Finance Committee, 27 Nov 2001

Meeting date: Tuesday, November 27, 2001

Official Report 184KB pdf

---

Private Finance Initiative/Public-Private Partnership Inquiry

The next item on our agenda is our second session on financial issues relating to our private finance initiative/public-private partnership inquiry. I welcome three witnesses to assist us with the inquiry. Professor Andrew Bain is from the department of economics at the University of Glasgow, Paul Brewer is from PricewaterhouseCoopers and Bob Martin is from the Institute of Chartered Accountants of Scotland.

Welcome and thank you for assisting us with our inquiry. I offer each of you the chance to say a few words about your background as it is relevant to the PFI/PPP inquiry rather than to the specific issues that we are going to discuss today. We have gone through the papers that you submitted and we will question you on them.

I will make a brief statement about the context of the institute's evidence and about my involvement in that evidence.

ICAS has about 15,000 members, of whom about 5 per cent work in the public sector. A significant number of our members work in the private sector and in client advisory roles, as well as being associated with users of services. I have been involved in producing our evidence as a member of a sub-group that was convened by the institute's public sector committee. I became a member of that sub-group largely because I am one of the institute's public sector members and have an involvement in and experience of PFI—at the user end—through my employment as an accountant in the national health service in Lothian. The sub-group comprises people who represent all forms of advisory members. It considered written evidence from members of the institute's public sector committee, which is one of a number of committees that the institute uses to consider aspects of public life that have an impact on the work of its members.

I will put PricewaterhouseCoopers in context. We are the largest professional advisory firm in the world; we are also the largest such organisation in Scotland. We have been closely involved in PFI/PPP since its inception. Our role is to advise public and private sector clients, helping to develop the initiative so that it works more effectively, and to establish best practice in developing projects so that they deliver the results that are expected from them.

In Scotland, we have given advice on a broad range of projects, including the Glasgow, Edinburgh and Falkirk schools projects, Law hospital, Kilmarnock prison and the wastewater projects that have been implemented by East of Scotland Water and West of Scotland Water. Our role is principally to act as financial advisers. We assist the public sector in taking forward the financial aspects of the procurement—in developing the structure of a PFI scheme and the effectiveness of the payment mechanism that governs the relationship and flow of funds between the public and private sectors, for example. We also act for the private sector, helping it to respond effectively to the public sector's requirements and to raise the finance that funds the projects.

We work with a number of overseas Governments that see PFI as a success story in the UK and seek to initiate their own PPP programmes. Although the emphasis of our work as advisers is on the procurement stage of the project, we believe that it is important for us to look at the operation of projects in order to feed back the lessons that come from those projects into the planning stages. Members may have received a copy of our recent survey, which examines how those who have been involved in PFI projects perceive the success of those projects.

I have been an academic economist for most of my career, I suppose. A long time ago, I spent a couple of years at the Bank of England. I was chief economist at Midland Bank for six years and have been on the board of Scottish Enterprise, so I have had a bit of experience outside academe as well. My main academic interest in economics is in the financial sector. In the 1970s, I took part in the Wilson committee, which looked at the UK's financial sector, and I was involved in some of that committee's proposals. I have also done other public sector work. For example, I worked with the Monopolies and Mergers Commission and I am now a member of the Competition Commission's appeals tribunal.

My particular expertise in the context of the PFI/PPP inquiry arose from a study that I carried out some years ago for the Higher Education Funding Council for England—the study was really for all the higher education funding councils—on capital funding for universities. At that point, I suppose that the funding councils had about £1.3 billion of private funding—they had used private funding quite widely. As a result of that study, I was asked to take part in a Department for Education and Employment and Committee of Vice-Chancellors and Principals working party that considered the applicability of PFI in the higher education sector. I think that I managed to influence the tenor of that report.

We have a number of questions on various themes to put to you.

Good morning, gentlemen. My question is primarily for Professor Bain, but the other witnesses may come in if they wish. PFI/PPP has often been promoted as a way of managing public finances better, in that it can reconcile the public's demand for increased capital investment with avoidance of difficult impacts on public finances. Why is it better to go down the PFI/PPP route than to take traditional procurement options?

Financing is relatively unimportant. Much more important are the other aspects of PPP that relate to the way in which projects are controlled and to the transfer of risks in such projects. The funding does not seem to me to matter very much. Practically, it is probably difficult to get the other advantages that go with PPP without having an element of private financing as part of the project.

Are you saying that if the element of private financing were missing, some of the benefits or advantages would be lost?

It would be difficult to set up a project in the way that such projects are now set up unless an element of private finance was involved. It would be harder to disassociate, or separate out, such projects from what has traditionally happened with public sector projects. The PricewaterhouseCoopers paper brings out a lot of the advantages, although perhaps some of the disadvantages should be brought out a little, too. Although it should theoretically be possible to capture those advantages however a project is financed, I do not think that that is how things work. An element of private finance is an aid to dealing with projects in the ways that they are dealt with through PPP schemes.

Can you explain why?

When projects come under public sector control, or have public sector financing, one comes up against overall aggregate capital controls from the Treasury. In the past, there have been individual controls on particular operations. I give as an example the system under the University Grants Committee, which was disastrous. The UGC controlled the amount of funding and went through the detail of the project, with the result that the buildings that were constructed during the 1960s and 1970s were poor. If we can get away from those constraints, people will be able to focus on what is required by the public sector entity in the long run and we will ultimately get a better project. That is the great advantage of PPP. I do not think that that could happen as long as the Treasury maintained overall control of a project.

Are you talking about Treasury control of the borrowing requirement rather than about the revenue spend?

There are two levels. The Treasury is concerned about the aggregate public sector borrowing requirement. As a result, the Treasury rationed funds for a lot of public sector organisations, which led to far too little investment in infrastructure in the UK. PFI projects have released bodies from that constraint in a fruitful way. Furthermore, when direct Government expenditure was involved, there was a great deal of detailed control—I am tempted to call it interference—from various levels of Government of what people wanted to do. That was not fruitful.

If the local authority or whatever body was involved could borrow and was then willing to contract out the design build and construction servicing, why should that be subject to Treasury interference any more than if the body had sourced the capital privately?

If such activity is not subject to Treasury interference, there is less of a problem. From working with other bodies, I have found that there has frequently been an element of interference from the Treasury and perhaps even the Scottish Office. One had to go through various layers of reassessment and attempts to change things for what appeared to be not very good reasons, with the result that the projects were not as good as they might have been. Removing projects from direct public sector financing would allow us to handle matters more rationally.

There were three phases in university building. You referred to the first phase, which happened in the 1960s and early 1970s. At that time, the Treasury's interference in the universities was enormous. For example, I sat in the court of Heriot-Watt University, which had to turn away a £500,000 donation because the Treasury decided that the building specification was too high, even though the costs to it would have been lower than for an average student residence. The same thing happened at the University of Glasgow.

In the 1980s, the business expansion scheme—BES—was a system of public procurement that allowed the universities to raise money on the market. Was there still Treasury interference in the building programmes at that stage? It seems to me that that scheme constituted a different relationship between the universities and the market.

The business expansion scheme centred wholly on the private funding of university buildings. It was almost entirely directed at residences and was done in a way that was entirely free of Treasury control.

You mentioned interference in the 1960s and 1970s in the minutiae of the building plans. I was certainly aware of that interference. At the time, one could not spend more than a certain amount a head, even though such an approach was totally counterproductive in terms of building maintenance in Scotland, as it did not take account of the cold climate, for example. As that interference was not present in BES funding, why did such a procurement system, which allowed the universities to raise funds from the market and to guarantee a return to the investors, not work effectively? Why did we need to go on to PFI?

I think that the business expansion scheme was effective. The working party in which I was involved indicated that there was no good reason for replacing BES funding with something else. The scheme worked perfectly well in that situation.

I understand that some universities may be considering refinancing projects that were BES funded through something like PFI. In other words, they will sell on residences under a long-term contract for someone else to provide the accommodation, presumably because that is how they want to manage their balance sheets. I am all in favour of that. However, I should point out that BES was a rather good model of private, not public, financing.

The distinction between capital and revenue finance is not always as clear as it might be. Indeed, if we pursue the PFI/PPP route to its natural conclusion, there will be no capital finance. As the Treasury has interfered before in determining the overall amount of capital and the design and operation of the building, what is to prevent it from interfering with the revenue consequences of PFI projects? The capital may come from private finance, but the revenue consequences will still fall on the public purse. In fact, I think that the Treasury already interferes in that respect.

As far as I understand how things work at the moment, Scotland receives a block grant, which means that it would not be the Treasury but the Scottish Executive—and ultimately the Scottish Parliament—that would determine the funding for, for example, health organisations. Although the Treasury might wish to set some limits on the commitments that particular organisations can enter into for the future, that will be done on a broad basis instead of through simple detailed interference.

Do you have any advice about the proportion of the budget that should be spent on service infrastructure and long-term maintenance costs and on the proportion that should be set aside for services that are not so capital dependent?

I am certainly not in a position to give any general advice on that point. One would be required to examine the details of each service in order to reach a sensible view on that matter.

What I was trying to get at was, if the principal advantage of using private finance is to bypass central Government interference in capital spend, what could be gained from dealing with controls at that end rather than at the revenue end?

I am sorry; I did not say that that was the principal advantage. I think that it is an important advantage, but there are others.

The constraints on capital expenditure for the benefit of public sector organisations are pretty artificial. For example, the PSBR is a rather artificial accounting concept. In economic terms, it does not make much difference whether the resources used to build a hospital are financed privately or publicly. The same building resources are being used. However, there is a notion of a PSBR that needs to be controlled. One could debate whether it is sensible to have any overall limit at all; in fact, I think that such a limit is sensible, but for political rather than economic reasons. It is difficult to constrain overall expenditure proposals unless some limit is set for people to operate within. However, from an economic point of view, I do not think that such limits exist. It would be an advantage if one could finesse other constraints that exist for other reasons and get more capital into the public sector at a time when such expenditure has lagged behind what it should be.

I am interested in your comments about the PSBR. You mentioned that the rationing of funds has had an effect on public sector capital projects over the years. The Treasury justification for PFI rules seems to lie in the golden rule that the Chancellor of the Exchequer has stressed in relation to sustainable investment and public sector net debt as a proportion of gross domestic product. Does any logic underpin those rules?

There is a logic that says that current expenditure should be financed from taxation. However, there is much more debate about how much of capital expenditure should be financed by the current generation of taxpayers and how much should be financed by future taxpayers. There are no golden economic rules on that.

The golden rule applies to current expenditure and taxation. In the past, when everything was counted together, it was much easier to squeeze capital than current expenditure, which meant that capital expenditure was cut. PFI/PPP offers some escape from that system, if only because the Government needs to be more disciplined. If the public sector does not finance much capital expenditure, it is compelled to cut current expenditure if cuts are needed. That is as it should be if the Government does not want to raise taxes.

You said that the level and limits of the PSBR were not specified and that that was a political rather than an economic decision. Has the recent expansion of PFI projects affected that situation? If PFI continues to expand, will it be possible to consider the PSBR on more economic terms?

The PSBR is probably always going to be more important politically than it is economically. PFI has grown quite a lot, which means that we are well within Maastricht limits at the moment—they are not a problem. If we added back all of PFI, I do not know what the figure would be—perhaps one of my colleagues here could tell you—but I suspect that it would be close to the Maastricht limits. PFI is a device to ensure that the figure remains inside those limits. The important rule is not the overall PSBR limit; it is that current expenditure is financed from taxation.

Are you aware of the way in which the share of the PSBR that is available to the Scottish Executive is determined?

Not in detail.

That is not the subject of this inquiry. We can deal with that another day, perhaps, in writing.

The constraints on the revenue side, as a consequence of the move to PFI, are inevitable, especially in circumstances in which there is a block grant. Would it be wise for the Government to arrive at a guideline figure on that?

Not an overall guideline figure. Let us consider the position of a health trust. It has to think about how much of its funding it wants to commit in long-term contracts and how much it wants to retain to use more flexibly. That decision should not be taken at an aggregate level for Scotland as a whole; it should be a decision for the health trust, which has to satisfy the needs of a specific area.

That question was aimed at Professor Bain because it concerns a macroeconomic issue. However, if Mr Martin and Mr Brewer want to add anything to his answer, they are more than welcome to do so.

The discussion began with a question about the need to include private sector finance and it has focused on the management of public sector finances. At the micro level, the involvement of private sector finance in PPPs and PFI projects is as important as, if not more important than, the Government's financial management, although for other reasons. Private sector finance enables a project to be structured so that a single private sector operator takes responsibility for all aspects of the delivery of the service that is required. The fact that its own capital is at stake provides an immensely strong incentive for the operator to meet that requirement. Moreover, the involvement of significant amounts of private sector finance imposes disciplines at the inception of the project that the public sector would have found difficult to introduce of its own volition. Quite often, the need to conduct full diligence on a project is prioritised over the need to meet timetables, which is often the principal driver for those delivering capital projects in the public sector.

Professor Bain talked about giving individual NHS trusts discretion in committing to capital projects. However, that would be to attribute to trusts more independence of action than they have. Capital and revenue funds are largely allocation based, using a formula that is determined by what is committed to specific projects known at the macro level in Scotland. Individual trusts have some discretion regarding smaller projects based on formulae that are determined by the Scottish Executive health department.

At the macro level, there must be central monitoring of the implications of recycling annual capital charges in the public sector into annual payments for contracts funded by the private sector. The problem is that health sector finances treated capital charges on a circular basis. We were given money that covered capital charges in our allocations and we tended to pay that back in cash to the Scottish Executive budget. Under the PFI regime, savings from capital charges are recycled to pay cash out to a private sector provider. At the macro level, monitoring must take place to ensure that the Executive's health department takes account of that in setting new capital limits for the health service as a whole.

Thank you. Let us consider a specific aspect of PFI and PPP projects. In earlier evidence, we heard that it is essential that the public sector contractor ensures that the initial conditions are clearly specified. Some would regard that as a disadvantage, as it can slow down the process in the initial stages. However, it is also thought to have advantages in the long term. Do you agree that specifying the initial conditions carefully is important? If so, what initial specifications are critical to the success of a PPP or PFI project?

It is important to focus on two aspects of the initial conditions. The first is the public sector's specification of what it needs from the assets or services that it is seeking to procure. That may sound glib, but PFI has the discipline of ensuring that public sector specifications are fully thought through at the outset of the project. PFI does so not by requiring the public sector to say, "We require a hospital that is two storeys high, with this many general wards and operating theatres," but by requiring thought to be given to the functional requirements—the aspects of the building that make it work well. That is a different kind of specification from the kind that one would give to a builder who was structuring a normal building contract. Because the specification is drawn up in a different way, it requires serious thought to be given to a building at the outset. Hospitals do not perhaps provide the best example, as large hospitals are such immensely complex projects that they could never get off the ground without deep thought. In other areas, however, PFI has imposed a greater discipline.

The second aspect is those features of the relationship that relate not just to the functional service, but to the PFI relationship. It concerns the powers of the public sector to enforce the specifications; the financial deductions that will be imposed on a contractor who fails to meet the specifications; and what happens if the contractor fails completely to deliver, or, for whatever reason, the public sector defaults. Those issues require careful thought, but there is now significant standardisation as sectors learn from past projects. My recent experience of projects is that those issues are now addressed a great deal better and are less of a practical barrier to the quick completion of projects. The drawing up of specifications for individual projects always needs to be thought through extremely carefully. However, in sectors where PFI is taking a hold, each project is learning from past projects and completing that part of the process more effectively.

I agree with Paul Brewer. The specification is important and takes a long time to draw up, especially in areas such as the health service. We are not just talking about building a hospital—a place built out of bricks and mortar; we are talking about providing facilities for the appropriate treatment of patients. In best-value terms, that means that there must be considerable consultation with the local population and the health service professionals about how services are to be delivered most appropriately.

There are examples of compliance with those requirements. For example, the acute services review in the Lothian Health Board area commenced and reported in 1993. The full business case for the new infirmary was closed four years later and the handover of the first phase of the building is imminent, as members will be aware. The process is slow, but it is deliberate and ensures that there is full consultation with all the parties involved.

Is there evidence that the amount of time taken in putting together the specifications was shorter in the days before PFI was on the scene?

I do not know whether PFI has necessarily lengthened that process. There has been a catalytic process in relation to PFI. First of all, the public sector has decided that it needs to meet the challenge of improving its services. Following that it is forced, through the completion of the public sector comparator, to build a full specification of its requirements and, through competition or otherwise, to challenge the private sector or its own resources to deliver the deal that will give the best value for money for patients and taxpayers.

It is difficult, if not impossible, to set the initial specification in stone. Often, the specification or the user requirements change during the implementation phase or the planning process. Do PPPs and PFIs offer enough flexibility to deal with that?

There are two sides to that. It is important to involve users and think carefully about the specification. When the bricks start to be laid, you have to know what you need. Traditionally, in the public sector, a great deal of cost overrun arose from the fact that there was a lack of discipline and users were able to change the requirements once a contract had been let. The discipline of having to think through the specification fully before the PFI contract is let is beneficial in that regard. The corollary of a PFI contract is that cost overruns do not impact on the public sector, which is an immense step forward in terms of procurement, because decisions about the specifications must be complete before the contract is let.

Once the PFI contract has commenced, there is scope for variations. Some PFI contracts include specific sums to allow users to vary the specifications. It has to be recognised that, when it comes to specifying something as complex as a school or a hospital, nobody is perfect. There must be give and take. In relation to more substantial variations arising from changing needs over the years, it is often said that PFI is inflexible. However, one must remember that a traditional procurement regime and a PFI scheme both involve the building of an asset that is designed to last for about 60 years. Change is a difficult issue in capital projects of any sort and the fact that change has to be implemented within the framework of a PFI contract does not impose any significant constraints that would not have applied under a traditional contract.

On the initial specification, the PricewaterhouseCoopers submission says that, prior to the commencement of procurement, all decisions must be based on

"a robust assessment of all relevant factors. This assessment must include the view of stakeholders and potential service providers."

The ICAS submission perhaps contradicts that when it says that consultation on services and service delivery should be independent of financing and should take place before decisions are taken on the financing of a project.

How can you involve stakeholders—the users and the staff—at an early stage in the process without contravening the concept of commercial confidentiality?

I tried to explain this earlier but obviously did not do too well. In relation to hospitals, PFI would be used to deliver buildings and services. The other challenge is to do with reconfiguring the way in which doctors deliver the current service. The business case and the acute services strategy in Lothian are components of a best-value regime that was designed to challenge the previous regime of service delivery with new performance targets to be achieved by clinicians and nurses through a different way of working. That process requires significant consultation with the service users, which was largely achieved by involving staff in clinical task groups that examined the delivery of the clinical specification of the hospital. That is not necessarily in conflict with the commercial aspects of an individual deal for the delivery of the facility in which the users are involved in setting the specification and expect the private sector contractor to deliver to that specification.

I agree with that. There are not, nor should there be, any limits on consultation at the point of setting the specification. If a project is designed to meet public sector service requirements, the users of that facility—those who work in the institution or are consumers of the services—need to be taken fully into account. If the end-product does not meet their requirements, it has failed. In relation to commercial confidentiality at the point of setting the specification, the public sector must manage the consultation process in order to balance expectations about financial constraints against the logistics of gathering many opinions while meeting the project's timetable. That is not a PFI issue.

In the procurement phase, bidders will come up with differing proposals. It is important that, as far as possible, the stakeholders' views on those proposals are taken into account. At that point, a genuine issue of commercial confidentiality can arise. For example, in the case of the Glasgow schools project, the successful bidder came up with a proposal that involved 12 new schools when the council expected to get only two new schools. That was a unique selling point in that competition and the competition would not have operated effectively had that information been revealed to the other bidders.

Those who are consulted during the competition period need to be aware of the confidentiality requirements. Although it is a matter for public sector judgment in individual circumstances, that need for confidentiality puts practical limits on the extent of consultation and, clearly, public consultation is not possible. The private sector is becoming more relaxed about the extent of consultation and the commercial confidentiality surrounding the competitions.

On that point, you mentioned that your company was involved in the Glasgow schools project, which is an issue that has been raised before. In your submission, you refer to

"representatives of teachers and school boards on school projects".

You explain that, in the Glasgow example, public involvement was specifically limited for reasons of confidentiality. You talk about school board representatives not being able to divulge information. In the case that I am aware of, that caused great resentment locally, because some of the facilities being provided in the new schools that were being built were not in line with what the community as a whole wanted.

I am aware that there has to be a limit to the information that can be given. How do you get over that problem? Should that be dealt with at the design stage in a design specification? Should the local authority demand a certain level of facilities for the community and then invite the bidders to meet that, rather than have bidders competing to provide more—or, in this case, fewer—facilities to achieve preferred bidder status?

It is the former: as far as possible, the requirements of the project, including parameters and limits on the final configuration, should be specified at the consultation stage. In that instance, judgments had to be made about whether the extent and location of swimming pool facilities, for example, were to be specified absolutely. The bids all came back within the specification, so clearly decisions were taken at the specification stage that some stakeholders were not satisfied with.

On the consultation about new or refurbished schools, it was certainly the council's view that there were significant advantages in having new schools as a teaching environment. There were other consequences for the configuration of the schools, which again perhaps did not meet the requirements of all users. The public sector managers of the project were put in a position where they had to decide between bidder A's proposal not to build new schools and bidder B's proposal to build new schools. Immediately after that decision—following the announcement of the preferred bidder—the preferred bidder went into a series of consultations to advance the project.

We should not become too parochial.

Both the public and the private sectors have learned a great deal from their initial experiences of PFI/PPP. What impact has that had on the most recent PFI projects? Are they better designed contractually? Can you provide specific examples of how the early experiences have informed and helped with more recent projects?

One could take the custodial services sector—prisons—as a case study. The involvement of the private sector was significantly more expensive in the early projects of a programme initiated in England and Wales. Although there was significant overseas experience of designing, building and operating prisons, the private sector in the UK did not have an operation of that nature.

When the private sector saw that there was potentially a programme of new prisons, its response was to look hard at how to design, build and operate prisons. More specifically, the early prisons were closer to traditional construction. From that, the private sector moved to a highly modular approach to construction. It was prepared to invest money up front on developing a much cheaper method of obtaining the same quality of building because it knew that there was likely to be a stream of new projects. The capital cost of building prisons through PFI tumbled and has become significantly lower.

That is a good example, in that HM Prison Service has been willing to identify the lessons that it can take on board from what the private sector has done. The programme is constantly under review; the service constantly considers where the market has got to and whether it is still doing things the right way. At the moment, the decision is to continue with a programme of prisons along the same lines, but it is important that the public sector keeps reviewing where the markets have got to and where the balance between public and private input should lie.

Would not that happen under any circumstances? A vast reduction in cost would be expected with a programme that involved building 10 or so prisons—all designed in the same way, or at least using the same building blocks—regardless of who built them or how they were financed.

That is quite correct, but in the past it could not be assumed that new prisons would be designed in the same way. Because the private sector had no control over their design and operation, there was a much greater diversity prior to the introduction of the PPP programme. The private sector brought its own models to the services that were being provided. Although it might be debatable, there is no evidence that the programme would have been delivered in such an innovative way if responsibility had not been ceded to the private sector as a result of PPP.

Although you concentrated on the private sector, the public sector has also had opportunities to learn from the PFI/PPP experience. How can the public sector ensure that that experience is more widely shared throughout the sector, and how can it be translated into improvements for public sector procurement?

As I understand it, the PFI/PPP route accounts for less than 10 per cent of the total. Presumably, such experience can help better to inform public sector procurement as well. Do you have any evidence to suggest that? What impact will such experience have on drawing up public sector comparators? As a private sector company sharpens its pencil, the public sector comparator should sharpen its pencil too. When we come to assess whether we are getting value for money, that should be a moving experience rather than one that takes place against some fixed historical arrangement.

In your experience, has the public sector reacted in the kind of way that I have suggested?

As you rightly say, there has been an education process. I do not think that there has been a formal huddle within the public sector on how to consolidate that experience across the range, either within health or local government. The fora for the exchange of ideas in the various accounting institutes represent a route that can be followed without specific public sector involvement. That is why we have a public sector committee within the ICAS to disseminate those ideas—not only among a small group, but around the membership—by means of technical newsletters and briefings.

The public sector had a private finance unit within the then Scottish Office health department. Expertise on the legal side resides within the central legal office of the Common Services Agency. From experience, I know that we have had to consult with external advisers on the legal side. I also find that interaction with those on the legal side in the public sector has been positive. They have come up with some good ideas on, for example, managing contracts of a softer nature in information technology.

Will you give some specific examples? We are trying to obtain appropriate case studies, because most of the stuff that we get tends to be anecdotal.

The best example of public sector dissemination of education is through the accumulation of guidance from experience. Over the past four or five years—ever since the inception of trusts, capital charging and PFI—more detailed and helpful guidance has been given to public sector managers and finance staff on specifying the financial and economic aspects of reviewing projects. That experience is out there. There is an education process, and we are expected to comply with that guidance. We therefore achieve a uniformity of projects that the Scottish Executive health department can review and evaluate. That is probably the best way in which we can bring about a learning process.

Having said that, there are problems. For instance, if we take one of the biggest projects, the new Edinburgh royal infirmary, the managers who have been involved in that trust have been subjected to other political and personal changes. In the space of about four years there have been four chairmen, three chief executives, four finance directors, two human resources directors and two directors of nursing. Such disruption means that the education process within an organisation and throughout organisations tends to get dissipated. That is a risk when public sector organisations have to go on to manage and monitor the performance of contracts after they are signed and delivered.

That has been suggested to us elsewhere. Is it a wise use of experience in the public sector to keep chopping and changing? How will we ever learn if we do not have a certain amount of stability?

It is important to keep a core bank of expertise at the centre so that the potential educational value of the PFI experience is not lost.

I would go a little further than Mr Martin in his specific reference to the health sector. If one considers other sectors, the point about public sector expertise not being vested in people whose jobs are rotated at regular intervals is an important one. However, that has been recognised in Scotland. There has been less than the normal regularity of rotation of civil servants in and out of the private finance unit of the Scottish Executive. We have found that valuable because a good level of knowledge has accumulated and is available.

On people implementing new projects and learning lessons from those who have done it already, there has been encouragement for the sharing of lessons. If one takes the schools projects, in the local authority sector there is a regular forum of the individuals leading the projects, which is focused on learning the lessons and explaining, for the benefit of the next project, how one project has dealt with its difficult issues. It picks up on all the non-project specific issues. My understanding, from people I have worked with, is that that kind of forum is a good way of spreading the more practical lessons. The standard guidance that is available to all public sector bodies is also a strong step forward. What it avoids, in the generic areas of PFI, is the wheel being reinvented on each project. Ironing that aspect out of the process has been beneficial for PFI.

Finally, the lines of communication are shorter in Scotland than they are in England and Wales. The Scottish Executive private finance unit is closer to projects and the individuals who are implementing the projects. That has been valuable. If one considers the requirements for the outline business cases that local education teams are being required to propose in mid-December, a lot of the emphasis is on "How have you learned from the lessons of past projects? What are the practical issues that you need to address and demonstrate now, when you are going for initial approval? Have you thought about how you are going to address those issues?" Things are moving forward in that respect.

We have talked about specification and what happens when you get through to the competition phase. However, once the contract is awarded and you are moving into its implementation, what are the critical factors that will make that implementation a success or failure? Can you give us some examples?

The first general point is that in procuring big, difficult, complex capital projects, PFI will suffer from many of the same problems as any other form of procurement. I see that there have been a number of press reports about the progress of the royal infirmary and Glasgow schools. Those are some of the biggest, most complex capital projects that have been developed in Scotland. The fact that a project has a PFI structure will not wish away all the practical difficulties that the contractors face on the ground. Much of the press coverage has labelled the practical difficulties as PFI difficulties. It is important to recognise that.

On the implementation of the project, one of the important aspects is that the capital project is closely monitored and the whole life-cycle maintenance programme is carefully monitored by the financiers as well as by the public sector. That aspect of implementation is self-monitored within the PFI structure. It is structured so that if there are difficulties in delivering to the specification as well as the payment mechanism, which has financial consequences for the contractor for failing to deliver, financiers should always be looking out for problems, because they are first to get hit when projects start to face serious difficulties. Teams are engaged throughout the life of the project to undertake that monitoring.

On the public sector's role, it is essential from the outset that the focus is on how we make a successful partnership or relationship work for the life of the contract. If there is one aspect of the structure of PFI that is not geared up to that, it is the fact that it is creating an arm's-length contractual relationship in terms of the structure. The more forward-thinking procurement authorities realise that the contract is one aspect, but it is in substance a partnership that will not work effectively around invoking a contract. It will work effectively around constant consultation and each party understanding the other's needs and difficulties, to ensure that, before one even considers the contract, the relationship works well. Well-structured PFIs have an implementation phase structure that ensures that the dialogue is constantly kept up, and that problems are identified and dealt with as far as possible before they become issues.

Many people are concerned about PFI projects becoming arm's-length projects. There are issues about how they remain accountable to the end user. You have talked about how that can best happen. Are you saying that much of that is within the design of the original project and the approach that is taken towards partnership?

There are strong incentivising factors in the design of the original project. An example of a project that has gone horribly wrong is Eurotunnel. Users can catch Eurostar at Waterloo and find themselves in Paris. They pay a fare that is determined more by market forces than by the needs of financing the project. Eurotunnel's shareholders and debt providers have taken a horrible loss but the end user is relatively unaffected. It is part of the structure of PFI projects that it is always those who finance the project from the private sector side who, in the event of catastrophic project problems, absorb the financial cost. That is a scenario that a public sector manager operating his project does not even want to contemplate. It is important that there is a focus on making things work well day to day.

While PFI contracts have fundamental strengths to protect the public sector from the consequences of failure, day to day it is much more about the general incentives and payment mechanisms. Therefore, if relatively minor aspects of the specification are not dealt with, there is a meaningful financial penalty. There are also reporting and monitoring mechanisms to ensure that things are picked up and dealt with quickly.

I want to home in on a specific area where we know that there are problems: the cleaning of hospitals. The original system was that the cleaners were part of the team—they were employed by the public sector, which determined what the cleaners would do. The next stage was external procurement. The services were contracted out and, when the specifications were wrong—as they have been on a number of occasions—we ended up with an uncontrolled, unmanaged situation in which the specifications were met but the hospitals were not clean.

Correct me if I am wrong, but I understand that under the PFI contract, the design of the wards and facilities will maximise the cost savings on maintenance and cleaning. The facility will be designed to be easily cleaned because that will keep costs down.

However, beyond that, the private contractors in the PFI partnership will still be contracting out their services in the same way as before. How do we control that better than we have done in the public sector or the intermediate external procurement system? Having got the design part correct, what is the advantage to the public sector of using external services for 30 years? Presumably, we cannot order parties to the contract to change it—that would be a matter for the private provider.

What you are saying is correct, with one or two provisos. You are right to say that under a public sector comparator there is still an option for the public sector manager to determine whether particular services—defined as non-core services—are contracted out. That decision will continue to be made on an individual basis according to the circumstances facing the managers at the time and what managers feel is appropriate for the service to patients. Ultimately, the patients in hospitals are the customers and they want clean hospitals.

In my experience of specifications under PFI contracts, the initial contract period provides an opportunity to examine the value for money delivered by a contractor. That is done by scoring the quality of service against the specification. If that performance does not meet the specification, as Paul Brewer said, there are contract penalties that have a financial impact on the service provider. Therefore there is an incentive for the service provider to put things right.

There are also opportunities for the public sector to market test the price throughout the stages of the initial contract. Again, therefore, the public sector has the opportunity to address issues of quality and value for money during the initial period of the contract.

As Paul Brewer also said, it is important that managers and service providers meet regularly to monitor individual issues that might affect the quality of service. An example of such issues is the actual levels of sickness and absence compared with what was assumed when the specification was made. We are dealing with employees who have been protected under the Transfer of Undertakings (Protection of Employment) Regulations and it is important that any issues arising under those regulations are dealt with on a day-to-day basis.

I think that Paul Brewer said that in the case of Eurotunnel the risk was borne by the shareholders. Is it the case that if Eurotunnel went belly up, the UK and French Governments would have to jump in and set up a mechanism for running the channel tunnel? Politically, it would be unacceptable to close the tunnel down.

That is the point that I was making. If services are being procured by the public sector, there are checks but it is important that the services continue. Eurotunnel had catastrophic financial overruns that were absorbed by the banks and the shareholders. Those overruns would otherwise have fallen to the public purse.

If a PFI project suffers similar difficulties, the financial losses and cost overruns will be absorbed by the private sector under the PFI structure. The private sector has no contractual right to ask for more money. Then we might be faced with the scenario where the private contractor fails to deliver or stops providing the service. There are protective mechanisms in the PFI contract to put the contract back out to tender, if the service can be delivered by the private sector. All the financial losses that would be faced by the public sector in doing that will fall on the contractor whose finance was at stake in the first place. If we take that route, the public sector is kept whole.

Alternatively, if a private contractor fails to provide a service, an expert valuation of the project can be done so that a compensation sum can be paid to the operator. Following payment of that sum, the asset comes back to the public sector. The compensation is calculated to leave the public sector no better or worse off.

I do not think that those financial mechanisms change the substance of PFI contracts, which is that key public services must be the responsibility of the public sector. However, the mechanisms provide a lot of protection before we reach the point where the public sector has to step back in.

The PFI mechanism protects the public sector from limited cost overruns. However, when those overruns become large, as they did in some information technology contracts, the contract eventually becomes unenforceable and the public sector picks up the bill. Therefore, we should not exaggerate the extent of protection.

The quality of service has been discussed. I find it remarkable that people should want to put out to tender 30 years' worth of services, except perhaps building maintenance. However, it probably makes sense to outsource cleaning under a three-year contract. If people are not satisfied, they can then fire the contractor.

For services such as cleaning, PFI contracts tie bodies into using a particular contractor and make it much harder to get out of arrangements. That is all right as long as the relationship is going tolerably well. However, I suspect that if relationships become difficult, the public sector will find the PFI contract harder to handle than a straightforward outsourcing contract. There are horses for courses and I suspect that the PFI goes too far in some directions; some things are better done in other ways.

I have a final point on flexibility. Twenty-five years down the line, it will be easier to adapt and change if the public authority owns the building rather than having to negotiate with its private sector partners. Those partners might have slightly different agendas and relationships might be rather difficult. Everything can be written into a contract, but it is difficult to be flexible when everything is written down in a contract.

One has to watch out for a possible loss of flexibility when using long-term PFI contracts. After all, in others areas of business, there is a lot of criticism of long-term leases for buildings because they tie businesses in and are said to be unfair. We are trying to change the nature of those contractual relationships to make them more short term but, in the public sector, we are trying to make similar contractual relationships more long term. That does not quite stack up. Issues must be watched. The jury is out and will probably be out for at least another 15 years, after which we will know whether it is working.

Is it the case that the contractors whom we are discussing will not be around in 15 years, far less in 30 years? They will have been taken over or merged and the original identification with the project will have been lost. Given the rate at which local authorities and health boards are reorganised, those bodies will not be around in their current form either. In 15 years, the two partners to the contract will be totally different. Might that not have an impact, too?

It might well do.

We heard how PFI/PPP projects are better partly because they allow better project management techniques, impose more discipline and adopt life-cycle costing techniques. If the public sector adopted such techniques, could it not get close to the results that PPP projects are alleged to deliver?

It is important to keep that point under review. Experience shows that the manner in which the over-life responsibility for a project has been put out to tender in the private sector has generated different approaches to dealing with the balance between initial expenditure and over-life maintenance. Design innovations have been generated and change in the provision of public sector assets has been accelerated in a way that would have been difficult to achieve in the public sector.

There is one structural area in which it is difficult for the public sector to emulate PFI results. In any major capital project, the public sector has a relationship with the private sector. The public sector does not build and in some cases does not maintain. The splitting of design and build responsibility from maintenance responsibility leaves a situation in which the contractor who undertakes the design and build does not have the ultimate responsibility for the quality of their work. Potential difficulties arise from construction contracts in respect of the defects period of design and build contracts. During that period, there is scope to take action against a contractor for work that does not meet specifications. The contractor does not have to work in his own building, so if the design causes functional difficulties, that falls straight back on to the public sector. If there is a break between the design and build, and the maintenance contract, the operator does not have to pick up the maintenance costs that are associated with his own design.

Are you saying that PPP is needed to overcome the inadequacies of the private construction industry?

If a private contractor builds something to a public sector specification and does so extremely well, that does not mean that the contractor must pick up the consequences down stream of relatively high maintenance of the components that they have used. Under PPP, they would have to think not only about using low-maintenance components or economising on the capital build as there would be better value in a more intensive maintenance programme, but if they get things wrong, they would suffer the consequences.

You are saying that private contractors will build better if they know that they rather than the council will have to look after what they have built.

One can distinguish between the design and the build. The private contractor can design and specify projects better and can build to that specification.

Is there anything inherent in the public sector that means that it cannot bring in projects on time and on budget?

There is nothing inherent in the public sector that means that it cannot do that. The public sector contracts a great deal of the work on projects to the private sector. As a means of managing private sector relationships, that is more effective and provides stronger incentives for the private sector to deliver the public sector's requirements.

I agree with Paul. It is important to distinguish between a public sector comparator for projects that follow closely the methods of a private contractor—and are built to the kind of specifications that we are talking about—and the matter of handling risks in the longer term. The public sector must consider how to utilise an asset in 25 or 60 years and whether it can offset the risk in some way. The public sector must specify a quality of construction that ensures that projects will have a flexible use further down the line. When the public sector decides that a building is no longer needed for the purpose for which it was built, it can walk away by giving a private sector provider the use of the building and the opportunity to make revenue out of it. We must adjust the public sector comparator and the anticipated costs by taking account of the risks of future obsolescence or inappropriateness for purpose. The private sector has shown that it is willing to take on that risk.

Design, build, finance and operation involve four different disciplines and different skills, which will not necessarily be found to the same degree in one organisation. Why is the private sector better placed to offer those skills as a package than the public sector? Given that when projects are built it is likely that the asset might be sold on in future, why should we pass that income stream on to the private sector as opposed to the public sector?

Professor Bain mentioned that compulsory competitive tendering was a disaster for cleaning and catering, particularly in the health service, because standards dropped and it was difficult to enforce higher standards. Why should we be confident that in the long term—towards the end of a 25 or 30-year contract—we will still be able to enforce standards? Is CCT the best way?

There are two reasons. CCT is a management process and it is important—

As Professor Bain said, one alternative that was suggested was that operations that were not connected with the long-term maintenance of a building might be better dealt with through a shorter-term contract. However, when it was done that way, standards were not maintained.

That is right. However, if managers are given flexibility to make such decisions, good decisions will be taken in partnership with the staff. That has been an increasing trend in the health service latterly through the development of partnership working and the publication of guidance on staff governance and partnership working in the new health plan.

We can achieve value for money in two ways: first by choosing the best and most economic offer on the table and, secondly, by ensuring that, where appropriate, risks are transferred to the private sector. It is then incumbent on the deal makers to deliver a contract and project structure that enables value for money to be demonstrated in each individual business case. If risks are appropriately transferred to the private sector, which has shown that it is capable of building in project mechanisms to manage risk, that minimises the public sector's exposure to the downside risks of inappropriate builds and further maintenance costs down the line. The most important aspect of private sector involvement is the lay-off of risk.

Brian Adam asked why the public sector cannot bring together the design, build, finance and operation of projects. One must look at the types of projects that have been undertaken in Scotland and elsewhere. The design capability must often come from the private sector because the nature of very large capital projects is such that they are relatively rarely undertaken by the public sector. The build needs to come from the private sector. We have discussed the reasons why the finance comes from the private sector, but the question of value for money should also come into the financing of such projects. The operation needs to come from the private sector because heavy maintenance duties do not exist in many parts of the public sector—although the public sector is often responsible for more routine responsive maintenance. Throughout the design, build, finance and operation spectrum, a significant private sector involvement would be required in any event.

The question is not simply whether the public sector can take that on, but how it can organise its internal relationships and its relationships with the private sector to make the whole blend work most effectively. Design, build, finance and operation are components of every capital project.

The design, build, finance and operation of the Skye bridge is an example of the point that I was trying to make earlier about how the contract can move fairly quickly. In the initial stages, an agreement was entered into with one organisation, but the agreement was then moved on to another organisation. Are there not risks involved in moving the contract on? Why is the contract not tied to the consortium or company that is doing the design, build, finance and operation? The contract seems to be between whoever owns the contract and the public sector. We may contract with one consortium in year 1 but be on to a second consortium in year 5 and a third by year 15.

There are different ways in which that could happen. The shareholders in the special purpose company that was formed to take the project forward could sell their shareholding to a third party, or the companies that are involved in the project could be taken over. However, whoever the shareholders are, they still have an economic incentive to ensure that the whole thing works effectively. The important point must also be made that in every PFI contract there are controls over changes in shareholder identity.

My point is that the contract is not only about money, but about the relationship between the contractor and the client. If the contractor has the right to sell on the contract, the risks end up back with the client. Where is the public sector security in that? How much finance should the public sector hold back to cover that risk?

For example, people have tried to look into the Skye bridge contract, in which there has been considerable movement in terms of who operates it and who has done, and still does, the financing. I asked questions about that, but found it difficult to get answers. Many political questions are involved, which are not your responsibility, but the fact remains that some development companies are not especially interested in operating the facilities in the long term and have the idea at the back of their mind that they will move the contract on. Where is the public sector risk in that, and how is it accounted for, in terms of overall value for money?

The first way in which ownership, or the identity of the contractor, could change is when the project company changes. The project company then has subcontractors who are responsible for delivering the service. If a subcontractor—with whom the public sector generally has the day-to-day relationship—changes, the overall requirement remains for the service to be delivered to the same standard. The question is whether it is important to the public sector whether contractor A or his successor contractor B paints the white lines, or maintains the surface of the Skye bridge, if the same standard of service is delivered.

The important question is whether the public sector can be assured that the service will be the same if a subcontractor changes. Again, there are contract protections to ensure that a significant subcontractor cannot just change without the public sector having a say. If it is decided to change a subcontractor and the successor subcontractor fails to deliver to the required standard, the project company will be subject to all the financial deductions that are due. The banks, or whoever the lenders are, will have their money at stake if the project fails to deliver. As a result, financiers whose money is at stake have generally looked extremely hard—probably even before discussion has taken place with the public sector about changing the subcontractor—at the consequences of such change for the contract and for the public sector service.

Even the financing of the project might move. The bankers will generally underpin the whole contract. The risk for the public sector of being involved with one financier might be considerably different from the risk of being involved with another financier.

Was there any consultation with the public sector on the various changes that took place in the financing and operation of the Skye bridge contract, which is one of the longer-established PFIs in the country? Is there anything in that contract that says that the public sector must be consulted over changes or that it has the right of veto? Has the Skye bridge contract offered any protections to the public sector as a consequence of the various changes that have happened subsequent to its being signed?

Unfortunately, I do not have any personal knowledge of the Skye Bridge contract.

That is not what I was asking.

Nor do I have such knowledge. In a typical set of PFI contracts, the single contractors and the other parties that are involved will agree on the financiers' or the banks' step-in rights and the circumstances under which they will apply. Those rights determine the relationships between the banker, the contractors and the users, who are the public sector bodies. It is important, nonetheless, to ensure that the bankers do not jump in with their size 10 boots the moment that one minor problem appears in the contract. That is what that set of PFI agreements is about.

The circumstances in which the public sector can expect bankers to secure their cash-flow expectations under the project agreement are set out in the contracts. That is the level of protection that the public sector has at present.

Every successor banker to a project will be subject to the same terms and conditions. Ultimately, bankers suffer a high level of—

The point that I was trying to make was not that the contract might change, but that the partners in that contract might change, and that the risk that is involved for the public sector client may well differ, depending on who the contractor or financier is.

If I have an arrangement with a major building society, which sells on my mortgage to another lender who then changes the terms of that mortgage, what rights do I have? What happens if, all of a sudden, the Nationwide Building Society no longer wants my mortgage and sells it on to a junk-bond company, which increases my interest rate according to its assessment of the market rate?

In general—

I was just trying to highlight the circumstances in which there may be a public sector risk that you are not addressing.

I said "in general" but, in fact, the rule that I expect would be followed under all circumstances would be that the contract would specify that no changes to the financing elements of the contracts that might have an adverse effect on the public sector may be made without the public sector's consent. Bankers cannot unilaterally raise the interest rate—I guess that they could unilaterally drop the interest rate—without the public sector's consent. If, in all circumstances, the financing agreements that underlie a PFI were to change in a way that might potentially—not actually—have an adverse effect on the public sector, there would be a right of consent on the part of the public sector. Otherwise, the change would be void.

The transfer of risk, which distinguishes PFI projects from every other type of financing that has gone before, seems to be the most important thing in the contracts that we are discussing. The Treasury talks about optimal risk allocation, and we have heard much evidence from witnesses about different risks, some of which are easily quantifiable, others less so. What measurements can we make to determine whether risk has been allocated optimally?

One has to analyse the nature of the particular risk, then ask oneself which of the many parties that are involved in the contractual arrangement concerned would be best placed to handle that risk and which could minimise the risk through its own actions. One then tries to allocate the risk to the party that is in the best position to minimise it. That is how we reduce the risk to society as a whole.

How do we ascertain whether that has been achieved? I do not have a simple answer to that question. Historically, we could examine the extent to which particular risks had arisen in contracts and what they had cost. If we were to consider another group of PFI projects, we could find out whether similar risks had arisen and whether they had cost less as a result of the projects' having been carried out differently. However, I do not know how we could do that respectively.

We have received a lot of evidence about IT, which can be a difficult area in both public and private procurement—there is no difference. Because of the rapidity of change, IT is a problematic area, in which risks are substantial and difficult to quantify.

At the other end of the scale, risks may arise in relation to the maintenance of a building. Those risks are not at all difficult to quantify, so why bother transferring work in that area? How do we measure the optimal risk in the middle?

The risks that are attached to the maintenance of a building are harder to quantify than Dr Simpson suggests they are. They depend on how the building in question has been constructed. If cheap materials have been used, as was the case in the 1960s, we end up with the risk of high maintenance costs in 2000; if better materials had been used in the 1960s, we would now have lower maintenance costs. Builders were aware of those risks, but I am not sure that the clients were aware of them. I am certain that the UGC either did not know or did not want to know about them.

An arrangement under which the person who designs the building, decides how it will be constructed and takes on board such a risk will, in the end, lead to a lower total project cost, compared with an arrangement in which the risks are left with different parties and are not entirely transparent.

I fully agree with the empirical analysis that one has to consider at the outset who is in the best position to control the risk. Now, given the history of PFI projects, that analysis is being tempered with experience. There will always be difficult areas. If we consider, for example, a facility for the care of mental health patients, the private sector might be concerned about taking on the risk of damage to the facility, although that risk will be substantially influenced by the quality of the company's design and the materials that it uses. Usually, there will be a debate. In some areas, a sharing mechanism will ensure that both parties are incentivised to work together to mitigate the risk. On other questions, such as how long it will take to construct a building and what the cost of that will be, a great deal of the risk is, self-evidently, best controlled by the private sector.

PFI, which started with that empirical analysis, has now been tempered with experience. As the private sector grows in experience, it may be appropriate for the sector to bear additional risks.

Generally, the technique involves analysing the risks that pertain to a project and the likelihood of those risks occurring. Once the risks have been costed and the chances, or betting odds, of them happening have been applied, it is a matter of arriving at a set of expected values for those risks occurring. The extent to which the project divides the various risks between the public sector user and the private sector operator of the facilities must then be determined. Having taken a view on that, one would then find out whether the expected cost arising from those risks in the public sector comparator was higher or lower. One would then take a view on whether the risk had been transferred appropriately.

The techniques for estimating risk factors in the public sector have been, empirically, relatively crude, to the extent that there are no sophisticated simulation models to apply risk to the various factors, because of the lack of previous history. It will be some time before any such models reach an appropriate level of sophistication.

The technique that most public sector procurers adopt under such circumstances is to consult their advisers on the specific risks, whether they be technological obsolescence or design risks in the building, and to estimate the impact of those risks. Although that is an inexact science, the process is gone through and is rigorously tested in the review process.

PFI projects have now been going on for about four or five years. As we move forward with PFI, do we require a simulation unit? Should the centre have some sort of simulation process under which the matters that we have been discussing may be reviewed in order to improve our ability to predict?

As a bank of experience is built up, it will become possible to conduct that exercise. I do not profess to have the necessary expertise, and I am not sure whether the expertise to do that exists centrally. However, I am convinced that the health department is capable of compiling the requisite information from the projects that have been completed, in order to develop and circulate among the public sector the very models that we need to evaluate the various projects.

When appraising a PFI and deciding whether to proceed with it, you would obviously take several factors into account. Among those factors, how sensitive is the net present value to assumptions about the cost of capital? In your experience, what are the most sensitive factors?

The net present value of the project is sensitive to the cost of finance, but one needs to sub-analyse that, because a number of factors underline that cost.

The greatest single influence is the market cost of money. For projects that are being financed through bonds, it is the yield on Government gilts and for projects that are being financed through bank debt, it is generally the long-term price of LIBOR—the London interbank offered rate—that is the measurement of the cost of money in the bank financing market. That is the biggest influence on the cost of finance.

The margin that banks charge on top of LIBOR or that bond issuers charge on top of the gilt price is a smaller influence. One is looking at most bank-funded projects being funded for a margin of less than 1 per cent or, in bankers' terminology, 100 basis points. The very cheapest projects may be 10, 20 or, in some circumstances, 30 basis points below that, but the money markets can move by that amount over a relatively short period. That is a comparatively small influence.

The other factor in the cost of financing is how the financing is structured within the special purpose company. That is determined principally by the detailed terms and conditions from the banks about how much surplus they need in the income flows, over and above the amounts that are required to service debt and the amounts that they require the operator to set aside as a reserve for future maintenance costs. Those terms and conditions have every bit as much impact on the net present value to the public sector as the difference between one bank's margin charged on funds and that of another. There are a number of factors in play—the focus tends to be on the banker's margin.

Would it be fair to say that, because of the way that these things can fluctuate over a period, this is more of an art form than a science?

It is not an art form. As financial advisers, we would distinguish one banker's terms from those of another. One can quantify the impact of banks that would have onerous requirements on a project and select the one that offers not just the lowest margin but the prospect of the lowest net present value to the public sector. Anyone who was able to understand in advance the way in which the financial markets move would be very fortunate and prosperous.

I deserved that.

We have dealt with most of the areas that we wanted to cover, so I will bring the item to a close. I thank the witnesses for coming along this morning and for their full and helpful answers.