House of Lords - Broadband for all - an alternative vision

Broadband for all - an alternative vision - Communications Committee Contents

Chapter 2: Background and policy context

The communications revolution

12. It has become a cliché to state that recent years have seen a revolution in communications. Information can be stored, replicated, communicated and shared in ways which were unthinkable just a few decades ago. Commercial and social transactions are more efficient, connected enterprises have a global reach and machines can process and analyse information on a superhuman scale. Almost any information can be represented digitally. We live in an information world without frontiers, in which anyone can publish, anyone can consume, anyone can copy, and anyone can modify and reuse. The advent of broadband has played a crucial role in this evolving landscape. Moreover, extrapolating developments over the past 20 years, analysts foretell, for example, a future where appliances in our homes and at work, like fridges and desktops, are connected and enabled as interactive multimedia devices, and far more beyond, bringing benefits to society. Such an ever changing world would require enhanced broadband provision, offering an explanation for, and a justification for supporting, increasing consumer demand for faster broadband in telecoms networks today. The continuous development of broadband infrastructure and policy-making in this area therefore is of critical importance, requiring strategic, long term planning and collective working between governments, regulators and industry.

The UK's broadband performance

13. Like other technological advances, broadband has inspired devoted evangelists and we received evidence from a number of them. To their disappointment, when it comes to the most highly quoted broadband league tables, the UK does not—on most indices—have a place on the podium. Take average connection speed; according to the most recent Akamai 'State of the Internet' report, the UK ranks 16th in Europe and 25th globally.[3] Some who contributed evidence to this inquiry have lamented that the UK is therefore "achieving speeds more than three times slower than South Korea."[4] Moreover, in the UK, 14% are receiving speeds of less than 2 megabits per second (Mbps),[5] regarded by the Government as the minimum speed which allows you to use the internet reasonably effectively. Such a situation has caused much discussion of the dangers of the so-called 'digital divide', the dislocation between increasingly active users of the internet and users whose connections or whose skills do not allow them to harness the potential of the internet.

14. It has, however, also been pointed out to us on a number of occasions that the UK's performance in terms of the internet economy can be seen as hugely encouraging, trail-blazing even. Recent research by the Boston Consulting Group put the percentage of the UK's GDP derived from the internet economy at double the G20 average. The internet's contribution to UK GDP in 2010 was more than that of construction and education, and online retail is expected to account for up to 23% of total UK retail in 2016. The report concludes that, "the UK has become a nation of digital shopkeepers."[6] Whilst there may seem to be a contradiction here between the UK's moderate broadband connectivity and its economic performance, it can be explained by the fact that much e-commerce activity does not actually require superlative broadband speeds. The conclusion to be drawn is that the UK has vast potential in this sphere. If the UK can enhance its broadband provision, then further economic benefits will follow; certainly the Government's strategy is based on the belief that enhanced broadband provision will be a catalyst for economic dynamism. Equally, it could be argued that the UK's strong performance in the internet economy means that the UK has more to lose if it falls behind.

15. There are six important indicators to bear in mind when considering broadband performance, four of which go beyond speed:

(a) Speed (maximum and minimum, upload and download);

(b) Symmetry (a symmetric connection has equal speeds for upload and download);

(d) Jitter (also known as packet delay variation, the variability of latency);

(e) Reliability of service and the length of time required to resolve problems.

(f) Contention (the number of users sharing the same link to the internet)

16. Speed alone is therefore not the only determiner of broadband performance. Symmetry, latency, jitter, reliability and contention are also critical; what use for instance is a fast connection if it is only achievable at certain times of the day?[7]

Broadband infrastructure: Where is the UK starting from?

17. At first blush, statistics about international broadband download speeds might suggest that the UK need look no further than South Korea for lessons on how to build broadband infrastructure. That is misleading for one simple reason, articulated by Francesco Caio, author of the 2008 Department for Business, Enterprise and Regulatory Reform report on broadband:

"There are other countries that have the benefit of leapfrogging the development stage of the incumbent because they are just late to the party and it is good for them, and clearly Korea is one of them."[8]

By contrast, the UK does have an incumbent communications infrastructure. Given limited resources, this somewhat removes the reason for building a new one—however technically brilliant it could be—from scratch. When it comes to raising the standard of broadband in the UK, it is vital to understand where we start from and the way in which different communications technologies carry data in different ways.

18. In the beginning there was copper, first used for the 19th century telegraph. The physical medium over which most people receive their internet connection in the UK is a twisted pair of copper wires.[9] Twisted pairs carry data through electromagnetic waves confined within the metal. The copper telephone network was ideal for phone calls which require only low frequency waves. The development of digital subscriber line (DSL) technology in the 1990s used higher-frequency waves to transfer data, initially at moderate speeds. While technological improvements have increased DSL's maximum download speed, higher speeds can only be carried over ever shorter distances using copper wire, and the capacity of all but the shortest copper telephone connections, it is safe to say, is now being exceeded by demand for more data-intensive applications.

Optical fibre

19. In addition to copper pairs, a range of other broadband access technologies are available. Coaxial cables also use copper to carry electromagnetic waves, but they can carry higher frequencies—and so higher data rates—over longer distances. Optical fibre networks, however, carry optical signals along glass fibres thinner than a human hair. The cable network, now mostly owned by Virgin Media, uses a mixture of optical fibre and coaxial cables. The backbone network connecting our telephone exchanges to each other, and to the global internet, is already almost entirely optical fibre. Many argue that the future depends on further deployment of optical fibre.

20. An optical fibre is a hair-thin rod made of glass along which light of different wavelengths ('colours') can travel. Investment in fibre is said to be future-proof in the sense that fibre offers data rates far in excess of current and predicted future demand, and in that there are no proposed technologies that can offer comparable data rates over long distances. To illustrate, the copper technology in use today is around 100,000 times as fast as Morse code, and is approaching its physical limits. The bandwidth limits of fibre are around 100,000 times those of copper. Fibre technology, it is therefore claimed, will accommodate our bandwidth demands for decades to come.

The network architecture and the options for fibre deployment

FIXED LINE

21. Much of the current debate about broadband policy therefore revolves around questions to do with how deep into the network fibre should be deployed; how close it should be to the end user.[10] In beginning to understand this issue, it is important to have a grasp of the network architecture and, crucially, to note the distinctions between:

· core communications networks, which carry data around the UK and around the world;

· the middle mile networks (also referred to as backhaul) which connect communities to the core network, and;

· local access networks (also referred to figuratively as the 'final mile'[11]), which provide the final link (sometimes referred to as the final drop) from a local telecoms exchange or cable television hub into a premises.

It is generally acknowledged that the use of fibre in the core and middle mile networks is needed to provide any modern communications. An issue therefore arises of whether, and to what extent, fibre, given its technological superiority, is also used for the final link to a premises, instead of copper.

22. There are two major technological options for fibre deployment in the access network, and these place different constraints on the commercial organisation of infrastructure provision and service delivery:[12]

(i) Fibre to the Premises (FTTP), also referred to as Fibre to the Home (FTTH), where each customer has a fibre link coming directly into the home or business, providing the highest data rates and reliability—reportedly the gold standard—and;

(ii) Fibre to the Cabinet (FTTC), where fibre runs from the exchange to street cabinets, but existing copper is used for the final link into the home. Since the length of copper wire used is shorter than if it were to run all the way from the exchange, higher speeds can be achieved, but data rates are still ultimately limited to a considerably lower rate than is achievable with FTTH.

FIGURE 1

Network Architecture

23. FTTP can achieve data rates many orders of magnitude greater than FTTC, but it is very expensive to install new fibre links to existing premises. The Broadband Stakeholder Group (the Government's advisory group on broadband) estimated in September 2008 that national deployment of FTTC would cost £5.1 billion, while taking fibre to every UK home could cost as much as £28.8 billion, with the largest single cost component being the civil infrastructure.[13] New-build sites, however, can have FTTP installed at least as cheaply as a copper cable. Critics of FTTC argue that while FTTC is cheaper to install in the short term, it may prove more expensive in the long run to upgrade FTTC to FTTP. If the deployment of FTTP technology is chosen, there are still a number of options available for how the network might be structured (detailed in Chapter 4).

MOBILE, WIRELESS AND SATELLITE

24. There is also a role for wireless technologies which can be used as an alternative to copper or fibre for the final link to the premises. Wireless access networks, based on radio waves travelling freely through the air, offer a solution where installing fibre would be prohibitively expensive, for example in sparsely populated rural areas. There is also increasing use of broadband through the mobile network, where the link to the fibre network occurs at the local cellular base station. In rural areas, however, there is normally no nearby fibre network to link up to; in densely-populated areas demand for mobile broadband is already putting a strain on capacity, which is limited by the amount of radio spectrum available. Developments in wireless technologies such as 4G mobile broadband will allow wireless to achieve greater data rates, but they will not achieve faster data rates than fibre in the access network. Wireless technologies therefore seem to us to have a complementary role, standing in for fibre where there is none, and supplementing it where there is.

25. Satellite technology is also a technology option, but it has high latency (it is a long way to space and back). Furthermore, the total bandwidth available is spectrum-limited just as for terrestrial wireless, so it is not possible to have too many users in any one cell (the footprint on the ground of a single satellite beam). Moreover, satellite typically has even more stringent volume limits than mobile.

The Government's strategy

26. Against the background sketched above, in December 2010 the Department for Culture, Media and Sport and the Department for Business, Innovation and Skills published the Government's broadband strategy, in which it set out a commitment to deliver the best 'superfast' broadband network in Europe by 2015. In order to judge what 'the best' means, the Government adopted a scorecard which will take into account four key indicators: take up and coverage, speed, price and choice. Ofcom expects to publish the data for the 'best in Europe' scorecard this summer.

27. The Government's specific target is to provide superfast broadband, defined as 24Mbps[14], to at least 90 per cent of premises in the UK by 2015 and to provide universal access to standard broadband with a speed of at least 2Mbps by 2015. It is not wholly clear from Britain's Superfast Broadband Future why these speeds were chosen, but it is generally understood that they were not entirely arbitrary. Steven Unger, Chief Technology Officer, Ofcom, told us: "the rationale for… 2Mbps was that that was sufficient to support broadcast quality video."[15] In turn, the 24Mbps target seemed to relate to the highest headline speed achievable over copper lines from the exchange. In addition to the UK's targets, there is also a European dimension. The European Commission has said that: "By 2020, all Europeans should have access to internet of above 30 Mbps and 50% or more of European households have subscriptions above 100Mbps."[16] The Government have not put forward specific measures to reach these European targets, which could be perceived as very ambitious. Indeed, Ed Vaizey MP, Minister for Culture, Communications and the Creative Industries, described them as "challenging".[17]

28. In terms of progress towards the UK Government's targets, Ofcom figures report that the coverage of broadband at 2Mbps is 86% of existing connections.[18] They also report that coverage of superfast broadband is around 60% of UK premises. There are striking regional variations: 94% of premises in Northern Ireland have access to superfast broadband, but for Wales and Scotland only 30 to 40% have access.[19] While the Government's strategy is a UK wide strategy, the devolved administrations in Wales, Scotland and Northern Ireland have all identified broadband as a key policy area and have developed initiatives for their own nations which complement policy at a UK level. This report does not scrutinise in detail the strategies of the devolved administrations nor the plans of individual local authorities in England; it focuses on the overarching UK strategy and the progress made in implementing it.

29. For the Government, the drivers for its broadband strategy are economic growth and wider societal benefits. Unveiling the strategy in 2010, the Secretary of State for the Olympics, Culture, Media and Sport, Rt Hon Jeremy Hunt MP, stated boldly:

"A superfast network will be the foundation for a new economic dynamism, creating hundreds of thousands of jobs and adding billions to our GDP. But it is not just about the economy, around the world there are countless examples of superfast broadband helping to build a fairer and more prosperous society, and to transform the relationship between Government and citizens. And shifting Government services online will save billions of pounds of taxpayers' money."[20]

30. Britain's Superfast Broadband Future stated that while the UK was in a sound position, with a competitive market and 71% of UK households having broadband access, it was clear that the UK's use of the internet was going to require improvements to the network. The Government maintained that while its strategy was technology neutral, the deployment of fibre deeper into the network seemed inevitable.[21] The Government also acknowledged that the business case for rolling out broadband in less densely populated areas was challenging.[22]

31. The Government's strategy and targets are therefore predicated on the basis that the market will deliver some, but not all of the investment required. The critical difference between deployment of 1^st generation broadband and the current superfast deployment was the existence of BT's copper network which has been used to deliver several generations of broadband technology by simply upgrading the active equipment in homes and exchanges. Progress towards superfast provision now requires new fibre in the local access network and the middle mile because copper can only carry superfast speeds over a short distance—a kilometre or less.

32. There is a clear social cost of weak broadband performance in pockets of the UK. It is widely accepted that the market will deliver improvements to at least two thirds of households (58% of premises already have superfast coverage). Reaching the remainder, however, as the 'final third' of the population, requires Government support or a socially corrosive digital divide will follow. As with much of the nomenclature in the field of broadband infrastructure, the 'final third' can be misleading. It refers to the third of the population to whom infrastructure providers are less commercially motivated to build new network connections due to the weakness of the case for investment. For purposes of illustration, it would be helpful to show just where these individuals are located in the country. However, it is difficult to identify them on a map for a simple reason: when we have tried to do so, we have found that they are located almost everywhere.[23] This is because people without adequate broadband infrastructure are often surrounded by it. The final third must not be understood, therefore, as a geographical area. Nonetheless, it must be said that the geography of rural areas does make them the hardest to reach.[24] Broadband performance is already lower in areas of low population density across the UK because the copper lines are longer. Furthermore, there is a weaker commercial case for private investment in rural areas due to the high cost of building new networks where there is a large distance between premises.

33. To this end, the Government have allocated £530m within the lifetime of the current Parliament to stimulate commercial investment in rolling out superfast broadband in areas where the case for commercial investment is weak or non-existent. It has also indicated that a further £300 million may be available for investment in broadband up to 2017.[25] The Government see the lack of any commercial interest in deploying superfast broadband in the final third as a market failure that warrants state intervention. The important point to note about the Government's strategy is that they are focusing resources on the local access network, as this is the most challenging investment case for the market.

34. Britain's Superfast Broadband Future was clear that the Government's strategy would draw on the ethos of the 'Big Society.' Communities would be enabled to influence or take part in extending access networks; community need and aspiration should drive forward the process, not decisions made in Whitehall.[26]

35. Broadband Delivery UK (BDUK), a unit within the Department for Culture, Media and Sport (DCMS), is responsible for managing the Government's broadband funding. Britain's Superfast Broadband Future envisaged that the £530 million of funding would be released by BDUK in waves, beginning with four pilot areas (Cumbria, the Highlands and the Islands, Herefordshire and North Yorkshire), to local authorities and the devolved administrations. To access funding for roll-out, local authorities in England were asked by the Government to prepare local broadband plans for approval. Moreover, local authorities in England and the devolved administrations have to put in place matching funding from their own resources, and possibly try to access EU funds (the Government have suggested that the European Regional Development Fund could provide up to £100 million[27]). These various sources of public funding combine to fill the gap between the proposed private investment and the total required. In addition, the Department for the Environment, Food and Rural Affairs (Defra) have a further discrete broadband fund of £20 million specifically for rural areas, called the Rural Community Broadband Fund. It aims to enable communities outside the 90% coverage areas to have superfast broadband services if they can demonstrate local need or demand, feasibility and cost-effectiveness.

36. The issue of state aid has become a pivotal factor in the implementation of the Government's strategy. The European Union recognises that state aid may be needed to deliver a broadband infrastructure that will maximise societal benefits. In essence, a 'market failure' exists if markets, without intervention, fail to deliver an outcome that would yield the highest possible welfare benefits for society. This may arise for instance in terms of socially profitable investments not being undertaken. In such cases, the granting of state aid may produce positive effects and overall efficiency can be improved by adjusting incentives for firms.[28] Such public financing must respect EU competition and state aid rules. These are complex. However, the underlying principle is simple. It is to ensure that no aid granted by a Member State or through state resources in any form whatsoever may distort or threaten to distort competition by favouring certain undertakings or the production of certain goods.[29] This end is typically achieved by the imposition of various 'remedies'; for example, by imposing conditions to ensure that competitors have 'open access' to infrastructure created with state aid. BDUK have decided, with encouragement from the European Commission, to put in place a single umbrella scheme for the benefit of all local broadband projects. The details of this scheme are currently the subject of negotiations between the UK Government and the EU Commission.

37. As of 10 July 2012, a total of 44 out of 45 local broadband plans, detailing exactly how local authorities in England will roll-out superfast broadband in their areas, have been approved by the Secretary of State. However, action is stalled pending resolution of the state aid impasse.

38. The Government have also established a separate fund to create 'Super-Connected Cities'—many of which have pockets of low internet connectivity adjacent to areas of strong provision—with access to speeds of at least 80-100 Mbps across a wide area of the cities. In the Budget earlier this year, the Chancellor of the Exchequer revealed that the UK's first Super-Connected Cities—Birmingham, Bristol, Leeds & Bradford, Newcastle and Manchester along with the four UK capital cities—had all successfully bid to become Super-Connected Cities with so-called 'ultrafast' fixed broadband access, and large areas of public wireless connectivity.[30] Ultrafast broadband is defined as having a minimum download speed of at least 80Mbps.[31] The cities will share £100 million (this is a separate pot of money from the £530 million for the 'final third') to help deliver plans to use super-connected status to boost growth, attract new businesses and enhance the way services are provided and accessed. In addition, the Chancellor has announced that a new £50 million fund would be created to bring ultrafast broadband to further UK cities.[32] Together, the Government claim, the proposals involve providing ultrafast broadband access to around 1.7 million premises and 200,000 businesses by 2015 while almost 3 million residents would have access to a municipal wireless network.[33]

39. In a further strand of digital policy, the Government's £150m investment in the Mobile Infrastructure Project (MIP) is intended to extend and improve mobile coverage—that is to say, extend the mobile network to so-called 'not-spots' by subsidising the construction of mobile masts and other infrastructure in remote locations. Some areas of the UK are not provided with any mobile coverage by mobile network operators and other areas receive low quality coverage which results in a poor level of service. In certain areas of the UK—particularly rural areas—there is a limited commercial case for market-driven private investment to improve coverage and quality of service.

Industry approach

40. There are two major local access networks (the final mile) for broadband in the UK: BT's copper telephone network and Virgin Media's cable television network which almost entirely runs in parallel with BT's network.[34] BT is, however, undertaking a very significant upgrade of its network and has already committed £2.5 billion to roll out fibre which will deliver speeds of up to 80 Mbps to two-thirds of UK premises by the end of 2014. BT is bidding for public funds (from the £530 million pot) and working with local authorities to extend roll-out into areas where the commercial case for investment is more difficult (the final third). If successful in most of these bids, BT claims that it can deliver superfast broadband to more than 90% of UK premises. BT states that its commercial roll-out of fibre is progressing rapidly, with more than 10 million of the UK's 28 million premises now passed by a BT fibre optic connection. BT's fibre deployment is primarily by means of FTTC with its copper network being used for the final link from the local cabinet to the home. It is Openreach, the subsidiary of BT Group, which owns and manages the access network infrastructure.[35] Notably, BT has stressed that investing in fibre access networks is a high risk investment:

"Our fibre business case has a pay-back period of about 12 years. That is to say we do not get our money back for 12 years, and that is on the assumption that we achieve the volumes of customers we hope to achieve in our business case. That is a long-term investment that most commercial organisations would not tolerate."

Significantly, BT has indicated that in order to support the Government's policy objectives they are "willing to spend a further £1 billion or so of BT's capital to match Government funding to do that, to roll it out into the final third, and to get as far as we possibly can into the final third..."[36] While BT's fibre deployment is primarily by means of FTTC, it has recently launched a FTTP range of products and has announced plans to make a form of FTTP available 'on demand' in areas covered by its FTTC network.[37]

41. Virgin Media's cable network covers around 50% of the UK population. Over the last 20 years, Virgin states that the cable industry has invested over £13.5 billion in a fibre-rich network that has the capacity to deliver superfast broadband to around 13 million homes across the UK.[38] Virgin Media is currently offering speeds to consumers of up to 100 Mbps, which it refers to as 'ultrafast'. Virgin has no current plans to expand its infrastructure footprint substantially.

42. Fujitsu has also been engaged in the Government's plans and the BDUK procurement process. In April 2011, it announced plans, in collaboration with Virgin Media, TalkTalk and Cisco, to deliver next generation internet services to 5 million homes in rural Britain, describing the collaboration and build of a new superfast, fibre optic broadband network as "a ground breaking and innovative alternative to BT Openreach."[39] In the majority of areas, Fujitsu said that it would provide fibre directly to the home, rather than to the local street cabinet. Recent reports, however, suggest that Fujitsu has withdrawn from the BDUK procurement for the time being, albeit that it is committed to bidding in the future, causing concerns about the competitiveness of the tendering process and leaving BT as the only provider that has so far secured public funds.[40]

43. Companies in this sector have clearly been reluctant to participate in the Government's procurement programme and bid for funding. Prior to Fujitsu's recent decision to withdraw for the time being, seven of the original nine contenders had already withdrawn from BDUK's framework procurement process. Geo, for example, cited a number of concerns, including the accusation that the gap funding model (see Box 4 in Chapter 4) adopted by BDUK and local authorities favours the incumbent.[41]

44. The regulatory framework for the industry designed and enforced by Ofcom is based on a number of European Union (EU) directives which have been implemented into UK law by the Communications Act 2003.The framework is based on competition law principles, meaning that market interventions are only permissible if a market assessment concludes that one or more companies hold a position of significant market power (SMP) in the market. The obligations set by Ofcom are currently targeted at addressing identified market power in specific, narrowly-defined markets. An impact/cost-benefit assessment must demonstrate that any intervention is proportionate and reasonable. The regulatory framework has a specific and defined purpose and its implementation is subject to legal challenge.

The regulatory framework will be elaborated on in detail in later chapters. In particular, the final paragraphs of Chapter 4 consider the implications of the fact that since May 2011, Ofcom has had the power, by virtue of Article 12 of the Revised EU Framework Directive, to impose regulatory obligations with regard to infrastructure sharing in the broadband market without reference to significant market power. This seemingly small change may have a significant impact on policy (see paragraphs 221 to 236).

3 'Akamai, The State of the Internet: 4th Quarter 2011 Report, May 2012. Available online: http://www.akamai.com/dl/whitepapers/akamai_soti_q411.pdf?curl=/dl/whitepapers/akamai_soti_q411.pdf&solcheck=1&. It is worth noting that there are a host of broadband measures and indicators, and methodologies vary. The overarching point here, without wishing to get involved in an analysis of the merits of different datasets, is that there is clear room for improvement. South Korea and Japan have repeatedly been drawn to our attention as world leaders; South Korea has the highest average connection speed in the world and cities in South Korea and Japan hold many of the top spots in the rankings. Back

4 Pitchup.com Back

5 Ofcom, Communications Infrastructure Report 2011-Fixed broadband data, July 2011. Available online: http://stakeholders.ofcom.org.uk/binaries/research/broadband-research/Fixed_Broadband_June_2011.pdf Back

6 The Boston Consulting Group, The internet economy in the G-20, March 2012. Back

7 South West Internet CIC Back

8 Q 128 Back

9 Sometimes aluminium is used. It has lower performance than copper. Back

10 There is some possibly counter-intuitive terminology used in this regard. While 'deep into the network' might imply the part of the network towards the core, in fact, it is generally understood to refer to the end part of the network, the 'final mile' or local access network. Network engineers view the world from the centre of their network, so they use 'deep into the network' as a city dweller might use the phrase 'deep into the countryside'. Back

11 This usage is figurative. It refers to the final leg of connectivity to an end-user. This is often more than a mile in length, and may need to be made much shorter than a mile to deliver faster speeds, depending on the technology used. Back

12 With reference to written evidence received from the Parliamentary Office of Science and Technology. Back

13 BSG press release, 'BSG publishes costs of deploying fibre based superfast broadband', 8 September 2008. Available online:

http://www.broadbanduk.org/component/option,com_docman/task,doc_view/gid,1035/Itemid,9/ Back