Public Accounts Committee
Oral evidence: Lessons learned: Delivering value from government investment in major projects –
HC 456
Wednesday 20 March 2024
Ordered by the House of Commons to be published on 20 March 2024.
Members present: Dame Meg Hillier (Chair); Paula Barker, Mr Mark Francois, Peter Grant, Ben Lake, Anne Marie Morris, Matt Warman.
David Fairbrother, Treasury Officer of Accounts, was in attendance.
Questions 1-80
Witnesses
I: Nick Smallwood, CEO, Infrastructure and Projects Authority, and Conrad Smewing, Director General Public Spending, HM Treasury.
Report by the Comptroller and Auditor General
Lessons learned: Delivering value from government investment in major projects (HC 554)
Witnesses: Nick Smallwood, CEO, Infrastructure and Projects Authority, and Conrad Smewing, Director General Public Spending, HM Treasury, gave evidence.
Q1 Chair: Welcome to the Public Accounts Committee on Wednesday 20 March 2024. Today we are looking at how the Government have delivered value for money from major Government investment projects and the lessons learned from that. Our thanks to the National Audit Office, which has looked at a number of genuinely successful programmes to see whether, after they have been delivered, there has been a proper evaluation and discussion about whether they achieved their aims.
Navigating the unique and novel challenges of major programmes is crucial to securing the best value from public money, but all too often we see projects and programmes that are poorly managed, poorly delivered and often delivered late and over budget, which has a huge impact on the taxpayer and the user of whatever that project is. We want to know that the Government are looking, as the Committee has, at how they better evaluate the outcomes of these major programmes. We are keen to hear from our witnesses today about how together we can help educate Whitehall and the system to deliver better projects.
I am pleased to welcome our two witnesses today. The Infrastructure and Projects Authority’s chief executive, Nick Smallwood, is with us. He is joined by the director general for public spending, Conrad Smewing, from His Majesty’s Treasury. To be clear, Cat Little is in the throes of moving over to the Cabinet Office as Permanent Secretary. We have had information that your role has slightly changed, Mr Smewing. Can you explain to us what your role now includes?
Conrad Smewing: Cat Little is in the process of moving into the Cabinet Office. I am stepping up as co-head of the Government Finance Function and filling one of her shoes there. We will have another co-head of the Government Finance Function—there is a process to sort that out now. In terms of the Treasury’s responsibilities for public spending, I am taking on most of the day-to-day responsibilities that Cat undertook. Beth Russell is taking over public spending—
Chair: Beth Russell, the Second Permanent Secretary, based at the Darlington campus?
Conrad Smewing: Yes, that’s right.
Chair: Cat Little goes, and how many take on her job? That is some testament.
Conrad Smewing: Indeed. As I say, big shoes to fill.
Chair: Thank you. It is helpful to know. You have other things on your plate, other than what we are discussing today. People will be coming in and out of the room, because there are problems with queues getting into the buildings—apologies to anyone who has had that challenge. I will turn straightaway to Peter Grant MP to kick off.
Q2 Peter Grant: Good afternoon to our witnesses. I apologise in advance: I may need to leave at some point. I am supposed to be at another Committee later, and I am trying to find out if I am more desperately needed there than here. The two Chairs will be in discussion about the transfer fee later.
To start from the general principles, some people might think it surprising that the NAO thought that it had to include in the Report the idea that you should start from the change that you want to deliver and think about how best to deliver it, and then arrive at the project, whereas it sometimes seems as if the politicians announce the project and leave you guys to work out what that will mean. First, is that a fair comment? Is it too easy for politicians to decide on a big idea and to announce it before they have thought it through?
Conrad Smewing: The way the system is set up, the kind of guidance in the Green Book and the theory of how this should work, very much start from exactly the same place as the NAO: you start from what you are trying to do, rather than any individual solution, and then you set your strategic objectives, work out a long list of options which could meet those objectives, and down-select to a shortlist. That is how it is supposed to work, and I think that in the majority of situations it does work like that. That is not to say that there are not some circumstances in which people alight on a particular project that has some sort of momentum before you start, but the strength of the framework should be there to make sure you are assessing that properly against all the other options you have, and if it is genuinely the best-value-for-money way of pursuing your ultimate goal, select it only then.
Nick Smallwood: I absolutely agree with the NAO’s observation, to the point that we have spent a lot of time in the past two to three years developing tools that take you on exactly that journey of being clear about outcomes. We developed the project outcome toolkit, the Project Routemap tools to use for project set-up and Opportunity Framing, all of which challenge stakeholders to be clear on the objectives they are trying to deliver. Getting alignment of stakeholders is key; when we don’t do that, typically we have worse outcomes.
Q3 Peter Grant: The examples that the NAO chose to look at in more detail are essentially ones that appear to have gone well, so they are looking to see what lessons have been learned when things have worked. Obviously, if things don’t work, there is a different set of lessons to be learned. To look at two examples, the Millennium Dome and the London Olympics, most of us instinctively think that we know the potential benefits of hosting the Olympics or a major sporting event, and you will regularly see enthusiasm from governing bodies saying, “We would really like to host the world championships,” or whatever. It is harder to see how anyone at some point said, “Wouldn’t it be great to have a great big upside-down soup bowl planted on the Greenwich peninsula? What a great project that would be!” How do you put together a process or a system for evaluating those potential projects when trying to deal with such completely different sizes, scales and natures of beasts?
Conrad Smewing: What you are trying to do in your business case options appraisal is, first, to clarify what your strategic objective is up front. That could be a wide range of things. It could be economic regeneration in the East end of London, increasing participation in sport or a combination of those things. There is no perfect yardstick; you need to establish what your actual objective is and assess options against that. When we revised the Green Book a couple of years ago, we tried to make it clearer to people that you start with an objective and then find the best-value-for-money way of delivering that objective. Some of these objectives are impossible to compare against. It is very difficult to compare the nuclear deterrent with regeneration in the East end of London. Once you are clear about what the objective is, it is much more tractable to make an assessment of costs and the range of benefits you are likely to get, including much wider benefits. You then go through your business case process of working out how you will actually deliver it.
Nick Smallwood: The nature of the projects that we are delivering through the Government Major Projects Portfolio is complex and quite broad, and it is really challenging for Departments to fully embrace the potential upsides. We want them to think about social value and the five-case business model—we want them to think beyond how they have traditionally looked. The examples in the NAO Report were all very positive upsides. I take the Jubilee line every day from Canary Wharf, and I don’t think the Jubilee line expansion ever envisaged the scale of the benefits to the country of the Canary Wharf development. We at least want to make a stab, and it is just as well if it is a conservative stab, rather than an overly optimistic stab. A number of tools are being developed, and there is lots of work with the Evaluation Task Force to try to improve how we perform in that space.
Q4 Peter Grant: One of the issues that the NAO have picked up in relation to a number of these projects is that very often the actual long-term benefits take a lot longer to be realised than you might think. You can look at a nice new shiny building, or a lot of nice new buildings, but the actual long-term economic benefit to real live people might come 5 or 10 years down the line. Is that a fair observation for the NAO to have made? This question is perhaps more difficult: what have you done to change that from looking at the experience of these other projects?
Conrad Smewing: The NAO are absolutely right to say that, particularly with some of the bigger, wider economic regeneration benefits from these big projects, you are looking at decades, not a short-term period. It is inherently uncertain. At the outset of the project, you can make an assessment of the likely regeneration benefits, but you will never be really able to know for sure. The really important thing in those circumstances is making sure you have the structures set up around the project in order to maximise the chance of delivery of those benefits. There are a couple of references in the Report to having development corporations set up for the land around the big stations on HS1. Having those kinds of tools to ensure you are maximising the chance of getting the wider benefits is an important part of it.
Nick Smallwood: I think there is more we can do. The gate 5 review typically closes out a project at completion of the asset build, if you will, but that is the start of the journey because that is when you actually have something that can realise value. It is about getting more robust and more professional benefits realisation plans, and agreeing how long we are going to check the economic benefits for. All too often we are too short-focused; we don’t look at the longer term. We need that learning. There should be a feedback loop to help us understand, for example, what we did really well on HS1 that we can repeat on HS2.
Q5 Peter Grant: When it comes to shorter-term economic forecasts we have the Office for Budget Responsibility, which was set up essentially to make sure there was some kind of basis for the facts that politicians can then argue about. Is there anything similar for major projects for when a Minister stands up and says, “It will deliver 20,000 new jobs within 5 years”? Is an independent assessment of those figures made at the time?
Conrad Smewing: It is certainly not within the purview of the OBR, who are going to be looking at the aggregate level. Ex post—once the project is completed—the gold standard you should be looking for in an evaluation is having an independent evaluation that compares what has actually been delivered against a realistic counterfactual. Before the project has started, it is definitely harder. You can make an estimate of costs and benefits, and the way in which the approval process works and the way in which the Treasury, the Infrastructure and Projects Authority and others interrogate the sponsors of projects tries to ensure that you are being realistic on costs and likely benefits. That is a bit of a safeguard.
There is also in this space the independent National Infrastructure Commission, which gives advice on the overall infrastructure needs—economic infrastructure in particular—of the UK economy. That presents an independent analysis of likely needs and therefore likely benefits, so that is one body that is helpful here.
Nick Smallwood: Maybe I can add to that. One issue that we are working on right now is building a benchmarking data hub. We are really excited that we should be retaining all the data around our capital projects within government to inform us for the next project or for similar projects; we typically don’t do that today. The IPA has now built a hub, subject to funding, and I will expand and really populate that so that we will have access. That should really help us with the should-cost modelling of projects, rather than relying on the supply chain to tell us.
Q6 Peter Grant: Finally from me, how good are we at allowing for the whole-life cost of a project? You mentioned, for example, that when you have built the building, that is the end of the project. However, the building must still be maintained, and in the case of the Millennium Dome nobody had thought about what to do with it once we had finished having millennium events in it. Are we any better at envisioning what the longer-term future will look like and what resources will be needed to make it happen?
Conrad Smewing: Whole-life costing is central to the appraisal process; all the comparison of options will be looking at whole-life cost. For something like the Millennium Dome, when you are thinking about what the opportunities might be for using it further down the line, that can be a pretty uncertain thing to judge at the point of the business case. However, in general, thinking about the whole-life cost, in particular the costs of running and maintaining, is central to the options appraisal.
There is a further question there about ensuring that you manage those whole-life costs once the project is in flight and ensuring that you are putting the right amount of investment into maintaining the assets and not starving them of maintenance funding. That is an important thing post appraisal and post completion of the project where we need to ensure that we have the right incentives and structures to do it. Nick, do you have anything to add?
Nick Smallwood: I think it is an area that we can do a lot better in. The net zero agenda is driving us to lower operating costs and be focused in that space, which is good. The Construction (Design and Management) Regulations require you to think about the disposal and dismantling of an asset at the end of life, so we are improving—the anecdotal evidence that we have from the private finance initiative offboarding programme shows us that we have not actually done a very good job of long-term, whole-life costing and maintenance, so it has a focus now to improve, going forwards.
Q7 Chair: Thank you. PFI is a really good case in point. We have looked at that, and we are quite concerned about how much planning needs to be in place. We will revisit that report, but are you aware of any evaluations being done about the end of PFI that would inform future programmes?
Nick Smallwood: Yes. My team is doing a lot of work on doing the health checks as we offboard PFIs, looking at what best practice looks like and what we are learning from the different asset classes. We have quite some data together now within the IPA.
Q8 Chair: Will that include going into the design side of it? Obviously, with things such as ongoing maintenance costs, if you have a ceiling so high that you must have scaffolding up to change a lightbulb, it is very challenging for the organisation running the building. Will you be looking at all those aspects as well as at the PFI denouement itself?
Nick Smallwood: I think the overarching messaging is about running costs, rather than design. We have seen typically that projects have been built to time and cost, which is encouraging. However, I would have to check if we have access to the design information.
Q9 Chair: I remember that there was some discussion a few years ago, when Michael Gove was Education Secretary, about banning curves in buildings or something. I cannot remember the detail, but under the headline there was an attempt to try to standardise certain things to make them cheaper to run. Is that something that is built in as well?
Nick Smallwood: That is absolutely built into a number of programmes. We are seeing that in the prison building programme, where “design one, build many” is the philosophy. The hospital building programme is moving to a kit of parts, if I dare mention it.
Chair: People may want to see our Report on that as well.
Nick Smallwood: That programme is now making good progress, actually, on the Hospital 2.0 design. That will drive down the cost per square metre. We are thinking about the operating cost, and we are also thinking about net zero. I think the school building programme is probably ahead of both prison building and hospital building. It is building today net zero schools, which is really commendable.
We are seeing that standardisation increase. The more you move to modern methods of construction and off-site fabrication, the more opportunities come. The more you use digitalisation and data—“design one, build many” and use the same data—the more you get standardisation.
Q10 Chair: We have lots of discussions about standardisation, but one thing that came up when we visited one of the hospitals that will be under the 2.0 programme is that there are not many manufacturing sites for modular, off-site build. When you are looking at these things and learning lessons—particularly at the IPA, but it would also feed through to the Treasury—where in the system do you send the suggestions or the learning? If we are going to do this and it is cheaper to standardise, cheaper to design one and build many, there is therefore another bit of the system that needs to think about supporting or encouraging a factory or producer.
Nick Smallwood: Well, indeed. You have to do the homework on what makes economic sense. Just to do off-site fabrication as a stand-alone could cost a lot more money, actually.
Q11 Chair: I think it does in the short term, doesn’t it?
Nick Smallwood: It does in the short term, so you have to have scale—you have to have economies of scale and repeat work. That is about having a supply pipeline to the same MMC supplier so that it has continuity of work.
Q12 Chair: That is beyond you, though, isn’t it, because it’s a policy decision?
Nick Smallwood: That is then in the Departments, on the individual projects, but we do challenge, so that is one of the challenges for the hospital programme. It will have to stand up the supply chain. That does not exist today, but there is a willing market.
One of the benefits you get from off-site fabrication is that you avoid the problems of the weather. You have different access to a different market of skills, and you have repeatability, so you can automate, so your unit costs can come down and certainty goes up. It is attractive to industry if you give them that pipeline of certainty.
Q13 Chair: It’s interesting that it is a policy decision but you are seeing the problem, as you have well described. Mr Smewing, the IPA says this. You at the Treasury have some hand in this. What do you do with that information?
Conrad Smewing: One of the important links here is often through the spending review process. What we tend to do is involve the IPA in both considering the deliverability of the bids and looking at the cost estimates and whether it could be done more efficiently—essentially, improving the productivity of the capital budget using these kinds of things. That is a very early stage in the process of thinking about setting up a programme or a set of projects, and you can inject the thinking at that point.
Also, in the early stages of the approval process, the IPA and the Treasury are hand in glove on making sure that we think that the programme is set up to deliver the best value for money. If that involves pushing further down the standardisation route, or not, that is the opportunity for us to—
Q14 Chair: If I am being uncharitable about the Treasury, you are a bit of a cash-in, cash-out economy, in that you grant the money and you expect it to be spent in the year. If you are looking at a long-term programme like modular building that is more expensive up front, who is going to get the benefit of a supply chain? It might be the NHS in this case, because there would be enough of a volume, but equally it could potentially be schools and other buildings. How are you building in the extra costs that might be there to pump-prime a bit of industry to deliver modular buildings or anything else into the business plan? It might not be cost-effective in the beginning, but it might be over a longer period.
Conrad Smewing: We would typically try to identify the way of delivering a programme that gives best value for money, check against affordability, and then size the scale of the programme to do that, so that you are not sacrificing better value for money, because the up-front cost is lower. It is often the case that when the Treasury is thinking about an investment that is going to pay off in the medium to long term, we will set aside extra money, either from central funds or as part of the spending review, to pursue that.
I agree there is an inherent tension between affordability and, in some cases, delivering investments that will pay off. In those circumstances, you have to direct your resources to the highest-return investments available to you.
Q15 Chair: We have talked before, and you have been in the room, about cross-Government working and bidding jointly to the spending review, but relatively few do that. In terms of delivering on major projects, cross-Government working can be very important. Do you think enough happens in looking at major projects? Do you have any good news stories, and where are the challenges still?
Conrad Smewing: It is definitely a difficult problem. Lots of people have identified the problem exactly and the difficulty that where there are future benefits for one section of Government, the costs are falling on another, and it can be difficult to bring people together.
It can also be difficult to ensure the proper level of prioritisation. If individual Departments recognise that it is a good thing to do, but it is not right at the top of their priority list, it can fall off the agenda. One of the areas where I think we did this well in the last spending review, on which we will definitely be looking to build and expand for the next, is the Joint Combating Drugs Unit. We encouraged all the Departments involved to come together with a joint bid. They set up the right governance framework, with all the Ministers on a cross-ministerial board, to do that investment where, if it is successful, you are essentially doing preventive spending that impacts all over different Departments and saves downstream costs.
Chair: We have looked at that separately as well.
Conrad Smewing: Yes.
Q16 Chair: In this case, let’s take the modular building, for example. If it is working for hospitals, you could apply it to other areas, but at the moment, there does not seem to be much read-across from different Departments.
Nick Smallwood: There is, actually. Maybe I can give a couple of examples where we have cross-Government opportunities. The IPA chaired the Government construction board, at which we have those very conversations about transforming infrastructure performance, of which MMC is one of the opportunities. There are good conversations and cross-learning at that forum.
Q17 Chair: Would that include skills and industry?
Nick Smallwood: Indeed. In fact, just this week we have kicked off market analysis of skills and skill shortages over the next five years, because we think there are going to be some pinch points. The cross-Government construction board is one example. We also have the Project Council, where the chief project delivery officers are now accountable to me for driving performance improvement in their Departments. All the Departments are represented, such as the CDDO and the Treasury, and we get together quarterly to really challenge each other and share best practice.
The No. 10 delivery unit is now quite mature. We get the central teams together, scrutinise project delivery and just challenge: “Is there something that the centre could do in terms of an intervention that we are not doing?” There are other forums, like the 25-year environment plan board, on which I sit; again, all Departments are represented and we have a good conversation about the net zero challenges. We are starting to see that cross-Department conversation maturing and happening. It is not perfect, but I have seen huge change in the past two years.
Q18 Chair: And you have been in post since 2019, so five years?
Nick Smallwood: Correct.
Q19 Chair: In that time, have you seen some change in the delivery approach from No. 10?
Nick Smallwood: Yes.
Q20 Chair: You are saying now that what is in there is working better than—
Nick Smallwood: Absolutely. The work that Dr Emily Lawson did to set up the delivery unit has been really good.
Q21 Chair: That is helpful, because in an election year there is all sorts of discussion about delivery.
We have seen different models of governance for some of these projects that the NAO has highlighted. What do you think works best in governance terms?
Nick Smallwood: Having the competent people present at the right phase. It does not help putting in a delivery expert post-FBC. You need to be in the planning phase to ensure that you are setting up the project for success. I think the Crossrail “lessons learned” made some really good observations. We can probably strengthen our guidance in that space, if I am honest.
Pockets of good practice exist, but we have not brought it together into one single space. The IPA is preparing and delivering some training material for project boards of what “good” looks like. I think we just need to challenge: “Have we captured all the Crossrail lessons and put it into practice?”
Q22 Chair: We certainly see—with the UK Government investments and some others, and some of the non-exec appointments recently—some of the right people coming in. You made the point about having the right skills in the right place, but even at a non-exec level, this Committee has seen a lot of non-execs either not turning up—sometimes because they are from a Government Department and have other things on their plate—or staying on beyond their competency when the project has changed. You said that you may be changing the guidance. Can you give a bit more clarity about what you might be doing, when you might be doing it and what you might want to say?
Nick Smallwood: We have only just had the Crossrail report this week and we are busy scrutinising that, so it would be wrong for me to give you a timeline for when we are going to fix something. Realistically, I think we do need to cover non-exec roles and governance boards. I have seen a number of senior boards set up across Departments on some of the big energy projects, for example, and they are having a real impact. An officials’ discussion to make sure—
Q23 Chair: When you say senior boards, is that officials only, or are they bringing in outsiders?
Nick Smallwood: Officials only, but to inform them of how confident and competent they think the arm’s length body is in delivering the outcome. That is something I did not see on HS2, for example, but now on Sizewell, on CCUS and on others, they are quite impactful boards. I think that that is the best practice that we need to encourage others to take on board.
Q24 Chair: Do you think that there is a deficit of skills and understanding about what non-executives do inside the civil service? The non-execs brought in from outside are obviously in that bracket. When people are appointed from a Department to be a non-exec, as I say, we have seen too often that the attendance is not there. Just not being in the room is obviously one problem, but they might not always be following it as closely as they should be. They might just be there as a nominee but are not really engaged. Is that an issue? We have seen it. Is that something you have seen, too?
Nick Smallwood: I do not see it regularly and repeatedly; I have seen it in pockets. Right now, I am being asked through the IPA to give an opinion on the non-executive appointments for HS2, which I think is good. The fact that they have asked us is really good. I am now the Government rep on Sizewell as a NED. With that comes legal responsibilities, so not attending is not okay.
Chair: Good—I am glad that you have said that very clearly. We think so too, but it is heartening that you say that. That is very helpful.
Q25 Ben Lake: Thank you, gentlemen, for coming in this afternoon. I would like to touch on stakeholder engagement. Mr Smallwood, you have already mentioned the importance of different governance models and making sure that people are involved during the planning stage, the delivery phase and then post delivery.
Paragraph 20 of the NAO Report mentions that “it is rarely the case that a project delivery team alone can deliver the full value of a project. Our case examples show that delivering value often requires a range of organisations and stakeholders to work together and have a clear understanding about their roles.” Paragraph 23 states, “Our case examples indicate that in some cases it may be more effective for an organisation outside of the project delivery team to take responsibility and accountability for ensuring that the project delivers value”, post delivery. How effective is Government working with external stakeholders in particular to making the most of assets once the project has been delivered?
Nick Smallwood: It is a challenging space, for sure, on some of the major infrastructure projects, but the national infrastructure planning process requires that engagement—informal, formal. The planning inspectorate will require specific engagements.
To go beyond that to make sure that the key stakeholders are involved is key. Where I see projects doing a really good job of their planning preparation, they go that extra mile to engage the key stakeholders at the centres, as well as the supporters. If you then take that on to who has influence over value, where you see opportunities for growth, for example, on the HS2 stations, there has been good engagement in the communities and councils around all four stations, because that is really where there is huge potential value to come. The last figure I saw was of the order of £20 billion. Where we saw it in Ebbsfleet on HS1, I think we saw real benefit. That takes energy and time, but it is time well spent, so we need to encourage it to happen more often.
Where I see real best practice is where we are now embracing digital technology. You can actually visualise the end result of a project, which means you cannot rush to put spades in the ground. You need to do the detailed design work and the front-end development to a level of maturity where you can show a visualisation of what the end result would be. Where that has been done at A303 and the Lower Thames Crossing, they have used it to really good effect in the planning cycle, getting stakeholders on board and getting positive support for what essentially could have been a blocked project. There is lots more we can do and lots of best practice we can learn from.
Q26 Ben Lake: You referred to examples of best practice. Is that now being rolled out in a systematic way across Government? It strikes me as eminently sensible that you are identifying those stakeholders very early on.
Nick Smallwood: I don’t think it comes naturally to everyone, but we do have the assurance gate reviews that the IPA convenes, and we do challenge to make sure that we are learning from one project to the other.
Q27 Ben Lake: Just so my understanding is clear, once you have identified the relevant stakeholders on a particular project, are they then involved throughout the delivery phase so that they have a good understanding of the asset that they will potentially be inheriting?
Nick Smallwood: Absolutely. Take a project like Sizewell. That has a local liaison community forum set up in the local town that will stay for the life of the project.
Q28 [1]Ben Lake: I am sorry to labour the point, but are liaison committees or similar forums starting to be applied across the way we deliver projects?
[Witness Q28 – issued correction on 20th November 2024, read the correction here]
Nick Smallwood: It would be wrong to say that it always happens without fail. It needs to be encouraged to happen without fail, because you have to embrace your stakeholders.
Conrad Smewing: On things like the really big transformational rail and transport projects, there is an emerging best practice of having some kind of board that can bring together all the local authorities and quite a wide range of central Government Departments. You need to drive value from things, so it is not just the Department for Transport; it is also housing, economic regeneration and jobs. There is a Treasury East West Rail growth board that brings all those people together to make sure that while you are building the railway, you do your best to get all the broader benefits around it. That has built on the kinds of models that you saw for HS1, which are mentioned in the Report, and things like that.
Q29 [2]Ben Lake: Thank you, Mr Smewing. That is very helpful. I think you are saying that the Treasury recognises the way in which some of these major projects can drive growth and regeneration. With that in mind, how does the Treasury encourage Departments to put in place the measures and investments necessary to really maximise the growth? The NAO Report referred, as you have done, to projects like High Speed 1 and the benefits at St Pancras, Ebbsfleet and so on. How does the Treasury go about encouraging other Departments to really recognise the potential contribution of these major projects for growth?
[Witness Q29 – issued correction on 20th November 2024, read the correction here]
Conrad Smewing: The example that I just given absolutely hits that point. The Treasury set up the East West Rail growth board to have all the relevant Departments and the local authorities on it, because it is central to the Treasury’s conception of these major projects that they will deliver much broader economic benefits and broader growth.
It is not always the case that you have a project large enough that you want the Treasury to corral everyone. In those cases, the Departments for Transport and for Levelling Up, Housing and Communities will co-ordinate themselves. But on the really big projects, there is often some Treasury involvement.
Q30 Ben Lake: Mr Smallwood, paragraph 15 of the NAO Report says: “The importance of taking steps to maximise the use of an asset, and therefore the potential value from the investment also came through from some case examples”. It specifically cites Diamond Light Source as “an important piece of UK national infrastructure.” With particular reference to Diamond Light Source, although I am sure there are other examples, what more do you think Government could do to maximise the use of their assets to drive growth and regeneration?
Nick Smallwood: We touched on that a little bit earlier: the five-case business model and thinking beyond the delivery of the asset specifically, but also challenging what social and economic benefits could accrue from it, being broad in the thinking and driving project teams to do that early in the piece. That then gets translated into a benefits realisation plan, and then you can really start to have the dialogue with the right stakeholders. We need to do a better job in that space.
It is difficult. With some of these projects, you see them delivered and then the lightbulb goes on and people realise that there is a side benefit they had not considered. We want to do a better job of trying to think more broadly earlier. It is very difficult for some projects, and very easy for others. With a rail project, housing is an obvious one at stations. That is very clear. But what else might accrue? Are there benefits via reduced energy consumption? Those do not come naturally. We need to really challenge project teams to think more broadly.
Q31 Ben Lake: Thank you, that was a very useful answer. I agree with you that it is a challenging way of looking at a project. Do you have any thoughts on which Government Department, or which element of a project team, is best placed to do some of that thinking?
Nick Smallwood: It does sit with the Departments. I draw attention to both the Department for Transport and the Department for Energy Security and Net Zero. They are taking benefits realisation seriously. They are pushing that agenda, getting the training through and trying to up their game, because of course they are two Departments with very high capital spend. We need to encourage that to happen across the piece.
Q32 Peter Grant: Have you made any progress in making sure that the budget includes a realistic amount of money and time to carry out proper evaluation where the project is under way?
Conrad Smewing: It is not so much a budgetary issue as one of ensuring people have the right focus on it. As a proportion of the cost of these major projects, the evaluation will be quite small. I think we have made some progress, but there is definitely a lot more we can do. You will know that back in 2019, when we looked at the level of evaluation of plans across the Government Major Projects Portfolio, the level was quite low—I think it was something like 8%. That was one of the catalysts for setting up the Evaluation Task Force—I think it was in SR20—as a means of holding people to account and making sure they are following the guidance and have evaluations, but also helping them to design a proper robust evaluation of a major project. That sometimes does not look straightforward from the out, and these can be difficult projects to properly evaluate. The ETF has made a load of progress. It has engaged with several hundred different programmes and projects across—
Q33 Chair: Who is on the Evaluation Task Force? What professional skills are on there?
Conrad Smewing: They are professional social researchers in the main, and also economists. It is headed by Catherine Hutchinson, who used to be in the analysis function of a Department, so she knows how the individual directors of analysis in individual Departments think about things and how they operate. Her staff tend to be professional evaluators who are there to offer advice on how you would design things. This can be how to design a study, but also how to think about the roll-out of your project, programme or policy so that it is more evaluatable.
Q34 Chair: Do you have policy professionals in that at all?
Conrad Smewing: Yes, there are policy professionals in there as well. We have very close links between the spending teams in the Treasury and the ETF, so that when it comes around to spending reviews, we often bring the ETF in to give us its assessment of the state of evidence in a particular area or on a particular policy. You need the full life cycle, that the Treasury is both asking for the evidence on how effective this project or policy is going to be and ensuring—
Q35 Chair: You are trying to drive the behaviour. I remember that both the head of the policy function, Tamara Finkelstein, and the deputy head of the policy function, Susan Acland-Hood—both permanent secretaries—have talked about the need for looking at policy from end to end when it is developed. That is, delivery through to evaluation. Do you think that is getting better?
Conrad Smewing: I do think it is getting better. The ETF has been very successful. There is always more to do here, and these evaluations always take longer than—
Chair: Than the politicians want them to take!
Conrad Smewing—the political cycle would want. Nevertheless, there are things that we are setting up around them, including particularly transparency in where these evaluations are being performed or not performed, and what the outcomes are. The ETF is setting up a registry of all the evaluations, which I think it is going to produce later this year. That is quite a powerful tool.
Q36 Chair: How do you make sure this is not cut? I will come back to Mr Grant shortly. I remember that when I was a Minister, there was a cut to the Departments. When I was a science Minister, the social science fund was in the frame, because it was something that people would not really notice if it was cut. I pled against it being cut, saying that it would cause problems down the line and that we needed to think long term about a gap in knowledge and information. It was relatively small, but there is always a risk, isn’t there, that it does not seem important when you are trimming costs on a programme?
Conrad Smewing indicated assent.
Chair: How are you going to make sure that the evaluation is written in? As Susan Acland-Hood said to us when we were looking at this in her Department, it is a rounding area in a budget. It is relatively small, in terms of the overall cost—how are you making sure that the ETF is embedded?
Conrad Smewing: The Treasury is a huge supporter of all of this stuff, and so we are going to be frowning on anyone who suggests saving through, as you say, a rounding area in the budget. Cutting social research is not going to be something we are recommending. There are a couple of other points here. In some areas, it has been helpful for us to have budgets for evaluation held centrally. The ETF has an evaluation accelerator fund where, if for one reason or another resourcing is a problem, Departments can bid into that. So that is another way that we can support people.
Q37 Peter Grant: As a Committee, we often find that Departments push back very hard, even on projects of much smaller scale and complexity than we are talking about here. They push back very, very hard. The idea is that there should be robust quantitative evaluation built in at the start and then monitored and reported on later on. The NAO have highlighted that in essence there has not really been a culture change in the last 10 years. What do we need to do to change the culture so that the people in charge of setting up and starting to deliver these projects that cost billions of pounds of public money understand from day one that they have got to build in the evaluation at the start and have to be held properly to account against that evaluation throughout the project?
Conrad Smewing: I will start and then Nick might want to come in. There is definitely lots that we can do at the centre. The Treasury and the IPA can make things clear with our hard levers around approvals, conditions and spending reviews, but also in the way that we conduct ourselves, demand evidence and use that evidence in making decisions. I think that is an important thing, as are these sorts of discussions and the NAO’s Report. The people who run these major projects want to make a difference in the world and be able to show that they have done so.
There are really encouraging case studies in the Report, like the Diamond Light Source study. They have done the project and done a really good evaluation of what the impact of that has been, which has helped make the case for further investment in the Diamond Light Source to follow on. That kind of cycle can change people’s minds.
Nick Smallwood: Conrad and I have written to all the SROs—we started this week—requiring them to demonstrate their benefits realisation cases, because there simply are too many gaps in the GMPP. It is the responsibility of the SRO to take ownership of it. We have done a lot of work in the IPA to strengthen the training that goes to the SRO community, to increase their length of tenure in role, and the time commitments that they make on the projects. I think that is having a positive effect. There is more to do. It is very clear where the accountabilities sit and we need to chase that up now so that they start to show us the evidence of the evaluation plans.
Q38 Peter Grant: Mr Smewing, you said that the people in charge of these projects want to be able to demonstrate that they have delivered. Is it the case that they want to be able to demonstrate that after a very robust and independent evaluation process? Or do they simply want to be able to say, “We have been a success, and nobody can prove otherwise because you have not collected the information”?
Conrad Smewing: Well, that is where we need to have a decent standard of what is required to prove that you have actually made a difference and had a positive impact. I definitely agree with you about simply producing something that is a bit self-serving and saying, “Here are the outcomes without a proper estimate of the counterfactual and an attempt to demonstrate the causality of the impact of the project on it.” That is not good enough.
The ETF and the Treasury try to set out what good looks like in an evaluation so that we can say, “You are right. You have made an impact.” I do have some sympathy, particularly on some of the really big projects in areas like Defence, where it can be very difficult to set up a proper counterfactual. As we mentioned earlier, the timescales are so long that it is very difficult to have what you might think of as a gold standard evaluation. But there are always things that you can do short of that. Are you delivering the outputs that you expected? How has the asset that you have constructed performed? There are always things you can do.
Nick Smallwood: One area that is fundamental to doing a robust evaluation is getting the cost estimate right in the first place. That is a skill as well as an art, and an area we need to do better. We are weaning project teams off what has historically been a fixation on reference class forecasting, which is fine in the early phases, but when you have a developer—
Q39 Chair: Will you explain that?
Nick Smallwood: Reference class forecasting is looking at something that is maybe similar and applying the cost logic to your future project, perhaps often missing the unique specifics of location and the detail around that asset. It is much more straightforward and accurate to have a bottom-up cost estimate once you know what it is that you want to deliver. We wrote cost-estimating guidance a couple of years ago, which I think is world class. We developed a community of cost estimators across Government—across all 19 Departments. Now, I think it is the time to really home in on improving the quality of the cost estimates that come, rather than relying on the supply chain to tell us what they think it is going to cost. There is a real opportunity there with data and digitalisation to look at how the system performs. You should be able to analyse how accurate you are on school building, on hospital building and on linear infrastructure, and then have a feedback loop. We simply have not collected the data historically to do that.
Q40 Chair: So that is the benchmarking hub?
Nick Smallwood: The benchmarking hub would allow us to do that, and feed into the whole of the cost-estimating process.
Q41 Chair: What is the timeline for that?
Nick Smallwood: The benchmarking hub is built. We just need to populate it with Department data.
Q42 Chair: Okay, so what is the timeline for when it will begin to be more useful?
Nick Smallwood: I think later this year, subject to budget and resources to populate it.
Q43 Chair: Obviously, the more data that goes in, the more useful it will be.
Nick Smallwood: Absolutely.
Q44 Peter Grant: We have already heard the most recent figures the NAO had in December 2019: 8% of the spend and major projects had robust impact evaluation plans in place, but 64% did not appear to have any evaluation arrangements in place. What would you consider to be the limits of acceptability with those figures? If 8% is not acceptable, how high does it need to be to be acceptable? I am assuming you do not think that having 64% of a £400 million spend unevaluated is acceptable. What is the maximum level you would allow that unevaluated spend to be before you would say that it is not acceptable practice?
Conrad Smewing: I think it is not acceptable for any of those major projects to have nothing in place. There is going to be something that you can do on all of these, even for those major projects, which are often in defence, where having something that a social researcher would regard as a robust evaluation is not going to be possible. The way to look at this is to grade the quality of the evaluations. I do not think you are going to get a gold-standard evaluation on all these things. That said, 8% is definitely too low. Over time, you ought to be able to get to 100% for something on all of these things. We are in the process of re-doing the work that led to that original 8%. We are reviewing and reassessing where the GMPP stands on this. We will see how far we have come, and how far our admonitions and levers have been successful.
Nick Smallwood: I think most Departments have gaps, but the dominant numbers sit in the Ministry of Justice and the Ministry of Defence. The MoD has certain circumstances that we can’t fix, but it can still do something. The MoJ is actually being quite positive and is now on the case to review how it can do a better job of benefits realisation. I think we are going to see some short-term improvement.
Q45 Peter Grant: I apologise if I did not catch your answer earlier, but do you know where those December 2019 figures are now? Do you have an update on the 8% and the 64%?
Conrad Smewing: We are doing the work required to get that update now. I hope that later this year we will be able to come out with that.
Q46 Chair: Can you write to us with that? It can then be on the end of this report.
Conrad Smewing: Absolutely. Yes.
Q47 Peter Grant: If you look at the last annual report of the Infrastructure and Projects Authority, almost exactly 90% of projects that are big enough to be included in your report were assessed as either red or amber. The Committee has spent time looking at a number of those in some detail. With almost every one of those projects that we have looked at—although we might not have worded it exactly the same—we have pointed to the basic lessons in this report. There are catastrophic failures to comply with some of the basics. For example, with the emergency services network, the stakeholder consultation was just not good enough. If we are not robustly and almost aggressively evaluating these projects as they go on, how can we hope to learn from the mistakes of the ones that have gone disastrously, and learn from the good practice of the ones that have worked?
Nick Smallwood: Can I just check the figures you quoted? My figures suggest that 89% of projects are either amber or green, so only 11% are red. That has been consistent for the past two years, although we are seeing an increasing trend in red-rated projects. We have now implemented the response to red programme, where we look at red-rated projects and challenge them against the recommendations. We are actually seeing that 70% of those projects that receive a red rating can get to amber or green within 12 weeks, which I think is quite good. I think that the system is now functioning as intended, which is to not let projects convince themselves that they would be okay but rather be robust in an assurance review and challenge them on deliverability, preferably before FBC, so that we have the chance to fix them.
Many of the projects that you see are post FBC, and I am afraid that we cannot assure goodness into a badly developed project. The ESMCP is a good example of where, if we had our time again, we wouldn’t have done it the way that we did it. But, for the new projects coming down the track, I think we are seeing a different level of maturity of development, so a lot of the really knotty ones are turning out to be red, some as a result of high inflation or of overly ambitious ministerial schedules. There is always a good reason why they are rated red; it is less so that they are “not in control,” for want of a better expression.
Q48 Peter Grant: I don’t know that the Committee would accept “overly ambitious ministerial schedules” as being a good reason; it may be an explanation, in that we can understand why it has happened, but I think that part of our intention in doing all of this work is to find ways of making sure that Ministers are a bit more realistic about what they promise.
Just to clarify the figures, the figures that I have from the 2022-23 annual report were that it was about 9% red, 81% amber and 11% green. So, if you take red plus amber together, that is about 90%, and if you take amber plus green, it is about 92%.
Chair: Yes, so similar to what Mr Smallwood said.
Peter Grant: Are you able to tell us how many of that 9% red in 2022-23 have moved? I think you said that most of the ones that are red now should be able to move into amber or green fairly easily. How many of that 9% have moved?
Nick Smallwood: Off the top of my head, I think that we managed to move about 70%. But we have seen new reds appear, of course, and we have seen some of the red-rated projects ultimately deliver on their plans, so it is an evolving picture.
Q49 Chair: We have talked a lot about the business planning of it, but obviously there is then the delivery of a project, and you still have oversight at that point, Mr Smallwood—we will come on to the Treasury’s oversight at that point. What is going well and what isn’t, from after the business case? I don’t know what gate that would be, but that is after the business case and before it is completed. Where are the weaknesses in that system, and have we got the right skills in the civil service to deliver these programmes?
Nick Smallwood: We have done a lot in the skills space, in terms of developing and launching the projects academy, where we now assess and accredit our project-delivery professionals and SROs.
Q50 Chair: This is the Saïd Business School?
Nick Smallwood: This is the Oxford Saïd Business School for the major project leadership academy, and we use Cranfield for the project leadership programme. We have four levels, starting from foundation and going up to master practitioner. We are just about to get through 1,000 assessed and accredited professionals. But it is not immediate. You have to do a lot of work to get accredited, because you have to demonstrate your competence; you have to have attended the MPLA, for example, if you are master practitioner, and you have to be a chartered project professional if you are a project director who wants to be master. It is quite a journey to get through.
Q51 Chair: How long does that journey take if you are someone reasonably senior?
Nick Smallwood: It would take a year or so to get through, unless you already have the track record and evidence.
Q52 Chair: Okay—a year or so, so it is not so long.
Nick Smallwood: No. We have an ambition to get another 1,000 through by the end of March 2025. I think that is credible, but we do have 16,000 professionals that we need to get through, so the focus needs to switch now to the GMPP and where we are spending the most money. That is our ambition for the next 12 months.
Therefore, on skills, we are moving, but to your point on whether we have the right skills, I think that we need far more technical and engineering skills. We really have a gap there, and an overdependence on the supply chain.
Q53 Chair: But don’t you think that the civil service recruitment process is unlikely to bring in people at a junior level who will grow in that? You are looking at people coming sideways into Departments.
Nick Smallwood: We have a number of science, technology, engineering and mathematics graduates who are coming in, which is refreshing to see, but to be honest we don’t have enough at the more senior levels, operating the largest programmes and projects, and we certainly need them. That is why assessment and accreditation is so important. Claiming that you have been on the course but not having the skills to deliver is not good enough; we need to have the assessment.
A lot is happening on skills, but what could we do better? I think that we need to be much more intelligent clients. By that, I mean being smarter in how we go to the supply chain and manage our contracts. We are seeing some really great practice there, such as the PPP contract at Sellafield, and lot of good thinking is going into the hospital programme and how to deliver against this huge challenge. We did not have that level of rigour on HS2 and we have seen the result. Competence in the supply chain itself is generally a mixed bag, so you have to show up as an intelligent client to manage it. That is the gap. I think the biggest gap that we will have to resolve in the next few years is data and digital. That world is upon us, and we need to embrace it and use it to our benefit.
Q54 Chair: All the things that we have highlighted as well are critical. You talked earlier about senior responsible owners, and we have just talked about skills. It is perhaps a bit unfair of me to ask this, because I assume that you are not senior enough to answer for the whole of the civil service, but do you think that the rewards are in the right place? You are getting people through the Major Project Leadership Academy, but we often see that the route to senior civil service roles is not through these specialist technical skills. Do you think it is a problem that we do not have enough people at the most senior level who have actually delivered projects?
Nick Smallwood: It is a problem that it is within our gift to solve. One of things I have challenged is whether we have the right level of civil servant as an SRO on a major project or programme. It does not have to be a junior to the next level up; you can have parity of grade. That is one lever we can pull for the appropriate projects. We have also secured pivotal role allowances for competent assessed and accredited SROs. I would very much like to get the same in place for the programme directors, because we are seeing an increasing gap growing between the private sector and civil servants.
Q55 Chair: You mentioned Defence earlier as a big challenge, and it has repeatedly been a challenge. They are obviously long projects, and because they are so hugely expensive, a small delay or error could build a hospital or a school—we are talking about those sorts of figures. Do you worry about what is going on in Defence, Mr Smallwood, and what impact is the IPA having on defence projects?
Nick Smallwood: For Defence, they are showing up in large numbers to get through assessed and accredited schemes. Our MPLA has a very healthy cohort of MoD representatives, and the same is true of the project leadership programme. Governance is a challenge, and we are looking at how we can simplify some of the governance models. An example would be in defence nuclear, where some good work is being done by Maddy McTernan to streamline that get the right SRO in the right place doing the right work, which is an opportunity to improve. We have gone from almost no engagement with the IPA five years ago to absolutely open engagement. I have so many requests for help and input. It is a very healthy and positive engagement.
Q56 Chair: That is good. You are presumably cleared to look at every bit of what the MOD is doing, or your teams are.
Nick Smallwood: Within the resource constraints I have within the IPA.
Q57 Chair: But you have access to all areas; there is no hidden bit of the system that you are unable to access.
Nick Smallwood: That’s correct.
Q58 Chair: That is helpful to know.
To pick up on another point, earlier you talked about social value. When we looked at High Speed 1 and High Speed 2, the benefit-cost ratio was interesting. For rail, at one point the benefit-cost ratio took out the time on train travel because it was not valuable, whereas those of us who travel know that the time we are away from our offices and our phones is sometimes quite a useful time to work. Sometimes these benefit-cost ratios do not take into account the wider social value of job creation in an area. Is there a better way the system could be capturing social value? We have the legislation that says we should be doing that, but is there a better way it could be done? Are there any areas for improvement there? Mr Smewing, I’ll start with you.
Conrad Smewing: It is definitely the case that the benefit-cost ratios in these projects cannot capture all of these things, which are very difficult to quantify. This was always the case. It is one of the reasons why in the update of the Green Book we were crystal clear that when you are thinking about project selection, you should be thinking about what your strategic objectives are, which will often include delivering that social value, and choose the options that deliver that and not be a slave to the different benefit-cost ratios, which, although useful, are only one lens to look at how to think about a project.
On how to capture it better, there are a couple of directions in which you can go. It is possible within the quantitative appraisal to better estimate these broader social values, recognising that while they are going to be much more uncertain than direct monetary values, it is better to attempt to quantify and weight than not to try. That is something that we can do, and the Green Book team, who work on this all the time, are definitely looking at ways to do that better.
The other way—coming back to where we started with the NAO Report—is to think about what you are trying to achieve, which will include quite a broad range of things, and then see which options are really going to achieve that.
Q59 Chair: If we take the Olympics as an example, job creation and community sports activity were a part of that. I am also thinking of Ebbsfleet, but that is a slightly different issue. Those were long-range aims, and frankly a lot of them have not been delivered. I know the Olympics case well because some of the Olympic Park is in my constituency.
Going back to some of the points that Mr Grant raised about evaluation, how far on is the evaluation going, is it going far enough, and what is the consequence? We delivered the Olympics, and everyone said “Fantastic, we had a great games and the volunteers loved it.” My part of London got regenerated, and we managed to get someone to buy the stadium, which was the big challenge. It’s all, “Tick, tick, tick—gate 5 reached and delivered,” but actually gate 5 has not been reached because the jobs and sporting activity across the UK have not materialised in the same way, as we have covered in this Committee. What is the consequence for a big project when it delivers on one level but does not deliver all that social value that was supposed to be a part of it?
Conrad Smewing: There are two things here. We have had the Olympics and we cannot do that over again, so this is about lessons learned for future similar projects.
Chair: There were the Commonwealth Games.
Conrad Smewing: I know the DCMS—
Q60 Chair: How are you capturing those lessons? That might fall into benchmarking data—I will come back to Mr Smallwood on that.
Conrad Smewing: I am not an expert on this, but my understanding is that the DCMS did a review of how it had delivered and what the benefits of a range of these things were—the Commonwealth games, the Olympics and so on—to try to capture the lessons learned for doing those types of thing again. I suspect that some of those lessons are a bit more specialist to that particular kind of project.
The other thing here is something that we were talking about earlier: whether we have the right co-ordination around a project to make the most of it, which includes investment in housing and other regeneration. If you are putting a lot of money and effort into a big new project in a particular area, you need to have all the stakeholders lined up and a proper plan for how you will capitalise on it.
For instance, in some of the big science projects that the STFC do, you are investing in a piece of science equipment that is going to attract researchers and so on, but also you need space for spinouts. You need support for people who are going to develop companies that come out of that project, and you need enterprise zones around it; you need that kind of co-ordination with it.
Q61 Chair: Will the benchmarking data include longer range evaluation?
Nick Smallwood: It doesn’t yet, but there is no reason why we could not consider that as a proposal.
Q62 Chair: It is a real bugbear for constituency MPs when we are told that a project is being delivered in our area and that there will be “great jobs” created, but then you dig in and realise that postcode jobs do not mean someone is local. They can be living in a guest house or a hotel and be counted as local; there are all sorts of ways that the system can work.
The Olympics had a very hard deadline, so it was really hard to butt against it; if you needed a concreting specialist, you needed a concreting specialist now. I will not go into the pain of that, but we have all seen it in our own local areas. We are facing an election this year and both the Chancellor and shadow Chancellor are talking about economic growth, so job creation is really critical, but that is a long-term thing. A six-month job is different from a 10-year job.
Nick Smallwood: What I am seeing is some good emerging work on our projects and programmes that are coming through development now. The prison building programme for HMP Fosse Way has created 460 construction jobs, and 71 of them were for ex-offenders—they really went out of their way to maximise that. That is now continuing in the programme.
Q63 Chair: And are you watching to see if they keep the jobs?
Nick Smallwood: That is where the long-term focus needs to be. In Sizewell, because it is a very long-term project, they are looking at STEM graduates and what they can do for the local community’s employment opportunities, because there is a need for the project directly, so there is a win-win for the community as well.
I am generally seeing the UN sustainable development goals as a sort of checkpoint. It is a really good tool just to challenge and check which ones are relevant to your project, and to check that you are maximising the benefits. Sizewell has a particular problem with water, because Suffolk is really short of water, so they are thinking very hard about how they build this massive project, with huge amounts of concrete to be installed, without adding to the water challenge.
I see it every day on new projects. How we capture that long-term, I think we need to take away and think about.
Conrad Smewing: I am very keen to take that away and think about it. One thing that we could usefully do from the centre is lay out some guidelines on how you might assess and present these kinds of evaluations, because having seen quite a lot of jobs created calculations before, I know that there are a lot of different ways of doing it—
Chair: You are as cynical as we are, but you are too professional to say it.
Conrad Smewing: But there is a reality there that if you look at it in a consistent and sensible way against a proper counterfactual, you can have more confidence in those numbers.
Q64 Chair: And, of course, jobs created locally, where people live and they end up staying, can be very helpful.
I just wanted to go back to this point about gate 5—the end of the programme. You say that that is not enough. Picking up particularly on what you are saying about skills and jobs, do you think that there is perhaps an opportunity for a gate 6—that longer term view? Would you want to call it something like that, or just—?
Nick Smallwood: It might be and maybe it is delivered by others, but I think—
Chair: Ten years on, looking to see—
Nick Smallwood: —the benefits realisation plan needs to go beyond gate 5. If the plan is clear about who is going to do what by when and in what time period, I think that would capture it.
Q65 Chair: That is helpful. I touched on transport earlier. The Jubilee line Ebbsfleet was very slow; the development around Ebbsfleet didn’t happen until a long time after High Speed 1 was created. However, it’s well-known that transport drives opportunity and jobs. So, are we capturing the real benefits?
We covered a bit of this earlier, but do you think, Mr Smewing, that there are better ways in which we ought to assess transport projects? They are great in terms of levelling up or equalities, access to education, access to jobs, housing and other regeneration opportunities around that. We are beginning to get an idea of what they can deliver, because there have been a number of projects, rail and road. Do you think that we could do better at measuring their success?
Conrad Smewing: The first thing to say is that in my experience the Department for Transport’s ability to quantify and assess the benefits of its projects is one of the best in Whitehall. They have a really robust and clear methodology that is widely understood and that the Treasury is comfortable with.
It can be a little bit on the cautious side—I think they would probably say so themselves. They are quite ready to say, “This is the benefit-cost ratio, but there are a large number of unquantifiable benefits.” And as I was saying earlier, although it is right to say that those kind of unquantifiable benefits in the end will be not knowable to the same level that you could, for instance, estimate journey times or journey time savings and those kinds of things, you probably could get a bit further in saying, “This is our kind of ballpark figure, or estimate, under these headings, of what you could do.” However, I know that the DFT is far more expert in this than me and they have very good people who think about it.
Q66 Chair: It is interesting that all the Metro Mayors talk about transport as a major plank. Even if they don’t have it in their powers, often they want it, because they see it as an economic regenerator, but that is a slight segue.
Almost finally from me, which bits of the system do you think need to change and how? What would be your top three asks of any bit of Government. Both of you are in a prime position already to be able to indicate or control—perhaps cajole is a better word than control. What do you think, Mr Smewing?
Conrad Smewing: You go first, Nick.
Chair: It’s not a trick question.
Nick Smallwood: Thinking aloud, really the focus that would have the most impact would be to get a stronger grip of the largest of the Government’s major projects—those mega-projects or giga-projects that really are very challenging and difficult—and to increase the rigour and assurance around those, to increase the quality of cost-estimating and to track and challenge them through their lifecycle, rather than leaving it to arm’s length bodies. I think there is much more we can do in that space.
Q67 Chair: That would be for you to do, do you think?
Nick Smallwood: That would be for the IPA to do.
Conrad Smewing: Within that, I think that cost estimation is one of the things. How we think about that and apply it in our decision-making and spending reviews, and in the Green Book, is still not quite cracked. That is partly for reasons that are outside the control of some of these projects. If you have a big wave of inflation, then you have a big wave of inflation.
Q68 Chair: You can account for that.
Conrad Smewing: Exactly. We talked about evaluation a lot. I still think that is not where it needs to be. If I could wave a magic wand, that is where I would wave it.
Q69 Chair: We have covered evaluation quite a lot but, in terms of making sure that we are changing the system—we have heard from the IPA—is there anything in terms of financial levers that you think could incentivise Departments to do a better job?
Conrad Smewing: One area that is really worth our looking into, and we are looking into it for the next spending review, is duration of budget setting. It is pretty widely recognised that there are benefits to having longer-time budgets, particularly for larger projects, and we definitely agree with that. At the moment we have a patchwork system. Some areas have quite long visibility on their budgets and some less so. We could do with a bit of levelling up. Obviously, there are trade-offs there because you lose some fiscal flexibility in doing that. The Treasury needs to think about that. That is an area that we should definitely look at for the next spending review.
Q70 Chair: There is optimism bias in nearly everything we look at. With failed projects, people always said, “It is going to be fine.” Often, one of the sources of that over-optimism is the politicians. Our colleagues, the Ministers, are having to make decisions. That over-optimism is partly—there is a lot of academic writing on this—because if you are trying to sell the narrative of a policy, to tell the public it is going to cost £x billion is hard. Sometimes the point at which you have got to get the public on board may be a point before you have a business case, because it is a circular thing, isn’t it—which comes first, the chicken or the egg? Am I being pessimistic to think that there will still always be a challenge, with politicians either not wanting to be honest with the public or not being able to be honest because, to get the idea off the ground, they have to float the idea before there is a cost attached, so everyone is always talking about low costs from the beginning, not realistic costs?
Nick Smallwood: We should always talk about ranges until we have done the work. It is very notable that HS2 was 10% mature by the time we took the FBC, so we really did not know how much it was going to cost. Would you have made the decision differently had we known that?
Chair: FBC is final business case, I should say for people listening.
Nick Smallwood: Yes, final business case.
Chair: It was only 10% mature at final business case.
Nick Smallwood: I think being honest about ranges, and how accurate those ranges are, is helpful until we have done the work.
Conrad Smewing: I very much agree with that.
Nick Smallwood: Reality will always come out at the end. That is what I have seen on major projects.
Conrad Smewing: Transparency on the vintage of estimation is a friend here as well.
Q71 Chair: You speak our language, Mr Smewing. We will be pressing the Treasury for more transparency on these things from now on then.
Conrad Smewing: The publication of the GMPP and the publication of business cases—all of that is helpful.
Q72 Chair: And I think what you are saying is that there should be an honesty with the public that you cannot do things on the cheap. If it is going to be big, it is going to cost, but it is about keeping control of the cost.
Q73 Peter Grant: Do we just need to accept that there will always be a limit? Think, for example, of a person who delivers a project that, on the day it has finished, looks a great success. They have got their knighthood, they have got their promotion to a big job somewhere else, and over the next 10 years, if the real benefits don’t appear, they have gone and they have got the benefit from it. Do we simply need to recognise that that will always be an issue, and therefore there will be a limit on how closely we can get people to follow what we would both agree is best practice all the way through this?
Conrad Smewing: I am not sure I agree with that entirely. If you have set up a project in the expectation of benefits and you have delivered what the project is supposed to have delivered—although it will always be the case that, at the time, you are not really in control of the long-term benefits that it is intended will come 5 or 10 years down the line, if you have built the power station that you were intending to build and it is churning out the power, I think that is okay.
Nick Smallwood: I agree. I think it is about delivering on promises, because the external environment can always change over a 10 or 15-year period, but to have a robust plan from the outset and deliver the plan to the point of completion is fundamental, and we are not always doing that. That is step one. Step two is having that longer-term evaluation plan. But wars in Europe weren’t predicted; the 40% inflation since 2020 on material costs alone was not predicted in any of the projects that took off in 2020.
Peter Grant: This may be just an age thing, but I’m not sure I would really call 5 years “long term”, but we will leave it at that.
Chair: Well, from one age range to another.
Q74 Ben Lake: In answer to a question from the Chair on the challenges that government face when delivering major projects, Mr Smallwood, I think you mentioned that one challenge is the dependence on the supply chain for skills. Could you elaborate on what you meant by that?
Nick Smallwood: We are seeing an unprecedented ambition in terms of infrastructure spend in the next five years. We are already seeing competing management teams, so the same names appearing on bid lists in different projects. We are seeing discrete skills shortages; with the welding out of frigates in Rosyth, welders are in short supply. We are seeing civil steel fixers in short supply, which is what encouraged me to take on the work to do a market analysis of capability and capacity. It is not just in the core trade skills, but actually with white-collar workers as well. We are seeing huge external influence from the likes of the massive investment in Saudi Arabia; there is a design house in London where they are taking several hundred designers out of the UK market. We just need to be cognisant of that and, against that ambition to build more than we have ever built, understand whether we have enough capacity to really deliver it. I don’t know what the answer is, but I fear there are going to be some pinch points.
Q75 Ben Lake: As part of the important work that you are undertaking, will we also have a better idea of what impact, if any, that will have on cost and prices? It strikes me as though there is likely to be a higher demand on a very finite number of individuals.
Nick Smallwood: You are absolutely right, so how we manage that space is really important. In the hospital building programme, if we do not get the balance between speed and building market capacity, we are going to see inflationary pressures in that specific market. We need to guard against that. It is absolutely a risk.
Q76 Ben Lake: Very briefly, on the white-collar skills that you mentioned—I will ask another question in a moment—is there any merit at all in looking at whether there is a way of increasing the Government’s own skills in that regard—their own capacity?
Nick Smallwood: You make a really good point. I think we have an overreliance on some parts of the supply chain. We have a large number of contingent resources working on projects. I am of the view that a smaller, more competent and capable onus team is better than a larger team with weaker skills. I think we need to work through that as we go forward, and be smarter at how we manage the supply chain, picking the right partners to do work that they may be better placed to do than we have been perhaps doing in government. That is all in the mix.
Q77 Ben Lake: That is very interesting. It is quite a large piece of work that you are undertaking there, and a big study. Do you anticipate completing it by any particular date?
Nick Smallwood: We are hoping to have some early results by the summer—June or July time, I would hope. It is not a short piece of work.
Q78 Ben Lake: No. I will just go back to what you mentioned about SROs, Mr Smallwood. In your experience, how prompt are SROs in coming back to any concerns raised by the IPA?
Nick Smallwood: We are seeing a very good response, actually. An example of that would be the 70% of projects that returned from red to amber or green within 12 weeks, so really taking on board the advice and recommendations from independent assurance gate reviews that the IPA has organised. We are seeing increasing understanding of what assessed and accredited really means, and quite a good representation now appearing at the Major Projects Leadership Academy. I am requiring all SROs to go on the MPLA—every one. They will not get accredited unless they have been. We are not seeing any pushback. It is teaching them the skills that are necessary to deliver what are really complex and often unique projects anywhere in the world.
Q79 Chair: I have been to the MPLA a couple of times, and one session was very interesting; the room went silent as the former Deputy Chair and I came into the room, because they were discussing how pesky politicians in politics would slow down projects. It then ended up being an interesting discussion about how you need to build in the political challenges of buying in public support or stakeholder support and so on, as well as the constituency MPs concerned—all those sorts of things that matter as well. Obviously, you are looking primarily at the technical end and the delivery end, but do you have any of these discussions with Ministers to ensure that you are building in those other harder-to-pin-down challenges?
Nick Smallwood: Absolutely. We actually started ministerial training to have that discussion, and we would like to resurrect that and do more.
Q80 Chair: Would the Ministers do it with civil servants, or are they separate?
Nick Smallwood: It was just Ministers, with civil servants having the dialogue and delivering some masterclasses. They were very well received. I would like to start that again. I mentioned that we are doing the training of the delivery boards—the decision-making boards—because I think there is a gap there in knowledge and understanding. But you’re absolutely right: all key stakeholders need to be on the same page if you want to deliver a really successful outcome.
Chair: I will throw out a final challenge; I am not expecting an answer on this. Richard Bacon, who was then the Deputy Chair of the Committee, and I were very strongly of the view that all MPs should have the opportunity to go through this, because, as we have seen over the years—particularly in the last couple of Parliaments—you really don’t know who is going to be a Minister. There is no guarantee. I see Mr Lake looking hopeful, from his small party. We have seen stranger things happen. I just want to throw that out there: there needs to be a better literacy about what is possible and what is not. Anyway, I am not expecting you to answer that; that would be a bit unfair.
Can I thank you very much indeed for what is been a really interesting discussion? It may seem a bit esoteric to people watching, because we weren’t discussing a big particular failure, but we know that getting things right—getting those ducks lined up, to quote Professor Grube at Cambridge University—is really important, and that data and delivery, which is your end of it, is really key to getting that across the board. We hope that when we produce our report with recommendations, they land in the right place with the right intent, because whichever party or parties win the next general election, the British public need projects delivered on time and on budget. Every pound wasted is a pound that could be spent on something else or put back in the pockets of our constituents.
Thank you very much indeed. The transcript will be available on the website, uncorrected, in the next couple of days. Thank you to our colleagues at Hansard for that. We will be producing a report at some point after the Easter recess.