CORRECTED TRANSCRIPT OF ORAL EVIDENCE To be published as HC 1634-i

HOUSE OF COMMONS

ORAL EVIDENCE

TAKEN BEFORE THE

PUBLIC ADMINISTRATION COMMITTEE

PRE-APPOINTMENT HEARING FOR THE POST OF THE CHAIR OF THE UK STATISTICS AUTHORITY

TUESDAY 6 DECEMBER 2011

ANDREW DILNOT CBE

Evidence heard in Public

Questions 1 - 62

USE OF THE TRANSCRIPT

1.    

This is a corrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.

2.

The transcript is an approved formal record of these proceedings. It will be printed in due course.

Oral Evidence

Taken before the Public Administration Committee

on Tuesday 6 December 2011

Members present:

Mr Bernard Jenkin (Chair)

Alun Cairns

Paul Flynn

Robert Halfon

Priti Patel

Examination of Witness

Witness: Andrew Dilnot CBE gave evidence.

Q1 Chair: May I welcome you to this pre-appointment hearing for the office of Chair of the UK Statistics Authority? Could you, at the outset, please identify yourself for the record?

Andrew Dilnot: I am Andrew Dilnot.

Q2 Chair: Thank you very much. I think that you wanted to say a few words to open.

Andrew Dilnot: Yes. I was asked at the panel interview for this job to speak for five minutes about why I wanted the job, what I thought the challenges were and how I would address them, so I thought that it might help if I were to run briefly through that statement.

The simple answer to the question, "Why do I want the job?", is because I think that statistics is the most important thing that we face. There is no decision that any of us can take that is not, in the end, based on statistics. "Who to vote for?" is a pretty important decision. If you are running the Bank of England’s Monetary Policy Committee; if you are the Office for Budget Responsibility; if you have responsibility for climate change, poverty alleviation, transport policy or pensions policy; or if as an individual you are trying to work out where to live, or if as a business woman you are trying to work out what business you might want to run and where you should site it, you are dependent on statistics. We can’t do anything if we don’t know what the world looks like, so statistics are unbelievably important.

I have spent the whole of my career-31 years now-working in this area, trying to get, use, analyse and communicate data. I spent 21 years at the IFS and I have studied 23 consecutive Budgets and five general elections, all trying to work with what was at that stage a particular set of economic and social data. I have spent time in Oxford; I have worked on radio programmes for the BBC about numbers and statistics; I have worked as Chairman of the Statistics Users Forum; and recently, in the last year, I have worked on the Commission on the Funding of Care and Support, the commission looking at long-term care. All of those activities have been trying to get the very best out of data, and in all of those activities independence and integrity are vital. If we can’t trust the numbers, then we’re nowhere, so I’ve tried very hard throughout that time to stand up for that independence.

I love this material. I care about it deeply, and not just the numbers, but the way in which they are used and interpreted. The science of statistics is really important to me. We have in this country huge and marvellous richness of data. There is an astonishing array of data available, but by and large it is not used as well as it could be, inside Government or outside Government. We have a pretty good Act-an Act that I think we can be pretty proud of-but we need that, because trust in statistics is so low. So I think that there is a huge scope to do good in an area that I care deeply about.

What are the challenges? I think that there are three. One is to do with trust, user engagement and communication, and we have a long, long way to go on all of those issues. The second is making sure that we actually have the impact that we ought to in the development and evaluation of policy. And the third area of challenge is how we manage in an era where public spending is quite constrained, where there is growing complexity in the world and where there is technological change that we need to manage and use effectively.

How would I do that? To begin with, I think that I would get the whole team of the UK Statistics Authority board, the national statistician, the head of assessment, the Director General of the Office for National Statistics and the whole of the staff of the UK Statistics Authority and the ONS to feel that we have a shared vision, where we can be confident and ambitious.

Then I think that we need to reach out to the public, to Parliament, to Ministers, to the private sector, to the civil service, to the Bank, to the media and-a sector that is a particular passion of mine-to schools and universities. We desperately need to get young people more engaged with statistics than they have been.

I would want to be a guardian. I think that one of the things that has been really important about the first few years of the UK Statistics Authority is the way that it has established its credibility and reputation, so that if something is going wrong the authority says something about it. That is something that I would want to continue, identifying bad practice and identifying changes that are needed, and trying to create a culture where the expectation is that if-as a Minister or anybody else-you do something statistically wrong, you will be shown to have done something wrong. So let’s create a culture where people do the right thing first, because they know that if they don’t do the right thing somebody’s going to be coming down on them.

The second thing is that I would like to be a champion. I want us to start thinking about statistics positively and see the positive value of them: the excitement, the power that they create, and the sense of surprise and delight. Above all, I’d love us to think of statistics not by saying, "Statistics", but by saying, "Statistics-what fun, what enlightenment, what excitement, what surprise and what power there is there."

Thank you very much.

Q3 Chair: Thank you for that. So, as Chair of the authority, what is your most important role?

Andrew Dilnot: What is my most important role? I think that it is creating a sense of the significance of statistics and the consequences of their misuse. You may not like this answer; you might have wanted me to choose either champion or guardian, but I think the two are intimately linked. You can be an effective guardian only if you have made people realise how important statistics are. It is about helping people to understand their significance, the necessity for good statistical analysis and the consequences of going wrong.

Q4 Chair: How do you plan to maintain the independence of the UK stats authority?

Andrew Dilnot: It seems to me that the stats authority has made a marvellous start in the first four years. It has established a very strong sense of independence. It has done that through making very good judgments in a politically sensitive area, and I want to go on doing that-to make appropriate judgments that are entirely impartial and entirely outside the political realm, but that are none the less firm. In the many years I spent at the IFS, I had a great deal of experience of doing that with Government parties of both major political persuasions. It is a matter of seeking very seriously engagement with the statistics and with what the statistical issues are, of making a careful assessment of what has and has not been said, and, where something has been done inappropriately, of saying so very loudly.

Q5 Chair: I can attest that your predecessor-the current holder of the office-is regarded in Whitehall with a mixture of fear and outrage because of some of his interventions. Under what circumstances would you intervene on Ministers and others, and what would be the manner of those interventions?

Andrew Dilnot: I would seek to intervene if it seemed to me that the code of practice that has been agreed and published was materially abused-if something had been done with statistics that did not fit in with the code of practice and that seriously misled. As to the range of ways in which that has been done so far, Sir Michael has largely intervened through writing letters when things have got to the point where he has felt the need to intervene publicly, and that seems appropriate. It is hard to imagine the particular circumstances that will come up in the future. I have looked back at the interventions that the authority has made over the last three or four years. Although I have not, of course, had all the detail for all of those, it has not seemed to me that any of those have been inappropriate; they have been brave and impartial, and, as far as I am aware, there has been no significant comeback from any of those criticised that might undermine that judgment. I would hope to carry on that record of extremely effective, impartial, professionally-based judgment about what it is and is not appropriate to say.

Q6 Chair: How do you think you can assist Parliament in the scrutiny of statistics?

Andrew Dilnot: Perhaps the most important thing about this Act is that the Statistics Authority is answerable to Parliament, largely through this Committee, although through other Committees as well. Building strong relationships with Parliament is important. Building a strong relationship with this Committee is something I would very much hope to do if I were appointed to this post. Some of that relationship can be built through meetings like this; some of it, I would have thought, needs to be maintained and enhanced by regular reporting directly to this Committee of any issues that come up. I imagine that is done already, but I certainly would have thought that if something that looked like a breach of the code occurred, and the authority felt that was the case, it should be communicated to this Committee. I would be interested in the possibility of more regular, perhaps informal, meetings with some members-either the Chair or other members of the Committee. Relationships are very important here, and a relationship of trust needs to be built up. I am sure it exists already.

Q7 Chair: So you think the Committee itself could be seen to be more proactive in this role?

Andrew Dilnot: I am not sure that it would be fair to ask the Committee to be proactive-I imagine you have much to do already. It is about making sure the Committee is absolutely fully briefed at all times, and perhaps about raising not simply issues that come up when there has been a clear breach, but more strategic issues about the way in which we might like to move, the whole direction of statistical work in the UK and things we might like to see done. For example, there is the way analysis and data are considered in the civil service. These are important issues, which go beyond the crucial but quite narrow matter of identifying breaches of the code.

Q8 Alun Cairns: Would you use the Committee-should you be successful, obviously-where you thought a Department, a Minister or someone else might be misinterpreting statistics for their own advantage?

Andrew Dilnot: I would certainly hope-I hesitate to say "use"-to work with the Committee on a wide range of matters. My guess is that it would rarely be the case that, where a particular breach had occurred, there would be time to swing the machinery of the Committee into action. I think it more likely that it might be interesting to bring to the Committee wider ranges of issues. I am sure we will have the chance to talk about pre-release access later on. That might be one such example.

I also have a wider concern about the role of analysis and data within the machinery of Government. That is the kind of issue where, if there was work being done in the authority, it might be interesting to have a discussion with the Committee about what the appropriate way of bringing that more to Parliament’s attention might be.

Q9 Chair: You are very reliant on ONS and UKSA staff for advice, but will you be using other sources of information?

Andrew Dilnot: I will certainly be reliant on ONS and UKSA staff for advice. My impression is that those staff are of very high quality. I have myself a long-running record of interest in these matters and will want to get my hands very dirty indeed. I will be interested in the possibility of engaging the UK academic scene as well. That may be being done already, but the UK has some outstanding academic statisticians and I think that there may well be a role for getting advice and involvement from groups such as that. I have a very high view of statistics. I think that it is an extremely important profession and I want to get input from all parts of that profession, not just from the civil service.

Q10 Chair: Finally in this section, do you have an idea of what you want to achieve in your first year and how you will measure success?

Andrew Dilnot: There are some obvious prerequisites, such as ensuring that the independence of the authority is maintained and ensuring that, if any breaches occur, they are responded to appropriately, but, most of all, what I would love is a greater sense of confidence within the authority and the statistical service and a greater sense among the whole population-both private sector and public sector, and both individuals and institutions-of the value of statistics and the need to engage with what the data tell us, so that we have discussions that are based not on knee-jerks, but on the vast amount of data that are out there. I would love there to be more active use of the astonishing array of statistics that there is. That will be a good measure of success over the first, second, third and fourth years. We ought to be able to make progress pretty fast.

Chair: Being a statistician, I am sure you would find some means of measuring it.

Q11 Alun Cairns: I should say, as a Swansea boy, it is nice to welcome another Swansea boy to the Committee.

Can I press you on time commitment? You will be familiar with the reduction in commitment from three days to two days a week. Can you tell us whether you think that will be sufficient to conduct the tasks and keep the enthusiasm that you have talked about?

Andrew Dilnot: Of course, the post this time around has been advertised as two and a half days, not two days. It went from three days to two days and now it is back at two and a half days. There is a lot to be done, but I am confident that it can be done in two and a half days or roughly that. I am sure that there will be weeks when much more than that is necessary. If a particular issue blows up, there will be times when more time might be necessary, but, on the whole, my view is that two and a half days will be enough to get this done.

If the Queen chooses to appoint me on the advice of the Committee and Parliament, I will be resigning from almost everything else that I do. I plan to move from St Hugh’s College to Nuffield College next August, but I think that I will be resigning from all my other commitments. I will be resigning as a pro-vice-chancellor of the university and I will be resigning as Chairman of the Statistics Users Forum. I will be resigning as a trustee of the Nuffield Foundation and I will be resigning from any other smaller committees that I sit on. I am confident that I will have the space, and I am also confident, which I think is more your question, that two and a half days a week is enough, on average, over a year, to get this moving.

Q12 Chair: Can I just press you a little more on that? Your predecessor does more than two and a half days a week and has always argued that the time commitment needs to be more than two and a half days a week. If you find the same, would you be able to make that extra commitment?

Andrew Dilnot: Yes, but I don’t to expect to find the same. I have talked extensively with Sir Michael and I have looked at the job description. It is a non-executive chairman’s role, so my own view is that it should be possible to do it in that. But yes, I have done many jobs now and a similar question might have been asked, for example, about the care commission over the last year. Sometimes you need to put some more in and you do it, and I would do it.

Q13 Chair: We have consistently reserved our position on whether two and a half days a week is sufficient.

Andrew Dilnot: Sure.

Q14 Chair: I imagine we are minded to reserve our position until you have proved that that is the case.

Andrew Dilnot: Yes. And that would seem to me entirely right. The job was advertised at two and a half days a week so that is the post to which I have applied. But if I am appointed, which I very much hope I will be, then I will make this work.

Q15 Alun Cairns: Can you envisage a situation in which two and a half days is not sufficient and what action would you then take?

Andrew Dilnot: Well, I can certainly envisage a situation where in a particular week or several weeks a set of issues comes up which means that two and a half days a week is not sufficient. Of course it is possible that on average two and a half days a week is not sufficient. That is not my expectation and that is not the job for which I have applied. If that turned out to be the case then I would have to re-evaluate what to do about it.

Q16 Alun Cairns: Do you envisage taking on any additional roles? You talk about resignations from many of the roles that you currently hold. Do you envisage taking on any other role to go with it?

Andrew Dilnot: No. One of the entirely peripheral benefits of this from my perspective is that I think it would give me the absolutely perfect excuse to say no to all of the other things that I was asked to do. So, no, if I was to take this on then I will, of course, maintain my commitment to Nuffield College, to which I go in August of next year, but that is the only other post that I would have.

Q17 Alun Cairns: Do you see a risk of a conflict of interest between the roles that you currently hold and the time that they might wind up, or even some of the work that you have done within those roles?

Andrew Dilnot: I am always hesitant to say an absolute no, but I cannot see any. I see no conflict of interest between any of my past roles and the role of the Chairman of the UK Statistics Authority. In all of the roles that I have taken, I think part of the duty has been to be aggressively independent. The 21 years I spent at IFS, from 1980 to 2002, was the period during which the IFS’s reputation particularly developed and the reputation that the IFS sought to develop was precisely one of absolutely rigorous impartiality and independence, complete abstraction from the political fray while being entirely engaged with it. The commission on the funding of long-term care, which I have been involved in over the last 18 months or so, is also very explicitly independent of the political debate. So, no, I don’t see any conflict between anything I have done in the past. The role as Warden of Nuffield, absolutely not. It seems to me that Nuffield College exists to champion the social sciences-statistical analysis is crucial across all of the social sciences-but it does that without fear or favour in an entirely independent way. So I don’t think there is any conflict of interest and I am confident that the potential of conflict for time is something that I can manage.

Q18 Alun Cairns: Can I pursue the line about conflicts? UKSA is both a producer and a regulator of statistics. Do you see any risks for conflict within its role and how will you best balance your priorities and responsibility between the two roles?

Andrew Dilnot: I think this is an important point. It is quite an unusual set up for UKSA to be, as you say, both producer and regulator. The way that UKSA has managed that in its first four years has been to achieve, it seems to me, just about as much separation as you can by having two deputy chairmen, the ONS board somewhat separate from the authority itself, and a member of the authority board being the chairman of the ONS board. My impression and the view of those that I have spoken to about this is that that has worked pretty well. But you are right to point out that there is a potential conflict facing all members of the authority in their capacity as both regulators and producers. Now we can see in the actions of the UKSA that there have been occasions when the authority has criticised the ONS as producer. But the fact of criticism does not demonstrate that there is no conflict. If I am appointed, one of the things that I shall certainly want to do in my first few months is talk more with my colleagues on the authority and on the ONS to make sure that we have a line arrangement that works, because it is an unusual set-up and it merits further analysis.

Q19 Chair: Are there Chinese walls? Do you think there should be more physical separation? Is it possible physically to separate the roles?

Andrew Dilnot: The honest answer is I don’t know. It is possible that physical separation would be valuable and that is something to be looked at. I don’t know. It is something that I would want to work on once I were part of the organisation and able to discuss it.

Q20 Chair: Would it be fair to say that this an issue of concern to you?

Andrew Dilnot: It is an issue of potential concern. I have seen no evidence that it is causing a problem, but it seems that a reasonable person could, in principle, say, "Well, that is a bit unusual," as the authority has already said. It is in here-it is in the Act. The authority has come up with its best attempt at making sure that the arrangements work and it is confident that they have done so. I have seen no evidence that they do not, but it is a sufficiently unusual arrangement to merit further study.

Q21 Alun Cairns: Can I pursue the sources of statistics-the ONS, the Government Statistical Service and the various departments that throw out statistics? Have you assessed any strengths or weaknesses between the various sources? Is there anything that you might want to develop further in the role?

Andrew Dilnot: A very large number of assessments have been published by the authority since it got under way-we are well beyond the halfway point. By the middle of next year, I think just about every single national statistic will have been assessed.

My reading of the assessments that have been published so far is consistent with some of my greatest enthusiasms. The one consistent relative failing is in communication and explanation. That seems to cut right across from the ONS into the Government Statistical Service and it runs across the gamut of statistics. As I have said, my feeling is that the statistics available in this country are a great delight. If you know what you are looking for and if you can find your way around the ONS website, there is almost nothing that you cannot find out about, but for many people, those two ifs are quite a big obstacle. I would love to see a transformation as soon as possible in the way in which we engage with users and in the communication of our statistical material. That is true as much for the GSS as for the ONS.

The Statistics Authority is already working on this-there is a great deal of work on user engagement-but there is recognition that it is one of the really important next steps. We are hiding our light somewhat under a bushel and that is something I want to get some progress on.

Alun Cairns: Thank you.

Q22 Paul Flynn: When the Statistics and Registration Service Act 2007 went through this House, there was hardly a flicker of interest here and even less in the press, except for one article that I read and repeatedly quoted, which was written by you. It expressed what we have heard today-this wonderful enthusiasm for statistics, which you regard as a turn-on and sexy. I am sure that my many constituents who watch these sessions live will be applauding now and there will be gales of approval coming up the M4 because of the enthusiasm you have shown towards statistics. In that article, which I quoted repeatedly to the intense boredom of everyone else, you said that the Act was-I don’t remember the exact words-one of the most important that was likely to be introduced by the Labour Government. In their 13 years of office, this was the big Act. Do you still believe that, and was your enthusiasm justified by subsequent events?

Andrew Dilnot: I think I do still believe that, because it was a turning point. It has made possible something that otherwise was not possible, which is the recovery of a sense of independence and integrity for statistics. It seems that in the first few years of the authority, great progress has been made. There were a number of sceptics four years ago who were not sure whether the chairman of the authority would be able to do the job effectively. It seems that we have heard almost nothing but praise for the authority’s work so far. We have an extremely strong base from which to build. We have a long way further to go and we will come later to some problems that still exist in the Act. Yes, I do think that it is worth holding this up next to the independence of the Bank of England. The independence of the Bank of England gave a particular role in the setting of certain types of economic policy to something other than political control. This Act gives the control and management of statistics back to Parliament, taking it out of political hands. I think that was a very, very important step.

Of course, there has not been as much progress as I would have liked so far. The area where there has been least progress has been in communication, but even there I think we are making progress. If we look, for example, to the BBC website, these days you see some really marvellous representations of data. Just this week, there is an extremely good set of stuff on road accidents, which allows you to zoom in on interactive maps, see where they have occurred and what the changes have been.

There is potential for doing some really important stuff, and I think that that can transform our democracy. I really believe that if people are to understand whom to vote for and what is going on, they need to know what is happening, and by and large, they do not. It is worth saying that by and large it is not even the case that parliamentarians always know-they do not know what they have done. They do not know the data. How can we govern, if we do not know?

Q23 Paul Flynn: Because, I’m afraid, government is based on prejudice and not on reason or evidence.

Andrew Dilnot: Well, that needs to come to a stop.

Q24 Paul Flynn: In fact, it is going backwards. On e-petitions, we have handed over the power to the tabloids to make decisions, but I share your enthusiasm.

It might have made sense in biblical times to count the entire population, but is it not a wasteful nonsense now to take a sample of the entire country, instead of getting the same information from a 1,000th, 10,000th or 100,000th of that sample? Would you regard it as a failure if you do not get rid of the 10-yearly census and replace it with something likely to be more accurate and would not introduce, as the census does, its own distortions, when the forces of the Jedi knights and Pastafarianism decide to register some point by describing themselves as Jedi knights or Pastafarians? The whole thing is a nonsense. It should have stopped. It is a waste. It is there because of inertia, surely. You will stop it, won’t you?

Andrew Dilnot: There is an extremely important process called, "Beyond 2011", which the UK Statistics Authority is already engaged in. It will report back by 2014. My view is that it would be disappointing if in 2021 there were not a way of getting hold of the data that we crucially need. We absolutely, crucially need good estimates of the size of the population and its distribution across the country and by various types, so that we can allocate funds appropriately. Yes, I, with you, would be disappointed if by 2021 there were not a more effective and probably cheaper way of getting the data than simply going out and counting everybody. It is worth remembering that counting everybody turns out to be quite tricky.

There is some lovely evidence from the US census, which in 1970 suggested that there were slightly more than 100,000 people aged over 100 in the US, whereas the U.S Census Bureau thought that the answer was actually 3,000. Ten years later, the census itself produced an answer of 30,000, whereas the U.S Census Bureau thought the number was probably nearer 5,000. Just asking people does not necessarily give you the right answer and you cannot ask everybody unless you know where they are.

My view is that we desperately need a proper address register. I know that work on that is going on at the moment, supported by the Government, which I welcome. My expectation would be that once a proper address register is in place, it should be perfectly feasible to combine that with various forms of administrative data to get us timely and entirely satisfactory estimates of the size of the population and its distribution. Yes, I would be very sorry if we had not found a better way by 2021 of finding out the information we need.

Q25 Robert Halfon: Do you agree that it should be compulsory? Is it wrong that it is compulsory?

Andrew Dilnot: I certainly think that it is important that we are able to count everybody. Whether we should compel people to fill in forms, depends on what is being asked. We need to be able to compel people to know whether people are here or not. Once you go beyond that, you get into trickier issues about personal freedom, and those are things that-

Q26 Robert Halfon: But the compulsion element is very authoritarian, and given the length of the form and its detail, it seems very un-British-if you understand where I am coming from-to force people to fill out lengthy forms for no, or very little, personal benefit.

Andrew Dilnot: Yes, because it is not no personal benefit. In the end, we all benefit from funds being well allocated. I am afraid to confess that in my household there was a fight over who got to fill in the census form, but I do recognise that a long form might seem oppressive, particularly if there were any sense that the information could be got in another way. Part of the point of Beyond 2011 is to say, "Of course there is some information that we need to have. Let’s be clear about what that is, and then let’s find the most efficient way of doing it." I do not image that our successors will be sitting here in eight years’ time having a conversation based around something like the old-fashioned census.

Q27 Robert Halfon: So you think it could move to there being more of a sample element, as with opinion polling and how the BBC judges viewing figures?

Andrew Dilnot: I think it can. It is complicated. We absolutely have to have an address register, because you cannot have a sample unless you know what it is you are taking a sample from, and at the moment that is one of the things we get from the census. Then there is a wide variety of administrative data, including school and GP data, which ought to cover a lot of areas, but we need to make sure, for example, that we capture the institutional population correctly, and we need to do all we can to capture marginal bits of the distribution. My sense is that by targeted specialist investigation we can probably do better than we can with just a census, and that seems to be the consensus.

Q28 Paul Flynn: I do not know if you have been following the situation between Sir Michael Scholar and Boris Johnson. There is some dispute, and Sir Michael Scholar has been sending letters to Boris Johnson, whose response has been to describe Sir Michael as a socialist or left-wing, or something along those lines. How would you deal with that situation, if you had a response from Boris?

Andrew Dilnot: I think the phrase used was "Labour stooge".

Paul Flynn: "Labour stooge", yes; that is right.

Andrew Dilnot: I was subjected to that kind of experience frequently, from all sides, when I was the director of the IFS. I have been accused of being clearly left-wing, clearly right-wing and also clearly a member of the Liberal Democrats. You make very sure that you are right. If you are wrong, you have to say, "Actually, we were wrong," and if you are right you do not engage with the political discussion; you simply say, "These are the data." I have had that even in this context.

Q29 Paul Flynn: Knife crime is an example; when knife crime falls, the perception of the Daily Mail is that knife crime is going up-exactly the opposite. There is no truth behind it, and no rational basis for it. There are other subjects on which public prejudice points in one direction and the facts point in the other. How far do you go in standing up for truth and justice against the almighty voice of the Daily Mail?

Andrew Dilnot: I think you always stand up for truth. In the end, truth and a reputation for independence are very powerful, but you cannot create them overnight. Certainly my experience at the IFS was that it takes many years of being shown to be independent, impartial, outside the political debate and simply concerned with professionalism, and I think that the first four years of the UK Statistics Authority have been so impressive because that reputation has been built rather quickly. You just go on saying, "This is what it is." What Sir Michael and his colleagues have done for this institution is that they have created that reputation already, so if the current chairman of the UK Statistics Authority says, "Actually, not so," that attracts a fair amount of attention and clearly is something that is uncomfortable for those thus told off.

Paul Flynn: Thank you.

Q30 Robert Halfon: Can we turn to the website issue? I was very excited by what you said in your opening remarks, and in some of your replies to Mr Flynn, about how you see statistics. I admit that I am a simple person-I am not very numerate-but I have been trying to work out how to use your ONS website while you have been speaking, on this BlackBerry tablet. I have had about eight clicks and have not been able to find any statistics yet. Why is it so user-unfriendly, and what are you going to do about that?

Andrew Dilnot: Why is it so user-unfriendly? There is a truly vast amount of data on it, and my perception is that the particular difficulty at the moment is the search function.

Q31 Robert Halfon: It is almost as bad as the House of Commons website, and that’s saying something.

Andrew Dilnot: What you and I want to do is type in "unemployment" on the front page and we want to go immediately to a chart that shows us what has happened to the core measure of unemployment over the last 15 years, and which then allows us, if we want more detail, to click on it and go to youth unemployment, male unemployment, female unemployment, and be able to download the data.

Q32 Robert Halfon: And local.

Andrew Dilnot: And local, and all those neighbourhood statistics. At the moment, you type in "unemployment" and you get thousands of options. There is a desperate need to get that sorted out. The ONS has a web recovery programme in place. They have allocated some of their most senior staff to it and they are getting on with it. My own view is that I want to see this given an even higher priority. My feeling is that too much weight is given to the needs of official users.

Official users do not mind so much about the website, because, by and large, official users in Departments know that there are particular data sets that they want. They can get to them because they know exactly where they are. The people I care about even more are the citizen users. At the moment, the citizen user is essentially held away from all this data by the difficulty of access, so one of the things I will argue for very strongly is a new prioritisation for all forms of communication.

Q33 Robert Halfon: Why not go one stage further and democratise the statistics service and turn it into an open source? Obviously, you will have your links to the official stuff, but let people comment and provide their own statistics. Citizens who are in the know and those who are not could actually comment on what they think of the statistics and offer alternative models. At the moment, there is no interaction at all between the individual and this very bland, boring website.

Andrew Dilnot: I think more democratisation is a very good thing. As chairman of the statistics user forum for the last couple of years, that is the kind of thing we have been trying to get going. It is not terribly easy, but it is a good thing to do.

I think the Open Government website is another good initiative. That is data, and I am all for as much data as possible. Information is always good, as is interaction and comment on it. It is something different. The ONS website itself is trying to present statistics with some interpretation. We need both those things. In the Open Government website, we need as much data as we can get out there. Let us get it out there and let people talk about it, think about it and analyse it.

We also need for the mass citizen user, the 60 million of us, something where they can go and find out the numbers. We are collecting all these numbers. This stuff should be available to them in ways that are interpretable.

Q34 Robert Halfon: What I am trying to say is that your website should be more Wikipedia rather than encyclopaedia. At the moment, it is not even an encyclopaedia. It should be really interactive. Although you are the experts, you are not the all-seeing truth, so therefore people outside may have alternative models. Yours will still be the official statistics, but let us have real interaction. I do not just mean a better website that is easier to use, but offering people a real say in what they think you are saying, and having blogs, twitters and Facebooks, and being really interactive and saying, ""What are your thoughts on our statistical announcement we made today? Please put them here."

Andrew Dilnot: I am all in favour of that. At the statistics users forum and at the Royal Statistical Society, we are about to put up in the new year a thing called StatsUserNet, which will try to encourage exactly that kind of thing for people to interact, have discussions and all that kind of stuff. I am a bit agnostic about whether that is the right location for it or whether it should be independent like the Royal Statistical Society or as part of the ONS website. I would like to talk about that some more. Perhaps we might even meet up about that.

Q35 Robert Halfon: For example, there could be a blog on there, and people could at least comment on the blogs.

Andrew Dilnot: That is something I would just like a bit of notice on, because one of the things about an official site is that there are all kinds of moderation and things that need to be thought about quite carefully. I am certainly not opposed to it, but it is something that I would like to think about some more, and maybe we could even talk about it.

Q36 Robert Halfon: Will you give us a guarantee that in, say, a year’s time, we will at least have a better website?

Andrew Dilnot: Yes-if I am appointed. If I am not appointed, that would elude my grasp. If I were to be appointed and the website was not better, I would certainly have failed. We must make an improvement. We really must, not least because the stuff is so brilliant. If you know how to get around it, there is no subject of public policy where in two or three hours you could not produce a really compelling presentation that shows what has happened over the last 15 or 20 years and what the big issues are. It was just inaccessible to the population as a whole.

Q37 Priti Patel: In light of your desire to have more engagement with the public on all matters statistics and the piece around education, everything that we have seen thus far indicates that public trust in official statistics is quite low. What is your own assessment of that? You touched on your vision in your opening statement, but how far are you prepared to go, in terms of active engagement? I would not mind hearing a bit more about your vision for schools, in particular.

Andrew Dilnot: Trust is low, and my own view is that it is unfairly low. There is a common perception that statistics are manipulated either by civil servants or politicians. I have worked in this area for 30 years, and I have never seen evidence of the statistics themselves being manipulated. There is common evidence about publication times and the selective release of stuff, but the idea that the numbers themselves might be mucked around with I have not seen evidence of. However, because of the way in which we do these things-because of the issues around publication, which I am sure we are going to come to-there is that perception that it is controlled. The only way we can improve that is by demonstrating that it is not thus, which is why the guardian role is so important.

How might we engage? In particular, let us talk about youngsters. One of the things that made me most unhappy in the past couple of years was when, a couple of years ago, a friend of one of my teenage daughters, who was doing economics A-level, had an essay to do about unemployment, and he said, "Could I talk to you about that?" I said, "Of course; I would love to talk to you about it, but"-he was in our house-"just go and get the unemployment data, print them off, have a look at them and then we’ll talk about them." He came back an hour later and, rather as with Mr Halfon, he could not find them. I thought, "No, no!" Here was somebody who wanted to go and engage with them, and he could not.

You may have heard of the "Target Two Point Zero" competition that the Bank of England runs with schools, where they encourage sixth-formers to form pretend versions of the Monetary Policy Committee. They then compete against one another in a national competition, and the winners come to meet the Monetary Policy Committee. I would love to have a similar competition for schools that was just called "Tell Me a Story", where we said to groups of schoolchildren, "Go to the Office for National Statistics website, go to the Government Statistical Service website, and tell us a story about any aspect of the country in which you live, using statistics. You’ll compete against one another, and we will bring you together at the end of the year." We would see if we could persuade the BBC to give us 10 minutes on the "Today" programme for the winner-just to start people realising how much you can do, if you actually go and start looking at the world that we live in. That is the kind of thing that I would love to do. Whether we can do that, I have no idea, but that is the sort of thing that I would love to get going, to use this rich treasure that we have.

Q38 Priti Patel: What about wider public engagement? You have referred to that several times. There are normal citizens who probably feel quite cynical about statistics and-dare I say it-even the way that politicians and newspapers present statistics, and incorrect statistics as well.

Andrew Dilnot: They are right to feel cynical about the way newspapers and, sometimes, politicians use statistics. I have in my bag my favourite newspaper of the last two years-I shall not name the newspaper. Less than two years ago, one of our big broadsheet newspapers had a great big headline-a 36-point font headline-"Public pensions to cost you £4,000 a year". They had divided £9.4 billion by 26 million and got an answer of nearly £4,000. The answer is nearly £400, but it was the absolute banner headline on a broadsheet quality newspaper, and goodness knows how many people that has to go through. That shows you what is wrong with our statistics.

How do we get better public engagement? The way in which most of us, most of the time, engage with statistics is through the media. That is the way in which most of us consume most of the statistics we consume, and we are consuming statistics every day as readers of newspapers, listeners to the radio and watchers of the television news. You cannot not be a consumer of statistics-that is what we all are. There is work to be done training journalists and engaging more with the media. That is something that the Statistics Authority and the GSS are already doing. I think that there is much more to be done there.

I think that it would be possible for statisticians to be better at engaging with the media-to be a bit more adventurous, to be a bit more bold-and then we could have people consuming statistics without realising it. Often, when you see a really good piece of journalism on the television or in a newspaper, you are consuming statistics without realising it, because the statistics are integrated into the story rather than being used rhetorically, as they so often are.

We need to go on working-training journalists, training politicians and training civil servants. I continue to feel that, at the highest levels of the civil service, analysis is not given the significance that it should be. All too often, statistics are what you put in after you have worked out what the policy should be. That is something we need to work really hard at with the Cabinet Secretary and the Permanent Secretaries. It is a very, very serious matter. There is a sense that statistics are what come afterwards, and that is quite wrong. How can you possibly make policy if you do not know what it is you are making policy on? I would like to work with Jeremy Heywood and others to try to boost the sense of confidence of analysts and statisticians within the civil service.

Q39 Chair: I confess that I have drafted a speech and then asked for the statistics afterwards. I then found that I could not say what I had been intending to say, which is very instructive. May we go on to pre-release? Do you think pre-release is necessary for good government?

Andrew Dilnot: No. My personal opinion is that I would like to see all pre-release access removed. A reasonable person could reasonably believe that some limited pre-release of, for example, data that might move markets might be appropriate. It might be sensible for data that was likely to move markets to be available to the Bank of England and some Treasury Ministers for an hour or two or three before release. But I simply cannot understand why in the case of, for example, reoffending statistics, pre-release access to the last adult reconvictions release was granted to 23 people. I just do not understand how a reasonable person could reasonably say that that was necessary.

My view is that the view that has traditionally been taken by political parties when in opposition, which was admirably articulated by Mr Vince Cable in the run-up to the last election, is the right one: pre-release is bad. When you read through the Statistics and Registration Service Act 2007, you read a great deal of material that is extremely strong about independence and the significance of independence. Then you get to section 11-"Pre-release access"-and it is a little bit like reading Augustine’s confessions, "Make me chaste, but not yet." We are in favour of independence, independence, independence, except in the area where we think it might be awkward for us.

I do not think that pre-release is a good idea. I can see that some people might reasonably think that some limited pre-release might be appropriate, but that should be much more limited than what we have at the moment. My view is that some of the arrangements we have at the moment are part of bringing the game into disrepute. One way in which the Government could help the authority and help themselves is by being brave enough to say, "Actually, we can see that the authority is trustworthy. We are trusting the authority to champion and guard our statistics. We will trust it, under the tutelage of the Public Administration Select Committee and Parliament, to act sensibly as far as pre-release is concerned."

Q40 Robert Halfon: What about if it is market sensitive-the argument that you need to give the Government time to prepare because of sensitivities in the market?

Andrew Dilnot: I can see how a reasonable person could reasonably think that. My own experience-perhaps I am speaking from personal bitterness-is that I had to comment on 23 consecutive British Budgets without any pre-release access. I can see that there might be cases where a very limited period of pre-release access might help and I would be perfectly willing to discuss that matter with Treasury Ministers and officials and Bank of England officials. But the idea that 24 hours is necessary seems absurd. Even if you believe that, in the case of market-sensitive issues, we need pre-release, three hours seems to me at the absolute outside of what could possibly be necessary.

Chair: I do not think we need to spend any more time on that.

Q41 Paul Flynn: As a Swansea boy, you must regard the move to Newport as like being upgraded on a plane, especially as you will be under the benign watchfulness of the quality local MP. How much time do you expect to spend in Newport?

Andrew Dilnot: As a Swansea boy, I would have thought of a move to Swansea as being upgraded, Mr Flynn. I do not know how much time I expect to spend in Newport. I certainly expect to be there regularly. I expect to be in London regularly; I expect to be in Titchfield regularly; I expect to visit the devolved Administrations. Until I am doing it, I do not know. I have talked with Michael Scholar and looked at his travel programme. He seems to have been in London significantly more frequently than in Newport and Titchfield; I do not know. One thing that I can say is that I have looked at the way Sir Michael has run his time, and he has not come to London very frequently-a couple of times a month.

My preference would be to come to London on a fixed day every week at least, so that there is one day a week when all of the authority staff who are based in London would know that they could get at me, and then to be in Newport and Titchfield and elsewhere as seems to be appropriate, but I just do not know the answer to that.

Q42 Paul Flynn: The Royal Statistical Society is concerned about the status of statisticians and points out that, 30 years ago, senior statisticians in Government Departments were at director level, whereas none of them are at that level now. Do you think there has been a falling away in the status of statisticians in Government?

Andrew Dilnot: I think there has, but I have to say that I have not been inside Government enough to know that. I am also aware that there are some very senior figures who are not statisticians, but who care passionately about the use of statistics. That is what matters to me most of all. If none of the permanent secretaries was a statistician, but if all of the permanent secretaries really cared about the use of statistics and analysis, I would be content.

It seems to me that in general across the Government Statistical Service and the whole of the Government service analysis and statistics are not given the weight that they should be. I cannot speak with real knowledge about whether things have got significantly worse over the last 30 years, but given that the scope for the use of analysis and statistical data has increased so much, I do regret that we do not see that having a higher profile, and I would dearly love to see it have a higher profile.

Q43 Paul Flynn: In 1988, a group of statisticians came to me to complain that the supervision of their Department in government was moving from one Department to the other; it was from the Cabinet Office to the Treasury. Their complaint was that they were being moved to the office that had the greatest interest in distorting the figures.

I wrote to Mrs Thatcher at the time, who was greatly shocked by this unworthy suggestion that any Government Department would want to distort figures in any way. The statisticians were, quite reasonably, worried and were saying that if the wonderful, pristine figures that they turned out were being turned into garbage by the Government machine, it was lowering their professional status and the whole point of their existence. Who was right: the statisticians or Margaret Thatcher?

Andrew Dilnot: I do not know. I would need to know what the particular issue was.

Q44 Paul Flynn: In general, who should-

Andrew Dilnot: My general view is that our whole political system-civil service and politicians-does not pay adequate attention to the matter of statistics and does not look, as it should, to statistics to help form policy judgments. Part of the responsibility for that must lie with the statistical service itself. The statistical services need to do a better job of presenting material in a way that makes it compelling, so that it will be better used.

Q45 Chair: Can I express some disappointment with Budget Red Books? Twenty years ago, I was able to make some sense of them, but they have become compendiums of adventure and description, and the tables seem to be arbitrary these days. The comparisons with previous Red Books seem impossible to make. Is this something that you are going to be looking at?

Andrew Dilnot: It is something that I certainly have looked at. Indeed, one of the things that the IFS did in the early 1980s was first produce the "Green Budget"-the precursor to the pre-Budget report-and, in January 1982, I can remember sitting over back copies of the then Red Book, which was un-interpretable. It was just numbers, and it was very hard to reconcile them. I thought that in the 1980s and 1990s very considerable progress was made towards documents where it was quite easy to reconcile the various tables. You could actually see the information that you wanted. They related to one another.

Then there was a shift, under both parties of both Governments, towards stuff that tries to put a narrative that fits in with particular political statements being made. We have rather lost the neatness and coherence of the tables. I will not promise that that is something we would look at; I would need to look to make sure it was in our remit. It certainly seems to me that it is an example of how statistics can stop being useful. Certainly, when I look at a Red Book these days, I find it much harder to trace the numbers through.

Q46 Chair: Do you think the narrative, as presented in the Budget Red Book, moves many points on the opinion polls?

Andrew Dilnot: This is not a party political comment, and I will not identify the individuals concerned, but I think the best joke about tax in my 30 years was made the day after a Budget-I will not say which one-when the Leader of the Opposition stood up and challenged the then Chancellor about a particular set of figures. The Chancellor rejected the challenge, and the Leader of the Opposition said, "But it says in the Red Book-or, as it appears to be in your case, the unread book-a, b and c." I do not think many people read the Red Book. In fact, Chairman, I suspect that you and I may be part of a small group that would even be willing to admit ever having done so.

Q47 Alun Cairns: Can I just press the point about consistency, in relation not only to the Red Book but to statistics across the board? If they are to be valid-we go back to your competition in schools, where individuals would want to compare different times in the output of data-do you not accept that there needs to be consistency in terms of the reporting as well, although the measuring might change?

Andrew Dilnot: Consistency, particularly consistency in time series, is absolutely vital. One of the other things that I slightly regret is that our whole statistical culture has moved almost to an obsession with whatever the latest data, released at 9.30 this morning, tell us. Almost all the interesting statistical material is that which looks over a longer time period. Consistent time series are really vital, which is one of the things that we must not lose. We have developed in the appendix tables in the Red Book some marvellous long runs of data on tax as a share of national income, public spending and public sector borrowing. Those are all really valuable, and we need to be able to hold on to them.

Q48 Paul Flynn: Michael Scholar complained about the lack of fast-stream entrants in the civil service who are statisticians. He compared the membership of the Government Statistical Service with that of the Economic Service, and the Statistical Service is very much out-numbered by the Economic Service. That deterioration is apparently continuing. Shouldn’t it be the other way round? Isn’t this something that you should tackle?

Andrew Dilnot: Statistical skills are really important, so it is very important that we have plenty of statisticians, but it is also very important that generalists, economists and others are statistically alert. I almost care more about that. Of course we need lots of statisticians, because statistics is beautiful and powerful. However, we could have lots of great statistics, but if nobody is taking any notice we are wasting our breath. I am as passionate about the need to get non-statistical professionals to recognise the importance of the statistics as I am about just getting more statisticians.

Q49 Paul Flynn: When or if a professional statistician in the Department is being leaned on to break the code in order to serve his Department head rather than to obey the code, what can you do to help?

Andrew Dilnot: If such a person were to write to me or contact me in any way, which I hope she or he would, I would leap very strongly to their defence, because that seems to me entirely wrong and inappropriate. How that should be done, and whether it would be appropriately done through their head of profession, their Permanent Secretary, the Cabinet Secretary or in public, I do not know. That would be a matter for judgment.

Q50 Paul Flynn: Is there anything else you can do to raise the standing of statisticians?

Andrew Dilnot: I hope that we can boost confidence by the array of reaching-out measures that we have described. I hope that we can continue to build the relationship, if I am appointed, with this Committee, to make the Statistical Service feel that it has an unusual privilege of having direct access to Parliament. There are very few other bits of the civil service where you have direct access to Parliament, and that is a pearl of great price. A lot of it, I think, is about building confidence so that people recognise the treasure they have got, and I hope other people will recognise it too.

Chair: We move on to cuts in the Statistical Service.

Q51 Priti Patel: In light of your earlier comment that you felt there was not enough focus from public policy officials and perhaps Ministers as well in Government Departments on statistics and official statistics, do you have any concerns or observations around departmental budget cuts and the impact that will possibly have on the collation of statistical information?

Andrew Dilnot: The public finances do not look pretty. In the 30 years I have been working in this area, I do not remember the public finances ever being this problematic. It would be naïve and rather wrong for the Statistical Service broadly to think that it can escape the consequences of those. It does seem to me that it is very important that, as reductions in spending come through, they are managed effectively and, ideally, that they are managed across the whole piece. I know one of the things that Sir Michael Scholar has emphasised is that he would like to see discussions about cuts in particular statistical areas-particularly areas of the Government Statistical Service and Departments-at least brought to the consideration of the authority.

The authority has conducted a series of expenditure reviews. In some cases, in particular some of the health statistics, the authority has expressed regret that reductions in spending-decided in that case I think by the Department of Health-will lead to the loss of long-running time series on matters related to health. Some of it might even have included some smoking statistics. Those are matters where, in an ideal world, we would have some cross-Government consultation, so that we can optimally choose where to make the savings. There will have to be savings.

This is perhaps a controversial thing to say, but at the margin there are less valuable data series that we collect-I do not want to try to identify one now. There are some data series that are absolutely crucial. There will be a distribution, and there will be some data series that are less valuable than others and, when there is less money around, it may be appropriate to lose them. We just want to ensure that we are losing the least valuable across the whole piece. That would seem to argue for some cross-Government consultation.

Q52 Robert Halfon: Do you envisage having to make redundancies at the Statistics Authority in order to save funds?

Andrew Dilnot: My impression is that total staff numbers have been declining. They will decline anyway because of the census. There is a big jump in numbers every 10 years because of the census. There have been some, though a relatively small number-about 60-compulsory redundancies in the past year. The expectation is that that will tail off. Thereafter, I hope it would not be necessary to make redundancies, but I couldn’t promise that.

Q53 Priti Patel: Do you think there is new potential in this role to get more involved in budget round discussions with each Department around statistical information gathering, if there is going to be a significant hit? To your point about losing valuable information, how bad would it have to be for you to say enough is enough?

Andrew Dilnot: It would be over-ambitious for the Statistics Authority to try to get involved in the budget round negotiations with every Department, but I do not think it would be over-ambitious for the authority to be consulted where a Department was proposing to cut the spending for a particular significant data set. I think I would characterise it in that way.

When would one say enough is enough? In the end, budgets are set by Government. I think my sense would be that if the authority felt that something really damaging was going ahead, that it was hard to justify, then my natural first port of call would probably be to get in touch with the Chairman of this Committee and say that we have gone through this set of processes and discussions, we feel that this is sufficiently worrying that it is important for us to approach Parliament about it, and then seek the advice of the Chairman and the Committee.

Q54 Chair: Remembering the heady days of May 2010 when we all expected the economy to grow by 13% or more over the lifetime of the Parliament, and the Government wanted to consider that prosperity alone does not deliver happiness, the coalition committed itself to promoting quality of life as well as economic growth, which are all very good things. The ONS established the National Well-being Project to try to measure what people feel about their well-being. The ONS is currently consulting on the domains and headline indicators for measuring national well-being. We know that this has become very fashionable across many other countries and bodies throughout the world. Are you optimistic about the success of this project?

Andrew Dilnot: Well, I am optimistic that it will remind us that there are valuable things other than simply measuring GDP that can tell us about what is going on in a country. Simply measuring GDP does not provide us with an adequate way of understanding what is going on in the economy, even if looked at just as the economy. People have often said that the recession we have just had was the worst since the 1930s. It is true that GDP fell by more than it did during the past two recessions, but unemployment rose by significantly less. Somebody who might be unemployed might think that the level of unemployment was a better way of measuring the seriousness of a recession. In part, the project is a reminder that there are many different ways of measuring all sorts of things, and GDP is one relatively easy to understand-I will not say it is easy to measure because it is actually very complicated. There are other measures such as per capita GDP, per capita net income, or health-we would all recognise that people’s health status is relatively measurable, although less easily than GDP, and that is another important aspect.

Much of the attention has not focused on the idea that we might want to look at a wider range of indicators-we would all recognise that is a sensible thing to do- particularly on more subjective indicators. We all think that we can measure somebody’s health or income, but we are-rightly-anxious about whether asking me how happy I feel will produce data that are comparable with asking the Chair how happy he feels. I am an economist, so I am always optimistic. My sense is that this area is not uninteresting, and as we get richer and richer as a society, it is important to look at a wider range of indicators to assess what is going on.

Some of the data published at the end of October-a wide range of information-were news to me and I thought they were interesting. The thing I liked most, which I printed out last night by accident, was a table showing the distribution of hours of work. I printed out the Excel spreadsheet and I then realised that I was printing 79 pages of numbers. After 25, I stopped and turned it into a picture, which makes it quite a lot more useful. It shows that the proportion of the total labour force working more than 45 hours a week has fallen from a peak of about 26% in 1997, to about 19% now. I did not know that, and although I have been immersed in statistics, I was caught up in the widespread prejudice that we are working harder and harder-well, apparently not; fewer and fewer of us are working more than 45 hours a week.

There is merit in that, and we will get interesting information from some of the subjective data. They do not, however, tell us how happy we are, and one thing I am really clear about is that I hate with a deep loathing composite indices and attempts to take more than one source of data and multiply them together with weights to come up with rankings.

Q55 Chair: So, the well-being index-

Andrew Dilnot: This is a pile of nonsense. They are always a pile of nonsense. Some lovely work was done by some academics in York which showed that a country’s performance in the World cup was a perfect indicator of the quality of its health service. You could change the weights in the World Health Organisation health indices so that they precisely mimicked how your country did in the World cup. It is just nonsense. There is great value in getting more than one source of information, and we can learn something by asking people how they feel. Because we are British, we will tend to say that we feel a bit miserable, and we can learn about how that changes over time. We might even be able to learn something about how it varies across the country-the week after Shane Williams has scored another brilliant try, the Welsh among us will feel very happy. We should not, however, invest too much interest in it.

Q56 Chair: Is it something that should guide spending and policy decisions, or is it just a quirky, interesting thing?

Andrew Dilnot: Looking at a wider range of indicators should guide spending and policy decisions. Simply looking at GDP rather than life expectancy, health experience and so on-those are relatively measurable, important, clear and objective elements of our lives, and it is good to look at the wider range of stuff.

I am, as yet, unsure about what we can learn from the more subjective stuff, but I certainly think it is worth looking at. We know that a non-negligible group of the population experience mild to moderate mental health difficulties or depression, and we should look at whether there are aspects of that that we can learn more about. That seems to me valuable, but we are a long way from knowing what to do about it. In particular, we are not going to get to a happiness index. I want to say very loudly now: the idea of a happiness index seems to me, frankly, really rather silly for the population.

Q57 Paul Flynn: During an election, I felt outgunned in a forum by a candidate who promised that, if he was elected, the whole population of Newport would be suffused with bubbling bliss-I had the minimum wage to offer. As it turned out, the voters tended to believe me rather than what the Natural Law party said-

Chair: But that might not be cause and effect.

Paul Flynn: This gives me a sense of well-being because the work on well-being is being done in Newport. But what do you expect that the Government will do when they have the statistics? Will they have a Happiness Bill? What will they do?

Andrew Dilnot: I imagine that what they will do when they have these statistics is say, "These will be interesting as we see them build up over the next 10 years." Until we have a reasonably long time series, I do not think that there will be very much that is very striking about them. I also do not think that they will be terribly expensive.

Q58 Chair: Well, it is costing £2 million. Is that too expensive?

Andrew Dilnot: The programme is costing £2 million at the moment. I do not know. The programme has been set up-let us see what it produces and see whether there is something interesting. In general, I think we should not be opposed to innovation. Let us see whether we get something interesting out of it.

Q59 Chair: Is this not something of an exercise to prove that we heartless GDP-driven politicians actually do think about things other than money? Is the interest of politicians in a happiness or well-being index about ingratiation, rather than in anything of real statistical value?

Andrew Dilnot: I am not sure. To the extent that we have got any information out of it so far, it is that young people and older people describe themselves as being happier than those of us in the middle. I think that that is quite surprising. Many of us might have thought that young people would describe themselves as happy but are rather surprised-I am rather surprised-that we see that among those over state retirement age. So it is not impossible that insights will flow from this that are quite important-what might be said, for example, is that perhaps what that shows is that forms of community interaction that are easier when you are not in full-time paid employment actually turn out to be really important to people. That does not mean that we should therefore say, "Well, stop working", but it might mean that we can think about ways of facilitating those kinds of interactions for people of working age. I do not know, I speculate, but would be wary of saying, "It’s unhelpful", because we might learn things from it. At the moment, I would not want to say that I thought it was the most important thing that we were doing.

Q60 Paul Flynn: Is not the experience of communities living in the war-as I did, and have vivid memories of-or under communism and then a democratic system that, in general, they prefer the equality of misery, whether warfare or communism, rather than the inequalities of prosperity? It is the sense of fairness that matters.

Andrew Dilnot: It is just possible, Mr Flynn, that that could drag me away from statistical matters into political matters, which I choose to eschew.

Paul Flynn: They are more interesting though, aren’t they?

Q61 Robert Halfon: May I just go back to the schools issue? I have just read a pamphlet that said that roughly around 50% of children do not get grades A to C in maths or English, which to me is a criminal statistic, if accurate. What would you do to improve the teaching of maths and statistics in schools?

Chair: Sounds like an education question to me.

Andrew Dilnot: I think that that is largely beyond my remit, but let me say one thing. People are far too scared of numbers. You do not have to be mathematically gifted to work out when a number is big or little. An example I use is of, many years ago now, a Government who announced that they were going to transform child care by allocating an extra £300 million to provide 1 million child care places over five years, and this was widely described as being fantastic. Well, £300 million to provide 1 million child care places-you do not need to be good at maths to work out that that is £300 per child care place. Three-hundred pounds per child care place over five years-again, you do not need to be good at maths to realise that that is £60 a year. Sixty divided by 52 is more difficult, but it is clearly between £1 and £2 a week. You didn’t need to be any kind of mathematician as long as you had the confidence, when faced with a serious-looking, smartly dressed person telling you something, to say, "Really?"

That’s what I would love to see in our schools. Rather than numbers coming at people in an aggressive way, people should feel released to say, "Well, does that make sense?" You do not have to be able to do a simultaneous equation; you just need to be able to do a little bit of simple arithmetic and not be frightened. People are much more capable than they believe. That’s the world that I’d love to get to.

Q62 Chair: One final and rather brutal question. Were we not to recommend your approval for the post and the Government went ahead with your appointment, would you accept it?

Andrew Dilnot: I think it would depend why you did not recommend that I be appointed. If you were to recommend that I were not appointed, I would want to look at why. If I thought you had good points, I would say, "Well, actually, they’re right, I shouldn’t do this." If I thought you did not want me for reasons that I thought were not really relevant, then I would be happy to go ahead.

Chair: Thank you very much for your time. We will decide whether you have made us happy or not.

Prepared 16th December 2011