Next Generation Networks : Responding to recent developments to protect consumers, promote effective competition and secure efficient investment
Summary
Summary
1.1 The adoption of Next Generation Network ('NGN') technology promises to be a positive yet disruptive trend in the telecoms industry. The technology has the potential to bring significant benefits to citizens and consumers through new and improved services, and lower prices due to the likely greater efficiency of a multi-service network. It also has the potential to alter the prevailing model of competition in the telecoms sector. For these reasons, understanding NGN developments continues to be of vital importance to consumers, industry and Ofcom, and to the design and implementation of effective and sustainable regulation.
1.2 In 2004, when details first started to emerge of BT's plan to build an NGN through its 21st Century Network ('21CN') programme, NGNs were seen as perhaps the most important development in telecoms since privatisation. At the time, it was thought that they might represent a change of such magnitude as to require a different approach to regulation.
1.3 Since that time, with experience of real-world implementations of NGN technology, it has become apparent that the move to NGNs is not likely to occur as the step change that was once expected. It now seems more likely that NGNs will be adopted gradually, forming part of the wider evolution of network technologies, and with many opportunities for changes in direction along the way.
1.4 NGN technology is being adopted alongside fixed and mobile access network upgrades, and alongside equally important developments outside the telecoms sector, perhaps most notably in IT. It is within this revised outlook that Ofcom is now considering the potential impact of NGNs on regulation.
1.5 The purpose of this consultation is twofold:
1.5.1 First, to present our response to recent NGN developments, including the latest revisions by BT to its plans for 21CN, and to the related concerns raised by stakeholders.
1.5.2 Second, in the light of recent developments to provide an update on our thinking as to how consumers should be protected during the migration to NGNs.
1.6 In addition, we discuss some of the possible longer term implications for regulation of a widespread adoption of NGN technology. In particular, we consider whether there is anything Ofcom should be doing today to cater for this future world, and hence to better serve the interests of citizens and consumers.
1.7 We recognise that the interaction between NGNs and Next Generation Access ('NGA') technologies, which can be used to provide super-fast broadband services, raises important regulatory issues about the future model of competition in the UK telecommunications market. In particular, it is not yet clear what sort of regulated wholesale products should provide the basis for competition where fibre is used in the access network, on either a fibre-to-the-cabinet ('FTTC') or fibre-to-the-premises ('FTTP') basis. This is a fundamental strategic issue which is not addressed in this consultation, but which will begin to consider in the forthcoming Wholesale Local Access and Wholesale Broadband Access market reviews. We expect to publish consultation documents on these market reviews in the near future.
We think the existing regulatory priorities for NGNs continue to be appropriate
1.8 Our last consultation and Statement on NGN issues, in 2005 (-1-) and 2006 (-2-) respectively, focussed on the impact of 21CN on consumers, regulated access and interconnect products, and the consequences for the model of equivalence as implemented in the Undertakings offered by BT and accepted by Ofcom in September 2005. (-3-) The main policy objectives established in these documents can be summarised as follows:
- to provide incentives for efficient investment in NGNs;
- to promote effective competition based on NGN infrastructure; and
- to protect consumers from disruption during the transition to NGNs.
1.9 We consider that these objectives remain the priority for NGN regulatory policy, and they underpin our consideration of the issues raised in this consultation.
Super-fast broadband is being prioritised over NGN investment
1.10 NGNs are generally understood to refer to networks using the Internet Protocol ('IP') capable of being used for both voice and data, and in which there is some control over Quality of Service ('QoS'). At a more technical level, NGNs feature a common transport layer which physically carries packets of data, and a separate control layer which provides the intelligence to specify, control and manage the services contained within the data packets.
1.11 The business case for building an NGN generally rests on two benefits it brings to operators. First, a single network is cheaper to build and run than the current approach of having a separate bespoke network for each service. Secondly, NGNs can make it cheaper and faster to develop and deploy new services, thus making NGN operators more responsive to customer demands and therefore more competitive.
1.12 With the downturn in the economy, and increasing restrictions on the availability of capital, Communications Providers ('CPs') are having to re-prioritise investments. The cost savings brought about by NGNs only become apparent after a period in which costs may increase. In addition, the benefit of being able to develop new products faster will only have a material effect if there is a market opportunity for these as yet unidentified new products. In the short term, therefore, there are risks that NGN investment will deliver neither cost savings nor new products. As a result, CPs are increasingly looking for safer investment opportunities, focussing on projects which are more likely to deliver new or enhanced services, and extending the lives of existing assets.
1.13 As noted in our recent statement on super-fast broadband (-4-), 'super-fast broadband services, and the networks required to deliver them, continue to grow in importance and interest to consumers, industry and politicians alike.' This growing interest has seen Next Generation Access ('NGA') become the priority for future network investment. The shift in priority away from standalone core NGNs partly reflects the acceptance that the future of access networks will be increasingly fibre-based, and therefore any new investment in copper-based networks, and products that are dependent on copper access, will have a limited life-span.
BT has fundamentally changed its plans for 21CN
1.14 These developments have culminated in a fundamentally different outlook for NGNs in the UK. For the past five years, since BT announced its intention to build 21CN, the expectation has been that in the not too distant future, BT would replace its Public Switched Telephone Network ('PSTN') in its entirety. Following a strategic review of its plans for 21CN, BT has decided to step back from this vision of a complete replacement of its PSTN.
1.15 It is now expected that parts of BT's PSTN will be replaced as and when needed, for example when equipment reaches the end of its useful economic life. The focus for future investment is on upgrading the access network with FTTC and FTTP. These deployments are likely to be accompanied by core NGNs to deliver telephony and other services. However, the design of this future all-fibre NGN could be very different from the architecture originally envisaged for BT's 21CN and also used by other fixed-network CPs.
1.16 This change in outlook has created considerable uncertainty. Although most CPs expect NGN technology to be adopted in the future, it is no longer possible to say with any degree of certainty how or when this will happen. In addition to the uncertainty itself, three other issues have been raised by CPs which concern regulation. These relate to the impact of NGNs on the voice interconnection regime, the impact of 21CN on the Undertakings, and the possible need for a new voice-only access product from Openreach.
BT's revised plans for 21CN are likely to have implications for the Undertakings
1.17 Under BT's original plan for 21CN, it was to deploy Multi Service Access Nodes ('MSANs') in local exchanges throughout the country, to provide both voice and broadband services on copper lines. This deployment would have represented a significant change from today's network in which these services are provided on physically separate voice and broadband equipment.
1.18 This 'converged' MSAN approach is not fully compatible with BT's Undertakings. The Undertakings require that BT use an Equivalence of Inputs ('EoI') product from Openreach in producing its wholesale services. BT currently uses Shared Metallic Path Facility ('SMPF') on an EoI basis in the provision of wholesale broadband. SMPF relies on the fact that a copper line is split to allow the use of physically separate equipment for voice and broadband. The converged MSAN combines the functions of the separate voice and broadband equipment, and therefore no longer requires a split copper line. This convergence would have meant that it was no longer possible for BT to use SMPF, or an equivalent broadband only product, on an EoI basis.
1.19 Under BT's revised 21CN plan, the converged MSAN approach is only likely to be used in a small number of exchange areas. For this reason, in most parts of the country, the current model of competition established by the Undertakings, based on Wholesale Line Rental ('WLR'), SMPF and Metallic Path Facility ('MPF') products, can continue. In terms of upcoming developments, however, it should be noted that BT's plans to build NGA networks are likely to have implications for this model of competition.
1.20 To cater for those exchange areas in which converged MSANs will be deployed, BT and Ofcom will need to agree a change to the Undertakings. The process for dealing with such variations to the Undertakings is well established, and Ofcom will consult on this as soon as there is a firm proposal from BT.
1.21 Some of the issues raised by BT using converged MSANs could also have consequences for the model of competition established by the Undertakings. For example, if BT Wholesale were to launch a new 21CN-based voice product, or a converged voice and broadband product, it would be necessary to consider what the upstream inputs to those products should be. We will address issues of this sort as and when they arise.
The case for a new voice access product from Openreach will depend on demand
1.22 For some time it has been suggested that there should be a voice-only passive access product from Openreach the voice equivalent of SMPF for broadband. Whereas MPF provides access to the whole copper line, SMPF provides access to just the broadband part of the line, but is only available on lines with voice service enabled. This voice service is always based on WLR one of the EoI products provided by Openreach.
1.23 In addition to these services, some CPs have suggested that there is a requirement for a new product which would provide access to just the voice part of a line, but only on lines where broadband is being supplied using SMPF. In essence, it is the passive copper access service which Openreach implicitly uses to create WLR on lines with SMPF. This potential new product has been, and is here generally referred to as xMPF. (-5-)
1.24 Based on discussions with stakeholders, we understand that the most important benefit to CPs from using xMPF would come from an increased margin relative to using WLR. However, it has also been suggested that in order to make xMPF financially viable, a greater margin would be needed between MPF and WLR.
1.25 Due to the close physical similarity of the products, it seems possible that the costs for xMPF and MPF would be very similar. Therefore, pending further analysis, it appears reasonable to assume that, if xMPF was offered by BT, its price would be similar to that of MPF. This would suggest that an increased margin for CPs using xMPF might be achieved by either increasing the WLR price or reducing the MPF price. Neither of these suggestions are considered in this consultation since the regulation of MPF charges has been recently covered in the Openreach Financial Framework review (-6-), and our proposals for the WLR charge control are currently out to consultation. (-7-)
1.26 There remains the question of whether, given MPF pricing and our current proposals for the pricing of WLR, and assuming that the price of xMPF would be similar to MPF, there would be sufficient demand for an xMPF product. Our preliminary view is that, if there is sufficient demand at these prevailing prices, the matter could be resolved through the existing Statement of Requirements ('SoR') process which CPs use to request Openreach to develop new products. Equally, we believe that the current SoR process should be sufficient to cater for future variants of xMPF not explicitly considered in this consultation. We would, however, welcome stakeholder views on this assessment, and we will be keeping the matter under review.
Interconnection arrangements should not act as a barrier to investment in NGN technology
1.27 NGN deployment will also bring about a transition from interconnection of voice services based on Time Division Multiplexing ('TDM') to IP-based interconnection.
1.28 From a technical perspective, there has been significant progress. Against a challenging timescale, NICC Standards Limited ('NICC') (the UK forum for interoperability standards) has delivered two releases of an entirely new suite of interconnection standards for IP interconnection.
1.29 Under BT's original plan for 21CN the rapid migration of voice services would have set the pace of the transition to IP interconnection. BT's revised plans mean that TDM and IP will coexist for the foreseeable future, and other operators may lead the adoption of IP interconnection.
1.30 The arrangements for interconnection between networks using different technologies are set to become an increasingly important issue. In addressing this issue, we need to ensure that regulated interconnection products, and regulation of interconnection more generally, does not act as a barrier to efficient investment in new technology.
1.31 Accordingly, we would welcome stakeholder views on the following issues:
1.31.1 In a mixed TDM/IP environment, which network should provide the interworking function to convert traffic between the two standards, and how should the costs of providing this function be recovered? This issue arises in relation to both traffic between BT's 21CN and other CPs' TDM networks, and traffic between BT's TDM network and other CPs' NGNs.
1.31.2 Should BT be required to offer an IP equivalent to TDM interconnection and, if so, when?
1.31.3 Does the principle of reciprocity continue to be appropriate in a mixed TDM/IP environment? The principle of reciprocity is based on the idea that the regulated charges of the incumbent are a reasonable proxy for the costs of an efficient network operator. However, if BT continues to originate and terminate most of its traffic on a TDM network, its regulated charges may not reflect the efficiently incurred costs of another CP with an NGN.
1.31.4 What further technical standards work is required to enable the transition to NGNs?
1.32 Our intention is to develop our thinking in relation to these issues. This will help to inform future work on interconnection in the context of relevant market reviews.
Uncertainty surrounding BT's plans could have a negative effect on investment by other operators
1.33 It has long been recognised that BT's plans for 21CN, and network development more generally, have a significant impact on investment by competing CPs. All CPs need to interconnect with BT, and the majority also need to use BT's network to access customers. Therefore, changes to the design of BT's network have a direct impact on the businesses of competing network operators.
1.34 BT's recent change to its plans for 21CN has created considerable uncertainty. It has become very difficult for CPs to know what an efficient design will be for their networks given the uncertainty about the topology of the BT network, the geographic location of the points of interconnect, and the technology for interconnection.
1.35 BT is now adopting a much shorter planning horizon. Whereas previously it had planned 21CN on a rolling five year timeframe, it is now looking only 12-18 months ahead. Beyond this time horizon there is no confirmed investment or network upgrade plan. This may be a realistic and optimal approach for BT's own needs, but it also has consequences for investment planning by other CPs.
1.36 More generally, the industry is having to cope with uncertainty created by the current economic climate; greater costs of financing investments; technology risks around the future of telephony; and considerable technology and commercial risk associated with building NGNs and NGAs. Many of these factors have contributed to BT's change in strategy.
1.37 Due to the combination of the general economic climate and the lack of knowledge of BT's future network architecture, there is a risk that BT's fixed network competitors defer investment in NGNs. We are therefore considering what options there might be to improve the investment incentives for all CPs. In making this assessment, we will need to gauge the extent of the risk to investment.
Convergence and bundling during the transition to NGNs will also raise challenges for consumer switching processes
1.38 It is already evident that the transition towards NGNs will be accompanied by a trend towards service bundling at the retail level, and an increasingly complex range of wholesale products. In this situation, there is a risk that the process of switching between retail providers will become more difficult for consumers, and that this could deter switching and harm competition. This risk is exacerbated by the fact that there are already different switching processes for landline, broadband and mobile services.
1.39 Ofcom is undertaking a separate project as part of its migrations work, which will consider the extent to which there is a need for harmonisation of switching processes across different services. The project is examining a broader set of issues around optimal switching models for transferring services between different CPs. We will seek to ensure that decisions made in relation to the design of NGNs and NGAs take into account the emerging evidence from the broader project on migration processes.
NGNs represent a significant technology change, and consumers must be protected from undue disruption
1.40 When Ofcom last consulted on NGNs in 2006, one of the key areas of focus was the potential impact on consumers. As a result of this work, we agreed the following three principles to guide our activities:
- the services offered to consumers on NGNs should at least be equivalent to their existing services;
- consumers should not suffer any detriment during the transition to NGNs, for example, due to loss of access to emergency services or degraded call quality; and
- any changes to end-user services should be fully explained to consumers.
1.41 These principles can be seen as a response to BT's original 21CN plans. These included the proposal that the migration of lines to 21CN would be provider-led, with BT transferring wholesale services to the new network while minimising disruption for consumers. Although BT now appears to be adopting a slower approach to migration driven by individual customer demand for new services, we feel that the principles continue to be appropriate to ensure consumers are adequately protected.
1.42 Although the roll-out of NGNs has been much slower than expected, extensive testing of customer equipment has taken place in recent years. This has highlighted a number of compatibility issues. For example, a significant proportion of security, fire and social telecare alarms connected to the telephone network are sensitive to the increased end-to-end delay of NGNs, and may therefore not operate reliably in certain circumstances. The relevant industry associations are aware of this issue and are co-ordinating further testing activities while helping their members to assess the risk and to plan mitigation activities. The full extent of the issue is currently unclear, but the move towards a much slower demand-led migration to NGNs certainly helps by providing more time to locate and fix specific customer problems.
1.43 This specific issue highlights our general approach in implementing the principles mentioned above. We will continue to monitor developments concerning customer equipment compatibility very closely, and will work with stakeholders to ensure that appropriate solutions are found before migration to the new networks takes place.
In the longer term, NGNs may drive new models of competition, but this does not necessitate immediate changes in regulatory strategy
1.44 Looking further ahead, the direction of network evolution is highly uncertain. It seems likely that NGN technology will eventually be adopted for voice services, but the manner in which this will occur is not yet known.
1.45 In theory, NGNs increase the scope for non-network based competition. The separation of conveyance from service control that is inherent in NGN design creates the potential for new models of competition. In these new models, innovation is controlled and delivered by software development rather than the network infrastructure investment which is required today. In the extreme, the competition model in the telecoms sector may begin to resemble that found on the Internet more closely. This envisages network operators focussing on the provision of generic conveyance services, whilst a multiplicity of independent service providers develop and deliver rich applications which run over these generic conveyance networks.
1.46 However, the NGN designs proposed by CPs to date would tend to indicate that the separation between conveyance and service control will be less than complete. Network operators are likely to retain control of some services, such as guaranteed-quality voice, in a manner similar to today. In this way, there would continue to be significant benefits to vertical integration, and so it may be less likely that an independent application-based service market will develop.
1.47 Our preliminary analysis suggests that the intense competition in value added services that run over networks, which has been made possible by the Internet, is a powerful force that will shape a market-led outcome without a need for regulatory intervention. We would, however, be interested in stakeholder views on the future direction of change, and the regulatory issues to which it may give rise.
Footnotes:
1.- 'Next Generation Networks: Further Consultation', 30/06/2005. See http://www.ofcom.org.uk/consult/condocs/nxgnfc/
2.- 'Next Generation Networks: Developing the regulatory framework', 07/03/2006. See http://www.ofcom.org.uk/consult/condocs/nxgnfc/statement/
3.- For further details on BT's Undertakings see http://www.ofcom.org.uk/telecoms/btundertakings/.
4.- 'Delivering super-fast broadband in the UK', http://www.ofcom.org.uk/consult/condocs/nga_future_broadband/.
5.- Although the general concept of xMPF is well understood, consideration of its practical implementation leads to a number of different potential products. In this consultation we focus on just the most immediately relevant and best understood cases.
6.- The Statement can be found at http://www.ofcom.org.uk/consult/condocs/openreachframework/statement/statement.pdf. This decision has been appealed in The Carphone Warehouse Group plc v Office of Communications, case number 1111/3/3/09, http://www.catribunal.org.uk/237-4154/1111-3-3-09-The-Carphone-Warehouse-Group-plc.html.
7.- See http://www.ofcom.org.uk/consult/condocs/wlrcc/.
The full document is available below
