Official Report: Minutes of Evidence

The Chairperson (Mr Humphrey): The officials joining the meeting today are Mr Mike Brennan, the permanent secretary and accounting officer at the Department for the Economy; Mr Nigel Robbins, project director, Project Stratum; Mr Trevor Forsythe, project manager, Project Stratum; Mr David Hamill, project manager, superfast rollout programme (SRP); Mr Andrew Field, programme director, Project Gigabit; and Mr Billy McClean, area development lead for Northern Ireland. Also in attendance is the Comptroller and Auditor General, and joining us remotely is Mr Stuart Stevenson.

Can all our witnesses hear and see us OK?

Mr Mike Brennan (Department for the Economy): Yes, Chair, we can hear you fine.

The Chairperson (Mr Humphrey): OK. Thank you very much.

Members, with your permission, I invite Mr Brennan, the permanent secretary and accounting officer in the Department for the Economy, to make an opening statement. After that, I will open the meeting to members for questions.

Mr Brennan: Thank you, Chair. I just hope that you can hear me OK.

The Chairperson (Mr Humphrey): We can hear you fine, thanks.

Mr Brennan: Good afternoon, folks. As the Chair said, I am Mike Brennan, the permanent secretary in the Department for the Economy. The Chair also indicated that we have a number of other witnesses with us today. We have Nigel Robbins, the project director of Project Stratum; Trevor Forsythe, the project manager of Project Stratum; and David Hamill, the project manager of the superfast rollout programme. In addition to those colleagues from DFE, we are joined by two colleagues from Building Digital UK (BDUK). They are Andrew Field, programme director of Project Gigabit; and Billy McClean, the Northern Ireland area leader for BDUK.

At the outset, our Departments welcome the Audit Office (NIAO) reports on broadband investment in Northern Ireland and Project Stratum. It is always valuable to have independent scrutiny of any government-funded projects to ensure that best practice is followed and to inform future policy in any area of our Department's endeavours. DFE is always open to considering all possible options and exploring whether and how government should intervene in supporting improvements in our broadband infrastructure.

The Department for Digital, Culture, Media and Sport (DCMS) representation here today is critical because, as you know, telecoms policy is a reserved matter. DFE has only a very limited role under the Communications Act 2003. Much of the discussion today will probably focus on what are reserved policy matters.

My starting-point observation is that the recent projects delivered in partnership with BDUK have already improved superfast broadband services to over 67,000 premises across Northern Ireland. The economic and social benefits to businesses and citizens are well understood. Indeed, in 2018, the National Infrastructure Commission stated that digital connectivity is an essential utility, just like water or electricity.

The need for access to good broadband connectivity has never been higher or better understood than since the advent of COVID. DFE is determined to continue working very closely with BDUK and other key stakeholders to consider how best to take forward policy in this area. However, the challenges for government are many-faceted. The challenges include the need to ensure that resources are used to the best advantage in an area where technology is moving at an incredibly fast pace. There is also a need to predict what the needs of business and the citizen will be over the medium to long term. Also, policy and projects must be implemented within any applicable state aid compliance arrangements that exist. Finally, there is a need to stimulate commercial investment where there is a market failure without, at the same time, substituting an investment that the private sector would have taken forward.

As you know, Project Stratum is our current broadband intervention, and it is the largest publicly funded telecoms infrastructure project ever undertaken here. It is no exaggeration to say that Stratum is transforming the conductivity fortunes for all our citizens, but particularly in rural areas, where the lack of access to good broadband is most felt. In that context, it is important to remember that 97% of Stratum's interventions are in local rural areas. Fibrus has already delivered gigabit-capable broadband to over 22,000 premises here, ahead of schedule. However, the Audit Office report on Project Stratum is indeed timely. Of course, we welcome the positive references to contract award and management approval processes with references to best practice techniques in procurement. We also welcome the acknowledgement that the delivery fully complied with the 2016 national broadband scheme aid requirements. However, it is fair to say that it was not an easy procurement to deliver and the team had to address a number of complications that emerged from bidders late in the process. The project progresses; indeed it has been enhanced. Just this week, it was announced that an additional £32 million of funding has been provided by the UK Government and the two devolved Departments to see a further 8,500 premises now being made eligible.

I will pause there, Chair, and look forward to the discussion as it unfolds this afternoon.

The Chairperson (Mr Humphrey): Thank you. Do any other members of your team wish to make any additional remarks before I bring members in for questions?

Mr Brennan: Not at this stage, Chair.

The Chairperson (Mr Humphrey): Before I bring other members in, I will ask a question. In the Northern Ireland broadband improvement project (NIBIP), the Audit Office report shows a huge difference between the 117,000 properties that the Department originally thought its investment would improve broadband access to and the actual amount of 37,500. Can you explain to the Committee how the Department got that so wrong?

Mr Brennan: I am not sure that I would say the Department got it wrong, Chair. Those two figures represent the evolution of the procurement process. I will bring in my colleague Nigel in a second. The 117,000 was the Department's initial assessment of premises that could be brought within scope at the very start of the procurement process when it was trying to identify the impact in terms of coverage. As the procurement process moved through its various stages and data became available on, for example, the cost to deliver to rural areas in Northern Ireland, the extent of rurality in Northern Ireland, the fact that we were moving to different degrees of technological upgrade and a whole range of other cost items, the assessment of the premises that could be captured at the final stage of the procurement was much reduced. I will bring Nigel in to go into greater detail.

Mr Nigel Robbins (Department for the Economy): Thanks. Can everyone hear me? I noticed that I was frozen on the camera. Is the audio coming through?

The Chairperson (Mr Humphrey): We can hear you fine. Thank you.

Mr Robbins: OK. Thank you. It is a pleasure to be here. I am joined, as Mike outlined, by David Hamill, who was the project director for the previous broadband projects. It is true that the process that set out to capture all the eligible premises and the end result was slightly different. We will talk more about the Project Stratum process a little later. David, perhaps you can take the Committee through the process for the previous projects, please.

Mr David Hamill (Department for the Economy): Yes, certainly. The original initial business case for NIBIP was one of the earliest projects under the framework contracts with BDUK. In preparing the business case, one of the sources of data that we used to try to establish and estimate how much it would cost to intervene in those rural areas was one of the previous projects that the Department had run successfully: the next generation broadband project. The ballpark costs and coverage that had been provided on that project formed part of the basis for the estimated costs under NIBIP. That is why it was initially thought that we would be able to reach 117,000 premises with the funding available and, in addition to that, the level of contribution that we might expect from the supplier. When the bid was received from BT, it was considerably lower than what we had originally hoped for. That bid then went through a detailed process of benchmarking and comparisons with about 40 other projects under the same framework contracts across the rest of the UK to establish whether the costs and coverage that had been proposed by BT were actually in line with what was being seen elsewhere, and, at the end of that process, it was established that they were. So, yes, our initial estimates were considerably out, but the Department used the lessons learnt from that in taking forward the follow-up superfast rollout programme, and, in that case, we were much more conservative in estimating the costs to deliver to those rural areas and the likely contribution that we would get from the supplier. Indeed, when that went out to tender, BT exceeded our expectations because we had so much reduced what we thought we would get.

The Chairperson (Mr Humphrey): The term "considerably out" is much better than "slightly different", as was initially used in the answer. I do not think that a disparity of 79,000 households is "slightly different", with all due respect.

The Northern Ireland Audit Office report says that the Department is due to receive £14 million in clawback from BT. What is the Department doing to get the money back? Once you have got it, what will you spend it on?

Mr Brennan: The £14 million refers to clawback that applies to the two schemes: NIBIP and SRP. The actual allocation across the schemes is £6·4 million of clawback due to NIBIP and £7·6 million of clawback attributed to SRP. However, that £14 million is not available to BDUK or the Department at this stage because it is tied into when those two contracts formally end. From recollection, the NIBIP contract formally ends in March 2023 and the SRP contract formally ends in December 2024. There is an ongoing constant process of evaluation of what the gainshare clawback mechanism will be. Colleagues can elaborate, but, from recollection, on NIBIP, there was approximately £1·7 million of clawback, and that was rolled into SRP. The money is rolled forward depending on the status of the contract and is available for the Executive and BDUK to use in future investments.

The Chairperson (Mr Humphrey): So, on the £14 million, one of the contracts runs out in March 2023 and the other runs out in December 2024. Will that mean that none of the £14 million will be brought in before the beginning of 2025?

Mr Brennan: From recollection, the £14 million does not all accrue to the Executive. In total, some £9 million of it accrues to the Executive to be reinvested and the rest is for allocation to BDUK to use. Somewhere in the order of £1·7 million of that clawback has already been utilised. I will bring David in just to confirm that.

Mr Hamill: Thanks Mike. Just to clarify the question, the NIBIP contract completes at the end of March 2024. At that point, we will get back any clawback due from that project. We do not have to wait until 2024-25.

The Chairperson (Mr Humphrey): So it is March 2024, not 2023? Is that what you are saying?

Mr Hamill: Sorry. I will just check.

Mr Robbins: It is March 2023 for NIBIP.

Mr Hamill: Yes, sorry, it is March 2023 for NIBIP. At that point, we get back the clawback for NIBIP. It is not until December 2024 that we get the SRP money, but we do not wait until December 2024 to get all the money back.

The Chairperson (Mr Humphrey): How will you spend it? Are you making plans to spend what you have clarified is £9 million coming back to the Northern Ireland Executive? Are you making plans to spend that money?

Mr Hamill: The Department gets back £9 million, but it is not ring-fenced for telecoms. It will be down to the Department, possibly the Department of Finance and what the Executive want to do with that funding.

Mr Brennan: From a DFE perspective, we will do an evaluation of where we are with Project Stratum and those premises remaining uncovered and the extent to which we might seek to ask the Executive to utilise some of that to cover those individuals.

The Chairperson (Mr Humphrey): I will pause my questioning there.

Ms Hunter: I hope that you can all hear me OK. Thank you to the panel for being here this afternoon. I am based in the very rural constituency of East Derry, and I have seen the impact of poor broadband on communities here, especially with the digital divide and with education throughout the pandemic. I have a keen interest in the project, and I thank you for your comments and your clarity so far.

I have two questions. One is about the potential for overbuild, with broadband infrastructure being updated twice for the same premises. The Department has said that it is managing that risk. Can you provide a bit more clarity on that?

Mr Brennan: I will ask Nigel to address that.

Mr Robbins: I am happy to address that, Mike. Thank you for the question. Yes, there is a mechanism in the Project Stratum contract to de-scope premises if there is a risk of overbuild, for example, or if a premises is identified on the ground as derelict, which sometimes happens with data, just as there is a mechanism to bring premises into scope. I can expand on that, if you wish, in relation to a number of premises that were indicated as commercially viable in the intervention area. There is a mechanism to address those premises that we feel are eligible candidates to be de-scoped.

On the process that we have adopted, we thank the NIAO for acknowledging the best practices that the Department has stood over which have underpinned the development of Project Stratum. Those same best practices have influenced key decisions at every step of the way. They include challenging our own recommendations, casework committee review, project board scrutiny, an advisory panel through the procurement process, and Department of Finance approvals, before we even get to the BDUK state aid assurance approvals. The same mechanism has come into play here, when we have looked at premises that some may say are candidates to be de-scoped. The Department's overarching aspiration is to maximise broadband across the intervention area to eligible premises, but to do so in a way that protects the integrity of the process and preserves the ring-fenced intervention area that has been created through the state aid open market review and public consultation, and taking into consideration additional information provided to us by a major infrastructure provider.

All those things have been taken into consideration when we have made decisions or recommendations for the Department to consider, and we can stand over the target intervention area and feel that all those premises are eligible for intervention. There would be a risk to those premises — they would be commercially vulnerable rather than commercially viable — if they were de-scoped to the degree that some have suggested.

Ms Hunter: Thank you very much, Nigel. I have one further question about the homes that will not have their broadband upgraded. I think that 3,400 premises are without a proposed infrastructure solution. Is that correct? Is the Department looking into that?

Mr Robbins: At the point of the Project Stratum contract award, there were 2,500 premises that required further funding. One thing that you will hear me say a lot is that the £165 million for Project Stratum was never a budget. We have communicated that from the very beginning, when there was £150 million from the confidence-and-supply agreement. It is not a budget, but it is a terrific sum of money to address a problem and to close a broadband connectivity gap that has existed in Northern Ireland for too long. I live in Northern Ireland, I am proud to have called Northern Ireland my home for 10 years, and I would like to see Northern Ireland leading rather than playing catch-up. That sum of money helps us to close the predominantly rural digital divide that has existed here.

So some 2,500 premises were out of scope, despite the £165 million and a £48 million investment from the successful bidder. We have a headroom funding provision in the contract that allows us to utilise funding of up to £35 million to go to a £200 million cap under state aid rules, and we were always considering whether to draw on that. We certainly needed to do so for the 2,500 harder-to-reach premises. We then identified additional premises that were eligible, and that was the announcement made this week to bring 8,500 into scope. All but 37 premises that have been identified through the process will therefore benefit from gigabit-capable broadband. The 37 premises are hard to reach because they are all surrounded by bodies of water, but the successful bidder, Fibrus, will be conducting site surveys this quarter to propose a solution.

To address your question about other premises, some are captured, if you like, in a process. As I said, everything that the Department does is aligned to strict protocols in a process. The state aid process says that, if a premises can receive a qualifying broadband service from a wireless internet service provider (WISP), we must give the WISPs every opportunity to show evidence that they can provide a robust, compliant and consistent broadband service. There is a measuring period of three years after public consultation when we hold the WISPs to account, and we are coming to the end of that process. There are some premises where WISPs have claimed a service, and many of those WISPs have provided a very important service — ten years ago, that was all that premises could receive — so they have every right to show that they can provide a service. If they cannot, and if the evidence is not there at the end of the under-review process, the Department has the opportunity to consider recategorising them as eligible for intervention. We are going through that process now, and that will capture all the premises.

There are some premises that have been dropped from an infrastructure provider's commercial roll-out plans, which we may come to later. Commercial operators can change their plans; there is nothing to stop them from doing that, but it can be frustrating. The data also needs constant analysis to ensure that premises are not left behind. The Department is focused only on that unwavering good governance process and on ensuring that premises are not left behind. We very much focus on, if you like, the negative: all the premises that require solutions.

To answer your question, the under-review process will draw its conclusions later this quarter. We will have more to say about premises that have been dropped from commercial plans, perhaps a little bit later.

Ms Hunter: Thank you very much. That really helps to provide extra detail. I will come back quickly to the homes that you mentioned that still face challenges in harder-to-reach areas. Is there a way to find out where, geographically, those tough spots are located?

Mr Robbins: Yes. It is safe to say that they are all pretty tough. This is quite a unique intervention area for Project Stratum because, as we often say, it is over 97% rural. That means that it is in the Northern Ireland Statistics and Research Agency (NISRA) band H category, which is villages and hamlets of populations of 1,000 or less and sparsely populated countryside. As we all know, Northern Ireland is different to England, certainly, and to other parts of the UK in that regard. Northern Ireland has the longest telephone line links and four times more telegraph poles per capita for that reason of sparsely populated premises. That means that, in more than 97% of the intervention area, two thirds of premises cannot access speeds of 15 Mbps, let alone 30, and many are on the lower end of that.

As Mike stated, broadband is a utility. In fact, many members of the team are in the same position. We all are citizens of Northern Ireland, and everybody deserves the very best broadband service regardless of where they live. There are some harder-to-reach premises among the 2,500 that we have brought into scope. The contractor requires a little more time to deploy to them, because new infrastructure is required to get to them. The contractor will utilise existing infrastructure where possible, for example by renting from BT. There is a very good relationship, we are told, between the BT physical infrastructure access (PIA) team and the Fibrus deployment team, so that is a positive. We all want the same thing. It is also terrific for Northern Ireland that we have BT, Virgin and Fibrus competing against one another. We have a dynamic here that has put Northern Ireland ahead of the rest of the UK, with 77% of premises being gigabit-capable. Stratum has raised that to 87%, but that is not the case in rural Northern Ireland, where 30% of premises still cannot access superfast speeds. That is what Project Stratum seeks to address.

Ms Hunter: Thank you.

Mr Trevor Forsythe (Department for the Economy): May I come back to your query about overbuild? We have been looking at that carefully. We have asked the supplier whether it has come across any evidence of overbuild in its network deployments to date.

It has reported to us that it has not seen any evidence that widespread infrastructure is being deployed in the areas in which it currently operates. We will keep an eye on that, however, to reassure ourselves that that is the case.

As Nigel said, this is a rural, including the very remote band H, but it also includes some small hamlets and clachans, as they might have been called. We should bear in mind what some people might regard as rural. We are looking at the overbuild issue, however, to make sure that we can manage it.

Mr Robbins: Finally, if I may, to touch on that, when we were looking at which additional premises to bring into scope, we did not include those premises for which we have received recent data to indicate that a commercial solution is available, so something like 2,000 premises were not considered. Of all the 8,500 premises that we have added, fewer than 100 are not band H. That is the typical profile of those premises.

Three years ago, we established a forum for other drivers of public broadband intervention, which included the teams behind the development of the city deals, the Full Fibre Northern Ireland Consortium and our colleagues at DCMS and Ofcom. The purpose of establishing that forum is that it is in our interests to share information where we can and where the data allows — this is not a poker game, after all — and to try to avoid any overlapping or gaps. As a result of that, we de-scoped some premises from Project Stratum that the Full Fibre NI team was going to pick up earlier. We are very much looking at opportunities where we can de-scope and make savings for the public purse and, at the same time, very rigorously ensure that the process is protected and that premises are not left behind as a result of this much-needed intervention.

The Chairperson (Mr Humphrey): OK, thank you. Quite a bit of mention has been made of premises. I have a question for Mr Brennan, before I bring Mr Muir in. How did the Department go about identifying premises for improvement under NIBIP and SRP phase 2? How did you work to ensure that the only targeting of upgrades was at premises where it was not commercially viable?

Mr Brennan: OK, Chair. I will bring in David in a second. NIBIP and SRP were a progression from the previous next generation broadband scheme that the Department ran in 2009. We then moved on to the BDUK national framework and adhered to those framework guidelines. David, do you want to give greater insight into the definitions that we had to adhere to, such as its interrelationship with what was eligible for state aid?

Mr Hamill: Yes, certainly. BDUK put in place an open market review (OMR) process that all the local authorities had to carry out, whereby we determined all identifiable postcodes across Northern Ireland and how many premises were in every postcode. We then did a consultation with all the key industry providers, asking them to provide us with detail of where they currently provide access at superfast speeds and of any plans to improve their infrastructure over the next three years. We took that information and carried out a public consultation to verify it. That was then reviewed along with BDUK's intelligence on the market. Out of that came details of premises that were eligible for improvement. The same process was done for both NIBIP and SRP. An open market review process was also done for Project Stratum.

The Chairperson (Mr Humphrey): Thank you very much.

Mr Muir: Thank you to everyone for joining us in this inquiry. This is an important issue for me, because it is about the use of public funds and ensuring that value for money has been procured. I have a couple of questions about that. Public money has been used to support the roll-out of broadband, but surely that could have been done on a commercial basis, meaning that public funds may not have been required. That is the core of this inquiry and of our consideration today.

It is important, first, to get an understanding from you of what lessons have been learned in relation to the issues raised by the Audit Office report. I would like to hear a bit more about that.

My other question relates to the formula that was used for Project Stratum. I want to know why the Department used such a basic formula to score that. I consider it to be basic and would have expected to see a procurement process that was a bit more complicated to ensure that there was a better outcome and that public funds were better safeguarded. I want the Department for the Economy to tell me why such a basic formula was used.

My final question relates to the Land and Property Services (LPS) database. A significant number of properties was missed. What due diligence was undertaken in relation to the LPS database?

Mr Brennan: You have flagged a significant number of issues there. I will deal with them as quickly as I can, and then maybe invite a few colleagues in.

On the first issue of value for money and intervening in cases where, as you seem to suggest, the private sector may have wanted to be involved anyway — in other words, involving it where there was no market failure — the first point is that there is a radical difference between suppliers making commercial decisions in an urban environment and doing so in a rural environment. The private sector had stepped up to the plate, as it were, in the roll-out of broadband to commercial settings in Northern Ireland. That clearly suggests that there was no market failure in providing in that environment.

As we have touched on already, the costs to the private sector to provide acceptable standards of broadband in the rural environment are significantly higher. The imperative was to make broadband available to the rural community in Northern Ireland. We went through a process, which was a UK-wide process, for the NIBIP and SRP to roll out broadband in the standard way. BT was the only company on that framework during that period. Actually, it was not the only one at the start: there were other providers, such as Fujitsu, that decided to remove themselves from the framework. BT was left. I take comfort from the fact that, in the 11 cases, I think, in total, where local bodies in GB decided to provide broadband outwith the BDUK framework, all 11 procurement processes ended up appointing BT. BT was awarded those contracts through an entirely separate procurement process. That gives me one degree of comfort.

The other degree of comfort that I have with regard to value for money is that a significant clawback gainshare mechanism was built into the contracts that allowed the Department to undertake a scrutiny exercise to determine, for example, whether costs were eligible or a degree of supernormal profit had been earned by BT. Therefore, if the costs that were incurred by BT were lower than expected, or the take-up by premises was higher than expected, the taxpayer had a mechanism by which to recoup some of the investment.

The other issue with regard to value for money, particularly when we got into Project Stratum, was that a significant checking mechanism was built in by BDUK and the Department for verification and vouching of payments. Maybe we could get into the detail of how that works. From the taxpayer's perspective, how costs are actually approved is reassuring. BT gets paid only for what it actually delivers. As I said, significantly, those investments are in extremely rural areas.

You referred to a "basic formula" that determined how the tender was awarded. I am not sure that I would accept the word "basic", because the Audit Office report does not fully detail the actual process that was used for the weighting and scoring of the two tenders that came in under Project Stratum. I think that one table is presented in appendix 5 at the end of the Project Stratum report. I will bring Nigel in to give you significantly more insight into the methodology that was employed and how the awarding of the contract to Fibrus was done across a number of different metrics, not just one. The idea that marks could go from 30% down to 29%, just on the basis of coverage making a single adjustment, masks the other issues that were brought into play in the weighting and scoring for the award.

Mr Muir's final point was about the LPS database. Obviously, the Pointer database is used by LPS as a rates assessment and payment mechanism. It was never constructed for this particular purpose. That is something that I came across in relation to COVID grant payments, as well. The LPS database is a very significant and useful database for the wider public sector, but it is constantly evolving. We take the evidence as it evolves, and that leads us to readjust where we are. As Nigel said, as part of that readjustment process, somewhere in the order of 6,000 properties out of the 8,500 referred to in the announcement earlier this week were added as a consequence of reassessments of what the LPS Pointer database was suggesting to us.

I will go back to the point about the basic formula, because we need to shed a bit more light on that in order to give you some comfort. Nigel, do you want to say why the calculations were much more complex?

Mr Robbins: Yes. If I may, Mike, I will touch on Mr Muir's first question: he asked three very important questions. The issue about whether the Department is using public funding sensibly is raised a lot. That is understandable and appropriate, and it is referenced in the NIAO report particularly in relation to premises that BT indicated were commercially viable and which, we could see, were in the intervention area. It is important that the Committee is aware of the process.

Again, underpinned by governance that led to the Department applying the "line in the sand" principle — that is key, and it is important that everybody is aware of it — and as the NIAO report establishes, we had a situation where the invitation to tender (ITT) process — the second stage of the procurement — was launched, we had a competitive bid process, and we were delighted with the outcome. That followed the market warm-up exercise, where 13 telecommunications entities were interested in Project Stratum, and we had a selection and questionnaire process, all on the back of the open market review and public consultation. At that point, we could see that the intervention area was 97,000 premises strong.

Based on the outline business case (OBC), we knew that more funding would be needed to get to all of them. The OBC indicated that a combination of £165 million of public funding and the supplier contribution would allow us to get to around 74,000 premises, which is, if you like, an important benchmark.

Going into the ITT process, BT conducted a national data refresh exercise. We were made aware of new data, and BT indicated that a number of premises could be considered to be removed from Project Stratum on the basis that more accurate speed-hub data could indicate that premises were compliant or that, in many cases, premises were part of the commercial, full-fibre roll-out plans for BT.

We engaged with BDUK at that time and removed 18,000 premises from the intervention area. That was the right thing to do, based on the premises data that BT provided. Those premises were removed, and that is why the intervention area was reduced to 79,000 premises. That was in January 2020, right at the beginning of the ITT process. Not all of the bidders were happy with us doing that, but we felt that it was important that all bidders had the same information on a level playing field. We extended the ITT process to take that into consideration. A new date for the ITT deadline for the submission of bids was set, which was 5 May — 16 weeks after that OMR data information was provided by BT. We then had an evaluation process. I will come to that, because it is fair to say that it was a very complex evaluation process. It would be helpful to the Committee if I do not ramble on for too long, but I would like to take you through that process. Coverage is a key component, but 24 detailed criteria were assessed.

However, I go back to the data issue. On 5 May, we received bids, and an evaluation was conducted over the lockdown summer of 2020. All the evaluation panel members were in blissful isolation, evaluating over an eight- to 10-week period before consensus. That was remote and was managed by the Central Procurement Directorate (CPD). Letters of intent were issued on 10 September. Ten days later, BT provided information to the Department that indicated that 20% of the intervention area was commercially viable. The information was provided in a spreadsheet with "fibre to the premises" (FTTP) alongside premises, and there was a letter, which was caveated to a certain degree, that stated that "plans are subject to change". When we saw that 16,000 premises overlapped with the intervention area, we engaged with a number of key stakeholders.

Getting back to the governance process, between September and October 2020, we engaged with BDUK, the public spending directorate, the Economy Committee, Ofcom, a member of the Cabinet Office and the National Competence Centre in BDUK, before drawing on the line in the sand principle, which is set out in paragraph 56 of the national broadband scheme 2016. That indicates that, if premises data is provided at a late stage after public consultation, contracting authorities or local bodies can elect not to take that data into consideration if they feel that there is no credibility or not enough credibility associated with it and premises could be vulnerable. That is a state aid mechanism to protect and preserve the integrity of the process and the ring-fenced intervention area.

In our view, it was unusual practice. BT behaved impeccably throughout the process, except in that one area, where it was unusual to state that 20% of an intervention area that comprised premises that a supplier had sought public funding to get to were now commercially viable. It may be the case that funds that had been earmarked for Project Stratum were being diverted, or there may have been other motivations. It is not for us to say. If the roles had been reversed, BT had been successful and the other party was doing the same thing, we would have applied the line in the sand principle. BT would be familiar with that, as it applied it 10 times in public interventions across the UK.

That is an important point. For those 16,000 premises — it could be 10,000 or 20,000 premises — the same process was applied. Everybody with whom we engaged accepted the Department's rationale, and we have moved on from that. BDUK's chief executive also engaged with BT's UK head of infrastructure, in February last year, and was surprised that BT was resisting our recommendation to apply the line in the sand principle and recommended that it did so. In fairness to it, BT accepted that line in the sand principle, because it is there for those reasons.

We have concerns about some premises that were dropped from BT's commercial roll-out plans, including some that we removed during the data refresh exercise. I go back to the points about following the process, playing with a straight bat, and doing everything to try to protect the intervention area and to ensure that nobody is left behind. Sorry, that was a long answer, but it addresses the point that is referenced in the report.

If you will bear with me, on the scoring mechanism, coverage is, of course, key. We want to maximise coverage across the intervention area. We say, "maximise". There was not a starting point of 100% coverage, because £165 million was not enough budget. We knew that coverage would potentially get to about 74,000 premises. We were delighted with the excellent bids from BT and Fibrus. They were highly credible bids, but there was quite a gap in the overall scoring.

How did we get to that gap? It was by assessing and evaluating 24 detailed criteria. The short answer is that the Department had "absolute confidence" — to use the ITT terminology — that Fibrus would get to 76,000 premises with a viable and complaint FTTP solution, applying all the other criteria. The longer answer is that coverage was weighted much higher than any other criteria. However, taking into consideration previous NIAO reports, we also weighted half the coverage weighting, financial model, and commercial sustainability and viability. Those were very important considerations. The other criteria included environmental impact, wholesale access pricing and stakeholder engagement. An enormous amount of detail was required. Of course, we do not compare bids when we evaluate; we evaluate what is in front of us, based on evidence and detailed responses to the criteria-based questions.

Of the 24 criteria, the two bidders scored equally in 14 areas, where they were neck and neck — 11 of those were judgement-based scores and three were pass-fail — and it was a dead heat. BT scored higher in two areas, one being coverage, as it proposed 79,000 premises, while Fibrus proposed fewer. Fibrus scored higher on eight key criteria, including financial model, commercial sustainability and viability, and environmental impact.

Those are very important criteria, and they combine to form the whole. When automated scoring is applied to coverage, it has to be supported by all the other information that will give the Department enough confidence to say, "This public money can be used sensibly, because this successful bidder will deliver what it said on the tin". That was the outcome. While we accept that coverage being subject to an automated score may look as though something as important as coverage is being simplified, it is underpinned by all those detailed criteria that would give the Department confidence, or less confidence, that it would be delivered.

The scores were 98·88 for Fibrus and 85·44 for BT —

The Chairperson (Mr Humphrey): Mr Robbins, we need answers to be much more succinct than they have been hitherto. I am not trying to guillotine discussion, but we need answers to be succinct. A number of members are still waiting to ask questions, and then we have a report to conclude. I ask you to please be more succinct in your answers. Thank you.

Mr Robbins: I will, Chair. Some of the answers are complex, so forgive me for going on, but I certainly take that on board and will be as brief as possible.

Mr Muir: I have one follow-up question. There is, as has been outlined, a clawback mechanism in relation to BT. How is that operating, and is the Department confident that it will receive the funds?

Mr Brennan: We have touched on that already. A figure of £14 million from the clawback mechanism is referred to in the Audit Office report: £6·5 million from NIBIP and £7·5 million due under SRP. They are constantly reviewed. Of the £14 million, something in the order of £9 million is due to the Executive, with the balance going to BDUK in DCMS. We hope to ask the Executive for it to be invested in telecoms in Northern Ireland. We will not have the final pictures on that until we get to end points for the two contracts, which, as I said earlier, are March 2023 and December 2024.

Mr Muir: OK. Thank you.

The Chairperson (Mr Humphrey): Does that conclude your questioning, Mr Muir?

Mr Muir: As you said, Chair, there were very long answers to the questions, and concerns resulting from the Audit Office report remain about value for money. Responses have been given, but the fact is that a contract was let to Fibrus, and then BT said that it was going to cover a significant number of properties. The public will have concerns about that and the fact that money was spent. That remains a concern.

Mr Forsythe: I will add to that. Prior to us beginning the process, the CBI reported on whether investment was likely to take place. Its conclusion was that, without some sort of public investment, it was unlikely that the commercial operators would move into some areas of Northern Ireland. The CBI produced that report just after the confidence-and-supply funding was announced.

Mr Muir: OK.

The Chairperson (Mr Humphrey): Mr Muir, as I said, I am not guillotining discussion, and, if we need to have further sessions, discussions or, indeed, correspondence, we will discuss that later in the meeting.

Mr Irwin: The basic scoring mechanism has been touched on. I have been in business all my life. If I go out to tender for something, I expect the tender to be like-for-like. It seems strange that the two tenders were not like-for-like on the number of houses covered and that one proposed to cover 2,500 fewer houses at a cost of approximately £25 million. It just seems a bit strange that you can use a basic scoring matrix on something when it is not like-for-like.

Mr Brennan: I am not sure that the assessment was on a like-for-like basis.

Nigel set out that there were 24 aspects to the scoring. Coverage was one dimension of that. Nigel may want to give some insight on that.

Mr Irwin: The difference in coverage alone involved £25 million. Is that right?

Mr Brennan: Nigel, will you explain the scoring differential between BT and Fibrus, how it was weighted and the fact that the £25 million refers to getting to the overall wider coverage.

Mr Robbins: Yes. The weighting took coverage into consideration. I will not repeat myself about the scoring process, but, in brief, it is fair to say that, based on the detailed information and evidence provided by the successful bidder, the Department had 100% confidence that that bidder would deliver to the stated 76,000 premises. Clearly, it had more confidence in that than in the other bidder's proposal to deliver to 79,000. That is not to say that it was not a credible bid from BT; a score of 85 is very high. A score of 98 is considerably high, but coverage of 97%, which was the successful bidder's proposal, is high, and it reached more premises than we were expecting with the available funding.

Premises have been removed from BT's commercial plans, which equates to around 4,000. That is more than the 2,500 premises that we need to fund as a result of the Fibrus proposal. That is another matter, which we may or may not come back to later. I do not know whether Trevor or others have anything to add, so I do not get carried away.

Mr Forsythe: The criteria were clearly set out in the invitation to tender documentation. Twenty-four areas were examined and assessed; coverage was just one of those. The state aid rules that applied to the decision outlined the areas that had to be assessed and gave some indication of the weightings that could be applied. During the design of the procurement exercise, we spent time looking at how those should be weighted and ended up with what was published in the ITT.

There was a difference in the like-for-like scoring that we looked across. Fibrus scored higher than BT in other areas of its bid. We based it on the information that was presented to the evaluation team. We are doing a like-for-like comparison based on the information in the bid documentation that was submitted to the evaluation panel during the procurement process.

Mr Irwin: Thank you. I have one further question. You may not be able to answer it. In the contract that was signed, is there any scope for add-ons? I live in a rural constituency, and we have one situation in which there is a row of houses getting Project Stratum. One house has been newly built in the past couple of years. The cable goes by the front door, but Fibrus will not connect the extra house. The cable goes past the front door, but it does not seem to be willing to do the connection. There may not be anything in the contract that allows anything to be added to what is in place. Is that right?

Mr Forsythe: Fibrus has said to us that it is continuing to keep that situation under review. At the moment, it is very mindful that it was initially contracted to deliver a service to 76,000 premises. That has been added to with the extra 8,500 premises. Fibrus is very focused on committing its resources to those premises at this stage. That is the focus of its efforts. However, it is looking at policies for what it terms "new builds" as well at this stage. We expect to see suggestions from Fibrus in the next few months on how that might be managed.

Mr Irwin: OK. Thank you. Hopefully, that will help.

Mr Beggs: Thank you for your evidence so far. I am curious, Mr Brennan: how have we come to a situation in which there is a public subsidy for the introduction of broadband, when one of the telecoms suppliers has indicated that a significant part of the area in which it intends to install it is commercially sustainable on its own without public subsidy?

Mr Brennan: I cannot comment on what has driven BT to take those commercial decisions. As the guys said, when the bids were being constructed, BT previously indicated that they had to be included in the target intervention area, because they were not commercially viable. That is an internal BT discussion, and it would need to answer for why it has taken that decision.

As I said, from a taxpayer perspective, we have built in quite detailed verification systems. We have also built in systems to scrutinise all costs. Open book principles apply, and clawback mechanisms are built in. We can look at and benchmark the costs that Fibrus incurs, and we can take BDUK's expert advice on whether those are appropriate. We therefore have all sorts of challenge mechanisms built into the contract to give assurance to the taxpayer that value for money is being delivered in areas that, as we said at the start, are rural and less likely to be commercially viable. BT itself said that they were not viable when it put its bids in. As I say, that question is really an internal one for BT to answer.

Mr Beggs: There are implications for the public purse if two systems get built: we will have spent public money unnecessarily. Do you accept that?

Mr Brennan: Theoretically, if two parallel systems were built, there would be a question over why BT would want to do that if the taxpayer is providing infrastructure on a particular street, lane or whatever. I am not sure that it will exist or be a real problem that we have to address.

Mr Beggs: I understand that, if a private company decides to install fibre on its own — I am thinking of Virgin, which has done extensive installation work in my constituency — it gets a certain number of years in order to try to recoup that investment. Is that correct?

Mr Brennan: I am not sure —

Mr Beggs: It would not face competition from others.

Mr Brennan: I am not sure of the nature of the Virgin contracts. My presumption is that the provider of the wholesale network has a contract, but there are terms that have to apply for the provision of retail broadband so that anyone can use it. I will bring Nigel in on that one.

Mr Robbins: To go back to the 1,600 premises, in some ways, it is tantamount to the bidder for a bridge contract being unsuccessful and then proposing to build 20% of the bridge on the easy side of a river. I am not being flippant. That is because those 1,600 premises were required in order to build the viable financial model that formed the basis of BT's bid and Fibrus's successful bid. If you removed them, it would be more difficult to get to the harder-to-reach premises and would be more expensive, because they are widely distributed and dispersed across the intervention area.

We conducted an impact assessment. The issue could have delayed procurement. We might have even had to reset and repeat the entire procurement, because the commercial viability of the awarded contract would have been brought into question, and there would have been less contribution from the successful bidder and many stranded premises.

Again, getting back to the process and the governance to protect the integrity of the awarded contract and the process, if those premises were commercially viable, it would have been appropriate for BT to advise us of that 16 weeks prior to the submission of its bid or during the procurement process, when a channel of communication was open through CPD and eTendersNI in the form of clarification questions. In defence of BT, it is a commercial operator and may have decided that it wanted to go to those premises to make the landscape more interesting. Also in defence of BT, the OMR was conducted a long time ago. There will be more agile OMRs in the future. In January 2020, we accepted updated information as part of a data refresh exercise. The information provided in September was out of sync with that and was unusual.

Mr Beggs: To avoid all this, if the contract had been protected in the conventional manner that other new networks were protected, would that not have solved the problem?

Mr Brennan: I am not sure what you mean by that, Mr Beggs. How would a different contract —

Mr Beggs: You are guaranteed use of a network without competition for a certain period in order to recoup costs.

Mr Forsythe: In this contract, there is not that protection for exclusive use of the network. One of the conditions of deploying this network is that other network operators or broadband service providers can avail themselves of it, whether or not the service on them is being provisioned by Fibrus as a wholesale operator. There is also an option that passive infrastructure should be made available to other operators, if they want that, on a fair and equitable basis, which is one of the requirements of the state aid decision.

Mr Brennan: Another important aspect to bear in mind is that the contract was awarded to Fibrus in November 2020, but the state aid approval regime lapsed, I think, the following month — December — so that would have had significant implications for procurement and delivery. I will bring in our BDUK colleagues. Billy or Andrew, do you want to say something about where you think the procurement would have been if we had missed the state aid milestone?

Mr Andrew Field (Department for Digital, Culture, Media and Sport): I will come back on that one, Mike. One of the roles that BDUK fulfils is to act as the national competence centre for state aid under the EU regime and now under the UK subsidy control regime. At the end of 2020, we were moving out of the EU arrangements and into the new UK arrangements. The state aid decision to which Nigel referred was made in 2016 by the European Commission. That is referred to as the national broadband scheme 2016. That effectively delegated responsibility from the European Commission to BDUK to confirm compliance with the EU state aid regime, and that lapsed at the end of 2020, which was coincident with exit from the EU. Subsequently, there was a bit of a gap, which has now been filled by the new UK subsidy control arrangements, which, in reality, are very similar to the old EU regime. If the Stratum contract had not been agreed in November 2020, there would almost certainly have been a delay in early 2021 before the new requirements and regime became clear.

Mr Beggs: Thanks for the information. I am astonished that BT will be able to access this fibre and yet it is considering installing its own fibre. That seems very strange. I assumed that BT would not have been able to get access to it.

On a different area, 6,000 premises were not on the LPS database. Presumably, those are all premises that were not paying any rates.

Mr Forsythe: No, the issue relates to how the premises had been classified on the Pointer data set. We had to go back and use rating information to try to merge those and use that as part of the evidence to include the additional 6,000 premises.

Mr Beggs: Can you please explain how they have been classified? I do not understand that, if they are not paying rates.

Mr Forsythe: No, we used rating information to include those premises. They were on Pointer. They had been classified as not built or approved, and we then used the rating information to confirm whether that was the case. That is how those additional premises came in under the project.

Mr Beggs: Is that vacant sites or derelict property?

Mr Forsythe: All we can say is that the Pointer data set stated that they were not built or approved but that there was something on that site.

Mr Beggs: Right. OK.

The Chairperson (Mr Humphrey): It seems an extremely high figure to have 6,000 premises across the piece in that category. Do you agree, Mr Forsythe?

Mr Forsythe: We were surprised at the numbers, and we have been working with LPS to understand how the Pointer data set gets updated and refreshed and how to improve the quality of the information that we extract from the data set.

Mr Beggs: I will turn to the areas that have been designated for Project Stratum. How did you select those areas as qualifying with less than 30 Mbps? I ask that as someone who has less than 10 Mbps and does not appear to be included in the area.

Mr Forsythe: The process is set out in the national broadband scheme 2016. The first part was to consult the infrastructure providers that we were aware operated in the area. We went to the likes of Openreach, Virgin Media and some of the wireless infrastructure providers to ask whether they could provide information on their network coverage. We combined that with the property information and involved people from the Ordnance Survey to help to match those records. That gave us the initial list of postcodes where we believed that intervention was required. We then undertook a public open market review. We published that information on our website and invited responses. We liaised with councils to ask whether the lists seemed appropriate. We invited members of the public to inform us of any premises that they thought should be included. We had an advertising campaign across the Province, and we wrote to key stakeholders to inform them of the process and gave them opportunities to inform us if we had omitted anything or any areas or premises that should be included.

Mr Beggs: OK. I obviously missed all that. With regard to basic broadband of more than 2 Mbps, Northern Ireland has only 98% whereas GB has 100%. At the completion of this project, what percentage of properties in Northern Ireland will remain untouched by Project Stratum or without even basic broadband?

Mr Forsythe: A whole raft of technologies is now coming into the market. At the end of this process, we expect that a basic broadband service should be available to most premises across Northern Ireland. It will also depend on whether we can identify retrospective premises. We will be reliant on information from the communications market review and the work of Ofcom in informing us of the availability and roll-out of services. NIBIP is focused on providing superfast services to premises in Northern Ireland.

Mr Beggs: What is your target? I have not heard a figure. Will it be 100%?

Mr Forsythe: I will give some insight into how transformative Stratum and the previous interventions have been, Mr Beggs. The Ofcom figures from September last year show that, for example, gigabit-capable premises in Northern Ireland is 73%. The UK average is only 40%: England is 38%, Scotland is 47% and Wales is 30%. For access to full fibre, we are three times higher than any other region at 67%; the UK average is only 24%.

Mr Beggs: I am asking a question about the lower end. Will we achieve 100%? Will all those properties with very poor service have an improved service? What is the target figure, or is there no target?

Mr Forsythe: I am trying to understand what you mean by "improved service".

Mr Beggs: At present, 98% of properties have at least 2 Mbps in Northern Ireland. In GB, it is 100%. What is the target figure at the completion of this project?

Mr Forsythe: All premises that want a basic broadband service can avail themselves of the universal service obligation from Ofcom and ask for that service to be deployed by the designated infrastructure provider, which happens to be Openreach in Northern Ireland.

Mr Beggs: You cannot say that it will be 99%, and you are not saying that it will be 100%.

Prior to Project Stratum, there were several isolated rural communities. For many, the multinational telecom companies chose not to invest. There was no broadband and frequently there was no mobile phone signal. They were very isolated, and small independent telecom entrepreneurs installed Wi-Fi broadband using innovative methods very efficiently. That was subsidised by BDUK public money.

My question to BDUK is: do you accept that this is likely to write off all that previous public investment? As I understand it, the process involved to protect an area or customers meant that you had to publicly disclose who they were. If you did that, and there were clusters, BT would subsequently target those areas. Given the amount of administration involved and the risk of showing your hand to your competitors, many of those Wi-Fi networks, funded by the public purse, will ultimately be written off.

Mr Field: I will respond to that. A process of upgrading technology is going on across all parts of the UK as we go to superfast and then gigabit coverage. Stratum is an example of a project that is providing both superfast and gigabit coverage. The process that others have described asks the suppliers to confirm their current and planned coverage so that the publicly funded interventions go only to the areas where there is no coverage. It is up to the suppliers whether they respond to those requests for information. When the information is provided and comes into the Department, it is published in a public consultation. It will say which premises have, or are planned to get, coverage and which have not. It will not say which providers have the plans, but an operator could work out which providers operate in which parts of the country.

If we are to have a process by which we identify which premises need public intervention to get coverage, there has to be transparency of that coverage for premises. That requires the operators to take part. If they do not take part, there is no information from them that we can use. That is the way the process works.

The other part, obviously, is that, if previously funded projects, by whatever route, are providing only basic level broadband services while the ambition is to provide either superfast or gigabit, there is now the potential that those interventions will overbuild those earlier networks because the earlier networks no longer provide the speed capability that we now expect.

Mr Beggs: Just to be clear: I am talking about networks with a capability of greater than 30 Mbps.

Mr Field: Where a network operator providing more than 30 Mbps responded to the open market review and public consultation process, those premises should have been excluded from the intervention area.

Mr Beggs: My question to Mike Brennan is: do you accept that, if those networks responded, highlighting who their customers were, those customers would then be likely to be targeted by BT, should Stratum not cover those areas?

Mr Brennan: I am not sure that I am in a position to comment on the extent to which BT adopts a predatory practice like that. That is a question for BT to take a view on.

I will follow on from Andrew's comment. As you move towards seeking to deliver gigabit coverage throughout Northern Ireland, you would expect providers with a lower standard to find themselves displaced from that market.

Mr Beggs: I fully understand that things need to change. Given, however, that public investment has gone in, you would expect that funding to have a window of opportunity within which to provide before being overwritten.

That is fine, thank you.

The Chairperson (Mr Humphrey): I am not aware of any other member who wishes to ask a question. No indications have come to me from within the room or online. At this point —

Mr Boylan: Chair, if I could —

The Chairperson (Mr Humphrey): Yes?

Mr Boylan: Sorry, Chair. I have been using that many devices that I have been in and out of the meeting and missed some of the conversation. I did not have the ability to use the raise hand function on the system. May I ask a few questions?

The Chairperson (Mr Humphrey): Yes, that is fine.

Mr Boylan: Thank you, Chair. As an MLA, I have been dealing with this for a long time and have asked a number of questions of the Department in relation to it. Mike, can you go right back and identify where you got the database information from? There seems to have been an issue with how data was collected in the first place. I heard that it was collected through the rating system. Can you tell me where you got the information from in the first place? That leads into the question of why the procurement process was started when we did not have all the data. I am well aware of all the issues with band H and NISRA, but I want an overview. Unfortunately, because I have been in and out of the meeting, I have missed some of the answers. Could you quickly identify for me where you gathered all the data in relation to this process?

Mr Brennan: This is a very quick summary of the process. At the start of the consultation process, the various internet providers were consulted to establish the area that was eligible for coverage, given that the need was to roll out into largely rural areas of Northern Ireland. The target intervention area was largely developed by the likes of Openreach and Fibrus. The key players determined the target intervention area. At a later stage, the work of LPS, for example, was, essentially, to cross-reference and feed into that to establish the identity of individual properties. That is a very simplistic, high-level understanding of it.

Mr Boylan: I appreciate that. So, that was the starting point. I ask in the context of having asked a question of the Department and received the answer that it was not satisfactory. We now find out that we did not have all that information. Is that a fair assessment, on reflection? Band H settlements of 1,000 people or fewer were more important. Most members wanted to see intervention in the open countryside. As soon as the programme was rolled out, my office was contacted on a number of occasions by people who said that a fibre artery was being put down on their road but that premises on that road were still going to miss out. That should not have been the premise of the programme in the first place.

Mr Brennan: The premise of the programme was to reach places that were not commercially viable for the likes of BT to go to. As we said earlier, we were trying to focus on townlands and hamlets with fewer than 1,000 people. It is fair to say that, over the last couple of years, it has emerged that there are incredibly isolated properties. I think that it was Nigel who mentioned that there are properties on islands, for example. Things like that emerged that will not have been identified. In summary, yes, over the last couple of years, properties have emerged that probably were not embraced when the original target intervention area was drawn up as part of the procurement process. Is that a fair summary, Trevor?

Mr Forsythe: Yes. We had to look at which property records/information was available to us. We looked at systems that were being published. The data set that seemed to have what we were seeking is called Pointer. We spoke to colleagues in Ordnance Survey of Northern Ireland (OSNI) and told them that this is what we wanted to do. It was decided that we could use that data set. We believed that it was accurate, up to date and regularly refreshed. What has come to light is that some of the records contain anomalies. Where buildings have been designated as under construction or not approved, it turns out that they are built and approved. That may also be to do with how the whole database is updated. It relies on a wide number of stakeholders to update information, including councils, Building Control and other stakeholders. It is a massive data set to keep up to date.

As I mentioned previously, where it has emerged that fibres are going down roads on which there are premises, we have been looking at what those situations are. We are working with Fibrus to see whether it can come up with some sort of policy to allow it to connect those properties. We expect that it will be able to give us some options in the next few months. I mentioned that earlier in this session.

Mr Boylan: Trevor, have you worked out a percentage of premises across the board that will not have a fibre solution, meaning that they may need some kind of satellite system or different technology?

Mr Forsythe: No, that is, perhaps, our next project, which is Project Gigabit. That is the UK objective to provide fibre to some 85% to 95% of premises across the UK. That is what we are looking at next. At the moment, we do not have precise figures. When you talk about fibre solutions, that means that you need to consider the large number of premises that are currently getting services through fibre to the cabinet (FTTC). We do not yet have those details.

Mr Boylan: I have two final points. I have an answer from the Department that says that Newry and Armagh, for example, will reach the 97% target when this programme is finished. Can you stand over that? That is what the Department said about Newry and Armagh. I cannot speak for any other constituencies, but I have been following the Fibrus programme and the broadband programme for a number of years.

Mr Forsythe: With that number, that is next generation access, so that will be a service of 30 Mbps or better, and the technology used to deliver that could be could be a mix of fibre to the home, fibre to the cabinet or even services provided over cable TV, such as the service provided by Virgin.

Mr Boylan: Can you give us the detail on the original funding package and what it will end up as? What was the starting figure for the programme, and what will the final figure be?

Mr Forsythe: I do not have the initial figure to hand. We will need to come back to you on that.

Mr Brennan: Are you looking for the costs of Stratum in total?

Mr Boylan: Yes, the overall costs.

Mr Brennan: Nigel can correct me on this, but I think that, up to last week, the total project cost was £211 million. Then, earlier this week, we announced the additional £34 million to cover the extra 8,500 properties. That brings the total to just under £250 million. Approximately £46 million to £47 million of that comes from Fibrus, and the rest of it is taxpayer contributions.

Mr Robbins: That is correct, Mike. There is £197 million in public funding. That is the contract value under the extension, and, as Mike said, the complete costs are just under £250 million — around £249 million. Those are the network build costs, including the supplier contribution. The cost per premises of superfast intervention area premises is £2,330, and that is considerably lower than similar interventions. Interventions are all different. In Ireland, under the national broadband plan, the costs are about £4,700. With the most rural lot in Scotland, the costs are close to £5,000. The value for money (VFM) considerations and the investment in future connectivity for Northern Ireland are sound.

Mr Boylan: Are you saying that you can stand over it being value for money, Nigel, yes?

Mr Robbins: Absolutely. I believe that, on value for money, the NIAO report says that it is too early to draw conclusions but comments on the coverage, which we have tried to redress. Over the longer term, we believe that the value for money of Project Stratum will most certainly stand the test of time. When you look at the cost-benefit analysis of the superfast programme across the UK, you can see, for example, that for every pound of public subsidy, the longer-term value will be between £3·60 and £5·10. That is already starting to pay for itself. In addition, the benefits to e-learning, e-education and health and well-being, as well as the economic benefits to businesses, will certainly be part of our longer-term value for money assessment, along with the VFM assessments that BDUK conducts when looking at the bids. It conducts a bid comparison report, which will combine to form the VFM view in the longer term.

Mr Boylan: OK. Thank you.

The Chairperson (Mr Humphrey): I will finish, if I may, with a couple of quick questions to Mr Brennan, as the accounting officer in the Department. At the time you were drawing up the target intervention area, did you understand that there were likely to be deficiencies in the LPS database? With hindsight, should you have done more due diligence on the information to ensure that the figures were more accurate?

Mr Brennan: My understanding is that the issues around the LPS Pointer system were not identified and were not understood, and our awareness of them has evolved over only the last two or three years. As I said earlier, we also encountered those issues in a number of COVID grant schemes that used LPS Pointer. It was used for a rates system, and all public bodies now use it for something that was never intended. It is getting more accurate and getting better, and a lot of DOF investment has gone into it. The quick answer to your question is that, no, we were not aware of those issues. The best that we could have hoped for was that the likes of BT and the companies bidding into it had a perfect understanding of what should be in the target intervention area, and that has not come to pass. We know that anomalies exist, and that is what we are trying to address.

The Chairperson (Mr Humphrey): You say that you were not aware of it. When did you become aware that there were such wide-scale inaccuracies in the LPS database?

Mr Robbins: We became aware of it after the contract award. I go back to the basis that helped to form the intervention area. There was an open market review, as we have established, in the summer of 2018. Prior to that, the team had engaged with LPS and an LPS specialist. The terms of reference were drawn up, and it was identified that a premises with a category of built and approved would capture premises as accurately and robustly as possible for the purposes of the intervention. That is the basis of the Pointer data set that the Department has been drawing on.

At the point of state aid public consultation, we received fewer than 30 enquiries from citizens indicating that they were not included but were living in legitimately built and approved premises and paying rates. We dealt with those on a case-by-case basis. Following contract award and the widely publicised hyper-fast address-based link that the contractor manages, the number of enquiries increased quite substantially, and we realised that there were more anomalies than we might have expected. As Mike said, the LPS data set was not intended for these purposes, but it is still the most definitive address-based data set in Northern Ireland, and it is used by other Departments and the emergency services. There are close to a million pieces of data, and we drew on about 850,000 pieces, so you would expect some anomalies. However, we were surprised by the number of premises that were not categorised as built and approved. LPS relies on information to be fed into the system by councils and Royal Mail, and, for some reason, it was taking a considerable number of years, in some cases, for those premises to be updated. We became aware of the issue and consulted LPS, the OSNI team and others, and they have been very cooperative in helping us to arrive at solutions and put forward other data sets, including the valuations lists and unverified premises lists. We replicated the OMR process for premises that, based on data from infrastructure providers, could not receive speeds of 30 Mbps or above. That formed the view that 6,000 premises were eligible, based on the same overarching policy that was conducted during the OMR.

There are other data issues that we want to continue to work through with LPS and councils. Some premises that are categorised as built and approved may be derelict. We are working with the contractor to identify those on the ground and de-scope them, but we would much rather do that in a more efficient manner and identify all of them, rather than doing it on an area-by-area basis. As Mike said, the quality of the data is improving. We all want to work collaboratively to ensure that it improves.

The Chairperson (Mr Humphrey): As I said to Mr Forsythe, in the grand scale of the overall scheme, 6,000 seems huge in a country the size of Northern Ireland.

A number of mentions have been made of the Northern Ireland Audit Office report. Did the Department inform the Audit Office of those problems so that they could be reflected in the report?

Mr Brennan: Chair, are you referring specifically to the problems in relation to the Pointer database?

The Chairperson (Mr Humphrey): Yes.

Mr Brennan: Go ahead, Nigel.

Mr Robbins: The terms of reference of the audit were, of course, to look at the procurement process and contract management. We engaged very openly and transparently with NIAO colleagues during the audit. There were some questions that were not raised by the audit team that we certainly would have responded to. The matter regarding the 8,500 premises that were referenced in the report was mentioned in a press release that was issued by DCMS in the summer. It alluded to the additional funding and the 8,500 premises. No further —

The Chairperson (Mr Humphrey): Mr Robbins, we will speak to the Comptroller and Auditor General and his team later, and we can ask those questions of them. I am asking you and Mr Brennan whether the Department advised the Audit Office. You said that it could have asked questions. Did you provide the information?

Mr Robbins: We provided information regarding the premises data that we continued to review, and we said that we were keeping under review the £24 million associated with the 2,500 premises. We were working through a live procurement issue and seeking approvals related to the funding allocations from DFE and DAERA to add to the DCMS funding allocations. We were also going through approvals processes, which culminated in state aid assurance and DOF assurance. As soon as that process was completed, we provided all the information that we thought pertinent to the NIAO prior to this Public Accounts Committee hearing. Once we had secured the funding for that process, we also responded to questions and had a telephone call so that it could have all the information to hand.

The Chairperson (Mr Humphrey): Did you provide the information to the Audit Office during the process of the inquiry and production of the report, rather than that being done subsequently? Was it provided at the time?

Mr Robbins: It was not provided in as robust a fashion as it was at the point of approvals, but other areas, such as premises under review, were not discussed, and they were live procurement issues as well. That is why we felt that it was appropriate to bring this to its attention as soon as approvals were secured outside of the procurement process that was being audited.

The Chairperson (Mr Humphrey): Do Mr Donnelly or Mr Stevenson, who is Treasury Officer of Accounts, want to comment?

Mr Kieran Donnelly (Northern Ireland Audit Office): The first that I heard about the anomalies with the LPS tracker system was in the Department's press release last week. I would expect that type of material information to be disclosed automatically, without my people asking for it. That goes without saying; it is as simple as that.

The Chairperson (Mr Humphrey): I have to say, as Chair of the Committee, that I think that there is some veracity in that comment, gentlemen.

Mr Stevenson, is there anything that you would like to say or any questions that you would like to ask?

Mr Stuart Stevenson (Department of Finance): I will add a very brief contextual point. A couple of members asked about clawback, and it is worth referring to the in-year monitoring guidance that governs that area. It is refreshed annually. There are, however, some long-standing conventions, one of which is the area that covers reduced requirements, and I do not expect that convention on clawback issues to change between now and when the clawback issue is considered. The advice is that Departments flag up any reduced requirements in excess of £1 million. That feeds into their public expenditure process. In the early part of the relevant financial year, we would expect a clawback of that nature to be flagged. That would feed into the Executive's decision-making process on allocations. Nothing precludes the Department for the Economy from bidding for that money, but it will probably be an Executive decision as to whether to return it to a Department. That will depend on the other pressures that exist at the time and on the strength of that bid. I thought that would be helpful to set that out at this stage.

The Chairperson (Mr Humphrey): OK, thank you very much. Thank you, Mr Brennan, for your attendance today as the permanent secretary and accounting officer at the Department for the Economy. I also thank the local departmental representatives and those from the national DCMS for their time and contributions this afternoon. Good afternoon, gentlemen, and thank you very much. Mr Stevenson, you may leave us as well, and thank you very much for your time.