Finance Committee

Meeting date: Tuesday, February 23, 2010

Official Report 514KB pdf

---

Forth Crossing Bill: Financial Memorandum

Just to give the committee a financial context for the matter in hand, we say in the policy memorandum that our present estimate of the total Scottish Government capital budget for the period of the crossing’s construction is that it will be in the range of £2.6 billion to £2.9 billion a year. Although that estimate takes account of the Chancellor of the Exchequer’s pre-budget statement of December 2009, the figure will not be confirmed until after the completion of the next United Kingdom spending review, which we assume will start soon after the UK general election and conclude in autumn 2010.

In paragraph 143 of the policy memorandum, we suggest that annual expenditure on the Forth crossing will peak at £500 million. However, subsequent work on the phasing of the project suggests that it may be possible to smooth the costs over the years of construction to something of the order of £200 million in the first year, £350 million in the four following years and £100 million in the final year. On that basis, there would be four years in which the expenditure on the crossing would peak at some 13 per cent of the Scottish Government’s capital budget.

Given the expected reductions in the capital budget, ministers are planning to determine capital spending on the basis of priorities across the whole Scottish Government, which would mean no more automatic rolling forward of departmental or portfolio capital budgets. Ministers are clear that the Forth crossing will be one of their highest priorities when it comes to the capital side of the next spending review.

I can understand the complexities involved, but what does history tell us about how accurate past estimates of optimism bias have been in such projects?

I will start with an area that Mr Dow touched on in his remarks about the phasing of the costs. Although the overall total is very important to us, given the scale of the project the proportion of spend in any one year is significant both for this project and, more important, for other projects. Mr Dow talked about a pattern of spending and I do not think that it would be unfair to characterise what he said about smoothing the spending as aspirational. If my calculations are correct, that was for an outturn of £1.7 billion. If the total cost of the replacement crossing were to grow towards the upper end of the forecast, would the phasing be the same? In other words, would the £200 million and the £350 million simply be uprated by the same factor or would the costs bunch differently within the years?

That is primarily an engineering question. My understanding is that a project of this nature has an optimum duration, so it would probably bump up the costs within the same timeframe rather than stretching the timeframe.

Perhaps I can help here. The project comprises three contracts: the principal contract for the bridge and crossing; and two smaller contracts, for the new junction on the M9, and for an intelligent transport system north of the crossing in Fife. The bulk of the costs will be absorbed in the principal contract.

In turn, the principal contract has three parts that need to be considered. The first is the construction of the bridge. We are presuming a five-and-a-half-year contract period for that, which is driven by the length of time that it will take to put in the foundations, build the towers and put the deck in place. That will have a certain progression, so there is a degree of confidence in the expenditure rate.

There are also road works to the north and south of the Forth. The road works to the north are fairly complex and need to be done so that they are finished at the same time as the bridge is built because of the way in which they weave into the junction at Ferrytoll. We expect those to take about three and a half years. The lump of money for that will be spent at the back end of the five-and-a-half-year period.

The other section is the road works to the south of the Forth, on what is in effect a greenfield site. When everything else is done, the bridge will have to connect to the road network. However, that section of road can be built at any time during the five-and-a-half year period, and it might take two to two and a half years. The positioning of that chunk of work will give us the opportunity to move prices around. Left to its own devices, a contractor would need to think about whether such a chunk of work should be put off to the very last because it does not need to be done until then, or whether it should be brought forward so that there is a continuous workload of road works; various strategies can be deployed. We are engaging with the contractors on that with the aim of creating as smooth a cash flow as we can, but with the underlying advice that we are looking for the most efficient and economical construction process that we can get.

To answer the question, we are still discussing the issue with contractors but there are certain clear elements that drive the whole process.

15:00

There is one answer for Scottish budgets and one answer for Treasury budgets. For Scottish budgets, the new international financial reporting standards apply but, for Treasury purposes, we in effect go back to the old regime.

It is a novel concept to have two sets of books to move things back on balance sheet, but anyway—

Yes. Why is that the case?

You have inspired a sub-question.

Is that an option? Are you saying that it is an option for the UK Treasury but not for the Scottish Government?

No. We do not believe that any design changes will be required at this stage. As I explained, we went through an elaborate process of identifying the best option for the location of the crossing and determining whether it should be a bridge or a tunnel and, if a bridge, what sort of bridge it should be. The Cabinet Secretary for Finance and Sustainable Growth recommended that it should be a cable-stayed bridge at Queensferry. In the following year, we looked at how the Forth road bridge could contribute to the overall solution and value managed our proposals to match that. A great deal of thought has been given to what the new crossing should include. The resilience that we would expect of a crossing that is due to last 120 years has been built into that. We are not expecting any significant changes to be made.

I would not wish them on you.

I will start by looking at the funding method. Previously, you were asked why alternative options, such as use of the non-profit-distributing model, were ruled out. Table 1 on page 36 of the policy memorandum summarises the procurement options. Can you clarify for the committee the reasons for not using an NPD/PPP approach?

So the model was not excluded for accounting reasons; the reason was the condition of the market.

The primary reason at that stage was the condition of the market and the certainty of the design and build approach.

I am surprised to see that, in 2008—when some of the work was being done—PPP deals totalling £408 million were signed in Scotland and Transport Scotland said that using an NPD model for the Borders railway and the Aberdeen western peripheral route was the best way forward. Is that consistent?

What independent report indicated that? Was it simply your judgment?

Was the same basis used for making the recommendation as was used for the Borders railway, which is a PPP project?

I cannot answer that; I am responsible only for the Forth replacement crossing.

The conclusion that you reached on the funding mechanism for other schemes was that they could go forward because they were less expensive than the Forth crossing. Is that correct?

This is something that I get quite angry about, because there are all sorts of opportunities for the situation to be miscast. I welcome the opportunity to explain what the differences are between major bridges across the world. Accounting practices of countries come into it. For example, there is no taxation whatsoever on Stonecutters bridge, and no financial charges are applied to its costs. You will find that, despite what you might think, every country has its unique way of accounting for things.

The other characteristic is that it is not an open book situation. In many countries, all you will ever see, if you are lucky, is the tender price, not the outturn price. That comes into the next item: the contract strategy. As John Howison has explained, we are adopting an approach that gives price certainty post-tender. Other bridges have been constructed on very much more of a cost-plus or target basis, by which I mean that the tender price bears no resemblance to the outturn cost. I am afraid that you cannot make people tell you what the outturn costs are; they just will not do it. With very large projects there is always a problem with finding out what the end cost is. I think that that is what you are suffering from with many of these situations.

I turn from the financial infrastructure to the physical issues of building a bridge. Many of the people who talk about bridges in the discussions that I read in the press do not understand the anatomy of bridges. For example, our bridge is a bridge. It virtually goes from bank to bank, and on any world standing it is a big bridge. Many bridges that are cited in the press, such as the second Severn bridge or the Øresund crossing, have a small bridge in the middle, but our bridge is two spans of 650m with back spans that are almost 2km long. The Øresund bridge, by comparison, is less than half of that. What makes Øresund is all the approach viaducts that come to it.

Calculating the cost of the project divided by the length of the bridge does not give you the right picture, because Øresund is overwhelmingly a very cheap viaduct with an expensive bridge—we are not in that position. We designed the Øresund bridge, which I sometimes describe to friends—disingenuously—as a bridge over a puddle. Not only is the sea shallow but the foundations were predictable because of the consistent geology. The foundations solution for Øresund was simply to float out caissons and sink them to the sea bed. That is not the case with the Forth bridge. The Forth has variable geology and deep water—it shelves away quickly. One can become beguiled by the mudflats on either side of the Forth, but two deep channels run between them. In comparing the Forth with Øresund, the first thing that one has to do is to factor out the viaducts and look at the cost of the bridge component. When one does that, one finds that the Forth bridge compares very comfortably.

I am sorry if I am going on too long, but I hope that this is helpful.

There are also the physical constraints of the location. We are building at 58° north, which is quite a hostile environment, so we have built that into our cost calculations. As I have said, we produced a bottom-up estimate, in which we made allowances for downtime because of weather and so on. Although Hong Kong, for example, is subjected to typhoons, its climate is benign and predictable for most of the time, so there you do not get the extra zeros.

Last but not least is the functionality of the bridge. Our bridge has been designed effectively for three-lane running, yet when I hear reports they always say that our bridge has only two lanes in each direction. We designed the bridge with three lanes in each direction because we want to use the additional lanes for buses. Comparing our bridge with bridges that have three lanes in each direction and saying that our bridge has only two lanes in each direction totally misrepresents the design and functionality of the bridge that we have embarked upon.

In summary, you have to consider the accounting practices in each country that you are looking at. You have to consider the honesty of the costs that you are looking at. As the committee can see, ours are very transparent. You also have to look at the geology of the area, the depth of water and the functionality. If you take those factors into account in a qualitative way, you will find that this bridge is no more expensive than any of the other European bridges.

I should also say that we designed Stonecutters bridge. The labour cost environment in China is totally different, and Hong Kong is now dominated by China.

All sorts of subsidies go into certain materials in the economy—I am referring to no one particular country—all of which distort the profile. I can assure you that, on the basis of the analysis that we have done, this bridge is not a very expensive bridge; it ranks right alongside its peer group.

Current directors of the SFT were involved in the analysis that we did when they worked for a previous organisation, but the SFT’s contribution has been far more targeted on helping us with governance and examination of the process than on the funding aspects of the project.

The decision on funding the bridge was taken in December 2008. From that point onwards, we have been making progress in finalising the design and the procurement processes. At that stage, we set in train a route, which we have followed.

It would be a bit strange if, following a year-long process to find a preferred bidder, you did not think that you had done it properly, would it not?

It will not be a matter of whether we think that we have done it properly; the SFT will provide an independent view on whether it has been done properly, which will serve to advise ministers and the Cabinet before they sanction the award of such a significant contract.

Mr Dow mentioned the ability to “smooth” the path of capital spend. Can you give us more explanation of how you think you will be able to do that? If I picked you up correctly, instead of there being a £500 million peak, the cost will be £200 million or £350 million in each of four years. Is that right?

15:30

Yes.

I hope not.

Is Mr Dow going to enlighten me on the discussions with the Treasury?

I am indeed. The Cabinet Secretary for Finance and Sustainable Growth wrote to the Chief Secretary to the Treasury on 27 November 2008. The proposal at that stage was that the Treasury should allow the Scottish budget to be reprofiled to bring money forward to the period 2012-13 to 2015-16, which will be the key construction period, with a corresponding reduction in the following years through to 2031. The reply from the Treasury stated that it was not possible for it to allow any advance of budget cover in the way that had been suggested. In correspondence and at a meeting in March 2009, the Treasury made some other suggestions as to how the Scottish Government could find the budget cover for the project. I can list the suggestions if you want, but the conclusion is that there is no “get out of jail free” card.

The bill is for a specific piece of infrastructure. We are talking to Fife Council, the City of Edinburgh Council, West Lothian Council and the south east of Scotland transport partnership about a public transport policy. Indeed, a number of the public transport elements that were announced following the strategic transport projects review will form part of the project. They all have their own routes to securing consents and to procuring land and they do not need to rely on the bill to do that. The bill simply covers the bridge, its approach roads and essential infrastructure, such as the modifications that we will have to make to the park-and-ride site at Ferrytoll. The bill does not cover remote park-and-ride sites, which can be developed independently.

I am probably going to illustrate my ignorance of public transport strategy, but as I understand it, you propose to widen the hard shoulder, as it were, so that it can take buses. It was not part of the original thinking to say that, if public transport cannot use the existing bridge, it will use the new bridge.

In the design-and-build contract, there is a five-year defects maintenance period. In effect, a retention will operate so that the contractor is not paid the full amount of the construction price until the works have gone through that five-year maintenance period. Funds will be released as the maintenance period progresses, which is what the £196 million figure to which Mr Chisholm referred shows.

Yes—that is a very superficial way of looking at the situation. It makes headlines, but misses out many issues. John Howison explained the bridge’s functionality, so it is wrong to report matters in the way that you have described.

People are worried about the overall cost, given the squeeze on capital budgets that is coming, and some people have been critical of the bridge because of that. A new entrant among the critics was in Friday’s Scotsman: Bill Jamieson, its executive editor, suggested that we could delay the bridge for a little while and see whether we really need it.

The focus is always on the suspension cables—the main cables—of the bridge. One thing that is not reported widely enough—Barry Colford would be able to speak with more certainty on this—is that the catalogue of issues with the existing bridge is very, very long, with the cables being just one of them. There are expansion joints at 60ft intervals across the bridge as you go across—the bump, bump, bump—that are reaching a state at which there has to be some major works on them. That is evident if you stand on the bridge and see a 40-tonne truck go across it. The other issues are related to the anchorages. There is a long catalogue of issues with the existing bridge. It is wrong to consider that it is simply a case of replacing some cables.

The true story—the better picture—is that to bring that bridge back even to the level of functionality that it had when it was constructed, we would have what is estimated to be between eight and 10 years of delays. Our studies have reinforced FETA’s studies about the delay and disruption to the crossing of making those repairs. I want to reinforce that the issue is not just the cables.

There was market testing in relation to the construction companies that would front up a consortium, but not in relation to the financial markets.

So, there was no market testing of the financing options. Is that correct?

That is an option within the contract that we are procuring. Again, we are exploring with each participant whether value for money might be gained through providing an advance payment. If an advance payment is made, it will be underwritten by an on-demand bond so that any money that is paid is not at risk.

Have advance payments of such a scale been made on previous projects in Scotland?

Paragraph 295 in the financial memorandum states:

“Our approach provides for a lump sum design and build contract”.

That is not technically true. Once the principal contract has been concluded, the taxpayer will pay for it.

It is not a single lump sum, but it is a fixed amount of money.

Mr Dow described the context in which the estimates for the overall capital budget have been made. I understand the caveats that he placed on those estimates. Is the cap £350 million or £300 million?

It is £350 million.

You said that the figure represents 13 per cent of the Scottish capital budget. Is it correct that the calculation includes capital grants to local authorities and capital for the health service?

I do not have that percentage to hand. You were correct to say that the figure that I quoted includes the elements to which you referred.

I will provide some background to our approach to the matter. The construction industry has a series of indices for commodities and activities—for example, labour costs or the price of fabricated steel. We analysed the historical record for about 20 of the different commodities that make up the project, looking at trends over the past 18 years. We coupled that analysis with an examination of what we think the industry and the world economy—we are dealing with an international project—will do over the period during which the crossing is constructed. That involved some crystal-ball gazing, but we have built up a picture of what economic activity will be over the next eight years. That has given us a view of what we think median construction inflation will be in each of the commodities.

Interestingly, some of them—construction plant, for example—have been remarkably flat and there is no reason to believe that they should be otherwise. Others are much more volatile. We have made a projection for each of the different commodity areas, then we have summed them, taking account of their contribution to the project. That then gives us this trend, which we have also had independently reviewed to see whether our median is correct. John Howison explained to you how we then looked at the extremes of the trend, which means looking at historical trends.

I guess that it is one of the elements of a long capital project that you would have concerns about anyway. It just struck me that it is quite difficult to have any certainty in the current economic climate.

I would like to go back to something that Sharon Fitzgerald talked about. She explained to Malcolm Chisholm about retention and said that it would be in place for five years. Am I right in thinking that in the normal scheme of things any major design defects are not covered other than by indemnity insurance?

So that would cover major design defects in the longer term.

Yes.

That certainly answers my question and is probably the answer that I expected. This is a major capital project in Scotland. Plenty of evidence everywhere supports beauty in design as well as functionality—I see Mike Glover nodding; as an engineer, he is obviously interested in design. It would be a shame to compromise on beauty in design, its effect on wellbeing and its associated benefits for the surrounding areas.

That is why the design is very prescriptive. It has been presented to Architecture and Design Scotland, which has also considered the landscaping setting. That is in the bill and in the environmental statement.

The issue is how to deliver it at best cost.

Convener, Mr Glover is desperate to say something.

We have rightly got to the heart of the matter. I will allow two very quick questions to finish off.

For the spending review, it would be necessary to examine projects in the round and across the Scottish Government—everything within the wider budget that you identified.

You have estimated costs of £10 million for compensation but, since the memorandum was published, you have received 89 objections. Are there likely to be any cost implications from those?

Compensation will be dealt with by the lead committee. No doubt it will pursue that important issue in detail.

Do the witnesses have any final comments to make?

No.

Item 2 is evidence on the financial memorandum to the Forth Crossing Bill. The committee agreed to adopt level 2 scrutiny for the bill, which means that we agreed to seek written evidence from financially affected organisations and to take oral evidence from Scottish Government officials. The written evidence has been circulated.

I welcome David Dow, finance team leader for transport at the Scottish Government, and, from Transport Scotland, Sharon Fairweather, finance director; Sharon Fitzgerald, commission, procurement and contract adviser; Mike Glover, commission project manager; and John Howison, interim project director for the Forth replacement crossing. Do any of our witnesses wish to make an opening statement?

14:45

I am wary of financial predictions that suggest things. Obviously, we are all concerned about estimates that fail to be delivered, and projections for things that end up costing far more than planned. I want to ask you about two specific issues. First, what are the key risks for over or underspend, and how does the risk allowance provide for them? Secondly, how accurate have Transport Scotland’s estimates of optimism bias been in past projects?

The estimating that we have undertaken for the scheme has been undertaken by people who are accustomed to working on this size of project and who are familiar with the scale of the difficulties in front of them. The estimates have been broken down into what we expect the cost to be, based on our price data levels at 2006 prices. The process then rolls those forward to today’s prices using indices of movements in materials, labour and plant, which reflect what has happened over the intervening years. The estimates are then moved forward to outturn costs, when the pound notes are actually spent. The process is relatively simple and logical.

You asked about the risks and the accuracy of the estimates. They are one and the same thing. At this stage, we are fairly confident about the underpinning cost levels of the project, but the big issue will be the value that the two bidders put on the works when they submit their tenders in December. Thereafter, the form of contract that we are using has a good record for cost control. Having used that type of contract before, we would expect any overrun of the outturn price over and above the tender price to be limited to about 4 per cent of the tender value. The one exception is that, because of the duration of this contract, we will be taking risk for the fluctuation in prices of materials as we go along. We will be very much in the hands of inflation during the period following the award of the contract. Mike Glover might want to say more about that.

That is very helpful. The policy memorandum mentions the decision to select the conventional procurement method as opposed to any of the other variants that were selected. From memory, I think that the global costs under conventional procurement relative to public-private partnership were there or thereabouts the same.

When we had a debate on the funding of the bridge last January, the cabinet secretary majored on the budgetary impact of using PPP, because of the application of the international financial reporting standards. I think that I am correct in picking up from paragraphs 125 and 126 of the policy memorandum that, because of the move to Eurostat guidance for PPP, there would not now be that budgetary problem in using PPP. We therefore come back to the fundamental decision about the merits of private finance in respect of the certainty of contract, people being willing to bid and, of course, the overall cost to the taxpayer. Can you confirm that that assessment is right in respect of the budgetary impact? Am I right to say that, under current guidance, the figure of £2 billion would not be brought on to the balance sheet in 2016 or whenever the bridge is completed?

Thank you for that information. If it is possible for you to supplement it in writing, that would be helpful. We appreciate the tutorial.

It might be helpful if you amplify that point in writing.

One thing that strikes me about the financial memorandum is that the percentage optimism bias is higher for the bridge than for the network connections. I have never been involved in procuring a bridge before, so I have no particular expertise in the area but, intuitively, that seems to be understandable, because you would assume that the bridge is probably more difficult to construct than the network connections. However, in the case of the risk allowance, the opposite seems to be true: the percentage risk allowance for the likely cost of the network connections seems to be higher than for the bridge. What drives that in respect of your assumptions? At first glance, from a layman’s perspective, one would not expect that.

In quantum, there is more risk money against the bridge. As members are aware, the bridge accounts for about 70 per cent of the total cost. As John Howison said, a bridge is quite a discrete entity and constructing one is almost like production engineering, in the sense that the project can be broken down and specific risks can be identified. On land, unforeseen ground conditions dominate, especially on the east coast. The geology across the Forth is very variable. One of the biggest elements of variability is the area’s industrial heritage. It is not readily acknowledged that certain areas were large areas of industry. That is especially true of the area around St Margaret’s marsh, underneath which there are many old railway sidings. A greater risk attaches to foundation works there, because it is not a greenfield site by any stretch of the imagination.

Paragraph 269 of the financial memorandum states:

“The governance structure and the project’s disciplined approach resonate with the recommendations of Audit Scotland, as they relate to its recent review of major capital projects in Scotland, and the conclusions and recommendations of the Holyrood Inquiry.”

This may be an example of officialese confusing me, but does the statement that

“The governance structure and ... approach resonate with the recommendations”

mean that they are fully compliant with the recommendations or just that they are somewhat in tune with them?

Has the likelihood of significant changes being made to the design before or during construction been estimated? I was part of a painful process in relation to this building.

Does that reason still stand?

Yes.

So you are saying that it was not possible for there to be separate contracts for the north and south network connections.

That is not the same as it not being possible.

Is the value for money assessment guidance used consistently across projects?

Yes.

It has been consistent.

Yes.

The value for money assessment from Transport Scotland says that option B—NPD/PPP—is better value for money. I am exploring why it was excluded from this project when it was not excluded from other projects. If we are determining the robustness of the finances, that is relevant. It is up to you, convener.

Joe Fitzpatrick?

Most of the questions that I wanted to ask have been asked. When the bill is discussed by the public, it is the headline figures that are looked at. Sometimes, the headline figures for this project look to be considerably higher than those for other projects, such as Stonecutters bridge in Hong Kong. Why do they have different figures? We see how much a bridge cost in Hong Kong, compared with what this bridge will potentially cost.

On a point of order, convener. I hope that we are not talking about bridges for which Transport Scotland does not have responsibility.

I hope not. It is a wonderful question.

It is clear that you enjoy the work that you do, but we will finish off there, because other folk have questions.

My recollection is that, at that stage, the SFT was not staffed up to provide such advice.

But given that the SFT exists and that we are told that it has experts in public procurement and all the rest of it who are assisting the Government to get best value for money, has the SFT never been asked to offer a view on whether the Government’s approach is still the right way to fund the bridge?

What role will the SFT have in the final stage? Its submission says:

“A further review will be undertaken by SFT in early 2011 prior to the appointment of a preferred bidder.”

What will its role be then?

Indeed. It is a matter of budgetary management and anticipating a case in which the project is ahead of schedule. We had that with the M74—I am sorry, that is another contract. It is not unheard of for a project to be ahead of schedule and, under the Scottish Government’s budgeting rules, we would have to accept that work was being done. Whether or not the money was being paid for or accrued, it would be a budget hit, so we would have to be aware of that as a pressure from the project and ensure that something else happened somewhere else within the Scottish Government that allowed us to cover those costs without breaching the budget limits.

Two of the questions that I was going to ask have been answered by the previous two witnesses, but I still have a couple of questions. Table 2 of the financial memorandum provides an illustrated spend profile that covers the years 2007-08 to post-2016. The estimate for post-2016 is £196 million. The obvious question is to ask what that will pay for, if the bridge is completed by 2016 as planned.

Sharon Fitzgerald will talk about retention money, which was the basis on which that sum was put forward.

If the dehumidification of the cables proved to be successful, would that not change the scenario somewhat?

It is worth noting that the cables are already compromised—they have already lost their strength. The most that dehumidification can do—we sincerely hope that it will work and we are expecting it to work—is to prevent further deterioration of the strength of the cables. In that situation, we would be able to apply public transport to the bridge, without having to replace the cables.

Sorry—I have lost my train of thought.

I want to go back to the method of funding. Was a market-testing exercise carried out before the recommendations that excluded a non-profit-distributing option were put forward to ministers?

We thought that we had got a full enough understanding of the contractors’ issues, and of the balance between long-term maintenance and the construction process. Members will have gathered from the reporting of the exercise that it was done at a time when the financial markets were still operating effectively, although we quickly ran into a deterioration in the financial markets. Any financial testing would simply have supported our recommendation that PPP looked like good value for money at the time. In the end, the decision was taken against the background of a different and quickly changing situation.

I am asking about the matter because of an answer that Sharon Fairweather gave. The same value-for-money assessment guidance was used for other schemes in Scotland. People were told to use the same approach, but a market-testing exercise on the method of financing for one project concluded that the NPD approach was the right way forward. That conclusion was reached at the same time as you were doing work on the project that we are discussing. I do not know why there are differences.

I am asking such questions because all that we have is a presumption by Transport Scotland about the project. However, it has been said that NPD clearly offered better value for money in another project in which Transport Scotland is involved at the same time as that project.

Let me try to explain again. We went through a comprehensive exercise to compare various financing forms using the same type of analysis as had been employed elsewhere. That exercise reached a conclusion and reported to ministers around October 2008. That market testing involved looking at construction industry firms as the promoters of a PPP-type exercise, but it presumed, without actually talking to specific banks, that the market would be able to deliver funds for such a scheme. The report concluded that PPP and conventional funding could be expected to provide more or less the same good value for money if we used some of the more innovative options that we considered and which could pull the project off balance sheet. At the time when ministers took the decision, the full deterioration in the financial markets had become clear. At that stage, we were involved in looking at other PPP schemes in parallel and we were aware of the difficulties in providing financing.

As David Dow mentioned earlier, we are currently conducting a competitive dialogue with two tenderers. We are discussing with them their technical solution programmes and how their proposals would sit with the budget figures that have been mentioned. Currently, we expect that there will be a mechanism within the contract that will set annual caps, but the level of those caps will be set over the process of the dialogue. The caps will be set in cognisance of the proposals from each of the tenderers.

That has not been done on this scale, but it is not uncommon for there to be advance payment for certain types of project. For example, up-front payment for manufacturing, mobilisation and design of long-lead items has been made, when it has been needed.

16:00

I am sorry, but I do not understand your point.

So when you said that it was a case of prioritisation, you were including all the capital grants to local authorities. The Government points out to us regularly that local authority and health capital expenditure accounts for the majority by far of capital expenditure in Scotland. What percentage of non-health and non-local government capital expenditure does the Forth crossing project represent?

You may provide us with a written response, instead of answering the question off the cuff.

It has been quite a long session. Linda Fabiani will ask the last question.

I hope that that is not encouragement to be quick.

I will begin with a couple of brief financial questions, to ensure that I understand matters correctly. I have listened with interest to what has been said about the financial markets and the changing economic climate as preparation for this long project has proceeded. What is the witnesses’ feeling about the integrity of the construction inflation element that has been included in the terms of the contract? Can anyone be confident about it?

Other defects would be covered under the Prescription and Limitation (Scotland) Act 1984. The defects maintenance period is really about having access to the contractor on site to deal with any snagging items or defects that arise. Insurance could be used not just outside of the five-year period but for issues beyond that. The Transport Scotland team is considering the insurance package for the bridge in detail.

First, it is worth saying that because we need to go through a bill process for approval, the design will be fairly prescriptive at the end of the day and the differences between one bit and another are unlikely to be substantial. Secondly, the pre-qualification process considered the standing and the capabilities of the contractors and their principal suppliers, including their designers. We are talking about designers who are among the top dozen or so in the world, so we are confident that the standards that they will bring to the project will be very high. We also have a specification that we require the contractor to deliver. Those things—pre-qualification surveillance and the design process—can be put together.

The finished product will be very much driven by the lump sum price. However, we have looked at how we should judge the contract and we recognise that there are, shall we say, non-cash costs that need to be taken into account. We focused on four of those: the carbon content of the product, because it can be delivered in different ways with different levels of carbon consumption; the social responsibility element, including employment and training, because we expect a project of this size to leave a training legacy; the contractor’s approach to risk; and the contractor’s approach to management of the process. We have regard to the marking of those elements in the quality assessments, bearing in mind that each point of quality that is added is a surrogate for a pound of money. I hope that that answers your question.

Absolutely.

I have just one question, which I do not really expect anybody to answer—it is about a long-abiding passion. I am probably back to addressing Mr Howison. Academic studies over the years and many other written pieces have covered the fact that throughout Europe and the world costs are often underestimated in major public capital projects, because everybody knows that a project must be finished once it is started. I know that putting outturn costs in the public domain is always difficult, but you guys know them—let us not pretend that you do not. How much notice has been taken of the history of capital projects in the UK and in Scotland—the procurement method for the Parliament building is a prime example—and how confident are you that the outturn will match the estimates?

We are dealing with the unknown unknowns, although the Treasury has given considerable guidance on the outturn of various projects.

The optimism bias caters for the situation. I ask Mike Glover to say something about how the optimism bias has been considered and altered as we have gone through the various stages of the process.

The bulk of the money that we put aside for compensation covers situations in which we intend to buy land. About 45 per cent of the objections that we have received relate to the effect of what we shall call a South Queensferry bypass—the fact that the road works to the south run past the residential areas. People there would be entitled to submit claims—we call them part 1 claims—for injurious affection. We have examined the design of the works and put in mitigation measures to reduce the noise as much as possible. From our experience of building that type of road, I do not expect substantial sums to arise in relation to part 1 claims.

There is a difference between not expecting them and getting them. The £10 million is what you know at the moment. Can I take it from your answer that you expect that you might have to spend some more money on compensation but not a lot more?

I will provide some background on the project. The need to consider a replacement crossing arose from the findings back in 2004 on the condition of the Forth road bridge’s cables. In December 2007, the Cabinet Secretary for Finance and Sustainable Growth announced in Parliament the conclusion of quite an extensive study, which was that a replacement cable-stayed bridge should be built west of Queensferry. At that time, the price was between £3.4 billion and £4.2 billion.

Moving the story along, a rather more encouraging prognosis for the Forth road bridge indicated that, despite its serious shortcomings, it could have a role in the emerging solution for a resilient crossing. Value management identified the managed crossing scheme, which was brought to Parliament in December 2008 and had a reduced cost of £1.7 billion to £2.3 billion. We considered a wide range of funding options and promoting entities, which culminated in the consideration of four options that would be promoted by Scottish ministers. Of those, the conventional design and build procurement was identified as providing good value for money and being deliverable within the prevailing financial circumstances.

Our procurement strategy for the project is founded on the creation of a contract delivering effective responsibility and transfer of risk, which will be procured under effective competition. As a result of an extensive marketing process, an informal pre-qualification and an assessment of bidders, we have now engaged in a 12-month bidding process with two consortia comprising leading European and international contractors, each of which is supported by a raft of world-class consultants. Scottish ministers have confirmed their intention to fund the project from the Scottish budget.

I want to say two things. First, our estimates are audited independently. I emphasise that we arrive at the cost by what I would call a resource-based estimate. In other words, we calculate exactly the manpower, plant and materials that are required, and we build the estimate up from the ground, so we are relatively confident with that. We carry out rigorous risk analyses, in which we list the risks to the project and consider the cost range of those risks.

The other thing that I would like to focus on is the certainty of scope of work. One of the things that go wrong with projects that go off the rails is a lack of clarity on the scope of the project—in other words, scope growth. We have put a lot of effort into the definition of the scope, and have been very precise about what the project contains. I would always caution any client against allowing the scope to grow a little bit here or there, because therein lie the seeds of overrun. I hope that that helps.

Optimism bias is about reflecting the risks that we do not really know about at the moment. The known risks are covered in the risk analysis. The optimism bias covers things that may or may not happen. The two big issues at the moment are, first, the view that the contractors will have on how much they need to allow for the risk in constructing the project; and, secondly, the inflation that will occur after the tender has been received and the contract awarded. As I said before, our experience of contracts that we have awarded on the basis of all-inclusive prices that take inflation into account—they tend to be shorter-duration contracts—is that we have limited the overrun to 3 to 4 per cent. We are therefore fairly confident about the scope at that stage. The big imponderable will be the value that contractors put on the tenders compared to the allowance that we put on underlying costs.

I wonder whether Mr Dow could clarify that a bit further. Some colleagues are desperate for more information about the issue.

The UK Government and the Scottish Government have both agreed to go to an international code of practice for accounting, which gives you the answer of bringing something such as this on to the balance sheet. However, there is also a European set of rules for national accounting purposes and statistical purposes; it operates on a different basis and that is the basis on which the Treasury has to see financial business. That provides the opportunity to take something off balance sheet; it is a different set of rules. For the Treasury rules, it is by risk and for the Scottish Government rules—the international rules—it is by control. I am not sure whether that explanation helps.

No. The rules are the same for both; both tiers of Government have the same choices.

The first point that we need to make is that optimism bias and risk are reviewed regularly as the process goes forward. Essentially, both start at a relatively high level, and as you know more and more about the project they come down.

The issue with a bridge is that it is a more defined structure. You know where you are going, you know the foundations that you are looking at, you know what a tower looks like and you know what a deck looks like. Road works always contain an underlying degree of uncertainty, because you are putting the road into the natural ground conditions. There is a greater proportion of underground works with road works than with a bridge, which, once the foundations are laid, is out of the ground. That is why the risk relating to the road works is greater.

The recommendations are recommendations. We have looked at the underlying problems that were identified and the recommendations that were made in the Holyrood inquiry, in particular. We have worked out whether we are likely to get into the same situation with the Forth crossing contract and, if there is a risk of that happening, how we will deal with it. We have addressed the recommendations one at a time and have an answer to them. We can provide you with that answer after the meeting, if you wish.

You will recall that, when the work was being finalised, we ran into a rather unfortunate period in our financial history, when the funding that tended to support NPD and PPP projects on the best terms disappeared. In particular, bond funding disappeared following the removal of the underlying monoline insurance industry, and banks were starting to run into problems. For the Forth crossing project, we would have expected to borrow for PPP or NPD purposes at fine rates and to have a lead provider that would bring in a consortium to provide the finance, but that approach became extremely difficult.

One issue is the size of the finance that must be amalgamated. A deal was done at that time for the M80 project. The timeline will confirm that finalising the finance for that took much longer than was normal for such deals by Transport Scotland, even though that project’s value is significantly lower than that of the Forth crossing. At the same time, the Highways Agency ran into difficulties in trying to close its PPP for the M25 contract. That route was not considered to provide the certainty that we need for the Forth crossing contract, given that we wish to complete the crossing before the risk of compromise on the existing bridge starts to materialise.

The bill parcels the different works in different ways, but the project was considered only as one whole project for the parcelling of works and was considered only for one direct or NPD procurement. Is it correct that it was impossible to divide the contracts for the bridge works and the approach road works? That has been done with other projects around the world.

We must be careful about effective interfaces in civil engineering projects—one tries to avoid interfaces between contracts. I took the view that a fairly major contract in the overall project to deliver the bridge and the approach roads was likely. We considered hiving off elements and we have hived off the Fife ITS and the junction 1A works to provide a more balanced response throughout the industry.

The difficulty of separating the southern bridge works, for example, is that the site for constructing the southern approach viaduct to the main crossing—which has to be integral with the main crossing—is required for those bridge works and the southern approach roads. The approach road contractor’s works could end up cutting off access that is required for the bridge works.

Would it have been impossible to operate in another way? No. Would it have been wise to operate in another way? Certainly not.

15:15

Yes. Our conclusion was that it would have been very unwise to approach it in that way.

It would have been very unwise to separate them.

It was a judgment by me and our consultants, which took account of bilateral discussions that we had with a number of major international constructors.

Did Transport Scotland recommend the method of funding to ministers or did ministers decide on that?

Transport Scotland made a recommendation and ministers accepted that recommendation.

Every project is different and has to be looked at in the context of its own unique circumstances. Different recommendations were made to ministers around the Borders railway. To return to John Howison’s point, the capital requirements for the Borders railway are significantly smaller than the capital requirements for the Forth road crossing, which means that issues of the availability of funding through the financial markets are very different.

I remind members that we are here to discuss the Forth bridge project and not any other projects in which Transport Scotland is involved. It is unfair to bring in other projects. Please stick to the Forth bridge situation.

We are getting dragged into the same problem.

That is a wonderful question to answer. I must ask for the committee’s patience.

You will have seen the submissions that we have received, one of which was from the Scottish Futures Trust. What discussions, if any, did you have with the SFT on the funding package for the bridge?

So the SFT was not called on to offer its expensive expertise on whether building the bridge as a straightforward capital procurement project or as a public-private partnership was the right way to go. That was purely Transport Scotland’s decision.

Right. You say that the SFT is involved in governance and process. What role does it have in the process that you are going through of considering the two short-listed tenders?

It looked at the process that we adopted to bring the tenderers on board and at the process that we were to embark on for the procurement exercise before we invited the tenders. It provided us with advice on that, some aspects of which we were directed to fulfil before we took the step of inviting the tenders. The adoption of other parts of that advice was an on-going process. The SFT revisited us to assess our approach to taking up its recommendations.

Its role will be to look at the process that we have been through and at our assessment of the tenders, to satisfy itself that that has been done properly and adequately, and to endorse—or not—the move towards awarding a contract.

The peak will be a cost of £400 million. What has got you thinking that you can do that?

John Howison has largely covered that ground. We can discuss that with the bidders during the coming stage and suggest to them that that would be our preferred option. We do not want to force them to stick to a £350 million ceiling if that will give bad value for money, but we want to see whether they can arrange the sequence of their work on site without blowing that ceiling.

I do not want to interrupt you, but Mr Glover has explained the geology, geography, the weather conditions and all the rest of it. It might be—who knows?—that we have three years of good weather with not bad winters and good summers, which would allow work to progress quickly. In that case, you would be spending as you went along. Would you say to the contractors, “Carry on, boys—but you’re only going to get £350 million this year”? Is that the process that you are going through to keep the money at the level that you describe?

The matter is negotiable with the contractors. There is absolutely no doubt that, once the contract has been awarded, it will be essential to monitor progress monthly. It is unlikely that we would want, or be able, to tell a contractor to down tools because they were reaching the ceiling.

I understand that such matters are examined weekly by the director of finance in another place at another time, and that the Cabinet Secretary for Finance and Sustainable Growth looks at them monthly. We are interested—at least, my side is—in the amount of slippage in a number of capital projects and how that slippage could be used elsewhere. However, that is another debate, which we will not go into today. What discussions have you had with the Treasury and our European friends about the possibility of grants and other finances?

I will start on the European front. The route is part of the trans-European transport network—TEN-T—and is, therefore, eligible for grants for construction purposes, which are broken down into grants for preparatory works and those for the actual construction works. Obviously, we are at the preparatory works stage. We submitted applications for grant in 2008-09 and in 2009-10. In both cases, although there was recognition of the merits of the case that we put forward, it was not selected as a project that Europe wished to fund.

It must be said that Europe’s funding for the project is somewhat less than one would imagine that it should be. It would probably take too big a chunk out of European finances to fund a project as large as this.

Fife Council’s submission to us states:

“The Financial Memorandum does not make any allowance for the delivery of a Public Transport Strategy.”

Why is that?

May I explain that? The original concept was that the existing bridge might no longer be there, so we proposed a bridge with functionality that included dual two-lane carriageways, hard shoulders, pedestrian and cycle tracks and dedicated provision for public transport. By 2008, however, the prognosis for the existing bridge was much better. The work that we did to examine that with the Forth Estuary Transport Authority gave us confidence that the bridge could be used not only for buses but for trams—should that be justifiable in the future—and for pedestrians and cyclists. As far as we could foresee, that position was secure for the future, which meant that the functionality of the new bridge could be reduced.

I am sure that you will appreciate that the last thing anyone wants is to arrive at the Ferrytoll park-and-ride site on a windy winter morning to find that the buses are not running just as the car that you came in has driven off back to Dalgety Bay or somewhere. We accept that use of the Forth road bridge for public transport for most of the time will be a helpful contribution, and that there should be flexibility. We therefore considered the hard shoulder on the new bridge and said that, during periods of high winds, it could be used by buses, which would be diverted on to it. During such periods, the lane will act as a bus lane and not as a hard shoulder. That can be accommodated by virtue of the intelligent transport system, which will provide priority for buses onto the bridge and off it at the other end.

We also took the view—notwithstanding our confidence in the existing bridge—that because the bridge is to last 120 years, we could not rule out the possibility that something might happen to it in the future. In considering the width of the hard shoulder, we therefore decided to take into account not only its use by buses, but the possibility of the existing bridge not being able to take trams. The new bridge has therefore been designed with sufficient width in its hard shoulders to allow it to take tram-based light rapid transit, should that be justifiable on financial grounds. However, it would not, of course, be able to take light rapid transit and buses and act as a hard shoulder all at the same time—those are options that we can do within the cross-section. That is the key to the cross-section that we have chosen for the new bridge, which will have the flexibility to adapt to different traffic uses in the future.

That deals with the question.

Mike Glover’s long answer to Joe FitzPatrick’s question was very helpful—I was going to ask something similar. There are many technical questions, many of which have been asked, but from the public’s point of view—as we saw on the front page of one of the Sunday newspapers—the issue is the cost of the bridge. Mike Glover dealt with most of that, but the headline point was the cost per kilometre of lane. Is that, as it were, superficially true from your point of view but explained by all the factors that you have put forward?

Obviously, John Howison dealt with the hard-shoulder issue, and I think that most people accept the logic of that. However, the implication was that if there is no hard shoulder, but just a two-lane bridge, the cost would be reduced significantly.

If there is no hard shoulder, we will have failed to provide the future proofing that is needed. For example, the hard shoulder on the Severn bridge is only 2m wide, so it cannot be used for anything else. We set out to create a hard shoulder that could be used short term for public transport if there was a problem with the FRB. The hard shoulder will provide in the longer term other functionality that will inevitably be required. The life of the bridge will be 120 years, because we are designing it for maintenance and replaceability. It will not suffer from the sorts of problems that the FRB has suffered from because first, the new bridge will be a cable-stayed bridge, which means that each of the cables can be replaced in time, and secondly, the construction of the steel box is a box rather than a series of truss members. There are all sorts of reasons why the bridge that we are proposing should give you a great deal of confidence in the future proofing.



15:45

That is allowed. We do it ourselves.

It is old age.

Corrosion is a related issue. Corrosion leaves pitting in the wires and makes them susceptible to microfractures and so on. We may be able to stabilise the humidity regime and stop further corrosion, but we do not know whether that will stop the cables from continuing to deteriorate.

The main issue is that 86 per cent of the strength of the bridge is required to keep the bridge up, and only 14 per cent is directed towards carrying the traffic on it. We have already lost 8 per cent of that strength. By the time dehumidification works, we might have lost 10 per cent. That leaves the bridge in a good position in that if we take the loading of the traffic off it and put the light public transport loading on to it, we will have balanced it and regained the same factor of safety that it had to start with.

There is not only an issue about strength: there are also operational issues. The existing bridge is two lanes only, without hard shoulders. There are a significant number of breakdowns, and the more we use the bridge the more we experience the difficulties that breakdowns cause on it. With the weight that the bridge now carries, the road surfacing lasts for less and less time—I think that there are seven or eight years between resurfacing—which has a disruptive effect. Even incidents such as people climbing the catenary wires tend to result in the bridge being closed. Operationally, considering the importance of the traffic across the Forth, it is not a reliable bridge.

At £2 billion, the overall cost is of concern to people, but I suppose you have almost the opposite reaction to the basic cost of £543 million. Is that kind of discrepancy between the basic cost and the final cost quite common in similar projects?

Yes. I fully appreciate the problem, but the great advantage in this discussion is the transparency about how we get from one figure to the other. It can be seen how each of the component parts builds up, but I am afraid that the public are not aware of that. There is a big difference between bridge costs and project costs. That is the issue. However, it is difficult to get that concept across in open discussions.

No. The promoters of most transport public-private partnerships tend to be contractors who construct. Therefore, work was undertaken on whether a PPP approach was appropriate, having regard to the balance of maintenance activities and construction works, but there was no direct market testing of the availability of finance while the exercise was being undertaken.

Why not?

I am sorry; it is obvious that I have not made myself clear. There was market testing that reflected on the procurement method. That was done only in the construction market, which would have fronted up a PPP process; it was not done in the financial market, which would have provided the cash. The reason for that was the presumption that, at the time when the exercise was undertaken, there would not have been a distinct problem with providing the cash.

The value for money and the deliverability or timing of projects are two different things—

I am sorry for interrupting, but none of the answers so far has been about deliverability. Paragraph 129 of the policy memorandum states:

“the recent dislocation of the finance market has made the delivery of funding problematical with fewer financial product options and less appetite for the consolidation of senior debt by a lead provider.”

I am simply quoting the Government’s evidence, which states that an alternative funding model was excluded because of a presumption about the state of the financial markets at the time. However, in the same financial climate—in which there was, presumably, the same smaller appetite and the same uncertainty—Transport Scotland put forward an alternative funding model for another project of up to £300 million. That is relevant because—as I will come on to—the funding model will have an impact on other parts of the capital budget.

I might ask questions at another time about the choice that was made for that other scheme.

Appendix D, “Securing value for money from procurement”, details the method for paying the contractor, which Ms Fitzgerald touched on in her response to questions from Malcolm Chisholm. I seek clarity on what we were told about the phasing of budgets. First, have ministers in effect set a cap of £350 million for annual expenditure on the project?

Will there be any payment in advance for any works?

Yes.

I missed your question.

It was about the construction inflation element that you have had to include in your estimates for the tender costs.

We looked at data sets of 10-year rolling inflation over an eight-year period and averaged out the figures to arrive at a mean. We also looked at the high and low borders around that figure. The inflation bracket that we have—from high to low—is based on what has happened in 10-year rolling periods over the past eight years. Mike Glover will answer the question at greater length.

That brings me to a serious point. I know that the package is a design and build package and that different elements of the tenders will be weighted—I hope that it will not just be the cheapest option that wins. How much weighting will be given to the design element—for the bridge and its approaches and the cohesiveness of the design across the three contracts that will be let as part of the project?

My plea is that the project should not just be engineer driven—sorry, Mr Glover.

My career has been spent designing signature structures—I did St Pancras station, for example, and I want the bridge to live up at least to that. As John Howison emphasised, in the employer’s requirements to contractors we have been as prescriptive as it is wise and sensible to be. We have described the geometry that we expect in the towers and the profile of the deck, for example. For the contractor, the exercise is in achieving the most effective way of dealing with the logistics of a large bridge—how it will be built and bought. We have protected the design intent—the essence and the iconic nature—as much as anybody sensibly can.

We can see the vision clearly. Does Linda Fabiani wish to ask another question or can we finish? I ask for a short question.

The team that is dealing with the project has wide experience in delivering projects to cost and to programme—for example, I was the technical director of the Channel tunnel rail link for 12 years. One learns about proactivity—being ahead of, rather than behind, developments. That is the essence of how big projects are delivered on time and under budget. As I said, it is important to control the scope of the work. If a bright idea is added to the project late in the day, that is a warning sign and something to be careful of.

I return to the issue that John Howison passed to me—optimism bias. It is a sensible and wise approach and mixes qualitative and quantitative elements. One considers and works through the certainty that one has about some of the decisions that one has made, which gives a measure of the uncertainty that one has. The optimism bias that we have at the moment represents the stage of the project that we are at.

16:15

I have a quick question to clarify something that Mr Dow mentioned before about the impact of the expenditure on other parts of the capital budget. Will only the priorities for capital projects within the transport budget be examined or will capital projects in local government and health also not be protected from a consideration of the priorities?

The amount that is put in is an estimate. What it actually costs at the end of the day will be a measure of what it actually costs at the end of the day. That will be based on prices of housing and land at the time and the extent of actual disturbance at the time, which will be measured by use of the road once it is opened.

If you wish to clarify any matters in writing to the committee, please do so. The matter is important, and the committee wishes you wisdom and success in your work. Thank you for appearing today.