Q11 Mr Bacon: I would like to explore more. I have two more questions. One is about the reset button, or the stop button. I have hankered for the centre to have a reset button for many years, or even a big red stop button. If Departments knew that the centre had that power at the outset when they were first thinking about a project and came to talk to the centre about it, do you think it would alter their behaviour at the beginning?
Dr Van Grondelle: I can only give a very generic answer. Project by project it is different, and within the project it is different person by person. Therefore I go for a very generic answer. I see very often that projects start and fail for all kinds of reasons, which are either beyond the Department's control or because individuals within do not feel empowered to flag the shortcomings, for whatever reason. It can be very personal reasons, institutional reasons or a competency reason. I have yet to come across, in 20 years, projects where the project team was particularly delighted that it did not go well. There is always a cost: an emotional cost, a cost to a career and a cost to the overall success and standing of the project. I do not think it would result in flippancy in approving projects.
Generally speaking, front-end loading of projects is a good idea. Spend time thinking about it. Get the contract strategy absolutely right. Challenge it, wait for a few weeks and then think about it again and get some fresh eyes to challenge you again. The more you do of that up-front, the more likely it is that your project remains under control. It is not a guarantee, but the more likely it is.
Q12 Mr Bacon: One more question. I want to ask you about the process of project review and RAG ratings-not about specific projects, but generically. We have been pursuing this issue for many years and we have seen RAG ratings occasionally published for particular projects like, for example, the Rural Payments Agency some years ago. The answer that one hears frequently is that the only trouble with public discussion of such RAG ratings is that it means people will not be honest in their reviews and you do not really get what you need to get. By the way, there was an interesting radio programme with Alan Mulally of Ford, which I sent to Bob Kerslake, the head of the civil service, the other day. Alan Mulally had a worldwide conference of managers meeting every Thursday morning at 7 o'clock to have a phone-in. They had RAG ratings for everything around the whole planet in their business. Everything, on every single index, was green. He said, "Guys, this is really interesting. Fantastic visibility. I've just got one question: if everything is green, why are we about to post a loss of $17 billion? Is there anything at all that is not going well?" It took a while before somebody put their hand up and said, "I've got a particular problem here." He said, "We sorted it out in 12 seconds, because everyone was in on the call." He eventually got everybody on board and then, a few weeks later, with the process being pushed and continued, he got what he called a rainbow of honesty, when there was real clarity. Then, and for the first time, they were able to manage the business. My concern about the answer-I fully understand about keeping things hugger-mugger, quiet and closed, otherwise you don't get the honesty-is that we have tried that, and it has not worked for years and years. We still get these car crashes, one after another for many, many years, under Governments of both parties. It seems to me that projects are like anaerobic bacteria; if they are weak, they can only survive in dark corners where no light is shined upon them. Isn't it the case that strong projects are improved by exposure? First of all, they are able to defend themselves and if they were not as fit and strong as they should be in defending themselves, the fact that they know that they have to makes them stronger and tougher, with bigger muscles. They end up doing a better job precisely because they are in the public gaze-in front of Parliament, journalists and everyone else-and they are able to defend themselves.
Chair: There is no prejudice in that question, but go on, give us a non-prejudiced answer.
Dr Van Grondelle: There is no hard and fast answer. I do know one thing: if you do assurance on projects, then 50% of it is science and 50% is art. Fifty per cent is a scientific, systematic review approach, and 50%) is, as an assurance professional, that you know when you are being told a fib.
Q13 Mr Bacon: The art is about the people.
Dr Van Grondelle: It is about the people, the skill and not necessarily trusting that beautiful score card with all the green lights that you have been given. It is about realising very quickly whether you are in an open and honest debate with the project team, or whether the project team is spinning you a yarn. It may not necessarily be negative. Sometimes project teams have the feeling that they cannot do anything else; they did it out of loyalty to the project or the team, or because they are not seeing a way out. They do not know where to reach for help to reset an underperforming project. It is an extremely painful position to be in as a project manager if you are not bringing home the goods. If you can combine assurance with a potential intervention option, you give them a way out. Then it comes down, ultimately, to systematically applying the assurance, which goes without saying is necessary, but also-
Q14 Chair: You have not actually answered the transparency issue. Is it better to be open, or does that mean that people hide the truth?
Mr Bacon: Isn't it true that sunshine is the best disinfectant?
Dr Van Grondelle: I have seen the exact opposite of that sometimes. It depends, again, on the individual project. I have come from two projects recently. One was held in full openness and was a complete disaster as a result, because there was constant intervention- I add that this was a private sector project and not a public sector or Government project-but I have seen in the same company a project, run by a different team, that had never been visited by the board, the CEO or MDs, and it was perfectly run, because they were allowed to get on with it quietly. The answer is yes, transparency by definition is better, because it should result in the full public gaze, full scrutiny and therefore full interaction, but transparency should never descend into interference and intervention because that will destabilise the project. Very often, that effect does happen-lots of scrutiny, lots of visits, lots of governance committees and lots of interventions, and as a result the project goes off the rails.
Q15 Meg Hillier: I am interested in your views about embedding insurance teams in projects, because there is a danger that they can go native, but I guess there are also some benefits. I wonder what your views are on the different approaches. Do we have an embedded team, ad hoc drop-ins or deep diving into projects?
Dr Van Grondelle: I am a great proponent of embedded assurance teams. I have done it for a very long time in my career. You do need to protect them because, first, they can go native and, secondly, they can also be seen to have gone native without having gone native. The individuals need to be protected and nourished, and they must have a very strong reporting line back to their parent organisation-the MPA in this case. If you have the resourcing to do it, the presence of an assurance team gives you much more intrusive assurance-it give you that ability to hear the little rumour, to listen to the individual employee who is disgruntled and saying, "Look, you should really look at this bit of the schedule or that bit of cost." I find it, personally, a very powerful way of doing assurance. There is a secondary effect, that it has a deterrent effect. The fact that the assurance people are permanently present means that, first, they tend to be more accessible-I am generalising terribly-and, secondly, assurance is continuous. I am always a great proponent of continuous and embedded assurance provided that it can be resourced.
Q16 Meg Hillier: An alternative way of doing it perhaps is to work on the culture of an organisation. I say this because we had the opportunity last week to meet the man from NASA who has been doing a lot of work because of some of the terrible, tragic problems that they had. In one case, because they had tried to reduce resources, a junior engineer had seen some data but not appreciated that the data required quick action and did not have the authority to take it. That has encouraged everyone at very junior level to raise their hands and say they are worried about it. Is that an alternative to embedding, or is that something that should happen alongside it? What do you feel about how the public sector works in that respect in terms of the culture?
At the moment, if you are a civil servant, going to your Minister or somewhere up the line with a failure is not going to be good for your career, whereas, if you tell everyone that it is all right and it is not, you do not get much comeback either. The incentive is perhaps not to be honest about a problem. That is a bit of a generalisation but you know what I mean.
Dr Van Grondelle: You raise a number of points. I think the culture is extremely important. I see that the shorter projects are the less time is taken to embed a culture. Priority tends to be given to safety culture, preventing accidents and injuries and that kind of thing. That is already a very substantial challenge. A good project manager or director manages culture continuously. Depending on the pressures that he or she is under it does not always mature to the point that the junior engineer feels comfortable telling the No. 1 man or woman the bad news. As a general observation, does bad news travel more quickly in the private sector? I don't necessarily think so. Sometimes it does, depending on the personalities and the culture that has been created; sometimes it doesn't. If you look at NASA, I refer to the Challenger and Columbia disasters for example, and the Apollo 1 disaster, going back even further. Time and again it resulted in soul-searching; time and again it resulted in managing the culture of the organisation as a whole, and the culture of the project as a subset of that. I find, being head of joint ventures, that even some of the largest multinationals struggle to manage the culture of a project or a joint venture because it becomes arm's length and is also a temporary phenomenon. I keep referring to it as the Portakabin syndrome. Why would you paint the Portakabin and keep it clean and tidy, because we are going to take it away in two years' time anyway? That sounds one more point that the major projects authority, in my view, plays a role in setting the culture of projects in government, by being there.
Q17 Matthew Hancock: I want to ask one question. You talked earlier about the intervention. Could you describe what a big, heavy, serious intervention in a failing major project looks like?
Dr Van Grondelle: In very generic terms, I would almost split it in two phases. One question is: can we effectively intervene in this? At that point you are almost in a medical analogy: the doctor comes in, the GP says that things are not going terribly well and we should have a chat with a specialist. Is there an illness, yes or no? Can we do something, yes or no? At that point they take a second decision to scope out what that intervention looks like. You either do it or you elect that it is unsalvageable or not worth the effort. That would be roughly the decision process around intervening in these projects. Is that always successful? No. Sometimes you settle for an 80:20 solution because that is the best you think collectively, either in the public or private sector, that you can achieve.
That may still be better than the alternative. The alternative, which you always bear in mind, is just to abandon it, walk away and take the cost, take the hit on the bottom line, which may not be the best outcome financially from a public scrutiny perspective. Each and every one of them you assess whether it can be salvaged; what it would take to salvage it and then what you need to do. I was in Qatar last week doing exactly that. That was fortunately a project where we came to the conclusion that it could be salvaged. You then come up with a very detailed plan collectively, usually involving the project team, the assurance bodies and whoever is helping or authorised to reset. That can go down to changing personnel and the contracting strategy. Sometimes it means taking hits, whereby you exit one contractor and go for another. You pay a penalty to do that but it still might be a better overall outcome. That is where you go into the detailed design of what the intervention looks like.
Q18 Chair: I take from this-this is your opportunity to add anything you want-that the authority to reset is important; resources are important; and leading a culture within the organisation is important. Those are the three messages I have taken from your contribution. Is there anything you want to add?
Dr Van Grondelle: The continuous assurance discretion is important. There are two angles to that, and one I have not mentioned yet. There is continuous assurance of projects by embedding people. There are also now technologies available, in which I am not a specialist, such as data analytics that allow you continuously to check elements of a project and detect, for example, fraud, incorrect payments and underpayments. Professional colleagues of mine were recently deployed at the telecoms sector and they were absolutely amazed to figure out that even with very large companies deploying these techniques, which are not visible-they are not a threat or even a distraction to the project team; they take place almost behind the scenes-very substantial incorrect payments could be recovered. There was a fair amount of that going on in just about any project, whether it was public sector or private sector-incorrect payments, early payments or payments that are not due. If you take a data analytics look, without interrupting the project you can suddenly claw back substantial sums of money without having disrupted the project in any way. That is the second element of continuous assurance-embedding people. If you have them, fantastic; manage them well and keep them safe. The other thing is using certain analytical technologies to keep an eye on the projects as they move forward. That is good bang for the buck.
Chair: Thank you very much indeed. That was very helpful. Apologies again for keeping you waiting at the start.
Dr Van Grondelle: It is a real pleasure to be here.
____________
Examination of Witnesses
Witnesses: David Pitchford, Executive Director, Major Projects Authority, Cabinet Office, and Sharon White, Director General, Public Spending, HM Treasury, gave evidence.
Q19 Chair: Welcome back. We are now taking evidence from you in public. Let us just start by putting on the record that, as you know, David, from your session with us about nine months or a year ago, we are very supportive of the concept of a Major Projects Authority having teeth at the centre of Government and working closely with the funder, which is the Treasury. We are very supportive of the principle, and we note from the Report that good progress has been made.
Then come the "howevers", and I am afraid I have to start with some of them. It is difficult to know where to start. In paragraph 1.5 on page 14 of the Report, the NAO define an integrated assurance system, and presumably you would agree. But we have not really got that so far, have we? I accept that it is a "so far," and I really want you both to talk a little bit about what needs to be done actually to get that integrated assurance system embedded in the whole of Government.
David Pitchford: That is a fair analysis, Madam Chair. The numbers come out at roughly 62% of Departments have got an adequate IAAP, some of them much better than others. For example, MOJ have not only got it for their major portfolio projects but they have embedded this right across the Department, so all their projects are being run against an IAAP, which is the best side of town. We have had some terrific successful results such as the DWP IAAP for Universal Credit. As I mentioned to you in private session, that is a very complex project so the plan is very complex, but they have gone at it in a very strong way.
In other Departments it is not as well regarded. My own belief is that it is not so much inertia because they do not want to do it, but there is not a broad understanding of just what a powerful management tool an integrated assurance and approvals plan can be. Basically, that is because not only does it set up the fundamental approval points and the assurance activity that must go on before it, but if the achievement milestones for the project-such as calling in tenders, letting projects, and refunding stages-are folded into the same thing and then aligned with the Department's own internal audit processes, the integration becomes amazingly productive. If you can then fold it into what we hope will be an exponential improvement in the standard of management information that informs all that, we can get a much better outcome.
I have to say that we aren't anywhere near concluding that process. When we first started out, it was new. There have been some difficulties in trying to get enough time to spend on it, because of the prioritisation that I mentioned before, but we are more aligned with the Treasury in the approvals and assurance process than may be apparent, and certainly more than we were two years ago. We are getting better and better at this. The alignment of the MPA clusters with the spending teams within the Treasury is much better, and now we have them mirroring each other, not only in terms of title, but how they approach Departments.
So, I do take the challenge. We haven't got there yet, and one really strong thing that we will be driving towards in the second year of our operations is to elevate the thinking about this to introduce the concept as a management tool, and not just something that you need to do to get approval from the Treasury.
Q20 Stephen Barclay: You mentioned the MOJ as being the best for this, so which is the worst?
Chair: I was going to ask whether you are able to name those that have not got it.
David Pitchford: There are several Departments that have not embraced it as yet. I would rather not name them at this point if I can, because it is important that we maintain relationships with them to try and bring them to it, rather than label them as recalcitrant.