Finance Committee, 20 Nov 2001

Meeting date: Tuesday, November 20, 2001

Official Report 223KB pdf

---

Private Finance Initiative/Public-Private Partnerships Inquiry

I formally welcome all the witnesses who will give evidence today.

As ever at the start of these meetings, I remind everyone that all mobile phones must be switched off and any pagers switched to buzz. The committee has one absentee, Donald Gorrie, who has a clash this morning with the Procedures Committee. I hope that he can join us later. For the benefit of the people in the gallery and our witnesses, I should point out that we are joined by Professor Peter Jackson of Leicester University, who is our adviser to the inquiry.

We have received written submissions from all the organisations represented this morning. I offer the witnesses the opportunity to make a brief opening statement. It would be helpful if you could restrict your remarks to a couple of minutes; we will raise other issues in the questioning that will follow. I will call the witnesses in the order in which their names appear on my script. First on my list is Andy Wynne, who is the head of public sector technical issues for the Association of Chartered Certified Accountants.

I would like just to pull out a few points from our submission. We must recognise that there has been a lot of controversy since the inception of the private finance initiative. For example, there was sustained local opposition to the Skye bridge project. One of the most significant factors locally is that, as a recent survey showed, only one in 10 senior public sector financial managers expressed strong support for the statement:

"PFI and other forms of public-private partnerships are having a beneficial effect on public services".

However, the question whether PPP or PFI will ensure value for money is still largely a political rather than a professional opinion. For example, in a recent report, the Audit Commission for England and Wales stated that it was

"too early to say whether PFI contracts generally offer the public sector long-term value for money".

However, as the UK Government is committed to PFI, there is a question about how much flexibility the Scottish Parliament has in determining whether public sector investment should happen through PFI or other public-private partnerships or through traditional direct investment.

The Scottish Parliament might consider it worth while to make a case for greater freedom to fund direct capital investment. The Commission on Public-Private Partnerships, which reported in July 2001, said that there should be

"an evidence-based approach to policy. A commitment is necessary to pilot, monitor, and systematically evaluate a spectrum of partnership arrangements. Depending on the evidence that emerges PPPs could be rolled out or rolled back".

I understand that the National Assembly for Wales has said that it will now use direct investment instead of PFI, at least for health service projects.

I will highlight a few points about the management of PFI projects that we raised in our submission. First, it is very important to carry out a sensitivity analysis of the PFI option appraisals by considering the various factors that could affect whether PFI would provide value for money. For example, we should find out whether certain risks are transferred to the private sector partner and perhaps what rate of interest is used.

Secondly, we should develop and maintain public sector expertise. At the moment, there are relatively few private sector players in the market and they are clearly experts at these kinds of arrangements. However, on the other hand, there is quite a fragmented public sector. In such a situation, it is important for the Scottish Parliament to ensure that the knowledge and expertise that is gained through the various PFI projects is maintained within the public sector. Perhaps there should be a central resource of management consultants and advisers who are able to work with the managers of the individual public sector entities using the PFI route. For example, the National Audit Office recently stated:

"There is no substitute for the knowledge that is acquired while developing and negotiating a PFI deal. The public sector needs to retain staff with experience of PFI deals and to use the experience on future deals."

Finally, I want to quote David Hinchliffe, the chairman of the House of Commons Select Committee on Health. In a foreword to a recent report on PFI and PPPs, he said that

"talk of value for money, risk transfer and contract transparency can be so much hot air in the real world where fragmented and under-resourced public agencies are forced by stringent fiscal constraints into long-term dependency on powerful private sector providers".

Thank you. The next witness is Philip Grant, who is head of infrastructure finance at the Bank of Scotland.

I will keep my comments brief and focus on what we fund and why we fund it. I will also make a few comments on the issue of value for money in PFI/PPP and on the cost of private capital, which is a feature of the on-going debate.

By way of update and context, I should point out that the recent merger of the Bank of Scotland and Halifax plc has led to the formation of HBOS plc, which is now the fifth largest banking group in the UK. The merger group has its headquarters yards away from here on the Mound and employs about 60,000 people worldwide, more than 14,000 of whom are in Scotland. We are strongly committed to Scotland and its economy, infrastructure and communities.

Bank of Scotland corporate banking is responsible for the group's merged infrastructure funding activities and I have a team of 43 professionals based in Edinburgh, London and overseas developing PFI/PPP. Since 1992, we have been at the forefront of the development of PFI in the United Kingdom, and in Scotland in particular. We have participated in the funding of 84 PFI projects—more than any other financial institution. Across the UK, we have more than £2 billion committed to PFI special purpose companies—SPCs. More than £1.2 million of that is drawn and invested.

In addition to being lenders in a traditional banking sense, we are shareholders in about 25 special purpose companies. We are active investors and I and selected experienced colleagues serve as directors for special purpose companies in which the Bank of Scotland is a shareholder.

In Scotland, we have funded 19 projects in a range of sectors including hospitals, schools, prisons, waste projects and roads. I shall focus on education as an example: in Glasgow, we underwrote the £310 million project in support of the new build and refurbishment of 29 schools. That represented all of the city's secondary stock, plus one primary and one nursery school. The project also included the provision of information technology services and facilities to the city's 30,000 secondary pupils and the maintenance of all the schools and facilities for 30 years.

Only last week, with the same partners—Amey plc and the Miller Group—we completed the funding of 10 new primary and two special schools, as well as a secure unit, community centre and the refurbishment of four more schools for City of Edinburgh Council. That project will provide a high-quality learning environment for 11,000 pupils and staff across the city.

The Bank of Scotland was also the lead funder for the Falkirk schools project. That was the first schools package in the UK, with five new schools providing facilities for 5,000 pupils, including a new special needs school at Larbert.

In March this year, we completed the funding of the new build and extension of three schools in Banff and Oldmeldrum in Aberdeenshire. Those schools will provide modern facilities for almost 2,000 pupils, including special needs facilities at Banff Academy.

All that adds up to new facilities for 50,000 Scottish pupils and their successors. All those projects were developed and delivered in close partnership with the local authorities concerned. We believe that those partnerships have contributed to the enhancement of the educational environment for teachers and pupils alike. The SPCs have binding contractual commitments to maintain the quality of the facilities over a 30-year period for a fixed cost to the public sector.

The quality of the facilities will be maintained for generations of pupils to come. At the end of the project terms, the assets will revert to the public sector.

I refer to the health sector. The newly opened 630-bed Wishaw general hospital was developed as a partnership between Sir Robert McAlpine and the Bank of Scotland. If members have time to visit, they will find a 21st century environment for the delivery of health care. That hospital replaced the hospital at Law, which had been constructed as a temporary measure during the second world war. The SPC is committed to maintaining the availability and quality of those facilities to the highest standard and thus assist Lanarkshire Acute Hospitals NHS Trust to continue delivering the highest standards of care to the people of Lanarkshire.

I will stop you there because we need only brief introductory statements. I am sure that you will be able to bring out your other points during questioning.

I invite whichever one of the representatives from the Chartered Institute of Public Finance and Accountancy is going to speak to do so now.

Thank you. I am the technical director at CIPFA. I am joined by my colleagues, David Dorward from Dundee City Council and Lynn Brown, head of corporate finance at City of Edinburgh Council. Both my colleagues are practitioners and have first-hand experience of working with PFI schemes.

Briefly, CIPFA made its position clear in its publication on PFI and PPP. I will leave that document with the clerk. We see it as one of a range of financing and procurement methods available to public services. We do not therefore necessarily see it as good or bad, but we do believe that PFI has acquired a poor reputation with some finance directors. That is partly because when it was introduced, there was a duality of policy objectives. Secondly, it is often presented as the only alternative to traditional public finance.

When PFI was introduced, the duality was partly connected with the need to reduce the public sector borrowing requirement and, at the same time, to achieve value for money in the delivery of public services. In the early days, there was much pressure to achieve the first objective and reduce the amount of public sector borrowing as a proportion of gross domestic product. There was some pressure to get schemes through, and those schemes did not always represent value for money.

The second issue is perhaps more relevant today. In many parts of the public services, PFIs are often presented as the only option. For us, the main issue is the control frameworks within which public services are financed. In the local government sector, for example, for many years the capital control system has effectively put a cap on the amount of public sector investment that local authorities can pursue from traditional methods of finance. Currently CIPFA is working on a UK-wide basis on a new prudential approach to capital investment in local authorities that would remove the complex rules that currently exist, put in place a system that would allow local authorities to borrow on the strength of their revenues and financial position, and give authorities more freedom, within certain guidelines and with certain safeguards. Such an approach would enable authorities to use more traditional sources of financing and allow for a real option appraisal, rather than just a choice between two alternatives.

We are aware that there is considerable interest in different vehicles for providing public services. We welcome that, as we do not believe that anyone profits from having only one way of doing anything. We would like a more positive view to be taken of public-public partnerships that allow different parts of the public services to get together, on both a capital and a revenue basis, to provide services. Throughout the UK there is a new emphasis on cross-cutting services. In that scenario, public services must have the ability to work together in a way that they cannot at the moment. Whatever vehicles are selected for the delivery of particular public services, attention should be paid to issues such as governance, accountability and the danger of creating a democratic deficit.

I ask Bill Davidson of NorthLink Orkney and Shetland Ferries to address the committee. Am I right in thinking that you will speak in a personal capacity, based on your experience over a number of years?

Yes.

Good morning, convener. My name is Bill Davidson and I worked for 12 years in the health service, for 10 years as a management consultant, largely working with the public sector, and for four years in corporate finance, where I was heavily involved in the delivery of PFI and PPP projects to many parts of the public sector. I was then involved in the creation of one of the very first wider markets initiative companies, which are the next thing on from PPP. I reached a financial close on that deal just before Christmas last year. In June this year, I was appointed chief executive of NorthLink Orkney and Shetland Ferries. I am here as an experienced practitioner, but not as a representative of KPMG, which was my employer at the time that I wrote my submission.

As someone who has worked in the public sector and has been involved in trying to deliver projects, I have experience of both traditional public sector finance-based procurement and PFI/PPP-based approaches. I have been actively involved in PFI/PPP projects, and although those have been painful along the way, in every case the end result has been a far better product for the public sector consumers of the service. As a taxpayer, I believe that I am getting far better value for money from PFI/PPP projects than from the many public sector-funded projects that I have seen in the past. I would be happy to expand on that and to answer any questions that members have.

I thank all our witnesses. We have a range of questions to put to you.

As has been stated, many of the arguments relating to PFI/PPP and more traditional methods of finance are political. What do you see as the major improvements that are offered by taking the PFI/PPP route? If there were unlimited funds or easier access to funding for capital projects in the public sector and there were real choices, would significant benefits still be offered by taking the PFI/PPP route instead of using public sector funding?

I could list benefits of the PFI/PPP approach from my experience, but I would like to highlight two in particular. First, the whole-life cost of a facility is considered—the provision of an asset and a service—and there is no concentration on the initial capital cost. I can take members to a hospital in Scotland that was built under the old lowest-capital-cost regime. It has cost an absolute fortune to run and maintain from the day it was opened. There have been maintenance, heating and lighting costs, but the layout of the wards is such that staffing levels must be much higher. The head of one patient and the feet of two others can be seen from the nurses' station. Consideration of lowest capital cost meant a lowest-capital-cost product, which inevitably costs a fortune to run.

The new Edinburgh royal infirmary provides an example in which the capital cost was increased to provide additional facilities to allow the hospital to be run much more cheaply over the length of the concession period. Consideration of the whole-life cost of a facility means taking into account the full arithmetic relating to a hospital.

Are you referring to Ninewells hospital in Dundee?

No.

I was just checking.

Why is one approach related to the public sector and the other related to the private sector?

Traditionally there has been an emphasis on lowest capital cost in the public sector rather than on the whole-life cost of a facility. That was a failing in the way in which projects were carried out in the public sector.

Could not we simply insist that if the public sector itself were to procure, it should consider the whole-life cost, too? Why should a project go to the private sector before that approach is taken?

I agree that if the public sector took a PFI/PPP approach to projects and considered whole-life costs, there would be considerably better value for money over the life of those projects.

The second benefit that I wish to highlight is that one of the disciplines that PFI/PPP projects bring is the necessity to nail down all the details before the first brick is laid. In my experience, political and other influences were inevitably brought to bear to get on with particular projects that had public sector financing. With PFI/PPP, I have come across occasions where pressure has been brought to bear, but if a project was not right from a commercial point of view, people would not proceed with it until everything was sorted out. The time that is spent up front sorting out all the details before going to contract, therefore, pays handsome dividends in respect of the project being delivered to budget and—frequently—ahead of time.

I will not accuse you of being a late convert, but when you were involved at the front line in the public sector, did lifetime costs feature in any of the projects with which you were involved? Did they feature in preparatory discussions with clients, or were they excluded from the discussions?

We never considered the full lifetime costs at the NHS. Capital costs were considered. Frequently, unusual decisions were made so as to spend capital budgets by the end of the financial year. That was a result of how the NHS worked at the time.

When I was a management consultant and we bought large projects—particularly IT-type systems—with public sector finance, we considered through-life costs, but we frequently did so in respect of maintenance contracts and things of that ilk. In around 1992 or 1993, we introduced the concept of trying to assess the financial benefits of introducing new technology—for example, would six people be able to do the work of eight? That was introduced as a cost-benefit analysis and there was a lot of scepticism about the deliverability of the benefits in the public sector.

Brian Adam wanted to say something, but I noticed that Lynn Brown offered to comment on the same question. I invite her to do so. Brian Adam will then have an opportunity to say something.

I am speaking from a City of Edinburgh Council perspective about what happened in our recent PPP. At the outset, there must be a public sector comparator that considers the whole life-cycle costs if unlimited money were available from the public sector. That is compared with the PFI model. The difference is risk transfer. Risk is moved to where it can be best managed. The bottom line is that if the public sector comparator provides better value for money than the modelling on PFI, one should not go ahead with PFI. There should be incredible detail so that anyone can see the whole outline and final business case. That should be considered at the outset.

Mr Davidson suggested that one of the advantages of PFI/PPP is nailing down the whole project in advance, but given the length of these contracts, which tends to be 30 years, the services that are to be delivered will change. How do you allow for the flexibility that will be needed in the system to accommodate those changes? Some of the changes that will happen in the public sector procurement process will be as a consequence of changes in the nature of the service delivery.

Secondly, with regard to the lifetime costs maintenance cycle, what incentive is there for the operator in year 27 of a 30-year contract to continue to maintain the asset when that asset will not be in the operator's ownership?

You are right that there will be changes. The health service is a classic example of that. In the 1960s, lots of people had to go into hospital for major ulcer operations, whereas nowadays that is done almost on a day-case basis. The incidence of disease and the way in which we treat it change considerably, and the types of facilities that we need change considerably, so the buildings must be designed to be flexible.

I remember working on one project where the architects were talking about the wards having rubber walls. They meant that the walls could be moved so that the configuration of wards could be changed relatively easily. The contracts always have provisions to allow for such changes to be made and for the way that the works that have to be done are costed. On the services side—the provision of catering, cleaning and so on—provision is made for altering the quantity of services that are provided to reflect changes in the business. All the details are provided at the outset. The contracts contain detailed mechanisms for altering such things.

Your second point was on the incentive for the operator to maintain the assets in good condition. There are strict provisions in the contracts for that. Effectively, the asset will be returned at the end of the period, in most cases almost as good as new. That leads to situations, such as with roads contracts, where the contractor has to resurface the road just before he finishes maintaining it. He is contractually obliged to do lots of things, and he has to do them according to certain requirements, so he cannot wear out the asset towards the end of the contract and walk away from it. The financing provisions that are in place are strict, to ensure that he fulfils his contractual obligation.

Can you give us some examples of where that has happened, or are you just saying what should happen in future? Is it correct that none of those contracts has got to the end yet?

That is the problem. There are no examples to point to where we can say, "That was a 15-year deal and they carried out refurbishment just before they left." Financing, for example, for road resurfacing is built into the financial plan. The money is there to be drawn down at the appropriate period, to fund the costs of the works that are to be done.

I will address that final point. With the projects that we undertake, it is normal for the public sector to establish a standard of facilities at the point of handover. That normally includes quite a large amount of what would be termed life-cycle spending towards the back end, particularly from about year 25 to the end. The funders, for example banks or fund providers, ensure that during the project there is sufficient reserving of cash so that at the point when work is required, cash is available to be spent on the facility, because it is important to them as lenders and shareholders for the contractual obligations on standards to be met.

The situation is monitored continually for the funders by technical advisers, who report to the banks and shareholders on the physical condition of the facilities and how they compare with what is defined in the contract. So there is a mechanism to ensure that the terms of the contracts are met. You are right that time will tell, but there are parties to those projects—banks, shareholders and so on—who have a financial interest in the contractual obligations being met.

PFI/PPP has not been running for an enormous length of time and, as you say, none of the contracts has come to an end. In your opinion, has there been any significant change, since PFI started off, in how organisations set up arrangements and put them into place? You mentioned Edinburgh and Glasgow schools refurbishment.

PFI/PPP covers many different things, such as roads and the provision of information technology services. Are there some kinds of project that it is more appropriate for and others that it is less appropriate for?

I think Vernon Soare mentioned a change that we have seen recently. One of the original arguments for PFI was that it would allow investment in public services without its being considered part of the public sector borrowing requirement. There has been a significant change away from that. With most PFI projects that are being signed off now, my impression is that the investment is considered part of the PSBR in the sense that PFI is being recognised as a way of financing the investment. In essence, PFI is just another way of borrowing money.

Since June, most PFI projects in the NHS are on balance sheet, so the asset is considered to be part of the trust's assets and, along with that, is part of the PSBR. There never was a problem with lack of public finance. Since 1945, the public debt—as a percentage of GDP—has fallen from about 45 per cent to about 30 per cent. The Treasury stated that a prudent level would be about 40 per cent. There is substantial leeway for direct public investment in infrastructure projects.

I wonder whether you could identify what sorts of areas are more appropriate for PFI/PPP. In the first session of our inquiry, one of the witnesses said that he believes that PFI works well with projects such as prisons and roads, but less well with hospital and school projects. I am aware that that contradicts some of today's evidence, particularly Bill Davidson's. Is there are a general view? It is clearly not a case of one size fits all.

I would like to amplify that. Across the piece, are we learning which risks it is more appropriate to transfer? Rather than asking about schools, hospitals or roads, we should ask whether we are learning for which things the private sector manages the risk better and for which things the public sector manages it better—or is learning to manage it better.

I am not sure that that is the question that I am going to answer.

If you answer the first one, you can pass on the other one.

I will deal with whether we are learning about PFI projects. In Dundee, we had a PFI on a waste energy plant back in 1993-94. The contractual basis of that and the negotiations that we went through were complex. In our current work for education—on schools—I have noticed a significant improvement in the pre-tender stage and in the structure of the contracts and the negotiations. In that respect, the Scottish Executive has got its house in order by promoting PFI, which it was not doing five or six years ago.

Andy Wynne brought up the question whether the public sector has sufficient resources to invest. I feel that it is unfair to make a comparison with the public sector's performance on the traditional procurement method because comparison would be based on a situation in which the sector had limited funds to invest in infrastructure. In the early 1990s, the public sector experienced significant reductions in capital allocations—particularly to local authorities—to invest in infrastructure.

We are comparing that position with the present one, when we can go out and enter into deals on PFI projects that are worth £60 million or £90 million. The two are not strictly comparable. If a local authority were able to go out and borrow to the same level and degree, I am convinced that it would consider life-cycle costs. There would then be a realistic comparison with the PFI model.

I will take up Lynn Brown's point: the public sector comparator is essential in ensuring that if a PFI project is to proceed it gives the public authority value for money.

The public sector comparator is an important issue. Is one always available? We discussed the matter at a previous meeting and that was identified as the problem. Can you construct a reliable model?

One of the issues with the public sector comparator is that you are often looking for a new method of service delivery that involves capital and services being wrapped up together and provided through a partnership. It can be difficult to find a public sector comparator that allows a like-with-like comparison, because PFI brings a new slant to delivering a service. Therefore, you may not always be able to find an exact public sector comparator.

I can speak only for the City of Edinburgh Council. We would not go ahead without the public sector comparator, because it is crucial to prove to the Scottish Executive and to our auditors that we are achieving value for money. I cannot speak about elsewhere, but the assumption is that the money is available. You then consider the transfer of risk and who would best manage it, to produce the bottom line figure.

Have you always been able to put together a public sector comparator?

Yes. You have to do it before the outline business case, before you go to the market and say that you are interested in going into a PFI project. The bidders and those who want to enter into those projects with us are also looking for that. The public sector comparator must be in place for the outline business case. It is updated for the final business case, when the deal is signed.

When Andy Wynne speaks, I ask him to clarify a point that he made earlier. You talked about some projects being on balance sheet. Will you explain what you meant by that? I thought that they would necessarily be off balance sheet. Perhaps I picked you up incorrectly.

There are two closely related issues. One is whether the asset will be considered an asset of the public sector entity: will the hospital building be considered to belong to the trust and appear on its balance sheet. Connected to that is whether it is considered to be part of public sector borrowing and within public sector borrowing limits.

In the past, it was said that if from an accounting point of view the deal can be constructed so that the asset is not on the hospital's balance sheet, the financing of the project is not considered to be part of the public sector borrowing requirement, so the Government can achieve the twin objectives of substantial public sector investment and a low level of public sector borrowing. Is that clear?

I thought that you said earlier that, since June, most NHS projects have been on balance sheet. Does that mean that they are not PFI projects?

No. Before then, a condition of a project's being considered a PFI project was that it was off the balance sheet and therefore not part of public sector borrowing. Since then, Alan Milburn has said that that condition is not necessary; a project can still be considered a PFI project if the building is on the balance sheet of the trust, for example. The capital investment will now be included in the public sector borrowing limit.

Does that mean that, in these cases, the asset—the hospital or whatever—belongs to the health board once it is built and that what remains is a maintenance contract for the lifetime of the health board's asset? I cannot see how it can be on the balance sheet if it does not belong to the health board.

There is a difference between the accounting treatment of the asset and the legal position. It is possible for the asset legally to belong to the private sector consortium and still be on the balance sheet of the public sector organisation.

Let me describe the experience in Scottish local government. If a PFI scheme was put on the balance sheet, it would score against the local authority's capital consent, which would defeat the object. No authority could afford to initiate a PFI scheme on the balance sheet. Therefore, one of the priorities is to get the external auditor for each public sector body to certify that a scheme is off the balance sheet.

That was our understanding, which is why we were surprised by what Mr Wynne said. As this is a crucial point, I shall allow a couple more questions on it. I ask David Davidson and Brian Adam to ask their questions briefly.

I understood that what a council would be looking to put on the balance sheet was the fact that a facility was available. In other words, a facility could be treated like a long-term lease and there would not be a capital transaction on the balance sheet. That is what has caused the confusion. The fact of the council having a legal right to use the facility has a value, but it is not a cost to the service—it is a notional situation. I presume that, under resource accounting and budgeting, if there is a charge at all it will be an annual charge based on the usability of the facility rather than the capital transaction costs. Is that correct?

There are two, different, situations. In PFI projects, the asset may or may not be on the balance sheet. For most PFI projects involving the health service, prisons and roads, the assets are on the balance sheet of the public sector entity and are part of the public sector borrowing requirement. As David Dorward said, the exceptions involve local authorities—perhaps in the building of schools. In those situations, local authorities do not have a choice. This is important and it goes back to the public sector comparator. The choice is not between using public finance and using a PFI or PPP; the choice is between investing in schools and other infrastructure assets or not investing. The public sector comparator is not a level-playing-field option appraisal between direct investment and a PPP; it is an exercise that is necessary to show the Treasury that the PFI option will provide value for money and will be affordable.

We are in a vicious circle—at least, we have been. For the investment to go ahead, the public sector entity has to demonstrate that the PFI option provides value for money. Many practitioners do not believe that that is an objective assessment—they believe that it is one way—as they have to try to prove that the PFI option is cheaper. The result is that public sector entities have to present evidence to the Treasury over and over again to show that their PFI will provide value for money, which reinforces the Treasury's view that, indeed, PFI provides value for money.

It is important to have the option of direct public sector investment. Such investment would mean that a proper, objective option appraisal would be undertaken and that real evidence would be available before the event to show whether a PFI would provide value for money. An exercise should also be undertaken afterwards to show whether the PFI option actually provided value for money.

The discretion that local authorities and other public bodies have over maintenance budgets has meant that choices have had to be made between maintenance and services. Although there is considerable wisdom in building in lifetime maintenance costs, there may be a significant adverse effect on other services that are provided by such bodies because they will no longer have that discretion. When will that start to bite?

I am not sure whether that point was aimed at a particular witness or whether it was just a statement.

No one wants to answer that question.

Before we move on to questions about best value—

Convener, Lynn Brown wants to give an answer.

One of the key things is affordability. What the council can afford needs to be considered. For example, City of Edinburgh Council can afford its current schools PPP. The council is not losing the discretion to spend on other areas. The PPP actually helps. Because 16 per cent of Edinburgh's schools will be new, money will be released that can be invested in the maintenance that is required in the schools that are not new. The PPP allows a switch of resources.

I worked in the public sector for 30 years and it seemed to me that maintenance costs were cut every time there was a financial problem. We are now seeing the eventual costs of that: a totally disastrous infrastructure. Often, the infrastructure requires total replacement. I experienced that in a number of sectors.

Building in maintenance costs as an absolute and as part of a contractual obligation that cannot be got out of will mean that the quality of the service and the maintenance of the building will be excellent. That will mean that the people who work in the building are working in a good environment. The failure to do that over the years has been disastrous for staff morale. It is almost impossible to say how such things should be quantified in financial terms, but such imponderables are crucial to these schemes.

We must restrict ourselves to questions rather than statements, otherwise we will not get through all our questions.

Mr Grant, you mentioned that you have been involved in projects with the Bank of Scotland over more or less the whole spectrum of PFI/PPP projects. Where is PFI most appropriate? Where is it inappropriate—although I appreciate that you may not want to say that. You state in your submission that PFI/PPP

"is now central to the development of essential Infrastructure and the provision of high quality public services in Scotland, the UK and beyond."

In Scotland, PFI accounts for only 10 per cent of annual capital spend. Given the fact that the proportion is relatively small—although it may grow—why do you describe it as central?

The proportion is small. When people are exposed to the statistics about the level of public sector procurement that is still carried on within a set framework, they are surprised that the public sector has not completely committed all its procurement options to PFI.

It often appears as if that is the case.

In our experience, the extent to which PFI works well is largely a function of how much experience people have of dealing with PFI. Earlier, you asked where the improvements have been. The improvements have been in projects that are based on accommodation: hospitals, schools, prisons, court buildings, police stations and so on.

Over the past five to eight years, both the public sector and the private sector have built up a large amount of experience of dealing with such projects. On both sides, there is an understanding of what the private sector is able to deliver and what the public sector's expectations are. It has been a learning process on both sides. That means that an Edinburgh schools project closes a lot more quickly and in a much more standardised contractual framework than did a Glasgow schools project. It means that time is saved and that the project is efficient in terms of procurement costs. There is a learning process. PFI is most appropriate where the service output expectation is clear and understandable and where the risks involved in delivering that output are manageable by the private sector.

Banks and other funders have generally found IT projects difficult because the risk that is involved in service delivery—which normally involves technical research and development of an IT product—cannot easily be passed to the private sector. The risk is not easily passable because of the way in which the requirements for IT projects need to be defined and delivered to expectations. That is true not only of public sector procurement, but of private sector procurement. Organisations, including banks, have cost overruns and experience problems when procuring IT. The public sector should not be accused of being deficient in its procurement of IT.

As more projects are undertaken in a sector, there is learning on both sides, which makes PFI able to deliver much more efficiently.

The convener mentioned that only about 10 per cent of capital investment comes from PFI. People may not be surprised that so much public procurement is still financed by traditional methods because the cost of capital that comes from borrowing with a Government guarantee is much lower than the cost of capital from the private sector.

The counter argument to that is that the private sector will make efficiency savings that will outweigh the difference in the cost of capital, but I return to the point that my colleague David Dorward made: to some extent, a false comparison is being made. Over recent decades there has been a lack of public sector investment, coupled with a squeeze on revenues. One could argue quite convincingly that the public sector has not had the opportunity to demonstrate what good publicly financed procurement can achieve with reduced capital costs and a level playing field. Perhaps more leeway needs to be given to the public sector—through public-public partnerships, for example—so that it can show what can be done through more traditional means of procurement.

I am not saying that PFI should not be used: I am saying that we should not jump to conclusions based on what has happened in the past and that we should take lessons from what could be delivered over the next few years.

In an earlier answer, you mentioned different methods of delivering public services. Can you expand on that? What else did you have in mind?

We are keen that the capital control framework in local government should be freed up, so that local authorities are not tied in either to providing a service themselves or to opting for a PFI project.

There is considerable interest in different forms of joint venture. The best-value initiative has created a great impetus for joint working of all descriptions. Some of that is taking place through local authority companies. In other ventures, the public, private and charitable sectors are coming together to deliver projects. We welcome that. We feel that attention needs to be paid to issues of governance and accountability. I do not want to digress further on the balance-sheet treatment of assets, but there are some accounting treatments that need to be sorted out when different parts of the economy are brought together. It is important that we get those right, if only to ensure proper stewardship for taxpayers.

My colleagues have some experience of different types of vehicle. I will now hand over to Lynn Brown.

I would like to set out for the committee the alternatives to PFI that we have used in Edinburgh. We support CIPFA's view that PFI is one among a range of options.

In Scotland, unlike in England, it is possible to set up council companies that do not count against an authority's capital regime. In Edinburgh, they include Edinburgh Development and Investment. EDI is part owner of New Edinburgh Ltd, which built the Gyle centre and which, with the Miller Group, is extending the Gyle's office space. That development has provided 15,000 jobs in Edinburgh. City of Edinburgh Council also wholly owns the Edinburgh International Conference Centre and has developed the Exchange area, in partnership with local enterprise boards. Our Dynamic Earth and Edinburgh Festival Theatre were joint ventures by the council, again with local enterprise boards.

Transfers have been mentioned. We transferred all our sport and leisure facilities into Edinburgh Leisure, which is a not-for-profit organisation with charitable status. That secured 600 jobs and no closures. Finally, we merged the King's Theatre and Edinburgh Festival Theatre and we will enter new, leased office accommodation. The public sector has a range of partnerships available to it. We are fortunate in Edinburgh to have the assets to invest in land and so on. Scotland is definitely in a better position than England to develop things such as council companies.

Those are useful examples. Am I right in thinking that the council establishes not-for-profit companies?

No. New Edinburgh Ltd, which developed Edinburgh Park, is 50 per cent owned by the EDI Group. It has shared dividends—it is a dividend distribution company. We are developing the waterfront with Lothian and Edinburgh Enterprise Limited. We are a 50:50 shareholder and the assumption is that we will reinvest in the waterfront.

Do you see EDI and similar companies as just an alternative form of public sector investment?

We see that as an alternative form of partnership between the private and public sectors. It is an option for us for public investment in large services.

But it releases local authorities from the constraints that they are under in terms of having direct investment into new capital projects.

It does not count against our capital consent but it is a different regime. It works in the private sector with a board of directors and so on. It is wholly owned by the council—we are the only shareholder.

When it comes to deciding which partnership you are going to go into—you described a number of them—do you consider the end project and a means to deliver it, or do you consider the avenues and what you can deliver through those avenues? You have obviously developed a critical management style in your department.

We look at what the council wants to achieve with its investment, then we consider the best vehicle to do that. The bottom line is always what the council hopes to do with investing its assets.

So basically you consider the end product then the way to deliver it.

The best value for money for delivery, yes.

Your leisure services are a not-for-profit, charitable company. Did you compare that against the possibility of a private finance initiative for those services? Why was that model chosen? Was it a political decision or was there a financial basis for it?

We looked at that back in 1997/98. We had to ensure that we were working within the compulsory competitive tendering legislation. We considered different options and the one we went for was the one that the council felt was the best option in terms of what it wanted to achieve with its leisure services. That option was able to deliver enough investment for the council to reinvest and have no closures.

I am just wondering whether some of the other witnesses might say that that was not the most efficient or financially beneficial way of doing it.

Your inquiry is happening at an interesting time because a lot of work is going on in the public and private sectors to develop the PFI model that we are discussing today. From a local authority requesting that a director be appointed to the special purpose company to represent the council's interests, to a local authority wishing to have a partnership where the council makes a form of investment within the project to ensure that its interests are aligned with that project, there is an increasing appetite among local authorities in particular to become more closely involved in the delivery of their projects through a partnership model.

Organisations such as Partnerships UK, which is the Treasury task force, are undertaking a number of pilots to bring local authorities to the market, where that broad term "not for profit" and other partnership structures are being developed. Partnerships UK aims to align the local authority's interests with the private sector in a different way and introduce models that are fundable by the private sector.

We are fully engaged with organisations in the public sector in continuing to review new models and new ways of funding. A lot of research is being done on new partnership structures that will involve local authorities much more directly in their projects.

You have given us considerable food for thought, but we have to move on, as there are a number of areas that we want to cover.

I would like Lynn Brown to expand on the different models. How do you judge best value when choosing a model? How can we determine which of the options of PPP, PFI and the public sector route offers best value? Are there any projects whose outcomes have clearly demonstrated that better value has been obtained through the use of private finance than would have been the case through the use of public finance?

That is a difficult question to answer. With regard to the regeneration of the city—Edinburgh Park and the EICC were very much a part of that—the council did not have a shopping list of options from which it chose PFI. The council owned the land and wanted to control what happened in that area. The option that it chose of having a wholly owned company with expertise brought in from the private sector suited that purpose.

The council did not use a public sector comparator; the information came from Treasury task force guidance. Funders were reluctant to get involved in PFI and the Treasury task force guidance is excellent and helps local government to make progress by setting standards that local authorities can adhere to when they are negotiating with the private sector. The public sector comparator does not come into every option; it is part of PFI and allows such schemes to be judged. I do not know whether what I have said helps you at all.

There are two aspects to making the decision between direct public sector investment and PFI or PPP. On one side, there are the figures that the accountants have provided. One will be told that project A will cost one sum and project B will cost another sum. On the other side, there is the subjective element that relates to which project is thought to be most likely to produce better services. It is important to recognise that both sides are involved in the decision. It is easy to look at the figures and tell which project is cheaper, but the decision about the quality of the services must be a political one, as it does not involve the comparison of like with like.

For example, in the comparison between the refurbishment of a number of city centre hospitals and the development of a hospital on the outskirts of the city, difficult decisions have to be made. One of the considerations is how convenient it might be for people to get to the hospital, especially those who do not have cars. If a hospital is in the city centre, people will be able to reach it by taking a bus. However, if a hospital is on the outskirts, it might take two bus rides to get there. The cost of that is difficult to factor in. There would be a series of such decisions.

Another decision about hospitals is how many beds there should be. Clear evidence shows that, in PFI hospital schemes, the number of beds that are proposed in the original outline business case is significantly reduced in the deals that are struck. Decisions about how many beds are required could be based on advice from medical professionals, but that is not as clear-cut as financial evidence. One must try to weigh up those two forms of evidence, which means that it is difficult to achieve a balance in making such a decision.

That point is important. There is a suspicion that the pint pot is determined and then things are squeezed into it. If a PFI hospital has problems, bed numbers and staffing are reduced. That happened, for example, with the new Edinburgh royal infirmary, which is one of the models that the committee will consider. Because PFI hospitals seem to be more expensive, the result is a high-quality but smaller service. Is that valid?

I worked in the NHS and spent about 15 years doing a variety of projects for it, so I know it fairly well. You are correct that a typical PFI hospital has a reduced number of beds in comparison to the existing bed stock. One of the characteristics of the NHS acute sector is that its use of beds is not very efficient. The reason is that most hospitals are not oriented to the type of health care services that are delivered today. Research shows that, in an average acute hospital, a person who goes in as a cold case for an elective procedure can spend up to 80 per cent of their time waiting for something to happen. An awful lot of time and effort is wasted in current NHS hospitals.

The efficiency drives of the past were department focused rather than patient focused. Not long ago, the pharmacies in some hospitals closed at midday on a Friday to save money. That meant that no patients could be discharged after about 10 am on Friday because they could not get take-home prescriptions. The result was that patients stayed in the ward until Monday. I could give endless examples of that type, but I could take members to hospitals in which people have been able to provide more efficient services that are more patient friendly. Most patients do not like spending time in hospitals.

Most NHS hospitals have been around for a long time. Some hospitals in Scotland were designed in the 1970s, built in the 1980s and opened in the 1990s—they were classic two-phase constructions. I was involved in commissioning a hospital with 400 beds. While we were planning to open the new hospital, we suddenly twigged that it did not have a mortuary; the mortuary was to be completed in the second phase, but the architects had not been commissioned to design the second phase. Such practices from the old public sector procurement process still dog the health service. That hospital still has to make do with temporary arrangements.

In a lot of cases, the efficiency of the delivery of the service has been impacted on by past decisions. In a PFI environment, the contractor is incentivised to provide assets and facilities that give the best-quality service in the most efficient manner. The staff in such hospitals end up delivering patient care rather than making do with facilities.

The best PFI hospital schemes are those that engage with clinicians for some time at the design and bidding stages to ensure that the clinical aspects of the care model that is wanted are represented in the design. A partnership between trusts and the companies that provide the services is required to ensure that the care model and the service delivery model are compatible. A lot of work is put in at the trust, clinical and private sector levels to ensure that good schemes are compatible at the design stage.

I have a point about affordability and how the public sector comparator drives value for money. The comparator of affordability is not a soft number for the private sector to hit at. A lot of work goes into optimising the quality of the service and the delivery costs to meet that comparator. The other dynamic that has not been mentioned is that the private sector option competes not only with the comparator, but with other private sector options that are aimed at optimising service, cost and all the other related social issues that may be considerations in the public sector.

I will pick up on the two examples that Bill Davidson gave. One was about a pharmacy closing at noon and the other was about a hospital with no mortuary. I am puzzled as to why those situations would have been different under PFI. I presume that the example of the pharmacy has something to do with running an existing establishment—what you describe is just inefficiency in running the establishment. Were you suggesting that the contractor should run the whole hospital and therefore that contractors are more efficient?

On the second example, what are you saying would have been different about how the contract for the hospital was arrived at? If the people who designed the hospital did not realise that a mortuary was needed, why would the outcome have been any different if they had negotiated with a private contractor to build what they had designed?

I will respond to the second point first. I was using that situation as an example of a hospital that was built through the old public sector finance route. There were only enough funds to build the hospital in two phases. Funds were allocated for phase 1, which was built. Four hundred patients were to be transferred into that building and services were to be delivered there, but phase 2 had not even been designed, let alone built. The hospital was going to operate on two sites, several miles apart. No one had provided for a mortuary in phase 1. The scheme was originally a whole, but it was cut in half.

Are you saying that split schemes would not exist under PFI because enough money would be available for the whole scheme?

Yes. On the pharmacy, I was drifting towards the point that, when PFI began, there was a lot of discussion about how far down the service spectrum to take it. Consider all the services in a typical hospital. We start with maintenance, engineering and cleaning. We continue through the service spectrum to, for example, laboratories, the pharmacy, radiology and the provision of nursing services and then to the doctors and surgeons. There was much debate about how far through that spectrum the outsourcing of services to a contractor should go.

The Government of the time decided that anything with a clinical content would be retained within the national health service. That includes services such as pharmacies. If Boots, a typical pharmacy contractor, said that a customer had to hand in their prescription and come back four hours later because it took them that long to make up the prescription, it would not be in business for long, as it would not be providing the service that customers require. It is not unusual in NHS hospitals for the pharmacy services not to match the needs of the rest of the hospital and the patients because they are run to provide as low cost a service as possible. My argument—it may be slightly provocative—is that we should consider contracting such services out as part of the support that doctors and nurses need to deliver the required patient care.

That argument would apply equally well further up the spectrum. The logic would presumably be to contract out the nursing and clinical services as well, would it not? Go on: be provocative.

I accept that, if we want to push the argument to its limit, we will likely end up with what I describe as the Kwik-Fit approach to medicine. Thirty years ago, we took our cars to the local dealer for everything, but now we go to different specialist outlets for tyres, exhausts, brakes, windows, electrical work and body work. Those places specialise in small parts of the overall care to our cars, but we still have general dealers to whom we take the car for the complex parts. Most NHS hospitals are clogged up with people who do not need high-tech expertise to deal with the routine work. From the perspective of efficiency—I accept that there are many other factors—that is not a good way of delivering health care services.

Given your experience of PFI and PPP, what is your understanding of the costs and time associated with determining whether such a project is worth while? In particular, I would like you to apply your mind to the value of the current process for determining public sector comparators. For example, is the discount rate realistic or sensible?

I will give a swift response and then allow others to respond. A characteristic of a PFI or PPP project is the phenomenal amount of time, effort and money that goes in before the contract is signed. As a result, the period up to the signing of the contract is considerably longer than for a conventional project. However, our experience shows that the construction phase of projects—be they roads or hospitals or whatever—is diminishing dramatically because more time has been spent on the planning phase, which means that more work can be done in parallel and the asset can be delivered much more quickly. I feel that the time and money that is spent up-front on individual projects is well worth while to ensure that all the details have been sorted out before people rush into digging foundations and laying bricks.

Is it right and fair that the significant costs involved in that part of the process are not considered in assessments of whether the public procurement route or the PFI or PPP route should be taken?

The costs are obviously included in the successful bidder's price. He has to recover his costs—

Yes, but the point that I am making is that those costs are not included in the assessment of which of the two routes should be taken. They are discounted.

Yes, there is an interesting point on whether an allowance should be made for those costs in the public sector comparator. We talked a lot about public sector comparators on the way to this meeting. They are not a soft touch. The data that one puts into them have to be based on something. For example, in a roads project, one has to consider the evidence of previous experience, late delivery or over-budget delivery before one puts numbers into the public sector comparator—

In the A74/M6 PFI, the costs of the development process were not included in the public sector comparator. Audit Scotland felt that that was inappropriate. When that factor was included, the benefit—or otherwise—to the public purse was marginal. You have answered the point about time scales, but I want to ask all the witnesses whether the costs of the development process should be taken into account.

You probably are correct. The only point that I would make is that, as more experience has been gained, the amount of effort and time required for the process has reduced rather than increased.

For the City of Edinburgh Council, the rules changed halfway through. The auditors said clearly in the code of practice that development costs had to be written off in the year that they were incurred. Previously, local authorities had assumed that they could do that over the 30 years of the concession.

Development costs are taken into account, perhaps not in the public sector comparator, but in the affordability factor. In Edinburgh, we had to consider whether we could afford the costs. At present, Audit Scotland recommends that about 4 per cent of the total costs of the project should be for development work. That is quite a low figure, but it will be achieved only when we learn from the different PPP projects that we have been involved in.

Philip Grant mentioned that the City of Edinburgh Council had benefited from the experience of Glasgow City Council. That was certainly the case. Things were sharper because we knew what to expect. At officer level, Glasgow was very helpful in advising us on things to look out for.

Development costs are taken into account when affordability is considered. However, whereas costs would once have been written off over 30 years, now they cannot be.

Are you saying that PFI or PPP is becoming a cheaper product to develop and therefore a more affordable option?

It is becoming cheaper to develop. Whether it is more affordable depends on the affordability factor for the particular project. For example, the City of Edinburgh Council was lucky in having a number of surplus sites that were attractive to developers. That put our affordability at a certain level. Each project in each council will have a different affordability factor.

Do the other witnesses agree that, across the country and in different sectors, there is a trend towards reducing costs in the new partnership exercises?

There is a clear trend. Efficiency is coming through in the procurement programme—in the bidding and development costs of projects. To be honest, there is still a long way to go. On average, projects take 18 months to develop. The private sector must go through various stages starting with the OJEC—the Official Journal of the European Communities—all the way through various acronyms. Costs may be incurred at four or five stages until the preferred bidder is chosen, and all that cost must be taken at risk.

If the project does not proceed, costs are written off. Many consortia bid in a range of sectors and for a range of projects. I am sure that contractors and other sponsors will talk to the committee about the liabilities that they incur on bidding costs. Everyone is incentivised, particularly in the private sector, to shorten the development period for a project and to reduce costs. That makes our projects more affordable and makes the unitary charges that we can apply more acceptable.

We used to say that any PFI project would progress as fast as the slowest person—the most inexperienced. Sadly, it has been my experience that people from the public sector are frequently the most inexperienced. When people turn up for a new project, the financial, legal and technical people know one another from previous projects and know exactly how far the banks will go, but the public sector people are not very good at building on the experience that they gain from one project and taking that to another. The one exception is in the Ministry of Defence's purchasing of major equipment. It has a central pool that moves from project to project. In Scotland, health bodies and local government share little information.

That situation is changing. The City of Edinburgh Council, its officers and their advisers have been more than a match for the private sector in their ability to keep pace with the project and to deliver to the timeline and to the expectations of all parties. The Scottish Executive and local authorities have done much to have the public sector and the people who represent it operating efficiently through the process.

It must be appreciated that elements of the development review would have to be conducted even in a traditional procurement process. If a large scheme to build eight or nine secondary schools were to be delivered through traditional procurement, a large element of the development costs would still be incurred. Not all the development costs of a PFI should be considered additional costs.

I am happy to accept that, but the initial experience with PFI was that it incurred considerable extra costs that were not included in the assessment of whether a project should proceed.

I will return to your initial question. Different options must be considered. People should consider not just one figure, but a range of figures. You mentioned the discount rate, which has been 6 per cent since 1991.

The rate is far too high, given the current fiscal climate and likely fiscal climate in the future. That fact significantly disadvantages the public sector comparator.

A range of scenarios should be considered. We could say, "If we had a 6 per cent discount rate, perhaps the PFI project would be X per cent cheaper, but what would happen if the rate were 5 per cent or 4 per cent?" Other options could be considered in a similar way. For example, you talked about the initial professional fees. We can argue about whether they should be included in considerations, but perhaps we should consider both positions. We can make comparisons with and without all those professional fees included.

The matter is not straightforward. A range of comparators and factors should be used. We should be able to say, "If we assume this, that is the result." We are dealing with a long time, and many variables are involved. A sensitivity analysis that considers the options should help in deciding whether the PFI or public sector procurement method provides better value for money.

From a private sector point of view—

I am sorry, Brian, but we are against the clock. Lots of information has to be obtained, but you must move on to another question.

I will make a couple of other points on the process. We are talking about acquiring public assets and services, so the user or the citizen ought to have a voice. How does that work? There has been considerable public disquiet before and after some projects. How does such consultation fit in with the pattern of PFI/PPP?

A PFI or PPP is a commercial arrangement. What is the impact of commercial confidentiality after the event? I can understand why it is necessary in the bidding process, but once the contract has been entered into it is sometimes difficult to access such information, which would allow proper judgments to be made on whether we had value for money.

I have a comment on the involvement of citizens' or users' views. From local government's point of view, the best-value initiative has brought into sharper focus the need to consult citizens. I am not saying that that was not done before, but best value has introduced a new framework in which to do it.

CIPFA has consistently advocated that, when a PFI deal to bring new facilities into a local area is being considered, a framework of consultation along the lines of best-value principles should be adopted. That should ensure that the project meets not just the requirements of the private sector partner and the representatives of the local authority, but acceptance in the locality.

Much work has been done on what constitutes proper consultation. I will not go into that, but the important point—we made it in our introduction—is that, whatever vehicle is chosen to provide and procure public services, people must get the governance and accountability right, because public money is involved and it is meant to satisfy public needs. It is therefore incumbent on any partnership—be it a PFI or another partnership between the public, private and charitable sectors—to ensure that it does justice to consultation.

Brian Adam suggests that commercial confidentiality is a problem. I suggest to the committee that it recommend that, when such comparisons are undertaken, the details of the option appraisals—the comparisons between the PFI project and the direct public investment—are made available at some point. That would allow research to be undertaken to consider across the board the relative costs of PFI and direct public investment.

As for accountability, there is an inherent problem with taking the PFI route. For example, a local authority is responsible for providing and maintaining its schools and is accountable for that. It can be taken to task and can take action as appropriate. Once people go down the PFI route, however, they cannot do that so easily. That is because they are dependent on the intricacies of the PFI contract and its flexibility. If people decide that PFI provides value for money, one of the costs of going down that route is some loss of direct public accountability for the provision of services.

Before I bring in Lynn Brown, I should say that that is my experience of PFI contracts. I will be interested to hear the experience of other witnesses, in particular Mr Grant, as he mentioned the Glasgow schools project.

In Glasgow, the public wanted to know what new schools were to be built and what facilities they would have. The public wanted to participate and to become involved in the community assets of the schools, but they were shut out of the process. They asked certain questions and were told that the information was commercially confidential. The bidders would not say what they would provide; they would not even give an outline. I did not understand why they could not have said, for example, whether they would provide a swimming pool. How can that aspect of PFI be got around?

I will invite Mr Grant to respond after Lynn Brown, who offered to do so first.

I will speak from an Edinburgh perspective. The point about consultation is a good one. Consultation is crucial to PPP being taken forward positively.

The City of Edinburgh Council education department does a fantastic job. We are rationalising a number of schools, including a number of special schools, each of which has its own special difficulties. We have done that without one delegation going to council. The education department put a lot of effort into consultation, which is crucial. Representatives of the education department, including a councillor, went out to school boards and public meetings.

I am not sure how the health service could be brought on board. However, consultation routes are open to local authorities because they are democratically based, their meetings are held in public and they have school board regimes in place.

On confidentiality, when we took our final awarding of the contract to council, it was what is known as a B agenda, which means that it had to be heard in private. Before that, all the members, including the opposition groups, were briefed and were given access to the full final business case. That allowed them to ask questions. They had access to all the information that they needed to make the final decision.

In many ways, if a local authority is involved in a school project, the private sector regards the local authority as the body that represents the interests of the communities that it serves.

The convener referred to a situation in Glasgow, which would have arisen when the consortia were still competing. At that point, the detail of their bids could have been picked up on by other bidders, which could have led to a loss in the value of their bid. In such situations, it is difficult for the consortia to consult. However, once a preferred bidder has been announced, the service offering and the quality and configuration of the schools may be opened up to some level of public consultation.

That was the point at which the problem to which I referred arose. Once the preferred bidder had been announced, and the plans had emerged, local communities said that they wanted more than that, or something different, but by then it was too late to influence the process.

It all comes down to the public sector stating its service and facilities requirements. The private sector consults extensively on its design. It holds focus groups, goes out into the communities and holds meetings. It wants to gather information so that it can produce the optimal solution.

The public sector also has to take on board the fact that it is delivering the services that are required by the communities that use those services. The private sector has an expectation that, when those services or asset requirements come forward, they are the function of a defined process. If PFI has done anything in the past five years, it has brought people closer to focusing on what they want to achieve.

In Dundee, PFI has delivered through a joint-venture company in which the local authority is a 40 per cent shareholder. The deal for the plant was completed in 1997 and the construction was completed in 2000. Since then, six-monthly reports have been presented to the local authority, in open session, on what has taken place at the plant and the financial arrangements between the local authority and the joint-venture company. We have also set up a better-neighbourhood agreement with local community groups, so that the joint-venture company can hold regular meetings with the local community about the operation of the plant. We did that not because the plant is a PFI vehicle, but because we wanted to take on board the local community's concerns about the type of plant that was being provided under the PFI venture.

I want to move on to what happens once a decision—whether PFI or PPP—has been made. In your experience, what are the critical factors that make the public-private split a success and what evidence do you have to demonstrate that point?

The co-operation between the local authority—or the trust, or whatever the public body is—and the private sector is increasingly being held up as a sign of success. Good schemes are those that are regarded as genuine partnerships. "Partnerships" tends to be an overused word in this sector, but in some projects there is a close relationship between the public and private sectors. For example, we are involved in the funding of Law hospital, which opened earlier this year. The management of that project is a shared responsibility between the trust and the special purpose company, or SPC. Senior representatives of the trust and representatives from the private sector meet at a monthly forum that has a set agenda of reviewing the quality of service delivery and making specific recommendations. That is a genuine partnership. Everyone relies on the contractual framework, but overlaying that are strong partnership relationships. When people rely on just the contract and the contractual relationship, that level of co-operation and understanding is lacking and that can cause problems for a scheme's operation.

Do you have to tick particular boxes to assess where the project is going after the contract has been signed? Are there specific points that are critical if both sides are to feel that the project is going where it is supposed to go?

Yes. On any scheme, after a contract has been signed, the next major milestone is commissioning—the point at which the asset or service is provided. There has been much learning from experience over recent years, in particular over the past 12 months, about the importance of co-operation during transitional phases and during the commissioning of assets. That is a key point. Once the asset is operating, the whole risk matrix changes. Up to that point, there is, in effect, a construction environment in which there is a well-established means—transferred straight from the construction industry—of monitoring progress and risk in delivery.

PFI is developing a new way of assessing the risks of operating a hospital facility, a prison, or a group of schools. We are adopting new ways of managing such assets and of managing and monitoring service performance. We are also ensuring that SPCs are appropriately resourced to manage and monitor not just the closing-out projects and building assets that we have talked about, but the physical delivery of the services. Good projects are now regarded as those that have such a matrix in place between the private and the public sector and are operating well at that level.

I will answer, then ask Lynn Brown to talk about performance indicators and other matters in some of the contracts in Edinburgh.

Philip Grant pointed out that, before the final PFI agreement is signed, outcome and output measures will be agreed. At least a percentage of the contract payment will be determined by whether those outputs and outcomes are achieved.

It is important to think about how the success of a PFI project can be measured. Short-term success may be measured by examining whether the targets that were agreed in the contract output measures have been met, but longer-term success is harder to measure. As has been mentioned, the PFI projects will run for 20 or 30 years and many changes will take place during that time. It concerns me that a large number of bilateral agreements are being created between trusts, local authorities or Government and the contractor. Fifteen or so years into that 20 or 30-year period, as the population changes and advances are made in medical science, will it be possible to reconfigure the way in which services are delivered when the local authority has to make sense of a large number of PFI agreements? That might not be quite the area that you asked about, but it is important to think about how the success of a contract can be measured.

Lynn Brown will talk about the situation in Edinburgh.

It would be helpful if she also talked about the effectiveness of the monitoring arrangements.

Philip Grant talked about a spirit of partnership and not relying on a contract. That is key, but a contract is still needed because the process is output-driven. The regime that is linked to that means that, if our partner does not deliver, we have comeback because of the contract. We would not go to each page of the contract to win our point, but the existence of the contract helps in any argument.

The situation with schools is different because, although the running of the building is transferred, the educational attainment of the children is still the responsibility of the local authority. The statutory indicators that deal with class sizes and so on still apply. There is a slight split between recognised indicators and outcomes.

The council would measure the success of the project by examining whether the outputs were being delivered without a lot of input on our part. We are at the beginning of the process of using PPP in education, but in other areas where we have not used PPP, we measure success by whether the outputs that we wanted to be achieved have been achieved.

As that answer demonstrated, it is difficult to point to a success in relation to PFI. It is still early days. The first PFI contracts were signed in 1997, which means that the oldest are only four years old. Given that the contracts are supposed to last for up to 35 years, four years is not enough time to see how things are going.

To be negative, a range of PPP and PFI projects have clearly failed: Railtrack was declared insolvent; the Public Health Laboratory Service has had problems; the Contributions Agency and the Passport Agency computers have had difficulties; and a number of private sector interests that provide housing benefit services to local authorities have pulled out early in Lambeth and Southwark. It should be borne in mind that, even after the decision has been made to go ahead with a PFI or PPP option, there is a risk that the private sector provider will not continue to provide the service over the 30 years of the project.

Is there a danger that, because a council that has entered into a few PFI arrangements has tied up a large part of its revenue stream over an extended period, the flexibility of its budgeting arrangements will be lost and it will be locked in to a spending plan for a long period, even if circumstances change?

That is a problem. We said earlier that it could be argued that an advantage of PFI projects is that they will avoid the backlog maintenance situation. In the health service, individual health trusts have chosen to cut back on maintenance and to maintain services in other ways. If those trusts feel that they do not have enough money, they have chosen not to maintain their buildings sufficiently but to spend the money on nurses, doctors, accountants and administrators.

The position is similar in council housing, where there is a significant backlog maintenance situation. In Glasgow, the tenants are being offered the opportunity to transfer out of local authority control, with the promise that their houses will be properly maintained. It is swings and roundabouts. PFI locks bodies into an arrangement in which adequate maintenance will be done, but at the cost of lack of flexibility. The hospital may be well maintained, but the trust might not have the resources to maintain the staffing levels that it feels are necessary in the clinical areas. A political decision has to be made as to which scenario is preferable.

Bill Davidson's submission states:

"the public sector is committing itself to 25 years of recurrent funding of the PPP unitary charges—regardless of whether the serviced asset is still required or not."

You mentioned IT services. Could you give other examples of services that might not be required? A school, prison or hospital would be required. For what other services might the need decline during the period when payment was to be made?

When I wrote that, I was thinking not so much about the IT side, but about school facilities that might be affected by demographic shifts.

We have worked on projects where schools had been built for previous generations and, as a result, the kids were being bussed from one side of the motorway to the other and all that sort of nonsense. The need for facilities like primary schools waxes and wanes. Similarly, certain types of health service facilities may not be needed in the future to the same extent as they are now. Not that long ago, for example, psychiatric patients were all institutionalised in psychiatric hospitals, but we have now moved to community-based care, so those facilities are no longer required. Who can say what will happen in relation to correctional services, prisons and so on? Will we need the same number of them in the future?

My point was that, the more of the budget that is committed to 25 year deals, the less flexibility there is. The other point is one that was mentioned earlier: if we do not spend on maintenance now, the hole in the roof gets bigger and the problem becomes very expensive to fix when we eventually have the budget to do so. Fixing the hole now would give better value for money.

Various points have been made about accountability, relationships and, from the public perspective, who is responsible for what. The Edinburgh model covers a wide variety of partnerships. How do you decide who is accountable for what in building your models? What experience have you had with that? When Mr Wynne gave his introductory evidence, he suggested that the Parliament would be the final arbiter of which models were to come into play. The purpose of this exercise is to try to get a feeling for that. When we discuss this subject, accountability is always mentioned in relation to public perception and how people respond. How do you define and handle accountability in the early stages of establishing different types of partnership?

Accountability is an issue. The City of Edinburgh Council follows the guidance in an Accounts Commission document, "Code of Guidance on Funding External Bodies and Following the Public Pound". We are aggressive about accountability, because public money is involved.

We can put in place anything that we want to ensure that public money is delivered properly. If a company is being signed up, for example, memorandums, articles of association and whatever else is needed should be set up and rights of access to the books should be ensured. Things have to be right from the beginning. The bottom line is that we must ensure that we can follow public money, because the auditor can follow it.

The council tends to think that it is accountable. If there is a wholly owned council company, the council has an involvement and is held to be accountable in the public eye. Things must be clear and in place to ensure that accountability can be delivered for councillors.

I will answer the question, but I want to say something first about tying up maintenance or revenue budgets. There must be a strategic view at local public service body level, and at a wider level, of what PFI and more traditional forms of public sector procurement are used for. If people take a piecemeal approach and decide to sort out a backlog with PFI, that would be poor in two ways. First, a lot of revenue budget may end up being tied up in PFI payments. Secondly, a strategic view on prioritisation will not have been taken. What is the wider picture? What cross-cutting service picture is emerging on how to progress?

On accountability, my colleagues and other witnesses made the point that it is important to make it clear at the start of a contract who is responsible for bearing which risk and who gets the benefit.

Another point that has been made is that the refinancing of PFI is important. With the early PFI contracts—although I am not sure that this applied to many of them—the refinancing benefit tended to go the way of the private sector partner rather than the public sector. If one is careful and clear up front, there should be no problem. The private sector can price in that risk and allocate money to cover it. The question of who bears responsibility when things go wrong should be sorted out beforehand, but we hope that such a situation will not arise.

Public finance should be cheaper than private finance, given the risk-spreading capacity of the public sector and its access to tax revenues. Is not that a strong argument in favour of the use of public rather than private sources of finance?

We are intrigued by that question—it is a chestnut that comes up regularly. We undertook a simulation before I left KPMG. Members should not ask me too many technical details about that, because staff did it for me. We took the figures for a conventional PFI deal that we had just closed with bank finance—it was sorted out as a complete deal—and compared them with the figures that we would have got if we had financed the deal with public sector borrowing at the rate at which the Government could borrow in the market at that time. We compared the net present value over the 25 years of the concessions. We took all the capital expenditure, and all the life-cycle and maintenance costs and discounted them back for common comparison. The difference between the two deals was 3 per cent. That would be the potential saving over the life of the project of moving from standard commercial bank finance to public sector finance.

Philip Grant and his colleagues do not like this, but PFI projects have increasingly moved away from bank finance into the bond market, where we were heavily instrumental in introducing many new types of financing to do with bonds—wrapped bonds and index bonds, for example. I will not bore the committee with details, but, in essence, large life companies end up sitting with huge pots of money to invest on behalf of people who want to buy annuities, for example, in a market in which there is not nearly the amount of Government gilt that there used to be. PFI-type projects with the right risk profiles and the right insurance wrappings could be considered as pseudo-gilts. There is, therefore, cheaper money available than the banks can lend. The life companies are genuinely looking for 25-year investments and do not want those to be paid off early.

Frequently, life companies are being asked to provide pensions with index-linking, so they are looking for a revenue stream that is index-linked as well. That allows index-linked bonds to be used, which means that the interest rate that is paid on day one of the construction project is lower than it would be with conventional bank finance, but the cost goes up each year in line with the retail prices index. It can be demonstrated that the income of national health service trusts, for example, has always gone up by at least the RPI, so their ability to pay year on year is not impacted, but the amount that they have to pay on day one is much reduced, because they start off at a much lower interest rate when they have the maximum amount of borrowing and the maximum amount of rolled-up interest. By using those types of instruments, it is possible to get close to the point where there is not much material difference between conventional Government borrowing and a PFI-type deal. The banks, I might add, are fighting back hard. Competition in this area is in the taxpayer's best interest.

I notice that a representative of one of the banks wants to speak, no doubt about the banks fighting back.

Four years ago, for example, the blended cost of capital of the whole private funding package in a PFI scheme would be about 9 per cent. The cost for transactions closed in the last few months of a scheme will be closer to 7.2 per cent, which is because of a number of factors. The first factor is competition between funders. The margins for debt that are provided by banks like ours have dropped over the same period from about 1.5 per cent to less than 1 per cent. That is at the lower end of where a commercial bank will drive for return. Secondly, bond markets have developed. Wrapped bonds and index-linked bonds have emerged in competition with banks. We compete with them vigorously. The result of that is to introduce more competition to drive down the cost of private finance.

When you get down to those levels of capital, through to about 7.1 per cent, competitive market finance drives the cost of capital to a point at which the project must deliver value beyond the cost differential. It should be borne in mind that the cost of private finance takes into account the risk that is associated with a project, and the premium for that risk. The cost of a benchmark gilt takes into account no risk at all, other than the fact that taxpayers in the UK will continue to pay taxes to service debt.

I was surprised that in the two accountancy papers we received there was little content from an accountancy point of view on how risks are valued in the process. How do you value risk in your competitive marketplace? Is it a science or an art form?

It is a competitive art form in many ways. In such circumstances the private sector is being asked to take a commercial view on the risk—over an extended period of time, as you have noted—and to take that risk away from the public sector and take it on itself. There is extensive due diligence with technical and professional advisers on all aspects of schemes, such as life-cycle costs and various other risks. In those projects an inordinate amount of time is spent validating the assumptions that are used in them and in making judgments on risk. Extensive sensitivity analyses are done on complex financial models in order to understand the impact of costs being greater and the cost of service delivery being greater, and what that does to the robustness of the project.

It must be borne in mind that those projects include no recourse. We lend to them on a non-recourse basis, so if the project is not robust and does not deliver, we have nobody else to whom to go to get the money to repay the debt. The parties to whom we subcontract the building and services, for example, are liable only to the extent of their contracts. We cannot go to them and ask them to help us out because the project is not robust. We spend a lot of time examining and sensitising the risk, which is mainly cost risk and delivery risk. We ask whether the parties can physically deliver to the public sector what they are signing up to.

In many respects, I would argue that the interests of the banks in the sector are closely aligned with those of the public sector during the life of the project. However, it might take time for people to take that argument on board. It is in a bank's interest for a special purpose company to perform to the letter of its project agreement. If it does not, its revenue will reduce and it risks being unable to service our debt.

Members should be assured that special purpose companies in PFI are closely monitored by their funders and bankers at operating and financial levels to ensure that they deliver on their contractual obligations. Our interests are aligned for the full 25 to 30-year period. Four years ago, banks were lending for only 15 years; now they are going to 30 years. That has contributed to the existence of more affordable projects. The loan period might be for 25 years, but someone in the bank will still be waiting at the end of that time. They will ensure that spend is being made to rectify the quality of the assets, that the service is still being delivered and that the quality of the assets is not resulting in the payment mechanism penalising the company and the bank.

In our submission and publication on PFI, we make the point that there has recently been an appreciable difference between the cost of Government-backed borrowing and borrowing from the private sector. That might change in future, but that is why we have said in the recent past that it is probably not a good idea for a public service to go into a PFI deal if it is looking for a building or something like that. That is especially true if one assumes that after getting the right contractor to build it, the contractor will manage the project with you so that you do not get a bid-cost overrun.

Empirically, one of the interesting facts about the local authority sector throughout the UK is that the Public Works Loan Board is the main vehicle for local authority borrowing. Those loans are offered at very competitive rates. I am straying out of my territory, but that is probably one of the reasons why a bond market has not emerged in the local authority sector.

Profit is inherent in the PFI system and is generally assumed to be much more than the cost of capital. I have recently seen a couple of estimates of such profit in the order of 12 per cent to 15 per cent. We should ask ourselves how the PFI option is more efficient. In the prison sector, efficiency has been brought about by employing fewer warders, warders working longer hours for less money, and non-provision of pension schemes that are as attractive as those in the public sector. That is one of the costs of going with PFI. It might well mean that there is less employment than in the public sector comparator.

The PFI datasheet on the Scottish Executive website says:

"East Ayrshire is an area of high unemployment and the prison makes a significant contribution to the local economy with the majority of staff being from the local area."

Clearly, if you are going to go down the route of PFI, one of the costs is lower employment at lower wages, probably with less advantageous pension schemes. That might well result in later costs to the state in terms of providing those employees with the necessary living standards when they retire.

That is a major aspect of the problem and is one of the main reasons for public opposition to PFI and PPP projects. It is only part of the equation but it is an issue. We are particularly interested in the protection of employment of public sector employees.

Bill Davidson's submission says that he believes that

"it is not the role of the tax-payer funded public sector to create and fund more jobs than are necessary."

I do not think that we would disagree with that. However, in many cases, the number of jobs that are being carried out after the implementation of a project is significantly fewer. Not only are the jobs fewer in number, but the conditions of service are poorer. One can understand a reduction in the number of jobs, but not a reduction in the conditions of service, pension provision and so on. Those issues must be dealt with satisfactorily to overcome one of the major obstacles in the public's perception of PPP and PFI. I ask Mr Dorward to give us Dundee City Council's view.

I do not think that a PFI project will necessarily mean a reduction in the wages or pension rights of individuals who are employed by that company.

Not necessarily; however, the evidence shows that that often happens.

Control of that rests with the public authority that is letting the contract. That authority must make it clear in the project specification that the contractor who wins the contract must pay fair wages and comply with the current situation regarding pensions. The whip is definitely in the hand of the public authority. If the authority is remiss and allows such situations to go unmonitored, I am sure that the private sector will be only too ready to abuse its position. However, it is the public authority's responsibility to ensure that the contracts that it enters into under PFI protect staff in those circumstances.

That is important. I accept that.

The transfer of employees is an important part of the PFI process. It is covered by the Transfer of Undertakings (Protection of Employment) Regulations and is an extension of the outsourcing part of those. An interesting pilot scheme is taking place—in which three of our hospital bids are engaged—using the retained employment model, which is being delivered through the Department of Health. According to that model, management of employees passes to the private sector, but their terms and conditions and their employment status remain with the public sector. We are awaiting the decision, which will be given in the next few weeks, on whether our projects will have to progress in that pilot scheme using the retained employment model. The ways in which management can address your concerns about the terms and conditions of employees are being considered.

That is an important issue. We will have a session specifically on employee-related issues later in our inquiry.

I would like an explanation of what a wrapped bond is. We are moving from PFI to PPP, and you have hinted that there could be further developments in that area. What is your view of the not-for-profit trust approach, especially as it is linked to the idea of raising funds through bonds?

A wrapped bond is simply a bond that is credit enhanced. In other words, the holder of the bond benefits from the covenant or the quality of the person who is wrapping the bond, who effectively guarantees the bond. An organisation with a AAA rating guarantees that the bond and the holder of the bond do not have to rely on the performance of the project, but on the performance of the wrapper. The AAA entity wraps round the credit risk, and that reduces the cost.

Such organisations that operate in the UK are mainly American-based. There is a market for municipal bonds in the States, which developed the concept of wrapping, and there is a cost to that. The organisations get a return for their wrap. The cost of the bond is normally priced from the gilt and there is the spread, which is the AAA margin. There is also the cost of the wrap on top of that. That is what a wrapped bond is. It attracts people to come to AAA status companies. They do not give too much thought to the project; all the due diligence and the risk of the project are taken by the wrapper.

I would not say that organisations come to a project without thinking about it too carefully. Life assurance companies and pension funds do not invest their money lightly, because they must protect it. The wrapped bond means that in effect they are not exposed to the risks of the project. Somebody else provides them with a guarantee that, if the project goes pear-shaped, they will still be able to get their funds back. Those funds are frequently people's long-earned pension funds that they are looking for a decent return on over a long time.

I apologise for the pun, but it is time for us to wrap up this meeting. We have covered a wide range of issues, and this has been only the second evidence session of our inquiry. The information that the witnesses have given us in written and oral form will be very helpful, and we appreciate it. At some stage, we might ask for further information if we want clarification of points or issues that have developed in subsequent questioning. Thank you for the time that you have given in preparation for this session and in answering our questions.