dispute between cable & wireless and t-mobile about mobile

Dispute between Cable & Wireless and T-Mobile about

mobile termination rates non-confidential version

This is the non-confidential version. Confidential information

and data have been redacted. Redactions are indicated by []

Final determination and statement

Publication date: 20 May 2009

Determination to resolve a dispute between C&W and T-Mobile about mobile termination rates

Contents

Section Page 1 Summary 1

2 Background 3

3 The dispute 5

4 Ofcom’s statutory obligations and regulatory principles 16

5 Ofcom’s analysis and proposed decision 20

6 Responses to the Consultation 51

7 The Further Consultation 75

8 Summary of conclusions 88

9 The Determination 94

Annex Page

1 Cost Annex 97

2 Revised estimate of the cost of termination for C&W’s FMC Service 105


1

Section 1

1 Summary 1.1 This dispute concerns the termination rate payable by T-Mobile (UK) Ltd (T-Mobile)

to Cable & Wireless (C&W) for calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W customers using its fixed to mobile convergence (FMC) service. These payments are known as mobile termination rates (MTR).

1.2 On 8 September 2008, the parties to this dispute agreed an interim MTR of 1.2ppm, subject to this rate being adjusted retrospectively following a determination on the issue by Ofcom.

1.3 On 24 November 2008, we received a submission from C&W requesting that Ofcom handle, consider and determine this dispute. On 26 November 2008, we sent a non-confidential version of C&W’s submission to T-Mobile, asking them to comment on the scope of the dispute. On 3 December 2008, T-Mobile provided a response regarding the scope of the dispute.

1.4 Our powers and duties to resolve certain disputes are set out at sections 185-191 of the Communications Act 2003 (the 2003 Act). In accordance with Section 186(4) of the 2003 Act, on 16 December 2008 we decided that it was appropriate for us to handle this dispute, informed the parties to the dispute of our decision and published a Competition Bulletin entry setting out the scope of the dispute.1

1.5 In resolving this dispute, we have considered our general statutory duties and Community obligations under sections 3 and 4 of the 2003 Act. In the context of this dispute, we have had particular regard to our primary duty under section 3(1)(b) of the 2003 Act to further the interests of consumers in relevant markets, where appropriate, by promoting competition.

1.6 On 20 March 2009, Ofcom issued to each of the parties in dispute and to 3 parties that asked to be considered as an interested party a non-confidential version of a draft determination and explanatory statement (referred to in this document as “the Consultation”) setting out its provisional findings for the resolution of the dispute. The Consultation was published on our website on 23 March 2009.2 We published an amended version of the Consultation on 27 March 20093

1.7 In provisionally determining the appropriate resolution of the dispute, we considered the 6 principles of pricing and cost recovery established by Ofcom as an appropriate basis for the framework to set a MTR which is reasonable as between the parties and satisfies our general statutory duties and Community obligations. In consequence of that assessment, we provisionally decided that it was appropriate to directly link C&W’s MTR to a regulated rate. We considered which rate would be appropriate in this matter and provisionally decided, in the interests of cost minimisation, effective competition and reciprocity, that it should be the lowest regulated rate.

. Ofcom asked for comments from all stakeholders by 10:00am 30 March 2009.

1 See: http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ocases/open_all/cw_01004/ 2 http://www.ofcom.org.uk/consult/condocs/cwtm_termrates/determination.pdf 3 This amended version contained an adjustment to paragraph 5.35: http://www.ofcom.org.uk/consult/condocs/cwtm_termrates/determination_updated.pdf

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ocases/open_all/cw_01004/�

http://www.ofcom.org.uk/consult/condocs/cwtm_termrates/determination.pdf�

http://www.ofcom.org.uk/consult/condocs/cwtm_termrates/determination_updated.pdf�


2

1.8 In summary, based on the submissions of the parties and the evidence gathered in this dispute, for the reasons set out in the Consultation, our provisional conclusion was that:

“As from [the date of the final determination] and until the implementation of the CC Determination,4

1.9 In the light of responses to the Consultation received from stakeholders, on 15 April 2009, Ofcom issued a further consultation (referred to in this document as “the Further Consultation”) on the appropriateness of time of day (ToD) charging and the appropriate means for its implementation (if appropriate). A revised version of the Further Consultation was published on 17 April 2009.

, C&W is not entitled to charge for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in relation to C&W’s customers using its FMC an amount in excess of 4.71ppm, which is our current best estimate of the TAC for Vodafone and O2 for 2009/2010, as specified in the CC Determination, converted into nominal terms”.

5

1.10 Ofcom has carefully considered stakeholders’ responses and conducted further analysis in order to reach its final determination of this dispute. For the reasons given in section 6, Ofcom concludes that, based on the submissions of the parties, the evidence gathered in this dispute and the responses received to the Consultation and Further Consultation, as from 20 May 2009 and until such time as alternative charges are in place between the parties, C&W is not entitled to charge for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in relation to C&W’s customers using its FMC service a weighted average charge in excess of 4.71ppm and shall set the relevant time of day charges according to the following traffic weights: (i) 65% for Daytime Traffic

On 15th April 2009, in light of the need for further consultation in this matter, Ofcom declared that exceptional circumstances had arisen in accordance with section 188(5) of the 2003 Act.

6, (ii) 20% for Evening Traffic7 and (iii) 15% for Weekend Traffic8

1.11 The background to this dispute is set out in section 2. The history of this dispute is set out in section 3. Section 4 sets out the statutory obligations and principles which apply in resolving the dispute. The analysis and reasoning underpinning our proposals for resolving the dispute in the Consultation are set out in section 5.

.

9

1.12 Ofcom’s determination, which takes effect on 20 May 2009, is set out at section 9.

We address stakeholders’ comments to the Consultation and set out our responses in section 6. We address the Further Consultation and stakeholders comments to the Further Consultation in section 7. A summary of the stakeholders’ responses and Ofcom’s response is set out at section 8.

4 The Competition Commission (CC) Price Control Determination (the “CC Determination”), available at http://www.catribunal.org.uk/files/CC_Determination_1083_H3G_1085_BT_220109.pdf. 5 http://www.ofcom.org.uk/consult/condocs/cw_tm_mobileterm/dispute1.pdf 6 Daytime Traffic means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using its FMC service during the period of time between 08.00 and 18.00 on Monday to Friday. 7 Evening Traffic means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using its FMC service which is not either Daytime Traffic or Weekend Traffic. 8 Weekend Traffic means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using its FMC service during the period of time between 24.00 on Friday and 24.00 on Sunday. 9 Note that this section sets out the original analysis and reasoning underpinning Ofcom’s draft determination as contained in the Consultation.

http://www.catribunal.org.uk/files/CC_Determination_1083_H3G_1085_BT_220109.pdf�

http://www.ofcom.org.uk/consult/condocs/cw_tm_mobileterm/dispute1.pdf�


3

Section 2

2 Background Mobile voice call termination

2.1 This dispute concerns the termination rate payable by T-Mobile (UK) Ltd (T-Mobile) to Cable & Wireless (C&W) for calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W customers using its fixed to mobile convergence (FMC) service. These payments are known as mobile termination rates (MTR).10

2.2 In order for customers of different networks to be able to call each other, telecommunications networks, including mobile networks, need to be connected to each other, either directly or through a transit operator. In practice, network operators conclude interconnection agreements, setting out the terms and conditions on which they will interconnect.

2.3 The parties to this dispute have agreed a direct interconnection agreement.

2.4 According to C&W’s submission, T-Mobile has refused to accept the rates proposed by C&W on 15 May 2008 for the termination of calls to customers of C&W’s FMC service, equal to the termination rate payable for the “BT Fusion” product,11

2.5 On 8 September 2008, they agreed an interim MTR of 1.2ppm, subject to this rate being adjusted retrospectively following a determination on the issue by Ofcom.

which is 6.418ppm across each time period (i.e., [ ] for Daytime Traffic; [ ] for Evening Traffic; and [ ] for Weekend Traffic). On 30 May 2008, T-Mobile rejected the MTR proposed by C&W and suggested a MTR equal to 1.2ppm across all time periods. Despite further negotiations, the parties to this dispute were unable to reach an agreement.

C&W service

2.6 C&W describe its FMC service as a telephony service that combines the benefits of fixed and mobile telephony. The customer can use a single mobile handset in the office that uses pico cells operating over C&W’s DECT guard band spectrum and is connected back into the customers’ fixed line network. Away from the office the connectivity is provided over one of the existing mobile networks under a roaming agreement established between C&W and its partner operator ([ ]).

2.7 We understand that the C&W FMC service will be Global System for Mobile communications (GSM) based and will offer 2G telephone services, short messaging service (SMS), telephone message, general packet radio service (GPRS) and data. C&W is engaging on the rollout of its FMC service with its first customer, [ ]. C&W explained that [ ] will be replacing their fixed telephones with GSM functionality.

10 T-Mobile acknowledges that the C&W FMC service is a mobile service (see paragraph 3.44). 11 The “BT Fusion” service uses conventional GSM handsets which route calls via the fixed broadband network while the user is indoors, and hands the call over to the cellular network when the user is outdoors. We note that Orange Unique (http://www.orange.co.uk/unique/) is another example of an unlicensed mobile access service launched in the UK.

http://www.orange.co.uk/unique/�


4

2.8 C&W told us that the FMC service is targeting large corporate customers. It is intending to target existing and potential virtual private networks. Its focus is essentially on businesses with 1000 or more devices; which amounts to the largest 3000 companies in the UK. 80% of customers of the FMC service are expected to be existing C&W customers. C&W see this as an opportunity to add to their data connectivity offer; the FMC service is not likely to be sold by itself.

2.9 C&W told us that the service was launched in September 2008. We understand this to be a product launch and that operationally the service is undergoing trials with only a handful of handsets making test calls with no customers using the service on a commercial basis. C&W told us that it is likely that the service will commence operations on a commercial basis later this year.

Description of T-Mobile

2.10 T-Mobile is the UK subsidiary of T-Mobile International AG, which in turn is owned by Deutsche Telecom.

2.11 T-Mobile’s main business activities include public mobile communications network operation and the provision of mobile network communications to the public.


5

Section 3

3 The dispute 3.1 C&W notified its MTR for the termination of calls to customers of C&W’s FMC service

to T-Mobile on 15 May 2008. This rate was equivalent to the rate charged by BT for its “BT Fusion” product.12

3.2 On 29 August 2008, T-Mobile wrote to C&W setting out its proposals to open the C&W mobile number range at 1.2ppm, subject to this rate being adjusted retrospectively following a determination on the issue by Ofcom. C&W and T-Mobile reached an agreement on 8 September 2008 that an interim rate of 1.2ppm should apply between the parties until the resolution of this dispute by Ofcom. On 24 November 2008, C&W submitted a submission for Ofcom to handle, consider and determine a dispute.

T-Mobile rejected this rate on 30 May 2008 and proposed a rate of 1.2ppm. On 1 July 2008, C&W rejected the rate proposed by T-Mobile. On 18 July 2008 T-Mobile responded to C&W’s letter and re-iterated its proposal of 1.2ppm. On 22 August 2008 a meeting was held between the parties which included a proposal from T-Mobile to open up C&W’s mobile number range on the basis of an interim rate, subject to Ofcom determining the appropriate rates for new mobile entrants.

3.3 Sections 185 to 191 of the 2003 Act set out Ofcom’s dispute resolution powers. They apply to disputes relating to the provision of network access and to other disputes relating to the rights and obligations conferred or imposed by or under Part 2 of the 2003 Act. Section 186 of the 2003 Act requires Ofcom to resolve a dispute referred to it under section 185 once it has decided in accordance with section 186(2) to handle the dispute. Ofcom’s remedial powers for resolving disputes are set out in section 190 of the 2003 Act.

3.4 Having considered the parties’ submissions and subsequent information, Ofcom was satisfied that the dispute the parties had asked it to resolve is a dispute between communications providers (CPs) relating to network access, and that the matters in dispute would not be resolved through further negotiation between the parties. On 16 December 2008, Ofcom decided that it was appropriate for it to handle this dispute for resolution. Ofcom informed the parties of this decision and published details of the dispute on its website.

Scope of the dispute

3.5 The scope of the dispute was to determine the termination rate payable by T-Mobile for voice calls originating on T-Mobile’s network and terminating on C&W’s network in respect of C&W customers using the FMC service.

3.6 In line with its standard procedures in disputes, Ofcom invited comments from stakeholders on the scope of the dispute as originally published. T-Mobile provided a response with its views on the scope of this dispute on 3 December 2008.

3.7 This Determination only applies to the scope of this dispute.

12 The proposed rates were: Daytime Traffic [ ]; Evening Traffic [ ]; and Weekend Traffic [ ].


6

Stakeholders interested in the outcome of the dispute

3.8 Upon opening this dispute for resolution, Ofcom invited interested stakeholders to express an interest in the outcome of this matter. Three stakeholders expressed an interest in the outcome of this dispute:

• Hay Systems Ltd;

• Hutchison 3G; and

• Stour Marine Ltd.

Further information provided by the parties

3.9 Ofcom invited T-Mobile to comment on C&W’s submissions. T-Mobile provided a response with its views on the scope of this dispute on 3 December 2008, followed by its initial response to the referred dispute on 9 January 2009.

3.10 On 8 January 2009, Ofcom sent C&W a notice under section 191 of the 2003 Act requiring it to provide documents and information in connection with this dispute. C&W responded to Ofcom’s notice on 14 January 2009. A second formal notice was sent to C&W on 2 February 2009. C&W responded to Ofcom’s second notice on 4 February 2009.

3.11 We requested further clarification of C&W on 9 February and 19 February 2009. We received this information on 9 February and 2 March 2009 respectively. We held a meeting with C&W on 27 January and two telephone discussions on 6 February and 3 March 2009.

The submissions of the parties

3.12 This section outlines the parties’ arguments presented prior to the issue of the Consultation.

C&W’s arguments

3.13 The following paragraphs set out C&W’s arguments. For the purpose of summarising C&W’s arguments we have adopted the headings used in C&W’s submission.

Relevant products or services

3.14 C&W submit that the relevant service is mobile call termination (MCT). C&W describe MCT as the service necessary to connect a caller with the intended recipient of the call on a different network.

3.15 C&W argue that MCT is the relevant service in this dispute because the FMC service to which the rates in this dispute relate is provided using mobile handsets and telephone numbers allocated for use with mobile communication services.

The ex-ante conditions to which this dispute relates

3.16 C&W submit that it does not have significant market power (SMP) and neither party to the dispute has any SMP conditions that relate directly to the dispute. C&W noted that it is not subject to any regulatory obligations to set its rates in accordance with charge control obligations. C&W also note that T-Mobile and the four other MNOs


7

have SMP in the market for wholesale MCT on their own networks and submit that the background to that regulation is therefore relevant to this dispute.

3.17 C&W outlined the market reviews that have taken place in relation to the provision of MCT. C&W drew our attention in particular to the second review of the mobile termination market contained in Ofcom’s “Mobile Call Termination Statement” published on 27 March 2007 (the Mobile Call Termination Statement).13

3.18 C&W observed that the Mobile Call Termination Statement did not make a determination of SMP in respect of any provider of MCT services other than the 5 MNOs.

C&W observed Ofcom’s finding of SMP for each of the five MNOs and the imposition of charge controls for the supply of MCT by each of the five MNOs.

3.19 C&W noted that the Mobile Call Termination Statement included a cost model to identify the costs of providing MTR which could be recoverable as part of the termination charge (the long run incremental cost (LRIC) model). A range of network costs were identified as costs which could be recovered as part of the MTR. C&W submitted that H3G’s costs stemming from its national roaming agreements were not included in the LRIC model because at the time of the market review H3G had already significantly extended its network coverage and was relying less on its national roaming agreement to terminate traffic. According to C&W, given the size of the volume of traffic being terminated by means of national roaming, no allowance for the recovery of national roaming costs was granted by Ofcom in setting the termination rates of H3G.

3.20 C&W also referred to the appeals that have been lodged with the Competition Appeal Tribunal (CAT) against a number of determinations of disputes which had previously been referred to Ofcom by BT and various mobile operators relating to termination rates: T-Mobile (UK) Limited and others v Office of Communications (the TRD Judgment).14

Appropriateness of the LRIC model

In C&W’s view, the TRD Judgment was clear that Ofcom must have regard to its overriding duties under sections 3 and 4 of the 2003 Act. C&W submitted that, in deciding what was a fair and reasonable rate, Ofcom should have regard not only to what is fair and reasonable as between the parties, but also must have regard to what is reasonable within the context of carrying out its wider regulatory duties.

3.21 C&W submitted that the same LRIC model used for setting termination rates for established operators with SMP is not appropriate for setting rates for a new service such as its FMC product before a market review or a finding of SMP. The basis for this view was because C&W is not subject to an SMP determination and the service in question is nascent, with subscriber take-up and costs uncertain. C&W have suggested that, given that no market failure has been identified, a LRIC model is not required and that imposing such a measure would be disproportionate.

3.22 C&W also indicated that applying a LRIC model to new market entrants would be inconsistent with Ofcom’s previous policy regarding the regulation of new services or new entrants. C&W drew comparisons with Ofcom’s approach to regulation of 3G

13 Mobile Call Termination Statement, 27 March 2007: http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf 14 [2008] CAT 12, 20 May 2008, available at: http://www.catribunal.org.uk/files/Judgment_TRDs_200508.pdf.

http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf�

http://www.catribunal.org.uk/files/Judgment_TRDs_200508.pdf�


8

services when charge controls were not initially imposed on 3G call termination, but were only applied once the market had developed and costs were more certain.

3.23 C&W argued that the determination of charges based upon LRIC cost models would act as a constraint on C&W’s ability to recover its start-up costs relating to this service. C&W contended that the ability to recover start-up costs is an essential element of allowing competition and innovation in the market.

3.24 C&W added that, given that the service has been recently launched, it is difficult to calculate the costs of the service as it is new and has uncertain economics. C&W concluded that, even if it were practical to develop a robust LRIC model, it would not be proportionate when set against the small traffic volumes that will be carried in the early phases of the product’s life.

The inclusion of national roaming charges within a LRIC model

3.25 C&W utilise a national roaming agreement to provide coverage outside of its network deployed for use within the office. C&W submitted that, whilst it does not agree that LRIC should be applied to the FMC service, in the event that Ofcom does decide, either now or in the future, that it should base FMC termination rates on the cost of the service, it would not be appropriate to exclude roaming charges, as they form a necessary and key cost of providing the service.

3.26 In C&W’s view, the fact that national roaming charges were not included in the cost model in the Mobile Call Termination Statement does not mean that such services should not be recoverable as part of the termination rates for its FMC service. C&W explained that national roaming charges were not included in the cost model for H3G because, as the network coverage increased, reliance on 2G roaming was declining. It was anticipated that by the end of the charge control period the number of calls terminating using the partner network would be very small.

3.27 C&W contrast its FMC service against the service provided by H3G, specifying that its FMC service will not be rolled out nationally and that the proportion of national roaming will not reduce as subscriber numbers on the mobile network increase. C&W contend that national roaming is a direct cost of the provision of the FMC service and as such should be recoverable in a cost model in the same way as other direct costs are. C&W add that it is essential that any cost model should take account of different services and the different costs associated with each service.

3.28 C&W have submitted that a MTR which did not allow for recovery of roaming charges would likely mean that C&W would make a loss on a significant number of calls made to its FMC customers. C&W stated that the provision of a service, which for at least a significant period of time results in a considerable loss, would not be commercially viable, resulting in its exit from the market for the provision of the FMC in the long run.

3.29 C&W conclude their submissions on national roaming by describing it as an efficiently incurred cost. C&W have stated that the use of a national roaming agreement is the most efficient way to provide the FMC service. C&W submit that its only alternative would be to roll out a national network using its guard band spectrum, which would be impractical and result in significantly higher costs and hence even higher MTR. On that basis, C&W contend that roaming charges are a legitimate and efficiently incurred cost and should be recoverable as part of its MTR.


9

Benchmarking

3.30 C&W quoted the TRD Judgment, where it stated that:

“…Benchmarking is a useful tool and Ofcom should consider the value of comparisons put forward by the parties and what they show about the reasonableness of the charges or other terms and conditions” (paragraph 186).

3.31 It is C&W’s position that benchmarking will assist Ofcom to determine a fair and reasonable rate in the current dispute. C&W observe that the end user will not perceive any difference between the mobile termination service provided by the FMC service and that provided by the other MNOs. C&W noted that there are a number of different benchmarks that reflect different sizes of operation, different phases in service life cycle and different technologies used.

3.32 C&W observed that some of the most obvious termination rates to be used as benchmarks had been set in accordance with price controls. These price controls, in C&W’s view, were set after extensive review of their specific circumstances of the operators concerned and taking into account their ability to drive greater efficiency during the charge control period. C&W distinguished its services in this instance, due to being a new entrant without a price control, and warned of the potential regulatory burdens that would follow from linking a new entrant’s rate to a regulated rate, imposed on an operator following a finding of SMP.

3.33 C&W submitted that its proposed MTR (6.418ppm across each time period) is fair and reasonable as it falls within the range of termination rates that currently exist15

3.34 C&W note that the termination rate of H3G is higher in part due to its greater use of expensive 3G spectrum which is not relevant to the FMC service. C&W claims that it is relevant to consider the other factors explaining their higher termination rate, namely that they are comparatively new in the market with smaller market shares than their rivals.

in the market for wholesale MCT. C&W observes that the four 2G/3G mobile operators are well established and have seen their termination rates fall significantly since the launch of their respective services. C&W contends that the fact that its proposed termination rate is only slightly above the other mobile operators indicates that this proposed MTR is not too high.

3.35 A particular benchmark that C&W identified is the rate applicable to the “BT Fusion” service provided by BT (identical to the MTR proposed by C&W – i.e. an MTR of 6.418ppm across each time period). C&W observed that the rate has been agreed by other CPs as being fair and reasonable. C&W added that, whilst BT Fusion’s termination rate is higher than the 2G/3G MNO rate, it has not necessitated an increase in retail prices for calls to different mobile networks to take into account the cost of calls to the BT Fusion product. C&W suggested that, if this same rate is charged for its FMC service, that similarly it is unlikely to give rise to a need for any changes in retail rates.

Current position

3.36 C&W told us that an agreement on termination rates between the parties to the dispute was required to open up the C&W number range on its network. Additionally,

15 As of the date of C&W’s submission on 24 November 2008.


10

T-Mobile made the agreement on termination rates a precondition of the mobile number portability (MNP) agreement between T-Mobile and C&W. C&W submitted that, in order to prevent a delay of the FMC service, or launch without connectivity to and from T-Mobile, an interim rate was agreed between T-Mobile and C&W at the level proposed by T-Mobile (1.2ppm). Both parties have agreed that the rate eventually determined by Ofcom will be applied retrospectively from the date the interim rate was agreed.

How the dispute should be resolved

3.37 C&W request that Ofcom resolve the dispute using their powers under section 185 of the 2003 Act to:

• determine that the rates proposed by C&W are fair and reasonable and that they be applied from 8 September 2008 until the date of the determination and thereafter until such time as the parties agree otherwise; or

• in the alternative, determine the proper amount of the charges for termination provided by C&W from the 8

September 2008 to the date of the determination

and thereafter until such time as the parties agree otherwise;

• for the purposes of giving effect to that determination, give a direction requiring T-Mobile to pay the termination rates as proposed by C&W, or such other amount as Ofcom may determine, plus interest to C&W by way of an adjustment to the underpayment made by T-Mobile.

T-Mobile’s arguments

3.38 The following paragraphs set out T-Mobile’s arguments prior to the issue of the Consultation. Note that, for the purpose of summarising T-Mobile’s arguments, we have adopted the headings used in T-Mobile’s submission.

The scope of the dispute

3.39 T-Mobile suggested that the scope of the dispute be defined as the determination of the level of the termination rate charged for the termination for calls originated on T-Mobile’s network and terminated on C&W’s network.

Market review

3.40 T-Mobile considered that the dispute relates in principle to the application or otherwise of cost based termination charges. T-Mobile did not propose that C&W be made subject to an SMP designation or that its termination rates should be assessed according to the strict application of the LRIC model. Rather, T-Mobile contended that termination rates should reflect the costs of termination if they are to be reasonable. T-Mobile referred to the TRD Judgment where the CAT stated that a full SMP review was not required in order to resolve a dispute. T-Mobile submitted that, when it purchases a service, that service should be priced at a reasonable rate and that it should therefore substantially relate to the costs of providing it.

3.41 T-Mobile has not suggested that C&W should set its termination rate in strict compliance with the LRIC model or that its MTR should be subject to a price control regulation through an SMP designation. T-Mobile referred to the European Commission’s Recommendation on Market Definition and Ofcom’s previous analysis of the market for voice call termination on individual mobile networks (former Market


11

16, now Market 7)16

3.42 T-Mobile submitted that it follows that each network, including C&W, prima facie enjoys SMP in respect of call termination on its network.

. T-Mobile observed that Ofcom’s analysis concluded that there are separate relevant markets for MCT on each MNO’s network, regardless of the technology used for termination; and that all MNOs have 100 % share of the market for termination on their own network.

3.43 T-Mobile did not suggest that Ofcom should conduct a market review and make an SMP determination in respect of C&W, or that such an exercise is required to determine the dispute. T-Mobile submitted that Ofcom’s powers of dispute resolution under section 185 of the 2003 Act are distinct from its ability to set price controls pursuant to a market review, although each enables it to determine an appropriate MTR.

The nature of the C&W FMC service

3.44 T-Mobile submitted that C&W is not proposing a new or innovative service in that it is a 2G GSM service that utilises the same technology, handsets, cells, etc as existing 2G mobile services. T-Mobile do not characterise the C&W product as a “convergence” of fixed and mobile services, as it uses two mobile networks to provide coverage. T-Mobile also suggests that there is nothing novel about the combination of different networks to provide a service. T-Mobile described the use of pico cells for localised coverage combined with the use of macro cells for wider coverage as an approach that has been adopted by the five national MNOs for some time.

3.45 T-Mobile notes that the C&W FMC solution is based on existing technology and that it is aimed at corporate customers, not the general consumer, and is not intended by C&W to be a public consumer service.

Principles that apply to the regulation of new entrants’ termination rates

3.46 T-Mobile proposes that Ofcom continue to approach the resolution of disputes regarding the reasonableness of termination rates in a manner that is cost related as per the TRD Judgment.

3.47 T-Mobile rejected the suggestion by C&W that, as a result of their new entrant / innovative service status, there should be some form of regulatory forbearance. T-Mobile suggests that any issues arising from new entrant status can adequately be taken into account by Ofcom.

16 As provided for in the Framework Directive (Directive 2002/21/EC), the European Commission (the “Commission”) has adopted a Recommendation on relevant products and services markets (“the Recommendation”) which identifies markets within the electronic communications sector, the characteristics of which may be such as to justify the imposition of regulatory obligations. The Recommendation’s Market 7 (former Market 16) is the market for voice call termination on individual mobile networks: “Recommendation on relevant product and service markets within the electronic communications sector susceptible to ex ante regulation in accordance with Directive 2002/21/EC of the European Parliament and of the Council on a common regulatory framework for electronic communication networks and services” http://eur-lex.europa.eu/LexUriServ/site/en/oj/2007/l_344/l_34420071228en00650069.pdf:.


12

The costs that are relevant to the determination of termination rates

3.48 T-Mobile outlined its view that allowing asymmetric termination rates must be justified and that the justification can no longer be based on “new entrant” status per se, but must be related to differences in costs. T-Mobile referred to a paper of the European Regulators Group (ERG) where the ERG did not support the contention that asymmetry and higher call termination rates should be logically allowed for new entrants or smaller players.17 T-Mobile also indicated that the paper stated that, if asymmetric rates were imposed, they should only reflect differences in costs.18

National roaming

3.49 T-Mobile submitted that the choice to provide a national roaming service outside of their network is a matter for C&W and its customers. T-Mobile contended that “recipient network/its customer pays” (for the additional costs of roaming) is an internationally recognised model in call forwarding for both GSM networks and VoIP operators. T-Mobile argued that this is appropriate because the originating network (i) does not know which network the call is ultimately routed to and (ii) should not be required to pay for the use of two networks, where roaming is involved at the receiving party’s/network’s election. Put another way, T-Mobile argues that the roaming charge should not be included in C&W’s termination charge as this would effectively result in this charge being passed to the caller (not the recipient).

3.50 T-Mobile pointed to the setting of the H3G price control, which takes no account of national or international roaming. T-Mobile highlighted a specific category of “other costs”, including roaming and interconnect charges, which was used in Ofcom’s analysis of the Mobile Call Termination Statement. T-Mobile explained that Ofcom did not find these costs relevant to the consideration of the cost of call termination, as they were not costs that were incurred in relation to the MNO’s own network (at paragraph A6.20).

3.51 T-Mobile considered that it is possible, if the MTR was set at the level of roaming costs as proposed by C&W, for C&W to earn excessive profits on the core service of termination on C&W’s spectrum. T-Mobile point to the direct harm to callers of excessive termination charges through higher retail charges contending that the recovery of C&W’s roaming costs would risk distortions to competition, such as allowing C&W to cross-subsidise its retail prices in competition with other operators. T-Mobile suggest that this would enable C&W to expand through artificial regulatory advantage rather than from offering better services or better value to customers.

3.52 T-Mobile submitted that C&W appear to make no allowance for the fact that, despite justifying their high termination rate largely on the basis of national roaming, a significant proportion of their traffic will be terminated by themselves, without these roaming costs. T-Mobile suggested that until C&W has been operational for a period of time and can generate robust figures for the amount of traffic terminated on either network, it will be very difficult to attempt to blend the costs of termination on the C&W network and the costs of termination on their partner’s 2G network. T-Mobile surmised that the only appropriate method to determine this matter is to pay the cost of calls terminated on C&W’s network only.

17 “ERG’s common position on symmetry of fixed call termination rates and symmetry of mobile call termination rates” (ERG(07)83): http://www.erg.eu.int/doc/publications/erg_07_83_mtr_ftr_cp_12_03_08.pdf 18 Ibid pages 5 and 85.

http://www.erg.eu.int/doc/publications/erg_07_83_mtr_ftr_cp_12_03_08.pdf�


13

3.53 T-Mobile concluded that the originating network (i.e. T-Mobile) should not pay the costs of roaming and accordingly these costs should not form part of the assessment of network costs when determining an appropriate termination rate. T-Mobile suggested that a national network is not required by C&W, and that C&W should charge lower termination rates to reflect the lower costs of network access provision. T-Mobile contend that if the C&W FMC service is indeed innovative and efficient, then C&W must be able to propose retail costs significantly below those of the MNOs, and in that way enjoy a margin to recover the additional associated costs of national roaming, rather than recovering these costs from its competitors.

Benchmarking

3.54 T-Mobile submitted that the CAT does not regard benchmarking against the price control caps of operators such as T-Mobile to be appropriate when determining the reasonableness of termination rates, particularly in situations where this would maintain termination rates significantly above costs and deny purchasers the benefits of potential reductions.

3.55 T-Mobile added that, to the extent that any form of benchmarking is appropriate, the relevant benchmark should not be the current termination rates of the MNOs. T-Mobile submitted that none of these companies provide only localised coverage using a DECT guard band network. T-Mobile stressed that, if benchmarking is used to establish the appropriateness of the C&W rates, then these rates must be benchmarked against those for similar operators.

3.56 T-Mobile rejected comparisons of the C&W FMC service with the BT Fusion service. It argued that comparisons were inappropriate because the technology of BT Fusion never fully functioned and calls to BT Fusion numbers were delivered via BT’s roaming partner, Vodafone.

The likely costs of voice call termination on the C&W network

3.57 T-Mobile’s assessment of the reasonable costs incurred for the MTR for a DECT guard band operator is 1.2ppm.

3.58 T-Mobile indicated that C&W has not invested, nor intends to invest, in a national network on the scale of the established MNOs. C&W has no coverage obligation, but intends to restrict its rollout to specific sites of pre-acquired customers.

Network costs

3.59 T-Mobile contended that the costs of a network rollout at a local level for a DECT guard band network are not comparable to those of the mainstream MNOs, even accounting for scale/coverage. T-Mobile outlined that the relevant network costs include: site costs, equipment costs, backhaul, buildings and RAN planning.

3.60 T-Mobile contended that the GSM licence for DECT guard band spectrum was acquired for relatively little cost by C&W and does not currently attract Administered Incentive Pricing (AIP).

Share of spectrum licence costs associated with termination

Share of administrative cost associated with termination


14

3.61 T-Mobile suggested that there are no indicators that C&W’s administrative costs should not be compared to that of T-Mobile.

3.62 T-Mobile argued that they and other GSM MNOs are granted inclusion of a network externality surcharge (at the time of T-Mobile’s submission). This surcharge aims to rectify the externality that occurs because of the social benefit that occurs when individuals have a mobile phone and other people are able to contact them. According to T-Mobile, Ofcom has explained that the main purpose of their network externality in the UK surcharge is now not to increase the penetration of mobile phones, but rather to help subsidise mobile phones for current subscribers who are unwilling to pay enough to renew their subscription.

Network externality surcharge

3.63 T-Mobile understands that C&W are not aiming to increase the number of mobile phone subscribers in the UK, but rather are aiming to provide a replacement service using the existing mobile subscribers of alternative national operators, using a SIM to be inserted in the subscriber’s existing handset. T-Mobile submits that there is no social benefit in the form of a network effect that occurs when C&W increase their subscription base, as these customers are already contactable through their existing providers. T-Mobile therefore do not consider that a network externality surcharge is appropriate for the C&W service and rates this as zero in comparing the two companies’ termination costs.

The sustainability of the service

3.64 T-Mobile addressed C&W’s assertions that its services will not be sustainable at lower termination rates. T-Mobile disagreed with this argument on the following basis:

• T-Mobile argued that the termination rate suggested by C&W has been set through benchmarking, in which rates are set independently of costs and profitability and are not calculated on the basis of meeting a particular threshold for commercial sustainability.

• According to T-Mobile, this implied that there is no flexibility in C&W’s business case for the product (i.e. such as an allowance for an increase in call origination charges). If this is the case, T-Mobile suggested that it may be indicative that the product or service is inefficient, since it is not competitive at marginal costs. T-Mobile stated that it ought not to be required to subsidise an inefficient new service or fund competitors’ initial investment.

• C&W asserted that it does not benefit from the scale efficiencies of the incumbent operators. In T-Mobile’s view, this overlooks the substantial size of C&W.T-Mobile also submit that C&W’s assertion downplays that, although its FMC product is a new product for C&W, it does not require an entirely separate business to be established. T-Mobile argued that, even if it could be established that the FMC service derived no benefit from its place within the C&W group, Ofcom’s approach to the initial inefficiency of networks should remain the same and its method of Economic Depreciation should be used to deal with the issue of any under-utilisation of the C&W network.

Number portability

3.65 T-Mobile understood that C&W propose to acquire a significant proportion of their customers through number portability. In the case of each such customer the


15

recipient operator network will receive the termination rate of the donor operator network. T-Mobile does not propose how this revenue is accounted for in its proposed MTR. T-Mobile submitted that Ofcom should account for the expected revenues that would be received by C&W through ported numbers where they receive regulated terminated rates irrespective of the network that the call terminates on.

Manner of implementation

3.66 If Ofcom’s determination adjusted the current agreed interim rate, T-Mobile did not consider that any determination of any balance payment or interest to be necessary or appropriate as the parties have expressly agreed that the interim MTR was subject to the possibility of making a retrospective adjustment payment in the light of Ofcom’s determination in due course. Furthermore, T-Mobile submitted that its customers make no calls to C&W numbers and accordingly any adjustment payment is likely to be very low and a retrospective assessment is not a matter that it would be proportionate for Ofcom to determine.


16

Section 4

4 Ofcom’s statutory obligations and regulatory principles 4.1 The starting point for the resolution of any dispute is for us to consider our dispute

resolution powers, statutory obligations and regulatory principles. This section sets out those obligations and principles which are then taken into account in the resolution of this dispute in section 5.

4.2 Sections 185 to 191 of the 2003 Act set out our dispute resolution powers. They apply to disputes relating to the provision of network access and to other disputes relating to the rights and obligations conferred or imposed by or under Part 2 of the 2003 Act. Section 186 of the 2003 Act requires us to resolve a dispute referred which meets the requirements of section 185. Our powers to impose remedies to resolve disputes are set out in section 190 of the 2003 Act.

4.3 Our dispute resolution powers in the 2003 Act derive from the European Common Regulatory Framework, in particular, the Framework Directive and the Access Directive.19

4.4 Article 5(4) of the Access Directive is implemented through the dispute resolution procedures set out in section 185 to 191 of the 2003 Act and Article 8 of the Framework Directive has been implemented in section 4 of the 2003 Act. Under section 4(2) of the 2003 Act, we are required to act in accordance with the six Community requirements when exercising our functions under the 2003 Act in relation to disputes referred to it under section 185. The six Community requirements set out in section 4(3) – (10) give effect, amongst other things, to the requirements of Article 8 of the Framework Directive and are to be read in accordance with them.

In accordance with Article 5(4) of the Access Directive, Ofcom is required to resolve disputes in relation to access and interconnection in accordance with the policy objectives of Article 8 of the Framework Directive.

4.5 In summary, the Community requirements are requirements:

• to promote competition in communications markets.

• to ensure that Ofcom contributes to the development of the European internal market;

• to promote the interests of all European Union citizens;

• to act in a manner which, so far as practicable, is technology-neutral;

• to encourage, to the extent Ofcom considers it appropriate, the provision of network access and service interoperability for the purposes of securing efficiency and sustainable competition in communications markets and the

19 Directive 2002/21/EC of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive): http://eur-lex.europa.eu/pri/en/oj/dat/2002/l_108/l_10820020424en00330050.pdf; Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access to, and interconnection of, electronic communications networks and associated facilities (Access Directive): http://eur-lex.europa.eu/pri/en/oj/dat/2002/l_108/l_10820020424en00070020.pdf

http://eur-lex.europa.eu/pri/en/oj/dat/2002/l_108/l_10820020424en00330050.pdf�

http://eur-lex.europa.eu/pri/en/oj/dat/2002/l_108/l_10820020424en00070020.pdf�


17

maximum benefit for the customers of communications network and services providers; and

• to encourage such compliance with certain international standards as is necessary for facilitating service interoperability and securing freedom of choice for the customers of CPs.

4.6 In the context of this dispute, the following aspects of the policy objectives of Article 8 of the Framework Directive are of particular note:

• the promotion of competition is to be achieved by, inter alia, ensuring that users derive maximum benefit in terms of choice, price and quality and that there is no distortion or restriction of competition; and

• the contribution to the development of the internal market is to be achieved by, inter alia, ensuring that, in similar circumstances, there is no discrimination in the treatment of undertakings providing electronic communications networks and services.

4.7 Section 3 of the 2003 Act sets out our general statutory duties which must be taken into account in carrying out our dispute resolution function under Chapter 3 of Part 2 of the 2003 Act.

4.8 Section 3(1) of the 2003 Act sets out our principal duties to be taken into account in carrying out our functions:

“(a) to further the interests of citizens in relation to communications matters; and

(b) to further the interests of consumers in relevant markets, where appropriate, by promoting competition.”

4.9 The things which, by virtue of its principal obligations, we are required to secure in the carrying out of our functions include, according to section 3(2) of the 2003 Act:

“(a) to the optimal use for wireless telegraphy of the electro-magnetic spectrum;

(b) the availability throughout the United Kingdom of a wide range of electronic communications services;

(c) the availability throughout the United Kingdom of a wide range of television and radio services which (taken as a whole) are both of high quality and calculated to appeal to a variety of tastes and interests;

(d) the maintenance of a sufficient plurality of providers of different television and radio services;

(e) the application, in the case of all television and radio services, of standards that provide adequate protection to members of the public from the inclusion of offensive and harmful material in such services; and


18

(f) the application, in the case of all television and radio services, of standards that provide adequate protection to members of the public and all other persons from both –

(i) unfair treatment in programmes included in such services; and

(ii) unwarranted infringements of privacy resulting from activities carried on for the purposes of such services.”

4.10 Section 3(3) of the 2003 Act provides that in performing our principal duties, we must have regard, in all cases, to:

“(a) the principles under which regulatory activities should be transparent, accountable, proportionate, consistent and targeted only at cases in which action is needed; and

(b) any other principles appearing to Ofcom to represent the best regulatory practice.”

4.11 Section 3(4) of the 2003 Act sets out a number of principles which we must have regard to in performing our principal duties where it appears to Ofcom that they are relevant, including the desirability of promoting competition in the relevant markets and the desirability of encouraging investment and innovation in the relevant markets.

4.12 In performing the principal duty of furthering the interests of consumers specifically, section 3(5) of the 2003 Act provides that Ofcom must have regard, in particular, to the interests of those consumers in respect of choice, price, quality of service and value for money.

4.13 Where it appears to us that any of our general duties under section 3 of the 2003 Act conflict in a particular case, we must secure that the conflict is resolved in the manner we consider best in the circumstances.20 Similarly, we must secure that any conflict between the Community requirements set out in section 4 of the 2003 Act is resolved in the manner we consider best in the circumstances.21 Where it appears that a general duty under section 3 of the 2003 Act conflicts with one or more duties under section 4 of the 2003 Act, priority is given to the duties set out in section 4 of the 2003 Act.22

4.14 We also exercise our regulatory functions according to the following regulatory principles:

23

• We will regulate with a clearly articulated and publicly reviewed annual plan, with stated policy objectives;

• We will intervene where there is a specific statutory duty to work towards a public policy goal which markets alone cannot achieve;

• We will operate with a bias against intervention, but with a willingness to intervene firmly, promptly and effectively where required;

20 Section 3(7) of the 2003 Act. 21 Section 4(11) of the 2003 Act. 22 Section 3(6) of the 2003 Act. 23 http://www.ofcom.org.uk/consult/condocs/plan/annual_plan/regulatory_principles.pdf

http://www.ofcom.org.uk/consult/condocs/plan/annual_plan/regulatory_principles.pdf�


19

• We will strive to ensure its interventions will be evidence-based, proportionate, consistent, accountable and transparent in both deliberation and outcome;

• We will always seek the least intrusive regulatory mechanisms to achieve its policy objectives;

• We will research markets constantly and will aim to remain at the forefront of technological understanding; and

• We will consult widely with all relevant stakeholders and assess the impact of regulatory action before imposing regulation upon a market.


20

Section 5

5 Ofcom’s analysis and proposed decision Which are the relevant factors to be considered in order for Ofcom to resolve this dispute?

5.1 After consideration of submissions received from the parties and evidence gathered during the dispute, Ofcom sent the Consultation to the parties to the dispute and to three interested parties on 20 March 2009 and published the Consultation on our website on 23 March 2009. The Consultation set out Ofcom’s preliminary conclusions on resolution of the dispute together with Ofcom’s analysis and reasoning in reaching its provisional conclusions.

5.2 For clarity, this section sets out the original analysis and reasoning underpinning Ofcom’s draft determination (which also appeared at section 5 of the Consultation). Any subsequent changes to our analysis and reasoning are set out and explained in section 6.

5.3 Having taken account of representations from the parties and our duties as set out in the 2003 Act, we have considered the appropriate means of resolving this dispute. We decided that the 6 principles of pricing and cost recovery established by Ofcom were the most appropriate framework for resolving this dispute because they would allow us to adequately ensure our regulatory objectives by accurately weighing up a number of factors, including costs, distribution of benefits and competitive effects, which are relevant for us to set a MTR that is reasonable as between the parties and satisfies our duties. We noted that these are also important factors to be considered in order to ensure that our determination will further the interests of citizens in relation to communication matters, as well as the interests of consumers in the relevant markets, where appropriate by promoting competition. Furthermore, we noted that the 6 principles provided for an appropriate set of objective criteria against which we could comparatively assess the possible options that we had identified to determine this dispute, including the different MTRs suggested by T-Mobile and C&W respectively. Finally, we also considered that the 6 principles attach some importance to practicability, which is also a relevant factor for us in order to meet our statutory duty to resolve this dispute within 4 months and consider the ease of implementation of the outcome as between the parties.

5.4 From that basis, we have also considered the extent to which any other factors may be relevant to the outcome of the dispute which would more adequately ensure our regulatory objectives. We have then considered the extent to which our proposed outcome is consistent with our statutory duties.

The six principles of pricing and cost recovery

5.5 The six principles of pricing and cost recovery were developed by Oftel in the context of number portability, endorsed by the Monopolies and Mergers Commission24 and have subsequently been used by Ofcom in analysing various pricing issues25

24 Telephone Number Portability: A Report on a reference under s13 of the Telecommunications Act 1984 (MMC, 1995):

,

http://www.competition-commission.org.uk/rep_pub/reports/1995/374telephone.htm#full 25 See for example: ‘Determination under Section 190 of the Communications Act and Direction under Regulation 6(6) of the Telecommunications (Interconnection) regulations 1997 for resolving a dispute

http://www.competition-commission.org.uk/rep_pub/reports/1995/374telephone.htm#full�

http://www.competition-commission.org.uk/rep_pub/reports/1995/374telephone.htm#full�


21

including setting charges for CPS26, the 2006 WLR charge setting exercise27 and the resolution of a dispute between BT and Telewest about the geographic call termination reciprocity agreement28

5.6 The six principles of pricing and cost recovery are:

.

i) Cost causation: costs should be recovered from those whose actions cause the costs to be incurred;

ii) Cost minimisation: the mechanism for cost recovery should ensure that there are strong incentives to minimise costs;

iii) Effective competition: the mechanism for cost recovery should not undermine or weaken the pressures for effective competition;

iv) Reciprocity: where services are provided reciprocally, charges should also be reciprocal;

v) Distribution of benefits: costs should be recovered from the beneficiaries especially where there are externalities; and

vi) Practicability: the mechanism for cost recovery needs to be practicable and relatively easy to implement.

5.7 The application of any one of these principles to the relevant circumstances can sometimes point in a different direction to other principles. But the set of principles provides a framework to identify such trade-offs and to facilitate the use of judgement to strike an appropriate balance in reaching conclusions.

Other factors to be considered

5.8 We have also given consideration to the general guidance provided by the CAT in the TRD Judgment to determine a rate which is reasonable between the parties, in taking into account our statutory duties. We have therefore considered the following in reaching our draft determination:

i) an analysis of each side’s argument for a particular rate; between Orange Personal Communications Services Ltd. (‘Orange’) and British Telecommunications plc (‘BT’) concerning the cost sharing arrangements for Customer Sited Interconnect (‘CSI’) links connection and rental charges’, 19 November 2003: http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_663/. See also ‘Direction concerning ADSL Broadband Access Migration Services, 13 May 2004: http://www.ofcom.org.uk/consult/condocs/bam/statement/; and a Determination to resolve a dispute between Tiscali, Thus and BT concerning ADSL Broadband Access Migration Services’, 9 August 2004: http://www.ofcom.org.uk/consult/condocs/bam/statement/. Determination to resolve a dispute between Opal Telecom and British Telecommunications PLC (Openreach) about LLU bulk migration charges, 2 June 2006: http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_889/determin/determination.pdf. 26 Final Determination on costs and charges for the provision of permanent carrier pre selection - 2 September 2002: http://www.ofcom.org.uk/static/archive/oftel/publications/carrier/2002/pcps0902.htm 27Wholesale Line Rental: Reviewing and setting charge ceilings for WLR services, http://www.ofcom.org.uk/consult/condocs/wlrcharge/statement/statement.pdf 28 Final determination published on 16 June 2006: http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_890/determination.pdf

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_663/�

http://www.ofcom.org.uk/consult/condocs/bam/statement/�

http://www.ofcom.org.uk/consult/condocs/bam/statement/�

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_889/determin/determination.pdf�

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_889/determin/determination.pdf�

http://www.ofcom.org.uk/static/archive/oftel/publications/carrier/2002/pcps0902.htm�

http://www.ofcom.org.uk/consult/condocs/wlrcharge/statement/statement.pdf�

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_890/determination.pdf�

http://www.ofcom.org.uk/bulletins/comp_bull_index/comp_bull_ccases/closed_all/cw_890/determination.pdf�


22

ii) an assessment of costs; and

iii) the relevance of any benchmarks.

5.9 Our approach, as set out below, provides an analysis of the arguments of C&W and T-Mobile for an MTR of 6.418ppm (across each time period) and 1.2ppm respectively. We also conducted an assessment of the relevant costs and considered whether there are any relevant benchmarks which could be applied to this case.

5.10 In assessing costs in this case, we have conducted an assessment of C&W’s costs and of the likely efficient level of termination costs for an operator such as C&W providing a service similar to the FMC service with national coverage. To gain a better understanding of the costs C&W incurs when a call to its FMC service is terminated, we completed a high level assessment of termination costs based on C&W’s business plan. We also assessed the evidence of C&W’s costs submitted by T-Mobile. In addition to this, we compared the outcome of our analysis with the efficient unit cost of termination by a national operator as estimated by the MCT cost model (see Annex 1 for further information).

Assessment of Costs

5.11 With regard to benchmarking, at paragraph 186 of the TRD Judgment the CAT stated that:

Benchmarks

“Benchmarking is a useful tool and Ofcom should consider the value of comparisons put forward by the parties and what they show about the reasonableness of the charges or other terms and conditions being proposed.”29

5.12 The parties have provided their submissions on the usefulness or otherwise of other termination rates to be used as a benchmark.

5.13 C&W contended that benchmarking will assist Ofcom to determine a fair and reasonable rate in this dispute, encouraging Ofcom to consider useful comparisons – those that reflect different sizes of operation, different phases in service life cycle and different technologies used. A particular benchmark that C&W identified is that of the MTR charged by BT for terminating calls on BT’s network in respect of BT customers using its “BT Fusion” service (identical to the MTR proposed by C&W). C&W observed that the rate has been agreed by other CPs as being fair and reasonable. C&W add that whilst BT Fusion’s termination rate is higher than the 2G/3G MNO rate, it has not necessitated an increase in retail prices for calls to different mobile networks to take into account the cost of calls to the BT Fusion product. C&W suggest that if this same rate is charged for its FMC service, that similarly it is unlikely to give rise to a need for any changes in retail rates.

5.14 T-Mobile did not consider that a benchmark was appropriate in this matter. T-Mobile added that, to the extent that any form of benchmarking is appropriate, the relevant benchmark should not be the current termination rates of the MNOs. T-Mobile stressed that, if benchmarking is used to establish the appropriateness of the C&W MTRs, then these rates must be benchmarked against those operators using a similar technology. T-Mobile rejected the use of the BT Fusion service as a

29 Ibid.


23

benchmark due to issues with the functionality of underlying technology (see footnote 4). [ ]

5.15 In order to resolve this dispute, we have considered the relevance of benchmarks for the disputed termination charges. There are many different benchmarks for a MTR, both on a national basis and international basis.

National Benchmarks

5.16 We have considered whether a fixed termination rate would be a relevant benchmark in this matter. In this regard, we note that the C&W’s FMC service offers greater functionality than a fixed service as it can be used outside of the customers’ premises. Additionally, C&W’s FMC network inside its customers’ premises actually utilises a mobile technology (through its own pico cells and national roaming agreement) which means its traffic-sensitive costs are likely to be higher than those of a fixed network. Whilst some aspects of C&W's FMC service may be seen, and intended, as alternatives to fixed line services, this is also true of some services offered by the established MNOs. As a result, we do not consider a fixed termination rate to be a benchmark that would apply in this case due to the fundamental differences in the nature of the services offered by the C&W FMC service and a fixed network.

Fixed termination rate

5.17 We considered whether the BT Fusion rate (i.e. 6.418ppm across each time period) would be a relevant benchmark. C&W has identified the BT Fusion product as another fixed to mobile convergence solution and in C&W’s view a level of MTR which has been agreed by many other CPs as being fair and reasonable.

BT Fusion rate

5.18 We do not regard the BT Fusion termination rate as an appropriate benchmark because it is not subject to any regulatory control or assessment and therefore we have not determined the extent to which it reflects the incurred or efficient costs of the service. Further, we note that to date this product has not achieved commercial significance, which may also bring into question its relevance as a benchmark. We also consider that the underlying technology of the BT Fusion service significantly differs with the FMC offering of C&W. We consider them to be a different commercial proposition. The BT Fusion uses conventional GSM handsets which route calls via the fixed broadband network while the user is indoors, and hands the call over to the cellular network when the user is outdoors, whereas the FMC product uses pico cells operating over DECT guard band spectrum while the user is indoors and connectivity is provided over one of the existing mobile networks under a roaming agreement outside of the office.

5.19 As a result of these considerations, we do not regard the BT Fusion MTR as being an appropriate benchmark in this matter.

5.20 We then considered the extent to which regulated MTRs may be a relevant benchmark in this case.

A regulated MNO rate


24

5.21 We have taken account of the general guidance provided by the TRD Judgment30

5.77

, and have approached with caution the option of benchmarking the MTR for terminating calls to customers of the FMC service against the regulated MTRs applying to the established MNOs. In particular, we were mindful to ensure that C&W would retain incentives to minimise costs (see paragraphs to 5.90). Our considerations as to whether a regulated MTR is an appropriate benchmark are outlined in the following paragraphs.

5.22 We noted that the regulated rates may be an appropriate benchmark, since there are substantial similarities between the nature of the termination service provided by C&W and the services provided by the established MNOs:

i) it is expected that the C&W FMC service, as a result of its own network investment and utilising its national roaming arrangements, is a fully mobile service allowing its customers to make and receive calls while on the move in the same way as do the services offered by the five incumbent MNOs;

ii) in addition, the underlying technology of the C&W service is a cellular wireless technology like that of the other mobile services;

iii) C&W has obtained a spectrum licence, has been provided with a mobile number range, has undergone a mobile number portability process with the other MNOs; and

iv) in addition, the service is marketed to the consumer as a mobile service and calls are made and received on a mobile handset.

5.23 Since the regulated rates for the MNOs are designed to reflect the efficient costs of termination on a mobile network, they are likely to be relevant benchmarks in determining the appropriate level of MTR for C&W.

5.24 In particular, the Mobile Call Termination Statement concluded a market review into MCT charges which found that each of the 5 incumbent MNOs, namely Vodafone, Orange, T-Mobile, O2 and Hutchison 3G, had significant market power in the market for wholesale mobile voice call termination provided to other CPs by the relevant MNO in the United Kingdom.

5.25 Within the Mobile Call Termination Statement, we used a cost model to derive the cost to a network operator of providing voice termination services, using the cost standard of Long Run Incremental Cost plus mark-up to contribute to common costs recovery (LRIC+). We continue to hold the view that a LRIC+ methodology constitutes the most appropriate means of determining the efficient levels for charges on mobile voice call termination services. The primary objective of the model is to

30 See paragraph 186: “Benchmarking is a useful tool and Ofcom should consider the value of comparisons put forward by the parties and what they show about the reasonableness of the charges or other terms and conditions being proposed. Nevertheless, the Tribunal considers that benchmarking against a price control cap set as an SMP condition needs to be approached with caution. Price controls are set on the basis of information about costs available at the start of a period to be covered by a market review and such controls will extend over a number of years. The regulatory intention is that such controls encourage undertakings bound by them to reduce their costs over the period so as to maximise profits. Any such reductions in costs will then be taken into account when the controls are reviewed and revised for a subsequent period of years. It is important therefore not to allow benchmarking against actual or proposed price controls to be used in a way which deprives the undertakings of the benefits of cost reductions and other efficiency savings which such controls were intended to encourage.”


25

assess the network costs to a single network operator of delivering voice services over 2G and/or 3G mobile networks.

5.26 We therefore imposed SMP conditions on each of the 5 incumbent MNOs, including charge controls on the supply of MCT by each of the MNOs as from 1 April 2007. Those charge controls provided for an annual target average charge (TAC) of 9.1ppm for Hutchison 3G, 5.7ppm for O2 and Vodafone, and 6.2ppm for Orange and T-Mobile during the first year of operation of the controls. Annual TACs for subsequent years were to be reduced according to a glide path such that, by 2010/11, they would be equal to 5.9ppm for Hutchison 3G and 5.1ppm for each of O2, Vodafone, Orange and T-Mobile.31

5.27 In the Mobile Call Termination Statement, Ofcom concluded that the 2G/3G MNOs should be required to reduce their charges in subsequent years in line with a smooth glide path of four equal percentage reductions, the steps to be calculated with reference to the applicable TAC for the final year of the charge control (2010/11), taking the headline level of the charge controls currently in force.

5.28 The Mobile Call Termination Statement concluded that the appropriate MTR for Hutchison 3G should be a substantial charge reduction in 2007/8 followed in subsequent years by a smooth glide path, such that charges in 2010/11 align with the cost-based target for that year. The conclusions of the Mobile Call Termination Statement for the 5 incumbent MNOs are shown in Figure 1.

Figure 1: Table of charge control conclusions following adjustment for notice period in the Mobile Call Termination Statement

Average regulated charges in 2006/7

First year (2007/8) target charge (nominal)

Second year (2008/9) percentage reduction (i.e. X in RPI-X)

Third and fourth year (2009/10 and 2010/11) percentage reduction (i.e. X in RPI-X)

Final charge in 2010/11 (real 06/07 prices)

Vodafone and O2

5.6 5.7 3.2% 2.5% 5.1

T-Mobile and Orange

6.3 6.2 5.8% 5.3% 5.1

H3G Not regulated

9.1 15.1% 11.8% 5.9

Source: The Mobile Call Termination Statement 2007, Figure 9.6

5.29 We note in this regard that the Mobile Call Termination Statement is currently the subject of appeal proceedings before the CAT brought by Hutchison 3G and BT in May 2007.32

31 These charges for 2010/1 are expressed in real terms, 2006/7 prices. 32 Case 1083/3/3/07 and Case 1085/3/3/07.

We recognise therefore that caution must be applied when testing the regulated rates since, in the event that the current appeal proceedings are


26

successful, the target average charges (TACs) may differ from those which Ofcom has determined.

5.30 As part of the appeal proceedings concerning the Mobile Call Termination Statement, the CAT referred the price control matters related to the appeals to the CC in accordance with section 193 of the 2003 Act.

The CC Determination

5.31 On 22 January 2009, the CC issued the CC Determination, in which it set out its own determination of the price controls for the four years from 1 April 2007. The CC Determination stated that the TACs in 2010/11 should be 4.0ppm for O2, Orange, T-Mobile and Vodafone, and 4.4ppm for H3G in 2006/07 prices and for the preceding years as shown in Table 2.

Table 2: CC Determination revised charges (ppm in real terms, 2006/7 prices)

2007/08

2008/09 2009/10 2010/11

Vodafone & O2 (900/1800 MHz operators)

5.2 4.7 4.4 4.0

T-Mobile & Orange (1800-MHz-only operators)

5.7 5.0 4.5 4.0

H3G (3G only operator)

8.9 6.8 5.5 4.4

Source: Table 16.1 of the CC Determination.

5.32 We recognise that the CC Determination will not be finalised until the end of the litigation process. However, in considering relevant benchmarks, we consider that, for the purposes of determining the appropriate benchmark rate, the CC Determination represents the best estimate of the level of regulated termination charges. Ofcom is required by statute to resolve the dispute within four months, a period during which it appears unlikely that the process will be resolved definitively and we therefore consider it appropriate, in considering whether regulated rates are appropriate benchmarks, to use the outcome of the CC Determination in resolving this dispute.

5.33 In light of our conclusion that a regulated MTR could be a relevant benchmark in this case, we have therefore considered which rate of the three regulated TACs – the 900/1800 MHz operators; 1800-MHz-only operators; or, a 3G only operator – is the most appropriate benchmark.

5.34 In the CC Determination, differences were allowed for a later entrant, H3G, in terms of both costs (reflected in the difference in TACs in 2010/11) and a separate glide path. Whilst C&W is clearly a more recent entrant, still we believe that the CC Determination’s reasoning is consistent with the lowest of the three rates being the most relevant benchmark in this case because of the reasons outlined below.

The reasons for identifying the lowest regulated rate as the most appropriate benchmark


27

5.35 The differences between the three glide paths reflected primarily differences in the level of charges at the start of the control and the desire to allow “sufficient time for operators and customers to adjust to new levels and structures of mobile charges and take these changes into account in their business plans and planned capital expenditure”.33

5.36 In addition, we understand that C&W does not need to incur investment in advance of acquiring a customer because it installs the pico cells in the customer’s premises. Given the likelihood that C&W’s costs are lower than the proposed regulated rate, we consider that a charge at the level of the lowest of the three rates does not create a risk of undue disruption.

C&W’s FMC service is yet to commence commercial operations, although it has already made plans and started to acquire customers.

5.37 We considered also whether we ought to adopt the rate of C&W’s roaming partner [ ] as an appropriate benchmark. We noted that these roaming rates can vary and [ ]. The variation in these rates may be an incentive for C&W to negotiate with the cheapest network if there is no “pass-through” (i.e., if we do not directly link C&W’s MTR in relation to the FMC service to the MTR charged by its roaming partner). It is also possible that C&W may choose to switch to a different roaming partner. As a result of these considerations, we do not consider that it is practicable to adopt C&W’s roaming partner’s rate as a benchmark.

5.38 Therefore, we believe that the most relevant benchmark is the lowest of the regulated rates, i.e. the TAC for the 900/1800 MHz operators (applying to Vodafone and O2). We assess whether this rate is in fact an appropriate rate to be applied in this case in the light of Ofcom’s six principles of pricing, under paragraphs 5.45 - 5.129). Given the date when we expect to resolve this dispute, the most relevant TAC is that for 2009/10.

5.39 We have not identified any other national benchmarks which might apply in this case.

International Benchmarks

5.40 We have also considered whether any international benchmarks (i.e. MTRs applied in other jurisdictions) may be of relevance.

5.41 The parties did not submit to us that any international benchmarks were appropriate to the determination of this matter.

5.42 We conducted some research as to whether there are any relevant MNOs in other jurisdictions utilising guard band spectrum in the same manner as C&W. We are aware of a non-profit organisation called “Fixed-Mobile Convergence Alliance” which focuses the interests of companies which either deploy, state an intention to roll out, or encourage what is described as an FMC service.34

5.43 We have maintained awareness of other MTR regimes and MTRs in other jurisdictions through our work on the mobile sector assessment project.

It is not clear, however, that there are any immediate similar services that are a useful comparison to the specific C&W FMC service.

35

33 Op. cit paragraph 13.4b.

There

34 See, for instance, the members of the Fixed-Mobile Convergence Alliance (FMCA) (http://www.thefmca.com/about-us) which has the purpose of encouraging the seamless integration of mobile and fixed-line telephone services. We note that BT is a member of the FMCA. 35 See further at: Mobile citizens, mobile consumers Adapting regulation for a mobile, wireless world, 28 August 2008: http://www.ofcom.org.uk/consult/condocs/msa08/msa.pdf

http://www.thefmca.com/about-us�

http://www.ofcom.org.uk/consult/condocs/msa08/msa.pdf�


28

appears to be no immediate international comparisons which might be appropriate for this dispute. Therefore, we do not consider that there are any relevant international benchmarks.

5.44 In light of the above, we have concluded that the most relevant benchmark in the present case is the regulated TAC for 2009/10 in the CC Determination which applies to Vodafone & O2. We have therefore considered this as one of the options for resolving the dispute.

Application of the six principles of pricing and cost recovery to this dispute

5.45 The following section considers the application of the six principles of pricing and cost recovery to the four options that we have considered for setting C&W’s termination rate in this dispute:

5.46 1. the termination rate proposed by C&W, equal to 6.418ppm;

5.47 2. T-Mobile’s estimate of the cost of termination on a DECT guard band network, equal to 1.2ppm;

5.48 3. our best available estimate of C&W’s termination cost, between 2.61-4.14ppm36

5.49 4. to benchmark against the MTR set in the CC Determination for Vodafone and O2 in 2009/10 of 4.4ppm (2006/07 prices).

; and

5.50 Options 1 and 2 are based on the parties’ arguments which were discussed in section 3. Option 3 is based on our consideration, given the time and evidence available, of C&W’s likely termination costs. Option 4 is based on our view of the most appropriate benchmark for this case. Our assessment of all four options against the 6 principles of pricing and cost recovery is outlined below.

Cost causation

5.51 The cost causation principle states that costs should be recovered from those whose actions cause them to be incurred. Since it is generally efficient for charges to reflect costs, it is usual to give most weight to this principle unless there are good reasons for not doing so in a particular case. Additionally, the TRD Judgment stated that Ofcom should consider whether an analysis, however broad brush, of the relationship of prices to costs is necessary.37

5.52 In Ofcom’s “Determination to resolve a dispute between BT and Telewest about a geographic call termination reciprocity agreement” (June 2006), Ofcom stated:

“In this context, BT, as the originating operator (calling party), is causing the costs of termination on Telewest’s network to be incurred, and thus should be the party responsible for bearing the costs”

5.53 Charges for termination which reflect the efficient level of costs incurred when a call is terminated on C&W’s network would be consistent with this view of cost causation in that T-Mobile will be responsible for the (efficiently incurred) costs caused by it.

36 As set out in paragraph A1.11. 37 See number 2 at paragraph 184.


29

The inclusion of national roaming costs

5.54 For any call terminated outside the coverage of its own network, C&W has entered into a national roaming agreement under which C&W is charged a roaming rate by its roaming provider. Ofcom has considered the arguments of the parties as to whether or not this is a relevant cost to be taken into account in assessing the efficient level of costs for the termination of calls to C&W’s customers using its FMC service.

5.55 In arguing that national roaming costs are irrelevant in determining a termination rate for this dispute, T-Mobile states:

‘..the rationale for termination rates is to ensure that the terminating network covers the costs of termination on its own network – it is not to enable the terminating network to provide termination on other networks, or additional services for its own customers….’

5.56 However, without this roaming facility, C&W customers would only be able to make and receive calls when in the area of C&W’s own network coverage. Therefore, the roaming agreement and its associated termination costs form an unavoidable component of the cost to C&W of providing a national mobile termination service. For this reason, we believe that it is appropriate to include the national roaming costs in our estimation of the incurred cost of termination of calls received by C&W’s customers using its FMC service.38

5.57 C&W argued in their submission that, if Ofcom decides to base ‘FMC termination rates on the cost of the service then it would not be appropriate to exclude roaming charges in the manner that T-Mobile suggest because:

• If the roaming costs are excluded and the rates set as proposed by T-Mobile it would result in the service making a loss and hence being unsustainable. As such the rates would be a barrier to investment and competition;

• It would be inconsistent with Ofcom’s policy to increase innovation and competition in the mobile market for which the auction of GSM Guardband spectrum was a key component; and

• The cost of roaming is an efficiently incurred cost of providing the service. It would be both impractical and more costly for C&W to attempt to provide national coverage using its GSM Guardband licence.’

38 In its submission, T-Mobile also argued that Ofcom’s decision not to take account of national roaming in the setting of H3G’s price control (see A9.30 at http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf) indicated that Ofcom does not consider national roaming to be relevant to the consideration of the cost of call termination generally. However, in setting H3G’s price control, Ofcom chose to exclude roaming costs from the cost model because the volume of traffic that would be terminated whilst roaming would be small (H3G stated in March 2006 that its 3G network provided 88% population coverage) and the inclusion of this cost in the model would have made no difference to the assessment of the average cost of termination on H3G’s mobile service. In contrast, C&W’s expects between [ ] of calls to be terminated whilst roaming, and the inclusion of national roaming costs makes a substantial difference to the assessment of the average cost of termination on C&W’s FMC service. Therefore, the factors which made national roaming irrelevant to the setting of H3G’s price control do not apply to the determination of a termination rate in this case.



30

5.58 We agree that it is desirable for providers to be able to bring a service to market in the most efficient way possible, and it is important that innovation and competition are not stifled by inflexible charging arrangements. We therefore consider that the cost of national roaming is a relevant factor to be included in the cost analysis, to the extent that it is an efficiently incurred cost of providing the service.

5.59 As illustrated by the parties opposing views above, in the case of national roaming, two interpretations of cost causation seem possible. It could be argued that the incremental cost of termination of a call to C&W’s FMC service via roaming is caused by the caller’s decision to make the call, and that they should therefore bear the cost of this decision. Alternatively, it could be argued that the roaming charge is caused by the decision of the C&W customer to roam outside the coverage area, and that they should therefore bear the cost of this decision.

5.60 We believe that callers should bear the cost of their decision to make a call to a mobile phone, as they do to all other UK mobile services, up to the point where the call recipient’s actions cause the call termination cost to rise above what a caller would expect to pay for termination on an efficient national network. In other words, a provider should be free to determine the balance between (i) building their own infrastructure and (ii) paying for the use of existing infrastructure of their roaming partners, and should be able to benefit from any cost savings they are able to make, but should not be able to charge more than the efficient termination rate if this balance results in costs which are higher than an efficient national operator.

5.61 Therefore, as indicated in paragraph 5.56, we believe that the cost of national roaming should be accounted for in C&W’s termination charge to the extent that it does not increase the termination charge above that of an efficient national mobile operator.

5.62 T-Mobile go on to argue in their submission that:

‘...even if it is appropriate to pass on the costs of termination on another network, C&W can only do so in respect of the traffic that is in fact so terminated, and so either (i) not apply the same charge to all traffic [or] (ii) significantly reduce its termination charge to reflect the proportion of traffic terminated on its own network.’

5.63 As regards the analysis of C&W’s costs, we agree that the proportions of traffic terminated on its own network and under the national roaming agreement are relevant. T-Mobile’s suggestion in point (ii) is likely to be more practicable than the suggestion in point (i) and also more consistent with the termination rates of the other MNOs, which do not reflect variations in the costs of terminating individual calls. Accordingly, we have based our cost analysis on the calculation of a range of cost estimates which take into account C&W’s view of the likely proportion of calls to be terminated via its roaming agreement (see table A1 in Annex 1).

The efficient cost of a national operator

5.64 Although C&W’s own network infrastructure alone does not provide national coverage, FMC customers are provided national coverage through C&W’s roaming agreement. As the end users (both callers and call recipients) receive the same benefits regardless of whether national coverage is provided by an operators own network or through national roaming, we believe that the efficient cost of providing national coverage is the relevant basis for analysis in this case.


31

5.65 Therefore, to inform our analysis of the likely efficient costs of termination on a national network, we considered output from Ofcom’s MCT cost model which suggests that the average network unit cost of an efficient national operator is 3.7ppm. 39 This figure refers to network costs only. An additional allowance should also be made for non-network costs in order to be consistent with the MCT charge control. Ofcom used a mark-up for non-network costs of 0.3ppm for 2G/3G operators in the charge control. Therefore, the estimated efficient total termination cost for 2G/3G MNOs is 4.0ppm (in 2006/07 prices).40

5.66 However, we note that a number of factors make the applicability of the MCT cost model to C&W’s operations uncertain:

• C&W’s network rollout will cover only its clients’ premises and coverage on the FMC service will be provided through its national roaming provider’s network in all other areas. Therefore, because C&W’s FMC business model allows it to invest in infrastructure only when it is certain it will be needed, it is likely to have lower fixed costs than existing MNOs.

• However, because C&W pays a charge for national roaming, its variable costs could be higher than those experienced by MNOs (see table A1 in Annex 1). The creation of a national service which relies on national roaming adds uncertainty to C&W’s FMC business proposition because (i) the average cost of service provision is determined by the proportion of calls terminated by roaming, which C&W cannot predict with certainty and (ii) their business proposition relies on C&W’s ability to gain and retain a national roaming agreement with one of the MNOs on suitable terms.

• The efficient costs of termination as estimated by Ofcom’s MCT cost model are based on the qualities of an efficient MNO, and thus assume access to identical technology. Although there are substantial similarities in the services, there are some differences of detail in the technology employed by C&W.

Our best available estimate of C&W’s termination cost

5.67 As set out at paragraph A1.11, a rate based solely on our estimate of costs, most consistent with the cost causation criterion, would lie between 2.61ppm – 4.14ppm (in 2007/2008 prices).

5.68 Our estimate of the cost to C&W of terminating a call on its own network is significantly smaller than the cost to C&W of terminating a call via its roaming agreement. Therefore, as well as being sensitive to C&W’s projections of their future call volumes, 41

39 The MCT cost model uses 2006/07 as its base year, therefore all numbers in this section are expressed in 2006/07 amounts and need to be inflated to 2007/08 numbers for comparison with the results from C&W’s business plan.

our estimate is also sensitive to C&W’s assumption of the balance

40 As explained in the section on benchmarking (see paragraphs 5.30 - 5.31), the CC Determination determined that the TACs in 2010/11 should be 4.0ppm for O2, Orange, T-Mobile and Vodafone. The 4.4ppm determined for Vodafone and O2 in 2009/10 represents a glide path towards this efficient rate. As discussed under the effective competition heading, we believe that it would unfairly disadvantage C&W to be awarded a lower termination charge than that awarded to efficient MNO providers. Therefore, as previously stated, we believe the MTR set in the CC Determination for Vodafone and O2 in 2009/10 of 4.4ppm (2006/07 prices) is an appropriate benchmark to consider for resolving this case. 41 For example, holding the assumed proportion of FMC calls terminated on the C&W network constant at 50%, if the projected number of calls terminating on C&W’s FMC service were halved the


32

between FMC calls terminated on the C&W network and calls terminated on the network of C&W’s roaming partner. The range of our estimates of C&W’s termination costs in relation to the FMC service, 2.61ppm – 4.14ppm, is based on C&W’s projection that the proportion of calls to their FMC service that would be terminated via national roaming would be between [ ] %.

5.69 For completeness, it should be noted that, if we based the cost estimate on the assumption that C&W used in their business plan, that [ ]% of calls to the FMC service would be terminated via roaming, the result would be[ ] ppm (see A1.11). However, given the uncertainty about the amount of traffic terminated via roaming (as acknowledged by C&W), we believe that it would be prudent to consider our range of estimates based on the more flexible assumption of the [ ]% traffic volume terminated via roaming provided by C&W.

5.70 C&W’s proposed termination rate, 6.418ppm, falls significantly above our cost estimate range. Even if 100% of calls to C&W’s FMC service are terminated via roaming, our estimate would still be lower than this rate.

5.71 On the other hand, our estimated range is significantly higher than the cost estimate provided by T-Mobile (1.2ppm). In fact, our sensitivity analysis shows that, even if all calls to C&W’s FMC service were terminated on C&W’s own network, the resulting estimate ([ ]ppm) would be higher than T-Mobile’s proposed rate.

5.72 The estimate of the efficient cost of termination on a national network provided by the MCT cost model, 4.2ppm in 2008/09 prices (or 4.0ppm in 2006/07 prices), 42 is slightly higher than the top of our range of estimates for the likely average cost to C&W of terminating a call on its network (2.61ppm – 4.14ppm). Our estimate of the cost of termination of calls to C&W’s FMC service is equal to 4.2ppm at the point in our estimate range where the proportion of calls terminated via roaming is assumed to be [ ]%.43

5.73 We have also considered relevant benchmarks. A particularly relevant benchmark is provided by the regulated rates for termination on the networks of the MNOs. The lower of these rates, which is the regulated Vodafone and O2 termination rate as determined by the CC, equal to 4.4ppm in 2006/07 prices

We believe that it is unlikely that such a high proportion of calls would be terminated via roaming, given the incentive C&W has to minimise the proportion of calls to its FMC service that are terminated via roaming.

44

resulting constant estimate would be 4.50ppm (assuming a decrease in call volume will not decrease C&W’s own network costs, but will proportionately decrease roaming costs). If the projected number of calls were to be increased by 50% the resulting revised constant estimate would be 3.25 (assuming an increase in call volume will not increase C&W’s network costs, but will proportionately increase roaming costs). 42 The MCT model uses 2006/07 as its base year, therefore all outputs from the MCT model are expressed in 2006/07 prices and need to be inflated to 2007/08 prices for comparison with the results from C&W’s business plan, which uses 2007/08 as its base year. 43 Due to the existence of certain fixed costs in the provision C&W’s own network infrastructure, the average unit cost of termination of calls to C&W’s FMC service is assumed to increase above the roaming termination rate in instances where very high proportions of calls are terminated via roaming.

(or 4.58ppm in 2007/08 prices), will approximate closely to the efficient costs of a national mobile operator (4.0ppm in 2006/07 prices), as it is itself based on the output of the MCT cost model. This lower rate corresponds to our estimate at the point where [ ]% of calls to C&W’s FMC service are terminated via roaming. C&W’s costs can amount to a figure above the regulated rate with the addition of fixed network and other costs.

44 See 16.22 at http://www.catribunal.org.uk/files/CC_Determination_1083_H3G_1085_BT_220109.pdf.



33

5.74 Therefore, while T-Mobile’s estimate (1.2ppm) is likely to underestimate the cost of termination of calls to C&W’s FMC service, both the rate proposed by C&W (6.418ppm) and the MTR set in the CC Determination for Vodafone and O2 in 2009/10 of 4.4ppm in 2006/07 prices (or 4.58ppm in 2007/08 prices) are likely to over estimate the cost of terminating a call to C&W’s FMC service.

5.75 Additionally, our range of cost estimates is sensitive to the cost allocation assumptions we have used as well as to the numbers provided by C&W’s business plan, which are particularly uncertain given that the service is not yet fully established. Therefore, these cost estimates should not be thought of as particularly robust or precise.

5.76 Furthermore, we need to balance cost considerations with other factors when resolving a termination rate dispute consistent with our duties. Therefore, we have considered it prudent not to rely on our cost assessment in isolation, and have considered a number of additional relevant factors, as described below, in establishing this draft determination.

Cost minimisation

5.77 In the context of this dispute, the principle of cost minimisation implies that the termination rate set should facilitate productive efficiency by providing an incentive for costs to be minimised.

5.78 Setting a termination rate higher than efficient costs could risk encouraging (inefficient) entry by operators whose costs are higher than those of (efficient) existing operators; this would reduce static efficiency.45

5.79 In this case, while we are uncertain what the average cost of termination will be for calls received by C&W’s customers using the FMC service specifically, we believe that the MCT cost model gives a robust estimate for the efficient cost of termination for an established MNO offering national coverage.

On the other hand, setting a rate that is lower than efficient costs could risk discouraging entry by operators whose costs are efficient, which would reduce dynamic efficiency. In addition, if there are entry barriers, entry may not occur even if charges are above efficient costs.

5.80 If a new entrant (or established firm) is able to charge the industry regulated rate regardless of the technology and business strategy it chooses to adopt, it will benefit if it is able to lower its costs below those of its competitors. This allows participants in regulated markets to retain an incentive to carry out the investment, innovation and market research, and adopt new technology, that could result in long-term industry cost reductions and product improvements. The importance of retaining these incentives was highlighted in the TRD Judgment:

"It is important therefore not to allow benchmarking against actual or proposed price controls to be used in a way which deprives the undertakings of the benefits of cost reductions and other efficiency savings which such controls were intended to encourage.”46

45 Static efficiency refers to maximising output with a given amount of resources, while dynamic efficiency allows consideration of the potential for increasing resources over the long term through innovation and investment. A competitive market is more likely to encourage investment and innovation, thus there is a relationship between increases in competitive pressure and increases in dynamic efficiency. 46 See number 11 at paragraph 186.


34

5.81 Consequently, it could be argued that setting a new entrant’s (or existing firm’s) charges equal to their costs, even when their costs are lower than that of other operators, might weaken the incentive for long term cost minimisation because lower cost operators would not benefit from the cost savings they make.

5.82 Likewise, if a firm is able to charge the actual cost of termination regardless of its relationship to the industry regulated rate or to the industry’s efficient costs, the incentive to minimise costs in the long term through investment, innovation and creativity in serving consumers could also be jeopardised, as the firm does not benefit from reducing its costs nor suffer detriment from incurring inefficient costs. Such a charging strategy could in turn harm consumers by allowing inefficient costs to be passed on into call prices.

5.83 Therefore, even though we consider it likely that C&W’s termination costs are lower than the MTR set in the CC Determination for Vodafone and O2 in 2009/10 of 4.4ppm (in 2006/07 prices), we believe it is desirable from a cost minimisation perspective to set C&W’s charges equal to this benchmark, rather than to C&W’s actual costs.

5.84 This reasoning is supported by the ERG’s ‘Common position on symmetry of fixed call termination rates and symmetry of mobile call termination rates’, which states that:

‘Unlike a unique efficient termination rate level, asymmetric TR pricing does a priori not favour productive efficiency. In particular, even if it ensures every type of operators (efficient or not) to recover their incurred costs, it imposes a constraint on more efficient operators to subsidize the relative in efficiencies of their competitors. Consequently, incentives to deal with inefficiencies may be reduced and passed on to down stream markets…’ 47

5.85 In paragraphs

5.86 – 5.90 below, we assess the four options that we have identified to resolve the dispute against the principle of cost minimisation.

5.86 Option 1 - T-Mobile’s cost estimate, 1.2ppm, is below the output of the MCT cost model for an estimation of the efficient cost of termination for a national operator (i.e. 4.2ppm in 2007/08 prices), and is also likely to be below the average cost of termination for C&W’s FMC service (2.61ppm – 4.14ppm), even when we completely exclude the impact of national roaming from the cost analysis (see paragraph 5.95). Therefore, the rate proposed by T-Mobile would not provide a sufficient price signal to encourage firms who are able to provide termination services at a cost lower than, or equal to, existing firms to enter the market. Additionally, this rate would not allow firms such as C&W to benefit from any cost reductions it might be able to achieve relative to its competitors due to adopting an innovative technology or business strategy.

5.87 Option 2 - Since the termination rate proposed by C&W, 6.418ppm, is significantly above the MCT cost model’s estimate of the efficient cost of termination for a national operator (4.2ppm in 2007/08 prices) as well as the regulated industry termination rates (between 4.4ppm and 5.5ppm in 2006/07 prices)48

47 See page 5 at

this rate might not provide a sufficient price signal to ensure that new entrants only enter the market if they are

http://www.erg.eu.int/doc/publications/erg_07_83_mtr_ftr_cp_12_03_08.pdf. 48 See 16.22 at http://www.catribunal.org.uk/files/CC_Determination_1083_H3G_1085_BT_220109.pdf

http://www.erg.eu.int/doc/publications/erg_07_83_mtr_ftr_cp_12_03_08.pdf�


35

able to provide termination services at equal or lower cost than existing firms. In other words, it might allow some inefficient entry.

5.88 Option 3 - The wide range between the upper and lower bounds of our estimate of the cost of terminating calls to customers of C&W’s FMC service (2.61ppm – 4.14ppm) arises because of the uncertainty surrounding the proportion of calls to C&W’s FMC customers that will be terminated via national roaming. This would make it difficult to determine an appropriate cost based termination charge for C&W’s FMC service. Additionally, as discussed above, setting a new entrant’s (or existing firm’s) charges equal to their costs, even when their costs are lower than that of other operators, might weaken the incentive for long term cost minimisation because lower cost operators would not benefit from the cost savings they make.

5.89 As the proportion of calls terminated via national roaming has such a significant impact on the average unit cost of providing termination services (see paragraph A1.9), it is reasonable to expect that C&W will seek to minimise the use of national roaming (given any of the termination charge options under consideration). It may therefore structure its retail prices to discourage customers from receiving large numbers of calls while roaming. As a result, it is not unreasonable to expect that C&W will work to keep the percentage of calls terminated via national roaming at the lower end of their projections or below (i.e., equal to, or lower than, 30%).

5.90 Option 4 - According to our cost estimates, the fourth option set out above – to set C&W’s termination charge equal to the CC Determination for Vodafone and O2, 4.4ppm in 2006/07 prices (or 4.58ppm in 2007/08 prices) – would allow C&W to make a profit on termination if it is able to keep the percentage of calls terminated via national roaming below [ ]%. This option would provide C&W with an incentive to minimise termination costs and allow it to retain any benefits resulting from its choice to adopt cheaper technology and to differentiate its service, should it be able to lower its termination costs below those of its competitors. Therefore, this option seems to be the most desirable from a cost minimisation perspective.

Effective competition

5.91 Consistency with this principle requires that the mobile termination rate set does not undermine the pressure for effective competition.

5.92 In the section on cost minimisation above, we discussed how the level of the determined rate will affect the prospects for entry. A rate set below the average cost of termination of calls received by C&W’s customers using the FMC service might jeopardise the ability of C&W to enter the market. However, setting a rate above the industry’s efficient costs would increase the risk of allowing inefficient entry.

5.93 In this section, we consider the possible competitive effects of our four stated options for resolving this dispute, including any competitive distortions that could arise if C&W’s termination rate was set significantly below or above its efficient costs.

New entry

5.94 As described in paragraph 2.6, C&W’s FMC service utilises the DECT guard band spectrum which it acquired in Ofcom’s 2006 auction. Our attitude to competitive entry through the auction of DECT guard band spectrum reflects the principles of Ofcom’s “Awards Programme Approach”49

49 http://www.ofcom.org.uk/radiocomms/spectrumawards/awardsapproach/

. Our general approach is to allow the market to


36

determine how spectrum is used, taking account of the benefits of competition and the development of innovative services.

5.95 In general, we consider that there is scope for benefits to consumers from further competition in mobile services (recognising the extent of competition that already exists). The entry of C&W is one possible example of such further competition.

5.96 In its submission, C&W explains that:

‘The FMC service is a telephony service that combines the benefits of fixed and mobile telephony. The customer can use a single mobile handset that will operate over the fixed network whilst in their office location, but then seamlessly switches to a mobile network when the customer is away from their office location.’

5.97 Through the provision of a service which allows business customers to gain from a single phone the functionality which previously required two phone contracts and two handsets (one fixed and one mobile), the introduction of C&W’s service could increase the convenience of business communication services for its customers and increase choice for consumers more generally. Whilst we make no judgement on the extent to which C&W’s service is beneficial for competition and consumers other than through the addition of a further operator, the approach adopted by us in the resolution of this dispute should not act as a deterrent to entry.

Potential distortion of competition through non- cost reflective and non-reciprocal charges

5.98 T-Mobile argues that the current C&W MTR does not reflect the efficient costs of termination of calls for a DECT guard band network operator and thus would potentially distort competition in the retail market.

5.99 In retail mobile markets, operators typically compete by offering relatively low subscription prices which might be financed by termination rates which are above cost.50

5.100 Likewise, if an operator is forced to charge a termination rate significantly lower than its costs of termination and in doing so makes a loss or earns lower profits on termination than its competitors, it could be undercut by them on prices in the retail market. This could distort competition to the extent that the inability to match retail prices reflects a termination charge below efficient costs, and below those of its rivals who are providing an equivalent service, rather than competition on the merits by incumbent firms.

Therefore, if an operator is able to charge a termination rate significantly higher than its costs of termination and in doing so earns larger profits on termination than its competitors, it could profitably undercut them on prices in the retail market. This could distort competition, to the extent that the ability to offer lower retail prices reflects a termination charge above efficient cost, and above those of its rivals who are providing an equivalent service, rather than competition on the merits (such as more efficient costs for providing a similar service).

50 The interaction between the two sides of the market for mobile calls (callers and call recipients) will generally lead a substantial portion of excess termination profits to be competed away in the retail mobile market through, for example, lowering prices to attract more mobile customers. This phenomenon is referred to as the waterbed effect. The control on mobile termination charges was set so as to prevent termination charges being set at excessive levels in order to fund low retail prices, which would have resulted in an inefficient structure of prices even if no excess profits were made overall.


37

5.101 Therefore we are mindful of the fact that, in general, setting a termination rate as close as possible to the efficient costs of termination, and setting the same rate for operators providing the same service, would minimise the potential for distortion of competition.

5.102 We considered whether setting C&W’s termination rate at a level significantly different from its average costs would give rise to any distortions of competition in the retail mobile market between C&W and other mobile service providers and have reached the following conclusions:

• Setting a termination rate which is below C&W’s average termination costs and below the regulated rate, such as the rate suggested by T-Mobile, 1.2ppm, would distort competition against C&W (see paragraph 5.100);

• Setting a termination rate which is above the regulated rate and above C&W’s average termination costs, as is likely to be the case if the rate was set at 6.418ppm as proposed by C&W, would distort competition in favour of C&W (see paragraph 5.99); and

• According to our cost analysis, C&W’s actual termination cost is likely to be lower than the regulated rate, 4.4ppm in 2006/07 prices (or 4.58ppm in 2007/08 prices), which we are proposing to use as a relevant benchmark. Setting a termination rate which is above C&W’s average termination cost but equal to the regulated rate would advantage C&W but would avoid callers to C&W’s FMC service paying a higher underlying wholesale termination rate than to call other mobile services. This would not distort competition, since there would be competition “on the merits”. Other MNOs that are as efficient as C&W would have the ability to competitively respond, since they receive the regulated rate. In other words, C&W would be advantaged only if its costs of providing a national mobile service are lower than its competitors, not because of any favourable treatment in the regulation of termination charges.

5.103 We have also considered whether there is a risk of “cherry picking” by C&W. In general, cherry picking occurs when new entrant operators choose to compete against incumbents, who are constrained to offer uniform national tariffs, only in (generally urban) areas which can be served at relatively low cost. The cost of C&W’s own mobile infrastructure is relatively low and, in theory, C&W could reduce its total costs by targeting customers who are unlikely to roam very much, and/or by tariffing in a way which discourages roaming. If C&W then received the regulated charge for termination, this could then enable it to undercut the MNOs in retail markets.

5.104 However, the marginal profit which an incumbent MNO would earn from terminating an additional call made within a particular coverage area is equal to the difference between its own regulated termination rate and its own costs of termination in that area. Therefore, even if C&W’s customers are concentrated in low cost areas, should they choose to, the MNOs have the ability to put themselves in a similar position to C&W. Even short of matching C&W’s FMC service, the incumbent MNOs have the ability to respond to competition from C&W, such as through using price discrimination or through targeting their marketing activities, to attract a similar customer base.

5.105 The C&W FMC proposition is currently targeted at business users, and offers both the opportunity to reduce costs by replacing the fixed line infrastructure of the customer and to increase functionality of their telephone systems. By converging the


38

two connections (fixed and mobile), a business user can in effect take their desk phone, with all its functionality, with them wherever they go. However, the MNOs already offer similar benefits via different technological approaches. For example, the MNOs have for many years provided mobile packages designed for corporate enterprises. For example, Vodafone provides Blackberry handsets and the back office systems that facilitate linkages into corporate email and corporate directories, with calling and data packages customised to maximise the value to the enterprise, while minimising costs.

5.106 This reasoning suggests that C&W could be disadvantaged relative to T-Mobile and the other MNOs in serving its targeted customer base, were we to set C&W’s termination rate according to option 3, i.e. our best estimate of C&W’s efficient termination cost, which is below the MTRs of the incumbent MNOs. This is because, if the incumbent MNOs matched or otherwise responded to C&W’s service (as indeed they might already do), they would be receiving a higher termination rate than C&W. Linking C&W’s MTR to our best estimate of C&W’s termination costs might enable the MNOs to profitably offer retail prices that C&W would be unable to match, not because of inferior performance but because of the disparity in termination rates. Overall, we consider that there is a risk of a distortion of competition against C&W under option 3.

5.107 The ability of the MNOs to match or otherwise respond to C&W’s FMC service, if they and C&W receive a similar termination rate, substantially reduces the potential concern about cherry picking. Overall, we consider that the risk of an anti-competitive distortion of competition through cherry picking in such circumstances would be low.

Potential distortion of competition through charges at C&W’s incurred costs

5.108 We also considered whether setting C&W’s termination rate at a level equal to its average costs would give rise to any distortions of competition in the retail mobile market between C&W and other mobile service providers. As mentioned before, the average cost to C&W of terminating a call received by one of its customers using the FMC service is uncertain, since it will depend, among other things, on the actual amount of traffic terminated on C&W’s own network or via roaming.

5.109 The approach in Ofcom’s 2006 “Determination to resolve a dispute between BT and Telewest about geographic call termination reciprocity agreement” supports the determination of equivalent termination rates. In that decision, we said:

“Ofcom considers that…if all call termination charges were based strictly on incurred costs, there would be a distortion of competition. If one CP, through being more efficient, were able to deliver calls more cheaply than another, the CP benefiting from this efficiency and lower cost would not be the more efficient CP which has reduced termination costs, but the less efficient CP since it is buying the cheaper call termination service. The less efficient CP would therefore gain a competitive advantage, in the sense that it would make smaller outpayments to the more efficient CP and would be able to offer its own customers cheaper calls (than if its prices were based only on its own network costs)”. 51

51 See Ofcom’s 2006 determination to resolve a dispute between BT and Telewest about geographic call termination reciprocity agreement.


39

5.110 Setting a termination rate equal to that of C&W’s competitors will not distort competition between C&W and the MNOs, because competition would be “on the merits” between providers of a mobile service with national coverage. In other words, as providers would receive matching conditions, they would all be able to benefit if they were able to reduce cost below that of their competitors.

5.111 Additionally, as discussed at paragraph 5.106 above, if we were to set C&W’s termination rate to reflect our best estimate of C&W’s efficient termination cost, which is below the MTRs of the incumbent MNOs, competition would be distorted against C&W, as they would be receiving a lower charge for an equivalent service and might therefore be unable to match the MNO’s retail prices.

Summary of competitive effects of the four options

5.112 Option 1 - The rate proposed by C&W of 6.418ppm is above the industry’s efficient termination costs (4.2ppm in 2007/08 prices) and the future regulated MTRs of the incumbent MNOs. As such, it would likely allow inefficient (as well as efficient) entry and runs a risk of some distortion of competition in favour of C&W in mobile retail markets.

5.113 Option 2 - We are also concerned about the distortion of competition that could result if we were to set C&W’s termination rate to 1.2ppm, which is likely to be an underestimation of C&W’s average termination cost as well as the efficient cost of termination by a national operator. This outcome could cause detriment to consumers by forcing C&W to charge higher retail prices than it otherwise would have or deter C&W’s entry into the market altogether. Therefore, setting C&W’s termination rate at the rate proposed by T-Mobile would distort competition against C&W and thus be inconsistent with the promotion of effective competition.

5.114 Option 3 – Setting a rate equal to our best estimate of C&W’s average termination costs, between 2.61ppm and 4.14ppm, would mean setting a termination rate for C&W which is lower than the regulated MNO termination rates. Therefore, setting C&W’s termination rate according to this option could give rise to a distortion of competition against C&W and thus be inconsistent with the principle of effective competition (see paragraphs 5.106 and 5.111).

5.115 Option 4 – Setting a rate equal to the benchmark of the CC Determination for 900/1800MHz operators in 2009/10 of 4.58ppm (4.4ppm in 2006/07 prices) could favour C&W, as this rate is likely to be higher than the cost of termination of calls to C&W’s FMC service. However, we consider that this option allows competition on the merits, and retains incentives for cost minimisation and efficient entry by allowing C&W to benefit if it is able to reduce its termination costs below that of its competitors. Therefore, on balance, we consider that option 4 is the most consistent with the principle of effective competition.

Reciprocity

5.116 The principle of reciprocity requires that, where services are provided reciprocally, charges should also be reciprocal. As both parties to this dispute supply a national mobile service, and therefore offer a termination service which appears identical to end users, there is a strong argument that their termination charges should be equal.

5.117 For the reasons set out by Ofcom in its determination to resolve a dispute between BT and Telewest about geographic call termination reciprocity agreement (2006), when charges are based on each operator’s own costs rather than on reciprocity,


40

higher cost providers are effectively subsidised by lower cost providers. This is because the costs of the higher termination cost network would be passed to the subscribers of the lower cost network when they call subscribers of the former.

5.118 The range of estimates for the average costs of termination on calls received by a C&W’s customer using the FMC service are related to certain components of its current business strategy and its business plan projections, including its current roaming arrangement and the projected proportion of calls terminated via this agreement.

5.119 In general, it seems reasonable for the termination rate to be independent of the operator’s business strategy. For example, it would seem unreasonable to reward greater reliance on roaming or a higher cost roaming agreement with a higher termination rate or penalise it through a lower rate. Incentives to adopt an efficient mix between own network expansion and the purchase of roaming services are promoted by setting charges on the basis of an exogenous benchmark.

5.120 We are aware that, historically, there have not always been reciprocal termination charges for mobile termination. However, we stated in the Mobile Call Termination Statement that, without fettering our discretion, in the event of imposing price controls for new entrants, we are of the view that it is desirable for new entrants’ MCT charges to be aligned with those of incumbent suppliers.52

5.121 We note that the ERG’s ‘Common position on symmetry of fixed call termination rates and symmetry of mobile call termination rates’ states that asymmetric termination rates may be justified:

We also noted in the Mobile Call Termination Statement that we would anticipate further convergence in MNOs’ mobile termination rates.

• ‘to take into account differentiated conditions of spectrum allocation’;

• to encourage the development of a new entrant on the market, which suffers from a lack of scale due to late market entry. Indeed, this allows higher expected profits in the short term and induces a more intense competition in the long term to the benefit of end users. In other words, a regulator may allow asymmetric rates for a limited time period – thus trading off short-term inefficiency for long-term objectives (i.e. dynamic efficiency).’

5.122 However, we do not believe that either of these conditions apply to this case because of the following reasons:

1) the DECT guard band spectrum used by C&W was bought at a lower cost than the spectrum used by their competitors;

2) the UK mobile market is sufficiently competitive so as not to require entry assistance in order to promote dynamic efficiency;

3) we believe that C&W’s termination costs are likely to be lower than the lowest regulated rate as determined by the CC, 4.4ppm (in 2006/07 prices) and,

4) the unique nature of C&W’s service is such that its network costs do not suffer from its lack of scale – as it provides network infrastructure only for customers it has

52 See number 6, at page 158.


41

already acquired and the unit cost of its coverage outside its own network does not vary with call volumes.

5.123 Option 4 – benchmarking against the MTR set in the CC Determination for Vodafone and O2 in 2009/10 of 4.4ppm (2006/07 prices) is the option most closely aligned with the principle of reciprocity because T-Mobile is also subject to a regulated rate which is itself very close to this option.53

Distribution of benefits

5.124 If, as asserted by T-Mobile and as indicated by our cost analysis, C&W’s termination costs are lower than those of T-Mobile and the other MNOs, there is a question of how the benefits from these lower costs should be distributed.

5.125 In arriving at our proposed determination, we considered the possible impact – on the distribution of benefits between T-Mobile, C&W and their respective customers - of setting C&W’s MTR under each of the four options identified:

Option 1

• Option 1, which would provide the highest termination rate for C&W of the four options (6.4ppm), will distribute the benefits of C&W’s lower termination cost to C&W. As this rate is above the MNO’s termination rates, it could also lead to some redistribution of profits to C&W from the operators purchasing termination from C&W, which includes its competitors. Customers of C&W’s FMC would benefit, since we would expect a significant proportion of C&W’s termination profit to be passed through to them. However, callers to C&W might face high prices caused by the termination charge (which would be high relative to the termination charges of the incumbent MNOs after implementation of the CC’s determination).

Option 2

• Option 2, which would provide C&W with the lowest termination rate of the four options (1.2ppm), would distribute the benefits to T-Mobile and the other operators purchasing termination from C&W. As this rate is likely to be below C&W’s average termination cost, it could lead to some redistribution of profits from C&W to the operators purchasing termination from C&W, which includes its competitors. Customers of C&W’s FMC would suffer relative to other options. But callers to C&W might benefit from lower prices (to the extent that retail prices charged by originating operators reflected differences in termination rates).

Option 3

• The adoption of option 3 would involve setting C&W’s termination charge to exactly reflect the cost of terminating a call received by C&W’s customers using the FMC service. Assuming we had adequate information to accurately set such a rate, this option would likely set a termination rate for C&W that is below the termination rates of its competitors. Callers to C&W might benefit from lower prices (to the extent that retail prices charged by originating operators reflected

53 The CC’s Determination allowed T-Mobile and Orange of 4.5ppm (in 2006/07 prices) for 2009/10 and a rate of 4.0ppm (in 2006/07 prices) for 2010/11, which will be equal to that of Vodafone and O2 in that year.


42

differences in termination rates). But this option would distribute all the benefits of any differential between the termination costs of C&W and its competitors to C&W’s competitors and their customers. In other words, C&W and its customers would not receive any of the benefits, should it be able to reduce its costs below those of its competitors.

Option 4

• If the average cost to C&W of call termination in relation to its FMC service is lower than the regulated rate (which according to our cost analysis would be the case as long as less than [ ]% of calls to the FMC service are terminated via roaming), option 4 would distribute all the benefits to C&W and its customers of any cost differential between its termination cost and the termination cost of its competitors. Callers to C&W might pay higher prices than under Options 2 and 3 (to the extent that retail prices charged by originating operators reflected differences in termination rates). However, under this option, callers to C&W would pay no more for the underlying wholesale termination rate than for calls to other mobile networks.

5.126 C&W, as a provider of a new service in an effectively competitive market, is likely to a significant extent to use higher termination profits, such as those it might receive under options 1 and 4, to offer lower retail prices to its customers. Therefore, any termination rate higher than C&W’s costs are likely to ultimately distribute the benefits to C&W’s customers. Likewise, any options suggesting a termination rate lower than C&W’s costs are likely to ultimately impose costs on C&W’s customers.

5.127 To the extent that T-Mobile and the other operators will pass the lower termination rates they pay to C&W under options 2 and 3 onto their customers, in the form of lower retail prices for calls to C&W, these options will distribute the benefits from C&W’s lower cost structure to callers to C&W. It is also possible that T-Mobile and other operators would pass these benefits on to their customers in ways other than reducing the cost of calls to C&W customers. To the extent that T-Mobile and the other operators will not pass on lower termination rates into lower prices, they will benefit rather than consumers.

5.128 In summary, a termination rate for C&W which is higher than the industry’s efficient costs is likely to benefit C&W’s customers, but may result in higher prices for callers to C&W. Using the benchmark (i.e. option 4) seems desirable from a distribution of benefits perspective because it would allow C&W’s customers to benefit from any lower costs but would avoid callers to C&W paying more than to call other mobile networks (to the extent that originating operators reflect relative termination rates in their retail prices for calls). Lower termination rates could alter this distribution of benefits, increasing benefits to callers to C&W (to the extent that originating operators reflect lower C&W termination rates in lower retail call prices) and reducing the benefits to C&W’s own customers.

Practicability

5.129 Consistency with the final pricing principle requires that the termination rate determined is practicable and relatively easy to implement. In arriving at our proposed determination, we considered the practicality of implementing the four options for C&W’s termination rate:

• Adoption of the termination rate proposed by C&W, 6.418ppm, would be practicable to determine and easy to implement as it would set the termination


43

rate payable by T-Mobile to C&W equal to that rate currently agreed between C&W and the majority of the other operators.

• Setting C&W’s termination rate to the cost of termination on a DECT guard band network as estimated by T-Mobile, 1.2ppm, would be practicable to determine and the easiest of all the options to implement, as it would simply involve retaining the status quo.

• Setting C&W’s termination rate to the benchmark of the CC Determination for 900/1800MHz operators in 2009/10 of 4.4ppm (2006/07 prices) would be practicable to determine and reasonably simple to implement.

• Due, in part, to the practical difficulties in deriving a specific robust estimate of C&W’s costs, setting C&W’s termination rate equal to its costs is the least practicable to determine and most difficult option to implement. Additionally, as explained above, this option is not robust to changes in C&W’s strategy, as the basis for the cost calculation would fall away should C&W alter its roaming agreement and/or expand its own network. We consider a solution that necessitates re-determining and/or re-negotiating a termination rate every time an operator changes it business strategy to have significant practical disadvantages.

Summary of the application of Ofcom’s principles of pricing and cost recovery

5.130 The outcome of our analysis, which applied the six principles of pricing and cost recovery to the four options for setting C&W’s termination rate in this dispute, is summarised below.

Option 1: The termination rate as proposed by C&W, 6.418ppm

5.131 The charge proposed by C&W is likely to exceed its average termination costs (and so is not supported by the principle of cost causation). As this rate is higher than those of C&W’s competitors, and the efficient termination cost of a national operator as estimated by the MCT cost model, it entails the highest risk of inefficient entry and distortion of competition in favour of C&W (and is therefore contrary to the principles of cost minimisation, effective competition and reciprocity).This rate would result in distribution of all the benefits of any lower termination costs to C&W and its customers.

Option 2: T-Mobile’s estimation of the cost of termination on a DECT guard band network, 1.2ppm

5.132 It is likely that T-Mobile’s suggested rate underestimates C&W’s efficient termination costs (and so is not supported by the principle of cost causation). This rate would also be lower than the MTRs of all C&W’s competitors (contrary to the principle of reciprocity). The implementation of a rate below the industry’s efficient termination cost would not provide a sufficient price signal to encourage firms who are able to provide termination services at lower cost than existing firms to enter the market (contrary to the principles of cost minimisation and effective competition). Setting C&W’s rate at such a low level would distribute the benefits of any lower costs to T-Mobile and the other operators purchasing termination from C&W and could cause detriment to consumers by forcing C&W to charge higher retail prices or deterring C&W’s entry into the market altogether (distribution of benefits, effective competition).


44

Option 3: Our best available estimate of C&W’s termination cost, between 2.61ppm – 4.14ppm

5.133 In theory, this option would send the most cost reflective price signals and is the most consistent with Ofcom’s cost causation principle.

5.134 However, the cost evidence considered in our analysis has significant limitations in providing a robust cost estimate. Therefore, the theoretical support for this option from the principle of cost causation is less strong in practice and faces significant concerns about practicability.

5.135 Furthermore, setting a new entrant’s (or existing firm’s) charges equal to their costs, even when their costs are lower than that of other operators, might deter investment, innovation and creativity in serving consumers (which would be counter to cost minimisation).

5.136 Setting C&W’s termination rate equal to cost would (assuming we could arrive at an accurate cost estimate) distribute all of the benefits of C&W’s likely lower cost structure to T-Mobile and other operators, and callers to C&W to the extent that such operators passed on lower termination rates into lower retail prices (distribution of benefits).

5.137 As there is significant potential for error in our cost analysis, the consequences for competition of setting C&W’s termination rate equal to our estimate of costs depend on the rate selected. However, even if we could be sure that our estimate of C&W’s termination costs is accurate, setting C&W’s MTR lower than those of all the incumbent MNOs would risk a distortion of competition against C&W and allow C&W’s competitors an advantage because of their higher termination rates (effective competition). Additionally, this option is not robust to changes in C&W’s business strategy, particularly the terms of its roaming agreement (and impacts on the following principles of pricing: practicability, reciprocity, effective competition).

Option 4: the benchmark of the CC Determination for 900/1800MHz operators in 2009/10 of 4.4ppm (2006/07 prices)

5.138 This benchmark appears to be higher than the likely cost of termination for C&W’s FMC service. This option would distribute all the benefits to C&W and its customers if C&W is able to reduce its termination cost below that of its competitors (distribution of benefits, cost minimisation).

5.139 A lower termination rate than under this option might allow callers to C&W to benefit, to the extent that originating operators reflected a lower termination rate in a lower retail price for calls to C&W. Nevertheless, under this option, the termination rate would not cause callers to C&W to pay more than for calls to other mobile networks, to the extent that originating operators reflect relative termination rates in their retail prices for calls (distribution of benefits).

5.140 We have considered whether a termination rate above efficient costs could create a potential risk of distorting competition in favour of C&W. However, incumbent MNOs have the ability to match or otherwise respond to C&W’s service. Additionally, as C&W provide a mobile service with national coverage, this represents competition on the merits because it provides a comparable service to the incumbent MNOs. Therefore, we consider the risk of detrimental distortion to competition to be low under this option. However, a cost based termination rate that is below the rate


45

received by operators providing an equivalent service would create a risk of distorting competition against C&W.

5.141 This option reduces the risk of distortion of competition against an efficient provider and avoids distorting incentives to C&W to alter its business strategy. Under this option, like charges for like services are applied and therefore the principle of reciprocity is met.

The preferred option

5.142 Our preferred option is option 4, that is to determine that the MTR charged by C&W for the termination of calls originated on T-Mobile’s network and terminated on the C&W’s network in respect of C&W’s customers using the FMC service may not exceed our current best estimate of the target average charge (TAC) for Vodafone and O2 (this being the lowest regulated MTR).

5.143 We consider that this option would be reasonable and fair between the parties since it allows efficient costs to be borne by those responsible for incurring them, whilst also providing incentives for those costs to be minimised. Cost minimisation relates to the signals for efficient entry and the ability of more efficient providers to benefit from their greater efficiency.

5.144 We regard this option as being likely to enhance effective competition as it relates to competition on the merits; because operators receive the same (or similar) termination rates, there is an advantage to a lower cost operator and a disadvantage to a higher cost operator. Therefore, we consider that this option reduces the risk of distortion of competition against an efficient provider and avoids distorting incentives to C&W to alter its business strategy.

5.145 Our preferred option is consistent with the principle of reciprocity, which is relevant to this case as the C&W FMC service, like the incumbent MNOs, provides a termination service on a national basis. From the perspective of the end-user of termination (i.e. callers to the C&W FMC service and incumbent MNOs), these services are reciprocal.

5.146 Additionally, our range of cost estimates is sensitive to the cost allocation assumptions we have used as well as to the numbers provided by C&W’s business plan, which are particularly uncertain given that the service is not yet fully established. Therefore, as these costs should not be thought of as particularly robust or precise, we have not sought to rely on our estimate of cost in determining this matter.

5.147 Furthermore, our preferred option seems desirable from a distribution of benefits perspective because it would allow C&W’s customers to benefit from any lower costs but would avoid callers to C&W paying more than to call other mobile networks (to the extent that originating operators reflect relative termination rates in their retail prices). We also believe that this option is practical to implement as between the parties.

5.148 We believe that the interests of consumers are better served overall by this option compared to the others for the following reasons:

• It furthers the interests of consumers through the promotion of competition and the availability of a wide range of electronic communications services by


46

facilitating efficient entry into the mobile market, without encouraging inefficient entry;

• It allows C&W’s customers to benefit if C&W is able to lower its average termination cost below the regulated rate, without causing callers to pay more than to call other mobile networks; and

• It reduces the risk that any party with efficient costs is placed at a competitive disadvantage because of a difference in termination rates.

5.149 Ofcom therefore proposes to resolve the dispute by determining that the MTR charged by C&W for the termination of calls originated on the T-Mobile network and terminated on C&W’s network in respect of C&W’s customers of its FMC service may not exceed the TAC for Vodafone and O2, this being the lowest regulated MTR.

5.150 As explained in paragraph 5.32 above, we recognise that the TAC for Vodafone and O2, as set out in the CC Determination, may be the subject of review by the CAT on judicial review grounds and will not be finalised until the end of the litigation process. However, since we are required by statute to resolve this dispute within four months, we have referred to the MTRs set out in the CC Determination as our current best estimate of the regulated MTRs.

5.151 In particular, we noted that the TAC for Vodafone and O2 in 2009/10, as specified in the CC Determination, is 4.4ppm (in 2006/07 prices).

5.152 An adjustment was required to convert this option into nominal terms to account for three years of relevant inflation. We consider that the methodology for converting the regulated MTRs into nominal terms should follow that used in implementing the CC Determination. Our indicative calculations are attached at Annex 3.54

5.153 Once the CC Determination has been implemented following the ongoing litigation, we would expect the parties to apply the final TAC for Vodafone and O2 and make any necessary adjustment. However, we also recognise that, if the TAC for Vodafone and O2 resulting from the implementation of the CC Determination is different from our current best estimate of it (4.71ppm for 2009/2010) and the parties are unable to reach an agreement, they may refer a further dispute to us.

This shows that 4.4ppm (in 2006/07 prices) converts into 4.71ppm in nominal terms. Therefore, we propose to resolve the dispute by determining the MTR applicable between C&W and T-Mobile in respect of C&W’s customers of its FMC service to be our current best estimate of the lowest regulated rate, which is 4.71ppm.

Assessment of the preferred option against Ofcom’s statutory duties and Community requirements

5.154 We have carefully considered the powers, obligations and duties detailed in section 4 in deciding on the appropriate means of resolving this dispute. In particular, we have considered the relevance of our primary duties and of the Community requirements to this dispute.

5.155 We consider that the following duties are of relevance to this dispute:

(i) the duty to further the interests of citizens (i.e., all members of the public in the United Kingdom) in relation to communication matters (section 3(1)(a));

54 See Annex 3 for “Indicative calculations requested by the Tribunal at CMC on 2 February 2009”.


47

(ii) the duty to further the interests of consumers in the relevant markets, where appropriate by promoting competition (section 3(1)(b));

(iii) the duty to secure the availability throughout the United Kingdom of a wide range of electronic communications services (section 3(2)(b));

(iv) the duty to have regard to the desirability of promoting competition in relevant markets (section 3(4)(b));

(v) the duty to have regard to the different interests of persons in the different parts of the United Kingdom, of the different ethnic communities within the United Kingdom and of persons living in rural and in urban areas (section 3(4)(l));

(vi) the duty to have regard, in particular, to the interests of consumers in respect of choice, price, quality of service and value for money (section 3(5));

(vii) the duty to promote competition (section 4(3), setting out the first Community requirement).

(viii) the duty to secure that Ofcom’s activities contribute to the development of the European internal market (section 4(4));

(ix) the duty to promote the interests of all persons who are citizens of the European Union (section 4(5));

(x) the duty to ensure technology neutrality (section 4(6)).

5.156 We consider that the duties set out at (ii), (iii), (iv), (vi), (vii) and (x) are of particular relevance for resolving this dispute since both the parties have recognised that the resolution of this dispute would have an impact on competition and, therefore, on the offer of electronic communications services to consumers in terms of choice, price, quality of service and value for money. The preferred option gives rise to the lowest risk of a detrimental distortion of competition. Additionally, we consider it highly likely that, as a new entrant in an effectively competitive market, C&W will pass on a significant proportion of any cost savings it is able to achieve to its customers in the form of lower prices. Therefore, in this case, we consider that the potential for customer benefit from this option outweighs the risk of any detriment to competition. Callers to mobile would pay no more for the underlying wholesale termination rate than for calls to other mobile networks.

5.157 We recognise that through the provision of a service which allows business customers to gain the functionality – which previously required two phone lines and two handsets (one fixed and one mobile) – from one phone line and handset, the introduction of C&W’s service has the potential to increase the convenience of business communication services for its customers and increase choice for consumers more generally.

5.158 In developing our approach, we have also retained an approach which is competitively neutral to ensure a level playing field for the provision of services whilst ensuring that our approach promotes competition through the development of new and innovative services. Part of the C&W FMC proposition uses DECT guard band spectrum which was awarded to C&W, alongside 11 others, following Ofcom’s auction in May 2006. We have an interest in ensuring that the determination in this matter does not deter the use of the recently auctioned guard band spectrum through efficient and innovative entry into the market.


48

5.159 In general, we consider that there is scope for benefits to consumers from further competition in mobile services (recognising the extent of competition that already exists). The entry of C&W is one possible example of such further competition.

5.160 We have also sought to adopt a technology neutral approach which does not favour the use of any particular technology (whether that used in T-Mobile’s network or C&W’s network). In doing so, we have ensured, to the greatest extent possible, that our approach does not favour the technology of either C&W or T-Mobile. We have approached the resolution of this dispute from the perspective of consumers with respect to the services being offered rather than the underlying technologies.

5.161 We consider the duties set out at (i), (v) and (ix) are of relevance to this dispute given that the introduction of the C&W FMC service would be consistent with promoting the interests of citizens by encouraging a variety of services which may lead to cost benefits to consumers in the long run. It furthers the interests of consumers through the promotion of competition and the availability of a wide range of electronic communications services by facilitating efficient entry into the mobile market, without encouraging inefficient entry.

5.162 We further consider that the duty set out at (viii) above will also be fulfilled in this case since our approach may encourage new entry into markets in the United Kingdom which potentially could allow operators from other Member States to enter the UK market.

5.163 In addition, we consider that the duties set out in section 4(7) and (8)of the 2003 Act is also relevant, namely the duty to encourage the provision of network access and service interoperability for the purposes of securing efficiency and sustainable competition in communications markets and the maximum benefit for the customers of communications network and services providers. We consider this duty to be of relevance for resolving this dispute since this dispute concerns the service of call termination, which is essential for encouraging interoperability between different networks, so that the customers of one network can call, and receive calls from, the customers of other networks. Further, given that the service of call termination facilitates the development of communications between customers of different networks, we consider it relevant also for the purpose of development of the European internal market.

5.164 We also consider that our duties set out in sections 3(2)(a), 3(4)(d), 3(4)(f) and 4(6)of the 2003 Act may be relevant to the resolution of this dispute, namely those to:

(i) secure the optimal use for wireless telegraphy of the electromagnetic spectrum (section 3(2)(a));

(ii) have regard to the desirability of encouraging investment and innovation in relevant markets (section 3(4)(d)); and

(iii) have regard to the different needs and interest, so far as the use of the electromagnetic spectrum for wireless telegraphy is concerned, of all persons who may wish to make use of it (section 3(4)(f)).

5.165 We consider these duties to be of relevance for resolving this dispute since C&W is in the process of investing in the development of a network to offer mobile services by using the DECT guard band spectrum licence granted by Ofcom in May 2006. Our approach ensures that operators holding a DECT guard band spectrum licence will not suffer from disincentives to enter the market by rolling out networks, thus


49

ensuring that their interests are supported. As a result, those holding licences will be encouraged to make use of those licences, ensuring optimal use of the spectrum. This in turn will encourage investment and innovation as operators roll out network in competition with existing networks.

5.166 Finally, we consider our duties set out in section 3(3) of the 2003 Act to be relevant, namely to have regard to the principles under which regulatory activities should be transparent, accountable, proportionate, consistent and targeted only at cases in which action is needed, as well as any other principles appearing to us to represent the best regulatory practice. In developing our approach, we have considered all relevant previous decisions made by Ofcom, the CAT and other relevant bodies in order to ensure that the proposed approach is consistent with previous regulatory practice.

5.167 Ofcom considers that this document clearly sets out the parties’ arguments and Ofcom’s reasoning that leads to this proposed conclusion, and notes that the parties will have an opportunity to comment on Ofcom’s proposals, and that this supports Ofcom’s duty to ensure that its regulatory activities are transparent, accountable, proportionate, consistent and targeted. Our resolution is targeted in that it seeks to resolve the dispute as between the parties only (indeed Ofcom’s powers in dispute resolution may not go beyond this).

5.168 We do not consider that the duties set out in the following sections are of relevance to the resolution of this dispute since they relate to matters which are not covered by this dispute:

(i) sections 3(2)(c) to (f);

(ii) sections 3(4)(a), (c), (e), (g) to (k) and (m); and

(iii) section 4(9).

How to implement the proposed outcome

5.169 Ofcom’s powers in order to resolve disputes are set out under section 190 of the 2003 Act and include a power to make a declaration setting out the rights and obligations of the parties to the dispute and to direct the parties to enter into a transaction between themselves on such terms and conditions as Ofcom may determine.

5.170 Ofcom’s powers to resolve disputes are limited to imposing obligations on the parties to the disputes only. Ofcom therefore propose to use Ofcom’s powers under Section 190(2)(c) of the 2003 Act to give a direction that, until the implementation of the CC Determination, C&W is not entitled to charge for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers of the FMC service an amount in excess of 4.71ppm, which is our current best estimate of the TAC for Vodafone and O2 for 2009/2010, as specified in the CC Determination, converted into nominal terms.



50

5.172 In determining the dispute, we will also need to decide the date from which the final determination should apply. Given that the parties have an agreement to an interim rate subject to retrospective adjustment and the likely small levels of traffic on this service to date, we consider that the determination should apply from the date of the final determination in this matter.

Proposed resolution

5.173 Based on the analysis set out at in this section, our proposed resolution is:

i) As from 20 May 2009 and until the implementation of the CC Determination, C&W is not entitled to charge for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in relation to C&W customers using its FMC service an amount in excess of 4.71ppm, which is our current best estimate of the TAC for Vodafone and O2 for 2009/2010, as specified in the CC Determination, converted into nominal terms; and

ii) That this resolution shall apply from the date of the final determination.


5.175 Ofcom proposed to adopt the draft determination that accompanied the Consultation by the statutory deadline for resolving this dispute which is 15 April 2009.


51

Section 6

6 Responses to the Consultation 6.1 In this section, we address each stakeholder’s comments to the Consultation and

Further Consultation in turn and set out our responses. Then, in the light of that discussion, we set out our conclusions.

C&W’s responses to the Consultation

6.2 C&W argued that it should receive a higher MTR than we proposed in the Consultation. It advanced a number of detailed arguments to support its case which we address in turn below, under the following headings:

• Assessment of C&W’s incurred termination costs

• Estimate of efficient termination costs for a national operator

• The relevance of different charge control and glidepath allowances for H3G

• C&W revised proposed rates

• Timing of the decision

• Variation of charges in line with those of Vodafone and O2

• Variation of charges with time of day and compliance

Assessment of C&W’s incurred termination costs

6.3 C&W argued that we should revise our estimate of C&W’s incurred termination costs as follows:

C&W’s arguments

i) Termination costs when roaming

ii)

: In the draft determination, only the direct cost of roaming was allocated to calls terminated via roaming. However, C&W explained that calls terminated via roaming also cause costs in C&W’s own network. C&W proposed that we allow for recovery of the costs incurred in its fixed network and of its non-network costs on calls that are terminated via roaming;

Inclusion of ‘customer site installations’ in capital costs: C&W argued that our estimate of C&W’s capital costs (of which 25% would be attributed to our estimate of C&W’s termination

iii)

costs) should include the costs of the installation of pico cells and the IP-VPN, which C&W considers as equivalent of mobile masts and backhaul transmission’;

Inclusion of IT and product development in the capex figure: C&W note that IT and product development expenditure have not been included in our estimate;


52

iv) Asset lifespan

6.4 C&W also argued that, even if we were to make the suggested changes, the resulting estimate would still be based on a very simple cost model, with very uncertain assumptions. Therefore, C&W suggest that we should make allowances for variations in assumptions and rely on an outcome towards the upper end of the likely range. C&W also suggested that the range of estimates should consider changes to traffic volume assumptions and variations in the amount of roaming traffic. They suggested that a reasonable range of estimates for its termination costs is between 3ppm and 7ppm.

: C&W submitted that we should use a 5 year average asset lifespan, rather than 10 years as we had assumed, as 10 years is too long for the FMC specific investments.

6.5 On this basis, C&W argue that there is a strong likelihood that the rate proposed by Ofcom will fail to allow it to recover its efficiently incurred costs.

6.6 Ofcom has considered the points raised by C&W in the following paragraphs. To inform our analysis of a reasonably plausible range for C&W’s incurred termination costs. However, we have not reached a conclusion about whether these costs have been efficiently incurred, that is, whether they are the minimum given C&W’s business model and technology. We consider that this approach is appropriate given that we remain of the view that we should not rely on our estimate of C&W’s incurred costs in setting C&W’s MTR.

Ofcom response

6.7 Termination costs when roaming

6.8

: We have revised our unit estimates of C&W’s termination cost to include an allowance for the relevant costs of its fixed network and for non-network costs on minutes that are terminated via roaming.

Inclusion of ‘customer site installations’ in capital costs

6.9

: We have included ‘customer site installations’ in the capital costs used to calculate our revised estimates of C&W’s termination cost (25% of our estimate of the capital costs of the FMC service has been allocated to our estimate of termination costs). However, as C&W anticipate that some of this cost will be recovered through other charges to the customer such as installation charges and rental fees, the inclusion of 100% of this expense in our calculations of the capital costs of C&W’s FMC service has resulted in a likely overestimation of C&W’s termination costs.

Inclusion of IT and product development in non-network costs:

6.10

We agree that some aspects of IT expenses should be attributed to the non-network component of termination costs. However, it is likely that ‘product development’ expenses would be associated with outgoing calls and customer acquisition, not termination costs (as such expenses are likely to be related to C&W’s own subscribers or to attracting new subscribers not to callers to C&W which have subscriptions with other providers). In any case, we used a proxy for the total non-network unit cost in our estimate (see Annex 1). Therefore, all non-network costs associated with termination are already provided for.

Asset lifespan

6.11 As a result of the adjustments discussed above, our revised estimate of the likely range of C&W’s incurred cost unit of termination has increased from 2.6ppm –

: Given the additional information that we have received in relation to the assets employed in C&W’s FMC business, we considered it appropriate to adjust our estimates to reflect an average 5 year asset lifespan.


53

4.1ppm, to 3.2ppm – 5.0ppm (in 2007/08 prices). The revised estimates are explained in greater detail in Annex 2.

6.12 We agree with C&W that these results are highly sensitive to call volume and roaming proportion assumptions. However, given the proposed method of setting C&W’s MTR, C&W will retain an incentive to discourage the termination of calls via roaming. We therefore believe that it is likely that this proportion would be closer to the bottom of the range (i.e. we consider it likely that the proportion of calls to C&W’s FMC service terminated via roaming would be close to, or lower than [ ]%).55

6.13 We believe that only efficient costs should be reflected in the charges set for termination. This is consistent with the principles of pricing which Ofcom believes are appropriate to resolve this dispute, for the reasons set out in section 5 above. Therefore, it is important to draw a distinction between C&W’s incurred costs and the efficient costs of providing their service. The MCT model provides an estimate of the efficient cost of termination for a provider of a mobile service with national coverage (see paragraph

In any case, we do not propose to base our determination on our estimate of C&W’s incurred cost.

5.65). The revised estimate of C&W’s costs included in Annex 2 is an estimate of C&W’s incurred costs, based on their business plan, including their call volume projections. As noted in paragraph 6.6, we do not take a view on whether these are themselves efficiently incurred given C&W’s business model and technology.

6.14 Our estimate of C&W’s incurred costs remains lower than the MTR proposed as long as less than [ ]% of calls to C&W’s FMC service are terminated via roaming. We believe it unlikely that more than [ ]% of calls to C&W’s FMC service would be terminated via roaming given C&W’s incentive and ability to influence the level of roaming by its customers. Additionally, it is unlikely that if more than [ ]% of calls to the FMC service are terminated via roaming, the associated cost to C&W should be considered to be efficiently incurred because these costs would be more than our benchmark for the efficient termination costs of a national mobile operator.

Estimate of efficient termination costs for a national operator

6.15 C&W submitted that they support Ofcom’s use of benchmarking combined with an estimate of cost in order to judge what a reasonable termination rate might be. However, C&W argue that the estimate of the cost of termination for an efficient national operator, resulting from the use of the MCT model, fails to take adequate account of the specifics of the service that C&W offer. In particular, C&W argued that their efficient termination costs might be higher than the MCT cost model estimates for the following reasons:

C&W’s arguments

(i) the market share

55 For example, even if the total call volume projections were reduced by 50%, we believe a plausible estimate of C&W’s incurred termination cost would be closer to 4.9ppm than the 7ppm suggested by C&W. 4.9ppm represents our estimate of C&W’s termination cost under the assumptions that 1) C&W’s call volumes are 50% of the projected call volumes they have included in their business plan and 2) that [ ]% of calls to C&W’s FMC service are terminated via roaming.

of the C&W FMC service will be materially different to those of the established 2G/3G mobile operators; and


54

(ii) C&W’s time of day usage is concentrated during normal working hours, while the MNOs’ usage is spread more evenly over all times of the day.

6.16 To the extent that C&W’s costs are subject to economies of scale, then its average incurred costs will tend to be higher the smaller is its market share. However, a lower market share does not necessarily mean that C&W’s average incurred costs will be higher than those of the other MNOs.

Ofcom response Market Share

6.17 C&W does not need to incur investment in advance of customer demand as sites are rolled out in the premises of clients that it has already acquired. In other words, C&W’s network capacity growth is linked directly with traffic volumes. Therefore, C&W is significantly less disadvantaged by its smaller market share than an operator with a full national network which has to build capacity in advance of demand.

Time of day

6.18 A mobile network must be sized to cope with the busiest hour of its operations. Therefore, we agree that if C&W’s call volumes are more concentrated at particular times of the day this could drive up its own network costs relative to other providers.

6.19 [ ]. The direction of the overall effect on C&W’s average costs of an increase in the number of calls terminated in the ‘busiest hour’ is unclear therefore. We also note that, as the proportion of calls which are terminated on C&W’s own network increases, C&W’s average unit termination cost decreases (see Annex 1).

6.20 In addition, our cost estimate is based on C&W’s current business plan, which has been prepared in light of C&W’s view of its customer profile and the costs of providing enough capacity to supply its ‘busy hour’. Even taking this into account, our estimate indicates that C&W’s average incurred termination cost on its own network is likely to be below that of the MNOs.

The relevance of different charge control and glide path allowances for H3G

6.21 C&W notes that, in the charge control for H3G Ofcom allowed H3G time to achieve efficient scale. C&W argue that, as the same types of considerations are valid for C&W’s costs, Ofcom should acknowledge that C&W is at an early point in its product lifecycle and allow time for C&W to achieve its efficient scale in this determination.

C&W’s arguments

6.22 C&W also argues that the fact that the FMC service has not yet begun commercial operations is irrelevant to the consideration of whether they should be granted a glide path. C&W state that their business case was based on their best view of the market at the time, investment has been made on that basis and their proposed termination rates were accepted by some operators.

6.23 We do not agree that the charge control on H3G is a precedent for the way in which C&W’s termination charge should be set. We note that C&W is in a significantly different position to H3G: H3G rolled out a network with substantial national coverage

Ofcom response


55

using relatively new technology with initially low customer numbers, in comparison to the level of its investment, relative to the other MNOs. This is in contrast to the ability of C&W to match investment to demand (see paragraph 6.17).

6.24 Additionally, C&W is able to benefit from economies of scope since overheads and other common costs can be shared between its FMC service and the fixed line and data services it already provides. H3G did not have this opportunity when it entered the mobile call market in the UK.

6.25 We do not agree that C&W’s position in its product lifecycle should be a relevant factor in considering efficient costs. Due to its employment of economic depreciation, the measure of costs derived from the MCT model is not sensitive to the point at which an operator is in its lifecycle.56

6.26 C&W has not yet commenced commercial operations, and so our proposed charge should not create a risk of undue disruption. We recognise that it might require some revisions of current business plans, but we consider this appropriate to the extent that previous plans were based on unreasonably high termination rates. Additionally, as we believe that the current incurred cost of termination on C&W’s network is unlikely to be higher than the average efficient costs of a national operator (see paragraphs

Therefore it is consistent with the MCT methodology that a new entrant would recover less of its initial investment in the early years of operation, when its customer numbers are lower, and a greater amount in later years as its customer base expands.

6.21- 6.24), we consider that setting a glide path from a rate above this level would be a form of entry assistance in this case, which would not be appropriate in the circumstances of a competitive market such as the UK mobile market.

C&W revised proposed rates

6.27 C&W argues that their efficient termination costs are higher than the regulated rate for Vodafone and 02 in 2009/10, but lower than the rate set for H3G for that year. Therefore, C&W states that a benchmark set somewhere between the lowest and the highest TAC may be more appropriate. On this basis, C&W propose a MTR of about 5.2ppm or 5.3ppm.

C&W’s arguments

6.28 We do not believe that setting the MTR for C&W according to C&W’s revised proposal is an appropriate manner in which to resolve this dispute for the following reasons.

Ofcom response

i) As explained in Annex 2, this proposed rate is higher than our revised range of estimates of C&W’s termination cost, given reasonable assumptions about the extent of roaming. Our modelling suggests that C&W’s costs would only be as high as C&W’s proposed rate if [ ]% of calls to C&W’s FMC service are terminated via roaming, whereas C&W has indicated that it believes that between [ ]% and [ ] % of their incoming calls will be terminated via national roaming); and

ii) The MTR proposed by C&W is significantly higher than the efficient cost of termination as estimated by the MCT cost model.

56 The timing of cost recovery under economic depreciation seeks to smooth the path of an asset’s cost recovery over time by linking it to the use of or extraction of value from that asset.


56

Timing of the decision

6.29 C&W argued that it would be more equitable to benchmark C&W’s rates against the TAC contained in the Mobile Call Termination Statement rather than using those set by the CC Determination. In this respect, C&W also submitted that there is no scope within the wording of the draft determination for any readjustment to be made to C&W’s termination rate to take into account any variations in the CC Determination following a subsequent legal challenge.

C&W’s arguments

6.30 On 2 April 2009, pursuant to section 195(4) of the 2003 Act, the CAT upheld the appeal against the Mobile Call Termination Statement and remitted the matter to Ofcom with directions, in accordance with the CC Determination.

Ofcom response

57 On the same date, Ofcom implemented the CAT’s directions by adopting a number of revisions to the Mobile Call Termination Statement,58

6.31 The implementation of the CC Determination by the CAT and subsequently Ofcom also addresses C&W’s concerns regarding any further challenges to the CC Determination.

including a revised TAC in 2009/2010 for Vodafone and O2 equal to 4.71ppm (in nominal terms). All the established MNOs are therefore now subject to the revised TACs set out in the CC Determination. C&W will not therefore be disadvantaged as compared with the MNOs since the TAC against which we have benchmarked C&W’s MTR is now in force.

Variation of charges in line with that of Vodafone and 02

6.32 C&W state that it is unclear from the draft determination whether Ofcom anticipates that as the TAC rates are amended in 2010 in line with the CC determination that C&W will need to amend its own rates to be in line.

C&W’s arguments

6.33 C&W argue that they should not be made subject to Vodafone’s charge control either directly or indirectly as this would effectively impose a charge control on C&W despite the fact that they are not subject of an SMP finding.

6.34 In the interests of promoting effective competition and cost minimisation, it is our intention to benchmark C&W’s MTR with that of Vodafone/ 02. Recognising the fact that Ofcom is not seeking to impose a charge control on C&W through its determination of this dispute, Ofcom does not, however, propose to impose a glide path on C&W equivalent to that set out in the MCT Statement for Vodafone and O2. However, it follows from the reasoning included in the Consultation and in this document that we would expect C&W’s average MTR to fall in line with Vodafone/02’s for the period of the charge control. Should parties be unable to agree termination rates in future years, we would consider carefully whether this expectation should be reconsidered in light of the circumstances at that time.

Ofcom response

57 http://www.catribunal.org.uk/files/Judgment_1083_1085_MCT_02.04.09.pdf. 58 http://www.ofcom.org.uk/consult/condocs/mobile_call_term/CTMAmendment2009final.pdf.

http://www.catribunal.org.uk/files/Judgment_1083_1085_MCT_02.04.09.pdf�

http://www.ofcom.org.uk/consult/condocs/mobile_call_term/CTMAmendment2009final.pdf�


57

Variation of charges with time of day and compliance

6.35 C&W questioned how compliance with the determination will be judged. C&W explained that their current termination rates use a time of day based approach. C&W indicated that they would not find it desirable to be forced to charge a constant termination rate across all time periods and would favour an approach under which we determined separate day, evening and weekend rates.

C&W’s arguments

6.36 Following receipt of C&W’s submission on this issue and further discussions with C&W, we considered that it might be appropriate for the resolution of this dispute to allow C&W to charge different rates according to time of day.

Ofcom’s response

6.37 In order to allow stakeholders the opportunity to comment on our proposals, on 15 April 2009, we issued a Further Consultation requesting submissions on (i) the appropriateness of allowing C&Ws charges to vary by time of day and (ii) the manner in which such charges might be set. The Further Consultation also included an explanation of our preferred option for resolving this issue.

6.38 The responses to the Further Consultation as well as our conclusion on this matter are discussed in section 7.


58

T-Mobile’s responses to the Consultation

6.39 T-Mobile argued that C&W should receive a lower MTR than we proposed in the

Consultation. Like C&W, it advanced a number of detailed arguments to support its case which we address in turn below, under the following headings:

• The nature of the C&W FMC service

• The appropriateness of the six principles as a framework

• The application of the six principles

• Ofcom’s benchmarking is inappropriate

The nature of the C&W FMC service

6.40 T-Mobile is concerned that Ofcom is extending regulation designed for the MNOs to new entrants, who have very different cost structures and supply services which may also differ in significant ways, particularly regarding the geographic area covered, or the degree of mobility. T-Mobile argues that this new entry should be matched by more detailed and sophisticated application of termination regulation. Therefore, T-Mobile considers that the regulation of the MTRs for new service delivery models warrants thorough examination as part of Ofcom’s new market 7 review.


6.41 T-Mobile suggests that in the meantime Ofcom can address the risk of market distortion by capping new entrants’ charges at the level of costs that they incur in terminating services on their networks.

6.42 Ofcom is required to resolve this dispute in a necessarily limited time. As T-Mobile recognises, given the limited time and scope of the dispute resolution process, we cannot determine the wide ranging issues affecting all termination providers and undertake the significant analysis required in the context of a market review. The market review process allows Ofcom the time and scope to consider more sophisticated regulatory remedies and options including consideration of whether recent market developments necessitate changes to the current regulatory regime. However, dispute resolution needs to fit within the framework of the current regulatory regime.

Ofcom’s response

6.43 We remain of the view that it is appropriate to resolve this dispute by benchmarking C&W’s MTR against the lowest regulated MTR. This is because, for the reasons set out in paragraph 5.22, C&W’s FMC service is a national, fully mobile service, allowing its customers to make and receive calls while on the move in the same way as the services offered by the five incumbent MNOs. The charge control on the MNOs restrict their MTRs to an appropriate level for such a service, taking into account the efficient costs of provision.

6.44 We do not believe setting termination rates equal to C&W’s incurred costs to be the appropriate methodology for resolving this dispute. An approach in which termination


59

charges were set equal to the operator’s own incurred costs could allow inefficient operators to recoup termination costs higher than those of their competitors, thereby creating a risk of distortion of competition.

6.45 Additionally, we have insufficient information to accurately determine C&W’s incurred costs or its efficient costs, given its business model. C&W’s FMC service is not yet operational which means that no reliable data on its actual costs are available. The difficulty of forecasting costs is exacerbated by the uncertainty of the proportion of calls to C&W’s FMC service that will be terminated via roaming (see cost Annex 1). Furthermore, as noted in paragraph 6.6, we have not reached a conclusion on the extent to which C&W’s costs are efficiently incurred Therefore, we do not consider it appropriate to resolve this dispute by ‘capping new entrants’ charges at the level of costs that they incur in terminating services on their networks’, as suggested by T-Mobile.

The appropriateness of the six principles as a framework

6.46 T-Mobile acknowledged the usefulness of a framework for the analysis of a dispute, but considered that the 6 principles of pricing and cost recovery were only likely to accord with Ofcom’s duties in certain cases and could be inappropriate in other cases. In particular, T-Mobile submitted that the principles are inappropriate in the determination of MTR, particularly in the case of new entrants where services and infrastructure are typically not characterised by service and price reciprocity.


6.47 T-Mobile argued that the application of the 6 principles has almost always been in cases relating to cost allocation between parties where (i) there is common infrastructure and common costs; (ii) there is service equivalence; and (iii) the same infrastructure/service is used by both parties. T-Mobile argues that in disputes over MTR, these characteristics are not met.

6.48 In considering the appropriate resolution of this dispute, we have taken into account the relevance of all of the six principles. For the reasons set out below, we remain of the view that the six principles provide an appropriate framework for the resolution of this dispute in a manner consistent with our duties. The relevant statutory obligations and regulatory principles that we have taken into account are described in section 4.

Ofcom’s response

6.49 In any case, we regard C&W’s service as a national mobile service similar to that provided by the MNOs. We have taken this into account in applying the six principles, though we do not agree that their applicability is limited in the way suggested by T-Mobile. The principles are sufficiently general that they are capable of providing a framework in a wide variety of situations with careful application to the applicable circumstances.

Cost causation 6.50 Generally, it is efficient for the party causing costs to bear them, and for the price of a

service to reflect the cost of the resources needed to provide it. Prices which reflect costs enable markets to work efficiently, allocating resources to the services which consumers value most. This is to the benefit of consumers and so is consistent with our duties under Section 3(1)(b) of the 2003 Act. Also, allowing operators to recover the costs efficiently incurred in providing a service will encourage investment and


60

innovation in relevant markets, being therefore consistent with our duties under section 3(4) of the 2003 Act.

6.51 Indeed, unless operators are able to recover their efficiently incurred costs, including a reasonable return on capital employed, investment and market entry will be stifled. The cost causation principle is therefore consistent with the duties to promote competition (section 4(3) of the 2003 Act) and may lead to benefits to consumers in the long run (section 3(1)(b) of the 2003 Act). Moreover, ensuring that inefficiently incurred costs are not recovered from customers is also clearly consistent with the duty to further the interests of consumers with respect to price, quality, choice and value for money (section 3(5) of the 2003 Act).

6.52 T-Mobile has not put forward any reason why this approach is not appropriate in the consideration of MTRs. For the reasons set out above, we therefore remain of the view that it is suitable to include this principle within the framework for resolution of this dispute.

Cost minimisation

6.53 The cost minimisation principle states that ‘the mechanism for cost recovery should ensure that there are strong incentives to minimise costs’. The incentive to make cost reductions will encourage investment in innovation and new technology in order to lower costs. Therefore, the principle of cost minimisation is consistent with our duty to further the interests of consumers taking into account the desirability of encouraging investment and innovation in relevant markets under section 3(4) of the 2003 Act.

6.54 Consideration of this principle is also consistent with our duty to further the interests of consumers having regard to their interests in respect of price and value for money under section 3(5) of the 2003 Act since a reduction in costs overall may in turn result in a reduction in prices for consumers.

6.55 Therefore, we believe that the consideration of this principle is an appropriate inclusion within the framework for resolution of this dispute. We do not consider that there are any specifics of the markets for MTR which make this approach unsuitable in the context of the current dispute.


6.56 The effective competition principle states that ‘the mechanism for cost recovery should not undermine or weaken the pressures for effective competition’. Therefore, the inclusion of this principle in our framework for analysis is consistent with our duty to promote competition (section 4(3) of the 2003 Act). We remain of the view that this principle is relevant to a consideration of MTR in the context of the present dispute.

Distribution of benefits 6.57 This principle states that ‘costs should be recovered from the beneficiaries especially

where there are externalities’. We consider that, to make an assessment of whether a charge is fair and reasonable between the parties in line with our statutory duties, it is relevant to consider how the benefits of any transaction at that charge are distributed between the parties. Taking the distribution of benefits into account is consistent with our duty to have regard to the interests of consumers in respect of choice, price, quality of service and value for money (section 4(3) of the 2003 Act).


61

6.58 We do not consider that there are any reasons why this principle would not apply in the context of a consideration of MTRs and remain of the view that the distribution of benefits principle is a relevant factor to be taken into account in resolving the dispute.

Reciprocity

6.59 This principle states ‘where services are provided reciprocally, charges should also be reciprocal’. We consider that the application of the reciprocity principle, where relevant, achieves fairness as between the parties and avoids distortions to competition. This is because reciprocal charging means that operators who supply each other with equivalent services will each pay, and receive, the same amount for those services.

6.60 We therefore believe that the consideration of this factor within the framework for resolving this dispute remains appropriate and consistent with our statutory duties. In resolving the dispute we have considered the extent to which reciprocity is appropriate in the specific circumstances of the assessment of MTR and see no reason to depart from this approach.

Practicability

6.61 This principle states ‘the mechanism for cost recovery needs to be practicable and relatively easy to implement’. Practicability is a relevant factor for us in meeting our statutory duty to resolve disputes within 4 months and consider the ease of implementation of the outcome by the parties. We have therefore attached some importance to determining a solution that is practicable to determine and reasonably simple to implement (see paragraph 5.129 above).

6.62 We do not consider that T-Mobile has advanced any reasons as to why practicability should not be taken into account within the framework for resolution of this dispute and remain of the view that it is a relevant factor which ensures an appropriate outcome.

The application of the six principles 6.63 In the next section we respond to T-Mobile’s arguments relating to the application of

the six principles to this dispute specifically.

Cost causation

6.64 T-Mobile suggested the following approach to national roaming costs. Its preferred option is to cap C&W's termination rate at an estimate of the efficient costs of termination on C&W's network. T-Mobile's view now is that this should be interpreted to mean the cost of termination in urban areas derived from the MCT model, which is approximately 2.9ppm. This would apply to all C&W terminated traffic, including traffic terminated via roaming. T-Mobile also suggested an alternative option under which, for traffic terminated via roaming, C&W would receive the benchmark MCT rate and, for traffic terminated on the FMC network, C&W would receive 2.9ppm. Ofcom would seek periodic confirmation from C&W of the proportions of traffic eligible to receive each rate.


The inclusion of national roaming


62

6.65 In relation to termination services provided while a customer is roaming, T-Mobile argued that it is a customer choice in the same way that international roaming is a customer choice. In each case it is the account holder and/or handset user that determines whether to make or receive calls outside the coverage of its home network. On this basis, T-Mobile argues that Ofcom should treat national roaming charges in the same way as those for international roaming and, further, that this implied that they should not be taken into account in setting C&W’s termination rate.

6.66 T-Mobile stated that, by including roaming charges in its estimate of C&W’s costs, Ofcom had been inconsistent with the decision not to mandate national roaming in the DECT guard band auctions and submitted that the inclusion of the costs of national roaming in the cost of termination was therefore wholly inappropriate.

6.67 T-Mobile added that, to the extent that it is appropriate to include such costs, Ofcom has inadequately considered the degree to which these costs actually affect the average cost of termination and the extent to which C&W and its customers are able to and have incentives to minimise roaming traffic.

6.68 T-Mobile also stated that it was appropriate to exclude any roaming charges that C&W may pay that would result in charges above the national MCT rates. T-Mobile stated that, as Ofcom has already determined that the national MNOs can effectively supply mobile services nationally at the MCT rate, this rate represents the upper cap on the efficient cost of provision.

Applicability of Ofcom’s MCT model to this dispute 6.69 T-Mobile stated that originating networks should pay charges that related to the

efficient costs of the specific termination service. It interpreted this to mean that C&W’s termination rate should be set using Ofcom’s MCT model and, specifically, that the calculated costs for the urban geotype should be used. It argued that the costs of termination in urban areas were most relevant given that C&W would be targeting corporate customers and locating equipment on their premises.

6.70 T-Mobile made the following points in relation to the applicability of Ofcom’s MCT cost model to this dispute:

• The detailed information incorporated into the MCT model is capable of producing reasonably robust cost estimates by geo-type;

• The MCT model provides a readily available tool for the settlement of call termination disputes;

• Ofcom had previously applied the MCT model to the analysis of costs entirely unrelated to call termination, which would suggest that it is more robust and adaptable than Ofcom suggests;

• Any concerns Ofcom has about the applicability of the MCT model to C&W’s operations are unfounded, particularly as C&W will in fact benefit from the economies of scale and scope of its roaming partner.

Ofcom’s response

The inclusion of national roaming


63

6.71 We believe that it is desirable for providers to be able to bring a service to market in the most efficient way possible. A service provided by a combination of a network with limited coverage and roaming may be efficient and we therefore consider that the cost of national roaming is a relevant factor to be included in the cost analysis, to the extent that it is an efficiently incurred cost of providing the service.

6.72 Whether these costs are in fact efficiently incurred will depend on whether, overall, C&W can provide this service the same or less cost than its competitors. In determining this dispute we therefore considered a measure of efficient costs, which we based on the regulated MTRs, to be a more relevant and reliable estimate of C&W’s termination costs than our estimate of C&W’s incurred costs.

6.73 The more relevant question still is how the costs of roaming should be treated in setting charges. We believe that the most appropriate approach is not to set C&W’s termination charge equal to its costs (including or excluding roaming) but to benchmark it to the lowest of the regulated MNOs MTR. This of course does not depend on the likely proportion of calls to C&W’s FMC service that would be terminated via roaming. We believe it would be inappropriate for C&W to recover above the efficient termination cost of a national operator regardless of what percentage of calls are terminated via roaming. Therefore, we have not had to reach a view as to the precise extent to which the cost of national roaming should be incorporated into our estimate of C&W’s costs.

6.74 An advantage of our approach is that C&W has the incentive to minimise its costs, including roaming costs (see paragraph 5.72). But in any case, we do not think it is desirable to use an estimate of the actual proportion of calls to C&W’s FMC service that are terminated via roaming when setting C&W’s MTR. This is because C&W has not yet begun operations and we cannot therefore predict what percentage of calls will be terminated via roaming and it is not practical to determine this dispute in a manner that requires frequent revision.

6.75 We do not agree with the inference T-Mobile draws from the regime for international roaming charges. The difference is related to the information which the caller has when he or she makes a call. Mobile phone customers choose to use a service that is more expensive than a fixed line. Likewise, callers to mobile phones make a decision to call a service that is more expensive than a fixed line. The current termination regime allows for these additional termination costs to be recovered from callers to mobile services, as long as they are aware they are calling a mobile service (i.e. they are calling a mobile number) because they receive the associated benefit of calling a higher quality service (i.e. a service with greater coverage) relative to a fixed service.

6.76 Callers to a mobile whilst the recipient is roaming internationally pay only the national termination charge, even though they benefit from the additional coverage. This is because a caller does not know whether the call recipient is roaming internationally and cannot therefore make an informed decision whether to incur this additional cost. Conversely, callers to C&W’s FMC service will be aware that they are calling a mobile service and will (to the extent that MTRs are reflected in retail call prices) pay no more to call FMC customers than they will to call any other mobile service.

6.77 Our attitude to competitive entry through the auction of DECT guard band spectrum reflects the principles of Ofcom’s “Awards Programme Approach”.59

59http://www.ofcom.org.uk/radiocomms/spectrumawards/awardsapproach/

Our general approach is to allow the market to determine how spectrum is used, taking account of


64

the benefits of competition and the development of innovative services. Therefore, we did not mandate national roaming in the DECT guard band auctions and nor did we preclude the owners of DECT guard band spectrum from deciding to establish national roaming arrangements in bringing a service to market. The fact that roaming was not mandated does not in any way imply a judgement that it was an inefficient way of providing a national mobile service in combination with spectrum acquired through the auction. Rather it is indicative of a technology neutral approach which allows the spectrum holder to decide on the best way to deliver the service it wishes to supply.

Applicability of Ofcom’s MCT model to this dispute

6.78 We also believe that benchmarking C&W’s termination rate with the lowest of the MNO regulated rates is consistent with the principle that originating networks should pay termination charges that relate to the efficient costs of the specific termination service, the service in this case being termination on a mobile service with national coverage.

6.79 This means that our determined MTR does not necessarily reflect the exact efficient cost of termination of calls to C&W’s FMC service, as indeed it does not exactly equal the costs of any of the MNOs to which the controls on termination charges apply. It represents the costs of an efficient, average supplier of mobile termination, rather than each MNO’s particular customer mix, for example.

6.80 We also note the following:

• We are obliged to balance cost considerations with other factors when resolving a termination rate dispute. In particular, consistent with the general guidance provided by the CAT in the TRD Judgment, we have considered an analysis of each party’s arguments, the relevance of any benchmarks and consistency with our duties in addition to a broad brush assessment of costs in reaching our determination;

• Limitations in time and information prevented us from carrying out a full analysis of the efficient cost for the provision of termination for callers to C&W’s FMC service specifically;

• There are limitations in the scope and available remedies for dispute resolution. The consideration of any option which would significantly change the charging regime for mobile termination, for example by disaggregating charges by location of customer type, would be more suitable to the larger scope of a market review (see paragraph 6.42).

6.81 We further disagree with T-Mobile’s assertions that the MCT model, disaggregated by geo-type, provides an appropriate basis for the settlement of this dispute since the usefulness and relevance of estimates from the MCT model of costs in urban areas is subject to the following limitations:

• The MCT model, while providing a robust estimate of the efficient national average cost of termination for an MNO in aggregate, was not designed to calculate with precision the cost of termination in specific geo-type areas and is not as reliable when it is used this way. In particular, the inputs on which the model is based were not disaggregated in this manner and there has been no calibration of model outputs with accounting data or asset counts at this level of granularity;


65

• Additionally, there are many uncertainties regarding exactly how an efficient geo-type-specific operator would operate. The implicit assumption we would be making if we were to rely on the MCT model in such a way is that the costs of a local operator are similar to those of a national operator with all of the associated economies of scale and scope. This is not necessarily appropriate to a stand alone venture;

• C&W’s FMC service provides callers with a mobile termination service with national coverage, therefore it is more appropriate to use an estimate of efficient costs which incorporate the average cost across all geo-types than to use a disaggregated estimate;

• It is unclear where C&W’s future customers will be located and therefore unclear in which geo-types C&W will roll out its physical infrastructure. Therefore, even if we ignore the fact that C&W are providing a service with national coverage and exclude all coverage provided by national roaming from our analysis (which we do not propose to do), benchmarking against output from the MCT model that has been disaggregated by geo-type may be misleading.

6.82 We agree that C&W might benefit from the economies of scale and scope of its national roaming provider, to the extent that it is able to negotiate a roaming charge that is close to the efficient costs faced by its roaming partner. This is consistent with use of the MCT model at the national level to provide an appropriate estimate of the efficient cost of termination of calls to C&W’s FMC service. C&W may however incur additional costs associated with managing its own network.

Cost minimisation

6.83 T-Mobile argued that if termination rate regulation sets these rates regardless of costs, and in all cases above them, then it creates no additional encouragement for the pursuit of cost reductions and efficiency, and serves only to preserve inefficient pricing. T-Mobile suggests that we have calculated the efficient cost of delivering the services and that setting C&W’s charges at this level (Option 3 in the Consultation) would retain C&W’s incentives to develop genuinely innovative, lower cost ways of service delivery.


6.84 T-Mobile also submitted that Ofcom’s arguments in favour of allowing charges above estimated efficient costs are not consistent with how regulators generally apply RPI-X regulation. Further, they argued that cost minimisation should be accorded relatively little significance in relation to determining MTRs because competition in relation to services other than termination (such as network access and network origination services) can be expected to drive operators to adopt the least cost means of service provisions.

6.85 There are two components of the application of the cost minimisation principle to this dispute i) incentives to

Ofcom’s response

minimise costs and ii) incentives for efficient entry

i) Incentives to

. It is useful to address these two components separately:

minimise costs - As it does not allow for C&W’s termination charge to vary directly with its incurred cost of termination, benchmarking against a regulated rate retains C&W’s incentive to minimise its termination cost. However,


66

we acknowledge that our determination is not unique in this respect - in fact any option under which charges are not tied to actual incurred costs would provide this incentive.

To clarify this point, we believe that this option is consistent with the first component of the cost minimisation principle, but do not assert that it is the only option that could be consistent.

ii) Incentives for efficient entry

To clarify, our method of determining this dispute is consistent with the second component of the cost minimisation principle if C&W is providing its service more efficiently. If C&W is providing a ‘cheaper’ service, then our method could allow some inefficient entry to the extent that MNOs are unable to competitively respond to such entry. However, the MNOs all incur these same lower costs when serving C&W’s targeted customer base. Therefore, in this circumstance we would consider that it is still desirable to benchmark C&W’s rate with the regulated rates for competitive neutrality reasons (see paragraph

- From our analysis of C&W’s incurred termination costs, it appears likely that C&W’s average incurred termination costs will be lower than that of the MNOs (see paragraph A2.6). It is unclear whether (a) this is due to C&W providing its service more efficiently (in which case they should benefit from their reduced costs) or whether (b) as argued by T-Mobile, they are providing a ‘cheaper’ service (presumably by serving a customer base which is cheaper to serve), although they are, in principle, providing a similar national mobile service.

5.106).

6.86 Due to the limited time and information available, we did not carry out a comprehensive assessment of the efficient cost of termination on C&W’s FMC service specifically (see Annex 1), that is, the minimum cost given C&W’s chosen technology and business model. However, to inform our analysis of the likely efficient costs of termination on a network with national coverage, we considered output from Ofcom’s MCT cost model. We also carried out a high level assessment of C&W’s incurred termination costs.

6.87 The MCT model’s estimate of the efficient termination cost of a national operator, 4.0ppm (in 2006/07 prices) and the benchmark we have proposed to use in determining this dispute, 4.4ppm (in 2006/07 prices), fall within our revised range of estimates of the incurred cost of termination for C&W’s FMC service, 3.2ppm – 5.0ppm (in 2007/08 prices). Therefore, although we have not resolved this dispute in accordance with Option 3 of the Consultation (i.e., setting a rate equal to our best estimate of C&W’s average incurred termination costs), the rate we have determined is also consistent with Option 3, under assumptions which are within our reasonable range.

6.88 We do not agree that our approach to cost minimisation incentives is inconsistent with standard approaches to RPI-X, or that such incentives are not relevant in this case. Our approach is consistent with the approach taken in setting the mobile termination charge controls for example. Thus, in footnote 145 to paragraph A5.104 of the Mobile Call Termination Statement we set out the principle that “to the extent that MNOs are able to achieve greater efficiencies than the average efficient operator benchmark, it is reasonable that they should gain the associated benefits”. The relevance to MTRs of incentives to reduce costs is also clear elsewhere in that statement (see for example paragraphs 8.68 and 9.6 – 9.8).


67

6.89 To the extent that T-Mobile’s argument is that the costs of a DECT guard band operator are now the efficient costs of supplying mobile termination for the purposes of setting a charge control, it would, by implication, mean that all MNOs should receive lower rates based on the costs of using DECT guardband spectrum in certain areas. It would not be appropriate to set a charge for just one operator on this basis because this would create a risk of distortion of competition.


6.90 T-Mobile submitted that there is a fundamental flaw in Ofcom’s reasoning about effective competition given that Ofcom could equally put such arguments in relation to allowing BT’s fixed termination charges to be set at the level of mobile termination charges.


Fixed operators

T-Mobile cannot match the C&W FMC service 6.91 T-Mobile added that under the proposed approach, C&W will receive an MTR based

on the costs of investing in a national network while terminating the majority of calls on its own infrastructure, leveraging existing fixed assets which are likely to cost a fraction of the cost of a national network.

6.92 T-Mobile also argued that the national MNOs cannot duplicate this approach because they have to ensure national coverage for mobile services and cannot address only a small target customer base. It argued that C&W can cross-sell at low cost to existing closed user groups that will themselves be naturally inclined to minimise off-net calls.

Asymmetric regulation

6.93 T-Mobile argued that Ofcom is applying asymmetric regulation in permitting termination rates that are unrelated to the efficient costs of termination on the relevant network. T-Mobile argues that, by allowing termination charges unrelated to costs, Ofcom is in effect sponsoring new entry by allowing for the subsidisation of new entrants’ retail offers.

Competitive distortions 6.94 T-Mobile states that the proposed approach effectively requires callers to cheaper

networks to pay more than the costs of the calls they initiate, and in doing so subsidise the outgoing calls of the dialled network’s customers.

6.95 T-Mobile submitted that Ofcom has ignored statutory obligations in relation to the encouragement of investment and ignored the continuing investment in network infrastructure the national MNOs make, compared with the use of low powered 2G spectrum.

6.96 T-Mobile submitted that national operators, such as T-Mobile, use their income from termination charges, which are above cost in some areas, in order to recover their overall costs of national service provision. Therefore, the MNOs have not been allowed any excess margin which they could use to cross-subsidise retail offers in urban areas. As a result, T-Mobile considers that Ofcom’s approach would give C&W


68

alone excess margins and, in doing so, Ofcom would create competitive distortions that threaten current efficient consumer outcomes.

6.97 Fixed line services which do not offer mobility do not compete in the same market as mobile operators. Our reasoning under the effective competition principle could not therefore justify setting BT’s fixed termination charges at the level of mobile termination charges. The reasoning applied in the Consultation (see paragraph

Ofcom’s response

Ofcom’s flawed reasoning of this principle

5.100 above) relates to providers of mobile services with national coverage competing with each other, not to providers of fixed call services. C&W’s service is a fully mobile service with national coverage not a fixed line service (see paragraph 6.42).

6.98 Furthermore, to the extent that there is substitution between fixed and mobile services (albeit not to an extent which would place them in the same market), competition between fixed and mobile operators was taken into account when developing the current MNO charge control against which this determination is benchmarked.60

T-Mobile cannot match the C&W FMC service

6.99 We appreciate that due to licence obligations it might not be possible for T-Mobile to replace its existing network with technology similar to C&W’s. But this does not mean it cannot compete effectively with the C&W FMC service. We believe that T-Mobile can offer differentiated services to different customer groups, and in particular services or tariffs that may target business customers, in addition to the other services it provides. We also understand that T-Mobile can install pico cells, such as those employed by C&W in providing its FMC service, in high density areas where the use of such technology could lead to cost savings.

Asymmetric regulation 6.100 Setting C&W’s termination charge equal to the lowest regulated MNO termination

rate, which is close to our best estimate of C&W’s efficient costs of termination, does not constitute entry assistance. Rather, this determination seeks to put C&W on a level playing field with the MNOs with which it will compete in the provision of its FMC service.

6.101 It is possible that C&W’s business strategy (i.e. targeting predominantly business customers) results in it facing a lower average efficient termination cost than that faced by the MNOs. However, we do not believe that C&W’s strategy should be considered sufficiently different as to justify using a different cost benchmark than that used for all other operators with national coverage. In any case, the time and information limitations of dispute resolution mean that this is not possible in practice.

6.102 As noted above, the MTRs of the MNOs reflect the costs of an average efficient operator, rather than each MNO’s particular customer mix. In not reflecting C&W’s customer mix in C&W’s average MTR, our resolution of this dispute is therefore consistent as far as possible with the regulation applying to the MNOs.

60 See paragraphs 7.57- 7.63 at http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf



69

6.103 Additionally, if we set C&W’s average MTR lower than the average MTRs of the incumbent MNOs (even if C&W’s cost of termination is lower than that of the MNOs) this could distort competition against C&W (see paragraph 6.109).

Competitive distortions 6.104 We have considered T-Mobile’s argument that our approach would make T-Mobile

uncompetitive vis-à-vis C&W or unable to recover the costs of providing termination on its national network. In doing so we have considered it important to distinguish between: (i) the question of whether termination revenues are sufficient to recover termination costs, and (ii) the waterbed effect – i.e. the fact that termination charges above (or below) cost are likely to lead to lower (higher) retail prices to mobile customers.

6.105 The primary factor relevant to T-Mobile’s ability to recover its costs is its own MTR. Our proposed approach does not change T-Mobile’s MTR, nor the feature that it is currently averaged over all customer types and so is likely to be above cost for some customers (for example some business customers) and below cost for others. T-Mobile’s termination charges will continue, on average, to be at a level sufficient to fund the costs of an efficient national provider of termination. To the extent that investment is currently funded by termination revenues, T-Mobile will be able to continue this.

6.106 The other of T-Mobile’s concerns relates to the potential for a differential waterbed effect for business and other customer types, through the competitive process. That is, profits from MTRs above cost for business customers may be competed away through lower prices to those customers, initiated by C&W and then responded to by other MNOs. If this effect is sufficiently material, there could also be a corresponding effect in the opposite direction for other customers.

6.107 The first condition for this to occur is that C&W’s MTR is above its incurred costs. Otherwise, C&W would not be able to subsidise lower retail prices, funded by termination profits. The limited evidence that we have on C&W’s incurred costs suggests that this condition might apply (see paragraph 6.85), but that it does not necessarily do so.

6.108 We consider that our proposed approach could allow C&W to earn a profit on termination relative to its incurred cost (see paragraph 6.85) (although, because of the limitations in our cost estimates, we cannot ascertain what this margin would be, and it will also depend on C&W’s efficiency). This means that there is the potential for a differential waterbed effect, although it is not certain that it will occur. The size of any differential waterbed effect would depend on the extent to which C&W’s incurred costs were below its MTR and its impact in the market (such as the volume of customers it attracts and the extent to which it elicits a competitive response from the MNOs). However, given that MNOs already target corporate customers specifically (see paragraph 5.99), a differential waterbed might already exist between business customers and other customer types.

6.109 We acknowledge that all the options available to us carry some disadvantages. However, we have had to weigh up the risks of alternative options in relation to the circumstances in this particular case. For example, we also consider that C&W could be disadvantaged relative to T-Mobile and the other MNOs in serving its targeted customer base, were we to set C&W’s termination rate below the MTRs of the incumbent MNOs. At the margin, if T-Mobile receives a rate in excess of its efficient costs of serving business customers but C&W does not, it might enable T-Mobile to


70

profitably offer retail prices that C&W would be unable to match, not because of inferior performance but because of the disparity in termination rates. Therefore, if a deaveraged termination rate was applied to C&W, this risk of a distortion of competition against C&W would be present. This is because in the current regulatory regime for other MNOs (which is beyond the scope of this dispute) the termination rates of MNOs against whom C&W might be competing are not also deaveraged by customer type. We therefore believe that it is desirable to benchmark C&W’s MTR against the relevant MNO rate, for competitive neutrality reasons.

Reciprocity

6.110 T-Mobile contends that the C&W service is not equivalent to that provided to the national MNOs, stating that it is simply a means of leveraging C&W’s fixed network and existing backhaul capacity from client premises and is available only to corporate customers that use C&W’s fixed line services.


6.111 T-Mobile argued that Ofcom’s stated desire to see new entrants’ charges aligned with those of incumbents does not address the fundamental question of whether it is appropriate to do so given substantial differences in the nature of the service being provided.

6.112 T-Mobile also submitted that it is likely that the costs of termination on the C&W FMC network are similar to the price of termination on a fixed network.

6.113 C&W’s service is a mobile service with national coverage. We consider that both parties to this dispute supply a national mobile service (see paragraphs

Ofcom’s response

5.116 and 5.145). In light of this, we remain of the view that the services offered are sufficiently similar in nature to justify consideration of the reciprocity principle.

6.114 We do not believe that C&W’s decision to target business customers sufficiently differentiates its business strategy as to justify using a different cost benchmark than that used for all other operators with national coverage. Additionally, we consider that the MNOs can also target business customers (see paragraph 6.101). Therefore, in this case, we consider it appropriate to align the new entrant’s charges with those of the incumbent MNOs, consistent with the principle of reciprocity.

6.115 As discussed at paragraph 6.85 the benchmark we have used falls within our revised range of estimates of the incurred cost of termination for C&W’s FMC service. According to our cost analysis, even if all calls to C&W’s FMC service were terminated on C&W’s own network the cost of termination is likely to be higher than the cost of termination on a fixed network (see Annex 2), contrary to T-Mobile’s assertion (see paragraph 6.127).

Distribution of benefits

6.116 T-Mobile stated that, in the context of a competitive market, consideration of the distribution of benefits between competitors is prima facie irrelevant, since the benefits will be distributed to the customer in any event.



71

6.117 T-Mobile argued that the Consultation’s discussion of the criterion of distribution of benefits ignored the benefits to call recipients. T-Mobile stated that, while it will generally be efficient to price termination at the level of efficient costs, the existence of calling externalities may warrant some of the cost of termination being recovered from the call recipient. Therefore, T-Mobile argues that there is potentially a case for termination charges to be set below, not above, efficient costs.

6.118 T-Mobile considered that Ofcom’s application of the principle of distribution of benefits is flawed since it results in allocation of 100% of the benefit of C&W’s lower costs to C&W and its customers, and the beneficiaries of termination would not receive any benefit from C&W’s lower termination costs, as set out at 6.57.

6.119 We agree that in a competitive market the benefits of lower costs are likely to be passed on to consumers. However, in determining an MTR that is fair and reasonable we consider it relevant to consider how alternative solutions would distribute any benefits from C&W’s possible lower incurred costs.

Ofcom’s response

6.120 We consider that if mobile providers are able to achieve greater efficiencies than the efficient operator benchmark, it is reasonable that they should gain the associated benefits, in order to retain incentives for efficient entry.

6.121 We consider that the distribution of benefits principle supports our methodology for resolving this dispute as it allows C&W’s customers to benefit if C&W is able to lower its costs below that of its competitors but avoids callers to C&W paying more than to call other mobile networks (to the extent that relative termination rates are reflected in retail call prices).

6.122 See paragraph 6.104 - 6.109 for consideration of possible distortions to competition arising from setting charges higher than costs, and see paragraph 6.83 for consideration of lower costs resulting from efficiency as opposed to lower costs resulting from serving a ‘cheaper’ customer base.

Practicality

6.123 According to T-Mobile, given that Ofcom was in fact able to calculate the likely efficient costs of termination, it arguably cannot assert that practicability prevents it from adopting Option 3 of the Consultation.


6.124 As discussed at paragraph

Ofcom’s arguments

6.85, we have estimated a range for C&W’s incurred

6.125 Additionally, this range does not necessarily represent the

unit cost of termination of between 3.2ppm – 5.0ppm (2007/08 prices). We have not, however, derived a robust cost estimate from this analysis as we do not know i) where C&W’s average termination cost lies within this range and ii) whether the range is in fact accurate.

efficient cost of termination for C&W’s FMC service. Our best estimate of the efficient cost of termination for C&W’s FMC service is the MCT model’s estimate of the efficient total termination unit cost for 2G/3G MNOs, 4.0ppm (2006/07 prices).


72

6.126 T-Mobile has suggested that the cost of termination in the urban geotype provides an appropriate estimate of efficient costs. For the reasons set out above (see paragraph 6.81), we do not consider that this provides a robust estimate.

Ofcom’s benchmarking is inappropriate

6.127 T-Mobile submitted that Ofcom identified that the efficient costs of termination on the C&W DECT guard band network are likely to be substantially below those on the 2G/3G national networks, but nonetheless applies a uniform rate across these different technologies.


The benchmark is inappropriate

6.128 In T-Mobile’s view, a DECT guard band network is inherently cheaper than a national 2G/3G network. T-Mobile submitted that the relevant benchmark is the urban costs of the 2G/3G networks, not the national rate.

6.129 T-Mobile also argued that Ofcom has ignored C&W’s proposed business model, commenting that C&W will deploy infrastructure only in the premises of existing corporate clients, which means that it will be able to ensure that its investment is specifically tailored to the traffic volumes of that customer.

The potential detriment to consumers of significantly above cost MTRs

6.130 T-Mobile considers that Ofcom has not sufficiently addressed the concerns outlined in its own consideration of the potential detrimental impact on consumers of significantly above cost mobile call termination rates, as set out in Section 7 of the Mobile Call Termination Statement particularly in relation to the five headings identified in the document.

6.131 T-Mobile considers that Ofcom has failed to adequately address: (i) on what basis cross-subsidisation by C&W will not create a distortion in retail competition: (ii) whether this distortion will be outweighed by the alleged wider benefit of allocating the benefits of lower costs exclusively to C&W; and (iii) how this choice leads to a lower level of distortion than the alternatives.

6.132 We noted that C&W’s likely lower incurred termination costs may be a result of C&W (i) providing its service more efficiently or (ii) serving a ‘cheaper’ customer base. However, C&W is providing a national mobile service in essence similar to those of the MNOs. Therefore, we believe that benchmarking to the lowest regulated MTR is the appropriate method for determining this dispute under either scenario (see paragraph

Ofcom’s response

The benchmark is inappropriate

6.87).

6.133 We disagree that the relevant benchmark in this case should be the urban costs of the 2G/3G networks for reasons which are set out above at paragraph 6.87.

6.134 In explaining the reasons for identifying the lowest regulated rate as the most appropriate benchmark for C&W’s MTR, we have acknowledged that C&W will deploy infrastructure only in the premises of existing corporate clients, which means


73

that it will be able to ensure that its investment is specifically tailored to the traffic volumes of that customer (see paragraph 5.36).

The potential detriment to consumers of significantly above cost MTRs

6.135 We believe that our approach is compatible with Ofcom’s statements of principle in the Mobile Call Termination Statement for a number of reasons. Firstly, it is unclear whether benchmarking to the lowest regulated MTR actually does afford C&W an MTR significantly above its termination costs (see paragraph 6.85).

6.136 Secondly, in acknowledging the risk that this determination might allow C&W to recover a termination charge higher than its average incurred cost of termination, we have sufficiently addressed the concerns as set out in section 7 of the MCT Statement, namely61

• Excessive prices overall – to the extent that relative termination rates are reflected in retail call prices, our determination avoids callers to C&W paying more than to call other mobile networks (see paragraph

:

6.121);

• Inefficient structure of prices – our determination ensures that C&W’s termination rate is benchmarked to the termination rates of the MNOs. Therefore, it also addresses the danger that excess termination profits could be used to subsidise retail prices to an extent which could result in a distorted pricing structure. We consider the possibility of a differential waterbed effect (see paragraphs 6.107 - 6.109);

• Distortion of consumer choice – we have considered the risk of distortions to competition under each option (see paragraphs 6.104 - 6.108);

• Inequitable distributional effects – we do not consider that our proposed determination will lead to inequitable distribution effects (see paragraph 6.121);

• Risk of anticompetitive behaviour – we consider that C&W could be disadvantaged relative to T-Mobile and the other MNOs in serving its targeted customer base were we to set C&W’s termination rate below the MTRs of the incumbent MNOs (see paragraph 6.114).

6.137 We have addressed concerns that possible cross-subsidisation by C&W could create a distortion in retail competition (see paragraph 6.109) but, because T-Mobile and the other operators can compete effectively against C&W’s offering (see paragraph 6.114), we consider this risk to be sufficiently low. We have also acknowledged that all the options available to us carry some disadvantages (see paragraph 6.109).

6.138 We have also argued that it is appropriate to allow C&W to retain the benefits if they are able to provide national mobile termination services at lower cost than their competitors (see paragraph 6.121).

6.139 Finally, as discussed in paragraphs 5.112 to 5.115, we consider that the alternative options would risk distorting competition against C&W and may risk deterring efficient entry. Therefore, we consider the risk of a detrimental distortion of competition arising from our chosen option for determining this dispute to be lower than all other options available.

61 See paragraphs 7.29-7.73 at http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf

http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf)paragprahs�


74

H3G’s responses to the Consultation

A market review would be appropriate

6.140 In its response to the Consultation, H3G stated that it believes there is significant regulatory and commercial uncertainty in relation to Ofcom’s policy with respect to the provision of MCT services by new entrant operators who are currently charging unregulated termination rates.

H3G’s arguments

6.141 To the extent that H3G’s own SMP designation remains, however, H3G remains of the view that the issue of new entrant MCT and the appropriate regulation thereof should be dealt with by means of a market review (and notes that H3G was subject to a market review less than a year after commencing operations).

6.142 As previously stated, Ofcom has a statutory obligation under section 188 of the 2003 Act to resolve disputes that it accepts within four months. This does not allow enough time for an analysis akin to a market review. However, if in the future Ofcom deems it appropriate to undertake a market review, then it will consider this matter at that time.

Ofcom’s response

6.143 Ofcom is conducting an assessment of Ofcom’s overall approach to regulation of the mobile sector in the medium term. In addition, Ofcom is currently planning to publish a preliminary consultation on Mobile Call Termination during the second quarter of 2009, as a precursor for a Market Review covering mobile termination for the period beyond 2011.

Conclusion on responses to the Consultation

6.144 For the reasons set out above, Ofcom does not consider it appropriate to depart from the position adopted in the Consultation with respect to the average MTR applicable for the resolution of this dispute. Accordingly, we consider it appropriate to benchmark the MTR in respect of C&W’s FMC service to the lowest regulated rate, which is that applied to Vodafone and O2. As set out in the Mobile Call Termination Statement, that rate is set at 4.71 pence per minute.

6.145 However, as set out below in section 7, Ofcom has considered the extent to which C&W should be allowed to vary its charges by time of day whilst maintaining an average charge of 4.71 pence per minute.


75

Section 7

7 The Further Consultation Time of Day weights

7.1 C&W’s response to the Consultation queried how compliance with the determination would be judged, in particular whether the determination would allow C&W to vary its charges by time of day (Daytime, Evening, Weekend) whilst still being consistent with a weighted average charge of no more than 4.71ppm (the rate proposed in the draft determination). C&W also expressed concern that, should they not be allowed to vary their charge according to time of day, the restrictions placed on them would be more stringent than those imposed on the incumbent mobile network operators’ (MNOs) under the charge control against which we have proposed to benchmark C&W’s MTR.

7.2 In order to resolve this dispute, we considered that it may be appropriate to allow C&W to vary charges by time of day in order to ensure that it is not subject to greater restrictions on its pricing freedom than those applied to the regulated MNOs.

7.3 In order to allow stakeholders the opportunity to comment on our proposals for resolution of the dispute, we therefore issued the Further Consultation in order to determine (i) the appropriateness of allowing C&W charges to vary by time of day in resolving this dispute, as well as (ii) the manner in which the time of day weights for such charges might be set, if appropriate. Paragraphs 7.4 to 7.31 repeat the analysis and reasoning in the Further Consultation. We then set out the responses and our further analysis.

The reasons why it may be appropriate to provide C&W with greater flexibility

7.4 It is common practice for telecoms operators’ wholesale and retail tariffs to vary by time of day according to a “tariff gradient”. Tariff gradients are employed to increase the efficiency with which network capacity is used. In general charges are highest during the daytime peak, when demand to use the network is at its highest, and lowest in off-peak periods in the evening and at weekends when demand is relatively low. This ensures that the costs of network capacity are recovered primarily from peak users whose demand causes the capacity to be installed, and also allows operators to send price signals that promote the usage of that capacity in off-peak times, when it might otherwise be spare, and consequently can be used at very low (marginal) cost. To the extent that these more accurate price signals would affect caller behaviour, they could allow an operator to provide the same level of call termination minutes more efficiently – i.e. if some of the calls that would have been received at peak times are received at lower cost times of the day, the operator would be able to supply the same number of call minutes on a lower capacity (cheaper) network.

7.5 The incumbent MNOs, whilst subject to a charge control which sets a headline TAC, are able to vary their charges by time of day, provided that the weighted average charge reflects the headline TAC. This allows them flexibility and the efficiency benefits associated with tariff gradients as described above.

7.6 Considering C&W’s response to the draft determination, we therefore considered it appropriate to explore options which would allow C&W to vary its charges by time of day in order to allow it similar flexibility in determining its own charging structures.


76

This is particularly relevant in this case where we proposed in the draft determination to determine a MTR for C&W’s FMC service which is benchmarked against the regulated TAC for Vodafone and O2.

Time of day splits

7.7 If we were to resolve the dispute by allowing C&W the flexibility to vary its charges by time of day, whilst maintaining an average termination charge across time periods, we might need to determine the appropriate traffic weights or ‘time of day splits’ between day, evening and weekend for incoming calls of C&W’s FMC service in order to assess how C&W might be permitted to operate a time of day pricing policy.

7.8 Under the charge controls in place following Ofcom’s market review of the Mobile Call Termination market62

7.9 As C&W’s FMC service has yet to be launched on a commercial basis, no traffic information is available for the preceding year and it is therefore not possible to set the weights upon which C&W’s time of day variations are based in the same manner as the calculation of the AIC for the MNOs. Accordingly, in determining this dispute in a manner which gives C&W the flexibility to vary charges by time of day or by setting maximum charges by time of day we need to estimate C&W’s day, evening, weekend call splits for its FMC service.

, the MNOs are required to use the proportions of termination minutes in the previous year received during the day, evening and weekend charging periods as the weights to compute the weighted average interconnection charge (AIC). This AIC must comply with the TAC for the current year.

Determination of a proxy for the time of day traffic weights of C&W’s FMC service

7.10 The evidence that is available to assess traffic splits reliably is inevitably limited, given the lack of evidence from C&W’s FMC service in commercial operation. In arriving at our view of a reasonable proxy for the traffic profile of C&W’s FMC service, we have considered the following evidence:

(a) The average time of day traffic profile of the total consolidated traffic of the five MNOs in 2006/07: (Day - 48%, Evening – 29%, Weekend – 23%)63

(b) The time of day traffic profile of C&W customers’ outgoing calls to mobiles

64

(c) The time of day traffic profile of C&W customers’ incoming calls from mobiles

: (Day – [ ] [% 55%-65%], Evening – [ ] [% 15%-25%,] Weekend – [ ] [% 10%-20%])

65

7.11 We do not consider that any of these time of day splits provide a robust forecast of the time of day traffic profile of calls terminating on C&W’s FMC service which will

: (Day – [ ] [% 50%-60%], Evening – [ ] [% 15%-25%], Weekend – [ ] [% 15%-25%]

62 http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf 63 This information was collected by Ofcom to check the MNO’s compliance with their current charge controls. 64 This profile is based on all C&W services other than the FMC service. The FMC service is not yet operational. 65 This profile is based on all C&W services other than the FMC service. The FMC service is not yet operational.



77

serve predominately business customers. Since C&W’s FMC service is expected to serve mainly business customers (who are likely to receive a greater proportion of their calls in business hours – because this is when they will be at work – than the average mobile user), it seems likely that C&W’s incoming call profile will have a higher proportion of day time traffic than that of the MNO average traffic profile. MNOs will typically serve a more diverse base of customers including business customers and individual consumers and therefore option (a) is unlikely to provide a reasonable proxy.

7.12 All of the other traffic splits, (b)-(c), have a higher proportion of day time traffic. Traffic split (c) considers calls from mobile, but not calls to mobile and is therefore also not considered a reasonable proxy for calls to business customers on C&W’s FMC network.

7.13 Traffic split (b) reflects calls to mobile from C&W’s customers. We understand that C&W’s customers are typically business customers and therefore this traffic split may be closest to the correct forecast to be used in estimating the likely traffic received by customers of C&W’s FMC service. We recognise, however, that this forecast is based upon C&W customers alone and may not therefore reflect the mix of callers of other networks to C&W’s FMC service. However, given the limited information available, we consider that this is likely to provide the most accurate proxy for traffic received by the C&W FMC service and have therefore based our time of day calculations on this traffic split.

7.14 Given the inherent uncertainty in such a forecast and the underlying confidentiality surrounding the exact figures for C&W’s network, we have considered it appropriate in this case to round the exact split for option (b) to provide the following weighting: Day – 65%, Evening – 20%, Weekend – 15%. We consider that the difference between the traffic weights set out above and the exact figures contained in option (b) is sufficiently small so as not to have a material effect on the outcome of the dispute. This involves a significantly higher weight on the day time period than the average time of day traffic profile of the total consolidated traffic of the five MNOs, as seems plausible given the expected customers of the FMC service.

Options for implementation

7.15 In the Further Consultation, we set out 4 Options:

• Option (i): determining a flat rate (e.g. 4.71ppm) across all times of day;

• Option (ii): determining an annual average weighted charge across all time periods on the basis of actual traffic. Compliance with the average charge would be calculated at the end of a 12 month period, using C&W’s preceding actual year call volumes attributable to T-Mobile;

• Option (iii): determining the average weighted charge (for example 4.71ppm) and the time of day traffic weights (for example – Day – 65%, Evening – 20%, Weekend – 15%) and allowing C&W to set and change their time of day pricing structure as it sees fit, within those constraints;

• Option (iv): determining specific maximum time of day call charges based on (a) the average charge, (b) the underlying traffic assumptions (Day – 65%, Evening – 20%, Weekend – 15 %), and (c) the time of day pricing structure incorporated in C&W’s proposal.


78

Option (i)

7.16 Option (i) would be the least complicated to determine and to implement of all options considered. It would also provide certainty that C&W’s average charge recovered will be equal to the determined rate in this matter.

7.17 However, this option would be significantly more restrictive than the charge controls under which the MNOs operate since it allows no flexibility to C&W in its charging structure. Therefore, it would appear to discriminate against C&W if Ofcom were to restrict the ability of C&W to vary its own charges whilst a regulated MNO, against whose charges we propose to benchmark C&W’s MTR, is afforded greater flexibility in its own charging structure. This is particularly the case where T-Mobile, the other party to the dispute, itself benefits from such flexibility.

7.18 Additionally, varying charges according to time of day traffic levels allows operators to send more accurate price signals and increase the efficiency of their operations. Therefore, preventing C&W from varying their charges by time of day could lead to C&W missing out on possible efficiency benefits.

Option (ii)

7.19 Option (ii) would be relatively simple to determine, as it would not require us to take a view on the appropriate time of day weights. However, under this option, T-Mobile could only calculate compliance with the average TAC at the end of the 12 month period once the volume of traffic it had sent to C&W’s FMC service was known.

7.20 This option would however provide certainty that C&W’s average charge recovered will be equal to the determined rate in this matter.

7.21 As this option involves using weights which will remain uncertain until the end of the 12 month period, it would be complicated for C&W to implement. In order to implement this option C&W could have to vary their charges throughout the year if their forecasts are not accurate. This option might also require a balancing payment at the end of the year between C&W and T-Mobile at the end of the 12 month period, in order to ensure compliance with the determination, if C&W has over-recovered at the end of the 12 months. These complications would tend to reduce the price signal benefits of varying charges by time of day, because T-Mobile would be aware that it would recover any overpayment it made to C&W and thus have less incentive to take any action to reduce the proportion of calls its customers made to C&W in peak (high cost) times.

7.22 C&W indicated to Ofcom that, if we were to resolve the dispute by adopting Option (ii), they would be forced to set a constant rate across all time periods in order to ensure compliance and stability of charges. Where this is the case, this option would result in the same outcome as Option (i).

Option (iii)

7.23 Option (iii) would allow C&W the flexibility to vary its charges, provided the charges set are consistent with the average charge, and the time of day weights as determined and would therefore provide C&W with a similar termination charge arrangement to that of the MNOs.

7.24 However, due to the uncertainty of the C&W FMC service’s traffic profile (and therefore the significant risk of inaccuracy of the determined weights) resolving the


79

dispute in this manner gives rise to significant risks, including that C&W would be able to over recover – as they will have the ability to vary charges in their favour once their traffic profile becomes known to them, by increasing the charge for that time(s) of day (Day, Evening or Weekend) which the determined weights have underestimated.

Option (iv)

7.25 Option (iv) would be relatively simple to determine, as it would result in defined maximum pence per minute rates for day, evening and weekend, and would give both parties certainty going forward.

7.26 Allowing for charges to more closely reflect the traffic volumes would allow C&W some of the associated efficiency benefits of more accurate price signals.

7.27 However, this option would not provide C&W with flexibility to vary its charges (for example by increasing daytime rates and reducing evening and weekend rates). The option would also not provide certainty that C&W’s average charge recovered will be equal to the determined rate since the actual amount received by C&W might vary from that average rate if its traffic profile departs from the weights on which these maximum charges have been calculated. Nevertheless, we consider that there is a lower risk of over-recovery than under Option (iii), because the limitations on C&W’s flexibility to vary charges will mean that they do not have the ability to vary charges in their favour once their traffic profile becomes known to them.

Ofcom’s preferred option in the consultation

7.28 On balance, we considered Option (iv) to be the most appropriate option to resolve this dispute as it allows C&W to retain some of the benefits of time of day pricing, but minimised the risk and uncertainty that necessarily result from basing a determination on incomplete information.

7.29 We considered that C&W’s FMC service will provide a mobile service with national coverage in competition with the established MNOs by using a different balance of technology to that used by the MNOs. Our preferred option allowed, as far as was practically possible, C&W’s FMC service the same pricing structure as is available to the MNOs. Therefore, we considered that our preferred option, Option (iv), complies with our duty to take account of the desirability of carrying out our function in a manner which, so far as practicable, does not favour (a) one form of electronic communications network, electronic communications service or associated facility; or (b) one means of providing or making available such a network, service or facility as set out in section 4(6) of the 2003 Act.

7.30 The proposed option is aimed at providing C&W with the benefits of time of day pricing that are available to its competitors and therefore enables a level playing field in which the C&W FMC service can operate. Therefore, we also consider that our proposed approach complies with our duties not to discriminate and to promote competition (as set in sections 3(1)(b), 3(4)(b) and 4(3) of the 2003 Act).

7.31 In addition we consider that it is fair and reasonable between the parties to propose Option (iv) as this option will provide certainty to both parties and balances the risk of over- or under- recovery with the desirability of allowing C&W’s charges for the FMC service to more closely reflect network costs, therefore allowing C&W some of the associated efficiency benefits of sending more accurate price signals.


80

C&W’s response to the Further Consultation Ofcom should be more cautious of the risk of under- than over-recovery

C&W’s argument

7.32 C&W believe that Ofcom’s preferred option under the Further Consultation (i.e., Option (iv)) gives rise to a high risk of under-recovery of efficiently incurred costs.

7.33 C&W argue that, given the age and relative size of the FMC service, the risk of market distortion from over-recovery is very small. Conversely, C&W argue that the risk to their operations and for signals for future investors/market entrants from under-recovery is significant. Therefore, they argue that Ofcom should err on the side of ensuring that they give the best opportunity for investments to be recovered.

7.34 We have reconsidered the risk of under- or over-recovery in the light of C&W’s comments. We distinguish between (i) the risk of over or under recovery relative to C&W’s termination costs at the determined average termination rate (4.71ppm), and (ii) the risk of over or under recovery relative to this average arising from the way charges are allowed to vary by time of day. The Further Consultation was only concerned with the time of day issue, issue (ii) and this element alone is considered here.

Ofcom response

7.35 The traffic profile underpinning our proposed time of day (ToD) gradient, which was the subject of the Further Consultation, is our best proxy for the likely traffic profile on C&Ws FMC service. As such it is intended to balance the risk of over- and under-recovery relative to the benchmark rate of 4.71ppm. This risk arises because the traffic mix used to assess compliance differs from the actual traffic mix in the year in question. This risk is especially relevant here as, in advance of commercial operations, there is no actual traffic data for the FMC service on which to base the traffic profile.

7.36 We believe that the under-recovery which, under Option (iv), could result if C&W’s proposed traffic weights (for example) were correct, is unlikely to be material given the small volume of traffic expected in the first year of operation. Traffic is expected to grow after the first year but it will by then be possible to use weights derived from actual prior year traffic data and consequently it will not be necessary to use a proxy. Ofcom would expect that the parties would take account of actual traffic in determining the appropriate MTR for C&W’s FMC Service in future years.

7.37 Moreover, revenue from termination charges is only one source of income from customers for this service and the effect of different time of day weights on termination revenue is itself likely to be small. In any case, revenues in the first year are of much less significance to the return on investment than those in later years, when customer numbers are expected to be higher. We therefore believe that the impact of variations in first year time of day weights on C&W’s return on its investment in the FMC service is likely to be very small and that there is no unacceptable risk of creating a disincentive to investment through our choice of time of day profile.


81

7.38 However, we accept that the risk of over-recovery to an extent which could distort the market is similarly small. This is because we believe that a proxy solution in which Ofcom determines the weights used to calculate the weighted average termination charge will only be needed for the first year of operations. The volume of traffic terminated on C&W’s FMC service in the first year is likely to be very small, as noted above. In the light of this, we no longer believe that it is necessary to limit the extent of potential over-recovery arising from incorrect weights by setting charge ceilings for each time of day period, as well as traffic weights.

Ofcom’s proposal allows less flexibility than the MNOs against whom C&W are being benchmarked

C&W’s argument

7.39 C&W argues that it would be unfair for Ofcom to link C&W’s MTR to Vodafone’s charge control but allow C&W significantly less flexibility than Vodafone in setting ToD rates. C&W considers that this would therefore place it at a disadvantage compared to Vodafone despite the fact that the C&W service was only just being put into service.

.

7.40 We agree that, if possible, C&W should be given the same level of flexibility to set time of day charges as the MNOs. However, the options available in dispute resolution are inherently more limited than those for charge controls because the resulting determination applies only as between the parties. Additionally, the MNOs charge controls use prior year traffic volumes to establish compliance with the TACs, whereas this information is not available for C&W’s new FMC service.

Ofcom response

7.41 We have therefore had to consider whether it is possible to afford C&W the same level of flexibility that is available to the MNOs, at least for C&W’s first year of operations. In doing so we have taken account of the risk that setting the wrong weights would allow C&W to recover more than the determined average charge. In the further consultation we proposed to limit this risk by determining charge ceilings for each time of day period (Option iv). As noted in paragraph 7.38 above, in the light of responses we no longer believe this to be necessary however and so agree that C&W should be given as far as possible the same flexibility as the MNOs.

C&W’s preferred option

C&W’s argument

7.42 C&W believes that only Option (iii) provides C&W with close to the flexibility afforded to the MNOs. However, C&W argues that the ToD profile proposed by Ofcom is inappropriate.

.

7.43 Due to the uncertainty of the C&W FMC service’s traffic profile (and therefore the significant risk of inaccuracy of the determined weights), Option (iii) carries some risk of over-recovery, as C&W would have the ability to vary charges in their favour once their traffic profile becomes known to them, by increasing the charge for that time(s) of day (Day, Evening or Weekend) which the determined weights have

Ofcom response


82

underestimated. As noted above, however, we now believe that the extent of possible over-recovery is not sufficient to justify additional restrictions on C&W’s ability to vary charges by time of day, compared to the controls on the MNOs’ termination rates. Therefore, we now believe that Option (iii) is the most appropriate way of resolving this dispute.

C&W’s proposal for time of day weights

7.44 C&W submits that the traffic weights proposed by Ofcom are likely to overestimate the traffic that the FMC users will receive during daytime for the following reasons:

C&W’s argument

i) Business users often use their work phone in the evenings and weekends too; and

ii) C&W’s first customer, [ ], operates a 24/7 business.

7.45 Therefore, C&W argue that the FMC ToD profile is likely to be closer to the average MNO profile than the C&W fixed to mobiles profile.

7.46 The ToD split used under our preferred option in the Further Consultation (ie, Day – 65%, Evening – 20%, Weekend – 15 %) is , in our view, the best available proxy for the time of day traffic profile of calls terminating on C&W’s FMC service. Since C&W’s FMC service is expected to serve mainly business customers (who are likely to receive a relatively large proportion of their calls in business hours), it seems likely that C&W’s incoming call profile will have a higher proportion of day time traffic than that of the MNO average traffic profile. Ofcom notes in this regard that, in its response to the Consultation, C&W stated, as set out at paragraph 6.15 above, the time of day usage by customers of its FMC service would be concentrated during normal working hours, while MNOs usage is spread more evenly over all times of day. We further note that T-Mobile suggested to us that the daytime proportion for its corporate/business customers is significantly higher than the average (and the proportion under our preferred option). Therefore we disagree that the MNOs ToD profile would be an appropriate proxy for the FMC ToD profile.

Ofcom’s response

T-Mobile’s response to the Further Consultation

Determining a rate by time of day is outside the scope of this dispute

7.47 T-Mobile argues that there was no failure in negotiation between the parties as to a ToD split and therefore no dispute between the parties as to how the rate should be recovered. T-Mobile argues that by determining the detail of the commercial arrangements between the parties as to how the determined MTR is recovered Ofcom is extending the scope of the dispute and application of the final determination in a manner that is contrary to its statutory dispute resolution role and that may prejudice T-Mobile.

T-Mobile’s argument.


83

7.48 In referring this dispute to Ofcom, C&W submitted that T-Mobile had refused to accept the MTRs proposed by C&W, equal to the termination rate payable for the BT Fusion product. These rates were expressed as charges varying according to the ToD [ ], resulting in an average rate of 6.418ppm (based, according to the information provided by C&W, on comparison with current ToD profile for calls to mobile networks from the C&W’s network). T-Mobile proposed, instead, an average rate of 1.2ppm, without specifying any ToD profile. Given the inherent interrelation between the ToD charges and the average rate, and considering that T-Mobile did not accept the ToD charges proposed by C&W, it follows that the parties’ disagreement concerned the average rate as well as the ToD charges.

Ofcom response

7.49 The scope of this dispute is not necessarily confined to the determination of an average charge or a fixed charge to be applied across all time periods, since the termination rate payable by T-Mobile for voice calls originating on T-Mobile’s network and terminating on C&W’s network in respect of C&W customers using the FMC service may be expressed in various ways, including by means of ToD charges. Therefore, given that we consider it appropriate to afford C&W some degree of flexibility in varying its charges by ToD, we remain of the view that our proposed outcome falls within the scope of this dispute, as defined on 16 December 2008.

Exceptional circumstances

7.50 T-Mobile argues that Ofcom failed to establish a basis for exceptional circumstances that would meet the threshold for delaying a final determination in accordance with section 188(5) of the 2003 Act.

T-Mobile’s argument

7.51 As explained in the Further Consultation, we considered that C&W’s request to be given some flexibility in varying its charges by ToD may have some merit. However, since the Consultation did not propose different termination rates for different times of the day, we noted that stakeholders had not had the opportunity to comment on the appropriateness of ToD charging, as well as the appropriate means for its implementation (if considered to be appropriate). Therefore, we did not consider it appropriate to proceed with a resolution of this dispute until such time as the parties have been afforded the opportunity to comment on Ofcom’s amended proposals, which meant that we were exceptionally required to resolve the dispute in more than 4 months because of the need for further consultation.

Ofcom response

T-Mobile’s preferred option

7.52 T-Mobile considers that, should Ofcom decide to amend the draft determination in its final determination, Option (ii) – (i.e., determining an annual average weighted charge across all time periods on the basis of actual traffic) – is the most appropriate option. T-Mobile interpreted Option (ii) to mean that the average charge would be



84

calculated using current year weights for the first year of operation of the FMC service and prior year weights for subsequent years.

7.53 T-Mobile argues that Option (ii) ensures C&W has the same flexibility as its competitors and allows the terminating network to ensure cost reflective pricing and promote efficiency. T-Mobile also states that balancing payments, which are required at the end of the first year under its proposal, occur regularly between interconnect partners and are not problematic.

7.54 We agree that the most appropriate manner in which to allow C&W to vary termination charges by ToD is to use prior year traffic volumes as weights, because this is the approach for the benchmark rate we have used (Vodafone and O2’s regulated rate). We therefore agree that T-Mobile’s proposal is likely to be appropriate for year two onwards, once prior year traffic weights for C&W’s FMC service can be calculated. However, as this is not possible for C&W’s first year of operations, we should identify an alternative approach to afford C&W the possibility of using ToD charges, until such time as alternative charges are in place between the parties.

Ofcom Response

7.55 We believe that T-Mobile’s preferred option, which involves using weights which will remain uncertain until the end of the 12 month period, would be significantly more burdensome on C&W than either Option (iii) or Option (iv). This is because it would require C&W to set charges before the traffic volumes which would be used to assess compliance with the maximum annual average termination rate would be known. C&W would therefore have to set charges on the basis of forecast volumes which, if wrong, could result in the average charge being either above or below the permitted maximum.

7.56 This means that it is likely to be difficult for C&W to achieve compliance without a number of price changes towards the end of the year, or may require some form of balancing payment to T-Mobile in the event that the average charge is above the permitted maximum (with, in T-Mobile’s view, interest at a rate sufficient to penalise inaccurate forecasting). Such an outcome is therefore also likely to be less practicable to implement for C&W. C&W have stated that if Option (ii) was implemented it would choose to adopt a pricing strategy identical in nature to that required by Option (i). As set out above, Ofcom considers that Option (i) would not be appropriate in this case since it would be more restrictive for C&W than the restrictions on ToD pricing by the charge control which is used as a benchmark.

7.57 As the MNOs’ controls use prior year weights with no need for a forecasting or repayment mechanism, we consider this option would unfairly discriminate against C&W relative to the MNOs. Indeed, one of the reasons for using prior year traffic weights in the price controls on the MNOs is to avoid the need for such forecasting or repayment mechanism. Additionally, we consider the implementation of this option to be disproportionate, particularly as traffic terminated by C&W is likely to be relatively low in its first year of operations, whereas from the first year onwards the parties will have the possibility of using the information about the traffic volume of the preceding year in order to adjust the relevant ToD charges accordingly.

T-Mobile does not support the preferred Option in the further consultation



85

7.58 T-Mobile argues that Option (iv) is less consistent with Ofcom’s duties than Option (ii) because it allows less flexibility and a higher risk of over-recovery.

7.59 T-Mobile states that ‘it is simply not possible to create a close benchmark for a service that has not yet been put in effect’. Therefore, according to T-Mobile, as Options (iii) and (iv) would be based on incorrect assumptions, they are unsuitable methods to determine this dispute.

7.60 T-Mobile states that, despite Ofcom’s expressed desire in the Further Consultation to find a solution that is robust to changes in business strategy, Option (iv) is based on an estimation of C&W’s ToD split associated with its current business strategy.

7.61 As set out above, we consider that Option (ii) would place an unfair burden on C&W relative to the current arrangements of the MNOs. It would be less practicable than the other options and more burdensome on C&W. Therefore, we remain of the view that Option (ii) is inconsistent with our statutory duties.

Ofcom response

7.62 We also consider that, given the low expected traffic volumes in the first year of operation of C&W’s FMC service, any over-recovery under either Option (iii) or Option (iv) is unlikely to be material.

7.63 The ToD traffic profile we propose to use is intended only to apply for the first year of operation of C&W’s FMC service. We would ordinarily expect the parties to put in place alternative ToD charges based on the reasoning set out in this determination on the basis of the actual traffic received by the FMC users in the preceding year (i.e. a similar approach as for the price controls on the MNOs) once this information will be available. Therefore, we consider that our determination of this dispute is robust to changes in C&W’s business strategy.

The specific call profile proposed by Ofcom is inappropriate

7.64 T-Mobile believes Ofcom’s proposed incoming call profile to be inappropriate for the following reasons:


• T-Mobile’s service and the FMC service are mobile services, but the ToD split on

which Ofcom’s proposed weights are based do not include calls from mobiles.

• T-Mobile’s own corporate/business customers display an approximate 75% (Day):15% (evening):10% (weekend) split (according to a high level review). Therefore the proposed split might allow C&W to over-recover.

7.65 In the absence of a traffic profile for mobile to mobile calls by business customers, we think our proposed profile is the most appropriate of the options set out in the further consultation. This is because, unlike the others, it relates to calls to mobiles made by business customers. In our view it is therefore likely to be a more

Ofcom response


86

reasonable proxy for calls to C&W’s FMC service, which is in essence a national mobile service, than alternatives based on calls to fixed lines, even those which include calls from mobiles.

7.66 In general, it seems reasonable to assume that fixed line business users will make a higher proportion of their calls during the day than mobile users because calls from business fixed lines are inherently more likely to be made during office hours. If so, the fact that the benchmark used does not include calls from mobiles does not mean that it underestimates the proportion of calls likely to be made to C&W’s FMC service during the day. We do not therefore agree that the proposed incoming call profile is inappropriate.

7.67 We have considered T-Mobile’s suggested traffic profile, as well as the suggestion from C&W. However T-Mobile itself has acknowledged that its suggested proportion of 75% derives from a “very high level review” only (and in any case, there could be material differences between the traffic profiles for T-Mobile’s corporate/business customers and the customers of the FMC service). Therefore, we believe that we should not rely on it in this determination. It is not clear to us that the proportion of day traffic for the C&W FMC service will be as high as the proportion of day traffic received by T-Mobile’s own corporate/business customers, since all other available proxies suggest lower proportions

7.68 Ofcom has reconsidered the options for implementation which were set out in the further consultation in the light of responses. Our intention is that the method for calculating C&W’s average termination charge should as far as possible resemble the controls on the MNOs, for reasons of competitive neutrality and consistency with the chosen benchmark rate. Compliance with the controls on the MNOs’ termination charges is assessed using prior year traffic weights to calculate a weighted average charge. Charges may vary by time of day provided the weighted average is less than the maximum allowed.

Conclusions on the options in the light of the C&W and T-Mobile responses

7.69 Option (i), setting a flat rate to apply at all times of day, would give C&W significantly less flexibility than the MNOs. We did not favour this option in the further consultation. It received no support from respondents. We believe it would not be a fair and reasonable way to resolve this dispute.

7.70 Option (ii), determining a weighted average charge on the basis of current year traffic volumes, would be disproportionate and unduly burdensome on C&W. This is because current year volumes would not be known at the time C&W sets charges, requiring it to do so on the basis of a forecast of traffic, with a mechanism for reimbursement of any overpayment resulting if the forecast is inaccurate. The MNO’s controls are based on prior year weights not current year weights. We do not favour Option (ii) therefore.

7.71 Option (iii), of all the options, most closely resembles the controls on the MNOs’ termination charges. It is therefore most consistent with competition on the merits. This option requires us to determine the weights used to calculate C&W’s average termination charge for the first year. If we do not set the weights correctly, C&W may be able to recover more than the determined average charge. However, because of the low volume of the traffic in the first year, we do not believe that any over-recovery would be material. We note that the MNOs may also be able to exploit differences between prior year traffic weights and current year traffic profiles to their advantage.


87

7.72 We therefore consider that, in adopting Option (iii), we comply with our duties not to discriminate and to promote competition (as set out in sections 3(1)(b), 3(4)(b) and 4(3) of the 2003 Act). We therefore consider that this option is fair and reasonable as between the parties and is therefore appropriate for the resolution of this dispute.

7.73 Option (iv) places additional restrictions on C&W compared to Option (iii) in order to limit the potential for over-recovery which might arise if the weights used to calculate the average charge are incorrectly set. In the light of responses, we no longer believe that these restrictions are necessary. This option would not be fair and reasonable, therefore, as it would give C&W less flexibility to vary charges by time of day than the MNOs. Ofcom has therefore decided not to adopt this Option despite it having been presented as Ofcom’s preferred Option in the Further Consultation.

7.74 We consider the weights put forward as Ofcom’s preferred option in the Further Consultation to be fair and reasonable in the light of the arguments of the two parties. We note that C&W intends to serve business customers and anticipates that the traffic profile will be concentrated during the day. Whilst recognising the limitation in any forecast of traffic profile for a service which is not yet operational, Ofcom is of the view that the profile put forward in this regard is likely to be reflective of C&W traffic, taking into account that the weights on which we have based our determination fall within the range determined by (i) the weights proposed by C&W (ie, the MNOs’ weights) and (ii) the proportion of traffic received by T-Mobile’s own corporate/business customers based on its very high level review and are consistent with C&W’s view that its traffic is likely to be concentrated during the day to a greater extent than then MNOs.

Conclusions on traffic weights in the light of the C&W and T-Mobile responses

7.75 On this basis, and given the limited information available, we consider that the ToD traffic profile of C&W customers’ outgoing calls to mobiles proposed in the Further Consultation in connection with Option (iv) (see paragraph 1.16 (b) of the Further Consultation) remains the most suitable proxy for traffic received by the C&W FMC service and have therefore based our ToD rate calculations on this traffic split.

7.76 Those weights are:

• 65% Daytime Traffic;

• 20% Evening Traffic; and

• 15% Weekend Traffic.

7.77 Finally, we would ordinarily expect that the parties would take into account the basis for Ofcom’s calculation in determining future MTRs for the C&W FMC service. As set out above, Ofcom would ordinarily expect the parties to take into account C&W’s ToD weights in line with prior year traffic volumes once 12 months of data is available when determining rates once such data is available. Therefore, C&W’s concern is relevant to C&W’s first year of operations only.


88

Section 8

8 Summary of conclusions Summary of Ofcom’s response to T-Mobile’s revised proposal in respect of the Consultation

8.1 T-Mobile submitted that national roaming should be excluded from C&W’s MTR and that the MTR should be set equal to the efficient costs of termination on mobile networks in urban areas, as estimated by Ofcom’s MCT model. This would mean setting C&W’s MTR to 2.9ppm (in 2006/07 prices).

8.2 Alternatively, T-Mobile suggest that we should consider the costs associated to the actual traffic terminated via national roaming (by seeking from C&W periodical confirmation of the actual proportion of traffic terminated via roaming) and that the national MCT rate should be applied to actual traffic terminating via roaming only, whereas the efficient costs of termination on mobile networks in urban areas, as estimated by Ofcom’s MCT model, should be applied to the traffic terminated on C&W’s own network.

8.3 We do not consider that determining the dispute by setting C&W’s MTR to 2.9ppm (in 2006/07 prices) would be fair and reasonable as between the parties because this rate is significantly lower that the rate other mobile operators with national coverage, including T-Mobile, receive for termination.

8.4 Additionally, we do not believe that the efficient costs of termination on mobile networks in urban areas, as estimated by the MCT model, is a meaningful or accurate estimate of C&W’s efficient costs:

• C&W’s FMC service provides callers with a mobile termination service with national coverage; therefore it is inappropriate to use an estimate of efficient costs which is disaggregated by geo-type;

• We do not know in which geo-types C&W will roll out its physical

infrastructure in the future. Therefore, we do not know which of the numerous MCT geo-type estimates would be the most appropriate;

• The MCT model was not designed to calculate with precision the cost of

termination in specific geo-type areas and is not as reliable when it is used this way (see paragraph 6.81).

• It is not clear how an efficient geo-type-specific operator would operate. The

implicit assumption of the MCT model (i.e. that the costs of a local operator are the same as the relevant localised costs of an MNO) is not necessarily appropriate to C&W’s business structure (see paragraph 6.81).

8.5 In relation to T-Mobile’s alternative suggestion, we do not think it is appropriate

to vary C&W’s MTR according to the actual number of calls terminated via national roaming. The cost to MNOs of terminating individual calls varies by a number of factors but the current regulatory regime does not differentiate the MTR according to location or any other factor. We believe it would be impractical and discriminatory to differentiate C&W’s MTR by the location of the


89

call recipient (according to whether calls are terminated via roaming or on C&W’s own network).

8.6 Currently all mobile operators, including T-Mobile, receive MTRs in some areas (or for serving some customer types) which are higher than the efficient cost of termination for those areas/customers. Setting geographically/customer type disaggregated rates for one mobile operator without setting them for its competitors would be discriminatory, would risk deterring efficient entry and could distort competition.

8.7 These risks would be present because in the current regulatory regime (which is beyond the scope of this dispute) the termination rates of MNOs, against whom C&W might compete, are averaged across all customer types and areas. This might enable them to profitably offer retail prices that C&W would be unable to match, not because of inferior performance but because of the disparity in termination rates.

Summary of Ofcom’s response to C&W’s proposal in respect of the Consultation

8.8 In its response to the draft determination, C&W argued that their efficient costs

of termination are higher that the regulated rate for Vodafone and 02 in 2009/10, but lower than the rate set for H3G for that year. On this basis, C&W proposed that we determine an average termination charge of about 5.2ppm or 5.3ppm in finalising this dispute.

8.9 We have not accepted this proposal because we do not consider that determining the dispute in this way would be fair and reasonable as between the parties for the following main reasons:

(i) We consider that C&W’s efficient costs of termination for its FMC service are unlikely to be higher than the average efficient costs of a national operator. Therefore, we consider that a rate of 5.2ppm for C&W is likely to be significantly above C&W’s efficient costs. This rate is also above our range of estimates of C&W’s incurred costs;

(ii) To the extent that retail prices for calls to C&W reflect the underlying MTR, a

rate of 5.2ppm would cause the prices paid by callers to be higher than the price payable for calling other mobile networks, whereas from the customer perspective calls to the FMC users are similar to calls to other mobile networks;

(iii) This rate is above the regulated MTRs of the incumbent MNOs, as determined

by the CC and implemented by Ofcom. As such, it could allow inefficient entry and it risks distorting competition.

Summary of the rationale for our final determination to benchmark C&W’s average MTR against the lowest regulated MTR

8.10 We believe that the most appropriate manner in which to determine this dispute

is to benchmark the average MTR charged by C&W for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W customers using its FMC service equal to the lowest regulated MTR (i.e. 4.71ppm in 2009/10 prices).


90

8.11 We consider that it is fair and reasonable as between the parties for the following main reasons:

(i) It furthers the interests of consumers through the promotion of competition and the availability of a wide range of electronic communications services by allowing C&W to enter the market to compete in the mobile voice call market;

(ii) It allows C&W to retain and pass onto its customers any relative benefit

resulting from its choice to adopt cheaper technology and to differentiate its service;

(iii) The termination rate would not cause callers to C&W to pay more than they

would for calling other mobile services (with the precise outcome for callers depending on the way in which originating operators sets retail prices);

(iv) We consider that this rate should allow C&W to earn a profit on termination

relative to its efficient costs. C&W’s efficient costs of termination are unlikely to be higher than this rate and the measure of cost we are using already includes a reasonable return on investment. We acknowledge that there is a risk that C&W’s efficient costs might be lower than this rate, but we are not certain about the size of any such gap, given the limitations of the available evidence, and we consider that this is nonetheless unlikely to create distortions of competition;

(v) Since this approach provides C&W with the efficient termination rate for a

service with national coverage, it is robust to changes in C&W’s business strategy and to subsequent network rollout;

(vi) The approach does not change T-Mobile’s termination rate, or the feature that it

is currently averaged across all locations and customer types (and so likely to be above the average cost of providing termination for some customers and below the average cost for other customers). Therefore, T-Mobile’s termination revenues will continue to be sufficient to fund their termination costs (even if there were a differential waterbed effect between some customer types);

(vii) We do not believe that there is likely to be a resulting detrimental distortion to

competition against T-Mobile (or other MNOs) from benchmarking C&W’s average MTR against the lowest regulated MTR, even to the extent that it might allow C&W a margin above its efficient cost of termination. In particular, this is because T-Mobile and other MNOs, against whom C&W might compete, would also receive the same (or a higher) rate for terminating calls to business customers. We believe that the risk of detriment from distorting competition under our preferred option is smaller than under the other options considered.

Ofcom’s conclusion on the preferred option of benchmarking C&W’s average MTR against the lowest regulated MTR

8.12 Having considered stakeholders’ comments on the Consultation, we conclude

that the preferred option is to determine that the average MTR charged by C&W for the termination of calls originated on the T-Mobile network and terminated on the C&W’s network in respect of C&W customers using its FMC service may not exceed 4.71ppm (in nominal terms), this being the Target Average Charge (TAC) for Vodafone and O2, which is the lowest regulated MTR.


91

Variation of charges by time of day66

8.13 C&W suggested that we should determine this dispute by setting C&W’s MTR either:

C&W’s proposal

(1) According to Option (iii) in the Further Consultation (i.e., determining the average weighted charge and the ToD traffic weights and allowing C&W to set and change their ToD pricing structure as its sees fit within those constraints), but using the average weights of the MNOs’ ToD splits rather than the weights as proposed by Ofcom;

or

(2) According to Option (iv) of the Further Consultation (i.e., determining specific maximum ToD call charges based on the average charge, the underlying traffic profile assumptions and the following of day pricing structure: Day - 1, Evening - 0.75, Weekend - 0.67), but using a TAC equal to 5.2ppm or 5.3ppm rather than the 4.71ppm proposed by Ofcom.

8.14 T-Mobile argued that determination of how C&W should vary its charges by ToD falls outside of the scope of this dispute.

T-Mobile’s proposal

8.15 T-Mobile proposed that, if Ofcom is to determine how C&W is to vary charges by ToD, we should implement Option (ii) (i.e., determining an annual average weighted charge across all time periods on the basis of actual traffic).

8.16 We have not accepted either of C&W’s proposed options. We believe that under C&W’s Option (1) there would be a significant risk of C&W recovering above the TAC of 4.71 because C&W’s FMC service is expected to serve mainly business customers (who are likely to receive a great proportion of their calls in business hours than the average mobile user) and therefore it seems likely that C&W’s incoming call profile will have a higher proportion of day time traffic than that of the MNO average traffic profile.

Ofcom’s conclusion

Summary of our response to C&W’s proposal on ToD

8.17 We believe that C&W’s proposed Option (2) is inappropriate as it targets an average charge above that which we consider the most suitable TAC for resolving this matter (see paragraphs 1.5 and 1.6 above).

Summary of our response to T-Mobile’s proposal on ToD

8.18 The incumbent MNOs are able to vary their charges by ToD, provided that the weighted average charge reflects their headline TAC. This allows them flexibility and the efficiency benefits associated with tariff gradients. As we have

66 See http://www.ofcom.org.uk/consult/condocs/cw_tm_mobileterm/dispute1.pdf

http://www.ofcom.org.uk/consult/condocs/cw_tm_mobileterm/dispute1.pdf�


92

determined that it is appropriate to benchmark the C&W’s MTR against the regulated TAC for Vodafone and O2, we consider it appropriate to allow C&W a similar opportunity to realise these efficiency benefits, in line with the flexibility afforded to the MNOs.

8.19 T-Mobile’s preferred option, Option (ii) under the Further Consultation, involves using weights which will remain uncertain until the end of the 12 month period and would therefore be complicated for C&W to implement, since C&W could have to vary their charges throughout the year if their forecasts are not accurate. This option might also require a balancing payment at the end of the 12 month period. Therefore, we believe this option would be unduly burdensome on C&W. Additionally, we consider that the implementation of this option to be disproportionate, particularly as traffic terminated by C&W is likely to be relatively low in its first year of operations.

Summary of the rationale for our final determination ToD split

8.20 We believe that the most appropriate manner in which to allow C&W to vary termination charges by ToD is to determine the average weighted charge and the time of day traffic weights, and allow C&W to set and change their time of day pricing structure as it sees fit, within those constraints. This most closely resembles the controls on the MNOs’ termination charges and is therefore most consistent with competition on the merits.

8.21 This option requires us to determine the weights used to calculate C&W’s average termination charge for the first year. If we do not set the weights correctly, C&W may be able to recover more than the determined average charge. However, because of the low volume of the traffic in the first year, we do not believe that any over-recovery would be material. We note that the MNOs may also be able to exploit differences between prior year traffic weights and current year traffic profiles to their advantage. We therefore consider this option is fair and reasonable as between the parties.

8.22 As set out at paragraphs 7.75 to 7.77 above, Ofcom considers that the best estimate available for the traffic weights of C&W’s FMC service are 65% Daytime Traffic, 20% Evening Traffic and 15% Weekend Traffic. In considering this proposal, Ofcom has taken into account representations made by the parties as to the likely profile of a service targeted at business customers.

Ofcom’s determination

8.23 Ofcom’s powers in order to resolve disputes are set out under section 190 of

the 2003 Act and include a power to make a declaration setting out the rights and obligations of the parties to the dispute and to direct the parties to enter into a transaction between themselves on such terms and conditions as Ofcom may fix.

8.24 Ofcom’s powers to resolve disputes are limited to imposing obligations on the parties to the disputes only. In light of the above reasoning, Ofcom considers it appropriate in the context of this dispute to make a declaration setting out that for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using its FMC service, C&W is entitled to charge an amount not exceeding 4.71ppm and shall set the relevant


93

time of day charges according to the following traffic weights: (i) 65% for Daytime Traffic; (ii) 20% for Evening Traffic; and (iii) 15% for Weekend Traffic.

8.25 Given that the parties have an agreement to an interim rate subject to retrospective adjustment and the likely small levels of traffic on this service to date, we consider that the determination should apply from the date of the final determination in this matter.

8.26 Ofcom’s determination is at Section 9 below and takes effect on 20 May 2009.


94

Section 9

9 The Determination Determination under sections 188 and 190 of the Communications Act 2003 (“the 2003 Act”) for resolving a dispute between Cable & Wireless UK (“C&W”) and T-Mobile (UK) Ltd (“T-Mobile”) concerning the termination rate payable by T-Mobile for calls originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using its FMC service.

WHEREAS

(A) section 188(2) of the 2003 Act provides that, where Ofcom has decided pursuant to section 186(2) of the 2003 Act that it is appropriate for it to handle a dispute, Ofcom must consider the dispute and make a determination for resolving it. The determination that Ofcom makes for resolving the dispute must be notified to the parties in accordance with section 188(7) of the 2003 Act, together with a full statement of the reasons on which the determination is based, and publish so much of its determination as (having regard, in particular, to the need to preserve commercial confidentiality) they consider appropriate to publish for bringing it to the attention of the members of the public, including to the extent that Ofcom considers pursuant to section 393(2)(a) of the 2003 Act that any such disclosure is made for the purpose of facilitating the carrying out by Ofcom of any of its functions;

(B) section 190 of the 2003 Act sets out the scope of Ofcom’s powers in resolving a dispute which may, in accordance with section 190(2) of the 2003 Act, include:

a) making a declaration setting out the rights and obligations of the parties to the dispute;

b) giving a direction fixing the terms or conditions of transactions between the parties to the dispute;

c) giving 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 to the dispute to the other, giving a direction, enforceable by the party to whom sums are to be paid, requiring the payment of sums by way of adjustment of an underpayment or overpayment;

(C) on 24 November 2008 C&W submitted a dispute with T-Mobile to Ofcom for resolution;

(D) on 16 December 2008 Ofcom decided that it was appropriate for it to handle the dispute, and informed the parties of this decision;

(E) on 16 December 2008 Ofcom published details of the dispute on its website and invited comments from stakeholders on the scope of the dispute;

(F) on 16 December 2008 Ofcom set the scope of the dispute to be resolved as to determine the termination rate payable by T-Mobile for voice calls originating on T-Mobile’s


95

network and terminated on C&W’s network in respect of C&W customers using the C&W FMC service;

(G) a non-confidential draft determination was sent to the parties on 20 March 2009 January 2009 and published on Ofcom’s website on 20 March 2009;

(H) on 15 April 2009, Ofcom issued a further consultation on the appropriateness of ToD charging and the appropriate means for its implementation (if appropriate). On the same date, in light of the need for further consultation in this matter, Ofcom declared that exceptional circumstances had arisen in accordance with section 188(5) of the 2003 Act;

(I) in order to resolve this dispute, Ofcom has considered (among other things) the information provided by the parties and Ofcom has further acted in accordance with its general duties set out in section 3 of, and the six Community requirements set out in section 4 of the 2003 Act;

(J) a fuller explanation of the background to the dispute and Ofcom’s reasons for making this determination are set out in the explanatory statement accompanying this determination; and

NOW, therefore, Ofcom makes, for the reasons set out in the accompanying explanatory statement, this determination for resolving this dispute:

Declaration of rights and obligations, etc

1 As from 20 May 2009 and until such time as alternative charges are in place between the parties, C&W is not entitled to charge for the termination of calls originated on T-Mobile’s network and terminated on C&W’s network in relation to C&W’s customers using the C&W FMC service a weighted average charge in excess of 4.71ppm and shall set the relevant time of day charges according to the following traffic weights:

• 65% for Daytime Traffic,

• 20% for Evening Traffic and

• 15% for Weekend Traffic.

Binding nature and effective date 2 This determination is binding on C&W and T-Mobile in accordance with section

190(8) of the 2003 Act; 3 This determination takes effect on the date of the final determination; Interpretation

4 For the purpose of interpreting this Determination—

a) headings and titles shall be disregarded; and

b) the Interpretation Act 1978 shall apply as if this Determination were an Act of Parliament.


96

5 In this Determination:

a) ‘2003 Act’ means the Communications Act 2003 (c.21);

b) ‘C&W’ means Cable & Wireless UK whose registered company number is 1541957, and any of its subsidiaries or holding companies, or any subsidiary of such holding companies, all as defined by section 736 of the Companies Act 1985, as amended by the Companies Act 1989;

c) ‘the C&W FMC service’ means the fixed to mobile convergence communications service provided by C&W;

d) ‘Daytime Traffic’ means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using the C&W FMC service during the period of time between 08.00 and 18.00 on Monday to Friday;

e) ‘Evening Traffic’ means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using the C&W FMC service which is not either Daytime Traffic or Weekend Traffic;

f) ‘T-Mobile’ means T-Mobile (UK) Ltd whose registered company number is 02382161, and any of its subsidiaries or holding companies, or any subsidiary of such holding companies, all as defined by section 736 of the Companies Act 1985, as amended by the Companies Act 1989;

g) ‘Ofcom’ means the Office of Communications; and

h) ‘Weekend Traffic’ means call traffic originated on T-Mobile’s network and terminated on C&W’s network in respect of C&W’s customers using the C&W FMC service during the period of time between 24.00 on Friday and 24.00 on Sunday.

Neil Buckley

Director of Investigations

A person duly authorised in accordance with paragraph 18 of the Schedule to the Office of Communications Act 2003

20 May 2009


97

Annex 1

1 Cost Annex Assessment of C&W’s costs

A1.1 To gain a better understanding of the costs C&W incurs when a call to its FMC service is terminated, we completed a high level assessment of termination costs based on C&W’s business plan. We also assessed the evidence of C&W’s costs submitted by T-Mobile. In addition to this, we compared the outcome of our analysis with the efficient unit cost of termination by a national operator as estimated by the MCT cost model.

A1.2 This annex describes the following analysis of C&W’s costs in relation to the termination of a call to its FMC service:

a) An estimate of the cost of termination which combines:

• the call volume estimates and mobile business plan cost projections included in C&W’s business plan;

• a proxy for the cost of existing fixed line infrastructure and operating costs that will be used for the FMC service, but have not been included in the mobile business plan; and

• the cost to C&W when a call is terminated via its roaming agreement.

b) Evaluation of the cost evidence submitted by T-Mobile.

c) A comparison of the estimates under a) and b) with the estimated unit termination cost of an efficient national operator.

Estimated cost of termination on C&W’s Network

A1.3 The first eight rows of table A1 contain estimates of FMC termination costs on C&W’s own network. The last row contains estimates of the full cost to C&W of the termination of a call to its FMC service (i.e. the weighted average of the cost of termination on C&W’s network and the cost of termination via C&W’s roaming agreement).


98

Table A1: Estimated cost of termination (ppm)67

Notes Year 1 Year 2 Year 3 Year 4 Year 5 Constant rate1

Opex estimate [ ] [ ] [ ] [ ] [ ] [ ]

Depreciation estimate 2 [ ] [ ] [ ] [ ] [ ] [ ]

Cost of capital estimate 3 [ ] [ ] [ ] [ ] [ ] [ ]

Fixed network proxy 4 [ ] [ ] [ ] [ ] [ ] [ ]

Network costs estimate 5 [ ] [ ] [ ] [ ] [ ] [ ]

Non-network cost estimate 6 [ ] [ ] [ ] [ ] [ ] [ ]

Spectrum cost estimate 7 [ ] [ ] [ ] [ ] [ ] [ ]

Cost estimate – C&W network only [ ] [ ] [ ] [ ] [ ] [ ]

Cost estimate – including roaming 8 [ ] [ ] [ ] [ ] [ ] [ ]

1. In using the MCT cost model to assess the costs of termination for the regulated charges of the 5 incumbent MNOs, Ofcom used economic depreciation68

4. This price of 'Inter-Tandem Conveyance (Medium distance) on BT's carrier price list.

, which provides a more stable termination cost per call minute over the lifetime of a network. This technique is difficult to apply to C&W’s business plan, which covers a shorter period of time than the lifetime of the network. Therefore, for simplicity, to provide a rough proxy for the unit cost implied by C&W’s business plan under the economic depreciation methodology, we derived the constant ppm rate over the five-year period of C&W’s business plan (which, given the forecast volumes, is projected to recover the same amount of cost over the 5 years as the straight-line depreciation figures in NPV terms). 2. The CTM methodology provides for a variety of asset lives for over 60 assets. C&W’s balance sheet is not disaggregated to this level. However, it is likely that the majority of the C&W’s mobile business asset types would fit into categories that the CTM methodology would depreciate over at least 5 years. Therefore, this figure has been calculated assuming 5 year asset lives. 3. This figure has been calculated using the CTM allowance of 11.5% for cost of capital.

69

6. The figure in C&W’s FMC business plan includes the incremental administration costs of providing the mobile component of the FMC service only. Therefore, because that figure is an underestimation of the actual non-network costs of providing the FMC service, we have used the unit cost allowance included in the CTM methodology as a proxy for the total non-network component of the unit cost of termination for C&W’s FMC service here.

5. 25% of network costs have been allocated to voice termination.

70

67 Cost and call volume estimates are based on C&W’s business plan. All amounts are expressed in 2007/2008 prices. 68 The timing of cost recovery under economic depreciation generally varies from that under accounting depreciation: typically, accounting methods take the actual price paid for equipment (or its replacement cost) and divides this by the expected equipment life to reach a depreciation charge for the year. Economic depreciation seeks to smooth the path of an assets cost recovery over time by linking it to the use or extraction of value from that asset.

7. This figure is very close to zero due to the low average price of spectrum sold at the 2006 auction. The allocation for the 2G spectrum cost in the incumbent 2G/3G MNO’s costs of 2G termination is 0.16ppm. 8. C&W informed us that they have projected that [ ]% of their calls will be terminated via their roaming agreement. Therefore, the overall cost estimate has used the centre point of this assumption and attributed roaming costs to 50% of the projected calls and C&W’s own mobile business cost estimates (plus a proxy for the existing fixed component of their network as well as a proxy for administration costs) to the other 50% of projected calls.

69 See http://www.btwholesale.com/pages/downloads/service_and_support/pricing_information/carrier_price_list_browsable/C1.rtf 70 See 9.64 at http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf

http://www.btwholesale.com/pages/downloads/service_and_support/pricing_information/carrier_price_list_browsable/C1.rtf�




99

A1.4 Care is required with the interpretation of these figures. In calculating the above cost estimates, we used a number of assumptions based on Ofcom’s view of the termination costs of MNOs. It is unclear whether these assumptions are appropriate in estimating the cost of termination on C&W’s network.

A1.5 Additionally, because C&W were only able to provide us with the incremental costs of the mobile section of their business, the cost information we were provided does not include any allowance for the cost of conveying calls over C&W’s fixed network for termination. Therefore, we sought to estimate this cost by using BT's published wholesale charges for conveyance over its network as a proxy. As the inter-tandem conveyance market is effectively competitive, we believe these charges to be a reasonable proxy for the cost of conveyance over an efficient network.

A1.6 In order to identify which of BT’s wholesale charges would be an appropriate proxy, we discussed with C&W the way in which calls to FMC customers are routed over its network. C&W told us that calls to its FMC customers from customers of other networks will be handed over to C&W on a "near-end handover" basis and will in all cases be routed via C&W's MSC in Birmingham. From there calls will be routed onto the customer IP-VPN and then terminated on a mobile handset. C&W also told us that, because of the limited number of its switches which are connected directly to its FMC network, the "vast majority" of calls will be routed over more than two switches in its fixed core network.

A1.7 The relevant portion of the call is from the point of handover to C&W (near to the point of origination) to the MSC in Birmingham. The BT wholesale service which seems to us to be the closest approximation to this is inter-tandem conveyance. Due to the unavoidable absence of any information on the actual routing patterns of calls to C&W's FMC service, we have assumed that all calls use "inter-tandem conveyance medium”.

A1.8 C&W’s FMC business plan includes only the incremental administration costs of providing the mobile component of the FMC service. Relying on this figure would have led us to underestimate the actual non-network costs of providing the FMC service. Therefore, we have used the unit cost allowance included in the CTM methodology as a proxy for the non-network component of the unit cost of termination for C&W’s FMC service.

A1.9 Additionally, our estimates are sensitive to C&W’s projections of their future call volumes. 71

A1.10 We estimated the cost to C&W of terminating a call on its own network to be lower than the cost of terminating a call via its roaming agreement. Therefore, the unit cost estimates are sensitive to C&W’s assumption of the balance between FMC calls terminated on the C&W network and calls terminated on the network of C&W’s roaming partner. Figure A1 shows the effect of varying this assumption on the constant termination cost estimate.

Given that C&W’s FMC service is still in its testing phase, we cannot determine the accuracy of these projections.

71 For example, holding the assumed proportion of FMC calls terminated on the C&W network constant at 50%, if the projected number of calls terminating on C&W’s FMC service were halved the resulting constant estimate would be 4.50ppm (assuming a decrease in call volume will not decrease C&W’s own network costs, but will proportionately decrease roaming costs). If the projected number of calls were to be increased by 50% the resulting revised constant estimate would be 3.25 (assuming an increase in call volume will not increase C&W’s network costs, but will proportionately increase roaming costs).


100

A1.11 Table A1 above presents a constant rate termination cost estimate of [ ]. However, C&W told us that they expect between [ ] of their traffic to be terminated via their roaming agreement but used the central 50/50 split for the purpose of their business plan calculations. Therefore, it would be more accurate to consider that the cost of terminating a call to C&W’s FMC service is in the range between 2.6ppm and 4.1ppm (see figure A1).72

Figure A1: Sensitivity of result to roaming proportion estimates1

[ ]

1. This sensitivity analysis assumes that a decrease in the volume of calls terminated on C&W’s own network will not decrease its own network costs. If C&W’s own network costs vary with volume, the unit costs when the roaming proportion is between 0% and 50% would be underestimated, and the unit costs when roaming is between 50% and 100% would be overestimated. Evaluation of cost evidence submitted by T-Mobile

A1.12 T-Mobile included its own assessment of C&W’s likely costs in its submission. T-Mobile based this assessment on a comparison with the components of its own termination rate.

A1.13 T-Mobile referred to the following estimate of the likely cost of termination on a DECT guard band network, which they originally provided in their submission to the MCom/T-Mobile dispute.

Table A2: T-Mobile’s detailed termination rate assessment Termination rate breakdown (all in ppm) T-Mobile Guard band Network costs 3.08 0.5

72 We note that C&W will have an incentive to keep their roaming costs as low as possible, and that they could influence this though choice of customer as well as through their retail pricing strategy. Therefore, it is possible that the proportion of calls terminated on C&W’s FMC network via roaming [ ].


101

Spectrum costs 1.45 0.00011455 Network externality 0.3 0 Admin costs 0.3 0.3 5.13 0.80011455 50% uplift: 1.200171825 T-Mobile offer

1.2

A1.14 C&W’s network rollout will cover only its clients’ premises. Coverage on the FMC service will be provided through its national roaming provider’s network in all other areas. Therefore, because it has not invested in a national network on the scale of the established UK GSM network operators, T-Mobile argue that C&W’s network costs would be significantly lower than its own.

Network costs

A1.15 T-Mobile’s submission outlines that the relevant network costs, including site costs, equipment costs, backhaul, buildings and RAN planning, would all be significantly lower for C&W than that of full power 2G/3G networks, even accounting for scale/coverage.

A1.16 T-Mobile estimates that the equipment costs of a local DECT guard band network would be 0.25ppm but their submission does not disaggregate this estimate between the components of network costs. T-Mobile adds 0.25ppm for the BT single tandem charge (their estimate of backhaul cost) to get a final network cost estimate of 0.5.

A1.17 Our analysis of C&W’s business plan indicates that its network costs are likely to be significantly higher than the 0.5ppm estimated by T-Mobile (see table A1). However, our analysis does support T-Mobile’s assertion that C&W’s network costs are likely to be lower than T-Mobile’s (and the other MNOs).

A1.18 T-Mobile points out that C&W’s GSM licence for DECT guard band spectrum was acquired at relatively little cost and does not currently attract administered incentive pricing (i.e. a methodology to set licence fees based on opportunity costs). Accordingly, T-Mobile included a very small allowance for spectrum in its estimate of the cost of termination on a DECT guard band network.

Spectrum licence costs

A1.19 We note that, even if C&W’s entire spectrum cost (i.e. the average price paid for a DECT guard band licence in the 2006 auction) was allocated to termination in the first 5 years of C&W’s operations, the resulting unit cost would be less than 0.001ppm. Therefore we use an allocation of 0.0ppm for spectrum costs in our estimation of C&W’s termination cost.

A1.20 In its analysis of 3G spectrum costs, the CC Determination stated that in a competitive market one would not expect the sum of network costs and spectrum costs to be different for services that are essentially homogeneous73

73 See 2.3.3 at

. It is possible that the significantly lower cost of DECT guard band spectrum, to some extent,

http://www.catribunal.org.uk/files/CC_Determination_1083_H3G_1085_BT_220109.pdf



102

balances the higher cost to C&W of providing national coverage by relying on negotiating a national roaming agreement.

A1.21 T-Mobile states that it is not aware of any reason that an administrative cost comparable to T-Mobile’s would be inappropriate in the case of C&W. Therefore T-Mobile’s methodology allocates an administration cost equal to its own (0.3ppm) to its estimate of the likely cost of termination on a DECT guard band network.

Administrative costs

A1.22 Our analysis of C&W’s administrative costs results in a significantly lower figure, 0.01ppm. However, this is based only on the incremental administration costs included in C&W’s mobile business plan. If C&W’s existing administration arrangements will be used to support the FMC service, then this figure would be an underestimation of the true administrative costs of termination on C&W’s FMC service if costs are based on a fully allocated cost basis.

A1.23 Therefore, in our estimation of C&W’s termination cost, we used the unit cost allowance included in the CTM methodology as a proxy for the total non-network component of the unit cost of termination for C&W’s FMC service.

A1.24 T-Mobile state in their submission that C&W are: “aiming to provide a replacement service using the existing mobile subscribers of alternative national operators, using a SIM to be inserted in a subscriber’s existing handset.” Therefore, T-Mobile argue that there would be no network externality when C&W increase their subscription base, and thus does not include a network externality surcharge in its estimation of C&W’s termination costs.

Network externality surcharge (NES)

A1.25 The CC Determination recommends the removal of the NES from MNO termination rates; we have not included a NES in our estimation of the cost of termination on C&W’s network.

A1.26 The sum of all inputs as estimated by T-Mobile is 0.8001ppm. T-Mobile multiplied this estimation by 1.5 ‘to counter any inaccuracies in T-Mobile’s estimation’, leading to a final estimate of 1.2ppm.

T-Mobile’s final estimate

A1.27 T-Mobile argue that C&W’s termination costs are likely to be even lower than this figure because it has:

• nil site costs: C&W installs its base stations in the premises of its clients;

• nil backhaul costs: C&W uses the existing landline connection of its clients and

• high network usage: C&W only rolls out sites in the premises of clients that it has already acquired: there is no speculative network development and network rollout and capacity can be tailored specifically to traffic.

A1.28 We agree that these factors are likely to facilitate low termination costs on C&W’s network. Our estimate of the cost of termination on C&W’s own network, 2.25ppm (see table A1), is significantly lower than the cost of termination on the networks of T-Mobile and the other MNOs, but higher than T-Mobile’s estimate.


103

A1.29 Significantly, these estimations are based only on the direct costs of running C&W’s own network only and therefore exclude the costs to C&W of providing termination outside the reach of its own network.

A1.30 The costs to C&W of providing full national coverage, which it does through a national roaming agreement, are significantly higher than the cost of termination on its own localised network. T-Mobile excludes these costs from its estimation and therefore significantly underestimates the average pence per minute cost of termination on the FMC service.

National Roaming

A1.31 C&W has a national roaming agreement, under which for any call terminated outside of its own network’s coverage C&W is charged [ ]. C&W estimates that between [ ] of calls to the FMC service will be terminated on its network. Therefore, the inclusion of the roaming component of termination costs significantly increases the estimated average ppm cost of termination on C&W’s FMC service.

A1.32 Without its roaming facility, C&W customers would only be able to make and receive calls when in the limited area of C&W’s own network coverage. Therefore, the roaming facility forms an unavoidable component of the cost to C&W of providing a national mobile service. See paragraphs 5.54 - 5.61 for discussion of the relevance of roaming costs in the setting of the FMC termination rate.

The efficient cost of a national operator

A1.33 Although C&W’s own network infrastructure alone does not provide national coverage, FMC customers are provided national coverage through C&W’s roaming agreement. As the end users (both callers and call recipients) receive the same benefits regardless of whether national coverage is provided by an operators own network or through national roaming, we believe that the efficient cost of providing national coverage is the relevant cost consideration in this case.

A1.34 Therefore, to inform our analysis of the likely efficient costs of termination on a national network, we used the model of mobile termination costs which Ofcom developed to set charge controls for the five incumbent MNOs (i.e., the MCT cost model).74

A1.35 The MCT cost model’s estimate — based on Ofcom’s medium traffic scenario and excluding 3G costs and call volumes — suggests that the average network unit costs of an efficient national operator are 3.7ppm.

A1.36 These figures refer to network costs only. An additional allowance should also be made for non-network costs in order to be consistent with the MCT charge control. We used a mark-up for non-network costs of 0.3ppm for 2G/3G operators in the charge control. Therefore, the estimated efficient total termination cost for 2G/3G MNOs is 4.0ppm (in 2006/07 prices).75

74 The MCT cost model uses 2006/07 as its base year, therefore all numbers in this section are expressed in 2006/07 amounts and need to be inflated to 2007/08 numbers for comparison with the results from C&W’s business plan.

75 The MCT cost model estimates the network costs of termination for an efficient MNO, and thus assumes access to identical technology as well as the economies of scale and scope obtained by MNOs. It is not clear to what extent these parameters or the general cost allocation assumptions, such as the mix between incoming and outgoing calls as well as the mix between data and voice calls


104

Comparison of cost estimates76

A1.37 As explained in paragraph A1.11, our best estimate of the cost of termination for C&W’s FMC service is a range between 2.6ppm and 4.1ppm:

• The efficient cost of termination on a national network as estimated by Ofcom’s MCT charge control methodology, 4.2ppm (4.0ppm in 2006/07 prices), is higher than this range. (Our estimate corresponds to the MCT estimate at the point in the range where [ ] of calls to C&W’s FMC service are terminated via roaming).

• The Vodafone / O2 MTR termination rate as established by the recent CC Determination, 4.58ppm (4.4ppm in 2006/07 prices) is also above this range of estimates (our estimate corresponds to this benchmark at the point where [ ] of calls to C&W’s FMC service are terminated via roaming).

• C&W’s proposed termination rate, 6.418ppm (across each time period), is significantly higher than all estimates in this range (our estimate would be lower than C&W’s proposed rate even if all calls to C&W’s FMC service were projected to be terminated via roaming). C&W’s proposed MTR is significantly higher than the efficient cost of termination as estimated by the MCT cost model and is higher than all the regulated termination rates contained in the CC Determination for 2009/10 and 2010/11.

• Our range of estimates is significantly higher than the cost estimate provided by T-Mobile (even if all calls to C&W’s FMC service were terminated on C&W’s own network the resulting estimate, [ ], would be higher than T-Mobile’s proposed rate). T-Mobile’s estimate is also significantly lower than the efficient cost of termination as estimated by the MCT cost model and all the regulated termination rates contained in the CC Determination.

which were designed to approximate the use of MNO’s networks, apply to C&W’s service. Additionally, the efficient non-network termination costs for a small specialised service may not be equal to those for an established MNO. 76 The numbers in this section are expressed in 2007/08 prices unless otherwise stated.


105

Annex 2

2 Revised estimate of the cost of termination for C&W’s FMC Service Assessment of C&W’s costs

A2.38 This annex contains the adjustments to our estimates of C&W’s incurred termination costs (see Table A2). These costs were discussed at paragraphs 6.3-6.13 above.

A2.39 These revised estimates are different from the unit cost estimates included in our draft determination. For example, our constant unit cost estimate including roaming costs – implementing the assumption of a 50/50 split between calls terminated via roaming and calls terminated on C&W’s own network – has increased from 3.37 to 4.14.

A2.40 The difference between this estimate and the estimate included in the Consultation arises from77

• a change to the depreciation rate assumption from the original draft. We have adopted an average five year asset life for our estimation – to take account of the types of assets, namely pico cells, employed in C&W’s service – which is a departure from the 10 year asset life assumption we used in the draft determination; and

:

• our underestimation of the cost to C&W of calls terminated via roaming. In the draft determination, only the direct cost of roaming was allocated to calls terminated via roaming. However, C&W informed us that ‘in practice even those calls terminated via the roaming agreement drive significant cots in C&W’s own network as well as roaming charges’. Therefore, we have revised the unit cost estimates to include the fixed cost proxy and the non-network cost proxy as suggested by C&W;

• the addition of components of C&W’s capital expenditure, including base station and installation costs, which equates to £21,582,000 expenditure over the five years78

.

77 The revisions of our estimates of C&W incurred termination cost are a result of comments we received on our draft determination from C&W. Given the time limitations of the dispute resolution process, we did not seek to arrive at definite conclusions regarding the appropriateness of these cost revisions, but have included them to inform our analysis of a reasonably plausible range for C&W’s incurred termination costs. Additionally, we have not reached a conclusion about whether these costs have been efficiently incurred. We consider that this approach is appropriate given that the method with which we have determined this dispute does not rely on our estimate of C&W’s incurred costs. 78 As C&W anticipate that ‘some of this cost will be recovered through other charges to the customer such as installation charges and rental fees’, it is unclear what proportion of this expense should be attributed to the FMC service . Therefore, the inclusion of 100% of this expense in our calculations of the appropriate network costs on which to base our estimate of C&W’s incurred termination costs has likely resulted in a likely overestimation of C&W’s termination costs.


106

Table A2: Revised cost of termination estimate (ppm)79

Notes Year 1 Year 2 Year 3 Year 4 Year 5 Constant rate1

Opex estimate [ ] [ ] [ ] [ ] [ ] [ ]

Depreciation estimate 2 [ ] [ ] [ ] [ ] [ ] [ ]

Cost of capital estimate 3 [ ] [ ] [ ] [ ] [ ] [ ]

Fixed network proxy 4 [ ] [ ] [ ] [ ] [ ] [ ]

Network costs estimate 5 [ ] [ ] [ ] [ ] [ ] [ ]

Non-network cost estimate 6 [ ] [ ] [ ] [ ] [ ] [ ]

Spectrum cost estimate 7 [ ] [ ] [ ] [ ] [ ] [ ]

Cost estimate – C&W network only [ ] [ ] [ ] [ ] [ ] [ ]

Cost estimate – including roaming 8 [ ] [ ] [ ] [ ] [ ] [ ]

1. In using the MCT model to assess the costs of termination for the regulated charges of the 5 incumbent MNOs, Ofcom used economic depreciation80

4. This price of Inter-Tandem Conveyance (Medium distance) on BT's carrier price list.

, which provides a more stable termination cost per call minute over the lifetime of a network. This technique is difficult to apply to C&W’s business plan, which covers a shorter period of time than the lifetime of the network. Therefore, for simplicity, to provide a rough proxy for the unit cost implied by C&W’s business plan under the economic depreciation methodology, we derived the constant ppm rate over the five-year period of C&W’s business plan (which, given the forecast volumes, is projected to recover the same amount of cost over the 5 years as the straight-line depreciation figures in NPV terms). 2. We have adopted an average five year asset life for our estimation (this is a departure from the 10 year asset life assumption we used in the Consultation). 3. This figure has been calculated using an assumption that 25% of network costs should be allocated to voice termination.

81

6. The figure in C&W’s FMC business plan includes the incremental administration costs of providing the mobile component of the FMC service only. Therefore, because that figure is an underestimation of the actual non-network costs of providing the FMC service, we have used the unit cost allowance included in the CTM methodology as a proxy for the total non-network component of the unit cost of termination for C&W’s FMC service here.

5. 25% of network costs have been allocated to voice termination.

82

A2.41 Table A2 above presents a constant rate termination cost estimate of [ ]ppm. However, C&W told us that they expect between [ ] of their traffic to be terminated via their roaming agreement but used the central 50/50 split for the

7. This figure is very close to zero due to the low average price of spectrum sold at the 2006 auction. The allocation for the 2G spectrum cost in the incumbent 2G/3G MNO’s costs of 2G termination is 0.16ppm. 8. C&W informed us that they have projected that [ ] of their calls will be terminated via their roaming agreement. Therefore, the overall cost estimate has used the centre point of this assumption and attributed roaming costs to 50% of the projected calls and C&W’s own mobile business cost estimates (plus a proxy for the existing fixed component of their network as well as a proxy for administration costs) to the other 50% of projected calls.

79 Cost and call volume estimates are based on C&W’s business plan. All amounts are expressed in 2007/2008 prices. 80 The timing of cost recovery under economic depreciation generally varies from that under accounting depreciation: typically, accounting methods take the actual price paid for equipment (or its replacement cost) and divides this by the expected equipment life to reach a depreciation charge for the year. Economic depreciation seeks to smooth the path of an assets cost recovery over time by linking it to the use or extraction of value from that asset. 81 See http://www.btwholesale.com/pages/downloads/service_and_support/pricing_information/carrier_price_list_browsable/C1.rtf. 82 See 9.64 at http://www.ofcom.org.uk/consult/condocs/mobile_call_term/statement/statement.pdf.





107

purpose of their business plan calculations. Therefore, as in the draft determination, it would be more accurate to consider that the cost of terminating a call to C&W’s FMC service is in the range between 3.2ppm and 5.0ppm (see figure A2).83

A2.42 Additionally, our estimates are sensitive to C&W’s projections of their future call volumes.

84

Figure A2: Sensitivity of result to roaming proportion estimates1

Given that C&W’s FMC service is still in its testing phase, we cannot determine the accuracy of these projections.

[ ]

1. This sensitivity analysis assumes that a decrease in the volume of calls terminated on C&W’s own network will not decrease its own network costs. If C&W’s own network costs vary with volume, the unit costs when the roaming proportion is between 0% and 50% would be underestimated, and the unit costs when roaming is between 50% and 100% would be overestimated. Comparison of cost estimates85

A2.43 As explained in paragraph A2.4, our best estimate of the cost of termination of calls to C&W’s customers using its FMC service is a range between 3.2ppm and 5.0ppm.

• The efficient cost of termination on a national network as estimated by Ofcom’s MCT charge control methodology, 4.2ppm (4.0ppm in 2006/07 prices), falls within this range. Our estimate corresponds to the MCT estimate at the point in the range where [ ]% of calls to C&W’s FMC service are terminated via roaming.

83 We note that C&W will have an incentive to keep their roaming costs as low as possible, and that they could influence this through choice of customer as well as through their retail pricing strategy. Therefore, it is possible that the proportion of calls terminated on C&W’s FMC network via roaming [ ]. 84 For example, holding the assumed proportion of FMC calls terminated on the C&W network constant at 50%, if the projected number of calls terminating on C&W’s FMC service were halved, the resulting constant estimate would be 5.7ppm (assuming a decrease in call volume will not decrease C&W’s own network costs, but will proportionately decrease roaming costs). If the projected number of calls were to be increased by 50%, the resulting revised constant estimate would be 3.91 (assuming an increase in call volume will not increase C&W’s network costs, but will proportionately increase roaming costs). 85 The numbers in this section are expressed in 2007/08 prices unless otherwise stated.


108

• The Vodafone/O2 MTR as established by the recent CC Determination, 4.58ppm (4.4ppm in 2006/07 prices), also falls within this range of estimates. Our estimate corresponds to this benchmark at the point where [ ]% of calls to C&W’s FMC service are terminated via roaming.

• C&W’s revised proposed termination rate, 5.2ppm, is higher than all estimates in this range. Our estimate corresponds to C&W’s proposed rate where[ ]% of calls to C&W’s FMC service are terminated via roaming. C&W’s proposed MTR is significantly higher than the efficient cost of termination as estimated by the MCT cost model and is higher than the regulated MTRs for four of the five MNOs contained in the CC Determination for 2009/10 and for all five of the rates for 2010/11.

• Our range of estimates is significantly higher than the cost estimate provided by T-Mobile (1.2ppm). Even if all calls to C&W’s FMC service were terminated on C&W’s own network the resulting estimate, [ ]ppm, would be higher than T-Mobile’s proposed rate. T-Mobile’s estimate is also significantly lower than the efficient cost of termination as estimated by the MCT cost model and all the regulated termination rates contained in the CC Determination.

dispute between cable & wireless and t-mobile about mobile

Documents