3.1 The regulatory regime summarised in paragraphs 2.1-2.4 was devised to maximise competition in circumstances where the Government had decided to limit the number of mobile licensees to two, and where one of these (neither having yet launched their networks) was controlled and backed by the dominant fixed line PTO. Eleven years later the market shares of the two original licensees are roughly equal, while two further networks have been licensed and in their first two years of operation have between them captured nearly 16% of the total mobile telephony market. Moreover, the two new operators are now accounting for nearly half of net new cellular connections (over 42% in the first quarter of 1996). Thus there are now four mobile operators in strong and increasing competition with one another.
3.2 This makes it increasingly anomalous to maintain detailed regulation of the way in which mobile airtime is sold, with Oftel as competition authority effectively deciding the appropriate market structure, and more appropriate market structure, and more approriate to withdraw the from this, as Oftel is progressivley doing in relation to fixed telephony, and relay instead on general provisions to deal with anti-competitive behaviour and abuse of a dominant position.
3.3 Consequently it has become difficult to justify special rules for mobile telephony except to the extent that the objectives of fully effective network competition, the most efficient distribution of airtime and the best deal for the consumer clearly cannot be achieved and maintained without them. In particular, there is no justification for continuing to prescribe and manage the structure of a market by means of artificial rules when the circumstances those rules were designed to meet have ceased to exist, and the mobile market is now very different from what it was in 1985 or even the early 1990s.
3.4 Against this background a number of questions arise which are addressed in the following sections:
(a) Is there any longer a need to retain licence controls regulating the way in which the mobile network operators provide their services to the public, or to impose a particular retailing structure on the industry? Is inter-network competition now strong enough to justify reducing reliance on detailed regulation of the way in which airtime is sold in order to ensure the best deal for the consumer?
(a) Is there any longer a need to retain licence controls regulating the way in which the mobile network operators provide their services to the public, or to impose a particular retailing structure on the industry? Is inter-network competition now strong enough to justify reducing reliance on detailed regulation of the way in which airtime is sold in order to ensure the best deal for the consumer?
(b) in the light of the answers to (a):
3.5 To answer these questions requires a closer look at how competitive the mobile telephony market really is. Which operators enjoy a position of significant market power, and how is market power likely to influence their market strategies and competitive behaviour? What external factors constrain competition? Tied in with any examination of market power will be an assessment of the definition of the relevant market.
3.6 The first question is whether it is justifiable to regard mobile telephony as a separate market from other forms of telecommunications. It is certainly possible to envisage a situation where the fixed and mobile markets converge, and it would then be necessary to look at competition issues as arising across that wider market. However, for the moment mobile telephones offer an important range of functions and facilities which are not available from fixed link telecommunications (ie mobility) or other forms of mobile communications (such as radiopaging or PAMR). Thus mobile telephony is still generally regarded as complementary to other forms of telecommunications rather than as a competitor, and it is this additional service that consumers are prepared to pay for. Thus these products are not close substitutes for mobile telephony, even though to varying degrees they compete with aspects of the service provided by it. In general, therefore, Oftel considers it appropriate to view mobile telephony as a distinct market.
3.7 Oftel views the relevant geographical market for mobile telephony as the national market, ie the UK. The licences for the mobile networks are national in scope and within the UK the geographical location of the use of the services does not generally determine the price. Although international roaming agreements between national operators mean that GSM phones can often be used abroad, competition between service providers is for customers in their own countries rather than at a cross-border level.
3.8 However, the question also arises whether the mobile telephone market should itself be divided into sub-sectors. The functionality of analogue TACS, and digital GSM and DCS 1800 services is broadly similar, and many consumers may make their choice solely on the basis of entry and running costs. These services, however, are by no means entirely substitutable for each other. They have a number of technical differences which other consumers may regard as important. They require different types of handset specific to their technologies, so that switching between these technologies will generally impose a cost on the subscriber. In addition there is currently no provision for non-geographic number portability.
3.9 Speech quality may vary between digital and analogue services and digital services are much more secure against interception and fraud than analogue. International roaming is now very widely available on GSM but is extremely limited on TACS and (for the moment) on DCS 1800. Another factor is the Government's recently announced aim to migrate all users to the digital services by 2005 and to close down TACS services by that date. A number of handset manufacturers have ceased production of analogue handsets.
3.10 These factors, coupled with fast pace of change in the mobile market, has led some to argue that the digital and analogue mobile technologies should each now be viewed as markets in their own right. In the analogue segment, Vodafone and Cellnet are the only two network operators but in the digital segment Vodafone and Orange are each estimated to have 27-29% of total digital subscribers, while MOTO has 23% and Cellnet has 20%.
3.11 The key issue in market definition in this area is the extent to which the price of analogue telephony is a constraint on digital telephony prices and vice versa. The above discussion indicates that analogue and digital telephony are differentiated through factors such as quality of service, security, ability to "roam" abroad etc. However, competition between service providers for customers centres on the buying decision itself and at that point it is not clear that customers make a distinction based on the underlying technology: rational consumers would take into account the whole range of factors. In addition, existing analogue customers are likely to regard the digital services as simply an enhancement or an upgrade of their existing mobile telephony service and therefore to regard them as effective substitutes.
3.12 In assessing the force of competition, particularly as it affects network attitudes towards service provision, Oftel considers that the broad market definition remains at present the most appropriate one. Consumers are primarily interested in buying mobile services, not technologies, and for most of them the prime influences on their choice of network are likely to be entry and running costs, so long as functionality is broadly comparable. It would therefore be wrong to look at market definition solely from a technology perspective.
3.13 Having put forward a definition of the relevant market, the issue of whether any mobile licensee has a degree of market power can be considered. In the context of the mobile market with its limit on the number of competing operators the issue of market power relates to the ability of the incumbent operators to impede or prevent the new entrants from competing effectively. An analysis of market power will take into account a range of structural and behavioural factors. The important factors that would need to be considered are the structure of a given market (ie the market shares of firms, the degree of concentration and how those market shares have changed over time); the nature of competition in that market (eg the extent of product differentiation) and the significance of barriers to entry. The ability to set prices, and otherwise act, without reference to competitors is also an important indicator.
3.14 The first consideration bearing on competition is the limited availability of radio spectrum, which constrains the number of licences for mobile networks it is possible to grant. The number of licensees is currently four, and seems likely to remain so for some years to come. However, even given these constraints, there is no reason why a substantial degree of network competition should not be achievable in a market with four players. Indeed, as already noted, there is evidence that competition between the mobile operators is now becoming quite intense. This points towards taking steps to align the fixed and mobile regulatory regimes earlier rather than later, retaining in the mobile network licences only such additional provisions as are necessary to ensure that the "special rights" deriving from privileged access to scarce spectrum are not abused.
3.14 The first consideration bearing on competition is the limited availability of radio spectrum, which constrains the number of licences for mobile networks it is possible to grant. The number of licensees is currently four, and seems likely to remain so for some years to come. However, even given these constraints, there is no reason why a substantial degree of network competition should not be achievable in a market with four players. Indeed, as already noted, there is evidence that competition between the mobile operators is now becoming quite intense. This points towards taking steps to align the fixed and mobile regulatory regimes earlier rather than later, retaining in the mobile network licences only such additional provisions as are necessary to ensure that the "special rights" deriving from privileged access to scarce spectrum are not abused.
[Note from paragraph 3.14. A recent Government Consultative Document on future allocation of mobile phone spectrum (Mobile phone Spectrum over the next Decade, issued by the Department of Trade and Industry on 23 February 1996) envisages that all remaining available spectrum in the 900 and 1800 MHZ bands will be reserved for the use of the existing four operators and explicitly rules out any prospect of new licences until spectrum for "third generation" (FPLMTS) systems becomes available in the early years of the next century.]
3.15 What should such provisions amount to? The need arising from these special rights is to ensure the fullest and widest possible distribution of mobile airtime, in order to make the best possible use of the allocated spectrum. Incentives need to be in place to ensure that operators compete on overall price levels as well as by niche tariffing. It is desirable to promote the achievement of these objectives through competition and this supports the case for retaining for the time being the obligation to supply at the wholesale level, the prohibition of undue preference and undue discrimination, and other provisions designed to ensure that independent retailers can obtain airtime on the same footing as a TSP or DSB.
3.16 The next considerations are structural. The mobile market has a number of important structural characteristics which affect the way in which the network operators will be able to compete with each other. The two incumbent operators not only have a large installed base of customers but they also have strong vertical links through their own TSPs who control the majority of customers. The substantial base of analogue customers also means that the incumbents are in a position to use some of the profits from this installed base to fund the acquisition of new customers and they also have a base of customers to migrate to their digital services. This means that they are in a position to influence the competition for new and migrating customers. These factors would suggest that the incumbent operators, Cellnet and Vodafone, currently have a degree of market power compared to both MOTO and Orange in the mobile market.
3.17 In the mobile context it can be argued that, while the freedom to cross-subsidise and offer other forms of preferential treatment might involve increased integration into service provision and a reduction in the number of independent service providers, it would not necessarily lead to the demise of the latter. There is some evidence, if not conclusive, that the economies of scale available in mobile service provision are considerably less than those arising in network operation, and that the potential cost savings from network integration into service provision are not very significant. Network operators can be expected to choose the distribution structure best calculated to attract and retain customers, and it would be logical for them to continue to sell through independent service providers so long as they were efficient, innovative and entrepreneurial, and better at controlling costs than TSPs. The financial results of service provision businesses show a very mixed pattern. Both the tied and the independent sector offer profitable and unprofitable examples. Service provision businesses linked to the networks have by no means always been as successful as independents in controlling costs and producing a profit.
3.18 On the other hand, the new entrants MOTO and Orange are still relatively small in relation to the mobile market as a whole. Any tactics which successfully impede the access of these companies to the established retail channels may therefore be attractive to the dominant operators, who have an interest in keeping the new entrants weak. Given a continuing imperfect level of consumer knowledge about mobile telephony and the key role of service providers and the high street phone retailers in guiding customers in their initial choice of network, a strategy of network integration into service provision, and the building of the strongest possible links with the high street multiples, may be seen by the dominant networks as a good way to achieve this objective. Revenue from existing customers, tied to Cellnet and Vodafone by factors inhibiting active customer choice, such as lack of number portability, coverage requirements or long-term contracts, can be used to finance the acquisition of new customers at low or even negative profits.
3.19 Oftel is committed to encouraging a more innovative and competitive market in services offered over telecommunication networks, and set out policies for achieving this in relation to fixed networks in a recent Consultative Document. Taking all this into account, it does seem that there are still some benefits to competition to be gained from retaining for the time being a sufficient level of regulation to safeguard the access of service providers to airtime on fair terms. Oftel therefore believes that the broad framework of network licence conditions governing the relationship between operators and service providers should continue to apply for the time being. However, the level of competition, and the corresponding impact on the market, is clearly moving towards the point at which the detailed framework of rules could be dismantled altogether, and this point may not be very far away. The mobile regulatory regime should therefore be kept under review with the objective of aligning it with the general framework of telecoms regulation as soon as competitive conditions permit.
3.20 There may, however, be valid arguments for modifying the detail of the controls and the way in which they are applied, to take account of current market conditions. The question also arises whether, in the light of the considerations set out above, it is appropriate that these rules should continue to apply with equal force to all the mobile network operators and their TSPs, or whether coverage should be restricted to networks having significant market power. At present the fair trading rules supporting independent service provision apply to all four mobile network operators, and are implemented mainly by applying the Oftel formula and monitoring arrangements to the TSPs, including direct sales organisations ("DSOs"), associated with all of them: Cellnet and Vodafone with their market shares in mobile telecommunications (measured by subscribers) of over 40% each; and MOTO and Orange with less than 10% each. (Application of the rules to Orange was scheduled to begin in May 1996.)
3.21 To discuss these issues, it is necessary first to examine the motives which might lead any network to cross-subsidise its DSO or TSPs.
3.22 As already noted, when cellular networks began operations in the United Kingdom in the mid-1980s, the Government took the view that, while the number of networks to be licensed would be limited by spectrum scarcity and other considerations, competition could be injected into the sector by encouraging a number of service providers. For this reason, as already noted, Cellnet and Vodafone were obliged to sell airtime through service providers, and forbidden to discriminate between their own tied service providers and independents.
3.23 This approach reflected a view that service provision is a potentially competitive activity capable of sustaining a reasonable number of separate firms. The relative ease of entry into the business would prevent any firm from exercising market power and earning above normal profits for a sustained period.
3.24 If service provision could be characterised in this way, then there is no obvious means by which a network operator, or any other firm, could increase its profits by cross-subsidising a service provider, driving out competitors and raising prices. Subsequent entry by new firms would quickly eliminate any excess returns, and make the strategy unprofitable.
3.25 Yet evidence, supported by certain of the monitoring returns, has been presented to Oftel of continued cross-subsidies to TSPs. There is a number of possible explanations for this.
3.26 The first is that losses are incurred in the start-up period, but eliminated when entrants have accumulated a reasonable market share. In recognition of this, Oftel has allowed MOTO and Orange a two-year derogation from application of the Oftel formula, which has now expired. However, this leaves open the question whether the length of the derogation for new operators is appropriate.
3.27 The second is that TSPs, like other service providers, are susceptible to losses either as a result of management miscalculations or of temporary excessive price cutting in the service provider sector. If this explanation were the correct one, then regulatory intervention would not be required.
3.28 A third possibility is that entry into service provision is not so easy as described above. It might therefore be possible to establish some market power in the sector. In such circumstances, network operators in particular would have an incentive to expand their market share in service provision, as a combination of market power at the network level in service provision would give an integrated firm greater control over the overall market: the argument that an oligopolist in one activity can extend its control by gaining a degree of market power in a vertically related sector is widely accepted by industrial economists.
3.29 Fourthly, network operators may be prepared to cross-subsidise TSPs in order to retain their dominance at network level, particularly at times when it is being threatened by new entrants. This process would work as follows. By cross-subsidising TSPs and retailers, a network can steer new customers in its own direction. It may be worthwhile to do this, even at the expense of the TSP making a loss on new customers, if network entrants (also competing for the same customers) are thereby weakened. At the same time, existing networks are able to make profits from their existing customers, who would incur costs in switching to a new supplier. On this view, a cross-subsidy to new customers gained by a TSP is a form of "incremental predation": in order to weaken competitors. Losses are taken (at any rate by the TSP) on new business, but these losses are counterbalanced by profits from existing subscribers.
3.30 Fifthly, network operators may be prepared to cross-subsidise TSPs in order to acquire existing independent service providers at a cheaper price. This process would work because as independents are subject to a margin squeeze they become less profitable and therefore of less value. The investments made by service providers can then be acquired by network operators for less. (It should be noted that this process would not result in the service provider's customers' becoming less profitable for the network - indeed, they would tend to become more profitable). Once the service provider's business is acquired the network is in a position, if it wishes, to set about switching any customers on a competitor's network over to its own (for example, by persuading the now tied service provider to raise retail prices for calls on the competitor's network and/or to reduce them on its own network). This action also has an additional beneficial effect (for the established networks) because it makes it more difficult for new independent service providers to become established.
3.31 According to the third to fifth explanations above, it would only make sense for a network to cross-subsidise a service provider either if it enjoyed significant market power at the network level, which could be enhanced by some degree of control over the service provider market, or if it were cross-subsidising new retail customers to weaken entrants at the network level.
3.32 In none of these cases could a network operator without significant market power gain advantage from cross-subsidising retail sales. Such advantages would be confined to network operators in a dominant position. Thus, if either of these explanations is correct, use of the Oftel formula could be confined to TSPs owned by or linked to network operators with a position of significant market power. At present, the practical effect of applying this approach, which Oftel believes to be the right one, would be to subject only Cellnet and Vodafone and their DSOs and TSPs to the Oftel formula, while at present releasing MOTO and Orange (and their direct sales organisations) from compliance with it. The coverage of the formula would, however, change as network operators achieved or ceased to enjoy significant market power.
3.33 The complaint by Talkland International and others primarily concerns issues of unfair cross-subsidy. However, the mobile licences contain a number of other conditions aimed at protecting the position of service providers. These include:
3.34 These conditions are important in maintaining access for service providers to mobile services and are designed to ensure a fair competitive environment. Oftel believes that they should continue to apply, at any rate until network competition is complete. However, the achievement of effective network competition may be inhibited if network operators cannot exercise some control over how their product is presented in the marketplace. If they are to be able to differentiate their products and promote them effectively to consumers, they need to be free of regulatory constraints on "through-branding", ie the ability to develop a consistent retail proposition and ensure its uniform presentation to the public. The new mobile networks have been able to do this during the period up to the present during which their services have been marketed only through a DSB, but (for largely historical reasons) it is more difficult under the existing regime for the established network operators to do so.
3.35 There is in fact no specific provision in the mobile licences preventing the mobile network operators from including in the contracts they offer to service providers contractual conditions requiring them to adhere to a particular marketing strategy. The operators are permitted under their licences to refuse to provide services to any service provider who refuses to sign a contract in circumstances where the operator has published standard terms and conditions. Nor is there any licence requirement on the operators to have such terms and conditions approved by the regulator. However, service providers are equally at liberty to refuse to sign contracts to which they object, so that a network seeking to impose wholly unacceptable terms could find itself faced with losing the custom of all its independent resellers. The extent of such a risk would depend on how reliant the network is on independent channels of distribution in relation to tied channels and direct selling. One important constraint on a network operator including contractual provisions to control brand promotion, where he judges that he has the commercial scope to do so, is the danger that the regulator will, on investigation of a complaint, find such provisions to be in conflict with the obligation to supply and strike them down. Oftel has up to now generally taken the view that contractual conditions which merely act as incentives to promote the operator's brand are permissible, but that conditions threatening cessation of services unless specific targets are met, or specific strategies complied with, may not be. The time may have come to allow the commercial process alone to determine what is agreed by contract.
3.36 In the light of the above discussion Oftel has reached the following conclusions on how the licence controls supporting a fair trading regime for mobile service providers should be applied:
Oftel believes that the following licence conditions designed to ensure fair trading and safeguard the position of service providers should for the time being continue to apply equally to all network operators, including MOTO and Orange:
- the obligation to provide services to service providers for resale on written request;
- the obligation to connect individual customers of a service provider to the network and keep them connected;
- the prohibition of undue preference and undue discrimination;
- the obligation to publish the charges, terms and conditions on which the Licensee provides mobile services to service providers (ie its wholesale prices), and not depart from such published charges, terms and conditions;
- the requirement for accounting separation.
The inclusion of contractual provisions in the operator's contracts with service providers which require adherence to strategies (including tariffing strategies) designed to promote the operator's brand should not be regarded by the regulator as in conflict with the obligation to provide services. Such provisions, however, should not extend to control of retail prices.
Oftel believes that mobile network operators with significant market power in the overall mobile market should continue to be prevented from unfairly cross-subsidising any direct retailing business which they may conduct (ie a "Direct Business" as defined in the current mobile licences), or any similar activities where conducted by a member of their Licensee's Groups.
Oftel has concluded that where an operator has no significant market power, it will not consider cross-subsidies to be unfair. In practice this conclusion means that the controls should continue to apply to Cellnet and Vodafone Ltd, but not to MOTO or Orange until such time as they achieve a position of significant market power.
Oftel has not yet taken a final view as to whether amendments to the licences of MOTO and Orange are necessary to effect this change. But in policy terms the concept of "unfair" in relation to cross-subsidy is viewed as only arising when the business receiving the subsidy is retailing the services provided by a network with significant market power and is owned or controlled by that network or its parent Group.
The regulatory regime for mobile service provision should be kept under review, with the general objective of progressively withdrawing, as network competition develops further, from detailed control of how mobile services are distributed to the public and relying instead on general provisions for the control of anti-competitive behaviour and abuse of a dominant position.
Oftel invites views on these conclusions. It would also welcome views on:
(a) the point at which MOTO and Orange (or any subsequent new market entrant) should be deemed to have achieved a sufficient position of market power for any cross-subsidy to be deemed unfair, and, correspondingly,
(b) the point at which any reduction in the market power of Cellnet and Vodafone might be considered sufficient to justify no longer deeming any cross-subsidy by them to be unfair;
Any such views will need to take account of Oftel's approach that markets cannot be defined on the basis of technologies.
3.37 Under the present arrangements for monitoring the Oftel formula MOTO became subject to the financial criteria from October 1995 and Orange will do so from May 1996.
In the case of both MOTO and Orange the Director intends to suspend the application of the Oftel formula used to implement the unfair cross-subsidy condition until the period for responses to the present Consultative Document has ended and a decision has been taken in the light of them.
Both operators, however, will be expected to continue to provide quarterly returns to Oftel to enable any cross-subsidies to their TSPs to be monitored.