Chapter 1     Introduction

Chapter 2     Oftel’s conclusions and the way forward

Chapter 3     Reviewing the policy

Chapter 4     The service to be provided

Chapter 5     Guidelines for operators

Chapter 6     Pricing

Chapter 7     Co-location

Chapter 8     Implementation process

Chapter 9     Oftel’s approach to BT’s ADSL roll-out

Chapter 10   Other Issues

Annex A       Responses to July consultation

Annex B       Technical explanatory document

Annex C       Option 2 Legal analysis

Annex D       Option 2 Cost Benefit Analysis

Annex E        Option 2 Charge for co-location

S1. This policy statement sets out Oftel’s decision to require BT to make available its local loops to other operators and the timetable on which this requirement will be introduced. Oftel has concluded that this action is necessary to introduce competition into the provision of higher bandwidth access. This is not the end, but the beginning of a process to facilitate the delivery of high-speed information age services to consumers. First BT, Oftel and the industry will need to work together to ensure the successful implementation of this policy. Following its introduction, Oftel is committed to undertaking regular reviews to ensure that the regulatory framework encourages the development of a fully competitive market in higher bandwidth to the benefit of consumers.

S2. High speed, high quality electronic information is essential to the development of the information society in the UK. Always-on unmetered Internet access, interactive home shopping, and video-on-demand are just some of the services that will transform the way we access information and communicate. A variety of new technologies such as digital subscriber lines (DSL), cable modems, broadband wireless, third generation mobile and digital TV offer the opportunity for the widespread delivery of these services.

S3. Oftel expects that, in time, a number of different delivery routes for higher bandwidth will be available. But for now, DSL technology, which enables higher bandwidth access to be delivered over ordinary copper telephone lines, offers the greatest potential to deliver these new services to smaller businesses and the public at large. Currently only BT provides the ubiquitous network that could deliver this access to the mass market.

S4. BT has recently begun trials of ADSL in some areas and has announced its intention for a full launch of these services in March 2000. Oftel welcomes this development. At its launch, BT’s proposal will bring ADSL capability to around 6 million households. This is a significant step forward in the delivery of higher bandwidth services to consumers. Oftel will continue to monitor BT’s roll-out against the requirements of its licence.

S5. But BT alone providing access is not sufficient to ensure competitive prices and choice for consumers, nor to encourage innovation in the provision of access. Giving other operators the opportunity to upgrade the loop and take their own decisions on when, where and how to roll out, is required as well.

S6. Oftel has therefore concluded that BT should make its local loops available to other operators to allow them to compete directly with BT in providing higher bandwidth access. This will be delivered through a form of local loop unbundling referred to in Oftel’s consultations as option 2. This document sets out how and when this new service will be made available and the advantages it will bring to consumers:

• Local access lines or ‘unbundled loops’ and co-location will be available to operators with interconnection rights and obligations

• Loops will be made available at a cost-based price, allowing for a reasonable element of profit (Long Run Incremental Cost plus a mark-up to cover common costs). An indicative price will be published by April 2000.
• Oftel will set out clearly the requirements on BT through a new condition in BT’s licence. Oftel intends to begin formal consultation on the licence condition by April 2000.

• The service will be introduced by July 2001 and earlier if possible. BT and industry groups have agreed a timetable that will allow the delivery of unbundled loops by this date.

• Oftel will conduct a review of its policy on access to bandwidth four years after the introduction of unbundling and every two years thereafter.

S7. This is a key decision for the development of telecoms in the UK. It sets the framework for the competitive delivery of information age services to consumers. Consumers will be able to take a variety of services together with telephony over the local loop direct from new providers. This will lead to competitive prices and greater choice and diversity in the products on offer.

S8. It also presents a major challenge for the industry. BT and other network operators have been working over the past few months on processes to enable the smooth introduction of unbundled loops through option 2. Important inputs are needed from operators by the end of January 2000. The continued co-operation of operators and their participation in industry groups is essential in ensuring that this important new initiative is delivered on time.

Back to Contents

This document presents Oftel’s conclusions on bringing higher bandwidth services to consumers.

Chapter 2 sets out Oftel’s objectives; the background to the proposals set out in the July Consultation ‘Access to Bandwidth: proposals for action’; the analysis that Oftel has conducted on the costs and benefits, and Oftel’s conclusion that the provision of unbundled loops (option 2) is necessary to promote competition in higher bandwidth access.

Chapter 3 sets out the procedures and timetable by which Oftel will review the effectiveness and continuing appropriateness of the policy and the tests that will be applied to assess whether any change to the policy is required.

Chapter 4 describes the service (partial baseband leased circuit) that BT will need to provide to other operators. It also sets out the legal principles behind the requirement to provide this circuit.

Chapter 5 sets out guidelines for operators on the availability of loops and conditions of use. It gives details of who will be entitled to purchase the service, how the service may be used, and where the service will be made available.

Chapter 6 describes the pricing principles that will apply to provision of the local loop. It confirms the cost-based (LRIC) approach set out in previous consultations, and confirms that there will be no requirement for a geographically averaged tariff.

Chapter 7 gives details of the co-location facilities that BT will be required to offer as an adjunct to the option 2 service, and the pricing arrangements that will apply.

Chapter 8 gives an update on the implementation procedures Oftel has put in place to ensure that option 2 is delivered by the target date of July 2001.

Chapter 9 details the progress made by BT in introducing its ADSL service and Oftel’s approach to monitoring and assessing this service to ensure compatibility with BT’s licence obligations.

Other issues are examined in chapter 10. These include consideration of how best to bring higher bandwidth to rural areas and the economically disadvantaged.

Back to Contents

2.1 Higher bandwidth access is of fundamental importance to development of new information society services. Technologies such as DSL, cable modems, third generation mobile, broadband fixed radio and digital TV will enable services such as always-on unmetered high speed Internet access, interactive audio-visual services and video-on-demand to be accessible to a wide audience.

2.2 Enabling these services to develop to their full potential is central to Oftel’s primary goal of promoting choice, quality and value for money for consumers.

2.3 The best way to achieve the variety of services that consumers want at reasonable prices is to promote effective competition in the provision of access to and delivery of these services. Regulatory intervention should only be undertaken where it is justified – that is in circumstances where left to its own devices, the market would not deliver the best outcome for the consumer, and regulation can bring a clear benefit.

2.4 In examining the case for action, Oftel has considered the level of demand in various segments of the market, the supply routes available and whether there are barriers to the competitive delivery of higher bandwidth access and services. The conclusion is that regulatory action is needed to introduce competition into the upgrade of the local loop. This is based on extensive consultation with consumers and the telecoms industry and on further analysis of the risks of action and inaction and costs and benefits.

2.5 Throughout the consultative process the focus has been on the delivery of services over BT’s copper local loop, upgraded to provide higher bandwidth capability. The rationale behind this approach has been that:

• BT’s local network is ubiquitous;

• BT has a very strong position in the local access market, with limited direct competition;

• BT is likely to remain the primary supplier of local access in the near future.

2.6 Oftel expects that other technologies also have the capability in the medium term to provide widespread access to higher bandwidth – cable, satellite, and broadband fixed wireless in particular. These technologies and others may in time provide a real alternative to higher bandwidth delivered over the local loop.

2.7 But in the near future Oftel believes that the local loop, upgraded using technologies such as DSL, offers the greatest potential to deliver higher bandwidth access to the mass market. It is for this reason that access to customers through the local loop is the focus of this document. When other technologies, such as those described above, develop to the extent that there is effective competition in the provision of higher bandwidth access Oftel can withdraw from regulation.

2.8 This Statement applies only to BT’s copper loops, and not to local fibre connections, such as those that are supplied direct to some businesses. Oftel does not rule out consideration of fibre in the future, should market conditions suggest it is necessary, but does not consider it appropriate now to take regulatory action in respect of fibre.

2.9 Oftel began its consultation with the publication in December 1998 of an exploratory document that looked at whether there was effective demand for higher bandwidth services from consumers, SMEs, Service Providers and Other Licenced Operators (OLOs) and at whether there were any barriers to the efficient supply of these services. It focused in particular on the supply of these services via BT’s local loop, and proposed five options to tackle any barriers found to exist. These options ranged from unbundling of the local loop (options 1 and 2) to solutions consisting in access to customers over BT’s upgraded network (options 3,4 and 5).

2.10 Responses to the consultation suggested that there was potential demand from consumers for higher bandwidth services and effective demand from SMEs that remained unmet. Service providers and OLOs were also of the view that there was demand from customers that they wished to address, but that they were unable to gain access to these customers. Respondents claimed that BT’s control of local access lines (the local loop) was creating a barrier to the delivery of higher bandwidth services. The remedies that received most support were:

• Option 2 (Partial Baseband Leased Circuit), a form of unbundling in which BT makes local access lines available as leased circuits to other operators, was identified by many operators as being the most attractive option. It would allow operators the flexibility to create their own access offerings choosing when and where to roll out and the technology with which to upgrade the lines.

• Option 4 (Permanent Virtual Circuit), where service providers and operators access customers over BT’s upgraded network, was cited by many service providers as their preferred option. This had the advantage that access to a number of customers could be gained without the need for significant investment in infrastructure. The disadvantage is that it offers less flexibility than option 2.

2.11 In July Oftel published a second consultation that detailed these findings and set out Oftel’s preliminary conclusions. It noted that while BT was due to announce plans for roll-out of ADSL that were similar to option 4, this would not in itself be sufficient to ensure the competitive delivery of higher bandwidth access. It therefore proposed that to introduce competition into the upgrade of the loop, unbundling through option 2 should be mandated and made available from July 2001.

2.12 The July document invited comment on the overall proposals and on specific issues such as pricing and availability of unbundled loops. The document also identified technical and operational issues that would need to be addressed before loops could be made available. Two industry groups were set the task of resolving these issues and producing guidelines and recommendations for the implementation of option 2. The groups began work on these tasks in August.

2.13 Responses to the July consultation generally endorsed Oftel’s proposals, with useful additional feedback on specific issues:

2.14 The majority of network operators agreed with Oftel’s view that there is a need to take some form of regulatory action to meet the demand for higher bandwidth services. They supported Oftel’s proposals for option 2 and agreed that the service should be priced at LRIC with a mark-up for common costs. Many expressed concern that action was not being taken sooner to introduce this service and argued that Oftel’s target date of July 2001 be brought forward to ensure that competition was introduced at the earliest opportunity. A number of operators observed that, particularly in the time before option 2 was introduced, the only access route open to them would be BT’s ADSL offering which they felt to be insufficiently flexible to satisfy their needs.

2.15 Oftel’s proposal to mandate option 2 was well received by service providers, as a solution that would allow real competition to develop. It was hoped that option 2 would lead to lower tariff packages than those available through BT’s ADSL product and result in increased choice and value for money for consumers. Again concern was expressed at the proposed timescale for introduction of option 2.

2.16 In general consumers and consumer groups supported the proposal to open the local loop to competition as a mechanism to prevent BT from extending its strong position in the telecoms market into the provision of higher bandwidth. Consumers pressed for Oftel to ensure that all sectors of society were able to gain access to higher bandwidth services particularly those in rural and economically disadvantaged areas.

2.17 In its response BT continued to question the rationale behind Oftel’s proposals and specifically identified the proposed LRIC pricing of the option 2 loops, and use of the loops for telephony and to provide service to large businesses, as areas of concern. BT also proposed that the requirement to provide option 2 be specifically set out in a new licence condition.

2.18 A summary of the responses received is set out in Annex A. Oftel has considered these responses carefully in reaching the conclusions set out below.

2.19 Since the publication of Oftel’s July consultation, BT has begun trials of its ADSL service, similar to option 4 of Oftel’s consultation, and has released further details of plans for a widespread roll-out across its network.

• As required by its licence, BT is making wholesale products available to service providers and other operators on the same terms as are available to their own in-house service provider.

• Trials of BT’s wholesale ADSL products began in October and November in limited areas and with a selected number of participants. The trial prices for these wholesale products range between £60 and £150 per month per end user.

• BT is aiming to launch the wholesale products that it is currently trialling in late March 2000. By this time BT expects to have upgraded over 400 exchanges, which will serve around 6 million homes and businesses. Further expansion of the ADSL service is planned, but is dependent on the success of the initial roll-out phases.

2.20 Oftel welcomes BT’s decision to respond to customer demand and offer services providing higher bandwidth access. Oftel recognises that this is a major investment programme, which should make ADSL available to a significant proportion of the population in a relatively short timescale. Oftel will be monitoring the service against the requirements of BT’s licence to ensure that the service is supplied to competing service providers on non-discriminatory terms, and that reasonably required interconnection products are available to operators. Further information on BT’s products and Oftel’s approach to assessing BT’s roll-out is set out in chapter 9.

2.21 However, while welcoming BT’s roll-out, Oftel does not accept that the availability of BT’s wholesale service, similar to option 4, will in itself deliver a fully competitive outcome in the market for higher bandwidth access and services. For BT alone to roll-out DSL technology is not sufficient because, in the absence of widespread alternative access mechanisms:

• Provision of access to customers and the technology used to enable access would depend on BT’s judgement alone. BT could set the pace of delivery and dictate the precise technologies used, denying other operators the chance to compete on equal terms and depriving consumers of choice;

• If BT delayed roll-out or did not react to demand for particular types of service, demand from consumers and SMEs could remain unmet;

• BT would be able to set a price for the services above that which would be charged in a fully competitive market, to the detriment of consumers.

2.22 Oftel has considered how these problems might be addressed. There are two main options.

2.23 Firstly Oftel could dictate the roll-out of BT’s product offerings, establish milestones and control prices. But this would essentially mean Oftel substituting its own commercial judgement for BT’s. This is an unattractive option for Oftel, at best a pale substitute for competition, and at worst, could result in significant distortions in BT’s roll-out plans leading to stranded and uneconomic investments. It could also distort the competitive landscape, discouraging others from rolling out in competition with BT.

2.24 Secondly, Oftel could take action to introduce competition into the upgrade of the local loop by mandating option 2. This offers the advantage that it allows operators to make their own commercial decisions as to where to roll out and which technology to use. And, importantly, in allowing others to compete with BT in offering higher bandwidth access, Oftel could hold back from detailed regulation of BT’s ADSL roll-out. BT would be subject to competitive constraints and would be able to respond according to its own commercial judgement. Other operators and service providers would be able to make their own investment using unbundled loops if they found BT’s products unattractive to them.

2.25 Oftel recognises that there are important considerations that need to be taken into account with option 2. There is a risk that if inappropriately applied, particularly if loops were made available at too low a price, it could have the effect of deterring investment in alternative networks. But Oftel considers that this risk can be minimised by careful application of pricing principles.

2.26 Overall Oftel considers that the second of these options, to take regulatory action to introduce option 2, is likely to deliver the best outcome for the development of competition in the provision of higher bandwidth, and will bring the greatest benefits to consumers. It will also place competitive constraints on BT’s offering so removing the need for strict ongoing regulatory constraints that might otherwise be necessary.

2.27 Oftel has also conducted a cost benefit analysis to assess whether the benefits of introducing option 2 outweigh the costs. Oftel considers that the main costs associated with the implementation of Option 2 are:

• An increase in the average cost of DSL services due to loss of economies of scale. One of the possible effects of this could be to jeopardise the viability of making DSL technology available to the mass market. To examine this in more detail, Oftel commissioned an independent assessment of the impact of option 2 on the commercial viability of BT’s planned ADSL roll-out. The results of the study suggest that there is a continued case for BT’s roll-out even with a substantial reduction in customer numbers;

• Costs are also incurred through the need for services ancillary to option 2 such as co-location and order handling.

2.28 However these costs must be weighed against the prospective benefits that will accrue through the introduction of competition in the provision of higher bandwidth access. The main benefits that would be brought about by competition are that:

• In a competitive environment operators will have an incentive to increase the efficiency of their operation. This will result in a reduction in the costs of providing a service;

• Competitive pressures will lead to a reduction in operators’ profit margins and lower prices for consumers;

• Competition will foster an environment in which operators innovate to compete for market share. The availability of new types of services is likely to increase demand for higher bandwidth from consumers, leading to an overall expansion in the market.

2.29 The source of the information used by Oftel to calculate the costs and benefits of Option 2 as set out above is a model of an ADSL roll-out completed for Oftel by Analysys. The result of this cost benefit analysis is to show very significant net benefits of between £120m and £440m, depending on a range of assumptions and the expected timing of competitive alternatives to copper loop technology. The full cost benefit analysis can be found at Annex D.

2.30 On the above analysis, Oftel’s conclusion is that option 2 and option 4, the latter delivered through BT’s voluntary roll-out, together present the best way of ensuring maximum availability of higher bandwidth at reasonable prices.

2.31 The availability of option 2 will bring competition into the market for higher bandwidth access, promoting innovation in access mechanisms and allowing operators other than BT to make their own decisions as to when and where to invest and offer service. This will be complementary and present a competitive spur to plans by BT to roll out ADSL to a large proportion of the population. This two-track approach will best meet the needs of consumers and SMEs requiring low-cost access to a variety of higher bandwidth services.

2.32 Operators and service providers are keen to see the early introduction of option 2 and some consider the date proposed in Oftel’s last consultation of July 2001 to be too late. But this is the maximum timetable. Oftel would like to see the introduction of option 2 as soon as is practically possible. There are three elements that need to be resolved before it can be successfully introduced:

• Guidelines on operational processes such as co-location and order handling need to be completed. These issues are being examined by a subgroup of Oftel’s Operator Policy Forum (OPF);

• Recommendations for a spectral management plan to minimise interference between lines need to be adopted and published. These issues are being examined by a subgroup of the NICC, an advisory body to Oftel on technical matters;

• BT needs to develop their systems to enable them to process orders for option 2 loops (PBLCs) and co-location. This work is dependent on inputs from the two groups above.

2.33 Oftel is committed to ensuring that these three elements are completed in time to allow a full trial of the option 2 service by the end of 2000 and launch of the service by July 2001. This will need considerable commitment both from BT and other market players. Oftel is confident that this will be forthcoming and therefore considers the targets are realistic for ensuring a smooth launch. But if the industry discussions progress sufficiently that outstanding issues can be resolved and support systems built in a shorter timescale, Oftel will bring forward the implementation date for option 2.

2.34 Further details of the implementation procedure and current timetables agreed by BT, the OPF and NICC working groups for completion of their tasks are set out in chapter 8. These timetables will ensure that option 2 is introduced by the target date of July 2001, but may be revised if it is found that an earlier implementation date can be practically achieved.

2.35 Full contribution from the industry is essential in ensuring the successful introduction of option 2. Oftel is therefore encouraging operators that wish to make use of option 2 to participate in the Operator Policy Forum focus group looking at operational matters. Operators wishing to do so should contact the Chairman:

Greg Mook, MCI Worldcom on greg.mook@wcom.co.uk

2.36 Additionally operators wishing to participate in or follow the progress of the NICC group looking at technical issues should contact the Chairman:

David Davies, BT on david.r.davies@bt.com

2.37 Option 2 is a service provided by BT to other operators. It is not in itself a service that consumers will buy or have to order themselves. The operators providing the higher bandwidth services will manage all this. However, consumers need to be aware of a number of important options in the arrangements that may be offered.

2.38 Most consumers are likely to take asymmetric DSL services such as ADSL. These services are designed to allow the simultaneous use over a single line of higher bandwidth and analogue telephony. Thus consumers taking service from operators taking option 2 loop, may be offered a package of telephony and higher bandwidth services. A consumer may:

• Have a single line going into his/her premises carrying both telephony and higher bandwidth services from another operator, and cancel his/her contract with BT;

• Have two lines going into his/her premises with one carrying the BT telephony and the other carrying service(s) from the other operator;

• In certain circumstances it might be possible to have a single line carrying both telephony provided by BT and a higher bandwidth service from another operator. To simplify the contractual arrangements it is likely that the new operator will bill for both telephony and higher bandwidth and handle other customer service issues.

2.39 Symmetric services such as HDSL require a number of telephone lines (2 or 3) to be dedicated to them, and are not designed to run simultaneous with telephony over the same wire. Thus customers taking symmetric services will need to retain their existing telephone line or make an alternative arrangement for the provision of telephony.

Back to Contents

3.1 Chapter 2 explains the reasons why Oftel believes regulatory intervention to require BT to make available its local loop is appropriate. However, Oftel hopes that the market position will change over the course of time. Indeed Oftel expects that, through increased competition, the option 2 policy will facilitate this market development. In particular Oftel expects that, with the expansion of the market, alternative technologies such as cable modems, third generation mobile, broadband fixed radio, and access through digital TV will become more widespread and available to a larger proportion of the population. Those who have built out their networks to take advantage of unbundled loops may also at some date choose to build out their infrastructure further to the end-user rather than continuing to rely on BT’s loops.

3.2 For these reasons, and to examine the efficacy of the particular option 2 solution proposed, Oftel considers it necessary to hold a regular review after the introduction of option 2. To provide sufficient certainty for those intending to make investments, Oftel proposes that the first review should be held four years from the introduction of option 2. Pricing is one of the key elements of the option 2 policy and this timing is also likely to coincide with any Price Control Review. Thereafter reviews will be held every two years. The review of Oftel’s Access to Bandwidth policy will examine:

1. the state of the market for higher bandwidth access, and in particular the state of development of alternative infrastructures to BT’s local loops capable of delivering higher bandwidth; and
2. the efficacy of the option 2 policy in enabling operators to deliver higher bandwidth to consumers and SMEs. This will include examination of the product offered and the pricing regime.

3.3 Taking these factors into account, the review will consider whether any change is needed to Oftel’s policy on the provision of PBLCs at LRIC prices. Any changes to the policy will only be made after due consultation and will be introduced only after an appropriate notice period. (Oftel’s current view is that notice of some 2 years would be appropriate. Existing contractual arrangements would be undisturbed.) Amendments would be appropriate to the market conditions at the time, but measures that might be considered include:

1. an adjustment to the product required to better meet the needs of consumers and operators;
2. adjustments to the pricing structure of unbundled loops to reflect a market in which competitive alternatives were available; and
3. notice of termination of the requirement on BT to provide loops.

3.4 Oftel understands that it is crucial to the success of option 2 that operators are confident that the product will be available over a reasonable period of time. Oftel is committed to the continued availability of the option 2 service for as long as market conditions suggest that it is necessary.

Back to Contents

4.1 This chapter describes the services that Oftel considers that BT will need to offer operators under option 2.

4.2 Option 2 refers to a type of leased or ‘private’ circuit known as a Partial Baseband Leased Circuit (PBLC). This consists of a copper link extending from the customer premises to the local exchange (commonly referred to as a ‘local loop’). The circuit is partial because, rather than connecting one customer with another, it provides only a portion of the link and is designed for connection with another operator’s system for onward conveyance. The circuit is baseband because it extends over the entire available spectrum, from DC(0Hz) upwards, and can thus be used to provide services that make use of all the available frequencies on the line.

4.3 The operator purchasing the service will be able to install equipment at the local exchange and customer premises to enable the loop to deliver a range of services. There will be no restrictions on the technology used to upgrade the loop, but the operator will need to adhere to certain technical standards to ensure there is no interference between the loops. This is discussed below. In the immediate future DSL is likely to be the technology used to deliver higher bandwidth and other digital services (including digital telephony) over the local loop. The operator may also deliver traditional analogue phone services similar to those currently being provided by BT.

4.4 Loops sold as PBLCs will need to meet a given quality standard to enable them to be successfully upgraded and used to deliver higher bandwidth services. This quality level will be measured in terms of a number of characteristics, which may include the length of the loop, its tendency to radiate unwanted emissions and the material from which it is composed (eg copper). The quality determines whether the loop can support certain technologies, such as DSL, and also determines the performance level achievable over the loop when upgraded. For instance in rural areas because of the longer length of the loops, the maximum data rate using current technologies may be less than 200kbit/s. The NICC group is currently considering what would be an appropriate quality standard to set for PBLCs.

4.5 Because operators could place different types of equipment on the PBLC to provide higher bandwidth, there is a risk that crosstalk between the equipment could cause mutual interference. This risk can be minimised by specifying a spectral management plan and deployment processes for the attachment of equipment onto the loop. The exact specification of the plan will depend on a defined set of objectives. Oftel’s high level objectives are that the plan should be capable of delivering a high level of customer penetration (this is likely to be delivered primarily through asymmetric technologies such as ADSL) and should require the minimum of management processes.

4.6 The NICC DSL task group will develop a precise plan and deployment processes based on these objectives, and will, subject to Oftel’s agreement, publish guidelines, which operators should follow when upgrading the PBLC.

4.7 Further details on line characteristics, spectral management plan and other technical issues are available at annex B.

4.8 It is expected that to provide the PBLC, BT will in most circumstances be able to make use of ‘spare pairs’ in its access network, just as it does when customers request a second telephone line in their homes. Once provisioned and tested, the customer can use the new PBLC, as enhanced by the other operator. It will be for the customer to decide whether to cease any existing service(s) that he/she has from BT. In some cases, where no spare pair is available and the customer does wish to transfer all service from the current line, it may be possible for the PBLC to use the existing access circuit, but BT and the new operator using the loop will need to ensure the continuity of 999 access. The important point is that a PBLC does not necessarily need to use the existing customer telephone line, but can use any suitable line that can reasonably be provisioned.

4.9 The primary scenario that Oftel has considered is that an operator purchasing the PBLC service will supply both telephony over the low frequency portion and DSL services over the higher frequencies. This is the way that BT itself will be making use of DSL upgraded lines and should bring the maximum economic value from the line.

4.10 However consumers have suggested that they would like to have the option of continuing to take their telephony from BT while using a competing operator for higher bandwidth services. Some operators have also suggested that they would like to focus on the provision of higher bandwidth services, while BT continues to provide telephony.

4.11 In responses to the consultations, operators suggested this could be done through a variant of option 2. This variant could be implemented in two ways:

a) the line is split before the main distribution frame (MDF) and the other operator purchases only the higher frequency portion of the line while the low frequency analogue voice channel is directly routed to BT; or

b) the other operator purchases the whole circuit and splits off the lower frequency telephone band for connection to BT’s exchange. The operator would then resell BT’s service.

4.12 Further details of these options are given at annex B

4.13 Of the two solutions described above, Oftel considers that the first is likely to cause considerable operational difficulties, as there are effectively two companies having simultaneous use of a single line.

4.14 But Oftel sees no reason why the second variant cannot be successfully implemented. Oftel’s current view is that the appropriate mechanism for achieving this should be a private contractual arrangement between BT and the other operator.

4.15 In order to enable operators to plan their roll-out, BT will need to provide to other operators information on the location of their local telephone exchange buildings and the precise areas they serve. BT will also need to provide, on request, feasibility data for each proposed circuit (such as the Red-Amber-Green system used for their own roll-out) so that operators can assess whether they are likely to be able to provide a given customer with DSL.

4.16 The general requirement for BT to provide the PBLC service will be established through BT’s licence. Oftel expects that further detailed elements, such as those on the quality of the line and adherence to the spectral management plan, will be mutually agreed and covered through contractual arrangements between BT and operators taking the service. Operators will also need to ensure that their service is compliant with any standards laid down by the Radiocommunications Agency on radio emissions levels.

4.17 The PBLC described above is a telecommunication service as defined in the Telecommunications Act 1984. Oftel can therefore require the provision of the PBLC through existing conditions in BT’s licence that relate to the provision of telecoms services on reasonable request. However Oftel recognises that this is an important decision for all concerned, and that the exact nature of the service should be clear both to BT and operators. Oftel is therefore proposing that, for transparency and the avoidance of doubt, a new condition be inserted in BT’s licence that sets out the services that BT will need to provide and any relevant conditions. The insertion of a new condition in BT’s licence will also enable Oftel to set out a clear pricing regime for the service.

4.18 Oftel has already begun discussions with BT about the form this new licence condition will take. Oftel intends to consult informally on this new condition and publish it for formal consultation by April 2000. There will then be a statutory consultation period of at least 28 days, and Oftel will allow another 14 days for comments on comments.

4.19 As with other leased circuits, the PBLC remains part of BT’s network, and is managed and maintained by BT. Thus Oftel expects that the contract between BT and operators taking the service will cover in detail BT’s obligations to test the line, correct line faults and maintain the quality of the line at a given level. These are all necessary elements in the provision of the service as described above. Equally there will be contractual obligations on the operator using the service to adhere to the spectral management plan and deployment processes to prevent cross-talk between lines. These are reasonable measures to ensure that the integrity of BT’s network is maintained and BT would be entitled to disconnect any operator infringing these requirements.

4.20 Details of the contractual arrangements are currently being considered by the OPF focus group tasked by Oftel with resolving the operational issues associated with option 2. Further information on the work of this group and on the NICC group examining technical issues is given in chapter 8 on implementation.

4.21 Oftel will resolve any disputes that arise over the provision of unbundled loops, such as refusal to supply or late delivery of the service, through the usual compliance procedures.

4.22 It is in all operators’ interests to adhere to technical requirements such as the spectral management plan and deployment processes, and Oftel hopes that this will largely be self-policing. But if an issue nonetheless arises – for instance where an operator claims that it has wrongfully been disconnected by BT – the matter can be referred to Oftel.

4.23 Further details on the legal framework are set out in annex C.

Back to Contents

5.1 This chapter gives guidance to operators on the conditions that will apply to the provision of option 2. It sets out:

• Which telecommunications operators will be entitled to make use of option 2 loops

• Where loops will be made available

• Circumstances in which loops may not be available

• Restrictions on the uses to which the loop can be put

These and other matters will be covered in the proposed new BT licence condition.

5.2 The service will be available to those listed as having interconnection rights and obligations under Schedule 2 of the Telecommunications (Interconnection) Regulations 1997 No. 2931. The rationale behind this is that:

• The PBLC is only capable of being used when interconnected to the system of another operator. It should thus be made available to Schedule 2 operators who have rights and obligations to interconnect under the Interconnection Regulations;

• Oftel considers limiting BT’s requirement to provide loops to Schedule 2 operators is reasonable without being over-restrictive. Indeed, this is not a major restriction in practice, as most operators that provide basic conveyance services (‘network services’), and would thus be able to make use of the loop, are eligible for inclusion in Schedule 2.

5.3 Oftel considers that competition in higher bandwidth access through unbundled loops should be available in all areas where there is reasonable demand. Some respondents to the consultation suggested that to promote the development of alternative access routes, access to loops should be restricted in areas where other access options such as cable modems or access through BT’s DSL service, were available. Oftel has considered this argument, but has concluded that such a restriction would deprive customers in these areas of the technological innovation and competitive benefits of option 2. In particular:

• Customers in areas where option 2 was not available might not have access to the full choice of higher bandwidth technologies and equipment solutions that would have been available otherwise;

• In areas where BT had rolled out there would be less incentive on BT to innovate, improve service and lower the prices of its own comparable ADSL offering;
• The pricing of option 2 loops has been set at the level it would reach in a market where loops were competitively supplied. If other technological access solutions are equally or more efficient than DSL, they should not be adversely affected by the availability of option 2. Indeed the availability of option 2 could provide a competitive spur to the development of other forms of access;

• Oftel wants to see effective competition in the provision of higher bandwidth services and this is best promoted by enabling all those that can efficiently serve a market to do so.

5.4 While, as described above, there will be no general geographic limitations on where BT should be required to supply loops, there may be a small number of instances in which BT could argue that it is not possible or reasonable for them to provide the requested service:

• If demand for loops at a particular location was low, BT might reasonably argue that demand was not sufficient to justify the arrangements necessary, such as provisioning of extra staff, to introduce option 2 into that area;

• BT is required to provide lines at a given quality level and should take reasonable measures to condition lines to enable them to meet this quality level. But in rare circumstances there may be no lines available that can reasonably be conditioned to provide the service requested. This might happen, for example, where lines were of extreme length. Oftel would not, however, consider poor quality of local infrastructure or the use of line-doubling equipment on existing lines to be reasonable arguments as to why the service could not be provided.

5.5 An indicative list of circumstances in which BT might be excused from the obligation to provide telecoms services such as the PBLC is set out in BT’s licence. This is discussed in more detail at annex C.

5.6 The driver behind Oftel’s proposals is to open up competition in the provision of higher bandwidth services to consumers, and perhaps particularly SMEs. This objective was stated in the original document, and the conclusions set out in this document are designed to achieve it.

5.7 On this basis, BT has argued that it should be entitled to impose a contractual condition preventing use of the loop for analogue telephony alone (where the loop has not been upgraded). Oftel’s view is that such a restriction would be inappropriate and unnecessary because:

• There is little demand for unbundled loops for telephony alone. If an operator wants to deliver single line analogue telephony there are in nearly all cases much cheaper effective alternatives already available, such as indirect access;

• Using the local loop would also mean that the other operator would need to invest in infrastructure to the local exchange, and in co-location. This investment could be disproportionate to the returns that could be achieved through using the loop for single line telephony only, and would be unlikely to produce the returns needed to justify the investment;

• If in special circumstances operators can make practical and efficient use of the lines for telephony alone, it should result in increased competition and benefits to customers.

5.8 For these reasons Oftel does not intend to permit a formal restriction. However Oftel will keep developments in this area under close review. In particular Oftel will consider the likely impact on the use of option 2 loops of any changes made to other regimes such as the indirect access pricing regime. If Oftel were to propose changes in other regimes in the course of the price control review that seemed likely to introduce material distortions into the market through use of option 2 loops for telephony, Oftel would consider introducing some mechanism (such as an additional mark-up on the LRIC basis for pricing option 2) to remove any such distortion.

5.9 While it is likely that most operators will be offering higher bandwidth access over the loop of one kind or another and that has been the focus of Oftel’s consultations, technologies such as DSL may also be used to provide other innovative services. For instance technology is being developed that will allow the delivery of multiple digital voice channels over a single upgraded line. Or IP could be run over DSL to deliver multiple lines or a combination of telephony lines and higher bandwidth services. While the provision of multiple voice channels is not itself ‘higher bandwidth’, it does make use of the higher frequencies made available through the upgrade of the loop. Oftel is keen to encourage the innovative use of DSL technology to provide services such as these, which may bring to households and small business the benefits that larger businesses currently enjoy. Thus Oftel does not consider at present that there should be any restrictions on the uses to which the upgraded loop may be put.

5.10 It is expected that unbundling of the local loop will largely be of benefit to residential consumers and SMEs, but Oftel does not consider it necessary to restrict use of the loops to serve only customers falling into these categories. This is because Oftel considers that:

• It is impractical to make a distinction between one type of customer and another. Any distinction created would be to some extent arbitrary and open to dispute;

• While many large businesses are competitively served, this is not universally so, particularly in the case of branch offices. Thus customers that might truly benefit from the availability of competitive DSL could be disadvantaged by such a restriction;

• The product by its very nature is likely to be more attractive to consumers and SMEs. It is unlikely to meet the requirements of those large businesses which require dedicated high bandwidth services of the type more effectively supplied by fibre leased lines. Even if it does meet these high specifications in certain circumstances, it is likely that cheaper competitively supplied alternatives would be available to these businesses. There is therefore no need to restrict the availability of DSL;
• If for certain purposes and in certain locations, DSL technology is of use to larger businesses (where other leased lines may not be appropriate or may not be available) it would seem unreasonable to deny them the benefits.

Back to Contents

6.1 As explained in chapter 2, the rationale for requiring BT to provide option 2 is based on BT’s strong market position in the provision of local access lines and the need to promote competition in the supply of access to higher bandwidth. This strong market position would enable BT, in the absence of controls, to set prices for local access lines at a higher level than they would be in a competitive market and make profits in excess of those an operator in a competitive market could expect. For this reason Oftel has concluded that it is necessary to set the price of these lines. Oftel will set out the basis on which the initial price and any ongoing adjustments will be established in the proposed licence condition. When Oftel consults on the proposed licence condition, it will also publish an indicative price for option 2 loops: the final price Oftel determines will be based on this but updated to reflect latest data. The pricing regime for loops will be reviewed regularly through the procedure set out in chapter 3 and may be adjusted to reflect changing market conditions.

6.2 The December Consultation Document set out three options for charges for BT’s loop under Option 2:

• Line rental less retail costs (retail minus);

• Long Run Incremental Cost (LRIC) plus a mark-up to account for costs that are common to the line and other BT services (the ‘LRIC +’ basis); and

• LRIC plus lost profit from line rental and calls (efficient component pricing rule, ECPR).

6.3 The response to the Consultation Document indicated general support for LRIC as the correct approach to charging for the local loop. This was consistent with Oftel’s own view, and Oftel proposed this pricing regime in the July Consultation Document. After some further analysis of the options, in the light of responses to the July consultation, Oftel has decided to confirm LRIC + as the pricing regime for the local loop.

6.4 The reasoning behind this decision is that:

• A LRIC + based charge reflects replacement cost, which is the economic value of a line in a market where loops are competitively supplied.

• LRIC + takes account of the structure of telephony tariffs

6.5 It could be argued that the charge for each line should reflect its economic value, as given by its profitability, under a pricing arrangement known as Efficient Component Pricing Rule (ECPR). A strict interpretation of this approach would require the setting of a separate price for each individual line and would clearly be impractical.

6.6 The setting of a uniform charge for all lines (although see the section on geographical averaging below) is much more practical and in Oftel’s view, a simple and appropriate approach is to set an average price that reflects the economic value of lines in a competitive market. The economic value of the average line, according to the reasoning above, in a competitive market should be LRIC plus mark-up. This, in Oftel’s view, gives the appropriate charge for an unbundled line under option 2.

6.7 Oftel recognises that it is possible that the combination of a single price for lines, reflecting average profitability, and a tariff that creates varying degrees of profitability per line, might create incentives for operators to target the more highly profitable customers. If this were to turn out to be the case, and if the result were to be that BT was unable on average to recover its costs plus a reasonable return on capital, Oftel would consider whether it was appropriate to address this through an additional mark-up on the LRIC-based price for the use of the loop.

6.8 The charge for the loop will reflect both the capital and operating costs associated with providing a line of a minimum quality from the end customer – home or business – to the local exchange or concentrator site. Operators may also need to purchase additional elements to enable them to make use of the loop. These could include, for example, a tie cable linking the operator’s equipment inside the exchange to their system outside. Oftel considers that the basis for charges for these necessary elements should also be the efficient costs of providing these elements calculated on a LRIC basis.

6.9 An additional issue in the pricing of the loops is whether a single price should be set for the entire country (geographic averaging) or whether variations in prices should be permitted where these variations reflect different costs in different areas. Oftel proposed in the July consultation that, in the case of option 2 loops, geographic averaging should not be a requirement on BT. The benefit of this approach is that, as individual prices would more closely reflect costs, it would encourage efficient investment decisions, particularly by providers of alternative infrastructures. Following responses to the consultation, Oftel has decided to confirm this approach.

6.10 Oftel expects that the initial price established will be a single average figure. However, it will be open for BT to argue for variations in the price. Clearly BT would need to provide adequate supporting evidence to demonstrate that any proposed price variations did indeed reflect underlying costs.

6.11 The pricing that Oftel has proposed for PBLCs (at LRIC) differs from the pricing regime applied to other leased circuits, which are currently not subject to a price control.

6.12 The reason that Oftel has chosen to examine the price of this type of leased circuit separately from others is that the opportunity has only recently arisen for these circuits to deliver higher bandwidth services to consumers and SMEs. Before the advent of DSL technology there was not much interest from operators in making use of the loop as there are other ways, notably by indirect access arrangements, to deliver telephone services to consumers and SMEs. Thus it was not necessary for the local loop to be offered as a leased circuit.

6.13 Now that demand for this type of circuit has arisen in order to provide high bandwidth services, it is necessary to examine BT’s market position in the provision of these new circuits to determine which pricing arrangement would be suitable. Oftel’s conclusion, as is stated elsewhere in the document, is that BT has a very strong market position in the provision of local loops, which would allow it to set prices above those in a market where they were competitively supplied.

6.14 Oftel has therefore proposed that to promote competition in the upgrade of the local loop, and in the long term provide to a stepping stone from which operators can roll out their own infrastructure, copper loops should be made available on a LRIC + basis to other operators.

6.15 This Statement does not address the pricing of other types of leased circuit. Oftel has just published a Statement that examines the state of competition in the leased lines market and announces a review of the regulatory framework relating to national leased lines.

6.16 As well as upgrading the loop to provide ADSL is possible to upgrade lines (eg using HDSL) to provide symmetrical services. This technology is already used by BT in some circumstances to provide leased circuits of around 2Mbit/s in its ‘Megastream’ product range. It is hoped that the ability of other operators to make use of DSL to provide leased lines will further stimulate competition in the leased lines market and in particular pass on any savings that might be made in using this technology.

Back to Contents

7.1 In addition to purchasing the PBLC, operators will need to co-locate at or close to the exchange or concentrator buildings to enable them to interconnect with the circuit. This chapter describes the types of co-location that may be available to an operator and the pricing regime that will apply.

7.2 Co-location takes place when one or more operators rent space at a site to install equipment and make use of centralised facilities. There are three main types of co-location that could be used by an operator taking unbundled loops:

• Physical co-location – this involves renting space at a BT local exchange or concentrator building. A separate room might be provided for each operator or several operators might elect to share a single room with cages protecting the equipment. Operators would need reasonable access to the room in the BT building in which their equipment was sited;

• Distant co-location – this involves renting space at a nearby independently managed site connected to BT’s local exchange by a tie circuit. No direct access to BT’s facilities would be required, but a suitable adjacent property would need to be found, developed and managed for this option to be made available;

• Virtual co-location – an operator could arrange for BT to manage and maintain its equipment on its behalf. This would require the operator to relay specific instructions to BT on the set-up and configuration of the equipment.

7.3 Oftel considers that, to provide a level playing field between BT and other operators, BT should provide physical co-location, where space is available. This is because the alternatives, in Oftel’s view, do not in all circumstances provide a comparable service. If distant co-location is used, the connection between an operator and the end-customer will inevitably be lengthened, and this could limit the bandwidth that can be provided to end-customers over the loops. Virtual co-location could also be problematic as BT engineers might need to be trained, potentially at considerable expense, to operate equipment that was not familiar to them.

7.4 Some operators may nonetheless choose to use a distant co-location facility, where one is available, because of the greater flexibility it may offer, and Oftel will ensure that this option is also available to operators.

7.5 Operators may also find that virtual co-location is a practical option where the equipment they use does not differ substantially from that used by BT. Oftel expects that a service of this type could develop through commercial arrangements between BT and other operators.

7.6 BT’s licence obliges BT to provide appropriate points of interconnection, information and other services (such as co-location) necessary for interconnection where reasonably required. Oftel has established that provision for physical and distant co-location is reasonably required in order to enable operators to interconnect appropriately with the PBLC service. Additionally Oftel will set out further specific provisions relating to co-location in the proposed new licence condition. If disputes arise, Oftel’s intention is that these should be resolved through standard interconnection dispute resolution procedures.

7.7 Oftel expects that further detailed arrangements for physical co-location, such as those on security and the provision of power and other essential services to operators will be contained in the contract between BT and the other operator. It is expected that the contract will also detail the process through which BT would give notice of any changes to the co-location arrangements. BT has indicated that it plans to consolidate its exchange facilities in certain areas, and it will need to give adequate notice to operators co-locating within the buildings of any planned closures and the alternative arrangements available.

7.8 The detailed arrangements for co-location are currently under consideration by the OPF focus group.

7.9 As physical co-location space may be limited, Oftel expects that operators using it will need to demonstrate that they are making effective use of the space in conjunction with the purchase of option 2 loops, and not using it unnecessarily for activities that could be conducted elsewhere. Thus it would not in Oftel’s view be reasonable for operators to use space for the installation of switching or independent routing equipment (not included within the DSLAM). This limitation would apply at physical co-location sites, but clearly at an independent distant co-location facility any rules or restrictions on usage would be set by the manager of that facility.

7.10 Since the publication of the July document, Oftel has been considering, in consultation with BT and operators participating in the Operator Policy Forum option 2 focus group, possible pricing arrangements for physical co-location. The issues Oftel has considered include:

• The extent to which operators should bear the cost of co-location.

• How physical co-location should be arranged.

• If operators should bear the cost, and there is a shared co-location area, how the cost of setting up the area should be split amongst the operators.

7.11 In the light of the analysis produced by Oftel and responses from the OPF group, Oftel has concluded that:

• The cost that a fully efficient operator would incur in providing co-location should be borne by the operators using it.

• If the space available for co-location is able to hold the equipment of more than one operator, it should be up to the operators to decide whether to have separate rooms or caged equipment. The charge for co-location should reflect this choice.

• If operators have caged equipment within a separate, but shared room in an exchange, the charge for co-location should be set on the basis of average cost but perhaps with some allowance made for the additional risks an operator will take in being the first to co-locate at a particular site.

7.12 The full version of Oftel’s paper on co-location pricing is at annex E.

7.13 As well as the ongoing costs associated with dealing with operators and running the services, BT will incur costs in setting up the order handling processes necessary to deliver both loops and co-location. Oftel has applied the principles used to examine cost recovery for co-location to the issue of recovery of these additional costs. Using these principles, Oftel has concluded that operators should bear the reasonable costs of Option 2. The reasonable costs of Option 2 are those costs that an efficient operator would bear as a direct result of the introduction of Option 2. Oftel would expect these costs to be calculated on a LRIC basis.

Back to Contents

8.1 Before option 2 can be implemented a number of technical and operational issues need to be resolved. To address these issues two industry groups have been established. The Network Interoperability Consultative Committee (NICC) DSL Task Group will address and give guidance on all technical issues and the Operators Policy Forum (OPF) Option 2 Task Group will address all operational issues. It is essential that these groups work closely with BT, as their output will affect the requirements of BT’s operational support systems (OSS).

The specific issues that are to be addressed by these groups and the impact they have on BT’s OSS, including indicative timescales, are outlined below.

8.2 The NICC DSL Task Group will produce a spectral management plan for services to be connected to BT's local access cables and this will include connection and deployment processes.

8.3 These rules and processes are dependent on a set of objectives set out by Oftel. These objectives include the aim of a penetration level for upgraded loops that is as high as possible amongst residential and SME customers. Further details of Oftel's objectives are set out in annex B.

8.4 To ensure a speedy introduction of option 2 this group will initially concentrate on frequencies up to 1.1 MHz, which is sufficient for current ADSL technology, while protecting frequencies above this. The final output will cover the frequency range up to 30 MHz. Table 1 shows the agreed work plan for the NICC group. The group will also feed information to BT before the final completion dates so that BT has sufficient time to develop its OSS.

8.5 The OPF Option 2 Task Group will produce guidelines on processes and procedures for all aspects of co-location and circuit ordering, maintenance and billing.

The output from this group will be a set of documents addressing the following areas.

8.6 Table 2 shows the agreed work plan for the OPF group. As with the NICC group this group will be providing information to support BT’s OSS development before the final completion dates.

8.7 The introduction of option 2 creates two main new products for BT, the PBLC and co-location space, and a number of related products. The creation of these products and the anticipated demand for them will have a major impact on BT’s existing OSS. The systems that are likely to be affected will include order handling, provision, test and maintenance, fault handling and billing.

8.8 BT has produced a time plan for OSS development that will enable option 2 to be available by July 2001 (Table3). Under this time plan BT will need to begin OSS development in February 2000, and will thus need inputs from the OPF and NICC task groups by the end of January 2000.

8.9 A simplified version of BT’s OSS development time plan is shown in Table3. The four general OSS identified consist of many individual systems, so the completion dates given indicate when all the individual systems will be available.

8.10 To ensure a smooth roll out of option 2 services a trial programme will be run in parallel with the OSS development. The trial will start in July 2000, finish in June 2001 and will consist of three stages. The first stage will last until December 2000 and will test the plan and build for implementing co-location. The second stage follows on from stage 1 and lasts until March 2001. In this stage OLOs will take charge of a small number of PBLCs to allow technical tests of operators’ DSL equipment and manual trials of the pair provision, test and repair processes. The third stage will be a customer trial and will test BT’s automated ordering and provisioning systems. This will last until the middle of June 2001. If the trials succeed earlier, this timetable will be brought forward.

8.11 To provide co-location space BT will need to plan and build facilities in the exchanges affected. On the timetable outlined above BT should be ready to accept orders for co-location space by 1 January 2001. Oftel is monitoring BT’s progress closely to ensure that, given appropriate inputs from the industry groups, this timetable is met and there are no unjustifiable delays.

8.12 The final stage in delivering option 2 to end-users is for BT to provision PBLCs and for OLOs to install and configure their equipment. BT will be in a position to process orders for PBLCs using automated systems by 18 June 2001.

8.13 Completion of the above processes should mean that operators can begin offering services based on option 2 from 1 July 2001 at the latest. If however, progress is faster, this date can be brought forward. On launch of the service, BT has proposed that deployment initially be limited to areas of greatest demand and then phased in to other areas over a period of several months. Oftel will monitor this closely.

Back to Contents

9.1 As discussed in chapter 2, many respondents to Oftel’s first ‘Access to Bandwidth’ consultation said that they wanted BT to upgrade its network to provide DSL capability and to provide wholesale products to enable service providers and operators to provide high bandwidth services.

9.2 In July this year BT announced plans to roll-out ADSL technology. BT has announced the roll-out of three ADSL wholesale products, which are aimed at OLOs and SPs, including BT’s own SP and has just started trials of the products. The trial products are:-

• Videostream

• IPstream

9.3 BT began limited trials of its Videostream and Datastream products on 25 October 1999 and its IPstream product on 22 November 1999. BT intends to launch full products in late March 2000. By this time, BT expects to have upgraded over 400 exchanges which will serve around 6 million homes.

9.4 Oftel understands that BT will shortly announce details of wholesale products that are suitable as the basis of services for domestic consumers.

9.5 Oftel welcomes BT’s roll-out plans, which should present a significant step forward in the availability of higher bandwidth services for small and medium-sized businesses and consumers in the UK. Oftel’s aim is to ensure that in providing its ADSL products BT complies with its licence obligations. For example, BT is obliged not to show undue preference to its own SP business or exercise undue discrimination against other SPs/OLOs. BT must also comply with its obligation to provide reasonably required interconnection services.

9.6 Oftel has had a number of meetings with BT to discuss, amongst other things, its systems for handling orders. Oftel also monitored the selection process for the trial participants and will continue to meet with BT to discuss the progress of the trials and its preparations for the launch of the ADSL products next year.

9.7 Oftel has also invited comments on BT’s roll-out plans from operators and service providers and raised the subject for debate in Oftel forums. The two main subjects that operators and service providers have raised are the interconnection arrangements and the pricing of BT’s wholesale products.

9.8 The present trial products can be fully sourced from BT, including the interconnect from the various geographically deployed Points of Presence (PoPs) to the SP. However, operators must be able to make use of their own infrastructure from their own network up to the point of interconnection with BT. Any requests from operators for interconnection at other points that would enable Operators to make efficiency gains through the use of their own networks should be made to BT in the first instance. Oftel will consider any dispute referred to it in accordance with its normal compliance procedures. If demand for connection at another point is reasonable, taking all the relevant circumstances into account, BT would be required to provide the service under its current licence.

9.9 Some service providers and operators have expressed concern about the level of BT’s trial prices, outline details of which are given in paragraph 9.2. BT has not yet announced indicative launch prices, but it is expected that these will not differ greatly. Prices for end-users are likely to be higher than these figures as service providers, including BT’s own SP business, purchasing the BT ADSL products are likely to need to add an amount to cover their costs and make a return on the services (such as high-speed Internet access) they provide to customers over BT’s upgraded line. It is however possible that some service providers may also be able to use additional revenue streams, such as online advertising and e-commerce to lower the prices charged to consumers.

9.10 Oftel will continue to monitor the pricing trends and structures that emerge as the trial progresses. However, Oftel’s current view remains as stated in the July consultation document – that price control of BT’s prices for its own ADSL roll-out would be inappropriate. Price control at the start of a high risk and very expensive investment is likely to deter that investment and to limit the roll-out to the detriment of consumers. Instead Oftel wants to encourage roll-out to as wide a coverage as possible. Equally Oftel does not want users to be deterred by high prices – although of course prices must cover costs or otherwise development of competition will be hindered.

9.11 While Oftel does not intend to impose price controls, there are constraints on BT’s pricing freedom. First, under its licence BT must offer wholesale access products to competing service providers on the same terms as its offers to its own service provider business. It must also offer any wholesale access product for which there is reasonable demand. If it did not, Oftel would take enforcement action. Secondly, as stated in the July consultation document, Oftel would be able to take enforcement action if there were any evidence that BT were abusing a position of dominance. Thirdly, the prospect of competition from Option 2 and other competing delivery routes will be a constraint on BT’s pricing behaviour. Oftel expects that prices will fall as take-up increases, as technology develops and as competition grows – from other delivery routes such as cable and from the prospective introduction of Option 2. Any perception that prices for DSL are ‘high’ will encourage faster deployment of these alternatives. Further, it is in BT’s interests to attract as many users as possible – in other words not to deter users by high prices.

9.12 Oftel will look at BT’s prices carefully to ensure that it is complying with its licence obligations. As noted above in 9.12, BT has suggested that apart from subscriptions from consumers, there may be other potential revenue streams, such as revenues from e-commerce and advertising. If BT confirms that it will be making use of these revenue streams to reduce its prices to consumers, Oftel will want to be assured that these revenues are realistically assessed and that BT has fully accounted for its costs.

Back to Contents

10.1 Oftel has considered the position of Kingston with regard to local loop unbundling. It has been suggested that, as an operator with significant market power in the Hull area, a similar obligation should apply. Oftel’s preliminary view on this matter is that while Kingston enjoys market power in Hull, as BT does for the remainder of the country, the positions of the two companies are sufficiently different that it should not be assumed that a requirement on BT should automatically be passed to Kingston. The reasons for this are that:

• There has not as yet been significant demand from other operators for access to Kingston’s loops;

• The administration involved in establishing LLU in Hull is likely at present to be disproportionate to the benefit achieved. It may not therefore be reasonable to require Kingston to supply this product.

10.2 Oftel will continue to keep the situation under review and will give due consideration to any reasonable demand for access to Kingston loops

10.3 This statement does not cover the issue of whether other operators with local access networks capable of broadband delivery such as cable providers should be obliged to provide access to their end users (through an option 4 scenario) or offer unbundled local loops to competitors. Oftel will be consulting on access to other broadband networks shortly.

10.4 In their responses to the consultation, consumers stressed the importance of higher bandwidth access being available to, in particular, consumers in rural areas and those on lower incomes. Oftel shares this objective. Ultimately all consumers should have the opportunity to use broadband access to receive the new Information Age services.

10.5 Regulatory action now to require such provision would be premature because:

• The introduction of DSL will require considerable investment by BT, other operators and service providers. At this early stage in the development of a new market, it is vital to ensure that there are no unnecessary regulatory burdens on business. Rather barriers to the development of competition in the market for higher bandwidth access should be removed. As is normal with the introduction of new services, Oftel expects that the first stages of roll-out will cover areas where it is most profitable to introduce service. However Oftel expects that competition and choice will lead to more consumers taking up the service, so driving down costs which in turn will increase take-up, so increasing the incentive to roll-out to other areas which will become economically viable to serve. Regulatory action too early could deter investment and prevent this positive development of service over ever-increasing areas;

• Second, Oftel’s objective as described in the preceding chapters is competition between different routes for broadband access. To regulate to require one particular route to be rolled out might deprive consumers of choice in the future;

• Third, requiring roll-out in areas which are not yet economic to serve will mean higher prices.

10.6 Nevertheless Oftel intends to monitor developments closely so as to track progress towards broadband access for all. Other initiatives are also relevant here:

• Oftel has been in talks with BT and other operators about providing special tariffs for higher bandwidth access to schools and public institutions. This would allow for the provision of high speed Internet access points to school-children and the public at large. Special tariffs for dial-up and ISDN Internet access are already available to schools and Oftel has proposed in a statement on November 4 that similar products should be made available to public institutions;

• Funds are being made available by the European Community and UK Government agencies such as the Welsh Development Agency to support the roll-out of advanced infrastructure in some remote and rural areas. Similar initiatives are also underway in Scotland.

10.7 Oftel is already undertaking a review of the universal service obligation. Oftel has made clear that, while its current view is that including broadband access within the universal service obligation is not appropriate now, it will keep it under close review and will re-examine the case for inclusion in the light of developments in the market.

10.8 The obligation on BT to supply loops for the provision of the PBLC service does not affect BT’s current universal service obligation to supply telephone lines on reasonable request. Ultimately a change in the competitive landscape might require a different approach to the funding of universal service. Oftel is currently undertaking a review of BT’s universal service obligation and will continue to review the situation in the light of market developments.

Back to Contents

A1 In July 1999, Oftel put out for consultation the proposal that BT should be required to permit access to its local loop.

A2 Oftel received over 55 responses to the consultation document from a wide variety of sources including.

• Telecommunications operators

• Equipment suppliers

• Service providers

• Television companies

• Consumer groups, including,

- Individual consumers concerned about the price of access to higher bandwidth services.

- Groups with special needs.

- Telecoms advisory bodies and organisations representing rural or economically disadvantaged    areas.

A3 On the whole Oftel’s proposals as set out in the consultation document has been well received with the majority of respondents welcoming Oftel’s proposal to mandate some form of access to BT’s local loop (option 2 partial baseband leased circuit as set out in the July consultation document) rather than leave it to BT to offer higher bandwidth services via it option 4 offering (permanent virtual circuit access as set out in the July consultation).

A4 Key issues raised in response to the consultation include

• The timetable for the implementation of option 2. Many respondents questioned Oftel’s proposal to introduce option 2 by July 2001.

• The price of BT’s wholesale option 4 product.

• The need for Oftel to be actively involved in the OPF sub group and the NICC sub group looking at technical, operational and procedural issues

• Oftel’s decision to mandate access to BT’s local loop should be reviewed sometime between three and five years.

• Prices for option 2 should be based on LRIC.

The information below is a brief summary of the responses received.

A5 With one or two exceptions, the majority of network operators that responded to the consultation agreed with Oftel’s view that there is a need to take some form of regulatory action to meet the demand for higher bandwidth services rather than leave it to BT to roll out and develop higher bandwidth services. They suggested that the price for option 2 should be based on LRIC, it should be reviewed sometime between three and five years and should be available to consumers, SME’s and large business users.

A6 The major concern seemed to be that the proposed timetable for the full implementation of option 2 by July 2001 would give BT a head start in the provision of higher bandwidth services. This, they suggested, would serve to reinforce BT’s dominance in the access market. The proposed timetable needed to be brought forward by between six to nine months or BT should be prohibited from rolling out it option 4 product until such time as the option 2 product was available. Operators were also concerned that the price for option 4 based on retail minus was prohibitive and that BT’s products did not give network operators any flexibility in offering innovative higher bandwidth services to customers.

A7 A selection of these comments is highlighted below

• Oftel has failed to consider/overlook a number of fundamental economic points relating to the decision to mandate LLU, primarily the nature of risk associated with network investment to deliver broadband services. Oftel’s proposals to base the price for option 2 on LRIC is derived from a regulatory need to track/prevent cross subsidy. An approach based on ECPR is better as this leads to an economically efficient outcome as it makes proper allowance for the risk and investment associated with broadband development.

• Network Operators are very concerned about the RA’s proposal to introduce a radio interference standard to control emission from wired telecoms network. The proposed standard is far to onerous and could also impact on existing services such as ISDN. Respondents have questioned the need to set such a low emission level and also express grave concern about the way the RA is consulting the telecoms industry. They would like the RA to embark on a proper public consultation so as to resolve the debate in a satisfactory manner.

• Network operators recognise that a substantial amount of work has to be undertaken before option 2 can be made available. Operators intend to work within the NICC and OPF groups to ensure that this happens. It is not at present clear if the timetable is too long or too short. The OPF and the NICC sub groups need to identify what is realistic before a decision on timing can be made. Oftel also needs to clarify the legal vires for the mandating of LLU as soon as possible as it would be very unfortunate if the roll out of option 2 was held up by legal arguments.

• Oftel should secure written undertakings from the BT board that they will ensure that adequate resources will be committed to ensure that option 2 is fully implemented by July 2001. Oftel should also modify BT’s licence to comply with these actions.

• Option 4 Prices – the current proposal of Retail Minus allows BT to keep all the profit for itself and is anti-competitive. Oftel should mandate pricing in one of the following ways.

• Prices should be LRIC based with an appropriate rate of return to reflect BT’s investment risk or

• Prices should be based on retail minus with a requirement that BT’s service business should earn an appropriate rate of return as an independent and profitable entity.

• Distant co-location is a viable alternative to the complications of co-location within BT’s buildings.

• Option 2 should not be restricted to copper. If fibre has been deployed then it should be made available.

• Option 2 should not be restricted to consumers and SMEs. The regulatory framework should reflect the view that option 2 should be available to everybody.

A8 As stated in the previous summary of responses (Annex A July Consultation) the key issue for network services and enhanced service providers was the need for the mandation of some form of open access to BT’s network. Therefore Oftel’s proposal to mandate option 2 was generally well received by this interest group, as being a solution that would allow real competition to develop. This, they suggested, would lead to the development of innovative new service provision and lower tariff packages for consumers, and would, in turn lead to an increase in choice, quality and value for money for all customers. Concern was expressed about the proposed timetable for the introduction of option 2 and about BT’s proposed wholesale price for its option 4 products.

A9 There was no convergence of views from this group with some respondents welcoming Oftel’s proposals and others arguing that it would lead to the provision of inefficient telecoms services. However, the main consensus from this group was that due thought should be given to resolving/addressing all the technical and procedural issues within the NICC and OPF sub groups. Appropriate standards should be used and the integrity of the radio spectrum should be maintained.

A10 The affordability and accessibility of higher bandwidth services was the central theme of the majority of respondents from this group. In general they agreed with Oftel’s proposals to mandate some form of access to BT’s local loop as this would stop BT form increasing its dominant position in the telecoms market and lead to increased choice for consumers. Specific issues commented upon were the need to ensure that all sectors of society were able to gain access to higher bandwidth services particularly those in rural and economically disadvantaged areas.

An electronic version of the majority of the responses can be found on the Oftel web site.

Back to Contents

B1 Option 2 is where BT provides a leased line between the end user's premises and the local concentrator/exchange building.

B2 The other licensed operator (OLO) leases the Partial Baseband Leased Circuit (PBLC), which extends from the Network Terminating Point (NTP) at the end user's premises to a NTP at the local concentrator/exchange building, or at an alternative interconnection point specified by the OLO.

B3 By installing equipment at each end of the PBLC the OLO can convert the circuit into a high-speed data link. In order to complete this high-speed link the equipment at the exchange end must be connected to a broadband network.

B4 The most likely physical arrangement for option 2 is that BT will extend the circuit from their Main Distribution Frame (MDF) and terminate it on a distribution frame that the OLO will have sole access to. The OLO would then be in a position to configure their distribution frame and equipment and connect to their network. Figure 1 shows a schematic for physical co-location.

B5 Under this arrangement BT would be responsible for the integrity of the circuit but not the equipment connected to it. However because of the potential problems that can be caused by cross talk, the connected equipment must operate and be connected within the agreed rules. These rules are the subject of the NICC DSL Task Group as part of the implementation process.

B6 It is considered that the PBLC is similar to BT's existing analogue private circuit products, especially the baseband private circuit. The main difference is that BT's existing products provide a connection between two premises and thus use two local end links, ie the circuit connecting the end user's premises to the local exchange. Whereas, option 2 only uses one local link and this is why it has been termed Partial Baseband Leased Circuit.

B7 Three different forms of co-location have been identified: physical, distant and virtual. However, only the first two of these are being addressed by Oftel as virtual co-location is considered to be a matter for commercial agreement between BT and other operators. A brief description of each form of co-location is given below.

B8 Physical co-location is where the OLO rents space in BT’s exchange building in order to locate the equipment needed to interconnect with the PBLCs. In this arrangement the OLO will own and maintain its equipment and will thus require access to the exchange. A schematic of this arrangement is shown in Figure 1.

B9 Distant co-location, sometimes referred to as adjacent co-location, is where the OLO locates its equipment in a building that is independent to BT's exchange. In this arrangement a tie circuit is required between the MDF, in BT's exchange, and the location where the OLO's equipment is housed. BT will be responsible for installing this tie circuit which could be extended all the way to the OLO's premises or terminated in a street cabinet to give the OLO the flexibility to extend the circuits itself. As with physical co-location the OLO will own and maintain its equipment, however they will not need to rent space from BT or have access to BT's exchange. A schematic of this arrangement is shown in Figure 2

B10 A perceived disadvantage to distance co-location is that as the length of access pairs is increased the potential achievable performance is reduced.

B11 Virtual co-location is where the OLO contracts BT to house and maintain its equipment. One of the perceived problems with this form of co-location is that BT staff would need to be trained to operate the OLO's equipment, if it is different from BT's, and this could make it an expensive option. Of course the OLO could elect to use the same equipment as BT or even allow BT to supply the equipment, if this happens then this arrangement begins to look similar to Oftel's option 3.

B12 In this arrangement the OLO may own the equipment but will not be required to maintain or operate it, thus after the OLO's network has been extended to BT's exchange the OLO will rarely, if ever, need access to the exchange.

B13 Virtual co-location is considered to be a matter for commercial agreement between BT and other operators.

B14 Many variants of DSL technology can be provided simultaneous to the existing telephony service, on a single access pair. Thus existing access pairs can be upgraded to a broadband data channel whilst maintaining the telephony service. This arrangement makes good use of the existing infrastructure and could therefore ultimately provide better value to the customer.

B15 If an operator is to provide a traditional telephony service and a data channel then clearly they are likely to configure their services so that they share a single access pair and this is the arrangement adopted by BT for its ADSL products. However if different operators were to supply the telephony service and the data channel over the same access pair, it would be necessary for these two operators to share a line.

B16 To enable two operators to share a line, the line would need to be split at the exchange end and possibly also at the customer end, so that both operators can gain access. Technically, splitting a line is not particularly complicated, however one of the operators must provide and be responsible for the splitter and this can raise a number of difficulties.

B17 The difficulties arise because one of the operators would be dependent on the other to correctly implement and maintain the split. This means that the dependent operator would not be able to fully test the line and thus may have difficulties with fault finding and ensuring quality of service. Additionally the operator controlling the split would only be able to fully test the line by interrupting on the other operator’s service. Finally line sharing between independent operators could lead to customer confusion in the event of a problem.

B18 Many of these issues relate to operational difficulties between the two operators sharing the line. Oftel therefore considers that for maximum simplicity any line sharing is best arranged through an operator purchasing the line as a PBLC in its entirety, splitting and passing the low frequency portion back to BT for the onward conveyance of telephony The other operator is then effectively reselling BT’s telephony service, whilst providing the billing and customer service functions.

B19 Voice over DSL (VoDSL) is currently in an advanced state of development with prototype systems up and running. To achieve VoDSL, the analogue to digital conversion process must be performed at the customer's end, such that the voice can be transmitted as a data stream over the DSL link. Moving the analogue to digital conversion process to the customer end is not new and has been employed by ISDN phones for some years. The main difference between ISDN phones and VoDSL is that an ISDN B channel is 64 kbits/sec and is thus directly compatible with the PSTN, whereas the DSL link is likely to have a capacity greater than 64 kbits/sec. Even if VoDSL is coded at 64 kbits/sec, which seems sensible, it will only occupy only part of the DSL channel and therefore must be separated and conditioned before it can be connected to the PSTN. The PSTN interconnection could take place within the local exchange or any other suitable interconnect point.

B20 An alternative to physical interconnection with the PSTN is to provide voice over IP over DSL thus the interconnection would take place at the transport/network layer.

B21 One of the limitations with VoDSL is that the service requires power at the customer end, thus in the event of a power failure the service would be lost. Unless there is a one-hour battery back-up, this type of service is not suitable for a customer’s primary telephony service as it does not meet the 999 access requirements.

B22 The PBLC will be required to meet a certain level of quality and this level must be maintained. Before an operator takes delivery of a PBLC, BT will be required to perform a pre-qualification test to ensure that it meets the required quality level.

B23 If the required quality level is set too low it may be difficult to upgrade the PBLC, using current technology, to a standard expected by end users. Equally if the required quality level is set too high then a lower proportion of the existing access infrastructure will be able to meet it and this will impact the potential penetration level of option 2. Because of this relationship, quality level and required penetration level must be considered together.

B24 Essentially there are two alternative methods that could be adopted for specifying the required quality of a PBLC.

• Bench Mark Performance, ie minimum data rate using reference equipment, as currently adopted for voice bandwidth data.

• Electrical characteristics, ie attenuation, return loss, cross talk, balance, etc.

B25 Superficially the first option appears to be the simplest, both in terms of specifying and enforcing and it has clear synergy with the existing minimum data rate standard. However the second option is probably the most rigorous and flexible in terms of future developments, as electrical characteristics are technology neutral.

B26 As indicated there are many features and parameters that could be used to describe the quality of a PBLC and the identification of which of these should be used and the values associated with them is one of the areas being addressed by the NICC DSL Task Group.

B27 Although BT's copper access pairs are substantially under-utilised, which is why there is so much scope to upgrade them using technologies such as DSL, they do have a limit and this limit is determined predominantly by cross-talk. Cross-talk describes the situation where signals on one access pair are coupled into adjacent pairs within the same access cable. Strictly speaking cross-talk is not limited to a single access cable but in reality this situation will dominate and hence becomes the limiting factor.

B28 To ensure that BT's access network can successfully be used to provide broadband access to a high proportion of the users connected to it a spectral management plan is needed. A spectral management plan needs to consider three interrelated areas and these are identified below.

• Frequency mask for the individual access pairs, ie maximum allowable signal power as a function of frequency.

• Cross-talk within the access cables. Consider mixing different services within an access cable.

• Electromagnetic radiated emissions. The frequency mask(s) will need to be considered in conjunction with the characteristics of the access network to ensure the radiated emissions are kept at an acceptable level.

B29 The NICC DSL Task Group has been actioned to produce a spectral management plan for BT's copper access network. The output from this group will define the frequency mask(s) and the rules for connection, in accordance with the first two points identified above. However they must be mindful of the radiated emissions limits and these are currently being address by the Radiocommunications Agency (RA).

B30 Before the plan and rules can be developed the objectives for them need to be agreed. These objectives will specify the characteristics and amount of the data to be carried on BT's access network. For example how much data is required to travel upstream and downstream (symmetric/asymmetric) and what potential penetration level should be assumed? These objectives need to consider current and anticipated future requirements of end users but at the same this must be balanced against the broader objectives of the Access to Bandwidth initiative, which is to make broadband access widely available.

B31 Oftel’s high level objectives for the spectral management plan (SMP) are:

• High customer penetration

• Technology neutral

• Minimal management processes

B32 It is Oftel’s goal to ensure the availability of broadband access to the largest proportion of the UK population as possibly. Ideally this would be 100%, however as discussed in paragraph B23 this is unlikely to be achievable using only BT’s copper access network, instead a trade-off between the quality of the PBLCs and the potential penetration level is required. Based on consultations with various technical groups Oftel considers that a customer penetration level of 80 – 85% should be used as the design objective for the SMP.

B33 To ensure that the SMP is flexible, non-restrictive and contains a degree of future proofing Oftel believes that it should be defined in a technology neutral fashion. For this reason Oftel would favour the use of power spectral density (PSD) masks to describe the signals that can be connected to the network at different locations. It is recognised that at the present time ADSL and HDSL are likely to be the main technologies deployed, indeed HDSL in two and three pair form is already widely deployed in the network. It is therefore considered that the PSD masks should allow for the deployment of these technologies.

B34 In order to minimise the management costs associated with option 2 and to ensure its timely introduction, Oftel believes that the SMP should be produced in such a way as to avoid the need for complicated and restrictive management processes. To this end mix quotas for certain services within a binder group and binder group configuration control should be kept to a minimum and if possible avoided altogether.

B35 The deployment of equipment that does not comply with the SMP has the potential to not only interfere with other systems but may also be interfered with by the other systems. This potential situation demonstrates that all operators that take up option 2 will have a common interest in ensuring that the SMP is correctly implemented. Oftel would therefore encourage BT and other option 2 operators to mutually agree on the suitability of new technologies before deployment.

B36 To promote competition and innovation in the equipment market it is suggested that equipment manufactures be allowed to sponsor tests to demonstrate compatibility with the SMP.

B37 As discussed in paragraph B35, correct implementation of the SMP is of common interest to all operators. Oftel therefore considers that self policing should be adopted by the industry at this stage. However, if at a later date there appears to be an unacceptable number of problems and disputes Oftel will re-address this issue.

B38 All electrical equipment radiates electromagnetic energy, regardless of whether this was the design intention or not. This radiation needs to be controlled (limited) so that the performance of systems that use the radio spectrum, ie communications and beacons, is not unduly degraded.

B39 The proposed use of BT's copper infrastructure to provide broadband access raises new, and as yet unresolved, issues in terms of radiated emissions. BT's access network is ubiquitous and therefore any radiation from it will also be ubiquitous, this differs from previously considered sources, which tend to be relatively localised. Also the anticipated technologies will operate in the frequency range below 30 MHz, and there appear to be no suitable European EMC standards that address radiated emissions in this frequency range.

B40 To address this shortcoming the RA has produced a draft enforcement standard, MPT 1570, Radiation Limits and Measurement Standard, July 1999. It is not the intention that systems need to comply with this standard before they can be deployed instead this standard will be used to determine whether enforcement action can be taken in the event of a complaint.

B41 Originally MPT 1570 was produced to address power line transmission (PLT), this is where the mains power lines are used as the access route, and has only recently been considered applicable to DSL deployment. For this reason some of the details of the standard are being re-addressed with DSL deployment in mind and also potential DSL users and manufactures are being consulted.

B42 Many of the technologies being considered to upgrade the local access network actually require the professional installation of a splitter at the customer premises. Because of the time and cost implications associated with this requirement it has been considered that it could potentially limit the roll out of broadband access. For this reason customer installable solutions have been investigated. The main thrust of these solutions is to remove the need for a splitter at the customer end and thus they can generically be referred to as splitterless DSL. There are two main concerns with splitterless DSL and they are.

• Interference issues between the DSL modem and existing customer premises equipment (CPE).

• DSL data being conducted on the 'uncontrolled' user wiring which may give rise to potentially high levels of radiated emissions.

B43 The interference issues can be addressed by ensuring that the DSL modem has a high pass filter installed, which is normally the case, and by installing an in-line low pass filter with each CPE, effectively producing a distributed splitter. The in-line low pass filter would be a plug in device that could be installed by the customer. However, the emissions problem is more fundamental and the only way to reduce it is to change the characteristics of the signals transmitted into the wiring.

B44 DSL-Lite is essentially a cut down version of standard ADSL, it uses lower transmission power and only operates in the lower part of the frequency range. These features reduce the potential operating distance and data rate respectively. Because of this reduced performance and the conception of distributed splitters the deployment of DSL-Lite is likely to be limited. Instead operators may choose to deploy standard ADSL with a distributed splitter arrangement. BT is already considering this option.

NICC DSL Task Group – Report on the use of DSL Technology in the UK

• Part 1: Interference Issues

• Part 2: Interoperability Issues for DSL-Lite Technology

• Addendum to Part 1: Interference Issues

• Addendum to Part 2: Interoperability Issues for DSL-Lite Technology

• Conclusions of the NICC DSL Task Group

Radiocommunications Agency – Radio Technology and Compatibility Group

• Draft MPT 1570, Radiation Limits and Measurement Standard – Electromagnetic radiation from telecommunications systems operating over material substances in the frequency range 9 kHz to 300 MHz

Back to Contents

C1 The option 2 service that Oftel intends to mandate is a Partial Baseband Leased Circuit linking the end-user premises to BT’s local exchange buildings. This section sets out the legal framework through which Oftel will require the introduction of this service.

C2 It is important to bear in mind that while commonly referred to as ‘leased’ circuits, PBLCs and other types of circuit provided by BT do not in fact involve a leasing arrangement in the strict legal sense. Rather the circuits are provided as a telecoms service to other operators and remain part of BT’s ‘applicable systems’ and are managed and maintained by BT under the terms of the contract with the purchasing operator.

C3 As there are already conditions in BT’s licence that relate to the provision of telecoms services and other ancillary services, the requirement for BT to provide the PBLC service and attendant co-location can be exercised through BT’s licence as it currently stands. However, as Oftel wishes to set out the obligation clearly for the avoidance of doubt and establish price control on this service, Oftel intends to put forward a new licence condition that will specifically set out the relevant service obligations. This will be done under Section 12 et seq of the Telecommunications Act 1984 (The 1984 Act). The analysis that follows details existing general licence conditions that are relevant to this new service and the process Oftel will follow to propose the specific amendment to BT’s licence.

C4 The PBLC service as described in option 2 is managed by BT and consists in the conveyance by BT’s system from the local exchange to the customer of matters mentioned in Section 4(1) of the 1984 Act. It is thus a telecommunication service as defined in Section 4(3) of that Act.

C5 This service can be mandated under condition 43.2 of BT’s licence, which requires BT to supply telecoms services not consisting in voice services on reasonable request. Condition 43.2 provides:

C6 The Director General considers that in relation to the provision of PBLCs, the conditions set out above have been satisfied. The Director considers that, from representations from operators and through the consultation process, demand for this service has been clearly identified (see chapter 2) and that this demand cannot currently be met by other means. The Director General therefore accepts, in general terms, that a request from an operator for this service is reasonable.

C7 There may be certain circumstances in which it might not be reasonable to expect BT to provide these circuits. These circumstances are set out in general terms in condition 64.5 of the BT licence and include circumstances in which:

1. There is no reasonable demand for the service
2. Provision of the service would raise health and safety concerns
3. Where the Licensee is unable to obtain anything necessary to provide a service of the quality or standards required by the person who requests provision of the service. In the event of a dispute, the Director General of Oftel’s decision as to whether anything is necessary shall be final.
4. Where the Director General is satisfied that it would not reasonably be practical to provide the service requested at the time or place demanded
5. Where revenue gained is not likely to cover the costs reasonably incurred in supplying service to a particular area

C8 In order to make use of the service an operator purchasing it will need to connect its system to the BT-provided circuit, one of BT’s applicable systems. This interconnection of the two systems is provided for in Condition 9.1 of BT’s licence. Condition 9.1 provides:

C9 In addition to interconnection itself, BT is required to provide ‘other services’ to secure that points of connection between BT’s Applicable Systems and the other operator’s telecommunication system are established. Condition 9.1(b) of BT’s licence provides that BT shall negotiate to provide:

C10 Because of the nature of PBLCs and the uses to which they may be put, the network connecting apparatus of the OLO needs to be as close as possible to the end of the circuit. The Director General thus considers that physical co-location facilities established by BT at the local exchange or concentrator site are reasonably required as an adjunct to interconnection. BT should also provide interconnection facilities for operators choosing to use a distant co-location facility at a point outside the BT exchange building. The characteristics of these types of co-location are discussed in more detail in Annex B.

C11 In certain circumstances it may not be reasonable or feasible for BT to provide co-location at its exchange or concentrator site. Such circumstances are outlined in general terms at Condition 9.4. This provides that:

C12 These circumstances could arise for example where BT could demonstrate that there was insufficient space to support physical co-location. In these cases BT would still be required to provide a tie circuit link to a distant co-location facility established by operators wishing to make use of the PBLC service.

C13 Condition 9.2(b) provides that BT:

C14 Oftel intends to draw upon these procedures in resolving any disputes relating to interconnection or attendant co-location that may arise.

C15 As discussed, Oftel considers that it can require BT to provide the PBLC and attendant services through BT’s existing licence. But Oftel recognises that the introduction of option 2 is a significant decision, and that it is important for all that the requirement for BT to provide the service is clearly and unambiguously set out. Therefore, without prejudice to the relevant provisions set out above, Oftel proposes, for the avoidance of doubt, to set out in the licence specific details of the service to be provided.

C16 The licence condition that Oftel intends to introduce will include as a minimum:

• Description of the service to be offered (The PBLC)

• Details of which operators will be entitled to make use of the service (Schedule 2 Operators)

• Circumstances in which it is to be provided

• The pricing regime

• Description of the co-location services to be provided

C17 The statutory consultation period provided for in section 12 of the Telecommunications Act requires a minimum of 28 days for interested parties to make representations on the proposed modifications, and Oftel’s practice is to allow further time (usually 14 days) for comments on these representations. Oftel intends to consult informally early in 2000 and publish the proposed licence condition for formal consultation by April 2000.

C18 The unbundling of BT’s loop raises the prospect that other operators and end users might install equipment that could interfere with services being supplied to other customers on BT’s network. This includes other customers taking DSL services and customers taking standard telephony. It is clearly essential that appropriate enforcement action can be taken against operators that cause this type of interference or harm to the network.

C19 Oftel expects that the contract between BT and OLOs taking the PBLC service will specify the criteria that equipment attached to the network will need to meet. It is intended that these criteria will be based on independent reports to be published by the NICC DSL task group, and will cover rules both on the characteristics of the equipment that may be deployed and the manner in which it may be deployed.

C20 The conditions on spectral management and deployment processes set out in the contract will need to be consistent with or refer to the Essential Requirements provisions set out in article 13 of the RVTD and transposed as Condition 20 of BT’s licence. This provision requires that operators set out transparent and objective criteria for the circumstances in which they would expect to be able to restrict access to and use of their network in order to maintain network integrity. These criteria need to be published in advance and open to consultation. They may contain rules relating to the connection of telecoms equipment (such as DSL) to BT’s network, and to home networking equipment where it is configured in such as way that emissions could leak into the network.

C21 Providing BT has set out suitable objective criteria in advance, BT will be entitled to disconnect DSL equipment where it infringes the requirements of the spectral management plan and deployment processes. Equally redress will be available to other operators if BT takes unjustified action to disconnect an operator. Oftel would be able to investigate such claims as a breach of BT’s licence.

C22 In addition, equipment attached to the customer premises end of the circuit (telecoms terminal equipment) will need to meet the essential requirements set out in article 3 of Directive 99/5/EC (RTTE), which will be incorporated in UK law from April 2000. Under the RTTE, DSL equipment will only be required to meet the compulsory requirements relating to safety, electro-magnetic compatibility and effective use of the radio spectrum at article 3.1(a) and 3.1(b). Oftel is examining the case for recommending that an application should be made to the Commission, under established RTTE procedures, that DSL equipment should also be required to meet the additional essential requirement set out in article 3.3(b) of the RTTE that terminal equipment must not harm the network or its functioning.

Back to Contents

D1 The costs and revenues used in the CBA are from a model of an ADSL roll-out constructed for Oftel by Analysys. This model examines the costs and revenues that would result from rolling out an ADSL service to each exchange that Analysys estimates that it is profitable for the incumbent to roll out to.

D2 The model splits customers into three categories:

• Business;

• SOHO (small businesses and home workers); and

• Residential.

D3 For each category of customer, Analysys has estimated the number of customers taking up the service. The prices used are estimates based on discussions with BT on the trial prices and likely product offerings. The Analysys model has assumed that residential customers will opt for a 500k product. Two 500k products have been assumed, with the difference being in the contention ratio. It is assumed in the base case that 70% of residential customers opt for the lower priced 500k product, with the remaining customers opting for the more expensive 500k product.

D4 Customer numbers and prices, along with the annual revenue from the service, are shown in Annex 1.

D5 The costs incurred in supplying an ADSL service can be categorised as:

• Customer premises/terminating equipment;

• DSLAM;

• DSLAM – ATM link;

• ATM network; and

• IP network.

D6 The Analysys model includes the first three of these categories, but not the costs of the ATM and IP networks. It seems reasonable to include the costs of the ATM and IP networks in the CBA because the tariffs used in the Analysys model are BT’s original Option 4 prices, and these prices include conveyance on the ATM/IP networks. The inclusion of the ATM and IP networks is important when considering both the costs of Option 2 that arise due to an increase in average cost, and the benefits of Option 2 that arise due to an increase in cost efficiency. Analysys has not provided an estimate of these costs, so the costs used in the base case are estimates supplied by BT. An estimate of the total cost of the ADSL service (Analysys estimate plus ATM and IP networks) is shown in Annex 1.

D7 The first step in the CBA is to define a base case against which the costs and benefits of introducing Option 2 can be measured. The base case should represent the costs incurred and revenues gained by BT from its ADSL roll out if Option 2 does not happen.

D8 To define this base case, one needs to make a judgement about the level of competition that BT’s ADSL roll-out would face in the absence of Option 2. This is subject to considerable uncertainty. This is partly because it is not clear how close a substitute for ADSL other technologies such as cable modems, digital satellite, third generation mobile and broadband fixed radio access will become, and partly because there is uncertainty about the speed at which these other technologies will be introduced.

D9 This uncertainty can be dealt with by constructing a central base case and then testing alternative base cases as part of the sensitivity analysis. The central base case assumes:

• Option 2 is introduced in July 2001;

• At the second review of Option 2, which takes place in 2007, Oftel takes the view that the market will be effectively competitive in four years from the date of the review;

• BT is not required to make any further loops available to Option 2 operators;

• After a period of four years from the date of the review, Option 2 operators lose the right to use BT’s loops.

D10 Two alternative base cases are tested as part of the sensitivity analysis. These are:

• Option 2 remains the only effective competition to BT’s ADSL service throughout the period examined by the CBA. This scenario could arise if there are no close substitutes for ADSL. It could also arise if competition to ADSL exists from other technologies, but there are not sufficient substitutes to create effective competition.

• At the first review of Option 2, which is due to take place in 2005, Oftel takes the view that the market is effectively competitive. No further loops have to be made available to Option 2 operators. Four years from the date of the review, Option 2 operators lose the right to use BT’s loops.

D11 Option 2 imposes a resource cost on the economy because:

• Increase in average cost. If all costs are variable and subject to constant returns to scale, the average cost of supplying customers will be independent of the number of competitors in the market. However, when costs are fixed, or subject to economies of scale, a reduction in the number of customers supplied will lead to an increase in the average cost. So the impact of competition will be to increase the average cost of supplying an ADSL service.
  
• Option 2 handling and collocation costs. Option 2 imposes costs that are over and above the costs that would be incurred under Option 4. For example, Option 2 will require development of BT’s operational systems (eg order handling, fault reporting) and product design and development. Option 2 will also require work to take place on BT’s exchanges in order to allow collocation.

D12 The benefits of competition can be split into Type I benefits and Type II benefits. Type I benefits accrue to those customers who take the service of a competing ADSL supplier. These benefits are:

• Greater cost efficiency of new entrant;

• Lower price of new entrant due to reduction in the profit margin and greater cost efficiency

• Increase in the size of the market due to innovation in services.

D13 Type II benefits accrue to those who continue to take the service of BT as an ADSL supplier. These benefits arise due to the fact that BT will respond to competition by:

• Increasing cost efficiency

• Reducing price because of a reduction in the profit margin and greater cost efficiency

D14 It is normal in a CBA to separate out the Type I and Type II benefits. However, for simplicity it is assumed in this CBA that the result of competition is to force BT to match the efficiency and price of the new entrant. This enables us to examine the effect of greater cost efficiency and lower prices on the market as a whole.

D15 Option 2 will require new entrants to construct their own DSLAM-ATM links and ATM/IP networks. As discussed above, this can impose a cost if it causes a loss in economies of scale. However, it can also bring a benefit. Once Option 2 is withdrawn, Option 2 operators will be able to continue using their ATM and IP networks, and may also be able to continue using the DSLAM-ATM links. This mitigates the need for other operators to invest in these assets, and so brings a benefit in terms of saved investment.

D16 In order to measure the costs and benefits associated with the introduction of Option 2, a number of assumptions have to be made. The main assumptions used in this CBA are shown in Table 1.

D17 The CBA examines the costs and benefits from 1999/2000 to 2019/2020. A real discount rate of 12.5% is used. This gives a conservative calculation of net benefits, since it could be argued that it would be appropriate to use a lower rate, perhaps as low as the Government’s Test Discount Rate of 6%. A lower discount rate would increase the present value of the net benefits.

D18 A central estimate of the net present value of the implementation of Option 2 is obtained from the CBA. This central estimate is converted into a range in order to take account of the uncertainty surrounding many of the assumptions. The alternative assumptions used to produce this range are shown in Table 1.

D19 The additional costs incurred due to competition can be estimated by comparing the total cost of the service as supplied by one firm with the total cost if a number of firms are involved in supplying the same number of customers.

D20 The scale of the duplication of costs will vary with the number of competitors; the greater the number of competitors, the greater the duplication of cost. In the CBA the central assumption used is that there are three competitors in the market, each supplying a third of the demand.

D21 There are two effects that can lead to an increase in cost per customer when competition is introduced. These are:

• Loss of economies of scale

• Indivisibility of assets

D22 In the CBA, only the loss of economies of scale is considered to be a cost of Option 2. Current indivisibilities are likely to disappear as operators use the lower capacity equipment that is currently available, and manufacturers adapt their products to meet the remaining needs of smaller operators offering an ADSL service.

D23 The Analysys model indicates that cost inefficiency is driven by losses of economies of scale and indivisibilities in DSLAMs, DSLAM-ATM links and development.

D24 Ideally, one would hope to separate out these two effects and include a direct estimate of the costs associated with the loss of economies of scale. However, this is not possible using the Analysys model. The model was not designed to look at competitive supply of ADSL, so it does not take account of cost reductions that could be made as customer numbers fall.

D25 For example, it may be possible for an operator to use a lower capacity DSLAM-ATM link than that included in the model. It is assumed in the model that all DSLAM-ATM links are 155Mbit/s. It may also be possible, either now or at a later date, to use a smaller DSLAM than is assumed in the model. It is assumed in the model that a DSLAM serves 128 customers.

D26 The inability of the model to separate out these two effects means that we must make an adjustment to the figure the model produces as the cost of Option 2 that arises from inefficiency. It is assumed that 50% of the cost is due to losses of economies of scale. This assumption is tested as part of the sensitivity analysis.

D27 Estimating the costs that are incremental to the introduction of Option 2 is more difficult. This CBA includes BT’s estimate that the additional cost involved in setting up the Option 2 service will be £10m. It is assumed that this cost is incurred in 1999/2000.

D28 This figure does not include the costs incurred due to collocation. These costs will be one-off, such as the need to construct walls or cages to isolate OLOs’ equipment, and on-going, such as the additional costs involved in escorting representatives of the OLOs through BT’s exchanges.

D29 The CBA includes one-off costs of £5000 per exchange where Option 2 is demanded. Option 2 is assumed to be demanded only in those exchanges that the Analysys model indicates would be profitable for BT to roll out its ADSL service to. This gives a total of 2240 exchanges, leading to a total one-off cost of £11.2m. This cost is assumed to be incurred in 1999/2000. The on-going cost is assumed to be £1000 per year for each exchange. This leads to a yearly cost of £2.24m. This cost is assumed to be incurred in each year from 2001/2002.

D30 Table 2 shows that the central estimate of the costs of Option 2 is a net present value of £210m. Most of this cost is dues to losses of economies of scale. This accounts for 90% of the total cost of Option 2.

D31 The first benefit is greater cost efficiency of both the new entrant and BT. Estimation of this benefit involves examining the total annual cost of an ADSL service, which is dependent on the number of competitors, and applying a reduction on the basis of a reasonable assumption about the extent to which competition may result in greater efficiency.

D32 The central assumptions included in the CBA are:

• The presence of three competitors in the market will lead to costs being 10% lower in each year up to and including the year of the review (2007/8).

• Some benefit from cost efficiency continues until Option 2 is fully withdrawn.

• The benefit from cost efficiency reduces linearly from 10% in 2007/08 to 0% in 2011/12.

D33 The second benefit is the reduction in price. A price reduction leads to higher demand for the service and this leads to a welfare gain. The demand curve used in the CBA is in the form:

q = a – bp

with an elasticity of demand of 0.5 at the prices shown in Annex 1.

D34 The central assumptions are:

• Competition will lead to all competitors charging a price that is 25% lower than the price contained in the Analysys model in each year until 2007/08.

• Option 2 will continue to have some impact on prices until it is fully withdrawn in 2011/12.

• The impact of Option 2 on prices reduces linearly from 25% in 2007/08 to 0% in 2011/12.

D35 The third benefit is the growth in the overall size of the market that occurs due to innovation. Innovation will occur due to Option 2 operators having freedom over all technical aspects of the service provided to the customer. For example, an Option 2 operator could use an alternative technology over the local loop.

D36 An increase in market size brings four benefits:

• Benefit to customers who purchase broadband services over the local loop as a result of the innovation that is a feature of Option 2.

• Benefit to customers who are already purchasing ADSL, but who switch to the new innovative services that they value more highly.
• Benefit to producers from supplying more customers.

• Benefit to all customers due to a reduction in average cost.

D37 Using the same demand curve, it is possible to calculate the benefit to both consumers and producers of an increase in demand.

• The central assumptions used in the CBA are that:

• In each year up to 2011/12, demand is 10% higher than in the base case.

• The benefits from innovation do not disappear with the removal of Option 2.

• The benefit from innovation falls to 5% in 2011/12, and declines linearly until it reaches 0% in 2016/17.

• There is no further benefit from innovation after 2016/17.

• Higher customer numbers tends to lower average cost. The erosion of the benefit from this cost reduction mirrors the erosion of the benefit from innovation.

• 10% of existing ADSL customers switch to the new innovative services. This assumption is tested as part of the sensitivity analysis.

D38 One can estimate the benefit arising from this source as the residual value of these assets at the time when Option 2 is fully withdrawn. In the CBA it is assumed that this will take place in 2011/12. It has also been assumed in the CBA that these assets have a working life of twenty years.

D39 The result of estimating these benefits is shown in Table 3. As the table shows, the central estimate of the CBA is that Option 2 will bring benefits in the order of a net present value of £430m.

Central estimate of the CBA

D40 Table 4 shows the impact of bringing the estimates of costs and benefits together. The central estimate of the CBA is a net present value of Option 2 of £220m.

D41 However, this central estimate relies on all of the assumptions in the CBA being entirely accurate. Given the degree of uncertainty involved in modelling the future behaviour of any market, this is unlikely to occur. It is more realistic to examine how the result of the CBA changes as these assumptions vary by reasonable amounts, and produce a range estimate of the impact of Option 2.

D42 Table 1 set out most of the assumptions used in the CBA. These are tested as part of the sensitivity analysis, along with:

• Alternative base cases

• The percentage of existing ADSL customers taking up innovative Option 2 services.
  
• The percentage of the cost inefficiency of Option 2 that is accounted for by the loss of economies of scale.

D43 Annex 2 shows the results of this sensitivity testing. The main results from the sensitivity analysis are:

• The result is robust for changes in any single assumption;

• The net present value of Option 2 is most sensitive to the extent to which effective competition from other technologies exists.

• It is also sensitive to the percentage of cost inefficiency due to the loss of economies of scale, the number of operators competing in the market, and the growth in the market;

• The net present value of Option 2 is relatively insensitive to the demand elasticity used;

D44 The table shows the impact of changing one assumption at a time. This analysis shows that the net benefit of Option 2 is at its lowest where the introduction of Option 2 stimulates only a 5% growth in customer numbers. In this scenario, the benefit from Option 2 is around £120m. The net benefit of Option 2 is at its highest if the innovation the occurs as a result of Option 2 leads to 25% of existing customers switching to the new service, which they value more highly. This scenario produces a central estimate of the benefit of Option 2 as being £440m.

D45 It is possible that more than one assumption will prove to be inaccurate. To test for the robustness of the CBA to multiple changes in assumptions, two scenarios have been constructed.

• Scenario 1: In this scenario, there are four competing operators, the market grows by only 5% as a result of the introduction of Option 2, and no existing Option 2 customers switch to the innovative services introduced by Option 2 operators. This gives a net present value of the introduction of Option 2 of £11m.

• Scenario 2: In this scenario, there are two competing operators, the market grows by 15% as a result of the introduction of Option 2, and 25% of existing customers switch to innovative services which they value more highly. This gives a net present value of the introduction of Option 2 of £496m.

D46 The CBA produces a central estimate of the net present value of the introduction of Option 2 of £220m. The assumptions underlying this central estimate are subject to considerable uncertainty. It is therefore prudent to test these assumptions and consider the result of the CBA to be within the range indicated by that sensitivity testing. The sensitivity analysis shows that this central estimate is robust to any single change in an assumption, and also to multiple changes. The range indicated by the sensitivity analysis is of a net present value of introducing Option 2 of £120m to £440m.

D47 One of the main uncertainties in the CBA is the extent to which other technologies will become effective competitors to BT’s ADSL service. The extent of competition to BT’s ADSL service and the length of time before this competition becomes effective, will have a significant impact on the result of the CBA.

D48 However, Option 2 brings some benefit even if its impact is only to bring competition forward by a few years. On the other hand, there is much to lose from not introducing Option 2. If Option 2 is not introduced and other technologies do not establish themselves as effective competitors to BT’s ADSL service, the welfare cost is likely to be significant.

D49 In conclusion, the result of the CBA supports the proposal to introduce Option 2. Under most reasonable assumptions Option 2 results in a significant net benefit.

Back to Contents

E1 One issue associated with Option 2 is how to deal with the requirement that BT allow OLOs to locate their equipment in its exchanges. This sharing of the BT exchange by other operators is termed ‘co-location’. Co-location will require space in BT’s exchanges to be rented out to OLOs wishing to offer services in that exchange area. It will also require BT to maintain the area in terms of light, heating, etc.

E2 This paper examines three issues relating to co-location, namely:

• To what extent should OLOs bear the cost of co-location?

• How should physical co-location be arranged?

• If OLOs should bear the cost, and there is a shared co-location area, how should the cost of setting up the area be split amongst the OLOs?

E3 The paper concludes that:

• The efficient cost of co-location should be borne by the OLOs.

• If the space available for co-location is able to hold more than one OLO, it should be up to the OLOs to decide whether to have separate rooms or caged equipment. The charge for co-location should reflect this choice.

• If space is not available in an exchange, OLOs should be offered the option of distant co-location.

• If space is available in an exchange, OLOs should be able to choose between physical co-location and distant co-location.

• If OLOs have caged equipment within a general area in an exchange, the charge for co-location should be set on the basis of average cost but with some adjustment for risk.
  

E4 There are six principles that Oftel commonly uses to analyse cost recovery:

• Cost causation                         Costs should be recovered from those whose actions cause the costs to be incurred at                                                   the margin;

• Distribution of benefits            Costs should be recovered from the beneficiaries especially where there are                                                   externalities.

• Cost minimisation                    The mechanism for cost recover should ensure that there are strong incentives to                                                   minimise costs.

• Effective competition              The mechanism for cost recovery should not undermine or weaken the pressure for             effective competition.

• Practicality                              The mechanism for cost recovery needs to be practicable and relatively easy to                 implement.

• Reciprocity                               Where services are provided reciprocally, charges should also be reciprocal.
  

E5 When applying the principles, it is generally sound to start with cost causation on the grounds that economic efficiency is enhanced by requiring parties to pay for costs which they directly cause to be incurred. The other principles are then considered, to see the extent to which this starting point may require modification.

E6 The first issue to consider is the extent to which the OLOs should bear the costs of co-location. Costs arise both as one-off costs, such as those associated with preparing exchanges for co-location, and as on-going costs, such as those associated with maintaining the area of the exchange used for co-location. These costs could be borne entirely by BT, entirely by OLOs, or shared between BT and OLOs.

Cost Causation

E7 BT incurs the costs arising from co-location only if other operators demand use of its local loop. So it is these operators who cause the cost to be incurred and under this principle it is these operators who should bear the costs.

E8 There will be a direct benefit to customers who choose to use the services that OLOs are providing over BT’s local loop. However, there will also be a benefit to those customers who remain with BT. This benefit arises because the ability of OLOs to offer services over BT’s local loop will increase competition for both telephony and broadband services. This would indicate that the cost of co-location should be shared between the OLOs and BT.

E9 It is BT who is in a position to determine the size of the cost of co-location. Recovery of at least some of these costs from BT is required if BT is to have an incentive to minimise these costs.

E10 Unbundling of the local loop will allow OLOs to offer both telephony and broadband services to the customer in direct competition to BT. When BT offers services over its local loop it will bear the costs involved in its use of the exchange. So for effective competition the OLOs should also pay for their use of the exchange.

E11 The easiest system to implement is one where BT bears the cost of co-location. This would prevent the need for an assessment of the scale of these costs and the design of a mechanism for recovering them. However, the difficulties associated with recovering the costs of co-location from OLOs are not insurmountable, so this principle does not give a strong steer.

E12 Reciprocity is not a relevant principle in determining the recovery of co-location costs. While BT may have to open up its local loop to OLOs, there will be no automatic requirement for OLOs to open up their local loops to BT.

E13 The starting point for any analysis of cost recovery is the principle of cost causation. Only if the other principles point strongly towards another method of cost recovery should cost causation not be followed. In this case, cost causation would indicate that the costs of co-location should be recovered from OLOs. This conclusion is supported by the principle of effective competition. The principles of distribution of benefits and cost minimisation would indicate that at least some of the costs should be borne by BT. The principles of practicality and reciprocity do not give a strong steer.

E14 The main rationale for not following the principle of cost causation would be the desire to minimise the cost of co-location. An alternative, if imperfect, way of encouraging cost minimisation is through regulation. It is hoped that charges for co-location can be decided through commercial negotiation between BT and the OLOs. However, in the absence of agreement Oftel would have a role in setting charges, and any charge set by Oftel would be based on the efficient cost of co-location. This should provide some incentive for BT to minimise the cost of co-location.

E15 It therefore appears that, on balance, there is there is not a strong enough rationale to deviate from the principle of cost causation. The benefits gained from ensuring that consumers face the right price signal and that competition is not distorted would appear to outweigh the costs involved in having a weaker incentive for cost minimisation and not following the principle of distribution of benefits. This would indicate that OLOs should bear the reasonable cost of co-location.

E16 Co-location can be physical or distant. Physical co-location occurs when an OLO’s ADSL equipment is placed within one of BT’s exchanges. Distant co-location occurs when an OLOs equipment is placed in a separate building and a link is provided to the BT exchange. It would seem reasonable for BT to offer physical co-location to the extent that there is spare capacity within an exchange. It would also seem reasonable for an OLO to have the choice between physical and distant co-location.

E17 Both the amount of space available for co-location, and the physical structure of that space, will vary from exchange to exchange. In some exchanges the available space will be a series of separate rooms each able to house only one OLO. Other exchanges will have vacant areas with the capacity to house several OLOs. Many exchanges will have a combination of these two types of accommodation.

E18 If there is more than one OLO in an area set aside for co-location, measures must be taken to prevent OLOs having access to each other’s equipment. There are two ways in which an OLO’s equipment can be made secure. The first is to build a room for each OLO to place its equipment in. OLOs would only be allowed access to the room they have been allocated. The second is to place a cage around the equipment belonging to each OLO. OLOs would only be allowed to enter the cage around their own equipment.

E19 There is likely to be a trade-off between security and the cost of co-location. Placing equipment within a room is likely to provide greater security than placing it within a cage, but it is also likely to cost more. It seems reasonable that each OLO be able to make this trade-off for themselves, rather than BT imposing a solution. This could be implemented by BT having a range of prices for co-location. For example, there could be different prices for:

• Co-location within a room built at the request of the OLO; and

• Co-location within a cage built at the request of the OLO.

E20 As discussed above, it seems reasonable for the costs of co-location to be borne by the OLOs. The pricing of co-location is relatively simple if each OLO is located in a separate room. The OLO will pay the cost it causes BT to incur. However, the pricing issue is not so simple if OLOs are sharing an area within the exchange. This will be the scenario if a BT exchange contains an area large enough to house more the one OLO, and the OLOs choose to have a cage around their equipment rather than a separate room. There are two types of costs that we need to consider:

• Variable costs – costs that vary with the number of operators collocating. Variable costs can exhibit constant returns to scale (constant average cost) or be subject to economies of scale (decreasing average cost) or diseconomies of scale (increasing average cost).

• Fixed costs – costs that are independent of the number of operators collocating.

E21 The most significant fixed cost is likely to be the set-up cost involved with preparing the area of the exchange for co-location. However, there will also be on going fixed costs because the entire co-location area has to be maintained (eg, heating, air conditioning, lighting) regardless of the number of OLOs who have chosen to collocate in that area.

E22 There are two ways in which the costs of co-location could be split between operators. The first is to charge each operator the incremental costs that its decision to collocate causes to be incurred. The second is charge each operator average cost. That is, to split the total cost of co-location between the operators.

E23 These two methods of recovering cost are identical where all costs are variable and subject to constant returns to scale. However, complications set in where some costs are fixed or are variable but subject to economies of scale. It is useful to examine both incremental cost and average cost with respect to the six principles of cost recovery.

Incremental Cost
E24 In the presence of costs that are fixed, or variable but subject to economies of scale, a charge set on the basis of incremental cost would lead to differential charging. If some costs are fixed, they will be borne entirely by the first operator because they would be incremental to its decision to collocate. If some of the costs are variable but subject to economies of scale, the incremental cost will fall as the number of operators rises.

E25 This charging structure would comply with the principle of cost causation. Each operator would bear the costs that it caused to be incurred. However, it would not comply with the principle of distribution of benefits. The benefits gained by the customers of the first OLO are no greater or lesser than the benefits gained by the customers of those OLOs who subsequently choose to collocate. Nor does it comply with the principle of effective competition. The principle of effective competition would indicate that the costs should be recovered in a way that does not distort competition between the OLOs. If the fixed costs are of a significant size the first OLO to collocate will have a higher cost base than its competitors. If there are significant economies of scale, each OLO will have a different cost base, with the first OLO again having the highest cost base.

E26 The principle of cost minimisation is not relevant in this scenario because the responsibility for cost efficiency is with BT (and Oftel). Neither is the principle of reciprocity. In terms of practicality, this system would be relatively easy to implement.

Average Cost
E27 Another option for the charging structure is to set the charge at average cost. This would involve dividing the total cost that BT incurs by the number of collocating operators. The effect of taking this approach is to average out the costs that are fixed or variable but subject to economies of scale. This will lead to each operator facing the same charge.

E28 This is not strictly compliant with the principle of cost causation. Strictly speaking, the first operator to request co-location causes the fixed costs to be incurred. However, it is compliant with the principle of distribution of benefits. The customers of OLOs benefit equally and have to bear the costs equally. Operators would be competing on a level playing field, so it is also compliant with the principle of effective competition.

E29 The principle of cost minimisation is again irrelevant because it is BT (and Oftel) who is responsible for cost efficiency. The principle of reciprocity is also irrelevant.

E30 Practicality is an issue for a charge set at average cost. The difficulty the system raises is that the charge to each operator will change as the number of operators collocating changes. For example, suppose that the following scenario exists:

Fixed cost of co-location = 12

Variable cost of co-location = 20 per operator

E31 If there is only one operator choosing to collocate it will have to pay its variable costs plus all of the fixed costs. It will face a charge of 32 (12+20=32). Now suppose a second operator wishes to collocate. It will also pay its variable costs of 20. However, the fixed costs should now be split between the two operators. The second operator will therefore pay a total of 26 (6+20=26) and the charge to the first operator should be reduced to 26 (6+20=26). If a third operator requests co-location the fixed costs are split three ways and the charge to each operator is 24 (4+20=24).

E32 These adjustments could be made in either of two ways. They could be made by BT altering its charges as and when the number of collocating operators changes. Alternatively, the first operator could continue to pay all of the fixed costs, but receive side payments from other operators to pay for their share of these fixed costs.

E33 The only way in which to make this system easier to implement is to decide how many operators to share the total cost between before the service is offered. This would allow the ex-ante setting of charges. However, the problem with this is that the number of operators who may wish to collocate at any exchange in unknown and will vary over time. If the estimate is different to the actual number of operators collocating BT will either over or under recover its costs

E34 One could deal with this uncertainty by simply dividing the total cost of co-location by the number of operators that it is physically possible to collocate in the concentrator unit. For example, in the above scenario if there were room in the concentrator for ten operators to collocate the cost to each operator would be 21.2. However, this system would mean that BT could only recover its costs if the maximum number of operators chose to collocate.

E35 The above analysis indicates that recovering costs on an incremental cost basis is complaint with a strict interpretation of the principle of cost causation. However, it is not compliant with the principles of distribution of benefits and effective competition, whereas a charge set at average cost is.

E36 As discussed above, there has to be a strong reason for moving away from the principle of cost causation. It would appear that, on balance, the arguments in favour of an average charge do provide a justification for moving away from cost causation. This indicates that a charge based on average cost would seem to be preferable.

Adjustment for risk?
E37 However, there is a further principle to consider. This is the principle that those who take risks in order to provide services to consumers should be rewarded for doing so. Otherwise there is no incentive to take the risk.

E38 There are two risks that may be associated with early co-location. The first is the risk associated with being the first operator to offer any service. A second risk exists if the charging system is on an average cost basis. The cost the first operator will face at any time is a function of the total number of operators co-locating, but this is an unknown factor. So the earlier the operator collocates, the greater the uncertainty over the long-term cost of co-location.

E39 The average cost charging system does not provide an incentive for operators to take the risk of early co-location. Those who take risks face the same charge as those who do not. An adjustment to take account of risk could be made. This would involve reducing the charge to the operators who co-locate early, and increasing the charge to the operators who delay offering services. However, the issue becomes complex because of the need to link the reward to operators to the additional risk they take. This involves the construction of a risk profile.

E40 Alternatively, one could produce an incentive for early co-location by charging operators as if they had chosen to co-locate from the start. So any operator who did not co-locate at the start would, on requesting co-location, face back charges for use of the exchange. This would mean that there is a disincentive to delay the decision to co-locate.

E41 If one starts from the principle of cost causation, the appropriate charging system is incremental cost. However, there do seem to be strong arguments for deviating from this principle. Such a system is not compliant with the principles of distribution of benefits and effective competition.

E42 A system whereby each operator is charged the average cost of co-location would comply with the principles of distribution of benefits and effective competition. However, practicality becomes an issue in this charging system. The options seem to be either to try to predict the number of operators who will wish to collocate, or to have a system whereby the charge to each operator varies with the number of operators collocating. The first of these options makes it difficult to ensure that BT exactly recovers its costs. The second introduces a layer of complexity and may lead to uncertainty for operators.

E43 However, these issues of practicality are not insurmountable. It would therefore seem that there is a strong rationale for deviating from following the principle of cost causation, and that a system based on average cost would be optimal.

E44 Charging on an average cost basis does not provide an incentive for operators to take risks in order to provide services to consumers. The incentive to take risk could be improved by adjusting the charging system to reduce the cost to the operators who choose to co-locate early. This would involve the complex task of producing a risk profile. Alternatively, operators could be charged as if they had been co-locating from the start. This would provide an incentive for early entry.

E45 Within the EU Denmark, Germany and Austria have implemented unbundling. The countries intending to implement LLU include the Netherlands, Italy, Spain, Ireland and France. Outside the EU the main example of local loop unbundling is in the US. Oftel has contacted each of these countries to request information on, amongst other things, how the charge for co-location is, or is intended to be, structured.

E46 Table 1 shows the responses Oftel has received. The table shows that, in each country for which we have information, the chosen charging system is average cost. The analysis earlier in this paper identified two ways of dealing with fixed costs when implementing an average cost charging system. These are:

• First operator pays the whole of the fixed charge. The second operator pays the first operator 50% of the fixed charge; the third operator pays the first two operators 33% of the fixed charge, etc.

• Each operator is charged the same amount.

E47 The table shows that both of these approaches have been used in other countries. Germany has implemented the first of these systems and Ireland is consulting on which of these two systems to use. The remaining countries have opted for a system whereby each operator is charged the same amount. The general approach is to share the total costs of co-location between operators according to the proportion of the floor space of the exchange they occupy.

E48 However, this raises a further issue of how the total cost of co-location is to be shared by the operators is determined. The total cost will vary with the number of operators choosing to collocate. There are two different approaches that seem to have been used. The first is to take the total cost to be the cost the incumbent would incur if the space allocated for co-location were to be fully utilised. This could lead to under recovery by the incumbent if the actual number of operators choosing to collocate is less than the maximum and the incumbent does not have an alternative use for the space. The second approach is to make a forecast of the number of operators who are expected to choose to collocate. This could lead to under or over recovery by the incumbent.

Back to Contents

Asymmetric digital subscriber loop (ADSL) – a technology that allows the use of a copper line to send a large quantity of data (eg a television picture) in one direction and a small quantity (eg a control channel and a telephone call) in the other

Asynchronous Transfer Mode (ATM) – the internationally agreed basis for broadband ISDN. A technology that enables all types of information (data, voice and video in any combination) to be transported by a single network infrastructure.

Bandwidth – the physical characteristic of a telecommunications system that indicates the speed at which information can be transferred. In analogue systems, it is measured in cycles per second (Hertz) and in digital systems in binary bits per second. (bit/s).

Broadband – a service or connection allowing a considerable amount of information to be conveyed, such as television pictures. Generally defined as a bandwidth > 2Mbit/s

Carrier Pre-Selection – A service which enables customers to get their calls carried by a different operator through their existing phone lines without having to dial any extra digits.

Concentrator – the part the local exchange which is positioned close to the customers. It is sometimes within the local exchange, and sometimes located remote from the local exchange.

Copper line – the main transmission medium used in telephony networks to connect a telephone or other apparatus to the local exchange. Copper lines have relatively narrow bandwidth and so have limited ability to carry broadband services such as video unless combined with an enabling technology such as ADSL.

Crosstalk – the phenomenon whereby telecommunications signals on one circuit can leak across to another, potentially degrading the performance on the affected circuit.

Deaveraging – the move from a single averaged tariff or charge to separate ones based on differing geographical areas or types of service.

Digital – the coded representation of a waveform by binary digits, as opposed to analogue which is the direct representation of a waveform.

Digital Local Exchange (DLE) – the telephone exchange to which customers are directly connected.

Digital Main Switching Unit (DMSU) – a trunk exchange primarily used for connecting long distance calls.

Digital Subscriber Loop (DSL) – A family of technologies generically referred to as DSL, or xDSL, capable of transforming ordinary phone lines (also known as ‘twisted copper pairs’) into high-speed digital lines, capable of supporting advanced services such as fast Internet access and video-on-demand. ADSL (Asymmetric Digital Subscriber Line), HDSL (High data rate Digital Subscriber Line) and VDSL (Very high data rate Digital Subscriber Line) are all variants of xDSL.

E-Commerce – The action of buying online or establishing an online store-front. Also, using technology to speed up and make more efficient the transaction of commerce at all stages of the process from production to delivery.

Efficient component pricing rule (ECPR) – A rule for determining interconnection prices, under which the price is composed of the incremental cost of providing the interconnection service plus the profit (including contribution to common costs) that the network operator foregoes by selling interconnection to another operator rather than a service to the final customer.

Geographically averaged prices – prices established by averaging the costs of network elements across the country so that customers in different areas of the country do not pay different rates.

Incremental costs – the capital and operating costs that arise as a result of the provision of the ‘increment’. In contrast to fully allocated costs, the incremental costs include only those costs that are caused by the provision of the increment. So long as revenue exceeds incremental costs, the company increases its value by providing the increment.

Indirect access – where a customer’s call is routed and billed through operator A’s network even though the call originated from the network of operator B. (It is the generic term for both easy access and equal access Carrier Pre-Selection)

Integrated Services Digital Network (ISDN) – a network based on the existing digital PSTN which provides digital links to customers and end to end digital connectivity between them. ISDN2 provides a maximum bandwidth of 128kbit/s.

Interactive services – this term covers two forms of interactivity. The first is where viewers use the remote control to click to applications, which are included in the broadcast stream. The second form of interactivity is where the modem is used to communicate with a remote server.

Interconnection – the physical and logical linking of telecommunications networks used by the same or a different organisation in order to allow the users of one organisation to communicate with users of the same or another organisation, or to access services provided by another organisation. Services may be provided by the parties involved or other parties who have access to the network.

Internet – a global network of networks, mainly narrowband, accessed by users with a computer and a modem via a service provider.

Internet Protocol (IP) – A set of instructions describing how to address and transfer information across a network. The Internet is a public network consisting of many interconnected IP networks.

Local loop – the access network connection between the customer’s premises and the local PSTN exchange, usually a loop comprised by two copper wires.

Long-run incremental cost (LRIC) – the cost avoided through no longer providing the output of a defined increment; for example, the cost of call conveyance is the cost which would be saved in the long-run if this service was no longer provided

Main distribution frame – the apparatus in the local concentrator (exchange) building where the copper cables terminate and cross connection to other apparatus can be made by flexible jumpers.

Mbit/s – Mega (million) bits per second. A measure of the speed of transfer of digital information.

Modem – a device which converts digital signals from a data-transmitting terminal into modulated analogue signals which can be carried by a public telephone network.

Narrowband – a service or connection allowing only a limited amount of information to be conveyed, such as for telephony. This compares with broadband, which allows a considerable amount of information to be conveyed.

Network Interoperability Consultative Committee (NICC) – is leading UK telecommunications industry group which deals with the technical issues associated with network competition. It produces voluntary technical specifications on behalf of the industry and is a source of advice to the Director General of Telecommunications on the harmonisation of interconnect arrangements

Network Termination Point (NTP) – the edge of a network at which the network is connected to other networks or to terminal apparatus, eg a telephone.

Optical Fibre – cable made of glass fibres through which signals are transmitted as pulses of light. It is a broadband medium that can easily provide capacity for a large number of channels.

Partial baseband leased circuit – where a leased circuit extends from one customer to the local exchange and the bandwidth on that circuit extends down to zero Hertz.

Public Switched Telephone Network (PSTN) – the complete network of interconnections between telephone subscribers.

Radio Spectrum – the range of wavelengths used, for example, for broadcasting radio, terrestrial television and satellite television. Usable wavelength ranges from about 100 KHz to about 400 GHz although there are as yet no broadcasts above about 12 GHz.

Service provider – provider of telecommunication services, or services with a telecommunication service component, to third parties whether over its own network or otherwise.

Spectral management – managing the type and quantity of devices connected to the copper pairs, through analysing the impacts of the power and frequencies transmitted, so as to eliminate, as far as possible, any unwanted interference between services on the cable; and between those services and external radio services. Also, to optimise the effective use of technologies to deliver the maximum overall benefit from the cable concerned, for example, by deploying a higher density of lower capacity devices compared to a lower density of higher capacity devices.

Video-on-demand – a programme or film sent independently to a customer in response to his individual request. This contrasts with broadcast television, which is sent simultaneously to all customers able to receive it.

Back to Contents        Back to Home Page