Good afternoon. My name is Scott Coates and I am the CEO of the Wireless Infrastructure Group, an independent British wireless infrastructure company that builds and operates communication towers and fibre networks.
Q In your written evidence, Mr Coates, you talked about the need for greater diversity in the ownership of mobile infrastructure. Does the Bill go far enough on that?
We welcome the measures in the Bill to improve the speed at which infrastructure can be deployed and to improve the economics of deploying the infrastructure. It is critical to understand that there are different ways of deploying infrastructure. There are different ownership models, for which the Bill could have different impacts. When I say “infrastructure”, I mean the kind of mobile and fixed infrastructure that you see in the field, whether that is cables, ducts, cabinets or communication tower facilities.
There are two different types of owners of those types of infrastructure. First, the vertically integrated players are effectively building and operating that infrastructure for their own networks, primarily, and their business case is based on their economic use of that infrastructure. Secondly, you have a growing pool of independent infrastructure companies, of which we are one. We are very different from the traditional, vertically integrated players in that we are investing in infrastructure not for our own network, but to provide access, on a shared basis, to all other networks.
If I talk about mobile infrastructure, around a third of the UK’s communications towers—of which we think there are around 27,000 in the UK—are independently operated. It is really interesting that, globally, there has been a very firm shift over the past decade towards more independent operation of such upstream digital infrastructure.
Currently, more than 60% of all communication towers globally are held in an entity separate from the networks that use them. In countries such as India or the US, that figure is somewhere between 80% and 90%. There are real benefits that flow from the independent ownership of infrastructure. We are trying to do more in the UK, but the UK currently lags behind in the global statistics I mentioned.
I do not think that the Bill does anything to encourage more independent infrastructure. The Government’s policy position at the moment is very clear: they want to maintain investment incentives for independent infrastructure. To achieve clarity on this requires that the Bill is worded very carefully.
When we deploy our tariff facilities and infrastructure on or adjacent to land, as things are now one of the definitions of UK land often covers things that sit on that land. One of the potential risks is that if the activities we engage in and the facilities that we deploy are not carefully carved out, they risk being treated as land. Under the new valuations principles in the communications code, that potentially risks giving them no value or low value, which would obviously be devastating to investment appetite. The consequence of that would be further concentration of infrastructure ownership in the hands of the larger, vertically integrated players who have different incentives from us when they approach this.
Q So there is potential for this to get worse, but what could be done to actually encourage more independently owned infrastructure?
We would like to see a carve-out that is as clear as possible for the activities that we are engaged in. We would like to see it made absolutely clear that the communications code, which is a compulsory purchase tool to bring land into the telecoms sector, does not drift beyond that focus and risk entering into what is really Ofcom’s territory, which is to govern the relationships between telecoms companies.
Dr Whitley, if I may jump to part 5 of the Bill, we heard earlier that there were concerns that the Government have not taken sufficiently into account safeguards around privacy and personal data. Do you think that this strikes the right balance between open policy-making and privacy?Q
My main concern with part 5 is that the detail is just not there. The codes of practice that one would expect to have there, which would give the details about how privacy might be protected, are not present. We have been involved with the privacy and consumer advisory group. As far as I can tell, we had our first meeting with the team who were developing these proposals back in July 2013. We said from the very beginning that we want detail, because when we have specific details we can give advice and suggestions and review it, but we have never had that level of specific detail.
Whether it is in primary legislation or in codes of practice, my personal view is that you need a certain level of detail to be able to make an informed decision. Otherwise there will be some vague position of, “We will share some data with other people within Government. Trust us, because we are going to develop some codes of practice that will be consulted on and will then be put in front of Parliament. There will be protections and it will all be fine”. We are saying that there are lots of different ways of doing that. The earlier you give us at least a first attempt at those details, the better we can improve it.
It depends. There has been talk along the lines of there being codes of practice and liaison with the Information Commissioner’s Office, so at a very high level there has obviously been some discussion. But at the very specific level—for example, the civil registration clauses talk both about allowing a yes/no check around whether there is a birth certificate associated with a family, while on the other hand there will be bulk data sharing within Government so that different Departments can know stuff and possibly make things better for society.
One half of that seems to be quite specific, and you can see how it could well be designed as a simple “Does a birth certificate exist for this person?” and the answer is yes or no. The privacy protections around that are reasonably well known and not very much data is being shared. Then the other illustration just says, “we will share these data with other bits of Government” and there is nothing there about what kind of privacy protections might be put in place. There are many different ways in which that can be done, but until we have some specific details, we cannot give you sensible reviews as to whether that is a good or not so good way of doing it.
There is no doubt that for the last 5%, maybe a greater proportion than that, wireless technologies have a significant role to play. Six of the seven trials run by the Department for Culture, Media and Sport earlier this year were of a wireless-based structure. I think there is a role for it. It is also interesting, as you look beyond 10 megabits to the future when universal service means something far more substantial than that, that a new disruptive technology is coming.
Everyone is talking about 5G; it does not really exist at this stage, but we know it is going to be ultra-high bandwidth, ultra-low latency, with the potential to be a disruptive technology and replace fixed line to the home. Some countries around the world that have not had the wave of fixed line technology roll-out will be moving straight to wireless as their domestic broadband service.
With 5G, it is really hot and talked about now, but it is still some way off. Mobile operators will market it strongly and talk about it strongly, but there was something last week from France Telecom admitting it does not know what it is yet, and that is the substance of the matter.
As we come in to the early 2020s, at the beginning of the next decade, we will start to see something. Interestingly, the infrastructure that is going to enable it is starting to go down now, so particularly in urban areas; as the concentration of cell sizes needs to get smaller and smaller, the infrastructure needed to power faster 4G services will ultimately be the infrastructure used to power 5G.
Coming back to the structure of the industry, it is critical that there is a competitive infrastructure market for 5G. As a new technology that is a combination of wireless and fibre, it has the opportunity to have multiple infrastructure parties competing. It also carries the risk of being a monopolised infrastructure.
In terms of using wireless to achieve USO, mobile as a technology has a very clean and efficient way of pushing out coverage to rural parts of the population, and that is through the licences. There is another major round of licensing, with something called 2.3 and 2.4, which is coming soon.
There is also 700 MHz, which is a really powerful frequency for delivering coverage into rural areas and which has already been licensed in many European countries. It is not licensed here yet, but the rules of those licences create an opportunity to get coverage out to the most rural parts of the country. You could do things like in Germany, where they said rural areas have to be covered before urban areas. That is the most efficient way of unlocking coverage from a wireless perspective in rural areas.
One of the biggest challenges facing coastal and rural communities like mine is the problems with undulating coastlines and areas of outstanding natural beauty. I am interested in your thoughts on how we can strengthen the Bill to make sure we get out to some of the rural areas left behind in the pastQ .
I refer you back to the last question. The most efficient way to deal with that is through the licences. There is licensing coming up that will create an opportunity. Unfortunately, it is going to be a few years before the airwaves that deliver that are available for deployment.
There is a lot of activity happening in the sector at the moment. The mobile operators are very busy investing in their networks and we are working hand in hand with them to help them deliver that. I know we are building new towers in coastal areas right now; I do not know if we are building one in your constituency. So it is getting better. Bear in mind that the Government struck a deal with the mobile operators 18 months ago and the operators are busy investing on the back of that. In the last 4G licence, when the 800 MHz got auctioned, one of the licence lots, bought by Telefónica, required it to cover more of the country, so Telefónica is investing on the back of that as well.
Q I want to push Dr Whitley on the privacy question. I think that what you are asking for, a code of conduct and some clarity, is reasonable, but equally, we cannot know what the demands and the questions might be going forward, or the data requirements. I look back on where Government do share data, querying the national insurance database, or, indeed, the Government ID project, where DVLA records were queried as a measure of identity, it all appeared to be fine, there were no issues of privacy or data loss, to my knowledge. In a way, should we not be taking on trust—I know that trust is a word people never like to use with Government, whereas we trust corporates all the time with all kinds of data—that we have not had a problem and that the right rules and procedures and the spirit of privacy will be protected?
You have highlighted a very privacy-friendly way of checking data that says, somebody has a database and you look it up and you say, “This particular person, or this particular attribute, is it true, yes or no?” Referring to the previous evidence session and the question, “Is this person over 18 and therefore able to access?”, yes/no seems a perfectly reasonable way of doing that and that is the kind of thing that we have been encouraging Government to do. As you say, the Verify programme uses exactly those kinds of checks. The problem is that, without that level of detail, it is not at all clear that that is going to be proposed for all parts of the data sharing. Again, with the civil registration data, they say explicitly, “We want to do bulk sharing” and that is, by definition, not a yes/no check. That is, “Here is a set of data that we have that we think will be useful for your Department to match against and thereby tailor particular services.”
As the National Audit Office reported a few weeks ago, there were 9,000 data incidents within Government in 2014-15. If you start just moving the data around, you really run the risk of data incidents of varying levels of severity, and if you do not have that detail you have to rely on trust. Is it not better to have that detail, so you can say, “This is what we want to do, this is the way we are thinking of doing it”, and ask experts, not only in PCAG but in general, “Do you have any issues or concerns about that and, if you do, what alternative ways might there be for addressing those?”?
Q Do large corporate families do that? Nobody ever reads the Ts and Cs, but if they do, do you give explicit permission for your data to be handled around the Facebook family, for example, in the way that you suggest Government should specify? That is just a question from ignorance.
I do not know exactly how Facebook would handle it, but even if you are not worried about the data breach and data loss issue there is just a simple efficiency thing: it is a lot easier to have small pieces of data—yes/no, they are interested in this form of cat food, they are interested in those kinds of holidays, therefore target adverts based on that—than sending huge swathes of data to other parts of the system for duplication and therefore increasing the risk of data loss.
Yes. From my perspective you start with a privacy concern that says, minimise the data that you are handling, do not have it in duplicate locations all over, but a consequence of starting with that privacy concern is that you also have very clear operational efficiencies; that you are not duplicating data and you are not having large amounts of data in your system, because the more data you hold, the more likely it is that there will be a breach, an attack, an accidental loss or whatever.
Mr Coates, will you expand a little on your experience, internationally, of licence requirements in broadening coverage to rural areas? What is the specific benefit of independently owned infrastructure for rural communities, in bringing access to places which struggle with mobile signal today?Q
I am going to pick on two countries that we have looked at making investments into. Germany, which I mentioned earlier, has an outside-in policy, so you have to cover their rural areas with your new batch of spectrum before you are allowed to deploy it into urban areas. France has got a very interesting model, in which they have compartmentalised the whole country. At the moment the Ofcom licences ensure that Scotland, England and Wales have their own targets, but if you break it down even further, the demands become higher on those targets. We have seen some targets in France where, by compartment, we are looking at 99.6% coverage by 2027. They have given the industry a long time to reach that target, but it is very bold. If people knew it was going to get better, maybe it would become a bit more understandable. This is not like changing a lightbulb; this is infrastructure that needs to be built.
I think there are three benefits of independent infrastructure. First, there is clear evidence that it enables better connectivity. Because our infrastructure is operated independently of a network, we do not have any of the conflicts of interest that normally exist in the vertically integrated model, in which the infrastructure owner is forced to provide access to their competitor. Because we focus only on infrastructure—it is our core business model—we tend to build better infrastructure, and we share it with more networks. There is evidence out there. Ernst and Young looked at this last year and studied independent communication tower ownership across north America and Europe. They compared it with communication towers that are owned as part of mobile networks. They found that there are twice as many networks using the independent infrastructure, compared with the vertically integrated owned infrastructure. That is twice the productivity coming off a piece of infrastructure, which is transformational, when it comes to enabling connectivity, particularly in rural areas.
The second benefit is around investment. At the end of the day, solving these problems comes down to investment. Independent infrastructure opens up a whole new channel of investor and brings a different type of investor into our industry: long-term, low-cost-to-capital infrastructure investors who are targeting infrastructure only. They do not want to invest in the retail operations or in buying premiership football rights; they want to invest purely in infrastructure. We can be a conduit to bring in that capital to invest in infrastructure. Earlier this year, after 10 years of various rounds of financing, our business announced a major fundraising transaction with a UK blue chip infrastructure investor—3i Infrastructure plc—and a north American investor that invests on behalf of state pension plans. That is exactly the kind of capital you want—long-term, patient capital—fuelling the growth of infrastructure.
The final benefit is in and around competition. We create competition at the infrastructure level. On the fixed-line side of the market, you can see some of the challenges from a lack of competition. But we also enable competition at a retail level, because our infrastructure is open for everyone to use. Mobile operators are the biggest users of our infrastructure, but well over 100 different network use our infrastructure in rural areas. Sometimes that can mean a local wireless broadband company that simply cannot afford to build its own infrastructure and would find it very difficult to get access to a mobile operator on a piece of infrastructure. On average, every one of our towers in the UK supports a non-mobile operator network running over it. Those are the three benefits of independent infrastructure.
There was a Coates who played for Liverpool. He was from Uruguay, so they called him Co-ah-tez.
Q He will again.
I would like to ask you about the USO, and then I would like to come on to the mobile environment. I have a problem with the USO not just because of the lack of ambition and what 10 megabits means for people living in those areas, but because the tactical low-speed USO will not push fibre a lot further. The lines between wired and wireless are blurring all the time, so would a more ambitious USO with faster speeds help you, in terms of pushing fibre further and putting other infrastructure out there?
I think it comes down to the cost element. The further out you go with fibre, the more expensive it becomes. Our infrastructure in rural areas tends to be bigger pieces of infrastructure, so quite often there is fibre coming through it or it links to a site that has fibre, and that creates more bandwidth to power the wireless services coming over it. More generally, I would say that the USO is a start. No one is going to be happy with 10 megabits in a few years, but I would say that you need to start somewhere and it needs to be manageable from a cost point of view.
Q I will not ask you, then, whether the Scottish Government’s policy to have 30 megabits everywhere is more appropriate. I think that everyone is in agreement that the electronic communications code needed to be reformed, and there are some welcome measures in there, but as an independent infrastructure provider, do you honestly think that that will lead to more coverage by mobile providers, or will it simply give them a better bottom line?
There are certainly measures that will make it easier to get rid of the bottlenecks and faster to resolve disputes. Running cables to connect up mobile sites has been a real challenge, so being able to fix those problems—that is not really about economics; it is about having faster resolution. Some of the pricing elements I do not think will have a material impact in rural areas when the commercial case to invest is not really there for the mobile operators anyway. The only way you can deal with that is through the licences. The new code will help to remove some of the ransom costs that we see in the industry and certainly give us a much more powerful weapon against those, but on a day-to-day basis, we do not expect to be moving towards compulsory-based conversations with our customers. The industry needs to work on a voluntary basis. That is absolutely essential; it is how it works everywhere else in the world. We have busy infrastructure facilities. We are there on average every 12 days. We need to have a good partnership with our land providers. The code is a really helpful and powerful new tool of last resort, but our whole industry needs to maintain a voluntary basis of engaging as our MO for dealing with landowners.
Q Thank you very much. Mr Coates, I thought you gave a great overview of why independent infrastructure is really important. You obviously feel a concern, so is there specific wording that you would like to see in the Bill that we could discuss at the next stage to ensure that you are protected and the value of your assets is not lost?
Thank you for that question. We have had a really engaging journey with the Minister’s officials. They have been very diligent and transparent in engaging with us all the way through this fairly long process on the communications code. Our concern generally is that there is a fine line between the technical drafting that says that what we do on land is not covered by the communications code, and the risk of a legal challenge that it might be and might have nil or low value. What we have really asked for is as much clarity as can be provided. That will help to enhance the investability of our business. We are in a different place from the mobile operators and some other network providers, because we do not get any economic benefit from our own infrastructure; it is built for other people to use, so we are not a net user of infrastructure.
Q So that is a, “Yes, if possible, please.” It is okay; do not answer that. You have already answered. My final quick question is this. Although this is not retrospective, is there any case for excluding existing sites, if this is really about building out more network, in terms of the valuation element, given that a lot of those sites are actually on publicly owned land?
There is certainly a difference in the substance of a transaction when you are approaching a farmer, a sports club, a university or whatever and asking for access to build a new piece of infrastructure where there is new coverage, and you are having that negotiation in the context of a new communications code that has tighter reference points on pricing. You will have more leverage for that conversation. You will still end up, I believe, paying them a rate way in excess of what zero value would be because that is just how you have those conversations, but it will be less than what is paid today, that is for sure, because you have got this new reference point. The substance of that is very different from the substance of a voluntary agreement you entered into with a firm six or seven years ago and that comes up for renewal in two to three years and the infrastructure is already there.
I think it is important that we have a robust set of tools as an industry but, as I mentioned earlier, it is equally if not more important that the industry acts responsibly and avoids behaviour such as forcing situations where they need a new compulsory purchase tool, even though they have already got access today. There is definitely a way of engaging on existing sites that should be a bit different from new sites, as part of a package of trying to maintain the voluntary support of the land and property sector for our industry.
Q Could you set out in more detail—you have already gone into this a bit—about what you mean when you say that the code should include land owned by the infrastructure providers but not the apparatus, and the distinction there in the written evidence?
It comes back to this. Under UK property law, anything that affixes to land could be considered land. At the moment, the code effectively is to regulate land coming into the telecom sector, not to regulate the relationships between telecoms companies. It carves out from land the apparatus.
I am advised that there is a risk of ambiguity. That is probably the best way I could describe it. It may be challenged down the line. This is an evolving and dynamic industry where we don’t exactly know the physical things we are going to be deploying in future. There is a risk that some of the things we do might receive a challenge that it is land not apparatus. I do not know.
Is a new runway at Heathrow infrastructure or land because it sits on top of land? Is the national grid transmission network an infrastructure asset or land because it sits on land? It is a fairly technical point. Like all these things, once the lawyers are running around looking at them, they will find concerns.
All we are saying is that we invest over 20 to 30-year horizons. The more clarity that can be provided is helpful. We acknowledge and clearly appreciate the intent behind Government policy to protect investment and passive infrastructure but more clarity around that will only help the investability of what we do.
Q Thanks. I am also grateful for what you said about the team at DCMS, who will have picked up on your kind words I am sure. I wanted to follow up on 5G. You talked earlier about the 5G roll-out. This is a bigger-picture question. What do think the Government need to be doing now to ensure that we are in the lead when it comes to the roll-out of 5G?
People must be exhausted with hearing about the challenges with Openreach and what can be done there. The key thing is to help facilitate our competitive market for infrastructure. So 5G has the ability to be driven by the mobile operators, by the fibre players, by independent infrastructure companies. If you look at the US, half the small cells that power 4G and 5G are actually going in by independent infrastructure players; mobile operators as well as fibre players are in there, too.
It comes down to helping to facilitate as competitive a market as possible. We have started deploying infrastructure in at least one city in the UK: 4G initially, but it will lead to 5G. We would love to be able to get a competitive basis of access, or any access, to BT ducts. We cannot do that, despite the fact that they can access every single piece of our infrastructure.
That is one thing. The other thing is around the planning permissions for affixing equipment to lampposts. We are working in Aberdeen and I have to say that we have had a fantastically positive experience with the local council, which has been amazing and very supportive in everything we have been trying to do there. That experience is not shared across other councils in the UK.
Q Thanks very much. Dr Whitely, would you say that, done right and should the codes come out right, the clauses in the Bill have the potential to improve public services through better use of data?
Absolutely. You could have a side question about whether, for example, focusing on subsidies from energy providers is the best way to deal with fuel poverty, but in terms of that specific focus—if it is done right—then, absolutely. Our concern is that we just do not have the detail as to whether or not it is going to be done right. That has been the frustration over the last three years.
Q I want to talk about the spectrum licensing issue. We spent a lot of time in earlier sessions talking about the minimum average speed, particularly for SMEs, as being 10 megabits per second and whether or not that was ambitious for the future.
You talked about the outside-in licensing regime that could be possible—and is possible in other countries since it is being deployed, particularly for new tech and for the 700Mhz and the 5G licensing that will come. If that approach is adopted by the UK Government in terms of licensing, is it your belief that it would make that inequality almost go away and that it would deliver much greater equality across the pace of speeds for people to access business and other methods that they need?
If a policy objective is to ensure that rural areas get a high quality mobile signal, then forcing the industry to invest in rural areas—and effectively funding that by allowing them to pay less money for the licences that they acquire—is the most efficient way to deliver that. It would have positive outcomes, for sure.
Q So it would achieve that aim, in your view, and it would to a great extent future-proof the need to go to that level of where you are going from 10 megabits per second to a higher level, and then a higher level again. Is that correct?
Yes. The industry invests in order to stay competitive in areas where the market is working, and—where the licences oblige them—to invest in areas where the market is not working. The infrastructure needed to support some of these new services needs to be high bandwidth to support that, which will then support the uplift into the future in quality and speed of service.
This is not about big data but data-sharing, but there are opportunities there for big data to be used. There are questions about how you manage it and about how you handle it.
One of the other things that I am involved with is a steering group for the Administrative Data Research Network, which is where administrative data can be used by researchers in very strictly controlled environments to answer interesting research questions, generate hypotheses and explore those hypotheses by matching data from various different Government datasets. But that is done in a very locked down, secure environment with no mobile phones and no taking out of data and so on. So there absolutely are opportunities, but doing it right is what I particularly care about.
Q We are one of the most sophisticated digital economies on the planet and we have some of the brightest brains on the planet. Surely we can work this out.
Yes. The process has been going on for three years and we still do not have codes of practice. That is the bit that puzzles me. If we have all these brilliant brains can they not put together even a draft code of practice, so that we can know what we are talking about?
For example, in the consultation around fuel poverty, it talked about gathering data and matching up potential houses and individuals who might benefit or be at risk, and it says that they will inform the licensed energy suppliers as to which of their customers should receive assistance. That, to me, sounds like a push: “Here is a big set of customers that may or may not belong to your company. Check through that list to see whether or not any of them are your customers and give them a fuel discount.”
But then a couple of paragraphs further on—this is the consultation relating to the proposals—the Government would simply have an eligibility flag along with customers’ names or addresses for doing that. Even in the consultation, it does not seem that these brilliant minds have been applied as well as they could be.
Q Once we work that out, which I am confident we will, where are the opportunities? Where is the up side? Where is the positive stuff coming out of this? How can Government be better as a result of this? I am always an optimist.
Done right, there are fantastic opportunities. Government is digitising. The GDS has got lots of experience about how to manage and handle and do attributes checking, which is what most of this is. There are definitely opportunities and the skills, but somehow something has gone wrong with regard to these proposals.
It is not as if the proposals have been rushed through in the past few minutes. We have been looking at these and asking for more details since July 2013 and we are still here without even a resemblance of a code of practice. Part 5 has six codes of practice that need to be developed and none of them is here. Yes, please, but some detail. I am academic; I want to see the detail.
Q As you say, it is an enormous shift in terms of data sharing within Government. Clause 29 would allow personal data on citizens to be shared if there is a
“contribution made by them to society” or wellbeing to be gained. That basically covers anything, doesn’t it? Why have the Government not produced even a draft code of practice at this stage? How can we possibly be expected to vote on this while plainly placing blind faith in the Government?
You are basically saying what I was going to say. If you compare the comprehensive replies that Mr Coates has been able to give, talking about very specific details, with the vague “we don’t know anything” comments that I have made, you see that it is a real problem and also an issue for more general scrutiny of technological issues. If you do not have details about the different mobile phone frequencies that you are talking about, you cannot make detailed policy. Yet when it comes to data sharing, there is a sense that it will all work out in the end because we have the right people to do it.
We have consistently said—the Privacy and Consumer Advisory Group particularly, because we have this existing relationship with Government, but civil society and experts more generally—that we are more than happy to engage. We have repeatedly said, “Give us some detail. Don’t just come and talk about high-level stuff. Give us the detail and we will give you detailed comments to improve the process.”
That has worked very well in relation to the Verify scheme; that is privacy friendly and has a lot of support from the kinds of people who are very concerned about privacy. So the expertise is there and the working relationships are there. Give us an opportunity to help; we want to. It is just that we need something to work on.