Lord O’Donnell.

Q51 The Chair: Good morning and welcome to the Select Committee on Science and Technology. It is a public hearing. We are very pleased to have as our witness Lord O’Donnell, former Cabinet Secretary and, before that, Permanent Secretary at the Treasury.

Lord O’Donnell, as you know, we are conducting an inquiry into financing and scaling of UK science and technology companies, and are very interested to hear from you about the role of the Treasury in funding and financing research and development, and some of the cross-government mechanisms that might support a science and technology strategy. Can you tell us about your time in government at the Treasury, how it approaches funding for R&D and, more broadly, how it approaches innovation policies?

Lord O'Donnell: I am happy to do that. First, I will declare my interests. You think of me as a former Permanent Secretary at the Treasury and a former Cabinet Secretary, but I am also on the main board of Brookfield, which has a private equity arm which alone controls $147 billion-worth of assets, let alone all the other things that we do, so I do get very heavily involved in real investment projects around the world.

But you want to ask me about the Treasury. If you think of this room as the Cabinet Room then virtually all of you want to spend more money. You want to increase what your department is doing and the Treasury has to try and fund this through increasing taxes.

What is the one thing that makes that easier? It is a higher growth rate. People talk about bean counters in the Treasury. The Treasury loves growth—sustainable, long-term growth. That is the easiest thing in the world to make the sums add up. We are passionate about growth. That is why we have set up the enterprise and productivity team.

I know you are going to hear from John Kingman, and that is really important. The issue really is why have John and I written the same papers saying that there are all these barriers to growth and, frankly, we have failed? That is the key point I would like to make.

The Chair: Can you point to the reasons for the failure?

Lord O'Donnell: Yes. We could go through the various things in detail but they are all politically very difficult—they have failed at the political level. Planning reform would be classic; there are so many projects.

I am involved in one. I am on the committee of Wimbledon. We are trying to turn a private golf club into a park and grass tennis courts for qualifying and a stadium. It is taking years, and now they have announced they want to have a judicial review—not of us, but of the GLC for giving us planning permission. It is going to be in the two days in the middle of the Wimbledon Championships, so that is really great, is not it?

If you were serious, you would really tackle judicial review and really tackle planning problems. When I think about how long it takes to do anything in this country of any great note to our infrastructure, it puts people off dramatically.

Then you get people playing games. They start off by saying, “Here’s this project, it is great, do it”. But they have underestimated the costs dramatically and it turns out that it does not work the way they said. So departments play games to get them through.

I remember the London Olympics 2012. Wonderful. They said £2.5 billion; it was £9.5 billion at the end. It may have been a great return for money, but that is the kind of thing you are you are dealing with. MoD procurement—everything—turns out to cost rather a lot more than they said it would, so there are real problems in the system.

Q52 Baroness Willis of Summertown: I am coming at this as a scientist myself but we have heard from DSIT quite a bit. I think one of the areas we are trying to explore is who in the Treasury makes the decisions on science in particular and assesses the R&D, the impacts of R&D and the cost benefits of that. We are still very hazy on that whole question.

Lord O'Donnell: Are you thinking that the Treasury goes through this project by project? No. Remember, come June, there will be a spending review. They have already worked out what the number is for the total and the spending review is just a process of allocating the cake. The department will then have the money and will determine what it does with the money. The Treasury’s controls are admin fees and pay, and large or novel projects will go to it, so a lot of this is about departments.

Baroness Willis of Summertown: So the Treasury does not do the cost-benefit analysis?

Lord O'Donnell: That is a really good question. My first job in the Treasury was rewriting the Green Book to allow for risk and uncertainty on investment appraisals. When I did a review of it with the late Norman Glass, the biggest problem we found was that people were not using it.

I have since worked on revising the Green Book. I have helped the Treasury in the last few years post retirement on how you analyse things where the outcomes are not in money terms—so intangibles; the supplementary guidance on well-being, for example. Again, the biggest problem is great guidance not being used.

A government department should do an analysis using that investment appraisal, but when it comes to it, when I look at the cost-benefit analyses that are done, they are generally quite poor. Quite often that is because it is not quite clear what the desired outcome is. The one thing I would say is that the biggest problem is people are doing this and they are not clear what they are really trying to achieve. I think that is a problem with the mission boards at the moment. It is not going away; it is there at the macro level but also at the micro level. The Treasury is not going to be analysing every last piece of business, but there are some big allocations it can do.

Baroness Willis of Summertown: I have one other question related to that. So it comes through different departments, but there is also the timescale. What we are hearing from a lot of evidence—and I know myself—is that often it takes a long time. For scientific ideas or technologies—engineering biology, for example—we are looking at 10-year timeframes, which are not those typically being worked through in the Treasury. Maybe they are, and maybe the forecast is much longer term.

Lord O'Donnell: The standard way of doing these things is a standard cost-benefit analysis and what determines the issue you are talking about is the discount rate—oh, I was looking round for Lord Stern.

The Chair: Unfortunately he is not with us.

Lord O'Donnell: Nick would tell you that there is an interesting case for using a zero discount rate for really long-term things, and I think we should think about these things. But you then have to distinguish why it is zero for that group rather than for everything.

There is a Treasury scepticism about projects that involve very large amounts of money up front and very big gains later down the road. I would say if you are going to have projects like that, you need to have some way of making sure there are milestones along the way so that you are not just pouring more and more into these projects, which turn out not to deliver.

It is quite difficult, as you can imagine, as you know what the area of start-ups is like. We have dealt with some small start-ups where it has been a great idea, but an even better idea has come along so that that rate of return of 20% that you thought you were going to get actually does not turn out to be 20% at all because something even better has come along. DeepSeek is the classic.

Baroness Willis of Summertown: Is that also to do with the appetite for risk? What is the appetite for risk in the Treasury, would you say?

Lord O'Donnell: The appetite for risk is not bad. The biggest problem in the Treasury and the biggest problem for Government as a whole is that you are crucified for failure, and nobody gives a monkey’s about success. Parliament has set itself up with the National Audit Office that does big hearings, and you make your political career by saying, “This is terrible. You shouldn’t have done this”. It happens all the time, so you have built into the system a massive degree of risk aversion.

I am glad to say that the new leadership in the NAO has recognised this, but I do not know whether that will really change the way hearings in the other place go. But if you want to do something fundamental to change the risk aversion particularly among civil servants, can we kindly celebrate some successes? Can we kindly accept the fact that it is a really good idea that you failed because the Government—God bless the CDL—has said that test and learn is really part of it? Absolutely. If you are going to test and learn, you are going to fail quite frequently, and that is a good thing, not a bad thing. At the moment, the biases are massively towards “Let us just be safe”.

The area where I think it shows up is procurement. That very old-fashioned phrase, “No one lost their job for buying from IBM", is kind of a long way out of date now. We need to learn the lessons, particularly from Covid, where we looked at the regulatory rules and suddenly found we could do stuff in parallel. We can get vaccines regulated more quickly, completely safely.

Why do we have the old rules? Why can we not be much more innovative about saying that actually these regulations are just way too slow? They do not allow for innovation. I remember getting some difficult times from a Select Committee when we bought tablets for SARS, way back before Covid. It was basically, “This is precautionary, this could be terrible, and therefore, I am buying them”. It turned out that they were not needed; great, that is fine.

After that, we built up the PPE stocks but, surprise, surprise, the NHS is very strained for cash, so it ran them down before Covid. You definitely need to think about what incentive structures you are putting in there.

Q53 Lord Drayson: You have ranged quite widely in your initial answers there, and I would like to bring you back to the Treasury. Are you telling us that you feel that the Green Book is fit for purpose in 2025?

Lord O'Donnell: There are some things I would change about the Green Book. The Green Book was set up really to do it project by project and that does not work for innovative projects. The change I would make would be to say that we need a portfolio approach and are going to look at a group of projects, all of which, for example, are on how we might apply AI to the collection of taxes and how we might do AI in different departments. Then you would say, “We expect seven of these projects to fail”.

Lord Drayson: So it is a portfolio approach? Do you feel—I think you have mentioned it already—that it is the ability to properly take into account in a cost-benefit analysis the intellectual property, intangible benefits, and spin-offs? Do people blame it for reasons that are actually not true; they are just not using it properly?

Lord O'Donnell: Indeed. I have had this from the start. You try and say to people that to do a proper cost-benefit analysis you have to quantify these things and you hit all sorts of political obstacles. Let us take Covid. When you do the cost-benefit analysis of Covid and all the various things that were going on, you need a value of life. Politicians are just not going there, absolutely not.

Lord Drayson: NICE has that. There is a value of life.

Lord O'Donnell: Of course there is but if you look at the cost-benefit studies done, we did not use it.

Lord Drayson: Is that because of politics? Is that because of Ministers refusing to do it?

Lord O'Donnell: Yes. There is SAGE and all the rest. They could have asked officials to do an analysis. You could have done a proper cost-benefit analysis of closing the schools. Where it goes wrong is that you have really poor cost-benefit analysis done. That is either because they are not asking the right questions or do not have enough capacity.

At times, that is certainly true within the Treasury because it is a very constrained department, even though it is a lot bigger than it was in my day, and there may not be enough resources. But there are plenty of people out there. I have just stood down from running an economics consultancy, Frontier Economics. I have over 400 economists and we do proper cost-benefit analysis. We did stuff for DSIT on this very subject— which I hope you are reading, on innovation and scaling.

Lord Drayson: On the talent pool in the Treasury, given that the impact of technology and science on society is doing that, why has the Treasury failed to get in-depth proper scientific expertise inside the Treasury?

Lord O'Donnell: Number one—and it is not just true of the Treasury but of the whole Civil Service—is pay. It is very, very simple. If there was one recommendation you could do, this would be mine. We have the dumbest pay system you could imagine. Look at me: while I was a civil servant working very hard, I did not get paid very much. I could have made huge amounts more working in an investment bank and was tempted along the way. Now I am a retired civil servant, I am getting a big index-linked pension. What incentive structure system pays you most after you have left?

Imagine if you are a young scientist thinking of joining the Treasury and you are going to live in London because that is where Westminster and things are. Your pay is pathetic but you have got this great pension. You go to the building society. Are they going to give you a mortgage? No chance. They do not pay any attention to your pension. If we switched one thing, it would be to switch the pay/pension stuff.

Lord Drayson: My last question is: having left, have you seen how other countries have developed the capacity in their treasuries to do this well? Brookfield is looking around the world. Who has got their act together?

Lord O'Donnell: There is a very clear answer to that, because I did some work with the University of Oxford, the Blavatnik School, on who has the best civil services around the world. Number one is Singapore. The reason for that is the pay is massively higher, both for civil servants and for Ministers. It attracts people. It has much greater movement between the sectors. We have obsessed by saying there is a conflict if this person goes and works for an industry which actually gives you real knowledge; it is crazy. We are shooting ourselves in the foot there. if we have got to model on ourselves on someone, it is definitely Singapore, where the trust levels between Ministers and officials are really high.

Lord Drayson: Would you say that the UK has a cultural problem here in the relationship between its Civil Service, its Government and industry? It really struck me, being on the board of Airbus, the difference in the relationship between the executives of Airbus and the Elysee, the Chancellery and the British Government.

Lord O'Donnell: Yes, because if you take the French system—the énarques and all the rest—they move in and out and have done both things. I have worked with senior French civil servants and know Airbus well, so yes, the relationships are very different and much stronger.

I think we need more mobility between the sectors. But some of the best civil servants who worked for me have now gone outside, some to Airbus, some to the big four, all the rest of it. They all say, “I am only doing it for a few years, and then I will come back in”. Surprise, surprise, when they are faced with getting their pay cut to a third of what it is, they decide to stay in the private sector so we do not get them back, so we lose all that interaction.

Q54 Baroness Neuberger: To some extent, I am channelling my inner Lord Stern, because I am no expert, and Nick certainly is. Obviously, you have answered part of this, but we are interested because the UK has seen stagnant productivity, so it is putting pressure on the fiscal sustainability of the Government. How much do you feel that the Treasury really owns the productivity problem, or is it just concerned with balancing the books? You said a bit about that. How do we embed the long-term thinking and investment in R&D within the Treasury? It is a key question.

Lord O'Donnell: That is the key question. The idea that the Treasury is just about balancing the books really frustrates me. I mean, come on; that is so ridiculous. If you look at what I wrote in the FT recently about the need for greater investment—

Baroness Neuberger: We have read that.

Lord O'Donnell: Good. It is absolutely crucial. We have very low investment. I could go back to 1997, when Gordon Brown came in. The problem we had then with public investment was that capital and current was put together. Every time there was a budget crisis, we cut the capital projects. That was the way you did it. With separating the two out and having a golden rule, lots of things have changed, which has improved things.

We had one-year spending reviews, moving it to three. I have urged that we have 10-year spending reviews. Nobody listened—which is nice. But if you are going to plan these things, you have to come back to the question about how you get this long-term thinking. If you set up the structures to be so short-term, that is what you are going to get, so a 10-year spending review would make a huge difference. I strongly support that.

I know you are going to talk to John Kingman, who ran the productivity and growth team in the Treasury. God bless him. He and I have put forward paper after paper as to how you improve growth, and we failed. It is a bit like how I have told Chancellor after Chancellor about the pay/pensions split, and I have failed—for the stupidest of reasons. They are accounting reasons.

If we increase the pay of civil servants, this shows up as an increase in the deficit. I say, “Yes, do not worry”. The Chancellor will say that this is offset by the fact that we have reduced the pensions, so that will reduce our long-term costs—no.

Baroness Neuberger: It does not make any difference, does it?

Lord O'Donnell: It does not make any difference, because your debt numbers do not include the pension liabilities. Why do they not include the pension liabilities? Because we have an unfunded pension liability scheme. Therefore, they say that the markets will notice debt not moving. I think the markets would be a bit smarter than that.

This leads on to your questions about Singapore. If you think about how you do this right, you end up thinking about unfunded tax liabilities. Can we fund them? Can we set up some kind of project or national wealth fund that would actually then invest? Again, the classic is GIC in Singapore which does fantastic things. So, there is lots we could do, but good luck getting politicians to buy it.

Baroness Neuberger: How would you—I think you have probably tried—change the way that policies in the Treasury are formulated? That is partly a political question. You gave the example of valuing life. If you shape the policies, how do you then have the battle discussion—or whatever—with the politicians to get what you want through?

Lord O'Donnell: Number one is making sure that you have the right goal, that you actually understand. You need to say to the Treasury, “Here’s what we want you to achieve”. At the moment, as far as I can understand, the Treasury is trying to achieve living with the fiscal rules and the fastest GDP growth rate in the G7, which is clearly not a sensible target. We can do that, if we just open all the coal mines. It is no problem. Then, I could deliver the fastest growth in the G7, just like that. So, they do not mean that.

If you have an outcome that you do not really mean, then you are in a mess. So, if they were to say “this is what we want as the outcome”—some major companies which are big players in the tech space, to increase employment in tech and to rebalance our economy towards this—then that would involve some really difficult decisions.

I spent Saturday in this place some time ago discussing Scunthorpe and steel. We were all told that this is a strategic requirement. Really? We have a National Security Act now, which has a really important part to play in determining what really is strategic. If you have a budget for this, do you want to spend £700,000 a day on the Scunthorpe steel plant?

Q55 Lord Borwick: You complimented the French and the Singaporean systems because of the trust between the civil service—

Lord O'Donnell: Singapore.

Lord Borwick: And you complimented the énarques for the way—

Lord O'Donnell: For the movement back and forward.

Lord Borwick: But is that not rather similar to the Eton and Oxford-type old tradition of the British Civil Service, where the senior ones seemed to be educated very similarly to the senior decision-makers in politics?

Lord O'Donnell: You are saying this to a boy who was state-educated and went to Warwick as his first university.

Lord Borwick: Exactly. You are an example of how a more diverse system works. Does it work better?

Lord O'Donnell: No—it has got worse, and I will tell you why. I refer you back to my previous answer about pay. Do you know who stays in the Civil Service now? The people who cannot get jobs anywhere else. That is not good for anybody. Let us be honest—it is poor performance. The Civil Service itself says that this is a big problem.

Secondly, they stay in because they have public spirit, and they have the bank of Mum and Dad—so because they can afford it. I refer to my own family in this case. If, on the other hand, you have come from a working-class family, you have worked your way up, you are trying to get a flat in London, you have no deposit, you are on Civil Service pay, then you join the investment bank and you get out.

So, we are losing all the diversity from the system, and it is a tragedy. I am not talking about fancy DEI or anything, I am talking about, at the bottom line, money. It is crazy. The backgrounds of those who stay in are going to become much less diverse, and we are going to lose the good ones.

Lord Borwick: Back to pay. To move on to the main question about investment, an earlier witness suggested that there was a conflict between the sale of gilts and encouraging people to invest in British companies for the future, that the two are in competition for the same investment funds. Do you think that is fair?

Lord O'Donnell: Obviously, there is competition. For me, the big issue—which is a really important point—is that if you look at our pensions and insurance companies, they are full of gilts. I can now refer to my experience with Brookfield. Brookfield has bought lots of insurance companies in the States. The thing about insurance companies is that they have got a set of liabilities which are very predictable. You know what they are, pretty much. You have some longevity risks, but not much else. The way they are investing their money is incredibly risk averse.

We have bought them, and it is fantastic business, because we can invest the money far better and get a greater return. We know what the liabilities are. I honestly do not understand precisely why they are so conservative in their investment structures. It would be very good to talk to John Kingman as the chairman of L&G. It is about getting them to invest in private equity, venture capital and all of these things. They seem obsessed with holding rather more gilts than I would think. If, in the end, that means we have more money in VAT and slightly less demand for gilts, that is a trade-off I would take, definitely.

Lord Borwick: But it is a function of the return, is it not?

Lord O'Donnell: I do not think so, because curiously enough, as for foreign owners, when I look at these things, I think that I would not run them like that.

Lord Borwick: The investors are looking over their shoulders at the regulators and the comments that the regulators would make for them if they got it wrong.

Lord O'Donnell: But how can you get it wrong? You know that the long-term return from equities is much higher than the long-term return from bonds, so how are you going to get it wrong? You might get it wrong if—

Lord Borwick: By your investment going bust.

Lord O'Donnell: Some will go bust but, over the long run, you will get a higher rate of return if you have a higher equity share and a higher share of the scale-ups and the other bits.

That is the whole point about being big: you can take failures. That is fine. In fact, you are bound to have some failures, and that is good. So we probably need a bit more foreign ownership, not less.

Lord Borwick: We should celebrate it when someone is investing in Britain.

Lord O'Donnell: Absolutely. If you look at the work that Jonathan Haskel did on productivity and go back to see which UK-owned firms and companies improved productivity, you see that they tend to be foreign. They come in with new ideas and new ways of doing things, and they increase productivity. On Unilever, I am sure he would tell you that. Using Japanese techniques, they improved efficiency dramatically; I should not say this in front of someone who knows a lot more about the car industry than me.

Q56 Lord Drayson: Do you not think that the way in which the UK has conflated investment with ownership and control has massively undermined our self-confidence as a nation? You mentioned DeepMind, which sold for £400 million. Control of this company takes place elsewhere. We have seen that with company after company, and infrastructure after infrastructure. How does that get taken into account in an assessment of whether inward investment is a good thing or whether inward investment should be balanced with a strategic long-term view of what is in this country’s long-term interest? We seem to be a complete outlier compared to other countries, in that basically everything is up for sale here.

Lord O'Donnell: In the National Security Act—this is obviously past my time—there are ways that that could be used to control things. To be perfectly honest, I was surprised that Arm was allowed to be sold. I thought that was a strategic interest.

Lord Drayson: Is there a strategic interest in having Palantir running a federated data environment for the NHS?

Lord O'Donnell: That is an interesting question about whether Palantir does it better than anyone else. In all these things you need to think about the risk. The tech giants of the world are either US or Chinese. China is definitely an unreliable partner, and the US—well. How will we do it? We need to play in this space. We are about to try to use AI to enhance everything, and if we say, “No foreigners allowed”, we will not make progress.

We need to think about the areas where we definitely need strategic control, and we need a proper analysis. For all these sorts of things, I love the idea of a National Security Council structure, where the committee that would look at it would be the group of politicians who are relevant in the different areas, plus the key delivery agencies or Permanent Secretaries.

That is why the NSC works—because it has the agency heads and the JIC heads, as you will know. You get all of them around the same table, as opposed to—dare I say it?—what I mostly lived through: Cabinet committees with all the Ministers round the table, having been briefed by their civil servants, who are having a meeting over there. Can we grow up? Can we put the two together? In Singapore, you would not know who the Ministers and civil servants in the room are when you enter a meeting with them. I am a fellow of the Civil Service College in Singapore—I should own up to that—and I work with it closely. It does it much better.

Q57 Lord Ranger of Northwood: I apologise for being late this morning. You mentioned the French and how they operate. I spent 10 years in a global French IT business, so I have seen the relationship between industry, senior civil servants and the French Government—it blends, as you said, and there is a lot of understanding between them.

We were also a strategic supplier for the UK Government. I dealt with the Cabinet Office extensively during that period. But, coming back to Palantir and other organisations, the challenge is: if we have to play in this space, as you rightly say, and do not have a sovereign base to lean on, what are we going to do to help that situation? For example, the French prioritise the development of their sovereign capability in these various technologies, and they have been doing so for a pretty long time, so they are ahead.

When I was on the other side and was engaging with British Government, I found that we were not looking at that. We were not looking at our longer-term strategic positioning. There are UK businesses that have the potential to be of size and scale. Speaking as someone from industry and still advising industry, the constant feedback that I hear from smaller-scale British businesses is: “We do not get our look in”. It is not about winning those immediate contracts; it is to help develop for a longer-term play. How can we do that better?

Lord O'Donnell: There are two things. One is that, as I said earlier, I would change the procurement rules to make them much more like they were during Covid. I would relax some of the rules and think about accepting the fact that we will take more risks in this space and back some small-scale people, and some of them will go bust. We would not then be hauled and crucified in front of Parliament. That requires some changes in the other place. So, yes, I completely agree with you.

The other part of it, of course, would be the industrial strategy. You want the long-term strategic view: what are the things that we should care about? What are the things that we strategically need to make in the UK? We are short on rare earth materials, and there are all sorts of other things we will not have here and will be reliant on. Do you think you can have an independent nuclear deterrent without co-operation with the US? Dream on.

So there are definitely areas where we need to have that serious conversation and that need to be put in the industrial strategy—what are the things that we think, strategically, we need to invest in as a country from the bottom up to develop our own expertise? I completely agree with that.

The Chair: Baroness Northover, you want to pursue the question of industrial strategy.

Q58 Baroness Northover: I do. Picking up on what you have said so far, I suppose one major reason why the Civil Service would be risk averse is that the politicians are risk averse, because we are in an oppositional political structure. So maybe we need more coalitions.

Setting that aside, as you know the Government are currently developing their industrial strategy, which we expect to be published imminently. From your experience of previous efforts, including in the coalition of course, what advice would you have for them in developing and implementing a long-term economic strategy like this? What have we learned from previous attempts to implement an industrial strategy? What worked and what did not?

Lord O'Donnell: That is a difficult question. For a long time, we did not really have industrial strategies, and it was hard to get that conversation going because there was this aversion. It dates back to Margaret Thatcher: “The Government should not be picking winners. We should let the market decide. Who are the Government to get into this? The market will sort out the winners and losers”, and you left it at that. We have moved on quite significantly from that to ask: what are the market failures? What are the areas we need to be involved in, strategically, from a security point of view?

The problem is that, sometimes, those conversations get buried and hidden, and it is not really explicit. It is all about trade-offs, and politicians do not like talking about trade-offs. They like win-wins. You say to them, “You can’t have 38 strategic priorities. We need a smaller number. Come on”. There are trade-offs between different things if you are going to go down certain routes.

If we were to, say, get our pension insurance companies to invest more in UK businesses—I saw reports in this morning’s papers about that and all the rest of it—they might invest less in gilts. That would mean that the debt interest bill, which is currently our second-largest spending programme, would go up even more. It is a trade-off. You have to take it. Grown up discussions of trade-offs would really help.

I think understanding the strategic side is our weakness is in science and technology. Patrick Vallance is a hero for me. He is exactly what we need. I have worked closely with Chief Scientific Advisers—it was Dave King in my day—and others, and that relationship is really important.

I was talking to someone from the Government Office for Science yesterday about the fact that scientists in government tend to be a bit purist—they tend to want to be in their camp. I have always contrasted this with the economists. I admit that I am an ex-academic economist; I was head of the GES. We are like Japanese knotweed; we get everywhere. We are quite happy to think about cost-benefit analysis of this, how you manage that and what about these intangibles.

Doing this properly is our game. Doing it badly happens quite a lot unfortunately. Again, I refer back to the fact that we do not get the best people because we do not pay the best. We need the scientists to get in there, mix it, be in a room, be prepared to challenge the politicians—to be prepared to say, “Actually, we need more scientists involved in this”.

We need to be proactive about how we hire them. GCHQ and the National Cyber Security Centre both do quite well in hiring really good scientists. Why is that? I asked Ciaran Martin, who set up the National Cyber Security Centre, and he said basically it has specialist pay which is above the norm. It is not the same as market pay, but it gives recognition.

There are people—and this is true of science and was true of me when I was working in the Treasury with some of our econometrician modellers— who, frankly, are really good specialists but could not manage a drink-up in a brewery. You need to find ways for these specialists to be retained, to do what they are really good at, and not put them in large teams where they spend their time managing, where they have no idea. It does come back to Ciaran’s description of an extrovert in GCHQ being someone who looked at your shoes when they spoke to you rather than their own.

Baroness Northover: There were the attempts, both in the coalition and by Greg Clark, to have industrial strategies. What did you feel worked? What did you feel did not work well? Those were attempts when people were signed up to the notion of an industrial strategy and we have that coming through again, which is very welcome.

Lord O'Donnell: I would say it was not as comprehensive as it needed to be. I agree with you. I really enjoyed coalition government because it was really interesting and novel and different, but I literally was there only for the start and getting it going, so by the end of 2011 I had retired. I cannot really say what worked and what did not and why.

Baroness Northover: I was not thinking of the coalition—I think the coalition worked very well too, except in the end. I was thinking in terms of the industrial strategies that came forward against the challenge of the Opposition or oppositional politics—where everything put forward will be shot down by the Opposition or is not enough or not the right things et cetera. Plus there are all the MPs wanting something in their area which is the thing that pulls the NHS apart. What did you feel worked well and what did not well work well about those ones? What could we take over into what is happening now? What ought our lessons to be?

Lord O'Donnell: I have to own up that, yes, I am in favour of proportional representation. You go into the Commons and the Opposition’s job is to oppose and that creates certain problems. I mentioned the joys of Singapore. There has just been an election in Singapore where, for the first time, they thought the Opposition would do quite well and the vote share of the Government went up.

You can have a democracy where only one party wins, and that gives you a lot of stability and therefore it has long-term industrial strategies. You could do it and could try and do cross-party consensus and all this, but I have tried that and within our system I have not found a way to make that work particularly well. It kind of works in some areas, maybe in security occasionally when you have briefings and can try and get everybody on side, but it is mixed.

Q59 Baroness Neuberger: I am an economist and I am married to an economist, but I wanted to pick you up on the grown-up conversations and the way we needed to have conversations in a different way. I really just wanted to see whether you might say a little bit more about how we could perhaps achieve that in this country, because you were beginning to and then and then you moved away.

Lord O'Donnell: The way to do that is through good old-fashioned things such as a Green Paper. Lots of people have written about industrial strategy—Andy Haldane has written some really interesting stuff on industrial strategy. I think we need a grown-up debate and I would like to have the private sector really involved in that. What would the private sector find really useful in terms of guidance, both at the small and the large level?

We could try and find ways to do that process, and through that come up with some ideas about what would work and what would be sustainable and that all the sectors would have buy-in to. Occasionally you see this. I think back to the pensions review that Tony Blair did which took a lot of time and got pretty much cross-party support. For big changes, you need to try and get that support.

If you have an industrial strategy, which by definition is a 10-year plan, and the Opposition are saying they do not agree with it and when they get in in X years’ time, which is a lot less than 10, they are going to have a different one, if I were advising Brookfield whether it is a long-term strategy it can rely on, the answer is “No, not really”. It is there as long as this Government survive. We need to find ways to work in a more consensual way.

Q60 Baroness Young of Old Scone: You talked about the movement towards more flexible thinking about backing winners, or at least marginally selecting sectors in which winners might exist and which are important for us strategically. On that basis, does the R&D tax credit regime need a major overhaul? It is a very broad brush. It is a big chunk of the R&D and innovation budget, and basically has no selection criteria whatever.

Lord O'Donnell: Everyone has ideas about spending and we can all spend more money on stuff. Whenever anybody asks me how we save money, number one for me would be looking at the back of those Treasury costings books and looking at tax, looking at all the credits, and the tax reliefs. R&D tax credits is a good example. They are £9 billion. Fraud is huge. I think that the HMRC study came up with £4 billion. Targeting is quite poor, I would say. I would think there is a really strong case for saying: is this really the right way to spend those many billions of pounds?

It is a bit like tax relief on pensions. We all thought, “Well, that is very sensible because that is encouraging people to save for their pensions”. But it turned out that the best thing we have ever done for increasing savings for our retirement was auto-enrolment. It was nothing to do with tax credits. It is massively cheap compared to tax credits.

So, yes, if you start with a problem and say, “What is the outcome we want?”—and then think about all the different ways you could do it—R&D tax credits would be one of them. But I would not start by saying, “Let’s just look at tax credits and think about how we might reform that”. I would say, “Give me the outcome you really want, coming from your industrial strategy, and then let us think about all the different tools that exist to achieve that outcome”—and not just start by thinking about a tax credit. It might be grants, or all sorts of things.

Baroness Young of Old Scone: Should there be a state and Treasury role if you were introducing more grants, directed investment programmes and things like that?

Lord O'Donnell: I would give that to DSIT, without a doubt. I would say to DSIT, “Here’s the outcome. Here’s the amount of money we have allocated to it. Get to it”—obviously with some milestones, reporting and all the rest of it. The Treasury is too small; it is doing fiscal and all sorts of things. It has multiple taxes and all the rest of it. If you want to really understand industry and you have got a department with the word “industry” in it, surely it is its job. Surely it has the experts. So I would definitely put all this with DSIT.

Q61 Baroness Neville-Jones: I will ask you about cross-government co-ordination, but I just want to probe you a little more about the institutions we have. You commented on the National Security Council not really playing the role it should, in the sense of including the economic side of things. The way I designed it, it should have done. Interestingly, we have achieved it in cyber but not across the rest of policy, particularly in relation to China. It is the single biggest threat we face, and we have no China policy and no consequences as a result. If we had one, we might, for instance, have had a policy on rare earths. Ministers are not using the tools of government that actually are available. This is not the blob.

Lord O'Donnell: I completely agree with you. The really interesting question is: how do you get the NSC into the economic sphere? Who are the people you would have to add around the table? Normally—you are right—you are talking about cyber and other things.

Baroness Neville-Jones: Letwin got in a bit, and he did it in cyber.

Lord O'Donnell: Indeed—you can talk about Huawei and how it will get involved. You have to have that. I would love there to be an NSC with a strong economic component to it, which actually would be the place where you would sort out these big strategic issues.

Baroness Neville-Jones: Talking about co-ordination across Government, you will not be surprised to hear that Lord Vallance agrees with you about the 10-year approach. He is explicit on the subject—so that is two powerful voices—and I know he takes the view that it has to run across finance, regulation, procurement and so on. If it will run across all those functions, it will have to run across the departments.

We are not brilliant at co-ordination or conducting joint policies. The lead government department and the way the money works—you have to account for every penny that you are responsible for—lean against co-operating and sharing or pooling financial responsibility. What would you try to get Government to do to make this a more successful area? Is DSIT the department that ought to be doing the co-ordination where we are, is it another department or is it the Cabinet Office?

Lord O'Donnell: You are right, but it is really important that people understand that this is constitutional.

Baroness Neville-Jones: It absolutely is, and that is part of the problem of why you cannot get it going.

Lord O'Donnell: We have a constitution that is all about departments, and there is very little about cross-department work. That makes life really hard, and you say to a Secretary of State—

Baroness Neville-Jones: It is the law.

Lord O'Donnell: Exactly: “I am looking after this. You want me to spend some money and he gets the political credit”—come on; it is not going to happen. So, yes, we should be thinking about ways to break down the barriers. The National Security Council definitely worked, and extrapolating that to lots more areas would be a really good idea. To give this Government credit, they thought that mission boards might do it, but I do not think they are they are the best way, to be honest.

Baroness Neville-Jones: Why?

Lord O'Donnell: These things are only as good as the outcomes. If your outcome is an input measure, you are not going to get anywhere. Some of the outcomes are contradictory between the missions. So I am not hopeful—but there we are.

You have to start with a big mission: what are the Government trying to achieve? For me, that would be very straightforward. It would be something like “Improving the well-being of the nation”. Fine—so how do you do that? Then you break it down into all these things. You need economic growth, but you need sustainable long-term economic growth of the right kind, as well as these other strategic things and all the rest of it. You would have a cascade downwards, with the slogans to go along with it. From “Fixing the foundations”, you move on to “Building better lives”. Then you would say, “Okay, how do we build better lives?” Then you move on to education—you talk about STEM and how you are delivering the right people coming through the system.

You could do all this, but we are lacking the vision. There is no big-picture stuff. You have individual little mission boards that are doing very narrow things. They are trying to get departments to come together, but departments are doing other things that are sometimes contradictory to that. This is where you may need the Cabinet to move to the A-team, as it were—this has often been thought about—as a small group that manages the vision bit and then basically delegates down and sorts things out. You need some people who are prepared to think about this. We have done small things—you will know about this—such as Foreign Office development putting money together into individual pots.

Baroness Neville-Jones: That was the one example I could cite.

Lord O'Donnell: Absolutely—that was in our day, and it was a success, but these are small. Where are we on rare earths and computer chips? There are all sorts of things that other countries are grappling with—the Inflation Reduction Act in the US was an attempt to think about this.

Baroness Neville-Jones: We do not have a framework for policy. We make it much harder.

Lord O'Donnell: That is right. You start with a big picture that has a vision, and you go from there. But we are long way away from that.

Q62 Baroness Walmsley: I will take you back to another kind of co-ordination. We have heard from you and others about the need to embed more academic experience, business and investment in the Civil Service. We have certainly heard your message about shifting the balance between pensions and pay. Is that the only answer, or can you think of any other ways of doing it, such as time-limited embedding, keeping your pay and moving back again, in both directions? Is there anything like that?

Lord O'Donnell: There is. You are absolutely right. Secondments would be good. At the moment, when you move out you have ACOBA—God bless. I remember, when I left, putting in an application to it and getting told off that I had not asked it about being a visiting professor at LSE—which paid zero, to be absolutely clear. I could not think what possible conflict there was. We need to task ACOBA and say, “We like the idea of people moving in and out. We want you to make that easier. We want you to encourage it”. We want both sides, the private sector and the public sector, to be grown up about this. Yes, there might be some grown-up conflict rules, but that is fine. But let us say that our outcome is that this happens more, not less.

Baroness Walmsley: We have heard about the benefits to the Civil Service of doing this. How can we make sure that business also feels a benefit?

Lord O'Donnell: It will—you would not believe this. I am lucky to be paid a lot by people who do not understand Government. They have no idea what you lot are up to—what we are up to. They say, “What do you mean, ‘Second Reading’? What’s the House of Lords got to do with this?”

I am embarrassed to say that it is really quite simple to help the private sector in dealing with the public sector. As I think anyone who has done both will know, sometimes you have to explain to them the total oddities of a Cabinet committee system, a write-round, what a spending review is doing and why this is deemed novel and original by the Treasury and therefore they will call it in.

So, there is a lot that you can do for the private sector, and once you work with them, it is great. What they also realise is, God, what a difficult job you have got, compared to the private sector. When you are in an organisation where the CEO has been in that organisation for ever, knows the business backwards, is an expert and has surrounded himself with really good people, it is quite easy.

Baroness Walmsley: They know nothing about how politics works. I think we have all had the questions, have we not?

Q63 Viscount Stansgate: I am sorry that we have to bring this to an end, but we are a committee that is discussing the subject, and we are going to draw up conclusions, as you would expect. We want to try and have an impact. What recommendations do you think we should make? Of course, you have interspersed in all your comments so far various things which you think we should do. What I am really asking you to do is to pull them together. We can go through our notes and the record and draw them out, but it would be better if you tried to summarise yourself what you thought were the most important recommendations you would like to make.

Lord O'Donnell: Curiously enough, I have a list. Number one is to review the government procurement rules; that is a real problem. Number two is the capital markets issue. For me, that is about why are our pensions and insurance companies so risk averse. I do not know the answer to that, but it would be really interesting to ask John Kingman that, particularly.

Lord Drayson: On your point about pay, do you think it applies to politicians as well as civil servants?

Lord O'Donnell: Most certainly. One of the aspects of Singapore is, as I think I said, that pay for Ministers and civil servants is much higher and is tied to private sector markers. So yes, absolutely.

Lord Drayson: In terms of the balance of the capacity within Government, you mentioned that we are lacking a vision. Do you think we need more capacity in No. 10, given the power that No. 10 has over the vision?

Lord O'Donnell: Yes, unambiguously. If you look at their capacity on the economy, it is very limited. So, yes, there is no question.

The Chair: Lord O’Donnell, I am sorry—you had a very important little list, so please continue.

Lord O'Donnell: Number three is handling failure. This comes back to the way the NAO and Parliament treat things. You start hearing of people wanting to know about all the successes there were in Government, what are the lessons that come from those, and why did somebody not fail anywhere, because you are supposed to do test and learn.

Number four would be to get the expertise in. Government needs to welcome STEM people more as well. There is a real question about this. I love talking to government scientists. I am hoping to meet the chief scientist very soon about children’s well-being. We actually talk the same language, because we all talk about evidence and data. We are analytical geeks. There should be a much closer relationship and there is not, but we both use the same scientific method, and we are both interested in evidence, not stories, so we should be in the same place.

My next one would be more global learning. For example, there is all this stuff about getting the long-term unemployed back into work, which is massively important, because you can do all sorts of things if you start addressing that. I was at a behavioural science conference yesterday at the LSE, and Denmark and the Netherlands have some amazing ideas about this. They have some amazing ideas in the Netherlands about how they do prisons. We need that global outlook, but we can be very insular when we look at what are the right answers to this.

Then, there is that question about—

Baroness Neville-Jones: We do not speak any languages, either.

Lord O'Donnell: Exactly. There is the question of whether there is a problem about the way we are regionally so biased towards London and the south-east, and all of the start-ups and all of the rest of it. There can be great ideas going on in our universities around the north, the Midlands, or wherever. If you talk to Jim O’Neill, he is very good on this. Sometimes, they are just not picked up, and the finance guys are all in London or Silicon Valley, so they are not getting through to them.

Those, for me, would be the biggest things you could do. Obviously, the pay and pensions on the expertise would be massive, as would finding ways to get foreign companies to persuade our pensions and insurance industry to be a bit more adventurous.

The Chair: Thank you very much. Ending on that excellent list concludes a really terrific session. We are very grateful that you came and shared all of your expertise; thank you so much.