Public Accounts Committee
Oral evidence: Crossrail, HC 2127
Wednesday 15 May 2019
Ordered by the House of Commons to be published on 15 May 2019.
Members present: Meg Hillier (Chair); Sir Geoffrey Clifton-Brown; Chris Davies; Chris Evans; Caroline Flint; Nigel Mills; Anne Marie Morris; Lee Rowley.
Sir Amyas Morse, Comptroller and Auditor General, Adrian Jenner, Director of Parliamentary Relations, National Audit Office, Lee-Anne Murray, Director, NAO, and Marius Gallaher, Alternate Treasury Officer of Accounts, HM Treasury, were in attendance.
Questions 1-229
Witnesses
I: Bernadette Kelly, Permanent Secretary, Department for Transport; Matthew Lodge, Director of Rail Infrastructure South, Crossrail Senior Responsible Owner, DfT; Tony Meggs, Chairman, Crossrail Ltd; Mark Wild, Chief Executive Officer, Crossrail Ltd; Sir Terry Morgan, former Chair, Crossrail Ltd; and Andrew Wolstenholme, former CEO, Crossrail Ltd.
Report by the Comptroller and Auditor General
Completing Crossrail (HC 2106)
Examination of witnesses
Witnesses: Bernadette Kelly, Matthew Lodge, Tony Meggs, Mark Wild, Sir Terry Morgan and Andrew Wolstenholme.
Q1 Chair: Welcome to the Public Accounts Committee of Wednesday 15 May 2019. We are here today to look mainly at the Crossrail project, but we are also going to touch on some issues around High Speed 2, which I know the permanent secretary is aware of. This is the second time in just a couple of months that we have discussed what has gone on with the Crossrail programme after the delays were suddenly announced last summer. It was announced that it wasn’t going to open in December last year, as planned. I pay tribute to the National Audit Office for its quick work to enable us to do that hearing and for its subsequent work, and to the London Assembly, which has done an excellent report looking at this from a slightly different angle, but giving us some very useful information.
Crossrail had always had a date to open of December 2018. There was a headlong rush for that, as the NAO Report highlights, but now we know that it is not that. I thank Crossrail Ltd and Tony Meggs, who hosted us on a visit to Tottenham Court Road station last week, where we saw some very interesting things. We are pleased that people were candid with us about what the issues were.
We are going to go into that, but I want to start with the permanent secretary on HS2. Before I do that, Ms Kelly, I just want to introduce all the witnesses. It is a long panel today. We will be directing questions at you individually, so you are not all expected, or will not all be allowed, to answer every question; otherwise, we would be here for some days. We are not planning to do that. From my left to right, we have Matthew Lodge, who is the director of rail infrastructure south and Crossrail, and the senior responsible owner for Crossrail at the Department for Transport—welcome back. Bernadette Kelly is the permanent secretary at the Department for Transport. We then have Sir Terry Morgan, who was the chair of Crossrail Ltd until last August. I think that is when you stood down—no, last July.
Sir Terry Morgan: I actually resigned in December.
Q2 Chair: In December—forgive me. I am muddling up who resigned or left when.
Andrew Wolstenholme is the former chief executive of Crossrail Ltd. He has appeared before this Committee before and has hosted us in previous times. He now works for BAE Systems in Barrow-in-Furness.
Andrew Wolstenholme: In London.
Q3 Chair: We might be seeing you with that hat on at some point as well. There is no rest, Mr Wolstenholme.
Tony Meggs is the current chairman of Crossrail Ltd. Mark Wild is the chief executive officer of Crossrail Ltd. Both of them were in front of us in the last few weeks.
Bernadette Kelly, can I just turn to you on the issues around HS2? Obviously, this Report throws up some potential challenges. Let me just get to some grassroots issues. You promised that you would publish an updated business case, including estimated scheme costs, as part of the spending review. We are hearing from a number of your colleagues that the spending review is uncertain—that is putting it politely—and we don’t know if we are going to be having one yet. Will you still be publishing a business case and costs for HS2, even if there isn't a spending review?
Bernadette Kelly: To my understanding, there is a spending review. We are certainly working on that basis.
Q4 Chair: Do you know when, then?
Bernadette Kelly: I am expecting it to happen in the autumn. We have received a letter—all Departments have received a letter from the Treasury in the last few days—setting out what they now want us to do. So all the signs are there that we are mobilising for a spending review being on the way.
Chair: Can I ask the Treasury Officer of Accounts? Do you have any further news on a spending review date?
Marius Gallaher: It is right that Departments have been asked to prepare for a spending review.
Chair: In the autumn? Is there any movement on what the autumn is?
Marius Gallaher: Not yet.
Chair: In civil service terms, that could take us into January, potentially.
Marius Gallaher: Yes.
Sir Geoffrey Clifton-Brown: That’s winter.
Q5 Chair: Well, we would say. I think the civil service’s autumn is a long period of time.
You are working on the spending review. Does whether or not the spending review takes place make a difference to you working up the business case, and particularly the estimated scheme costs, for HS2?
Bernadette Kelly: What we have committed to doing in the next stage of this project is proceeding to notice to proceed. That is an important milestone in the governance and delivery of the project. That is also highly relevant.
Q6 Chair: That would be for phase 2a and 2b as well?
Bernadette Kelly: It is for the project as a whole, although it is particularly significant in terms of the authorisation it gives to HS2 to enter into contracts for phase 1. That is the particular significance of notice to proceed. We are working towards that at the end of the year. We expect it to coincide with a spending review. Obviously, all the logic would point to doing a revised business case and so on at that point, rolling up the notice to proceed decision with the outcome of the spending review. That would be the neatest and most logical way to do that.
Q7 Chair: It doesn’t always work neatly and logically. If it doesn’t, what is the plan?
Bernadette Kelly: Not all things do. That is the basis on which we are working. If there isn't a spending review, which isn't currently what I would expect, we will need to reflect on what we can put into the public domain at that point that gives a genuine sense of progress on where we are on the delivery of the project.
Q8 Chair: If there is not a spending review, is there a possibility of further delay to HS2?
Bernadette Kelly: I don’t think that the delivery of the project is dependent on the spending review. The question is, when is it sensible to put into the public domain revised business cases and so on? It is more sensible to do so, obviously, when you have gone through those significant moments of public spending and everything else.
Q9 Chair: You and your Ministers have pledged that there is no money for HS2?
Bernadette Kelly: Yes.
Q10 Chair: They have to work within the current envelope? Obviously, delays can increase costs. Does the “no more money” diktat remain, even if delays are a result of the Government decision being delayed?
Bernadette Kelly: The Government have been very clear that £55.7 billion is the allocated budget for this funding and that remains the case until and unless any other decision is taken, which it hasn’t been. That is absolutely a key parameter and I don’t see it changing in any way other than in a spending review, and quite probably not then.
Q11 Chair: Fine. So the money stays? That is the answer to the question? That’s fine.
Bernadette Kelly: I am not expecting that to change—
Chair: I am aware we have quite a lot of witnesses on Crossrail.
Bernadette Kelly: Having said that, I would just say that we clearly do recognise that this is a project that is now in delivery, and we need to be constantly assessing progress against delivery. I recognise also that whether or not there is a spending review, we need to make public where we are in terms of overall delivery of the project—
Q12 Chair: That brings me neatly on to the challenge that Mark Thurston as chief executive has already compromised on the speed—by at least 30 mph or kph; I’m not sure which measure he was using—and the reduction in the number of trains, which could reduce capacity. Is there a point at which HS2 stops being HS2, if they compromise too much to stay within the envelope? How are you watching that?
Bernadette Kelly: What Mark Thurston has said is that there are cost pressures. Frankly, it would be very surprising in any project of this sort if those didn’t exist. He is now working through how best to assess those.
Q13 Chair: We know that, Ms Kelly. What, in terms of your position—
Bernadette Kelly: I don’t recognise that decisions of the sort that you mention have been taken. I think the CEO set out the sort of options that he would need to consider.
Q14 Chair: Are there options that would be beyond the pale for the Department for Transport for this project to continue and still call itself HS2?
Bernadette Kelly: What we always have to do with any project of this sort is to look at the budget and the scope and consider how we ensure that we have a deliverable project. When you look at scope choices, you also have to consider whether changing the scope reduces the benefits more than it reduces or saves costs.
Q15 Chair: Which will feed into the business case?
Bernadette Kelly: Those are exactly the assessments that are currently being made about this project—the sensible steps that need to be taken to manage cost, the impact those steps will have on the benefits and whether we are still confident that this project has a solid BCR. It has a BCR of 2.3 on current parameters, which is pretty solid.
Q16 Chair: On the current parameters—
Bernadette Kelly: But we will certainly look at any decision to make sure that it protects the benefits of this project.
Q17 Chair: So both scope and potential delay could have a very big impact? This project is a complex web, isn’t it? For example—
Bernadette Kelly: It is. Any—
Q18 Chair: Let me just finish. On train purchases, if they do not know by a certain point how many trains they are going to run, as well as some of the other issues we are discussing, they will not know which trains to buy. When is the point of no return for decision making, in your mind? What sort of window is there? I am not asking for a precise date.
Bernadette Kelly: There are multiple decisions that you are describing, the interdependencies of which need to be managed by HS2 very carefully, such as decisions on the specification of the track versus the specification of the trains. Those are being actively managed at any given point throughout the life of the project. I cannot give you a single decision point on those things. What exists is a project plan that identifies where those decision points are, the interdependencies and what the sequencing of them needs to be.
Q19 Chair: Again, as the accounting officer for the Department, you must be looking closely at the business case to make sure that the BCR ratio that you talked about stays at a good level.
Bernadette Kelly: Correct.
Q20 Chair: Is there a point at which you would have concerns?
Bernadette Kelly: Clearly, if this project was getting close to no longer having a positive BCR, absolutely, as accounting officer, I would have concerns. As I say, given that it has a BCR of 2.3, I think you would have to see a very large increase in costs and a very large decrease in benefits to get to that point.
Q21 Chair: We just spoke about the interplays. The Crossrail report that we are looking at in our main session today raises serious issues about integration challenges.
Bernadette Kelly: It does, yes.
Q22 Chair: There must be many lessons that you will hopefully have learned from the NAO Report and will have been picking up before the NAO Report. What are the key lessons that you will take out of the Report that might make a difference to how HS2 is delivered?
Bernadette Kelly: I very much agree that there are a huge number of lessons from Crossrail that we need to take into HS2. As you know, and as I have written to you to share, we have done a very thorough piece of work on lessons learned. We have produced a report that has some 24 lessons in it, which we are now ensuring are embedded in all our projects.
If I had to pick out the most significant lessons from that Crossrail report—I have thought about it a lot—I would pick out a number of practical ones and then a couple of more cultural and behavioural ones. An important lesson that we are thinking about constantly is the balance to be struck between the autonomy to give the delivery partner and the oversight that we as client and funder provide to that project, and also how you evolve that over the life cycle of the project. I think that is important.
A really important series of lessons is how you ensure at all points in the project that you have a board that has the right skills and that provides the right level of challenge to the executive. Allan Cook, the current chair, is strengthening his board to make sure he has the skills around the table to enable him to do that.
There is a really interesting question that needs to be resolved and that we need to think about going forward about whether you have, as it were, evidence-based ranges of opening dates for these sort of projects, with phasing options, as opposed to single dates. We have learned from Crossrail that a fixation on a single date, especially when it is set 10 or 15 years out, can actually drive cost and inefficiency and, in the end, delay.
Early consideration of systems integration is a very important point, and we are thinking about that. The west coast partnership franchise competition, which is out there now, is intended to bring a shadow operator in early precisely to help think about some of the challenges of running a railway at the end of this. The use of benchmarking also comes through strongly in our report. That has been a big theme of our work on HS2. I could go on to others.
There are two behavioural points that I would like to stress. There is a huge tension all the time in these projects—the Crossrail example is a good one—between balancing realism and optimism in delivery. You need a degree of can-do optimism. You also need realism, informed by facts and understanding of what is going on on the ground. We will certainly be taking that to heart in thinking about how we are dealing with HS2.
The other general point is just how you get trust and transparency working throughout the delivery of a really complex and difficult project. Sometimes it is quite difficult in a political system, when you are delivering controversial projects, to create an environment in which there is genuine trust and open communication at all times. We really have to have that for HS2.
Q23 Chair: You talked about changes at the board at HS2 being led by the current chair of HS2. Is there anything else that you need to see change significantly at HS2 to make sure that some of these problems are resolved or avoided?
Bernadette Kelly: I think a lot of this is already embedded in the things that we are doing, because you learn these things constantly. There is not a moment of truth in these things; it is a constant process.
Q24 Chair: But there is nothing else? There is no other Belisha beacon or warning sign?
Bernadette Kelly: No, but the challenge we need to keep very firmly in mind—we really do need to challenge ourselves, the company and the politics of HS2 on this—is this balance between being realistic about what can be delivered within the milestones set and not fixating on milestones in a way that ultimately causes perverse behaviours or inefficiency. That is the thing that I have taken to heart and that we need to reflect on as we take that project forward.
Chair: We will be talking about that in relation to Crossrail.
Q25 Sir Geoffrey Clifton-Brown: Ms Kelly, I am amazed by that list you have just given us. This Report has a lot to say about how contracts are let, and a lot of what has gone wrong with Crossrail is to do with how the contracts were let—whether they were let on fixed price; whether they were let with incentives built into them; the interfaces between different contractors; and the compensation arrangements. A host of things have gone wrong. Surely, those all read across to HS2, and the board of HS2 needs to be thinking very carefully about that.
Bernadette Kelly: I agree; you are quite right. The NAO Report has a lot to say, and I am sure my fellow panel members will be able to say more about the decisions that were taken in terms of how contracts were let. The NAO Report particularly identifies that a large number of projects were let and that the interface risks within them were not sufficiently managed. Clearly, that is an important lesson. I do not take from that that there is a single way to let contracts or to manage interface risks. I do take from it that whatever system is chosen by HS2 needs to reflect on whether the contractual arrangements, the arrangements for managing risk and the arrangements for managing interfaces are all under control. Those things will absolutely be being thought about very carefully by HS2 Ltd, which of course is entering into a lot of contractual arrangements at the current time. The structure of some of that looks quite different from some of the structures that were in place in Crossrail.
Q26 Sir Geoffrey Clifton-Brown: Can I probe you on one of the Chair’s other areas of questioning? Clearly, at the end of the day, how much money you put in has an effect on the type of railway you get out—things like the number of trains, the speeds and so on. Will Parliament or the public, or both, have an opportunity to influence those decisions before final decisions are made?
Bernadette Kelly: There are already requirements built into the hybrid Bill for phase 1, and obviously we now have a hybrid Bill going through for phase 2a. It is an interesting question. On the judgments that now need to be made about any changes to the specification of HS2 to ensure it can be delivered in an affordable way, those are the questions that Mark Thurston and the new chair, Allan Cook, are thinking about and assessing. The Government will ultimately need to take some judgments about what balance it wants to strike in this project between scope and cost. It would be quite difficult to deliver a project in a way that allowed the public to be consulted on at every single point in time, particularly given the extensive consultation that has taken place already on HS2 and the extensive parliamentary consultation through the hybrid Bill process.
Q27 Sir Geoffrey Clifton-Brown: Finally, if you wanted to cancel the whole of HS2, where does that point arise?
Bernadette Kelly: I have not assessed that. As far as I am concerned, the Government have been very clear that we are delivering this project. Therefore, that is what my Department and HS2 Ltd are working to do. With any project, if there is a go/no-go decision to be taken, which is the Government’s choice to make, given their priorities, you would assess the costs and benefits that would accrue if you proceeded and balance them against the costs that have already been sunk into the project. From an accounting officer perspective, I would need to make an assessment about what that told me about value for money, but that is almost a technical assessment. Right now, I am working hard on HS2 and working hard to deliver the project, because that is the Government’s clear and stated priority.
Chair: We will move to Caroline Flint, but can I urge you, Ms Kelly, to keep your answers a bit shorter? We will not get through our business if we carry on at this length. Ms Flint will ask tight questions, as ever.
Q28 Caroline Flint: Ms Kelly, could you tell me how the Department is assuring itself that the budgets for land and property acquisition is sufficient, given that NAO Reports have shown that the estimates are out of kilter and the “Panorama” programme indicated that there were completely misleading promises about what this would cost? Where are we on the budgets for land and property acquisition?
Bernadette Kelly: I confess I did not look into this in detail before today’s evidence session, but I did so at a previous point when it was looked at very carefully, in settling HS2 Ltd’s accounts last year. At that point, we looked carefully to ensure that the assurance of property estimates was sensible, was being updated and took into account independent advice. To my knowledge and understanding, although I can always write with further details, there is a thorough process within HS2 to assure itself regarding land and property estimates.
Q29 Caroline Flint: I am sure the Committee will take you up on your offer to write to us, not only about the amount of money and looking at the comparison with what was suggested before, but about the governance and how you are assuring yourself that they are identifying the right values against the lands and properties that will be affected. Moving on to a situation in my own patch, south Yorkshire, as a result of HS2 route phase 2b, which was an alteration to what was originally expected, the Shimmer estate in Mexborough has faced a huge amount of uncertainty. To give a real-life example, HS2 claimed for two years that only 16 demolitions were to be made on the Shimmer estate. In 2018 it had to revise that number up to 52, but that is still not an accurate reflection of the impact on that estate. How can we have confidence in the processes and methodology for determining what needs to be acquired at what cost, if some fundamentals such as that are going so wrong?
Bernadette Kelly: Again, I am conscious that I did not have a chance to look into the detail of that point before this Committee; I will do so in response to the question. I am obviously well aware that the Shimmer estate has been a particular area of sensitivity, that residents have been impacted by route decisions on HS2 and that there are extensive conversations going on between the company and those affected by HS2. Every effort is going into ensuring, first, that estimates are correct and, secondly, that residents and property owners affected by HS2 are treated as fairly and as well as possible. I am sorry I cannot give you a more detailed response than that. I will see what I can pull up.
Q30 Caroline Flint: Maybe I will just give you something to take away and think about. It is understood that, when the alternative route was being looked at, surveyors and others were looking at old maps that did not show the existence of the Shimmer housing estate, which is a new estate—that by the time the announcements were made you were aware, but when the plans were being put together, it was not even seen. I would really like you to come back to us in writing about how such a basic error could happen in planning such a major development. Do you accept that that is unacceptable and does not give us confidence in the way these areas are being surveyed for impact?
Bernadette Kelly: I will certainly write setting out the facts of how the Shimmer estate has been taken into account throughout the development of this project. I am not sure whether I can confirm the particular facts that you are describing, but I would be very happy to set out the facts as we and HS2 Ltd understand them, and the steps that HS2 Ltd has taken to manage the impacts of this on the Shimmer estate.
Q31 Caroline Flint: Alongside the examples I have given about the impact on those homes and the people who live there, the estimated costs of the new spur are spiralling as well. For example, the estimated cost of making Sheffield Midland and Chesterfield stations HS2-ready is £825 million, but none of that is included in HS2’s cost envelope. I can give you other examples. It would seem that any one of these factors would trigger a review of the business case of such a major project. Is that going to happen?
Bernadette Kelly: There are many estimates made by many parties relating to the costs of HS2 Ltd. Very often, we do not recognise those cost estimates or necessarily understand on what basis they have been derived. I am conscious that there are lots of alternative estimates out there, but we don’t necessarily recognise them or understand how they have been arrived at.
As for the business case, we absolutely will be revising it thoroughly in the light of our understanding of actual costs and benefits of the project, as we know and understand them, and as HS2 Ltd are able to advise on.
Q32 Caroline Flint: Moving on to skills, could you tell us a bit more about how you think the academies that have been established to provide the skills for HS2 are working? Where employees are taken on, what proportion of those employees are foreign workers as opposed to those already living in the country?
Bernadette Kelly: As you know, two skills academies have been established to help provide the supply of apprenticeships and skills for HS2 Ltd. I think that is a positive example of how a project like this can help to grow the supply chain and the skills base. There are now hundreds of apprentices working on HS2 Ltd. That may be something that Sir Terry has some recollection of as well, as I know he has been involved in these.
I cannot give you a number of foreign workers versus domestic workers. I am very happy to get that number and to give it to you, but I am afraid I don’t have it available.
Q33 Caroline Flint: If you could write to us on that and, if you haven't got the number for this either, can you write to us on it as well? It is about the places that are available at the various colleges. Obviously, we have the national rail college in Doncaster, but there is also what I think is still called the HS2 academy in Birmingham. Could you provide us with information about take-up of places, and whether they are being fulfilled?
Bernadette Kelly: Yes, I am very happy to do that.
Chair: Thank you very much, Ms Kelly.
Q34 Lee Rowley: I wanted to follow up on Ms Flint’s point from a moment ago. You said you don’t recognise a number of the estimates and statements around some of the costs. Chesterfield and Sheffield were referenced. As a Chesterfield MP, how are you expecting the electrification of that spur from Clay Cross to Sheffield, and all the reconfiguration at Chesterfield and Dronfield stations, to occur with no additional money?
Bernadette Kelly: I am wary of being drawn on very detailed things, lest I give slightly inaccurate answers.
Chair: Fine. Tell us what you can.
Bernadette Kelly: I think the general point is that the budget for HS2 Ltd assumes certain costs that will be paid by the company through the project to deliver that project.
Q35 Chair: Does it include the spur that Mr Rowley has just talked about? If you can’t tell us, okay.
Bernadette Kelly: I am sorry. If I have advance notice, I always come ready to answer these questions. At this level of detail, it is sometimes a little difficult when I don’t.
Q36 Lee Rowley: It is more the principle of assuming a greater scope without additional money, unless there was a lot of money splashing around in the first place. Otherwise, it is reasonable of me, as a local MP, to assume that it will either be done on the cheap, or badly, or end up in the same place as what we are about to discuss—Crossrail.
Bernadette Kelly: The budget for the project has been set in relation to the scope that exists. There may be aspirations—I am quite sure there are—to extend the project in ways that would deliver benefits locally and that would not be included within that budget; funding would need to be found from other sources for those benefits to be unlocked. That is the sort of general, in-principle point that I would make. The thing we must do, at all stages in this project, in taking it forward, is ensure that we are clear about scope, and cost and budget in relation to that scope. There is always going to be the potential for other things to be delivered outside that which deliver benefits, although they may not be funded from the same source.
Chair: I think you just launched a major bidding campaign from Chesterfield. Mr Rowley, was there anything else?
Q37 Lee Rowley: The question on capacity and speed: what is more important?
Bernadette Kelly: I think we increasingly see that the real benefits of this project are capacity. It is absolutely about a major uplift in the capacity of our national and inter-city rail network. Speed is important in so far as it improves connectivity and delivers journey time improvements, but I think in terms of how we talk about the project now, we really think about it as capacity, connectivity and growth.
Chair: Sir Geoffrey, a last, quick one?
Q38 Sir Geoffrey Clifton-Brown: I shall be asking the other witnesses questions about the contingency in Crossrail, but I want to ask you about the contingency in HS2. How much of that £56 billion figure is contained within contingency, and will that be adequate?
Bernadette Kelly: Contingency is included within the overall funding envelope for all phases of this. Clearly we will need to be reviewing that level of contingency as the project develops, but, yes, there is an allocation for contingency.
Q39 Sir Geoffrey Clifton-Brown: How much is it at the moment? Or please write to us.
Bernadette Kelly: Okay, I will. I am sorry; if I give you the figures, I will end up potentially giving you the wrong ones. There is an allowance for contingency for this project.
Q40 Chair: We need to move on now—other witnesses have waited patiently—to the knotty issue of Crossrail, so I want to start with you, Sir Terry Morgan, and Andrew Wolstenholme. When were you first aware that there were problems, Sir Terry, with the delivery timetable for Crossrail?
Sir Terry Morgan: Problems? To be frank with you, with the programme, there were always problems.
Q41 Chair: Right, but when did you realise there were problems that were so significant that they were going to delay the opening?
Sir Terry Morgan: I always think we had a very transparent relationship with the sponsors, but we formally came to a position about not being able to deliver the programme in July 2017.
Q42 Chair: That is when you formally came to a decision. Is that the first time that you were aware, as chair of the board, that there was a problem?
Sir Terry Morgan: We had, at every board meeting, discussed the condition of the programme and our ability to deliver it in 2018. In a way, I am repeating what the permanent secretary has just said, but I am the first to acknowledge that there is a very fine line—I think the culture we created in Crossrail was very helpful on many occasions—between can-do and optimism bias.
I would acknowledge that even in July we would have underestimated the scale of what remained to be done. Systems integration is clearly a very good example of that. Always, when we were getting into these difficulties, the executive would develop a mitigation plan, and normally associated with that would be, “And if we are late, what would be the cost consequences of that?” So there was always this calculation about cost versus time, and we also always took into account the strong sense of needing to demonstrate that we had done everything feasible to open the programme, as had been promised, in 2018. That was a very strong driver, both within the programme and politically, in terms of the sponsors that we dealt with.
Q43 Chair: Mr Wolstenholme, when were you first aware that opening on time was going to be a challenge?
Andrew Wolstenholme: I’ll answer that, but can I just thank you for inviting us here, Chair, and the Committee? It is very important that we understand what the lessons are, and for the NAO Report to highlight some of those. I spent seven years on this programme, and I would like to say how personally disappointed I am that clearly there is now an overrun. I am personally disappointed for the passengers who are not using that service now, and personally disappointed for the many tens of thousands of people who did a fantastic job in delivering the programme thus far.
On when we first realised, if you look at this holistically for a moment, we have something that is completely unprecedented that is bringing in new and novel technologies and has not been done in the UK before. The 42 km tunnels, the 10 new stations—these stations are between 20 and 40 metres underground.
Q44 Chair: We are aware; we have been to visit, and the NAO Report highlights some of those engineering challenges. I want to cut you short a bit, because we need to get to the main question.
Andrew Wolstenholme: But if you paint the picture holistically of how this programme has got to where it is over the past 10 years, as Sir Terry has said—you use the word problems, but I use the word challenges—there have been huge challenges throughout the programme, and what this project team has done consistently is make professional judgments about how it delivers remediation on those particular activities.
I can give an example to illustrate the sort of things we had to get over, which will lead into 2017. There was a one-year delay on the Moorgate shaft. Just before that, if you look at the NAO Report of 2014, we were reporting that the programme was only at 65% likelihood of completing, and that was in 2014. If you look at the examples of how we have delivered that, we will take you to 2017; there were three aspects of the programme in 2017 and the risk profile was perhaps designed in a different way from how it was earlier.
We then had three different supply chains, one representing the complex rail systems, one representing the stations and one representing the rolling-stock. There were three risks in particular that we were looking at very carefully. One was the whole sequence to lead into the energisations of the tunnels in the east, and that was very important, because that energisation—
Q45 Chair: Obviously, this has been your baby for a long time, seven years, and there was a lot of major and good engineering work that went on. The point is that it did not deliver in December 2018, and we need to understand when you knew there might be a problem with that delivery. What you have just highlighted is that, of course, there were big problems along the way, but the key question is when you, as a chief executive and as a board, knew that those big problems might lead to a big risk to the delivery date—a set delivery date, which is a challenge in itself.
I don’t want to go through every detail of the Report, because it is well written and we have all read it, but if we take figure 6, it illustrates an example. From September 2016, the risk exposure was quite some distance from the contingency money available, so there was obviously a big problem, which then widened by autumn of last year. That gap was not appreciably narrowing, on balance. There was obviously a mismatch between the risk exposure and the contingency funding available to do it. That gap was widening unhelpfully. Were you aware of that, Mr Wolstenholme?
Andrew Wolstenholme: We were aware of the risk exposure, but the outcome of where the cost risk is is all about the activities that happen on the ground. I was just explaining that in 2017 there were three issues driving that risk. The first was the critical path taking us to the energisation, which is important, because the energisation then gave you the eight-month window through which you did the dynamic testing. Of course, in November 2017 there was an explosion, and we can come back to the implications. The second critical path was going through the completion of the stations and the railway systems, and the third was bringing the complexity of the trains and rolling stock into the stage 2 and stage 3 scenario.
Q46 Chair: But these were known challenges or known risks. That was the whole challenge of the project—major, exciting, interesting, ground-breaking engineering works and IT and so on, but all integrating, and integration ultimately was Crossrail, or will be Crossrail.
Andrew Wolstenholme: Yes, but if I may say so, these risk are dynamic risks. You have a clean plan of how to start the programme off, and you find things such as Moorgate, and incidents that are unknown and unplanned, such as Pudding Mill Lane. Those are things you have to be able to accommodate as you go along. On illustration 6, I can tell you that in 2016 we were carrying a contingency of £859 million at P50. We had identified risks then of £763 million, so we were still covered in terms of those risks. The exercise we undertook to ensure that we had headroom within that contingency was extensive. Those are typically the sorts of things that we did: it was in five parts, with the aggressive pursuit of opportunities to win further headroom.
Q47 Chair: You had each of these problems, and it sounds as though each time you were reporting to the board, and the board and you were trying to manage those problems, but when did you first realise that they were so significant that there was a risk to the delivery date? It is relatively unusual to have a very set delivery date for a project of this complexity. As Ms Kelly highlighted, it was set a decade or so out. When were you aware that these things might cause a problem, and what were you telling the board at that point?
Andrew Wolstenholme: Up to early 2018, we were assessing what the risks were and the culture of the team. We realised the importance of the political ambition; we realised that the objective of Crossrail was to open in December 2018. We were assessing the risks and, as Sir Terry Morgan says, we were finding ways through the risks.
Q48 Chair: The can-do approach was perhaps over-optimistic in the end.
Andrew Wolstenholme: I think there is a difference between optimistic and can-do. Yes, it is absolutely true to say that the team was working across the complete enterprise of Crossrail to find ways through the challenges and the risks that we had, and that was the culture that we had. We can talk about that balance between optimism and over-zealousness versus can-do but up to that point, if you look at the history of Crossrail, we had continually found big, major engineering challenges for which no technical solutions existed, and that is what we sought to do.
Q49 Chair: That wasn’t a surprise. You are digging up a large chunk of central London for a start. Those known unknowns were going to be built in. Do you think you did not plan for those known unknowns?
Andrew Wolstenholme: I think we did plan for those unknowns but, despite the fact that you spend two years doing survey and remedial work, you still find things that you are not expecting. The context within which we are working on this massive programme that was setting new standards and achieving things that had not been achieved before—
Q50 Chair: But at that point you must have realised there was a risk to the delivery date because of these big challenges you were uncovering.
Andrew Wolstenholme: As I say, in 2017 the risk profile was different. There were the three risks around an explosion at Pudding Mill Lane, the complexity of finishing off the tunnels, and introducing the new rolling stock. Those were substantial risks, and the programme that we put together in March 2018 still found a route through in order to open in December ’18. This is, with respect, looking at the dials that we had at the time. With the benefit of hindsight, one might have read something different from those dial, but as one was sitting managing those risks going forward and understanding the complexity of those risks and the interfaces of how those risks were presented, there was a considerable amount of risk there.
At that stage, the programme team was still clear that there was a path through to December ’18. On the difference between its being risk-free, this programme has never been risk-free.
Q51 Chair: I don’t think we ever suggested that a project of this capacity would be risk-free.
Sir Terry Morgan: Could I give you an example? We had a number of phases. First of all, building the tunnels and starting some of the fabric of the station boxes. There were a number of ground conditions that were not what we expected them to be and we had to slow down certain parts of the programme. The pressure that we had in the programme and which the executive were expected to respond to was not just to say, “The consequence of this is that we are tunnelling late.” The answer was always, “You’re late, but what are we going to do to mitigate the risk?”
On tunnelling, in 2016 we finished to within two or three weeks of a plan we had laid in 2012, but there were consequences. If you take Moorgate in the way that Andrew described, it meant we could not progress the station box work, so in parallel with the tunnelling there was also work ongoing to actually look at the plan to complete the stations. It is true that as time moved on some of those aspects became tighter and tighter, but the pressure to try and meet the end date was still very much part of what we were trying to do.
Q52 Chair: Did you ever question that end date, Sir Terry?
Sir Terry Morgan: The answer is I can’t remember a board meeting where we didn’t. Indeed, where we were needing further assurance that the advice we were getting from the executive was robust, we had at least four independent reviews to assess the plan that had been presented to us by independent experts. To be frank with you, the project is so complex and so big that when you talk to an expert, not many people have actually done what we have done. What they did was evaluate the plan we put in place.
Q53 Chair: I repeat: we acknowledge that they were very big and interesting engineering projects, but today we are not here to talk about the challenging and interesting engineering. We are here to talk about the overall project management, which we recognise included that, but ultimately, it has not just overshot on time, but massively on budget, and has had a big impact, for instance, on TfL’s farebox for a start.
Sir Terry Morgan: That is not a consideration for the programme.
Q54 Chair: No, but what I am trying to say very clearly is that there are very big consequences to this delay on Londoners and those beyond London who will be using it. We need to focus. We need to put aside that we recognise that major and exciting engineering projects are involved in this and that, overall, it is a big and exciting engineering project; we are trying to focus on the value for money. So to pick up on that before we come to Sir Geoffrey, Mr Wolstenholme, you have just heard from Sir Terry that he was getting assurances from you as the executive that things were on time. Did you ever question the opening date? Did you ever have a conversation with Sir Terry about changing that date, to make it easier to deliver and to manage the risks that you were having to manage?
Andrew Wolstenholme: As Sir Terry said, we constantly questioned the programme and our ability to deliver that programme.
Q55 Chair: Specifically, did you look at the issue around the delivery date? Because that has been a big driver of some of these problems.
Andrew Wolstenholme: No. We didn’t ask permission to deliver this programme late. I will tell you why. Whenever we did the parallel studies as to how to deliver the December date, we also looked at the economy. We tried to do two things: first, to deliver it on time; and, secondly, to deliver it within the funding envelope.
All the studies showed that because this programme carries such a vast number of different suppliers in different combinations, it is very expensive to manage that overhead. So the most economic way through was not to delay; the most economic way through was always to find a way through those risks and challenges. Therefore, the shortest period on Crossrail, the shortest programme—this is normally the case on most programmes—is the most economic. Whenever we came up with studies saying, “Can we delay things? Can we move things back?”, that always added cost. The inference from that was no. We had a challenging programme; we knew what the risks were; we knew what the political ambition was; and the culture of this, to work right across the enterprise, was to find ways through those issues.
Now, as I said, back in 2017, and with the benefit of hindsight, there were things that happened that are unknown unknowns, such as an explosion at Pudding Mill Lane. The instinct of the team right then was to go and fix it. We did absolutely herculean things to find a way through that. Again, we submitted the plan to the board in March, which had allocated the right time for testing and commissioning, but had underestimated—this is with the benefit of hindsight—those activities that were yet to be completed on the stations. I am sure that Mark will talk about that later, but I would agree with him: we had underestimated, despite the fact that the information we had on our screens was coming from the specialist suppliers who had designed and put their programmes together. We had underestimated that.
The programme was therefore very tight in March. It carried lots of risks. The PRep would acknowledge the fact that it had carried risks and would be showing lots of red on the screens. Much of the information that the PRep got was derived from our own reports, so acknowledged that the programme was very risky at that stage. We openly and honestly, with transparency, discussed that with the board and the sponsor at the time.
Q56 Sir Geoffrey Clifton-Brown: Sir Terry, I fully acknowledge all the technical achievements of the railway—we have been through all this—but in 2016 you had burnt up all your contingency of £600 million. By the beginning of 2018, you were rapidly burning up £2 billion-worth of TfL’s contingency. In June 2018, I was on an official visit, and I was told then that everything was still on time, and on schedule to open at the end of last year, and yet we had a visit this last week and we could see just how much there is still to do. The opening time is likely to be delayed by two and a half years. Doesn’t all this really demonstrate that your board wasn’t on top of the job?
Sir Terry Morgan: It is obviously something I have thought about a great deal. One of the things that was always a factor—I would have said something similar to Andrew in the opening statement, but I would add this—was that all our plans were based on submissions from our supply side. We underestimated the system integration; of that I have no doubt at all. But the board itself, I still think was a good board, with the right skillset. I have obviously thought a great deal about that. We carried out our duties in the right way and we didn’t breach our contingency.
It’s a very difficult calculation to make, but all our costs to complete were based on some form of probability analysis, whether it was P50, P80 or P95. There were very distinct lines of engagement with the sponsors if we breached, whether we called it an IP0 at P50 or whatever, and we had a number of different trigger points. In terms of the numbers that you were quoting, we did not breach. If we did breach, we had a duty to—and always did—report back to the sponsors at the time we were reporting a breach. We did do that, as we actually had to access some of the reserve funding that had been made available by the sponsors as we were coming to draw down some of those contingencies to meet the date in December 2018 that we were trying to go after.
It was always the case that we were putting a lot of pressure on the supply chain to deliver their productivity. They did not. And I would argue that one of the risks you always take when you start moving the date is that the pressure comes off the supply side delivering on their promises. That’s always the risk that you take with these programmes. Getting them to actually get back to the pace we were looking for is very difficult.
Q57 Sir Geoffrey Clifton-Brown: Can I ask you a very simple factual question?
Sir Terry Morgan: Please.
Sir Geoffrey Clifton-Brown: At what point did you have a candid conversation with your sponsors—namely, TfL and the Department—about delaying the opening date?
Sir Terry Morgan: You have to bear it in mind that I’m the non-exec chair, so it would have been the executive who replaced Andrew who would have spoken. I spoke to the Secretary of State and made him aware at the end of July that we—
Sir Geoffrey Clifton-Brown: End of July 2018?
Sir Terry Morgan: 2018. I also—
Q58 Sir Geoffrey Clifton-Brown: Not until then?
Sir Terry Morgan: Not until then. There had always been a conversation around the status of the project, and we talked about some of the challenges that we were facing, but at that time, we still had a plan that I had tried to get validated independently, by experts, to support the board’s recommendations from the executive. We always had the evidence that supported the fact that we had a plan to open in 2018, yes.
Q59 Chair: But on 23 July you came to an all-party parliamentary group here, on Crossrail.
Sir Terry Morgan: I did.
Chair: And you told them that it was on schedule.
Sir Terry Morgan: It was a very difficult time, to be frank with you.
Q60 Chair: Sorry, it was a difficult time, to be frank with the Committee or with the all-party group?
Sir Terry Morgan: Well, verbatim, I talked about the fact that there was a huge challenge still to be undertaken before we could get to that position. We had a board meeting just after that. I saw the Mayor of London towards the end of July and I made him aware—well, with the chief executive. We made him aware that the plan for 2018 was no longer deliverable. I had this expected response from both sponsors: “It has to be done in 2018.” What else could we do? How do we throw money at it to get there on time?
Q61 Chair: So when you went to see the two sponsors at the end of July, they told you that you had to continue—
Sir Terry Morgan: Find a way of opening the railway. That was always the right pressure, by the way. I have absolutely no problems with it. I was never under any pressure to say something I did not believe was true. But that was always the expectation.
Q62 Chair: We may go back to the issue of the APPG in a moment, but Ms Kelly, just on that point, before I bring in Mr Davies.
Bernadette Kelly: My understanding—our records show that Sir Terry did indeed raise this in a conversation with the Secretary of State at the end of July. That was in the context of a meeting on HS2, but it came up in the margins. At that point, Sir Terry indicated that the board was now considering options around delay. This was the first time, I think, that the board had countenanced that December 2018 might not be met. Formal notification, I should say, of that did not come until August, after the board had actually formally considered the situation. I would just distinguish between those two things.
Q63 Chair: And do your records say what the Minister or the Department said back to Sir Terry?
Bernadette Kelly: I was not in the meeting, so I can’t give my personal recollection. I’m quite sure the natural reaction of the Secretary of State would be, “Do everything you can.”
Q64 Chair: But would it have been minuted? The civil service are excellent at minuting meetings.
Bernadette Kelly: Yes, it was; there are notes. It was kind of, “Do everything you can to see what opening is possible.” That was certainly the gist of it.
Sir Terry Morgan: I left no notes for the Secretary of State at that meeting and I have no idea what was minuted from that meeting, but what it very clearly said was: there is no feasible way of opening the railway in 2018. With the Mayor, it was different; we—
Q65 Chair: You said that very clearly: there is no feasible—
Sir Terry Morgan: I said that, but it was repeating something that had been said to the Mayor two days—three days—before.
Q66 Chair: To be clear, this is the Mayor you are talking about, not TfL, at this point.
Sir Terry Morgan: I’m talking about the Secretary of State.
Chair: Yes, sorry, the Secretary of State.
Sir Terry Morgan: I saw the Secretary of State on the last day of July, and it was to be a meeting about HS2, but we didn’t spend much time on HS2, as you might imagine. I had seen the Mayor three days before that. We had left a copy of the presentation material that the chief executive used. The words were very clear: delivery in 2018 was no longer feasible. And we offered three dates in 2019 that we thought we could match.
The work that was done, which the permanent secretary is referring to, was that we had not costed the consequences of the delay. We needed in August to go away and do the costings to support the late delivery.
Q67 Chair: Can I just be clear? I thought earlier you said that you were told to go away and find a way to do it, so can you be clear what “it” was? I was under the impression that you meant doing it by December 2018.
Sir Terry Morgan: The conversations were different. With the Secretary of State, it was, “Surely we should be able to open something,” and the answer was, “No. We’ve been through that.” To my mind, that was a justifiable response from the Secretary of State.
Q68 Chair: So they asked the question and you said no, very clearly.
Sir Terry Morgan: With the Mayor, it was different: “How can I throw more money at it? Can we actually incentivise the contractors more than we’ve already done?”
Q69 Chair: And what did you say to the Mayor?
Sir Terry Morgan: The answer was very clearly in the words, “We have gone through all of that, and it is no longer feasible to deliver.” I am very clear.
Q70 Chris Davies: Mr Wolstenholme, the Chair tried to get out of you for 20 minutes when you thought this had gone wrong. We tried last week to get out of you, Mr Meggs, when Crossrail will be opening. I am just wondering whether it is in the job description of the chief executive of Crossrail to avoid questions when they are put to you.
Can I just go back to you, Mr Wolstenholme? Do you agree with Sir Terry that it was in July of last year that this went wrong, and you realised it was going wrong and it wasn’t going to open in December?
Andrew Wolstenholme: I actually left the project in March 2018.
Q71 Chris Davies: But you must have seen, even in March—this was a massive project. You must have known when it was going wrong. You had been advising Sir Terry.
Andrew Wolstenholme: I have explained that point in time in 2017 when the risk profile changed. I have also explained that in March 2018, when we put the final programme down, along with a cost and a lot of the risk assessment, we explained that this was a very high-risk approach. No one was in any doubt in March 2018, which was my last month, that there was not a valid programme ahead, but it came with considerable risk. I think that is the point that we made when we spoke to the sponsors in January, February and March.
It was quite clear to my programme director, who took over from me from then, that there needed to be three test points through 2018 at which he measured either the maturity of the rolling stock or the maturity of the systems that supported the rolling stock, in the RailTrace and on the stations, and there needed to be a decision through 2018 as to whether this opened or not. That, as Sir Terry has said, was formulating in July of that year. In March, we had the programme. It had a high level of risk. That was explained to the board, explained to the sponsors, and that was the position we were in in March.
Q72 Sir Geoffrey Clifton-Brown: Sir Terry, at this really important meeting when you went to tell the Secretary of State that the thing was not possible to deliver by December 2018, you indicated just now that you gave him some dates when you thought it could be delivered. What were those dates?
Sir Terry Morgan: There were three dates that had different probabilities attached to them, but took you into the third quarter of 2019.
Q73 Sir Geoffrey Clifton-Brown: The project would be complete by the third quarter of 2019.
Sir Terry Morgan: The key to Crossrail was always around a date on which we could open what was described as stage 3.
Chair: The central bit.
Sir Terry Morgan: Stage 3 was the opening of the tunnels between Paddington and, basically, Canary Wharf. That was always the centre of attention, because that is where we had to get the rolling stock to work, it is where we had to get the stations complete, and it is where we had to get the systems integration in play. That was our key date. Stages 4 and 5 were beyond that, but to be frank with you, it was all about getting confidence into a programme.
Q74 Chair: So the third quarter of 2019 was when you expected that.
Sir Terry Morgan: Yes.
Q75 Sir Geoffrey Clifton-Brown: Nevertheless, you had been responsible for this project; for producing a railway. What were your predicted dates for stages 4 and 5—that is, to complete the railway—at that meeting?
Sir Terry Morgan: At that meeting, it was off into 2020. I do not have the data with me—I could provide a copy of it—but stage 3 was always absolutely key to the programme. Until we got stage 3 started, we really were not testing Crossrail as a programme.
Q76 Chair: Mr Wolstenholme, you left in March, and you said that you were raising warning signs that there were future target dates to meet certain milestones ahead. Are you saying, though, that when you left in March, everything was okay and there was a possibility that you could still deliver it by December 2018?
Andrew Wolstenholme: In March, with the information that we had, with the programmes that were put together from our supply chains, and with the time that we knew we needed to do the integrated testing, there was a considerable amount of risk. The discussion with the sponsors and the board is that we needed to test that risk through the course of 2018, in early summer and mid-summer. That was a clear indication that those risks were not certain to go one way or the other. There was still a lot of work to do, and a lot of risk to unwind. That risk came from two areas: one was the maturity of the train and rolling stock, and the other was the completion of the tunnels and the physical stations themselves.
Q77 Chair: Did any of your suppliers raise any concerns about the delivery dates and the milestones that you were setting?
Andrew Wolstenholme: I will focus on the area of most complexity, which was the completion of the SCADA system on the station. When you have put in the final pipes and wires—mechanical and electrical—then you put the complex systems overlay on it. We were using a specialist contractor who had done this. Of course, the novelty was that there were 10 stations’ worth of this to do. But based on the information that he provided in late 2017 on the then best definition of the scope that needed to be done, we were getting indications from programmes in our supply chain that the scope of work on each station was between three and six weeks. That is the information we were getting from our supply chain at the time, and that is what we loaded into our programmes. Of course, with the benefit of hindsight, that was materially inaccurate and wrong.
Q78 Chair: Who was doing an assurance of that? Sir Terry, you have talked about independent studies that were commissioned by the board. Did it look at the whole SCADA set-up? We saw some of that when we were in Tottenham Court Road. We recognise the complexity, even as lay people. Did you have people doing assurance about those deadlines?
Andrew Wolstenholme: I am not sure that in March 2018 we had an independent review of that particular area of scope. What we did say is that where we had complexity and needed assurance, we went out to go and find those reports. On several occasions, be it the system integration or some of our industrial partner’s Network Rail issues, we went out and had those separate independent reports to see whether the risks that we are looking at are the same risks that we are dealing with—is the knowledge from around the rest of the industry in the world comparable to that?
Q79 Chair: Were you personally walking the sites? Were you going and talking to the contractors on the ground about whether they could deliver this?
Andrew Wolstenholme: Constantly.
Q80 Chair: And they assured you, eyeball to eyeball, that they could deliver this.
Andrew Wolstenholme: Of course, when you can see physical activity, it is demonstrable. There is always an issue in the whole sector around how you measure progress in software and software codes, because a lot of the SCADA is about software and software codes. In order to take that risk off of the critical path, part of the integration strategy, which had several tiers on it, was to put all of this through an off-site laboratory. This is standard best practice in the industry. Thameslink had an extremely good off-site laboratory, and we took hold of the ideas. What that did in 2017 is begin to make sure that any issues and risks that there were would be detected earlier.
Q81 Chair: Sir Terry, did all these independent studies you have commissioned as a board flag up that some of these supplier promises were over-optimistic?
Sir Terry Morgan: No. These independent reports came in and were discussed by the board. We never kept those to ourselves; we always shared those with the sponsors. They reinforced the plan that was in play. As you might imagine, we had one or two issues in there that we needed to address—not least the behaviour of one of our contractors, and I think they are still not behaving in the way I would want them to. The behaviour of one of the contractors was very difficult, and that has always been an issue when you say contractors complain. In the 10 years in which I was on Crossrail, I was never approached about complaints about the supply chain. Did I hear rumours about what the supply chain thought? Of course I did, but I did not personally advocate an approach to concerns about the programme. Indeed, when in August I met a number of the key contractors who are still causing problems, two of them came to me and said, “Is there anything we can do to support an October completion date? We’ll do everything we can to support you.” My answer to that was just, “Get it done.” There was always a gap between the ambition and what got delivered on the ground, I’m afraid.
Q82 Chair: And you did not realise that till summer last year?
Sir Terry Morgan: As I said earlier, we put a lot of pressure on the supply chain to deliver—that was this whole thing about “can do”. We got many things done that people said were not possible. Of course, the job of a programme is to drive the programme to deliver beyond the expectations of what people thought was possible.
Q83 Chair: Whose decision was it? Was it at board level, Mr Wolstenholme? Presumably you recommended to the board that partners such as Bechtel were subsumed into the Crossrail team. We heard, when we went to visit Tottenham Court Road, this proud boast: “We’re all one team—we’re all Crossrail. It doesn’t matter which company we’re in.” That took away some of that independent assurance.
Sir Terry Morgan: Actually, we had five levels of assurance, compared with the Olympics, which had three. We had 150 audits in the last year of the programme. In the assurance process, we had the NAO, the IPA and all sorts of people.
Q84 Chair: Not everyone was necessarily looking at the whole thing. What about project management assurance?
Sir Terry Morgan: The problem, as Andrew said in the introduction, is that this is a digital railway. Nobody has done that in the UK before. There are very few people around the world—
Q85 Chair: Ultimately, it is project management. Bechtel are an international company that do project management, but then instead of project managing one step removed they were brought into the whole. Did that make a difference?
Sir Terry Morgan: I know you asked the question to Andrew, but I was actually in Crossrail when the decision was made to change the model of how Bechtel should operate. They won the bid. They put resource against what they thought it needed to do. Surprise, surprise: the costs started to escalate quite significantly. They wanted a lot more people. What we didn’t want, as we laid out much bigger contracts in the supply side than we had originally envisaged, was big contracts with a big contractor—Crossrail—sitting on top. We saw that as being very cumbersome administration, so we agreed to change it.
In a very simplistic way, to put this in context, we had probably 400 people from Bechtel. This is not a small piece of Bechtel inside the programme; this was very, very significant. We had a rule that we tried to operate that if we had appointments to make that we thought would last the lifetime of the programme, it was better value for money for us to try and recruit them and to create a more integrated team. That was the driver for doing so.
Q86 Chair: To be clear, these Bechtel people became Crossrail employees?
Sir Terry Morgan: They were. There was almost a proud boast, to be frank with you, that when you went on a site you wouldn’t know who was Bechtel and who was Crossrail.
Q87 Chair: That was exactly what we got, you will be pleased to hear, from the project manager when we were on site last week. But my point is, once you have wrapped them into the structure, you are losing the contractual bite that you have got over them.
Sir Terry Morgan: With 400 people from Bechtel they dominated the delivery side of the programme.
Andrew Wolstenholme: May I add something? If you are comparing a thin client—a client with very few people in—with a Bechtel, for instance on the ODA Olympic Games, the scope of the Olympic Games, which was £1 billion of revenues and £2 billion of transport programmes, all on the surface, with fairly simple logistics, are the sorts of projects that you can draw lines round, and you can articulate where those risk boundaries are. It is a very appropriate model to have an ODA-style layer of programme or project managers who you incentivise to do that.
Q88 Chair: The ODA being the Olympic Delivery Authority.
Andrew Wolstenholme: Correct. If you look at the Crossrail scenario, this is 42 km. There are 52 construction sites on Crossrail in order to get down to the access. The reason that I was explaining the sorts of risks that were going on in 2014-15 is because an unknown obstruction in Moorgate caused a tunnel boring machine 2.5 km away to stop, which was the right place to stop it. That created the circumstance where the whole of the tendered track installation programme was reversed.
I think the implication you are giving is: “Why didn’t you just ask Bechtel to take on that risk?” These are enormous risks, and the reason that we ended up with 36 different suppliers is because, if you look at the analysis that was done in 2010, the capacity of the industry to be able to take on risk, to take on skills and capability, was a very scientific approach. When those programmes were let over a five-year period, vertically and horizontally, we did a very careful piece of analysis to determine how strong the balance sheets were, how able they were to take on the technology, and how to optimise the interface risks between those.
We got this report looked at by an independent group of procurement experts, which acknowledged that the compromise that you have to make, in having the right size of contractors and understanding the risk between these different interfaces, was around what we had at the time. If I had that decision again, that is the answer you would get. There is no way that you can put one organisation with one balance sheet between the complexity of those risks and seemingly protecting you against managing the complexity of those interfaces.
Chair: We are going to come on to some of the odd incentives with payment in a moment.
Q89 Sir Geoffrey Clifton-Brown: Mr Wolstenholme, you will have seen, I hope, the NAO Report. It is an agreed report between the NAO and the Department for Transport. I would like to take you to paragraph 2.19 on page 25, which states: “Crossrail Ltd integrated contractors from the two partner firms”—that is Transcend Ltd and Bechtel—“within the central Crossrail team…this meant that Crossrail Ltd had few commercial levers to drive management and integration of the overall programme, and took on all the risk of doing so itself.” Is that not pretty damning of your strategy in this respect in not allowing the project managers to get on and do the job themselves?
Andrew Wolstenholme: We did not have a Carillion failure on our job. We were very conscious of the ability and capacity of our supply chains to take on risk. The form of contract is the NEC—the new engineering contract—and there are various formulas with it. You can fix the price. When you fix the price, you are certain of the scope, and normally you are certain of that scope because it is manufacturers and you can draw a line and a boundary around it. We did not have that position. I do not know of any option A contracts that we actually delivered, apart from moving the 4.5 million cubic metres of spoil, which was a very fixed volume, to Wallasea island.
It was appropriate to have a form of contract that was a target cost. That allowed you, at the level of scope definition that you had in the early days, to be able to get the contracting organisations in early to make sure that their ideas and innovations were locked in. Once you then agreed that scope, which still had many years of design definition to go, you then allocated a target cost. The reason that form of incentive was chosen was because those contractors then had their own interface schedule and scope of work. They were looking after their own target cost around that agreed number, and you have got to get that number in place.
Q90 Sir Geoffrey Clifton-Brown: It is a long answer to a simple question. The question is: was it wrong to let those two contractors off the risk of not delivering the job on time?
Andrew Wolstenholme: I think I have explained the size and the scale—
Q91 Sir Geoffrey Clifton-Brown: I am not interested in the size and scale. I am interested in how you structure a project like this.
Andrew Wolstenholme: If we went back and did it again, I believe that the right answer for the complexity and the number of interfaces is to understand the risk and then get the best people to manage that risk. You cannot do that by just putting them in a single office and saying, “That risk is yours.” The complexity and dynamics of the risk profile were such that you needed every day to manage that risk register and apply the appropriate technologies to manage that risk. I would not change it. If I went back and did it again, on balance I think we had the right number of contracts and the right contractual mechanism in letting them through a target cost. On balance, there were more examples where the combined team found a way through, rather than holding on the risk. There is no way—
Q92 Sir Geoffrey Clifton-Brown: We have exhausted this subject. Can I move on to a different subject? I made two official visits to Crossrail, one in January 2017 and one in June 2018. In both visits I was told how wonderfully all the civil engineering—the tunnelling and the rails—had gone. A huge amount of time was spent on the escalators. Not once was I told that there were problems with the systems engineering and the integration. Is it a fact—it is a useful lesson to learn for HS2—that you and your management spent too much time concentrating on the civil engineering aspects of this project and not enough time on the systems engineering?
Andrew Wolstenholme: I do not think it is as simple as saying, “Is it a fact?” If you look at how the transition of the programme went through, there were five different transitions to the programme. You are right in saying that the early three or four years were focused on the enabling works and then the heavy and complex tunnelling works. If you look at the programme in 2013-14, the organisation changed to ensure that we had the right leaders, supervision and skill sets to manage the new supply chains that were coming in. First, those represented the stations. Later on, we then had another organisation shift, in order to make sure we had the right skill set to look at the complex rail systems. The final organisation shift that we had enabled us to make the transition between a construction programme and Transport for London, and make sure that we had those skill sets to manage the remaining risk.
I very consciously looked at the organisational design, to make sure we had the right skill sets. If you go back to the beginning—I will give this one example—there were only a few dozen people who had the skill sets to dig soft tunnels. On reflection, if you look at the industry that is presented to us, there is still a huge skills shortage around certain elements that you need at certain times. I think that, on reflection, we should probably have matched the great work that we did in the tunnel with training in the academy, to make sure that the skill set was robust. With the strategic resource management at a level, there were competing programmes at the time and, yes, it is true to say that those resources that we needed in 2018 were scarce and difficult to get.
Q93 Chair: Before I move on to Mr Wild in a moment—I warn him of that—you have both talked about the challenges of system integration, including the fact that you have stations to complete, the lines—all of these elements and complexities. But it was always known that integrating it was one of the challenges. I still cannot quite understand why you did not appreciate the challenges of integration, particularly—as we saw from our visits—some of the challenges of the IT, at an earlier stage. Seriously, did it not ever keep you awake at night—that difficult last year, six months, or whatever time period—when you were beginning to have to integrate these systems? And the fact, Mr Wolstenholme, that when you left you still were not integrating systems in March for a December delivery, because you were not at that position yet—did that not sound an alarm bell to you, in all honesty?
Andrew Wolstenholme: The whole subject of integration is huge when you look at the enormity of this programme—
Q94 Chair: But that was the programme.
Andrew Wolstenholme: Right, so we set up something called the railway integration authority back in 2015-16. That authority looked over seven different geographical areas, through eight different stages, and the element that we are talking about is one of those elements within one of those areas.
Again, with the benefit of hindsight, three things were coming together on those stations that had never been done before. One is that there were 10 of them; and the second is the aspiration, which is quite understandable, of TfL to produce a digital railway. Thirdly, not only was that going to be controlled locally in the station environment, but it was going to be controlled remotely, and I think that people had not appreciated and not understood the complexity of that.
I also said that the relatively late arrival of the mechanical and electrical systems compounded three elements: shortage of skills; a misunderstanding of the complexity; and this happening relatively late in the programme. That is why this risk appeared relatively late—because these things only come late. This activity is the final activity that you do—
Q95 Chair: Sorry, but a late understanding of the complexity? You have just outlined very clearly why it was a complex project, although I have to say, regarding the number of stations, that there are plenty of metros and so on around the country that have built 10 at a time. There are examples, perhaps not all in the same place at the same time, of trains run on a digital basis without drivers, for example, although clearly they are not as complex as Crossrail.
However, you must have seen these complexities coming. You took on, with enthusiasm, this huge and complex engineering project as chief executive. You must have been enthused partly because of the complexity, surely. And you are telling us that you did not really understand that until just around the time you were leaving?
Andrew Wolstenholme: I think that—Mark can probably help out here—a normal station in a TfL environment would potentially have around 100 points where you monitor. These stations that we are discussing were specified with around 1,000 points to monitor.
Q96 Chair: This is the monitoring for the drilling?
Andrew Wolstenholme: No. This is monitoring for the supervisory control and data acquisition, or SCADA, system in a single station, which was of an order of magnitude more complex than a station that you would see on the network today.
Yes, with the benefit of hindsight, I think that no one—including the specialist supplier, who was actually allocated to provide this—had understood the complexity and the volume of work that needed to be done right at the end of the construction process.
Q97 Chair: Right. I have to say—Ms Kelly is a welcome recipient of our questions on the complexities of integrating complex rail projects; I am sure she enjoys it every time she comes—it is not a new thing that the integration of many rail projects above ground is challenging, so obviously below ground was more challenging. Sir Terry, as a board, do you think you were asking the right questions and getting the right assurances? If someone came to build an extension in my home and said, “Don’t worry. It will all be fine,” I woud be asking them a few more detailed questions to make sure they were really on the timetable and the detail. Were you getting the right information from the executive? Was your board up to it and asking the right questions?
Sir Terry Morgan: I think we had people on the board who were up to it, with the skill sets. We always thought very hard about who should be on the board and why they should be on the board. For example, we had a retired chief operating officer from MTR in Hong Kong, which, at the time we recruited him, was seen as the benchmark railway around the world. He joined the board.
Q98 Chair: Was he asking the right questions?
Sir Terry Morgan: Absolutely, and he still asks lots of interrogating questions. To be frank with you—I know this is not necessarily the thing that people want to hear—one of the things that stopped us having in-depth understanding of system integration was the lack of a train. There was an adverse notice issued to the sponsors in 2017 that the train we were expecting to see delivered, which was under contract with TfL, was not available. It is a digital railway, so the heart of the railway is in that train. That is what drives the system.
Q99 Chair: But there are so many other complexities. I don’t think the absence of the train was the sole thing.
Sir Terry Morgan: No, not the sole thing, but that train is probably the best part of two years late now.
Chair: I am not going to get drawn on the train. I am going to bring in Mr Lee Rowley and then Sir Amyas Morse, and then we are going to move on to some other subjects before coming back to Mr Rowley.
Q100 Lee Rowley: Mr Wolstenhome and Sir Terry, you have extensively explained how complicated this programme is and said that there is an incredible amount of complexity. It was perhaps unique in the world, in terms of its complexity. Why, then, does the NAO describe your plan as “aspirational” on page 7 of its Report?
Andrew Wolstenholme: The programme that we referred to—the MOHS, or the master operational handover schedule—
Q101 Lee Rowley: I just want you to answer the question. If you were trying to manage such a level of complexity, why did you have only an aspirational plan?
Andrew Wolstenholme: I have to say that I don’t agree with the word “aspirational”. The programme has more than 20,000 activities to it. It was built up bottom-up by the suppliers’ programme. It was very carefully composed in order to make sure that it served the purpose of being able to deliver the messages to the board, the supply chain and the audience that needed them. As you know, building up programmes for a programme the size of Crossrail means hundreds of different supply programmes coming together.
Q102 Lee Rowley: Yes, and unfortunately the NAO don’t think you had it. “Crossrail Ltd started to produce a detailed, realistic, bottom-up plan in late 2018”—after you left the company. They are indicating that they do not believe you had a plan that was sufficient to manage the complexity that you have been so keen to tell us about for most of the last hour and a half.
Andrew Wolstenholme: If you go back to the start of the programme in 2012—
Q103 Lee Rowley: I’d rather talk about the end of the programme, which is the problem.
Andrew Wolstenholme: I understand that, but the detail of the programme emerges as you go forward.
Q104 Lee Rowley: I don’t need any education in how programmes emerge. I want you to tell me why the NAO are wrong in their assessment, towards the end of this project, that you did not have a detailed, realistic, bottom-up plan until after you left the company.
Andrew Wolstenholme: With respect, they are referring to the element of work that was done at the time. I don’t think the NAO are saying that, for the whole of the eight years that provided an extremely successful outcome on those elements, we did not have a programme.
Lee Rowley: I’m just telling you what they have written. It is a very clear statement: “Crossrail Ltd started to produce a detailed, realistic, bottom-up plan in late 2018”, ergo you did not have one beforehand. The Department for Transport presumably agrees with that, because they signed this off.
Q105 Chair: Ms Kelly, while we are on that, do you agree with the NAO?
Bernadette Kelly: We signed the Report off. We have no factual disagreements with the Report. That was the NAO’s description; those are the NAO’s words.
Q106 Chair: Mr Lodge is nodding. He agrees with you.
Matthew Lodge: The MOHS was presented to sponsors as being the detailed plan about how we got the railway open. We assumed it was an integrated plan that would take the project from the civil side to completion and running the trains.
Q107 Lee Rowley: I am not sure I understand that. Do you agree that there was no detailed, bottom-up plan before late 2018 or not?
Matthew Lodge: The MOHS was presented to the sponsors—
Q108 Lee Rowley: Yes or no is the answer, not a detailed explanation. Do you agree with the NAO’s statement? It is a binary proposition—yes or no?
Matthew Lodge: Based on the Report the NAO put in front of us, I concur with what the NAO is saying—that the plans—
Q109 Lee Rowley: Excellent. Back to Mr Wolstenholme. The Department for Transport also agree that you did not have a detailed, realistic or bottom-up plan before you left. Why did you not?
Andrew Wolstenholme: We were able to provide comments to the draft NAO Report, and both myself and some of my project team disagreed, with great respect, on there not being a detailed programme. I accept that, for the thousands of different, complex areas of this programme, the details will emerge. I accept that a year before, in 2017, there was not the same crystallisation or granularity of the programme that was required a year later in order to deliver that detail.
I accept that, but that is how the layers, the detail and the granularity of programmes are produced. These emerge, and when the detail of the final supply chain is let and awarded, that is the point at which you can bring in that level of detail to your higher-level programme. In the early days these were concept designs. Later on these were detailed designs. Further on through the supply chain, you have the actual engineering drawings that you are going to deliver, and only at that point can you cascade these up into the final detail. I accept, for those elements of scope, that that detail emerged late in the day.
Lee Rowley: I was not seeking the final plan; I was seeking a detailed, realistic and bottom-up plan. The NAO does not think you have it.
Sir Amyas Morse: As I have listened to your testimony, I have heard that you did not realise how difficult the systems integration would be, you did not realise that the completion of the stations would be late, you did not realise that there would be site difficulties with ground state and so forth. All those added up to major difficulties that now mean the project will be late. You had the job of identifying things early enough to give assurance that the plan could be delivered. That was surely the job.
To say that this is so unique that it was a voyage of discovery all the way through is really not a reasonable position to adopt about a major project. You cannot say that you entered into it with a timescale, a cost and everything else, and that we accepted that as a kind of speculative set of indicators. That is not what we were hearing from you. We heard from you that this was all going really well and would be delivered on time until pretty late in the day.
We tried to explain in the Report that, by taking all the risk into Crossrail Ltd but leaving a lot of the variable factors under the control of the contractors, you actually did not know what was going to happen until quite late on. You have more or less admitted that, too. You were not really in control of the project—isn’t that a fair comment?
Andrew Wolstenholme: No, I don’t agree with that at all, with great respect.
Sir Amyas Morse: You don’t need to give me great respect. Just give me some evidence.
Andrew Wolstenholme: In taking on some of the huge challenges and elements that we had, no, we did not have the answers. We were not able to specify those things that we did not know. That is why I go back to many issues that occurred throughout the engineering. Perhaps we made that look quite easy, but we had to find ways around the challenges that we had, and we constructed a team that was able to find routes around those.
On the three risks that we had in 2017, by then there were only 13 months to go before that incident at Pudding Mill Lane. Right then we were able to reconstruct the programme of testing and commissioning, but still providing sufficient time. In all of these we had programmes from our specialist suppliers and dashboards to understand how those risks were dynamically changing at any time. Based on the information and expertise that we had, we made clinical, open and honest decisions that we put forward to the board and to the sponsors in order to move those risks forward. It is incorrect to say that Crossrail Ltd was managing the risk and that the contingencies were held apart.
Sir Amyas Morse: In July 2018, you estimated a £300 million cost increase and a nine-month delay. What has actually emerged is a £2.6 billion cost increase and a two-year delay, at least. That is not a project that is under control. I’m sorry; I know it is easier to talk with hindsight and I am sympathetic, but really that is a massive, massive difference. You cannot just say that was all a surprise.
Andrew Wolstenholme: I am not going to comment on the cost that has come out in July 2018.
Sir Amyas Morse: You don’t think they are legitimate? You don’t think they were already in the system and on their way through?
Q110 Chair: Sir Amyas, we are not having a conversation. Mr Wolstenholme, on the costs?
Andrew Wolstenholme: I can tell you where the estimate was in December 2017 and in early 2018, which was submitted to the Department for Transport and our sponsors. It looked at several different scenarios for how you would allocate that cost to the programmes. That exercise was called the book-ends, and those were the book-ends between £150 million and, I think, £320 million. That was the allocation that went up for additional funding, and that was in addition to the funding envelope. Those were the exercises I am very familiar with. I am not able to bring myself forward by nine months and to comment on the cost plan that has been put forward at the moment.
Chair: I will move on, then, to the current incumbent and Mr Wild. Let’s start with stations—forgive me; I forgot Mr Davies.
Q111 Chris Davies: My question is to Mr Wild, if I may. Mr Wild, I have been watching your facial expression as Mr Wolstenholme has been giving his evidence, and I must say that from this angle it is very interesting. You are now the chief executive; the confidence of the country is now in you to finish that project—I would say on time, but we do not know when that time is going to be. When you look back on what has happened, would you have done things the same way as the previous chief executive?
Mark Wild: I sat on the board alongside Tony and Andrew, so I know the project very well. I will give you my personal view, which is more aligned to the NAO’s: this is a failure and we must learn from it. When I look back at it, not with 20:20 hindsight but with the reality I face now, Crossrail had exceptional system engineering, a system of monitoring and a detailed plan, but all the dials were not telling us of the latent risk that did exist. I completely agree with Amyas that the costs that have come out were there anyway. The real issue on Crossrail, to me, in 2017 was that the system integration authority just did not have the appropriate skills, measurement or methodology. That is a huge lesson for future projects.
That is not a criticism of anybody. I think the Crossrail team were absolutely exceptional, but it is just a gap; something so big always needed a powerful system integration authority, not a railway integration authority. The difference is that system integration is much more about sensing these issues. I agree that the project should have spotted them; I think about it all the time. It is a systemic issue, so there is no person to blame. It is simply that the project did not have a system integration authority that could assess the risk.
Clearly these costs are real, we incur them every single day and they were there already. The lesson is for the future and we really must recognise it for future projects. My face, by the way, is nothing more than my face, so don’t worry about my face.
Q112 Chair: I think you will be in front of us a few more times to discuss these very issues. I will move on to the situation you have inherited. Let’s start with the stations. How much work is there to do at the stations? We were at Tottenham Court Road and we have heard about some of the others, but do you want to list, not every single station, but the work overall and any particular stations that still have to have major works done?
Mark Wild: There are three critical paths in this project to complete, and a sprint to Christmas: the stations, the routeway and the trains.
Q113 Chair: I want to talk about the stations, if we can talk about that.
Mark Wild: On the stations themselves, there are 10 stations and eight shafts and portals. They are in stages of completion. You went to Tottenham Court Road, which would be the most complete of our stations. Although it looks physically complete, the electrical engineering, the migration of activities, will take us until August or September. That is a good indication that our most advanced station will not be complete until August or September. They are doing very well at Tottenham Court Road. At the other extreme, Bond Street is 18 months to two years away from completion; that’s the reality. Its design is not complete and will not be complete until September or October. You have got that big range. In the plan we have come up with, we have a good plan that sequences the activities and that enables this railway to open with all the stations, excluding Bond Street.
Q114 Chair: Were you aware of the state of readiness of the stations when you took over in this new role?
Mark Wild: My objective is to look forward. I can only tell you what I did when I got in the seat. We spent at least 10 weeks going through, root and branch, counting every camera to do and every wire to install. It is fair to say—and we have to accept this on the programme—that we didn’t really have a full grip during early to mid-2018 of the work that was left to do. The position I am in now is that I would tell you exactly what is left to do on every station. I think that is a fair challenge that we must keep very vigilant about, going forward.
Q115 Chair: Just to be absolutely clear, when you took over were you under the impression that work was more complete than what you inherited?
Mark Wild: I have been surprised by the lack of completion and the supply chain particularly; their programmes have proven that there is more work to do than we anticipated.
Q116 Chair: Ms Kelly, were you led to believe that the stations were more complete than they were? Mr Wild has uncovered some problems, he says. Were you aware of that before?
Bernadette Kelly: We would have been receiving the same information as the company was providing about the state of completion at the stations.
Q117 Chris Davies: Mr Wild, you have indicated and said quite clearly that you have been on the board for two years. Why were you not made aware of how far behind this project was?
Mark Wild: On the stations? Are you talking about the stations particularly?
Chair: Yes, on the stations particularly.
Mark Wild: I think there was a real issue—a failing—that the project clearly did not have enough granularity of information coming from the supply chain.
Q118 Chris Davies: In your opinion, who should have reported that to the board?
Mark Wild: The supply chain clearly should have been bringing their plans up and they should have been reported through the programme director to the board. I think there was good transparency, actually, but what wasn’t there, particularly in 2018, was the ever-increasing optimism bias of recovery plans and mitigations. It was becoming increasingly difficult to say, “Well, I do 10 of these items every week. Suddenly I’ve got to do 50 items.” What really happened from early 2018 onwards was the hockey stick effect.
Q119 Chair: Sir Terry, on that point—that the board didn’t have enough information—do you agree with Mr Wild? Was the granularity of information missing?
Sir Terry Morgan: I am listening to a conversation that obviously happened after I had left. If you take Tottenham Court Road, I am very aware that we were planning an event with the contractor last June that presented the station in as complete a form as we thought it could be. At that stage SCADA was not able to be installed; that was something we were still phasing through.
Q120 Chair: So you are saying that you didn’t know that it was as unready as it was?
Sir Terry Morgan: If Mark were to talk about Bond Street, it was always reported quite openly to sponsors. Bond Street is a problem child, as was Whitechapel. Tottenham Court Road and Farringdon were in a much better state. We had problems at stations such as Custom House and Woolwich that were overcome.
Q121 Chair: But given that Tottenham Court Road won’t be ready until August or September this year—you left in December last year and Mr Wolstenholme left in March last year—you had assumed it would open in December last year, so you must have had some very positive information about it.
Sir Terry Morgan: You asked the question earlier if I visited the sites. I visited the sites probably every other week. I tend to look at progress—
Chair: It’s not so much the visits—well, it is a bit—as the metrics.
Sir Terry Morgan: If you visit, you physically see what is going on.
Q122 Chair: Obviously that is important, but what about the metrics that were coming up from the different suppliers? What was everyone telling you?
Sir Terry Morgan: At every single board meeting we probably spent the first two hours going through a tour of the programme, with metrics about completeness. If reference was made, for example, to escalators, it included a suite of assets in terms of their completion. So yes, that was reported to the board and indeed reported to the sponsors.
Q123 Chair: Did anyone at any point flag the fact that—as we saw when we visited—sometimes you have a small contractor having to ramp up or spread themselves across 10 stations and sequence their work with only a few people involved? Were you aware of the skills issues that Mr Wolstenholme has also touched on?
Sir Terry Morgan: On some examples, yes. There were some examples, but to be frank with you, over the life of the programme we were always told that we were going to have skills shortages. Many of them did not materialise, because we planned our way around them. A question was asked right at the beginning about what percentage of our workers were not British passport holders; 40% of the workers on Crossrail were not British passport holders. We had a skills programme. Andrew is quite right: we focused on tunnelling, because that is where we saw the shortage. We had done a national survey. We planned on 300 apprentices; we had 1,000. There were lots of things that we did as the programme progressed to try to mitigate the situation. These are good examples.
Q124 Chair: But some of these were very specific skills and very specific suppliers, so they would be a bit harder to—
Sir Terry Morgan: Later in the programme, to be frank with you, as the M&E work was done.
Q125 Chair: Did you not see any of that coming? You had very specific, technical suppliers.
Sir Terry Morgan: In that regard, we were given assurances by our contractors that they had the resources that they needed.
Q126 Chair: Contractors often say positive things because they want the contract. Presumably you were a bit more sceptical than just accepting their word.
Sir Terry Morgan: As if we would! No, we did not. We wanted evidence.
Q127 Chair: Good. Were you surprised when Mark Wild sat on the board alongside you and said that he did not find that there was enough granularity in the plans? Does that chime with your own experience, Sir Terry?
Sir Terry Morgan: I am disappointed, to be frank with you. As Andrew said at the beginning, I am disappointed that we did not meet the date. It was a very heavily driven programme, and I do reflect on—
Q128 Chair: Do you agree, in hindsight—if hindsight were a company, we would all be buying shares in it—that you needed more detail at board level?
Sir Terry Morgan: With the benefit of hindsight, there is always a balance to strike. I am in no doubt that some of why this is going back much later than I ever imagined is because we took the pressure off the programme. There was real pressure on the programme to get the job done, and when somebody says that there is some variability in the date, contractors—to go back to your point—will always take advantage of it. It is an opportunity.
Q129 Chair: We are going to come back to contractors taking advantage in a moment. Although I am tempted to put that to Mr Wild, I want to pursue the issue of the stations, which we were in the middle of—Bond Street, for example. Do you have an estimate of the extra costs for the work that has to be done at Bond Street, which was always a problem? Do you have a figure for how much it will cost to complete Bond Street?
Mark Wild: No, we don’t, actually. Bond Street remains—alongside the train integration with the signalling system—our biggest challenge. What we need to focus on with Bond Street is getting to a certain point of completion to allow the train system to run. It is one of the outstanding pieces of work that we aim to do over the next six to eight weeks.
Q130 Chair: So you will have a figure in six to eight weeks for Bond Street.
Mark Wild: We will.
Q131 Chair: What about Canary Wharf?
Mark Wild: Canary Wharf is an interesting case. It was built by Canary Wharf Group—Andrew and Terry are probably better people than me to speak about it. We have spent nearly £80 million on changing the equipment at Canary Wharf to make it equivalent to the safety standards required for a London underground station. It was completed in 2015, but we still have not finished the work at Canary Wharf—we will not finish Canary Wharf until September or October this year. It is fair to say that the quality of the work that we found in Canary Wharf, completed by Canary Wharf Group—before my time, actually—has had to have a wholesale retrofit, particularly in the safety systems.
Q132 Chair: You are saying that that is what it costs, but who is paying for it? Presumably Canary Wharf is.
Mark Wild: No, that is in our expected cost to completion.
Q133 Chair: So there’s no liability for Canary Wharf, even though they built it?
Mark Wild: We have some commercial conversations with Canary Wharf, but this is more Andrew’s gig than mine, because it was 2014-15. I believe that we accepted the station to some extent, so although there is some commercial discussion with Canary Wharf, my focus is on getting the job done. We have spent about £80 million.
Q134 Chair: Let me just ask Mr Wolstenholme about Canary Wharf, then—as you say, it was in his time.
Crossrail—under you, Mr Wolstenholme—accepted the station at Canary Wharf. Is that right? You took it over.
Andrew Wolstenholme: Canary Wharf is a very interesting subject, because the contract to let Canary Wharf was in 2009. It was the first part of the site. I arrived halfway through 2011. It was let early because it was on a private site that had the opportunity to start.
Q135 Chair: But they built it.
Andrew Wolstenholme: They built it. It was a design-build contract. The design for the railway station, as opposed to the top site, came many years later. This was a risk that the Canary Wharf Group and, I believe, the early part of Crossrail took: that it was always likely in the design-build not to meet the full specification.
Q136 Chair: But it would be a box that you could add things into. Was that the idea?
Andrew Wolstenholme: It was a box—it was fitted out. You could go and visit the upside of that box right now, but it went down into the passengers’ concourse spaces and certain elements conflicted, from a specification point of view. What is right for a shopping centre is not right for a rail system.
Q137 Chair: Even without the final specifications, some of those things around safety and fire safety are standard across the rail industry—that is one of the challenges of building a railway. When you took over, did it meet those requirements and specifications?
Andrew Wolstenholme: No. We had a programme of review and change—
Q138 Chair: Canary Wharf paid for that change, presumably?
Andrew Wolstenholme: They paid for some, yes, but as a commercial organisation, Canary Wharf Group say, “We are going to build and pay for what was in the original specification.”
Q139 Chair: Do you know what Canary Wharf spent on that? I recognise that with the distance of time and the fact that you are no longer in the role, you might not have the figure to hand.
Andrew Wolstenholme: I don’t know.
Q140 Chair: Maybe you can find out.
Andrew Wolstenholme: We can look it up.
Q141 Chair: Sir Terry, the board agreed—or had to—take Canary Wharf station back. Do you have any idea of the costs saved by Canary Wharf coming in?
Sir Terry Morgan: I was told not to guess, but—
Q142 Chair: Tens of millions? What order of magnitude?
Sir Terry Morgan: Maybe £300 million.
Chair: And it is going to cost £80 million—
Sir Terry Morgan: But then please look at the cost of Bond Street and Tottenham Court Road. That figure is what Canary Wharf has said.
Q143 Chair: But for £300 million, Crossrail took over a station that still needs £80 million of spending. Am I right?
Sir Terry Morgan: I am not sure that Crossrail has even taken it over yet. That needs to be clarified.
Q144 Chair: But you, as Crossrail Ltd, took it off Canary Wharf? Maybe I misunderstand. Mr Wild, is that right?
Mark Wild: That is correct, yes.
Q145 Chair: So it was adopted by Crossrail Ltd when you were the chair of the board, Sir Terry. Am I right?
Sir Terry Morgan: You might be right Chair, but the board’s decision was not to say, “Yes, we accept.” This was a—
Chair: A technical—
Sir Terry Morgan: Somebody will have to go away and check. I really don’t know the answer to that.
Q146 Chair: But it must have been reported to you that the station was now part of Crossrail. Surely the board would have been aware that a major station was completed. It was the first to be completed, was it not?
Sir Terry Morgan: Yes. We used it in 2016 as a celebration of Crossrail.
Q147 Chair: So you were celebrating a £300 million station that now needs an £80 million retrofit.
Sir Terry Morgan: I don’t know about that.
Q148 Chair: Mr Wolstenholme, did you get any independent assurances that the station met all the requirements and that recognised that some of the technical bits may have changed as time went on?
Andrew Wolstenholme: In 2017—I can look up the precise date—when the station was due to be handed back to Crossrail for care and custody until it was due to become operational, we did a full reference survey of the completeness and of its response to the specification of the design that emerged in the eight years since. It was decided that if we instructed Canary Wharf through a design build, it would cost more money than if we took it over and retrofitted the elements that had changed. I am not aware of the £80 million number that Mark has quoted. I am certainly aware that there were technical aspects that had to be refitted and retrofitted, which came at a cost. We worked very hard to make sure that costs that were due by Canary Wharf were paid by Canary Wharf. Some of those costs were risk exposures to the programme.
Q149 Chair: It is amazing. This project, as we know, is late and over-budget—as the NAO, the London Assembly and the Committee have highlighted—but the Canary Wharf situation is extraordinary. It was built by a third party, yet you did not highlight that as one of the risks in the previous hour or so of discussion. You had a station and knew that there were challenges with it. Did that not suggest to you that there would be problems?
Sir Terry Morgan: I would have to check the board minutes. I think that the number that was reported to the board was nearer to £30 million, but it has to be said that I don’t know.
Q150 Chair: I will stop quoting the figure, but regardless of the figure, it is the principle of the station being built by a third party and brought in. You had a physical station and knew that there would be challenges in putting the systems in. Surely that was a great big warning sign that it was, if nothing else, quite a complicated project and that even the finished station needed a lot of work to make it operable to deliver Crossrail. Did that ring any warning bells with you, Mr Wolstenholme? Obviously, you were dealing with quite a challenging and snagging project—to put it mildly—with Canary Wharf and commercial negotiations about who would pay for it. Am I right?
Andrew Wolstenholme: As I said, we were aware when taking over. I think that there is a lesson there. It was heralded as a great breakthrough to get a completely private sector organisation to start early and to deliver a complete railway station. The prices have been compared with other stations, and there are some advantages and disadvantages. One of the disadvantages when you hand over to a completely private organisation in an isolated environment is that you have to be very sure, eight years in advance of when it is going to be taken over, that you understand exactly what needs to be built. On some of the lines of delineation, had they done less of the fit-out work and left that for a single pass by the railway system engineers later on, that possibly could have been a better answer. But this was started in 2009 and I was aware that when it was handed over for care and custody in 2017 there would need to be a retrofit package. The precise nature and cost of that is something that one would have to ask Mark.
Q151 Chair: Remind us when you became chief executive.
Andrew Wolstenholme: In July 2011.
Q152 Chair: At that point, did you go and look at Canary Wharf or talk to Canary Wharf about what they were delivering?
Andrew Wolstenholme: At that stage, it was a big hole in the ground in the middle of a part of the waterway. It was at a very early stage indeed.
Q153 Chair: So there was plenty of time for Crossrail Ltd to have lots of negotiation with Canary Wharf as you were planning the new station so they were all—
Andrew Wolstenholme: As I think I have explained, to instruct a private developer around an emerging design as they are going on is a balanced decision, and it was not until probably 2014-15, which is when Canary Wharf Group was pretty much completing that—the facilities there have been open to the public now for three years—that the detail of some of what is being retrofitted now was emerging. The detail of what the design should have been was emerging—that is how this thing was set up—around the time that Canary Wharf were completing, so we knew that that retrofit had to come after it was handed over for care and custody by us.
Q154 Chair: We do not have time now, but I think Ms Kelly will have many lessons to learn from this for the Department. Before I pass over to Mr Davies, can I move on to the remuneration issues? Sir Terry, why did you let Mr Wolstenholme leave in 2018? Was that always the plan, or were there any particular reasons for it?
Sir Terry Morgan: It certainly was the plan. As Andrew indicated, for the whole of the Crossrail team, there was a very detailed transition plan. As far as I was concerned, we had Andrew and we also had the programme director, and at some phase in the programme the conclusion was we would not need both, so in terms of succession planning it was a non-risk route. In other words, the individual who then became chief executive was a guy called Simon Wright, who we had recruited from the Olympics. He worked on the executive and indeed was on the board for three years, so he was not new to the programme. He did not have to go through some learning curve; it basically was his programme.
Andrew and I had spoken about when was the right time for him to move. He was going to leave in December ’17. I had a conversation with him because of the problems that we were having at that time—emerging problems around things like the explosion. I asked him to stay a little longer, which he readily agreed to, and I am in no doubt at all that if I had asked him to stay longer, he would have done so. But in the meantime, quite rightly, he started thinking about his future. I guess that conversation took the best part of a twelvemonth. Obviously, there was a time when Andrew had to start thinking about his own future, and that is what he did.
Q155 Chair: Mr Wolstenholme, you asked the chair of the remuneration committee for your long-term incentive plan payment to be made in 2018. Was that because you were leaving at that point?
Andrew Wolstenholme: I did ask for a clarification. I was given that clarification from the RemCo and I accepted it, so I moved on.
Q156 Chair: To be clear, you had bonuses of £481,000 in 2015-16 and £160,000 in 2016-17. Given what we know now, Sir Terry, why did you agree those payments—particularly that last one in 2018—to Andrew Wolstenholme?
Sir Terry Morgan: Every year Andrew and I sat down and I did an appraisal of his performance, and we also updated the objectives for the coming year. That is what we did every year. The result of that was a map of objectives against performance. On that basis, we scored his performance against those objectives. They were audited to make sure that those metrics were correct, and we made adjustments as a consequence of that. To my knowledge, I only ever made one where I used a degree of discretion, because I wanted to reflect the outstanding performance the executive had achieved around the safety of the programme, which is just world class.
Q157 Chair: How many members of your board were on the remuneration committee?
Sir Terry Morgan: Four.
Q158 Chair: Were you a member of it?
Sir Terry Morgan: I was indeed. I was in attendance, I should add.
Q159 Chair: Right, so you would give advice about the appraisal process.
Sir Terry Morgan: There was that, and to be frank there was a consultation process that was in the framework agreement with the sponsors. Although, I suppose, in governance terms, it should have been the chair who spoke to the sponsors, because I had the relationship with the sponsors I took on that role for many years, to consult with TfL around the recommendations that the remuneration committee was making.
Q160 Chair: Ms Kelly, I know that your Department has given evidence to our sister Committee, the Public Administration and Constitutional Affairs Committee, about remuneration and bonuses. Do you have any comments about the level of bonuses at Crossrail, and whether you felt that the system was right and that the payments were justified?
Bernadette Kelly: Sorry—do you mind repeating the question? I couldn’t quite catch it.
Q161 Chair: You gave evidence to our sister Committee about remuneration and bonuses—someone in your Department did anyway. I will find the exact reference and whizz it round to you, so you can have a look at what someone—you, I’m sure—wisely said. You were talking about limits on remuneration, and it gives you some wriggle room for certain commercial elements of this. I will pass the exact copy over now.
I just wondered whether you think that there are issues here. We have also had issues with HS2, as you know, with redundancy payments. We have had some exchange about that. Do you feel that these remuneration packages were well handled? Do you have any comments to make about it?
Bernadette Kelly: It was not me who gave this evidence, actually, so I am reading it as I speak. My views are very much, as I indicated at the last hearing on remuneration, particularly in these arm’s length bodies delivering these large projects, which are very challenging roles, that we need to have a remuneration framework that allows us to recruit people with the right sorts of skills and experience.
In this instance, I would take the view now that there was insufficient sponsor oversight of, challenge of, or opportunity to be consulted upon the remuneration decisions. It is important to remember in this regard that our role was very limited. Crossrail is a wholly owned subsidiary of TfL. It had a slightly different role in being consulted on remuneration, but I would say that both sponsors had a very limited role, and the accountability in that sense was possibly not what I would expect to see.
As we now organise this within the arm’s length bodies that my Department is responsible for, I would say that certainly we ensure that we have an opportunity directly to participate in the RemCo in a challenge function, and an observer function. We very much see it as our responsibility to do that. Also, we probably take a different view about the balance to be struck between base pay and variable pay in incentive structures as well. I would say that the practice that we have in place for our other arm’s length bodies, where we have been responsible for setting the remuneration principles more directly, are somewhat different and give greater accountability.
Q162 Lee Rowley: I am still not sure that I have had a clear view from the people who have spoken this afternoon about when precisely they thought that this project, in their judgment, was moving off track—when the probability of not hitting the date exceeded the probability of hitting the date. Sir Terry, can you tell me which of those monthly board meetings, where you have told us you assess the risk every time, you walked out of and for the first time said privately to yourself, “I don’t think we’re going to hit December 2018”?
Sir Terry Morgan: I will try to be very honest with you. I was at Crossrail for 10 years, and in those 10 years I missed one board meeting, in June. I had a conversation that Mark was at, and I had a number of conversations with the NEDs. That was the first time that there was a real concern about whether we were getting ourselves into a state where the programme was becoming impossible to deliver. In July, we sat down formally and acknowledged that the things that had happened in that period had just not happened in the way they should have done. That is when.
Q163 Lee Rowley: So you are saying June, in your view.
Sir Terry Morgan: That is when the sense was that the status of this was not making the progress to support what our ambitions were for the programme.
Q164 Lee Rowley: Mr Wolstenholme, you left in March. Again, in your private judgment at the time, did you believe that this project would deliver by December 2018?
Andrew Wolstenholme: I was not anticipating the problems that were found on the stations in terms of the complexity of the SCADA, because we were reading the dials at the time.
Lee Rowley: Just a yes or no will do.
Andrew Wolstenholme: No, when I left in March, the critical path still showed a way through to December 2018. As I said previously, there need to be check points, and that was around the maturity of the train and the software.
Q165 Lee Rowley: The critical path could show a pathway of 1%. In your judgment in March, did you think it would hit December 2018?
Andrew Wolstenholme: I would not have been surprised had the rail not opened because of the maturity of the train. I was perfectly well aware that there were risks to deliver on the stations. There were three stations.
Q166 Lee Rowley: And was there a greater or lesser than 50% chance of December 2018 not being hit in your final meeting before you left?
Andrew Wolstenholme: I am not going to put a figure on it, but it was greater than 50%.
Q167 Lee Rowley: You don’t have to put a figure on it; you can just give a rough view on it. Was it greater or less than 50%?
Andrew Wolstenholme: Throughout this programme we have been asked to record P50. As you know, P50 means there is an equal chance of meeting or not meeting it. Back in 2012, we set P65. There was a greater than 50% chance, but there needed to be points in early summer and summer when those needed to be measured.
Q168 Lee Rowley: To summarise, your balanced judgment in March was that it would still hit December 2018. There was a greater than 50% chance that it would hit December 2018.
Andrew Wolstenholme: I put a programme down that led to 2018, plus the cost that went with it and the risks. The risks were very clear.
Q169 Lee Rowley: I just want an answer. I am not going to allow you to go all the way around these things again.
Andrew Wolstenholme: Yes.
Q170 Lee Rowley: I find that incredible. Two people are running the most complicated engineering and change management programme in the world—in your words—and you were not aware that this project was going to fall over a month before it fell over. I don’t get that.
Sir Terry Morgan: If you were in charge, what would you do? You would seek independent experts to give you advice about what they thought of the plan. We shared responses and each time they said, “This plan is deliverable.”
Q171 Lee Rowley: The PRep on the Jacobs report said in April 2018—we talked about this, although admittedly you were not here—that it was there in black and white. I had read that Jacobs report. The MOHS “remains highly ambitious...Delivery performance has varied across the rail systems and concerns remain with general progress...testing has proceeded more slowly than expected...incomplete infrastructure...further schedule pressure...uncertainty as to whether or not signalling installations can be completed in time...The Infrastructure Managers (IMs) remain extremely concerned at the delays...More generally, the accumulation of delay across all areas of delivery continues to threaten the start”.
Sir Terry Morgan: You could read one report. I can take you to a report from 2012 that would say almost the same thing. That is their job—it is their job to report to the sponsors about programme concerns. What the board expected the executive to do is mitigate those concerns. That is the basis of the programme.
Q172 Lee Rowley: Any change manager who saw a report like that would say that that project was in deep, deep trouble. They would expect the head of the programme, which you two were, to understand the deep, deep trouble more than a month before you decided to reset the programme. I don’t understand how you reconcile those two things.
Sir Terry Morgan: Let me say to you that every single big project I have ever dealt with always had levels of risk. The job of the executive is to mitigate those risks all the way through. You will always find—
Q173 Lee Rowley: It had 16 red statuses. There was no green status, and there were two amber statuses in that report. I do not understand how everything was sweetness and light in this project, to the extent where you were not clear in your own head that this project was going to fall over until a month before it fell over. I do not understand how you intellectually rationalise that.
Sir Terry Morgan: I don’t accept your premise. As I said, the PRep report—
Q174 Lee Rowley: It’s not my premise; it’s the report.
Sir Terry Morgan: As I said, the report was discussed by the board with the executive and with the sponsors, because it was a sponsor report. They had their own meetings with the PRep. If the PRep had said—well, we actually had the PRep come in with another independent report. Jacobs did another independent report, and again reported that while there was a difference in the delta about how much more money might be needed, they still supported the idea with a programme that could deliver in 2018.
Q175 Lee Rowley: Ms Kelly, you had the benefit of being able to talk to Mr Lodge. When, in your private views, did you think this project was not going to deliver in December 2018? What was the date?
Bernadette Kelly: As we have indicated—
Lee Rowley: Just the date is fine.
Bernadette Kelly: I think our level of anxiety about the delivery date of this project spiked significantly after the incident in November 2017. That was a very significant moment: we knew at that point that three months had been lost on the schedule, clearly with 13 months to go and already significant compression of risks. That was a very significant moment, and I would say that at every single point thereafter, we were challenging the confidence that Crossrail was reporting around whether delivery was possible. The PRep report was part of that assurance and challenge process.
Q176 Lee Rowley: But you were in the same meeting that I was in, where Mr Wild told us that the Pudding Mill Lane explosion was completely immaterial to the problem. His exact words were, “The risk profile at that stage was such that Pudding Mill Lane wouldn’t have mattered anyway.” How can you tell me, as the overall person in charge here who didn’t actually get worried until something blew up—
Bernadette Kelly: We were always challenging—because it was the job of the sponsors to do that—the delivery, both in terms of cost and schedule, at every point in the delivery of this programme. What I have said is I think our level of concern spiked at that point, because it was a significant incident. Whether this could have been delivered without that incident is a very fair question, and I think that was what Mr Wild was referring to.
Q177 Lee Rowley: No, Mr Wild said it couldn’t be delivered.
Bernadette Kelly: You asked me about the level of anxiety and concern that we had, and I am merely reporting—
Q178 Lee Rowley: I asked you on what date you, in your view, felt that this wasn’t going to be achieved, which you haven’t really answered.
Bernadette Kelly: In the end, we were asking the company to tell us whether it could be delivered, and in the end, we received a formal notification from the company—as you know—in August. What I would say is for many months before that, we had sufficient concerns to be continually asking those questions, and to be seeking assurances and independent reports to give better assessment and enable us to ask those questions on an informed basis with the company. However, ultimately, we were asking the company to tell us, “Can you deliver this or can you not by December 2018?”
Q179 Lee Rowley: It is an incredibly passive approach by the Department for Transport to say, “I was asking the company to tell me that they may have a problem,” in the same way as the Mayor of London has been incredibly passive in his approach. Putting aside the challenges I have with some of the statements today, do you not think it is a complete derogation of duty on the part of the Department for Transport to not just challenge, but actually understand what is going on here? We cannot sit here, several months later, with the level of problem that we have seen and just go, “Well, I didn’t realise there was a problem until something exploded.” It is just irreconcilable.
Bernadette Kelly: Again, you are presenting a wholly distorted view here. We were monitoring and challenging the delivery of this programme throughout—by the way, we have talked particularly about the delivery date, but the level of scrutiny around cost escalation had been increasing from the point at which the first intervention point was breached, which was back in May 2015. At that point, remedial cost plans and so on were put in place. If you actually wanted me to track back every intervention from the sponsors, I would be able to give you a long list from May—
Q180 Lee Rowley: No, I just want a date, which you still haven’t given me.
Bernadette Kelly: Okay. If we track back to cost as well as schedule, I would say May 2015, which was the point at which the zero intervention point was breached, was a very significant moment and marked a stepping up of sponsor intervention at that point—of the concerns, the challenge, the assurance, and the plans that we were seeking from the company to demonstrate the steps it was taking to keep the project under control.
Q181 Lee Rowley: Sorry to push you, but you have just indicated to me that 2015 was when you first started being concerned that the December 2018 date would not be hit.
Bernadette Kelly: No, what I have just said to you was that December 2015 was particularly significant, because that is the point at which IP0—the first cost intervention point—was breached, and therefore at that point, I think there was a particular challenge to the company around the cost, and the management of the cost, of the project. I then said our concern about delivery of the schedule spiked—I’m not saying it didn’t exist before, but there was a significant moment of additional concern—in November 2017, which was the Pudding Mill Lane incident.
Q182 Lee Rowley: When, in your view, did you decide that this project was not likely to deliver in December 2018? What date was it?
Bernadette Kelly: There is not a single date. At every sponsor board and in every conversation we had with the board and the executive team, we were saying, “We need to know: can you deliver this?” We were asking whether they needed to consider options around phased and sub-optimal opening, and we were asking them to present us their considered plans for what was deliverable.
Q183 Lee Rowley: Do you think your failure to draw a conclusion is a derogation of your duty to ensure taxpayers’ money was spent correctly? If you’re telling me there wasn’t a date, that means you failed to draw that conclusion.
Bernadette Kelly: I think the duty of the company was to—
Q184 Lee Rowley: No, no, your organisation, not the company—
Bernadette Kelly: No, it is a serious point.
Q185 Lee Rowley: No, I will talk to Crossrail in a minute. I am asking you what your duty was and why you chose not to draw a conclusion about when this project was about to go off a cliff, and you are still not answering. I would like an answer.
Bernadette Kelly: So I will give you an answer. As sponsors, we had a responsibility for setting dates, the key milestones for this project. There was also a clear responsibility, which was set out in the letter I have sent to the Chair and the Committee, on the company to seek and to request a change to those target dates if it believed they could not be delivered. That is the duty the company had and the duty the sponsor had. We would have considered such a request and indeed, in the end, did so at the point when it was received from the company. That is how the thing was set up.
Q186 Lee Rowley: What is the point of your role as challenge—it is in figure 2, which presumably you accept—as the Department if you do not draw conclusions from that challenge?
Bernadette Kelly: We were drawing conclusions from that challenge.
Q187 Lee Rowley: But you decided not to draw one about the key part, which was whether this would be delivered on time or not. You decided to outsource that back to the company, which you have just told me you had significant concerns over for up to four years before the company failed to deliver the thing that you wanted them to deliver, which was your major project. You can’t have it both ways.
Bernadette Kelly: What we had, as joint sponsors, was a system of sponsor oversight. The delivery of the project was the responsibility of the company. What we sought was for the company to give us its view, which we had independently assured, about whether delivery, and what delivery, was possible. There was a very robust system in place, with monthly sponsor boards at which the progress and delivery of the project were considered in very considerable detail.
Q188 Lee Rowley: But you still didn’t draw a conclusion.
Bernadette Kelly: I would say we were drawing that conclusion before the company reached that conclusion itself, but we needed the company to tell us and to advise us what the opening date for this project was, because that was its job.
Q189 Lee Rowley: A final question, because I am not sure I got an answer on that: who is responsible for this failure? I am happy for anybody to answer.
Bernadette Kelly: Okay, I will answer this question. I have answered it on several occasions, so I will continue to give the answer. I believe that the principal cause of this failure was a failure within the company to understand, manage, mitigate and escalate risks to the delivery date.
Q190 Lee Rowley: Sir Terry, do you agree or disagree?
Sir Terry Morgan: I don’t agree. What you want is a name—I can tell what you want—but there is no name. This is a hugely complex project and it is inappropriate to think you can name somebody to be ultimately responsible for this. This was a project that was a great project, or will be a great project when it is completed, and what needs to be done now, in my opinion, is to let the guys get on and do the job.
Q191 Lee Rowley: If the project is too complex to provide a name as to where the failure is, why is the project not too complicated to provide bonuses to those at the top, when they haven't delivered?
Sir Terry Morgan: I don’t think you are comparing apples with apples in that regard. I said when I answered a question earlier—
Q192 Lee Rowley: Well, I am comparing a project that is £3 billion in a hole.
Sir Terry Morgan: I said to you that every year, we sat down and set objectives for the programme. For the years that those were achieved, the bonus was paid. Where the objectives weren’t achieved, they were not paid. Andrew did not achieve a bonus every year—I can assure you of that.
Q193 Lee Rowley: A project that is very complicated gives out the sweeties when the perception is that it works, but doesn’t take the accountability when it stops working. Do we think that is acceptable?
Sir Terry Morgan: I think it is acceptable to pay when you make a commitment against certain objectives that you have set for remuneration and they are achieved; you pay.
Q194 Lee Rowley: Therefore the objectives were wrong, because the end of the process is a failure.
Sir Terry Morgan: We consulted on those objectives, and I was always very clear that the feedback that we got, we took into account, in terms of setting those objectives. Bernadette is quite right to say, for example, that when we breached IP0, and that was what we advised the sponsors, we adjusted the incentives to make sure that there were still stretch targets to achieve the sorts of things we were trying to achieve, in terms of value for money.
Q195 Lee Rowley: And here we are, two years later. We have another two years to go before it opens; dates that have not been confirmed for at least a number of the stations; and £3 billion of taxpayers’ money added to it.
Sir Terry Morgan: I can’t comment on the extra money.
Lee Rowley: You don’t have to comment on it; it’s a fact.
Q196 Chair: Before I move on to Mr Meggs about the future—just to warn him that I am coming his way—I want to pick up on the issue about the contractual arrangements. The London Assembly called it “a culture that...encouraged a denial of responsibility”. Contractors were paid, partly under Mr Wolstenholme and partly under Sir Terry, compensation for events that could be, in the complex structure of the 38 companies, down to another contractor not delivering. We have discussed already some of the complexities of those small contractors. Andrew Wolstenholme, do you think that was a good way of working? It seems to reward contractors who are then picking up the pieces and also being paid more for problems that they are sometimes part of.
Andrew Wolstenholme: The NEC form of contracts has, as part of its process, the process of compensation events. This is to make sure that you deal with issues as and when they arise. If you go back 10 or 20 years, on fixed lump-sum contracts, all the issues used to be put right to the back, and you would probably end up in court. The role of compensation events is an absolutely standard form of mechanism. If you look at every project that uses the NEC form of contracts, there is absolutely nothing to be worried about in using compensation events. The question is: if you have an event happening at Moorgate that affects the interface issues at Whitechapel, should that risk be passed across? The answer I give again is that because this railway is 42 km long, underground, with 10 stations, the geography of how that is set up means it is impossible to ask one organisation—
Q197 Chair: That’s fine. So where was the controlling mind when there was a problem with the tunnelling slowing down, which meant there was a problem with putting in a signalling system or whatever? Where was the controlling mind to look across to see where contractors could be best deployed, or, if they couldn’t be, to ensure that they had a legitimate reason for triggering a compensation event?
Andrew Wolstenholme: This was the purpose of the specialist called Bechtel and the partners they brought. They sat at a programme-office level looking at the optimisation of how we made these decisions.
Q198 Chair: They got payments under the system, too, didn’t they?
Andrew Wolstenholme: No, Bechtel did not get payments.
Q199 Chair: Sir Terry just nodded.
Sir Terry Morgan: They were incentivised.
Andrew Wolstenholme: They were incentivised. Whether they were in a single layer or individuals, the organisation got incentivised to help us make the right decisions.
Q200 Chair: To help reduce the number of compensation events?
Andrew Wolstenholme: Yes.
Q201 Chair: So Bechtel got a reward if they kept compensation events down. Did you have some sort of metric to measure this?
Andrew Wolstenholme: The Bechtel incentivisation was based on the outcome of the programme and where the cost was. Their reward was zero at the end of it because that was the outcome. Bechtel were incentivised to make sure that we made the right decisions in the face of how do you optimise this programme across these interfaces. The better those decisions, the less opportunity there was for individual contractors to raise compensation events, so the money side of it is an outcome of the activities that happened on site.
Q202 Chair: So who was overseeing? Are you saying that Bechtel was the controlling mind making sure that all these different systems integrated well? Where there was a problem, a most optimal decision was made—about, for example, where you deployed a small contractor who could not do their work because something else had happened, but who could be redeployed. Was that Bechtel?
Andrew Wolstenholme: As Sir Terry said, Bechtel were deployed in the various different programme teams. We split the programme up into seven different zones. I think Bechtel led five of those zones and Crossrail individuals led the other two. It was the best person for the job. The controlling mind on that zone was empowered to make certain levels of decisions—
Q203 Chair: Okay, so how did those zones interconnect? As you have repeated many times, there were 10 stations, 42 miles of tunnelling, and so on.
Andrew Wolstenholme: Eventually, they came up to the programme director and the project executive.
Q204 Chair: So to your executive team?
Andrew Wolstenholme: The executive team, yes, have oversight over certain levels of decisions.
Q205 Chair: And what timeframe? How quickly would that happen?
Andrew Wolstenholme: Well, it would depend on what the decision was. There were many decisions. You have to encourage contractors to solve their own problems.
Q206 Chair: Absolutely, but these were the more difficult ones that would come up to you.
Andrew Wolstenholme: They would have to go through the commercial steering group. If they were signed off by that group, they would come up to the executive at a certain level, and at a certain level they went to the board.
Q207 Chair: So all that means quite a long potential delay, because I do not suppose that the board was meeting every day. These things all added to delays, which added to the compensation payments, presumably.
Andrew Wolstenholme: Yes, but those decisions were prepared in advance of the last responsible moment.
Q208 Chair: Good. That is the bit that I was driving at: that people foresaw them.
Andrew Wolstenholme: Absolutely.
Q209 Chair: You are confident that they saw these coming.
Andrew Wolstenholme: Yes.
Q210 Chair: Unfortunately, we do not have time to go into that much more.
I want to turn to Mark Wild for a moment to ask about the problems that you have inherited; we do not have time to go into more of the problems, but we had a visit, so we can add that to our evidence. Most or all of the contractors are still working on the project. They are getting more money for mistakes that they were part of. You are inheriting this, and presumably you have to work with what you have got, but do you think that that is a sensible system? There seems to be no loss for the contractor.
Mark Wild: It is a long project and a lot of these contractors have been on it for a long time. My focus is on getting them over the line in the next year to two years. I think other people would be better at saying whether the various supplemental agreements, relief of risk and claims were the right thing to do. As we look forward, though, we have a good supply chain. The people in the supply chain are very different and disparate.
Q211 Chair: How are you incentivising them to work differently?
Mark Wild: The first thing that we would say about the programme is that, for the first time—I have looked at this in some detail, actually—we are much more converged in what we think the out-turn cost and the dates will be. At the moment, the gap between the contractors’ view of the AFCDC and ours is about £20 million or £30 million. A year ago, it was £400 million. The gap between us—the coalition—in our acceptance of the dates is much tighter, so I think we are starting from a much more productive point of view.
We look at each contract individually, and we will incentivise contractors to reduce our risk as we go forward. We are actually in a much more straightforward position than the one that Terry and Andrew faced, because the work ahead of us is quite discrete. Although it is interconnected, it is less variable; the civil engineering is finished, the tunnelling is finished and we are into system integration now. We will treat Siemens, Bombardier, Costain and Laing O’Rourke differently from one another. We have a collection of seven or eight large companies and about 20 or 30 tier 2s and tier 3s. Every one of them will have a discrete package and a discrete approach that we will agree with the board.
We will incentivise contractors if it reduces our risks, but other contractors do not need incentivisation. We are meeting the global CEOs of Bombardier and Siemens tomorrow, interestingly, and they just want to collaborate to get the project finished.
Q212 Chair: So it is more that they are partnering with you now—is that how you would put it?
Mark Wild: Yes. Another change from the past is that the train and the signalling system were a significant problem, but—for the very first time, I think—we have a converged software path between Siemens and Bombardier. There is no real need to incentivise those people, because their incentive is to get their products into service, but a civil engineering company or a tier 2 may require incentivisation.
It will be a horses for courses approach agreed with the board, but the most important point is that we now have convergence of dates and money. Generally, our dates in our programmes now are slightly beyond where the dates of the contractors are. That is simply a product of getting to the end of the job. Andrew did not really have that luxury; quite rightly, they were driving early dates and we had quite a large gap. We are in a very different situation from where we were.
Q213 Chair: Mr Meggs, I have one longer question for you and a number of quick-fire ones. You have heard Sir Terry say a number of times that the deadline kept contractors on track. You used to chair the Infrastructure and Projects Authority and you have run big projects. Do you think that you have taken your foot off the pedal by taking that deadline away? We will get on to the delivery time in a moment, but do you agree with Sir Terry Morgan that that was a helpful driver in making sure that contractors got on with it?
Tony Meggs: I think it was a helpful driver, up to a point, until it became unhelpful. It became unhelpful in 2018, when it caused things to occur that probably should not have occurred.
I think there is a big risk that we are running right now. We have put forward a realistic schedule and worked in a number of different ways, in different scenarios, to understand all the risk. We have been completely open and transparent about that risk and the range that it produces. We are working very hard with our contractors to complete the major works that we need to get done in the stations and the tunnels by the end of this year. We have lots of methods for applying pressure on contractors—not just a final date. We have a much more intense system of daily management of work activities and other things like that.
Secondly, everybody wants to get this project finished right now. Our future depends on getting it done.
Q214 Chair: What is your date estimate, or firm date, for opening the central section?
Tony Meggs: We have taken the advice from the Department for Transport’s lessons learned and created a window.
Q215 Chair: Right, so can you tell us on the record what the window is?
Tony Meggs: Yes. The window is three months around the end of 2020. Effectively, it is from the beginning of October to the end of March 2020. The reason why we have a window—
Q216 Chair: 2021.
Tony Meggs: 2021; I apologise. I will say that again for the record. It is from the beginning of October 2020 to the end of March 2021.
Q217 Chair: And that is for the central section. What about the whole line? When will we see that?
Tony Meggs: We estimate that the remaining sections will take six months for stage 4—
Q218 Chair: Just remind us which end stage 4 is.
Tony Meggs: Stage 4 is to the east. Then another six months for stage 5.
Q219 Chair: So people coming from the west to Paddington will have problems for a year after the central section opens.
Tony Meggs: The full service will not be running. We have more work to do on the exact details of those stations.
Q220 Chair: Although we picked some of this up on our visit, for the record, not all stations in the central section will be open when it opens. We know about Bond Street. Are there any other stations that won’t be open in that window from October 2020 to March 2021?
Tony Meggs: Our current plan is for them all to be open by the end of—
Q221 Chair: Including Bond Street?
Tony Meggs: Sorry, except Bond Street.
Q222 Chair: Except for Bond Street, all the others will be open in the central section. Are there any other stations on other bits of the line that will not be ready when those lines open?
Mark Wild: Network Rail in the west have just placed a contract for Ealing and contracts like that. At the moment, they haven’t quite got their programme sorted out with their contractors.
Q223 Chair: So there is a risk there. That is Ealing and—
Mark Wild: Ealing and—oh God, now you are asking. Both Ealing stations. Terry and Andrew will know better than me.
Q224 Chair: Perhaps you can write to us to clarify. The stations around Ealing, basically—somewhere in west London. I am from east London, so I understand that you might not know the detail of west London.
Mark Wild: They will be open, but they won’t be step-free. All stations will be open.
Q225 Chair: That brings me to the compromises. What other compromises are there? You have Bond Street not opening, and you have a couple of stations without step-free access in west London. What other compromises are you prepared to make to get this open on time?
Mark Wild: First, all scope will be delivered in Crossrail to the end—to stage 5. We aren’t taking any scope out; everything will be delivered. What we have done is taken a lot of the digitisation—the density of the non-critical digital assets—out of the programme, and we will put them in the year after we open the central section.
Q226 Chair: What does that mean for a passenger?
Mark Wild: It will mean nothing for a passenger. It means something for the operators—London Underground and TfL. They may need more operational costs—quite minor, actually, but they will certainly need more human beings doing something that the digital assets will do slightly later. That period will take about a year. It is a very sensible decision, because it greatly reduces the digital complexity and the risk profile.
Q227 Chair: So that’s a cost to TfL. TfL will only start getting their money from this—Londoners will only start getting their money back—in October 2020 if we are lucky, or on the outside March 2021.
Mark Wild: The whole concept of our programme is to get TfL revenue-earning as fast as possible, which mostly by the way is stages 4 and 5, so we have to bring that date as far forward as we can, which crucially depends on stage 3. The operational costs that I refer to are still large sums of money, but they are nothing in comparison with the delay of stage 3. They are inoperational staff mostly—people doing activities that they—
Q228 Chair: I know it is a smaller cost, but Londoners are seeing threats to bus services; there is a big challenge for Transport for London in balancing its budget. I am a London MP, and we are all concerned about the impact of this delay on Crossrail’s budget. What are the biggest risks? Mr Meggs, what keeps you awake at night?
Tony Meggs: First, at the technical level, there is a huge amount of work to do on the software of the trains. We are acutely conscious that that is not always entirely predictable. That is point No. 1. Secondly, there is safety, which we always worry about. The changing conditions in the railway, moving from a construction project to a project with trains running up and down while work is still going on, raises safety issues that we are always working on. Thirdly, it is right to say that this project lost momentum—it is inevitable. It was like a big balloon that was blown up really tight and then it was punctured, and a lot of air went out of the system. Terry made that point, and it happens.
The key thing for us right now is to make sure that our organisation and our contractors are working together as one team. Having the right levels of drive and can-do spirit that the previous team had is not a trivial thing. It is not a trivial thing to take a project in which a lot of people have been very disappointed and have taken very personally, and to get all those folks back working at full tilt. That for me is one of the biggest leadership challenges of this project. That is probably the thing—if I am honest—that I am most worried about.
Q229 Chair: Thank you. Andrew Wolstenholme, you said at the beginning that you deeply regret where things are at. Sir Terry, what about you, as former chair of the board?
Sir Terry Morgan: I put 10 years of my life into Crossrail. I am so disappointed by not getting to the line that we said we would. It still is, and will be, a great programme, so when it opens it will still be something that we can all be proud of.
Chair: Certainly, having visited it, we would agree. There are some major elements. It will be hugely beneficial to London. It will, I think, take a tenth of London commuters, maybe of the London population. It will be very important. But we are, as you can tell, hugely disappointed about a major project that has been so problematic, and some of this was so predictable, even from my layperson’s position. Sir Geoffrey was on this Committee 20 years ago and I have been on it for eight years, and we look at these projects all the time—the integration of big systems is not a new problem. We will be quizzing Ms Kelly lots on the lessons learned, many times.
In the meantime, Londoners are without this system—not just Londoners, but Londoners in particular—and Transport for London has a hole in its budget as a result, so the impact on citizens and passengers is severe.
We will produce our report at some point towards the end of June, probably—maybe early in July—and we will send you a copy. The transcript of this evidence session will be on the website. It goes up uncorrected in the next couple of days, and we will alert you to that as well. I thank you very much indeed for your time.