Oral evidence: Emergency Services Communications, HC 770

Oral evidence, 16 November 2016.

Witnesses

Vincent Kennedy, Simon Frumkin, Peter Aykroyd and Jo Beresford-Robinson.

Chair

Welcome to the Public Accounts Committee. We are here to question a number of people—our first panel and, later, officials from the Home Office—about the very important programme to upgrade the emergency services communications, based on a Report by the National Audit Office. Emergency services UK-wide vitally rely on being able to communicate with each other. They currently use Airwave, which is run on a digital radio system. We are clear as a Committee that any new system needs to be at least as good as, and ideally better than, the current system. An effective and secure system is vital for the protection of all front-line staff—police, firefighters and others—and to their contribution to protecting the public. Whether those individuals are in a rural location, in London, on the London underground or policing something like a football match, they need to have the same degree of confidence in the system. I am delighted to welcome our first panel, which includes representatives of some of those frontline organisations. We have Jo Beresford-Robinson, the area manager for the emergency services network regional programme in the East Midlands Fire Service. We have Peter Aykroyd, the area manager of corporate services at Derbyshire Fire and Rescue Service. Vincent Kennedy is the vice-president and general manager of Motorola Solutions UK, which, to be clear, has an interest in the new ESN, and also owns Airwave. Simon Frumkin is the managing director of the programme for EE, the new provider of the new system. Welcome to you all. We are aiming for this part of our session to last about half an hour. I will ask Nigel Mills to kick off.

Nigel Mills

So, Mr Aykroyd and Ms Beresford-Robinson, I think it is just over a year until you start transitioning to using this new system. Perhaps you could tell us whether you are excited at the prospect or rather nervous about that situation.

Peter Aykroyd

I think it is fair to say we are excited about the prospect. The nervousness comes with the challenge of delivering such a programme. Certainly, we look forward to those challenges. We are acting as positively as we can and doing as much as we possibly can to prepare for that change. Members will be aware of the challenging timeframes on the programme. We certainly look forward to the challenge and some of the benefits that we hope to see from that programme in future.

Nigel Mills

I think that, technically, every organisation gets a choice as to whether they switch over, and they have to be convinced the new one is as good as Airwave. You are sat here confident and optimistic that this will be as good as Airwave in just over a year’s time.

Jo Beresford-Robinson

If we follow what the business case said, that is what we are being assured. I am quite sure that all authorities—local fire, police and ambulance services—will maintain the fact that we will need assurance before our chiefs sign up to it, so nobody will be switching over until we are happy that we have the same service, or a better one than Airwave is providing us with.

Nigel Mills

The NAO Report tells us—and I can tell you from my experience in Derbyshire—that EE coverage is nowhere near the 97% that Airwave has. Are you not at all nervous, at this stage, that there might be some rather large gaps in rural Lincolnshire or rural Derbyshire, Mr Aykroyd?

Peter Aykroyd

I can pick that up, certainly for Derbyshire. Obviously there is concern over getting the right amount of coverage, and we would seek that assurance before we recommend moving on to the new network. Remember that the driving factor for us, from the fire and rescue organisation, is public and firefighter safety. We need assurance that we have that resilient, reliable network, such that the emergency communications work and flow. The key thing for me—certainly on the network—is that we have a good, early flow of information, and that we are kept informed of the network roll-out plans in the areas that are fully 4G compliant, so we can get out early doors and test that network, rather than condensing everything late in the programme and then trying to squeeze in a lot of testing in a short period of time. The earlier engagement we get, the better communication will be. We can start building that up. I think investment early on in the assurance and testing will pay dividends later in the programme, because we will have that assurance. Like any sort of audit process, if we get assurance when we do the audit in certain areas, it makes it easier for us then to accept that the network coverage is up to the standard we require.

Nigel Mills

But at this stage, you are both fully bought into having to switch over from Airwave, starting in just over a year—

Chair

Sorry, I think we are talking about figure 12 on page 35.

Nigel Mills

So you both still fully expect to be using the new system in a couple of years’ time—in effect, during 2018.

Jo Beresford-Robinson

We hope to be using it. I don’t know whether we expect to be. We expect that there might be some movement to the right, because we have already challenged the timescales and we have done some work around looking at planning and planning assumptions. I think we expect to be switching on and utilising the system, but not necessarily within the timeframes that were initially being discussed.

Nigel Mills

How much parallel running would you like to have available, so that you can make sure the glitches are ironed out?

Jo Beresford-Robinson

I think total parallel running, to be honest, until we are fully sure that the new system is fit for purpose.

Nigel Mills

What length of time would you like to have parallel running for before you would be happy to turn off Airwave in the East Midlands?

Jo Beresford-Robinson

Our understanding is that the Airwave system will not be turned off until ESN goes totally live. Bearing in mind that we have other regions—cross-border regions—that are going later than ourselves in the East Midlands, that parallel running will obviously need to go right until the very last board is switched over.

Nigel Mills

I understand that, but under the current timetable, you will have parallel running for the whole of 2019, so you get a year of parallel running. If that became six months, would you be saying that was unsustainable?

Jo Beresford-Robinson

As I said right at the beginning, as long as we can assure our chiefs that it works, we will be happy to turn over.

Peter Aykroyd

I agree with Jo; it is about that assurance. If we have six months of excellent, fault-free running on a new network, we would be more assured in saying, “Yes, that’s enough dual running.” If that is not the case and we have to switch across to Airwave on a regular basis, because the network is letting us down, I would not feel assured in recommending to the chief officer in an authority that they switch off Airwave.

Nigel Mills

The contracts kind of mean that Airwave gets switched off at the end of 2019. I am trying to work out what will happen if you are saying you think the timetable will slip to the right, and you are expecting parallel running to start in late 2018 now. If we are saying that we need at least six months, and it slips more than a few months to the right, the East Midlands has a problem—and, frankly, nearly every other region has a big problem at that point. Are you feeling a little nervous yet that turning off Airwave at the end of 2019 might be a little optimistic?

Peter Aykroyd

I think there is nervousness, but it is probably early in the programme for us to give a definitive answer on that. For us to make that sort of risk-based decision really depends on how the technology rolls out.

Chair

You have mentioned that you have raised this up the line in your own organisations. Do you feel there is enough control by you locally to make that call? In the end, it is the Home Office that makes the judgment about whether to extend the contract, potentially at huge expense to the taxpayer.

Jo Beresford-Robinson

I think the governance arrangements, the local authority arrangements—as opposed to the authority of the Home Office—are strong enough for them to state whether they are happy to move over. We need to ensure that the Home Office understands the processes that local authorities have to go through, in terms of having to sign up to this and sign it off.

Chair

Do you think that is true at the moment—that the Home Office is aware of that process?

Jo Beresford-Robinson

We could always engage further with them.

Chair

You are very diplomatic, Ms Beresford-Robinson!

Nigel Mills

I have one last question. How happy are you with the new devices that you will be having to use on the new system?

Peter Aykroyd

We do not have sight of the new devices yet, so we are not able to comment on that, unfortunately.

Nigel Mills

That is quite important, isn’t it, because aren’t they due to be available quite shortly for testing? Have you had any input into what the system requirements will be, and in terms of what you would like to see on the devices, or is somebody else choosing that?

Jo Beresford-Robinson

The sector has had input. I think it was about two years ago that the specification was developed. I know that we need to be looking at that specification again, just to revisit it, because of the reallocation of time to the right, and certainly the programme has had a change of tack, in terms of the procurement strategy, so we, the users, are waiting to see the devices.

Nigel Mills

Are you slightly nervous that you do not now get much say in procuring the devices? Somebody—probably Motorola—is going to produce a device, and your choice is to take it or leave it, really.

Jo Beresford-Robinson

Again, we want to be involved in the testing, the trialling, of those devices.

Nigel Mills

Time is a bit tight for that now, if we are going to start changing specifications for devices. We are meant to have them in a couple of months; they are meant to be out being tested. It is a bit late to start rewriting specifications, isn’t it?

Jo Beresford-Robinson

That is why our view as user organisations is that this programme will have to slip to the right somewhat.

Chair

How much have you both invested in other technology, such as iPads—technology other than Airwave—to deal with data in your organisations?

Peter Aykroyd

On the front-line fire appliances, we have a device that we call a mobile data terminal, which uses data transfer for updating risk information that front-line firefighters access. It is also used for status messaging—a typical message back to a command-and-control centre that an appliance or a vehicle is at the scene of an incident and when it is available to return. So we do use that technology already for a certain amount of data.

Chair

What about you, Ms Beresford-Robinson?

Jo Beresford-Robinson

I am overseeing all five fire services; Pete can obviously discuss the situation for Derbyshire. The five fire services are all very different, because of the different funding mechanisms, but in terms of the technology, I think they are all trying to advance around the use of data.

Chair

So you invested in that equipment, and of course it will be obsolete once the new system is fully up and running. Is that right?

Peter Aykroyd

We are not sure on that—again, decisions we are waiting for.

Chair

So you are going to do double-running locally, on your own initiative.

Peter Aykroyd

Well, for example, with mobile data devices, we need assurance or clarity as to whether the device can connect to the network. Will it meet security and testing requirements, or is there a way it can be connected without going through those sorts of stringent controls? We have not got clarity on that.

Chair

This is the new kit that we were just talking about.

Peter Aykroyd

Our existing kit. Can we connect existing equipment? Can we tether it to a router-type system, or can it connect to the network? We still have not got clarity on that.

Chair

So you might have to write all that off, but you don’t know.

Peter Aykroyd

Potentially, but we don’t know.

Jo Beresford-Robinson

What we are being told is that the priority devices, which are the MDTs—mobile data terminals—will be able to tether to the ESN devices. One of the issues that we have is that it’s 4G. Within the fire and rescue services, they have 3G and 2G kit, so that kit would become obsolete. You therefore need some capital in your programme to upgrade that.

Chair

Have you done assessments in your organisations on the cost of that? Are you able to provide us with that information on the cost of the write-off?

Jo Beresford-Robinson

Certainly some of the TomTom-type devices would have to be replaced, but in the main, fire services are keeping up with technology. I do not think that there is an awful lot of legacy that they would have to move.

Chair

It would be helpful if you could both write to us or give us an indication, and we might seek that from other sources, too. I just wanted to turn to EE and Motorola. First, Mr Kennedy, we just heard about the devices that are supposed to be ready and available by the end of 2017, as we understand it. How are you seeing progress from Motorola on getting those devices out?

Vincent Kennedy

I guess it is important to understand our role in this particular programme. There are a number of lots or sub-contracts. The Home Office is acting as the overall prime contractor. The lot we are delivering does not include devices, so we are not responsible for the devices. I can certainly comment, because I know the business, but our lot is the user services: SIM card management, data centres, help desks and so on. I can comment on the devices. There are two types. The data devices, which I think colleagues here are describing, are relatively straightforward. Yes, they have to be 4G, but data can attach to the new emergency services network relatively easily. The voice devices are more complex. When you move the voice service on to a commercial mobile phone network, the device has to act in a specialised way. What I mean by that is that if I picked up my phone and dialled your number, it would take a few seconds to ring. It might take six seconds to you or five seconds to Mr Morse. In this world, when I am the commander of a team at a firearms incident and I press the button on the device and say, “Don’t fire”, you instantly all have to hear the same thing. It is a big problem if you hear, “Don’t fire”, and another person hears, “Fire.” That is just an example, but the voice piece makes the device more specialised, and it has to work in a special way. The mobile data terminals that are there—it is really a new SIM card or a new modem to attach to the network. Again, it is not part of our programme at all, and you can check with the Permanent Secretary, but there are plans to try to reuse as much as—

Mr Bacon

You are referring to the push-to-talk latency? I am reading an article from “The Register”, which is an online IT journal. It gives the example of a commander in the armed forces with a team of snipers giving the command, “Don’t shoot”. It goes on to say, “With 4G you might hope that this is fixed”, but that it isn’t that simple. Are you confident that the latency can be low enough with 4G that it won’t be a problem, and that it will be as near to instantaneous as to make no difference? It’s not at the moment, is it?

Vincent Kennedy

I guess this is the first project of its kind. The NAO has said that it is an ambitious first-of-a-kind project. The context is that we have been investing in the R and D of mobile LTE solutions for public safety since 2010. We have done a lot of design and R and D work in this space, the reason being that it is the trend for the world market. It is where the public safety market is going. They will eventually all be using mobile broadband, mobile data and voice. If it works to design, the latency can be solved, but that is why the design is so strict around the devices and the network.

Mr Bacon

And we are not there yet.

Vincent Kennedy

Well, we are building. We are designing and building—

Mr Bacon

So it is correct that we are not there yet.

Vincent Kennedy

We can prove the latency in a lab environment, but obviously you need to get into the field and do the testing.

Mr Bacon

You need to prove it at scale and in real conditions.

Vincent Kennedy

Correct, and under pressure. That is why the testing that Mr Aykroyd referred to is crucial. These are big technology projects, but they are not like regular IT projects. The people who use this technology—their lives depend on the technology working.

Mr Bacon

How long will it be before you prove it to yourselves?

Vincent Kennedy

We have a testing regime all through the spring and summer. We will hopefully be ready for operational trials in the autumn.

Mr Bacon

In the autumn of 2017.

Vincent Kennedy

Yes.

Mr Bacon

And how long will those operational trials go on for?

Vincent Kennedy

I think that discussion is ongoing between the Home Office and the user groups. It will be months; it won’t be weeks.

Mr Bacon

So when we hear that the testing timetable has been compressed, which is something we hear very frequently, we should be worried.

Vincent Kennedy

The chief fire officer, the chief police officer and the chief ambulance officer have a duty of care to their people. They won’t accept the service until they are happy with the service. This world is customer-led. You have to prove to these people that if their lives depend on it, it will work.

Chair

Mr Frumkin, do you have anything to add?

Simon Frumkin

I would endorse exactly the same thing. We take our responsibility extremely seriously, and we won’t compromise on quality through the testing, which is why we are working very closely with integrated testing.

Chair

The Comptroller and Auditor General has a question.

Sir Amyas Morse

Just a quick question: this is a world first for you, and it is a new product area, but am I right in thinking that your success is dependent on the quality of the systems integration being provided by the Department?

Vincent Kennedy

You are right that it is a first of a kind.

Sir Amyas Morse

I am just repeating back to you what you have said to me, more or less.

Vincent Kennedy

There are a number of subcontractors. There is ourselves at the centre, there is EE on the coverage, and there will be somebody doing London underground and air-to-ground. When we come together and plug everything together, the end-to-end testing has to prove the system. While we will prove it to ourselves, the most important proof is that the users actually then scientifically test it and check if it is working the way it should.

Sir Amyas Morse

Forgive me; what I am asking is this: say you have got a lot of different components and you are responsible for delivering your bit. The system-integrating activity is happening at the Department. Is that right?

Vincent Kennedy

I guess they are overseeing the integration. We’re doing part of the integration. Mr Frumkin’s organisation integrates to our system. We integrate to whoever does London underground and so on. Obviously, they are overseeing it, and I believe they have hired a consulting company for Lot 1 to do the assurance, witness the test and make sure it is checked correctly. Ultimately, the arbiter of success is the user groups that will eventually test it. Once they are happy it is safe, they will use it.

Chair

Can I ask Mr Frumkin about the network? This is new and cutting-edge—the civilian network and the emergency services network will be the same. What will the impact of the volume of users be on the network? The emergency services get priority, as we understand it, but have you tested that to the volume that it can be applied to?

Simon Frumkin

Yes. It is important to emphasise that we are on track, in terms of our delivery. Most importantly, we have completed the system testing, which proves that the prioritisation features in the software and the technology features that are mentioned in the NAO Report actually operate as expected. These are new features of the 4G technology, and we have proved that prioritisation works.

Chair

So you have really proved it works? So, emergency services will get priority, but that could mean that some of the civilian network goes down in an emergency? If there was a big incident, like the 7/7 incident, which is obviously very much in our minds, and the emergency services get priority, that would mean the civilian network would go down during an incident of that size, if necessary, to maintain service.

Simon Frumkin

At the extreme, we would be able to prioritise all of the commercial network for the emergency services. We think that is very unlikely. Our network is dimensioned for 30 million customers, compared with 300,000 emergency services users. The chance of us having to repurpose and reallocate all of the spectrum on any cell, or range of cells, is unlikely, but we could do that if we needed to.

Chair

That is very helpful. Sometimes the public can be very useful in an emergency, too. I mentioned 7/7. As a London MP with a constituency next to what happened at Aldgate, we were very concerned about the whole issue that the emergency services couldn’t communicate underground. Of course, there was a review, and that’s why Airwave came in in the first place. When Airwave goes, there are still discussions going on about use underground. Where has that got? Are Londoners and others safe if an emergency happens underground?

Simon Frumkin

The Home Office is in discussions with TfL. We have been passed those discussions, in terms of potentially supporting the deployment of LTE technology within the underground. Those discussions are still ongoing between the Home Office and London Underground.

Chair

They are still ongoing, yet the roll-out is going to begin in a couple of years’ time—or less than two years’ time. It’s easy to say here that conversations are ongoing, but are you optimistic that those conversations will lead to a good solution for using this new system underground?

Simon Frumkin

Yes. As I have said, the main conversations are between the Home Office and TfL, so it may be best for them to comment. I know that they are looking at a number of different contingencies, and options to make sure that there is coverage one way or another on the underground, when the underground transitions.

Chair

So what would you say you need? We are going to be talking to the Home Office next. From your point of view at EE, what do you need to make sure that there is secure coverage on the Underground in the event of an emergency?

Simon Frumkin

Either we would need time to deploy LTE on the Underground, or the Home Office would need to agree an alternative solution with TfL.

Chair

Okay. We talked just now about the volume of emergency service users—about 300,000—compared with 30 million other users. Having been a customer of EE in the past I know there are times when it can be difficult to get a signal, and colleagues say that at football matches it can be difficult at half-time. Do you ever have outages at busy times, such as new year or coffee breaks in the Olympics?

Simon Frumkin

We are constantly upgrading our network and dimensioning it for the increased data use—obviously data usage is increasing exponentially. We are always in the process of putting more and more capacity into our network, both over—

Chair

Do you ever have outages? You must have some sometimes.

Simon Frumkin

We sometimes have outages, but we have a very targeted approach to tackling hotspots and sites that get into congestion, making sure that we upgrade them to stay ahead of the demand.

Chair

How are you making sure that you would be able to cope with the demand if there was an outage at the same time as there was an emergency?

Simon Frumkin

When we talk about outages, I am talking about a site going to congestion rather an outage. What would tend to happen is that if there are too many users or too much usage on a particular cell, the performance slows down. That is where the prioritisation features of the network come in, which mean that we can allocate the spectrum and dedicate the bandwidth for the emergency services, so that the performance does not degrade.

Chair

Just to be absolutely clear, that happens instantaneously, does it?

Simon Frumkin

We can do that dynamically.

Chair

So it is an automated system. It is not reliant on someone noticing and pressing a button.

Simon Frumkin

That is right, it is an automated system.

Chair

I was going to ask you something, Mr Kennedy, but tell me if it is outside your remit. You—Motorola—bought Airwave and there have been negotiations about what would happen if we went on beyond the end of 2019. Are you aware of any offers made by Airwave before you bought Airwave to continue running so that this programme could be brought in at a slower pace?

Vincent Kennedy

I am not aware of previous offers that the previous owners made. From reading the NAO Report, I can see there were some discussions around continuing the service and the pricing of the service, so I know as much as is in the Report.

Chair

There has been some discussion with you now, as the owners of Airwave, about what would happen. Will you tell us in your own words what would happen if the Home Office, or indeed the end users, felt that the 2019 cut-off was just too radical, too steep? What would your position be, and how would you make sure that our emergency services were supported?

Vincent Kennedy

Yes, we obviously had quite detailed discussions. In the process known as ‘change of control’. So Airwave is designated as critical national infrastructure in the legislation and it is a PFI company, so within the contracts you could not buy Airwave without the Home Secretary’s approval. Once you approach the Home Secretary to look for approval, clearly you get into a detailed discussion about what you are going to do with the asset, the ownership, the security arrangements and all of that. Our interest in it was really to have control of it, so that we could migrate the users more safely, keep the parallel running going and build what I would call the inter-op. Today, you cannot talk between the EE network and Airwave, and we are having to build a bridge so that you can actually talk across, because parallel running is no good if you cannot talk to the people who are on board it—if you are on one network and I am on the other. We are doing that, but we have also agreed that if necessary—the way that the regional migration happens is that it happens in a sequence. There are 12 regions, they start in 2017—

Chair

Yes. I think figure 12 in the pack illustrates this very effectively.

Vincent Kennedy

One of the things that could happen is that the last two regions, if there was an incident or something happened, could end up going a month or two into 2020. In parallel, we have to design and set in motion taking down the network in 2020, because the PFI contracts all run out on a single day at the end of 2019. So one of the commitments that we gave was that, if needed, in an unplanned or emergency situation, we would keep the last regions going until they were on the EE network. Really, you have to, because the emergency services cannot be—

Chair

You were saying that you would do that free and gratis.

Vincent Kennedy

No. We—

Chair

What is it going to cost the taxpayer?

Vincent Kennedy

We agreed a fixed price per region per month, but the difficulty—

Chair

Per region per month? So you could do it literally month by month.

Vincent Kennedy

To an extent. The difficulty I have is that it is quite a big operation—4,000 leases, 16 data centres and 500 people—so I have to start to set in train the take-down planning and the exit of that long before; that does not start on the 1 January 2020. It is possible, at least, to sequence the dismantling in a way that leaves a couple of regions going over a few months, but it needs to be at that order of magnitude. An unplanned, long extension is going to be very difficult—I would not say impossible—because it is an unplanned thing and you have landlords who want their sites back and you have to give notice well in advance. It is possible to do. It should not be necessary, but with public safety you always have to plan. You cannot leave the Met police without the network, so if they are the last region and they need two months in 2020, you have got to provide the service.

Chair

Are you planning on that? Have you got a plan for doing that if necessary, or are you waiting for the Home Office to make a decision?

Vincent Kennedy

We have a plan, and I think that the NAO Report picked up on the contingency plan. I know how I can terminate the network but still keep sub-regions running. We have worked out technically how. The network is eight separate pieces, so once we know the regions, we can work it out before we start the dismantling project. We know how we can do it, but what we do not yet know is whether it will happen and which regions it is. I think that one of the recommendations in the NAO Report was around just fleshing out the contingency plan. We have three years to do it, and we have an exit board with the Department of Health, the Home Office and the fire services, so there is an exit committee and we are getting into the detailed planning of the “what if” scenario—what if there was a terrorist event at the end of 2019 and it meant that you could not do anything for three months? We have to be ready for all those eventualities in this world.

Chair

We understand it would cost £475 million a year, if it were to be extended for a whole year and beyond.

Vincent Kennedy

That is the figure in the NAO Report. It is quite a derived figure because it is not really possible to go and extend everything for a year. In an unplanned scenario, you have to be in the middle of taking down the network. If that was the case, you would have to plan much—two years in advance you would have to say, “Actually, I need it for another year,” because if I give notice to all my supply chain, I cannot unwind that once I find out there is an issue.

Louise Bladen

I just wanted to clarify that the £475 million in the Report is a 12 month—a year’s—extension for all regions. We did appreciate that you could, as you say, do it monthly and do it by region, but we wanted to give that figure so that there is a total of what the maximum might be in the event of that happening.

Mr Bacon

What is the latest date by which one would have to give notice of the intention to extend, given what you were saying a moment ago?

Vincent Kennedy

The date we have agreed is a minimum of 12 months’ notice—so, December 2018.

Mr Bacon

By the end of December 2018?

Vincent Kennedy

Yes, by the end of December 2018 I would need to know because I have a big supply chain behind me that I have to give annual notice to.

Mr Bacon

You have a written agreement that by 31 December 2018, if there is to be an extension, you will be told?

Vincent Kennedy

Yes, I need notice a year in advance.

Mr Bacon

I am sorry. I appreciate that you need notice—you have told us that already—but my question is, have you got a written agreement that states that you will be given notice by 31 December 2018 if it turns out that there is a need to extend?

Vincent Kennedy

Yes, that was part of our agreement with the Home Office.

Chair

You talked about the dual running. This will require different kit that means that you can switch from Airwave to the ESN network, will it not? How far in advance is that?

Vincent Kennedy

Actually, from the user’s perspective it will be seamless. If I am on a tetra-radio and I push to talk and you are on the ESN network, your ESN device will hear me. We are doing it in the background at the switching level. We are connecting our Airwave switch to the EE switching network, so the radios they have today and the LTE devices they buy will be able to talk to each other without them having to buy any more kit.

Chair

What happens to the Airwave infrastructure at the end of the contract? You said you are winding down and giving sites back to landlords. So is that it—it’s obsolete?

Vincent Kennedy

That’s it. It is a 20 year old system; it has not been refreshed. It is the end of the PFI and the base stations and switches are 20 years old, so they are just about lasting until then.

Charlie Elphicke

A couple of questions. Mr Kennedy, you say this is an ambitious, first of a kind project. It is “technically cutting edge” and unproven and “has not yet been used, nationwide, anywhere in the world.” As things stand, are you on track for the transitions across the country on page 35 of this Report?

Vincent Kennedy

Again, with the part we are doing, the lot 2 part—the user services, the data centres, the push-to-talk technology—yes, we are on track. We have gone through the design phase, we are in the build phase and this month we have a major deliverable and will move into the test phase in the spring. So yes, we are on track.

Charlie Elphicke

If it falls behind, who takes the risk?

Vincent Kennedy

I guess the programme risk sits with the prime contractor, the Home Office. We have our own risk to manage. If we fall behind, we obviously manage our own contract risk in that respect.

Charlie Elphicke

Mr Frumkin, obviously your plan is to use these base stations, which give you superb coverage under this contract, for your commercial operations as well. Is that correct?

Simon Frumkin

That’s right.

Charlie Elphicke

That being the case, will other operators like Vodafone, Three and all the rest of them have equal access?

Simon Frumkin

Yes, they will. A state aid decision states that we are not getting a material benefit from the contract. Just over 250 of the sites we are building are being funded by the Home Office. The rest of the network investment and the coverage enlargement are being funded by us. For the specific sites the Home Office is funding we have committed to give access other operators.

Charlie Elphicke

In that case, can you explain why all the other operators are writing to us and indicating that what you said does not seem to match their current understanding?

Simon Frumkin

In fact, we met Vodafone yesterday and had a constructive meeting. There is a standard way of sharing infrastructure and we are talking to them about giving them even earlier notice than they normally get in terms of the infrastructure we are proposing to build. We will be giving them the right to share that infrastructure if they wish. Obviously, delivering the ESN coverage has to be our priority, so we will give them access as long as it doesn’t slow down our coverage roll-out of ESN.

Chair

We understand it costs about five times as much to retrofit technology to an existing site. Is that under your consideration? I appreciate you have to deliver the network, but are you considering that when you are planning and talking to other telecoms companies?

Simon Frumkin

Yes. I am not sure we agree with the figure for that cost. Clearly it depends on the structure we build and the circumstance, and who wants to come on the configuration but, as I said, we will give access—

Chair

I think that figure came from one of your competitors, so I am not surprised there might be disagreement about it.

Simon Frumkin

As I said, there is a standard process during which we notify one another about the infrastructure we are building. We are proposing to give even earlier notice than we normally do under the standard process and, as long as it doesn’t slow us down in the ESN coverage, we propose to give them access.

Philip Boswell

I have a question about coverage with a focus on Scotland and the rural UK. The last 5% to 10% of any major project is always the hardest and often the most expensive to complete. Your target for 97% coverage across the UK is committed to be reached in 2017. How much of that 3% is in Scotland?

Simon Frumkin

How much of the last 3% of coverage is in Scotland? Unfortunately, that is a calculation I cannot do here and now, but I am happy to provide you with a written answer.

Philip Boswell

Are you keen to hit 100%, and when?

Simon Frumkin

The requirements we have for the emergency service are the same as Airwave’s current threshold and that is based on road coverage. The formula is based on road coverage and that is the coverage we are rolling out.

Philip Boswell

Okay. My second question goes back to figure 12 on page 35, which shows large swathes of rural UK—Wales, Scotland and the south-west—being last to transition. This is a substantial part of the UK. The contingency illustrated in figure 12 is for transition only. Following on from Mr Elphicke’s earlier question, if you are not ready to shut down Airwave system in 2019, who pays the contingency cost specifically of parallel running beyond 2019 under your contract?

Vincent Kennedy

As I said, if the last region for some reason needed to go to February 2020, I believe the Home Office—you will have to double-check this with the Permanent Secretary—funds the region and I am not sure how it would work that out with the region.

Chair

That means the taxpayer.

Kevin Foster

That seems apt, given that the last region is the south-west, where I am one of the MPs. Looking at figure 12 on page 35, every area has a contingency period and a period of parallel running indicated except the south-west. How confident are we that that transition plan is realistic and accurate?

Vincent Kennedy

I guess it is a good question for the Permanent Secretary. [Laughter.]

Simon Frumkin

I would say from a coverage perspective that the south-west, like the rest of the country, is a region we feel confident and on track to deliver coverage for.

Kevin Foster

You say it is like the rest of the country, but you have only got to look quickly at the diagram to see the rest of the country has a period of contingency and parallel running. Why not the same for the south-west?

Simon Frumkin

I am referring to the coverage roll-out rather than the transition activity.

Kevin Foster

That is a rather vital point, though, wouldn’t you say?

Simon Frumkin

Well, clearly coverage is an area that it is critical that the emergency services get confidence in. We will be able to test the coverage in advance of the transition to give those in the transition greater confidence about the coverage.

Kevin Foster

Just to be clear, in every region except the south-west you have a contingency period and a period of parallel running planned for in the document. Do you think that is a secure position to be in, or do you think we might end up with the position that we have just had outlined that it might end up overrunning for a month or two in that area?

Simon Frumkin

As I said, from a coverage perspective we can be confident in terms of the actual transition and contingency. That may be a question to direct to the Home Office.

Chair

Finally, can I ask you two questions? It is a quick-fire round. What is your percentage of confidence that this will roll out in time? And what keeps you awake at night about it?

Jo Beresford-Robinson

If you are looking at the current timeframe, my confidence is quite low.

Chair

Below 50%?

Jo Beresford-Robinson

Yes, lower than 50% that it will be achieved within the current timeframe for the early adopters of the ESN. I think for the later transition regions I am probably more confident, but seeing as I am in one of the early ones I am less confident.

Chair

So what is keeping you awake at night?

Jo Beresford-Robinson

What is keeping me awake at night is the assurance that this technology is fit for purpose and that public safety and the officers’ safety is actually going to be there.

Peter Aykroyd

I agree with Jo: I would certainly be under 50% on that, probably down to about 30% on whether we will meet the existing timeframes. The thing that gives me the most concern again is being able to fully end-to-end test against that full 97% of network within the timeframe to give that reassurance and confidence to fellow officers and members of the public and the fire authority.

Vincent Kennedy

For the part we are responsible for—data centres to centralised services—I am 100% confident we are delivering. We are a year in and we are still on track. We have rolled out many public safety networks in many countries and the thing that always keeps you awake is the coverage testing and verification and the certification. That is the piece that you have got to allocate enough time to.

Simon Frumkin

I would agree. We have signed up to challenging coverage targets and that is what keeps me awake. I am extremely busy at the moment, as you can imagine. To give you the scale of it, we are upgrading between 100 and 150 sites to 4G every single week, so there is an enormous amount of activity going on on the network. We have moved from 70% to 74% coverage since the NAO Report was written, but we have an awful lot to go—

Chair

So how confident are you that you can actually deliver the 97% coverage?

Simon Frumkin

I am 100% confident we can deliver the coverage.

Mr Bacon

By when?

Simon Frumkin

Well, I am very confident we can deliver the coverage by the time we transition. We need to prove to the emergency services that that coverage is there for them to transition.

Mr Bacon

You were 100% confident and now you are very confident—there is already a difference there. The question several of my colleagues and I asked was: by when?

Simon Frumkin

I am very confident that we can deliver the coverage by next September, which is when we are required to deliver the coverage.

Mr Bacon

September 2017?

Simon Frumkin

Yes.

Chair

Most of our other questions are now for the Permanent Secretary and other colleagues from the Home Office. Can I thank you very much indeed for your candour and your time? The uncorrected transcript of this hearing will be available on the website in a couple of days and we will send you copies. If you have corrections, please let us know. The Report may not be out before Christmas, which is five weeks away, but it may be available early in the new year. We may call you back at some point because this is such a vital programme for the security of our emergency services. Thank you very much for your time. Examination of witnesses

Witnesses

Mark Sedwill, Stephen Webb, David Williams and Liam Maxwell.

Chair

This is our second panel looking at the upgrade of emergency communications, the Emergency Services Network, which will replace Airwave. I welcome our new witnesses, who are David Williams, Director General of Finance and Group Operations at the Department of Health—you are a frequent flyer, Mr Williams, so welcome back—Mr Mark Sedwill, Permanent Secretary at the Home Office, who is the overall responsible person for this programme; Stephen Webb from the Home Office, who is the senior responsible officer for the programme, so it is a big job for you, Mr Webb; and Liam Maxwell, the national technology adviser, who I suppose is the technological expert on the programme. We know that, unusually for NAO Reports, there was a strongly worded paragraph on page 13, and you did not accept the NAO’s judgment about the underrating of the risks. We have just had a pre-panel, some of which Mr Maxwell heard and might be worth recapping. Two witnesses at the user end ranked their confidence that this would deliver on time: one said below 50% and one was explicit in saying as low as 30%. On the technology side, there is a big challenge in getting EE’s coverage up from its current 77%—it said 74% as of today—to 97% by next September, which is less than a year, and developing all the devices that are not out there yet. That is also a concern. There is also a challenge in making sure all the users are ready—we have heard users today expressing scepticism—and a very ambitious timetable, which was one of our main concerns when preparing for this hearing. As well, if that timetable is delayed, there will be a huge potential cost to the taxpayer, which the NAO has quantified as £470 million if there is a full year, full UK-wide on-running, although I am sure you can come up with figures month by month, region by region. Nevertheless, that is the potential risk to the taxpayer, so we want to focus on it. This Committee looks at many programmes, so you will forgive us if we are erring on the side of scepticism at this point. We want to hear from you some assurances on how you are ensuring the project will deliver and that our emergency services will be safe. We are pleased, Mr Webb, that you have been the SRO for this programme pretty much from the beginning, which is almost unique, I think, Mr Bacon, in the history of the civil service. We welcome the fact, Mr Webb, that you have deep knowledge of this and we will be using that and probing you closely.

Nigel Mills

Mr Webb, could you tell us how you are feeling about this project. Are you confident it is on track?

Stephen Webb

I am very confident that we have a programme that is deliverable. We have some suppliers working well and working well together. We are currently working on some potential revised plans for the timescale, but overall I am very confident that we will get it in very close to the timescale that we previously agreed at the final business case. There will be a little slippage, I accept that. We accept the NAO Report’s finding—this was an ambitious timescale—but our argument has always been that it is best to go for an ambitious timescale because we have that contingency to allow us to buy as much extension as we need. We are going as fast as we can. If it slips, we know how much it will cost, but we want to minimise that as much as possible—partly for the savings, but largely because of the enormous operational benefits that this system promises.

Nigel Mills

What do you think are the big risks that might cause further slippage beyond what you now expect?

Stephen Webb

With all the individual aspects of the programme, I think we have got a series of deliverable plans. There is obviously a lot of them and they need to come together, so any delay in one can potentially cause knock-on delays in the others. For me, it is less the technical risk than the sheer logistics of it when we talk about coverage and some of the other areas—the Underground, and so on. We have a very strong team managing them. We are doing the best we can, but there can be factors outside your control that could then potentially have knock-on effects on the timetable.

Nigel Mills

When I first looked at this Report I thought that we are nowhere near having the coverage and we do not have the devices, so this looks like a pretty racy thing to take on and be the first in the world to have a go at. Are you still confident that you can achieve this?

Stephen Webb

I think that the thing about coverage that it is worth reminding ourselves is that the overwhelming majority of the coverage will be delivered by upgrading masts that are already there. It is fundamentally different from the original Airwave system, where they had to build it from nothing to 90-something percent coverage, as described. Here, we will get to 93% geographical coverage with upgrading the series of masts that we already have, whether those are the ones on the EE network or, potentially, some of the Airwave ones in the very remote areas. Although there is a lot of work to be done on some of those existing masts, they are really to bridge a very small gap at the end to meet that target of roads coverage. There is a lot to do, but it is not quite as daunting as it may seem. You heard from me earlier about the actual pace of the roll-out; they have got from nought to 74% in 4G in quite a quick time.

Nigel Mills

We hear quite a lot from BT/EE that they can deliver broadband and 4G coverage across the whole country, and yet our feeling is that they do not quite get there as fast as we would like or quite when they promise. You are moving away from a network with 97% coverage and 99% availability; it is quite a long shot for you to have this ready in a couple of years’ time.

Stephen Webb

I think that within a very short period of time we will be extremely close. We are down to those last numbers of sites that need to be built—roughly 500 by roughly 200 in our extended area of services—but that really is for a tiny proportion. The vast majority of this will be dealt with. Similarly, on resilience, we know what needs to be done and we know how many sites will need to be hardened. It is fundamentally a question of logistics. The other thing that is quite helpful about 4G compared to Tetra is that, because it is the international system that everyone is using, it is a lot more flexible. If there are coverage gaps, there are potential mitigations. We could potentially strengthen the signal in masts around; we could potentially bring temporary coverage in. We have a choice. We have a discussion with the users about what levels of coverage they are prepared to accept and, equally, what sorts of temporary solutions they might be prepared to accept.

Nigel Mills

You just said that 4G is a system that everyone is using, but actually emergency services around the world are not using 4G. In fact, they are using the one that you are trying to move away from.

Chair

South Korea—

Stephen Webb

They are very impressive in this area. I mean that the fundamental technologies—we are talking about temporary coverage and temporary masts—are being used by all the MNOs, all the mobile network operators, throughout the world. We understand exactly how much they would cost and where to source them from. It is very different from Tetra, which is a system that really nobody used other than the emergency services, which meant that internationally it was a relatively small market and it did not have that flexibility.

Nigel Mills

So you didn’t think, when you were first asked to be the SRO of this, “Why on earth have we gone down this risky, aggressive route? Why do we have to be first? Can’t we stick with what we’ve got, which we’ve paid for and know works, rather than take this on?”

Stephen Webb

I made the decision myself, rather than being asked to. The choices were all difficult. They were all high-risk options. There were different sorts of risks—legal risks, financial risks, operational risks and technology risks. If you look at the number of risks we have really successfully mitigated in the three or four years—I don’t want to give the impression that I am complacent about the level of risk in the programme now, but if you look at the risks we have moved away, they are very striking. Back then, we were not sure how successful the 4G auctions would be and how much uptake there would be. If it had been as unsuccessful or had as limited coverage as 3G, the solution we are following would not really have been feasible. We did not know that the international standards body were going to prioritise emergency services’ functionality in the way that we helped persuade them to do. We did not know we would be able to go in with £1 billion of investment over the spending review period, and we didn’t know how much appetite there would be in the procurement. We always had in the back of our mind the expectation that we might face a legal challenge from the incumbent. That might set in context how we have brought the risk down really quite significantly. We still have an enormous task to do. As I say, I see it largely as logistics. I think it can all be done. It’s now a question of timing and getting the order right.

Nigel Mills

Mr Sedwill, what were the main reasons why you chose to move away from Airwave, which had been paid for and worked, and try something new?

Mark Sedwill

It had been paid for up until 2020, and then procurement rules would have required us to re-procure Airwave. Fundamentally, this goes to the strategic solution, which is well described in the Report. If you look at emergency services generally, interoperable data are becoming absolutely central to the way they are going to need to operate over the next five to 10 years. We want them to be able to be away from the base station more often and to have data as they are going to respond to a crisis—for example, for a fire engine that is going out to get a data feed en route, to know whether there is someone in those premises with mobility problems, and for the police to have a much richer picture of the people they are dealing with. Data generally are going to be absolutely central to the emergency services and, indeed, other business models over the next five to 10 years, so we need a capability to provide them with data—and data that are interoperable, because they are going to have to work much more effectively together.

Chair

There seems to be common agreement across the piece that it is the right strategic move. Our concern, and Mr Mills’s questioning, is about the timescale and whether it is achievable.

Mark Sedwill

Well, Mr Mills was asking about this choice versus the existing risk. There are other reasons. In government, it is very rare—as you know from your experience, Madam Chair—that there is a single reason for going for anything. We looked at commercial angles, financial angles and so on. Certainly from my perspective, the fundamental reason for supporting this programme is that it is going to give them the capabilities they will need in the next five to 10 years. That is the strategic answer.

Nigel Mills

And you weren’t tempted by what some other places have done—creating data functionality on a separate network and keeping a separate voice network?

Mark Sedwill

That is one of the options set out in the Report, in a table that sets out the various different options. Mr Webb can perhaps answer this in a bit more detail. Of course, it would have been possible to run systems for a longer term in parallel. Other countries have gone for an entirely separate ESN network based on 4G-type technology, but separate from the commercial networks. That would have required quite a lot of building from scratch of the kind that we had before and would have been a lot more expensive, and there would have been other risks associated with that. I would not want to suggest that this is not a high-risk programme. Actually, we entirely agree with the NAO that this is a high-risk programme and have sought, as Mr Webb said, to mitigate that. We had to judge the blend of cost, capability and risk across all those options, and this came out as the right solution.

Nigel Mills

I suppose we need to come back on to what happens now. Your timetables are going to slip. You are world-leading—and on a compressed timetable that you seem to keep compressing. The testing phase and the parallel running phases seem to be being squeezed, rather than extended. Are you getting a little nervous that that is just too aggressive?

Mark Sedwill

It is aggressive, Mr Mills. Mr Webb touched on this a moment ago and may want to add something. Essentially, the risk is a financial risk rather than an operational one. If the roll-out timetable were to slip to the right, we would run Airwave for longer and we would be in parallel for longer. We probably wouldn’t be countrywide. As the Chair said at the beginning, there are significant costs but the ones identified are essentially a full-year cost if the entire programme slipped everywhere for a year, and that is not what we would expect. It is essentially a financial risk.

Chair

Do you have a range of financial risk that you are looking at?

Mark Sedwill

It would depend. For example, let’s say that the last region slipped past the end—then we would be talking tens rather than hundreds of millions, I think. Obviously, if there was a very substantial slippage in a big part of the country, we would be looking at something substantial and in the worst case the NAO figure.

Chair

Do you have a tolerance written into your future budget to cover the risks?

Mark Sedwill

Of course we don’t yet have a settlement agreed post-2020, for obvious reasons.

Chair

But you will be bidding? Are you factoring this into your financial planning?

Mark Sedwill

Of course. This is funded overall from within the overall police settlement for us, and similarly for other Departments. Of course we will need to build some of that in as we look at the post-2020 settlement.

Chair

I just want to bring in the Comptroller and Auditor General.

Sir Amyas Morse

You were saying that the risk is essentially financial risk. I just want to test that based on the earlier testimony. If it is a given that nothing will be rolled out in any of the emergency services unless the chiefs are happy, which is what we have been told just now, and they don’t think the functionality is what they want—they may have a functionality but they don’t think it is fit for purpose in their judgment—then you are kind of stuck, aren’t you? Is that merely financial risk or is it not a functionality risk? If they are doing their testing, they think, “This just doesn’t get to where we want it to get to,” even though you may think it does. If they don’t agree, that is their decision, isn’t it? Have I got that right?

Mark Sedwill

Mr Webb may well add something. I was actually answering a question about the risk of delay not the overall risks in the programme. You are absolutely right, Auditor General, there are commercial, technical, functionality and financial risks in the programme.

Sir Amyas Morse

Am I right about this particular risk?

Chair

Who calls the shots? That is really what the Comptroller and Auditor General is asking. Is it the user or is it the Home Office? The cost comes to the taxpayer whichever one it is.

Mark Sedwill

It does and of course it comes from the same budgets that fund the rest of policing, so there are strong incentives for all of us to work collaboratively and to agree this, but of course we are only going to proceed if we have, essentially, a critical mass of people who believe it delivers the functionality.

Sir Amyas Morse

Sorry, just to be clear then. I am not trying to put an extreme argument. I am putting an argument where reasonable people can disagree as to what is an acceptable functionality. You have got all these different bodies bringing together the service and then you have the chiefs, who are independent, and who are going to look at it and say, “I am comparing this to what I have got now. It is either at least as good in my judgment or not.” If they say not, they have the say. That is right, isn’t it?

Stephen Webb

Absolutely. We are not going to move from trials into roll-out unless they are comfortable and we are not ultimately going to turn Airwave off unless they are comfortable. It will be a collective decision. We will look at it. Everything we have seen suggests that this will be vastly superior—that it meets all the contractual requirements of Airwave on coverage, resilience and so on but is inherently a better system. It has the priority and the pre-emption; recent tests are suggesting the data will be both cheaper and twice as fast. In-building coverage will be far better than it was in Tetra devices. We are very, very confident that they will be pleased with it but equally—

Sir Amyas Morse

If they are not, that is their say.

Stephen Webb

If they are not, we have to have a collective discussion with them, as the Permanent Secretary said.

Mr Bacon

Can we be clear about what having a collective discussion with them means? The point that the CAG raises is fundamental and is also one that we have seen before, in the national programme for IT in the health service. The Department of Health and Connecting for Health foisted upon hospitals immature technology that wasn’t fit for purpose, while putting out lots of press releases saying it was fine. It wasn’t fine. I can see something similar happening here if we are not very careful. If it is not working properly, it is not simply a case of having what you call a collective discussion. It is whether the chiefs can say, “No, we won’t have it,” and that is final, until they are satisfied. That is the question.

Mark Sedwill

The answer to that, Mr Bacon, is yes—but it is collective. We have chief constables represented on the programme board. The NPCC, the chiefs’ council, has effective mechanisms, now with the police and crime commissioners, for making collective decisions. It is really about individual versus collective responsibility within policing and, indeed, the other emergency services. Of course, we are not going to proceed unless we have a consensus among police chiefs, and also other emergency services, on the reasons you set out—that this is providing the right level of service.

Chair

Can we just pin this down, once and for all? Mr Williams may want to come in as well. We’ve got an ambulance service, for argument’s sake in Mr Bacon’s neck of the woods, saying, “We don’t think we’re content with this.” Would they be able to call the shots and say, “We’re going to wait until we turn off Airwave”? In the end, that is your decision at the Home Office, isn’t it?

Mark Sedwill

It is. We’d need to resolve that at the time, and it would depend on the circumstances at the time. If some were behaving completely unreasonably, we would need to work that through with them, but that would obviously depend—

Chair

One would hope that our emergency services would not just behave unreasonably.

Mark Sedwill

Exactly. That is exactly the point. I think we are at risk of heading into a cul-de-sac here.

Mr Bacon

Let’s just explore it a little further. One person’s definition of behaving unreasonably is another person’s description of safeguarding the public. If it’s clear to an ambulance service in a particular region of the UK that what they’ve got is less good than what they have now, provides less coverage than they have now and is functionally less secure than what they have now, one view would be that they could reasonably say they didn’t want it until met the minimum standards. I hope you would have a very reasonable discussion with them. The issue is that we’ve seen this many times before in different domains.

Mark Sedwill

Mr Bacon, just to be clear—my point was that it would depend on the circumstances. If in those circumstances they said, “Here are the reasons this isn’t working for us”, that is, of course, completely reasonable. We would then seek to ensure the programme adapted to deliver the level of capability that they required. I think we are all basically talking about the same thing. There has to be a collective view of this; they have a responsibility and we have a responsibility.

Chair

At the moment, the general election is scheduled for May 2020, but who can tell; we have a Fixed-term Parliament Act. This is due to be rolled out just prior to a general election. You are a Permanent Secretary. If you are still at the Home Office at that point, there will be enormous pressure on you to make sure this gets delivered, let’s be honest.

Mark Sedwill

There will enormous pressure on me anyway to make sure that this gets delivered, absolutely rightly. Of course, there will also be enormous pressure on me to make sure it gets delivered without a whole load of criticism from different trusts’ chiefs—we are talking about 400-plus organisations around the country—and without noise from them saying they aren’t satisfied with the service. The pressures on us are to deliver it, to deliver it to the specifications that have been collectively agreed and to ensure it meets everybody’s needs. We absolutely expect that. There is no risk. One risk I can assure you we don’t have, Mr Bacon, is of our foisting this on everybody and putting out press releases saying it is fine if it isn’t. We at the Home Office take emergency services capability very seriously, for all the reasons you know.

Mr Bacon

Mr Williams, did you want to say something?

David Williams

Only to echo that point. That judgment at the local level by trust chief executives, from a health perspective, is against a context of user engagement in the specification, clear contractual standards and a programme of work to agree the test and assurance strategy. If the system doesn’t meet those standards, it is a perfect reasonably judgment for chief executives to take locally that the system is not ready. If it meets all of that and an individual chief executive has a challenge or a worry about transition, I think that would probably be a different conversation.

Mr Bacon

Mr Webb, you mentioned earlier that the business of getting from 74% to 97% was mainly a logistical one. I don’t want to misrepresent you, but you made it sound as if the problems were primarily logistical—“there’s a huge amount to do and it’s got to be done in the right order”—rather than that it is enormously technically complicated, yet the NAO Report tells us this is the most advanced system in the world and hasn’t been tried anywhere else in the world before in this way. Are you understating the other risks?

Stephen Webb

No, I think they are two different areas. I do not think coverage is technical issue; it is building 4G masts, which we have known how to do for a long time. That is just a question of getting it through the planning, the design and the build.

Chair

Just getting it through the planning.

Stephen Webb

Probably should not put it quite that way. There has been some very successful engagement and a lot of work. For example, the team was meeting all the heads of planning in Scotland last week, had a good meeting at the Scottish Parliament and with the Welsh Government, and has an MOU with the national parks in England. So a huge amount of work has already gone on in this area, which is challenging, but fundamentally logistical. The point of the programme that the NAO Report was making—about the technical—was more about the functionality, which is cutting edge and new, but again we are extremely confident that this is on its way to being delivered. To take push-to-talk functionality, I was using it last week—the early Motorola—and it is not the absolutely final version, but it is there and is working at the moment. Similarly, with priority and pre-emption, we have seen it tested at the facilities down in Bristol, so we know that works. Devices—I have a couple of devices, which are prototypes that are going out for upgrading to release 10 for testing next week. Again, we are very confident that we will have a good selection and strong competition when we go there next year. So this is all new, but it is progressing.

Mr Bacon

I have two more questions. In terms of the simpler bit, the masts, the bit that you said was technically less risky and challenging, do you have a plan in which you know what is going to be done where and in which month, and that gets you from where you are now, at 74%, to 97% by the date you need? Month by month, for the next 12 months, two years, or however long it is.

Stephen Webb

Yes. Some of the plans are at earlier stages than others. With some of the sites we are in early discussions with planning, so the plan may need to change.

Mr Bacon

Do you have a spreadsheet in which you have set out that in April 2017 you will have expected to have done X, Y and Z, and in August 2017 you will have expected to have done P, Q and R? Do you have something like that?

Stephen Webb

Yes.

Mr Bacon

Can you send it to us?

Stephen Webb

Yes, by all means. There is EE—most of them—and then a certain amount that we are doing through EAS. So EE has most.

Mr Bacon

I fully accept that, even if you assert that you are going to achieve something in August or July, planning reasons may change that slightly, but we would still like to see it. It would be helpful for the public. As far as the more technically complex stuff is concerned—

Chair

Mr Bacon, may we come back to this later? I have just realised that Mr Mills is back in, and he was mid-stream.

Nigel Mills

When I left off, Mr Sedwill, you were saying that it was a financial risk to keep Airwave going. How practical is it? I assume that they will be running it down, people will be wanting to leave and they will not want to be working for a place that will shut at the end of 2019, so it is probably not as simple as ringing up in November 2019 to say, “Would you mind keeping half the country going for six months because we’re not quite on track?” How do you cope with that?

Mark Sedwill

It may be touched on in the Report, but of course we will know earlier than that. We will learn a great deal from the early roll-outs, and that will tell us whether there is risk at the back end with the later roll-outs—whether it will take longer, or there are unforeseen technical challenges, and so on—so we should know, I would think, perhaps a couple of years ahead of that 2020 contract end whether or not we will have to invoke the requirement to extend Airwave. It would not be a phone call in November 2019, I guess is what I am saying.

Stephen Webb

You need to give them a year’s notice, very understandably. So we will need to decide in mid-’18 which regions we will need to extend and for how long, because you are absolutely right, that will be the challenge facing Motorola. But I think that gives them enough time to prepare.

Chair

Are you expecting to extend regions? From the way you just spoke, it sounded as if you might be.

Stephen Webb

We are likely to slip if—it comes under the point about compression of transition. If the users, very understandably, do not want to compress it, then there would need to be a certain amount of limited extension of Airwave. Equally, it is possible that they might be able to absorb it, not least because the revised plan that will be taken to users will involve us doing some more things up front, which otherwise would have been done during the transition period. But we are not going to put pressure on them. They have to be comfortable that the plan is viable and that they will be comfortable turning off Airwave at the end.

Charlie Elphicke

My sense, from everything I have heard, is that this is a programme that is going to end up falling behind, and there is a risk, if it does fall behind, of a 12-month extension of the Airwave contract, which is—the maximum at risk—£475 million. When I asked Motorola about this earlier—who take the risk—they were basically saying that it was the Home Office that takes the risk. My concern is that delivery seems to be partly in the control of Motorola—so if everything gets delivered on time then it is all fine and Motorola did well under contract, but if it does not, then Motorola do well under the extension. To what extent have Motorola got skin in the game to deliver this project?

Stephen Webb

It is a very fair question. I think they do. Obviously, there are clear contractual provisions around the existing contract. They are not going to be paid until that has been delivered, and they are putting quite a lot of investment in—there is a large team working very hard—so it would be quite painful for them, obviously, not to deliver on the current contract. As to an extension period, that could potentially be quite painful for them as well. As Mr Mills described, they may need to do some capital investment; they may need to keep some people on. The actual income they get from the extension may be less than the cost of running the network if we ask for it only for a few regions for a certain period. I think they have a strong incentive for the orderly wind-down of their network. But most of all, this is the future of emergency services communications. This is now pretty much their sole business. They will want to get it right.

Charlie Elphicke

I have one other question. Effectively, you are paying for EE to roll out full coverage for their network, which other mobile phone companies can click into. That is great from the point of view of emergency services, but also great for their commercial services. To what extent are you capturing for the taxpayer interest the value that they are going to be getting? They are going to end up with plenty of windfall. To what extent are you clawing that back and making sure that the taxpayer interest is looked after?

Stephen Webb

When we went out to procurement, it was very clear that it would be subject to state aid, which we secured, and it would be open to them to use this network and open it to their customers, so we believe that was represented in the bids that the MNOs put in. It is a very small number of actual customers. In the very remote areas that we are adding masts to, we are talking about potentially 60,000 people, so it is not a huge expansion of their customer base, but none the less it is there, it is a marketing point and so on.

Chair

It is within the contract, so there is going to be no dividend; it is written in.

Mark Sedwill

It was priced into the bids, I guess is the summary, and of course there is a reason why the commercial providers are not in the areas we are requiring them to roll out to—because it is not profitable.

Chair

We are not going to revisit the whole broadband debate today, tempted as we are. We understand that there may be some challenge on EU state aid. Is that right?

Stephen Webb

No, we have had—well, I have seen the submissions that came in from the other operators. We are very confident that we understand the state aid provisions and what we are doing is compatible with that.

Chair

Do you know what the level of fine would be if you were found guilty of giving state aid by the EU?

Stephen Webb

Actually, the risk under state aid rests with the people who get the benefit, which in this case would—

Chair

Would be EE, so not taxpayers but EE would foot the bill.

Stephen Webb

That is how it works.

Nigel Mills

If the reason for a delay is lack of network coverage in, say, rural Derbyshire, and East Midlands ambulance service and Derbyshire police and fire say no, do EE pay for the extra costs of keeping Airwave going until that is fixed?

Stephen Webb

We as in Derbyshire or—

Nigel Mills

No, do EE pay for the cost if it is their fault that the delay has happened?

Stephen Webb

If they have not met their coverage requirements, they will not be paid for the service—it will not be deemed ready—so yes, it will be quite a hit.

Nigel Mills

And the contract has enough of a penalty for them for a delay that it covers all the extra costs of keeping Airwave on in the east midlands for however long we would need it.

Stephen Webb

Well, you are talking about a substantial period of dual running. They are looking to deliver the coverage by September ’17, and Airwave will obviously be running for two and a bit years after that.

Nigel Mills

That was just giving you an example. But the penalty would be enough in a region if it was paid.

Stephen Webb

The penalty would not be getting paid. Their service would not be ready; therefore they would not be getting the core payment, which is really quite substantial.

Nigel Mills

But presumably the amount that you have to pay Airwave to keep running in a region is more than—it looks to be a bit of a penal rate compared with the ongoing running costs of the new system.

Mark Sedwill

That is only if we go beyond the end of course, because we will have quite a long period of dual running.

Nigel Mills

Except in the south-west.

Mark Sedwill

I live in the south-west, so I have the same interest as Mr Foster has. But there will be quite a prolonged period of dual running. As Mr Webb says, they have to deliver the service; if they do not deliver the service or do not deliver it to time, they do not get paid.

Nigel Mills

I suppose the reason for asking that question is that it’s not like you have procured this from one main contractor, it is their job to deliver, and if they don’t you sue them or they pay penalties. You are kind of managing lots of contractors yourself here, aren’t you? Was that an attractive way of running the project, rather than passing the buck on to somebody?

Mark Sedwill

Maybe I could have the first go at that. Of course, I have sat before this Committee dealing with precisely the reverse, when we have had a single prime contractor who has then been responsible for all the subsequent secondary and tertiary contracts. As we know, and as the NAO have evidenced not only for us but in other circumstances, that is a high-risk option as well. The overall change of methodology in Government towards disaggregating contracts and being explicit that the integration risk and, of course, the political risks and so on that follow that sit with the Department, has been generally successful. We are moving towards more agile procurements as technology moves on. I think this is a good example of that. It is actually not as disaggregated as many others, because of the scale of the issues, but overall it is a better way of proceeding than being totally dependent on a single contractor. We are very familiar with some of the risks that arise from that.

Nigel Mills

But don’t you create a new risk? Taking London as an example, say the Underground connectivity is not on time and one contractor says, “Well, our bits of the Underground are fine. It was linking them through to the EE bit or the control room that didn’t work,” and they then say, “No it was them who were way behind,” and they say, “You didn’t commission it in time” and so on. You end up with three different contractors all fighting over whose fault it was that it was late rather than somebody taking ownership of it.

Mark Sedwill

I agree in theory, Mr Mills, but of course we face those risks even if we have a single contractor, because you end up with exactly the same kind of arguments going on between the prime and the subs, and then between the prime and the Department. Again, I have sat here previously trying to explain some of the choices we have made in previous very big programmes that were designed in that way. I think Mr Webb will want to answer this—indeed, it is something that goes more broadly, so Liam Maxwell may well have a perspective on it—but obviously it means that the Department and the programme have to have the capability to resolve those issues and to understand where the technical pinch points lay so that they can be clear about whose job it is to fix whatever issues might arise. That is built into the programme.

Chair

I wonder whether Mr Maxwell would like to come in. Does this, as Mr Sedwill has suggested, reflect existing contractual arrangements in other parts of Government? From your perspective, have the Home Office followed the best advice?

Liam Maxwell

We talked a bit about this when we were going over the Aspire contract, which is a similar situation in which you have one large contractor and are moving to a more disaggregated field. The aim here was to deliver effective services to the emergency services using new technology. The route that was taken did move away from having that come from one single supplier.

Chair

Do you think this contract is a good example?

Liam Maxwell

I think this is a good example of using the right components and the right specialists to deliver the right services within it. Putting it all into one boat could very easily lead to the issues to which Mr Sedwill was referring. Sometimes the attraction of having what’s called one throat to choke in order to get your contract delivered—[Laughter.]

Chair

That is a good phrase for this Committee. Mr Sedwill has volunteered.

Liam Maxwell

It is one of those things where the contractor will turn around and say, “Well, you have one throat to choke and that’s much easier,” but actually all these contracts and service delivery units are disparate teams, and you can have difficulties within teams and within the same company delivering the services. It is not a panacea to say we want to go one way or the other. In this particular case, splitting it into the people who can deliver the specialised core services seems to have been the more attractive route to go down. As we went through our review of it—we were the challenging and support function—we were very comfortable with the way the Home Office approached this. I have to say that I have been in both sessions so far, and we are comfortable with the high level of risk that this offers and with the high level of assurance that the Home Office has tried to place on this. I don’t think it is a case in which people are saying, “We’re complacent. We think this is all a done deal.” We know that this has high-risk components.

Chair

We will come back to that later. I just want to alert everyone to the fact that we are expecting a vote in around six minutes. Mr Mills, carry on until we hear the Division bell.

Nigel Mills

Can we go back to the reason not to continue with Airwave? There is a bit of a feeling that the Government fell out with Airwave because of the price they wanted to charge for further extensions. Is that what drove the motivation for this project?

Mark Sedwill

It was a factor, Mr Mills, of course it was. It was a PFI deal that left us with no equity after 2020. Mr Bacon may have scrutinised it at the time.

Mr Bacon

I did.

Mark Sedwill

But actually, it was not—not really. I set out earlier the need to create interoperable data capability for emergency services. That could not have happened with Airwave; it does not have that capability. In any event, the contract was up in 2020, so even aside from the financial issues, we would have had to have run a new procurement anyway even if the technical solution we had chosen was less ambitious than the one we went for. As I was saying earlier, there is always a range of different factors with these things. I would not suggest that the financial position of the owner of Airwave at the time, Macquarie, was the decisive one, but it was a factor, of course.

Nigel Mills

I suppose we ought to be happy that the Government occasionally call the bluff of a monopoly provider and say, “If that is how you are going to behave, you can lose your business,” which is pretty much what happened to them here. The question is whether falling out with them led to taking a decision more quickly or more aggressively than perhaps you would have done otherwise. We have ended up with a quite radical and short timetable in which to do something that perhaps you might have wanted to wait a few years to do.

Mark Sedwill

I would not describe it as falling out. It was a difficult commercial relationship.

Chair

It might not have been you, Mr Sedwill. We understand that other bits of Government may have fallen out. Is that fair to say?

Mark Sedwill

It is beyond my knowledge. I genuinely do not know. We hadn’t, though. It has proved to be a difficult relationship. It proved difficult once Macquarie had taken over. They were highly leveraged, and therefore were dependent—and were telling the market that they were dependent—on charging highly profitable rates after 2020, and of course that was a factor. But 2020 was when the contract came to an end anyway, so as the accounting officer in good order I would have been wanting to run a procurement to bring the old contract to an end and have a new one in place in 2020. Sometimes we extend a bit, but you only really do that confidently if you already have a procurement in flight, because that mitigates your legal risk. We do need to move emergency services on to an interoperable data capability of the kind that, frankly, everybody else is already using, with the additional resilience that they need. Private sector and so on is what I meant.

Chair

I was going to say that I think this is unique in the world.

Mark Sedwill

No, I mean just in terms of shifting to a business model that the digital native generation are familiar with. It was a series of factors.

Nigel Mills

In the comparison between continuing with Airwave and moving to the new thing, I think you assessed benefits over the period to the mid-2030s or something, yet you have no contracts going forward that far. How do you factor in benefits that you do not have lined up?

Mark Sedwill

That is a more technical answer. I might leave that to the SRO. I think we followed a standard method of calculating this.

Stephen Webb

Yes, we made assumptions about what a base case would be. That is one of the issues, and we have had many debates with the Comptroller and Auditor General and the NAO. We assumed that the Airwave contract would roll forward at the current price with the continued indexation, and on top of that that people would continue to use commercial broadband. On the face of it, that seems like it would be an extremely profitable contract because a lot of the investment should have been paid for in the PFI. There would, of course, need to be a major tech refresh if they had kept it going until 2032, but the main point was that we thought that was a reasonable assumption to make because we did not see that we had any negotiating leverage to get a better outcome than that. In all the time, we never got an offer that involved any substantial discounts. The Permanent Secretary made a point about the level of leverage. The assumptions that seem to be built into the debt gearing of GDCL, the parent company, suggested that that was ultimately what we were going to get. Early days, we did look at whether it would be possible to put some tension on them and create an alternative Tetra provider. I spent a fair bit of money on analysis to see whether we could free up some other spectrum so we could get another operator in there. We just couldn’t make it work. We worked a lot with MOD colleagues and other areas, but we couldn’t get enough spectrum freed up. The way the contract, the spectrum licence and the PFI worked meant that, had we not gone with an alternative solution, we would have had very little negotiating leverage on the incumbent.

Nigel Mills

How do we know we are not going to get into the same position when the original contracts expire so that we do not end up with them hiking prices or trying to maintain high levels of price on this new function from this new system in the mid-2020s?

Stephen Webb

It is a very fair concern. The area we are moving towards is an increasingly standardised one. We are very optimistic that we will get a better deal. In a few areas we have had to go to pre-standards approach with a migration to fully standards-based stuff later. EE is offering basically a network and we should be able to recompete that. We were very careful. I was going to come in earlier about the disaggregation point. One of the advantages of the way that we have disaggregated is that it will make it much easier to recompete and reduce any dominance that the incumbents have. For Motorola as well, it is possible that quite a lot of things that are currently within the scope of the Motorola contract could be moved out as they become fully standardised and commoditised. They will have a strong incentive to add value and to be competitive on price.

Nigel Mills

That is one of the challenges of this contract structure, isn’t it? All the various local customers of this have probably got far less choice now of who they go with for handsets or whatever else than they had under the existing one. They are almost going to get told, “This is your service and this is the price.”

Stephen Webb

For Airwave there were really only two providers of handsets. They will have the full range of availability of devices. I think early on there will be relatively fewer, and we may centrally procure one or two of the main categories, but within a few years I would expect there to be a very wide range. It is essentially just a network with a SIM card and you can put it in.

Nigel Mills

Are you expecting anything from a Huawei to an Apple kind of device?

Stephen Webb

It is network agnostic. We are starting off with Android, Apple and IOS. In practice, I suspect nearly everyone is going to go for an Android system. Yes, there is a wide range of devices out there. I have a few in my bag, including some examples of what the ruggedised versions are going to look like. There is a lot of interest in this and a lot of development work going on.

Chair

But are they going to be credibly ready by this time next year? That is quite a big jump.

Stephen Webb

Yes, it is a big challenge. We have examples of devices that are about to be upgraded to at least 12. We have example prototypes of the ruggedised one with the external emergency button that you push to talk. We know what they look like. It is a question of then going into production.

Chair

I know the vote is approaching, which is why I keep looking at the monitor. Mr Mills, have you got any quick points that you want to bring in?

Nigel Mills

My limited memory of Airwave is that you were expecting device costs or annual running costs to be a lot lower than they turned out to be. In terms of assessing costs through to 2032 and this being £800 a year rather than £1,300, how much risk is there that all these things creep back up and you end up not getting the running costs down as far as you think?

Stephen Webb

The problem with Tetra was that, because it was a fairly small niche technology, you really did not get the economies of scale. On this, we are basically joining the conveyor belt of the general public. We should get the upgrades and economies of scale with efficiencies from that. Again, we have been relatively cautious in our assumptions about how much things will cost when they are recompeted. [Interruption.]