Home > Consultations > Consultation Documents > Digital Dividend Review > DDR

Executive summary

1.1 The radio spectrum is a scarce resource of enormous importance in the modern world.

1.2 In the UK , uses of spectrum like mobile communications and broadcasting account for about 3% of the economy – more than the electricity and water industries combined. Spectrum is also an essential input into numerous public services – from defence to the emergency services, and from scientific research to transport.

1.3 The demand for spectrum is growing fast. In part, this reflects rapid innovation in wireless technologies and applications of many different kinds. In part, it stems from the fact that wireless services have unique features that are valued by almost everyone – like mobility and convenience.

1.4 The rising importance of the spectrum means that the way that it is managed is a vital issue for advanced economies around the world. So one of Ofcom’s most important objectives is to ensure that the use of spectrum brings as many benefits as possible to the UK’s citizens and consumers.

1.5 This consultation is about how we should achieve this through the release of one of the most valuable spectrum bands likely to be available in the foreseeable future – the digital dividend.

The digital dividend

1.6 As its name implies, the spectrum comprises a range of radio frequencies – from the very low to the very high. At very low frequencies, signals travel a long way, but they have little capacity to carry information; at very high frequencies, the signals do not travel far, but they can offer huge capacity.

1.7 The most attractive spectrum offers a combination of range (propagation) and capacity (bandwidth) that makes it suitable for lots of different uses. Good propagation means that less infrastructure is needed to provide coverage, which reduces cost and improves service in buildings and in rural areas. Good capacity means that the signals can be used for services that involve carrying lots of information to lots of users – like high-quality voice, data and video.

1.8 It is generally agreed that the most valuable spectrum in the UK is between around 200MHz and 1GHz – which offers just this combination of range and capacity. At present, nearly half of this spectrum is used to broadcast analogue television – 368MHz, or 46%, of the 800MHz.

1.9 The use of the spectrum for this purpose dates back to the decades after the Second World War, when television was first made available on a large scale and preparations were made for the introduction of broadcasting in colour. There were many fewer uses of spectrum then than now, and networks were generally optimised to economise on infrastructure rather than spectrum.

1.10 The UK’s analogue television signals will be switched off, region by region, between 2008 and 2012. In principle, this means that all 368MHz might be available for new uses, but it was previously decided by the Government that 256MHz of the 368MHz should be used for digital terrestrial television (DTT) from digital switchover. This digital broadcasting will be provided by six multiplexes, each of which can carry a number of television channels and some other services.

1.11 This decision will allow digital terrestrial television to expand its coverage – to match that of analogue, at 98.5% of the population; and its capacity – to around 10 times that of analogue in most of the country, and around 5 times elsewhere.

1.12 At the same time, digital switchover will allow the remaining spectrum – 112MHz – to be released for new uses. It is this 112MHz that forms the core of the ‘digital dividend’.

Different types of spectrum

1.13 In fact, there are three different sorts of spectrum that are available for release, and we have been considering all three in the Digital Dividend Review.

1.14 All of these frequencies are located in that part of the radio spectrum known as UHF, or Ultra High Frequency – a name that dates from the days when little use was made of higher frequencies. (That has long since changed, though UHF spectrum remains exceptionally useful.) The UHF band is conventionally divided in Europe into channels of 8MHz each, with the part often used for broadcasting ranging from channel 21 at the bottom to channel 69 at the top – or 470MHz to 862MHz.

The three sorts of spectrum are:

• the 112MHz mentioned above. This comprises spectrum that will be cleared as a result of digital switchover, and that will be available UK-wide for new uses. This 112MHz comprises 14 channels of 8MHz, which is presently used for analogue television, and on a secondary basis for uses such as wireless microphones. Both primary and secondary uses will cease at switchover;
• two other blocks of spectrum that are used for other things, but have the potential to be cleared. These are channel 36 and channel 69. Channel 36 is currently used for airport radar, and channel 69 principally for wireless microphones. In both cases, it was sensible to review the future use of this spectrum at the same time as the rest of the band. The inclusion of these two blocks means that there is a total of 128MHz potentially available as cleared spectrum on a national basis; and
• then there is interleaved spectrum.This is capacity that will be available within the frequencies that will be used to carry the six DTT multiplexes. It is effectively ‘white space’ that exists geographically between the transmitters needed for those six multiplexes.

These three categories of spectrum – the cleared, the potentially cleared, and the interleaved - are together referred to in this document to as the available UHF spectrum or, more loosely, as the digital dividend.

Potential uses and users

1.17 We have identified many possible uses for this spectrum. But it is important to recognise that we cannot identify all the possible uses. It is very likely that more potential uses will emerge in future, as technology changes and innovators create new products. The benefits of these unknown uses could be as large as, or larger than, the benefits of uses that we can identify now.

1.18 The main uses we can identify as possibilities now are:

• mobile television – and other types of mobile video and multimedia;
• digital television channels aimed at a national market; these could be in
• standard definition (SD) like terrestrial services broadcast now, or in
• high definition (HD);
• digital television channels aimed at a local market – local television;
• wireless microphones, and other programme-making and special events (PMSE) applications, like in-ear monitors;
• broadband wireless applications – which could also be mobile;
• mobile communications – services like voice and data;
• low power applications – like hubs to distribute content around the home;
• services using satellite communications; and
• public safety services – like applications for the emergency services.

1.19 But this is not an exhaustive list, and even within this list there is a huge array of variants that could affect the use of spectrum. These include different standards or technologies that could be used; different geographies that could be targeted; different services that could be offered; and different user groups that could be served.

1.20 As a result, among the issues covered in this document are the relevance of this spectrum to the rural coverage of many services; and the way in which enhanced television services could be provided to viewers with particular requirements for accessibility.

1.21 The question of who uses this spectrum is yet another dimension. There is an even wider array of potential users of the spectrum than there is of uses – many different operators of many different services. And it can be argued that the different character of these organisations could also be relevant – such as whether they belong to the public, private, or voluntary sector.

Objectives and approach

1.22 Our objective in releasing the digital dividend is to maximise the value that the use of this spectrum is likely to bring to society over time. It is emphatically not our objective to manage the spectrum so as to raise revenue for the Exchequer – nor, given our statutory duties, is this a consideration that Ofcom takes into account.

1.23 The use of the radio spectrum has historically been highly controlled by regulators in the UK and around the world. Regulation has been “command and control” in style – with the regulator controlling who may use the spectrum, what services they may provide, and what technologies they may use.

1.24 This highly intrusive approach is no longer fit for the modern age. Many studies have shown how excessive regulation of spectrum has led to extra scarcity and less flexibility. Both competition and innovation have been damaged – as new entrants and new technologies have both struggled to gain access to markets.

1.25 The use of spectrum is so important in the economy that excessive regulation has very large costs. One study for the European Commission has estimated that reducing spectrum regulation could bring extra benefits worth €9 billion every year to the EU economy[^(-1-)]. Other studies in the US have shown similarly large effects.

1.26 Since Ofcom came into existence at the end of 2003, we have pursued a vigorous agenda for reforming spectrum management – to allow users much more flexibility to decide how spectrum is used, for what and by whom. We are implementing this by reducing regulation and making much more use of market mechanisms:

• allowing spectrum to be traded from one user to another;
• liberalising the use of spectrum, by removing restrictions that limit use to particular technologies or services;
• releasing spectrum that is unused promptly to the market; and
• cutting regulation, where appropriate, by reducing the need for licences to use spectrum.

1.27 At the same time, we recognise the fundamental responsibilities of regulation – to prevent one use of spectrum interfering harmfully with another, and to ensure fair and effective competition. We also recognise that in the past the way spectrum is used has been linked directly to public policy goals – and that the transition from one model of command and control, to another of market mechanisms, requires careful thought and preparation.

1.28 How do we maximise the value to society from the digital dividend? To help us answer this question, we have undertaken a very large amount of analysis and research over the course of this year. The results of that work are set out in this document, in the Annexes and in a number of accompanying documents.

1.29 The key components of our work have included: technical analysis of the ways in which the spectrum could be used; extensive research into consumers’ interest in various potential uses; assessment of the potential demand for different services; modelling of the likely value of the spectrum to consumers and businesses and society more generally; and consideration of the options for packaging the spectrum and the design of an auction.

Analysis

1.30 Some key points stand out from the work we have done.

We know the value of the spectrum is large - but the value of any one use is uncertain

1.31 We estimate the total value of this spectrum to consumers and businesses at £5-10 billion, though the figures are uncertain. Note that this is not an estimate of auction proceeds, but a figure for the total value to consumers and businesses over 20 years (net present value).

1.32 But there is huge uncertainty over the value of the individual uses we have identified. This reflects, above all, the uncertainty that is inherent in fast-moving markets, where changes in technology and consumer preferences are unpredictable. This is quite apart from the uncertainty over uses that, as yet, are unknown.

1.34 There is also uncertainty over the technical feasibility of the different uses. We know that the use of the interleaved spectrum is much more constrained than the cleared spectrum – although even this has alternative uses (such as DTT with local or sub-national coverage, and low power devices such as wireless microphones).

1.35 We also know that there are some important technical constraints on use of the cleared spectrum given the need to protect services using adjacent channels in the UK (mainly reception of digital terrestrial TV), and to respect international agreements. In particular it is likely to be difficult to use some or all of the spectrum for mobile devices transmitting to networks (uplink). Just how difficult is not certain, but transmissions in the reverse, downlink, direction should be feasible, and it may be possible to use spectrum in neighbouring bands for any required uplink.

Wider value to society is a significant issue for some uses – less so for others

1.36 Throughout the project we have tried to identify all the different ways in which using the spectrum could create value to us both as consumers and as citizens – including value that it is difficult or impossible to quantify, such as potential contributions to broader social goals like community cohesion or cultural self-expression.

1.37 We have found that there are some potential uses of the spectrum that could bring broader value to society – value that might not be captured in an approach based solely on markets. Local television is one example. Our research with citizens and consumers suggested that local TV could bring broader value to society through better awareness of what is going on in a given community – though the level of actual commitment to watching it was not always clear.

1.38 Similar issues arise in relation to other services. The research showed people felt that widespread availability of mobile broadband could be of broader value to society, for example in rural areas. More choice on the DTT platform could also be beneficial to society if the content was good – the quality of content was seen as critical.

1.39 High definition television, by contrast, was not identified in our research as a major source of broader value to society. A bigger choice of channels was rated more highly, for both citizen and consumer interests, and HD was seen mainly as a premium consumer product, similar to mobile television.

1.40 It is possible, of course, that attitudes to an innovation like HD may change. In time, consumers and citizens may come to expect some HD content to be available to everyone free of charge, just as colour is now. But the evidence that this will happen is not strong right now.

Wider value to society can be delivered in many different ways

1.40 We also found, importantly, that there are alternative ways of delivering the value to society that could be provided by these services. Using the digital dividend is just one option among many.

1.41 For example, local television – and other types of local digital content – could be delivered by platforms like broadband and satellite. Terrestrial television is only one option. Mobile broadband, similarly, can be delivered using a wide range of different technologies and spectrum bands.

1.42 The story is the same for HD. All broadcasters have a range of options for making their services available in this format – this includes the PSB broadcasters who want to provide it free-to-view. The options for them include using the extra capacity created by improvements to DTT at digital switchover (equivalent to another multiplex); upgrading the DTT platform; and using free-to-view satellite – quite apart from new platforms like IPTV.

1.43 For any service, and any provider, each option will have its own costs, benefits, and complexities. We found as much uncertainty about the extent to which the digital dividend might bring broader value to society as we did on any other issue.

In short, it is not possible for us as the regulator to ‘know’ the best use

1.44 There is so much uncertainty about the current and future value of different uses of this spectrum to society, that we, as a regulator, simply cannot say with confidence what will be the best use of this spectrum over coming decades. To claim that we could would be wrong.

Strategic options

1.45 So what should we do, given this background?

1.46 In the past, successive regulators and Governments have used spectrum as an instrument of public policy. Spectrum has been ‘allocated’ to preferred uses and users as a way of helping to secure the delivery of particular policy goals, and responding to the risk of market failure. Other uses and users have been excluded, and the preferred user has usually had the benefit of access without having to pay a market price.

1.47 Some have argued that we should take a similar approach to the digital dividend – and that we should allocate the spectrum to the particular uses or users that they favour, in order to promote public policy goals.

1.48 We have considered these proposals carefully, but we do not think this approach is right or sustainable in the digital age. There are many reasons for this, but the most important concern how we, as a society, can derive the greatest benefits from this valuable resource over time.

1.49 Using spectrum as a policy instrument may have been appropriate when the variety of uses was less than it is today – and when there was little choice about how to secure value for society. But neither of these conditions now holds, and it is vital that we create a framework for using spectrum that encourages good decisions about how to use it – and good decisions about how to deliver public services.

1.50 Our analysis shows that if we pick preferred uses or users:

• we will distort incentives. The uses and users that get preferred access will have less incentive to use spectrum efficiently. They will tend to use too much, relative both to other potential users of the spectrum and their own use of other inputs to deliver services;
• we will reduce flexibility. If we pick a preferred use and user when we award the spectrum, we will have to impose extra constraints on the way the spectrum is used. There will be less flexibility for the way that spectrum is used to change if circumstances change – for example if demand for the preferred use turns out to be less than expected, or alternative uses turn out to be more valuable;
• we will risk distorting competition, because the preferred users may gain an advantage, and risk reducing the scope for innovation in uses that cannot get access to the spectrum; and
• we will risk getting it wrong, by picking a use or user that turns out not to be the best. This risk is larger the more uncertainty there is in the decision.

1.51 We therefore favour releasing the spectrum in a way that imposes as few constraints on how it can be used as possible. Some constraints are unavoidable, to avoid interfering with other services and to meet international obligations. But these constraints apart, users should be free to decide how the spectrum should be used, for what, and by whom.

1.52 This is a market-led approach to spectrum, not a regulator-led one. But like all markets, the use of spectrum needs to be subject to some rules to ensure fair play. So we will, for example, be considering in detail how best to ensure that the release of this spectrum promotes competition in downstream markets. We will also look at whether to include conditions in licences that promote competition and guard against anti-competitive hoarding.

1.53 We have also given a lot of thought to two issues that we think do justify some intervention in a market-led approach. One is the problem of transaction costs: the fact that some valuable uses of spectrum involve thousands of users who individually use a small amount of spectrum, but for whom co-ordinating use could be very costly. This could be particularly relevant to some potential innovative uses. The other is the risk of disruption to existing users of spectrum: as most of the digital dividend will be cleared by DSO, this problem relates mainly to the many users of wireless microphones, who presently use the spectrum interleaved with analogue broadcasting. The implications of these points are discussed below.

Implications for policy

1.54 Our approach is consistent with Ofcom’s strategy towards spectrum since 2003 – and with Government policy as stated on several occasions. It does, however, have some big implications for public policy.

1.55 This approach implies that we do not favour using preferred access to spectrum as a public policy tool for securing benefits for society – but it does not imply that those social benefits do not exist. Indeed, it is vital that those benefits can continue to be secured in a world where spectrum is a flexible, market-based resource.

1.56 But to make sure this happens, the financial and institutional framework for providing public services will need to recognise that the way spectrum is managed has changed. There are, for example, many public sector organisations that use spectrum as an input, from the Ministry of Defence to Public Service Broadcasters. The funding and governance of these organisations needs to recognise the potential need to acquire spectrum, if this is the best use of the resources available to them. The same may be true in other sectors – like voluntary organisations.

1.57 The need for this change was a key theme in the Government’s response to the Independent Audit of Spectrum Holdings published in March 2006. This endorsed the extension of a market-based approach to spectrum management across the entire public sector outside broadcasting – including uses such as defence and emergency services.

1.58 A major programme of work is now under way to implement the findings of this Audit. Our view is that the principles recommended by the Audit should be extended to cover all possible uses of the digital dividend, including national and local television.

1.59 Ofcom is fully committed to understanding the value that using the digital dividend spectrum could bring to society, in all its forms.

Individual uses

1.60 We have looked in detail at how to apply our strategic approach both as a whole, and in the context of each of the many potential uses of spectrum we can identify now.

1.61 Some key proposals that bear on individual uses are as follows:

• wireless microphones for community use, and similar low power devices: because there are many thousands of small, independent, users in this category, we propose to make available channel 69 for wireless microphones and similar low power devices such as in-ear monitors; we also propose to deregulate access to most or all of this spectrum, by making access free, on demand, to users, without the need for a licence;
• wireless microphones for professional use, and similar low power devices: we recognise the risk of causing disruption to the large community of professional users of wireless microphones and similar low power devices such as in-ear monitors and talkback. This equipment is widely used in theatres, broadcasting and special events; so we will phase in changes to spectrum management here. The cleared spectrum will cease to be available as DSO occurs region by region across the country. This will require many users to retune their equipment or purchase new equipment to make use of different frequencies. However, we will ensure that spectrum continues to be available for this type of use from the new capacity that will exist interleaved with the six digital terrestrial multiplexes after switchover. We will ensure this continued availability for a transitional period, until at least 2012. We set out in this document different options for how this capacity could be packaged and released into the market, including by auction. We will work closely with the user community in developing these proposals and managing the transition to new arrangements;
• potential low power uses: we are keen to investigate other potential innovative uses of the spectrum, but so far we have received few specific proposals. We are seeking to gather more evidence through this consultation, so we can make a more informed judgement next year on whether additional spectrum should be set aside for possible low power uses. We are also undertaking a wider review of how to facilitate more licence-exempt use of spectrum;
• local television: local television is a likely use of the interleaved spectrum, but we think that making this spectrum available nationwide could make it difficult for potential local TV operators. So we propose to offer packages in the interleaved spectrum that are suitable for local TV. These could be geographical assignments based on main transmitter sites across the country. There could be 40 or more packages available, perhaps up to 100. We do not propose to limit use of this spectrum to just local TV, as there are other possible uses for this spectrum, and we propose to award the packages by auction; and
• broadcasting and other uses: we propose to package the cleared spectrum so that it is suitable for use by national DTT, for a variety of services including high definition; but other uses and users will also be able to acquire rights to use the spectrum – uses such as mobile television, wireless broadband, mobile voice and data, and new innovative applications that may emerge.

Timing

1.62 We plan to release all, or almost all, of the available UHF spectrum as soon as practical, so that the benefits of the spectrum to society can be realised as early as possible.

1.63 It is currently expected that the earliest date that an auction of the available spectrum could take place would be during the second half of 2008. This would allow for new services to be deployed in particular regions of the country as the analogue signal is switched off – without having to wait until DSO is finished in 2012.

1.64 We propose to integrate the award of channel 36 with the rest of the spectrum, unless awarding the latter is significantly delayed from the timing set out above. The reasons for this are twofold:

• before channel 36 can be used for any alternative applications, the existing user (airport radar) must vacate the spectrum and international negotiations must be concluded on the future use of the channel. As a result it is not likely that the spectrum will be free for new uses before late 2008; and
• given that channel 36 is a strong substitute and complement for the other nationally available UHF spectrum (the cleared spectrum), it is likely to be more efficient to award all of this spectrum in one integrated award.

1.65 We are also inviting views on the case for holding back a small amount of spectrum – cleared or interleaved – as an “innovation reserve”. This would be against the possibility of major technological developments – such as new low power uses – that could find it difficult to access the rest of the spectrum, even if it has been licensed on a flexible basis.

Spectrum requirements and packaging

1.66 We plan to award all of the spectrum on a UK-wide basis, except for packages in the interleaved spectrum that would be suitable for local television.

1.67 There is a large number of potential options for packaging the cleared spectrum. We have set out six options which capture the range of possibilities, as follows:

• i - a single lot of all the cleared spectrum under consideration in the DDR;
• ii - 3 lots : channels 31-37, 39-40 and 63-68;
• iii - 4 lots: channels 31-34, 35-37, 39-40 and 63-68;
• iv - 4 lots: channels 31-33 & 63-65, 34-37, 39-40 and 66-68;
• v - 5 lots : channels 31-33, 34-37, 39-40, 63-65, and 66-68; and
• vi - 15 x 8 MHz lots, being each of channels 31, 32, 33, 34, 35, 36, 37, 39, 40, 63, 64, 65, 66, 67 and 68.

1.68 We think that options iii, iv, v and vi are likely to be preferable, but we invite views from stakeholders on all the options. We will present more detailed packaging proposals in a consultation in or around July 2007.

1.69 In the interleaved spectrum, we propose to offer:

• 40 or more packages suitable for local television; these will comprise at least one assignment at each of a large number of main transmitter sites; and
• packages of interleaved spectrum suitable for wireless microphones; these would be nationwide and we would ensure continued availability of spectrum for PMSE use at least until the end of 2012.

Potential auction designs

1.70 There are a number of different auction formats available, which may be suitable for the award of multiple lots of spectrum frequencies. In selecting the appropriate format for an auction, it is helpful to consider four key choices in design:

• simultaneous or sequential sale of lots;
• single round (sealed bid) or multiple rounds (ascending bids);
• generic or specific lots; and
• individual lot or package (combinatorial) bidding.

1.71 Ofcom’s current view is that using a simultaneous, multiple round process is likely to be the most appropriate approach for the cleared spectrum. We will give more thought to the options for the interleaved spectrum where a simpler mechanism may be preferable.

1.72 We also believe that it is likely to be more appropriate to use specific lots rather than generic lots, due to the differences in constraints (and hence usability) of the spectrum under consideration in the award.

1.73 However, we have not yet come to a view as to whether it would be appropriate to use package bidding in this award.

1.74 A number of potential auction formats are described, along with the advantages and disadvantages of each. Ofcom has not yet concluded on the appropriate auction design for this award, and will issue more detailed proposals for consultation in or around July 2007. The potential auction formats currently being considered are:

• a ‘standard’ simultaneous multiple round ascending auction ( SMRA) with pre-defined lots, which could be augmented by either limited withdrawals or augmented switching;
• a SMRA with pre-defined lots and package bidding;
• a clock/sealed bid hybrid with each lot as a unique category (equivalent to having pre-defined lots); or
• a clock/sealed bid hybrid with a more limited number of categories of generic lots.

Usage rights and obligations

1.75 The main non-technical conditions that Ofcom is currently minded to include in the licences to be issued for use of the UHF spectrum are:

• licence term – a minimum term of 18 years, with an indefinite term thereafter, subject to revocation on 5 years’ notice; additional fees may be payable after the minimum term;
• tradability – the licences to be tradable;
• liberalisation – the licences to contain the minimum necessary technical conditions and not specify either type of equipment or services; and
• obligations relating to PMSE – the obligation to make some interleaved capacity available for PMSE services.

1.76 We will also be considering how this spectrum award should take into account the 2012 London Olympics and Paralympics.

Citizens and consumers

1.77 We believe that the proposals set out in this document will deliver significant benefits to citizens and consumers. This is because we think that the right way to further the interests of citizens and consumers through the digital dividend is not to impose restrictive regulation on potential uses and users, which can impose very large hidden costs. Instead we propose to create as much flexibility as possible for spectrum to be used in the most efficient way.

1.78 Under this approach, a significant amount of valuable digital dividend spectrum will be made available for new services, which should bring benefits to consumers and citizens through a greater potential for innovation, greater flexibility to achieve the best uses of the spectrum, and the potential to increase competition in communications markets.

Next steps

1.79 This consultation, published on 19 December 2006, lasts for 13 weeks. The closing date for responses is 20 March 2007.

1.80 We warmly welcome comments on these proposals. We recognise the complexity and importance of the issues, and we will conduct a substantial programme of stakeholder engagement during January and February 2007, to allow stakeholders to express their views on the proposals we have put forward.

1.81 Alongside this document, we are also publishing a number of supporting documents, including Annexes to this consultation document, a report by our consultants, the results of Ofcom’s consumer market research, and a number of reports covering technical and compatibility issues. These are available on the Ofcom website at http://www.ofcom.org.uk/consult/condocs/ddr/ .

1.82 We expect to release a statement on this consultation in or around July 2007. At about the same time, we also expect to publish a further consultation on some of the many detailed issues that will need to be addressed to take the award forward – such as detailed proposals on spectrum packaging and auction design.

1.83 An indicative timetable for the whole spectrum award is set out in Section 12. This timetable is subject to a number of external factors beyond Ofcom’s control (eg international developments), and so may be amended during the course of the award process. In particular, it is possible that in time proposals may be made for action at European level on the digital dividend, including the possibility of decisions by European institutions that could be binding on the UK.

Question 1 : This executive summary sets out Ofcom’s proposals for the release of the digital dividend. Do you agree with these proposals?

Footnotes:

1.-http://ec.europa.eu/information_society/policy/radio_spectrum/docs/ref_docs/secontrad_study/secontrad_final.pdf

The full document is available below

• Digital Dividend Review - This document consults on the proposed approach to the award of the digital dividend spectrum (470-862MHz) [pdf]
  Full Print Version

Back to top