Y Pwyllgor Cyllid - Y Bumed Senedd

Finance Committee - Fifth Senedd

15/05/2019

Aelodau'r Pwyllgor a oedd yn bresennol

Committee Members in Attendance

Alun Davies
Llyr Gruffydd Cadeirydd y Pwyllgor
Committee Chair
Mike Hedges
Nick Ramsay
Rhianon Passmore
Rhun ap Iorwerth

Y rhai eraill a oedd yn bresennol

Others in Attendance

Gwyn Llewelyn Cyfarwyddwr, Infrastructure Advisory, KPMG LLP
Director, Infrastructure Advisory, KPMG LLP
Stuart Pearson Senior Associate—Construction, Energy and Projects, Capital Law
Uwch Swyddog Cyswllt—Adeiladu, Ynni a Phrosiectau, Capital Law

Swyddogion y Senedd a oedd yn bresennol

Senedd Officials in Attendance

Bethan Davies Clerc
Clerk
Christian Tipples Ymchwilydd
Researcher
Georgina Owen Ail Glerc
Second Clerk
Ryan Bishop Dirprwy Glerc
Deputy Clerk

Cofnodir y trafodion yn yr iaith y llefarwyd hwy ynddi yn y pwyllgor. Yn ogystal, cynhwysir trawsgrifiad o’r cyfieithu ar y pryd. Lle mae cyfranwyr wedi darparu cywiriadau i’w tystiolaeth, nodir y rheini yn y trawsgrifiad.

The proceedings are reported in the language in which they were spoken in the committee. In addition, a transcription of the simultaneous interpretation is included. Where contributors have supplied corrections to their evidence, these are noted in the transcript.

Dechreuodd rhan gyhoeddus y cyfarfod am 11:00.

The public part of the meeting began at 11:00.

6. Cyflwyniad, ymddiheuriadau, dirprwyon a datgan buddiannau
6. Introductions, apologies, substitutions and declarations of interest

Bore da a chroeso i chi gyd i gyfarfod y Pwyllgor Cyllid y bore yma. Gaf i groesawu Aelodau a gaf i nodi bod clustffonau ar gael ar gyfer cyfieithu ar y pryd neu addasu lefel y sain, os ŷch chi'n dymuno defnyddio hynny? Gaf i atgoffa Aelodau i ddiffodd y sain ar unrhyw ddyfeisiau electronig? A gaf i ofyn a oes gan unrhyw Aelod unrhyw fuddiannau i'w datgan? Na. Iawn, ocê, diolch yn fawr. 

Good morning and welcome to you all to this meeting of the Finance Committee this morning. Could I welcome Members and could I note that headphones are available for translation or sound amplification, if you want to use those? And could I remind Members to put any electronic devices on silent? Could I ask if Members have any interests to declare? No. Fine, okay, thank you very much.

7. Papurau i'w nodi
7. Papers to note

Mi welwch chi fod yna bapur i'w nodi—eitem 7—sydd yn gopi o gofnodion y cyfarfod a gynhaliwyd ar 9 Mai. Ydy Aelodau yn hapus i nodi'r cofnodion hynny? Ydych. Diolch yn fawr iawn i chi.

You'll see that there is a paper to note—item 7—which is a copy of the minutes of the meeting held on 9 May. Are Members happy to note those minutes? Yes. Thank you very much.

8. Ymchwiliad i ffynonellau cyllid cyfalaf Llywodraeth Cymru: Sesiwn dystiolaeth 4
8. Inquiry into the Welsh Government’s capital funding sources: Evidence session 4

Ymlaen â ni, felly, at brif eitem y sesiwn gyhoeddus y bore yma, sef parhau â'n hymchwiliad ni i ffynonellau cyllid cyfalaf Llywodraeth Cymru. Ac ar gyfer y sesiwn dystiolaeth y bore yma, mae'n bleser gen i groesawu Gwyn Llewelyn, sy'n gyfarwyddwr Infrastructure Advisory gyda KPMG LLP, a Stuart Pearson, sy'n uwch-swyddog cyswllt adeiladu, ynni a phrosiectau gyda Capital Law. Croeso cynnes i'r ddau ohonoch chi.

Awn ni'n syth ymlaen i gwestiynau, os ydy hynny'n iawn gyda chi. Mi gychwynna i drwy ofyn: yn eich barn chi, beth yw manteision ac anfanteision defnyddio modelau partneriaeth cyhoeddus-preifat wrth gyflwyno prosiectau seilwaith?

We'll move on, therefore, to the main item of this public session this morning, which is continuing our inquiry into the Welsh Government's capital funding sources. And for this evidence session, it's my pleasure to welcome Gwyn Llewelyn, who is director of Infrastructure Advisory with KPMG LLP, and Stuart Pearson, who is a senior liaison officer on construction, energy and projects with Capital Law. A warm welcome to both of you.

We'll move straight to questions, if that's okay with you. I'll start by asking: in your opinion, what are the benefits and disadvantages of using public-private partnership models to deliver infrastructure projects?

Os caf i ateb gyntaf. Ac fe wnaf ateb yn Saesneg, os yw hynny'n iawn.

If I can answer first. And I'll answer in English, if that's okay.

Popeth yn iawn. Fe gewch chi wneud yn eich dewis iaith.

Yes, fine. You can contribute in your language of choice.

For me, public-private partnerships are the only feasible option for delivering large-scale infrastructure projects. We know that public finances are tight at the moment, so it's the only real solution, and we can deliver at scale by involving the private sector.

I think, if I can add: the decision on public or private finance can get a bit lost in the debate of one's cheaper than the other. For me, this is: how do we bring private sector expertise and public sector expertise together to deliver what's needed and find a pragmatic solution to deliver the infrastructure that's needed? We always say that, sometimes, perfect is the enemy of good, and there are times when we need new schools and we need new hospitals and we need new infrastructure and we have to find solutions to do it. What bringing the private sector can do and what the PPP model does, and what the mutual investment model does is actually better than some of the models that have preceded it across the UK, is that it's designed to create those incentives to align people. If we rephrased PPP as, 'Okay, private sector, you're going to build us a hospital'—and I'll use a hospital as an example—but 'We're going to build you a hospital', well, the private sector have always built our hospitals; we don't have a ministry of works. And 'The private sector are going to maintain the hospital'—well, generally they do that anyway. What we are doing, through the use of private finance is saying, 'Well, rather than pay you upfront for building a hospital, we're going to pay you based on the performance of that asset over time, and we are only going to pay you if that asset works and is delivered and meets the service specification delivered over time.'

And it's that payment-by-performance that creates that alignment of incentives between people and creates a risk transfer to the private sector that means we're getting an efficient flow of capital and protecting the public purse. That's what I see as the primary advantage of bringing the private sector together. There are lots of models through which we can do it. Your question was focused on MIM, but the broader committee is looking at financing more generally, and I think this has to be about: how do we best bring parties together to share expertise, share resource and share finance to get infrastructure delivered that's for the benefit of everybody over time?

Yes, I'm just fascinated by Mr Pearson's first response—that it's the only way of delivering large-scale infrastructure. Clearly, it's not, technically—it simply isn't. So, I was interested as to why you—. Obviously, you're quite deliberate in your phraseology—it wasn't an accident. So, why were you so bullish in that initial response?

It's because there are a number of infrastructure projects, not only in Wales, but further afield, that have taken time to get off the ground. For example, the M4 relief road is—

11:05

Yes, but that's because of political decision making; it's not about construction operation.

You're exactly right, but even after the sort of political decision making, there would be then a period of implementation and so forth and securing appropriate funds et cetera, et cetera. For me, it is time now to introduce an infrastructure pipeline and to deliver those projects. I think it's time for delivery, and I think that's why I made the bullish statement, really—because I see that MIM can deliver the ambitious project pipeline that is being spoken about at Welsh Government at the moment.

I'm interested in the approach that you've both taken in these introductory exchanges, because you're very bullish, and that's good. We see that from Ministers sometimes. But in terms of sharing, as you said, Mr Llewelyn—in terms of sharing financing, I think you said, as part of your response, from my point of view, as a taxpayer, what I'm interested in is Government seeking the cheapest forms of financing, because that gets the best value for my tax pound, and that is almost always Government borrowing rather than private borrowing.

If I can respond on that, I would make a distinction between the cheapest form of finance and best value. I'm with you on, 'We need best value', but it's not all about the cost of finance. The cost of finance coming with risk is different to someone else taking the risk and, therefore, charging a slight premium for it, and is that good value to do that? Someone else taking the risk that construction projects overrun—well, frankly they do. We can all say that the public sector can deliver projects, but there are frequent cost overruns on major programmes. Pushing that risk to the private sector because they are well placed to manage that risk and allowing them to have a slight premium—that doesn't make it bad value for money. It doesn't always make it good value for money either. We have to worry about just taking a cookie-cutter approach and rolling it out, but where the private sector is pricing risk effectively, there's no market failure in the financing market and the financing market is acting in good competition and working efficiently, there's no reason that Government should be borrowing cheaply, loading risk up on taxpayers, because that's all that borrowing cheaply is really saying, and trying to undercut pricing. So, just make the distinction between cost of finance and value for money, that's all—

I just wanted to pick up on your very categorical opening statement as well about the benefits of PPPs. Can I just ask—? You have thrown some caveats in in your subsequent answer, but were you talking in terms of principle or do you draw a distinction between how they should and could work in principle and how they work in practice? Even though you were talking about, 'Yes, they're good at transferring risk to the private sector, delivering by insisting on particular results', assessments of PFI in Wales and, increasingly, on a UK level have said, 'Well, yes, that might be the principle, but, in practice, they don't really transfer risk very well and they don't always deliver what we want in terms of performance either.'

There are good and bad examples. In my experience, and I can only talk to my experience of working in this sector and on these sorts of deals for 14 years, by and large, most of the sectors deliver well, and they deliver infrastructure that would not otherwise have been delivered, and they perform well and the assets are in high performance. There are assets that haven't worked. In some senses, you can't win and you can't lose in these things. If a risk materialises and the private sector loses money, people will point to—'Well, there's been a construction failure', but actually it's the private sector that are taking the risk. If anything, that's the success of the model that someone else has taken the risk where there's construction failure. What there is is a very weak counter-factual evidence base of what the performance is of publicly delivered and funded assets across the UK to compare side by side. So, it's always easy to point at discrete failures. My suspicion and my experience tells me that, if we were to look across a portfolio, I think we'd prove good value for money, but is the evidence base there to substantiate that? Not really, if we're honest. Coming from theory to practice, I think there is failure of the industry in many ways, and you're embarking on a new problem, you've got the benefit of learning the lessons of other people. We develop contracts that are very robust contractually. They are set out contractually to deliver what they are supposed to deliver, but they're complex, and we pass them down to NHS trusts, and to school boards, and to local government and say, ‘Manage this contract. Here’s a 200 page contract that you don’t really understand that’s been put together by a bunch of lawyers and advisers.’ And then, what happens is that the contract largely gets put to one side and deals get managed on the basis of custom and practice—and custom and practice that doesn’t always have regard to the complex contract that’s put in place. Now, sometimes that’s good. You know, there’s nothing wrong with custom and practice if it’s just two parties coming together and managing it well. But at times, it isn’t, and it’s often the private sector that takes the better advice and manages contracts a bit more officiously than the public sector does. So, that’s an implementation point. If you’re going to do it, make sure the skills are in place to manage those contracts properly.

11:10

[Inaudible.]—three questions. The first one: there are always revenue consequences of capital, and that takes money out of money for paying teachers, money for paying doctors and nurses. You can tell me if I’m wrong on this, but I would guess that anybody who is borrowing in the private sector is going to struggle to borrow it at less than 4 per cent. Those are figures that are publicly available. And their profit margin will be pushing 5 per cent. So, they’ve got 10 per cent on top of using available capital. Are those numbers reasonable or would you give different ones?

It depends on sector. It depends on the risk of a particular deal. A road is a bit different to a hospital, and a bit different to a pipeline of schools deals. But for an availability-based deal—the most vanilla form of MIM—I think an all-in cost to capital that’s debt plus private equity that’s coming in, as you’ve suggested, yes, below 5 per cent. If I had to pick a number in the current market, you’re probably around 4.5 per cent, and that would seem reasonable—4.5 per cent. Government capital isn’t free either; there’s a cost of borrowing associated with that, but the premium over it is your consequence.

But Government has some—[correction: some capital]. Government gets capital given to it as part of its settlement. So, that capital, it doesn't actually cost this Government any money; it costs Westminster money, which, although it's at absolute cost, is not one that really affects Wales.

The other question is: you talked about risk and the private sector taking risks, but am I not right in saying that, quite often, you use single-purpose vehicles in order to do these developments and that they don’t run on fixed costs? Because if things start to go badly wrong, they just close the single-purpose vehicle down. Now, if you’re saying that companies are prepared to do developments on the way that the Millennium Stadium was built and other places that were built like that, where all the risks state that you pay a fixed cost and all the risks go through the private sector, and the primary company behind the development is prepared to underwrite it—. Is that what you're saying? Because I find that very difficult to believe.

No. The structure of the deals—again, we're generalising a bit, if I take MIM as the example—will be what we call limited recourse finance. So, the finance will be delivered through some kind of corporate vehicle, whether it’s a partnership or a company, and it will be limited recourse within that. But it will be limited recourse and funded by the private sector. So, if you’re building a £200 million hospital, the £200 million will be submitted by the equity and debt providers to that hospital. That is money at risk. If there are overruns, it is up to them to meet those overruns to finance things, otherwise they will have lost the full amount of their capital that they have put into deal. So, yes, you can’t go and then ask—. You can’t oblige them to say, ‘Well, actually, the cost is now £500 million; give me another £300 million, please’. But they are contractually bound that their £200 million is at risk to sort the problem out, and, frankly, if they don’t, you take the hospital back in whatever state of construction or operation it is for the value of what’s there.  

Just one final point on this, because I think we might come to agreement at some stage. My experience of highway contracts especially is that they come and then—a low tender and claim seems to be a mentality that exists amongst a large amount of the private sector whilst doing work for local authorities and others. I remember once that they found water when they were building a bridge, and they put a claim in for finding water, or there was the time they tendered to build a pavement and didn’t put any kerbs in because they only said they wanted a pavement. I mean, are you telling me that people will do a fixed cost and they’ll take all the risks related to additional costs, and that if they find subsurface problems, then they would cover that cost? Or actually is the risk being shared, and, if the risk gets substantial, it goes back to the public sector not the private sector?

11:15

I think there's always—. The theory is, yes, it's a fixed-price contract. There are always caveats to that. In any construction contract, whatever the form of financing, whether it's public capital or private capital, there are always exceptions to that. There are usually specific carve-outs for things like ground condition risks that are unforeseen and not possible to see. But, the reason for that is quite clear: if you are asking a private financier to take risks that it has no way of managing and no way of knowing on a site, it's going to charge you for them and it's going to charge you more. Is that value for money? Often, the answer is 'no'. Really, this should be about finding the right balance of risk sharing with the private sector, to say, 'Well, what risks should the private sector take and they can price sufficiently?' and what shouldn't be. Let's find the hybrid model that says, 'Well, if there are ground conditions and that's going to push up the cost of finance, does it make sense for the private sector to be pricing those risks?' Let's look at these risks and these deals and find true partnership in a way that makes sense.

Did you want to come in, Rhianon? Very briefly, if you do, because we need to make progress.

Very briefly, I don't want to be out of time. You've mentioned there's no perfect model, except for—obviously, the MIM is an attempt to seal some of those holes. So, in regard to the theme within this, and in some of your narrative already you state that there's a concern over a future loss of infrastructure pipeline. So, in your opinions, do you view the mutual investment model as a model that will safeguard that pipeline moving forward, or do you feel there's more of a risk on investors, bearing in mind the 20 per cent equity holding for Welsh Government?

In general terms, it's a means of delivering that infrastructure pipeline. Coming back slightly to the risk point, it's a question of almost getting those projects ready to be handed over to a MIM funding model—

So my point really would be—and I'll try and be very brief—in regard to that preferred model, the mutual investment model, moving forward, does that make that future pipeline more likely in terms of external investors or less likely?

In my view, yes, provided that—

We touched on performance and I'm just wondering, in terms of monitoring performance, what role would key performance indicators have, for example, in terms of ensuring that a PPP project comes in on time and within budget. Just tell us a little bit about how we can—because you talked about payment by performance, in effect. Well, how best do we monitor that?

It's fundamental to the model, and it's fundamental that the model's not going to work if you don't get it right. That comes down to getting the right measures, and not getting too many measures. You can bombard things with contracts that measure everything and lose meaning. Let's work through what matters to you, what are the right things to measure that really fundamentally drive performance and get measures around them. Scaling those measures of how severe they are before they start to bite is a test that needs calibrating. Because perhaps if you make them too severe with too much bite it will pass on the cost to finance, and you'll say that's not good value. But, make them too loose and it doesn't drive the incentives to great behaviour. So, art or science, I'm not entirely sure, but it's something that's very, very important to get right and focus on. But then there's, 'The contract's great, we've designed the perfect set of measures, how do we monitor and put the team in place to deliver?' I think contract management is something that can be improved from our lessons.

I saw that in the written evidence actually. People tend to be in auto pilot a little bit, do they? Is that the danger?

Auto pilot and—. Contracts get passed down. The people that deliver a deal—these are at times 25 to 30-year deals—people will deliver a deal, they'll work hard for 18 months to get a deal done, and they'll move on in their careers, move on to something else. The detail of those negotiations and why something was done and what was there can get lost. It'll get passed on. Over time, it doesn't seem such an important contract. It'll get passed, passed, passed down the chain to new employees and the knowledge gets lost and the expertise gets lost. You've really got to invest in that and think about how as a Government you support the parties that are going to be running these contracts to be managing them properly.

I was just going to say, in my experience, there are two types of contract, really. There's the one type of once it's signed it ends up in the drawer and is not looked at again until things start going wrong, for what the recourses are. There's the second type, which tends to be associated with collaboration, which is a living, breathing tool to facilitate the end game of that project. So, having KPIs that monitor that, facilitate discussions around progressing that project and having monthly meetings that refer back to the contract is an effective tool of ensuring delivery and trying to avoid dusting off that contract out of the drawer and arguing about it at the end, which, of course, as lawyers we'll enjoy, but it's not in the best interest of project delivery.

11:20

The Welsh Government's strategy is to use up all its free capital—money comes in capital allowances first—and then go out and get additional money. Professor Holtham came along last week or the week before—well, he came along one week—and said that if you work out how much money you want, what you need, then you ought to be borrowing for projects where you can borrow at the lowest amount rather than using it all up first and then having to go out with what might be more expensive items to borrow for. Any views on that?

Yes. I agree with the Professor. The strategy of 'use your money first and then let's see' is a little bit shortsighted. I would say: look at your funding need, to the best you can forecast it; look at how much capital you think you have, and target that capital at where you think it can deliver the most at the best point in the market. For me, that's not necessarily where it can be deployed cheaper or where the private sector can deploy cheaper, it's: where are the points of market failure in the private market that the private market is going to struggle to deliver on a value-for-money basis? That's often the larger schemes that are harder to co-ordinate in the private sector, it's the slightly riskier schemes—with infrastructure, we like to think about buildings, but actually there's a lot of technology. Are they things that the private sector, the financing markets are going to struggle to get a grip with, struggle to understand? Target your capital at those things that the private sector's not going to do a great job on. Leave the private sector to do the stuff that it can do very well and will do better than you will.

Can I now move on to sharing risk, again? I'm not convinced that the private sector actually takes risks. What it does is that it knows it'll get paid because it's the Government and, if it's not being paid, then the level of problems are such that it not being paid is probably not the most important thing affecting that company at that time because the British economy has just collapsed. So, it knows it's going to get paid, and it puts in a number of additions and the costs, et cetera, but if the project starts going wrong, and we've had a large number of contracts, including some roads—in fact, including one road at the moment that has gone substantially over budget—if it was a MIM or PFI project, what's to stop the private sector just walking away?

The money that they've invested. They won't get it back if they do.

You're confident they won't get it back. If they spent £200 million developing something, they're not prepared to go any further with that project then they will not get any of that money back, even though they're leaving it partly done. This happens amongst private sector companies that are working with each other, leaving aside that people get staged payments. My understanding from the private sector, from my experience there, was that you only have very little money at risk at the end because you'd have had your staged payments as you've developed.

The principle that Government is not paying for an asset until it is delivered is established under MIM. So, if there is a construction failure, you are not—construction seems to be the focus; it's not just construction. But you will not be paying for an asset until it's signed off, completed, handed over and met certification tests. So, there is no—. Ultimately, the contracts are usually structured that there's some form of long stop date, so if there are delays, the private sector's got a period to fix. It might try and fix through putting extra capital in. Sometimes, things just take more money. If you wanted to step in in advance of that long stop date and say, 'Look, I don't trust that you're going to get there; I'm going to step in', then you're not giving them the freedom that they are allowed under the contract and would need to step in and make some settlement, that's for sure. But the risk of—. The challenge that's supposed to be there is that, if they fail, we have this concept of 'no better, no worse'. So, it's no better, no worse for you if they have failed or if they haven't. So, that's not to say there is zero payment, but your payment is such just to restore to the position that you would have been in if the asset had been delivered, so they are the ones taking the hit for risks that materialise.

11:25

Well a number of PFI companies—I haven't written down now—actually just closed a single-purpose vehicle down and they got paid a certain amount of money for the work that they'd done. If I do £200 million-worth of work, and I said I've completed the £200 million, if I can show my £200 million is a reasonable expenditure on what I've done so far, you're going to come back and either negotiate me giving you more money or me paying in that £200 million. Are you saying a court would say, 'You might've done £200 million-worth of work, but it’s going to cost another £100 million to complete. Therefore, you're not going to get anything'?

There is a concept of, in a PFI contract—and I'm just being a bit nervous because I can't quite remember the clauses in MIM in the deals that have been done—but there’s a concept of what we call compensation on termination. So, if you terminate a contract, and depending on whose fault that termination occurs depends on the regime that is put in place for how much compensation. You know, if it's compensated because you want to cancel voluntarily or because you've failed to make payment or whatever, then they get compensated at market value of what they've done. That’s established. If they default and you terminate because they have not met their obligations, then the price that is paid is a re-tender price. So, would someone step into the contract? And where there’s a market, where we think there’s a market, we actually run a tender process—it’s called a liquid market re-tender—of saying: what would it take for a party to come in and restate that contract to a performing asset? And the cost of doing that is deducted from the value of the asset, so that you're not out-of-pocket. So, in that sense, you are in no better, no worse. So, in that circumstance, it is absolutely the private sector that is taking the risk of that failure.

I'm enjoying this. Thank you to the two of you for being pretty candid on this area.

Just on MIM, you said, 'Let's look for a hybrid, let's just look for a system that works, that satisfies different viewpoints and different needs.' Do you think MIM is pretty effective as a hybrid in (a) ensuring the biggest possible savings for the taxpayer, even though knocking price as low as possible isn't the be-all and end-all, but also (b) in this business of transferring risk in the most sensible way? That's to the two of you.

The item I like about MIM personally is that, although there is a complex contractual structure, there are elements that are recognisable to the market. So, for example, the delivery aspect of it looks like what I see on a day-to-day basis: a sort of construction project with SPV employer delivering on a project. And because the market recognises that, in delivery it’s something that they're comfortable using, comfortable working among. So, I think that is a huge positive for MIM. And it also allows risk sharing to work in that environment as well, because, again, the market recognises that, they understand that sort of flow of risk, et cetera, et cetera.

And then, the secondary bit, because of the standardisation of the document, around that, sort of the funding element, that can then be pretty much standardised, which I think then allows—and my friend here will probably be better able to comment on that—allows the funding market to understand what it's funding into, what the risks are, because it then becomes a repeatable project. So, it has almost, in my mind, the best of two worlds: it has the traditional construction model at its heart, plus a funding model around it that helps move the funding off balance sheet, which is key. And so, for me, although on first thought, or first look, it looks like an intricate contractual structure, when you break it down to its constituent parts, it is a very usable model.

There are two points I'd like to make. One is the debate between the merits of standard form versus bespoke models for bespoke deals. And if there's an established pipeline—. We'll get different views. You can ask different people at KPMG who work in the industry and we'll all have slightly different views on this, and that's something that I think we have to recognise.

There is benefit to a standard form of long-life contracts: funders get used to certain contracts; you only have two negotiate them once; you don't have the same transaction cost every time you do a deal, because you've standardised contractual positions. Does that mean that the contractual provisions don't always quite fit as a tailor-made piece of clothing to the deal that you're doing? Not always. So, where you've got larger deals, there's always an element of, 'Well, can we get a bit more bespoke rather than just off the shelf?' But there's something to be said for off the shelf in terms of speed of doing deals.

11:30

Do you see MIM as off the shelf, standardised, in this context? Or is MIM a bespoke—?

I think it's somewhere between the two, because there is a lot of standard-form documentation. I can see what you're doing or planning to do with schools—that's actually been tailored quite a lot to fit a different model. So, that's what I'd encourage you to do is not just take something off the shelf and almost enforce it: 'If you want to build a hospital, you have to use this model; if you want to build schools, you have to use this model.' Let's have a transparent options appraisal that says does this model fit, is it the right thing for this deal. Sometimes, it'll be better to use public capital for certain reasons. And if we're using this model, is there anything in particular about this deal that we just want to tweak slightly? That was one point.

If I just finish the other point—and it's an honest reflection—of are these deals always done for the best commercial position. The enemy—and we've just got to be pragmatists here—the enemy of that is accounting treatment. There is a stated policy: we'll use our own capital if we've got it. The flip of that—you didn't quite say it—is we won't use this capital and we won't do these deals unless they're off balance sheet. And the balance sheet rules drive us to some contractual positions. Hand on heart, if we didn't have those constraints there, we might make some of the clauses a bit different and do a slightly different deal. So, we are driven by accounting treatment. Is that great value? Do we want that to be the driver of deals? I don't think you'll find anybody that says 'Yes'. We've just got to be pragmatic and practical as to well; if that's one of the constraints you're operating in, there are things we have to hit. 

Just very briefly. I fully understand what you're saying there about not relying totally on an off-the-shelf model but tweaking the individual contracts. Just going back to what you said earlier—I do that sometimes—just going back to what you said earlier, it becomes doubly important that in the future, years down the line, those details of that individual contract are logged somewhere and are known by the people who inherit it, because you said earlier about how that expertise can be forgotten. So, people in the future aren't always dealing with the exact details of the contract as they were originally set up.

For me, just to clarify, because we're simple souls as politicians, looking at MIM versus PFI, and in the context that PFI has fallen out of favour for both I think political—we don't like it much here, it doesn't feel right—and also I think increasingly financial reasons—the bills are too high, the transfer of risk isn't quite as we like—is MIM markedly better, if that's not too blunt a word, than PFI, the old style?

It's an evolution. It's an evolution of PFI. I think we have to—. It borrows a lot of material and principles from PFI—hopefully, the good principles—and seeks to address some of the key things that have been challenging: what terms have driven exorbitant cost of finance, what's been difficult to manage, what creates the inflexibility of contracts—that's a personal bugbear of mine in working in this industry—and how do we address that, how do we get you better transparency of contracts by giving you a seat at the table, a voice on the board, all those things. They are sensible, practical steps to evolve the model. Does it still have at its roots, in its genetics, PFI? Yes, it does.

There are people around this table who say exactly this: 'MIM is just PFI in another brand.'

I think the evidence for that evolution is how facilities management is dealt with. There's this idea of hard facilities management and soft facilities management. Hard facilities management being planned maintenance, which you can clearly scope for. With PFI, soft facilities management is what has attracted significant press interest. I think I was reading yesterday there was a charge of about £500 for changing a lock, which sounds exorbitant really just for what is a straightforward task. The reason it has ended up in that scenario is because that is an unpredictable facilities management item. So, because the risk has been shifted across, it then attracts a risk premium. So, by taking away that unpredictable facilities management out of the MIM model, that allows more certainty for the contractor to price for it. And the soft FM can be let on a separate contract, which will be at a lower value, and probably get better value for doing that.

11:35

Of course, one of the reasons why PFI fell out of fashion was because it is fair to say that Government didn't have the expertise that you have in your businesses to negotiate and to manage contracts. Contracts were agreed, were poorly managed, and as a consequence, some businesses made profits that were beyond that which was publicly acceptable. As a consequence, the public purse was felt to have been unduly impacted by those decisions, shall we say.

Now, I'm interested in MIM as an evolution of PFI. I take a very pragmatic view about this. I want hospitals and schools, I want trains on rails and the rest of it, and I think that there is a responsibility on the Government in its wider sense of governance to deliver those core services in terms of infrastructure, and then in terms of the services that are delivered via those infrastructures. From your experience of dealing with the Welsh Government, are you confident that the Welsh Government is in a position to be able to negotiate and manage such contracts?

What I can say is that the Welsh Government has taken good advice. It's not trying to do things on the cheap, which is good for founding a new programme. It's taking advice from established and experienced advisers—unfortunately, not from us, but that's a different point. But it's taking all the right steps to doing it properly. It's not rushing the programme, it's taking good advice, and it's bedding it down, and it's doing a root-and-branch review of models that have been done before in Scotland, in England, overseas, and thinking what's the right model for Wales. We haven't been involved in all those discussions to know would we have reached the same view on all of those points—I can't answer that question. But does it seem from the outside you're taking the right steps? I think so.

I think there are pockets of best practice, for example the procurement approach with Transport for Wales and the Cardiff metro. The Welsh Government funded a significant procurement process, which press-wise has been well received, and we'll see over the next few years how that evolves. I think Welsh Government's role there is to really share that best practice down at local government level, because local government procurement teams, for example, are restricted funding-wise, and are less comfortable with larger procurement. So, I think there are pockets of best practice that could be dragged across into the MIM model, and I think Welsh Government's role is to really promote that best practice and assist with that.

And in the MIMs, in terms of contractual relationships, I think, Mr Llewelyn, you mentioned earlier about having standardised contracts, which have advantages, but elements of disadvantage as well. My recollection many years ago now in the construction field is that new contracts, when they were introduced 20-odd years ago now, made a significant difference to the ability of clients to work with contractors to deliver projects to time, to budget, and that made a significant difference then to a range of public and private businesses and projects. Do you think that MIM has the capacity to create that sort of partnership approach, which isn't necessarily the adversarial approach that you sometimes have between clients and contractors, but a partnership approach where both have an equal desire to achieve those same objectives within budget, within timescales, and to quality standards?

I think, without being pedantic, whether I could land on the word 'equal', I'm not sure. Does the model incentivise both parties to come together to find solutions to problems? Yes. Does it go further than PFI has done in the UK in forcing that to happen? I think yes. So look; it's definitely the right thing. We've got to look at problems delivering infrastructure in the UK. We've got a UK construction market now that's on its knees. The idea of, 'Let's beat up our contractor market and tender them to low price, tender them to take all the risk, and then pat ourselves on the back that we've done a great job' and then, two years down the line, they're struggling, they're going bankrupt and deals are falling or they stop tendering—that's not a partnership model, either. So, we've got to work at a partnership model on a deal basis, but there's a market management piece of sensible governance of the construction industry, ways of interacting, Government setting examples on how to work with industry with fair allocation of risk, fair partnership, fair ways of managing contracts that creates a sustainable supply chain for the future. And I think MIM is good at doing that, the contracts are good at doing that. The proof is, actually, the people in the room doing it day to day, not just what's written on paper. 

11:40

I think the form of contract you're referring to is the new engineering contract, or the NEC. That contract alone I don't feel delivered the change. It was really a cultural shift to manage a contract in the way that that form of contract envisaged. 

Exactly. I agree, but it wouldn't have happened just with the contract alone. It needed the contract to trigger it, to move away from the sort of lump sum, fixed price, fixed scale, risk being shunted—

Exactly. And it encourages those more adult discussions, I think, around risk, so risk is not something that is viewed as something just to get rid of, it's viewed more as a cake to be shared around and the party that is best placed to take the risk should handle that risk.

Yes, and part of the—. I would expect your view would be that part of the role of Government, of course, is to de-risk projects, to use the power of Government to be able to eliminate risks at the beginning so that you can minimise difficulties in the market in terms of raising capital, but also then ensuring that the project itself is seen as a project that is attractive to market funding. 

And act as a co-ordinator and sponsor in a way that only Government can. 

And I think the de-risking point is key. There's no point in handing it over to a MIM model without going through that process of working out whether it's a feasible project to be delivered by MIM and to de-risk some of the elements, like those mentioned this morning: site condition, risk and so forth—to go into it knowing what the potential risks are and to avoid those sorts of cost overruns and time overruns that are associated with the construction industry at the moment. 

Very briefly. One of the recognised issues and safeguards within the mutual investment model would be that stringent operational monitoring over the longer term of the projects through performance management, through key performance indicators. So, within that, that's a recognised loophole that's been mitigated within the model. So, on your point about dusting off that live document, hopefully—and the proof is in the pudding, isn't it—it's going to be part and parcel of that, moving forward. So, I don't know what your view on that would be. 

Only repeating what we've said, to some extent: I think the contracts do that. Honestly, with most PFI contracts, the contracts did it fairly well, anyway. The proof is—

Yes. It's putting the management process and governance around it to make sure it's done—a move from theory to implementation. Good contracts can be poorly implemented. 

Thanks, Chair. The PFI contracts were largely criticised because of their complexity. Are MIMs simpler, and do you think that the structure of the MIM eventually will make them more effective in terms of management?

They're still at the complex end of contracts. Compared to the NEC construction contract, it's at the complex end of deals. And if you're trying to raise significant amounts of private capital and bring in all the contractual protections that you need, they're going to be at the complex end. Does it step down in complexity a bit from PFI? I think it replicates a bit in a well-understood market. Where you're fortunate in Wales and across the UK is there's a very well-established project finance market with a lot of expertise. You have the opportunity to do this, which many countries do not, because you've got the expertise to manage it. Hand on heart, could this model in a few years be said to be complex? Well, it is complex. You've got to recognise that when going into it and put appropriate arrangements around it.

It's one of the things that's been attempted to be addressed through the structure. We say PFI as if PFI is one thing, because there's lots of evolution of PFI.

11:45

Well, no, the right steps to make them more flexible have been put in. 

I think with any project or contractual structure that has a number of moving parts, it is going to then require a structure around it, which, at the outset, looks quite complex. But construction industry supply chain management, that is something that is regularly seen. People in that market are used to managing those interfaces, so there's nothing at that end of the MIM model that is particularly unusual, I don't see. I think, at first look, it may appear complex. I think when it comes down to constituent parts and each individual's role in there, then I think there is a bit of recognisable simplicity. And the flexibility element, for me, comes out of this de-risking element, and putting a good specification in as to what's required at the outset, and I think that's where the flexibility comes—it's that key de-risking and making it clear what needs to be delivered at the end. 

I think it was in your evidence—the written evidence, that is—that the biggest challenge for Welsh Government is whether sufficient revenue can be generated for assets to pay for them. Can you elaborate on that, and is it something that the Welsh Government should be paying a bit more attention to? 

It applies to infrastructure more generally across the UK. This session, and rightly, is focused on financing. How do you pay for an asset over time? Do you pay upfront fully or do you pay over time? But either way, you pay. And the real question for the infrastructure pipeline in Wales is: where is the money coming from to pay? It typically comes from users—a toll road or charging people for using certain types of infrastructure, or it comes from tax, because we don't like to toll roads generally. We don't charge people for using hospitals and schools directly, so it comes through tax. So, the real question is: are we creating the infrastructure that's generating economic growth, generating social growth, that's going to raise new revenues that will pay for it over time? If it does, then we can defer the cost or pay for it upfront. Really, this is about funding being, 'What's the benefit of these things and is it generating economic benefit that can be turned into real financial benefit to meet those payments over time, or is it not?'    

That's really a question that the Government should be asking, whatever it is.  

Whatever it is. It's agnostic to the type of finance being used, but it's a far bigger challenge than one of financing. The financing we can solve; the question is: is there enough tax take, frankly, to pay for the infrastructure pipeline that we want, or if not tax take, what other forms of revenue can we generate to do it? And that's very asset dependent.   

Thank you. In regard to the safeguards—the strings attached to the mutual investment model—my question is a little bit similar to the one I asked earlier, really, in terms of the ability to bring in investment around this model as a choice of vehicle for the future. So, in that regard, we've got a couple of very important projects ongoing at the moment. From your perspective still, what are the challenges in attracting investment around this model for the future?  

If I go first, I think the first thing to say is that MIM has been structured—and very deliberately, from the outside—to create a financial product that's investment grade, without getting too technical. So, the debt elements of these will hit investment grade thresholds that investors will meet. That's where there's liquid money, that's where money is very well priced. And there is no shortage of money, whether it's UK capital, overseas capital. We often hear people say, 'Let's chase overseas capital', but there's no real need. There's lots of capital around for well-structured, well-governed projects, and MIM delivers that. So, I'm not concerned from either an equity or a debt perspective that the money's there. So far, you haven't even embarked on any real big mega-projects of several billion that might have to test that theory again. For the size of the projects you're doing, there's plenty of capital. 

Where are the challenges? I come back to the construction industry, actually. That's where my biggest concern is. Through your market testing, before these projects, I suspect it's being tested and, hopefully, with the right assurances. Are the right contractors in place that can deliver, because private finance comes with strings attached for them as well? Where the investor is putting money in, they are taking a risk, but they do try and push that risk to the construction company that is ultimately the one, and they'll require liability caps. Typically—I don't want to over-generalise—but liability caps within construction contracts under project finance will be, order of magnitude, 50 per cent of the construction cost. So, there is recourse back to the construction companies, but overruns at 50 per cent, again with caveats. Now, the construction industry in UK is running at 2 per cent or 3 per cent margins, asking for 50 per cent cost overrun guarantees. A lot of the UK construction companies have said, 'No thanks. No thanks anymore; we've had our fingers burnt in that market. We're not well capitalised enough to take those sorts of risks.' There are lots of overseas construction companies now sniffing around—lots of Spanish contractors, French contractors; good parties, nothing wrong with them being overseas contractors. But I think there's something to be said for just making sure that works, and to explain that a bit more. When the overseas contractors come in, they'll still need to tie up a Welsh supply chain and a local supply chain, which is important. 

11:50

So, obviously, in regard to that 20 per cent equity share, and the potential for profit then to come back, what's your view on the national audit office statement that it is unclear why the public sector is willing to take on the risk of equity which is most exposed to project performance, because, obviously, it depends on your prism of looking through it, who we're protecting here? But, in regard to that fair partnership that you talked about earlier on, in the round, bearing in mind that there is no perfection, and no perfect model, do you feel that that's a fair comment from them? I don't know if you've got a view on that statement in regard to that 20 per cent equity.

I'm not that familiar with the statement that you quoted to fairly comment on it. Is the principle of equity—? Having the private sector at first loss, which is what equity is—. The first pound of overrun in any contract goes to equity. It takes quite a long time before the debt providers get hurt underneath it. Equity is the one taking the real ups and downs at the risk of first loss. Is it fair that they take a bit higher profit for it? I don't have any objections to that. This is about running a competitive process to make sure that every part of that supply chain, whether it's the construction company, the equity, the debt, is all competed, and fairly competed, not just driving everyone to lowest cost, to make sure that you're getting assurance that you're getting best value from the market. 

Okay. In regard to the type of projects—we sort of touched upon this—that you think are best suited to the mutual investment model, could you extrapolate a little bit further on your comments earlier?

Yes, sure. It's those that are scalable, so it's the big value projects, and it's those that have that can be clearly defined, I think. Highways are a very good example. It's clear what you require out of a highway project. It's clear, to an extent, where risks sit with highway projects. So, a project like that is ideal for the mutual investment model. Projects that are not are probably the more innovative projects, where money is being spent speculatively, say, with carbon reduction homes—something like that—to try and test a market. It would not be suitable for projects like that. So, I think it's 'scalable with certainty' are the projects that I see—

The only thing I'd add is stability of requirement over time. You're signing up to a 25-year deal on a contract. Yes, we'll build in flexibility on how you amend contracts, but you don't sign up for a 25-year deal if your requirement moves after too many years. So, stability of requirement. Where does that lead you? Well, roads are pretty stable. We can talk about impacts of autonomous vehicles and things, if you like, but the general feeling is we're still going to need roads and highways. We're still going to need schools. We're still going to need universities. We're still going to need hospitals. Will you need more protection for—. To take the context of a hospital as a living, breathing building, the building is only part of it; the people in it are much more important, and does the contract need to be a bit more flexible or at least more actively managed than a highway, a strip of tarmac, and just making sure that the tarmac's in good nick? I'm being slightly facetious, but you do different contract management for different types of assets, but stability of requirement is important. High-technology assets, evolving technology assets, or assets with unproven technology, they're probably not right for your MIM. 

Okay. So, in regard to the generic private finance initiative model of old, and the evolution in this moving forward for Wales, we're pretty much clear on what's best to be funded through any type of model around this, are we, in terms of what you've just stated? So, in terms of lessons learned around the projects that we're now carrying out in Wales, they're very similar to the—

I'd say similar, but there are lessons in it. Through the building schools for the future programme in the UK, you've got your schools, and we had a good go at putting ICT provision in schools into contract. I don't think it worked. That's an example of ICT provision being hard to specify, moves over time, evolves too quickly. Take it out. Take it out. What is stable—the building, the need, and the fabric of the building—but take out the soft FM of people trying to change locks and things.

11:55

So, my question, to be more clear really, is that, in terms of the current mutual investment model projects that are ongoing, really, those lessons are already learned from earlier on, retrospective projects, aren't they, in terms of what's suitable?

We're coming to the end of our session, but I just want to ask finally about any other particular PPP models. We've spoken about PFI and we've spoken about the mutual investment model. Are there any other particular models that you would encourage the Government to consider in terms of generating private capital investment?

I think that one that's worthy of mention—there are a few—is the regulated asset-based model. That's more of a true partnership of—. It works better for utilities, where there's a third-party income stream and things coming in, but where there's a third-party income stream, we can look at—. Well, we can allow the private sector to run and operate and charge for infrastructure. We need to regulate so that we maintain our control and influence through regulation rather than through funding, and I think that's a model that can apply and has particular uses, particularly around the energy sector and utilities sectors across Wales.

Okay. There we are. Can I thank you both for attending this morning? You will be sent a transcript of this session just to check for accuracy. We're very grateful to you not only for your written submission, but for the oral evidence you have provided this morning, which will inform us in our deliberations on this inquiry. Diolch yn fawr iawn. Thank you.

9. Cynnig o dan Reol Sefydlog 17.42 i benderfynu gwahardd y cyhoedd o weddill y cyfarfod, y cyfarfod ar 23 Mai 2019, ac ar ddechrau'r cyfarfod ar 5 Mehefin 2019
9. Motion under Standing Order 17.42 to resolve to exclude the public from the remainder of the meeting, the meeting on 23 May 2019 and the start of the meeting on 5 June 2019

Cynnig:

bod y pwyllgor yn penderfynu gwahardd y cyhoedd o weddill y cyfarfod, y cyfarfod ar 23 Mai 2019, ac ar ddechrau'r cyfarfod ar 5 Mehefin 2019, yn unol â Rheol Sefydlog 17.42(vi).

Motion:

that the committee resolves to exclude the public from the remainder of the meeting, the meeting on 23 May 2019 and the start of the meeting on 5 June 2019, in accordance with Standing Order 17.42(vi).

Cynigiwyd y cynnig.

Motion moved.

We'll now move into private session. So, I propose, in accordance with Standing Order 17.42(vi), that the committee resolves to exclude the public from the remainder of this meeting and also the meeting on 23 May this year, which is next week, and the start of the meeting after that, which is 5 June. Are all Members content? Yes. Thank you. We'll move into private session, then. Diolch yn fawr.

Derbyniwyd y cynnig.

Daeth rhan gyhoeddus y cyfarfod i ben am 11:57

Motion agreed.

The public part of the meeting ended at 11:57.