Public Accounts Committee
Oral evidence: Delivering major projects in government, HC 710
Wednesday 21 January 2016
Ordered by the House of Commons to be published on 21 January 2016
Watch the meeting: http://parliamentlive.tv/Event/Index/a71ca73a-6fef-44bc-a7a3-eb3efad2a96c
Members present: Meg Hillier (Chair), Mr Richard Bacon, Deidre Brock, Chris Evans, Caroline Flint, Kevin Foster, Mr Stewart Jackson, Nigel Mills, David Mowat, Stephen Phillips, Karin Smyth, Mrs Anne-Marie Trevelyan
Sir Amyas Morse, Comptroller and Auditor General, Adrian Jenner, Director of Parliamentary Relations, and Geraldine Barker, Director, National Audit Office, and Richard Brown, Treasury Officer of Accounts, were in attendance.
Witnesses: John Manzoni, Chief Executive of the Civil Service and Permanent Secretary, Cabinet Office, and Tony Meggs, Chief Executive, Infrastructure and Projects Authority, former Chief Executive Officer, Major Projects Authority, gave evidence.
Q1 Chair: Welcome to the Public Accounts Committee on Wednesday 20 January 2016. We are here to discuss Delivering Major Projects in Government, an National Audit Office Report, which focus on the early years of the Major Projects Authority. As Mr Manzoni certainly, and hopefully by now Mr Meggs, will know, the Committee has always been a big supporter of the Major Projects Authority and now the Infrastructure and Projects Authority. We are keenly aware of the very important task you have to improve project delivery within Government.
We start on the same page in terms of our aims. We are also very aware, of course, of the difficulty of that task. We spend a lot of time in this room and Committee Room 15 seeing the legacy you have inherited of projects that have been poorly planned and implemented, without regard to the good practice that does exist, not to mention the problems caused by projects being led by people without the right skills and experience, and the high turnover of senior responsible officers. I am saying things that we all agree on.
We are impatient to see changes in performance. It is important for taxpayers and the customers or clients—the public—who receive these services and benefit from infrastructure change that projects are delivered properly, but we cannot yet see enough signs of improvement to make us go skipping down the road thinking it is all sorted. We know that data is an issue and we want to touch on that. We want to quiz you today about the building blocks you have in place to make sure the best lessons are learned and that the worst projects are history, and we no longer see such problems routinely in Government. We acknowledge what you have done so far. We will challenge you on how much more you will be doing and we want some comfort that you will really get to grips with this and that we will actually see a change, as will the taxpayer.
Our witnesses today are Tony Meggs, chief executive of the Infrastructure Projects Authority, as it is now called, and John Manzoni, former chief executive of the Major Projects Authority and now chief executive of the Civil Service and permanent secretary at the Cabinet Office.
Mr Meggs, you have asked to give a brief opening statement. We don’t usually do this in this Committee because we are always keen to get stuck into questions, but as you are new in your post and it is such an important role, we are happy to give you a couple of minutes to make your opening statement. Perhaps you will do that briefly.
Tony Meggs: First, I should say that we very much welcome the recent Report from the National Audit Office. As an independent assurance agency ourselves, we obviously recognise the value of independent scrutiny of our work and indeed that of many thousands of people across Government who are involved in delivering projects.
Secondly, I should say that we are broadly in agreement with the major recommendations of the Report: to engage with projects earlier in their life cycles, to develop an effective prioritisation mechanism in Government, and to improve the quality of performance measurement. We are hard at work in all those areas.
Finally, I would like to comment on the statement in the Report that the Government does not have a good track record of project delivery. I certainly agree there is room for improvement—indeed, that is one of the main reasons for our very existence. We are working very hard across Government to improve the track record. We certainly don’t want to see a single pound of taxpayers’ money wasted, but by way of contrast, I should say that the projects we track are, by their definition, very complex in nature, and therefore almost inevitably will experience difficulties during the course of their lives. Coming from the private sector, I can testify to the fact that this is true outside Government as well as inside, and there is solid evidence to support that statement. This is not to justify or excuse compromised delivery, but to try to place it in context and to say that we will never be satisfied with our performance. We are continually striving to learn from everything we do, to take the initiative and to improve the Government’s capacity to execute projects as flawlessly as possible and as flawlessly as real life allows. I hope that was less than two minutes.
Chair: Thank you very much, Mr Meggs. We recognise the complexity of these matters.
We have three Members doing the main questioning today and I will hand over to Richard Bacon who will kick off with a slide show, so perhaps you would look at the screen behind my head. Do the witnesses have a handout?
Q2 Mr Bacon: Yes, they do.
Mr Meggs, the last thing you said about continuous improvement sort of relates to this, which is really about transparency. In the foreword of your most recent annual report, the Chief Secretary and the Minister for the Cabinet Office penned an introduction which states that the MPA’s traffic light delivery confidence assessments show real improvements and that while there is much more to do of course, “rather than burying our heads in the sand, we believe it’s far better to be open about the process. That way problems can be flagged and corrective action taken.” The questions I want to ask you relate to the point about being open about the process—and how open—so that corrective action can be taken.
These slides will take only about a minute, or a maximum of two minutes. The first is from an NAO Report 10 years ago, Progress in improving government efficiency, and shows two pie charts, the first December 2004 and the second December 2005, in which the red sliver grew from 3% of the total to 4%. John Oughton, who was then chief executive of the Office of Government Commerce, the MPA’s predecessor in many ways, was the witness. I asked him to explain which Departments of State were represented by the 4%, and particularly by the increase of one third from 3% to 4%. His reply was, “I can’t tell you that. It’s advice to the Prime Minister,” which obviously I found not particularly helpful.
On page 13 of the current Report you see figure 2, which lists all the projects. For example, the Department for Work and Pensions has six projects with a whole-life value of £22 billion, but from this slide you cannot see what those six projects are, or which of them are amber, green or red. If you look on page 27 there is a high-level view across the piece of 106 of the 149 projects that will deliver in the next five years and of which are red, amber and green, but again not at an individual project level. The MPA’s own report shows, on page 23, for a three-year period the way in which the red-amber-green moves from year to year for different Departments. So it is quite helpful, but again it does not show it at an individual project level.
This next slide here, which is from the National Audit Office Report on the delays in administering the single farm payment—this was published 10 years ago—does show, for very specific items such as resources, scope or the business case, how the red-amber-green status moves month by month. For example, if you look at the line in May 2005 you can see that the business case was at green, but in June it went to amber, staying there until August, when it went to red—and so on and so forth. The narrative that goes with it, in the same Report from 10 years ago, explains in more detail specifically about that project what was going on.
This slide here is from the Office of Management and Budget in the United States and it shows the entire federal Government. Down the left you can see Agriculture, Commerce, Defence, Education, et cetera, so it is high level, but it does have the specifics of what the red, amber and green are about at a high level—so competitive sourcing, financial performance, e-government and so on.
My question is: can this Government create an environment in which it is safe for people to be open and honest about problems with projects?
Tony Meggs: Yes—
Mr Bacon: And publish it.
Tony Meggs: I think we have created an environment in which it is safe for projects to reveal how they think they are doing. The review process, the assurance process, that we undertake is designed to create an environment in which people will be frank, open and honest about performance.
We do not publish those reports, because we believe that if we did so that would severely impinge on our ability to get full, frank and open conversation.
Q3 Mr Bacon: May I stop you at that point? That is precisely the point that I want to discuss. I have been listening to that answer for 15 years. I am not saying it is definitely wrong, but if it were definitely correct that not generally publishing those reports—the slide here is the DEFRA example I showed you from a National Audit Office Report 10 years ago, which was something of an exception because it came into the public domain—and therefore not impinging on people’s ability to be open and honest, had encouraged the kind of honesty that we want, then we would have had much greater success in the last 10 to 15 years than we have had. The reality, as we know, is that we have had a series of projects—the single farm payment and the Rural Payments Agency, the Child Support Agency, individual learning accounts, tax credits, the national programme for IT in the NHS, student loans, the FE colleges building programme and so on—where that has not been the case, despite the non-publication, which should have enabled the open and frank discussion that you say you want to see.
I shared with you yesterday by email the example from Ford. The chief executive, Alan Mulally, had an internal traffic light programme, and plainly, despite being internal to Ford—I am sure it was not published in The Wall Street Journal—even getting honesty inside the corporation globally was quite difficult, but when they did it was transformational. So forgetting for a moment the question whether the information is made completely public and put in the public domain, so Parliament can see it and the newspapers can see it—or not, but we will come on to that in a second—how do you ensure that you get real honesty, even internally? Because if you had had that honesty for the last 10 to 15 years, based on your statement that you do not go open because you would impinge on it, we would have had better success. How do you get that honesty internally?
Tony Meggs: Let me say a couple of things. First, the level of transparency that I see in Government as a newcomer is really quite high. Secondly, we publish ratings annually. We don’t publish them more frequently than that, but we do publish annual ratings, so there is a level of transparency around that. Thirdly, creating the environment of openness and transparency is part of the art of what the IPA does. The way in which reviews are conducted, the skill of the reviewers and the environment that they create, is what helps to ensure open and honest reporting. Honestly, I would say that the spectrum—or as Mr Mulally said, the rainbow of colours—that we have here is an indication—
Mr Bacon: “The rainbow of honesty” is what he said.
Tony Meggs: The rainbow of honesty—I thought it was a fine phrase. It is an indication that we are really not trying to hide anything. The one thing we hold sacred—if I can say that—is that the reports themselves constitute an open and honest conversation, and the publication of the reports themselves would interfere with that.
I have a specific example. We recently did some work in Northern Ireland—pro bono, I might say—with our hard-pressed staff. We did an extensive review for them on a series of projects in Northern Ireland and, unusually, we agreed in advance with them that the report would be published. The consequence was that it was not a good report. There was a massive amount of negotiation and a process of what I think is described in Government as Maxwellisation to make sure everybody was happy with what everybody had said. The report became a political football and did not achieve its objectives. The issue of the publication of the reports per se is one that we hold very dear. With respect to the traffic lights—the RAG ratings—we do publish them on an annual basis. I don’t think we are trying to hide anything.
Mr Bacon: Mr Manzoni, do you want to add anything?
John Manzoni: I agree with Tony on this. He cited an example where the publication of the report resulted in a less useful intervention in a particular project.
There is internal transparency, which is greater than the external transparency. There are quarterly returns internally. As you know, the amount of that which is released into the public domain has been carefully negotiated over a long period and is set in a Cabinet policy called the transparency policy. That has been a political agreement, and you know and we know that we are not at liberty to cross that boundary. Internally, I don’t think there are any secrets at all, actually. The quarterly data allows a good conversation to take place, but it takes more than just transparency to get good projects, as you know. That is perhaps why, if we had all that transparency—we haven’t had it for 10 years, but even if we had— there would need to be a number of other things in place, such as capabilities, clear accountability and a series of other things.
Chair: Which we are going to come on to in a minute.
Q4 Mr Bacon: I have one more question now, and then I will hand over to my colleagues.
Mr Meggs, you described the IPA as an independent assurance agency. I don’t think anyone would have said that English Partnerships, Partnerships UK or Infrastructure UK were independent assurance agencies. They were bodies within Government to make things happen. No one would suggest, to take a crude example, merging the Department for Transport, which is responsible for delivering big infrastructure projects—new motorways, or whatever it is—with the value for money team at the National Audit Office that studies transport, and expect it to cause anything other than problems. Why isn’t that analogy fitting for the merger that has now taken place between Infrastructure UK and the MPA? We thought of the MPA an independent assurance agency, but it now looks like it is at risk of becoming a cheerleader for projects that the other part of Government is trying to push forward and make happen.
Tony Meggs: I honestly do not believe that the merger of these two entities will create an issue. In support of that, it is helpful to understand the way that our reviews are organised. While the historical MPA, now the IPA, is the organiser of reviews across Government for major projects, they are independent in the way that they are constructed. We use outside parties to conduct the reviews—people with no vested interest whatsoever in obtaining any particular outcome. While the IPA runs those reviews and sets the terms of reference and so on, we use independent people from across Government, from many different Departments, and from outside Government. We have about 60 or so first-class independent assurers. So the process that underlies those is unimpeachable, actually. It is entirely independent.
Q5 Mr Bacon: What are the benefits of the merger of the two?
Tony Meggs: I think the benefits are that we can increase the level of support and intervention that we provide to projects. One of the things that we do in our work is that we do not just provide an independent assessment; most importantly, we also create a series of recommendations in every report. We will provide whatever support we can to Departments and individual projects in attempting to address those recommendations, and we will follow up on them. I think the level of expertise that we now have in the combined IPA—in the combined entities—will be improved.
The second thing that I would say is that, from the historical perspective, the amount of overlap between the projects that IUK has historically supported and those projects that are on the GMPP is not huge. It is about a dozen projects.
John Manzoni: If anything—these things are always a blend—in many ways, this will actually give the MPA more teeth in its assurance role. Right now, we use the Treasury finance gates for approvals for funding for the next phase of the project, so now it is the Treasury and the Cabinet Office saying, “By the way, the following conditions have to pertain before you go through this gate,” and there isn’t any space between the two in this new entity. I think that is actually going to strengthen that aspect of the MPA, not weaken it.
Q6 Mr Bacon: Is the merger saving money?
Tony Meggs: It was not designed as an attempt to save money. We are combining the groups. Both teams have a set of plans in place, and we are obviously looking at them again. We want to use the merger as an opportunity to strengthen a number of areas. I will give you an example. IPA has a very good product it has developed over time called the route map. The route map is a very good guide to how to set up projects, which is one of the areas that we struggle with. It is used in Government, but it is not widely used. We intend to use this as an opportunity to take that, by way of example, and apply it much more widely to projects across Government. We have not identified them all, but there are many ways in which we can use this merger to enhance the overall performance.
Q7 Mr Bacon: I will finish here, but the comment that Mr Manzoni made about the Treasury having a say is potentially extraordinarily important. That design is perhaps something we should always have had. Will you be able to evidence this and show us in the next six, 12 or 24 months or more how that process—having the Treasury teeth as well as the Cabinet Office teeth biting earlier in the process—will produce better outcomes? Can you think about how you might evidence that to us going forward?
Tony Meggs: We will certainly take that as a challenge.
Q8 Caroline Flint: Paragraph 11 of the NAO Report states that “35% of projects due to deliver in the next 5 years are rated as red or amber-red” and paragraph 12 states: “This is worrying given the number of projects to be delivered within this Parliament, and the large proportion of these which are transformation projects.” Mr Meggs, what trends or reoccurring themes have you identified in the 35% of projects that you rate as red or amber-red?
Tony Meggs: Some of the recurring themes include, first, a failure to set projects up correctly in the first place, which is an area that we are paying an enormous amount of attention to—I just gave an example of how we hope to enhance that. Secondly, another typical theme is resourcing. It has to be said that I have never worked in an organisation that had sufficient resources. It is a fact of life, but the problem can be acute in some areas. Thirdly—the Report correctly identifies this—the timescales of some projects are unrealistic. As we go through the planning process, which we have going on right now with single departmental plans, we are going to be looking very closely at the timescales for some large transformation projects to try to ensure that they are as realistic as possible.
Q9 Caroline Flint: That was very clear and helpful in that you identify three things: the initial planning and setting up of projects; the resourcing to help take the plans forward; and timescales. I want to come on to a few questions about the early planning of projects, but while we’re on the issue of projects that have been traffic-lighted red—if that’s the right phrase—I understand that part of the definition of what constitutes red is whether the project is actually deliverable. Therefore, I suppose that my link to that is to ask why it is that when a project achieves—if that’s the right word—a red, there isn’t a pause to review whether it is feasible to go ahead with the project or whether it should be cancelled. We could have that rather than—I am sure this happens in other countries—people thinking, “Let’s just put some more money into it to make it work.” That also links to the original cost estimates and the benefits that were sought as part of the project. What should be initiated when a project receives a red?
Tony Meggs: I think that for the most part, a red project needs to be—I suppose the term is reset. We certainly need to make a major intervention, which often means re-scoping and reassessing the whole project. It’s rare and unpleasant, but we have had to do that from time to time.
Q10 Caroline Flint: Can you give an example?
Tony Meggs: There is everybody’s favourite example of universal credit, which was a programme—I say this with some trepidation—that was reset very successfully. It got off to a false start and caused a lot of angst—before my time—and it was then successfully re-scoped and reset and is now on a very viable pathway. The Student Loans Company is one where we have had some difficulties and it is now in the process of redesigning its future plans. It does happen.
Q11 Caroline Flint: But, I think as you’ve indicated, it wouldn’t necessarily happen to the extent that it had to be reviewed and reshaped if there had been better early planning. In our recent sessions on universal credit, on e-borders and I think on the common agricultural policy as well, what the NAO and ourselves have identified is a certain amount of what I think is called optimism bias in the development of such projects. Alongside that is the business case structure that exists, presumably, to try to ensure whether it is the first stage, which is the strategic outline case. Then we go on to stage 2, which is the outline business case, and then stage 3, which is the full business case. Is that structure of looking at these projects good enough? Do you think it is good enough or do we need something before we even get into that that involves some early pre-planning, before any green light to go ahead should happen?
Tony Meggs: Our objective is to have a review of projects when they are in the early stages of conception—that is, as the policy is developed. Our ultimate objective—I have to be realistic about this—is to ensure that no policy that will lead to a significant project gets announced without some level of assurance being conducted to ensure that the project is deliverable. First, I should say that this has been a long-standing objective. We have starting gate reviews. We have—I have forgotten the term now.
John Manzoni: Project validation reviews.
Tony Meggs: Thank you. We have project validation reviews and we have just introduced something more nimble and agile called a critical friend review, which is designed to be in very early and to provide advice. As I said earlier, we will make much greater use of the Routemap.
Q12 Caroline Flint: You have already said of the Routemap—I do not want to misquote you—that that is almost an optional take-up. That is what you seemed to indicate in your earlier answer.
Tony Meggs: Yes.
Q13 Caroline Flint: And that, presumably, is about early planning and the lessons that every project should apply itself to. But that is optional, isn’t it?
Tony Meggs: It is optional. This is really an important point about how the IPA operates and how it seeks to operate. We set standards for Government for how major projects should be executed. The assurance process is, in part, measuring performance against those standards. The more things we make mandatory, the less successful we will be. That is based not only on my brief experience here, but on experience elsewhere. We have to demonstrate to Departments the value of that and create demand. Creating a demand is a much more powerful way.
Q14 Mr Bacon: On that point, I was interested to hear you say that making it mandatory is not the route. Plainly, you would not need to make it mandatory if everybody understood the importance of it. That must be a better way to do it, culturally. It is also true that however many senior officials in different Departments, police officers, generals and officials you have running different projects—as you know, we are very impressed by the MPLA process at Oxford—you still have this factor, which we could call the grit in the oyster or the bacterium coming into an otherwise clean laboratory. Members of Parliament, mostly, come in, become Ministers and mess it all up with their clever ideas that it can all be done much faster without understanding the importance of project validation and of a critical friend, and not understanding the impact of short-term political decisions on the management of a large organisation.
Now, you can do something about that because you know that is the case, particularly given that Members of Parliament are uniquely unlikely to have the experience required of running large organisations because they spent the best years of their lives—when they would have been acquiring that experience—shoving leaflets through doors. They are uniquely unlikely to have that. As long as you explicitly acknowledge that, you could do something about it. Since you know that is where the vast majority of Ministers will come from and they will, with very few exceptions, be in that category to some degree or other, why don’t you routinely expose all Members of Parliament to Routemap—to public validation?
Tony Meggs: That is a great idea, if I may say so.
Q15 Mr Bacon: Can we quote you in our Report?
Tony Meggs: You can—well, I think you can. I might get fired, but I think it is a great idea.
We are very conscious of the fact that successful projects create an environment in which to be successful. One of the most important things is for the most senior decision makers to understand the difficulties and the key criteria for making successful projects. We have started a process of ministerial workshops. We did one at Oxford recently with Ministers of State and other Ministers there; we had permanent secretaries—
Q16 Mr Bacon: You could catch them earlier, couldn’t you?
Tony Meggs: It would be great to catch them earlier, but—
Q17 Caroline Flint: May I just get back to the pre-planning? Sometimes Members of Parliament and Ministers inject a bit of what it is like in life outside Whitehall, or what it is like for constituents on the receiving end of the projects, which is a valid thing. As a Minister, I used to find that sometimes I was the one asking why the emperor was not wearing any clothes, while everyone thought it was all wonderful—but there we go; I am just speaking up for some of my colleagues on that front.
Just on this pre-planning, it would appear—something you have acknowledged yourself today, Mr Meggs—that the quality of planning and initial savings are not as good as they should be. As a result, unfortunately, the full extent of that poor planning might only come to light down the road, after many years in some cases. There are a number of projects—universal credit is a good example, or e-borders, but I do not just want to pick on those—that might outlive not only a Government, but the personnel working on the project. I just wondered whether you thought that the present way in which projects are reviewed needs to be packaged in a different way, not only in a more realistic real-time framework, but also building out in phases, so that it would enable the IPA, Ministers, MPs and others working on a project to have a better sense of what they need to achieve by a certain time in a clear way and a way that gives a sense of how the project is really developing, while allowing flexibility, so that if things are not shaping up that is the time to stop or change course. How do we do that?
Tony Meggs: Again, not to be too compliant here, I agree with that suggestion. John Manzoni and I, and my team, have had discussions recently about what we can do to get sight of projects at the very beginning, even before they happen—we are getting much better at that through the planning process that Mr Manzoni has introduced. Secondly, as an idea, can we create something in the early phases that is not quite as onerous, if you like, as being a fully-fledged GMPP member, but rather a sort of GMPP early stage, which would capture a large number of projects and subject them to a different form of review, appropriate for that stage?
Q18 Caroline Flint: Mr Manzoni, will you comment on this as well?
John Manzoni: Let me give you a slightly different frame. You have mentioned three projects in particular: CAPD, e-borders and UC—
Chair: You had better say “universal credit” for those following our proceedings.
John Manzoni: I’m sorry, universal credit. Those projects are interesting, but I think we are talking about two different things. Your question is: is the strategic outline business case/outline business case/full business case an appropriate mechanism and methodology? What is interesting is that if you think about those three projects, which you mentioned particularly and which had fallen into some difficulty, they sort of happened right at the moment that we were learning—not just Government, but out there, everyone—about how the new and digital technologies could be applied to big, big business processes. In almost every one of those cases, as it happens, the system was overambitious in its expectations of what those new technologies could bring and how quickly they could bring them. You could make that same observation about all of those three projects. It happened right at the moment that new processes and new ways of doing projects were being introduced into the system. What I mean by that is that the gated process had hitherto been completely best practice in all major projects across the private sector and the public sector, but it was then said, “We are Agile now; we are not Waterfall.” I think we overcompensated and misunderstood that alternative process. Now we know that no project is either Waterfall or Agile; they are a good mixture of both. The judgment and the decisions about how you do each front-end bit might be Agile, but some of it is Waterfall. Those three occurred right at the moment that everybody was learning that. That is not an excuse, but it is partly an explanation.
We are beginning to learn minimum viable product. Let’s do something in a more Agile way. Let’s create a minimum viable product and do it simply. We don’t know what the end state will be. Universal credit has been built in a minimum viable product way, in a digital sense, and now it is going to be expanded and built. Therefore, there is a change happening, and it is to do with the technological change that is going on.
Q19 Caroline Flint: But some concern has been raised about the business planning process, which, from what I can gather, absorbs a huge amount of time and energy from those involved in these projects. Of course, it is about looking at what is happening and making it right, but is the balance of that, in terms of really being able to monitor projects’ performance and viability, acting as a hurdle to better planning at a more practical level for these projects? Does it get in the way somehow? Does it become a bureaucratic nightmare, rather than do what it is meant to do, which is to make sure you have got all your ducks in a row as far as possible? Nothing is without risk, and you can’t answer every question, but you can have a better sense of what you can accomplish by when and how it will be realised.
John Manzoni: Some of these new projects have a different sort of business case. They have a programme business case, which has a different set of characteristics to the standard five-base model in the Treasury Green Book. We are beginning to apply different technologies and techniques to understand the planning of projects. We can come back to single departmental plans, but the planning of projects is evolving. That is an area in which we need to continue to inquire.
Tony Meggs: I would just say one other thing. I don’t want to give the impression that all projects are poorly planned. We tend to focus inevitably, and perhaps rightly, on the things that go wrong, but many more of our projects go right than go wrong, if you can make that simplistic distinction. There is plenty of evidence of very good planning going on. I would say that the Government has become—
Q20 Caroline Flint: Thirty-five per cent of projects due to deliver in the next five years are rated red or amber-red. One of the definitions of a red is that there is a question mark about whether it is going to be deliverable or not. I think that is significant. I understand that the projects you look at are among the most high-risk, high-cost projects. There are lots of other projects happening across Government that don’t come under your view, so to speak.
Tony Meggs: In the matter of planning and early project set-up, not all of those things that are red or amber-red are due to a lack of planning. There are other things that can happen.
Q21 Chair: In answer to one of Ms Flint’s points, you talked about the idea of an early intervention support stage at the very early stages of a project. It sounded like a nice idea. Have you got concrete plans for that, and do they include projects automatically going through that process, or would they come and say, “We want your help”?
Tony Meggs: We have concrete plans, and we are implementing a process of much earlier intervention. We have products, if you want to call them that—interventions—that are designed to look at projects much earlier on. The second thing we are doing, which is really important culturally, is to get much closer to policy makers. I believe that there has historically been a divide—I wasn’t here, so I should be careful what I say—between those who make policy and those who go out and do policy or deliver projects that originate out of policy. One of the important things we are working on is bridging that gap. For example, next month I am going to the policymakers’ professional forum—their annual get-together—and I am going to talk to them about this very topic. There are things that we are doing with respect to early intervention. With respect to the idea of formulating a new early version of the GMPP, that is something that at the moment we are looking at.
Chair: That is very interesting.
Q22 Caroline Flint: That is very helpful. When we came to Oxford it was very helpful to be part of that seminar. One of the presentations outlined how projects can often start with a certain view of the cost estimate, which can then become much bigger than the initial assessment.
How confident are you in the way that the cost estimates for projects are being pulled together? Is there a danger of underestimating a project cost, because that would lead to a more favourable appraisal of the business case and the use of public money, and value for money set against the benefits? What do you think about that? What could be done to achieve a more realistic estimation of cost?
Tony Meggs: Let me say three things. First, Treasury guidance recognises the inherent optimism of the human character and requires an adjustment to estimates to account for “optimism bias”. That is the methodology currently employed. Initial cost estimates are inflated by certain factors to account for and recognise that. That is one thing that is important.
Secondly, the merger with IPA does provide, at least in some areas, the benefit of the fact that they have some real specific expertise in the area of costs and cost estimating, at least for large infrastructure projects. That will be highly beneficial.
The third thing is assurance. The other thing is what I call benchmarking and what we refer to as reference cases, which is an effort. It is difficult, in the context of the multiple and varied projects of Government, to introduce an outside perspective of what something should cost.
In my old work, all projects were benchmarked against industry standards. There were companies that existed, purely and simply, to provide benchmarking data. So you knew what a project should cost and how long it should take. We can do that with some of the more traditional projects, and indeed we are doing that. With the more transformational projects it is very difficult to establish benchmarks per se, but that does not mean that we should not try.
Q23 David Mowat: Mr Meggs, you used a very interesting phrase in your opening statement: “compromised delivery”. Mr Manzoni a few minutes ago used the phrase, “We don’t always know the end state.” They both struck me as very good phrases and true, because things change. Even on successful projects, what you end up delivering is not necessarily what you started off to deliver, nor should it be. Opportunities sometimes arise that you had not thought of at the start and so on.
One way to manage that process is to be very vigorous around the whole area of benefits delivery. It strikes me that you have come in and set up this organisation, which is looking at the whole discipline of delivering projects, something that I think everybody would agree the Government and many organisations need. However, the Report is not particularly complimentary about the area of benefits delivery, from paragraph 2.17 onwards. I am interested to know your perspectives on that, whether there is more that the Government, individual programmes or you could do to get that discipline, and whether that could be the next thing you focus on.
Tony Meggs: We are doing and we have done some things over the last year, for which I claim no credit—the organisation initiated some work that John initiated—to provide a much clearer definition of benefits: how you characterise the benefits in terms of savings, wider economic benefits and, in some cases, social benefits. It is never easy. Again, unlike with oil and gas, you cannot turn everything into a net present value or how many barrels came out of the ground, but we have done a lot of work to develop and propagate a system of benefits description and measurement that is much more rigorous. We have now introduced that system into our reporting requirements, so we require Departments and projects to report those benefits against a framework. However, I accept the criticism: I think that historically we have not paid enough attention to reporting them. My own experience—
Q24 David Mowat: You refer to your previous experience, but it struck me as a little odd that an exit review is not a mandatory thing that a project does. There are different names for it, but you would think that was something that would just be a natural part of the discipline of project management, just like planning is. At the end, you would look back and say, “What did we learn? What did we deliver?”
Tony Meggs: Something like 80 out of 96 recorded in here had an exit review. I, like you, am disappointed that it is not 100%. I think it ought to be mandatory.
Q25 David Mowat: Why isn’t it? That is quite a simple thing, in a way. Is it beyond your—
Tony Meggs: I don’t know.
John Manzoni: I think it sort of is, and occasionally one slips through the net. But even when you do an exit review—this was Tony’s point—the trouble is that it is the underlying data of the original articulation of the benefits that needs to be higher quality. That is the work that is now going on and will emerge. These things take some time, because we collect the data. I think Tony’s team are doing quite a lot of work on firming up the quality of the data about benefits.
Q26 David Mowat: One of the memories I have from being on this Committee in the last few months—we have talked about universal credit a few times and I don’t want to pick on it, but it has an extraordinary business case. My recollection—I don’t have the numbers in front of me—is that £100 million a week is the benefit case in relation to universal credit; I think I am right about that. If that is true, it drives all sorts of behaviours, because it says that if you are a month late, it has cost nearly half a billion pounds. I wonder whether the discipline of that fact—I do not want to over-push it, because I do not want to make a cheap point—was also in, for example, the reset of universal credit, because it would drive, perhaps, how you delivered it if that benefit stream and the need to get it on stream was what was driving your delivery choices. I wonder whether we could do more, in terms of the disciplines that you are trying to get into the organisation, in that area.
Tony Meggs: We can do more. As I said, we have made some significant improvements. We want to create more transparency around it and we want to drive it harder. I think the NAO has said in previous Reports that pushing too hard for early delivery of benefits can be problematic. It is always about finding the right balance between front-end loading—doing all the right planning—and getting on with the job. Finding that balance is always challenging.
Q27 David Mowat: During that session, I think I asked the permanent secretary this: if the benefits are £100 million a week—this is horrendous for the country—who in your organisation is responsible for delivering those benefits? You have both come from BP. I imagine that if there was a benefit stream of £100 million a week there, you would put a team on it to make sure you got those as soon as you could. It was not clear to me that team existed in that project; it was just another thing that the project director had to take care of. Maybe I am wrong in that, and I do not want to go too far with it, but the point I am making is that the organisation really ought to be on that and to regard it as a massive priority to do what it can on delivery options, reset options and sourcing options, including the use of third parties, if you are driven by that £100 million.
Tony Meggs: I do agree with that. I would say—this is relevant to your point—that universal credit is doing more to understand and measure the delivery of benefits than any other programme I have come across in such a complex arena, so I could not fault them. How they have organised it, I am not sure, but I could not fault them for the amount of effort they are putting in to measuring, and ensuring they achieve, the benefits they are setting out to achieve.
Q28 David Mowat: Thank you for that. I have an observation on the MPLA process, which, as the Chair said earlier, the Committee very much supports—to get those disciplines into it. One thing that struck me, though, when I looked at one of the earlier figures in the Report about all the different projects you had, was that the skillset needed to deliver, for example, an aircraft carrier procurement—that may be one of the projects—Hinkley Point C or the Olympics is quite different from that required to deliver a transformation project. Personally, I think the transformation projects are just about the hardest thing anybody can do, and universal credit is a great example of that. Do you have an issue with the fact that, to an extent, you have oranges, apples and lemons coming through and that the interchangeability is not as much as you might expect?
John Manzoni: Let me try to answer that for you. I think you have put your finger on one of the issues that faces us as we go forward. As we increase the proportion of transformative projects in the portfolio—versus the building of bridges, submarines and ships—that does need a different skillset. Two of the highest priorities in terms of skill-building in the civil service today are commercial and digital technology. That is in the project space, but it is actually in the heart of the business space, so we are taking steps to significantly ramp up the pace at which we bring these things in. We don’t have them inside—we have to bring people in from the outside—and that is complex because it raises remuneration issues and all the things we talked about before. But we have proposals which are coming forward now around commercial and digital technology in particular and about how we bring in large numbers of experienced people in those two disciplines. They are slightly different: it is hundreds in the commercial area, but it is probably thousands in the technical area. So we have to bring those people in.
There are components of the MPLA which are focused on those aspects, as well as on the bridges, ships and submarines, so there is a bit of both in the MPLA. But, actually, the issue is: you have to have the raw material. We have a critical shortage of those two skills in Government today, and given what is facing us—you put your finger on it in the Report—we have to build those skills and bring them into Government in a significant way. We have spent the last decade, basically, atrophying those sets of skills, so we have to rebuild them.
Q29 David Mowat: My point—I do not want to take it too far, and I will move on—was that if the disciplines are quite different by type of project, then in terms of the MPLA, you lose a bit of the utility of bringing all these people together to share experiences. You might lose a bit of focus, because the things you need to learn and develop in one area are different in another. That is really all I was saying.
Tony Meggs: Two things. One is, just to be clear, that it is the MPA, or now the IPA. The LA is our leadership academy—just for the record. I would also say that the fundamental concept of providing visibility around projects through the GMPP remains a good concept, regardless of the type of project.
Secondly, notwithstanding the extraordinary range of types of project in Government, there are many commonalities: front-end loading; good planning; getting the right people; and getting the right timescales and so on. Many of the notions of good project delivery and project planning are common. John has succinctly illustrated that there are different skillsets—different skills required within those projects—and that we are light in those areas. But I am still confident that the overall framework, which is designed to create transparency and to provide a kind of early-warning system, remains valid. As we go forward, teasing out the differences between the different types of projects more and perhaps applying different performance criteria to these different types of projects may be necessary.
Q30 David Mowat: The final area I am interested to probe is that of organisation culture and whether or not the Government have got a culture that makes the delivery of projects uniquely difficult. Both of you have come from BP. Presumably you have come because you were regarded within BP as knowing about these things. BP does big projects and delivers them. What observations do you have about the things that Government could learn from BP, or perhaps vice versa, but probably Government could learn from BP in terms of the culture needed to deliver big projects? Perhaps you can answer that question first.
Tony Meggs: First of all, the differences between large organisations and Government are less than sometimes indicated. Secondly, all projects are political in one way or another in private organisations as well as in the Government sector. But I would say two things. The first is rigour. I am used to an environment that is a bit more rigorous in terms of doing things right at the beginning in terms of tracking performance and all of the things we have talked about today. So there is a need, generally speaking—this does not apply everywhere by any means, so I should be careful—for more rigour. Secondly—I think it is obvious—there is a big political dimension. Where I think it manifests itself most is in the sometimes aggressive timescales that are imposed on projects because they essentially fit into a political cycle.
If we could find a way—we do this obviously with very large, long-life projects—of structuring our projects to both accommodate the realities of the political cycle, but also recognise that things can take longer than that—that is, to segment them—and recognise that, for example, transformation projects are a continuous journey. So if we could break things up into more manageable discrete parts that can be accomplished perhaps not easily but within a particular timeframe, that would be a very good thing.
Q31 David Mowat: What I thought you might say something about is that to get to the very top of BP you are unlikely not to have led big projects. You will certainly have been on big projects and delivered big projects and had a track record of big project delivery. I wonder whether that is the case in the Government. If it is not the case, perhaps that creates a culture that might be along the lines of, “Projects are something to be avoided because they go wrong, so let’s just run a Department somewhere and wait till my turn comes”, and that creates a culture that makes it hard to succeed on these very hard things.
John Manzoni: At the top of BP you do not have to have run a big project, but you have to have deep experience of something. Just to complement Tony’s answer, I would say that particularly in the transformative projects and in the infrastructure projects, we hire outside people. We get them done. You mentioned some of them. The Olympics was done and HS2 is being done. We hire outside people. When we are doing projects inside—which is an increasing proportion of the portfolio as you have identified—we need experienced leadership and clear accountability to do them. That is the issue that Government face, and it is something that we have to do as Government: we have to adjust expectations. People with a deeper experience have to succeed. We have to promote people who have what I call a career anchor, as opposed to being the pure generalist. I think that would be my cultural answer, and it is a big deal, because we have spent 10 or 15 years evolving to where we are, and where we are is very important. It is just that we need to add some strands, and those strands all relate to delivery and experience.
Q32 Mr Bacon: Are you saying that it is possible to get to the top of Government without having deep domain experience?
John Manzoni: We have a particular domain experience. We have policy and the craft of government policy domain experience at the top of the civil service.
Q33 Mr Bacon: Good policies are implementable policies, aren’t they, as Sir David Omand once said? We have plenty of generalists among Members of Parliament; I am really talking about officials.
John Manzoni: A lot gets done.
Chair: Let’s not underrate the craft of government; we are a bunch of MPs after all. We are already about an hour in, so I urge questioners and our visitors to be as brief as you can. If questions can be directed either to Tony Meggs or John Manzoni, that will avoid duplication.
Q34 Mr Bacon: Mr Meggs, I was reading your “Reflections of a newcomer” blog, where you say that not only is it “an exciting and dynamic environment”, but “it is hard to do everything well when you’re trying to do so much.” The problem of prioritisation is not new, and it is not new in the private sector. Lou Gerstner said, in his book on trying to turn around IBM, that “making sure that resources are applied to the most important elements of the strategy is perhaps the hardest thing for companies to do”, so this is not a unique public sector problem. How do you help Departments to go about the process of choosing priorities, because that is essential?
Tony Meggs: First of all, this is, if I could say that, one of our highest priorities—to help prioritise. I think John Manzoni made the comment, long before I arrived, that Government try to do too much, and it is difficult to do too much to the highest standards.
What we are doing—and what we have done—in the last few months is introduce some guidance to Departments on prioritisation. Again, that is in the form of guidance, rather than instructions. We are using the single departmental planning process as an opportunity to help or urge Departments towards prioritisation by better matching up the resources they have with the projects that they have to do. Thirdly, we are really promoting the notion of portfolio management across Government. I would say that we are seeing signs of success in that regard. It is too early by a long way to claim victory, but I think we see increasing signs of Departments that manage their major projects and change programmes as a portfolio of activities. John and I talked to a Department a few days ago that has raised the issue of change project portfolio management to the highest level of the Department. Indeed, they have established a special meeting of the executive committee on a regular basis to ensure that those projects are prioritised and co-ordinated among each other—which is a major issue—and that the interdependencies are understood, so that they are managed as a totality. We have a long way to go, and I would love to be able to sit here and tell you that we have got rid of the bottom 20% of the portfolio, but I think we are making progress.
Q35 Mr Bacon: One more question about stable leadership, then I would like to hand over to Caroline Flint. We took a Report recently on the GP extraction service, which had 10 project managers in five years. Five years ago, the inter-city west coast franchising competition did not have a senior responsible owner at all for a period. The Bowman radio communication system from 10 years ago did not have a senior responsible owner for a period. The national probation service information system strategy from 15 years ago had seven programme directors in seven years. Whitehall simply isn’t getting the point about senior people being in charge and staying in charge visibly, is it?
Tony Meggs: I think it is starting to get the point, to respectfully disagree. One of the things that was introduced under John’s leadership was the so-called Osmotherly letters, which are now being published. Those are a contract between myself, permanent secretaries and SROs that specify how much of their time they will devote to a particular project and how long they are expected to stay in post. While it is too early by any stretch to claim victory, we are seeing signs of a decrease in the SRO turnover. As a data point, our last data return shows 7% turnover in the quarter. There will always be some. If you do a little bit of trivial maths, that shows an average life expectancy—no, that is not a good way of putting it; an average time in post—of three and a half to four years. Actually, that is not too bad. We are still struggling with project directors, who are turning over at twice that rate.
Q36 Chair: Can you say, just so that people understand the context, what salary or what level a project director would be on?
Tony Meggs: Typically, one level below an SRO.
Q37 Chair: So roughly what would they earn? Can you just give us a ballpark range so we can put it in context?
Tony Meggs: I am not good on Government pay scales.
Q38 Chair: Mr Manzoni should know this, because you know the civil service.
John Manzoni: I could make a dangerous and flippant remark here.
Tony Meggs: “Not enough.”
John Manzoni: Not enough.
Q39 Chair: Are they being poached by the private sector? I suppose that is really what I am driving at.
John Manzoni: No, I don’t think so. Some of them might be, and it is an issue. Really good project leaders in the private sector are as rare as hen’s teeth. They are probably really difficult to get a hold of and to hold. But look—Tony made this point—we went from 14% to 7% SRO turnover in the period 2012-13 to 2014-15. That is where we are now, so I think we are going in the right direction. We have the pivotal role allowance and various mechanisms to enhance pay for these particularly critical roles; there are 30 or 40 of them out there in the system. Increasingly, for some of the transformation projects, these will be slightly different-looking people with different experiences, and we do have to think about the pay for some of those people.
Chair: We might come back to touch on that.
Q40 Deidre Brock: Mr Manzoni, can you explain how the Osmotherly procedure works in practice? It seems to have made quite a difference in terms of accountability, and the NAO has recognised that.
John Manzoni: Yes, there was a conversation. It would be nice if we could create an environment in which the line of accountability went from the Secretary of State to the permanent secretary down to the project. We had to break that circle a bit, because the permanent secretary is doing a lot of other things and perhaps does not have the necessary skills, so we created an environment in which the SRO signs a letter outlining his or her accountabilities. It says, “Here you are. You’re going to take this project. It is in this state, and it is supposed to be delivering the following benefits. This is the timetable—see annexe A—and this is what you’re going to take.”
We have seen quite a lot of benefit from that. People actually say, “Hang on a minute.” It is helping clarify accountability. One reason why they have taken a bit longer than we thought is that some of the SROs—I completely applaud this—are saying, “I’m not going to sign that until the following conditions exist,” which is exactly what we want. Essentially, they say you could put an SRO here, accountable directly to Parliament for his or her project. That is a mixed blessing if you are an SRO, but it makes the point, and I think it has sharpened accountability. But these were only introduced in the last 12 months, and some of them are still being negotiated, for that reason.
Q41 Deidre Brock: But you do feel that they are starting to make a difference, and that they are part of the improvement that you are seeing?
John Manzoni: Absolutely.
Tony Meggs: Yes.
Q42 Caroline Flint: In June 2015, when the role of the senior responsible owners was looked at—I find that quite a weird title, because it sounds like a dog-owning pensioner or manager—it was still the case that a small number of SROs, I think it was 16%, were responsible for two projects. Given that, from what I understand, SROs not only have responsibility to lead massive projects but have other jobs to do within the Department, where are we on that now? How many SROs are currently having to oversee more than one project?
Tony Meggs: I do not have a number in my head, so I will have to get back to you.
Q43 Caroline Flint: Would you write to us with that? Part of the question is about personnel, turnover and people feeling burnt out or maybe not valued. If you are being asked to oversee two big projects, or maybe more in some cases, it sounds a bit like the amount of work involved in this is not being valued. The consequence is that it may be too much. Should there be a bar on any SRO doing more than one project?
Tony Meggs: If I could put this into a personal context—John Manzoni will have a similar experience—in my old life as a business sponsor, which to a degree is what an SRO represents, it was not uncommon to be responsible for a portfolio of quite big projects. The critical thing is that the project directors got more support and were more highly valued than perhaps they are today.
Q44 Caroline Flint: I understand that they are not like the project manager, because you have the project directors under that, but it cannot help if the SROs are having to deal with more than one project for which they are accountable and are having to come to Committees such as this to explain what is going on, given what was said earlier about a high turnover among project directors, too.
Tony Meggs: For our very largest projects—the top of the tier—it would be much more common for an SRO to be full time.
Q45 Caroline Flint: Perhaps you could write to us about the current situation with SROs, whether they are full time on projects or whether it is something they do on top of another part of their job, and how many projects they have some accountability for. Permanent Secretaries clearly move around Departments, and therefore they aren’t necessarily always subject experts. They may have been in Whitehall for a while, but they are not necessarily an expert when they move from the Home Office to Health or anywhere else. Mr Manzoni, do you think that on the job description for a Permanent Secretary there should be evidence that they have led on a major project or been an SRO?
John Manzoni: Directionally, yes. Specifically, maybe. Our Departments are all really quite different, okay? Some Departments predominantly do policy, and some of them manage almost all the implementation through arm’s length bodies. There is a different blend of experiences, and we have to be pretty careful.
Q46 Caroline Flint: To really give this the value and the status. We talked about this when we were in Oxford. In lots of organisations, including in politics, there are the policy wonks and then there are the doers who actually make things happen. Clearly, they are both essential, but isn’t it important to give this the sort of emphasis it needs by saying that we want Permanent Secretary candidates to have policy expertise but hands-on delivery expertise as well?
John Manzoni: That is why I said what I said. In a general sense, I completely agree with your sentiment.
Q47 Caroline Flint: In the future, could we see that as a pre-requisite of applying for a Permanent Secretary job?
Chair: You are head of the home civil service. You are in a position of power to deliver this.
John Manzoni: No, I am not. I am the chief executive of the civil service. The head of the home civil service, as I keep repeating to the Committee, is my boss.
Q48 Chair: Forgive me. We get muddled, as mere humble politicians. You are in a position actually to do something about this.
John Manzoni: Yes, and I think—
Q49 Chair: It is all very well saying that you agree, but what are you actually going to do about it?
John Manzoni: Let’s say I impose that criterion—then I suddenly haven’t got any Permanent Secretaries to be the next Permanent Secretary.
Q50 Chair: There you go—that says volumes.
John Manzoni: That’s not complete, but that’s why I am hedging—I am not going to promise.
Q51 Mr Bacon: There would be other people who would step up to the plate, don’t you think?
Caroline Flint: You might be able to get someone from outside the civil service.
John Manzoni: What we have got to build is the blend of those skills. We have some excellent operational leaders in our system. We have some really great people; and actually they have spent their life running big operations. Actually, where they fall—you say, “Actually, handling Ministers, with the greatest of respect, is a particular different skill”; and they don’t have that. So we have got to give people, from an early stage—
Q52 Chair: How long would it take?
Mr Bacon: They don’t have that because what? Because they are too blunt? Sir Amyas Morse’s predecessor, Sir John Bourn, wrote in the Financial Times that “The top jobs should go to those who have successfully managed programmes and projects—in health, social welfare and taxation as well as construction and defence. At the moment, they’re given to those best at helping their Ministers get through the political week.” Now, I think that’s just a truism, of how things have been for many years; but are the people you are talking about not good at handling Ministers because they are too blunt, because they are too attached to the truth, or what?
John Manzoni: No, not necessarily. At the very top of the civil service we have Jeremy Heywood and we have me; we have done it in two people. I don’t do what Jeremy does. He does what he does brilliantly; we have yet to see whether I do what I do brilliantly, but we have embodied it in two people, and actually I have a feeling that in some Departments that model could work as well.
Q53 Chair: With respect, Mr Manzoni, without wanting to flatter a witness too readily, I think you handle politicians quite well.
John Manzoni: So far.
Q54 Chair: What I am saying is you have project management. You came in new to the civil service from outside Government, but you seem to have understood the politics of Whitehall pretty quickly. I am saying bright people can pick these things up.
John Manzoni: And so one of the issues is that I do actually now sit on the selection panels for the Permanent Secretaries, which I think is an important thing to do, and I do inject those things—so that conversation is starting, but we can’t change it overnight. I am absolutely directionally in the place that you are, which says, “You know, we need a better blend of skill sets at the very top of the civil service,” because then that will encourage others, and that’s really important.
Q55 Caroline Flint: That sounds like progress could be made in this area. We don’t know quite how long, but as you said it is very important to indicate to those in the middle ranks in the civil service, and those coming in on the fast-track programmes—whatever—that this is what the expectation would be, and should be built into training at every level, I would have thought. That is something you could start influencing now. Isn’t that the case?
John Manzoni: Completely.
Q56 Caroline Flint: Great. I have just a couple more questions. A lot of the projects that come under your scrutiny are obviously big infrastructure projects and transformation projects—IT projects; but, thinking about some of the projects I worked on as a Minister, they also involved the public and their willingness to change their behaviour and make changes, too.
It does seem to me that there are a number of examples where that hasn’t been taken into account as a crucial part of the delivery. So, if I give an example—and this may not have come under you, Mr Meggs—when I was Public Health Minister, and we had to implement a policy on smoke-free England, it was going to affect every single public place in the country and vehicles as well, alongside that, and trains and buses and everything else. I have to say everyone in Whitehall and beyond worked really well on that—what I would call a big project. As a result, because we also engaged with the public’s role in this, both understanding the policy but also enforcing the policy, we have had very few prosecutions.
Now, I am thinking of a couple of other more recent things—for example, the sharing of patient records in the NHS. The public weren’t thought about in this as much as they should be, which led to a big uproar about where the data would go, and a six-month delay or pause in that project going ahead. If I think about pension policy, for example: we are currently having a big debate about women who are affected by changes in their pensions. Everyone agreed that the policy of equalisation was right, but actually you have to expect the public—in this case, women—to make adjustments to their working lives and their financial planning for their retirement, as well.
It does seem—this isn’t party political, because some of these problems have overlapped Governments—that along the way not enough attention was given to engaging the people who were going to have to make changes to make the policy work and make them aware of what they had to do. As you can appreciate, when it comes to financial planning, that is crucially important to how people are going to be looked after when they retire. Do you think that appropriate action and behavioural change by the public should be a key marker of success in some of these projects that we are talking about and should be embedded in all stages of project planning and delivery? They would not be the end consideration, but would be embedded within the project from the start.
Tony Meggs: I absolutely agree with that. It is one of the things that makes public projects so interesting and challenging. Clearly, good project practice says that you involve all of the stakeholders up front, which would therefore point you towards public engagement. There are two things. We might be getting a bit better at this. The digital business of Government is massively focused on understanding the interaction with the public and enhancing it. Along with that is the iterative or agile process, as John Manzoni described it, which allows you to get feedback as you go. I would say that universal credit is now in a place where they are very nimbly adapting as they get continuous feedback from the public. I agree that it is not perfect, that we have a way to go and that we make mistakes, but there is a much higher level of awareness.
Q57 Caroline Flint: Mr Meggs and Mr Manzoni, would you be prepared to have a look at where this part of effective delivery fits into the planning? You mentioned digital, but, as we know, not everybody has access to it. The recent report on the polling of the general election shows that it didn’t take into account a lot of older people’s points of view, and therein lies a tale. Would you be prepared to consider how that could be factored in more? Ultimately, we’re talking about public money being spent, but if the public, as the recipients of the effects of sometimes important policies and projects, aren’t embedded in the delivery process, effective delivery will be an afterthought and we will then get debates like we are having now on pension changes and how they’re affecting a significant number of women.
Tony Meggs: Yes is the answer.
Q58 Caroline Flint: Perhaps you could write to us about where you think that fits into the present system. Would that be okay?
Tony Meggs: Yes.
Q59 Caroline Flint: My final question is a postscript, really. You are mainly concerned with high-risk projects with huge costs associated to them, but there are many other projects that do not come under your view. What is the process in terms of sharing best practice among both major projects and other projects in Whitehall? If people are not following good practice, who picks them up on that?
Tony Meggs: Let me say three things. First, one of our most fundamental objectives in the IPA has to do with the MPLA and the PLP that we are now creating. It is to create awareness and understanding of the principles of good project and programme management in delivery across the whole of Government. The stuff that appears in the GMPP is in there for certain reasons, but I believe that the work that is going on to raise the status and increase skill levels permeates throughout Departments. It is not just focused on the GMPP.
Secondly, there are several levels of assurance, so the types of processes that we have developed and propagated to review very large projects get done within Departments for the smaller projects.
Thirdly, when Departments look at their project portfolio, they look at the totality. It is not perfect, but the disciplines of good project and programme management are being inculcated throughout the system, partly through the work that we do. We now have a network of heads of profession in every department, whose job is to identify talent and to promote good practice, so I think there are a number of things that are going on that are meant to be really fundamental to the whole set of activities and not just the ones that we scrutinise.
Q60 Stephen Phillips: A few short questions from me. Maybe we can focus on the questions, which are quite short; I would just like answers to them. Mr Meggs, of the major projects you are overseeing—I think it is £511 billion worth in the portfolio—which are the three that you are most concerned about and why?
Tony Meggs: This is always a very, very difficult question. Obviously I am concerned about very, very large projects such as HS2. I actually think HS2 has got a fantastic leadership team and is very well structured, but the scope and scale of it—
Q61 Stephen Phillips: HS2 is one of them. What are the other two that you are most concerned about?
Q62 Chair: What keeps you awake at night?
Tony Meggs: I worry about transformation, as the report indicates. An example would be courts reform.
Q63Stephen Phillips: So all transformative projects—?
Tony Meggs: No, not all transformative projects.
Q64 Stephen Phillips: You see, I am asking a very specific question and at the moment you are giving me a pretty woolly answer. I want to know which are the three projects you are most concerned about and why.
Tony Meggs: HS2, because of its scope and scale; courts reform, because of its complexity; and I think shared services.
Q65 Stephen Phillips: Why in particular in relation to shared services—?
Tony Meggs: Because it is not going very well at all. Again, I think it is a project that suffers from not having been well set up. All of the stakeholders are not properly engaged and some of the complexities are misunderstood.
Q66 Stephen Phillips: You said in your opening statement that now we have got this new Infrastructure and Projects Authority, you wanted to take the initiative in ensuring that projects are well delivered. What five things are you going to do to take the initiative, to ensure that very significant spending on these major infrastructure projects is well spent and the projects well done?
Tony Meggs: Well, the first thing that we have discussed today is front-end loading—the initiation of projects. The second thing is that we are going to continue to train and develop project leaders and bring new project leaders into Government, because great project leaders deliver great projects. The third thing we are going to do is to continue to encourage prioritisation, so that we are more focused on the highest value projects.
I would say that the fourth area that we are focused on is—
Q67 Stephen Phillips: If it is only three, then we can stop at three. If you want to write subsequently, because I rather put you on the spot by imposing the discipline of identifying five, I am very happy to have the other two in writing, or indeed another seven in writing. I just want to know what you meant when you said, “We’re going to take the initiative”.
Tony Meggs: Those are three areas. I think that one other area I would add is performance measurement. We are working very hard to improve, as recommended by the Report, the transparency and the quality of the performance reporting on our projects.
Q68 Stephen Phillips: The Infrastructure and Projects Authority is based in Oxford. Presumably, you are based there too.
Tony Meggs: No.
Chair: That’s the MPLA.
Q69 Stephen Phillips: I am so sorry. Forgive me. Well, let me ask this then. In a statement to our predecessors, which we admire so much that Mr Bacon in fact put it in his book, Oliver Letwin told our predecessor Committee in 2011 that “There is incredibly little good quality information in Whitehall, at the centre of Whitehall, about what is going on in Whitehall.” Is there better information within the Infrastructure and Projects Authority about what is going on in Whitehall? If there isn’t, how are you going to get access to the information that you need to ensure the transformation that is required in this area?
Tony Meggs: First of all, it would be inappropriate for me to comment on any comparison between what we have now and the past, because I was not here. I think that we have taken steps in the last two reporting periods—we get reports on a quarterly basis—to enhance the specific data that we are getting and one of the areas that I mentioned was around benefits.
We are doing a great deal of work to improve the quality of that data. By that, I mean we are making sure that all cost data, for example, is on the same basis, which it has not hitherto been. So I would say that we are making very good progress on that. I think the quality of the data is variable, but—
Q70 Stephen Phillips: How are you going to make sure that it gets better?
Tony Meggs: By the initiatives that we have got under way right now. First of all, we have set up a forum with all of the Departments to specifically focus on the issue of data quality and harmonising the data to make sure that we are comparing apples with apples.
Secondly, we are also working very hard on the analysis of that data, because data by itself is of no value. And we are working internally to agree on the appropriate performance metrics, getting the right baselines against which we can measure performance. Thirdly, we are actually engaging with outside parties, such as a number of academic institutions who have interests in project management, to help us make sure that our analysis is first class and to provide access to a wider range of public data to use for benchmarking and comparison.
Q71 Stephen Phillips: Mr Manzoni, can I ask you to look at the NAO’s Report starting at paragraph 2.17, please?
Chair: I should just say that we have got a vote coming shortly, so we will finish Mr Phillips’s questions if we can before we have to go.
Q72 Stephen Phillips: In paragraph 2.17, the NAO—I think correctly—reported that “The Authority collects information but does not report on projects’ progress towards achieving intended benefits as these data are not included in the scope of the Cabinet Office’s transparency policy.” It notes in the first sentence—this must be correct—that “If projects do not deliver their intended benefits they are unlikely to have provided value for money.”
If you do not have a consistent way to measure whether projects have delivered their intended benefits after they have concluded and been handed on, how are the Government and Cabinet Office supposed to know that public money has been well spent? Just as importantly, knowing that those benefits have been achieved in relation to a particular project may in future drive the manner in which other projects are set up. Not doing it, therefore, means that lessons that could be learnt are not learnt. What are you doing about that?
John Manzoni: As we have discussed, Tony has described at some length what he is doing to improve the quality of the data that the MPA is collecting about the projected benefits of any project and, of course, it is no good—
Q73 Stephen Phillips: That is about projected benefits, not about benefits after the project is concluded, which is what this section of the Report is about.
John Manzoni: If we did that, we would end up with a completely centralised economy with the performance of the entire Government being managed through the MPA as a measurement of benefits—
Stephen Phillips: No, that is not what I am suggesting, Mr Manzoni. You well know that—
John Manzoni: Once the project is done, then it is about the accountability of Departments to deliver the stated commitments inside the fiscal envelopes that they have got. So that is now—
Q74 Stephen Phillips: The problem with that, if I may say so to the chief executive of the civil service, is the second sentence of paragraph 2.18, where the NAO says, “We have reported in the past that they”—Departments—“often do not do this.”
John Manzoni: I agree. That brings me back to the single departmental plans, which is exactly what we have set in place this year for the first time, where we bring together both the commitments of any Department—manifesto commitments or other commitments—and the resources in order to deliver those. The single departmental plans are the starting point for a performance management process and performance management awareness that is weaker than it should be in Government. So we now have in place—we did not get it first time round—a single departmental plan per Department. As you know, the public version will be published on 22 January. There are, of course, more detailed versions which will not be published. That will be the basis on which we can then judge ongoing performance. That is quite a breakthrough. We have brought together outputs and inputs by Department for the first time in recent history, I think.
Q75 Stephen Phillips: So next time we discuss this—the first sentence of paragraph 2.19 states: “The Authority is not in a position to monitor whether the benefits of a project are realised once it is complete and activities have become ‘business as usual’”—can I expect that to have changed?
John Manzoni: I am not sure I agree with you because it is not obvious that—
Q76 Stephen Phillips: You don’t agree with me, or you don’t agree with the sentence in an agreed Report with the National Audit Office.
John Manzoni: I don’t think it is making a clear recommendation about whether the Major Projects Authority should measure it or someone else should measure it. At that point, when it is in normal course of business, I believe the Department should measure its performance. The question is: have we got a plan against which to measure that performance?
Q77 Stephen Phillips: At the risk of continuing the love-in—I think the Committee has agreed on quite a lot with you this afternoon—can we agree that somebody, whether the Department or the MPA, needs to measure whether the projected benefits are in fact delivered?
John Manzoni: I think that is a very sensible suggestion—back to the love-in.
Q78 Chair: Hopefully, other Departments will develop the capacity to do that because one of the points of Mr Phillips’s questioning is that if it is business as usual and it is bad business as usual, it is no good, is it?
Q79 David Mowat: If there was an exit review, which we agreed they should all have as part of the methodology—accountability for that delivery may be in the Department to answer—to answer Stephen’s point, there would clearly be documented benefits that would be written down and could be looked at. That would clear the point up, I guess.
Tony Meggs: I agree with that.
Q80 David Mowat: I want to ask one question before the bell goes. It is very hard to disagree with the whole concept of what you are doing—MPLA, better project management and everything else. Perhaps the only risk is that somehow what you doing—in a way this goes back to Mr Manzoni’s last answer—dilutes accountability in some regard. If a project goes wrong and they just say, “Well, the MPLA said it was all right”, who is responsible? There is a risk of that.
Mr Meggs, in five years from now or whenever it is you are doing something different, how will you know that you have done a good job? There are two answers to that. Figures 8, 9 and 10 in this Report could all be green, or you could somehow, as you said earlier, have got a better understanding of the disciplines of project management across Government, which in some ways is an easier objective because it is rather hard to measure. I am interested. What do you think you are accountable for?
Tony Meggs: First, let me just say parenthetically that if 8, 9 and 10 were green in five years, I would think I had failed massively.
Q81 David Mowat: So it’s not that then.
Tony Meggs: Because it would mean that we no longer had a truth-telling machine in operation. I should make that point.
Q82 David Mowat: It would not necessarily mean that. It might just mean the projects had gone better. You cannot just say that because it is red it means they are telling the truth. That’s not necessarily an achievement.
Tony Meggs: Obviously, but in a system of projects as large and as complex as these, as I said in my opening remarks, it is not conceivable that they would all be green.
Q83 David Mowat: Fine. Let’s agree with that. My point was more whether you consider yourself accountable. You mentioned three projects to Mr Phillips: HS2, courts reform and shared services. I don’t think many people would say that you are accountable for HS2, but perhaps you are. I don’t know.
Tony Meggs: I am not—
David Mowat: What I am getting at is whether your role is just an information role, getting more skills across the patch and all the rest of it. Or are you accountable somehow for driving better outcomes in five years, 10 years or whatever it is?
Tony Meggs: I hold myself and the IPA accountable for driving better outcomes, which is very different from being responsible for individual projects—it really is. We can measure—we don’t do it well enough yet—the performance of projects, and I would expect to see the performance of those projects, measured against criteria of time, quality and cost, to improve over time.
Q84 David Mowat: So are you developing—and it is not figures 8, 9 and 10—some kind of quantitative measure that you can use to measure yourself or your group in that way?
Tony Meggs: Yes.
Q85 David Mowat: Right. What is it?
Tony Meggs: Well, it is a measure of how we do as a system, with respect to schedule, cost and, most important, benefits.
Q86 David Mowat: Is that something the NAO looked at? I would have thought that might be something in this analysis.
Tony Meggs: I think the NAO said we do not have the data to be able to do that.
Q87 David Mowat: So in five years they will.
Chair: Perhaps Geraldine could explain that point from the NAO’s perspective.
Geraldine Barker: As Mr Meggs said, it is a system in development, so at the time of writing there was nothing for us to write about.
David Mowat: So there is a system being built to do it. I will not ask whether it is a major project.
Q88 Stephen Phillips: When might it be completed? So that we know there is a system there by which we can measure your performance.
Chair: Otherwise, we might put one in the Report.
Tony Meggs: I will be held to it. This is continuous improvement, here. By this time next year we will be in a much better shape to understand performance, but there are lots of things that we have to do in order to get to that point. The year after that it will be better; and the year after that it will be better.
Q89 David Mowat: Do you need a system? I am thinking about it. I know there are a lot of projects. I don’t know how many there are but you could have a spreadsheet, somebody getting the key numbers and a method of reporting. That doesn’t sound like a system.
Tony Meggs: When I say a system I mean a clear methodology. There are two things that are particularly problematic: one is establishing a clear base line to measure performance against; the second is the measurement of benefits in a quantifiable way.
David Mowat: If it is possible, I think the Committee would find it useful if you could write to us with what you consider is the way that we should measure you, because your role is so critical to much of the way that money is spent across Government. To give us something tangible like that would be very useful for future sessions. [Interruption.]
Chair: We have to suspend for the vote.
Sitting suspended for a Division in the House.
On resuming—
Chair: Another vote is expected. In the meantime, over to Richard Bacon.
Q90 Mr Bacon: I wanted to ask about the traffic light thing again, because somebody who has been watching this hearing texted me in the interim to say, “It’s all very well doing an annual thing, but the point is that people move on, and what you really need is real-time accountability.” That goes back to something that the National Audit Office said to me in an email about this, although not specifically about the traffic light: “The real issue is the general lack of good real-time management information on major projects. It should not require separate data-gathering exercises to establish costs.” Now, you as central scrutineers in an independent assurance agency would surely love to have good independent real-time data on every project as it was happening, wouldn’t you?
Tony Meggs: Let me say two things. One is that I do not think that on projects, unlike daily operations, that you need data points continuously. I do not think it is value-adding.
Q91 Mr Bacon: At all? That’s interesting, because when we did this Report—the one that Marc van Grondelle of KPMG was a witness for, Assurance for major projects—three years ago, 2 May 2012, it talked specifically about different types of assurance. It distinguishes between point-in-time assurance and continuous assurance, and the holy grail was continuous embedded assurance. Paragraph 1.15 states: “Assurance is...either planned or consequential, and...point-in-time or continuous.” Planned, embedded continuous assurance was the holy grail. Why don’t you think it has added value?
Tony Meggs: I think that as a central entity we would drown in information.
Q92 Mr Bacon: But if they are embedded, they are in the project.
Tony Meggs: What Departments do, and we are encouraging proactive portfolio management, ought to be more embedded, by definition, and ought to be more frequent, but I was addressing this from the central perspective and I do not think that we would add value for the public and for other consumers of our information by producing—
Q93 Chair: But, Mr Meggs, at the central point, you oversee lots of projects at departmental level. They could do it and publish it, surely. You do not have to do it all yourself.
Tony Meggs: We should be fair to Departments and say that they are monitoring their projects to greater or lesser degrees on a much more frequent basis than we get data, and I think that is appropriate.
Q94 Mr Bacon: In your blog, “Reflections of a newcomer”, you say, “The fundamental challenges of being in the centralised part of a largely decentralised function are not unfamiliar”. Presumably in BP—perhaps Mr Manzoni could speak to this, too—you did not drown in data; you made sure you got the right data, with rigour, at the right time. Correct?
Tony Meggs: Correct. And that is what we are aiming at here.
Q95 Mr Bacon: But you are not there yet.
Tony Meggs: No.
Q96 Mr Bacon: How long do you think it will be before you have all the right data at the right time with the right rigour?
Tony Meggs: As I said in response to an earlier question, this is a journey. In terms of cleaning up and making sure that we have consistency of approach, we are making a lot of progress. I think we will be able to start in the second quarter of the coming year.
Q97 Mr Bacon: Start what?
Tony Meggs: Start reporting internally on performance against the kinds of metrics that I am talking about. But it will not be perfect; it will take some time to bed the system down to make sure that the data is good. Secondly, and importantly, we will need a period of measurement to be able to establish any trends, because what we are looking for is not just absolute data, but the trends, and that is going to take, frankly, a number of years.
Q98 Caroline Flint: From what you have said, in the second quarter of this year you are looking, at least internally, to roll out some identified, agreed matrix for the data to test out how it will work in terms of helping with performance. That must mean that you have done quite a lot of work already on agreeing what datasets you need to do the best benchmarking on performance. Is that the case? That has happened, has it? It sounds like you are going to go live with something in the second quarter.
Tony Meggs: We have done a great deal of work. First of all, we increased the amount of data that we were asking for two or three quarters ago. Secondly, we are doing a great deal of work to make the data consistent and coherent. The third thing we have to do—this will be a challenge—is establish a clear baseline against which to measure future performance. That sounds trivial, but it is actually quite hard.
Q99 Chair: Can I ask quickly about the Crown Commercial Service? That was introduced to put people with business experience into Departments. How is that going? Is there any early evaluation?
John Manzoni: The Crown Commercial Service is part of the commercial function in Government. I am in the market right now looking for a head of that new commercial function to bring in from the outside. We have been building and have hired recently: I think we are up to 12 senior commercial specialists, who are now the heads of commercial in Departments—two or three of them are in several Departments. We are starting the process of building skills.
We have developed a proposal significantly to increase and accelerate that building of commercial skills across the system, which involves the central employment of senior commercial specialists across our system, but that involves a discussion around how they progress and how they are remunerated—all those things. Those things are not agreed yet, but they are in the final stages of proposal. We are transforming the procurement organisation, which is a slightly separate entity, into a more high-category focused, higher skilled entity. Your Committee is asking us about some of the contract management in commercial in a few weeks’ time, and there is quite a lot to report, I think.
Chair: We have other questions we want to ask, but realistically, with three votes now programmed, I do not think it is reasonable for anyone to keep it running on. We will drop you a short memo in the next couple of days with a couple of questions that we did not get to ask. I apologise that we have been interrupted. We did not predict this, otherwise we might have stepped things up a bit. We will write to you, and we would be grateful if you can respond in good time.
In the meantime, our transcript will be on the website in the next couple of days, and we will obviously send you a copy of our Report. Thank you for coming. As I said at the beginning, we wish you well, but we think there are big challenges, as you will have picked up from the flavour of some of our questioning about ensuring that this is deliverable. We want to ensure that you remain a scrutineer and do not become so much a part of the system that you are coming here to defend indefensible projects. Thank you very much.
Oral evidence: Delivering major projects in government, HC 710 1