England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE COMPETITION APPEAL TRIBUNAL
(MARCUS SMITH Q.C., CLIVE ELPHICK AND JONATHAN MAY)
[2012] CAT 1

		Royal Courts of Justice Strand, London, WC2A 2LL
		30th October 2013

Sir Bernard Rix:

1. This appeal concerns the correct application of section 86(1)(b) of the Communications Act 2003 (the "2003 Act"). The appellant, TalkTalk Telecom Group Plc ("TalkTalk"), submits that the respondent, the Office of Communications or "Ofcom", has misapplied that subsection in arriving at its 20 July 2011 charge control determination, controlling the prices in a defined section of the wholesale broadband access ("WBA") markets. TalkTalk also submits that the Competition Appeal Tribunal (the "Tribunal"), to whom it appealed from Ofcom's determination, arrived at the wrong answer by reason of misstating Ofcom's market definition. It is acknowledged by Ofcom that the Tribunal did misstate its market definition, but Ofcom nevertheless submits that the Tribunal still managed to come to the right answer in dismissing TalkTalk's appeal.
2. TalkTalk is a leading provider of retail fixed line rental, voice calling and broadband telecommunications in the United Kingdom. Ofcom is the national regulatory authority, responsible for regulation of the communications industry in the UK. Retail broadband services are the services provided directly to members of the public. They are to be distinguished from wholesale broadband services, which are the means by which communications providers sell access to broadband to each other.
3. The principal provider of wholesale broadband services in the WBA markets is BT. It is by virtue of BT's significant market power in large sections of the country that Ofcom has considered it necessary to impose price controls on it in those areas where it is strongest. The effect, however, of such price controls is to make it more difficult for TalkTalk to recoup the cost of competing with BT in such areas. Hence its interest in this appeal. There is intervention in this appeal by another competitor, British Sky Broadcasting Limited ("Sky"), which adopts but does not materially add to Ofcom's submissions on the merits. It has however addressed issues as to possible remedies, in the event that TalkTalk's appeal were to succeed in principle. As for BT, it has also intervened below, but has not appeared on this appeal.
4. The 2003 Act provides for ex ante (i.e. forward looking) regulation of communications providers that are found to have significant market power ("SMP") in a given market such that competition is not effective. A three stage process is involved whereby Ofcom (i) identifies markets in which it is appropriate to consider whether to make a "market power determination", i.e. a determination that a person has SMP in a given market; (ii) analyses those markets for the purpose of deciding whether to make such a determination; and (iii) imposes regulatory obligations, if it thinks it right to do so, including price controls, on persons found to have SMP in a services market: see sections 45(1), 79(1) and 87(9) of the 2003 Act. Each stage of the process is carried out by Ofcom publishing a notification following prior consultation. Section 84 provided for Ofcom to review, at such intervals as it considers appropriate, its identifications of services markets, determinations of market power and setting of SMP conditions. These periodic re-examinations of particular markets are known as "market reviews". The SMP conditions may include price controls.
5. SMP Guidelines exist to which Ofcom must have regard when identifying and analysing markets (section 79(3)). Ofcom must consider how matters can be expected to develop over the period leading up to the next market review. Paragraph 20 of the Guidelines states:

"In carrying out the market analysis … NRAs [national regulatory authorities] will conduct a forward looking, structural evaluation of the relevant market, based on existing market conditions. NRAs should determine whether the market is prospectively competitive, and thus whether any lack of effective competition is durable, by taking into account expected or foreseeable market developments over the course of a reasonable period. The actual period should reflect the specific characteristics of the market and the expected timing of the next review of the relevant market by the NRA."

6. Market identification and definition involve analysis of relevant services and a market's geographic scope. The higher the market share a person holds in a given market, the greater his market power will be regarded as being. Where SMP is found then, as stated above, price controls may be imposed.
7. In a typical case where SMP is found in an identified market and Ofcom considers that price controls ought to be imposed, the market power determination and the charge control determination will be contained in a single document, published and notified at one and the same time. However, the 2003 Act contemplates that a market power determination may precede a charge control notification: but in such a case Ofcom must ensure, before going on to impose its regulatory controls, that there has in the meantime been no material change in the market previously defined. This situation is catered for in section 86 of the 2003 Act, which contains the provisions with which this appeal is particularly concerned.
8. The importance of market identification, analysis and review as part of a market power determination is highlighted in sections 79 and 84, which (ignoring immaterial amendments made as of 26 May 2011) provide inter alia as follows:

"79(1) Before making a market power determination, OFCOM must –

(a) Identify (by reference, in particular, to area and locality) the markets which in their opinion are the ones in which in the circumstances of the United Kingdom are the markets in relation to which it is appropriate to consider whether to make the determination …

(3) In considering whether to make or revise a market power determination in relation to a services market, OFCOM must take due account of all applicable guidelines…

84(1) This section applies where OFCOM have identified and analysed a services market for the purposes of making a market power determination.

(2) OFCOM must, at such intervals as they consider appropriate, carry out further analyses of the identified market for one or more of the following purposes

(a) reviewing market power determinations made on the basis of an earlier analysis…

(5) Before carrying out a further analysis under subsection (2), OFCOM may review any decision of theirs identifying the markets which it was appropriate to consider for the purpose of carrying out an earlier analysis.

(6) Where, on such a review, OFCOM conclude that the appropriate markets have changed –

(a) they must identify the markets they now consider to be the appropriate ones; and

(b) those markets shall be the identified markets for the purposes of the further analysis."

9. In particular, section 86 provides as follows:

"(1) OFCOM must not set an SMP services condition by a notification which does not also make the market power determination by reference to which the condition is set unless –

(a) the condition is set by reference to a market power determination which has been reviewed under section 84 and, in consequence of that review, is confirmed in the notification setting the condition; or

(b) the condition is set by reference to a market power determination made in relation to a market in which OFCOM are satisfied there has been no material change since the determination was made…

(6) A change is a material change for the purposes of subsection (1) … if it is one that is material to –

(a) the setting of the condition in question…"

10. In the present case, Ofcom published its market power determination on 3 December 2010, in which it looked forward to the imposition of charge controls but wished to consult further on them. In the event its charge control determination was not published for nearly another eight months, on 20 July 2011. In the event, it was required (at least) to satisfy itself that there had been no material change for the purposes of section 86. It is not clear that when it published its market power determination without at that time imposing its price controls it realised or remembered that it would be necessary to pass over the section 86 hurdle en route to its charge control determination. Certainly there is no reference to section 86 in its market power determination.
11. In its July 2011 charge control determination Ofcom expressed itself satisfied that there had been no material change for the purposes of section 86. That was challenged by TalkTalk on appeal to the Tribunal, and is challenged again on this appeal. The Tribunal only avoided finding in favour of TalkTalk's appeal by redefining for itself the relevant market in which the price controls were set. That redefinition was not contended for by Ofcom, and on this appeal is repudiated by it. The Tribunal said that, if the relevant market were to be defined as it is agreed by both TalkTalk and Ofcom it should be defined, then "there would be a material change for the purposes of section 86(1)(b)" (at para 113 of the Tribunal's judgment). Ofcom nevertheless submits that that obiter conclusion of the Tribunal is wrong, and is seen to be so when consideration is given to the SMP that BT would continue to exercise in the relevant market pending further review (originally) four years after its initial review (i.e. by December 2014)^[1]. In the course of applying its own market definition, the Tribunal went at least some way towards acknowledging Ofcom's analysis, but, apparently, only on the hypothesis that the correct market definition is not that which Ofcom determined. TalkTalk submits that, properly understood, the Tribunal's judgment is in its favour, and that it is not permissible, for the purposes of section 86, to evade Ofcom's own definition of relevant markets.

Ofcom's market power determination of December 2010

12. In its market power determination Ofcom explains how the WBA markets work. In essence, WBA services are provided primarily by BT, with its vast and historic telephone network, to other companies, who use them to provide retail broadband services to consumers and businesses in the UK. To compete at the retail level, a new company would therefore have the options of either (i) taking all its services from BT (and thus simply being a competing marketer of such services); or (ii) building its own physical network, running all the way to the premises of its customers, as cable companies have done in some parts of the country; or (iii) using what is known as "local loop unbundling" (or "LLU"), whereby the competitor installs equipment at a local telephone exchange and then rents the copper lines connecting that exchange with its customers in that area. Ofcom concentrated on such unbundling as being the best means by which competition between networks in the provision of broadband services can be created. "Where this competition develops regulation of wholesale broadband is unnecessary" (at para 2.8).
13. In many telephone exchange areas, however, for instance where the consumer population is too sparsely spread, LLU is unlikely to be successful, with the result that direct competition between broadband networks is discouraged. In such areas "regulation at the WBA level is necessary to ensure that consumers can choose between differing retail offers" (at para 2.9). It follows that, in essence, each telephone exchange is its own universe, save that many such exchanges are sufficiently homogeneous, competitively speaking, to be amalgamated in their own, quasi-geographic, but not contiguous, market. Ofcom therefore investigated every telephone exchange in the country, 5603 of them in total, in order to allocate each of them to what it concluded should be four separate markets. One of these was centred on Hull, where for historical (but for present purposes irrelevant) reasons 14 exchanges are operated by a small network which is not BT's. We are not concerned with the Hull market, which serves only 0.7% of UK premises, but with the other three, named market 1, market 2 and market 3 respectively.
14. Thus, in a critical paragraph of its market power determination (para 1.19) Ofcom defined these three markets as follows:

"We conclude that there are four separate geographic markets, as follows⁵:

[Hull]

? Market 1: exchanges where only BT is present or forecast to be present (11.7 per cent of premises)⁶;

? Market 2: exchanges where two principal Operators (POs)⁷ are present or forecast and exchanges where three POs are present or forecast but where BT's share is greater than or equal to 50 percent (10.0 per cent of premises) and

? Market 3: exchanges where four or more POs are present or forecast and exchanges where three POs are present or forecast but where BT's share is less than 50 per cent (77.6 per cent of premises)."

15. The footnotes to paragraph 1.19 are important. They are as follows.

i) Footnote 5: "Our analysis of the sizes of each market is based on the latest information we gathered, which was in June 2010".

ii) Footnote 6: "In assessing forecasted plans we have only counted operators as present where they have firm plans to deploy in specific exchanges".

iii) Footnote 7: "We define a Principal Operator within section 3 of this statement as an operator capable of providing a material constraint in the market."

16. The consequence of these market definitions for the individual markets 1, 2 and 3 were spelled out in Appendices 1, 2 and 3 to the determination: where the "names" of the exchanges were listed one by one: 3,389 exchanges in market 1, 660 exchanges in market 2, and 1,540 exchanges in market 3. In terms of numbers of exchanges, market 1 is the most numerous: but in terms of the provision of WBA services on a nationwide basis, it appears that the market 3 exchanges are the most important, because Ofcom comments (para 1.7) that it found effective competition in almost 80% of the UK, but insufficient competition in markets 1 and 2, making up just over 20% of the UK.
17. Ofcom found that BT held a position of SMP in markets 1 and 2, but not in 3.
18. Ofcom went on, under the heading of remedies, to explain its thinking as follows. As to market 1, it said "there is limited prospect in the near term of any wholesale competition" (para 1.22). Therefore, it said it was imposing general access, non-discrimination and transparency obligations on BT, together with a requirement that charges should be based on the cost of provision. As to charges, it said that it had decided to impose a charge control "the details of which will be subject to separate consultation", to ensure that BT did not set excessive prices which would ultimately be passed on to consumers (para 1.23). As to market 2, it said that there would also be general access, non-discrimination and transparency obligations, but no charge control. It explained (at para 1.25):

"there may be potential for BT to raise its prices to an excessive level. However, there is some wholesale competition in Market 2 and the potential for this to develop further, though the extent of any such further investment is uncertain. We consider that an approach to regulation that promotes investment where it is economic in order to provide effective and sustainable competition is appropriate in Market 2."

19. So far, what Ofcom wrote was in section 1 of its determination, headed "Summary". At para 1.16, within this opening section 1, Ofcom referred to TalkTalk's announcement of further unbundling plans:

"On 16 November 2010 Talk Talk announced its intention to unbundle 700 further exchanges. We understand these plans are still in the process of development. We have considered the implication of this announcement on this market review."

20. Ofcom revisited its market definition in detail in section 3. In doing so, it explained its attitude to POs' expansion plans. Thus it made a distinction between committed plans in specific, identified, exchanges, and uncommitted plans of a more general, unspecific, nature. It refers to this distinction in a number of places, as follows:

"3.75 With regard to coverage and network expansion plans by the POs we received information on both committed plans (up to December 2010) and further uncommitted plans … We have used both of these to form a view on the potential for further investment during the period covered by the review … However, we have decided to only rely on committed plans in the exercise of counting the number of POs in an exchange for the purpose of the market definition."

Both here and in other citations below I have italicised wording (the italics are not in the original) which deploys the distinction between committed and uncommitted plans.

21. The same distinction is made in an important passage at paras 3.169-3.190, which, although lengthy, needs to be cited in extenso. There Ofcom discusses "developments since the second consultation", and focusses particularly on TalkTalk's latest information concerning its unbundling plans. Ofcom wrote:

"3.169 Just a few days before our statement was due to be published, on 16 November 2010, Talk Talk stated its intention to extend its LLU footprint…

3.170 Talk Talk has not currently committed to deployment in any specific exchanges. Rather it is in the process of assessing the feasibility of deployment in a number of exchanges. In its public statements it presented this deployment as a medium term plan. Further, it indicated that the implementation period for the full rollout could be around three years.

3.171 Consistent with our approach to uncommitted plans from other POs, we do not consider that it would be appropriate for us to attempt to select which exchanges Talk Talk may unbundle in the future, or the order they may unbundle them, as part of our geographic market definition exercise. This could lead us to assign exchanges based on an assessment that turns out to be incorrect.

3.172 In our view there are two possible approaches for us at this late stage in our market review. First we could delay publication of the statement until Talk Talk is in a position to provide firm plans for which exchanges it plans to unbundle and rough timescales for the completion of this. Alternatively, we could conclude the review if we consider that the conclusions remain apposite, taking account of Talk Talk's plan.

3.173 It is our understanding that it is unlikely that Talk Talk would be able to provide firm information on the exchanges it will unbundle for several months … In waiting until then, we consider it could be appropriate to also then gather updated data from other POs on their rollout plans. The information gathered through this process from Talk Talk and the other POs may suggest that further analysis and consultation is required …

3.176 Given the timeframes for assessing the feasibility of deployment and for the deployment itself, it is unlikely in our opinion that Talk Talk will be in a position to exert a practical restraint in any new exchanges for a period of six to nine months …

3.177 Our approach to market definition is that we should not count a PO as present in an exchange until the PO has confirmed specific rollout plans. As Talk Talk has not yet identified the exchanges it plans to unbundle, our current approach results in exchanges staying in the market to which they have been allocated.

3.178 In carrying out a market review we are required to take a forward look at how competitive conditions may change over the period of the review. Whilst accepting that deployment in Market 1 exchanges will have an impact, we are also mindful that, based on the timescales above, a significant portion of the market review period will be characterised by BT being the only provider even in exchanges that Talk Talk chooses to unbundle. As such, our regulatory approach needs to balance the potential for further competition towards the end of the review period with BT's position of being a monopoly provider for the earlier part of the review period …

3.179 If the outcome of waiting until February/March for Talk Talk to identify the specific exchanges it plans to unbundle was that we simply moved exchanges into the relevant market (that is, we did not change our approach to market definition, SMP or remedies), the main effect on those exchanges that move from Market 1 to Market 2 would be that they would no longer be included within the charge control we have decided to impose in Market 1. They would still be subject to all the other remedies such as cost orientation and non-discrimination that we impose on Market 2.

3.180 Alternatively, we have also considered whether the imposition of a charge control in exchanges where BT is the only PO but where future entry will occur is still appropriate. We think this is a useful exercise in the particular circumstances facing us because we need to assess whether the uncertainty of delaying the conclusion of the review is justified, or whether an immediate conclusion results in a regulatory outcome that remains appropriate even in the face of updated information available to us …

3.182 It is clear that at the start of the period covered by the review BT's position in Market 1 exchanges where Talk Talk subsequently deploys would be the same as that for all other exchanges in Market 1. BT would be the only provider and would, as such, face no competitive constraints. Based on the potential for migration of customers from BT wholesale products onto Talk Talk's own network, and considering the effect when a second PO is present in other exchange areas, we are of the view that even if Talk Talk deploys towards the start of the review period BT's market share would be likely to be at least 70 to 80 per cent in the exchanges where Talk Talk deploys at the end of the review period. The information available from Talk Talk indicates that deployment would be over the period of the review and so the effect on BT's share would be less than this in many of the exchanges. Where BT's share is at this level and it faces competition from only one other provider, a charge control may still be considered to be an appropriate remedy.

3.183 It also needs to be remembered that market definition is not an end in itself but rather is a means to setting market boundaries within which SMP and the need for certain remedies can be assessed. In carrying out a geographic market analysis where exchanges are grouped, it is inevitable that a range of exchanges with slightly different competitive conditions may be grouped together … But it could be argued that exchanges where two POs are forecast to be present (but only one is currently present) are also similar to exchanges where only one PO is present, so that they should be included within Market 1. We have attempted to address this by only including firm forecasts of PO rollout in our assessment. This effectively reduces the period when only one PO is present and increases the period when two POs are present and BT is subject to the constraint of the second PO. In the case of the exchanges that Talk Talk aims to unbundle, it is not clear these could be treated in this way, since the time when BT is the only PO would be significant when compared to the overall period of the forward look."

22. Ofcom proceeded to argue that similar considerations would arise concerning the possible redeployment of exchanges from market 2 to market 3, which again could not be done without the identification of specific exchanges, and could again lead to undesirable delay, re-consultation and reconsideration of remedies. Then, under the heading of "Conclusion", i.e. conclusion on the issue raised by "developments since the second consultation", Ofcom stated:

"3.187 We accept that Talk Talk's announcement indicates that the scope for LLU deployment in Market 1 is greater than we had previously considered.

3.188 The aim of geographic market definition is to assess the markets in which market analysis can be undertaken. The grouping is based on assessing the extent of heterogeneity between different exchanges. We have considered the implications of delaying our conclusions until information is available that allows us to analyse the specific list of exchanges that Talk Talk is to unbundle. This would lead to a period of uncertainty which would be open ended since we could not commit to the conclusions we would draw based on the information provided, or whether we would need to re-consult.

3.189 We also consider that for the period covered by this review the exchanges where Talk Talk intends to deploy will be likely to have similar conditions to the markets in which they are currently allocated. This is because of the rollout timescales which mean that Talk Talk are unlikely to have any exchanges unbundled within the next nine months, and rollout will be ongoing throughout the forward look period of this review. Therefore it is our view that exchanges allocated to Market 1 where Talk Talk subsequently deploys can, for the purposes of the market analysis exercise, be considered to have competitive conditions that are sufficiently similar to exchanges in Market 1 where Talk Talk does not deploy. In the next review, when Talk Talk's deployment has been confirmed, the effect of this deployment can be taken into account. A similar argument holds for the exchanges in Market 2. In the specific situation of this market review we do not think it is inappropriate that exchanges where BT is expected to have a very high market share and will face only a single competitor entering at some point during the market review period should be subject to a charge control. By comparison, exchanges in Market 2 where there are two POs present or forecast to be present are more likely to be subject to the constraint of the second PO for the entire period of the review (since the forecast includes only confirmed plans expected to be implemented by December 2010).

3.190 Therefore, we conclude our market definition remains appropriate. As we have set out in section 2, the option exists to commence the next review before the end of the forward look period we have taken in the event that material changes in the market occur."

23. "Conclusion on market definition" is then set out at paras 3.193-194. The language used there defined market 1 as "exchanges where only BT is present" (see also at paras 3.5, 3.84 and 3.109, where the same language is used). Such language contrasts somewhat with para 1.19, cited above, where the definition is "exchanges where only BT is present or forecast to be present" together with footnote 6 stating that "we have only counted operators as present where they have firm plans to deploy in specific exchanges".
24. The conclusions I draw from these passages in section 3 of the determination are as follows. First, a plain and critical distinction is made between committed unbundling plans, where the exchanges concerned had been specified and identified, and the recent announcement (of 16 November 2010) of unspecified, unidentified, and uncommitted further unbundling. Secondly, although additional reasoning is provided for not redrawing the market definitions – reasoning such as the uncertainties of allowing delay to the market review to occur, or the uncertainties of when and how uncommitted plans would materialise, and even such as the estimated ultimate effect in terms of BT's continued dominance in market 1 exchanges at the end of a review period (at that time viewed as stretching forward for four years) even on the assumption that the plans would be put into immediate operation – the plain fact remains that uncommitted plans which did not identify specific exchanges as being the site of further unbundling could not be fitted within market definitions which required individual exchanges to be identified and named as belonging within one or other of the three markets (as Ofcom recognised at its para 3.171). Plainly, Ofcom was unwilling to jettison definitions which depended on such identifications for the uncertainties engendered by TalkTalk's late announcement. Fourthly, however, Ofcom also appears to have considered that it ought to take account of TalkTalk's announcement, inadequate as it was for the purposes of the proposed definitions, in what was required by statute to be a forward looking review: it could not simply ignore the announcement. This was no doubt an additional reason why Ofcom considered the possible competitive effect of further unbundling.
25. Fifthly, the fact remained that the driving force of Ofcom's reasoning was the absence of any committed plans for any competitor to BT in the market 1 exchanges. Thus market 1 exchanges remained exchanges where there was simply no competitor, either in being or committed to being created. It followed that in the case of market 1 there simply was no constraint at all, for the present at any rate and for the foreseeable future, to BT's monopoly. Market 1 contained those exchanges where, as Ofcom summarised the position in its para 1.22, "there is limited prospect in the near term of any wholesale competition" (I emphasise the "any"). Or, as Ofcom put it in para 1.26, the distinction between market 1 and market 2 was that in market 2 Ofcom could assess "the likely constraint arising from current and future investment by other operators". In market 1, however, there simply was no such constraint. The mere possibility of uncommitted plans in unidentified exchanges did not constitute either presence of a competitor or even the forecast of such presence. Again, as footnote 6 at para 1.19 stated, "we have only counted operators as present where they have firm plans to deploy in specific exchanges". It may be noted, moreover, that the projected 70 to 80% figure for BT's domination of market 1 even on the most optimistic assessment of the advancement of TalkTalk's then uncommitted plans played no part in Ofcom's market 1 definition (compare the 50% figure built into the definitions of markets 2 and 3).
26. My understanding of the long passage set out above from section 3 of the determination is supported by further passages in para 5.87 (where Ofcom discusses "Remedies") and in Annex 3 (where Ofcom gives further details of its "Geographic analysis"). Thus para 5.87 reads as follows:

"5.87 We disagree with BT's argument that a charge control is not justified in Market 1. Whilst we recognise the difficulty in assessing precise returns in each market based on the data reported by BT, our main reason for the imposition of the charge control is the lack of competition to BT in the wholesale market. Since any provider that is offering service in Market 1 is likely to be doing so based on BT's wholesale inputs, in the absence of pricing regulation it is our view that BT's pricing would be unconstrained. BT would be free to raise its wholesale prices and any such increases would be likely to be passed on to customers. The opportunity for other CPs to deploy competing infrastructure in Market 1 is, as discussed in section 4, unlikely to provide a sufficient constraint on BT's pricing, even though we accept that Talk Talk's intention to unbundle a further 700 exchanges will mean that at some point during the review period there will be a PO other than BT present in some exchanges in Market 1."

In other words, there is no current competitive presence in market 1 nor any such presence committed to be there in the future: and even though the unspecific, indefinite possibility of such presence arriving at some time within the next review period may have to be taken into account, there is insufficient prospect of that to avoid the need for control pricing now.

27. Similarly, in Annex 3 the following is found (emphasis again added):

"A3.6 We have used data at two points to inform our geographic market definition: actual data from June 2010 and forecast data based on confirmed rollout plans by the Principal Operators (POs). Whilst not all POs indicated firm dates for the completion of this rollout we have assumed this will be largely complete by December 2010. Therefore the data shown for December 2010 below corresponds to the firm rollout plans of each PO.

A.3.7 Some POs provided forecasts of rollout plans beyond this. Again, some of these plans were open ended. These plans are not confirmed. We have, therefore, not included them in our final geographic market definition. However, for completeness we have included them in this Annex to show the effect that they would have on the geographic market definition if they were implemented in full. We have indicated these within the data tables as "Uncommitted".

A3.8 On 16 November 2010 Talk Talk announced plans to unbundle a further 700 exchanges. However, Talk Talk is still in the process of assessing these plans and has not provided us with a list of these 700 exchanges. We have not updated the data below with the information provided to us by Talk Talk even though it may be considered at this stage to be similar to the "Uncommitted" plans previously provided. We do not believe that this has a material impact because we do not take account of these uncommitted plans in our assessment of the presence of POs by exchange."

28. Bearing all this material into account, I consider that it is reasonably plain that for the purpose of market definition the uncommitted rollout plans are not taken into account, even though, for the purpose of deciding that control pricing is needed for market 1, as consequently defined, account is taken of the possibility that uncommitted rollout plans, if they materialise, may ultimately have some, but insufficient, effect on competition in the market. It is a matter of line drawing: where the rollout is committed, the presence of a competitor in the market is already assumed; but where the rollout is uncommitted, it is not. Given the need to identify and define the market exchange by exchange, some such line-drawing was ultimately inevitable. However, it also reflects the real world. Where BT is and will in the future remain the sole operator in any given exchange, there is no competitive constraint. Where, however, an exchange has been identified as one in which a competitor has committed plans to roll out unbundling, there is such constraint.
29. Finally, under this heading, there is no reference by Ofcom to the point that, in a situation where control pricing is left to be imposed by a subsequent determination, there is in any event a statutory requirement to consider whether a further review is necessary because of some material change which has occurred.

The charge control determination of July 2011

30. In the course of the anticipated consultation for the purpose of its charge control determination, Ofcom received from TalkTalk further information about its previously unspecific proposals for a rollout programme for the unbundling of another 700 exchanges. TalkTalk made it clear to Ofcom that it could have, in confidence, full details of the specific, identified, exchanges in which it would undertake unbundling. Ofcom's attitude was that such information was immaterial and therefore not required. It is common ground on this appeal that TalkTalk should be regarded as having given notification to Ofcom of the specific, identified, exchanges in which it was now committed to roll out the unbundling of a further 700 exchanges. It was TalkTalk's submission to Ofcom and to the Tribunal, and remains so on this appeal, that this further information made it impossible for Ofcom to be satisfied that there was no material change within the meaning of section 86 of the 2003 Act. TalkTalk's unbundling programme has continued during the contentious issue which then ensued and has resulted in this litigation. TalkTalk's evidence, in the form of a witness statement from Mr Heaney, demonstrated that already as at 12 September 2011, only a few weeks after Ofcom's charge control determination, 153 further exchanges, including 92 market 1 exchanges had gone "live", and hundreds more were already in the pipeline for going live that year. However, the imposition of price controls in the very market in which TalkTalk had committed to roll out its unbundling programme was said to put a potential burden on the economics of TalkTalk's programme with the danger of making rollout in a number of exchanges unviable.
31. The consequence for Ofcom, if TalkTalk were to have been right in its section 86 submission as to material change, is that Ofcom could not have proceeded to its charge control determination without updating its December 2010 market power determination by conducting a further review under section 84 of the Act (see section 86(1)(a)). The subsection (1)(a) and (b) methodologies are alternatives, one of which must be fulfilled, to ensure that a charge control will not be imposed save on the basis of a materially up to date market power determination. As the Tribunal itself was to say (at para 88 of its judgment):

"The purpose of section 86(1)(b) is to enable OFCOM to make a Subsequent Charge Control Notification without also conducting a contemporaneous market power determination. Instead, the basis for the Subsequent Charge Control Notification is the Prior Market Power Determination. Plainly, it only makes sense to rely on the Prior Market Power Determination where the circumstances between the issue of the two decisions have not changed in any significant way. That is the rationale for the "no material change" requirement in section 86(1)(b)."

32. Ofcom's reasons for rejecting the submission of material change are given in its charge control determination. Ofcom expressed itself as follows (I have again italicised passages where the distinction between uncommitted and committed exchange unbundling is recognised):

"3.40 Under Section 86 of the Act, before Ofcom can set an SMP service condition by a notification, which is separate from the notification making the market power determination, Ofcom needs to be satisfied that there has been no material change since the market power determination was made. Ofcom therefore has a statutory discretion which involves making a judgment.

3.41 Having considered the evidence, we are satisfied that since the market power determination in the WBA Statement, there has been no material change in the market conditions for the following reasons:

3.42 First, we do not consider that a proposal to rollout LLU-based services in a number of exchanges constitutes a change in the actual competitive conditions of the market, even if the plan for some of these exchanges is now said to be "firm". In defining the geographic markets, it is often appropriate to take a limited forward look of the market and include in the assessment exchanges where operators have confirmed roll out plans. In contrast, for us to satisfy ourselves that a material change has occurred, the appropriate question is whether an event has actually occurred that has materially changed the competitive conditions in the market.

3.43 Second, and in the alternative, in the WBA statement we assessed TTG's planned rollout and concluded that it did not merit a change to our market definition (paragraphs 3.169 to 3.190), to our SMP assessment (paragraphs 4.36 to 4.41) or to our proposed remedies (including a charge control) (paragraph 4.91).

3.44 In summary, we examined TTG's potential rollout and concluded that exchanges allocated to Market 1 where TTG subsequently deploys can, for the purposes of the market analysis exercised, be considered to have competitive conditions that are sufficiently similar to exchanges in Market 1 where TTG does not deploy. We based this on the fact that at the start of the period covered by the review there would be no competitive constraint on BT and that any potential future entry by TTG would only introduce a constraint for part of the period covered by the review. We said that at the start of the period, BT would be the only provider and would, as such, face no competitive constraints. Based on the potential for migration of customers from BT wholesale products onto TTG's own network, and considering the effect when a second PO is present in other exchange areas, we are of the view that even if TTG deploys towards the start of the review period, BT's market share would be likely to be at least 70 to 80 per cent in these exchanges at the end of the review period. The information from TTG indicated that deployment would take place over the period of the review and so the effect on BT's share would be less than this in many of the exchanges. Where BT's share is at this level and it faces competition from only one provider, a charge control may still be considered an appropriate remedy.

3.45 TTG has not provided any materially new information since its initial announcement of the rollout plans, to change our assessment in the WBA statement. Although we accept that TTG's plans have developed, in so far as TTG has now identified the specific exchanges which it intends to rollout to Market 1 and has started placing orders for some of these, this does not alter our assessment.

3.46 Our analysis shows that BT's market share in Market 1 is likely to remain above 85 per cent throughout the period of this control. Moreover, our analysis shows that, in those specific exchanges in Market 1 where TTG plans to extend its LLU network, BT's market share is likely to remain above 70 per cent throughout the entire charge control period. We note that the SMP assessment is carried out at the level of the market as a whole, and therefore, BT's market share, the number of the operators in the market and our view of the potential for further entry are not affected by the identification of the specific exchanges.

3.47 It is clearly possible during the period of the market review that an operator will rollout to further exchanges (as TTG proposed to do). Ofcom must in exercising its judgement whether there has been a material change do so in a way that allows the market process to function effectively in the interest of promoting competition for consumers.

3.48 Our conclusions are set in light of the recent EU requirement for market reviews to be carried out every three years.^[2] We reviewed the market in December 2010 and in July 2011 we are setting the detailed control.

3.49 For the reasons set out above, we are satisfied that there has been no material change in the market since the publication of the WBA Statement and therefore that the section 86 test is met."

33. It is not easy to encapsulate the essential reasoning of this passage, but I will seek to do so in the following observations.
34. First, on behalf of Ofcom Mr Josh Holmes accepts that the reasoning in para 3.42 is fallacious and is not relied on. It is accepted that a new announcement of market related activity may constitute a material change for the purposes of section 86 in what must after all be a forward looking analysis. I would add that the test of section 86 "material change" is not the same as "whether an event has actually occurred" that has materially changed "the competitive conditions in the market". Other considerations apart, it is not possible to consider whether conditions in the market have changed until the market is properly defined: and it is that anterior question which is being considered for the purpose of section 86(1)(b) ("in relation to a market in which … there has been no material change"). If it be the case that the market, as defined in an earlier market power determination, has materially changed, then it is on the cards that competitive conditions in that altered market have changed, and also that, even if some charge control remains necessary, the appropriate terms or extent of such control may well be affected.
35. Secondly, it was not in dispute on this appeal that the effect of the definition of "material change" found in section 86(6) was that propounded by the Tribunal at para 96 of its judgment, viz –

"a material change exists where

(a) it would cause the Prior Market Power Determination to be different in a material respect (i.e. one that is more than de minimis); and

(b) that difference is capable of affecting the setting of the Subsequent Price Control Notification."

There had been some dispute between Ofcom and TalkTalk before the Tribunal as to the effect of section 86(6), but that has not been renewed on this appeal.

36. Thirdly, Mr Holmes placed reliance solely, therefore, on paras 3.43/44 as containing the essential Ofcom reasoning in support of meeting its section 86(1)(b) hurdle. He submitted that these paragraphs introduced a second ("Second"), alternative ("and in the alternative") reasoning, expounded in para 3.44 and reaching back into reasoning found already in Ofcom's market power determination. He described para 3.44 as the "key paragraph" which described Ofcom's case on this appeal: viz, that the exchanges committed by TalkTalk's further rollout announcement had been tested and the key finding made that even in such unbundled exchanges BT's market share would extend to 70%- 80% at the end of the review period, i.e. in three (or four) years' time. He described this as part of Ofcom's economic analysis which had not been challenged.
37. However, I have difficulty in understanding how the first and second of Ofcom's lines of reasoning actually differ. Both depend upon a conclusory view that nothing new has been notified or occurred that had not already been given full consideration in the earlier market power determination. Given the intricacy of Ofcom's reasoning both here and in its earlier determination I can understand how Ofcom managed to persuade itself of this conclusion. However, I cannot agree with it. As I have analysed the market power determination above, a critical distinction is made between committed and uncommitted rollout plans. As to the latter, Ofcom goes into further rationalisation to explain why even a forward looking analysis is justified in discounting such uncommitted plans when it comes to the definition of market 1 and the application of that definition to individual exchanges. Now, however, in para 3.44 of its later charge control determination, although justifying its position by reference to its previous assessment, when rollout plans were uncommitted, it appears to conclude that it makes no difference that rollout plans are now committed in identified, specific exchanges, on the basis that even if deployment occurred earlier rather than later, BT would still be left with predominant market power of a 70%-80% share in each unbundled exchange.
38. In my judgment, however, the effect of this reasoning is both to abandon its previous definition of market 1 as containing those exchanges where only BT was present or forecast to be present (on the basis of committed plans) and to substitute for that definition a new definition which is essentially to the effect that market 1 contains those exchanges where, whether or not BT is challenged by another operator's committed unbundling in a specified, identified, exchange, BT's market share is projected not to fall below 70% by the end of the current review period. Ofcom has moved from a market which is essentially defined by the presence in an individual exchange of a second operator, actual or committed, to a market which is essentially defined by a forwardly assessed market share, as of three (previously four) years away, in the range of 70%.
39. Fourthly, such a development is to my mind inconsistent with the definitions of not only market 1 but also markets 2 and 3 in the earlier market power determination, and inevitably marks a material change in that determination as well as in the effect of any charge control based on that determination. It is inconsistent with the earlier definitions of markets 1, 2 and 3 because: (a) market 1 was not originally defined in any way in terms of a level of market share in a competitive exchange, but rather in terms of exchanges where there was no competition at all (present or committed) and thus no constraint at all by a competitor; and (b) markets 2 and 3, on the other hand, were (in part) defined in terms of both competition by one or more competitors within the individual exchanges and BT's market share (at above or below 50%, a figure which obtains at the start of the review period rather than one which is assessed at its end). It is however inevitable that with a market in which there is only one competitor (market 1), any loss of market share over an ensuing period because of what might occur in the future pending a review some years away could only be a forward projection. BT's opening share in market 1 is always 100%; and there is no qualification in the market definition as to BT's market share progress thereafter. An exchange within market 2, however, may start off in a position where BT's current market share is 100% but there is commitment by another operator to unbundle in that exchange, or where there are two committed competitors in addition to BT but BT's share is greater than 50% (and could in theory for the present approach 100%).
40. Fifthly, it has to be remembered that Ofcom's market power determination defines market 1 (as well as markets 2 and 3) in what are in effect two ways. This is because at one and the same time each exchange is a separate competitive universe but all like exchanges can be treated in the same way. That is an application of the principle of equality (also the principle of rationality): that like things be treated alike. Therefore it is possible for the definition of market 1 to remain the same at two different points of time, but for the application of that principle to the separate universes of individual exchanges to change, as the facts within or applicable to those exchanges themselves change. Thus, in its market power determination, Ofcom both propounded its market definitions, and also set out in its Appendices 1, 2 and 3 how those definitions were reflected in individual exchanges. Ultimately, in a very real and practical sense, the markets are defined by those allocations, albeit it has been necessary in reaching those allocations to apply rationalisations of those allocations to demonstrate that the latter are not simply arbitrary but are patterned in accordance with principles of rationality and equality. It follows that if the question is asked whether there has been a change over time in for instance market 1, it is possible to say both that the definition has not changed but also that the allocations have, as the facts have changed. However, what Ofcom has done in its charge control determination is to change its definitions while purporting not to do so.
41. Thus in the present case, what Ofcom has done, as the facts have changed, is to argue that the facts have not changed (at any rate materially), and it has given its reasons for explaining that, on a forward looking basis, but, in order to make its new rationalisation fit it has had to alter the definition of the market it is considering. Indeed, it has had to alter the definitions of more than one market, because an exchange in which TalkTalk became committed to unbundling in the interval between market power determination and charge control determination now falls within the definition of market 2 but is excluded from it.
42. Sixthly, the only way in which Ofcom could reason that it was not setting about a redefinition of its previously defined markets was to state that its previous definitions were set in stone so far as timing was concerned. In other words, a deadline came down in December 2010 (there is even the possibility that it regarded June 2010 as such a deadline) for the purpose of commitments for future unbundling, and no future commitments would suffice because they were outside that deadline. There are straws in the wind of such a rationalisation in the relevant paragraphs found in both determinations, but no attempt to explain why the difference of a few months in commitment and identification of the relevant exchanges makes for immateriality. Indeed, the essential argument in the market power determination had been that if TalkTalk's new information about further unbundling plans had amounted to a commitment in specific identified exchanges, then it would have sufficed to bring those further exchanges into the newly appropriate market allocations.
43. However, ultimately, no such deadline rationalisation is put forward: nor in my judgment could it be, which is very probably the reason why it was not: for the whole purpose of section 86 is to avoid a situation where there is a time-gap between market power determination and charge control determination – with the potential therefore for the facts to change – without a proper consideration of those new facts. Either those new facts must be immaterial, or there has to be a new review, in effect an updated review, of market power. Therefore Ofcom cannot claim that its earlier market power determination is set in stone pending the next review. As it is, there is an element of such, to my mind erroneous, reasoning in Ofcom's comments that there would be time enough to consider the position of the 700 further exchanges where unbundling was now committed – at the next review. On the whole, however, the reasoning of the charge control determination is that it makes no difference whether further unbundling is committed or not – for even at the end of the review period BT will have a market share in each exchange of some 70% or more. And that reasoning is simply inconsistent with the rationalisation of the market definition in the earlier determination.
44. Seventhly, it appears that one possible reason why Ofcom took the line that it did is because it considered that section 86(1)(b) gave it a "discretion which involves making a judgment". At this appeal Ofcom has not sought to support such a view of the subsection. On the contrary, it is common ground that the statute requires an objective assessment of the facts, which is fully subject to appeal on the merits. It is not suggested that Ofcom's decision is either a discretionary decision, or a decision which is subject only to a public law test of rationality. There is perhaps room for argument that the statutory language which refers to Ofcom being "satisfied" that there has been no material change makes Ofcom a public law decision-maker from whose decision an appeal can be taken on only public law grounds, rather than on the merits. However, the Tribunal pointed out that the appeal to it fell within section 192(1)(b) of the 2003 Act and as such, pursuant to section 195(2), shall be decided "on the merits". It followed that the more limited standards of review applicable to judicial review were not in play, citing Hutchison 3G UK Limited v. Office of Communications [2008] CAT 11 at para 164 ("The question for the Tribunal is not whether the decision to impose price control was within the range of reasonable responses but whether the decision was the right one"). However, the Tribunal also went on to direct itself that it might be very difficult for it to disagree with Ofcom "if all that is impugned is an overall value judgment based upon competing commercial interests in the context of a public policy decision" (per Jacob LJ in T-Mobile (UK) Limited v. Office of Communications [2008] EWCA Civ 1373 at para 31); and also that it should be cautious about disagreeing with Ofcom where there may well be no single "right answer" (see T-Mobile (UK) Limited v. Office of Communications [2008] CAT 12 at para 82).
45. No doubt this court should be even more cautious in such situations where both Ofcom and the specialist appeal tribunal constituted in the CAT have come to the same conclusions. However, in this case we are not concerned with fine judgments of competition analysis but with definitions of markets which are the essential building blocks on which that analysis is contingent.
46. That is a suitable introduction to the decision of the Tribunal itself.

The Tribunal's judgment

47. It needs to be said that much of the Tribunal's reasoning is in favour of TalkTalk.
48. First, the Tribunal accepted TalkTalk's submission that where a previous market power determination has been affected by altered circumstances which could affect a subsequent charge control determination (in section 86(6)'s language, could affect or is material to the setting of a condition in question), then it would be necessary to put into effect the section 86(1)(a) requirement of a further review of the previous market power determination. As the Tribunal said:

"92 … the rationale of section 86(1)(b) is to enable an SMP services condition to be imposed without OFCOM having to conduct a contemporaneous market power determination. The only reason that this is possible is that there is, in existence, a Prior Market Power Determination which is unaffected by subsequent changes. It is self-evident that the Prior Market Power Determination can only be relied on in this case, i.e. where it is unaffected by subsequent changes. Where the Prior Market Power Determination has been affected by subsequent changes, and these changes are capable of affecting the Subsequent Charge Control Notification, it is equally self-evident that the prior Market Power Determination can no longer serve as the basis for the price control.

93. We do not understand how, where a material change has occurred, it can be said that the Subsequent Charge Control Notification would not be affected or not materially be affected by the change, without actually carrying out a further market power determination."

49. I agree. I do not think this has been disputed on this appeal. In any case, on the facts of this case, if market 1 has been affected by TalkTalk's commitment to unbundle in 700 further identified exchanges, on any view of the section 86 test of materiality I cannot see how the alteration of market 1, as constituted in Appendix 1 to the earlier determination, can fail to affect the setting of a charge control by the subsequent determination. Ex hypothesi, the charge control will apply to a different cohort of exchanges. Since each exchange is in its way a separate universe, it is hard to see how any reallocation of a single exchange is other than material. Perhaps it could still be argued that materiality ought to be considered across the market as a whole: I am sceptical, but I need not decide that. On any view, the number of affected exchanges, and the number of potential consumers affected within those exchanges, make the scenario under consideration one of material change. That is what the Tribunal itself said on the hypothesis that market 1 did not apply to identified exchanges where TalkTalk was committed to unbundle (at para 105 of its judgment, where it considered a difference between Ofcom and TalkTalk about the actual number committed to unbundling under TalkTalk's new plans, and said: "Accordingly, even though we consider that Mr Heaney's figures overstate matters, that overstatement is in no way sufficient to reduce the change relied upon by Talk Talk to an immaterial one").
50. Secondly, and accordingly, the Tribunal found that if TalkTalk were justified in basing itself on the definitions of markets contained in para 1.19 of Ofcom's market power determination, then there had been a material change for the purpose of section 86(1)(b). The Tribunal stated, plainly:

"106. In our view, on the basis of the definitions of markets 1 and 2 contained in paragraph 1.19 of the WBA Market Power Determination, there has been a material change."

I have cited para 1.19 at para 14 of this judgment above. The Tribunal repeated its (obiter) point if anything even more strongly at its para 113:

"Clearly, if OFCOM had in fact defined Market 1 as comprising those exchanges where "only BT is present or forecast to be present", and – in the period between the WBA Market Power Determination and the WBA Charge Control Decision – it became clear that as regards 40% of those exchanges, BT was not the only communications provider forecast to be present, there would be a material change for the purposes of section 86(1)(b). That is obvious."

51. Thirdly, however, and critically, the Tribunal repudiated Ofcom's own definition of market 1 (and thus of market 2) as found in Ofcom's para 1.19. It did so even while stressing that "it would not be permissible for the Tribunal to re-visit OFCOM's market definition, and we have not done so …" (at para 108(c)). Nevertheless, the Tribunal did just that. It reasoned that para 1.19 was a misstatement of Ofcom's market definitions, and that on a consideration of Ofcom's market power determination as a whole and in particular its paras 3.169 to 3.183 (which I have cited above), Ofcom was employing a different definition of market 1, which it addressed as follows:

"110. In short, OFCOM deliberately decided to keep within Market 1 those exchanges in respect of which there was going to be unbundling by Talk Talk. It follows that the definition of market 1 in paragraph 1.19 of the WBA Market Power Determination is wrong. At the time of the WBA Market Power Determination, OFCOM included in Market 1 those exchanges in respect of which Talk Talk was minded to unbundle. In other words, the definition of Market 1 is not "exchanges where only BT is present or forecast to be present", but rather "exchanges where only BT is present or forecast to be present or where, during the period of the market determination, Talk Talk may (at some point in the future) be present."

The Tribunal repeats that conclusion in its para 114.

52. Thus the Tribunal repudiates Ofcom's own market 1 definition ("is wrong"), and reformulates a new market 1 definition for itself, while saying that it is not doing so and accepting that it is not entitled to do so. All that is in turn repudiated by Ofcom on this appeal. It follows that, unless there is some other basis upon which Ofcom's or the Tribunal's judgment can be supported, this appeal should be allowed.
53. It will, moreover, be observed that in its reformulated market definition the Tribunal stipulates that it was Ofcom's expressed intention, already in its earlier market power determination, to formulate a market definition which sought to eliminate from market 2 and retain in market 1 any exchange which after 3 December 2010 was identified as one in which TalkTalk was committed to unbundling. Not only is that something which Ofcom did not there do (although it sought to do so in its later charge control determination), but, if it had attempted to do so in its market power determination, then it would have been begging the very question which section 86 requires it to consider for the purposes of a later charge control determination.
54. What then is the basis upon which Ofcom seeks to support its own charge control determination and the Tribunal's judgment? It is, as already introduced above, that Ofcom had an alternative reason for finding no material change despite TalkTalk's new information, namely in its, Ofcom's, assessment that even in the newly identified exchanges in which TalkTalk was now committed to unbundling BT would retain a 70%-80% market share until the next review some years away. However, for the reasons stated above, I am unpersuaded that this assessment, expressly carried out in the market power determination in relation to unidentified exchanges to the unbundling of which TalkTalk was not committed, can be applied, as Ofcom has sought to do, to newly identified exchanges to the unbundling of which TalkTalk had become committed – and all for the purpose of seeking to show that after all it made no difference, despite the market definitions which Ofcom had adopted, whether or not TalkTalk was committed to further unbundling in specific identified exchanges.
55. In my respectful judgment, the difficulty of this reasoning is demonstrated by the Tribunal's own attempts to make use of Ofcom's analysis. I have already shown that the Tribunal considered that it was necessary to redefine the market power determination's market definitions in order to make good its conclusion. I now cite the passage which on this appeal Mr Holmes submitted was "key", viz –

"109. As we noted in paragraph 61 above, it is clear from paragraphs 3.169 to 3.183 of the WBA Market Power Determination (which are set out in paragraph 60 above) that OFCOM knew of, and considered, TalkTalk's rollout proposals – including the proposal to roll out in exchanges allocated by OFCOM to Market 1 – and nevertheless decided to continue to allocate these exchanges to Market 1. In other words, OFCOM factored into its Market 1 definition TalkTalk's intended further roll out, even though this roll out could not be said to be "committed" as OFCOM had defined that term. Even though Talk Talk's proposed roll out fell to be classified as "uncommitted", OFCOM clearly took it into account when considering the definition of Market 1. The definition of Market 1 – in terms of the identity of the various exchanges falling within it – remained unchanged because OFCOM considered that even if Talk Talk unbundled a significant number [of] Market 1 exchanges early on, BT's market share in those exchanges would remain so great that some form of price control would remain appropriate" (emphasis added)."

56. The difficulty with relying on this passage is that it is immediately followed by the Tribunal's para 110, cited above, in which the Tribunal found it necessary to redefine Ofcom's market definition. As para 110 began: "In short, OFCOM deliberately decided to keep within market 1 those exchanges in respect of which there was going to be unbundling by Talk Talk." But that is not correct. Rather, Ofcom kept within market 1 those exchanges where BT was currently the only operator and where there was currently no commitment for any other operator to compete. Just as Ofcom's charge control determination seeks to blur its market power determination and definitions so as to cover the new situation it found a few months later, so the Tribunal seeks to do the same – while expressly recognising that it should not be doing anything to redefine Ofcom's definitions. It is simply an error of the Tribunal to say that Ofcom "factored into its market 1 definition TalkTalk's intended further roll out". It did not, precisely because it was not committed.
57. A further difficulty is to be found in para 115 of the Tribunal's judgment, where the Tribunal appears to be uncertain about the significance of an announcement to unbundle further exchanges by a communications provider other than TalkTalk, in the period after the market power determination but before the charge control determination. The Tribunal says:

"The matter is far from clear, on the face of the WBA Market Power Determination. Had another communications provider, other than TalkTalk, announced plans to unbundle significant further (unidentified) exchanges after the date of the WBA Market Power Determination, then we are not persuaded that such plans would have been factored into OFCOM's Market 1 definition."

Perhaps this is too obscure to found a comment, but I am inclined to read this as expressing some uncertainty about the materiality of such an announcement by another potential competitor. However it is expressed in terms of "unidentified" exchanges and thus a situation where there is no commitment. Such a hypothesis would not therefore affect market 1. The Tribunal's uncertainty about even such a case gives me no confidence about its reasoning.

58. Mr Holmes submits that TalkTalk's appeal is an attempt to put form above substance. He submits that the market definitions are form, and that what matters is competition analysis substance, and that the relevant paragraphs of Ofcom's two determinations upon which it relies, set out above, provide the substance which shows that Ofcom and the Tribunal are right to say that there has been no material change. That is an attractive submission, especially when proper account is taken, as it must be taken, of Ofcom's and the Tribunal's expertise in this field. Nevertheless I have concluded that it runs counter to the statutory prescription. The substance of competition analysis may be helpful to the definition of a market or markets and the presence of significant market power within such markets. Where that determination and the remedies to be applied as a result, in the form of a charge control determination, are concluded and notified at one and the same time, the substance will properly guide the regulator to its conclusions. Where, however, the underlying market power analysis and the remedy analysis are done at different times, section 86 requires caution on the part of the regulator. It must perform an updated market power review unless it can satisfy itself that there has been no material change in the underlying market which can affect the setting of a charge condition. The provisions of sections 79 and 84 also underline the significance of the building blocks of market identification and analysis. In such a situation, Ofcom must be cautious not to elide a previous market power determination with its charge control determination, thereby disregarding the need for a current updated review, just because it senses that, although there has been a material change, it will not after all, at the end of the day, affect its remedies. But that is the issue, and one that a competitor who wishes to add to the competition, and can satisfy Ofcom as to his seriousness and commitment, is entitled to be properly consulted about in the new circumstances.
59. Mr Pickford, on behalf of TalkTalk, has referred us to provisions in the Framework Directive, the 2003 Act, and in Ofcom's own published regulatory principles, to the effect that there should be a bias against ex ante intervention, and that such intervention should be proportionate and justified and only imposed where there is no effective competition. In other words, as he submits, intervention should be no greater than is necessary. Section 3 of the 2003 Act refers to the need for Ofcom to have regard to principles under which "regulatory activities should be transparent, accountable, proportionate, consistent and targeted only at cases in which action is needed". Section 6 requires Ofcom to secure that its regulation does not involve the "imposition or maintenance of unnecessary burdens". Section 45(8)(a) requires as a condition for the imposition of an SMP services condition against a person that he must be found to have significant market power "in a specific market". Mr Pickford relies on such material as supporting his submissions that in this case Ofcom has erred in finding that there was no need for an updated review of the markets to which the newly identified exchanges were to be allocated. I agree that such provisions in general support his submissions. However, I have not thought it necessary to bring them into account in order to reach my conclusions, which I have arrived at on the reasoning of Ofcom's two determinations, as reviewed on appeal to the Tribunal.

Conclusion

60. It suffices to say that in principle I would allow this appeal by TalkTalk for the reasons stated above. As for the consequences of this conclusion, the court has heard some submissions from the parties, including Sky represented by Mr Philip Woolfe. There is dispute as to the correct order to make: it is pointed out that it does not follow, some years down the line and with a new review about to take place shortly, that the charge control determination of July 2011 should simply be set aside. There was discussion as to whether the issue of remedies following the allowing of the appeal should be remitted to the Tribunal, or even to Ofcom.
61. In my view, the parties should make fresh submissions to the court, in the light of this judgment and taking into account the current situation, as to its consequences.

Sir Timothy Lloyd

Introduction

62. This appeal is brought by the appellant TalkTalk against an order of the Competition Appeal Tribunal made on 10 January 2012, giving effect to a judgment delivered on that date with the reference [2012] CAT 1. The Tribunal consisted of Marcus Smith Q.C. as chairman, sitting with Clive Elphick and Jonathan May. By its order the Tribunal dismissed TalkTalk's appeal against Ofcom's decision dated 20 July 2011 (which I refer to as the Charge Control Determination) to impose particular charge controls on BT in one of the four markets which Ofcom had identified in the prior Market Review Determination on 3 December 2010.
63. On 21 March 2012 the Tribunal refused permission to appeal against its order: see [2012] CAT 8. Lord Justice Etherton refused permission to appeal to this court on a consideration of the papers on 10 August 2012, but at a hearing on 30 October 2012 Lord Justice McCombe and I granted permission to appeal. The appeal came on for hearing on 11 and 12 February 2013. It is a matter of great regret that it has taken us a long time to determine the appeal.
64. As my Lord Sir Bernard Rix has explained, the case arises from Ofcom's review of the Wholesale Broadband Access (WBA) Markets initiated in 2010, looking forward over four years. By the Charge Control Determination which is under challenge Ofcom imposed charge controls on BT in what they defined as Market 1 from 17 August 2011 (28 days after the date of the determination) until 31 March 2014. (By that time a three year pattern had been introduced under EU regulatory legislation: see the revised Framework Directive, article 6 of Directive 2009/140/EC.) We were told at the hearing that Ofcom had already embarked on the process of review of these markets for the following period, and that they planned to issue a consultation paper in May 2013, in order to be able to put a new regime in place by the end of March 2014, when the current charge controls will come to an end. In fact, as appears from Ofcom's public material, the relevant consultation document was published on 11 July 2013 and the consultation period expired on 25 September 2013.
65. Sir Bernard has set out the critical sequence of events and much of the relevant material from Ofcom's two documents - the Market Review Determination in December 2010 and the Charge Control Determination in July 2011 – and from the Tribunal's judgment. As appears from his judgment, the Tribunal took the position that Ofcom had expressed their determination of the respective markets (and in particular Market 1) incorrectly at paragraph 1.19 of the Market Review Determination. We were told that this was a point which had not been contended for by any party before the Tribunal. With respect to the Tribunal I agree with Sir Bernard that, on this point, the Tribunal was wrong.
66. Having come to that erroneous view, the Tribunal said, first, that if Ofcom had defined Market 1 as they stated in paragraph 1.19, then in their Charge Control Determination they would have been wrong to say that there had been no material change, for the purposes of section 86 of the Communications Act 2003 (the Act). Secondly, however, as Sir Bernard has shown, it went on to express differently the definition of Market 1 which it held that Ofcom had in fact reached, and held that, on this basis, there had been no material change.
67. I agree with Sir Bernard that the Tribunal went wrong in rejecting the definition of Market 1 as stated in paragraph 1.19, and in devising a different definition of Market 1 as being that which Ofcom really decided. However, the question before us cannot be resolved only by showing that the Tribunal's reasoning was wrong. It is quite possible that the Tribunal's reasoning was wrong but that it was right to hold that Ofcom had not itself erred.
68. We have to decide, not whether the Tribunal was wrong, but whether Ofcom was wrong. The appeal to this court is, of course, limited to issues of law, unlike the appeal to the Tribunal which was on the merits. However, in the present case the issue is whether, on facts which I do not understand to be in dispute, there was or was not a material change, for the purposes of section 86 of the Act, between the time of the Market Review Determination and the time of the Charge Control Determination. That is a question of law.
69. In order to explain my own view as to whether Ofcom were wrong in stating in the Charge Control Determination that there had been no material change, I need to start with some observations about the process which Ofcom had to undertake. A good deal of this has been covered in Sir Bernard's judgment, and I will try to avoid unnecessary repetition, particularly in the way of quotation.

The regulation of telecommunications markets

70. A common characteristic of telecommunications markets is that they have been dominated, indeed often monopolised, by state-owned undertakings, such as BT used to be in the UK. By the time that the Common Regulatory Framework was first put in place by the EU, by Directives made in 2002, technological and commercial developments had been such as to require and prompt the development of a more competitive market in respect of telecommunications and related fields including media and information technology. Since then, the pace of development in terms of technology and the development of the relevant markets has in no way abated. Given the circumstances affecting these markets, and the way in which commercial operators work, the relevant economic pressures are such that there is a constant need for national regulatory authorities (Ofcom in the UK) to keep the relevant markets under review. The legacy of BT's former monopoly is capable of enduring for a considerable time, in certain markets, not least because of the practical feature that BT owns the telephone exchanges and the hardware used for connecting telephone calls to or from landlines.
71. That position is not in itself inconsistent with other undertakings offering broadband services to customers. They can do so in a number of different ways. For present purposes the two that matter are by putting in place their own infrastructure (Local Loop Unbundling, or LLU) or by entering into a contract with BT to take broadband access over BT's equipment on a wholesale basis and using that to offer the relevant services to their own retail customers. (I am not to be taken to be attempting to describe these alternatives in other than the most general terms. The Tribunal described the position more fully in paragraph 21 of its judgment.) BT's position in this respect is one that is clearly capable of giving it significant market power (SMP), as explained in section 78 of the Act, depending on the circumstances affecting the relevant market in terms of the development of competition.
72. For reasons such as these, WBA markets are among those which Ofcom has to keep under review from time to time. The EU regulatory regime used to leave the frequency for such a review to the discretion of national regulatory authorities, but, as already mentioned, since 2011 the frequency has been fixed at every three years (absent exceptional circumstances).
73. In the nature of such things, such a review, while proceeding from the present situation, must look forward over the relevant review period to forecast how circumstances, especially competitive conditions, are likely to change in the relevant markets. That is inherently necessary as a precondition for the imposition of ex ante regulation, and it is made plain by the European legislation, and particularly clearly so by the Guidelines to which Sir Bernard has referred at paragraph [4] above, in paragraph 20 which is there quoted and also in paragraph 27. In carrying out this task Ofcom will rely on their knowledge of the relevant markets, on what they have been told by operators in the market and other related markets, and by other relevant persons or bodies, in particular in the course of the consultation process which is an integral part of the procedure, and on their observation, knowledge and experience of relevant matters generally. On the basis of all this, it is for them to exercise their judgment not only as to what the competitive conditions are in relevant areas of economic activity, but also as to how they are likely to change over the review period. On the basis of that assessment they must take a view as to how to define a market in which competitive conditions are, and are likely to remain, sufficiently homogeneous to be treated as a single market for these regulatory purposes.

The statutory provisions

74. Ofcom's Market Review Determination was a market power determination for the purposes of the Act. Under section 79 of the Act Ofcom must first identify the markets which in their opinion are those which in the circumstances of the UK it is appropriate to consider whether to make a market power determination, and then carry out an analysis of those markets. Consultation processes are laid down for these purposes. Section 84 provides for the review of a service market where Ofcom have already identified and analysed a market in order to make a market power determination. Sir Bernard has set out the relevant provisions of both sections 79 and 84 at paragraph [8] above. The WBA market appears to be a classic example of a market in which such reviews are appropriate and necessary.
75. So we come to the critical section, section 86, for the text of which see paragraph [9] above. The essence of this provision is that, if Ofcom is to set an SMP condition, imposed on an undertaking which has SMP in the relevant market, either it must do so at the same time as making the market power determination, or, if it does so later, it must then be satisfied that there has been no material change in the relevant market since the market power determination was made; if there has been a material change it must reassess the market before imposing the SMP condition.

The Market Review Determination

76. When Ofcom made their Market Review Determination in December 2010, they decided to impose a charge control on BT by way of an SMP condition in relation to Market 1. This was the first time that Ofcom had decided to do that in relation to WBA markets. It was therefore not surprising that they should wish to take time and undertake further consultation before specifying what the charge control would be. In other respects, Ofcom imposed immediate SMP conditions on BT as regards both of Markets 1 and 2, but these are not controversial.
77. Ofcom undertook the further consultation during 2011, and came to their decision by the Charge Control Determination in July. This said that there had been no material change since the Market Review Determination. The question is whether that was right. In order to examine that question, it is necessary first to refer to the Market Review Determination itself.
78. That determination was arrived at following a consultation process which commenced with a consultation document dated March 2010. This was then followed up by a further document dated August 2010. Nothing turns on the details of that process. Ofcom were very nearly ready to publish the Market Review Determination when, on 16 November, TalkTalk announced its intention to undertake LLU in relation to a further 700 exchanges. This announcement caused Ofcom to delay the publication of the Market Review Determination, and to enquire of TalkTalk for more detail of its plans.
79. Ofcom then revised the draft of the Market Review Determination, and issued it, as already mentioned, on 3 December 2010.
80. The Market Review Determination consists of a summary (section 1), an introduction (section 2), a long passage dealing with market definition (section 3), then a market power assessment (section 4) and a long section dealing with remedies (section 5), followed by four annexes, of which two are relevant for present purposes: annex 2 on market definition methodology and annex 3 on geographic analysis.
81. Ofcom found there to be effective competition in 80% of the UK, but not in the remaining 20%, made up of Markets 1 and 2 (as well as Hull). BT was found to have SMP in both Markets 1 and 2, but to differing degrees, so that although SMP conditions were imposed for both markets, a charge control was only to be imposed in Market 1.
82. There is no issue as to the product definition aspect of the market, summarised in paragraph 1.17. The issue is as to the geographic aspect of the market, summarised in paragraph 1.19, for which see paragraph [14] above, with the important footnotes set out in paragraph [15] above. The Tribunal appears to have overlooked the footnotes. It is true that they do not appear, as such, where the markets are defined in paragraphs 3.5 and 3.194, but the important point as to committed plans to deploy (footnote 6) is made at paragraphs 3.75, 3.177 and 3.183. It is also fair to say that there are minor discrepancies between the statements of the definition of the markets at different points in the Market Review Determination. If the document is considered as a whole, however, it seems to me that there is no real issue of definition, and that the text in paragraph 1.19, with its footnotes, in which the word "firm" is equivalent to "committed", is a fair statement of Ofcom's position in this respect. The Tribunal's statement at paragraph 61(b) that paragraph 1.19 is "obviously wrong" is, I have to say, itself obviously wrong, since it ignores the effect of the footnote and Ofcom's clearly stated decision to treat an operator as present in an exchange only if it was actually present or it had firm and committed plans to deploy in that exchange.
83. In the Market Review Determination, in the course of the summary as regards remedies, Ofcom went on to make the observations to which Sir Bernard refers at paragraph [18] above.
84. Section 2 of the Market Review Determination, Introduction, requires little citation. However, at paragraphs 2.27 to 2.33 points were made which, though perhaps obvious in the context of a review which has to look forward to what was likely to happen in future, as required by the EU Guidelines, are worth quoting:

"2.27 Rather than just looking at the current position, market reviews look ahead to how competitive conditions may change in the future. Our evaluation of the current market takes into account past developments and evidence. Then we assess whether any lack of effective competition is durable, by considering expected or foreseeable market developments over a reasonable period in the future.

2.28 The actual period used for this forward look should reflect the specific characteristics of the market and the expected timing for the next review. In this market review, we have looked at potential developments over the next four years.

2.29 In this market, the key anticipated change over the next four years is that a significant amount of NGA [Next Generation Access] infrastructure will be deployed. This will support 'super-fast' broadband services, offering higher speeds than have been experienced so far by UK consumers. However, there is uncertainty about the extent and timing of NGA investment. This makes it harder to foresee how the existing competitive conditions will change over the next few years. It is possible that the WBA market will change quickly in the future, for example as the speed of NGA deployment picks up.

2.30 However, based on past data and the information before us, we are of the view that competitive and technological developments in the UK are not expected to materially affect our proposed market definitions within a four year period. Although services that require higher speed access may evolve as next generation rollout develops, we anticipate that the majority of broadband users' requirements will continue to be able to be met using current as well as next generation network access and speeds.

2.31 We also consider a four year forward look to be reasonable in this case as this period provides a reasonable degree of regulatory certainty to stakeholders in the UK. Such certainty is especially valuable at this point in time as it provides the right context for investment decisions during this important early phase of NGA deployment, in which the future market for NGA services is not yet clear. Whilst investment in NGA deployment may be considered to be more relevant to the WLA market, the WLA and WBA markets are closely related and our analysis of the WBA market takes account of the state of competition in the WLA market. This period of forward look also covers the entire period over which we will set the charge control that we have decided to impose in Market 1.

2.32 The four year forward look that we have used allows for some flexibility in the date of the next WBA market review and allows for the review to occur before the end of this four year period. Given the potential impact of developments in this market in the next few years (such as additional LLU rollout and NGA deployments), we will monitor closely the WBA market, and we will consider the timing of the next market review accordingly.

2.33 In considering the timeframes for the next review, we will consider the new procedures and timeframes for conducting market reviews introduced by the amendments to the EU regulatory framework."

85. Thus, Ofcom took a view, explicitly, as to how circumstances in the relevant markets might change in the future over – at that time – a four year timescale, in particular as to how competitive conditions might change during that period. In so doing they had express regard to the desirability of regulatory certainty over a reasonable period.
86. In section 3, Market definition, the passage that matters for our purposes starts at paragraph 3.65, with the heading "wholesale geographic market definition". At paragraph 3.67 a fairly basic point is made:

"3.67 We would further note that the issue of geographic market definition involves an element of judgement since there is a wide variation in the competitive conditions across different areas of the country. We have sought a position that appropriately balances the competing views, considers fully all the available evidence and uses objective criteria to identify areas in which competitive conditions are sufficiently homogeneous to be regarded as a single market."

87. Sir Bernard has quoted paragraph 3.75 at paragraph [20] above, in which the point is made (for the first time, other than in footnote 6 to paragraph 1.19) that Ofcom will only rely on committed plans to deploy in a given exchange when counting the number of operators present in that exchange.
88. It appears that a good deal of the focus of Ofcom's second consultation was about the distinction between Markets 2 and 3, rather than as between Markets 1 and 2. At paragraph 3.107 Ofcom explained that in their second consultation document they had chosen a threshold of 50% as regards BT's presence in an exchange, both on the basis that this is the standard threshold at which SMP can be presumed, and also because it:

"is also high enough to allow for the effects of continued rollout by POs. In choosing this threshold we did not take an explicit view on either the likely fall in BT's share over the period covered by the view or the exact level at which SMP can be considered a risk."

89. Ofcom's approach to the analysis required in this process is also described in Annex 2. At paragraph A2.32 they described their methodology as involving three steps, the first of which is to select the basic geographic unit, in this case exchange areas. They continued:

"Second, the homogeneity of competition needs to be judged according to factors such as barriers to entry, the number of significant suppliers in the market, distribution of market shares and price-cost margins, and as such necessarily means the geographic market definition and SMP analysis are somewhat inter-related; and

Third, areas with similar competitive characteristics need to be aggregated in order to define the geographic areas over which to conduct the SMP analysis."

90. In Annex 3 they gave more detail of their process as regards geographical analysis. At paragraph A3.8 they referred to TalkTalk's announcement and to the fact that the relevant exchanges were not yet identified. In the previous paragraph they had referred to some Principal Operators having provided forecasts of rollout plans beyond December 2010 which were not confirmed and which therefore were not taken into account as showing that the relevant operator was present, or forecast to be present, in the particular exchange. However, for completeness Ofcom included these forecast plans in the Annex "to show the effect they would have on the geographic market definition if they are implemented in full". By contrast, Ofcom did not include the TalkTalk information in the Annex even though it might be regarded as similar to the uncommitted announcements by other Principal Operators. They said that they did not regard this as having a material impact because they did not take account of these uncommitted plans in their assessment of the presence of Principal Operators in any given exchange. Sir Bernard has quoted paragraphs A3.6 to A3.8 in his paragraph [27] above. The difference in treatment of the uncommitted announcements or forecasts by TalkTalk and those by other Principal Operators does not seem to me to be of any relevance given that Ofcom took no account of any such uncommitted plans when considering whether an operator was present, or forecast to be present, in a given exchange. At A3.35 Ofcom said in terms that they would use only committed forecast rollout plans to the end of December 2010 in deciding whether a Principal Operator was to be regarded as present, or forecast to be present, in a given exchange. The Market Review Determination was published at the beginning of December 2010, but it was based on data provided up to June 2010: see footnote 5 to paragraph 1.19 quoted at paragraph [15] above.
91. In the main body of the decision, Ofcom devoted a lengthy passage to the implications of TalkTalk's announcement of the intention to unbundle in 700 more exchanges. Sir Bernard has set out the relevant passage (paragraphs 3.169 to 3.183) at paragraph [21] above. As appears from this, Ofcom were faced with a decision as to whether to wait until TalkTalk could provide firm plans as to the exchanges in which it intended to unbundle, and indicative timescales for that process, or to conclude the review having taken account of TalkTalk's announced plans. At paragraph 3.173 they noted that TalkTalk could not be expected to provide firm information as to which were the relevant exchanges for several months, and that if Ofcom were to wait until then, they might also need to gather updated information from other operators, and that they might need to undertake further analysis and consultation. They expressed the view at paragraph 3.176 that TalkTalk would not be in a position to exert a practical constraint in any new exchanges for a period of 6 to 9 months. Rollout in some exchanges could take up to three years, though TalkTalk might seek to deploy more quickly than that. Given the approach already stated of counting an operator as present in an exchange only where it had firm committed plans for that exchange, TalkTalk was at that time not to be regarded as present in any additional exchange. At paragraphs 3.178 and 3.182 Ofcom made important points which, though Sir Bernard has quoted them already, I will repeat because of their significance:

"3.178 In carrying out a market review we are required to take a forward look at how competitive conditions may change over the period of the review. Whilst accepting that deployment in Market 1 exchanges will have an impact, we are also mindful that, based on the timescales above, a significant portion of the market review period will be characterised by BT being the only provider even in exchanges that Talk Talk chooses to unbundle. As such, our regulatory approach needs to balance the potential for further competition towards the end of the review period with BT's position of being a monopoly provider for the earlier part of the review period. …

3.182 It is clear that at the start of the period covered by the review BT's position in Market 1 exchanges where Talk Talk subsequently deploys would be the same as that for all other exchanges in Market 1. BT would be the only provider and would, as such, face no competitive constraints. Based on the potential for migration of customers from BT wholesale products onto Talk Talk's own network, and considering the effect when a second PO is present in other exchange areas, we are of the view that even if Talk Talk deploys towards the start of the review period BT's market share would be likely to be at least 70 to 80 per cent in the exchanges where Talk Talk deploys at the end of the review period. The information available from Talk Talk indicates that deployment would be over the period of the review and so the effect on BT's share would be less than this in many of the exchanges. Where BT's share is at this level and it faces competition from only one other provider, a charge control may still be considered to be an appropriate remedy."

92. Ofcom decided not to wait until more information was forthcoming, but to complete the review on the material then available. In itself, that is not criticised in these proceedings.
93. Ofcom's conclusion is set out at paragraphs 3.187 to 3.194, as quoted (as far as paragraph 3.190, which is what matters) at paragraph [22] above. They recognised that TalkTalk's announcement indicated a greater scope for LLU deployment in Market 1 than they had previously considered. They thought it unlikely that TalkTalk would have any exchanges unbundled within the next 9 months; rollout would continue throughout the remaining period of the review (to March 2014). On that basis they concluded that the competitive conditions in Market 1 exchanges where TalkTalk came to deploy in future could be considered to be sufficiently similar to those in which TalkTalk does not deploy. In terms, they considered it appropriate that exchanges where BT is expected to have a very high market share and will face only a single competitor entering at some point during the market review period should be subject to charge control. Accordingly they decided to maintain their formula for defining Market 1 without any alteration: to reiterate, it is exchanges where only BT is present (or forecast to be present), the presence (or forecast presence) of another Principal Operator in a given exchange being counted for this purpose only where the other operator has firm committed plans to deploy in that specific exchange. Since TalkTalk's plans did not then relate to specified exchanges and were not then firm or committed, Ofcom's use of this formula resulted in TalkTalk's future plans being ignored for the purpose of identifying the exchanges within Market 1.
94. In the last sentence of paragraph 3.189 Ofcom referred to their forecast as including only confirmed plans expected to be implemented by December 2010. For the most part they were working on information provided to them by June 2010, in the course of the consultation process. That accounts for the reference to looking forward to December 2010 but no further. Mr Pickford made a point about cases where such deployment had been forecast, but had not materialised by December 2010. I will come back to that.
95. The specific exchanges in each of the four markets were identified in Annex 1 to the Market Review Determination. No issue turns on the list in the relevant part of the Annex as it then stood.
96. In section 4 of the Market Review Determination, Ofcom addressed the specific question of whether BT had SMP in the relevant markets. At paragraph 4.9 they observed that in the 3,389 exchanges in Market 1 BT only faced competition from small operators which did not count as a principal operator or from Virgin Media in some cases (not relevant for present purposes). They recognised, as discussed in section 3, that over the period of the review, entry by other principal operators, such as TalkTalk, might provide greater levels of competition in these exchanges. They noted, however, that there were substantial barriers to entry, in the form of sunk costs which could not be recovered on exit, and economies of scale. They remained of that view despite the responses to the consultation process. At paragraph 4.39 they said this:

"Further, in exchanges that Talk Talk unbundles towards the start of the review, it may be expected that BT's share will fall to 70 to 80 per cent during the period of the review. Conversely, the decline in share will be less in exchanges where Talk Talk deploys later and there remain just less than 3,000 exchanges where BT remains the only PO and as such will maintain a market share of around 100 per cent. Therefore, on average, BT's share is likely to be above 80 per cent even by the end of this review in the exchanges allocated to Market 1."

97. Thus, they remained of the view that BT had SMP in the Market 1 exchanges, even after taking account of the greater potential for entry suggested by TalkTalk's intention to deploy in a further 700 exchanges: paragraph 4.41. They also remained of the view that BT had SMP in the Market 2 exchanges and that no operator had SMP in the Market 3 exchanges.
98. In section 5 of the Market Review Determination Ofcom dealt with the issue of remedies. Sir Bernard has quoted at paragraph [26] above a pertinent paragraph, 5.87. In turn, at paragraph 5.91 they said this:

"We consider here the impact of Talk Talk's announcement of its intention to unbundle further exchanges. Whilst entry by Talk Talk in some exchanges in Market 1 during the period of the review will provide some competition to BT, we remain of the view that a charge control is required in Market 1 because the deployment is as yet uncertain on an exchange level basis and the effects of deployment as a competitive constraint may only become apparent towards the end of the review period. Further, even taking account of Talk Talk's deployment, we anticipate that BT will continue to enjoy very high market share in Market 1."

99. At paragraph 5.182 they said that the remedies which they proposed to impose (including, but by no means limited to, charge control in Market 1) "will operate together effectively to ensure effective Network Access on terms and conditions that will enable third party providers to compete effectively with BT in Market 1". Nothing turns on the detail of the SMP conditions in question, either those imposed at the time of the Market Review Determination or those which were the subject of later consultation, including the charge control.
100. Ofcom dealt in some detail with the various conditions to be imposed in the latter part of section 5. When it came to charge control in Market 1, they said this:

"Aims and effects of the condition

5.289 As discussed above, in Market 1, BT is currently the only provider. We do not consider that future entry (for example by Talk Talk, or the threat of entry, will act to constrain BT's wholesale prices. As such, BT has the ability and the incentive to set prices above the competitive level. BT's competitors at the retail level would be forced to pay these high prices in order to provide service on a national basis. We therefore are of the view that ex ante pricing obligations are required to address BT's SMP in Market 1.

5.290 BT is currently the monopoly provider in Market 1 and, even when the potential for future entry is accounted for, BT's market share is likely to remain very high. It is therefore unlikely that BT will be incentivised to reduce its costs and set prices at the competitive level. It would be likely to be able to recover higher costs through higher prices charged at the wholesale level, which would ultimately be passed on in higher retail charges.

5.291 In addition there are significant costs related to the WBA market that are not specifically allocated to the different geographic markets. BT may seek to recover these costs, as well as common costs, through its prices in Market 1.

5.292 Imposing a charge control allows for these effects to be addressed. It will provide more certainty over the life of the control period about the maximum level of WBA charges. It will also result in prices being based on a forward-look view of the costs related to provision of service in Market 1 at the end of the period, taking into account efficiency improvements and possible future investment by BT that will be of benefit to consumers and citizens.

5.293 We will discuss the specific structure of the charge control in a separate consultation which we will publish shortly."

101. At paragraph 5.295, when discussing the legal tests which apply as regards the imposition of conditions, they said this:

"A charge control is objectively justifiable in order to restrict BT's ability to charge excessive prices to CPs that would ultimately be passed on to consumers in a market where BT currently faces no competitive or pricing constraints and where its pricing is unlikely to be constrained throughout the period of this review."

102. It is unnecessary to refer to other passages in this section in which they addressed others of the legal requirements for the imposition of conditions.
103. Accordingly, it seems to me to appear with reasonable clarity from the Market Review Determination that Ofcom defined Market 1 as consisting of exchanges in which only BT was present (among Principal Operators), or forecast to be present, such presence or forecast presence being assessed by taking into account actual presence or firm committed plans by a Principal Operator to deploy in a given exchange which were due to be implemented by December 2010, but no other plans. Of course Ofcom spent a good deal of time and space discussing TalkTalk's announcement, but it seems to me clear that, having undertaken that discussion, they put TalkTalk's announced plans into the same category as deployment plans by any other Principal Operator which were not firm and committed, nor specific as to the relevant exchanges. Thus, their analysis of TalkTalk's position did not alter their definition of Market 1. They were able to, and did, apply their definition of Market 1 without alteration or adaptation in respect of TalkTalk's position.

The Charge Control Determination

104. Ofcom then proceeded to consult about the charge control to be imposed on BT in relation to Market 1. The consultation document was issued on 20 January 2011, with responses due in by 31 March. The Charge Control Determination was issued on 20 July 2011. As already mentioned, it imposed charge controls on BT with effect from 17 August 2011 (28 days later) until 31 March 2014. That period, being somewhat less than three years, was consistent with the three year cycle by then applying under the EU regulatory framework. It also fitted with the balance of the "forward look period" which had been considered in the Market Review Determination.
105. The only issue taken with the Charge Control Determination is as to whether Ofcom was correct in reaching the conclusion that there had been no material change in the relevant market since the Market Review Determination. It is not necessary to consider any of the detail of the charge control itself, to which most of the determination was devoted.
106. Ofcom referred to the "no material change" requirement, in section 86, at two points in the document. The first is paragraph 2.16, as follows:

"Under section 86 of the Act, Ofcom can set an SMP services condition by a notification which does not also make the market power determination when the condition is set by reference to a market power determination made in relation to a market in which Ofcom is satisfied there has been no material change since the determination was made. We discussed the no material change since the market power determination for Market 1 in the paragraphs above. We considered the relevant legal tests for imposing a charge control as an SMP condition under section 87(9) of the Act in the January Consultation. In Section 7, we set out our reasoning as to why we consider our proposed charge control condition meets each of those relevant tests."

107. It seems to me that the statement that the no material change issue had been discussed in "the paragraphs above" must be an inadvertent slip, possibly in the course of editing or word-processing, since there is no previous reference to it in the document.
108. Section 7 of the document, referred to in the last sentence quoted above, dealt only with the other legal tests applicable to the imposition of the relevant kind of condition, not with the "no material change" question.
109. The passage in which Ofcom did address the question of material change substantively is in section 3, of which Sir Bernard has set out paragraphs 3.40 to 3.49 at his paragraph [32] above. Ofcom had introduced this passage by referring to responses by TalkTalk to the consultation in March 2011, and later representations in May and July 2011, in which it was contended that there had been a material change, on the basis that, if the Market Review Determination exercise were done again, on the same basis as before but by reference to the facts as they were by the time the point was being made, then the outcome of the review would be significantly different.
110. Sir Bernard has pointed out in paragraph [44] above that Ofcom's language at the end of paragraph 3.40 is inappropriate in referring to a "discretion which involved making a judgment". I agree with him that the issue of material change or no is not a question in relation to which Ofcom had a discretion. Either there had been or there had not been a material change. On the other hand, the determination of that issue does involve judgment, just as defining a geographic market in the first place does: compare Ofcom's paragraph 3.67 quoted at paragraph [86] above. Nothing else in Ofcom's Charge Control Determination indicates that they considered that they were exercising a discretion in relation to the question of material change. It seems to me that the reference to a discretion is unfortunate but that it does not show that Ofcom committed any legal error in addressing the question whether there had been a material change.

TalkTalk's deployment in Market 1 exchanges

111. By the time of the Charge Control Determination TalkTalk had not only identified the specific exchanges (or at least many of them) in which it intended to deploy by way of LLU, it had placed orders for the purpose as regards some of those exchanges, and it had in fact completed the process as regards some exchanges. Ofcom did not have specific details at the time of the Charge Control Determination, but it is not in dispute that if they had asked TalkTalk for details in the course of the consultation, the information would have been given. It is therefore right to proceed on the footing of what would have been told to Ofcom if the request had been made. Quite what that would have been would depend on the date as at which the request was made. TalkTalk's evidence before the Tribunal showed that by July 2011 TalkTalk had been able to complete the LLU process in 17 exchanges within Market 1 as it had been defined, and that by September 2011 the process was complete in 92 such exchanges, with delivery confirmed within 2011/2012 for a further 277 such exchanges. Therefore, by June 2011 the number of Market 1 exchanges in which TalkTalk was present or forecast to be present, on the basis of firm committed plans for specific exchanges, was significant, and more than could be regarded as de minimis.
112. It follows that, if the Market Review Determination process had been applied in July 2011 in the same way as it was in December 2010, by reference to information which TalkTalk would have supplied to Ofcom if asked, say as at June 2011, the outcome as between Markets 1 and 2 would have been different, and sufficiently different in terms of numbers of exchanges, and probably also of the size of the relevant body of customers, to count as a material difference.
113. If therefore the test for material change required a consideration of whether the result would be materially different if the earlier process were carried out again by reference to up-to-date information, then Ofcom were wrong to conclude that there was no material change. That is TalkTalk's contention, advanced by Mr Pickford. As I understand his reasoning, my Lord Sir Bernard Rix would hold that this is the correct approach.

Ofcom's approach as regards material change

114. Before stating my own view, I think it necessary to examine Ofcom's reasoning in the critical passage in the Charge Control Determination, as set out by Sir Bernard.
115. As he says, Mr Holmes, for Ofcom, did not contend that paragraph 3.42 of the Charge Control Determination is fully correct. He accepted that an announcement could constitute, or create, a change in the competitive conditions in the relevant market. The last sentence of paragraph 3.42 is too narrowly expressed.
116. As it seems to me, paragraph 3.44 is the more important passage.
117. It starts by referring back to the basis of the Market Review Determination. Having examined TalkTalk's potential rollout, Ofcom concluded that exchanges allocated to Market 1 in which TalkTalk deployed during the period of the review could, in terms of the market analysis exercise, be considered to have competitive conditions that were sufficiently similar to exchanges in Market 1 where TalkTalk did not deploy. Thus TalkTalk's forecast subsequent deployment would not make enough of a difference in the relevant exchanges to justify treating those exchanges differently from others in which TalkTalk did not deploy.
118. This conclusion was based on the fact that at the start of the period covered by the review there would be no competitive constraint on BT and that any potential future entry by TalkTalk would only introduce a constraint for part of the period covered by the review. At the start of the period, BT would be the only provider and would, as such, face no competitive constraints.
119. Looking at the possible or likely effect of deployment by TalkTalk, and based on the potential for migration of customers from BT wholesale products to TalkTalk's own network, and considering the effect of a second Principal Operator being present in other exchange areas, Ofcom said that, even if TalkTalk were to deploy early in the review period, BT's market share in the exchanges where such early deployment took place would be likely to be at least 70 to 80 per cent at the end of the review period. The information provided by TalkTalk had indicated that deployment would take place over the period of the review. Thus, Ofcom commented, the effect on BT's share would be less than this in many of the exchanges. This appears from passages in the Market Review Determination including paragraph 4.39, quoted at paragraph [96] above, and paragraph 5.91, quoted at paragraph [98] above. It is also stated in the conclusion as regards the need to impose charge control, at paragraphs 5.289 onwards, which I have quoted at paragraph [100] above.
120. In the Charge Control Determination they concluded paragraph 3.44 by saying that, in a case in which BT's share is at this level (at least 70 to 80 per cent) and it faces competition from only one Principal Operator, a charge control may still be considered an appropriate remedy.
121. Ofcom recognised, in paragraph 3.45, that TalkTalk had identified the particular exchanges in Market 1 in which it intended to deploy, and had started placing orders for some of these exchanges, but they considered that this did not alter their assessment of the competitive conditions in the affected exchanges. At paragraph 3.46 they observed that, in the affected exchanges, BT's market share was likely to remain above 70% throughout the period for which the charge control was to be applied. At paragraph 3.47 they said that, in exercising their judgment as to whether there had been a material change, they had to do so "in a way that allows the market review process to function effectively in the interest of promoting competition for consumers".
122. That last comment seems to me to make an important point. Ofcom must comply with general statutory duties set out in sections 3 and 4 of the Act in relation to the carrying out of their functions. Section 3(1) sets out two principal duties of Ofcom: to further the interests of citizens (defined as all members of the public in the UK) in relation to communications matters, and to further the interests of consumers in relevant markets, where relevant by promoting competition. ("Consumer" is defined for these purposes in section 405 (5) of the Act). Of course Ofcom must also act consistently with the other general provisions in the Act, derived from the Directives making up the Common Regulatory Framework, including the requirement that regulatory activities are to be transparent, accountable, proportionate, consistent, and targeted only at cases in which action is needed: section 3(3)(a). And they must comply with specific provisions of the Act, notably section 86(1) in the present instance. But the prime objective is to promote competition, above all in a market where an undertaking has SMP.
123. I have already discussed the forward-looking nature of the regulatory review, at paragraph [73] above. Ofcom undertook that task in making the Market Review Determination, and they are not criticised in that respect. Paragraph 3.67 (see paragraph [86] above) is one of the passages which shows them addressing these issues. Paragraph A2.32 in Annex 2 is also relevant here: see paragraph [89] above.
124. A crucial part of that exercise, as applied to an area of economic activity which is characterised by technological change, by commercial pressure to use and exploit such change, especially as a process of movement away from a situation of market dominance by a former monopoly supplier such as BT, and by active interest from retail consumers, is to make a forecast as to likely changes in competitive conditions. Ofcom made clear in the Market Review Determination the view that they took as to the likely effects of such future developments, and dealt in particular with the likely effect of TalkTalk's proposals as they would or might be developed over the review period. Their forward-looking approach is explained in general terms at paragraphs 2.27 to 2.33, quoted at paragraph [84] above. Their application of this approach appears from passages to which both Sir Bernard and I have referred, including in particular paragraphs 3.178 and 3.182 quoted at paragraph [91] above.
125. Ofcom is not criticised for having decided to proceed with the Market Review Determination despite TalkTalk's announcement of a major exercise by way of LLU, made at a late stage in Ofcom's review process. To decide not to wait for an uncertain period until TalkTalk's proposals were firm and specific was plainly a reasonable course, given that there would be many exchanges within Market 1 which would not be affected by TalkTalk's proposals, and the interests of consumers in those markets required the protection that Ofcom proposed to impose by way of SMP conditions. Although Ofcom did not mention section 86 in the course of the Market Review Determination, it was part of the statutory context, given that, also for obvious reasons, Ofcom decided to impose certain SMP conditions at once, but to defer the definition and imposition of the charge control condition until after future consultation.

Did TalkTalk's deployment amount to a material change?

126. Since it was in the nature of the relevant market that its economic and competitive circumstances were likely to change over time, in the course of the review period, and it was also inherent in the actual review carried out that it included an assessment of how those circumstances were likely to change, it seems to me that the fact that the circumstances did change cannot of itself be regarded as a material change for the purposes of section 86.
127. To test this proposition, I pose a hypothetical example. Suppose that Ofcom had taken and expressed the view in the Market Review Determination that by a given date, say the beginning of July 2011, TalkTalk would have deployed in 25 exchanges (not yet identified) in which no other Principal Operator than BT was present, and that by the beginning of October 2011 it would have deployed in a further 75 such exchanges, but that its presence in those 100 exchanges early in the review period would not be sufficient to act as a significant competitive constraint on BT in those exchanges during the review period, and that therefore this forecast and expectation did not justify treating those exchanges differently from others in which TalkTalk might deploy later and yet others in which it had no plans to deploy. (I ignore for this purpose the element of "firm committed plans to deploy" which was part of Ofcom's actual definition of the markets.) If the Market Review Determination had proceeded on that basis, it does not seem to me that, if Ofcom's forecast as to TalkTalk's pace of deployment had turned out to be correct, the fact that, by the time of the Charge Control Determination, TalkTalk had deployed in 25 relevant exchanges and was well on its way to deploying in a further 75 such exchanges could amount to a material change. That is so even though, if the Market Review Determination had been carried out for the first time at the later date, the 25 exchanges in which TalkTalk was already present by July 2011 would have fallen outside the definition of Market 1 and, depending on how any firm current plans were treated, so might the additional 75 exchanges. The same would probably be the case even if the numbers turned out to be slightly higher, so that TalkTalk had managed to deploy in, say, 30 or 35 exchanges by July 2011.
128. As I see it the reason for this is that the actual deployment would not have falsified Ofcom's forecast, and therefore would not have undermined the basis of the Market Review Determination or required it to be re-assessed in the light of changed circumstances.
129. Looking, then, at what Ofcom did say, at paragraphs 3.178 and 3.182 which I have quoted above at paragraph [91], a point made is that, for a significant part of the review period, BT would be the only provider, even in exchanges where TalkTalk does decide to deploy. Even where TalkTalk deployed early in the review period, that would not be likely to divert so many customers from BT to TalkTalk as to affect BT's market share enough to leave it without a preponderant market share. Not merely would BT retain SMP but, as Ofcom considered, its market share in those exchanges would remain at least at 70 to 80% by the end of the review period, even where TalkTalk deployed early in the review period. At paragraph 3.176 Ofcom had said that TalkTalk would not be in a position to exercise a practical constraint on BT in any new exchanges for a period of 6 to 9 months. To show that TalkTalk had deployed in 17 exchanges by July 2011 and in 92 by September 2011 (7 and 9 months on from the date of the Market Review Determination) does not seem to me to show that this assessment by Ofcom was erroneous in December 2010, or that it had turned out to be erroneous by the summer and autumn of 2011. At paragraph 3.189 they said that TalkTalk was unlikely to have any exchanges unbundled within the next nine months. As it turned out, TalkTalk was able to unbundle in 17 exchanges sooner than was there said to be likely, though not sooner than seems to have been envisaged as possible in paragraph 3.176. But a statement that earlier deployment is unlikely is not in itself belied by the event of some limited earlier deployment.
130. In what respect, then, was Ofcom's forecast of the way in which competitive conditions in the market would change falsified by what TalkTalk actually did between November 2010 when its announcement was made and July 2011 when the Charge Control Determination was made? I cannot identify any. As it seems to me, what TalkTalk did in that period was reasonably well within the scope of what Ofcom had foreseen. If that is so, I do not perceive a regulatory justification for requiring Ofcom to go through the market review process again by reference to the up to date circumstances, their forecast as to the future having proved to be reasonably accurate.
131. For TalkTalk Mr Pickford submitted that "material change" had the meaning given to it by the Tribunal at its paragraph 96, quoted by Sir Bernard at his paragraph [35] above. He argued that this definition is to be applied by considering what the outcome of the Market Review Determination would be if it were carried out again as at the time of the Charge Control Determination.
132. To the contrary Mr Holmes for Ofcom contended that in comparing the position as it was at the time of the Charge Control Determination with the terms of the Market Review Determination in order to decide whether there had been a material change, the latest position had to be examined by reference not (or not only) to what the factual position had been at the earlier date, but to what Ofcom had forecast in their earlier analysis and determination. He argued that, unless the later position falsified the earlier forecast to a material extent, showing it to be wrong to a significant degree, then there had not been a material change. On the particular facts, he submitted that Ofcom's earlier forecast, in the Market Review Determination, was not falsified by subsequent events up to the date of the Charge Control Determination, to any significant extent or at all.
133. Before I come to a conclusion on this issue, I must make some reference to the judgment of the Tribunal.
134. The Tribunal came to the point now in issue at paragraph 80 of its judgment. At paragraph 96, the Tribunal set out its view as to the circumstances in which a change will amount to a material change, as quoted by Sir Bernard at paragraph [35] above. It had set out some examples of changes which would not be material on this test. It rejected an argument which it said had been advanced by Mr Holmes for Ofcom, which required a consideration of the question whether the change is material to the question of whether or not to set the price control in the form in which Ofcom proposed to set it. As the Tribunal said, if there has been a change in a relevant market which might be material, you cannot tell whether it would affect the charge control decision without having first reviewed the definition of the market and the analysis of its relevant conditions. However, it also observed that whether there had been a material change in any given case depended on how the relevant market was defined in the first place.
135. It therefore turned to the definition of Market 1 in the Market Review Determination. As Sir Bernard has stated, and I have agreed, it had already said (but wrongly) that paragraph 1.19 of the Market Review Determination was not an accurate statement of Ofcom's definition of Market 1: see paragraph 61(b) of the judgment. This was carried forward at paragraph 110. At that point it redefined Ofcom's Market 1 as being: "exchanges where only BT is present or forecast to be present or where, during the period of the market determination, Talk Talk may (at some point in the future) be present".
136. As already noted, the Tribunal's error in rejecting paragraph 1.19 seems to have arisen from having overlooked footnote 6 as explaining the meaning of "forecast". On the basis of that footnote, all plans which were not firm and specific were to be ignored. That will have led to a number of plans being ignored on the part of other Principal Operators in deciding which exchanges were within Market 1. Correspondingly (though after a lot more by way of discussion and reasoning) it also led to TalkTalk's plan being ignored for this purpose. It is no more appropriate to refer to TalkTalk's possible future deployment, in the definition of the market, than it would be to refer to any other Principal Operator which had unspecific and uncommitted plans for such deployment.
137. I also disagree with the way in which the Tribunal put the point at paragraph 115. Ofcom did take TalkTalk's plans into account, in the sense of discussing them at some length, but they did not take them into account in defining Market 1 as such. The question for Ofcom was how they fitted into the framework which had already been formulated in the course of the consultation process. If another Principal Operator had made a similar announcement of further unbundling on a significant scale in unspecified exchanges otherwise within Market 1, there seems to me to be no reason to suppose that Ofcom would not have dealt with that in a similar way, discussing it as appropriate but ultimately treating it as affecting neither the definition of Market 1 nor the forecast of the competitive conditions that might be expected to exist in that market during the review period.
138. At paragraph 106 the Tribunal said that, on the basis of Markets 1 and 2 as defined in paragraph 1.19, there had been a material change. However, since this proceeded on a false basis as to what that definition was, it seems to me that it cannot carry any particular weight.
139. For those reasons, to my regret, I have to say that I find the Tribunal's judgment of less assistance than I would have expected in determining the issue which arises on the appeal.

The correct test for material change

140. In my judgment, Mr Holmes' contention as to how the material change test is to be applied in a case such as this is correct. It may be that there is no issue as between the parties as to the test for the nature of a material change, in the terms stated by the Tribunal at its paragraph 96 (see paragraph [35] above). But what is very much in dispute is how that test is to be applied. I consider that, if the test were to be applied on the basis of seeing how the Market Review Determination would have come out if it had been carried out at the date of the Charge Control Determination and by reference to the facts as they then stood, it would be capable of making a nonsense of the regulatory process. Ofcom could have taken full account of the probabilities as regards future changes (of whatever kind) in its Market Review Determination, and could have been proved to have been accurate in its forecast, but the fact that changes predicted had in fact occurred would mean that a new market review determination had to be carried out. This would be likely to result in unnecessary expense and delay, and therefore to prejudice the exercise by Ofcom of the function of promoting competition in order to protect the interests of consumers of the relevant services.
141. It may be said that, if all that had happened was exactly what had been predicted, the new market review and analysis might not be very time-consuming to carry out. But that cannot be assumed, because it would not be limited to taking account of material changes, that is to say substantial ones. The whole market would need to be reviewed, following a full consultation process as required by the legislation.
142. In the present case it seems to me that the changes that took place by way of deployment by TalkTalk in Market 1 exchanges (no others are relevant for this purpose) were within the scope of what Ofcom had forecast and taken into account in carrying out the market analysis set out in the Market Review Determination. It seems that TalkTalk may have been able to bring its first deployments to completion somewhat faster than Ofcom had envisaged. That depends on the view one takes of the discrepancy between paragraph 3.176, with its reference to TalkTalk not being in a position to exert a practical constraint in any new exchanges for a period of 6 to 9 months, which clearly allows for some deployment within 6 months, on the one hand, and paragraph 3.189 which speaks of TalkTalk as being unlikely to have any exchanges unbundled within the next 9 months. Whatever view one takes, it seems to me that having completed deployment in 17 relevant exchanges within 7 months and 92 within 9 months is not at all far from what Ofcom envisaged and allowed for, and is not outside the scope of Ofcom's forecast, on the basis of which they conducted their analysis of the relevant markets.

Conclusion: no material change had taken place

143. On that basis, therefore, I would hold that, on a proper comparison of the circumstances known (or as they ought to have been known) to Ofcom in June 2011 on the one hand with the terms of the Market Review Determination in December 2010 on the other, there had been no material change in the market by reference to which the Market Review Determination had been made. Accordingly, section 86(1)(b) was satisfied and it was open to Ofcom to proceed without more to impose the SMP conditions consisting of the charge control upon BT in relation to Market 1.
144. I therefore respectfully disagree with my Lord Sir Bernard Rix and would dismiss the appeal, despite the errors on the part of the Tribunal. In my judgment Ofcom made no error of law on the question whether there had been a material change for the purposes of section 86(1)(b). I do not accept his proposition at paragraph [38] that what Ofcom did was to substitute a new definition of Market 1 in the Charge Control Determination, made by reference to BT's market share, for that which they had adopted in the Market Review Determination. It seems to me that this mistakes the process of analysis, carried out by Ofcom in the Charge Control Determination to examine the material change question, for a process of redefining the relevant market. The references to market share are there by way of market analysis, to test whether the extent to which TalkTalk had implemented and developed its plans was such as to make the competitive conditions in Market 1, as defined, materially different, not from what they had actually been in 2010, but from those that had been foreseen and allowed for in the forward-looking analysis carried out as regards the review period in the Market Review Determination.
145. In my view, my Lord's conclusion wrongly overlooks the need to compare the position as at the time of the Charge Control Determination, not (or not only) with the factual position as it was at the time of the Market Review Determination, but with (or also with) the forecasts made in the market analysis in the course of the Market Review Determination. If those forecasts have not been falsified to a material extent by what has happened in the meantime, it seems to me that a material change has not occurred. On the same basis I respectfully disagree with what he says at the end of paragraph [40] and in paragraph [41].
146. I agree with Sir Bernard that, by and large, the observations of the Tribunal are not helpful in resolving the issue before us, though, as I have indicated at paragraph [137] above, I have less difficulty than he does with paragraph 115 of that judgment.

Other arguments

147. Mr Pickford advanced another argument to which I have alluded in passing, at paragraph [94] as regards a comparison with exchanges within Market 2 (under the Market Review Determination) where a Principal Operator had specific and committed plans as at June 2010, to deploy no later than December 2010, but where in fact it had not brought those plans to fulfilment by that date. He argued that this was a separate category of exchange properly analogous to those in which TalkTalk intended to deploy. He submitted that they ought to have been excluded from Market 2 and brought back within Market 1. I cannot detect this point as having been included in the grounds on which TalkTalk appealed to the Tribunal, nor is it in TalkTalk's grounds of appeal to this court for which permission to appeal was granted. Accordingly, this point could not have been argued in the Tribunal (and, so far as I know, it was not so argued) and it is not open to be argued in this court.
148. Mr Pickford also relied, as Sir Bernard has mentioned at paragraph [59] above, on the EU requirements referred to at paragraph [122] above, which can be summarised, at some risk of over-simplification, in the proposition that there should be no greater regulatory intervention than is necessary. He contended that Ofcom did not address the factors relevant in this respect in the Charge Control Determination. He put this point on the basis that Ofcom had adopted a different approach in the Charge Control Determination from that which they had used in the Market Review Determination. Since, for the reasons I have given, it seems to me that they did not do so, it also seems to me to follow that their analysis in the Market Review Determination as to the application of the relevant EU requirements applies just as well to the Charge Control Determination as it does to the Market Review Determination. If there has been no material change, there is no reason to suppose that the circumstances relevant to these requirements have changed, or that the application of the requirements would result in any different outcome, either.
149. For those reasons, I would dismiss this appeal.

Lord Justice McFarlane

150. Given the disagreement that exists between my Lords, Sir Bernard Rix and Sir Timothy Lloyd, it is necessary for me to explain in short terms the conclusion to which I have come on the core issues in this appeal. I am grateful to each of my Lords for the care that they have so obviously taken in explaining the complicated background to these proceedings and for the clarity of language that they have deployed in describing the reasons that lie behind their conflicting conclusions. It is not therefore necessary for me in this judgment to attempt to retrace any of material that has been so effectively already laid out by my Lords.
151. The essential difference that lies between the judgments of each of my Lords is whether, at the time of the Charge Control Determination, Ofcom substituted a new definition of Market 1 for that which had been adopted in the Market Review Determination [per Sir Bernard Rix at paragraphs 34 to 44] or whether the changed factual circumstances that by then existed were no more than the acting out of events that were reasonably within the market forecast made by Ofcom at the time of the Market Review Determination [per Sir Timothy Lloyd at paragraphs 140 to 146].
152. In his analysis, Sir Bernard cites a number of factors arising from the Market Review Determination but he lays particular emphasis upon the fact that the definition of Market 1 at that time related solely to 'exchanges where only BT is present' and in doing so Ofcom made a distinction between exchanges where another PO had 'committed' plans in relation to a specific exchange or merely 'uncommitted plans'. At paragraphs 20 to 23 the relevant passages are set out and the express references to the distinction between committed or uncommitted plans have been italicised. As is plain, by the time of the Charge Control Determination the situation on the ground had changed in that Talk Talk had committed to rolling out LLU-based services in a number of specific Market 1 exchanges. For Sir Bernard, Ofcom's decision to press on with the Charge Control Determination on the basis that its original definition of Market 1 remained valid is untenable [paragraphs 37 and 38] as the list of Market 1 exchanges was no longer confined to exchanges where there was no committed 2^nd PO in addition to BT. He holds that Ofcom were in reality now accepting a revised definition of Market 1 as containing exchanges where, whether or not BT is challenged by another operator's committed unbundling in a specified exchange, BT's market share is projected not to fall below 70% by the end of the review period; in consequence Sir Bernard holds that it was necessary for Ofcom to conduct a fresh Market Review Determination.
153. Whilst I understand this essential reasoning within Sir Bernard Rix's judgment, together with the other supporting points that he makes, and despite the true respect that I have for his wide knowledge and extensive experience of these matters, I prefer the analysis given by Sir Timothy Lloyd in his judgment. Although Ofcom did draw a line in the Market Review Determination between exchanges where a 2^nd PO was 'committed' and those which were 'uncommitted', and Sir Bernard is right to highlight that clear distinction in identifying exchanges at the start of the review period, Ofcom also made it plain that it was taking a forward view, that it anticipated that Talk Talk would choose to unbundle in some of those exchanges during the review period and that 'even if Talk Talk deploys towards the start of the review period BT's market share would be likely to be at least 70 to 80 percent in the exchanges where Talk Talk deploys at the end of the period' (MRD paragraph 3.182). Sir Timothy Lloyd is right to have highlighted paragraphs 3.178 and 3.182 by repeating them [paragraph 91] as they were expressly in Ofcom's contemplation at the time of the Market Review Determination. The principal purpose of the process on which Ofcom were engaged was to look forward over the three year period of the review and to apply, where necessary, price structure to the market in the interest of promoting competition; they were required to execute this task in relation to a market which was bound to change and develop throughout that period. Given that the definition of Market 1 readily contemplated that during the review period some of the 'BT only' exchanges would change to ones in which Talk Talk was active, I agree with Sir Timothy that the fact that, by the Charge Control Determination, some of those changes were already taking place cannot of itself by regarded as a material change for the purposes of section 86 [paragraph 126] and I agree that Ofcom's forecast of the market was not falsified by the events that had taken place on the ground when the Charge Control Determination was made [paragraph 130].
154. I anticipate that I have said sufficient to explain why I prefer the analysis of my Lord, Sir Timothy Lloyd. For the reasons which are fully rehearsed in his judgment, with which I agree, I too would dismiss this appeal.

Note 1   In its original market power determination Ofcom contemplated that its next review would be four years later. In its charge control determination it spoke of a three year review period as having been introduced by EU legislation.    [Back]

Note 2   At the time of the market power determination the review period was seen as extending to four years.    [Back]