A finding that H3G did not have SMP would not automatically read across either to the other MNOs in the United Kingdom or to MNOs in other Member States.

(a) Do OFCOM’s dispute resolution powers in this case amount to regulation of both BT and H3G or only of BT?

102. H3G characterised OFCOM’s role in determining this dispute as “finding the price above which H3G’s mobile call termination rates would be unreasonable i.e. the price level above which BT would be entitled to refuse to purchase H3G’s call termination service”. H3G argued that the right and proportionate way for OFCOM to resolve a dispute about whether BT is being asked to interconnect at a reasonable price is not by imposing a price but by making a declaration. OFCOM should declare what is the maximum reasonable price at which BT can be obliged to interconnect. The effect of such a declaration is not to impose any price control on H3G because the parties remain free to negotiate commercially above or below the price that has been declared. All that OFCOM has to do is to clarify the obligation of BT under its end-to-end connectivity obligation.

103. The Tribunal rejects this analysis of OFCOM’s role in dispute resolution. First, this is contrary to the wording of article 20 of the Framework Directive from which the powers in section 185 in part flow. That article refers to the NRA issuing “a binding decision to resolve the dispute in the shortest possible time frame”. That is not consistent with OFCOM’s powers being limited in this case to setting a price above which the parties remain free to negotiate.

104. H3G’s argument also conflicts with the terms on which the disputes were referred by it to OFCOM. In H3G’s letter of 19 March 2007 to the Director of Investigations at OFCOM, H3G set out the scope of the dispute between it and BT, Orange and O2. Under the heading “Remedies Sought” H3G said “Ofcom should resolve the various disputes by determining a reasonable price for the [mobile termination rates] as between (i) BT and H3G (ii) H3G and Orange and (iii) H3G and O2.” In respect of the first, H3G asked that it should be paid the same as Orange’s higher 3G termination rate assuming this was approved for Orange. In respect of (ii) and (iii) H3G mirrored BT’s request that the rate payable by H3G to its competitors for terminating H3G calls on their 3G networks should be no more than the existing 2G regulated rates. In the alternative H3G asked that OFCOM should determine a rate derived from the results of an appropriate implementation of OFCOM’s model for the charges. There was no suggestion here that OFCOM should set a range of rates or that it should be setting a rate which left the parties free to negotiate different, higher prices.

105. The powers that OFCOM can exercise once it has determined a dispute clearly point to it making a definitive ruling as to what the price should be. Section 190(2) of the 2003 Act provides that:

“(2) [OFCOM’s] 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.”

106. These powers, particularly the power in section 190(2)(d), clearly envisage that OFCOM rules on what the price should be and incorporates in that ruling an order adjusting the payments that have been made pending the resolution of the dispute. The Tribunal therefore concludes that it was a proper exercise of its dispute resolution powers in this case for OFCOM to set the price at which mobile call termination was to be sold by H3G and bought by BT.

107. Further, as is apparent from the Tribunal’s judgment in the Termination Rate Dispute appeals being handed down today, the Tribunal considers that OFCOM was wrong to focus to such an extent on the end-to-end connectivity obligation when investigating the dispute between BT and H3G. Dispute resolution is not intended to operate as ancillary to pre-existing regulatory constraints. It is a form of regulation in its own right so that the fixing of the price at the end of the dispute resolution process does amount to a form of price regulation by OFCOM of H3G. This does not mean that the regulation has to be of the same kind as is imposed as a price control condition on a party that is found to have SMP. As the Tribunal said in H3G (1), the dispute resolution power and the power to set SMP conditions exist in parallel. But it is nonetheless a form of regulation.

(b) Is this a form of regulation which falls to be disregarded under the modified greenfield approach in considering whether H3G has SMP?

108. Both parties argued that the Tribunal had decided this point in their favour in the H3G (1) judgment. It is necessary therefore to set out the reasoning of the Tribunal in that case in order to determine precisely what was decided on this point.

109. H3G’s argument in the H3G (1) case was different from the arguments that were put forward in this appeal. In the earlier case H3G argued that, at least in a case where the regulator has based a finding of SMP on the existence of power over price, it is relevant if not important to consider the effect of regulation or possible regulation on the entity in question. Thus one of the main 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. In that case, therefore H3G was

relying on the potential power of OFCOM to set a price control under its SMP powers in order to argue that it was constrained in its pricing and hence did not have SMP.

110. OFCOM’s response to this was described by the Tribunal at paragraph [89] of the judgment as follows:

“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.”

111. Having concluded that they were not bound by authority to accept the proposition, the Tribunal referred to the EC Commission’s decision in Case DE/2005/0144 RegTP decided on 17 May 2005. The Decision was one given pursuant to the Commission’s power to veto decisions taken by the national regulatory authority as to whether an undertaking has SMP (see paragraph [18] above). RegTP was the national regulatory authority in Germany and had provisionally found that 53 fixed network operators other than the incumbent did not have SMP in the market for call termination on their networks. The incumbent fixed network operator, DTAG, was under a regulatory obligation to interconnect with the other operators. The question was whether or not it was right to take this obligation into account in assessing whether DTAG had CBP for the purpose of deciding the extent of the MNOs’ market power.

112. The EC Commission stated at paragraph 22 of its decision:

“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 added).

113. Thus the Commission regarded it as 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.

114. The Commission stated further in paragraph 23 of its decision:

“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.”

115. The Tribunal in H3G (1) went on to conclude:

“98. … 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.

99. 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.”

116. OFCOM and the Interveners relied on these passages as establishing a general principle that all regulation, not just SMP regulation, potentially brought to bear on the party whose market power is being investigated should be disregarded.

117. OFCOM also relied on a passage in the judgment in H3G (1) where the Tribunal was considering whether the existence of clause 13 of the SIA enabling the party who wishes to resist a price change to refer the dispute to OFCOM also meant that there was no SMP. H3G relied on this as a further “regulatory presence” that meant that H3G could not set an excessive price and therefore had no SMP. OFCOM argued that this would amount to enabling an undertaking to contract out of a dominant position by contracting for a third party to resolve disputes. The Tribunal concluded that the clause 13 mechanism did not preclude a finding of SMP. First, they held that the contractual position went to the question of abuse not to the question of SMP. The second reason for rejecting the argument was expressed as follows:

“138. … (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.” (emphasis added)

118. H3G argues that it is clear from the Tribunal’s actual decision in H3G (1) that the Tribunal did not regard dispute resolution powers as subject to the modified greenfield test in the same way as SMP powers. As to paragraphs [98] and [99] of the judgment, H3G argue that these passages should be read in the context of the argument that the Tribunal was rejecting, which was limited to the proposition that prospective SMP regulation itself ruled out a finding of SMP. The Tribunal was not, H3G argued, laying down any wider principle than that.

119. H3G contrasts what the Tribunal said in paragraph 138(b) with the reasons given for remitting the case to OFCOM. The Tribunal held that OFCOM had erred in that although it had correctly (under the modified greenfield approach) considered the effect of the end-to-end connectivity obligation on BT’s

countervailing bargaining power, it had not taken into account the fact that the obligation to grant access was subject to the proviso that the terms and conditions had to be reasonable. OFCOM had therefore treated the obligation to interconnect as entirely removing any negotiating leverage from BT. This approach was wrong, as the Tribunal explained in paragraphs [132] and [142] of the judgment:

“132. … 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.

142. 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.”

120. The Tribunal therefore remitted the decision to OFCOM requiring it to reconsider its determination of SMP taking into account “the extent to which CBP exists in BT”. Why, H3G asks, would the Tribunal have remitted the decision to OFCOM if the Tribunal had decided that the proviso to the end-to-end connectivity obligation was ultimately ineffective as a constraint on H3G’s market power because the dispute resolution powers were to be disregarded?

121. In the Tribunal’s judgment, the H3G (1) case does not decide the point one way or the other. The point at issue in that case was whether potential SMP regulation precluded a finding of SMP and the Tribunal found that it did not. The Tribunal also found that OFCOM had erred in concluding that because BT was effectively bound to connect under its end-to-end connectivity obligation it did not have any CBP. This ignored the important point that the end-to-end connectivity obligation was subject to the “reasonable terms and conditions” proviso. The Tribunal did not decide whether this proviso had the effect of precluding SMP or not. That was precisely the question which was remitted to OFCOM for reinvestigation, the Tribunal making it clear that it was entirely open to OFCOM having examined the point, to conclude that it made no

difference and that H3G had SMP. OFCOM carried out the investigation required by the H3G (1) judgment and arrived at the conclusion that the E2E Proviso did not make any difference. It is the task of this Tribunal to decide now whether that reassessment is correct. We do not therefore regard ourselves as bound by the H3G (1) judgment either to accept or reject H3G’s argument on this point.

122. The Tribunal’s conclusion on this point is that the dispute resolution powers of OFCOM under section 185 of the 2003 Act should be disregarded under the application of the modified greenfield approach. The exercise of OFCOM’s dispute resolution powers is a form of regulation which has the effect of curbing H3G’s exercise of market power even though that may not be its sole or even main aim. If H3G were to propose an abusively high price for mobile call termination whether during initial negotiations for interconnection or during the currency of an interconnection agreement, BT could refuse to accept that price and refer the matter to OFCOM for resolution. As we have found, OFCOM would be required in resolving the dispute to set a price which was below the level which constitutes an abuse of H3G’s dominant position. If BT chose, for its own commercial reasons, to accept an abusively high price, and simply pass the cost on to its own customers, it would be open to OFCOM to intervene in the exercise of its powers under section 105 of the 2003 Act or to open an investigation under Article 82 EC or Chapter II of the 1998 Act. Alternatively a third party, for example a transit customer, who has seen an increase in the price of transit might bring an action under competition law challenging the price set by H3G. All these possibilities are ways in which regulatory intervention could be brought to bear to prevent H3G from charging an abusive price or from exercising its market power in some other way when setting the terms and conditions on which it supplies mobile call termination. It is important not to lose sight of the overall purpose of examining this question which is to determine whether the market for mobile call termination on H3G’s network is effectively competitive within the meaning of the Framework Directive and, if it is not, whether H3G has significant market power. The fact that a company with a large market share is constrained in its pricing decisions by the threat of ex post regulation of one sort or other does not mean that the company is not dominant.

123. We also accept the argument put forward by OFCOM that H3G’s interpretation of the modified greenfield test risks undermining the scheme of the CRF. We do not regard this point as having been made “in terrorem” by OFCOM and the interveners as H3G suggested. It is clear from the Framework and the Access Directives that the process of market review and the imposition of SMP conditions is intended to exist alongside the NRAs’ dispute resolution powers. Reliance on the dispute resolution powers to curb the exercise of market power is not a satisfactory substitute for proper ex ante regulation in the form of price controls in markets which are found not to be effectively competitive because of the existence of SMP. As BT put it “dispute resolution per se is not intended to “cure” existing underlying behavioural or structural imbalances in the market”. Reliance on dispute resolution powers would give rise to obvious uncertainties for the market as a whole and would be disruptive, costly and resource intensive for the participants in the dispute and those further down the supply chain. If it had been intended that SMP could not exist in markets where the parties had the right to refer disputes for resolution, this would have been made clear in the scheme.

124. We do not accept that H3G’s argument that the findings of SMP are “fact sensitive” overcomes this fundamental difficulty. H3G referred to a number of features of the market which, they submitted, distinguished their case from other MNOs in the United Kingdom and which distinguished the United Kingdom markets from the markets in other Member States. These included the regulatory structure, BT’s status as a price sensitive customer and H3G’s status as a new entrant with a small market share. We have considered all the points put forward by H3G in this regard in both their written and oral submissions and concluded that none of these factors overcomes OFCOM’s argument that to regard the potential exercise of an NRA’s dispute resolution powers as constraining market power would be contrary to the scheme for regulation established by the CRF.

125. Both H3G and OFCOM provided the Tribunal with information about the findings of NRAs in other Member States as regards whether a new entrant to the mobile market has SMP in the market for call termination on its own network. This showed that no Member State has thus far finally concluded that a new

entrant in a similar position to H3G in the United Kingdom does not have SMP in Market 16. No one has pointed us to any indication from the EC Commission during the consultations required by the Framework Directive that it regards the findings by the NRAs as unsatisfactory. However, we accept that the position is still fluid in a number of important Member States – even though this fluidity is in part generated by challenges to findings of SMP by companies within the Hutchison group itself – and that it would not be safe for the Tribunal to rely on this material to conclude that H3G’s argument must be wrong. Although we do not regard this material as supporting H3G’s case, we do not rely on it in rejecting its argument either.

(x) The initial negotiations

126. H3G alleged that OFCOM had erred in failing to appreciate that a proper analysis of the initial negotiations between H3G and BT when the first interconnection agreement was signed showed that H3G did not have market power. The initial negotiations between BT and H3G resulted in an SIA being signed between them on 13 August 2001. However, the price provisions in the agreement were not settled at that stage and further negotiations took place in the latter part of 2001 and early in 2002. In its Notice of Appeal H3G relied on the initial negotiations only in respect of the finding of SMP in the Reassessment Statement. But at the hearing, H3G confirmed that they relied on the negotiations in respect of the finding of SMP both in the Reassessment Statement and in the 2007 Statement.

127. H3G argued that the contemporaneous material before OFCOM and the Tribunal shows that when H3G first approached BT to conclude an interconnection agreement, they were under considerable time pressure because the agreement had to be in place in time for H3G to be able to conduct its planned tests of the network. H3G felt constrained to accept a price which was too low and which may actually have been below H3G’s efficiently incurred costs at the time. H3G also relied on the content of a letter to H3G from BT dated 11 August 2006 in which BT referred to its “commercial position” as being that it was not willing to pay more for 3G termination than it paid for 2G termination. As to the relevance of the initial negotiations to the question that OFCOM, and now the Tribunal,

has to address, H3G argued that the issue is whether, and to what extent, what happened in those negotiations “produced continuing effects” on the balance of market power between H3G and BT. In fact that price struck at that point remained in effect during the whole period covered by the 2004 Statement (apart from the increase to 16.6 ppm towards the end of that period). This means, H3G argues, that H3G should not be treated as having SMP during that whole period.

128. BT strongly contests H3G’s description of the events that led up to the setting of H3G’s price at launch and argues that the Tribunal has already made findings in the H3G (1) judgment effectively rejecting H3G’s version of events. More significantly BT and OFCOM both argue that the Tribunal’s judgment in the earlier appeal shows that H3G’s argument as to the relevance of the initial negotiations is wrong.

129. It is common ground that the question for the Tribunal to decide is not whether H3G had SMP at the time of the initial negotiations. Even if the OFCOM had concluded that there was no SMP at that time, it is still necessary to consider how that affects the question whether they had SMP in 2004. This was made clear in the judgment in H3G (1) where the Tribunal described the relevance of the material relating to the initial negotiations in the following terms (paragraph [74] of the judgment):

“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 … 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 …).”

130. In the later part of the judgment the Tribunal explained why it was remitting the 2004 Statement to OFCOM on this point:

“140. 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.”

131. The Tribunal set out various factors arising from the evidence and continued:

“141. 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.”

132. We accept the point made by OFCOM that its reconsideration of the relevance of the initial negotiations was limited to considering whether they cast any light on how the parties would negotiate if a situation arose in the course of the period of the price control. We agree with OFCOM that the initial negotiations are not relevant because the circumstance which might have given BT bargaining power in 2001 – namely the fact that H3G was anxious to avoid a delay to the start of its service – was no longer operative by the start of the period covered by the 2004 Statement. The Tribunal therefore concludes that even if the evidence established that BT had CBP at the time of the initial negotiations, this did not mean that it still had CBP during the period 2004-2007 or the period 2007-2011.

(xi) Excessive pricing by H3G

133. Having concluded that OFCOM was right to reject H3G’s argument that BT had countervailing buyer power sufficient to counteract the market power that H3G could exercise as a result of its market share, was OFCOM also correct in concluding that H3G had SMP in the market for call termination on its network?

134. There were two additional points made by H3G. H3G alleged that OFCOM erred in law by equating any price other than a strictly cost based price with a price which is appreciably above the competitive level. This point arose from paragraph 5.154 of the 2007 Statement in which OFCOM was referring to the likely outcome of the reference of a dispute as to what was a reasonable charge for the purposes of BT’s end-to-end connectivity obligation. OFCOM said:

“On this basis, a reasonable charge for BT to purchase MCT with a view to ensuring end-to-end connectivity may be at a price appreciably above the competitive level. As such, if a charge appreciably above the competitive level were in dispute, Ofcom considers it unlikely that it would insist on a strictly cost based charge (such as used in deriving cost benchmarks … to set efficient regulated charges in the charge control) ie a charge that was not appreciably above the competitive level”.

135. It emerged from OFCOM’s Defence that the paragraph was not intended to indicate that OFCOM regarded any price other than a strictly cost based price as being a price “appreciably above the competitive level” and hence an indicator of SMP: see para 102 of the Defence. OFCOM accepted therefore that a dominant undertaking does not necessarily abuse its dominant position by charging a price which is not a cost based price. This is indeed clear from other passages in the 2007 Statement such as paragraph 4.14 where OFCOM denies that it has equated SMP with the ability to persistently raise prices above costs by any amount.

136. Secondly, the Tribunal heard argument as to whether H3G had in fact imposed excessive prices for MCT in the past and whether this was an alternative basis on which OFCOM could found its finding of SMP.

137. OFCOM’s findings in relation to excessive pricing in the 2007 Statement were that the 2G/3G MNOs have previously (that is, before these charges were regulated by the 2004 Statement) sustained 2G termination charges significantly above a reasonable estimate of costs. Further, the underlying 3G charges proposed by the 2G/3G MNOs were substantially greater than OFCOM’s estimate of costs (in some cases more than double). OFCOM also noted that “The underlying charges of three of the 2G/3G MNOs are also substantially greater than H3G’s charges”: see paragraph 4.64 of the 2007 Statement.

138. It appears from this that whatever may have been the case with regard to the 2G/3G MNOs, OFCOM was not relying on the level of H3G’s prices as evidence supporting the finding of SMP on the part of H3G. OFCOM emphasised that although evidence that MNOs are able to sustain charges to an appreciable extent above the competitive level supports the view that the MNOs have SMP, it is not a prerequisite to a finding of SMP (see paragraphs 4.45 and 5.173 of the 2007 Statement).

139. The consideration of the reasonableness or otherwise of H3G’s prices is relevant to another aspect of this case, namely whether there was a risk, for the purposes of section 88 of the 2003 Act, that H3G would set prices at an excessively high level. It is convenient therefore to deal with pricing issues in that context rather than here since OFCOM did not purport to rely on evidence of excessive pricing by H3G in support of its finding that H3G has SMP.

(xii) The Tribunal’s conclusion on SMP

140. The Tribunal’s conclusions on the existence of SMP on the part of H3G in the market for mobile call termination on H3G’s network can be summarised as follows:

(a) OFCOM was correct in concluding that the availability of its dispute resolution powers under section 185 of the 2003 Act did not constrain H3G’s market power to a degree sufficient to preclude a finding of SMP because:

i. although OFCOM must not, when setting a price in the exercise of its dispute resolution powers, set that price at a level which amounts to an abuse of a dominant position, it might, depending on the circumstances, set a price which is appreciably above the competitive level;

ii. because OFCOM could set a price appreciably above the competitive level, the existence of the dispute resolution mechanism, and the parties’ expectations as to how OFCOM will exercise its powers, do

not mean that BT has sufficient countervailing buyer power to preclude a finding of SMP on the part of H3G;

iii. this is the case whether or not it would be ultra vires for OFCOM to set a price which could be regarded as an “excessively high price” in the context of the test set out in section 88(3) of the 2003 Act;

iv. in any event, OFCOM’s powers of dispute resolution constitute a form of price regulation on H3G which falls to be disregarded when assessing H3G’s market power, under the modified greenfield approach described by the Tribunal in the judgment in H3G (1);

(b) OFCOM was correct to conclude that the evidence of the initial negotiations between BT and H3G in 2001/02 did not indicate that BT had sufficient CBP during the period covered by the Reassessment Statement or the 2007 Statement to counteract H3G’s market power;

(c) OFCOM did not base its conclusion on H3G’s SMP on any findings that H3G had in the past charged a price appreciably above the competitive rate for mobile call termination;

(d) OFCOM was right to rely on other factors such as BT’s regulatory obligations regarding Carrier Pre-Selection and Indirect Access and more general commercial considerations which might weaken BT’s CBP;

(e) OFCOM was therefore entitled to conclude that H3G had and has SMP because of its 100 per cent market share, the existence of absolute barriers to entry and the absence of sufficient countervailing buyer power on the part of its main customer BT.

141. The Tribunal therefore dismisses H3G’s appeal in so far as it challenges both the Reassessment Statement and the finding of SMP in the 2007 Statement.

VI. THE DECISION TO IMPOSE A PRICE CONTROL REMEDY

142. The second main challenge by H3G to the 2007 Statement was its assertion that OFCOM was wrong to impose a price control remedy on H3G – the Appropriate Remedy Issue.

143. The 2007 Statement set a number of conditions for the MNOs including two conditions relating to mobile call termination rates, Condition MA3 relating to fixed-to-mobile interconnection charges and Condition MA4 relating to mobile-to-mobile interconnection charges. These Conditions control the average charges which the MNOs may levy. The price control was set for all mobile call termination without distinguishing in terms of price between termination on a 2G network and termination on a 3G network. Thus for the first time, termination on 3G networks was regulated through an SMP price control condition and H3G was subject to a price control along with the 2G/3G MNOs who had previously been regulated under SMP conditions as to their charges for 2G termination but not for their 3G termination.

144. The Tribunal’s task is to consider whether OFCOM was wrong to exercise its power under section 88 to impose a price control remedy on H3G. The question of whether the price control in fact set by OFCOM is the right one is a specified price control matter which must be determined by the Competition Commission. On 18 March 2008, the Tribunal referred the specified price control matters in this appeal to the Competition Commission and those matters are in the course of being investigated in that forum.

145. It is also important to note that H3G is not arguing that MCT charges should be unregulated generally. On the contrary, part of the H3G appeal seeks the imposition of a stricter price control on the four 2G/3G MNOs. H3G does not challenge the inclusion of 3G termination as supplied by the 2G/3G MNOs in the price control set for those operators. Rather H3G’s case is that because of its unique position in the market, it should be allowed to negotiate a price with its customers unconstrained by the kind of price control that OFCOM imposed.

(i) The regulatory background

146. Before examining H3G’s arguments on this aspect of the case it is necessary to describe the scope of OFCOM’s powers in setting a price control and some other aspects of the industry which are relevant to the grounds of appeal, in particular the current arrangements for mobile number portability.

(a) The setting of SMP Conditions

147. Section 45 of the 2003 Act provides for OFCOM’s powers to set conditions. Section 45(2)(b) provides for the imposition, among other things, of “an access related condition” and an SMP condition. Section 47(1) of the 2003 Act provides as follows:

“Ofcom must not, in exercise or performance of any power or duty under this Chapter –

(a) set a condition under section 45, or

(b) modify such a condition,

unless they are satisfied that the condition or (as the case may be) the modification satisfies the test in subsection (2).”

148. Section 47(2) of the 2003 Act provides that:

“That test is that the condition or modification is –

(a) objectively justifiable in relation to the networks, services, facilities, apparatus or directories to which it relates;

(b) not such as to discriminate unduly against particular persons or against a particular description of persons;

(d) in relation to what it is intended to achieve, transparent.”

149. The power to impose SMP conditions is provided in sections 87 to 92 of the 2003 Act. Before an SMP condition may be imposed, there must be a finding of SMP made in respect of the undertaking concerned (see sections 46(7) and (8)). Section 87 makes provision for the particular conditions which may be imposed by OFCOM on an operator which it has determined has SMP and section 87(9)

confers the power to set a price control condition. Sections 87(9) and 88 have been set out earlier (see paragraph [20]-[21] above) but since section 88 is key to understanding this part of the appeal we set it out here again:

“88 Conditions about network access pricing etc.

(1) OFCOM are not to set an SMP condition falling within section 87(9) except where-(a) it appears to them from the market analysis carried out for the

purpose of setting that condition that there is a relevant risk of adverse effects arising from price distortion; and

(b) it also appears to them that the setting of the condition is appropriate for the purposes of-(i) promoting efficiency; (ii) promoting sustainable competition; and

(iii) conferring the greatest possible benefits on the end-users of public electronic communications services.

level, or

(b) so impose a price squeeze,

as to have adverse consequences for end-users of public electronic communications services.

(4) In considering the matters mentioned in subsection (1)(b) OFCOM may-

(a) have regard to the prices at which services are available in comparable competitive markets;

(b) determine what they consider to represent efficiency by using such cost accounting methods as they think fit.

…”

150. H3G characterised these provisions as imposing stringent pre-conditions for the imposition of a price control condition because price control is a highly intrusive form of regulation which may have unintended adverse consequences, both for

the company being regulated and, more generally, for competition. H3G stressed that the implications for a business of not being able to set the price for its own service over a four year period in a dynamic and fast-changing market are very severe and we do not understand that OFCOM disagreed with this description of the statutory scheme.

(b) Mobile Number Portability

151. Mobile number portability refers to the process whereby someone who decides to move from one mobile phone service provider to another can choose to take their phone number with them to the new network. Where a customer of MNO “A” (“the donor network”) wishes to transfer to MNO “B” (“the recipient network”), the recipient network will advise the customer to contact the donor network in order to obtain from it a Porting Authorisation Code (PAC). The Code will normally be provided immediately over the phone but will be confirmed later in writing within two working days. The customer will pass that PAC to the recipient network. That network will then enter the PAC and the phone number into the industry porting system. From this point in the process the donor network has five working days to complete the port. The customer is not without a mobile service over these five days. They can either continue to use their existing mobile connection for a further five days or they may be given a temporary number by their new MNO. Thereafter the calls to the number are automatically received on the new handset.

152. Under the current system of routing calls, a call to a ported number will initially be routed to the donor network not to the recipient network. This is because the phone numbers were allocated in blocks to the networks and so are recorded by the industry as “belonging” to that network. The donor network will recognise the number as one that has been ported and will route it to the customer’s new recipient network. It is the donor network rather than the recipient network whose mobile call termination rate is charged to the network from which the call is being made (“the originating network”). The originating network therefore pays the donor network for the termination at the donor network’s mobile call termination rate and the donor network passes this money onto the recipient network minus a deduction known as the donor conveyance charge to

compensate the donor network for its costs of routing the call on to the recipient network. This means that where a customer moves to H3G from one of the other MNOs and ports their number, H3G only receives the balance of the other MNO’s mobile call termination charge (after deduction of the donor conveyance charge) rather than its own mobile call termination charge.

153. Where a customer switches to a new network without porting their number, the customer will be given a new number by the new network, calls will be routed directly from the originating network to the new network and the mobile call termination charge will be imposed by the network at its own mobile call termination rate. Under the current arrangements, it is therefore less advantageous for H3G if a customer moving to its network decides to port its number from its old network and conversely, under the current arrangements, the 2G/3G MNOs have a particular incentive to persuade H3G subscribers to port their numbers when they move as they then get the benefit of H3G’s higher termination charges. H3G charges and receives its higher MCT rate both on calls terminated on its 3G network and on calls terminated on the 2G network of its roaming partner.

154. OFCOM has carried out a review of the UK’s mobile number portability arrangements. On 29 November 2007 OFCOM issued a Statement entitled Telephone number portability for consumers switching suppliers – Concluding Statement setting out its decision as to how the system should be changed. The current system is to be replaced by a new direct routing system by September 2009. The time taken to port numbers under the current system must be reduced to two working days by April 2008 at the latest and further reduced to less than two hours by 1 September 2009. On 29 January 2008 Vodafone lodged an appeal against that statement with the Tribunal. Although H3G welcomed the Mobile Number Portability Statement as at least to some extent a positive development H3G argues that it will only be some considerable time after direct routing has been implemented (perhaps two years or longer) when H3G and the market sees the full benefit of improved mobile number portability.

(ii) OFCOM’s decision to impose a price control on H3G

155. Four sections of the 2007 Statement focus on the issues concerning the appropriateness of setting a price control. In section 6 OFCOM sets out its duties and objectives. Section 7 sets out the benefits of regulating MCT charges and discusses the issues arising if those charges remain unregulated. Section 7 also describes the welfare analysis that OFCOM carried out to identify any benefits from regulation, further details of this analysis are set out in Annex 19 to the Statement. Section 8 discusses the different regulatory remedies OFCOM has considered and section 9 discusses the detail of the price control that OFCOM has decided to impose. Further information about the cost modelling that has been used in setting the rates is set out in the various Annexes to the Statement.

156. For our purposes sections 6 to 8 are the most relevant. In section 6 having set out its statutory duties under sections 3 and 4 of the 2003 Act, OFCOM describes the relevant considerations to be borne in mind when deciding on appropriate remedies -

(a) seeking to promote the interests of consumers by ensuring prices are not excessive and resources are efficiently allocated;

(b) ensuring technological neutrality and avoiding regulatory distortion of MNO decisions about delivery of mobile termination – seeking to ensure MNOs’ incentives to use one technology (for example, 2G) over another (for example, 3G) are not distorted by regulation;

(c) encouraging investment and innovation in existing and new mobile services – seeking to ensure operators recover efficiently incurred costs; and

(d) ensuring competitive neutrality and avoiding economic distortions, for example in the downstream retail market.

157. In section 7 OFCOM describes the detrimental impact of excessive mobile call termination charges which it considers under five headings:

(a) excessive prices overall. Generally speaking, if MNOs set excessive prices for mobile call termination they may be able to earn excess profits at the expense of consumers. However this possibility is offset to a considerable extent in this sector because of the “waterbed effect”, a term used to describe the fact that, to some extent, the pressures of competition in the retail market may lead MNOs to use the profits they make, including profits from terminating calls, to offer more attractive benefits to consumers on that retail market. Although OFCOM doubted whether the waterbed effect was complete (that is, whether there would be sufficient competitive incentive for the MNOs to pass on all the profits they make from MCT to consumers by reducing retail prices) OFCOM concluded that it would not rely heavily on the need to avoid the risk of excessive pricing when deciding what conditions should be imposed.

(b) Inefficient structure of prices. OFCOM concluded that even if the waterbed effect was complete, leaving MCT charges unregulated would lead to an inefficient structure of prices in retail and wholesale markets and this would have a detrimental impact on consumers. The structure of prices would lead to over-consumption of mobile retail services and under consumption of other services provided by operators which pay mobile call termination charges, for example, fixed retail services.

(c) Distortion of consumer choice. Excessive termination charges feed through to higher retail prices for fixed-to-mobile and mobile-to-mobile calls between networks. However on-net mobile calls (that is between two subscribers on the same network) incur no explicit termination charge and mobile-to-fixed call termination is charged at a regulated rate. Excessive mobile call termination charges also enable reductions in the prices of mobile retail services. OFCOM concluded that consumer choices would be distorted between mobile and fixed calling due to distortions in the

relative prices of fixed and mobile services as the relative prices do not reflect the underlying resource costs.

(d) Inequitable distributional effects. As mobile call termination charges are a major component of the price of calls to mobiles, callers to mobiles may face excessive prices for some services and lower prices for others. Although mobile subscribers may benefit from having the high profits passed through to them in the form of lower mobile retail prices, the consumers who benefit are not necessarily the same people as the ones who are paying the extra charge, particularly the fixed network subscribers. The overall effect is likely, OFCOM concludes, to be detrimental to consumers.

(e) Risk of anti-competitive behaviour. The ability to set excessive charges for mobile call termination could also be used to distort and reduce competition in retail mobile services. In particular OFCOM identified a potential for anticompetitive pricing by the larger MNOs to the detriment of smaller MNOs and therefore to the detriment of competition. If unregulated, the MNOs could charge a lower mobile call termination rate to each other than they charge to the smaller network (although the smaller network could avoid this problem by using BT to transit its calls to the MNOs). However, to the extent that a reduction in mobile call termination charges lessens the risk of such behaviour, OFCOM considered that it was a further benefit to be derived from charge controls.

158. In section 8 of the 2007 Statement OFCOM set out the various regulatory options available including reliance on ex post intervention under competition law, imposing a condition that charges should be “fair and reasonable” and imposing a transparency obligation. OFCOM concluded that a direct charge control was the most cost effective option for constraining MNOs with SMP from setting excessive charges to the detriment of consumers. The price control was set for the period from 1 April 2007 until 31 March 2011. OFCOM concluded that the maximum average charge (referred to in the Conditions as the Target Average Charge or “TAC”) for the 2G/3G MNOs during the fourth year

of the charge control period should be 5.1ppm expressed at 2006/07 prices. The TAC set for H3G for the fourth year of the charge is 5.9ppm.

159. As well as setting the TAC which the MNOs must ultimately achieve, the Conditions also set the “glide path” controlling the maximum average price that the MNOs can charge in each of the first three years of the control period. By following annual reductions set by the glide path, the MNO arrives at the final TAC. For the four 2G/3G MNOs, OFCOM concluded that the charges should be reduced in four broadly equal percentage steps starting from the regulated 2G rate which applied in 2006/7 pursuant to the price control in the extended 2004 Statement.⁵ In the case of H3G, OFCOM applied a different glide path, requiring H3G to reduce its charges in the first year to an average of 8.5 ppm (at 2006/7 prices).

160. The two Conditions MA3 and MA4 are intended to set the same charges, the purpose of having two charges being simply to ensure that, because MNOs have the scope within the average set by the price control to set different charges for mobile call termination they should not be able to charge relatively high charges for terminating fixed-to-mobile calls to offset low charges for mobile-to-mobile.

(iii) The test for the Tribunal to apply

161. Part of H3G’s challenge to the 2007 Statement as regards the Appropriate Remedy Issue was expressed in terms that arguments put forward by H3G to OFCOM in response to the consultation document were not adequately investigated by OFCOM and that the reasons set out in the 2007 Statement for rejecting them were not explained in sufficient detail. In taking us to the passages in the 2007 Statement where OFCOM had dealt with H3G’s objections to the proposed price control, H3G argued that OFCOM should have undertaken much more analysis and investigation and should have described this process with fuller reasons for arriving at its conclusion that H3G’s point was unfounded.

⁵ The actual first year rates were adjusted to take account of the fact that the new capped rates came into effect part way into the first year.

162. Market review is already a complex and lengthy task. The current review took almost 22 months to complete and the 2007 Statement stretches to over 900 paragraphs plus a further 16 Annexes. Much of the 2007 Statement is devoted to issues which were hard fought by the MNOs during the consultation period but which are not part of this appeal, in particular market definition. Although H3G is clearly a major “stakeholder” in the review, if OFCOM were required to carry out all the investigations that the principal parties urge upon it and explain at length its reasoning for rejecting each of the points put forward by H3G at the level of detail H3G now appears to expect, the task of carrying out these market reviews would become unmanageable. In the circumstances of this review, therefore, provided that OFCOM summarises the main points put to it by the principal participants and states briefly why it accepts or rejects the point, it should not ordinarily be criticised for lack of analysis.

163. It is apparent from the 2007 Statement that OFCOM did consider the arguments put forward by H3G to the effect that it should be treated differently from the other 2G/3G MNOs and it accepted them to the extent that gave rise to the different Target Average Charge in the price control and to the different glide path. OFCOM did not consider that the differences justified leaving H3G unregulated.

164. However, this is an appeal on the merits and the Tribunal is not concerned solely with whether the 2007 Statement is adequately reasoned but also with whether those reasons are correct. The Tribunal accepts the point made by H3G in their Reply on the SMP and Appropriate Remedy issues that it is a specialist court designed to be able to scrutinise the detail of regulatory decisions in a profound and rigorous manner. The question for the Tribunal is not whether the decision to impose a price control was within the range of reasonable responses but whether the decision was the right one.

(iv) H3G’s case on the price control remedy

165. H3G’s Notice of Appeal (as amended) and its skeleton argument set out the background facts, as H3G see them, as well as setting out the more formal

grounds of appeal. It became apparent during the course of the appeal that the grounds of appeal relating to the Appropriate Remedy Issue overlap to a certain extent and the factual material presented by H3G was relied on as relevant to more than one ground of appeal. For example, H3G’s argument that the welfare analysis carried out by OFCOM was seriously flawed was relevant to its submissions that the price control was disproportionate because, on a proper analysis, the welfare gains from regulating H3G were very small. It was also relied on in support of the allegation that OFCOM had erred in its assessment of its duty to maximise the benefit to end users for the purposes of section 88(1)(b)(iii) of the 2003 Act. Similarly, evidence aimed at showing that H3G was in a different position on the market from the other MNOs is relied on as relevant both to the allegation that OFCOM erred in failing to apply the statutory tests to H3G individually and to the allegation of discrimination contrary to section 47(2)(b) of the 2003 Act.

166. To avoid repeating the arguments raised, this judgment deals with the points in the order which appears to the Tribunal to be the most logical, noting that the Tribunal’s conclusions on each point hold good regardless of the legal context in which the point is raised.

167. In outline, H3G’s arguments can be summarised as follows. First, H3G alleges that OFCOM erred in its overall approach to the question posed by sections 87 and 88 of the 2003 Act whether to impose a price control on H3G. OFCOM considered the benefits and detriments arising from regulation of the industry as a whole rather than the benefits and detriments arising from regulating H3G in addition to regulating the 2G/3G MNOs. This meant that the decision to impose a price control on H3G was “swept up” in the general “regulation versus no regulation” decision.

168. Secondly, H3G alleges more specific errors on the part of OFCOM in its application of the statutory provisions. The imposition of a price control was disproportionate contrary to section 47(2)(c) of the 2003 Act because OFCOM-

(a) compared the position in which all the MNOs were regulated by a price control with the position where none of the MNOs was regulated whereas it should have considered the continuation of the current state of regulation as the primary counterfactual;

(b) greatly exaggerated the benefits to be gained from regulating H3G; properly measured, the benefits of regulating H3G in addition to the other MNOs are rather insignificant;

(c) failed to take into account the substantial detriments which arise from the price control which deprives H3G of the freedom to set its mobile call termination rates at a level which enables it -

i. to act as the “maverick” competitor in the retail market competing vigorously by offering attractive retail packages to customers; and

ii. to recoup the loss of funds which arises because of the system of charging for mobile call termination for ported numbers;

(d) failed to take into account the serious adverse effect the price control would have on H3G’s financial position both in absolute terms (because H3G’s business in the United Kingdom is not yet profitable) and relative to its competitors – the price control was disproportionate in the basic sense that the cost to H3G outweighs any benefits that are gained from regulation.

169. H3G also alleges that the decision to impose a price control was discriminatory in the sense that it treated different cases alike. OFCOM failed to appreciate that H3G should be treated differently from the other MNOs because -

(a) H3G is a new entrant coming into a market which is already “mature and saturated” so that it can only compete with the established MNOs by attracting their customers away from them, not by attracting new users;

(b) H3G has a much smaller market share in terms of subscribers – this places H3G at a considerable disadvantage because it is a net payer of mobile call termination charges whereas for the other MNOs who have roughly equal market shares, their payments of each other’s mobile call termination charges are largely cancelled out by the payments they receive;

(c) H3G suffers from a traffic imbalance, which means that its subscribers make a much larger number of calls to other networks (for which H3G has to pay the mobile call termination charge) than the number of off-net calls they receive (for which H3G is paid its charge);

(d) further, the disadvantages identified in (a), (b) and (c) above are caused or aggravated by the distortion of the competitive structure of the market produced by the current arrangements for mobile number portability;

(e) in rejecting H3G’s arguments in favour of leaving it unregulated, OFCOM ignored academic literature and decisions from other Member States indicating that asymmetric regulation in this market may be advantageous – OFCOM wrongly based its reasoning (expressly or impliedly) on an assumption that asymmetric regulation was undesirable.

170. Finally, H3G alleged that OFCOM wrongly concluded that there was a relevant risk of adverse price distortion for the purposes of section 88(1)(a) of the 2003 Act. H3G argues that on a proper analysis of the factual background, there was no evidence to support a finding that there was a risk that H3G would fix or maintain its mobile call termination charges at an excessively high level so as to have adverse consequences for end users.

171. H3G also argued that the imposition of a price control was not appropriate for the purposes of promoting efficiency and sustainable competition or conferring the greatest possible benefit on end users for the purposes of section 88(1)(b) of the 2003 Act because the price control exacerbates the distortions already apparent in the market caused by the unsatisfactory mobile number portability arrangements. The points that were relied on in relation to this ground were also

relied on in relation to one or more of the other grounds and are dealt with in this judgment in the consideration of those other grounds.

172. In support of their arguments on this part of their appeal, H3G relied on the expert evidence of Dr Stephen Littlechild, Emeritus Professor of Commerce at the University of Birmingham and Director General of Electricity Supply between 1989 and 1998. Dr Littlechild is the author of a number of papers on price controls on mobile call termination charges and was an adviser to the Secretary of State for Industry on the privatisation of BT. By the time of the hearing, three witness statements had been lodged by Dr Littlechild. At the opening of his oral evidence he confirmed that he was not arguing a positive case that H3G should not be subject to a price control, rather his primary submission was that the arguments for a price control are not sufficiently strong to justify that as the obvious conclusion. In particular, he contends, OFCOM did not properly assess the disbenefits of regulating H3G.

173. In response, OFCOM filed two witness statements from Mr Geoffrey Myers who is employed by OFCOM as a Director of Competition Economics in the Competition Group. Mr Myers holds MA and MPhil degrees in Economics and has worked as a professional economist in the public sector for sixteen years, eleven of them with OFCOM. T-Mobile also lodged an expert report by Dr Mike Walker of CRA International which is an economic consultancy established in the USA. He is a D.Phil in Economics and holds several academic posts relating to the economics of competition law.

174. H3G submitted that the Tribunal should prefer the evidence of Dr Littlechild to that of Mr Myers or Dr Walker. As regards Mr Myers, H3G said that he cannot be treated as an independent expert because self evidently he is a senior employee of OFCOM and he was involved in the taking of this decision. There was no suggestion that Mr Myers was acting in bad faith but his evidence could not be treated as the evidence of an independent expert. H3G was more critical of Dr Walker whom they described as “an advocate for T-Mobile”. H3G gave examples where it alleges Dr Walker refused to accept propositions which might be adverse to the interests of T-Mobile even when it was plain that the position

he was maintaining was unsustainable and where he was “not as neutral as he might have been”.

175. The Tribunal considers that these criticisms are unfounded. All three witnesses were, in the Tribunal’s judgment, doing their best to assist the Tribunal and were aware of their duty to the Tribunal. We do not consider that it was inappropriate for OFCOM in this case to deploy an in-house economist as an expert. Although Dr Walker’s oral evidence was delivered in a rather different style from that of Dr Littlechild or Mr Myers we did not regard it as inappropriate or gain the impression that he was acting as an advocate rather than an expert. Further, the Tribunal in this case is well able to assess the merits and demerits of the arguments that the experts were debating. The Tribunal does not therefore distinguish between the value of the evidence in the manner proposed by H3G.

(v) OFCOM’s alleged error in applying the statutory tests to the market as a whole rather than to H3G individually

176. H3G’s first argument was that OFCOM failed to apply the tests individually as required by the legislation.

177. H3G’s arguments to OFCOM in response to the consultation documents are summarised by OFCOM in paragraph s 7.6 and 7.7 of the 2007 Statement:

7.6 H3G argued that Ofcom’s proposals were inconsistent with its statutory duty to promote competition, to further the interests of citizens by ensuring that high speed mobile data services are made available across the UK, and to encourage investment and innovation. In H3G’s view, Ofcom’s proposals would reduce competition, to the detriment of consumers. H3G also argued that, as H3G currently has only a 5% share of active UK subscribers, the impact of any reduction in MCT charges on callers to H3G’s network will be small, and unlikely to balance the negative impacts on competition which H3G envisages flowing from reduced MCT charges. In H3G’s view the impact on consumers of reductions in H3G’s MCT charges will also be further diluted by MNOs’ practice of offering bundles of “any network” calls, and also by the practice of FNOs not directly reflecting differences in MCT charges in their retail prices for calls to mobiles.

7.7 H3G claimed that Ofcom had failed to give proper consideration to these issues and had failed to conduct a proper cost-benefit analysis which assesses the relative size of these effects, particularly when considering the merits of the three glide path options set out in respect of H3G. H3G also argued that Ofcom’s welfare analysis, being based on a global analysis of all MNOs, fails to consider the specific position of H3G as a 3G-only MNO. H3G noted, however, that

Ofcom’s analysis indicated that there are still significant welfare gains to be made from reducing 2G rates, particularly given the relative volume of calls.”

178. OFCOM’s response was recorded at paragraph 7.28:

“Ofcom is not persuaded by H3G’s argument that, because it has relatively few subscribers, regulation will be of limited net benefit. First, H3G will grow over the period of the control and Ofcom does not consider that the consequent volume of terminated minutes will be insignificant. Second, Ofcom’s analysis of the financial effect on H3G at paragraphs 9.204 et seq. suggests that the proposed remedy will not undermine H3G’s overall financial position and, moreover, H3G’s reduced MCT termination revenue would be small compared to its overall revenues. Therefore, for the reasons argued here and in section 9 Ofcom considers that there are material overall welfare gains from the regulation of H3G.”

179. O2 argued in its Statement of Intervention that although H3G describes this ground of appeal as an error of law, its complaint in reality is that H3G’s arguments that it should be permitted to set unregulated mobile call termination charges due to its specific situation were rejected by OFCOM. There is some force in that argument. OFCOM acknowledged in the 2007 Statement that H3G had argued that it should be treated differently from the other MNOs and OFCOM clearly rejected those arguments. It is not right therefore to say that OFCOM simply assumed that H3G should be treated in the same way as the other MNOs without considering whether the differences between H3G and the others meant that H3G should remain unregulated. OFCOM did consider this and concluded that H3G should be subject to a price control, albeit one which was different from the control imposed on the other MNOs. It is for the Tribunal now to decide whether OFCOM was right to arrive at that conclusion by considering the particular points raised by H3G.

(vi) Was the imposition of a price control disproportionate?

180. H3G’s second argument on the Appropriate Remedy Issue is that the imposition of a price control was disproportionate and therefore contrary to section 47 of the 2003 Act. That section (set out in paragraphs [147]-[148] above) provides that OFCOM must not exercise its powers to set a price control condition unless it is satisfied that the condition is proportionate to what the condition is intended to achieve.

(a) The correct counterfactual for assessing the benefits of price control

181. H3G argues that OFCOM erred in comparing the consequences of regulating all the MNOs with the consequences of not regulating any of them. The correct comparison should be between the consequences of regulating the 2G/3G MNOs and the consequences of regulating all five of them. H3G referred us to OFCOM’s publication Better Policy Making: OFCOM’s approach to Impact Assessment published in July 2005. That document states that one of OFCOM’s key regulatory principles is that it has a bias against intervention and that it aims to choose the least intrusive means of achieving its objectives. In identifying options, the document states, OFCOM will consider a wide range of options including not regulating.

182. H3G relies in particular on paragraph 3.3 of the Better Policy Making guidance which describes the process of carrying out an impact assessment:

“At the outset we should identify the issue to be addressed and the options available to us. In doing so, we should continue to bear in mind the need for options to be linked with our statutory duties. We will start by considering the option of not changing the regulatory framework, either by not introducing regulation or by retaining existing regulation. This option – no new intervention – will generally be the benchmark against which other options are judged i.e. what costs and benefits would be incurred additional to those which would be incurred if there were no new intervention?” (emphasis added)

183. H3G argues that OFCOM’s approach in the 2007 Statement was contrary to this guidance. OFCOM considered the benefits of regulating all five MNOs against the counterfactual of not regulating any of them. But the existing regulation which OFCOM should have considered retaining and which should have been the benchmark against which the regulation of all five should have been judged was the regulation imposed by the 2004 Statement.

184. The Tribunal does not accept H3G’s analysis on this point. First the regulatory position which H3G was urging OFCOM to adopt in the 2007 review is significantly different from the position arrived at in the 2004 Statement. The result of the 2004 Statement was not simply that H3G was not subject to a price control and the other MNOs were, but that no 3G termination was subject to a price control. It is true that at the time, H3G was the only MNO offering 3G

termination and also that the 2004 Statement did not impose a price control on H3G in respect of its termination of calls on the 2G networks of its roaming partner. But over the period covered by the 2004 Statement the 2G/3G MNOs launched their 3G termination services and imposed or proposed to impose charges for them. It was clear by the time OFCOM was consulting on the 2007 market review that the number of 3G subscribers had increased considerably to several million and that H3G had a substantial share of those subscribers. It was no longer realistic to leave 3G termination free from price control. OFCOM’s function in this review was to determine how termination on 3G networks should be treated. H3G does not dispute that it is appropriate now to control the 3G termination prices of the other MNOs. Once it had become clear that prices for 3G call termination should be regulated for the 2G/3G MNOs, the maintenance of the 2004 Statement position was untenable.

185. Secondly the Tribunal agrees with OFCOM that once a particular regulatory control expires, the status quo established by that control has no better claim to legitimacy for the future than any other proposed regime in circumstances where the market conditions have changed substantially between the date that the earlier regulatory regime was imposed and the date when the matter is being reconsidered.

186. We therefore conclude that OFCOM was right in this instance to consider the matter afresh and not to approach its task by considering first whether the regulatory position under the 2004 Statement should be maintained.

(b) OFCOM’s assessment of the benefits of regulating the 2G/3G MNOs and H3G

187. H3G argues that OFCOM has overstated the benefits to be gained from regulating H3G. OFCOM’s analysis of welfare gains in the 2007 Statement was in part qualitative and in part quantitative. OFCOM identified five possible detriments that could be avoided by regulation; excessive prices overall, inefficient structure of prices, distortion of consumer choice, inequitable distributional effects and risk of anti-competitive behaviour. In relation to one of these detriments, the avoidance of an inefficient price structure, OFCOM devised a model which compares consumer welfare in two scenarios, one where there is

no regulation and no threat of regulation and one where mobile termination charges are regulated in the manner proposed. The description of the model used and the assumptions made is set out in Annex 19 of the 2007 Statement. OFCOM notes (paragraph A19.25 of Annex 19) that, despite objections raised by H3G, it is appropriate to model the aggregate impact of regulating H3G alongside the other MNOs since “this captures the overall consumer welfare gains of regulating all the MNOs at the proposed rates”. But OFCOM said that it had considered the impact on H3G specifically in addition.

188. In making a comparison between the two scenarios, OFCOM had to determine what the price for mobile call termination was likely to be in the absence of regulation and the absence of the threat of regulation. OFCOM calculated that the monopoly price termination charge (that is the charge that would maximise revenue for a monopoly provider) was 23.9 ppm. If the model was run using this price of 23.9 ppm there was a consumer surplus of £54.4bn. This could be compared with £55.9bn which is the result if the model is run using a termination charge derived from the average of the charges in the proposed price control (5.2ppm). The difference of £1.4bn between those two figures is the consumer gain considered to arise from regulation in the last year of the price control 2010/11 (in real 2006/7 prices and to the nearest £0.1bn).

189. OFCOM also carried out the analysis using a sensitivity figure, 14.5 ppm, instead of 23.9 ppm, being the average between the regulated and unregulated charges. This resulted in an unregulated consumer surplus of £55.5bn compared with a regulated surplus of £55.9bn resulting in a consumer surplus of £0.4bn in 2010/11 (in real 2006/07 prices).

190. In the body of the 2007 Statement OFCOM described this outcome in the following terms (paragraph 7.49):

“ … it is also important to recall the circumstances and purpose of Ofcom’s welfare analysis. Ofcom has estimated the welfare gains from regulating call termination by comparing a situation with unregulated (excessive) termination charges against regulated termination charges. This exercise only seeks to derive an order-of-magnitude quantification of the benefits of a more efficient structure of prices and does not include quantification of the benefits to consumers from addressing the other detriments of excessive MCT charges discussed elsewhere in

this section. The analysis compares an unregulated termination charge of 23.9ppm (based on an estimated monopoly termination charge) and a regulated termination charge of 5.2 ppm based on the weighted average (by termination volumes) of Ofcom’s proposals for charges set out in Section 9. The assumptions underpinning Ofcom’s welfare analysis are set out in Annex 19.

7.50 The estimated welfare gain amounts to £1.4 billion in 2010/11 and over four years of a hypothetical charge control assuming a smooth industry average glidepath down to the target charge from the monopoly charge amounts to approximately £3.2 billion in present value terms at the beginning of 2007/08. Ofcom reiterates that the purpose of this exercise is to derive an order of magnitude estimate and not a precise estimate of the overall gains from regulation compared to no regulation (or threat of regulation). Even if the unregulated termination charge were significantly less and fell mid-way between the potential monopoly level and the regulated level, the welfare gains remain positive and large. In this sensitivity, the welfare gain amounts to £0.4 billion in 2010/11 which translates to a gain of £0.9 billion in present value terms.”

191. It was common ground between the parties by the time of the hearing that the welfare analysis carried out by OFCOM was only directed at one of the five detriments considered by OFCOM as arising from an absence of regulation of mobile call termination charges. It was also accepted that the exercise had various flaws and that it could be useful only in giving a general idea of the value of the welfare gains to be derived from regulation.

192. H3G levelled two criticisms at the model. First they argued that OFCOM had chosen the wrong two scenarios to compare with each other. They argued that OFCOM should have compared a scenario where all the MNOs except H3G are subject to the price control being proposed in the 2007 Statement and H3G remains outside the price control with the scenario where all five MNOs are subject to the price control proposed.

193. Although OFCOM did not agree that this was the correct comparison, Mr Myers, in his witness statement carried out an exercise to measure the welfare gain comparing the two scenarios put forward by H3G. By the time of the hearing Dr Littlechild accepted the methodology used – dubbed by Mr Myers the “share of volume methodology”. This involves developing a welfare model specific to H3G by appropriate adjustment of the key inputs (volume, costs and prices) for the services included in the model. Using this share of volume methodology, Mr Myers calculated the estimated welfare gain relating to a more efficient structure of prices as follows:


H3G unregulated charge	Estimated welfare gain
H3G unregulated charge	2010/11	Present value over the price control period, 2007-11
Monopoly price, 23.9 ppm	£131m	£302m
Price proposed by H3G in November 2006, 16.6 ppm	£51m	£118m
Sensitivity in 2007 Statement , 14.5 ppm	£ 35m	£ 80m
Price before January 2007, 10.7 ppm	£ 13m	£ 29m

194. Clearly the amount arrived at as the estimated welfare gain depends on the figure which is used in the model as the figure which H3G would charge in the absence of regulation. This was H3G’s second criticism of the model and, by the time of the hearing, the main area of disagreement between Dr Littlechild and Mr Myers. Dr Littlechild considered that the correct figure was 10.7 ppm because this was the average of the charges that H3G had in fact set during the period when it was unregulated. He rejected the use of the 16.6 ppm as being a purely temporary price proposed by H3G as a result of the prices set by its competitors – this was not the termination charge that H3G would set on a continuing basis in the absence of a price control. On the contrary, Dr Littlechild said, “H3G would be either willing or constrained to maintain its termination charges at a level not exceeding the 10.7 ppm previously obtaining”. He was careful to say that this was not a prediction of what price H3G would charge if it were unregulated over the coming four year period but he said that this was the most appropriate price to include in the model. H3G also rejected the use of the monopoly figure of 23.9 ppm because OFCOM’s reliance on this price was inconsistent, it argued, with OFCOM’s acceptance that the price which the MNOs could charge BT was constrained by the E2E Proviso – OFCOM would never set the monopoly price as the “reasonable” price for interconnection under BT’s E2E connectivity obligation so it did not make sense to use this as the likely price H3G would charge in the absence of a price control.

195. Using the price of 10.7 ppm, the present value of the welfare gain derived from the H3G-specific, share of volume, model over the four-year price control period 2007 – 2011 would be £29m. £29m over four years amounts to a little over £7m per year and, Dr Littlechild calculates, represents a welfare gain of 15.4 pence per year per adult in the United Kingdom. He concludes that in his opinion a welfare gain of about 15 pence per person per year is not material “It is more reasonable to describe such a potential gain as negligible”.

196. Mr Myers in his second witness statement rejected this assessment by Dr Littlechild. First, Dr Littlechild treats the gains indicated by this model as the totality of the welfare gains arising from regulation whereas in fact this model only purported to measure one of the five potential gains. The other gains, such as avoiding competitive distortion or the distortion of consumer choice are not measured by this or any other quantitative analysis but are nonetheless real. Secondly Mr Myers disputes the use of the 10.7 ppm as the input for H3G’s unregulated price. Mr Myers points to H3G’s proposal to charge 16.6ppm in November 2006 as indicating “its willingness and ability to increase charges in the absence of price control”. Thirdly Mr Myers does not regard a welfare gain of £29m as negligible saying that although the amount is small in relation to the size of the market, it is not a trivial figure in absolute terms. He states “the interpretation of this amount depends on whether it is offset by effects in other directions … or whether it represents a lower bound on the net benefits of regulation.” Mr Myers’ view is that this is not offset by effects in other directions but is in fact reinforced by net welfare gains from other sources which had been subject to qualitative rather than quantitative analysis.

The Tribunal’s conclusion on welfare gains from regulation

197. The Tribunal does not agree that the welfare gains from regulation are negligible in this market. First, the Tribunal does not consider that the original exercise carried out by OFCOM in Annex 19 of the 2007 Statement was flawed. OFCOM was very clear about the purpose of the model – that it was only intended to measure the benefits from one of the five sources of potential welfare gain – and acknowledged the limitations inherent in such an exercise. In so far as H3G’s criticism of the comparison of regulation of all five MNOs with

regulation of none of them is based on the argument that the correct counterfactual was with the position under the 2004 Statement rather than with no regulation of any MNOs, the Tribunal has already explained why it rejects that criticism (see paragraphs [184]-[186] above). In so far as H3G’s criticism is based on the alleged distinction between H3G and the other MNOs, as is discussed further below, the Tribunal does not consider that OFCOM erred in rejecting the arguments of H3G that it should not be regulated.

198. The Tribunal considers that OFCOM’s analysis of welfare benefit as explained in Annex 19 was a useful and appropriate exercise and that it indicates, in broad terms, that there is a significant welfare gain to be derived from regulating the MNOs as compared with not regulating them.

199. As regards the H3G-specific “share of volume” exercise carried out by the parties for the purposes of this appeal, the Tribunal does not consider that there is one single “right” figure to include as the price which H3G would charge in the absence of regulation or the threat of regulation such that all other figures are “wrong”. It is useful to run the model with a range of figures to get an order of magnitude of the consumer surplus. In the absence of regulation and the threat of regulation, including the threat of regulation through the medium of the E2E Proviso or dispute resolution, H3G would have both the incentive and the ability to raise its prices significantly above its historic level. OFCOM was entitled to posit that there might be no reason for the MNOs, including H3G, to refrain from charging the monopoly price. It makes sense, therefore, to include that price as indicating the upper bound of the quantified welfare gain. Similarly it is useful to consider whether there is any gain even using a figure at the lower bound, namely the 10.7 ppm proposed by Dr Littlechild. At that lower bound there is still a welfare gain of about £30 million derived from regulating H3G. To this must be added the other welfare gains from the four other factors identified by OFCOM. The fact that these have not been subject to quantitative analysis does not make the gains from these sources any less real or significant.

200. Looking at the welfare analysis carried out by OFCOM over all the five elements, OFCOM has, in the Tribunal’s judgment, clearly established that there

are welfare gains from regulating the MNOs, and from regulating H3G. Whether these gains are offset by the costs of regulation or the disbenefits arising from regulation is a different question which is discussed further below.

(c) Does the price control undermine H3G’s ability to compete in the retail market?

201. H3G argues that OFCOM has failed to give sufficient weight to factors which indicate that there are significant disbenefits from imposing a price control on H3G. First, H3G says that the imposition of a price control prevents H3G from setting its mobile call termination rate at a level which generates enough income to support its effort to offer attractive retail packages and hence compete effectively against the 2G/3G MNOs.

202. Evidence in support of this aspect of H3G’s case was given by Mr Kevin Russell who is the Chief Executive Officer of H3G. Mr Russell has been working with H3G’s 2G and 3G international telecommunications operations spanning 15 countries for over 12 years and has been closely involved in the development and deployment of the Hutchison group’s global 3G strategy. He described in his first witness statement the likely effect of the price control and indeed the effect it had already had on H3G’s retail offering. He points out that the price control will substantially increase the net amounts that H3G has to pay to its closest competitors, the other MNOs. He summarises H3G’s position as to OFCOM’s decision to impose a price control in the following terms:

“33. I cannot understand how this approach might be thought to favour competition. The Decisions benefit the incumbents at the expense of H3G, the new entrant operator which is the one operator least able to deal with such an effect and still compete effectively. The result will be a lessening of competition, higher retail prices and delays to the wider adoption of mobile broadband technology.

34. The current position is that, in circumstances where [mobile number portability] is not effective, the regulatory arrangements are structurally tilted in favour of the incumbent MNOs because H3G, the new entrant, is subject to a price control and is therefore obliged to make payments to them which they can use to fund marketing and other strategies to maintain their share and defeat H3G’s attempts to win customers from them. This seems illogical to me.”

203. Mr Russell’s evidence is that H3G has little choice, in responding to the impact of the price control, but to increase prices for new contract customers and cut

costs. Their first step was to withdraw the successful “Double Minutes, Double Text” promotion which had been available on 12-month and 18-month contracts taken out on certain price plans between 4 July 2006 and 30 November 2006. As the name suggests subscribers on these contracts obtained double their normal volume of voice minutes and texts for the first few months of their contract, for no extra charge. These tariffs provided, Mr Russell says, great value for H3G subscribers and customer acquisitions rose significantly during the course of the promotion. These tariffs were withdrawn because this step would have a significant impact on H3G’s termination costs and withdrawal could be implemented quickly since it would not require a change to H3G’s billing system. Mr Russell states that he believes they were correct in their assessment of the effect of withdrawing this tariff on their termination costs:

“H3G’s traffic imbalance has shifted over the course of 2006 and 2007 and, as regards the change in outbound minutes, I believe that has been in large part due to withdrawal of ‘Double Minutes, Double Text’.”

204. Mr Russell also gave confidential evidence of other steps that H3G had taken to reduce its costs.

205. Mr Russell concludes that with H3G now adopting a less aggressive approach to pricing the downward trend in the tariffs of the 2G/3G MNOs, which he believes were introduced as a delayed response to H3G’s promotions, would slow or stall in 2008-9. H3G does not regard imposing any price control on H3G as proportionate given H3G’s current financial position and the market conditions identified.

206. The interveners denied that H3G played the role of maverick competitor and said that they did not particularly take account of H3G in deciding their own offers. Vodafone points to the fact that in his Report Mobile access and call origination services market – Identification and analysis of market and Determination of market power – Explanatory Statement and Notification (4 August 2003), the former Director General of Telecommunications concluded that the market was already competitive at the time H3G launched its service. Mr Tillotson who is the Consumer Business Unit Director at Vodafone gave evidence on this point. He points to H3G’s lack of success in winning a substantial proportion of the

gross new connections in the four years since it entered the market and concludes that this low rate of growth is difficult to reconcile with H3G’s contention that it has had a significant impact on retail competition.

207. As Vodafone put it in its Statement of Intervention, all the MNOs use any “surplus” revenues generated from the supply of mobile call termination services to fund competition to recruit and retain mobile subscribers. It would distort competition among MNOs and other service providers to allow H3G to levy excessive mobile call termination charges in absolute terms and relative to other MNOs, since that would enable H3G (but not other MNOs) to offer more aggressive discounts to subscribers than other (potentially more efficient) MNOs could afford to offer. There is no good reason, according to Vodafone, to assist H3G to compete, relative to other MNOs by allowing it to generate “extra” mobile call termination revenues to fuel competition in the retail end of the market. To do so merely rewards inefficiency and perpetuates distortions of competition arising from the present asymmetric regulation of H3G’s and other MNOs’ call termination charges.

208. Mr Barden de Lacroix who is the Director of T-Mobile’s consumer business also gave evidence that the MNOs are under pressure to provide close substitutes for each other’s retail packages in terms of handsets, inclusive minutes and other additional services. His evidence is that the effect of H3G’s ability to charge more for its mobile call termination has been to increase the costs borne by H3G’s rivals and to provide H3G with a significantly larger income stream from termination which can be used by it to help fund its activities in the retail market. H3G has used this advantage to offer high call volume pay-monthly packages which it is not commercially viable for T-Mobile to meet. Mr Barden de Lacroix says:

“Mr Russell’s characterisation of the imposition of charge controls on H3G as leading to a “cross subsidy by H3G to the incumbent MNOs” seems to me to be turning things on their head. Regulating H3G’s termination charges in the same way as the other MNOs will lead to the removal of the cross-subsidy by the other MNOs to H3G that currently exists.”

The Tribunal’s conclusions

209. The Tribunal’s conclusions on this aspect of the case are as follows. First the Tribunal recognises that the practice of levying mobile call termination charges that are much higher than short run marginal costs enables MNOs to use some of the income from those charges to subsidise (not using that term in a pejorative sense) the retail packages they offer. For example, it is now typical to offer a free or heavily subsidised handset on a 12, 18 or 24-month contract plus a number of minutes that can be used to call any mobile network at any time of the day. In deciding to regulate the MNOs’ charges OFCOM correctly took account not only of the fact that it had found each MNO to enjoy SMP on its own network but also of the fact that the revenues generated from mobile call termination charges are available in part to be deployed in financing customer acquisition activity in the retail market. It is important therefore to recognise that a decision to regulate some but not all MNOs would alter the conditions for competition in the market for the supply of mobile subscription and call origination services.

210. OFCOM argued that it is not in the wider interests of the public to have retail packages which are subsidised by high mobile call termination charges in an unconstrained manner. This is particularly the case because of the position of fixed line callers to mobile networks. OFCOM noted at paragraphs 7.54 to 7.56 of the 2007 Statement that while fixed termination charges are regulated at cost, excessive mobile termination charges amount to a transfer of rents from fixed-to-mobile operators. This is not an efficient allocation of resources and, in a situation where fixed and mobile operators may compete with each other more closely, could result in a competitive distortion with mobile retail prices subsidised at the expense of fixed operators. Even if MNOs represent the key to challenging BT’s continued control over the local loop, OFCOM concluded that it is not efficient for this competition to be based on excessive charges for mobile call termination -

“7.56 For the same reason, Ofcom does not accept H3G’s view that increased competition in mobile markets, founded on the unique ability of one player with SMP to set MCT charges without regulatory constraints, should be pursued as a regulatory objective. Neither does Ofcom accept H3G’s view that the objections to

this approach may be set aside in the case of an MNO with a relatively small market share as the distortions between fixed and mobile sectors remain material – particularly as H3G grows.”

211. The Tribunal agrees with this analysis. The fixed network operators pay the same mobile call termination charges as the MNOs and the customers of the fixed networks do not necessarily benefit from the “waterbed effect” feeding through to cheaper retail mobile call charges. This is an important factor pointing not only in favour of imposing a price control on mobile call termination charges generally but against leaving one MNO’s charges outside the control imposed on the other four.

212. Even if one considers only competition amongst the MNOs and other mobile service providers and not between mobile and fixed networks, the ability of H3G to offer cheaper retail packages because of the money they make on mobile call termination is not necessarily beneficial to the competitive process or in the long term interests of consumers. Mr Barden de Lacroix referred in his evidence to the H3G “We Pay” tariff whereby H3G subscribers on that tariff received a five pence credit to their mobile phone accounts for every calling minute they received. Whereas that tariff may have been beneficial to the customers who bought it, the Tribunal agrees that it shows that there is a significant distortion in the retail market brought about by the disparity between H3G and the other MNOs’ mobile call termination rates. The “We Pay” tariff was only feasible commercially because H3G’s mobile call termination charges were so much higher than the charges set by the other MNOs. Such a disparity is likely either to remain in place or to increase if the 2G/3G MNOs are subject to the price control set out in the 2007 Statement and H3G is not.

213. Whether any adjustment should be made to the price control actually set in order to alter the competitive advantage given to H3G in this manner is a matter for the Competition Commission to consider. We do not accept that it justifies removing all price control from H3G.

(d) Allowing H3G to recoup sums lost through the mobile number portability arrangements

214. At the hearing H3G put forward an additional point that arises from the way in which mobile number portability currently affects the amount that H3G receives for mobile call termination on its network for subscribers who have ported their numbers to H3G from a rival network. As explained earlier, when a subscriber moves network and takes their number with them, the call to that subscriber is still routed through the donor network, and then redirected to the new recipient network. The donor network charges the network of the subscriber making the call its own mobile call termination charge, not the charge set by the recipient network. The recipient network receives that charge, less a small conveyance fee from the donor network. This means that if, say, an Orange subscriber ports their phone number to the H3G network, H3G will only receive the Orange mobile call termination charge for calls made from other networks to that subscriber (minus the Orange donor conveyance charge), not the higher H3G mobile call termination charge. Mr Myers for OFCOM accepted in cross examination that the target average charges set for the MNOs in the 2007 Statement do not take into account the fact that for some of its incoming call traffic, H3G will receive only the lower charge set by the donor network and not its own higher charge. Conversely the 2G/3G MNOs get what H3G described as “free revenue” when a customer ports their number from H3G to another MNO because although the MNO receives the higher H3G charge, this is not taken into account in determining the target average price.

215. In the 2007 Statement OFCOM records that H3G made “strong representations” in relation to the existing mobile number portability arrangements, arguing that such arrangements could result in H3G failing to recover the efficient costs of provision in the case where the rate it receives for terminating a call is below its actual cost of terminating that call. H3G also argued that the existing porting arrangement gives an incentive to other MNOs to focus their customer acquisition strategies on H3G’s customers in order to benefit from the higher termination rate they would receive when a H3G customer ports a number to their network. OFCOM had already published a consultation document

proposing changes to the mobile number portability to introduce direct routing. On the same day as it published the 2007 Statement, OFCOM published a further consultation document on whether it was appropriate to introduce an interim measure that would address the revenue impact of the current indirect routing system.

216. In that consultation document Amendment to charge control on Mobile Network Operators OFCOM set out various options for dealing with this point. The option that OFCOM regards as best addressing the impact of the existing arrangement for charging for calls to ported-in numbers was to modify the charge control so that it takes full account of such imbalances and remedies them so that all MNOs’ average effective termination revenues would be based on the appropriate termination charge (as defined in their TAC) regardless of an MNO’s actual proportion of calls to ported-in numbers. OFCOM calculates that if this option were adopted, H3G would receive in the first year of the charge control between £20-30m more termination revenue compared with what it will receive under the price control mechanism in fact imposed in the 2007 Statement. To put it another way, because of the current mobile number portability charging arrangements, H3G is worse off by £20-30m because it is not free to adapt its prices itself to counteract the acknowledged effect of the current mobile number portability arrangements.

217. H3G described this consultation process as “stalled” because OFCOM has announced that it will not pursue the consultation process on this point pending the resolution of the wider challenges to the price control set by the 2007 Statement.

218. The Tribunal considers that it would be wholly inappropriate to conclude that H3G should remain outside the price control in order to enable it, in effect, to anticipate the outcome of the consultation by increasing its charges to make up the revenue “lost” to it because of the existing mobile number portability charging arrangements. OFCOM explained that the reason it did not take account of the point in setting the price control in the 2007 Statement was that this would represent a material change to the proposals set out in the September

2006 consultation. If OFCOM’s cost model has not fully taken account of the current practices regarding mobile call termination for ported numbers this is a matter which should be raised in the consultation. That consultation must be allowed to take its course, once it is resumed following the disposal of H3G’s and BT’s appeals against the 2007 Statement.

(e) Serious adverse effects of price control on H3G’s financial position

219. H3G points out that the reduction in its own charges is substantially greater, as a percentage of the current charge, than the reduction in the charges of the other MNOs. This means that the amounts that they will have to pay the other MNOs each month for mobile call termination after netting off the amounts that they are owed by those MNOs will be substantially greater than it is currently. Mr Russell gave evidence as to the effect that this would have on H3G’s payments. He compared the amounts that H3G currently pays per month on average during 2006, which was considerably more than the amount it currently spends per month on its own marketing. He estimates that the effect of the price control will increase H3G’s mobile call termination payments to other MNOs by […][^] for the period from 1 April 2007 until 31 March 2011.

220. H3G provided a witness statement from Mr David Dyson, the Chief Financial Officer of H3G. He described the profitability milestones for the H3G business, starting with the first service revenues (generated in March 2003) and moving ultimately to a position where there is a cash flow breakeven with a return on capital, that is, where cumulative received revenues exceed cumulative paid costs and shareholders and loan holders have been repaid in full all their funding with a reasonable return on their investment. Mr Dyson says that H3G has achieved a number of profitability milestones but by no means all of them.

221. Mr Dyson explores various models of how H3G could change its business to respond to the imposition of the price control. But all the strategies demonstrate a material adverse financial impact on H3G as a result of the price control actually imposed and, Mr Dyson argues, are indicative of adverse impacts of any price control within a reasonable range being imposed. He concludes that H3G is not currently profitable and that it is very hard to understand the rationale of

applying any price control on H3G when profitability has not been achieved and in light of the fact that the price regulation would involve a value transfer to H3G’s competitors.

222. This argument was put to OFCOM by H3G during the consultation process. In the 2007 Statement OFCOM did assess whether its decision “was likely to generate any financial effects which present an unreasonable adjustment for the MNOs”. OFCOM noted that of the five MNOs only H3G identified the financial impact of OFCOM’s proposals as an argument against intervention. H3G argued that the proposals would significantly reduce H3G’s revenues over the four years of the control and result in a transfer of revenue from H3G to the other MNOs.

223. OFCOM did not dispute that a reduction in termination rates would reduce H3G’s termination revenues compared to a scenario where it was not subject to a price control. However, OFCOM noted that H3G’s own business plan, which was completed in February 2006 did not forecast mobile call termination charges which persisted at the current rate. OFCOM argued that when one compares the rate set in the price control with the rate used in the business plan forecasts, the difference was substantially less than the figure that H3G relies on. OFCOM considered that the reduction in revenue was caused by the fact H3G’s current charges were “well above cost” and that the drop in revenue predicted by H3G reflected the degree of reduction in charges that was needed in order to bring the charges into alignment with an efficient cost benchmark.

224. The Tribunal agrees with OFCOM that the size of the drop in revenue predicted as resulting from the imposition of the price control is of limited relevance to the question whether it is appropriate to set a price control since it may, as OFCOM claims in this case, simply reflect the fact that the unregulated price was far too high. H3G’s profitability is also of limited relevance – the price control is not set to ensure that H3G makes any particular level of profit but is rather based on the costs of an efficient operator. There was no evidence that H3G would be forced to exit the market as a result of the price control. Even if H3G did exit the market there was nothing to suggest that the network would cease to be operated in the United Kingdom. The Tribunal does not consider that H3G’s arguments

as to the difficulty of achieving profitability justify leaving its mobile call termination charges unregulated for a further four years.

(f) Tribunal’s conclusion on disproportionate ground

225. The Tribunal concludes that none of the arguments raised by H3G establishes that the imposition of a price control was a disproportionate measure. It is for the Competition Commission to determine whether any of the points raised by H3G justify an adjustment to the level of the price control actually imposed.

(vii) Was the imposition of a price control discriminatory contrary to section 47(2)(b) of the 2003 Act?

226. H3G alleges that the imposition of a price control on H3G was discriminatory because OFCOM failed to recognise and take account of the important differences between H3G and the other MNOs and decided to impose the same kind of price control on them all. This was contrary to section 47(2)(b) of the 2003 Act which provides that OFCOM must not set a price control unless it is satisfied that the condition is not such as to discriminate unduly against particular persons.

(a) Should H3G be regarded as a new entrant into a mature and saturated retail market?

227. H3G argued that it should be treated differently from the other MNOs because it is a new entrant to a market which it characterised as “mature and saturated” when H3G launched its service. Whereas the other MNOs could build their businesses by attracting first time mobile phone users to their services, H3G could only build market share by persuading existing mobile phone users to switch service providers.

228. Mr Russell gave evidence on behalf of H3G on this point. His evidence was that Orange and T-Mobile joined a mobile market (in April 1994 and September 1993 respectively) with only limited customer penetration - of approximately 1.5 million to 2.3 million United Kingdom subscribers - and were able to achieve growth from new customers coming to mobile. The vast majority of UK consumers had no mobile phone at all. However, UK subscriber numbers had

already increased to about 50 million by the time of H3G’s full commercial launch in March 2003. Mobile phones had achieved much greater market penetration and a highly developed “pay as you go” market sector already existed. H3G therefore needed to acquire customers from incumbent MNOs to be successful in introducing its 3G technology. Moreover, the 3G handsets that H3G had to use were much more expensive than the 2G handsets used previously because 2G was an established technology and the 2G handsets were already being produced in large numbers.

229. The Interveners disagreed with the description of the market as “mature” and “saturated”. Mr Tillotson said that Vodafone’s estimates indicate that in the four years after H3G entered the market there were about 104 million gross connections. A “gross connection” represents an occasion when a new number is activated on a mobile network either when a customer ports their number to a new network or where they are allocated a new number. This figure therefore includes not only new mobile customers with first time connections but also customers who have switched provider and those who “churn and return” i.e. who remain with their existing provider with a different mobile number (for example to take advantage of a new retail package which applies to new numbers only).

230. Mr Barden de Lacroix on behalf of T-Mobile stated that the total number of mobile service subscribers has risen from 52.8 million at the end of 2003 (that is after H3G had entered the market) to 69.7 million by the end of 2006. These figures suggest that in that three year period the market grew by about 17 million subscriptions, that is more than 30 per cent. The figures also indicate that an increasing number of people are using more than one mobile phone, a trend that is encouraged by the popularity of devices such as “BlackBerry” type handsets, and MP3-capable phones.

231. The Interveners also point to the success of the mobile virtual network operators who compete with the five major MNOs for retail mobile phone customers. Vodafone estimates that Virgin Mobile, which was the first MVNO and launched in 1999, has a market share by subscribers of more than 6 per cent in the United

Kingdom or approximately 4.5 million active subscribers. Other MVNOs include BT and Carphone Warehouse as well as other companies with high profile brands such as Tesco and Asda. There are now at least 13 significant MVNOs in the UK market accounting for about 10 per cent of the retail market.

232. H3G also argued that they are being subjected to a price control after a significantly fewer number of years of operation than was the case for Orange and T-Mobile. However, there are some advantages to late entry into the market. Mr Barden de Lacroix’s evidence was that when T-Mobile entered the market in 1993 it had network coverage of just 20 per cent of the UK population and there was no mechanism for T-Mobile to insist that the other UK networks allow roaming on their networks for T-Mobile subscribers. In contrast when H3G entered the market O2 and Vodafone were subject to licence conditions requiring them to provide national roaming services to H3G if H3G was unable to secure such services commercially. H3G was able to enter into a roaming agreement with O2. This meant that H3G had 98 per cent coverage by population over the United Kingdom as soon as it launched whereas T-Mobile only achieved such coverage by 2005, some 12 years after its entry into the market.

233. In the light of the evidence adduced by the Interveners, the Tribunal is not convinced that in 2007 H3G can properly be regarded as a new entrant or that it suffered disadvantages on entering the market to the degree it claims. Even if it would be right, as H3G argue in their Reply on the SMP and Appropriate Remedy Issues, to leave “churn and return” figures out of consideration and even if a proportion of the new gross connections and additional mobile subscriptions are second phones taken up in a situation where the subscriber is likely to favour his or her existing network, the figures still indicate that the retail mobile market is far from stagnant. H3G is a member of a substantial group of companies with considerable experience in penetrating this market. Certainly in the Tribunal’s judgment this factor does not justify leaving H3G’s price unregulated.

(b) H3G’s smaller market share and the reasons for its traffic imbalance with the other MNOs.

234. H3G argues that it is still substantially smaller than the other MNOs because it has only 4 per cent of subscribers whereas the 2G/3G MNOs have roughly equal shares of the remainder of the market. This means that the adverse effects on its business of the price control are much greater than is the case for its competitors. An MNO with a small market share is likely to have a larger proportion of its calls made “off-net” rather than “on-net”, that is the proportion of outgoing calls which are made to someone on a different network rather than on H3G’s network is likely to be greater for H3G than is the case for the networks with a larger share of subscribers. Since no mobile call termination is paid when the call is on-net, this means that H3G pays mobile call termination charges for a higher percentage of its calls than the other networks who have more on-net calls. Of course, this also means that H3G receives a higher proportion of incoming calls “off-net” that is, from subscribers on other networks.

235. There was some discussion about the correct measure for market share in particular contexts. Dr Walker, T-Mobile’s expert witness, pointed out that H3G’s share of minutes called was 10.4 per cent in the final quarter of 2006 and its share of total market revenue (that is taking retail revenue and wholesale revenue from mobile termination charges) was more than 9 per cent in that quarter. These figures were not challenged by H3G.

236. OFCOM stated in the 2007 Statement that it was “not persuaded” by H3G’s argument that because it has relatively few subscribers, regulation will be of limited benefit (paragraph 7.28):

“First, H3G will grow over the period of the control and Ofcom does not consider that the consequent volume of terminated minutes will be insignificant. Second, Ofcom’s analysis of the financial effect on H3G … suggests that the proposed remedy will not undermine H3G’s overall financial position and, moreover, H3G’s reduced MCT termination revenue would be small compared to its overall revenue. Therefore for the reasons argued here and in section 9 Ofcom considers that there are material welfare gains from the regulation of H3G.”

237. H3G argues from this that OFCOM had concluded that at H3G’s current market share there was no justification for imposing regulation and that the only

justification put forward was that H3G would grow. We do not consider that that is a fair reading either of that paragraph or of the 2007 Statement as a whole. It is clear from the 2007 Statement that OFCOM did not regard H3G’s presence in the market now as so small that regulating it or not regulating it would have an insignificant impact on the market. Looking at the variety of measures of market share we do not think it was possible for OFCOM to conclude that H3G’s current market share was so low that its pricing decisions could not give rise to any of the detrimental effects identified. The fact that H3G has a smaller market share and might remain smaller than the other MNOs over the period of the price control is not in the Tribunal’s judgment a reason why its mobile call termination charges should remain unregulated.

238. H3G argued that the 2G/3G MNOs have “much stronger SMP” than H3G and that BT’s CBP in relation to H3G offsets or neutralises H3G’s SMP to a far greater degree than is the case for the other 2G/3G MNOs. H3G argued from this that because of its smaller market share the need for certainty in relation to H3G’s prices was less strong in respect of H3G. BT disputed this. Mr Amoss on behalf of BT stated that BT still requires certainty as to H3G’s mobile call termination rates regardless of the fact that they currently have only 4 per cent of the retail market by subscriber. Mobile call termination from fixed lines to mobile is still a significant service in terms of volume of calls, involving some 16,000 million minutes of calls a year. A price difference of 5 pence per minute would cost fixed line consumers about £800 million per annum and H3G’s share of this would still be a significant sum in absolute terms even if its market share did not increase substantially.

239. We agree with Mr Amoss’s assessment. Even with a market share of 4 per cent of subscribers (and a significantly higher share of minutes) H3G’s position in the market is sufficiently significant to justify regulating its charges.

(c) H3G’s traffic imbalance and the reasons for it

240. Underlying H3G’s arguments about its small market share, its status as a new entrant and the reasons why it should not be subject to price control is the case that H3G put forward about the disadvantages it faces because of the current

mobile number portability arrangements. Mr Russell gave evidence for H3G on this issue. He asserted that in switching between MNOs most customers do not want to change their telephone numbers because of the inconvenience this causes. Compared with arrangements in markets overseas, the UK system is flawed – it presents a significant barrier to growth and limits customer choice and freedom. He identified two particular problems with the UK system. First that it is “donor led” in that the customer must approach their existing operator to start the process. This gives the existing network a chance to try to win back the customer or persuade that customer to keep their phone number for incoming calls. Mr Russell compares this with other countries that are all recipient-led and thus avoid this problem. Secondly the process is cumbersome and takes too long. Same day porting is international best practice. For example in Australia, many ports are achieved within 1 hour and the regulator requires that 90 per cent be completed within three hours. In Ireland porting takes 20 minutes on average.

241. According to Mr Russell the inadequacies of the UK mobile number portability system have materially contributed to H3G having what he calls “a systemic traffic imbalance” with the incumbent MNOs. This meant, he said in evidence, that even at the level of termination charges set by H3G when it was unregulated, H3G was a net out payer of mobile call termination charges to the incumbents. The 2007 Statement will seriously exacerbate this effect. Over the period covered by the 2007 Statement H3G estimates that the price control imposed on H3G will leave it much worse off in terms of its overall net interconnect position. This is because although H3G’s payments to the other MNOs will diminish because their prices are also being brought down by the 2007 Statement, H3G’s own revenue from mobile call termination will decrease by a greater amount as a result of its price control.

242. Further the result of these problems, according to Mr Russell, is that H3G has achieved a substantially lower percentage of customers who port their numbers compared with the Hutchison group businesses in other jurisdictions. He also states that the UK system encourages what he calls “second hand set behaviour” whereby many customers retain their existing hand set and number to receive in-coming calls and use the H3G phone for making outgoing calls. This means that

although H3G earns revenue from the new customer from the retail sale of the outgoing calls it does not earn revenue from mobile call termination charges for the incoming calls. He points to the fact that H3G customers make substantially more outgoing calls than they receive, though the extent to which this is the case has reduced somewhat when one compares the figures for the first nine months of 2007 with the figures for 2006. H3G does not experience the same level of traffic imbalance in any of the other jurisdictions in which it has 3G operations and H3G relies on a correlation between this fact and the fact that the other jurisdictions have effective recipient-led mobile number portability arrangements as supporting its arguments. The Hutchison companies around the world have not, according to Mr Russell’s evidence, experienced the same degree of traffic imbalance despite the fact that they have followed similar entry strategies.

243. The evidence put forward by Mr Russell was, in the Tribunal’s judgment, largely undermined by the evidence put forward by OFCOM and the Interveners.

244. The Interveners relied on market surveys conducted by OFCOM which indicate that very few customers cite the length of time taken to port their number as a reason not to change network. OFCOM’s research The consumer experience – Research report (November 2006) shows that 95 per cent of consumers who had switched provider believed that switching was easy or very easy. Such evidence does not, of course, indicate how many people who did not switch suppliers were discouraged from doing so by the number portability arrangements. However, OFCOM’s November 2006 consultation Review of General Condition 18 – Number Portability (16 November 2006) included the results of a consumer survey which found that of those respondents who had not switched provider, only three out of 1,167 cited unprompted the time taken to transfer their number to a new network as a reason for not switching. Of those respondents who had switched provider, three quarters changed their number. Their reasons for not porting their number were varied but only 10 per cent cited unprompted the hassle or delay of porting their number as a reason for not doing so. Of those respondents who had switched provider and ported their number, 82 per cent expressed themselves to be “satisfied” or “very satisfied” with the time taken to transfer their number.

245. In its July 2007 consultation on number portability, Arrangements for porting phone numbers when customers switch supplier – A review of General Condition 18 (17 July 2007) OFCOM stated that it does not agree with H3G that present arrangements for routing calls to ported numbers necessarily result in significant unmet demand for porting. In a survey annexed to that statement it was reported that those customers who had switched providers and had changed (that is not ported) their number, only 8 per cent cited hassle or delay as the reason for not doing so.

246. H3G submitted its own survey evidence to OFCOM and appended it to its Notice of Appeal. This indicated a higher level of dissatisfaction with the current mobile number portability arrangements but still indicated that the majority of respondents who had ported their numbers were satisfied with the porting time.

247. Moreover, there was a great deal of evidence that in fact the numbers of people changing network are very high. Mr Barden de Lacroix on behalf of T-Mobile referred to a table contained in a report produced by Merrill Lynch dated 28 March 2007 entitled Global Wireless Matrix for Q4 2006. This table shows that the level of “churn” in the market has been between 30 and 33 per cent in each of the years 2005, 2006 and 2007. This does not support a conclusion that switching is being impeded by concerns about the portability process. Rather, as Mr Barden de Lacroix says the figures demonstrate that “there are, every year, many millions of consumers who re-evaluate their mobile phone subscription arrangements and are prepared to move to a different MNO/MVNO in order to get a better deal”. (§53)

248. H3G in their Reply cast doubt on this, asserting that “the vast majority of customers never switch networks” and referring to OFCOM’s recent survey data as recording that only 12 per cent of subscribers had switched network in the past 12 months. Dr Walker, for T-Mobile, explained during his cross examination why this survey evidence was perhaps less reliable than the other evidence put before the Tribunal which was based on actual reported disconnections gathered from the operators. In any event, H3G’s Reply does not properly describe the results of the OFCOM Research Report The Consumer Experience

(20 November 2007). The figures show that the percentage of people who had switched in the last 12 months as at the second quarter of 2005 and 2006 was 16 per cent and this had fallen to 12 per cent for the second quarter of 2007. The Report also stated that two in five consumers have ever switched suppliers, a quarter had switched once and 16 per cent have switched at least twice. OFCOM recorded that there had been a “significant increase” in the proportion of consumers who have switched twice. Further, customers with a contract (rather than ‘Pay As You Go’) are much more likely to switch – switching amongst these customers increased from 50 per cent in 2006 to 55 per cent in 2007. We do not therefore regard the Report as contradicting the points made by the Interveners and OFCOM.

249. H3G also criticised the figures relied on by the Interveners because they include “churn and return” customers who remain with their existing network but with a different mobile number. H3G argues, it is unrealistic to suggest that a customer who is loyal to a particular network but simply changes his phone number in order to move his subscription to a different package offered by the same network, presents a real opportunity for a competing network to win his business. We disagree. In our judgment, a customer who is willing to change his phone number in order to improve the package he is on with his existing network does represent a competitive opportunity for another network since he is clearly more price sensitive than another customer who stays with the same package year after year.

250. Mr Barden de Lacroix has analysed where the customers who have ported away from T-Mobile have gone. In the year to September 2006 only 14 per cent of customers who ported away from T-Mobile ported to H3G. This figure compares unfavourably with the 35 per cent who ported to Orange and the 31 per cent who ported to O2. It is also below the 17 per cent who ported to Vodafone. Mr Tillotson gave the equivalent figures for the destination networks for those who port away from Vodafone. These also showed that each of Orange, T-Mobile and O2 was significantly more successful in attracting customers than H3G. H3G’s competitors use the same porting system and are subject to the same port lead times and donor conveyance charges as H3G. That some

networks have been more successful than H3G in winning new customers cannot therefore be attributable to the mobile number portability system.

251. The Merrill Lynch figures to which Mr Barden de Lacroix refers show that H3G’s churn figures are the highest of all the MNOs albeit that its rate of churn has fallen from 7.1 per cent per month in 2004 to around 4 per cent per month in 2006. Even at a churn rate of 4 per cent per month, this suggests that H3G has been losing around half of its customers in the space of a year. Mr Tillotson for Vodafone compared the figures from the 2006 Annual Report of Hutchison Whampoa (H3G’s ultimate parent company) which show an average churn rate of 4.9 per cent monthly for the UK and Ireland with Vodafone’s churn rate for UK contract and pre-pay customers of 33.8 per cent annually. This may be because, as H3G put it in their Reply, H3G is more likely to have “value-seeking regularly churning customers” given that its customer base is necessarily made up of those who have chosen at least once to switch networks. But it still indicates that the mobile number portability arrangements do not appear to be dissuading customers from leaving H3G.

252. Mr Tillotson’s evidence was that most customers attach little priority to keeping their phone number. He points to the significant minority of Vodafone’s churn rate which is accounted for by “churn and return” customers. The fact that so many customers choose to change phone number even without switching supplier in order to take advantage of a particular retail offer illustrates the lack of importance that customers attach to retaining their mobile phone number. Mr Tillotson accepted that the donor led process did allow the donor MNO to try to win back the customer. But he said, unsurprisingly, that their efforts to retain their customers apply to all those who are switching whether they are doing so simply by terminating their contract with Vodafone and getting a new number from their preferred network or by porting their number to a competitor.

253. As regards H3G’s evidence about second handset behaviour Mr Tillotson said that from “a purely intuitive perspective” it seemed unlikely that there can be many customers prepared to tolerate the inconvenience of keeping two mobile phones indefinitely rather than wait five days to transfer their number. He

referred to market research commissioned by OFCOM on phone usage among H3G users and particular among those with more than one SIM card. The answers suggested that there are a variety of reasons why a customer might wish to have more than one phone, for example to split business and personal calls or to take advantage of free calls at different times. None of the reasons given had anything to do with the current mobile number portability arrangements.

254. Mr Tillotson also casts doubt on H3G’s comparisons with its experience in other countries where the MNO arrangements are faster and recipient led. He refers to an international benchmarking study published by OFCOM in its November 2006 consultation indicating that the UK compares favourably with other countries in terms of the target maximum porting period and the percentage of mobile phone users who have actually ported their number. Of the 25 countries included in OFCOM’s survey, only seven have a shorter maximum porting period than the UK and the propensity to port in the UK was in line with the average. Furthermore the evidence does not support a link between the proportion of customers who choose to port their numbers and the target maximum porting period.

255. In its Defence OFCOM set out a table comparing H3G’s average monthly incoming and outgoing minutes per subscriber with the published data for the other four MNOs. The table which showed the traffic levels calculated by reference to the active subscriber figures of the other MNOs and an estimate of H3G’s active subscriber base showed the following


	OTHER MNOs (Active subscribers)	H3G (Active subscribers)
INCOMING TRAFFIC	43 – 71 mins	[...][K] mins
OUTGOING TRAFFIC	80-123 mins	[...][X] mins

256. OFCOM submitted that this makes clear that H3G’s traffic imbalance is not caused by a deficit of incoming traffic per subscriber (as would be the case if the mobile number portability system and second hand set behaviour were the cause)

100

but is due to a significant surplus of outgoing traffic per subscriber for H3G in comparison with the other MNOs. The figures show that H3G’s average monthly incoming traffic per active subscriber was not much above the average monthly incoming traffic per active subscriber for the other MNOs. The figures for incoming traffic in this table exclude all on-net calls for H3G and the other MNOs. The figures given for outgoing calls exclude on-net calls for H3G but include on-net calls for the other MNOs. Figures for H3G’s incoming calls including on-net calls were not available to OFCOM.

257. H3G countered this point by arguing that because it has a smaller market share of subscribers one would expect it to have a larger amount of off-net outgoing calls because a smaller proportion of the calls that its subscribers make are on-net. A subscriber to H3G would necessarily therefore make and receive a higher proportion of off-net calls than a subscriber to one of the other larger MNOs. OFCOM however provided the Tribunal with a note showing that this did not account for the discrepancy that the figures illustrate. If one assumes that each of the four 2G/3G MNOs has a market share of 25 per cent so that the incoming call figures (which exclude on-net) represent only 75 per cent of the total traffic, one would have to gross up the figures in the table by 33.3 per cent to arrive at a figure for incoming calls including on-net calls. This would convert the 43 - 71 mins figures to 57 - 95. Assuming H3G has a 5 per cent market share, its grossed up incoming minutes figure would be [… ][K] mins instead of [… ][K] mins. As regards outgoing calls, only H3G’s figure would need to be grossed up to include on-net outgoing calls since these are already included in the figures for the other MNOs. This would convert H3G’s figure of […][X] mins into a figure of [… ][X] mins.

258. The table would therefore look like this:


	OTHER MNOs (Active subscribers)	H3G (Active subscribers)
INCOMING TRAFFIC	57 - 95 mins	[…][X] mins
OUTGOING TRAFFIC	80-123 mins	[…][X] mins

101

259. These figures still show that the traffic imbalance is caused by a substantially larger number of outgoing minutes rather than a substantially smaller number of incoming minutes. These figures, even once adjusted, do not support Mr Russell’s evidence that the cause of the transfer imbalance is “second handset behaviour” or otherwise the result of the mobile number portability arrangements.

- The Tribunal’s conclusion on mobile number portability and traffic imbalance.

260. In the Tribunal’s judgment the evidence provided by OFCOM and the Interveners is overwhelming in establishing that the mobile number portability arrangements are not the sole or even the major cause of H3G’s failure hitherto to grow its market share. There may be good reasons to reform the current system for mobile number portability, for example to bring the United Kingdom system into line with the arrangements which exist in other jurisdictions or otherwise to improve the arrangements in ways which bring advantages to consumers. Our concern in this case is to consider whether there is evidence that the current arrangements, which apply to all the MNOs, in fact have a particularly adverse effect on H3G because of H3G’s position in this market. The evidence we have seen shows that there is a large amount of subscriber switching in this market and many millions of opportunities for the MNOs to compete with each other for customers’ business. Mobile number portability may at most be a contributory factor towards the lack of customers prepared to switch to H3G but certainly does not play the role that H3G has claimed for it in these proceedings.

261. We do not find the evidence of H3G’s experience in other jurisdictions, where the mobile number portability arrangements in place are different, at all convincing. First, it is very difficult to draw any conclusions derived from two disparate facts plucked out of the information about a wide range of international markets. The OFCOM survey cited by Mr Tillotson also indicates that the inferences Mr Russell invites the Tribunal to draw from this information are at least questionable.

102

- Should the Tribunal remit the question of the cause of traffic imbalance to OFCOM?

262. H3G submitted that if the Tribunal concluded that H3G’s failure to expand its business was not caused by the mobile number portability problems then it should remit the case to OFCOM to investigate further what the cause of traffic imbalance was. Various other explanations for H3G’s limited market expansion have been raised. H3G sought to amend its Notice of Appeal to include an argument based on the allegation that the 2G/3G MNOs charge lower retail prices for calls to other numbers on their own network (on-net calls) than they charged for calls to numbers on competing networks (off-net calls). This, they wished to argue, discouraged customers from moving away from an MNO with a larger market share to one with a small market share. The Tribunal refused to allow H3G to amend its pleading to raise this point for the reasons set out in the judgment handed down on 23 November 2007: see [2007] CAT 33.

263. Mr Tillotson for Vodafone and Mr Barden de Lacroix for T-Mobile suggested various other reasons that they consider more likely to be the cause of H3G’s perceived lack of growth. The Tribunal made it clear at the outset of the hearing in January that there was not enough evidence before the Tribunal for it to be able to make findings as to aspects of H3G’s business that that evidence highlighted. Similarly there was insufficient evidence before the Tribunal from H3G as to what commercial strategy H3G had pursued since the launch of its service in the United Kingdom in relation to what groups of customers it had targeted and why.

264. The Tribunal is therefore in a position where it has found that the cause for H3G’s low market share and for its imbalance of traffic with the other MNOs is not caused by mobile number portability arrangements but it is not able to say what the cause is. If the Tribunal concluded that the reason for H3G’s small market share was relevant to the decision to impose a price control then it would need to remit the matter to OFCOM to investigate this.

103

265. However, the Tribunal is satisfied that there is no need to investigate further the causes of traffic imbalance or the reason why H3G has not expanded its market share.

266. We have set out earlier (paragraph [234]-[239]) why the Tribunal does not consider that H3G’s small market share justifies leaving its mobile call termination charges outside the price control. H3G also alleges that OFCOM’s rejection of the evidence that mobile number portability arrangements were responsible for H3G’s slow growth meant that the assumptions OFCOM used about likely growth in market share over the period of the price control were over optimistic. OFCOM’s model used for setting the price control assumes that H3G achieves parity of market share with the other MNOs by 2016 (see paragraph 9.43 of the 2007 Statement and paragraphs A5.38 and A5.79 of Annex 5 to that Statement). H3G argues that unless substantial changes are made to the mobile number portability arrangements, it is very unlikely that H3G will increase its market share to such a significant extent.

267. It is important to understand the basis for OFCOM’s assumptions underlying the price control. OFCOM was not trying to predict how H3G’s market share would grow over the price control period. Any such prediction would have needed to take account of H3G’s commercial strategy and any other factors peculiar to H3G. OFCOM’s assumption was rather that a reasonably efficient MNO would capture its fair share of churn in the market and that this would lead to it increasing its market share at the expense of its competitors over a given period. Whether or not H3G in fact does so is not, generally speaking, OFCOM’s concern – that will depend on how well H3G manages to compete with its rivals. The price control must be based on the assumption that H3G will compete as well as the other MNOs so that H3G benefits from over performing in relation to this assumption but is not favoured in the event that it underperforms.

268. This approach by OFCOM was entirely proper unless H3G put forward a reason why it would not achieve that growth even if it were as efficient as its rivals. If H3G had been able to establish that the mobile number portability arrangements presented a significant hurdle to any new entrant in growing market share, that is

104

certainly a factor that might have altered the assumptions underlying the price control. But once OFCOM had rejected that argument, rightly in the Tribunal’s judgment, it was not then incumbent on OFCOM to investigate every other alternative reason why a company in H3G’s position might gain more or less than its fair share of the business available over the period. There was nothing to undermine the reasonableness of the assumption that the efficient entrant would achieve market parity by 2016 and OFCOM was entitled to rely on that when setting the price control.

269. The Tribunal therefore does not consider that it is necessary or appropriate to remit the matter to OFCOM for further investigation.

(d) Assumption that asymmetric regulation is undesirable

270. H3G alleges that OFCOM’s approach to the question whether it should impose a price control on the 2G/3G MNOs but not on H3G was influenced by an assumption that asymmetric regulation is in general undesirable in this market. H3G argued that this ignored more recent academic literature which revises this conventional view and points in favour of asymmetric regulation for the benefit of market entrants. It is important to distinguish between two different questions here. First there is the question, which is for the Tribunal to determine, whether it is appropriate to impose what we shall refer to as “asymmetric regulation” in this market, that is to impose a price control on the four 2G/3G MNOs but not on H3G. Secondly there is the question, which is for the Competition Commission to determine, whether it is appropriate to impose what we shall refer to as “asymmetric price control” that is to differentiate between H3G and the other MNOs in terms of the Target Average Charge and/or the glide path set within the price control.

271. Mr Myers’ witness statement on behalf of OFCOM emphasised the competitive distortions that might arise from asymmetric regulation where the unregulated MNO will be able to earn a much larger profit margin on each minute of termination than the other four MNOs. The unregulated MNO would have a competitive advantage in the retail mobile market arising from its ability to exploit its market power in the wholesale mobile call termination market.

105

Mr Myers therefore supported the conclusion set out at paragraph 7.55 – 56 of the 2007 Statement where OFCOM said:

“ … it is not efficient for this competition to be based on excessive charges for MCT. For the same reason, Ofcom does not accept H3G’s view that increased competition in mobile markets, founded on the unique ability of one player with SMP to set MCT charges without regulatory constraints, should be pursued as a regulatory objective.”

272. Dr Littlechild in his third witness statement stated that at least one economist questions whether asymmetric regulation is inappropriate and whether symmetric regulation makes the situation better or worse. That economist is Dr Martin Peitz, and Dr Littlechild refers to two articles by Dr Peitz in 2002 and 2005. He also referred to two other European NRAs who have recently approved asymmetric rates for the benefit of a new entrant, or at least a smaller player, and a recent public consultation on symmetry and asymmetry of mobile and fixed termination rates issued by the European Regulators Group (ERG). Dr Littlechild concludes in paragraph 23(c) of his third statement that:

“To assess all these papers and decisions is beyond the scope of the present response to Mr Myers. I have referred to the ERG consultation, but otherwise do not proposed to discuss, defend or rely on any of the arguments contained in these papers. I cite them simply to demonstrate that the view taken by Ofcom and Mr Myers, and by [an economist cited by Mr Myers], is not the only one. The case against asymmetric regulation is not as straightforward as it may once have seemed, particularly where new entrants and the nature of competition are concerned. This correspondingly undermines Ofcom’s heavy reliance on the principle of avoiding asymmetric regulation because of its distorting effect on competition”.

273. In his oral evidence Dr Littlechild seemed to go a little further than this, stating that the documents he relied on espoused the view that there is an actual disadvantage in symmetric regulation and a positive advantage in asymmetric regulation, and that this undermines OFCOM’s reliance on the disbenefits of asymmetric regulation.

274. The Tribunal has considered the article by Dr Peitz published in 2002 Asymmetric access price regulation in telecommunications markets and his later 2005 paper Asymmetric regulation of access and price discrimination in telecommunications. The Tribunal does not consider that they support the principles proposed by Dr Littlechild. Dr Peitz is dealing with the position of a

106

new entrant coming into a market to compete with an established incumbent. It is not at all clear whether Dr Peitz would, if presented with details of the market in which the MNOs operate in the UK, consider that the points he raises in his papers have any application in the situation that the Tribunal is considering. It is certainly not right to regard Dr Peitz as recommending the outcome for which H3G contends in its current appeal.

275. We have also considered the European regulatory decisions to which Dr Littlechild referred. The first was a decision of the Portuguese regulator ICP-ANACOM which was proposing to amend the price control obligation imposed in the wholesale market for mobile call termination. ANACOM required a reduction by all the operators from the current mobile call termination rate of 11 Euro cents per minute but allowed Optimus to charge between 1.6 cents and 1.3 cents per minute more than the other two. ANACOM explained that this asymmetry in favour of Optimus as a transitory measure prior to a new market review exercise taking place before the end of 2008. The proposed asymmetry reflected Optimus’ higher costs, the fact that it entered the market six years later than the other MNOs and that it was disadvantaged by network effects and traffic imbalances.

276. The European Commission in its comments on these proposals asked ANACOM to reconsider the proposed asymmetry of mobile call termination rates in favour of Optimus. The Commission stated that “although in exceptional cases, asymmetry might be justified by objective cost differences (which are outside the control of the operators), the fact that an MNO entered the market at a later stage and that it has a smaller market share can only justify higher MTRs for a transitory period.” The Commission urged ANACOM to move towards reducing MTRs to the level of costs of an efficient operator which takes into account only those objective cost differences which are outside the control of the operators.

277. The second decision referred to by the parties was the decision of the French regulator, Decision No 2007-0819 of French NRA, ARCEP, of 4 October 2007. In France at that time there were three MNOs, Orange France, SFR and Bouygues Telecom. Bouygues Telecom had entered the market later than the

107

other two operators. In that decision, ARCEP concluded on the basis of a wide ranging review of market conditions that it was appropriate to grant a price differentiation to Bouygues Telecom to take account of the progressive nature of the process of convergence of prices towards costs. It set a price for an 18-month period of €0.065 per minute for Orange France and SFR and €0.085 per minute for Bouygues Telecom, indicating that it would reconsider the levels and therefore the asymmetry thereafter. This decision was, the Authority believed, fully consistent with the principle of symmetry in the long term.

278. Finally the Tribunal has looked at the public consultation document issued in December 2007 by the ERG on a draft common position on symmetry of mobile/fixed call termination rates. The paper notes that the Commission considers that termination rates should normally be symmetric and that asymmetry requires an adequate justification. Referring to new entrants, the paper states that the right of new entrants to recover their costs should be reconciled with the regulatory objective of achieving the maximum level of efficiency in the supply of termination services. Hence, asymmetries should not remain in force for too long and each operator’s termination rate should be brought down to the cost of an efficient operator as soon as possible. The paper notes that asymmetric termination rates may be justified, but stresses that this must be for a limited transitional period and that differences in prices must effectively reflect differences in costs.

279. In the Tribunal’s judgment, there is nothing in the European materials that supports H3G’s case in this part of its appeal. It is clear that what is being discussed is asymmetric price control rather than asymmetric regulation – there is no indication that it is appropriate to leave even a new entrant’s prices unregulated entirely as H3G suggest here. That question is simply not discussed in any of the material. Secondly it appears that the justification put forward for the asymmetry is that the new entrant is incurring higher costs as a result of objective circumstances beyond the control of the operator, making it appropriate to set asymmetric prices for a short time. However, in the present case, H3G does not argue that its prices should be left unregulated because its costs of 3G termination are substantially higher than those of the other MNOs. Its argument

108

is that it should be allowed to charge high mobile call termination rates in order to support its offering in the retail market so that it can compete more energetically with the other MNOs. There is nothing in the material we have been shown which indicates any support for asymmetric regulation based on that justification.

280. The form of asymmetric regulation which H3G urged upon the Tribunal goes far beyond anything that the material we have been shown contemplates. We do not agree with Dr Littlechild that asymmetric regulation is simply an extreme form of asymmetric price control so that the principles discussed in that material can be extrapolated to support H3G’s case on this point.

The Tribunal’s conclusions on the discrimination ground of appeal

281. The Tribunal concludes that there was no justification for treating H3G differently from the other MNOs in deciding whether to impose a price control and that the 2007 Statement does not discriminate against H3G within the meaning of section 47(2)(b) of the 2003 Act. This does not of course prejudge in any way what the Competition Commission may decide about whether there should be asymmetric price control in this case either to the extent that exists in the price controls set by OFCOM or to any different extent.

(viii) Was there a risk of a price distortion having adverse effects within the meaning of section 88(1)(a) of the 2003 Act?

282. According to section 88 of the 2003 Act OFCOM must not set a price control unless it appears to them from the market analysis carried out for the purpose of setting that condition that there is a relevant risk of adverse effects arising from price distortion. A relevant risk is defined, for the purposes of this case, as a risk that the dominant provider might so fix and maintain some or all of his prices at an excessively high level as to have adverse consequences for end-users of public electronic communications services.

283. H3G argues that OFCOM did not adduce any evidence to support the allegation that there was a real risk that H3G might raise its prices to an excessive level in the absence of a price control. On the contrary, H3G argues, there is plenty of

109

evidence to show that they would not do so. H3G relies on the fact that the initial price struck with BT when H3G launched its 3G service in March 2003 was the same as the 2G rates fixed by one of the 2G/3G MNOs (T-Mobile) and comparable with those fixed by another (Orange). In its Reply, H3G analyses the figures set out in Figure A13.4 of the 2007 Statement where OFCOM compares 2G/3G 1800MHz and 3G-only blended efficient charge benchmarks under a particular voice and data traffic scenario. That shows, H3G says, that in 2003/4 when H3G entered into its SIA with BT, the efficiently incurred cost benchmark for 3G only termination was over 10 ppm and indeed H3G argues that it is clear from the figures in the 2007 Statement that until about 2006 the efficiently incurred cost benchmark was above 10 ppm. Secondly, H3G argues that H3G’s 3G rates have generally been lower than the 3G element within the blended rate of the 2G/3G MNOs.

284. OFCOM’s primary argument in rebuttal of this point is that there is no need for it to adduce the kind of evidence to which H3G refers before it can be satisfied that there is a relevant risk of excessive pricing for the purposes of section 88. It needs only to show that the dominant provider has the ability to raise prices and that it has the incentive to do so. Here one is dealing with a dominant provider which has 100 per cent share of the relevant market and absolute barriers to entry. There are two incentives for an MNO to increase its prices because (a) this increases the costs of MNOs’ competitors and (b) it generates revenues which can be used by the MNO to compete in the retail market. OFCOM argues that those facts in themselves are sufficient to satisfy the test in section 88.

285. The Tribunal agrees with OFCOM’s analysis of this point. It is clear from the structure of the SMP regime that the provisions contemplate that something more is needed than a simple finding of SMP in order to justify the imposition of price controls – it is possible that a communications provider can be found to have SMP in circumstances where the test for imposing a price condition is not satisfied. OFCOM referred, as an example of such a situation, to its consideration during the market review leading to the 2004 Statement as to whether to impose a price control on H3G as regards 2G termination. We were referred to passages in the December 2003 consultation which show that

110

OFCOM decided that H3G had the ability to set excessive prices for 2G termination services (see paragraph 5.125 of the Consultation) but found that it was less certain that it had the incentive to do so because it was in H3G’s interest to migrate traffic as soon as possible from its 2G roaming arrangement to its own 3G network. This, along with the difficulties of establishing the costs of providing the 3G service led OFCOM to conclude that a cost based price control for 2G call termination by H3G was not justified.

286. But in the Tribunal’s judgment this does not mean that OFCOM must identify a period in the past when the operator’s prices were not constrained by regulation and attempt to assess how reasonable its prices were during that period. It is enough that OFCOM can conclude that there is an ability to charge high prices, an incentive to do so and no other factor in the market which would remove the risk of excessively high pricing in the absence of regulation. There is no need to show that the operator has charged or attempted to charge an excessively high price.

287. As regards other factors which might remove the risk of excessively high pricing, H3G also reintroduced in this context the arguments it relied on for its case on the SMP Issue namely that the E2E Proviso itself has the effect of preventing BT from having to agree to excessively high prices and so prevents the risk of such prices arising. In its Reply on the SMP and Appropriate Remedy Issues, H3G points out that the end-to-end connectivity obligation was designed to limit BT’s otherwise very significant bargaining power and to equalise BT’s position with undertakings seeking interconnection. In order to find that, absent a price control, H3G would impose an excessive or abusive price on BT, OFCOM must conclude that it would in a dispute between BT and H3G approve such a price as reasonable under BT’s end-to-end connectivity obligation. This argument fails, in the Tribunal’s judgment for the same reasons that the argument fails in relation to the SMP Issue; first because OFCOM might set a rate as a result of dispute resolution which is appreciably above the competitive level, although not abusively high and secondly because under the modified greenfield approach, OFCOM’s potential intervention to set the price must be disregarded when considering H3G’s likely conduct in the absence of regulation.

111

288. Looking at the conditions in this market, OFCOM was fully justified in concluding that the test in section 88(1)(a) was met.

289. As a secondary argument, OFCOM also relies on the pricing behaviour of H3G as evidence which shows that the test in section 88(1)(a) is met. In particular OFCOM points to H3G’s increase in its termination rate to 16.6 ppm as indicating H3G’s ability and willingness to raise its prices in the absence of regulation. H3G argued strenuously that this was not a fair interpretation of events. Mr Russell’s evidence was that H3G only sought to increase its own mobile call termination rate to achieve “a degree of parity in 3G rates” with the 2G/3G MNOs who had introduced a 3G element into a blended rate. H3G did not want to be “at a further competitive disadvantage” in response to the actions of the 2G/3G MNOs. H3G’s position had always been, however, that mobile call termination rates were set too high – that is why it had appealed against OFCOM’s decision in the BT Dispute Determinations, even though OFCOM had upheld the price that H3G sought to charge.

290. The Tribunal is not convinced by Mr Russell’s explanation of the H3G price rise proposed in November 2006. The perceived need to achieve parity with the 3G rates of the other MNOs in order to protect H3G’s competitive position can mean two things. First, H3G might have been concerned that if it failed to increase its own termination rate, its payment imbalance with the other MNOs would widen as a result of the price increases imposed by the other MNOs. Secondly it could mean that H3G was concerned that the MNOs would increase their revenues from mobile call termination and be able to use those sums in competing for customers in the retail market. But to safeguard H3G’s position with regard to both those concerns, H3G needed only to increase its rate to match the overall increase in rate that the other MNOs were imposing, not to match the 3G element in the blended rate. A much smaller increase in H3G’s rate to achieve parity with the blended rates rather than the underlying 3G rates of the 2G/3G MNOs would have cancelled out any widening of the payment imbalance between H3G and the other operators and have provided H3G with a matching increase in absolute revenues to support its own retail effort.

112

291. But whatever the reason behind the price increase to 16.6 ppm proposed by H3G in November 2006, the Tribunal accepts that this attempt by H3G to increase its prices to BT supports OFCOM’s contention that the test in section 88(1)(a) is satisfied here. Given the arguments put to the Tribunal by H3G regarding their poor profitability, OFCOM was fully justified in assuming that if H3G had the ability to increase profits by charging more for mobile call termination, it would be likely to do so.

292. As regards H3G’s argument that its prices for 3G termination may have been fixed below the rate of efficiently incurred costs in 2003/04 and that they have generally been below the underlying rate for 3G termination included in the blended rates of the other MNOs, OFCOM argues that this comparison is misleading. One should not just look at the period since the 2G/3G MNOs introduced blended rates but at the period since they introduced 3G termination. There have been periods in which 2G/3G MNOs were providing 3G call termination but were not charging any more for it than the regulated 2G price. During that period, H3G’s price for 3G termination was above the price charged by 2G/3G MNOs. This is particularly the case given that H3G charges its 3G termination rate in respect of calls which are in fact terminated on the 2G networks of its roaming partners.

293. Mr Amoss giving evidence for BT also disputed this contention by H3G. He points to the fact that although H3G’s mobile call termination rate has remained constant over the period since its service was launched, the overall rates of the other MNOs declined over the same period. O2 in its Statement of Intervention also argues that the fact that H3G’s termination rate has remained static while the 2G rates of the other MNOS have fallen is a very strong indicator that H3G’s mobile call termination charges are not constrained in any way by competition and that there is a real risk that H3G will charge excessive rates absent a price control.

294. In the light of the fact that the rates under discussion are the subject of the Termination Rate Dispute appeals the Tribunal does not consider it is necessary in this judgment to explore further the appropriateness of the comparisons being

113

made. It suffices to say for the purposes of this part of the judgment that the Tribunal has seen nothing in the history of the setting of charges for mobile call termination on 3G networks which indicates that the market has become effectively competitive or that OFCOM was wrong to conclude that there was a relevant risk of adverse effects.

295. Section 88(1)(a) of the 2003 Act requires not only that there is a relevant risk of adverse effects arising from excessively high prices but that these prices must be such as to have adverse consequences for end-users of public electronic communications services. This raises the question of whether the “waterbed” effect is “complete” – that is to say, whether the state of competition in the retail market is sufficient to ensure that all surplus profits earned by MNOs from mobile call termination are passed through to consumers in the form of lower retail call packages.

296. In the 2007 Statement OFCOM acknowledged that the retail market has become more competitive in recent years and it has become less likely that MNOs will be able to retain excess profits earned in supplying termination services. Nevertheless OFCOM states that it remains of the view that in a market with a limited number of network competitors, complicated retail tariffs and significant entry barriers the waterbed effect is unlikely to be complete. The Tribunal shares this view and further agrees with OFCOM that even if the waterbed effect were fully effective, excessive termination charges may give rise to other detriments, in particular distortions between fixed and mobile operators and inequitable distributional effects.

297. The Tribunal’s conclusions as to the undesirability of unconstrained mobile call termination charges being used to advantage one mobile competitor in the retail market have been set out earlier. In the Tribunal’s judgment these concerns amply demonstrate that there would be a relevant risk of adverse consequences for end-users if H3G were not subject to a price control in respect of its mobile call termination prices.

114

(ix) The Tribunal’s conclusions on the price control remedy

298. The Tribunal’s conclusion is that none of the reasons put forward as to why it was inappropriate to impose a price control remedy on H3G is well founded. In the Tribunal’s judgment, OFCOM was correct in deciding to set a price control for H3G as well as for the 2G/3G MNOs in the 2007 Statement.

VII. CONCLUSION

299. The conclusions arrived at by the Tribunal on all the issues set out in this judgment are unanimous. The Tribunal therefore:

(a) dismisses the appeal against the finding of significant market power in the Reassessment Statement; and

(b) dismisses the appeal against the 2007 Statement in so far as it comprises matters which are not specified price control matters.

Vivien Rose Andrew Bain Adam Scott

Charles Dhanowa Date: 20 May 2008

Registrar

115

	[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 Limited v Office of Communications [2008] CAT 11 (20 May 2008) URL: http://www.bailii.org/uk/cases/CAT/2008/11.html Cite as: [2008] CAT 11





	Neutral citation [2008] CAT 11 IN THE COMPETITION 1083/3/3/07 APPEAL TRIBUNAL

	Victoria House Bloomsbury Place 20 May 2008 London WC1A 2EB Before: VIVIEN ROSE (Chairman) PROFESSOR ANDREW BAIN OBE ADAM SCOTT TD Sitting as a Tribunal in England and Wales BETWEEN: HUTCHISON 3G UK LIMITED Appellant -v-OFFICE OF COMMUNICATIONS Respondent supported by O2 (UK) LIMITED T-MOBILE (UK) LIMITED VODAFONE LIMITED ORANGE PERSONAL COMMUNICATIONS SERVICES LIMITED BRITISH TELECOMMUNICATIONS PLC Interveners Heard at Victoria House from 24 January to 5 February 2008

	JUDGMENT ON NON PRICE CONTROL MATTERS (Non-Confidential Version)



	I. INTRODUCTION 1. This appeal is brought by the Appellant (“H3G”) under section 192 of the Communications Act 2003 (“the 2003 Act”). H3G is a mobile network operator which challenges certain aspects of two decisions adopted by the Respondent (“OFCOM”) on 27 March 2007. 2. H3G’s appeal concerns the prices that mobile network operators charge for mobile call termination (“MCT”). Mobile call termination is the process of connecting a voice call from the caller’s network to the recipient’s network. Consumers expect to be able to make calls from their fixed line or mobile phone to any other retail customer irrespective of the service provider (fixed or mobile) to which the receiving party subscribes. Network operators enter into contractual arrangements with each other for the provision of access to each other’s networks. Under those arrangements the terminating network operator makes a charge for each call terminated on its network, known as a mobile call termination charge. 3. The charge for mobile call termination is expressed in pence per minute or “ppm”. Usually the mobile network operators (“MNOs”) set different prices for terminating day-time, evening and weekend minutes. There are tens of billions of minutes terminated on the networks of the MNOs each year so that changes of a fraction of a penny in the rates make a difference of many millions of pounds in the income and expenditure of these companies. 4. In the United Kingdom there are two main forms of mobile network commonly known as ‘2G’ and ‘3G’. Second Generation or 2G networks were originally designed to support mobile voice calls and text messaging services using a radio transmission technology known as Global System for Mobile Communications (“GSM”). 2G networks were subsequently enhanced to support low speed mobile data services such as mobile internet access and picture and multimedia messaging services. Third Generation or 3G networks are aimed at supporting higher speed call services (for video telephony) and higher speed mobile data

	1



	(c) such rules as they may make for those purposes about the use of cost accounting systems; and

	(d) obligations to adjust prices in accordance with such directions given by OFCOM as they may consider appropriate. (10) The SMP conditions authorised by subsection (9) include conditions requiring the application of presumptions in the fixing and determination of costs and charges for the purposes of the price controls, rules and obligations imposed by virtue of that subsection.” 21. Section 88 of the 2003 Act then sets important preconditions which must be satisfied before OFCOM sets a price control authorised by section 87(9): “88 Conditions about network access pricing etc. (1) OFCOM are not to set an SMP condition falling within section 87(9) except where-(a) it appears to them from the market analysis carried out for the purpose of setting that condition that there is a relevant risk of adverse effects arising from price distortion; and (b) it also appears to them that the setting of the condition is appropriate for the purposes of-(i) promoting efficiency; (ii) promoting sustainable competition; and (iii) conferring the greatest possible benefits on the end-users of public electronic communications services. (2) In setting an SMP condition falling within section 87(9) OFCOM must take account of the extent of the investment in the matters to which the condition relates of the person to whom it is to apply. (3) For the purposes of this section there is a relevant risk of adverse affects arising from price distortion if the dominant provider might-(a) so fix and maintain some or all of his prices at an excessively high level, or (b) so impose a price squeeze, as to have adverse consequences for end-users of public electronic communications services. (4) In considering the matters mentioned in subsection (1)(b) OFCOM may- (a) have regard to the prices at which services are available in comparable competitive markets;

	9



	(ix) OFCOM’s second line of defence 95. The Tribunal has heard extensive argument on OFCOM’s second line of defence which can be summarised as follow. OFCOM’s dispute resolution function under section 185 of the 2003 Act is a regulatory function. The determination of a dispute between H3G and BT amounts to regulation of both those parties, not just of BT. In so far as the dispute resolution powers therefore constrain H3G from setting an excessively high price, they fall to be disregarded as a matter of law when assessing whether H3G has SMP. 96. This is because of the application of what is called the “modified greenfield approach”. The modified greenfield approach is a way of distinguishing between those regulatory constraints existing in a market which ought to be taken into account when assessing market power under the Framework Directive and those which should be ignored. The application of the modified greenfield approach in a situation where the regulator is considering whether SMP exists has been approved both by the European Commission in its monitoring of the decisions of NRAs under the CRF and by this Tribunal in the H3G (1) judgment. In the H3G (1) judgment, discussed in more detail in paragraphs [108] et seq. below, the Tribunal held that under the modified greenfield approach, the potential power of the regulator to impose SMP conditions on H3G must be disregarded when considering whether H3G has SMP. Conversely the regulatory obligation imposed on BT to provide end-to-end connectivity did not fall to be disregarded but should be taken into account when considering the scope of BT’s countervailing buyer power and hence the strength of H3G’s market power. 97. OFCOM therefore argues that whatever the Tribunal decides as regards the correct test applied by OFCOM when exercising its dispute resolution powers, it would be contrary to the scheme of the CRF for the potential exercise by OFCOM or any NRA of its dispute resolution function to be treated as precluding SMP on the part of a network operator. H3G’s reliance on dispute resolution in this manner would effectively make it impossible for an NRA to find that a wholesaler has SMP in so far as that wholesaler is operating in a

	41



	(i) The regulatory background 146. Before examining H3G’s arguments on this aspect of the case it is necessary to describe the scope of OFCOM’s powers in setting a price control and some other aspects of the industry which are relevant to the grounds of appeal, in particular the current arrangements for mobile number portability. (a) The setting of SMP Conditions 147. Section 45 of the 2003 Act provides for OFCOM’s powers to set conditions. Section 45(2)(b) provides for the imposition, among other things, of “an access related condition” and an SMP condition. Section 47(1) of the 2003 Act provides as follows: “Ofcom must not, in exercise or performance of any power or duty under this Chapter – (a) set a condition under section 45, or (b) modify such a condition, unless they are satisfied that the condition or (as the case may be) the modification satisfies the test in subsection (2).” 148. Section 47(2) of the 2003 Act provides that: “That test is that the condition or modification is – (a) objectively justifiable in relation to the networks, services, facilities, apparatus or directories to which it relates; (b) not such as to discriminate unduly against particular persons or against a particular description of persons; (c) proportionate to what the condition or modification is intended to achieve; and (d) in relation to what it is intended to achieve, transparent.” 149. The power to impose SMP conditions is provided in sections 87 to 92 of the 2003 Act. Before an SMP condition may be imposed, there must be a finding of SMP made in respect of the undertaking concerned (see sections 46(7) and (8)). Section 87 makes provision for the particular conditions which may be imposed by OFCOM on an operator which it has determined has SMP and section 87(9)

	59



	The Tribunal’s conclusions 209. The Tribunal’s conclusions on this aspect of the case are as follows. First the Tribunal recognises that the practice of levying mobile call termination charges that are much higher than short run marginal costs enables MNOs to use some of the income from those charges to subsidise (not using that term in a pejorative sense) the retail packages they offer. For example, it is now typical to offer a free or heavily subsidised handset on a 12, 18 or 24-month contract plus a number of minutes that can be used to call any mobile network at any time of the day. In deciding to regulate the MNOs’ charges OFCOM correctly took account not only of the fact that it had found each MNO to enjoy SMP on its own network but also of the fact that the revenues generated from mobile call termination charges are available in part to be deployed in financing customer acquisition activity in the retail market. It is important therefore to recognise that a decision to regulate some but not all MNOs would alter the conditions for competition in the market for the supply of mobile subscription and call origination services. 210. OFCOM argued that it is not in the wider interests of the public to have retail packages which are subsidised by high mobile call termination charges in an unconstrained manner. This is particularly the case because of the position of fixed line callers to mobile networks. OFCOM noted at paragraphs 7.54 to 7.56 of the 2007 Statement that while fixed termination charges are regulated at cost, excessive mobile termination charges amount to a transfer of rents from fixed-to-mobile operators. This is not an efficient allocation of resources and, in a situation where fixed and mobile operators may compete with each other more closely, could result in a competitive distortion with mobile retail prices subsidised at the expense of fixed operators. Even if MNOs represent the key to challenging BT’s continued control over the local loop, OFCOM concluded that it is not efficient for this competition to be based on excessive charges for mobile call termination - “7.56 For the same reason, Ofcom does not accept H3G’s view that increased competition in mobile markets, founded on the unique ability of one player with SMP to set MCT charges without regulatory constraints, should be pursued as a regulatory objective. Neither does Ofcom accept H3G’s view that the objections to 84

United Kingdom Competition Appeals Tribunal