Response by SPIG
To the consultative document issued by the Director General of
Telecommunications
June 1999
The Service Provider Interest Group, SPIG, represents the interests of businesses in the service industries that provide a significant element of telecommunications including content, data, Internet and mobile.
Many SPIG members are currently independent mobile service providers, or seek to become service providers (SPs), and wish to become MVNOs or SPs for MVNOs. We welcome Oftel's initiative in setting out the Inquiry and the opportunity to respond.
SPIG has held lengthy discussions with Oftel on the definition of "Service Provider". In essence, Service Providers interface with customers for the provision of airtime and other services. SPs may fall anywhere within the continuum of selling
While Oftel is seeking to set out and differentiate between the 'tiers' of SPs, in reality individual businesses may be in one, some or all categories.
SPIG believes that there is a need to regulate to avoid the anti-competitive practices that dominance in a market can lead to for
SPIG welcomes this new approach to competition in the mobile sector. The problems of access to radio spectrum for civilian needs are well understood and the timescales for release of new spectrum for telephony are not matching the market place demand. In addition the increasing public concern about the siting of radio masts will contribute to the difficulties of rolling out more radio networks. Maximising the use of existing radio infrastructure has to be in the public interest.
Although the concept of MVNOs is new, a survey of SPIG members has shown significant interest in taking the concept forward. With new uncertainty about the UMTS timetable, early introduction of MVNOs onto the GSM networks would give an additional layer of competition into the mobile market. This would benefit consumers by leading to reduced prices, diversity of services and opportunities for converged product.
Once established for the second-generation services the concept of MVNOs could be extended to third generation and pamr services.
SPIG is in agreement with many of the issues raised in the response by Cellcom Ltd, a SPIG member company.
SPIG believes that there is enough latent demand for MVNOs for Oftel to move forward to put greater definition into the regulatory structure, such as definition of an MVNO, the legal framework and regulated basis of charging. This would set a framework for greater competition in the supply of GSM services, and could then be extended to Tetra pamr and UMTS.
Response to the June 1999 Consultation Document by Cellcom Limited.
July 1999
1. Cellcom is an independent service provider ("ISP") and a licensed fixed operator in the UK. We have interests in mobile service provider businesses in France and South Africa. Cellcom was incorporated in 1984 and began trading as an equipment distributor and service provider when Cellnet launched its network in January 1985.
2. Today Cellcom is a service provider on the Cellnet and Vodafone networks, an Annex 2 operator providing national and international telephony services using indirect access on the BT network and, as we are excluded from service provision, a distributor for the direct services businesses of Orange and One 2 One.
3. In the three countries where we have interests we provide telecommunication services to some 400,000 customers.
4. We are particularly interested in the potential that Mobile Virtual Network Operators ("MVNO") could have in providing an increased choice of services to customers which, through increasing competition at the network level, could lead to a reduction in the prices paid for mobile services. We further believe that if properly implemented MVNOs will provide converged fixed and mobile services together with value added services currently not available to mobile users.
5. We welcome the opportunity to comment on Oftel’s Consultation Document.
6. Oftel takes the position that it would prefer that "commercial agreements" were entered into by Mobile Network Operators ("MNO") and MVNOs presumably in a similar form to the bilateral roaming agreements between MNOs in different countries. We are concerned that MNOs will not voluntarily negotiate agreements with potential MVNOs. We base our concern on experience and the fact that we are not aware of any precedence for such voluntary negotiation, indeed in our experience MNOs are unwilling to discuss providing services to third parties where the provision of such services will compete with their own retail services. Here we cite the functionality for pre-pay services as a present day example of unwillingness to provide. Consequently for MVNOs to become a reality Oftel/DTI will have to intervene in negotiations or alternatively, subject to legal powers, to mandate the provision of services to MVNOs either through licensing or by some other means.
7. We remain concerned as regards Oftel's views on retail minus pricing for MVNO functionality. We believe that it will be impossible to set and maintain such prices where the services being bought by MVNOs are, in essence, element based conveyance. However of greater concern is that without further guidance as to pricing it is simply impossible to construct a business case for a MVNO. This is bound to lead to uncertainty and reduce the level of interest of potential MVNOs. The Director General's Statement dated July 1999 following his review of the mobile market could have been more helpful in this respect.
8. The level of unbundling of MNO provided services and network functionality available to MVNOs will be key to their success or otherwise. The consultation document rightly does not go into the detail of network functionality and interfaces. We believe that issues of security and network integrity are likely to be a major concern for MNOs in their dealings with MVNOs especially as they are potentially unlicensed and unregulated. In this respect MVNOs are unlike overseas operators for whom end to end services are provided to their roaming customers. Given the potential different forms of access for MVNOs envisaged in the consultation document we are concerned that unnecessary and expensive technical barriers will be erected on the pretext of protecting network integrity. Even worse would be that functionality was simply not extended to MVNOs by MNOs.
9. Clearly MVNOs will offer similar, competing, voice telephony and unified messaging services as are offered by the MNOs to their customers. Given the appropriate inputs there is no reason why MVNOs should not compete on price with the MNOs.
10. In addition the MVNOs will be in a position to provide transparent services across several fixed and mobile networks enabling "find me" and "one number" services to be developed. With appropriate agreements in place MVNOs will be able to switch their customers between different mobile networks to overcome, amongst other things, differences in geographic coverage between networks.
11. MVNOs may compete in the provision of non-voice services ranging from SMS, cell broadcast and telemetry services through to bulk data transmission and information services. Radio access for MVNOs customers should not exclude access to GPRS and other faster data services to be developed in the future.
12. MVNOs would increase choice. We have argued elsewhere 1 that with the continuing vertical integration and the consolidation of end to end distribution under the direct control of the MNOs in a four firm oligopoly there has been a reduction in the choice of services offered to customers at the point of sale. The advent of MVNOs with their own distribution will serve to redress the balance and to widen the choice available to customers.
13. It seems to us that MVNO status will be sought by all those existing service providers 2 , some mobile independent service providers and other organisations such as those with a strong brand name wishing to expand their portfolio of services to mobile. All of whom are presently constrained by the mobile access bottleneck.
14. As MVNOs these organisations will require substantially different levels of access to the mobile networks. As presented in more detail in the consultation document this access will range between the extremes of:
15. The elements of the mobile network which would be needed by a MVNO fall into four distinct categories:
16. Subject to satisfactory input prices Cellcom would seek to exploit the market opportunities if access to the mobile networks as a MVNO is made available.
17. If MVNOs are authorised then the capacity of the two digits MNC within the 15 digit IMSI will soon become exhausted. It may be appropriate for MVNOs to share one or more MNCs. Given that the MSIN capacity of the IMSI is unlikely to be exhausted by a single MVNO then 3 or more of the higher digits of the MSIN could be used to uniquely identify a particular MVNO sharing a common MNC.
18. Such a change may be contrary to the GSM Specification which would then need to be amended. Indeed, the use of such codes might not be seen as unique by overseas operators who identify roaming users solely by MCC and MNC.
19. We are of the view for reasons previously stated that MNOs will not enter into commercial negotiation with the generality of potential MVNOs voluntarily. If such negotiation does take place it will be on a discriminatory basis with an MNO choosing the potential MVNOs with whom it wishes to negotiate. There are parallels with the negotiation of the supply of unbranded airtime to mobile service providers where with no obligation to supply a MNO has set arbitrary minimum volume levels designed to dissuade applicants.
20. We believe that there will be no alternative to regulatory intervention if would be MVNOs are to have access to mobile networks.
21. Oftel has recently 3 decided that the mobile market in the UK is a separate market which is not yet effectively competitive and that there remains an absolute barrier to entry at the network provision level. MVNOs could therefore bring benefits by increasing the competition in the provision of services by providing services which compete directly with the services provided by the four MNOs. This would have the effect of widening the number of operators in mobile as customers would not know, or need to know, which radio infrastructure was being used for the provision of any call by an MVNO. Providing that the MVNOs were not artificially constrained by retail minus prices as their inputs then some of the barriers to effective competition in mobile may be removed.
22. The question arises as to whether MVNOs would be prepared to invest in their own systems without mandated access to one or more mobile networks and some certainty in time that such access would continue to be available even if, in the Director General's opinion, the mobile market has become effectively competitive. The investments to be made by prospective MVNOs are at the "minimum form" level are substantial and not materially different from the investment made by MNOs except for the investment in base station sites and radio infrastructure. MVNOs will incur considerable marketing expenditure consistent with that of a new entrant MNO. With no practical examples of Oftel's proposed retail minus pricing regime it will be impossible to determine the returns which might be earned from such investments. This coupled with a lack of regulatory certainty is bound to deter potential investors in MVNOs.
23. We agree that switching costs for MVNOs customers will be reduced however this can only be the case if access is granted for MVNOs to two or more mobile networks. We believe that access for MVNOs should be provided by all MNOs. If access is to be given to two or more networks then Oftel will need to rethink its basis for mandating obligations on network operators based on market power triggers.
24. A component within the switching cost of a mobile customer today is the handset subsidy that would have applied on joining a network. This has historically been recouped by a minimum term contract, typically 12 months, between the customer and the relevant service provider or by SIM locking or proprietary software in the case of pre-pay handsets. These factors all make switching more difficult and expensive for the customer and give scope for anti-competitive behaviour which should be prohibited in relation to MVNOs.
25. Oftel does not attempt to define MVNOs. We believe that at some point it will be necessary to provide a definition. For example, if access for MVNOs to one or more mobile networks were to be mandated then, without a suitable definition of an MVNO, it would be possible for an MNO to become an MVNO on another network. Effectively providing national roaming.
26. Oftel does not provide a clear view of the legal framework it is considering nor does it give any indication as to what conclusions it might reach or by when. An appropriate legal and regulatory framework is an essential pre-requisite for any organisation considering setting up as a MVNO. The lack of a suitable framework across the EU is likely to make prospective MVNOs extremely wary especially in the light of the Sense experience.
27. We note Oftel's initial view that the logical principle for charging for access for MVNOs would be retail minus. We disagree.
28. We submitted detailed practical arguments against retail minus pricing in our response to the Oftel consultation on indirect access to the mobile networks in February 1999. For convenience we include these arguments as Annex 1 to this response.
29. In his Statement 4 the Director General indicated that he had considered all the responses including the detailed economic arguments put to him during the indirect access consultation by experts acting on behalf of interested parties, notably the mobile operators. He concluded that retail minus was the appropriate basis for charging for indirect access for mobile networks. Apart from retail minus charging as proposed for mobile indirect access the present consultation cites its use for the "Calls and Access" service provider product from BT and its application for 3G to 2G roaming in three or four years time. The only application today is therefore "Calls and Access" where from our own calculations and from anecdotal evidence we know that the proposition is not viable.
30. We return to our arguments in favour of elemental charging based on cost and a reasonable rate of return 5 as being the only practicable method of charging MVNOs for access to MNO functionality. This would be advantageous in assuring MNOs as regards the return on their investments and encouraging potential MVNOs to make their own not insubstantial investments to enter the market and increase competition.
31. We believe that MVNOs will have to be defined at some stage. However we think that except for the use of its own SIM a MVNO is indistinguishable from an interconnecting operator. At every level of access, other than the level where the MVNO appears to a network as the equivalent of an overseas network whose customers roam to that network, there would be a form of interconnection between the MNO and the MVNO that is included in the definition of interconnection given in Article 2.1(a) of the ICD. We believe that IN management plane functionality including access to databases is also within the scope of the interconnection definition and that this is an accepted view from DGXIII. The requirement from a regulatory perspective for the MNO to accept the MVNOs SIM card remains a problem.
Is there a basis for requiring MNOs to supply MVNO services
32. We say subject to legal and regulatory constraints that there are good competitive reasons for mandating MVNO access. At a stroke MVNO could remove the absolute barrier to entry arising from the finite available radio spectrum. All those MNOs that have stewardship of spectrum should be required to provide access on non-discriminatory terms to MVNOs. This would widen the distribution of quasi network services and provide price competition and choice at the retail level.
1. In our view it is wrong to assume that indirect access ("IA") provided at cost plus prices will enable service providers to grossly undercut the mobile operators’ competitive retail prices and attract significant numbers of customers. The issue at present is that the retail prices for calls are held, by tacit collusion amongst the operators, substantially above the competitive price. There is no precedent in the fixed network for IA operators forcing BT, CWC, any cable operators or other local loop operators to materially reduce their prices as a result of cost plus IA prices from BT.
2. It seems to us that in considering the impact of IA at cost plus prices Oftel has failed to take account of the costs to IA operators in making a real impact on the mobile market. Especially since two of the operators are deemed to have market power and the other two have several million customers between them, powerful brands and tied distribution channels. It appears to us that an IA operator would have to invest significant sums in marketing in order to create a brand and customer proposition that would attract any relevant numbers of the mobile operators’ existing subscribers. When we say relevant numbers here we mean numbers that would cause the operators to reduce their retail prices below the competitive level.
3. We fail to understand Oftel’s statement attributed to "potential IA operators" in that "They [the potential IA operators] expect these new services to be self-sustaining and capable of holding there own in the market place once they have been allowed to develop, without the needing a permanent prop from the arbitrage opportunity." We do not regard a fair margin that is the difference between cost plus wholesale prices and competitive retail prices as arbitrage margin or a prop to be taken away at sometime in the future.
4. Arbitrage infers that the IA operator simply pockets the difference between a buying and a selling price, incurs no cost and adds no value. This is patently not the case here. At the very least IA operators provide credit, billing and customer services all included in a service surround and incur the costs of marketing and distribution to attract the customer in the first place. An IA operator, by definition, has to have the use of a least one telephony switch and has to maintain an interconnection with and transmission between at least one other fixed network operator to deliver traffic to the called party. The IA operators costs and investments are pro-rata no different from those of the mobile operator. The significant difference is that the mobile operator is not reliant on making a return on outgoing calls alone but also has the benefits of incoming call revenue for terminating calls on mobile handsets and the monthly subscription paid by the customer for access to the mobile network.
5. With a reasonable cost of collecting calls from the mobile networks IA operators would be in a position to develop services that compete with value added services embedded in the mobile operators network, for example store and forward services, messaging services and gateway services. In addition they would have the opportunity to develop new services spreading across more than one network for example convergent fixed and mobile services for closed user groups where the customers use of infrastructure is transparent.
6. We note Oftel’s concern about the future investment in network infrastructure when IA is mandated at cost plus prices, we do not share this concern. In particular we believe it is wrong for Oftel to anticipate and indeed to speculate on the effect mandatory IA on the existing mobile networks may have on those making decisions on possible investment in new systems and those potentially bidding for spectrum for 3rd generation networks. We do not believe such speculation is within Oftel’s remit, especially since 3rd generation network services are some years away from competing for customers. The development of competition in the 2nd generation mobile market and implementing the provisions of the ICD are today’s issues.
Is indirect access a guaranteed or the best way of reducing retail prices
7. IA as an instrument solely to reduce prices is not what we would advocate and we agree with Oftel that customer prices can be reduced by regulatory means, for example a price cap. However, we say that IA at cost plus prices, in addition to providing for market based competition in call prices, has several other benefits in the mobile market. IA will: -
8. Whilst there is no guarantee that IA will deliver choice and innovative new services to consumers there is every reason to believe that it can. Without IA these services and the consumer choice definitely will not be delivered.
9. A competitive market is driven by customer demand. It is clear to us that the oligopolistic nature of the mobile market ensures that the market is not demand driven, the customer takes what the operator chooses to offer, at the price it is offered, and has little or no perception of alternative prices or services that are possible. A situation that would be remedied through IA operators providing a wide range of new innovative services. Some of these services will search out latent customer demand, others will not survive.
Effects of indirect access at cost plus prices on the mobile operators 10. All of the mobile operators are large businesses with, or with the immediate potential to have, high stock market valuations. Two of the operators make large profits, the other two have yet to reach profitability. However, in all cases the respective shareholders are free to deal with their investments based on the high valuations.
11. Oftel states that IA at cost plus prices could delay the newer networks' profitability for several years. We would say if the newer entrants chose, for example, to compete less vigorously for market share they could achieve profitability sooner.
12. In its analysis of the impact that IA will have on the market we believe that Oftel has over estimated the effects. Any new entrant IA operator would have to compete against the established brands of the operators and against their strong distribution channels much of which the operators themselves control.
13. As regards 3G investors, we agree that they should know in the preparatory period for the spectrum auction whether or not IA will be mandatory for certain operators. This factor precludes one of Oftel's options which is to delay its decisions on IA and/or price control whilst it assesses the results of market developments.
14. In summary we do not share Oftel's concern that IA, mandated at cost plus prices, will be material in determining operators future investment plans in their networks or in new entrants decisions to invest in 3G networks.
What is a retail price? 15. Retail prices are prices for bundles of things, bundles and prices are interdependent and different retail prices are chosen to fulfil different market needs. In telecommunications for example peak and off-peak charges are set to discourage or penalise usage when the network is busy and users have a need to make calls and to encourage and potentially subsidise usage when the network is idle. A seller makes an overall return on the combination of all its retail prices for the things it sells, each of the separate things making a different return or potentially no return at all. In the case of mobile telephony services a customer pays for usage on a three part tariff that comprises: -
Retail minus for indirect access operators will not work 16. Each of the mobile operators retail prices are determined by their services provision businesses (or should be) for the bundles of services they choose to offer to customers. It follows that such prices will be inappropriate for different or alternative bundles of innovative services that an IA operator may choose to offer.
17. A retail minus charging mechanism cannot take account of an efficient IA operator who, for instance, interconnects at more than one point with a mobile operator or whose customers’ usage is confined to a locality close to the point of interconnect. In both of these cases the IA operator causes the mobile operator to incur less cost in delivering calls to the point of interconnect. Issues like this can only be dealt with through element based charging which has absolutely no relationship to retail prices.
18. As discussed above mobile operators’ retail prices for call services are themselves interdependent and dependent on other things, incoming call revenue and customer acquisition costs for example. These other things are not relevant for IA operators. Mobile operators are by definition permitted to move costs between different parts of their retail prices. If a retail minus approach was taken to pricing calls for IA operators there is considerable scope for a mobile operator to increase its retained margin on those call types provided to IA operators and thus discriminate against them.
19. IA operators would also be susceptible to mobile operators predatory pricing particular services, international calls for example, that would otherwise be an attractive area for competing services.
20. In Oftel’s view of retail minus pricing for IA operators the mobile operator retains all of the margin it would have made on a call had the call not been carried by the IA operator. The IA operator’s margin coming only from the savings it can make on the mobile operator's avoidable costs. This is clearly inequitable and it is unlikely that even an efficient IA operator would have a viable business in these circumstances.
21. Oftel has demonstrated that it is unable to enforce measures designed to prevent anti-competitive practices by mobile operators.6 What confidence would IA operators have that any retail minus scheme would be regulated in a timely fashion if the mobile operators take any of the actions described above.
22. In a retail minus pricing scenario for IA operators we consider that: -
Who will seek MVNO status and what will they need?
We are of the view that providing such a wide range of facilities will be inefficient, impracticable and expensive for the MNO and hence for the generality of MVNOs. We would therefore favour a single proposition that provided for the greatest unbundling of network functionality as envisaged in (b) above. We think that as a minimum a MVNO should provide its own HLR and authentication centre.
SIM cards and Mobile Network Codes ("MNC")
Regulatory issues
Economic assessment
Legal considerations
Basis of charges for MVNOs
Answers to Oftel's questions not covered elsewhere
Definition of MVNOs
ANNEX 1
E. Prices for indirect access
Cost plus charges for use of the mobile networks
At present we do not see any other means of providing all of these competitive benefits at the same time as allowing market forces to determine competitive prices.
F. Alternative pricing methods for indirect access
In addition to the three part tariff the mobile operator also receives revenue for terminating incoming calls which may be used to defray costs and lead to higher profits being made on outgoing calls. It is significant that not one of these charges has any direct relationship with the cost of actually providing it. It is sufficient that the consolidated margin across all of the retail prices covers the costs and gives a return acceptable to the operator's shareholders.
Footnotes
Downloaded 16 Jan 2002