Methods of Funding Capital Investment Projects Inquiry
Good afternoon and welcome to the Finance Committee's 11th meeting in 2008, in the third session of the Scottish Parliament. I ask everyone present to ensure that all pagers and mobile phones are turned off, because they interfere with the sound system. We have received apologies from Tom McCabe. No other apologies have been received.
Today's meeting is the committee's third evidence session in our inquiry into methods of funding capital investment projects. We will hear from three panels of witnesses. We have with us our advisers, Nathan Goode and Marianne Burgoyne, from Grant Thornton.
On our first panel we have a range of witnesses from companies that have advised the public and private sectors on investment projects. Paul Brewer is a partner at PricewaterhouseCoopers; Amanda Methven is a partner at Dundas and Wilson; Jenny Stewart is head of infrastructure and government at KPMG; and Michael Watson is a partner at McGrigors. I welcome you all and invite each of you to make a brief opening statement.
Paul Brewer (PricewaterhouseCoopers):
As the convener said, I am a partner at PricewaterhouseCoopers. I work in the financial advisory leg of the business, in the part of the firm that is focused on capital procurement. We are focused primarily on giving clients advice on effective capital procurement and how to get value for money from it. We advise on that subject not just in the United Kingdom, but globally, although it is an area in which the UK is probably the biggest source of thinking and ideas in the world.
My experience includes a great deal of private finance initiative activity, as part of which I have acted for Government and the private sector. In addition, I have examined a range of other models, including the non-profit-distributing model that is used in Scotland; models that try to do similar things to PFI but without the finance; and partnering models, which do not have a financial element but which consider how the parties could work together more collaboratively to achieve the same value and effectiveness in their infrastructure procurement. The perspective that I bring, I hope, is one of different approaches to financing capital projects.
Amanda Methven (Dundas and Wilson CS LLP):
I am a partner at Dundas and Wilson, which is a legal firm. We have been heavily involved in infrastructure projects for around 15 years. In that time, we have provided advice on a range of such projects. We advise the public sector, in all its guises, on traditionally procured projects, privately financed projects, projects that involve variants of those models and joint ventures.
We act for Scottish Water, the Scottish Government, local authorities, universities and health boards, so we are well versed in the context in which public sector agencies operate and the different dynamic that affects their capital investment decisions.
We have been at the forefront of advising on private finance, which is a relatively small but significant element of capital investment, for the past 15 years, too. We have seen the policy develop from inception to becoming a mature market now, with the model itself being refined and variants on the model developing, such as the not-for-profit model. We are well versed in that and can comment on what works in the context of PFI, what lessons have been learned, where the model works and where it can be improved.
As head of infrastructure and government at KPMG, I am responsible for all our services to the public sector. I also sit in our global infrastructure projects group and have advised on major infrastructure projects for the past 15 years across a range of sectors: transport, health, accommodation, education, waste and so on.
The committee might be most interested to know that I led the financial advice to Argyll and Bute Council on the non-profit-distributing model and pathfinder there. I currently advise Transport Scotland on the potential applicability of that model to the roads sector. Some committee members might recall that my colleague Bill Howat and I gave evidence to the committee in early October as part of the budget review. I have a broad interest in the overall infrastructure plan for Scotland.
I will pull out two key points from the written evidence that we submitted. The first issue on the financing side is that no one size fits all, which is the message that I tried to give in our paper. For the £3 billion of capital investment per annum in Scotland, a range of funding models is available to us, and our paper considered seven of them, I think. The key issue is very much around deciding what model will offer best value for money, although there will be interesting debates around how that is calculated.
The second issue is one of delivery. Understandably, the new Government has its own priorities—it published its own infrastructure investment plan a few weeks ago. One of the issues for the market as a whole and for improving public services is that now is the time to up the pace in delivering the capital investment projects that will support the growth in the economy and improve our public services.
Michael Watson (McGrigors LLP):
I am a partner in McGrigors LLP, which is a law firm. I am head of the banking, energy and infrastructure team, which has been involved in public-private partnership/private finance initiative infrastructure projects over the past 15 to 20 years, and with PPP/PFI since its inception. We have seen Scotland become a market leader in the delivery of PFI infrastructure projects, with the highest number of projects per capita delivered to date. Our firm has been involved in more than 160 PFI transactions.
I will draw out three points from the paper that we submitted to the committee. The first point is about the role that Scotland has played in developing the PPP/PFI model. Over the years, that model has evolved, leading to a strong pipeline of delivery of projects. However, there is a bit of a hiatus at the moment in the delivery of projects, bearing in mind the long lead time that they have in procurement. Certainly, from the industry's perspective, we are hearing that people are keen that projects move forward.
Secondly, the model and its variants are developing, or moving forward, in 97 countries, and much of our practice internationally has developed as a result. As I highlighted in our paper, there is great interest in the model in France and Germany and further afield. That means that the market is competitive internationally for investment, contractors and so on.
Thirdly, I will pick up on the non-profit-distributing structures that are being promulgated in Scotland and elsewhere. In Northern Ireland, for example, we have worked with a number of non-profit-distributing structures in which there is less risk transfer from the public sector to the private sector, but a very low cost of funding. I know that a reduction in the cost of funding is one of the key objectives around the variants that are being suggested. There are models that we can learn from internationally and within the UK.
Given the range of expertise that the witnesses represent, I thank them for their presence. I will start with a general question.
Today we are focusing on the lessons learnt from recent experience. Some submissions mention the perceived benefits of the PPP/PFI model. For example, the Dundas and Wilson submission says that "improved deliverability" of projects and a focus on "whole life considerations" are among the claimed benefits of such models. Which of those benefits are necessarily dependent on the PFI model? Could they be replicated or achieved with any model if the right approach was taken?
I am currently working on a project that is trying to capture some of the benefits of PFI without using external debt, which is proving difficult to achieve. We must consider the components of the PFI model. The public sector buys complex assets but deals with a single party owned by the people who are identified, in financial terms, as the equity providers. The equity providers have a role in bringing together all the capabilities that are required to deliver the project. PFI contracting involves not just creating the asset or delivering the associated services in the short term but delivering the asset, providing associated services and maintaining the asset throughout its life. That is difficult to achieve without someone in the role of an investor with something at stake. With PFI, there is a holistic approach, whereby a single contract is entered into for the through-life services associated with an asset. It would be difficult to achieve that with another model.
On external finance, equity has an absolutely essential role to play. We have found that lenders bring an incredibly strong, effective focus. One of the reasons for that is that although they are not engaged actively in delivering the project, they have an enormous amount at stake in the outcome. Lenders could face significant losses if projects are not delivered effectively. That means that they take an extremely close interest in ensuring that projects are delivered properly. Consequently, they bring high standards of discipline to the delivery of projects, from the stage at which the private sector is bidding to deliver them through to their implementation.
Some departments, such as the Ministry of Defence, say that there are great benefits in that approach, but ask whether we can capture those benefits in other ways, such as bringing in that due diligence without a lender. They are attempting to do that outside the PFI model, but are finding it difficult to get the hard and objective focus on getting it right first time that a bank brings. Equity is the integrator. The banks bring the discipline.
The final thing that I think is difficult to achieve outside the PFI model is the durability of the structure. Given that the financial parties involved are extremely incentivised to sort out problems, the public sector is shielded, in a management sense, from having to intervene in problems. It would not be so protected if it contracted conventionally. There is an array of examples of PFI projects that have gone wrong for the contractor, from which the end customer would have suffered badly under traditional procurement methods. The financial parties that have an enormous incentive will go to enormous lengths to ensure that the contract is delivered to the contracted standards. It would be extremely difficult to achieve that outside a PFI-type structure.
I wholly endorse what Paul Brewer said. The public sector has learned lessons in the evolution of the PFI process. It is now a much more intelligent client when it procures capital projects, and it can bring some of that discipline to projects for which PFI is not appropriate. The examples that Paul Brewer gave are not relevant to all projects; obviously, PFI is appropriate only for certain types of project. However, the public sector can learn from the process that has developed around the delivery of PFI projects.
The public sector has become a more intelligent client in that it has learned about structuring projects, coming to the market with projects that are well developed, having a sensible pipeline of projects, holding gateway reviews, taking into account whole-life costings, and considering the entirety instead of just the cost of capital projects.
I should say that the advisers are entitled to disagree. If they want to make any comments, they should feel free to do so.
The key benefit of PPP structures is that projects are delivered on time and on budget. That is well documented, and it is true of most projects. In the 20 per cent of cases in which projects are not on time and on budget, they miss by only a fraction. To pick up on Paul Brewer's point, that is because the parties have incentives to work together to get things right. Because the private sector bears the risk, it is much more likely to deliver on time and on budget.
It is interesting to consider whether those incentives can be replicated in different models or whether that is impossible. The not-for-profit-distributing model does not have equity in it, so it does not have the equity driver of PPP structures. It is possible to work on ways around that. Ultimately, however, the important thing is the contractual structure and the contractual relationship between the public and private sectors, as Amanda Methven said. I would not want the committee to forget that, even in traditional procurement, there is in effect a large contract between the two sectors. What we are talking about is a different contract between the two, with the private sector bringing in the financing on top.
The question focused on whether PFI is the only mechanism under which the benefits can be achieved. I endorse what others have said about the benefits that it brings. It should be possible constantly to evolve the structure and improve aspects of that model, but I would not recommend a wholesale departure from the contractual structure, for a couple of reasons.
First, as the committee heard, PFI incentivises the delivery of projects on time and on budget and focuses on the whole-life cost of the project. That ensures that a school that is built and delivered on day one is returned to the public sector at the end of the 30-year contract in the same state in which it was delivered. That is a real benefit in relation to the overall costing and deliverability.
Secondly, I return to the point that I made earlier about domestic and international competition for capital investment, contractors and so on. In the past 10 to 15 years, the PFI model has been developed to a point at which it can deliver projects quickly and efficiently. If we move to a new structure, people will need to relearn and to develop new processes. That will have the disbenefit of slowing down the delivery process, and there is also a risk that Scotland will be a less attractive place for external investment.
We have heard—and it is evident from the written submissions—that lessons have been learned as PFI/PPP has evolved over time, whether because the public sector has become a more informed and astute purchaser or because competition in the private sector has driven progress. It would be helpful to have some concrete examples of how that has manifested itself. Also, at least a couple of the witnesses have said that more can be done to drive down costs or drive through more benefit for the public sector. Again, it would be helpful if they could expand on that point and give some tangible examples of where improvements can be made.
As everyone has said, the model has developed. On concrete changes, the obvious change has been to financing in the not-for-profit structure. The concordat between the unions and the previous Government helped to make PPP, as opposed to PFI, more acceptable to public sector procurers. Some issues that are related to technical designs have also moved things forward.
On issues to explore further, I have worked in a couple of areas on specific projects that have been helpful. One such development is where elements of the public sector have injected some capital into a project at the outset. If a public sector authority cannot afford to construct the whole scheme itself, it can inject some of the capital upfront into a PPP project.
Picking up on the international point, the other area that would be interesting to pursue is that of Government guarantees. On the international stage, we are seeing that people are not as worried about balance sheet issues on a number of projects that access cheaper funds because they have the backing of a Government guarantee.
We have been talking as if PPP is a major part of infrastructure projects, but across our programme, a lot of transport projects are done through Network Rail, which is a not-for-profit organisation. The large-scale investment in the water industry is all done through a contractual structure involving Scottish Water Solutions. In fact, PPP is a small element, and lots of other interesting things are going on. I hope that that answers the question.
A specific example of the evolution of procurement methods in an area that, in Scotland, has attracted some controversy but which has had a slightly different tale to tell in England is the story of the prison service and how, in England, it has used private finance in quite a sophisticated way.
The introduction of private prisons in England resulted in a couple of changes. The first is that that mobilised the market into taking more notice because it was being given the opportunity to do things differently and more effectively. The change helped Her Majesty's Prison Service to create a programme of asset delivery when it needed to accelerate the creation of new prison places. That programme and its delivery through PFI resulted in a very different delivery model and an enormously lower cost of delivery. Part of that low cost of delivery came from the freedom that PFI gave contractors to think about how the design and operation of prisons interacted, which allowed them to take a lot of the cost out of design. Another part came from the fact that there was a programme, rather than single, discrete projects, so people were willing to invest in getting better solutions because they could see a whole series of projects going ahead. After that initial, successful programme, HM Prison Service stood back and asked how it could use the lessons that it had learned to make it a better procurer. It took a lot of those lessons into conventional procurement and ran a mixed economy in which it tested PFI projects against non-PFI projects—it has continued to do that.
That demonstrates how the public sector can both learn and keep benchmarking PPP approaches against non-PPP approaches and ensure that its objectives are best met by the approach that fits the outcome that it is trying to achieve.
One aspect I would pick out to develop, which Jenny Stewart raised, is the move in PFI and PPP-related contracts towards a more effective partnering approach between the public and the private sector. Examples of that approach in Scotland can be found in some of the structures around the early NPD projects in which there was representation on the board of the company and more transparency and accountability in the way in which the company operated.
Another example from in England and Wales is the local improvement finance trust structure, in which there is a genuine partnership between the public and private sectors. That was always the intention behind PFI but, as it has evolved into PPP, its structures have become more sophisticated and effective, and there has been more in the way of partnership operations. Transparency and accountability are key aspects of that.
The traditional PFI model, or PPP model as we understand it now, has refined over time, and lessons have been learned. Analysis of some early projects will probably not be representative of more recent projects. Financing terms have considerably improved over time, and the cost of projects tends to be lower. There is much better understanding of the types of risk that it is sensible to transfer and of the types of project to which PFI is applicable. Such projects include those in which there is reasonable certainty about the need for the underlying asset and around which fixed-price contracting is deliverable.
On the question of the not-for-profit approach and refinancing, when we first started out on PFI there was no refinancing gain, and the public sector was not seeking to share refinancing gains. The policy on that has evolved over time. Now, projects are structured in such a way that the public sector can share in refinancing. The big benefit of the not-for-profit, or non-profit-distributing, model is that the public sector can now have a say in when refinancing can occur, which was an issue that had led to tension previously. Moreover, the gains from refinancing might be more illusory than they were for the early projects—margins are much finer now, and the gains might not be there.
The big benefits of the not-for-profit model lie, first, in capping the returns to investors—and the investing community is at a stage where it is willing to accept that; secondly, in the fact that the public sector has a share in deciding when refinancing can occur; and, thirdly, in greater transparency in how the special purpose vehicles function.
I will come to the subject of risk in a minute. First, I have some questions about the cost of capital. I am not sure that we always compare apples with apples when we compare the PPP model with the traditional method of funding. The KPMG submission, whether we agree with it or not, is well written, certainly compared with the Bank of Scotland paper, which is one of the poorest that has ever been presented to a committee anywhere. In the fourth paragraph of page 17 of the submissions paper, KPMG states:
"Now that PFI/PPP is well understood by the market the difference between PWLB and PFI/PPP rates have reduced considerably. In the last 6 months, we would typically see a difference of some 60bps (0.6%) above PWLB rates."
PWLB stands for Public Works Loan Board.
Later panels of witnesses will discuss how we measure the cost of capital properly—leaving aside the potential impact of the credit crunch—and in particular the question whether the average weighted cost of capital and the internal rate of return provide the best measure of the cost of capital. I am not sure whether you have had the chance to read the papers from Margaret and Jim Cuthbert. If we apply the net present value approach, we come up with a completely different answer compared with measuring the cost of capital using the average weighted cost and the IRR. We are always suspicious of jiggery-pokery from accountants and economists. I would like to hear your comments on whether that is the right way to measure the cost of capital.
Who wants to answer on measuring the cost of capital?
As the KPMG paper has been referenced, maybe I had better go first. No one has ever suggested that the cost of capital in a PPP or a PFI project would be less than it would be if the PWLB was approached directly—that has always been the case. The question is, what is the differential?
The second issue is the fact that the cost of capital is a small element in deciding the overall value for money of a project. Typically, in considering a scheme and making the comparison, it is necessary to consider what the various cost components on the public sector side are—what the capital cost is; what the associated operational costs and so on are; and how the cost is built up. Those costs can then be compared with the annual unitary charge that the private sector proposes. The private sector will say, "We will charge you X over 30 years." That comparison is the important driver of value for money. It takes everything into account, including the risks that are transferred from the public sector to the private sector and whether they outweigh the slightly higher cost of capital that is involved in going down the private funding route. The whole issue must be looked at in the round.
I want to take the opportunity to clarify the situation, as I have read a different analysis in the newspapers. The newspapers will say that the capital cost of a particular project was originally envisaged to be £100 million but it is actually costing X—say £300 million—under the PFI. On the face of it, that looks dreadful. However, that £100 million of capital investment would have paid only for the base building at the outset. The capital cost under the PFI/PPP is not just for the core building but for all the operational maintenance and so on for 30 years. I take your point, but it is important to compare apples with apples.
I hear what you are saying. However, if you compare the Cuthberts' papers—which separate out the capital cost of the building and the unitary cost of running the service over, say, 30 years—with the use of NPV and IRR instead of the typical measures, you find that there is a huge difference.
In addition, the capital charge that is made under the traditional method of funding must be brought in as well. You say that there is a difference of 0.6 per cent, but when the capital charge that is applied to PWLB projects is removed, there is a substantial difference.
I agree. The capital charges point applies only to the health service. You are absolutely right to say that it applies on either side of the equation, so it should not be taken into account. I have not examined the Cuthberts' papers or their methodology. I had a quick look at the papers this morning, and they are very interesting.
As a professional adviser, I advise the public sector on particular deals and say, "Okay, we have come to the end of the process. You now have a fixed price from the private sector. Does it stack up in value-for-money terms?" I calculate that on the basis of a comparison against a public sector comparator and I examine the overall value for money, not just the cost of capital, which is only a small part of the picture. Overall, I am happy to put my professional reputation on the line in relation to the projects that I have advised on and say that, in those particular circumstances, I am content that the analysis was sufficiently rigorous to justify a value-for-money equation in favour of PFI/PPP.
Once you have had a chance to read the Cuthberts' papers I would be interested to receive your comments on them in writing. In particular, I am interested in the impact on the statement that I quoted from page 17 of the paper.
It will not change the statement on page 17, which is a statement of fact about the cost of finance.
Okay. My other concern is the way in which risk is calculated. Can you compare how you would calculate risk in assessing a PPP project with how an insurance company would calculate that risk? What are the differences and what is the logic behind the way in which risk is calculated? If we consider Hairmyres hospital or many other projects, it is hard for the layman to see where the real risk was. What is your methodology for calculating risk?
Before that question is answered, I think that Paul Brewer wants to come in on the previous point.
I want to raise a slightly different issue to broaden the point that Jenny Stewart made on the cost of capital. In essence, Jenny's point was that cost can be measured quite easily, but that we also have to consider what is being bought for that cost. If you were to ask Treasury economists about the cost of capital in the public sector, you would find that the subject was very contentious. The cost of capital as reflected in gilts, with which the Government raises money, or in PWLB loans through the Debt Management Office, has to do with the rate at which Government can borrow money when it is taking and guaranteeing all the risks. If, in order to fund a new piece of infrastructure, a local authority were to raise a bond issue that was backed by the entire resources of that local authority, it would get a very fine rate, but that is because the local authority's balance sheet could be called on if the project did not work well and the debt was not repaid. A local authority pays a higher cost of capital to get the private sector to take all the risks of delivery, but it is buying something completely different from what it would be buying if it raised money against its own balance sheet to fund a project. You will see a difference in the cost of capital, but the outcome that you are buying is really quite different. A different value is achieved for each cost.
I want to draw out a couple of points. On transparency and accountability, I have not been able to track down any figures that would allow a PFI project and a non-PFI project to be compared directly. It would be interesting to have access to figures for a fully publicly procured project in order to compare the two types of project. For a genuine comparison, one has to be sure that one is comparing apples with apples, as Jenny Stewart said.
I return to the point about whole-life costing. A PFI contract makes the asset available to the public sector throughout the term of the contract, and the asset is delivered back at a certain standard. I speak as someone who went to a school that was built in 1972 but which has already been knocked down and replaced by a PFI school because of the deterioration in the building quality. Such considerations are a key element of any economic analysis of the benefits.
Are we going back to my question on risk?
I think that Derek Brownlee wants to come in.
No, the witnesses may answer Alex Neil's question.
I will not go into a huge amount of detail, but at the start of a PFI/PPP project you draw up a risk register so that you are absolutely clear about the variety of risks that you, as a public sector procurer, are seeking to transfer to the private sector. A main risk that we have already talked about is capital cost overrun. If the public sector procurer says that the project will cost £100 million to build but it actually costs the private sector £120 million, that is a private sector risk and there is not a problem. However, if the public sector body had continued to provide the project itself, it would take the risk of the cost overrun. You have to make a judgment on what the cost overrun might be for the particular project, so—
I am sorry to interrupt, but if I am a local authority issuing a building contract for a school, then—irrespective of how I fund the school—I do not need PPP to build in a stipulation that, if the builder is over time or over budget, it is his responsibility.
Yes, but quite often the public sector client decides to take a different view of what it needs. Paul Brewer alluded to that earlier. If, during the course of a project, it does that and asks for an extension, it—
I am sorry to interrupt again, but if that happens on a PPP project, the contractor typically adds on charges for the changes.
The client would have to negotiate—
Where is the risk?
The risk is on the overruns—
That is not a risk.
You can take a different view.
On a PPP project, we use a table to work out the risks, which are assessed and weighted, after which technical advisers are involved in taking a view on the potential costs. The value of the risk is then calculated in the equation.
I have advised some of my public sector clients that the value-for-money equation for a PPP did not stack up for them because they were not seeking to transfer much risk. In such cases, it is not worth going down the PFI or PPP route. That is why I said that each case needs to be considered on its merits. We talked earlier about some smaller schemes that are not, generally speaking, applicable for PFI/PPP. We need to make a careful case-by-case analysis of the risks that are involved in each project.
I want to respond not to the question that Jenny Stewart has answered on how risk is calculated and assessed but to the follow-up question: what is the risk and how is it different for the public sector? I draw Alex Neil's attention to distress projects, for example the East Lothian schools project. If the local authority had procured the project on a design and build basis, it would have exposed itself to massive cost increases. Given that the project was procured under PPP, the local authority was fully insulated from what can only be described as very significant cost increases.
I accept that, but the West Lothian College project went the other way. We also had the Inverness airport and Skye bridge projects, which cost us an additional £70 million to buy out.
Those are different projects. We are talking specifically about the transfer of risk in terms of inflation and cost overruns—
But in those cases, the risk was supposed to have been transferred and yet the taxpayer ended up having to pick up a £70 million tab.
If we are looking specifically at the question of cost overrun, the East Lothian schools project is a good example. The Inverness airport project falls into a different category. We are not talking about cost overruns in that example.
I will pick up on Alex Neil's question on how our way of assessing risk compares with how insurers assess risk. The techniques are broadly the same. Generally speaking, anyone who goes into a major capital project—it matters not how they procure it—needs to make a careful assessment of risk. Again, generally speaking, the assessment will be done by technical advisers with wide experience of capital projects. Insurers look for a strong evidence base and a sufficient flow of information on the risk and situations in which it occurs. Provided that they see a sufficiently identifiable track record of the outcome of those risks, they will provide insurance. After all, insurers insure an enormous variety of risk.
Insurers underwrite the extended warranties on simple assets such as electrical goods, but those that are involved in capital procurement projects have to deal with immensely more complex assets. In principle, we do things in a similar way to insurers, but we work in an entirely bespoke situation. For example, if one is creating a hospital, one is creating an asset that is complex in its design, operation, constraints and maintenance. We do not have access to the same sort of track record information that insurers have, which gives a statistical read-out of the risks and how they will turn out. We have to do a much more bespoke exercise for public authorities that need such advice. Generally speaking, assessing risk is a sophisticated process that is rooted in the reality of actual project outcomes.
Comparing apples with apples is easy; recognising what is an apple is slightly more problematic.
A central tenet in all the submissions is that there is no single perfect way of procuring or funding projects and that a project-by-project assessment is probably the best way to proceed. However, there is obviously a strong political wind behind the non-profit-distributing model. What would be the impact if that were the only variant on the table? Is it preferable to retain, as part of a range of options, what we might perceive as the more traditional PPP models and, if so, why?
Jenny Stewart and I both worked on the first model. I acted for the financiers and the bidder in putting together the financial structure that delivered the NPD model and Jenny acted on the procuring side, so we have slightly different experiences.
So far, the NPD model has been applied through to achievement of contract for schools, and the bidders have generally been participating in an extensive programme of school building, so they think that they understand the risks. As a bidder, you can take a view on the sort of risks that you are taking on in a PPP-type project. As a financier and provider of equity, although we talk about those projects being all debt, there is a junior layer of debt that is the last to get paid out. That plays the role of equity, so there is still equity, but its return is fixed.
From the perspective of project sponsors and the people who subscribe for equity, the model has been proven to be workable in a number of situations where the type of project is well known and they feel that they understand the risks. The model is less tested in riskier projects and on a large scale. The question is, would the financial markets step up to a very large number of such projects and accept the returns for the risks that they take?
The model has established itself initially and there is definitely more mileage in it, but there are probably limitations on the sectors in which it will work. Take the waste sector as an example. There are many technical risks in waste disposal projects, and in a number of situations the people who subscribe for equity—the people whom I described previously as the integrators, who bring all the services together—have to roll up their sleeves to sort out problems. In such riskier projects, I question whether there would be the same appetite to proceed on the basis of a fixed return as there might be for a schools project, which is more of a known quantity and less technically difficult.
The committee will have seen from the submissions—certainly from the PPP Forum submission, which I played a part in drafting—that there is open-mindedness towards the model, but it has not yet been tested in riskier situations or on a larger scale.
Paul Brewer is right about the NPD vehicle having been tested in only limited circumstances as a project vehicle, although there are many good examples of it working as a corporate vehicle, for example with organisations that have a lot of substance, such as Network Rail and housing associations. I think that the committee previously asked me about Scottish Water mutualisation. The model is tried and tested at a corporate level.
I agree with Paul Brewer that the model works on accommodation projects, and I helped my client to get a bankable solution. However, there are issues about whether the model is scalable, particularly around refinancing provisions and whether the market would bear that with more risky projects. Refinancing probably needs to be looked at from a scalability perspective. The model might be easily transferable to the roads sector, where there has not been a great deal of refinancing. I agree with Paul Brewer that the waste sector would be trickier—we covered some of that in a document that I wrote. There is more work to be done on developing the model. If it were the only model around—which was the original question—we might struggle, but there is a wide range of other options.
To pick up on an earlier point, the argument about capped returns is well understood. When I was in the Department for Transport early on in my career, we did the Dartford and Severn crossings on a capped return basis. There are plenty of models that we can consider.
The Scottish futures trust proposals mention raising funds for a basket or suite of projects. Would that overcome Paul Brewer's point about the risk on riskier projects? They would be attached to less risky projects so that the overall risk was lowered.
I will not comment specifically on the Scottish futures trust, but the concept of portfolio financing is well known and has been widely applied in the project finance sector when there are a number of projects with a similar risk profile, such as oil and gas projects. It potentially provides a number of advantages. When financiers are able to look at risks across a portfolio of projects, they can see the upsides and downsides of the risks level out and, therefore, may consider the overall economic proposition to be less risky. Depending on how the portfolio is structured, there may be some opportunities for the more successful elements of the programme to cross-subsidise the others.
Portfolio finance is a concept that has been talked about a great deal in PFI but not widely implemented. I know some businesses with portfolios of projects that have, in effect, created a portfolio debt instrument. There are examples of portfolios of debt that have been turned into debt-owning vehicles. That generic class does not have the best of names in the financial markets just now, but the holders of those debts have been able to invite people to accept lower returns because the risks are bundled.
There is potential scope for portfolios of projects but, to procure projects with a portfolio of debt, it is necessary to have a number of pretty consistent projects with pretty consistent risk profiles. I considered doing that for a very large programme of investment in primary schools a number of years ago. We secured the support of the financial markets to get something that, in cost of capital, was going to be marginally cheaper—not transformationally so, but slightly cheaper. However, the real advantage came in its replicability and its ability to get debt into projects more quickly and more effectively. It brought greater certainty to the debt proposition and it meant that lenders would negotiate a consistent position on each project, so we could get savings from the projects being done more quickly and more cheaply as well as on the cost of capital.
A portfolio works if the projects are similar and have a similar risk profile, but it becomes much more difficult with projects in completely different sectors. I considered that for the Northern Ireland Department for Regional Development, which was trying to apply a not-for-profit structure across a range of different asset classes, such as transport, schools and water. We found that, when the projects have different risk profiles, the cost of finance starts to go up. If the ultimate recipients of projects are different organisations—such as Scottish Water and a local authority—what is the interest to those organisations in sharing? They might perceive a downside—a risk—in aggregation across sectors. A portfolio has many benefits and is worth considering, but I caution that those issues exist.
So having homogeneity, rather than disparate packages, works.
We in Scotland have already aggregated projects. The large Glasgow schools project aggregated similar assets—schools—and was one of the first examples of such aggregation, so Scotland led the way on that. I agree completely that two assets with different risk profiles could not cross-subsidise each other, because the beneficiaries of the well-performing asset might be unhappy that their asset failed as a result of the poorer-performing asset.
We must remember who does the borrowing. In PPP projects, the equity provider—the company that will be owned—does the borrowing. If that arrangement is to work, it must work for that body, too. If a debt facility that looks as though it is better value is available, that body may well draw on that.
The European Investment Bank, which works on a not-for-profit basis, is not a portfolio facility, but it is a source of relatively cheap finance. Authorities encourage bidders to use it, because it makes pricing more competitive, but they cannot compel bidders to do that. If a large portfolio debt facility were available, it would have to work for everyone and deliver better value to the authority.
The Bank of Scotland's submission seemed to arouse ire in my colleague Alex Neil, perhaps because of the accusation that the Government might be putting political dogma ahead of a genuine attempt to secure best value in public sector procurement. Perhaps that submission goes further than others do, but a couple of the points that it makes have been brought up more widely and again today. The first is that there is a lack of clear detail about the Scottish futures trust and the second is that the pipeline of business is uncertain. Will anyone in the panel comment on the timeframe for providing more clarity and certainty, on the impact of the situation on the skills and expertise that we have in the country and which are deployed nationally and internationally and on the availability of capital, which is here, but which might not necessarily remain here, given the competitive circumstances in the market?
Our quest is always for objectivity. Who will respond?
At the beginning of the meeting, I touched on the fact that there is a bit of a hiatus. If we dramatically change models with which people are used to working, all participants will have a learning curve to follow before they reach the same position—we saw that when PPP/PFI was introduced. If we are developing aspects of the SFT proposal, it is important to do that in parallel with maintaining a pipeline of projects that might use a structure that is not dissimilar to what we have at the moment. The NPD model can help with that, as it provides visibility through capped returns but in a structure that is not too different from that in the rest of the market.
The convener mentioned the need for objectivity. Investors and lenders are always objective. Several forces—some of which are large scale and macro—drive the availability of private sector capital to be used for public sector capital projects. For example, the growth in interest in equity investment in infrastructure is partly driven by the fact that the pension funds that manage the assets that look after all our pensions need instruments that are very long term and yield consistently but are not high risk.
Market dynamics are in play in the availability of capital and, in general, public sector and other infrastructure assets are currently very attractive. However, the people who manage those infrastructure assets on behalf of the pension funds are tasked with making the returns that the end owners of that equity seek. They have to have regard to, on one hand, the prospective return from the investment and, on the other, the cost of making the investment. Members will have an opportunity to speak to representatives of the equity market in the next evidence session but, for Scotland to be attractive to investors, we have to ensure that the costs of doing business are not inordinate compared with the alternatives. It is about having effective programmes, keeping it simple and not changing models in ways that do not add value.
People need to see enough flow of opportunities to convince them that it is worth investing their time here. Similar considerations apply to the debt markets. Although banks clearly have a wider geographic presence than equity investors and have the infrastructure in their organisations to deal with projects in Scotland, they will disinvest from markets if they do not see a great deal of activity and will be relatively hard to tempt to invest again. The pipeline, future certainty, certainty about the type of procurement and the need within that for private sector capital, which will always apply only to a certain proportion, will be the drivers of investors' interest.
I will pick up on something that Paul Brewer said and hark back to the question that Derek Brownlee asked earlier—if the not-for-profit model is the only game in town for PFI projects pending the advent of the Scottish futures trust, will that deter any particular sector of the market? Although the not-for-profit model has advantages that I mentioned earlier, there is work to be done. I will not get too technical, but that model expands the public sector's right to have a say and, when a refinancing occurs, not just of the senior element of the finance but of the equity investment—the subordinate debt element—that element is pretty unattractive to institutional equity investors. Although a flow of projects has been undertaken to date—this goes back to what Jenny Stewart said about size—institutional equity investors are unlikely to sign up to the not-for-profit model in its current guise for bigger projects, so a bit more work probably needs to be done to soften the model at the edges if we want to maintain interest and competition in the bidding community.
I do not think that anybody picked up on the point about the skills and expertise that we have in companies here—present company excepted because I do not expect a flight of human capital. If the not-for-profit model is the only game in town, is there a risk that highly skilled individuals in companies will seek to relocate to where more of the projects are?
We talked about the investment side, but there is also the contracting side to consider. As Michael Watson said ably at the beginning, we are dealing with a global market. We are finding that contractors are taking very firm positions on where to invest at the moment. There is a skills shortage and if you ask any of the technical consultants, they will say that they are missing a whole tranche of engineering capacity. Things are very tight. Therefore, they have to deploy their resources very effectively.
The point is not as much about the funding model per se as it is about the general issue of being in a global market. Trying to attract people to Scotland is very important. Ultimately, it is about getting best value for taxpayers' money. Whichever funding route you go down, whether PFI or traditional, in order to get best value, the more effective the competition you have, the better. That is one way of guaranteeing it, whether you go for a traditional or a PFI contract. That is why it is important that we spend time attracting people. A couple of weeks ago, I arranged an event for John Swinney in London so that he could meet a range of investors and so that they could get a feel for what is happening in Scotland. Frankly, the focus of such investors is on the middle east, the States and Europe—particularly for road infrastructure projects—so attracting them will be a hard task.
We are in a genuinely global infrastructure market, in which the larger players can choose projects from around the world. In addition, the rate of growth of the UK infrastructure market remains very high. In that environment, the achievement of value needs to be managed carefully. Overall, we are an attractive market as a lot of investment is being made in infrastructure around the UK. However, if the growth in demand on the supply side is too fast, that will be inflationary. Generally, the rate of inflation in civil engineering and construction is a step higher than the rate of inflation in the wider economy.
As well as the finance issues that we have talked about, I would put the proposition that we need genuine competition so that we attract the players who have the sophisticated skills to deliver projects effectively. That is probably the biggest single driver of value from the private sector side of the equation. For the tier of projects at the next size down from the big civil engineering projects—roads, bridges, hospitals and so on—that directly attract the interests of the global players, many of the smaller construction groups in Scotland could play a valuable part in the delivery supply chain, but they may not have the top-end skills to deliver projects effectively. We need people who have those skills to manage the supply chain. If some of the bigger players choose not to work in Scotland, we will still have the contractors who can deliver the infrastructure, but I would question whether we will necessarily get the best out of them.
Elaine Murray will ask the final question for this panel.
I have a question for Jenny Stewart. Your submission states that if the Scottish futures trust was in the public sector and provided funding only for the capital phase, it would not provide much benefit over traditional PWLB finance. The submission continues:
"However, the Scottish Futures Trust could sit in the private sector … It could act as a promoter or procurer."
Could you expand on that by pointing to any other similar models that currently exist?
My submission was written just before the consultation paper on the Scottish futures trust was published. The consultation paper makes it clear that the Scottish Government wants the trust to sit in the private sector rather than be a public sector entity. If the SFT is to be a private sector body, it will not have access to PWLB finance and will need to work even harder to reduce the cost of capital. As we have discussed, that might be done by aggregating projects and so on. That development has taken place since our submission was written.
How would the cost of capital for that rate as compared with PPP/PFI? What would be the advantages or disadvantages of that?
To be honest, it is too early to tell from what we have seen so far. The consultation paper sets out six objectives for the Scottish futures trust, but some of those—for example, best practice—are just advisory in nature. At times, the SFT sounds more akin to Northern Ireland's Strategic Investment Board, but the SFT might go all the way to having full access to funds. It is very early days at the moment, but I would be interested to comment once I see more information from the Government.
No doubt we shall return to that issue at a future date.
I thank our quartet of advisers for their contribution, which will be very helpful to our inquiry. We will take a short interlude before we resume with the second panel.
Meeting suspended.
On resuming—
We come to our second panel of witnesses, all of whom have extensive experience in financing public sector investment projects in recent years. We have with us a mix of those who act primarily as lenders and those who invest in equity. They are Nigel Middleton, managing director of Barclays Private Equity; Darryl Murphy, managing director of infrastructure finance at HSBC; and Iain Wales, head of structured finance for the UK and Ireland at Dexia Public Finance Bank. I welcome all the witnesses and ask each of them to make a brief opening statement.
Nigel Middleton (Barclays Private Equity):
I am a managing director of Barclays Private Equity, focusing on the management of infrastructure funds. I have been involved in the development and financing of infrastructure since the early 1990s. I started working in the area at the Treasury, where I worked on the UK private finance initiative in its early days. Following that, I led an advisory business at PricewaterhouseCoopers. Since 2002, I have been an investor with Barclays Private Equity. Barclays Private Equity has been focused on infrastructure investment, including public-private partnerships and, within that, PFI schemes when they were prominent. It has also invested in approximately 180 infrastructure schemes, including PPPs and PFIs.
I have a couple of observations to make. During the past 12 months, most of our investments in the United Kingdom have not been in what could be called PFI but have been in strategic development partnerships—a better description—in joint venture with the public sector under an arrangement that is more strategically focused and which has greater transparency in management of the supply chain and financing. It takes a much longer-term view of the relationship between the public and private sectors than does the classic PFI, and sits very comfortably with us as an investment proposition because we are a long-term investor alongside the public sector.
I will leave it at that.
To give the committee some background, I head up our infrastructure team, which focuses on lending and advising the bidding consortia for PPP infrastructure projects. My remit is not just in the UK and Ireland but across Europe, and I have a global role in co-ordinating our infrastructure activities. It is important that the PPP concept is truly global for us, which was recognised in the earlier discussion.
I interact with many other parts of the HSBC group, including the equity fund, which is separate to my business. We are also increasingly involved in corporate activities. The important point is that we see PPP as being in line with our corporate relationships, so it is very important that we align ourselves where our corporate relationships are active.
The debate that the committee is having is, not surprisingly, the one that we are having with our client base, who are the contractors and sponsors in the sector, particularly those who have to date been focused in the UK. There is no doubt that my trite summary of that debate is that we have to adapt if we are to survive into the future. We have the benefit of a more global international market. Others have to recognise that the old-fashioned idea that PFI is the only way is no longer true; it is one way, but not the only way.
We are keen to contribute to the debate about the use of the private finance in infrastructure, but we welcome ideas where private finance might or might not be applicable.
Iain Wales (Dexia Public Finance Bank):
Dexia is a very large European bank that is somewhat unusual in that it has a much narrower business focus than other banks. Globally, it is the biggest lender to the public sector and is the greatest funder of public infrastructure, as we define it, around the globe. My responsibility is to manage the business in the UK and Ireland, but as part of an integrated global team that has its headquarters in Paris.
At the forefront of my mind is the fact that we are competing for the bank's capital in a global environment so, when I make a presentation to the credit committee, it will be one of maybe 40 presentations made on that day. I am constantly in a battle to allocate resources appropriately, so in the debate about the future funding of public infrastructure in Scotland, I am interested in the type of pipeline that it could produce for my institution, which is primarily—although not wholly—a senior lending institution.
We have been active in Scotland in the past and we have funded more than £500 million-worth of the PFI type of structure. However, I emphasise that, to the extent that Dexia's business relates to funding structures, it is about funding public infrastructure, whether it is PFI, PPP or the Scottish futures trust bonds. Whatever the funding structure is, you will always find Dexia there.
After the advisers, we now have the expertise of the lenders and the bankers. Time is money so, since you are all under pressure of time, I ask for succinct questions and answers.
I was interested in Nigel Middleton's comments. He hinted that he is into life after PPP and talked about a different kind of strategic partnership. I ask him to develop what he said and tell us a bit about the services that his company offers.
In essence, the models that we are interested in involve a much closer relationship with the public sector. To contrast them with PFI, I will identify what is typically seen to be not so good about that method. PFI is contractual, it has a single-project focus and the returns are excessive—I am not sure whether that is the case now, but it was certainly the case in the past. One can argue about why those excessive returns arose. Those are the bad things about PFI. As an investor, why are we concerned about excessive returns? If everyone focuses on excessive returns, the good features of PFI can be lost.
What is good about PFI and what would a strategic development partnership, as I envisage it, capture? One could overcome much of the confrontation in the classic contractual PFI arrangement by involving the public sector in different facets—both as a client and as a joint owner of the delivery mechanism. In that way, the public sector will influence and be able to observe what happens in the delivery mechanism. That influence extends to forcing, if need be, greater transparency and almost separating the delivery mechanism, as a developer/integrator role, from the supply chain. For example, in strategic partnerships such as the local improvement finance trust—LIFT—projects in England, which I think were mentioned earlier, the preferred model is that there is no fixed contractor or fixed provider of debt; instead, there is a partner from the private sector that takes on a certain element of development risk, but which focuses mostly on management of the process, in conjunction with the public sector. Private equity is brought in to bear risk in relation to the development activity and as a buffer during the construction stage of the projects. The terms on which the private equity is brought in are observable and transparent to all parties in the venture.
Will you send us some written detail on that? I know that you did not get much warning about coming to the meeting. If you could give us some follow-up details, that would be helpful.
I am happy to do so.
That is an open invitation to all the witnesses. After the meeting, if you wish to add anything, please do so in writing.
I want to follow up on that interesting issue. Nigel Middleton is talking about a form of public-private partnership—it has senior debt and equity. Its financing is similar, so it is a refinement of the old PFI, rather than something else.
Yes, but there are certain key differences: for example, the relationship with the public sector is different.
Some evidence that we have heard suggests that it is horses for courses in deciding which type of funding is suitable for projects. From the perspective of the lenders and investors, which models are more attractive?
I am an equity investor. In essence, the only ingredients that we look for in PPPs, which is an area of focus for us, are the ability to manage the risk that we assume and a process by which we can get a fair return for that risk. Competition does not concern us at all. We operate in an increasingly competitive market for our investment. As an investor, we are concerned about regulation of our return.
I will broaden that out from a senior lender's perspective. One of the key points is that the debt proposition could be adapted to many scenarios. The problem that we find is that people tend to like the PPP-type structure because, in relative terms, it gives a higher level of senior debt, so the overall cost of capital is lower. However, one has to be very secure in one's assessment of the risks that one is taking on. We like to surround ourselves with a lot of mystique, but the lending proposition is simple. We just have to understand exactly what risks we are exposed to.
In the PPP scenario, two key elements have traditionally caused problems. One is high levels of technology risk where there is uncertainty that a particular developer can deliver what it is purporting to deliver. The other element is exposure to market risk in projects such as rail projects. Such projects are not necessarily unfinanceable in themselves, but they are not financeable in the rigid confines of a PPP structure. If one wants a different risk profile, one has to move the model slightly. That level of pragmatism has been absent from many debates with the public sector.
Even within the definition of PPP, once you get down to the nitty-gritty of the detail, there is a wide range of structures. I refer specifically to the global development of PPP. I will give members a statistic that helps to illustrate that. Dexia has been the leading global PPP funder for two or three years now. Within that market position, three years ago, the majority of deals that generated that business would have come from the United Kingdom, but that amount will now be less than 40 per cent. That demonstrates how the PPP model has grown geographically in terms of the deal flow within countries where it has been under development for some time. Of course, the deals are all different. A deal in Japan does not look the same as a deal in the UK, Australia or France. However, we fund all the deals.
At the risk of presenting the hard-nosed bank's story, I will set out what matters to us. I have to run every deal through a returns model, as does every office around the globe. We have to meet threshold returns—that goes back to the comment about competition for banks' capital. Two key inputs to the returns model are the pricing model and the risk model, which sets out a rating. The rating takes into account everything, including the risk matrix, to which the previous panel referred. The risk either passes the test or it does not. You get to a level of risk where senior lenders are not going to be interested, such as where there is not a pass-through of risks to an appropriate counterparty, whether in the public sector or private sector through a subcontract. You get to a level where the lender would not be interested in the project at any price, but such projects would not represent good value for the public sector anyway. It all comes down to the returns model.
Darryl Murphy will never give me a clue about how his returns model works in detail and I will never give him a clue about how mine works in detail, but we all go through the same sort of process.
It is relevant to see the business in an international context. When we go to credit committees, that is how they look at the business. They feed all the structures, which differ in detail, into the same kind of model.
The NPD model has been used to try to address perceived flaws in the traditional PFI approach. However, there has been conflicting evidence on the current level of private sector acceptance of the NPD concept. What are your views on it?
There is a concern about whether the NPD model is a one-way bet—and not a very good one—for the investor.
More specifically, the risk capital that sits within the NPDO as the junior debt carries a coupon that, on the face of it, seems to provide a respectable and not unattractive return for that risk. Many investors are concerned that that investment can be refinanced at will, at no gain to the investor, at any time during the project. That does not necessarily fit comfortably with the idea of taking risk behind the investment. In effect, risk is taken ahead of the lending that the project will have to take on in the event that it encounters difficulties.
The difficulties that are encountered with PFI projects are real. PFI/PPP would not exist as a procurement mechanism if such difficulties were commonplace, but we have suffered financial loss and have had to put more money into a project merely to enable it to be delivered for the public sector. We have done so at no cost to, and without compromising, the deal with the public sector. The fact that we must evaluate such matters is why we have some concerns about the NPDO model.
In times of turbulence, does public sector finance not represent a safer haven? Is it not, therefore, more attractive?
Is the—
The finance sector has not exactly covered itself in glory recently—I am referring to credit crunches and so forth. Is not an offer of cash by a public sector organisation a relatively good prospect—a safer haven, so to speak?
I am happy to provide cash for a project that has public sector support. With the NPDO model, we assume that the risk lies in the possibility that there will be a contractor failure. Contractor failures happen—that is our concern. We must evaluate the reality of that risk against the return that is offered through the NPDO model. It is as simple as that.
From a senior lender's point of view, it is important to note that although the NPDO model has some complexities around it, it does not fundamentally change the proposition as regards debt. What does change—this is highly pertinent to Nigel Middleton's comments—is the financial investor community. We have worked with consortia that have bid for some of the NPD schemes that have been procured in Scotland and have noticed that the investment arms of some contractors have been less interested. However, fundamentally, such schemes still involve a building contract and a services contract which, one can argue, are what drives the competition. From a debt perspective, the key risks remain the same.
As Nigel Middleton identified, the focus is, rightly or wrongly, on deliverability of the asset so, in general—certainly from our credit perspective—the construction risk is the first port of call in the analysis. The only issue to figure out is what happens when things go wrong. There is a saying: "Know thy borrower." From a lender's point of view, it is extremely important that one understands the corporate governance of the entity in question. One needs to understand what the incentives are if something goes wrong. That is the focus of senior lenders. If everything happens as it should do, the models are not radically different. Our only concerns are about a possible lack of incentive to help fix a problem if an issue arises with the contractor or the service operator.
I will not repeat what others have said. The NPD model presents less of a challenge from a senior debt perspective than it does from an equity perspective. My concerns are around some of the detail and the extent to which the nature of control and incentivisation is adjusted because of the change in the structure and removal of the profit element.
However, a number of my equity clients are stepping back from Scottish NPD projects, in the hope that the refinancing arrangements in particular will be tweaked to their satisfaction. I will not get involved in that debate, because it is not really my pigeon, but I will be extremely interested in the outcome.
I am more concerned about the delay in the rolling out of projects in Scotland. I do not want to harp on about it, but I repeat what I said in my opening comments: it is difficult for us to see projects coming through the pipeline—whether they be NPD or some other kind of PPP.
I want to pick up on points that Mr Middleton made earlier. I know that you will provide written evidence, so you might be able to answer this question in it. Evidence from Canmore suggested that Scotland should not emulate England in setting up a centralised procurement structure, and the Scottish Government has laid heavy emphasis on trying to decentralise responsibility to local authority level. You said that part of the unattractiveness of PFI was that it had to do with individual projects rather than with strategic approaches. However, does that view go against the grain of decentralising and of allowing local authorities to tailor their projects and their funding models to fit their needs?
I do not think that the two ideas are at all incompatible. Models that we have seen working successfully also work quite well at local authority level. The singular difference from the classic PFI is that our model would not be focused on one project that was being built or refurbished but would be focused on a programme over a period of time. The procurement unit, as it were—the size of reach—can be local authority-sized or primary care trust-sized. It works.
Just a quickie. I want to go back to the issue of comparing the risk of PPP with that of traditional methods of funding through the Public Works Loan Board. During evidence from the previous panel of witnesses, Jenny Stewart of KPMG said that roughly 20 per cent of PPP projects have been either over time or over budget, but by a fraction—not by any significant amount. Have any projects that you have been involved in been over time or over budget? I presume that that is one means of measuring risk by proxy. To the best of your knowledge, what are the comparable figures for non-PPP funded projects?
I do not know the comparable figures for non-PPP projects. Studies have been done by the National Audit Office in England, and they showed some fairly stark differences.
In my experience as an investor in PPP, the figure given previously of around 20 per cent is broadly correct. The consequences of delay will be borne—although not completely—by the building contractor as the party that is responsible for delivering on time. That appears to be a sensible allocation of risk that certainly works in practice.
However, we have experience of projects that have not failed but in which the delay has been quite long. At that point we, as an investor, bear the risk. The reasons for that are various. There can be limitations on liabilities from the constructor, and there can sometimes be a lot of confusion because public sector clients might not be particularly well informed on how to manage their obligations and responsibilities during the construction phase. That can lead to dispute, which inevitably causes us—the SPV—concern, because we are in the middle of the PFI project and we are caught in the middle of the argument. We tend to suffer losses as a consequence.
We are incentivised to manage actively, even though at first cut our risk is not financial on delay. We are incentivised to manage projects in order to get them smoothly to completion on time, and to manage our public sector client to act in such a way that it will not expose itself to potential claims that then give rise to arguments and, ultimately, loss to all parties in the process.
The PFI model distinguishes itself when there is contractor failure, which I spoke about earlier. There is then a risk transfer—the risk is borne by the investor—that goes beyond any design and build model. It is a risk that we bear.
Due diligence is important before we lend money for a project, particularly in respect of delivery of an asset. The fundamental focus is the contractor's technical and financial capability. By virtue of the fact that we have done our due diligence, we believe that the programme can be achieved at a certain cost. Whether that happens 80 per cent of the time or more, that is what should happen.
Another problem is that it is difficult to obtain empirical evidence because large-scale failures are few and far between. A number of the headlines ones are mentioned in the written evidence that we provided. The important thing to note is that, from our point of view, such failures are no accident. We would not lend money to a project if we had any discomfort about it or we thought that the person could not achieve what they had set out to do. Obviously, there are a number of reasons for individual failures.
Our experience of delays is consistent with the statistic that was mentioned. The area that we are discussing is one in which private finance brings a lot of advantages, partly because Nigel Middleton and his equity colleagues know that their return is at risk, so they manage the contractor carefully, but also because the senior banks tend to police projects closely. During construction, there are monthly reports, and if there is any sign that a project is falling behind, even way before the intended delivery date, the bank will be on the phone to its technical adviser, the contractor and the equity people, saying, "We need a plan to bring this project back on schedule."
Thereafter, because of the nature of senior lenders, as soon as the due delivery date passes, the senior lender will insist that the contractor pay liquidated damages. The contractor is well aware that the lender will invoke the letter of the contract. That is an example of the extra controls that create the favourable situation of projects being on time.
Our universe is about 80 deals. Off the top of my head, I cannot think of a deal that was delayed sufficiently that we went to the first long-stop date, which would be after six months, other than one or two highly specialised deals that would probably not be financed now. I refer, for example, to the type of highly technical information technology or London Underground deals that were done about seven or eight years ago.
As there are no more questions, I draw this part of the meeting to a close. I think that we have finished roughly when Mr Middleton wanted, and slightly earlier for Mr Wales. Thank you for your helpful contributions to the committee's inquiry.
Meeting suspended.
On resuming—
Our third and final panel of witnesses comprises commentators on financial matters relating to capital investment. We have with us Jim and Margaret Cuthbert, from the public interest research network at the University of Strathclyde; Iain Docherty from the school of business and management at the University of Glasgow; Mark Hellowell, research fellow from the centre for international public health policy at the University of Edinburgh; and Jan Love, representing the PPP Forum. You are all welcome. I invite each of you to make a brief opening statement.
Margaret Cuthbert (University of Strathclyde):
My name is Margaret Cuthbert and I am an economist. I have held lectureships in econometrics and business management at the University of Glasgow, the University of Strathclyde and Heriot-Watt University. I have also worked in economics and statistics at the Organisation for Economic Co-operation and Development. For many years, I ran an economics business consultancy.
For the past 10 years, I have concentrated my research on public finance in Scotland. The committee may be familiar with some of the subjects on which I have worked, which include, besides PFI, water, free care for the elderly and the "Government Expenditure and Revenue in Scotland" report.
Under the Freedom of Information (Scotland) Act 2002, Jim Cuthbert and I have been able to access the detail of a number of the final business plans and contracts relating to PFI projects. Before last year, it was extremely difficult to get that information—the whole issue was shrouded in secrecy. Looking at those documents—the document for the Edinburgh royal infirmary project was around 9,500 pages long—has been revelatory. Our work has shown the detail of the financial projections and how people have managed to manipulate affordability and value-for-money calculations, which has had an effect on whether PFI schemes have gone ahead.
Today we have heard a great deal about issues such as financial pipelines. All that work is incredibly important and interesting, but we cannot put any weight on what one person says, as opposed to another, unless we have some of the hard facts. We can have armchair discussions all day long in which the argument is won by the person who talks the talk, but we have looked at the facts, which we want to present to the committee.
In the work that the committee has received from us, we have not been able to show everything. Work that we have in the pipeline shows the negative effect that PFI has had on skills and training, employment, research and development and the growth and survival of small businesses in Scotland.
Dr Jim Cuthbert (University of Strathclyde):
For most of my career, I was a statistician in Government. I ended up as chief statistician in the old Scottish Office and worked for a time as a statistician in the general expenditure policy division of the Treasury. Prior to that, I was an academic in the statistics department of the University of Glasgow.
For the past 10 or 11 years, Margaret and I have been working together on researching the Scottish economy. We got into the area of PFI through our interest in the effects of the current cost accounting rules in Government on utilities. We found that there are similar mechanisms in PFI.
Margaret has mentioned the importance of the freedom of information data that we have been able to get. In essence, those data provide some facts in what is otherwise a very smoky and non-factual area. We can hardly emphasise too much the importance of factual information. In our paper, we have provided evidence about six schemes and we have suggested that what is needed in relation to PFI is hard monitoring information, which has not been available to date.
I want to pick up on a specific point that Jenny Stewart made, which is that commonly when one considers the cost of PFI, the comparison is clouded by the cost of services. We have been able to separate out the service element of costs—that is life-cycle costs, maintenance and operations—from the non-service elements, such as debt service, taxation and dividends, so all our comparisons are completely unclouded by that service element confusion. In the light of Jenny's comments, I cannot emphasise that point too much.
The factual information is vital but the other benefit of the freedom of information data is that they confirm a hypothesis that we had about what was going wrong with PFI. A mechanism is at work, related to the way in which the non-service element of the unitary charge is indexed, which we argue leads to high levels of profit in PFI schemes. Typically, the non-service element of the unitary charge is indexed in such a way that it goes up, or may be flat, while senior debt charges, which initially are the major element of expenditure within that, go down. According to what Partnerships UK says about the operation of PFI schemes, that is not meant to happen, but it happens within each of the six schemes for which we have presented evidence. Hard evidence of that mechanism is one of the important things to come out of our work.
Dr Iain Docherty (University of Glasgow):
I will limit myself to a short opening statement. I am Iain Docherty from the University of Glasgow business school, where I am senior lecturer and director of the MBA programme. My research work for more than a decade has been mostly on the transport industry and specifically the formulation and implementation of transport policy in Scotland and in other jurisdictions within the UK. I am also a non-executive director of Transport Scotland, although anything I say today does not represent the official view of the agency.
Mark Hellowell (University of Edinburgh):
I am from the University of Edinburgh's school of health and social science. In the past few years, I have looked particularly at public-private partnerships in health care and more generally at the introduction of market systems into health care. My colleagues examine more broadly issues of relevance to public health and health care delivery.
PPPs have been the dominant form of large-scale capital investment in the national health service since the early 1990s. Although, in quantitative terms, PPPs make up a relatively small amount of overall public investment, where they are used they tend to be the dominant form of investment. The reason is that there are fiscal and balance-sheet reasons to use PPP. In essence, the incentive is a quirk of financial reporting standards.
Perverse incentives to use private finance are provided throughout the system, from the Treasury, through departments and devolved Administrations, down to public authorities. They are perverse because the underlying economics of PFI mean that it can deliver lower public debt or a lower impact on capital budgets only if it is a more efficient form of procurement, and the argument that it is a more efficient form of procurement than the alternatives is problematic in that it is based on the idea that the higher cost of public finance is simply a function of risk. In fact, the Government's own evidence, which was based on an examination of financing costs commissioned by the Office of Government Commerce, was that financing costs were in excess of what one would expect from aggregating the cost of money plus risk.
If the private sector is to be more efficient than the public sector in delivering public investment and projects, it must more than offset the higher cost of finance; however, its ability to do that has not been demonstrated, and instead we have claims about PFI's superior ability to deliver on time and to budget. It is important to point out not only that the research base is problematic in a number of respects, but also that the conclusions are mainly irrelevant to the issue of value for money. It may well be the case—in fact, it is likely to be the case—that PFI delivers on time and to budget in terms of post-contractual price certainty, but that is irrelevant to the issue of value for money, which depends on how the risk is priced.
More recently, we have had the NPD and SFT models. It is difficult to suggest, at this stage, that we have done much empirical analysis of those—we have not. However, from our point of view, NPD looks like a relatively minor variant of PPP. Meanwhile, the SFT plans are too vague to evaluate fully, although there are a number of possible issues for concern.
Good afternoon. The PPP Forum is an industry body that represents more than 110 private sector companies. Its primary function is to promote a better understanding of PFI/PPP. I am currently working for a senior lender, so providing debt financing to PFI projects. Previously, I worked on the equity investment side of things and sat on the board of a number of operational PFI companies. Before that, I worked as a financial adviser, advising both the private sector and the public sector, so I have an understanding of the value-for-money side of things, too.
I have seen the PFI industry develop and mature into a knowledgeable market sector with a high-quality skills base of companies and people, and I view Scotland as a real market leader. Historically, Scotland has had a clear pipeline of PFI work. It has had numerous and regular PFI projects available for bidding and delivery, which has resulted in the Scottish PFI sector attracting and maintaining the highest quality within it. We believe that there remains a place for PFI, and we would look to work with the Scottish Government to continue to develop the model in whatever form it may take. We hope that that can be done in a way that builds on the knowledge that already exists. The key aims will be to maintain momentum in the marketplace and to retain the workforce and the knowledge that we have created.
Thank you. Your introductions have launched us into deep and important waters. The main criticism of the "traditional" PFI model is that it can lead to "excessive profits". Some witnesses suggest that refinancing gains and property disposals—two of the main apparent sources of "excessive profits"—have already been addressed. Does the concern about profits relate only to the early projects, or is it still an issue?
You are right to suggest—as Amanda Methven said—that the analysis that we have done has been only of the projects on which we could get data through the Freedom of Information (Scotland) Act 2002. That means that they tend to be from back a bit. Nevertheless, some of them—such as Highland Council's PPP2 initiative—are not that far back and, for some of them, the schools have not even been finished yet. If anyone suggests that we analyse only the early schemes, we would say that it is about time that they presented us with information on the later schemes so that we can analyse them. Just now, they are hiding behind a veil of secrecy around later schemes. The profits in the later schemes that we have analysed are still substantial.
We do not know what is happening in later schemes. In our view, that points to a need to monitor better. The information that we have published in our paper includes the net present values of the non-service element of the unitary charge and the internal rates of return on broad sense equity, together with the average outstanding debt upon which that return is earned—that is the element that, typically, is forgotten in the debate. We have found that, typically, net present values of around 20 per cent on broad sense equity were earned on an average debt that is twice or more the original input of subordinate debt and equity capital. That is the element of profitability that is forgotten. Unless indicators like that are produced and published for all PFI schemes, we cannot know what is happening.
Another point to remember is that there has been some indication that internal rates of return have been dropping over time, but that does not mean that schemes' profitability is dropping because, as was pointed out in an earlier paper to which there may be reference in the committee literature, one can adjust the internal rate of return markedly by altering the profiling of senior debt. If one shortens the profile of senior debt, the effect is to lower the internal rate of return on equity while, at the same time, increasing the average outstanding debt so that the NPV and the profitability of the reward to the equity holders go up. We cannot consider hearsay evidence of a decline in internal rates of return alone; it is entirely consistent with the possibility that the profits earned by broad sense equity holders are increasing.
I take exception to your rationalisation of where the profits come from. Our work suggests that the major element that underlies them is the inappropriate indexation of the non-service element of the unitary charge. The extent to which that is being corrected—if it is at all, which one suspects it is not—needs to be examined.
The key is to get monitoring data. There is no mystery about it and no need for debate on it. If we dug through all schemes and published the monitoring data, the questions would answer themselves.
It is incredibly important that there is greater transparency in PFI. For researchers, the public in general and committees, it is impossible to find out rates of return on more recent projects. That said, there are reasons to doubt that returns have come down markedly. The National Audit Office reported in 2007 that, in its sample of schemes, the level of competition among bidders had reduced. The prices are essentially market prices and, if competition reduces, we would expect the market price to reflect that. There is an argument that, as PFI has developed, it has become more mature and competition has increased but, in fact, competition has reduced over time. That is extraordinary, given the amount of bureaucratic effort that has gone into streamlining the programmes and maximising market capacity.
The early excessive profits were the product of an early market; as the market has matured, the process has become better standardised, better understood and more accurately priced. Significant losses were also made in that early market, and it is important to make that point. It works both ways.
I can understand the frustration about transparency. It is difficult to analyse the information properly because not a huge amount is available, and I take Mark Hellowell's point about the more recent projects. However, the public sector gets its benefit from the private sector bidding competitively. The issue is how much information we can disclose without showing commercially sensitive information that our competitors will use in a bid. That competition is positive for the public sector, but perhaps the private sector could work a wee bit harder on transparency.
From a policy perspective, I will add another reason why continuing monitoring of the early projects is particularly important. A comment was made earlier about the long-term financial certainty of the early PPP/PFI deals generally having been a good thing. Certainty is sometimes a good thing, but it is not if it locks out flexibility to change the purpose of such fixed large capital investments midway through their lives.
Given the kind of civil engineering that is involved and the education and health buildings that PPP/PFI projects have been used to fund, we are already beginning to find, fairly early in the overall life of the assets, that developments such as a change of use or on-going mid-life refreshments—which are normal over a 30-year period—are costly, because anything that was not in the original projects tends to be charged at an premium rate.
That is particularly important given that the major areas of public sector delivery in which buildings are financed through PPP/PFI—health and education—are precisely those in which new technology and working practices are changing how services are delivered and in which the pace of change is the fastest. In the higher education sector, buildings become quickly outmoded as new teaching styles develop. I know that colleagues who do research into health delivery and policy find similar changes in how health care is delivered inside the assets—the buildings—that are delivered to the public sector through those financial mechanisms. From a policy perspective, it is important that monitoring is carried out of how the total life costs are mediated by that on-going intervention, which is absolutely critical to the effective delivery of public services, which are the end output. So the issue is not just about the numbers; it is a slightly broader debate about how we use assets and how fit for purpose they are over their lifetime.
The PFI market has had time to mature, so one would expect that that would reduce risk or the perception of risk and therefore lower costs. However, because we have been led down the path of bundling projects—for example, in the Highlands, six schools might be bundled together or it might be a large number of schools in Glasgow or Edinburgh—local Scottish businesses find the projects too big to go for, so only a small number of very large companies go for them. We all know about the Office of Fair Trading's recent comments on large companies bidding for projects and cover pricing. In our analysis, we also found that Balfour Beatty, in an internal document that somehow or other got on to its website, said that it was a virtue of PFI projects that there were so few competitors that companies could get a good return. I put it to the committee that all that has happened is that we have very few competitors and so the price has gone up.
The actual quotation from Balfour Beatty was:
"Tender costs and complexity reduce competition".
There you have a view from the horse's mouth on a competitive market in PFI.
On the size and bundling of projects, on the slightly larger projects—which give better economies of scale and therefore assist from a value-for-money perspective—many smaller Scottish contractors often work with larger contractors, perhaps from down south, or form successful bidding consortia. I guess that that is the flip side of the issue. The market has moved to address the issue of large projects. Attracting larger companies in Scotland can, in itself, create jobs and offices in Scotland.
The combined introductory remarks have bestirred the committee.
The witnesses have covered some of the points that I wanted to cover. I ask those of you who have strong concerns about the PPP/PFI model what alternatives are preferable. Is the non-profit-distributing model preferable or, from what you have seen of it, the Scottish futures trust? Alternatively, if PFI and PPP schemes must be on balance sheet, is it preferable to increase the amount of public loan from the Public Works Loan Board? What is your solution?
I do not want to propose an alternative model as such. I agree with the earlier remarks that a variety of models are called for. A number of steps could be taken that would greatly improve current models of PFI and, I presume, lead to a much better futures trust. Better monitoring measures could be developed, so that information could be collected and published for all schemes, future and past. Projects could be unbundled—as we said in relation to Balfour Beatty, bundling reduces competition and leads to a non-competitive market. Whitehall and Westminster could be tackled on the distortions that bias the current system against the public sector. Capital charges and VAT also have a distorting effect. Better central advice could be provided; it is ridiculous that the Treasury has not picked up the point about the need to quote average debts in conjunction with internal rates of return. Public sector training and expertise could be generally improved.
Better value for money could be secured by stopping the diversion of so much resource away from the Scottish budget to Whitehall by means of tax. The net present value of projected corporation tax payments to Whitehall in relation to the five Scottish schemes that we analysed in our paper comes to more than £100 million—that is equivalent to more than £100 million from the Scottish budget. In a more rational approach to procurement, such significant diversion of funds would stop.
Better benchmarking of the cost of service provision would also secure better value for money. In the case of the new royal infirmary of Edinburgh, the contractors who were hired by the lenders did some benchmarking and found that some elements of the NRIE costs were above national average. The public sector was not told about that—or at least it never picked it up. There is no reason why the public sector should not benchmark the cost of services better and ensure better value for money.
Monitoring statistics should be published for all past schemes and—by golly—if the schemes are really bad, why not reopen them? After all, whatever the legal position, Government is a monopolistic buyer of such schemes and has considerable leverage over suppliers, so it could get blood out of a stone in relation to past schemes that proved to be unduly profitable.
It is important that public authorities and others are presented with a plurality of options, so that they can make a rational judgment on the best solution for their project. The Treasury, in particular, makes the case that PFI is a small but important part of public investment, which implies that there is a plurality of options for public authorities and that there is no dominant form. However, that is not really true; if you want to do a large-scale procurement in health, education, transport and many other sectors, PFI is probably the only option that is currently available.
There is what is sometimes called a PFI monoculture, which is intrinsically problematic and leads to distortions throughout the system. We have witnessed such distortions in public sector comparators, which have been manipulated so that the PFI option seems the most cost effective. Research, not just by academics but by the National Audit Office, shows such manipulation.
The most important point is that there should be genuine diversity of options, whereby different sources of finance are made available. In that context, the decision last week on Glasgow's Southern general hospital was welcome. A huge scheme will go ahead on the basis of public finance—it is the first large whole-hospital scheme to do so since the early 1990s.
Can you send us a reference to the research that you mentioned?
Yes, sure.
Members would expect the PPP Forum to quite like PPP and PFI, but we are not beholden to PFI and are more than comfortable working with the Scottish Government to develop alternatives. As I said, the market has matured. The key is to build on current knowledge. NPD might be a starting point, although potential equity issues, which we mentioned, need to be addressed. A positive aspect of NPD is transparency. Public sector board representation assists transparency and promotes greater understanding of schemes when they are operational. I would advocate that we work together and try to develop the PFI model, possibly via NPD, perhaps going on to the SFT model.
Reference was made to the Office of Fair Trading inquiry into allegations of collusion. We should perhaps find out from the OFT when its inquiry is due to finish and whether it will finish before ours. That inquiry may or may not inform our conclusions.
I have a question for Jan Love and a wider policy question for all members of the panel, starting with Margaret and Jim Cuthbert.
Jan Love said, as a counter to the point about the excessive profits that were made in the early days, that there were also losses. In Scotland, apart from in East Lothian, where were there losses?
More widely, how have we got here? The submissions, particularly those from Jim and Margaret Cuthbert, refer to excessive costs, excessive profits and very high returns. Why have those high returns been allowed to happen? What has gone wrong in the system? Aside from Jim Cuthbert's mini-manifesto on the future, what do we have to do up front to take out of the system all the excessive profits and costs and what you are in effect saying is a lot of kidology, to enable us to assess projects properly?
I have given Jan Love time to think.
East Lothian is the easy example, because the public were made aware of the situation and there was transparency. In a portfolio of projects, some will perform better and some will perform less well.
You said that there were losses. Apart from East Lothian, can you give us examples of where there were losses?
As the funders said, everyone will go into a project with an expected return. I am talking about a real reduction in terms of profit return—whatever you want to call that. So there is loss—
That is a lesser profit than someone expects. You specifically said that there were examples of losses. A loss is not a reduced profit; a loss is a loss.
If you let me finish, I am saying that that there are reductions in the returns on financing—that is more at the shareholder level—and there are also reductions and losses at the contractor level. That is not made public knowledge, because it is commercial information that is for the companies themselves. For example, there are facilities management contracts that are more expensive to maintain than they were originally priced and there are life-cycle and whole-life costs that are more expensive. The build contract is the key one and on some projects the build is more expensive than was estimated. Although we are talking about excessive profits, a company will have a portfolio of projects. There are profits and there are, whether you call it reductions in returns or lower profits, or—
They are not losses.
They can be losses—I am saying that there are various levels.
I am asking for examples, apart from East Lothian. With all due respect, you raised the issue to counter the point about excess profit. You are saying that some people did not make an excess profit; they made a reasonable profit or perhaps less than a reasonable profit. You have not given us another example of a scheme where there was a loss.
There are schemes where people have made losses.
Where are they?
I cannot give examples, because they are individual companies.
Write to us and let us know whether you can give us examples.
That is a matter for the individual companies. If a company is undertaking a number of projects, some of them will not go well; the company takes the risks and it will have to bear the additional costs.
I owe Dr Docherty an apology because I missed him out on the previous question. If he would like to answer this one, I will come to Margaret Cuthbert immediately afterwards.
I will return to Elaine Murray's question first and link into Alex Neil's.
I would phrase the question about the right model slightly differently because doing so could reveal something useful. Given that the public sector has a fundamental competitive advantage because it has the lowest cost of capital, why can it not then deliver best value by doing the whole process itself directly? The answer is that public sector institutions do not have what the political scientists would call strategic capacity; that is, they do not have enough highly skilled people to do the project management tasks that are part of the whole-life delivery of a project. Market forces are at work in the private sector, which such people inhabit because they are vastly better remunerated there.
A concrete example of where that has been turned on its head is Transport for London, which ended up going to court against the UK Government over its preferred model for the financing of the tube redevelopment. Bizarrely, TFL has much more fiscal autonomy than the Scottish Government and Parliament have because its categorisation as a local authority means that it has significant borrowing powers that Scotland does not, and it levers those powers extraordinarily well. The press release on TFL's last bond issue, in December 2006, showed that the difference in the costs of capital between the TFL bond and Government borrowing was 0.29 per cent and 0.34 per cent for the two bond issues.
Is that why Gordon Brown fought tooth and nail to stop it?
That was after the court case was resolved. TFL was able to do that because, without doubt, it has the best concentration of skilled financial, engineering and transport professionals of any equivalent body in the UK, and it has the capacity to lever its powers to deliver traditional public sector borrowing effectively for the projects that it needs.
I will start with Elaine Murray's question because I found it very interesting. I do not think that I am placed to say which of the many models might be appropriate. Hopefully we can help by forensically examining past schemes so that you will be advised of what to look out for in new ones. Given human nature, many of the problems that we have found could well occur in later projects.
You will find it very interesting that, when local authorities and hospital boards were required to apply the two tests of affordability and value for money, they went through tremendous loops and circles to ensure that they passed those tests so that they could go for PFI funding. Some of the manipulations that they used are very worrying. For example, when the trusts concerned realised that the projects for Hairmyres hospital and Edinburgh royal infirmary could not pass the affordability test, they sought to attract a bigger bit of their neighbouring markets. Hairmyres hoped to attract a chunk of the Glasgow market and Edinburgh hoped to attract a chunk of the West Lothian market. If everyone played that game around Scotland, we would have too many hospitals for the population. It was suggested many times in parliamentary meetings that, in the case of Lanarkshire, some of the other hospital trusts might have been cut short or be losing out because of Hairmyres. The full business case shows that that is the case. Money was channelled from a general fund to make the project affordable. Such things are very important.
May I move on to Alex Neil's point?
I have got Joe FitzPatrick down to ask a question just now. Is it on the specific point?
My point has just been answered.
I am not quite sure that I understood Alex Neil's question. Perhaps he would repeat it.
How have we got to a situation where there are excessive costs and excessive profits? What has gone wrong in the methodology and what do we need to do to rectify it?
The other witnesses will add to this. The Government's accounting policies, which include current cost accounting and the capital charge, are the basis of an awful lot of the problems. As is the case with statistical surveys, accounting methods may be useful for some purposes but not for others. The capital charge procedure, which was introduced so that we would make proper use of Government stock, such as hospitals, schools and derelict land around the railways, and would realise that it was all costing us money, has infiltrated itself into being used for working out the affordability of a public sector scheme compared to a PFI scheme, which is not the way in which it should have been used.
I do not understand current cost accounting. Can you tell us why it is a problem?
Would you like us to produce a paper on it?
Would the committee like a paper on it?
Yes. That would be helpful.
We have published an analysis of the effects of current cost accounting on the water industry. The issues are in effect the same. The pricing method that is used in the water industry in England and Wales has the same impact as the capital charge in current cost accounting in relation to health PFI projects. The distortions are in effect the same, which is how we got into the subject in the first place.
Would it be possible to cover in the same paper the capital charge issue as well as current cost accounting?
The capital charge is a manifestation of current cost accounting.
Alex Neil asked how we got into this situation. I do not want to go over the same ground but, at a sort of meta level, it was a failure of Government. There was a train crash. There was a combination of a political push to get schemes off the Government's books, almost at any price, allied with a weak official response, weak central guidance and officials in health trusts and so on who were just swamped by the new initiative and could not cope with the opportunities that the expert private sector saw.
Alex Neil asked what we could do about it. There is the whole issue of how risk is handled. Handling risk on an expected value basis is a nonsense. If you hand out to five different PFI schemes the expected value of something going wrong—if there is a 20 per cent chance of the project going wrong, you will hand out 20 per cent of the expected cost—in four out of five cases, the project will not go wrong and that cost will be banked as a windfall profit. In the remaining case, the project will go wrong and the 20 per cent that you have handed out will not cover the cost, so the scheme will go bust. Under the current way of handling risk, a lot of schemes will have windfall profits and the odd scheme will go bust.
It would be much better to handle risk on an insurance-fund basis. That is where the Scottish futures trust offers an opportunity. The futures trust could sit at the centre of its web operating as an insurance fund. If it kept the risk premiums for each scheme and then paid out up to a defined limit when one scheme went wrong, that would not involve windfall profits for individual contractors.
We will await the paper from you.
You said that you believe that current cost accounting is a weakness. Do you have a view on what would be the best alternative to current cost accounting?
I would say historical cost accounting. We have developed our thinking on that in relation to utilities to a good degree. On utilities, one would do a modified form of current cost accounting that did not allocate as a reward to equity funders the whole inflationary increase in the current cost asset base. We have developed that thinking to quite an advanced level. One would want to do something broadly similar for PFI, so that purely inflationary costs were not lumped into the costs that the hospital trust or whatever had to pay.
Perhaps that could be covered in more detail in the paper that will be provided.
Can Dr Cuthbert do that?
Yes.
I will respond to the question of how we got here. The history of PPP/PFI provides a classic example of the failure of the policy cycle. The innovation started with the best of intentions, from analysis of the debate about risk, which we have heard all afternoon. However, the imperative to put major items of public expenditure off balance sheet took over and that momentum eventually got us where we are today.
Some really bad failures of that policy have occurred in the transport sector. It could be argued that the privatisation of rail infrastructure—handing the whole asset to Railtrack—was in effect a large PPP that went bust and which the state had to bail out. The same thing has happened with the collapse of the Metronet PPP concession in the London underground. That highlights the fact that the public sector always retains the bottom-line risk in such very big and complex schemes. We can have academic conversations about the pricing of risk in a project, but when a project goes wrong because the structure is incapable of delivering it, the public sector bails out the funders.
I mention that because, if we are to debate the ideal or best forms of public sector procurement and to think about radically new models, such as the Scottish futures trust, we should break into the policy cycle. Now is a good time to do that and to ask fundamental questions, such as whether we have perverse incentives in the system—such as the treatment of VAT—which mean that the wrong bodies try to do the borrowing; what the consequences are for the fiscal relationship between Scotland and the UK; and—the big debate of today—whose risk it is. Only if we work out who holds the risk will we reach any consensus about how to price risk and feed that into policy.
From a policy perspective, forming an objective framework by which we assess options would help. We have different political views, but we can at least have a fairly objective analysis of the benefits or otherwise of any option.
Can changes be made to the public sector comparator to make it work effectively? If not, why not? People have mentioned a basket of indicators for scrutinising the effectiveness or otherwise of PPP or any other funding method. What indicators should be available to allow us to assess any funding mechanism's success, failure, costs and benefits? Is there any reason why they could not be applied to projects that are procured in the traditional way through the public sector and which do not use private finance?
On the value-for-money test, we contend that the public sector comparator uses a discount rate that is far too large and which favours the PFI. That concerns current cost accounting and the capital charge and will be covered in the paper that we will present to the committee.
As I have said, although the projects that we have examined might appear to be affordable and to pass the value-for-money test, so many circles have been squared on the way through that we cannot look just at the indicators—we need to see the background papers. The Scottish Executive at least, if not the Parliament, should have a unit that has access to those papers and has the skills to analyse them. No one in this room—other than any adviser to a project—has ever seen the detail on a PFI, other than the six cases whose papers we have obtained under freedom of information legislation, yet here we are all sitting around wondering whether one scheme would be better than another. We are powerless without the data, so we must have them.
I will take the points one at a time. I will not go into any great detail on the public sector comparator other than to say that, from what I have seen of it, the process is subject to a number of guidelines. The process is robust and it is looked at by independent parties, such as Audit Scotland. In terms of the public sector comparator, those in the advisory community work on both sides of the fence—the private sector and the public sector. While they are pricing risk for the public sector comparator, they are also seeing that risk being priced on the private sector side. That is the level of understanding that is involved.
In terms of the basket of indicators, an assessment of the success or failure of a project needs to be done on the ground and, more important, by talking to stakeholders. Are the projects delivering the school, hospital or whatever? Certainly, having sat on the board of a number of PPP projects, I know the good-news stories. It is really encouraging to see the successful partnership of public and private.
On the public sector comparator, having been involved in the Dundee schools PPP project, I know that it was clear at the time that the schools would not be built other than by a PFI route. Meeting the public sector comparator test was critical to the project, as was meeting the affordability test. If those tests had not been met, the schools would not have been built. In the situation where only one option—a traditional PFI model—applies, can that result in the application of non-robust tests?
I cannot comment on that project. In my involvement in the PPP process, it has always been robust. Although, clearly, the PPP Forum is for PFI, we are for PFI when it fits and not for PFI regardless. You are absolutely right; we should do PFI when it is the right thing to do and only when it is the right thing to do.
I was about to bring in Dr Cuthbert again, but Joe FitzPatrick has a further point to make. I ask him to be brief.
I cannot say whether the process was robust or not. Being on the board, I was involved in making decisions, but the board did not have the information. That is the point that people are making. We were asked to make a decision on the basis of the question, "Do you want the schools? Yes or no?" If we had said that any part of the process had failed the test, the schools would not have been built. That is not an option when schools are falling down.
What I was going to say picks up precisely on that point. When you talk to people who have been involved in the process—and we have talked to several people in that situation—they make precisely that point. They say that they were told, "You have until 4 o'clock today. Unless you decide to put the project through the value-for-money test, there will be no new hospital." The pressure on those people is intense. Unless the pressure is removed, bad decisions will continue to be made.
When one looks at the detail of what was done at Edinburgh royal infirmary, one finds double bookkeeping on the capital charge. A whole lot of things were done that were very questionable. We need better levels of expertise and guidance. Our view is that we should set up a central unit along the lines of the Northern Ireland Strategic Investment Board. Indeed, the Scottish futures trust could fulfil that function. If we had a central unit with expertise and with no political pressure, its staff could go through deals as they were happening and say, "No. That is a nonsense—you do not have the evidence for that." If that were to happen, it would comb out an awful lot of the nonsenses that happen at present in the value-for-money comparisons that are made.
Our view is that it would be better to increase competition. At the moment, what is happening is a failure of the market. Competition can be increased first by unbundling things and then by increasing information. When you look at the dividends in the financial projections that were produced for some of these schemes, you see projections of an £89 million dividend from an equities input of £100. If it was obvious to people that that was the scale of profit, by golly, other competitors would come forward.
Our view is the opposite of the one that is commonly advanced: make the information on projected profits available, and the competition will work; conceal it, and the competition will not work.
I agree 100 per cent that there is a need for better, harder financial data to enable us to make the decisions more carefully. Alongside that, I argue that we also need to think smarter, as is Government edict these days. I am also conscious of the national purpose that we now have. When we consider the procurement of public sector infrastructure in particular, we need to think collectively about how we can build in the development of capacity in the private sector or public sector that builds economic growth. Therefore, we must begin to have a wider range of indicators as well. Wages in Scottish companies may be more directly recycled into the economy than investor profits—I am sure that that is true. It is also about building our own expertise in the private sector financial institutions and in Government to ensure that we sustain the ability to make better decisions in future. I strongly support and echo the comments that we need some kind of apolitical, arm's-length body that can do that kind of thinking.
To return to value-for-money testing, it is true that Audit Scotland, the Audit Commission and the National Audit Office have examined the public sector comparator process. It would be fair to say that they have been highly critical of it and made the point that, if somebody compares a realistic option with a non-option, they are pretty likely to come out in favour of the realistic option. Public finance is not a realistic option for large-scale capital procurements and, therefore, the answer is rarely that public sector financing will be the way forward. In fact, a former assistant auditor general of the National Audit Office said that the PSC process was based on pseudo-scientific mumbo-jumbo. He made that intervention, as he called it, in the Financial Times and he did it because he wanted to make the point that the process was an embarrassment to PFI.
In accordance with that, the PSC has, in large part, been removed from the process in England. There is no longer a PSC process running through procurement in England. There still is in Scotland but, in England, there is only a basic test at the outset of the project that, basically, will always find in favour of PFI as long as it is a large-scale infrastructure programme.
It is important to point out that the PSC is discredited among auditors and other independent evaluators.
Would you give us a name and reference for that? If you can do it now, that is fine; if you cannot, let us know.
The auditor who made the intervention in the Financial Times was Jeremy Colman.
I will make two points, if I can remember them—I am getting kind of old now. The first relates to scrutiny. It has been mentioned how difficult it is for councillors when a project comes up because they do not have access to all the information. When we interviewed people at the Scottish Executive's PFI unit, we had freedom of information data and could see how much dividend was being made on a £100 equity stake in Hairmyres. We could see what the return was over the years—it added up to £89.1 million—and how much was being made on subordinate debt. The officials at the PFI unit did not know that, so we suspect that we do not have a proper regulatory group even at a central level in Scotland. To some extent, it defies belief that those officials are going to countries in other parts of the world to speak about the qualities and good points of PFI when they do not even know the dirt in their own back yard.
I am aware that this market day is wearing late. Do our witnesses want to make a final contribution or are they happy?
I have a quick comment on what Margaret Cuthbert said. Given that we hear from the private sector an awful lot that we now operate in a constrained market in the UK and that minor variations on the PFI theme are likely to lead to a dramatic reduction in competition, it is strange that the Scottish Government's financial partnerships unit and Partnerships UK, which is part funded by the Scottish Government, are going around the world trying to generate new markets in PFI. Is that in the public interest of the people of Scotland? There is an anomaly in that. I am not sure what explains it, but it might be that it is not such a great idea.
PFI has a lot of positive aspects, such as increased diligence, risk transfer and performance incentives. I reiterate that we want to work with the Scottish Government to continue to develop the PFI model and maintain the workforce and the market momentum.
We have covered considerable and substantial ground. I thank all our witnesses for their contributions, which will help the committee in its deliberations.
As members know, we invited the Chief Secretary to the Treasury, Yvette Cooper, to give evidence to the committee. She has declined our invitation.
Meeting continued in private until 16:59.