Finance Committee, 18 Dec 2001

Meeting date: Tuesday, December 18, 2001

Official Report 391KB pdf

---

Private Finance Initiative/Public-private Partnership Inquiry

Good morning to the press and public and to the witnesses whom we have invited to give evidence in our private finance initiative/public-private partnership inquiry. They are Professor Allyson Pollock, Des Murray, Paul O'Brien, Richard Blackburn, Professor Paul Grout and Geoff Haley.

Our adviser has devised a range of questions for the committee to ask. As there are six witnesses, it is probably not appropriate for me to invite each of you to make an opening statement. We would like to go straight to questions. If issues in which you are interested arise in relation to questions, we would be happy to hear from you. We will direct some questions towards specific witnesses. Do not feel inhibited about indicating that you want to comment on a certain issue. We will perhaps have a round-up at the end, if some issues have not been explored.

Donald Gorrie has sent his apologies.

I will ask the opening question. PFI was intended to reduce costs and time overruns in the public procurement of capital projects, while improving value for money and offering innovative solutions for the delivery of public services. Has PFI lived up to those expectations? If not, why not? What evidence is there to support your point of view?

We must go back to the fundamental arguments for introducing PFI. The first is the requirement to constrain public sector borrowing below 40 per cent of gross domestic product for European convergence. The other argument is that PFI represents value for money in public procurement.

The major point on the first matter is that public sector borrowing is now down to about 35 per cent of GDP, so with the PFI schemes that are under way, we would easily stay under the 40 per cent barrier. The argument is irrelevant, so the debate moves on to the one about value for money.

The National Audit Office report for this year shows that, in some of the larger schemes, savings of between 10 and 20 per cent have been made through PFI. However, a recent Audit Commission report indicates that there have been savings of only 3 to 5 per cent on smaller PFI projects. That is concerning, given the fact that the European Union is examining the use of the negotiated procedure in the setting up of PFI schemes. If the negotiated procedure was suspended, the likely outcome is that transaction costs would rise by 5 to 6 per cent and we would quickly see any savings that had been accrued through PFI nullified. On top of that, the wider social, economic and environmental implications of PFI mean that any savings from PFI are minimal and very difficult to show.

We are seeing improved efficiency in the construction of public works. The traditional procurement method often led to large cost and time overruns. The Thames barrier is a prime example of that. The original capital cost was £54 million, but the project ended up costing £256 million after years of litigation. Under PPP, contractors are entering into fixed price contracts with limited variations; the projects are being completed on time and to budget; the private sector retains all the risk and no risk is transferred to the public sector; and the contractors are performing. From that point of view, there is now substantial efficiency in the construction market.

On the EU procurement rules, the problem is—as I stated in my submission—that the UK's government of the negotiated procedure is being disputed. The European Commission wants the simplified procurement procedure to be used rather than the negotiated procedure. If that procedure is used, all the bidders will have to go down to the final, best price offer. Instead of a preferred bidder being selected at a certain time, all the bidders will have to be involved in the finer points. It can cost £2 million to £3 million to bid for a £30 million to £40 million PPP project. Each consortium that bids will incur that sort of cost to get to a point at which all but one bidder will lose. That approach is completely unrealistic and unsustainable. I represent much of the construction industry—we have 169 members, including the banks and the contractors—and it is extremely concerned about the EU's decision. It will produce such tremendous costs that it will not be possible for consortia to recover them, so a lot of the bidding will stop.

I seek clarification on the first part of your answer. You put one of the conventional arguments that we have heard in favour of PPP—that, under conventional procurement, the definition keeps changing and there are huge overruns whereas, under PPP, the specification can be much more fixed, with few variations. Why cannot a fixed specification be arrived at through conventional procurement? What happens? Do both procedures start off with the same specification, but under PPP, it is not allowed to be changed, so that—at one extreme—a PPP might end up as something that is not wanted? At the other extreme, are people forced to make up their minds? Why can we not find a mechanism for forcing people to make up their minds under the conventional procurement procedures?

We have not found the correct mechanism under the public system. Often, that is because of the incorrect use of contracting procedures. For example, a lot of Government projects are let under the general conditions of Government contracts for building and civil engineering works 1, an amended joint contractors tribunal form or an Institution of Civil Engineers conditions of contract 5th edition. Those contracts allow variations and allow claims to be made. It is the claims that cause the problems. If there are variations, there are cost and time overruns.

In the PPP process, the design is finalised during the negotiating procedure. When the construction contract is signed, there may have been a year's negotiation of that contract linked to the project agreement. The design for whatever is being built—a road or a hospital, for example—is finalised and the contractors are not allowed to vary it because the funders have bought into that final design. If anything is to be changed, the funders have to be approached for permission; without that, nothing can be changed.

I do not understand why the second model can exist only with private capital and only with contracts that also allocate to the contractors the servicing and maintenance of the asset over its lifetime. Surely that model could be achieved equally with capital from the Public Works Loan Board and when maintenance and servicing contracts are allocated elsewhere.

If there was a more efficient procurement procedure in the public sector, it could.

Mr O'Brien is suggesting that the public sector borrowing requirement argument is out of the window and that, if Europe has its way, it will negate any cash advantage from PFI.

Potentially.

Mr Haley is suggesting that, if the EU gets its way and bidders are removed from the marketplace, the construction sector will no longer have a competitive edge. Does that link back to public procurement robustness—the client model for the tender? Is there a split in the evidence you are giving us? Is Europe meddling and is that costing the procurement system? Presumably, the same rules will apply to the public system when a project is put out to tender. Is there evidence to show that there is more to the matter than just the cost of capital? Is it more about the notional value of achieving the delivery earlier, on time and in a robust manner?

Whether you think that Europe is meddling depends on whether you believe that it is appropriate for large-scale public projects to be procured following negotiations with only one bidder—whether you believe that that is open, above board and transparent. I think that Europe is correctly questioning the use of the negotiated procedure in every instance. I do not believe that the negotiated procedure should be followed in every instance.

But it is not used on day one; it is usually used after initial negotiations with several bidders.

Yes, but we have seen evidence of a project in which only one bidder was involved. When it was completed, it was much bigger than it had been at the negotiated bidder stage. That process is flawed.

Does Mr Haley want to respond to the question?

It is a practical problem. There are different directives related to the EU procurement rules, which deal with different procedures. The whole point of the negotiated procedure is to recognise the complexity of a project, such as a PPP, after the pricings from all the consortia have been received in the bids. All the contract conditions cannot be finalised with all the different parties because, if there were four bidders, that would require four separate sets of negotiations in different rooms. First, private sector companies do not have the human resources to bid for all the projects that are proposed and to provide people to sit in all those meetings. Secondly, they do not have the financial resources to do that. The Government has accepted those facts and has said that it will not accept the decision by the European Court of Justice if it goes against the UK.

The question is, at what point should consortia be cut out of the process? A lot of planning for new developments involves a master planning process in which three or four people are involved. A selection is made and then somebody carries on with the project. The process must be competitive; the issue is when and how the selection should take place.

Under the simple procedure, companies go right down to the best and final offer, so every part of the deal must be negotiated before it is decided which offer is the best and final one.

You said that each of the three or four bidders might incur costs of £2 million or £3 million on a £30 million project. That is a disproportionate amount of the overall cost, which is why the private sector might not wish to be involved. Not all the bidders are involved until the final stage, so how much do the unsuccessful bidders spend? We have heard that the European Union does not agree with that system. Who pays for the wasted bids other than the private contractors who do not win the contract? I presume that the contractors recover the costs through the next bid. A principal part of many organisations' business is trying to win public sector contracts. If they have costs that are even a tenth of the £2 million or £3 million, they must recover it from a public sector contract at some point. Ultimately, who takes the risk?

We must remove one misconception. We are talking about a £2 million to £3 million bid cost on a £30 million to £40 million capital programme. We must take into account that a 30-year operating contract might be linked to the programme. We must consider the construction contract and the long-term operating contract. The sum of £2 million to £3 million is not that big for a 30-year operating contract.

Do you mean that, to fund the bidding process, we pay £2 million or £3 million over the odds for the operating contract?

That is the cost of bidding for some projects. One reason why the costs are high is that the documentation that is released by the public sector is often inadequate. For example, the national health service trusts that were involved in hospital projects issued documentation, but it took nearly three and a half years for the Department of Health to issue standard documentation. The standard documentation covers 70 per cent of a basic contract. In other words, 70 per cent of an NHS hospital contract is fixed; companies know exactly what the contract terms are and can estimate the cost and make a bid. Around 30 per cent of contracts must be negotiated.

For three and a half years, none of the terms were fixed and every contract had different clauses, which is why the initial procurement procedures for hospitals were bad. The procedures for roads were good. The Highways Agency issued the first roads contract—for the A69 Carlisle to Newcastle road—and that was improved for each of the next five bidding processes. The contract became standard and it was successful. The same is true of HM Prison Service.

Not all companies will go down to the wire with a bid that costs £2 million to £3 million because they might not be successful in securing the construction contract. What are the costs of the procedures for unsuccessful bidders and how do they recover those costs? Have companies left the marketplace as a consequence of failing to win bids? What does it cost a company to participate in a bid for a £30 million construction project?

The losers probably incur costs of £250,000 to £400,000 in going down the preferred bidder route.

Do companies that bid for such contracts drop out of the process as a consequence of a series of unsuccessful bids?

Some major companies—I cannot name them because some of them are clients—pulled out of the PPP or PFI process because it was too expensive. Companies that do not have a success rate of one in three or four pull out because there is no reason to stay in the marketplace.

A requirement for a success rate of one in three or four gives a limited market, which does not imply that there is much competition. It might be in the interests of the limited number of contractors not to have a true market, because of the significant costs of the bidding process.

The point of the procurement rules is to encourage competition. The move by the European Commission would discourage competition because it would mean that smaller companies would not be involved in the PFI process—they would withdraw.

I ask our two academic experts to review the evidence about PFI.

It is useful to start considering the issue with the example of refuse collection. Unlike many sectors in which every PFI is different, the interesting point about emptying dustbins is that roughly the same technology is used everywhere. A lot of research has been done on refuse collection and has found that what really matters is the fact that competitive offers are made. It is almost statistically impossible to find a difference in the reduction in cost, whether the public sector body which has done the job for a long time or a new private company wins the contract. However, the reduction in cost that comes from competitive tendering is around 20 per cent, which is large. That strong statistic jumps out all the time. The reduction is independent and seems to come from the competition rather than from the private sector, which is an interesting observation.

An interesting point about the history of the privatisation programme is that the industries that were involved had huge gains and were able to produce the same output with half the labour force —although there has been a dispute since 1984 about where the gains went. With PFI projects, the huge benefits do not jump out. One reason for that is that the bidders do not all jump in with offers that are 20 per cent less than the public sector comparator, which seems to be the evidence from the report carried out by Arthur Andersen. A reason for that is the one-size-fits-all model. Everything is discounted at 6 per cent whether it is a public sector comparator or a PFI project and regardless of whether it is a Ministry of Defence project or a roads project, which are radically different things. There is huge sensitivity to the discount rates because some projects are for 40 years and the benefits are spread over long periods. Until we deal with those issues, it is almost impossible to discover accurately the financial benefits of PFI projects. I have been making that argument for quite a while and I submitted evidence on it.

I understand that the Treasury is seriously considering the problem as part of the green book discussion. There is open dispute about whether the discount rate should be 6 per cent or 3.5 per cent. We can understand how small changes in the discount rate have a huge impact on the real cost of a project if we consider that a 40-year project might involve a payment for every car and lorry that uses a road.

I believe that until we get the discount rate sorted out it will be terribly hard to address whether PFI is cheaper. My personal view is that when the rate is sorted out, PFI will be shown to be better value than it is under the existing rules, although I accept that the situation could go the other way. The critical point is banging in a number of 6 per cent for every project, simply because that is the appropriate test rate to discount, means that it is impossible to marry up the figures and say unambiguously that one system is cheaper than the other. That is where the problems are.

In other words, the factors that determine the cost-effectiveness of PFI are governed by the financial arrangements, not by the relative efficiency of PFI as a mechanism for constructing a public works project.

The two go together—cost-effectiveness is not determined only by the financial arrangements. At one extreme, we may be seeing only the cases in which the PFI arrangement can deliver such a large gain that, even if the 40-year returns are discounted at the wrong rate, the project is still worth doing.

There may well be lots of other things for which PFI would be suitable. The argument is not one-sided, because the value-for-money rules do not truly appreciate quality differences. One of the criticisms that people have made of PFI is that it has not resulted in tremendously innovative ways of working, but until one comes up with a system that rewards people for being innovative, it is much easier to base the private sector project on what would have been delivered anyway. The movement away from the traditional cost-benefit analysis, which is what everyone used in the 1960s and 1970s, to the precise and simple financial rules, masks a huge amount of difference between projects. Until the problem is seriously sorted out, it will be hard to answer the question about the size of the gains.

It is important to remember that PFI is not a neutral financing mechanism. It fundamentally alters the accountability and nature of public services and we have quite a lot of data to show that. This morning, representatives of the construction industry claimed that value for money and efficiency have resulted from PFI, but no evidence was provided for those assertions.

One of the most worrying aspects of the situation is that the construction industry has an advantage over everyone in the room as it has seen the full business cases and the value for money details of schemes. I have tried to get hold of full business cases for the Glasgow schools, the Falkirk schools, PFI schemes for water and PFI schemes for transport. None of them has been made available for public scrutiny and so any discussion on value for money or efficiency becomes academic. The fact that that information, including the value-for-money analysis, is not in the public domain is a huge issue. When we get hold of those documents, often pieces of the data are kept back, which means that we cannot compare the cash costs of the public sector comparator with PFI. For example, we are given the net present costs for the PFI option but not those for the public sector comparator. One of the critical issues for the Finance Committee is to ensure greater accountability for public funds and services by ensuring that all the business cases, plus all the data, are placed in the public domain. That would allow us to decide the basis of claims of efficiency. That point is fundamental: how can we talk about value for money without such data?

I have submitted two papers to the committee based on the data that have been made available in hospital business cases. One paper takes apart the value-for-money case. My work was aided by the fact that the Westminster Health Select Committee has, under the public expenditure questionnaire, obtained some data on value-for-money claims and risk transfers. I will not go through the paper, but it is important to remember that value for money is simply an economic appraisal and does not give us an analysis of cash costs at a local level. Moreover, everything involved in the value-for-money analysis depends on the schedule of payments and the application of the discount rate and risk transfer. As Paul Grout said, the discount rate can alter the outcome, even by switching it by a tiny amount. When we examined the figures for six hospitals in the public expenditure questionnaire, we saw that the changing of the discount rate had had no impact on them and that the 6 per cent discount rate still came out in favour of the public sector comparator—that is in spite of all of the advantages of front-loading and the use of too high a discount rate. The element that swung the balance was risk transfer. I draw members' attention to the tables in my paper.

The function of risk transfer seems to be about disguising the true cost of PFI. Today, we heard talk of transaction costs, but they are only one of the four major costs that add to the cost of PFI; the financing costs can add 30 per cent to 40 per cent to the cost of the scheme. An examination of risk transfer shows that the costs after discounting for PFI are roughly equivalent to the net present cost of the public sector comparator. The function of risk transfer is to swing the balance in the direction of PFI. The difference between the two is exceedingly slight. That means that the whole of the Government's argument rests on a microeconomic case of risk transfer.

To what extent is the private sector taking on risks? It is important to realise that the claims of risk transfer are ex ante not ex post and that we have no evaluations of them. We know of four or five main issues in the hospital sector relating to the way in which PFI fundamentally alters the nature of procurement. First, there has been concern about the loss of control over public services and accountability. We see that by the fact that the full business cases are not even in the public domain. The second main issue, to which even the construction industry has alluded, is concern about the fragmentation of the planning process. There has been a move away from planning for needs to planning for the needs of the providers and private sector industry. The third main issue is one of cost, quality and access, at the bottom of which is equity—after all, public services are supposed to be about providing universal goods. We know that, with PFI, because the high costs have to be met out of a revenue budget rather than a capital budget, the amount of money that is available is squeezed. In health, that has had disastrous consequences for cost, quality and access. I commend our publications to the committee. We have summarised some of the issues and the claims about risk transfer.

When considering risk transfer, it is important to consider schemes that have failed. The Treasury and Andersen published a shocking misleading report, which claimed that there had been 17 per cent efficiency savings in the 29 schemes that the report dealt with. Most of those schemes had no data on risk transfer and the report rested on one scheme that showed 80 per cent of the so-called savings due to risk transfer: the national insurance recording system 2 project, which went disastrously wrong. NIRS 2 was a Contributions Agency scheme that was an absolute disaster. Provision of a replacement national insurance recording system was contracted out to Andersen. The system lost an estimated 5.2 million records and was closed with the consequent loss of billions of pounds of tax revenue. However, Andersen paid only £3.9 million in compensation for the delays. When we examine the claims about risk transfer, we must look ex post and in detail at some of the schemes that go badly wrong.

It is important to make the point about affordability. Value for money is simply a piece of economic sorcery. It is a black art that is designed to give the answer that is sought—any professor of accounting will tell you that. It is absolutely meaningless and is laden with political judgments. The issues for the committee are affordability and what the cash costs of PFI are to the public sector; the way in which PFI distorts the planning process; and fragmentation. We need to examine the relevant accounts to see how they translate in practice for the public purse.

We are not the Audit Committee so we are not considering past PFIs or trying to make judgments about those that did not work.

The Finance Committee is, however, involved in bringing public expenditure to account.

We are, but our focus is on trying to identify how the system works or could be made to work and what its underlying economics are. We are more focused on the comparative claims for PFI against public-sector procurement.

I want to pick up on risk transfer. Are you saying that risk transfer is always a paper exercise and that there never is any real risk to be transferred? From your research, can you identify different categories of risk transfer in different kinds of scheme?

That comes down to availability of data. One of the striking things about the Treasury and Andersen reports is that, although they considered 29 schemes, only 17 of those schemes had any data on risk transfer. Most of the so-called transferred risks were construction cost risks. There were no other risks. However, risk transfer adds 30 to 40 per cent to the price of a PFI project.

When we talk about risks, we should be dealing with probabilities, not uncertainties. First, in PFI there is no real risk methodology. Secondly, the proportion of risk that is transferred seems to vary enormously across schemes, but no account is taken of what the risks are, how they are justified, how they are quantified and how they come out so conveniently close to the price difference between the PSC and the PFI. Risk assessment is not a science in PFI. In this case, it has been used to disguise the true costs of PFI.

I wonder whether the construction industry has—from the receiving end—a view on risk.

I cannot look back at what has happened in various transactions because I agree with Professor Pollock. Much information is not disclosed, so it is therefore impossible to take a view. We are involved in projects only as members of our organisations or, as I am, as a lawyer for example.

I have spent 20 years in the construction industry so—talking from the industry's point of view—there are distinct transfers of risk taking place from the traditional form of contracting to the present form. The reason for that is simple; the contractor is dealing with a special purpose vehicle company, which has the assets and revenue stream of what will be constructed, for example a hospital.

The discipline that has come into the construction industry means that buildings will be finished on time and finished at the cost that is stated in the construction contract. There will be no variations. There is therefore clear risk transfer, from the point of view of the construction industry.

On the operations side, risk transfer is difficult to distinguish. In the National Audit Office reports, various examples are given in which there is more efficiency through the private sector—instead of the public sector—running certain public sector assets. I am not an expert on that so I would not like to comment. However, there are distinct and beneficial transfers of risk to the construction industry.

I have questions for the two academic witnesses. I will start with Professor Pollock. Claims are being made that projects are being delivered early. You have given us a fair bit of number crunching. Has any public-domain academic work been done that tells us what the opportunity costs are of not delivering or of delaying delivery of a service? I am not talking about construction costs or spending money on a building or a road. I want to know about any academic work that has been done on delivery of service. That information would influence what the committee is trying to achieve.

I will focus on hospitals. First, cost and time overruns in public procurement are running at about 8 per cent at the moment—that is for public procurement, not for PFI.

Secondly, cost and time overruns could be dealt with in a contract. It is not necessary to add 30 per cent to the cost of the contract. Overruns could be handled and dealt with through penalty clauses, which is what has been done traditionally.

Thirdly, many of the cost and time overruns are not the fault of the public body or the builders. If one examines the Audit Commission and NAO reports and evidence from select committees, one can see that in some cases the overruns were due to VAT changes or the stop-go nature of public funding. In general, the overruns were the result of circumstances that were largely outside the public body's control.

That does not answer directly the question about opportunity costs, but does so indirectly. The opportunity cost is that we pay an awful lot more for a PFI scheme—sometimes adding 30 per cent or 40 per cent to the price—to deal with an 8 to 10 per cent chance of a cost overrun.

You just admitted that you have not answered the question that I asked.

I have answered the question. Under PFI, you add an awful lot to the cost for risks that might never materialise.

That is not what I meant. I was talking about public service and whether any academic work had been done on, for the sake of argument, the benefits of early delivery of health care? If you can deliver better health care, people can get back to work and be economically active—quite apart from the personal benefit of not being ill. Has any work been done to show that it is good for the economy in the round, including the social economy—as has been claimed by both the Conservative Government and the current Administration—to deliver services early by using PFI to deliver projects earlier?

You forget that, in the case of a PFI hospital, you are removing 30 per cent of the beds and 20 per cent of staff budgets. Those costs are enormous, but that is not a benefit of delivering a service early, it is a benefit of delivering a much smaller service, decreasing access and quality and having a major effect on the work force.

We know that that approach creates an apartheid system that results in a work force of two, three or four tiers, which has real cost implications. Mr Davidson is right that not enough research has been done into that, but that is because the matter is not currently politically popular, although it is important.

The reduction in the number of beds has been negated by the medical profession, because they are treating people more quickly. That is a separate argument.

It is not a separate argument. One of the effects of PFI, because of the affordability issue, is that 30 per cent of beds are taken out in each area. That must be paid for from the revenue budget—which pays for patient care—and it has been shown that that distorts the planning process at local level. The effect of that is that smaller hospitals are delivered at a higher cost. Another effect of many hospital PFI schemes is a reduction in the clinical staff budget. When there is a new claim—a higher cost of capital on the revenue budget—the only thing that can go is the staff budget.

I will give the committee an example. On average, since capital charging was introduced, hospitals spend about 8 or 9 per cent of their revenue budget on capital charges. In a selection of schemes, when one considers the effect of PFI, one sees that such hospitals are now using 20 per cent of their revenue budgets to pay for PFI—that is the cost of capital and availability fees. That money is spent before a single patient has been treated. Where does the money come from unless there is a huge increase in revenue? The money must come from services and the 20 per cent reduction in staff costs.

I have two quick points to make. I am not aware of any academic work in the context of PFI. There is a huge amount of literature in a similar area, which relates to the value of improving road conditions. Of course, it is terribly economic—in the sense that all one does is add together the economic valuations—and it is still a hotly disputed area. The research that David Davidson is interested in is doable, although I am not aware of anyone who has done it; it would need quite a lot of information.

I am sorry to drag you back to the discount rate, but it is important. We do not know what the cost of a PFI project is unless we have the right discount rate. Professor Pollock pointed out that projects seem to be insensitive to discount rates. That was mentioned in the Andersen report. However, basically, everyone is using the same discount rate for everything that they discount and then changing that number from 5 per cent to 4.5 per cent to 6.5 per cent—individual discount rates do not change things. The reason why they do not change is that a different discount rate should be used according to what one is discounting—that is, according to whether we are discounting apples or oranges.

If one wants a public sector comparator for the cost of building a road, that is a matter of having a contract with McAlpines or another company over a period of two years; there is a relatively low cost cash flow. If one compares that with the payment that is made per lorry and per car over 40 years—this is a different animal altogether, with a different discount rate—that figure will be alarmingly sensitive to whether the discount figure is 6 per cent, 5.5 per cent or 6.5 per cent, because the contract is over 40 years. Furthermore, the figure should be different in a Ministry of Defence type of project, in which there is no individual purchase element and therefore not much randomisation compared with building a road or something of that sort. I understand that the Treasury is now including those sensitivities in the green book. As far as I am aware, nobody has taken apart a PFI project and used different discount rates for different cash flows, which any textbook anywhere will say is what should be done. Until we do that, we will not know whether PFI projects are more or less expensive than the PSCs.

Can all the witnesses tell us whether they agree with Professor Grout's conclusion in evaluating the PFI/PPP project discount rate? When we had this debate last year, the Government in Scotland did not agree with that conclusion and the Treasury does not appear to agree with it, because it is easier for the Treasury and the Government to deal with things centrally. Do the witnesses have a view on different rates that could apply in different sectors?

I will make a general comment. When bidding for those projects, one is faced with accounting procedures and the Treasury rules. One of the big problems is that, as the business case and the negotiations are gone through, one must fit in with all the different accounting procedures and rules. If you find that you are moving to one side or the other and are falling outside the rules, you must find a way of coming back in.

When we were doing PFI projects back in the 1980s, we had to use the same procedure for the Ryrie rules. When John Major was Chief Secretary to the Treasury, he abolished the Ryrie rules; nobody has questioned that since. We should consider the Treasury rules and question them. That might solve some of the problems. We must ask what is important, what is not important and what can be changed. Can we change the definition of what is on and off the Government's balance sheet?

The procedures here are different from those that are used in some other countries. The rules are surely not cast in stone. They cause problems, so why cannot we consider them afresh? Professor Grout's paper makes it clear that there are uncertainties. If you are bidding, you must follow the rules. An inquiry such as this could examine the procedures again and re-evaluate the worth of some aspects of accounting standards.

On confidentiality of contracts, would the construction industry be prepared to open up the contracts after they are signed, so that a proper assessment could be made of value for money and the other measures that indicate whether the public get an appropriate return for their money?

I would be interested to hear from the witnesses from local government about the figures that Professor Pollock, among others, has produced on the significant impact of PFI contracts on revenue budgets. In the health service, the cost of such contracts has moved from 8 per cent or 9 per cent to more than 20 per cent. Could the witnesses from local government indicate what impact that might have on their revenue budgets?

Confidentiality is a major problem. It means that many matters that are resolved in one set of negotiations are renegotiated in another set of negotiations. That was apparent from a series of NHS trust projects, when a series of hospitals were being negotiated at the same time, all with different advisers and different forms of contract. That led to a complete waste of money.

There is now an easier process, following the standardisation of 70 per cent of the contract conditions, and there is no reason why that process cannot be developed. Many matters that are now kept confidential in construction contracts and project agreements arise in project after project. Why are they kept confidential? I cannot answer that question. The confidentiality agreement is in the documentation that is produced by the public sector. Before someone bids for any of those projects, they must sign up to the confidentiality agreement and agree not to release confidential information. It is up to Government departments to decide to what extent they want to retain confidentiality in their agreements.

In the United States, agreements go into the public domain and are registered as soon as they are signed. The day after a deal is concluded, the successful bidder's competitors can view all the documentation, including the confidential commercial terms. That is a fundamental point. It will be impossible to undertake any studies until a way is devised to make some of the information available at least to people who are undertaking research to inform decisions on whether the procedure is successful.

From what you say, I take it that the construction industry has no objection to making that information available?

I cannot speak for the construction industry. Our members also include the major banks, the equity investors and Government departments such as HM Prison Service, the Office of Government Commerce and HM Treasury. I am not really here to speak for the construction industry; I am here to speak for the IPFA in general.

What information essentially remains confidential? You have highlighted the fact that, in the United States, nothing appears to be essentially confidential once the contract has been signed. What do the people whom you represent regard as being essentially confidential? I understand the argument for keeping innovative design solutions confidential. Other than such things, what is essentially confidential?

The only relevant matter that would be confidential in any agreement is consideration. The way in which the contract terms are interpreted later depends on what happens during the contract negotiation. During the course of a contract, it could not be known whether a contract term was causing a problem. The only truly confidential information is in the consideration. For example, how much is in the construction contract? Is it £3 million or £3.2 million? If that information was released with the contract terms, a competitor would be able to calculate—based on the terms and conditions—what return the contractor would receive on that contract.

That happens in the United States.

Yes. It happens on acquisitions, mergers and some agreements.

It does not happen on general contracts. Let us move on.

We have talked about the benefits that flow from better project management and from more exact specifications. It has been suggested to the committee that some types of contract are not suitable for PPP and that IT projects have a propensity to be disastrous, regardless of the system that funds them, because of the difficulties of specification. I am not sure whether we should conclude from that that we should shove all the disasters into the public sector; however, the suggestion is that PPP is not suitable for IT projects. Professor Grout also seemed to suggest that some waste disposal contracts are not suitable for PPP because they are so simple that there would be no benefits from more exact specification. Are there types of contracts for which—for whatever reason—PPP is not suitable? Mr Blackburn might want to answer that question, as Dumfries and Galloway Council has recently concluded a PPP for waste management.

I think that my colleague was talking about waste collection rather than waste disposal. Waste disposal is an activity that carries far more risk than waste collection, which is a fairly simple routine activity.

A question arose about the potential effect of PPP projects on other activities in the budgets of councils. The APSE is concerned that, when there is a funding gap in a PPP project, one of the mechanisms for trying to close that gap is not simply to pass over the buildings and the construction side, but to bundle up some of the services for 25 or 30 years on the operating side. Geoff Haley said earlier that the profit stream tends to come, by and large, from the operational side rather than from the design, build and finance. That is not surprising.

The question is; why are these services being bundled into the project and how are the profits being achieved? There is good evidence to suggest that, far from gains being created through productivity efficiencies, the gains are coming from labour-saving efficiencies. That is to say, they are achieved through reducing the terms and conditions of the staff who have transferred with the project. There is fairly good evidence that that has happened.

We are moving away from the era of compulsory competitive tendering, which brought fairly big changes to the most vulnerable employees in the workforce, such as females with part-time, part-year contracts. We are concerned that those people will suffer again as their terms and conditions are eroded still further as they are transferred with the PFI schemes. That raises a wider issue that is to do with anti-poverty action and social inclusion.

When Scottish Gas and other such organisations moved away from the public sector and became stand-alone entities, conditions did not necessarily ease as new management regimes were brought in with a remit to bring down costs. Is the situation partly to do with greater management efficiencies leading to different working methods? Do you have evidence about that? Many claims are made on the subject but I would like us to get beyond assertions. I see that Professor Pollock would like to speak.

My colleague, Dr Jean Shaoul from Manchester, has done some work on that issue. I can send the committee a table that details the effects of privatisation on utilities such as gas and telecommunications. There has been a 20 per cent to 30 per cent reduction in the work force and a commensurate rise in shareholder dividends. In labour-intensive industries such as health, education or the postal service, profits and savings can often be made only at the expense of the work force. Of course, that has wider implications for society. Good data are available on the impact of privatisation on the work force, but there are not enough good data recording what is happening on the ground.

Are you suggesting that public sector organisations employ more people than they need to?

In health and education, we know that staffing levels are the key to quality, but we also know that compulsory competitive tendering and best value regimes—certainly in England—do not consider the quality of the service.

With respect, you are swinging from one organisation to another in your explanation.

I do not think so. One of the things—

Let me finish. You used the gas and telecommunications utilities as an example and said that their work force dropped by 20 per cent to 30 per cent when they moved away from the public sector. Are you saying that, when they were in the public sector, they existed merely to provide employment irrespective of outputs? Do you think that that is a justifiable reason for employing people? You are using two different organisations.

I did not intend to. The important question that must be asked is; what has been the impact on the supply of those services as a result of shedding some of the work force? What is happening in terms of responsiveness, maintenance, repairs and quality? However, there has been no research into those issues because such research would not be politically popular.

Do you really think that empirical evidence would show us that the electricity supply of this country has deteriorated because of changes that were made in the past 10 years?

I am not an expert on electricity, but I am an expert on health and social services. In those areas, we are aware that efficiency savings, reductions in revenue and the other effects of privatisation have caused a major reduction in the volume and quality of services that can be offered to older people—or to the rest of the population. There is good, documented statistical evidence of growing unmet need and decreasing provision.

It is wrong to assume that all segments of the work force are affected. The market also comes into play. There is evidence that highly technical specialists have not suffered and that their terms and conditions might even have improved. The Association for Public Service Excellence is concerned for people with low skills and earnings who are down at the bottom. We are trying to sustain decent employment within the public sector.

We foresee terms and conditions being eroded by transfer, because there might be only two years protection under the Transfer of Undertakings (Protection of Employment) Regulations. That protection will not endure for the length of those extensive PFI contracts.

It is wrong to equate efficiency with cost. In a simple operation such as cleaning, 100 hours paid at £5 an hour will cost £500. An alternative supplier could give 120 hours for £4 an hour, which is less efficient in productivity, but costs less. We need a more differentiated analysis in that area of the effect on employment numbers. We must also examine the subcontracting policies and wage rates across different parts of the work force.

That highlights the need for evidence. Perhaps Professor Grout would like to say something about evidence.

I have a couple of points. Richard Blackburn was absolutely right to say that we mean waste collection rather than disposal. Waste collection is such an unusual experiment in assessing the effects of competition and the introduction of possible private supply because it is one of the few cases for which we have masses of independent observations.

I emphasise the point about distinguishing between employment numbers and employee conditions. Employee conditions in the utilities are no worse than previously. Indeed, if one goes high enough in the organisation one's conditions seem to improve significantly—right up to the top.

Employee numbers have fallen more than anyone would have thought. In addition, quality conditions are important and the output quality of those industries has risen amazingly. For example, if one assesses the electricity industry by such criteria as blackouts and brownouts or other statistics, or assesses the telecoms industry by the time it takes for a call to be answered or to fix a telephone fault, one finds that it is absolutely another world from 1983. The water industry regulator has battled with the European Union and the Environment Agency, claiming, at the most recent round of price controls, that because the required standards are so high he was unhappy to fund them. It would be hard to argue that the standards in any of the core utilities have not improved enormously as employee numbers have fallen. The gains have been made through the reduction in employee numbers rather than by a change in employees' conditions—that is a different issue that must be treated separately.

I am trying to reconcile that with what Professor Pollock said. Is it the case that there are some work areas in which one can have, for example, a big input of technology or a system of economic regulation because there is an income stream and a competitive framework which allows the logic that you outlined to work? How does that apply in other sectors such as health and education, in which the opportunities or capacity for technological change have historically been fewer?

That will differ from sector to sector. What I said was in response to the question about what has happened to the output quality in telecoms and the other utilities. I take your point that those matters are always sector specific. I return constantly to the point that PFI rules are one-size-fits-all rules. Until we sort that out, there are many questions that we cannot answer.

It is important to distinguish between technological advances from 1982 to 2000 and what is actually happening to the services. The telecoms, water and electricity industries are now going into major financial crisis. They also have a history of underinvestment over the past 10 or 20 years. Most of those industries are investing less than they ever did at any time when they were nationalised. The literature on that is good and I can pass it to the committee. If we have a discussion on the utilities, it is important that we examine them in more depth.

That would not apply to the water industry in Scotland.

It might be different for Scotland, but in England, the telecoms and water industries have problems.

David Davidson covered many of my questions. Professor Pollock said earlier that the real issue is cash costs and David Davidson asked about opportunity costs. A client might receive a facility on time and to the design that is specified and the maintenance regime of the facility might be such that the users are not demoralised by corridors that are not maintained or by grimy wards.

Surely there are issues other than the cash costs. Professor Pollock made the point that we pay a lot of extra money for improvements. How do we quantify those improvements? She gave the example of new hospitals that have 30 per cent fewer beds, but people in the medical profession might say that, because of medical advancement, new hospitals require fewer beds than older ones. My area has a new hospital that is a world leader in digital X-ray. That hospital was not there before, but it is now. You said that there are fewer beds, but new facilities involve significant advances.

Why could those advances not happen with public facilities?

There might be a reduction of beds in public facilities.

Yes, but the introduction of the capital charging regime and the switch to debt finance in the NHS had major consequences. The evidence shows that the switch precipitated downsizing in the health sector. That was exacerbated by PFI, which makes an additional claim on the revenue budget. Not many doctors would condone the decrease in beds, because we have a major crisis. Scotland is doing better than England, but the national bed inquiry said that no more beds can safely be closed.

Do you have empirical evidence that hospitals provided by PFI produce less of a service to the community than other hospitals?

Yes, we have lots of empirical evidence. Hospitals have opened in Durham, Carlisle and Dartford and the effect has been to reduce the volume of services provided and to reduce the case load that can be treated. Another issue is the diversion of funds intended for other NHS capital schemes and resources. Because new hospitals are expensive, they suck in funds in the form of subsidy from the Treasury—capital budgets that are intended for the rest of the NHS—and other services in the area must be closed. For instance, in Dartford, services for people with learning difficulties and mental illness were not reprovided because of affordability problems. When I walked round the new Edinburgh royal infirmary I learned that the PFI has major affordability issues and that the board is in financial crisis. Those issues must be examined. Without major injections of revenue, new projects will come at the expense of other services.

It is claimed that the NHS is treating more patients and carrying out more operations. However, you said that there is evidence that new facilities provide less of a service and cater for fewer people.

In order to make projects affordable, downsizing takes place up to four or five years before new hospitals open. Dr Matthew Dunnigan, who is a retired consultant in Glasgow, recently published a large two-volume report on Edinburgh, which shows that Edinburgh's capacity and its ability to treat patients are falling. Edinburgh is at saturation point and is turning off trade, as one of the chief executives described the situation. Patients cannot get in anywhere and they are not being treated. That is a natural experiment. The rest of Scotland has not had such serious bed reductions and case loads are rising.

I want to pick up on a point made by Professor Pollock and others. One of the effects of PPP is that a significant proportion of revenue budgets is tied up over the next 30 to 40 years. Although it could be argued that that money is being paid for the profits that should have been made during the construction stage, the assets will be maintained over the period of the PPP. That situation is forcing councils or other public bodies to make the hard choices that they have not been making for the past 30 years.

The reality is that lots of councils have not been maintaining their assets. That is why we have crumbling schools and hospitals. Therefore, although it causes the councils pain to pay whatever percentage of the capital budget they have to pay, they should have been doing it anyway, regardless of whether they are now being forced to do so by PPP.

There are many issues about longer-term constraints and the lack of flexibility of such lengthy contracts. Things change over time, as we are all aware. Another point is that we are making decisions today that we have to live with in future. That ties in to some of the points made earlier about local government not being shown the business case.

Is not it the reality that if you build a road, a hospital or a school, you know that you are going to have to maintain it throughout its lifetime? Politicians, however, have made the easy choices. It is easier to defer maintenance and spend the money on something that the public can readily see. The proponents of PPP allege that it is forcing the councils or public bodies to make those hard choices and do the maintenance that is required.

The other aspect of it is that, at present, local government does not really have a choice. If it does not use the money that is available through PFI to bring such projects on-stream, it has no other choice. That is a major factor in why some of those schemes are happening at present—there is nowhere else to go for the finance.

It would be interesting to see the data that separates capital costs from maintenance costs. Most of the schemes do not show how much is being sacrificed on maintenance costs.

The second issue is about the inflexibility of the contracts and how we cannot change our minds and decide one year to treat patients rather than pay maintenance if there is, for example, a shortfall in funding. That maintenance money is ring-fenced and guaranteed before a single patient can be treated or pupil can be taught.

Another issue is that we are undoing something very important by using PFI. When markets are introduced, public services lose some of their ability to cross-subsidise or risk pool. The essence of public services is risk pooling and cross-subsidisation. The expensive treatments cross-subsidise the less expensive. When services are unbundled, important efficiencies are lost, and the key efficiency is cross-subsidisation.

A good example of that is the current situation in the postal service. It has lost its monopoly status and its ability to cross-subsidise the less expensive parts of its business with the more expensive. That has grave consequences because the risks are then passed to individual users and society at large.

As well as trying to drive efficiency, we must ensure that we have appropriate quality standards. What consultation ought there to be with the end users of the services, not just the facilities?

As well as efficiency, there is the question of equity and fairness. There might be circumstances where some sectors of society have dropped out altogether. Professor Pollock was suggesting that there is evidence that particular parts of the health service are being sacrificed because we can no longer afford them. How equitable and fair are the PFI/PPP procedures? How do we build equity and fairness into those procedures? Can you give us any examples of consultation with end users and of steps that have been taken to preserve equity and fairness as part of the process?

As far as consultation with end users is concerned, there is very little empirical evidence in Scotland. A report was produced by the Public Private Partnerships Programme—or 4Ps—in England and Wales which examined the design process in PFI schemes and the consultation that had been carried out with the various stakeholders involved in that process. The outcome of its considerations was that the process was critically flawed. The buildings being produced and the services being provided in those buildings were, in many areas, not fit for the purpose or the customers for whom they were intended. The customers were not involved in any consultation process on the design and procurement of the premises or the function in question.

Was that the case before? I know of a lot of public sector buildings that are far from fit for purpose. Somebody got it wrong somewhere along the line previously.

Absolutely—that is a fair comment. From all the views that have been expressed here this morning—I do not want to put words in my colleagues' mouths—it is clear that there are critical problems with the process. There have been critical problems in the past, which we are facing again. Surely we must take the opportunity to—

Can you give us some specific examples, perhaps saying how the process might be altered in order to accommodate the needs of end users?

There is a simple comment to be made on the whole PFI procurement process: it is extremely complicated. As has been said previously, it is necessary to jump though various hoops to make progress with it. In Scotland, and indeed in the whole of the UK, we must meet the various partners—the private sector, the stakeholders, the customers, the councils, the NHS and so on—and simplify the process, which currently does not meet all our goals—

But the people you mention are not actually the end users. You are talking about the players in the game. I want to know whether you have any examples of how end users are being consulted as part of the process.

Yes.

If the current arrangements are unsatisfactory, what arrangements ought we to have for the future?

We have a process in place now, although we do not have legislation for it yet in Scotland. You will be familiar with it: it is best value. It exists in Scotland and across the UK and it is legislated for in England and Wales. Built into that framework is a duty to consult the customer, the community and the stakeholder. PFI/PPP decisions are procurement decisions, which are made at the end of an options appraisal exercise, or they should be. That options appraisal exercise should include detailed consultation with the end users. Perhaps the question should be whether that is the reality—it is not.

I have been reading about consultation with end users. Previous witnesses to the inquiry have said that, with the introduction of PFI, there has been a much more rigorous process of specifying the end requirement, which has necessitated consultation with all sorts of people, including end users. That process came in with PFI, but it did not, and does not have to be done only with PFI. Such consultation—indeed proper project management and proper project specification—could have been brought in by the public sector any time over the past 20 or 30 years.

If PFI/PPP were taken away, does any driver remain to ensure proper project management and specification in a building project or in providing a service?

If I may first return to an earlier point, best-value legislation will be in place here in Scotland next year, which will put a duty on local authorities to carry out this exact process.

To answer Elaine Thomson's point, firms must be seen to carry out, and must show empirical evidence of, consultation with their stakeholders and customers.

But surely the introduction of resource accounting and budgeting, whether you agree with it or not, would also provide a driver to ensure that units are properly maintained. If they are not essential units, they will be disposed of. That is the natural conclusion. There are other mechanisms, not involving PFI, which are effectively intended to achieve the same end, whether you believe them to be fundamentally flawed or not.

I have two points to make. Let us consider hospitals. First, it has recently been claimed in the papers that PPP hospitals are responsible for the shortage of beds. That is complete nonsense. When we negotiate a PPP hospital, we are constantly asked to change the specification to match the finance that is available in the business plan. If a trust decides to reduce the size of a ward by five beds, we do that. If we are told to reduce a service, we do that to match the finance that is available. When they sign PPP documents, trusts and the health department know exactly what they are getting.

Secondly, we moved from the private finance initiative to PPP because the Labour Government wanted a true public-private partnership, but we carried on using the same documentation. At the moment, concluding a deal involves very complex legal documents—financing documents, project documents and construction documents. A number of chief executives of national health service trusts have told me that they find those documents daunting and that that affects how they run the hospitals. At times they have to examine a document before they can decide what they can or cannot do.

A couple of chief executives have said to me that they would like NHS trusts to have some involvement in the special purpose vehicle companies that are set up to run hospitals. Some of them would like to have equity in SPVs. They would like to have a seat on the boards of SPVs, so that when SPVs meet they can be involved in discussions about those hospitals. We need to consider how we can introduce partnering arrangements on top of the contract documentation that already exists. We want to have the sort of informal partnership or alliance structure that many oil companies in the North sea have and that works satisfactorily. We need to develop that concept, because there are concerns and anxieties in the public sector. If a project is designated as a public-private partnership, it should be a true public-private partnership.

The Government is also learning from the hospital side. In the two previous hospital projects that were put out to tender, it opted to appoint advisers. In the case of St Helens and Salford, the Government was seeking hospitals that would provide complete primary care and intensive care. The projects that go out to the private sector for bidding were designed to provide complete care for the community. Such an approach involves consultation from day one, as well as appointing advisers. In the case of St Helens, there has been consultation with about 12 different organisations involved in health care in the area. The process embraces those organisations, the local NHS trust and union members, who have a seat on the committee and are able to express concerns and to see how the private sector reacts to those. That is a very positive step.

I want to return to consultation, as we are in danger of confusing two parts of the process. Surely PFI or PPP is a means to an end, not an end in itself. Public bodies should take soundings from all stakeholders to establish what they are trying to achieve. Having decided that, they should consider what options are available to them and what the best means are of achieving their aims.

One problem with PFI projects—Geoff Haley has already mentioned this—is that once we have moved from the first stage of consultation, planning and setting objectives to the means of delivery, a scheme can change. At that stage, schemes tend to be technocratically driven, with finance issues becoming more important. Once people have embarked on a scheme, they do not want it to fall through. The reason that there are concerns about consultation is that schemes tend to gain their own momentum and to become ends in themselves, rather than means to an end.

The question that I would like to ask is a bit off the wall. Are we moving towards setting up an independent body to assist the public sector, the private sector and the end user to decide what projects are suitable for PFI/PPP? Such a body would also evaluate risk—decide where the risk lies and how to cost it—and examine the service provision that comes out of the other end of the sausage machine. Is there any merit in such an approach? Are we moving towards that? There appear to be many tensions in the marketplace.

The Association for Public Service Excellence believes that there is a bit of a rush at the moment to get into PFI projects without taking time to consider them fully. That is not surprising, given that the financial and time-limited attractions are so great that many councils—and, no doubt, other public bodies—are using phrases such as, "It is the only show in town" and "We have to catch the boat before it sails" and so on.

The Association for Public Service Excellence argues that, as part of the process, a more careful examination of PFI needs to be undertaken—all the options need to be looked at. Some of the debate this morning seems to have gone along the lines of "PFI good" or "PFI bad". It is probable that, like everything else in the world, some PFI will be good and some will be bad. We should be able to differentiate between the two.

It comes back to the point that David Davidson made earlier about opportunity costs and the effects on other budgets. Although councils may not see it, they are being offered severe opportunity costs at the moment in the form of large subsidies to enter PFI schemes. Those multi-million sums could otherwise have been used to improve public services.

What kind of leverage is used to get councils into PFI schemes? Is the Government putting up specified sums as encouragement for authorities to go into PFIs, including education PFIs? What is the volume of that encouragement and how does it work?

Scotland's local authorities are split into 32 councils. Not all authorities will have the same level of expertise to develop specifications or to work up schemes. Should authorities pool specialist resources, or can another mechanism be introduced so that if councils go ahead with PFI schemes they meet the requirements? Does every local authority have to go through the same slow and painful learning process?

I cannot answer that, as I have not been directly involved in framing PFI projects. On the supply and demand side, £500 million is available to support submissions for the education PPP schemes that had to be in by 14 December. Applications for those schemes are oversubscribed by two or three times. I suspect that, for many of those councils, the applications were driven by the consultants who were brought in to assist them with the process, rather than by in-depth analysis carried out in-house. From what I have seen of PFI/PPP schemes, it would be useful to have mechanisms put in place to assist the process.

I want to ask about the cost to councils of the competitive bidding process to get access to the monies that you mentioned for schools projects. The process is grossly oversubscribed. A lot of time and effort has gone into it and a cost is attached to that.

One set of figures has been published. Yet again, they do not apply to Scotland—there is a lack of serious reporting in this area. The Audit Commission down south published the report and it examined consultancy fees for the NHS and for other public sector PFI projects.

The report identified that PFI delivered value-for-money savings of not 17 per cent, but between 3 and 5 per cent. However, those savings had to be offset against consultancy fees for the specialist advice on the projects. Those fees ranged from 2.4 per cent to 8.7 per cent of the capital value of the projects, giving an average of 4.1 per cent. That figure did not include management costs within the public sector organisations. Consultancy fees alone negated almost all of the value-for-money savings.

Has anybody done any work on what the costs are to the organisations?

Do you mean the councils?

Yes. The costs must be considerable. As Richard Blackburn rightly pointed out, the applications for school PFIs are oversubscribed by several times. Councils will have borne a lot of costs, which they will not be able to recover. That is at the expense of other services.

The idea for PFI did not drop out of the sky—the world did not just start. Councils have always spent an awful lot of resources year on year making unsuccessful capital bids. We have to keep some balance. That is how the world has always operated.

I want to come back to Geoff Haley on how we evaluate risk. Who decides the risk and is there a fair model? Is there any evidence to show that there is uniformity in how all the different players assess risk, who takes it and how it is valued?

The whole point about risk is how it is priced. I will give an example. When I acted for a consortium on the channel tunnel rail link, I produced a risk analysis that came to 258 pages. We had to go through all those risks and decide which risks could not be accepted by the private sector—and so had to go back to the Government—which risks were insurable and which risks we would take. We then priced the risks that we had decided to take. One has to do that calculation for every project. We work out the cost for taking all the risk and then tell the public sector organisation that if we take a certain risk the cost will be £3 million higher. We ask the public sector organisation whether it is prepared to take back that risk, in which case we knock £3 million off the price. It is as simple as that.

When the local authorities join the PFI bidding process, from day one they will need a considerable amount of help from a central body, so that they do not have to learn the process. Dealing with NHS trusts, I found that it took a good six months before a trust would understand what the process was about. That is six months of time wasted negotiating clauses that the trusts did not agree but that they had no chance of not accepting if the funders were to finance the project.

We have now established in PPPs what is acceptable to funders. If the funders will not fund something the whole project dies. There is a parameter of clauses that must be included in the deal. We now have fairly good standard clauses that are acceptable to both public and private sectors. Unfortunately, those clauses do not seem to be emerging in certain new sectors. It took us a good three and a half years to establish them in the NHS and they are now well established for roads and prisons. Unless something is done about it now, individual local authorities will go down diverse routes and we will have ridiculously high bidding costs as a percentage of capital cost. We need shorter bidding programmes, more standardisation and greater simplification.

Are you saying that we need a P book—a core procedural document to which people can refer as a guide? The game is changing from PFI to PPP and there is a different set of parameters. However, I presume that the valuation of risk is fairly standardised.

Yes, it is.

In the same way that the valuation of risk is standardised and so the cost of procurement becomes less, the experience must have led to the discovery that some of the risks were not real risks at all and can be written down. Conversely, there may be some risks that had been underestimated. Can you give us evidence of risks that are no longer regarded as significant—by the finance, construction or operating people—and so will lead to reduced private sector bids, and can you give us evidence of risks that have gone in the opposite direction?

I will give you an example. If a sector is successful—such as roads or prisons—contractors are keen to get into the sector and funders are keen to invest in or lend to the sector. The Prison Service has had several successful projects. In the last project, which was issued recently, the Prison Service insisted that the transfer of risk of riot went to the private sector. Up to that point, the risk of riot had not been accepted by the private sector. It was a Government risk; all the costs arising from a riot were uninsurable in the insurance market, so the Government would pick up the cost. The private sector consortium is now comfortable with taking that risk. In the last project, that risk is in the package and it is being accepted by the bankers. That is an example of where confidence has changed the risk profile of projects.

This question follows on from David Davidson's. A couple of years ago the committee discussed the role of Partnerships UK. I understood that part of its role was to disseminate best practice and act as an intermediary body for organisations considering PFI/PPP. Is it carrying out that role? What impact has it had?

Partnerships UK followed on from the Treasury task force. The task force produced a number of know-how publications, which have been useful and are classed as the bibles, but very limited standardisation has been introduced. The previous major standardisation was for the health service. The Public Private Partnerships Programme is working on some standardised documents for local authorities. I am not sure how far down the road it has got with those, because it has a small staff. It would be beneficial to have a larger staff in such organisations producing documents for use by other bodies. Otherwise, each individual body must produce its own documents, which is very costly.

I want to clarify a point on 4Ps. Whether it is time to wind the programme up is currently under review. It has been transferred over to the Improvement and Development Agency, which is based in London.

Many PFI/PPP contracts are for long periods of time. I understand that certain contracts, such as information technology ones, are for much shorter periods of time. What restrictions are there on flexibility within a contract? Is it possible to build in flexibility? Over a long time, the way that one wants to design and provide services may change significantly, because of technological innovation or for other reasons. Does PFI/PPP reduce innovation and flexibility?

Has PPP as it is now—compared to PFI as it was originally introduced—improved the ability to have flexibility over long periods of time?

The basic principle of PFI/PPP, as against conventional ownership, is not that PFI projects do not allow for flexibility—it is who has the property rights that influences how one deals with the need for flexibility when it arises. These problems are endemic in such models. All the contracts that are written for student halls of residence mention how many microwaves there should be in students' kitchens. Those contracts are for 25 years. Consider how many microwaves would have been in the contracts that were written 25 years ago—none at all, because nobody would have thought that that was appropriate. Incentives exist in the private sector to produce good ideas, but the private contractor has to sell the ideas to the public sector to persuade it to respecify the system.

So there is a trade-off: incentives are needed for the private sector to come up with ideas, but those ideas may get dulled if the private sector cannot implement them without getting an appropriate price from the public sector to respecify the contract. I do not know which contracts could be easily renegotiated and which ones could not, but I cannot see any problem with having new specifications for PFI and PPP projects. However, private contractors may have problems because their incentives are slightly dulled. They are in a weak negotiating position.

The public sector, too, is in a weak negotiating position, because it is tied to one provider. We can consider the example of a school project. Population trends show that populations can change a lot in 10 or 15 years, and schools sometimes have to close down, as we have seen in a number of areas. However, if the population changes when there is a 25 or 30-year deal, it may be very difficult for the public sector to get out of a contract without incurring very high costs.

Stability has to be part of a PFI project. A large sum of money is paid up front and the project is paid for over a long period. If there is to be national guidance on the appropriateness of PFI schemes, perhaps it should recommend that an activity that is liable to be subject to substantial technological change, or change in demand, should be closely examined before a decision is made on whether PFI or some other means is the best way to proceed. If there is going to be a lot of change, PFI may not be the most appropriate way to proceed.

What criteria might you use? One issue that has come up is the balance of capital repayment charges for the PFI against the available revenue budgets—the crowding-out issue to which Professor Pollock referred. How do we measure affordability of a capital project in terms of its long-term impact on revenue?

One criterion might be the total debt that is generated, but are there other criteria that might influence whether a public authority should embark on a PFI or PPP scheme as opposed to using other methods? Questions have arisen over whether the public sector comparator as it stands is an adequate test.

Contract theory says that risks should be allocated to those best able to bear them—and clearly the public sector is always best able to bear them, because the state does not usually insure itself against risks. However, other risks arise—the political risks of devolving responsibility and accountability for public services to the private sector. We are talking about markets. We know that there is nowhere in the world that delivers public services on the back of the marketplace. Evidence is now accumulating on the effects of PFI and of using markets. We have heard about primary and community care in hospitals being squeezed out in favour of acute services because of affordability problems, and we know about the segmentation of the risks of the work force and now the population.

I appreciate that the terms of the committee's inquiry are quite narrow and are limited to what will work with PFI. However, I put it to you that the political risks could be so great that you may have to consider going back to direct Government grant and proper Government capital funding. You perhaps should not limit yourselves in this inquiry.

I sense that things will unravel in the next 10 or 20 years. We are already seeing some unravelling with Railtrack. In Australia, because of the effects of privatisation, the Government had to step in and take La Trobe hospital back into public ownership. In the United States of America, where private markets have delivered health care for a long time, the effect has been that 50 million people carry their own risks and costs of care.

If we really believe in collective social solidarity—which has been Scotland's framework—we really have to consider the consequences of PFI, what the political risks will be, and whether we should think about reinstating public ownership and control, under proper frameworks for democracy and accountability, which have perhaps been weak.

That is a legitimate point of view, but what we really have to do is identify the parameters of PFI. It is only if we understand the economics of it that we can begin to calculate some of the political issues that you have referred to. Those issues will be decided not by the committee but by the Parliament and the Government. Our specific interest is in identifying the economic framework within which those decisions can be made.

You have answered my question about the kind of system that you would propose if you had a clean sheet of paper. Perhaps it would be useful to conclude by posing that question to the other witnesses. What would you all do? What is your route forward from where we are?

PFI should be an option that is available to bodies to use to procure goods and services, but it should be only one of a number of options.

If I were in the committee's shoes, I would be asking the Treasury to review some of the rules and to consider the potential framework. At the moment, pressure is coming from Wales for that. I come from a local government background. In the past year, there has been a green paper at Westminster and a white paper was issued last week. Part 2 of that white paper deals exclusively with local government finance issues. That is worth considering as part of a wider review of financial issues in Scotland.

Are you saying that the real issue is the availability of capital finance for local government and other public services.

Yes. We need to look at the issue in a wider context, rather than focusing purely on PFI.

There are clearly financial, political and philosophical issues involved, which the committee will no doubt consider. However, I suggest that it might be useful to spend some time considering what happens on the other side of the fence, to see how the whole process works in practice and how it is implemented. We have heard in evidence this morning that the wheel is being reinvented all over the place, and perhaps in some places people are making the wrong kind of wheel. Perhaps people are being driven in certain directions by the one-size-fits-all advice that they are given. We would like a more selective and discriminating approach, not just in the philosophy and policy but in the implementation. That would be worth examining and spending time on.

We are about to look at a case study-based approach, so I can reassure you on that point.

The private sector and private finance clearly have a role to play. It is important to distinguish between the private provision of public services, which is essentially what we are talking about, and the situation in America, where individuals pay for their own health care. That is why 50 per cent of the people are not served. The situation in America is not like the situation in the UK. In the UK, we are talking about the private delivery of hospitals that are paid for by the public sector.

There is clearly a role for the private sector to supply services. The real question that the Government is considering is whether to buy assets or to buy services. The answer is that that will be appropriate in some situations and not others. That is the central point that has not bottomed out. Private finance schemes do not work everywhere. In some places, they work terribly well. Roads are a good example. As a standard model, privately financed roads work very well, because that model has been well worked through. The differences between roads and other areas have not been picked up on.

Many of the problems that have been dropped on PFI in this morning's discussion are actually to do with how PFI is working within a rigid set of other rules. The crowding-out issue seems to be exactly such a problem. That is not a clear argument against private delivery in certain circumstances. It is just that, if we use PFI and do not change other rules to deal with it, we may end up with a situation that is not as good as it could be. That is a separate issue from the issue of whether there is scope for private finance in the delivery of some public services.

Whether people like it or not, PPPs are here to stay. If we consider the education sector, a UK Government adviser said that 90 per cent of schools are beyond their reasonable life; £3.5 billion will have to be spent on education between now and 2004, of which it is planned that a third will be invested by the private sector. Twenty-nine new hospitals are coming up for bidding over the next 12 months. The Government does not have the money to provide those facilities. If the public sector does not provide the equity investment and meet the debt, those schools and hospitals will not be built. It is therefore a simple choice: we either go the PPP route or we do not.

The second point is that about 65 countries around the world are adopting our pattern—the PPP concept. They are doing it in different ways, such as build-own-operate projects, but they are following a similar route to us and cannot see any alternative to moving from the public sector to the private sector. We should not just accept where we are at present but should say, "Well, we've got this system; it is not set in stone, so how can we improve it?" That is why I broached the subject of partnership. Can we develop true partnerships, with more flexibility? What are the problems of having public sector investment in some of those projects? What is the problem with having public sector bonds—or tax-exempt municipal bonds—as they have in the United States? Those are things that are sometimes considered by the Treasury and rejected immediately but without any informed discussion at a sector level. We need to review them all.

In addition, we are not really considering what is happening in other countries. Argentina is producing a PFI fund, which will be created by taking a percentage of fuel tax and building up a fund so that it can have a PFI/PPP programme. There are examples around the world that we ought to be considering. We should start to consider some of the public sector concerns and find solutions to them as soon as we can; otherwise we will go down a divergent route.

Incorrect criticism of PPPs that is laid at the door of the private sector causes uncertainty in the market. If we cause uncertainty in the market it means that the equity investors want a higher return and the lenders want a higher return on their debts. That is the last thing that we want to do. London Underground has been a typical example. It has caused a lot of concern in the marketplace because of all the disagreement on the structure for the London Underground PPP.

Railtrack has had a minor hiccup because it was a privatisation, not a PPP. The Government could not have done what it did to Railtrack if it had been a PPP because a PPP has a contractual structure. The Government would have had to discharge the debt and pay off the equity. We have to differentiate ourselves from Railtrack.

On the international dimension—although I hope that we do not go down Argentina's route—there was a question about the EU. It is important to note that we are a signatory to the World Trade Organisation and have signed up to GATS—the general agreement on trade in services. The WTO is committed to opening up public services, especially health and education, to the marketplace. A question to put back to the committee is how does opening up and exposing our public services to markets through privatisations, and exposing ourselves to the trade rules through our membership of EU level and GATS, make us vulnerable? We need to explore that. We might think that we can control the market locally through protectionist measures—we talked about Government procurement—but we may be into a completely different ball game as a result of signing up to GATS, which commit us to liberalisation or, in other words, to opening up markets in trade and public services.

That is a valid question on which to end, although I am not sure whether we can answer it here. I thank our witnesses for coming along. It was a vital debate and quite a substantial range of views were expressed. To a considerable extent, that range of views is probably reflected in the committee. We will consider how to take our inquiry forward and will almost certainly give more in-depth consideration to specific schemes, to try to get to the bottom of the general issues that you have raised and how they apply in practice. We may get back to you to ask for further information or clarification on the issues that that throws up. Thank you for your submissions. We will consider them and, hopefully, come to a view on the issue over the coming months.

Meeting adjourned.

Meeting resumed in private at 14:07 and continued thereafter in public at 14:35.

I welcome members of the press and public to this reconvened session of the Finance Committee. For agenda item 5, we will take further evidence for our PFI/PPP inquiry from six representatives: Tom Kelly, who is the chief executive of the Association of Scottish Colleges; Andrew Gordon, who is the chief executive of Canmore Partnerships; Ian McDonald, whom I know well in his guise as depute director of education services for Glasgow City Council; John Curley, who is senior education officer for Glasgow City Council; Keith Patterson, who is a partner of MacRoberts; and Sandy Bremner, who is the head of the public-private partnership unit at Miller Construction.

Given that there are six witnesses, we shall not ask you to make general presentations. As a start-off question, do you have experience or evidence to show that PFI provides effective delivery? Tom Kelly can start by talking about the further education sector.

Our sector is in a rather odd situation in that, although one of the earliest PFI projects in education was the one at Stirling, we are still at a fairly low point on the learning curve. We have had two other projects since, but we do not have a major consortium or any of the very large projects that have taken place since the one at Stirling.

I believe that PFI has delivered in the projects that we have had so far, which have provided good facilities that have substantially benefited the learners and the local community. For example, West Lothian College's move to its new location and facilities has transformed recruitment. A large part of that success must be attributed to the college's new capital facility, which has a new design and a new location. PFI has made it possible to get that facility perhaps sooner than might otherwise have been the case.

How many of those benefits could still have been obtained through more traditional funding of a capital project? That is a speculative question, I suppose. As we have used PFI for the project, it is reasonable to attribute at least some of the benefits to the way in which it was done.

We will hear from the public sector representatives; others can then follow on.

As you will know, our PFI projects were for accommodation services and IT services. I will address accommodation; my colleague John Curley will speak about IT services.

We are now 18 months through the initial phase of the accommodation services project and, with the exception of one school, we should be finished by July 2002. Although at present we are still in the middle of what is a gigantic and very complex building project that is spread over more than 20 sites, I can identify three important aspects from which the council has benefited to date.

First, the project is on programme. Schools are being completed as they were supposed to be and there has been no significant slippage. When a complex service such as education provision is planned, having schools completed on time is beneficial. For example, three new schools are due to be passed to us at Christmas this year, and they will be passed to us. By and large, the project is on time.

Secondly, as for Glasgow City Council's overall responsibilities in the next 30 years, it is clear that the risk is transferred from the council to the private sector during that period for the construction phase and thereafter for latent defects in all the buildings and so on. There is a genuine transfer of risk and responsibility for providing the council with 29 good secondary schools for 30 years.

Thirdly, another benefit is that the contracts are long term. We might have wanted that facility in the public sector. The ability to plan over 30 years and to have that responsibility over 30 years means that important decisions can be taken. Our experience is that a higher-quality finish is being applied to school premises in the fixtures, fittings and furniture. The successful contractor in the Glasgow schools project takes the view that a higher-quality finish across the board to all facilities, furnishings and fittings will mean that they should last a bit longer and need to be replaced less often. The quality has improved.

As Ian McDonald said, I am responsible for the information and communications technology part of the contract, which is much smaller than the buildings part. The PPP contract allowed us to take a longer-term view of ICT than would otherwise have been possible. We have a 10-year ICT contract. No form of funding other than a PPP would have allowed us to make such a contract, which has enabled Glasgow to meet the stretching targets that the Executive set us on the number of computers for youngsters and of an e-mail address and filtered access to the internet for everyone. We could not have met those targets early without that form of funding. Ian McDonald was correct to say that we transferred significant risk to the private sector, which can manage ICT much better than the council can.

I would like you to address the other side of the framework. Will you concentrate on your experience of PFI/PPP cost and time overruns as against traditional methods of procuring projects?

We have not experienced major time or cost overruns. We have three projects in Scotland: the Stirling centre, which Tom Kelly mentioned, Inverness airport terminal, and the new maternity and day surgery centre in Dumfries and Galloway, which will open next year. If those three projects illustrate anything, it is that there was a different way of doing what the public sector wanted. The Stirling centre is in a different place and probably looks different from most further education colleges.

Although we inherited the public sector's architects for Inverness air terminal, they eventually designed a different building. Highlands and Islands Airports Ltd has taken our service providers to run its new publicly funded terminals in Stornoway and Kirkwall, which seems to show a certain satisfaction with what we have done in the new terminal.

It is clear that no one in Dumfries and Galloway had ever thought of the design solution that we eventually proposed. PPP opens the service up. We have not been late and have not overrun to any marked degree.

It is probably clear that PPP has delivered more buildings on the ground than would otherwise have been the case. It is worth recognising that there are a number of different aspects to servicing a PPP project. Much of the initial focus has been on the delivery of the buildings and facilities. That is fine for the moment, but they are 30-year service contracts. The jury is out on whether service delivery and performance will be better or worse in future than it is now. It is too early to come to a conclusion on that. We need to assess the impact of the benchmarking and market-testing provisions when they are applied about five years into the project.

I would like to take up the point that Ian McDonald made about the Glasgow schools contract. Miller Construction is involved in that project, as shareholder and as constructor, and in the life-cycle element of the contract. Obviously, the contract is all about performance, and performance targets have to be met. There are stiff penalties if performance is not achieved. If accommodation is not delivered on time and if facilities are not serviced, there are stiff penalties from the public sector and from the funders. Performance has to be delivered and that is something that we strive to do.

With regard to the Glasgow schools project, mention was made of the improved quality of some of the fabric that was used. Andrew Gordon referred to the different design solution that was proposed for the Cresswell replacement maternity unit in Dumfries and Galloway. What I cannot quite understand is why those improvements are necessarily associated with PPP. Could not either of those projects have been done by public procurement or by a trust or arm's-length company being set up? Why is the success of those projects uniquely associated with the PPP solution?

I am happy to respond to that question. I qualified my response by saying that it would have been good to have the luxury of planning over a 30-year period in the public sector, but that has certainly not been my experience. In my experience, we have always had a long list of projects to be met through conventional procurement, and one is always trying to spread the money as far as it can go. In that situation, the overall level of quality is often a bit lower than one would like to be able to afford. I qualified my answer by saying that the big difference between a PPP contract and conventional procurement is that, in a PPP contract, there is the capacity to plan over a 30-year period and take a view that is 30 years long.

Is that because PPP allows you to avoid capital consent restrictions?

No. I was just speaking quite pragmatically. Previously, in the annual spending round, I had to take a very short-term view of what could be afforded, because I never had the luxury of looking 30 years ahead. Aside from the projects that were funded, there were always other projects that were not funded. I take a pragmatic view. As I said, we would all have enjoyed the luxury of being able to take a long-term view of an asset, rather than doing what had to be done because we were taking a short-term view.

Presumably that asset would be constructed out of your capital budget, not out of your revenue budget. Even under the old regime, it would be to your advantage to construct something that had lower maintenance costs rather than higher ones, providing that you had capital consent.

Logic would take you to that conclusion, but reality is somewhat different. One need only look at the large number of flat-roofed schools in the west of Scotland, which has inclement weather. Nobody would ever have wanted to build a school with a flat roof. People would always have wanted to build a school with a sloping roof, but roofs account for about 30 per cent of the cost of a new school building. My experience has been that, because we were always involved in short-term planning, we tried to eke out as much as we could from very scarce resources, rather than taking a longer-term view.

School after school was built with a flat roof, which has a long life-cycle maintenance cost attached to it. I am afraid that that was the result of good public servants having to take a pragmatic view of the available money. There is nothing inherently better about that. We just did not have the luxury of 30-year planning.

With regard to pragmatism and capacity in relation to the Glasgow schools project, how much of the drive towards that approach was related to the perceived over-capacity in Glasgow schools and the difficult political problem of trying to change that? Did not the PPP route of financing the refurbishment of the schools represent a convenient way of dealing with a difficult and intractable political problem within the city?

That is true, but we were open about the fact that there was 40 per cent surplus capacity in Glasgow secondary schools. That was a significant drag on the overall property costs that the council had to meet. The council recognised at an early stage that keeping open so many half-empty or three-quarters-occupied buildings was not a judicious use of public funds. Significant rationalisation was an important part of the Glasgow schools project. That was difficult to deliver politically, but it was delivered.

The council considered several options for realising its objectives over a short period. When it embarked on probably its biggest ever rationalisation exercise, it wanted to produce as quickly as possible high-quality new or improved buildings. The council did not want a long time scale for that. It took a brave political decision so that it could deliver an end result as quickly as possible and not have the problem dragging on for generations.

After a detailed feasibility exercise, the council determined that a PPP scheme was the only accessible source of public funding that would begin to deliver its ambition of improving significantly the remaining schools. Glasgow City Council has always been open about why it took that decision. It did not want to see matters dragging on for 10, 15 or 20 years, but wanted a solution to be delivered quickly. The council is always interested in considering what funding is available. It took a pragmatic view that, to proceed quickly, the only feasible answer was to consider a public-private partnership project.

I remember David Montgomery, who was depute director of education in the former Strathclyde Regional Council, arguing that within Strathclyde every school had a shelf-life of 400 years, such was the flow of funds for replacement.

Was that a flat shelf?

What about the different design solution that you said was proposed?

I do not know why that was so. Perhaps the PPP process is more challenging, so that when consortia are asked to think about a project, they think in a different way. The Stirling example is interesting, because if the project had been traditionally procured and conventionally financed, the FE college would have been sited on the ring road. One of our colleagues knew that the stone yard that was used for Stirling Castle was about to become vacant because the work on the castle was almost finished—after about 400 years.

Public procurement.

I do not think that it was a design-and-build project.

That yard was in public ownership, but the private sector flushed it out as a site. The FE college, which was represented on the committee, said that it preferred that site. The building that resulted is not what people might think of—rightly or wrongly—as an FE college, but it is a bold statement.

I do not know why no one had thought of our solution for Dumfries and Galloway Royal Infirmary. For 17 years, trusts and their forebears had considered what might be built on that site. For many years, as people talked about closing Cresswell Hospital and moving it on to the Dumfries and Galloway Royal Infirmary site, no one had apparently ever considered its being adjacent to the operating theatres—where we put it. I do not know why they did not consider that, but I know why we did. I know that all the public sector's comparative planning did not perceive that the answer could be what we actually did.

The process in Inverness was interesting. We rebriefed architects who had originally been briefed by our public sector client. We came up with a building that was more flexible and easier to expand. It is quite likely that most of an airport will not be needed in 25 years. One does not want to think about disposable architecture, but the architecture should be very flexible. An airport like the one in Inverness will not be like a tiny Glasgow airport or Edinburgh airport, because most of the planes that land in Inverness do not even reach the first floor.

Originally, the plan was for a two-storey airport, which is a mystery to me. The airport would never have needed air bridges because they would have gone right over the planes.

Such situations do not always happen. Sometimes there is an obvious design solution that it is right to implement. With at least two of our projects in England, I cannot conceive that the public sector would have built them in a different way; there was an obvious way to build them. However, something in the process seemed to make us more challenging and questioning. I suppose that there is a danger that we might get it wrong.

What is it in the process that leads to such challenges?

Colleagues might have a view on that. Perhaps it is just because it is a much more open brief. One is told, "This is what we would like to do. We have our own ideas. Here they are, but you can come up with anything you want." None of us wants to win by just being the cheapest; we would rather come up with something that is better, and we would rather make more money out of a better product. That is what makes us more inventive.

I was one of Andrew Gordon's competitors on the Stirling centre. We had a completely different solution on a different site. Andrew Gordon's company was the winning consortium. That demonstrates that two consortia looked at the project in a different manner and came up with two different innovative solutions. The private sector puts a lot into the innovative thought process. We demonstrated that with what we brought to the Glasgow secondary schools project. The original concept was for two new secondary schools. There ended up being 12 new builds—11 secondary schools and one primary school—because of the innovative approach to the process and the recognition of what the council was looking for in its output spec.

PFI completely changes the basis on which fundability is tackled in the public sector. Under traditional capital grants solutions, one is tied into annual budgets and a place in the queue. If a project requires too much in one year, one is sent back to rephase. If the overall cost of the project is starting to look too big, one is sent back. Those early stages, when a contractor is not involved at all, can be extremely protracted and difficult, and are followed by a rush to get on site at the last minute. In contrast, with a PFI project, one has to work to the fundability of the project over its life, which means that the balance shifts to the revenue side. That is particularly important in our sector. The issue of fundability becomes a revenue question rather than a capital question, and it is tackled differently.

I wish to pursue that. I understand what you are saying about annuality and a place in the queue. Does that mean that design issues and fit-for-purpose issues tend to be neglected, because everything is waiting to rise up until it is over the threshold for funding? Is what is new the fact that, in a sense, somebody actually goes in with a design brief, or the possibility of producing a design brief, for something that is out of the box of what could have been funded by traditional mechanisms?

With traditional capital projects, fitting the timetable and the capital ration became the preoccupations. I know of projects that I handled when I was working in the civil service that had been sent back for redesign four or five times before they got to site. The issue was about how much the project would cost, not the value that it would deliver over its lifetime. That is the liberating element. There has not been a lack of innovation or creativity in public services. There has been a difficulty in getting the opportunity to work with contractors in a different way.

On the back of the comments that Mr Kelly made about fundability, I return to Falkirk college's expansion into Stirling. I considered the project when I was on Stirling Council planning committee. We had nothing to do with the commissioning of the college; we were merely the planning authority.

Part of the drive that came across in the general presentation work that was done in advance of the formal committee meetings was that there was an opportunity for Falkirk college to take a risk in going into another area to operate a new venture. The risk was that if the college had done that with its own money and ownership long term, there could have been a downside, which would have come back to the matter of long-term affordability.

Was not the college making use of a scheme that was useful to it, in that it moved the risk of the building? The building that I saw as the final model could be used for other purposes. There was a built-in exit strategy for both the college and the PFI deliverer. The building had another use; it was in a location that was fairly close to the town centre and it would link into the town centre schemes.

We saw an early use of PFI to ensure that the project was affordable long term, but there was also the exit clause of the removal of risk for both the PFI deliverer and the college authority.

I ask Mr Gordon to expand a bit on that thinking and on why that was such a useful project for Falkirk college.

That thinking certainly reduces risk in two ways. It is an unusual element of the Stirling centre transactions that the college can hand back given parts of the building. There is a minimum notice period of three years. In a nutshell, the college can hand back big chunks of the building and never pay big chunks of the unitary charge.

There is flexibility in the building. It was designed to meet not just the college's needs, but a general office spec. The part of Stirling in which the college was built was slightly more rundown when we went to visit it than it is now. It has come up and the building has, in part, helped to bring it up. There is a self-fulfilling prophecy.

The hand-back option would not be possible in all projects, but it works in that project because an omni-purpose office space has been built.

Not having to do the building is a huge lifting of the management burden for a college such as Falkirk college or any other that is undertaking an entirely new venture. Opening a college is a big enough job. Being able to move in, recruit and teach is a huge task, without having to deal with the bricks and mortar. All the rest is done for the college so that it can concentrate on its real expertise and its core mission. That also reduces risk.

On the back of that comment, did the nature of the project make any difference to the price of the package? Did the fact that there was an exit strategy for both the college and the PFI developer reduce the college's costs?

It did not reduce costs markedly. That might not be what you would expect, but the reality is that most pricing for PFI transactions is determined by the requirements of banks in the first one or two years of the project, and meeting cash-ratio requirements. After those have been met, one finds that investors and everybody else can enjoy an adequate return. Although the building has real residual value, real residual value 30 years down the track, discounted back to the present day, is not a large sum. The amount that the college paid when it occupied all the buildings in year one was not reduced. If things had gone badly for the college, which they did not, they would have gone slightly less badly than they would if the college had procured the building directly.

It has been suggested to us that one of the benefits of PFI/PPP is that the project management has sharpened up considerably compared with normal public procurement. There is a much tighter specification of the product. It has also been suggested to us that certain projects—a few people have mentioned IT projects—are not particularly suitable for PPPs. Is that your feeling? You have given us what you consider to be good examples of PFI/PPP. Do you have any bad examples?

Some of the project management in the private sector is very sound. The private sector has a lot of expertise that we do not have in delivering large-scale ICT projects. We have a lot of confidence in the ability of the private sector to deliver ICT. The specification that we ended up with took almost a year to develop. It still leaves the vast majority of the risk with the private sector. We have retained very little of the risk in relation to the operation of the network, computers and software. That is of great benefit to us.

I do not have much experience of negotiating ICT contracts, because my background is in education. It was not something that we had done before. We had help from corporate ICT, but the project involves a huge network—possibly the largest schools network in Europe—and is a new venture for everyone. We might have got it wrong. If the private sector gets it wrong, it pays for that and the costs are severe.

Are you procuring an asset or are you contracting out a service? What belongs to whom at the end of the contract?

It belongs to us. We are buying a service. We do not pay until the service is in place. Again, that is unique. Usually, we would specify what we want, we would get what we asked for and if it did not work that would be tough—we would still have to pay for it. Under this type of contract, we described what we wanted to do educationally and left it to four bidders to come up with innovative solutions for how to deliver that and take the risks if it did not work.

A good example is internet access. At the moment, no one can guess how much bandwidth we will need to access the internet in the next five to 10 years. I could take a stab at that and buy the appropriate lines for the school, but in a year's time I might find that I do not have enough capacity. That element has been farmed out to the private sector, which supports that risk completely. If the network does not meet the specifications, the contractor must increase the bandwidth of the network so that it can cope with the demands. That is at their cost and at no cost to us. The risk that we have shifted on to the private sector is big.

Perhaps Ian McDonald will deal with the point about project management and specification.

I would be pleased to do that. The point is not about inherent quality; it is simply that project management is simplified in PPP projects. Traditionally, project management is more difficult at the interfaces between one contract and another. If we procure an accommodation service, it does not mean that we procure the repainting of a school, or the provision and moving of its furniture—all that is separate. When all the separate contracts come together is when it is difficult for the council to manage things.

When we procure an overall accommodation service and specify what the service should be like, we pass the project management to the contractor in a much more simplified form. The contractor will have complete responsibility for anything that happens. If we are dealing with one contract, and our school has been painted from top to bottom and damage is done to the pristine walls while the furniture is being moved in, there is no headache for the council. In a PPP project, the contractor is responsible overall, regardless of who damaged the wall. Project management can be simplified because the contract is for an accommodation service.

It is not a question of inherent quality—that people are better or worse; it is simply that the PPP as it is currently constructed simplifies project management and when things are simpler they tend to turn out better.

I return to a point that Keith Patterson made about the jury being out on the facilities management aspect of PFI projects. That is true. There are disadvantages to what I would call first-generation facilities management contracts, which are extremely detailed. Clients for whom the contracts were the first introduction to this way of working found it difficult to be certain that they were negotiating on the right basis. Some of the details are so prescriptive as to defy logic and the user requirement.

A good example is where the facilities manager is responsible for security of a building and no one is sure who is allowed to open the windows. It is essential that we share experience across public services. When the Scottish Parliament was first established, members had to legislate for situations that had not been thought about. In the past, the owners of a building were responsible for its maintenance; people are on a different footing if they work with a facilities manager.

We remain uneasy about that, as the contracts tie people down. Colleges require flexible user space, but there are uncontrollable risks, including damage to classrooms, and it is difficult to pin things like that down in the detail of contracts. The experience is wholly different and the relationship that emerges is also different.

Is that necessarily a function of PFI or are the positive things that you highlight a product of different ways of going about contracting, which move colleges away from the idea of being run entirely in-house to the idea of outsourcing project management and other operational issues to contractors? Is PFI a necessary component of what you suggest or is that just the way that the situation has evolved?

I agree with that analysis. I take a positive, long-term view of contracts that provide accommodation services. I have not linked funding sources to what, in our experience, has been of benefit to date. The nature of the contract is that it is a long-term contract for an accommodation service.

Facilities management is not my field, but I have a couple of general comments to make. Lack of performance leads to penalties. The whole PFI/PPP market has matured. It has been on the go since about 1992. Things have moved on since then and FM is an important part of the process. Output specifications have developed to assist in the understanding of the issues that have been raised today. Issues such as who is responsible for opening windows are becoming more defined in the output specifications and are regulated under the contracts. Those things may have been tiresome and problematic at the outset, but they are becoming less problematic. The market is becoming more sophisticated, as are the providers and the public sector in terms of the requirements of what needs to be delivered.

With facilities management, the client is confronted much more directly with the fact that the consequence of a transfer of risk is a transfer of control. If someone enters a PFI project for facilities management on any other basis, they are deluding themselves. What is required are ways of managing that transfer of control. Those who were first in, in our sector, have a lot to teach the rest of the sector before it goes down that route.

Is there sufficient consultation with end users in the design of PFI/PPP projects? Some interesting discussions have taken place about the involvement of end users in the Glasgow schools project.

It has been suggested that there may be a fixed budget and that, if variations are made to it, the savings will be made on the operations side—the services side. Inevitably, if fewer services end up being negotiated, as the detail is worked out, some services will not be provided. That introduces issues of equity and fairness. To what extent do you take those issues into account in your consultation with end users?

Significant and on-going consultation took place with the end users or, more correctly, with representatives of end users, since we could not begin to consult 2,600 end users. There were layers of consultation. We consulted the school management teams, school boards and the trade unions. We carried out a series of consultations and we also involved the end users in the final evaluation. We built in as much consultation as possible and fed what people wanted into the design of schools. Many attractive and unique features in Glasgow's schools are a tribute to the design input of users. For example, every school now has a large internal social space for pupils. School boards told us repeatedly that it rains in Glasgow, so it would be good for youngsters to have decent internal social space instead of having to go outside to find somewhere to huddle.

About 800 people were engaged throughout the consultation process. The picture may change because, for example, the head teacher when we started is not the head teacher today. People must take decisions on what is appropriate for school accommodation, because the budget is not elastic.

Is it not true that during the process, a significant number of facilities, such as swimming pools and similar amenities, were lost? Those who were engaged in the process were originally under the impression that those facilities would be there, but they were not delivered. How can you maintain credibility with end users? How do you ensure fairness for all interested parties?

There is not a difficulty with maintaining credibility among end users. I see the delight on people's faces when they move into new facilities. That is not an issue.

The council ended up with what it specified to begin with. It wanted 29 secondary schools, a primary school and a pre-fives facility. It banded all its schools according to the maximum number of pupils that the building could take. That is a progressive move in Scotland: it had not happened before. The maximum specification and overall accommodation required was detailed down to individual laboratories and so on.

I am aware of some of the public comment—informed and uninformed—that has been made about the Glasgow schools project. Most of the focus tends to be on the loss of some swimming pools in Glasgow. It is correct that the council did not specify the replacement of a number of swimming pools. That decision was taken after extensive consultation with schools, with our culture and leisure services department, which is responsible for all public facilities, and with bodies such as sportscotland.

We tried to ensure that, rather than simply replace what was there, the new improved facilities gave Glasgow a better strategic covering of all sports facilities. We identified that, although Glasgow was exceptionally short of good-quality games halls and good-quality Astroturf outdoor football pitches, it was not short of swimming pools. The council took the strategic view that it should improve the facilities not only in each school but across the whole city. Therefore, we decided that we would not replace, I think, six swimming pools, but that we would invest significantly in alternative leisure and recreation facilities. That is providing benefits. The most recent analysis of the city's provision showed that our ability to meet the demand for community use of indoor provision has rocketed because of the provision of 29 high-quality games halls that are attached to schools.

When we took that strategic view, we did not overlook the requirements for swimming provision in the city. Our decisions took into account whether there were public pools nearby that schools could access at no cost to the council or to the school. Transport provision was made available as well. We also did a cost-benefit analysis—from which we got the same results as another council in Scotland—which showed that a school swimming pool is quite a costly investment in relation to actual usage. Our figures showed that the cost was of the order of almost £40 per swim per pupil, whereas making provision for the activity at a public facility is a fraction of that cost.

We did not reduce facilities or reduce provision over the term of the programme.

Surely your cost-benefit analysis figures would have been available before, not subsequent to, the start of the process? Are you telling me that the primary driver for arriving at the final scheme was the operating costs of swimming pools compared with those for games halls or whatever other facilities were provided as an alternative? Did the decision come down to operating costs, not up-front capital costs?

What I said is that we took a strategic view of all leisure facilities across Glasgow and determined where the schools programme could remedy obvious difficulties in our overall provision, which is what has happened. By combining the facilities that are managed by culture and leisure services and those managed by the schools, we have given Glasgow a more equitable provision of leisure facilities of a much higher standard. Pupils and local people have benefited from community provision. We took a strategic view; our decisions were not driven by finance. I used the example of swimming pools because their provision is expensive in terms of overall usage.

Will you spell out for me why, given the fact that the survey that related the facilities that were required in schools to the facilities that were required overall had presumably been done in advance of the PFI project, the project specification started out with the swimming pools and ended up without them? You told us that the cost was £40 a swim. It looks awfully like your decision was based on operating costs.

No. The council was operating within an overall affordability limit. We have always been very public about that. We wanted to maximise the return to the council, to pupils and to people in Glasgow within that overall affordability limit. I said that our decisions on leisure and recreational facilities were taken against the backdrop of what was available across Glasgow. It was open to the council to specify that all swimming pools would be replaced, but that view was not taken at the outset of the project.

To answer your question in its totality, what was delivered was what the council wished to see delivered and what the council specified. The decision was taken with the broadest view possible. All analysis since has shown that the council will be in a better position to meet community demand for sporting and leisure facilities than was the case several years ago.

I think that David Davidson wants to shift the focus.

I touched on the movement and sharing of risk in the model that was mentioned earlier. The theory is that, under PFI/PPP, risk is put where it is best managed. We have been told in evidence over the past few weeks and again today that the amount of risk that is transferred is negligible; some have suggested that ultimately the risk comes back to the client. I think that there is evidence either way, but we would like to hear how you think the risk assessment for PFI should be conducted. How do you go about valuing the risk? Can any of the witnesses give us evidence about how that risk has been transferred in practice?

It is best if I answer that in terms of how the contractual framework deals with risk transfer. On the one hand, there is the question of what the risks are and how they are allocated between the parties. On the other hand, if a risk that has been allocated to the private sector is not managed properly and the contractors end up in breach, what are the consequences? Is there leakage back to the public sector? The way in which the standard contracts—those that are now in place—tend to work is that the compensation provisions provide that, on a private sector default, the overall exposure of the public sector to the project is no greater than it would have been if the project had continued. If the public sector has to take over the project because the private sector has not completed, has not made the billed specification—that happened with a project in England, for example—and is in default, and the contract has to be terminated, the private sector loses. The public sector probably suffers delay, but the private sector loses the money. That driver reinforces the risk allocation that is set out in the contract.

How do you put a cost on the risk that is transferred?

What tends to happen with most of the projects that are at outline business case stage at the Scottish Executive is that a team of technical advisers and accountants gets together. As a lawyer, I tend not to be involved in that process, but I know what happens. The accountants and technical advisers take a view on what the risk is, the likelihood of it happening and the cost if it does happen. They come up with a weighting figure to apply to that risk and to the cost in the public sector comparator. The public sector comparator and the net present value of the special purpose vehicle privately funded model that goes to the Scottish Executive will each have a cost category. In the public sector one, that will be unadjusted; in the private sector one, a view will be taken on the likelihood of, for example, a construction cost overrun and a percentage variation will be applied to that cost category. That is how we get our adjusted public sector comparator.

Perhaps we will hear Mr Kelly's view from the client end, but are you saying that it is all down to the advisers that both sides employ to negotiate, and that some kind of matrix comes out of that? Does not that make you think that perhaps there needs to be a national guideline? It sounds like overlap.

You brought up the public sector comparator. What value do you think should be applied, or are you happy that one size fits all?

On guidelines on valuing the risk allocation, I was talking about the process before procurement is initiated. That has to be re-evaluated at the end when the full business case is made. Most parties try to use values, which they believe are derived from evidence, that are, in a sense, national, but there is not a national guideline. There would be difficulties with that, because in some projects, specific factors affect the risk, the valuation and the likelihood of the event's occurring. For example, a construction project in the Highlands and Islands is likely to have a different risk assessment of overrun from that on a city project. A national guideline can be taken only so far, but it might well be useful.

We are discussing not only the allocation of risk between the parties, but whether the total amount of risk in a project is reduced through the specification. Does that happen?

At the moment, risk transfer is settled. The public sector has pushed quite a lot of risk on to the private sector. The private sector resisted some of that. We have reached a point at which we are more or less happy. We have no control over some stuff that is transferred to us, but we think that that stuff is so unlikely to happen that we accept it, because we want the business. Such risks do not affect prices, because we think that the events are unlikely to happen.

Guidance by sector is produced. In England and Wales, the NHS executive gives guidance on risk transfer. That is probably too structured and not project-specific enough—the pendulum has probably swung the wrong way. A strong argument exists for a common methodology, provided that such a methodology accepts that projects are different and encourages projects to be different. When you offend the norms, you must explain why.

It is suggested—it may or may not be true—that risk transfer was what made some early PFI projects work; it made them stack up. Those involved could work out the value-for-money gap, which gave the risk transfer. People say that that could not possibly have happened, but the process could be more open. There is no reason for us not to debate what the transfer should be. The private sector could say, "That risk is of little importance to the public sector, but we are worried about it. Why don't you take it back, and we'll charge you less?" We could have a bit of a discussion about it. Instead, in the NHS in England and Wales, we are told that a building cannot overrun by more than X per cent. I know of at least five projects that have overrun by more than that.

We have some local experience of that.

The exceptions will not make the rule. Public sector buildings are often well procured, but there are also some duff ones. The NHS has had some good buildings, but it has had some bad experiences too. The buildings that are involved are particularly complex and are easy to get wrong. There is a big risk that the buildings will be late and that they will not be right when the private sector thinks that they are ready. There is real value in getting rid of that situation. Does Sandy Bremner agree?

Yes. I agree with Andrew Gordon's comments about where risk stands. The deals have mechanisms for dealing with risk transfer to the private sector, particularly through funding. Funders must be given many performance guarantees. The contracts include long-stop dates for overruns on the contract and drop-dead dates too. They are all mitigated by various measures, such as guarantees, letters of credit or liquidated damages.

That might be financial mitigation, but the private sector has much to lose in not performing. It will mitigate that by other means rather than wait for the financial triggers. For instance, if a ward is late for a health project, a temporary solution might be found—if it is late because the air conditioning is not ready, temporary air conditioning could deal with the situation. Practical measures can be taken to mitigate such construction delay risk.

That applies equally to the operating side. If there were a power cut because somebody had cut a cable, transformers could be brought in to restore power. The private sector is aware of those matters and takes those risks on board. It does its utmost to mitigate such risk. That is something that has been taken away from the public sector.

Tom Kelly wants to say something from the client side.

There is a problem for a sector such as ours because, in practice, a college will be a client only once. Risk assessment can seem like a dark art that is hidden from everybody except the participants. It may be a black hole into which you will unexpectedly fall, but that has not been our experience to date. In our sector we have needed, and will continue to need, a degree of hand-holding on the client side, which can be provided both by employing one's own expert advisers and by having access to central expertise such as that offered by the Scottish Executive's PFI unit.

This is a difficult area. Risk—in the technical sense—is used in the assessment. I have looked at a risk assessment, but I am not sure that I have grasped the sense in which risk is taken to be relevant. The more practical issue for our sector is that of the transfer of risk, which I mentioned earlier, as against the transfer of control.

Revenue underwrite is a big problem, particularly in relation to financial risk. The contractors must be satisfied that the means are in place to pay for the contract over a continuing period. In the college sector, which has too many colleges that operate with a deficit and in which funding is guaranteed only a year ahead, the issue of revenue underwrite has had to be addressed on a case-by-case basis.

I wish to return to Andrew Gordon's point about good and bad projects, whether conducted under PFI or under traditional procurement arrangements. Previous witnesses have suggested to the committee that one of the impacts of PFI/PPP on traditional public procurement projects is that it has improved project management, specification and delivery—it has driven up quality in the traditional sector. Does Andrew Gordon agree with that?

That is not surprising. It illustrates that there are good projects and bad projects—a number of the responses to questions have illustrated that point. Some projects happen to be PFI/PPP, but if one has conventional funding and does not consult users, specifies badly or has bad project management in place, one will have a bad project with poor outturns. Experience of a good project affects everyone who is willing to learn. A good PFI project will be a good educational experience.

We can only speculate why PFI was invented, although I think that we have a good idea that it was led by the F of PFI. The most interesting point is how one can use PFI/PPP as a logical extension of outsourcing, in order to have the best teams—in both the public and the private sectors—doing what they are best at. The finance is interesting—it is surrounded by a whole industry—and allows large, almost infinite, amounts of capital investment, where annual budgeting would not.

Good projects, and how they are specified, are among the fundamental topics that we are discussing. We all know that, when projects go wrong, it is usually because one of the old bugbears has gone wrong, whether they are PFI or conventionally procured projects.

Can you give us some examples of projects that have gone wrong, and why they went wrong?

I certainly would not mention the project to which members have alluded, as I could not possibly win.

The channel tunnel is such an example. I was tangentially involved in it through advising one of the shareholders not to go into it. The project was technically outstanding—putting a tunnel under the channel was a miraculous engineering feat. However, the research into the market for the tunnel and the arrangements for supporting infrastructure to take people from the mouth of the tunnel—particularly on this side of it—were spectacularly poor. That was a private sector job that was arguably driven, in large part, by political will. There is much that was brilliant about the project, but it had some major faults where those involved simply got it badly wrong.

In terms of unforeseen risk, I suppose that the issue of asylum seekers, or economic migrants, could not reasonably have been forecast.

It certainly was not forecast. I remember the bidding document. It was obsessed with rabies. There were pages—huge sections—about rabies and rats and dogs coming in through the tunnel, but there was no mention of people coming in at all. And of course, all that the engineers were thinking was, "I wonder what's under the channel. Until we go in, we won't really know."

Supporting research was done on who would use the tunnel, on how much they would be prepared to pay, and on how they would get there on a train that went at more than 40mph, but it beggars belief that billions of pounds could have been spent with such enormous failures in intelligence.

That is a well-known example of a project that is not as successful as it might have been. I take it that there are other examples of unsuccessful PFI-type projects that, perhaps for reasons of commercial confidentiality, you are not prepared to share with us just yet.

I will give two other examples. The Stirling centre did not go perfectly. We were slightly late and we paid the penalty. It was acknowledged that it was an extraordinarily aggressive programme—no one had ever built a further education college in one year. We did not do so either—we did it in one year and two weeks, and paid the penalty. However, both sides were happy. The penalties worked and risk transfer clearly happened. That was largely down to an extremely ambitious programme and, on our side, a form of construction that took a risk. As it happened, that did not turn out in our favour.

The Inverness airport terminal, on the other hand, was finished on time, but we knew that the client was not ready to move in. We could have turned round and said, "How inconvenient. We've got a terminal and we'd like you to take it. You're contractually obliged to take it." We did not do that. We held on for a week until British Airways was ready. We did that because we believed that it was in our commercial interests to start off on the right foot with our only client.

When we finally opened, we could not get a liquor licence, because Highland Council does not sit very often. Our solution was to give drink away. We did not publicise the availability of alcohol, because we could not. The Press and Journal ran an article along the lines of "No licence—red faces at terminal", but the real story was that if you turned up at Inverness airport in the first week, you got a free drink. We gave it away. It was the most expensive round I have ever bought. [Laughter.] Although perhaps not as big a round as you might think.

I do not have an example of a bad PPP project to tell you about, but the public sector has learned a lot about developing contracts through the use of PPP. In Glasgow, we have negotiated a contract for ICT in primary schools. It is not a PPP, but we used the same techniques of output specification.

The difficulty is that the contract is short term because the funding is short term. We found that the private sector is much more reluctant to take on risk for a short-term contract because, of course, if things went wrong, it would have very little time to recoup money that it might well lose. On a 10-year ICT contract, the private sector would have more time to recoup money if something went wrong.

I would not like to comment on unsuccessful projects. An earlier question was about whether the public sector was learning anything about good practice from PPP and I think that the answer is yes. What supports our project—which is complex, multi-site, and covers ICT and accommodation—is a very efficient and well-run helpdesk. End-users have to make only one phone call to report a fault, no matter what it is—a broken window, a computer not working, or something to do with the standard of cleaning. We have benefited from a one-stop shop where people can register that something needs to be fixed and the council is considering how to replicate that service.

With PPP, the service offered by the helpdesk has been reinforced by the fact that people can track the issue that they have raised, that time constraints apply in responding to any difficulties and that there are financial penalties if those time constraints are not met. Although we would not go as far as penalising ourselves in-house, we are learning lessons from having a service that allows people to raise a problem quickly and easily and have it rectified or remedied.

For example, a few weeks ago, when we thought that we were facing a civil emergency—which in the end did not happen—and needed a secondary school for emergency accommodation, it took only one phone call from the council saying, "We might need a school at 10 o'clock tonight to house several hundred people," to make the arrangements. We were able to do that because the helpdesk operates 24 hours a day, 365 days a year. As a result, we are now considering whether other council services could benefit from such one-stop-shop support.

That example demonstrates the partnership that we had, which was the final element of the PPP. I will mention another example, this time in relation to ICT. We have some residual ICT contracts in schools and because equipment does not belong to the PPP it is very difficult for schools to know whom to contact when something breaks. However, in the spirit of partnership, the ICT providers very kindly allowed all calls to go through them and passed on calls about equipment that was not their own to the right providers at no charge.

PFI/PPP contracts can be extremely long term—you mentioned a 10-year ICT contract. For other kinds of project, the typical contract might be for 25 or 30 years. Over several decades many aspects, such as service requirements, population and so on can change. Do PFI and PPP contracts lock in a service in a way that means that flexibility might be lost? Furthermore, is it possible to build in maximum flexibility to cater for future changes? If so, how have you done that?

One of my more pleasant duties during the contract negotiations was a visit to Paris to see a number of schools that had been designed and built by a quasi-PPP. The big lesson that I learned on that visit was that that partnership had not thought through the question of how to build in change. Our contract recognises that things will change and as a result contains a change mechanism. Furthermore, the tendering arrangements for implementing that mechanism are in place, which means that benchmark costs have to be specified.

Some future changes—for example, changes in law—might be the contractor's clear responsibility. With the change mechanism, however, Glasgow City Council would be able to add a new subject to those already being taught in school as cost-effectively as by any other means. It is important to take the possibility of change very seriously at the onset of the contract. I was lucky to see the impact of the lack of a change mechanism in a different country. People must face up to the fact that things are dynamic, not static, and that the contract must reflect that dynamism. The change mechanism is therefore a central feature of our overall contract.

Obviously, your experience of implementing ICT via PPP was positive. Would it be fair to say that the success or failure of an IT project quite often has nothing to do with the financing, but with other project methodology and management aspects?

Yes, we both agree with that. As Ian McDonald said, the finance is just one part of a project. In our case, it allowed us to do something over ten years that we would normally have only two or three years to do. We have all learned how to write good contracts and to be tight on the specification. Lots of things sit away from the finance—that has been said by everyone around the table.

In our sector, the two biggest ICT projects have been done by traditional funding. Most colleges are linked for broadband to the super joint academic network—superJANET—which the universities first developed with finance from the funding councils. The university of the Highlands and Islands, which represents a major ICT project, has been funded from public sector sources rather than through PFI. In-house teams have been developed to manage those projects.

Apart from that, each college has developed solutions—some good, some not so good—in partnership with contractors. I support the point that a disadvantage of the traditional public funding route is that horizons, planning and relationships are short-term. I agree with what has been said about the Glasgow experience—that there are advantages in long-term relationships, particularly for ICT.

Taking capital items off the balance sheet is one of the attractions of PFI. What are the implications of that for revenue budgets? How far ahead have you projected? Will there be an impact on budgetary discretion in future and have you been able to quantify that? Could we use alternative vehicles—such as not-for-profit trusts—that would take the capital items off local authority or college balance sheets? What work has been done on that? Those questions are first for Ian McDonald.

Affordability was a big issue for the council on day one and the seriousness with which it was treated applied to the 30-year length of the contract. The council was comfortable that the affordability levels were set, as it could project forward what those levels would be over a 30-year period.

I was under severe instructions about the overall affordability limits that the project had to meet. Those limits were delivered on, so the council is aware of what the costs of the project will be on a year-by-year basis over the 30-year period. We made every effort to ensure that the affordability had a flattened profile, so that the project would not become an increasing burden to the council, but—relative to on-going inflation—a decreasing, although not a zero, burden. We took overall affordability seriously.

Any council would examine with great interest resources that would improve public service delivery. We would welcome any opportunity that enabled the Scottish Parliament to support local councils to further improve public services. We would not rush to make any value judgments for or against alternatives to PPP.

There are possible refinements of the PPP process, which—as we work through it—might give greater comfort to the general public and the public sector. Improvements can always be made. We would welcome any opportunities to give serious consideration to strategic alliances that would release additional funding to support what we all want—improved public services.

You said that you had parameters within which affordability could be calculated. What percentage of your revenue budget would you have deemed to be unaffordable?

Clearly, it was not a personal decision. We had a rough figure, which we would not have been able to go above. We knew what was affordable to the council, taking into account level-playing-field support from the Scottish Executive. We realised the efficiency savings to the council from rationalising and we knew what the on-going budgets were. When they were added up, we realised what the council could afford and we did not stray above those levels.

Are you saying that level-playing-field support was crucial to your calculations on affordability?

Absolutely. Without level-playing-field support, the project could not have happened. The council's ambitions would have been significantly reduced. It would have needed to consider much longer time scales. We would certainly not be sitting here today looking a few months down the line to everything being complete—we would be looking several decades down the line before everything was complete. Without level-playing-field support, the Glasgow project would not have been feasible.

In a sense, the level-playing-field support was an immediate capital injection.

Yes—it was absolutely central. It gave us the opportunity to realise our ambitions in a short time scale. We were comfortable in that the level-playing-field support would be available over the 30-year period of the contract.

I am not clear whether the level-playing-field support that you are talking about is a one-off cash injection or a guarantee over a longer period.

It is an annual payment to the council to assist in the overall annual charge. Level-playing-field support is known to the council on a year-by-year basis over the entire project.

It is important that we do not have access to level-playing-field support in our sector. The grant support percentage to which the funding council may go is limited to 50 per cent. We cannot get up to 80 per cent support, which is available to other sectors.

I would like to return to trusts. The UHI Millennium Institute is a sort of trust that operates on behalf of the colleges, which deliver the teaching—that had not occurred to me until the question was asked. It operates as a public service trust rather than as a PFI or PPP trust. We are moving towards more inter-institutional co-operation on capital projects—I am thinking of the Crichton project at Dumfries, which is used by a number of institutions. There is a willingness to consider that approach, but so far—certainly in our sector—only the public service approach is being considered and not yet a PFI approach. That may come before very long.

I want to explore level-playing-field support further. Are you saying that it is simply a fixed payment that the Executive will give you each year for the next 30 years? Is that how it works?

Yes. Councils were invited to make bids to the Scottish Executive three years ago, I think. Bids were made and ours was successful. The outcome was a year-by-year allocation to Glasgow.

I understand that. Along with the rate support grant, therefore, there is an element that reflects level-playing-field support in the project. In effect, part of the rate support grant is hypothecated for the project. Is not there a danger, as with all such hypothecations, that future Executives will simply consider the total of your rate support grant and that that might not increase in some years as much as it might have because the level-playing-field support has been taken as part of your rate support grant?

I am happy to respond to that. The reason that it was called level-playing-field support is that it was basically the same arrangement as for conventional funding of capital. When we made our bid three years ago, if councils got permission to invest capital funding in schools or other services, the council got pound-for-pound support. From that point of view, the mechanism was no different.

Level-playing-field support repays the capital debt over the period of the project. That debt is fixed for the period of the project. It does not fluctuate, so the element from the Scottish Executive will be more than sufficient for the council to meet its overall capital costs. The figures that fluctuate are those that are affected by inflation, not by the capital costs. Through the project, the council receives energy, cleaning and so on—all of those services are subject to inflation. For our project, the winning contractors had to bid a division of inflation—they had to bid for the inflation factor year by year, divided by a number greater than one, so that the answer was smaller. The on-going running costs of the project can increase, but they increase anyway. They are governed by having to be less than the rate of inflation.

Perhaps that is not the clearest answer. I am saying that level-playing-field support goes to the fixed element of the project in a year-by-year payment.

We might discuss that with you in more detail, but Elaine Thomson has a question on staffing.

People have suggested to the committee that one of the results of PFI/PPP has been a degradation in the terms and conditions of staff—lower pay, less security and so on. Andrew Gordon says in his submission that

"our experience is that public sector employees do not enjoy terms and conditions which in the round are materially better than those of equivalent private sector employees."

Will you expand on that? This morning, witnesses said that PFI sometimes has negative impacts, particularly for people at the bottom of the hierarchy.

We have experience of six projects. The submission mentions Inverness airport and the Stirling centre, which are both unusual. The centre was a start-up and had no existing staff and all Inverness staff were already in the private sector. We have considered the health service and its staffing with our facilities management partners and our experience is that health service workers in general are not paid better and do not have better terms and conditions of service than their peers in the private sector. Therefore, when we have asked our FM providers if they would be disadvantaged, they have said that they have no problems whatever with the terms and conditions of service. There is a suspicion—which may be unfounded—that the NHS is slightly soft on sick leave, but I have never seen any evidence for that. People who we were asking to take those people over told us that the basic terms and conditions did not pose a problem.

We routinely offer not to change the terms and conditions of employees for at least five years without their prior consent, because we do not want to create fear where there need be none and because it gives us a competitive advantage if we can show that we approach matters in such a way and can deliver a work force that does not fear transfer.

Before best value and contracting out were being reviewed, PFI could deliver more constant employment to a support worker than they then enjoyed, which is a paradox. Some support services were routinely market tested every three years, which meant considerable uncertainty and fear of transfer—rightly or wrongly. However, under PFI, there was potentially a 25-year employer with a 25-year contract. That is sometimes lost in the debate, which is contaminated by the debate on the politics of PFI.

Those who represent transferred workers will say in private—as they have said to me—that, surprise, surprise, they have known private sector employers who are as good or better than public sector employers, but that does not come out in public. I am sure that there are private sector employers who are worse than public sector employers, but I do not think that there is anything intrinsically wrong with private sector employers. Workers' rights are protected under TUPE and ensuring that fear is removed is part of the management of change—which is part of outsourcing.

Equally, in our experience, it has never been the case that soft services have made a capital project work, although I accept that that is the case for others. The public sector already procured those services relatively efficiently. If hospitals work on a design-build-maintain basis, they work whether soft services are put in or out. We have always had an open mind on that. Our view is that trusts should decide whether they want soft services in as a management strategy, not in order to make projects work.

Do the other witnesses wish to comment on that? Is that the experience elsewhere?

I support what has just been said. In our sector, the question whether soft services are included is separate from the question whether a building project will work. We are more suspicious that projects will not be so easy because of the unpredictability of use. We are in a volatile market and our projects are in centres of underprovision, which means that demand is strong. At the moment, the race is to keep up with demand. We have no experience of what happens when demand does not meet expectation, which is a different situation.

Support staff are a staffing issue that is likely to be sensitive for us. Managing a facilities management contract is different from running a building, whether it is leased or owned. We do not have enough experience to say what that will mean for staff. Such situations are dealt with case by case.

I will conclude the session by firing the same question at all the witnesses, beginning with Sandy Bremner. From your experience of the PFI process and of considering the options, do you have a proposal that would bring about greater value for money in public sector capital procurement? You can mention a big idea or a little idea or something that gets in the way of delivering better value for money.

It is difficult to give a specific suggestion. One experience that we all have is that the PFI market is maturing. There is a lot of experience in the private and public sectors and, like all things, as it matures it gets better—that is my theory.

Standardisation of contracts could benefit PFI and PPP. The standard form of contract was put together by the Treasury task force, but it is not always followed, nor should it be, because specific projects require specific contracts. However, there can be one form of contract for one bid and an entirely different one for another, which creates problems in understanding the contract and bidding processes. It also creates issues for funders, who wonder why the contracts are different and whether there is an inherent problem. Standardisation would give more comfort and help the process to move forward by smoothing the bidding and operational processes.

I did not think about the question in advance. Advantages are likely to be gained through small technical improvements in the procurement process. From my experience—which is evolving—the process does not operate well universally. Many technical improvements can and should be made by learning from experience. For example, streamlining and—more important from the public sector's point of view—the competitiveness of procedures can be improved. The procedures are competitive in some procurements but less so in others.

If PPPs are to be expanded, for example, into transport, the public sector should consider making a capital contribution to projects. It is often said that public sector funds are cheaper than private sector funds, which is true. However, if the overall value for money in PPP projects is established, that is because of the risk transfer. That principle will not be eroded if the public sector makes only a fixed capital contribution, which can be used to support a project that might not work otherwise. That has not been explored much, other than in cases where European funds may be available.

So you are saying that the public sector might be able to rebalance the partnerships with a greater capital contribution and that that would be advantageous.

Yes, it is worth thinking about.

I was going to answer a different question. I thought that you were asking what we have learned through this process that might help us in other forms of procurement. The one thing that I have learned is that if I can get longer-term planning for resources, I can do a better job in the public service. Whether that approach is called PPP or something else is irrelevant. If you can give public services the opportunity to make longer-term financial plans, it would be very helpful.

I agree that the standardisation of contracts would be helpful. It would be beneficial to us all if we understood what has been described as the black art of risk analysis. Risk analysis and risk transfer are difficult for us all and we would like to understand them better.

I would be more than happy to answer the question directly. All young people in Scotland deserve to go to a decent school building to be educated. Sadly, as we know, that is not the case, given the school estate that has been inherited by the Scottish Parliament and local councils. The investment that is about to be made by the Scottish Executive will make a significant improvement to many school buildings across Scotland.

In terms of better value, I would like to see more young people in more schools being able to benefit from the funding available to the Scottish Parliament. Two aspects are important if that is to be achieved. First, the money must be spent as efficiently as possible so that it goes as far as possible. There is no question but that larger projects—no matter how they are put together—are more efficient and deliver better-quality facilities at a lower cost. Secondly, we must remember that we are dealing with a general Scotland-wide problem. People may talk about the standards that are required under the contract, but it is important to ensure that the funding goes as far as possible by considering standards across all accommodation. We do not want to have one-off projects providing a lavish facility in one school that leaves the rest of the estate in a condition that is less than good.

To summarise, we look to the Scottish Parliament to ensure that the funding that can be made available for the improvement of schools goes as far as possible to provide as many young people as possible with an improved school facility.

Those are utilitarian principles.

I agree with encouraging standardisation and common methodologies. My only caveat is that that must still allow for the possibility of innovation. That will not happen in most cases, but we must keep open the 1 per cent possibility that it might. We do not want people simply to say, "We didn't do it that way last time".

We should follow the guidance on openness in calculating value for money. We do not just have to deliver value for money; we must be seen to deliver value for money. The whole PFI/PPP thing is probably seen by the public as a giant political fix for a giant political problem. As long as the debate is conducted on that level, we will get no further.

I take it that by that you mean there is no need for the level of commercial confidentiality that currently exists post-contract. Openness would allow people to measure the outcome properly.

As we said in our submission, there are very few exceptions to that. We cannot see why there should not be a comparison.

Is that view shared by the other private sector witnesses?

After the contracts have been signed, there is no reason why people should not see what is involved in those contracts.

My sector is seeking to enhance quality and capacity at an affordable cost. The problem has always been the funding gap. That is true whether we go for public finance or PFI. There is also a gap in expertise to be bridged. Others have suggested solutions and we have proposed some in our submission. Any new PFI client should be starting at a higher level of expertise and confidence in the process than was possible for the pioneers.

Are you saying that the gap is on the client's side and that we should consider providing training and support for clients going into PFI?

I think so. The problem has been with accessing that expertise at an early stage.

I thank all the witnesses for coming. It has been a valuable and interesting session. We are keen to learn from your experience and expertise.

The future of our investigation will involve more detailed consideration of specific PFI projects. We will not publish our conclusions immediately. We still have some way to go but we have learned a lot from what you have told us today. If there are further details that we need to follow up with you, we will be in touch. Thank you.

It is worth putting on record that Arthur Midwinter has been appointed as our budget adviser. The committee had bid for that appointment.

As I was a member of the Scottish Parliament Corporate Body until last week, it is appropriate that I stand down for the next item. Elaine Thomson will take the chair.