“Superfast broadband for Scotland: A progress report”
Agenda item 2 is evidence from Audit Scotland on “Superfast broadband for Scotland: A progress report”. I am delighted to welcome from Audit Scotland Fraser McKinlay, director and controller of audit; Angela Cullen, assistant director; Graeme Greenhill, senior manager; and Andra Laird, audit manager.
I understand that Mr McKinlay has a short opening statement to make on the report.
Thank you, convener, and good morning, members.
First, I pass on the Auditor General’s apologies for not being able to make the meeting. I am delighted to be here to present, on her behalf, her report on the roll-out of superfast broadband in Scotland.
Access to superfast broadband is hugely important for homes and businesses across Scotland, particularly in some of our more remote and rural communities. The report looks at progress made in developing a superfast broadband network across the country.
The work on building the network started in 2013, when the Scottish Government and Highlands and Islands Enterprise separately employed BT to build and maintain a superfast broadband network. The network will cost around £412 million over an 11-year period, with the Scottish public sector contributing around £165 million of that. BT is contracted to complete the building work by December 2017.
The report assesses whether the Scottish Government and HIE have in place clear plans and arrangements to build the superfast broadband network; it looks at the aims and objectives of the Scottish Government’s investment programme and the targets that have been set; it looks at the procurement and contract management arrangements; and it looks at what has been delivered to date.
I will summarise briefly the findings in the report under three main headings. First of all, we look at the achievement of the programme’s targets, aims and objectives. The Scottish Government’s stated ambition is to develop a superfast broadband network with the capacity to deliver broadband speeds of between 40 and 80 megabits per second to 85 or 90 per cent of premises in Scotland by the interim target of March 2016, and then to extend that to more than 95 per cent of premises by December 2017. We report that the contracts signed with BT do not guarantee that all users will obtain speeds of more than 40Mbps. Indeed, almost a quarter of premises may need to rely on further technological advances or further investment to achieve speeds in excess of 24Mbps. The Scottish Government and HIE are not yet able to state with certainty what broadband speeds they expect the contracts to deliver, as that cannot be determined until the survey work has been completed.
Secondly, we look at procurement and contract management. As the committee will be aware, BT was the only final bidder for each contract, but we are satisfied that both the Scottish Government and HIE project teams used a variety of approaches to assure themselves that BT’s bids were value for money. That included benchmarking BT’s costs against other United Kingdom broadband projects. The teams found that costs in the rest of Scotland are in line with those other projects, but costs in the Highlands and Islands are higher, which is perhaps not surprising. However, it is difficult to conclude definitively whether the Scottish contracts represent value for money, as BT was also the provider for the other projects across the UK. Contract management arrangements for scrutinising progress are good, although we also say that they are complex, and we have highlighted a risk that project teams may struggle to cope in their busier periods when workloads increase as the projects ramp up or when other demands are placed on them.
Thirdly, we look at the progress that has been made towards achieving the interim milestone, which is in March 2016. As at December of last year, BT was ahead of its contractual targets and, based on our calculations, we conclude that the Scottish Government will achieve its interim target to make broadband available to at least 85 per cent of premises across Scotland by March 2016. That calculation takes into account the fact that BT has exceeded its contractual targets up to December 2014, which is the last data that we looked at. It also assumes that BT will continue at least to meet its contractual targets between now and March 2016.
Finally, I will highlight a couple of key recommendations that the report makes. We recommend that the Scottish Government sets out more clearly what it expects the contracts with BT to deliver, not only in terms of both coverage and speed but to help manage the expectations of households and businesses across the country. We also recommend that the Scottish Government and HIE develop clear plans by June 2015 for the additional investment of £42 million in superfast broadband that was announced in February last year. The plans should balance the need to extend coverage to premises that have been excluded from the work and the need to increase broadband speeds in some premises.
As always, my colleagues and I are very happy to take questions.
Thank you, Mr McKinlay. The first question is from Mary Scanlon.
I hope that I will be forgiven for just focusing on the Highlands and Islands, given that Tavish Scott is at another committee dealing with amendments.
I want to pick up on paragraph 1 in the report’s introduction, which says:
“Highlands and Islands Enterprise (HIE) cannot yet state with certainty what broadband speeds they expect their contracts with BT to ultimately deliver.”
I find that quite difficult to understand. I apologise if I am the only member of the committee who does not know this. Is superfast broadband between 40 and 80Mbps, and is minimum broadband 24Mbps? Can you give me a definition of superfast broadband? We have been talking about it for so long, and I would like to understand that.
First of all, I assure you that it is not just you, because we have grappled with the very same question.
I thought that I was the only one.
There is no consistent definition of superfast. Different places use different definitions, and, interestingly, one of the things that we have recognised—this relates to our recommendation on the clarity of reporting—is that although lots of different words, such as superfast, high speed and ultrafast, are used, their meanings are not consistent or clear.
The UK Government defines superfast as having a speed above 24Mbps, which we reckon is a pretty reasonable benchmark. It is worth saying that the Scottish Government has been more ambitious than that in its vision for a digital Scotland: when it talks about the roll-out of superfast, it is talking about speeds of between 40Mbps and 80Mbps. In paragraph 3 on page 7 of the report, we talk through how different countries and different people use different definitions, but we think that the definitions of above or below 24Mbps are reasonable benchmarks for our purposes.
I read paragraph 3, but the point is that this is a report about superfast broadband roll-out, and a lot of expectations are raised when people in the Highlands are told that they are getting superfast broadband. Twenty-four megabits per second is considered superfast in England, but more than 40Mbps is considered superfast in Scotland. I do not think that I am the only member of the committee who does not have someone telling us at least once a week, “I was promised superfast broadband but this is what I have.” I will leave that issue there, now that I understand it.
When there is a contract with BT to deliver superfast broadband and it comes in at 15 or 24Mbps, does that mean that BT has failed? If the Scottish Government is setting a higher bar, how do we know how much superfast broadband has been achieved? How do we know where it is more than 40Mbps? I did not see that anywhere in the report.
I will ask the team to come in and talk in a little bit more detail about what some of the variables are in the speeds that are delivered to homes and businesses—the speeds that people expect and those that they receive at the end of the day.
The challenge is that there are lots of variables that affect the delivery of different speeds to homes and businesses, and it is difficult for BT to know exactly how broadband is going to be delivered until it has done the survey work and has started the work in an area. We recommend that BT needs to be clear about reporting because we think that, even with that restriction, it is still possible for BT to report more publicly and more quickly on the speeds that people can expect to be delivered and, indeed, on the speeds that people are receiving.
We could spend millions and millions of pounds with nobody in the Highlands and Islands having more than 40Mbps, but you could still tick a box and say, “We have delivered superfast broadband.”
The diagram in exhibit 4, on page 21, looks across the country—it is not specific to the Highlands and Islands. It sets out in broad terms what people can expect to get but, as you say, we cannot say as part of that how many of the 55 per cent—the blue bit that we see on the diagram—will get more than 30Mbps, and we cannot say how many will get more than 40Mbps. You are right: it is difficult for us, certainly, but also for BT Scotland and HIE to say.
Another problem is shown in exhibit 3. You could have one person not yet covered by the network but their next-door neighbour could have more than 40Mbps. MSPs find it difficult to deal with people who say, “Well, my neighbour has more than 40Mbps but I am not even connected.” I am sure that colleagues will come back to that.
I find paragraph 58 quite difficult. It says:
“there is a cap of £1,700 on how much BT can spend to connect each premise. If it is likely to cost more, BT must seek agreement from the Scottish Government and HIE and, where relevant, with each contributing council”.
People from Lybster in Caithness have got in touch with me to say that it costs £11,000 to get a BT phone line into a newly built croft house. Is that cap not a bureaucratic means of holding up progress? Seventeen hundred pounds is not a huge amount in a remote and rural area, yet BT has to get permission from the Government, Highlands and Islands Enterprise and the council. Why was the bar set so low—or am I being unrealistic?
I will ask the team to come in on the specifics of the amount if anyone has any further information about it.
Inevitably, there is always a balance to be struck between coverage, speed and cost. I think that the process is trying to ensure that if there are significantly increased costs to connect premises, particularly in remote and rural areas, there is some kind of control to make sure that people sign up to what is being spent. There is no doubt that, as the report says, there are some bits of the country that will not be touched by the programme, and that is why we have Community Broadband Scotland, for example, and why—
Do you agree that having to get permission from three organisations before going ahead is detrimental in terms of bureaucracy, and that it is a delaying factor in remote and rural areas?
I do not think that BT has to get permission from all three; I think that it would get permission from the relevant organisation, depending where the work was being done and which contract was involved. I will check with Graeme Greenhill about that.
In the Highlands and Islands, it would be HIE that would give approval if the likely cost was going to exceed £1,700. We are not aware just how often that has happened so far. We do not know how many times HIE has been approached by BT looking for permission to exceed the £1,700, or indeed what HIE’s response has been to such questions. As Fraser McKinlay says, there is a cap basically to ensure that coverage is spread out to as wide an area as possible.
I understand that.
I see the convener moving me on a bit, so I will ask my final question.
I have to say that I found the report slightly contradictory. I will give an example. Paragraph 68 says that
“BT had exceeded its contractual ... targets”
by 57,000 premises. However, as at December 2014, it was 14,000 premises behind the implementation plan. Paragraph 71 says that it has not achieved the target, so it changed it. I will lump in my final example. Paragraph 79 says that Highlands and Islands Enterprise paid BT £26 million less than planned. That is an awful lot of money. Can you explain why it did not pay the £26 million to BT that it had planned to pay?
10:15
I will briefly touch on the first bit of your question, which was about the difference between the contractual targets and the implementation plan targets. The contractual targets are, as the name suggests, the things that BT has to deliver as part of its contract. As we say in the report, if it keeps meeting those contractual targets, it will meet its overall target by the interim date of March 2016. The implementation plan targets are BT’s own targets and include some premises that are not included as part of the contract.
Does BT have targets that are separate from Highlands and Islands Enterprise’s targets?
There is a difference between what BT is contracted to deliver and what it is saying, in operational terms, that it wants to achieve. It looks a bit contradictory because those are two different things.
Does BT have two types of target?
Effectively. There is a contractual target, and there is an operational target, which BT has set.
No wonder I am confused.
I have been trying to get my head around it as well. The important thing to focus on is that if BT continues to meet the contractual targets, it will meet the overall ambition of reaching 85 per cent of premises by March 2016. The implementation targets that BT has set for itself would take it over and above that.
The Scottish Government reduced the implementation target. The report says that in paragraph 71.
As the programme is rolled out, BT, HIE and the Scottish Government are having conversations about the contractual targets and the implementation targets, about how many premises will be done and about where and how the programme will roll out. It is a continually moving feast.
Okay. What about the £26 million?
I will ask the team to come in on the £26 million.
The £26 million reflects work in progress. Basically, much of that underspend is because BT has not submitted the necessary invoices and evidence that it has completed the work up to that point.
It is encouraging to get a report that indicates that the Scottish Government’s projections are on target and within budget. Notwithstanding BT’s interpretation of the targets, it is encouraging to see that it has exceeded its own target, and to read comments such as:
“Arrangements for scrutinising for BT’s progress against the contracts are good”.
The programme appears to be well managed and well controlled, as far as it goes.
I will ask a question for my own information. Will upload and download speeds be the same under superfast broadband?
It depends on which package you buy from the internet service provider. Some will sell upload and download speeds at the same level, but most often the download speed is higher than the upload speed. However, it varies, depending on who you buy your service from.
I have always been curious about that. ISPs always quote the fast speed of their broadband, but, in actual fact, the download speed is frequently very much less.
It is usually higher than the upload speed.
Maybe I missed it, but the report does not mention resolving grey commercial areas or how that issue is being tackled.
We have not looked at that. If there is anything specific that you want us to have a look at, we can certainly take it away, have a look at it and come back to you .
I was just keen to know whether the programme is on target as far as grey commercial areas are concerned. My understanding is that the issue will be resolved possibly as early as early April, and I was just curious to know whether your research validated that. It is quite an important point.
Graeme, do we have anything on that?
It is not something that we looked at. As you know, and as the report says, some of the councils in the rest of Scotland area have been putting their own money into the projects specifically to address aspects such as grey commercial areas as part of their local economic development plans. However, we have not looked at the extent to which the issue has been resolved.
The issue is quite important for local communities because, obviously, until it is resolved, they will find it difficult to know how to move forward.
On page 9, the final sentence in paragraph 10 says
“Fibre ... gives better speeds than copper as it does not slow down over long distances”.
I have been hunting unsuccessfully for a diagram that I got from BT. According to my memory of the diagram, it seems that in fact the speeds drop off surprisingly quickly for fibre optic. If I remember correctly, the loss in speed is fairly steep after about half a kilometre.
I should say that my technical knowledge on this is a little bit sketchy in places.
Mine is pretty close to zero.
We are probably not that far away from each other then.
Can we be clear that we are focusing on the report and that the question is relevant to the report? There may be some technical aspects of this that are for—
All I am querying is the statement in the report that fibre optic will give better speeds and less drop off. I am curious as to the—
It is certainly the case that fibre is better than copper. One of variables that I mentioned is how far away your house is from the cabinet. The closer you are to a cabinet with a fibre connection, the faster your broadband speed will be. By no means is fibre perfect, but it is better than copper.
The report also says that
“BT encountered delays in obtaining marine licences for the subsea cabling”.
How many months’ delay did BT face in getting the licences, and what were the reasons for the delay?
Andra, can you help with that?
BT had planned to start doing the subsea cabling work earlier in 2014 than it was able to. The delay was about three or four months, but BT was able to catch up and do all the work this year. I think that the issue was that Marine Scotland was implementing a new process and it just took a bit of time to learn what that process was and go through it.
Is the delay resolved?
Yes. BT has done all the work that it planned to do.
The first bullet point at the top of page 32 says
“BT had difficulty getting access to privately owned land.”
I thought that, as a result of the legislation that was put in place, BT had some sort of pre-emptive right to enter private land for the purposes of communications.
Again, I do not know the detail of that; it just seems to be something that took BT a while to negotiate. I think that it is building infrastructure on private land, and that it has to go through a process to get that agreed.
Was the difficulty in one case or in a number of cases? How widespread was it?
We do not have the detail, but there was more than one instance.
Okay.
Overall, this has been a good report. When will you be doing a follow up?
Obviously, we will continue to monitor the programme as the contract rolls out through to the end of 2017. However, as to what else the Auditor General decides to do, I think that, apart from anything, we would want to see what the committee decides to do.
I will bring in Sandra White for a brief supplementary before I move to Drew Smith.
With regard to the commercial grey area that Colin Beattie referred to, I note that the last bullet point in paragraph 54 on page 24 of the report says:
“Twenty thousand premises in Edinburgh, Glasgow and Aberdeen will not have access to superfast broadband. This is because the commercial sector has decided not to invest in these areas”.
In parts of my own area and other areas, particularly in the city of Glasgow, where you would expect broadband, businesses have no such access. In fact, they have been told by BT that because of European Union procurement law it cannot be installed. Can you clarify that point? After all, such access is very important for small businesses in certain areas.
I ask Graeme Greenhill to comment on that.
In order to comply with state aid rules, which basically say that the public sector is not permitted to interfere in the free market, public sector contracts are essentially geared towards providing access to broadband in rural areas. In urban areas, the expectation is that the free market will provide broadband. With regard to the 20,000 premises that have been mentioned, the private sector has decided that there is not enough money to be made in investing in those areas.
You mentioned the private sector and competition. Another substantive question is about the contract itself; given that only BT has the contract, there is not much competition. Would local councils in those areas be able to step in and provide the moneys for small businesses that are not able to access broadband through this particular procurement contract?
I cannot give you a simple answer to that. Again, I think that state aid rules might be an issue that the councils would have to think about.
I wonder if you can briefly talk me through the section from paragraph 35 onwards that refers to the European regional development funding. We are talking about pretty significant public investment here; over the page, we find that public money accounts for almost 90 per cent of the funding in the Highlands and Islands and around 60 per cent in the rest of the country. Any opportunity to access some European funding would be very welcome, but what are the challenges in that respect? Why are we not going down that route and claiming as much as we possibly could?
The Scottish Government originally wanted to claim £20.5 million, but that figure has come down to £13 million. When it was looking for £20.5 million, the Scottish Government expected the EU not to classify the project as revenue generating; however, after negotiations, that is exactly how the EU classified it. The Scottish Government was not generating revenue, but BT was and, as a result, the intervention rate dropped. The Government had originally thought that the rate would be 40 per cent, which is where it got the £20.5 million from. However, the rate dropped to 25 per cent, and the funding was reduced to £13 million.
Sandra White asked about the extent to which local authorities might be involved. Can you give me a bit more detail about that? Under what criteria can local authorities put their own money in? At one level, you are interested in a consistent approach but, at the same time, the situation is different in each part of the country. Where do moneys to support infrastructure come from? I am interested in finding out how we support people, particularly in my part of the world, to access broadband, but that question is separate from whether the infrastructure is actually there. Are councils having to make a judgment between those two things or is there dedicated funding available for them to do it? What is going on in different parts of the country with regard to councils choosing to intervene?
It is worth reminding ourselves that there are two elements to council funding. As exhibit 2 on page 16 shows, there are what are called collective and individual council contributions. As part of the UK roll-out of superfast broadband, the Scottish Government got extra money and agreed with the Convention of Scottish Local Authorities to pass that money straight to the delivery of these projects; that money is what we call the collective council contributions. Beyond that—and the team will keep me right here—it was up to individual councils to look at their local areas and local economic regeneration and decide whether they wanted to put more money in themselves. I am not sure that there were any consistent criteria across the country; I think that it was up to individual councils to decide whether it was worth investing in an area and what the benefit would be from that investment. Is that about right, Andra?
Yes. I have nothing to add.
I am concerned that if one local authority is proactive and says, “We want to be involved in improving this situation and invest money in it”, and another says, “Well, we’re not willing to do that”, the outcome will be that the state might end up picking up the tab for the people who were less willing to assist at the time. I presume that you do not have concerns in that respect.
No. It is worth pointing out that every council area in the land, with the exception of the Western Isles, will have at least 75 per cent coverage. This is going to make a big difference right across the land, and it will go a long way towards achieving the objective of trying to bridge the so-called digital divide.
Within that, the issue is about what councils want to invest their money in and what their local priorities are, but it is worth bearing in mind that this is all about getting everyone up to a much better level. Beyond that, it is a matter for individual councils, and we have no particular concerns about the way in which that has come about.
10:30
We now have the Highlands and Islands contract and what is rather inelegantly called the rest-of-Scotland contract. The report says that that happened primarily because the process in the Highlands and Islands was at a more advanced stage, but on the face of it, it would have made more economic sense to have had one contract for the whole of Scotland. Is that a fair comment?
We looked at the issue quite closely, and I do not think that we reached that conclusion. As you have suggested, the starting point pragmatically reflected the fact that Highlands and Islands Enterprise was further down the road with this; however, it thought carefully about halting the Highlands and Islands contract and combining the two, but the trade-off was between the delay for the Highlands and Islands and whether that would have been more significant than any benefit gained from combining the two projects.
Later on in the report, we say that the governance arrangements are pretty sound. Although officially the two contracts are being operated separately, the contract teams are regularly in touch with each other; a lot of discussion and collaboration are happening between them; and at the top of the governance structure, there is a place where everything comes together. They have for what we think are perfectly sound reasons decided to go ahead with two contracts even though, as you have said, you would instinctively think that a place such as Scotland should be able to have one. Having looked at it, we think that the decision to continue with two was perfectly reasonable, and structures have been put in place to help manage the risk.
That is helpful. My final question is about the exit point for these contracts and when they complete. Will the outcomes at the end of these contracts be future proofed? Given that technology marches on, is it the case that the process is not going to end here? Will we get to a point where the two areas in question are broadly comparable, or will we still find major differences between the two if indeed we want to consider having one contract or putting a different model in place? What issues should we be interested in there?
As we have said, our calculation is that by the end point in December 2017, the target of having a network with the capacity to deliver these speeds will have been met. However, as we also point out and as you have rightly highlighted, technology moves on apace, and more investment and development will be required. The contract is designed to support that, and it contains incentives to support and encourage BT to continue to invest beyond its end. There is no doubt that if the targets are achieved—and at the moment we think that they will be—it will make a significant difference to the availability of broadband across large swathes of the country.
It is worth bearing in mind that, had none of this been done in the Highlands and Islands, only about 21 per cent of premises would have had access to superfast broadband. As a result of this project, the figure will be up to 82 or 84 per cent. That is a significant difference, and it is the reason for our recommendation that June of this year should be an important milestone for thinking about exactly what needs to be done beyond the current contract. After all, £42 million has already been identified. What has to be done to reach the bits that have not yet been reached and to look at speeds, too?
Perhaps I can chip in with some numbers to illustrate Fraser McKinlay’s comments. Paragraph 51 gives an indication of expected coverage at the end of December 2017, by which time all of the infrastructure is supposed to have been built, and makes it clear that there will be a difference in expected coverage between the rest of Scotland and the Highlands and Islands. Indeed, it points out that in the rest of Scotland a much higher percentage of coverage will be delivered by the private sector, which reflects the country’s geography and the areas where the private sector obviously thinks that it can make most money out of doing this itself.
Good morning, colleagues. I am grateful to Drew Smith for getting me exactly where I want to carry on from. To me, it does not matter whether we have 75, 85 or 95 per cent because I will always be worried about the next bit that we have not done. I am going to carry on being worried about this until we have arrived at 99 point something per cent, although I recognise that there will always be a cot somewhere on the top of some ben that we are not going to get to.
Could I start in the context of the report at page 37, paragraph 90 or so, where you talk about the modelling the benefits? I am wondering whether we have arrived at the point where “benefits” is just not the right word. Maybe we are at the point where this is essential and everybody needs it. A couple of weeks ago, we heard from NHS Highland that it wants to be able to do things by broadband; it wants to be able to interface with patients by Skype or whatever. We are rapidly reaching the point where “benefits” has become the wrong word and it is essential. Forgive the repetition but, as auditors, how do you see other people thinking? Are they thinking like that and responding to those kinds of thoughts, or are they still talking about the next 10 per cent when, really, it has to be everything?
It is worth bearing in mind that it is the Government’s ambition to have universal coverage by 2020 so, in its ambition, the Government is already there and this project is a big step in trying to achieve that. However, I think that you are absolutely right. In a sense, it is easier to think about the disbenefit of not doing it than the benefit of doing it because, as you say, for all sorts of reasons to do with telecare and learning and communities and inclusion, not having it is not an option. I absolutely agree with that.
Equally, though, as auditors we will continue to be interested in questions of value for money. There is £400 million plus going into this project, £165 million of which is from the Scottish public sector. It is important that the Government and HIE are able to articulate, not for our benefit for but for the benefit of the taxpayer in Scotland, what we think we are going to get for that money. What are the implications for healthcare? What are the implications for economic growth? What are the implications for learning and inclusive communities? I think that we all have a sense in conceptual terms that that is all a good thing. We think that there is more that can be done to pin some of that down and update it as it goes, once we know what the speeds are and what difference those are making. It is interesting that nobody has clearly articulated what the difference is between having a broadband speed of 17 to 24Mbps and having a speed of 24 to 40Mbps or 40 to 80Mbps. It would be very helpful to understand such things better
Yes, although I suspect that that will change as people learn to change the way they pixellate pictures and all the rest. The technology is always moving under your feet. Could I then ask you the corollary of that, which is going back to the whole system? As a chemical engineer in a previous existence, if I had to design a pipe system that got water from here to everywhere I would be perfectly capable, and so would my colleagues, of designing it back here in such a way that it would be big enough for everywhere once we finally got there. That is easy for water. I have no idea what the equivalent is in terms of broadband, but are you assured that people are asking the right questions to make sure that what they are putting in throughout the entire system is going to be big enough, strong enough and versatile enough to serve the whole community when we finally get to 99.99 per cent?
Again, I would be stretching it a little bit and getting into speculation, which is a sure way of getting a kick under the table from my team, if I started to answer that question. Having said that, I think that what the Government is committed to is, as we say in the report, a network that has the capacity to deliver those kinds of speeds. Progress is being made and significant work is being done. It is worth bearing in mind that the significant roots-up work that BT is having to undertake in the backhaul network, to use the technical term, that it is having to build, with undersea cabling and all those things, will stand it in good stead for whatever comes along in the future. I certainly would not want to give a cast-iron guarantee because that is not my job or my business, but we think that it will certainly put it in a much stronger position than it would have been in otherwise.
Fine, thank you. Lastly, I will come back to the issue that has already been raised about state aid rules. I have a suspicion that, where we are sitting, we are within a good seven iron of buildings that will not get broadband commercially, precisely for the reasons that you have indicated. Presumably, for the reasons that I discussed a couple of paragraphs back, there will come a point where the state aid rules have to be changed in such a way that it is recognised that for telecare, for example, even though the commercial suppliers will not get there, it is in our collective interest to get there. Do you see people asking the right questions in the right places about whether the state aid rules are going to change?
I am not sure that we have seen evidence of that specifically. We have seen evidence that the people involved in this exercise are very conscious of the pockets in urban areas that are not getting access at the moment. One of the interesting things for me in coming to this reasonably fresh is that there are pockets in areas that are not going to have access when you would absolutely expect them to. I think that that needs to be part of the conversation, for example about the additional £42 million. Is there anything that can be done in the context of the state aid rules? Who knows? If the state aid rules continue to be a challenge, that might be something that the Government and HIE and others want to look at. For the moment, I think that the focus is on delivering as wide a coverage as they can, given the constraints that they are operating under.
We have touched on this area, but I am looking for a bit of clarity about the future proofing of all the technology that is being installed. We have already touched on the subsea cabling work and the target that the Scottish Government has for the roll-out of this technology. There has to be continual investment anyway but, at some point in, say, 10 years’ time, if there had to be a major investment in the infrastructure, could the subsea cabling that is currently being installed be easily amended, or will it be capable of taking a stronger superfast broadband? Obviously, the target at the moment is between 40 and 80Mbps.
Team, can we help with that?
I am not strong on this. I think the backhaul network—the big cables that are being put in—is capable of delivering very fast speeds. The reason why BT is not predicting that for this project is what happens at the end of those cables where they get to the cabinets and copper wire takes over. Future investment to try to boost speeds might deal with that copper wire element.
The reason for posing the question is, first, I am not a technical geek in any way, shape or form, so I did not know what the answer was. Secondly, when it comes to subsea cabling, obviously Marine Scotland has to be involved, as does the Crown Estate. Obviously, there is discussion at the moment about the devolution of the Crown Estate to Scotland and about what happens to it after it is devolved to the Scottish Parliament. If the Crown Estate were to be devolved further to local authorities, that would then open up more organisations to get involved in the planning and licensing of subsea cabling. I was just trying to get a bit of clarity on what may or may not happen at some point further down the line.
Given the complexity and the work involved in doing subsea cabling, you do not want to do it every couple of years for sure. I think that BT has laid cabling that is as future proofed as you can make it, which will allow fast speeds up to 40 to 80Mbps. As we have said, looking 10 years ahead in a technology landscape that moves incredibly quickly is quite tricky and I guess that BT and the guys who are doing this for a living would probably be better placed to give you a more detailed answer to that, but certainly we have not picked up any concern that BT is going to have to go back and do some subsea cabling any time soon.
I have just one final point, if I may, convener. In paragraph 3, you provide the examples of the German and Swedish targets—Sweden is proposing to increase its target. Did you look at the levels of superfast broadband in any other EU member states for comparison?
10:45
Certainly, the information is available, but what we did in that paragraph was pick a couple of examples that give you an idea of the higher ambition. The EU has set the benchmark in asking for ultrafast broadband speeds for 50 per cent of households by 2020, I think. Everybody is trying to move in that direction.
I am aware that Estonia is one of the most advanced countries in the use of electronic technology. It even has electronic elections, so it is certainly something that it has made progress on. Did you look at what Estonia has done?
We have not as part of this work but, like you, I am familiar with it and my understanding of places such as Estonia is that, in a funny kind of way, it is easier to go from nothing at all to really fast. We have a network and an infrastructure in place that we need to upgrade and do things with. If you were starting from a blank sheet of paper, it would probably be easier to lay cables and to get connections that would be more fit for purpose for today’s usage, which the bulk of our network in this country was not designed for. If it would be helpful to the committee, we would be happy to look at some other countries and come back with some detail on the kind of speeds that they are looking at. That is no problem.
Mary Scanlon has a quick supplementary and then I have a final question.
Again, it is just to help me understand. At paragraph 51, the report talks about how many households and premises are signed up to the “broadband network” and then, at paragraph 53, it talks about “superfast broadband”. Just to help me understand, can we assume that when you are talking about connections to the broadband network, you are talking about less than 40Mbps, and when you are talking about connections to superfast broadband, like paragraph 53, you are talking about over 40Mbps? They tend to be interchangeable through the report.
I appreciate that, Mrs Scanlon. No. When we refer to the superfast broadband network in this report, we are using the cut-off of 24Mbps for that.
I thought Scotland was 40Mbps.
Yes. Given that there is no definition of what superfast is, what we have tried to describe is that the Scottish Government had an ambition to achieve between 40 and 80Mbps, but we think that 24Mbps is a reasonable benchmark in terms of what most people would describe as superfast. That is the number that we have used for the purpose of this exercise.
Paragraph 53 refers to “superfast broadband” and the preceding paragraphs just say “broadband”. Is that less than 24Mbps?
With apologies, I think that that is probably just us not being consistent enough in our language, Mrs Scanlon. We do not mean to differentiate between the two. We have just used a bit of shorthand there, so my apologies for the confusion there.
One assumes that Audit Scotland is just about numbers, but I am impressed by your knowledge of cables and various other things, too. You have done very well in answering our questions. Audit Scotland might want to look into state aid and the issue of small pockets, particularly in my area of Glasgow Kelvin and others where people are working more from home. You might want to look into their difficulty with accessing broadband.
Obviously, the Public Audit Committee is concerned with value for money and so on. The report is very good, and it seems to be working out fine. However, BT seems to be the only company that bids for and gets these contracts. In paragraph 55, the report says:
“HIE allowed two companies to progress to the competitive dialogue stage”.
Is there any reason why one of those companies pulled out?
The Scottish Government also looked at two companies—BT and Fujitsu—and one of them, Fujitsu, dropped out. Is there any reason why BT is the only bidder? Perhaps that is confidential information. In your opening remarks, you said that although you looked at other parts of the UK for comparison, BT was the only company that got the contracts there so there was no real comparison that you could make.
We have not commented on or gone into much depth about the market, but it is very striking that the market has produced only one provider. I think that there are 44 contracts in the rest of the UK, and obviously we have two here in Scotland. As you said, BT is it, which makes it difficult to assess value for money. As I said, we are satisfied within that framework that the teams have done as much as possible to assess and assure themselves that value for money is being delivered. However, there is no doubt that there is an issue with the market when BT is—apparently—the only company that is able to do this work. When our colleagues in the National Audit Office in London did a bit of work on this a year and a bit ago, they commented on the same thing.
The team will respond on the specifics of why some contractors came out of the process.
There is a bit of information on that on page 41, in endnote 14, which talks about the Highlands and Islands process. Fujitsu just withdrew, having decided that the job was not for it. Cable & Wireless’s solution was non-compliant; I would have to go back to Highlands and Islands Enterprise to check whether we can reveal what the difficulty was there, but perhaps we could follow up with a quick summary report on why it dropped out. The third company was Commendium; it progressed further and was left in the process with BT. It offered to deliver the broadband through power lines, and I think that there was some concern around whether, or how well, that could work in the Highlands and Islands. Having considered that, I think that the company dropped out at that stage. Highlands and Islands Enterprise made a good attempt at getting competition, but that is what happened to the different companies.
It is worth bearing in mind that these contracts are, by definition, the hardest bits to do. Obviously, there is much more competition in the majority of broadband provision in the country, which is commercially provided. In a sense, it was always going to be more difficult to attract people to the bits that the commercial providers do not go for. As Graeme Greenhill said, there are commercial reasons for that.
It would be good to get the report that you mentioned, and I apologise for not picking up on page 41 in that regard. It would be good to understand why BT is the only provider. That concerns me, and I know that it concerns others, too—you mentioned the National Audit Office.
I have another point about HIE; Drew Smith may have touched on this point, too. HIE went out to tender in 2012, but it did not go through the BDUK’s framework agreement—thankfully, I have looked that up and found out what it means—for the procurement. The Scottish Government went through that process just a couple of months later. The report says that the agreement was not in place at that time. Given that it was only a couple of months earlier, I wonder why the agreement was not in place. Would it have been easier or more cost-effective if HIE had gone through the framework agreement?
In response to Drew Smith, I think that you said the situation is being monitored and that HIE is absolutely fine with the two different contracts. As I said, I just wonder why HIE did not go through the BUDK’s contract.
I think that it was a matter of timing. Highlands and Islands Enterprise had started earlier, because it wanted to push ahead. It saw broadband as a crucial investment for the economic development of the Highlands, and decided that it would rather bash on than wait for the framework agreement to become available. It would be very difficult for us to say what difference it might have made had HIE waited for the framework to become available.
I have a final question that follows on from the contractual issues that Sandra White raised. Can I just play this out? There are bidders in the market looking at the contract, and BT is part of that. Given BT’s specialism and the fact that it owns the infrastructure, it has a significant advantage. BT then becomes the only bidder in town. At that stage, is there not a need to look at the contract that has been put out to tender, given that only one contractor is left in the running and that contractor owns all the infrastructure? Is that not a significant advantage in delivering the contract, given all the infrastructure and all the manpower that are available to BT? Does the contract not need to be renegotiated? I have seen it happen in a lot of situations that franchises have had to be renegotiated. Did that happen?
I am not sure about a renegotiation, but as I mentioned I think we can see in the contracting processes that the teams did a lot of work to assure themselves of value for money. There are also mechanisms in the contract process that help to ensure that costs are managed and benchmarked, and that there are good levels of control. There are also clauses about incentivisation. If contractors are paid only when things are delivered—as we said, there is a robust process for checking that the work they say has been done has actually been done—that means that, if they do not do the work, they do not get paid. That seems a sensible way to progress.
There is no getting round the fact that we have a network that was, historically, BT’s. It continues to own that network, and that is the environment in which we were working. Within that environment, we think that the project teams have done a good job and have taken a reasonable approach to checking value for money.
When tenders were invited, I suppose that nobody expected there to be only one bidder in the system—although perhaps we did. When we reached that stage, were there no options available that would enable us to say, “It might be much better value for money to look at another method of going forward with the project.” We have a bidder that, even at the very start, has a significant advantage over everyone else at the table.
As we go through a contract process like this, there is always a process of negotiation, discussion and challenge both ways. Perhaps I could ask the team whether they have any evidence of other options that were specifically considered; if we do not know, we can double-check that. Instinctively, I am not sure what other options there would have been. The situation is what it is, and BT has the infrastructure. Given the digital divide and the desire to expand coverage, the options were probably pretty limited.
There are analogies with this, where public money is involved. Bids have not come forward in franchise processes, although I appreciate that that might not be an exact analogy. If a bidder has been at a significant advantage, that situation has been looked at by the Competition Commission. There are examples in which public money has been spent and somebody has been at a significant advantage, and the public purse has had to look at that and say, “Could we not have renegotiated?” and we have done that.
Surely, if we end up with just one bidder, we have to look at doing that? I think you have already alluded to the fact that having one bidder on the table is not the ideal position for us to be in.
As you said, we have looked at other big infrastructure projects and other capital projects that have ended up in that place. The difference with some of those projects—off the top of my head, I am thinking about the Borders rail project, which ended up in that position—is that they are being built from scratch and there are other people out there who can do that.
There is a genuine issue in this case that is probably unique, and it is about the state of the infrastructure and BT’s dominance of the market. I mentioned the contract process because, in a sense, the renegotiation was happening all the way through to the point of the contract being signed. It was not a case of the contract being signed and then the other people dropped out. The project teams knew that they were operating with only one preferred supplier, so they were working to ensure that the contract that they signed was as competitive as possible in the environment that they were working in. I do not think that it was a case of signing the contract and then going back and renegotiating it. All of that work, on negotiating the best deal that they could get, was done in the lead-up to signing the contract in the first place, if that makes sense.
11:00
The contract was developed on the basis that people would bid for it and that, in the end, a final bidder would come forward. I suppose that that is how it would have been modelled—that there would be some competition—but that is not how it ended up. I am intrigued, and probably a bit concerned, by the fact that someone has not taken a step back and said, “Yes, BT are the only players in town here, but we are giving them a lot of cash.”
From a public perspective, there are issues about competition and about the roll-out of the project. As constituency and regional MSPs, we receive complaints, as Mary Scanlon said, about streets where some people have superfast broadband and their neighbours do not. There is a host of issues surrounding that, and the public funds are significant. Someone must have looked at the situation and said, “Is there something else we can do here?”
I agree with all that. In an ideal world, and from where we are sitting, there would be a much more competitive environment and a much more competitive process. That is not how it ended up. As I said, we can double-check to what extent there was a specific and explicit process of thinking about what else could be done if the work was not done by BT. If the committee were to decide to take evidence, I guess that there would be a question for the Scottish Government and for HIE on what their view was in dealing with the market situation that they found themselves in.
If BT owned the infrastructure, any other bidder would have to lease that infrastructure within their bid. Did you find any evidence of the leasing of the infrastructure being made prohibitive for any of the other organisations—Cable & Wireless, Fujitsu and so on—or was that ever found to be detrimental to them pursuing their bid? Was that one of the reasons for them dropping out? Obviously, BT owns the infrastructure, which puts it in a powerful position. The others have to rent BT’s cabling infrastructure. Was that found to be a factor in other organisations dropping out?
I think that I am right in saying that the contract specifically deals with that issue, but I ask Andra Laird to comment.
There are regulations in place. BT is divided, effectively, into two different bits. There is the bit that builds the infrastructure and then there is the retail part. The part of BT that has built the infrastructure needs to sell to BT retail, and the regulations are in place to ensure that the price at which it sells to its own retail division is the same price that it has to offer to other providers.
That is very helpful. Thank you.
Have the companies that expressed an interest initially and then did not carry that forward undertaken similar work in other countries?
I do not know.
We can check that for you, Mr McMillan, and see whether we can find out.
Thank you.
I thank Mr McKinlay and the team for their time this morning.
“Preparations for the implementation of the Scotland Act 2012”
We move on to item 3, under which we will consider submissions on the AGS report, “Preparations for the implementation of the Scotland Act 2012”. Members also have a further submission from Revenue Scotland, which was received on Monday.
Members will note that Revenue Scotland has confirmed that it will not be invoking its contingency plan for the information technology system, which will now be made available to users before 1 April 2015. Revenue Scotland has also committed to update the committee on its key performance indicators, following the discussions this month. Audit Scotland has also confirmed that Revenue Scotland will be subject to its audit reporting regime.
Given those responses, I am minded to note the submissions, but I would welcome comments.
I am keen to keep a watchful eye on the issue, but I am happy to note the submissions.
I think that the submissions have given us the reassurances that we were looking for. One thing that we need to look at is paragraph 32, which states:
“The Board is considering these issues”—
meaning various operational issues—
“and seeking to establish a way of presenting Revenue Scotland costs for scrutiny that will be more meaningful.”
That is obviously quite an important thing to achieve, and presumably Revenue Scotland is working with Audit Scotland on that, but maybe it is something that we should note for the future.
Are there any other comments?
I think that we should just note the submissions.
Further to Colin Beattie’s comment, we can have a look at the response in connection with the key indicators when it comes in. Is that acceptable?
Yes.
Do we agree to note the submissions?
Members indicated agreement.
Next
Section 22 Report