Q81 Joseph Johnson: Was it Parsons Brinckerhoff that encouraged the Highways Agency to have this inflated view of what it would cost?
Graham Dalton: The shadow bid model, which is the one you are referring to-the numbers are in figure 6-was actually put together by PricewaterhouseCoopers as part of the cost model.
Q82 Mr Bacon: Isn't it amazing that you spend all this money on consultants-£80 million all together- and still get it so badly wrong? What does that say about your ability to buy advice?
Graham Dalton: Can I just point out that we did have the maintenance report before? In the two and a half years that I have been in the agency, we have done quite a lot of work to improve the commercial management and estimating capability. Before, maintenance was actually on the major projects as well. The difference with this one is that we go out routinely. In 2008, we let three maintenance contracts, which are typically for five to seven-year terms. We have much better data about what those costs will be when we go and do the job.
Q83 Chair: Do they all come in below your estimated costs?
Graham Dalton: The winning bids on two of the three, if I recall, were below by a small margin, and our estimate was typically around the second bid level-so a spread of bids around our estimate. The difference between this and the maintenance contracts that we've bid more routinely is that this is asking contractors to take a view over 30 years of what it's going to cost them to either maintain or renew the asset and the structures. Routinely, we just take a four or five-year look ahead, or get rates for the type of work, however much of that work comes up.
Q84 Ian Swales: So was that a bad decision, to decide to do that, then? To go over 30 years?
Graham Dalton: This is the nub of what this contract is really about. The widening is an important part of improving the service to road users, but we are effectively pre-buying maintenance and transferring the renewals of asset risk on to the contracting consortium. The competition that we got, where we got two very tight bids, two very close bids, was an ambitious view on what they could do to operate and, really, their view on when they were going to have to renew assets, or the amount of work they'd have to do over that time. That's where the real competition comes in.
Q85 Mrs McGuire: How did you judge whether or not they were taking a pessimistic view of what needed to be carried out over a 30-year contract, which would inflate the cost of the maintenance, or an optimistic view? How did you benchmark their analysis?
Graham Dalton: Ginny Clarke was the chief highways engineer. I'll ask her to give the detail.
Ginny Clarke: We gave them an indication of what maintenance we would expect in the normal five-year cycle, and part of the tender they returned to us was their indication of how they would renew the asset over that 30-year period. So against the overall cost of operation maintenance they then told us an indicative basis on which they would renew the asset and maintain it. So we were able to make a judgment-whether they were actually underestimating, for instance, the sustainability of the road pavement, and when they would renew that. So that was part of the assessment of the tender and of the robustness of those bids.
Q86 Mrs McGuire: And they would build in a 30-year projection on construction inflation costs and so on into that?
Ginny Clarke: No. Effectively the cost model gave them high-level figures. What we asked them to do was understand the techniques they were going to use, rather than the cost of those-because the cost is the total cost; what we wanted to understand was what sort of work would they be doing. Would it be similar to the sort of things we might project over that period of time?
Q87 Mrs McGuire: Why did you go for the 30 years, which obviously you had little experience of, as against your normal projection of five or even 10 years? Thirty years, in terms of coming up with a figure for a highway, just strikes me as a little lengthy. I can understand it in terms of a building or something, but I don't quite understand why you would want to do it 30 years.
Ginny Clarke: Actually we had done our previous design, build finance and operate schemes on 30 years.
Q88 Mrs McGuire: So you weren't new at this game.
Ginny Clarke: No. They are now 10 or 12 years old, now, the first ones, so we have some understanding about how, over the period of time, the DBFO actually enacts those plans. Rather than the theory of it, what do they actually do in practice? That's the basis on which we did, again, a 30 year. It's also in relation to the financial model-
Q89 Chair: But you got it horribly wrong. So what have you learned? You got it wrong. You know you were out by between, whatever it is-I've lost the figures now-27% to 43%. Or 48%? I cannot read my own writing. You got it horribly, horribly wrong.
I'll tell you what it just makes me think. If any contractor came in, they came in lower. We haven't a clue whether at their lower price actually we're getting value for money.
Graham Dalton: Well you could look at it the other way and say that's actually the power of the competition and three strong bids coming in-
Q90 Chair: No, because we can have no confidence in your advice or your advisers' advice as to what is a real value-for-money figure for maintaining this bit of the M25 for 30 years. We just haven't got a clue-and no confidence, given your track record in failing to understand the real cost of it.
Graham Dalton: Well, the estimates came in very competitively. I think it's a good thing that they came in competitively. The thing we had to do is to really make sure that the assumptions that have been made in those bids and those tenders were actually bids that the contractor was going to be able to live with.