CW Talk: Solving 5G rural connectivity regulation or innovation?

This article is from the CW Journal archive.

The Speakers

Mansoor Hanif
UK5G Board member, previously Director of Radio Access Networks at EE and since this talk appointed as CTO of Ofcom

Richard Feasey
Formerly Group Policy Advisor to Vodafone for 12 years and currently on the Competition and Markets Authority Panel

Not surprisingly, the speakers had rather different views, Mansoor, speaking before his recent appointment as CTO of Ofcom, focused on the possibilities opened up by recent technology developments; whilst Richard questioned, first, if we really understood why rural coverage was seen as important; and second whether the regulatory structure that had been in place for the past 30 years is fit for the purpose. Both speakers have been kind enough to summarise their views for the Journal and we think this is a useful contribution to an important debate. If they stimulate thoughts and ideas, let us know, we are keen to continue the debate in these pages.

The past six years of rapid 4G network development and large-scale investment have put the UK back to a pioneering position in mobile connectivity and innovation. The UK now has the leading 4G subscriber base in Europe and has led the commercial deployment of technologies such as LTE-Advanced, Voice over LTE (VoLTE) and Voice over Wi-Fi. It will also be among the first countries in the world to transition its legacy voice-only Emergency Services Network (ESN) to full 4G capability. There has been a similar push to extend network coverage and reliability in rural areas, with a new aspiration to go beyond areas where people live and work (population coverage) to cover all areas of the UK territory – this year, UK networks will have reached an unprecedented 90 per cent of geographical coverage.

There are still many areas for improvement, in particular there is a widespread perception that rural coverage is poor. Coverage improvement in these areas is very recent and unevenly spread between networks.

It will take time to change public perception – especially where rural coverage has been extended by using the lower 800 MHz 4G frequency, which VoLTE-enabled devices can only access on certain networks. In many cases, this negative experience is due to patchy coverage in less remote, more residential rural areas within commuting distance of towns and cities.

One of the main factors is that antenna heights are restricted in the UK, with a standard antenna height of 15 metres being very common even in rural areas, whereas in many European countries a standard height of 30 or even 40 metres would be applied. Mobile masts are often surrounded by trees – many of which were planted over 20 years ago to lower the visual impact of the masts – unfortunately, local planners ignored the fact that trees grow but masts don't!

In those early days, council policies took into account widespread local opposition to the installation of new masts. Nowadays the situation has been reversed, with many of the same local town councils now pushing for new mast installations as mobile coverage has become more central to social and economic well-being, especially in rural areas. While there is still opposition, this is now from a minority in most places – but even then this may still block the installation of new masts. Mast height extensions are now allowed but are costly and take a long time due to the impact on existing services.

Other major obstacles include the cost and difficulty of site access and installation, and for providing backhaul. In some cases, extra dig costs alone for fibre-based backhaul can reach £80k or more – nearly triple – due to the long distances involved and lack of existing ducts. While much progress has been made in designing alternative backhaul solutions, the requirements have risen even faster and will remain challenging.

Due to this increased cost and complexity in rural areas, an upgrade or new site installation can often require three to five times more capital expenditure than an equivalent urban site – in many cases more than several hundred thousand pounds per site.

The UK is Pioneering the transition of emergency services to 4G

Building and switching-on is only the beginning, network reliability is increasingly important as remote communities become more dependent on connectivity. This is especially acute in remote areas due to the large distances involved, challenging transport issues, and relatively unstable power supplies in extreme weather, all of which lead to high costs for maintenance and operation. The inherent low population density per rural site often leads to a negative return on investment.

All the mobile network operators (MNOs) participated in the Mobile Infrastructure Project (MIP), a government-led initiative to improve rural connectivity by building shared, publicly-funded rural masts. The initial target of the three-year programme was to deliver 575 rural sites with £150m funding. Despite the government support, only 75 new sites were actually delivered. There are many factors which led to this disappointing result. The MIP Impact and Benefits report, published in October 2017, gives some insights to the economic challenges: the average site cost was £477k per site and 63 of the 75 sites built were considered commercially unviable based on an estimated 20-year profit of only £1k per premise covered.

As this was a publicly-funded project the site-selection and procurement process was hugely complex and led to many delays and drop-outs – this could be mitigated by careful selection of the appropriate public/private vehicle for the project, as was done for the Department for Digital, Culture, Media & Sport (DCMS) Local Full-Fibre Networks programme. Simply focusing on the economic model alone is not enough to resolve the rural connectivity issue – a broader approach which takes into account the latest technological and regulatory innovations is required. For example, alignment of technical specifications across four MNOs and other partners was very difficult, especially with legacy hardware and software, and limited technical options for shared configurations.

New virtualised radio access networks (vRAN), driven by the onset of 5G virtualised infrastructure, are now available and could make implementation significantly easier. There are other technical innovations which have since emerged and could eliminate many of the difficulties faced by MIP – these innovations should be considered, not necessarily to create a new MIP but to encourage greater co-operation to improve rural connectivity through innovation. Beyond MIP, alternative solutions, initiatives and debates in recent years include:

• Imposition of National Roaming Between MNOs
  This was strongly resisted by MNOs due to cost overheads and negative customer impact – but these views were based on the legacy roaming solutions available in 2015. Today, there are innovative commercial solutions (such as DENTwireless ) based on blockchain and dynamic smart contracts which are much more flexible and could be adapted for localised roaming models in rural areas, and could be more appealing both to MNOs and users. These new models should be evaluated.
• Enabling and Promoting Community-run Local Networks
  B4RN, CH4LK and other examples exist today but they cannot scale, owing to limited interaction with the MNOs due to the high cost of integration and the lack of carrier-grade, community-run networks.
  Here again, recent technical innovations offer new perspectives, including the Open Cellular project, tested in Scotland by BT with Facebook and Nokia, which offers a framework for a light-touch, community-run network model which can also be considered carrier-grade for sustainable operations
• Increasing Coverage Obligations
  The recently adopted 90 per cent geographic coverage obligation actually had limited impact since it was superseded by MNO commercial plans and the deployment of the Emergency Services Network (ESN).
  Imposing new coverage obligations in return for public-funding alone is unlikely to provide the desired results and will still face the obstacles described earlier. A broader approach is worth consideration where MNOs could be incentivised to co-operate with each other or with community-run networks through the development of innovative national roaming solutions, or in return for having the resulting coverage counting against any coverage obligation.
• Committed Public Sector Support
  The UK Government is committed to opening up sites to mobile through the revised Electronic Communications Code (ECC), reformed planning regime for mobile, barrier busting support, and the Digital Infrastructure Toolkit, as well as fostering better relationships with key landowners, such as National Parks, National Trust, and Network Rail. In addition, new initiatives have embraced the lessons from MIP: Scottish Government 4G Infill and R100 (Reach 100%) programmes, DCMS 5G Trials focusing on the remaining 4G pain points including rural, rail and roads. Convergence and alignment between 4G/5G (5GTT), Full Fibre (LFFN) and Broadband (BDUK Superfast) programmes could produce high impact in rural areas.
• Focus on Innovation to Drive Geographic Coverage to 95% and Beyond
  There is growing recognition that simply building larger towers in increasingly remote areas is neither desirable nor feasible. MNOs have increasingly engaged in initiatives such as the Scottish Innovation Partnership (SIP) and the Telecoms Infrastructure Project (TIP) driving innovation in rural connectivity. The ESN programme is deploying pioneering LTE-Advanced technologies for coverage and resilience (satellite, meshed networks, airborne coverage, rapid response vehicles, vehicle gateways) which could be applied to rural connectivity as a whole.

The rapid advance of 5G is driving an ecosystem of innovations which will arguably have an even greater impact on rural connectivity than in urban areas due to the high-cost and lower average uptake of rural connectivity:

• Network Slicing Makes most effective use of underused network resources (RAN, core and transport), especially high efficiency for remote areas where usage is sporadic.
• vRAN Allows dynamic and remote reconfiguration for multi-host RAN. Perfect for rural areas where customer access is often sporadic, allowing a higher ratio of sharing and more flexible reconfigurations.
• mmWV/Beamforming 3D beamforming can significantly increase the quality and reach of fixed wireless access (FWA) solutions for rural areas on high towers. It also increases the possibility of effective integration with airborne and low Earth-orbit (LEO) satellites and is particularly effective for tracking vehicles along roads and railways.
• Multi-Access Orchestration Ability to Integrate and orchestrate various access technologies, including Wi-Fi, Lifi, TV White Space (TVWS), MulteFire and various other unlicensed technologies. Important in rural areas to make the best use of all available options – which are likely to be scarce.

I have spent a lot of time trying to understand why markets fail to meet the needs of consumers are supposed to do. Behavioural economics tells us it may be because consumers do not act in ways they say, or think, they will – there is a disconnection between thought and action. This is also true of politicians and policymakers and, for me, a classic illustration of this is the 'problem' of rural connectivity.

The problem arises, in part, because 30 years ago we decided to allow markets and privately-owned firms to take responsibility for delivering digital communications services (and a lot of other 'essential public services' like energy and water) but there has been little or no effort to reset popular expectations about the resulting outcomes. The result is we still expect communications companies to behave like the formerly publicly-owned utilities, but to do it more efficiently. That means we expect networks to be available everywhere, and to everyone, at the same price, ignoring the huge geographic differences in costs and returns.

There is also a 'problem' because there is a popular belief that 'connectivity' is important for economic and social inclusion, and that rural connectivity is needed to avoid further rural depopulation. But there is no precision about what we mean by connectivity: is it connectivity or mobility? In-building or outside? Available or affordable?

Moreover, there is a lot of anecdotal evidence but not much data about the benefits of connectivity for rural areas. The Real Wireless study from 2014 for the Scottish Government is one of the few and claims productivity benefits for 98 per cent 4G indoor coverage (itself quite ambitious) of only £300m per year, adding only 0.025 per cent to the growth rate of the Scottish economy.

In the UK in general, the Mobile Infrastructure Project assessment from 2017 for the Department for Digital, Culture, Media and Sport found that 75 rural mobile sites yielded a negative net present value (NPV) with benefits of 84 pence for every pound spent. Even if we accept the study's assumption that consumers value the incremental rural coverage at an additional £13.40 per month; and businesses at £29.60; these figures are not very compelling.

What evidence I have seen suggests it is the urban poor, not rural dwellers, who feel most digitally excluded. Many rural households retain their non-digital or pre-digital lifestyles and so value connectivity less than their urban counterparts. Urban visitors to rural areas (like MPs) seem to be the ones that complain the most.

If there is actually demand for rural coverage, and assuming consumers are prepared to act accordingly, why doesn't the competitive market rise to meet it? Responding to consumer needs is what, after all, competitive markets are supposed to be good at, compared to old-school, public sector provision. I don't fully know the answer but I observe that pricing for mobile services in the UK does not allow the mobile operators to monetise demand for incremental coverage – the extra £13.40 or £29.60 I referred to earlier. Customers pay the same price wherever they use the service and nobody offers a cheaper, 'low coverage' version in the UK or charges higher prices for calls made from rural areas.

This is not the case in Australia, where Telstra can, and does, charge a significant premium for its vastly superior geographic coverage, as do Verizon in the US. In my view the UK is just too small in geographic terms to allow any operator to open up an unmatchable coverage gap, relative to its rivals, and charge for it today. Coverage competition does not arise as a result.

Missed opportunities
The UK operators have also been understandably wary and reluctant to engage with politicians to solve the connectivity problem. Universal service is viewed as an unwelcome financial cost for little commercial benefit, a service which has a habit of expanding over time and where outdated commitments (payphones) are difficult to cast off. The existing players are always reluctant to abandon existing business models and new players see only niche commercial opportunities in rural areas.

Some opportunities have been missed. For example, 2G national roaming as proposed by the Government a few years ago would, in my view, have meant better coverage for UK mobile customers – some of which should have translated into higher average revenue per user (ARPU) for all operators – and would have had little impact on the competitive positioning of any one player. BT and other fixed operators have also failed to challenge misconceptions or pushed aggressively for regional pricing of broadband to match costs better (the outcome of a normal competitive market). This would be a long campaign but one on which it would be I think it would be worth embarking.

It is the urban poor not rural dwellers who feel most digitally excluded

The UK industry is not unique in this regard. Throughout Europe, the telecoms industry has been wary of working with government. It has not argued strongly for state aid in telecoms, with the result that a tiny share of EU structural funds – about three per cent – are today allocated to broadband. This compares to over 50 per cent for renewable energy and environmental projects.

If rural connectivity really is important then the industry (and the government) ought to be arguing for its fair share of the available resources. It has failed to do that so far.

When the UK Government does intervene, is it more ambitious than its counterparts? When David Cameron announced the 10 Mbps Universal Service Obligation (USO), only Spain, France and Finland had targets and none were more than 1 Mbps. Similarly, the Broadband Delivery UK (BDUK) scheme of 2011 was among the earliest in Europe.

Rural superfast coverage in UK is near 90 per cent today, but half of EU28 countries are still below 50 per cent including Finland, Sweden and France.The results of the Government's efforts have been mixed: Lord Carter's scheme distributed only 10,000 out of a possible 300,000 vouchers and the Mobile Infrastructure Project (MIP) only built 75 out of 575 planned mobile sites.

The BDUK scheme has successfully pushed VDSL connectivity to over 80 per cent of UK households, although over £500m of the £1.7bn paid looks to be unnecessary and has been clawed back. The BDUK scheme, however, has left the government with a huge challenge for fibre transition because it did nothing to challenge BT's dominant position.

So what can we do?
We need to be very clear about our expectations and ask some difficult questions. Is rural coverage more important than driving adoption rates? I doubt it but politicians always have an unfortunate bias for the infrastructure and supply side. Must coverage be better in the UK than Sweden? Is it sensible to disregard huge geographic differences in network costs and have every household to pay the same, or are we in the business of narrowing gaps?

Once we are clear about our goals, we should be prepared to pay for them, recognising that the market and existing players cannot deliver – and we should stop waiting for them to do so. We have to take the task of solving rural coverage out of the everyday competitive process.

We should be more radical about the business models used. Think about wholesale-only models, public/private partnerships to solve co-ordination issues on transport links and other synergistic infrastructure, plus franchising and tendering of rights to serve designated areas and unlock innovation. We should use public money for strategic purposes, including influencing how the market develops in competitive areas, as well as just filling in gaps. This means finding billions, not low hundreds of millions. Government major infrastructure projects involve £225bn to 2021, so I don't see why £5bn to £10bn for connectivity objectives is implausible.

We need to start aligning our actions with our words.

Mansoor Hanif
Chair of the UK5G Testbeds & Trials Review Working GroupExecutive Director, Neom (former CTO, Ofcom)

Mansoor has 25 years experience of planning, building, optimising and operating mobile networks around the world. In 2011 he joined EE and was accountable for the technical launch of 4G as well as the Integration of the 2G/3G Orange and T-Mobile networks. Mansoor was Director of Radio Networks and a board member of MBNL (the joint venture of EE with H3G) until 2016. At BT he was Director of the Converged Networks Research Lab from 2016 to 2018. From September 2018 to December 2019 Mansoor was the CTO of Ofcom, the UK telecoms and media regulator. As CTO he was head of the technology profession across the organisation, ensuring that the technology aspects of all Ofcom’s policy work are informed and robust, and was a member of the Policy and Management Board. He also lead Ofcom’s activities in network security & resilience as well as engagement with CTOs at our stakeholders, industry bodies and with Government. Outside of Ofcom, Mansoor is patron of the ITP, a member of the Steering Board of the UK5G Innovation Network, and on the Advisory Boards of the Satellite Applications Catapult and UCL Electrical and Electronic Engineering Dept.

Richard Feasey
Formerly Group Policy Advisor to Vodafone for 12 years and currently on the Competition and Markets Authority Panel

Richard is currently a Panel Member at the UK's Competition and Markets Authority and advises a number of public and private sector organisations. He was formerly a special adviser to the House of Lords EU Sub-Committee and is currently an adviser to the National Infrastructure Commission. He is a Research Fellow at the Centre on Regulation in Europe and lectures from time to time at University College, Kings College London and the Judge Business School. Richard had a 25-year career in the telecommunications industry and was Director of Policy at Vodafone plc between 2001 and 2014, where he was responsible for Vodafone's engagement with competition and regulatory authorities across Vodafone's global footprint. www.fronfraithltd.com