To be published as HC 910-i

House of COMMONS



Public Administration SELECT Committee

Appointment of the Chair of the UK Statistics Authority

Tuesday 10 May 2011

Sir Michael Scholar KCB, Jil Matheson and Richard Alldritt

Evidence heard in Public Questions 1 - 104



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.


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

Oral Evidence

Taken before the Public Administration Select Committee

on Tuesday 10 May 2011

Members present:

Mr Bernard Jenkin (Chair)

Alun Cairns

Paul Flynn

Robert Halfon

Kelvin Hopkins

Greg Mulholland


Examination of Witnesses

Witnesses: Sir Michael Scholar, KCB, Chair, UK Statistics Authority, Jil Matheson, National Statistician, and Richard Alldritt, Head of Assessment, UK Statistics Authority, gave evidence.

Q1 Chair: Sir Michael, Jil Matheson and Richard Alldritt, welcome to this session. Would you identify yourselves and each of your roles for the record, please?

Sir Michael Scholar: Yes. My name is Michael Scholar and I am the Chair of the UK Statistics Authority.

Jil Matheson: I am Jil Matheson, the National Statistician.

Richard Alldritt: I am Richard Alldritt, the Head of Assessment in the Statistics Authority.

Q2 Chair: Sir Michael, I believe that you have one or two opening remarks. If you keep them very brief, we would be very grateful.

Sir Michael Scholar: I simply wanted to say, Chair, that perhaps the single most important feature of the Statistics Authority, in our view, is the independence that the Statistics and Registration Service Act accorded to us. We think that is most important to our whole operation. It, of course, rests on our accountability to Parliament, so I am very glad to be here and I will try to answer your questions as best as I can.

Chair: Thank you very much. We are very grateful for your being here and for the work that you have been doing. We are going to start on the question of the census.

Q3 Robert Halfon: Good morning. May I ask why we have the census in this day and age with open information?

Sir Michael Scholar: We need the census to have a robust estimate of the total number of people who live in this country, and also to have good estimates of where they live. That is primarily and essentially the purpose of the census, although it does give us information on a granular basis, and on a detailed regional basis, that cannot be obtained by other methods.

Q4 Robert Halfon: Is it right that this will be the last census?

Sir Michael Scholar: The Statistics Authority has expressed the hope that it will be the last census. I think that the Government have gone a little further than that and said that it is their firm intention that it will be the last census.

Q5 Robert Halfon: And if it is the last census-something you described earlier as so essential-why was there a need to do this census?

Sir Michael Scholar: Because if it is the last census-if there does not need to be a census on the conventional lines that we have been used to since 1801, and if there does not need to be one in 2016 or 2021-it will be because the ONS has developed better methods of making those estimates that I referred to before.

Q6 Robert Halfon: But do you not think that it is wrong that the census is somewhat authoritarian and that people are compelled to fill out their personal details?

Sir Michael Scholar: It has been a feature of censuses in this country for many decades, and I believe it has helped to ensure a high count in the censuses that have happened.

Q7 Robert Halfon: I saw one advert for the census near my constituency that suggested that you had to fill out the census because otherwise you would miss out on public spending. Surely that is the wrong way to encourage people, because whether or not they fill out the census does not really affect the public spending rounds. In essence it is a bribe; that must be the wrong way to encourage people to do it.

Sir Michael Scholar: I do not think so. Every year about £100 billion of public expenditure is allocated in various ways by the Treasury-through local authorities, through the health service and so on. If there are not good population estimates, there will be serious misallocation of those funds. You are absolutely right that there is no reason to think that the total of public expenditure would be different, but if, say, more money is being allocated to a local Authority than is needed, and not enough to another local Authority where it is needed, there is a misallocation of resources. I think that that is a perfectly reasonable argument to put to those who are being asked to fill out the census.

Q8 Robert Halfon: You have agreed that it is likely that there will not be a census in the future, but the public allocations to various areas will still happen in the future, so the link between public expenditure for your locality and whether you fill out the census must be a false one and the wrong way to advertise it.

Sir Michael Scholar: I have not said that it is likely that there will not be a census. I expressed the hope that there would not be a census on conventional lines, and I expressed that hope because the census is a very expensive operation. If it were possible to have estimates that were more up to date than are possible if one is doing the census every 10 years-if it is possible, as it were, to have a running tally of the number of people in the country, their characteristics, where they are and so on-that would be better, but we do not have the capability to do that at the moment.

Q9 Robert Halfon: In the last census, 300,000 people apparently described themselves as Jedi Knights in the question about their religion. Does that not show a healthy scepticism among the British people because of the compulsion to fill it in? Is not the census rather an unnecessary authoritarian measure in the sense that people are being forced to give out their personal details to the state and facing penalties if they do not?

Sir Michael Scholar: I do not know the motivation of the people who wrote Jedi Knights in 2001. It was particularly strange, because they were not compelled to answer that question. The question on religion was in 2001-it has been in this year’s census-voluntary. However, as you say, a number of people did write that in 2001.

Q10 Paul Flynn: Can I say a word of thanks and congratulate you personally and the Authority on your work in establishing the Authority as an independent and courageous voice that has been critical of the Government, the Opposition and anyone else who tries to manipulate statistics? I think we have taken a step along the road of restoring public confidence in statistics, which is crucial to having policy making that is based on evidence. However, may I join my colleague in suggesting that there really is no possibility of having a census in the future based on the biblical methods of 2,000 years ago of gaining information by asking everyone? There is no other investigation you take that involves anything but a sample of the population-one in 10,000 or perhaps one in 100,000-to give a result that would be, considering the costs, almost as good as having a census of the entire population. The only reason why it continued-these criticisms have been made over the past five years-was inertia in the system. People decided, and no one had the courage to stop it and say, "No, there is a better way of doing this. Let’s take a sample". It will not happen again, will it?

Sir Michael Scholar: The answer, as I have discovered from talking to professional statisticians here and abroad, is that it is not straightforward to dispense with that kind of enumerative census. Some countries do it every five years; we have chosen to do it every 10 years. The process of using the surveys that you describe is highly useful for gathering information, but that needs to be pinned down from time to time with the kind of census that we have had in this country. We would be in a stronger position to manage without this kind of enumerative census if we had an address register in this country, which we do not. We would be in an even stronger position if we had a population register, which we do not. I think that the population register is something that would be highly desirable from a statistical point of view. I imagine that the House of Commons might be unwilling to introduce a population register, but I am very happy to see that the case that we have made as an Authority for the creation of an address register has at last been accepted by the Government in recent months. I think that that will be a very important element for moving forward in the way that Mr Halfon was inviting us to do.

Q11 Robert Halfon: May I just ask how you work out the questions on the census? For example, if I am not mistaken, there is no ethnic recognition of Sikhs on the census whereas there is recognition of other ethnic groups. How do you make that decision?

Sir Michael Scholar: Well, on the Sikh point, there was a place for recognition of Sikhs in the religion question, although not in the ethnicity question.

Q12 Robert Halfon: But they argue that it should be done ethnically. Who makes that decision? Why do you choose some ethnic groups and not others?

Sir Michael Scholar: Parliament made the decisions, because we put the questionnaire to Parliament for its approval, as we are obliged to do under the Census Act. However, we did so on the basis of testing and consultation. Perhaps Ms Matheson would like to fill that out a bit.

Jil Matheson: The process of determining the content of the census is a lengthy one that involves a lot of public consultation, a lot of engagement with users of very different kinds and, I have to say, a huge demand-a demand that far exceeds the capability of any household or individual to be able to fill in. So we go through a process of testing. The form that we have had in 2011 was extensively tested. Of course, what we had to do was to make recommendations to Parliament based on a combination of: the case that had been put to us by users; whether there was any other sort of information available; why it was essential to have it for a total population and for small areas and so on; the ability of people to fill in the form; the length of the form; cost; and so on. In terms of the ethnicity question, we could have had a page of ethnicity classifications. What we did was look, as I say, at user need and people’s ability to fill it in, but also at the numbers in the population. So, you have a cutoff that recognises the largest groups and then the ability for people to write in for the other categories.

Q13 Robert Halfon: But the Sikhs are inexorably exercised about this point. There is quite a high proportion of Sikhs in the United Kingdom and they regard themselves as an ethnic group, not just a religion, and yet they are not allowed that option. They have no say in how that decision is made.

Jil Matheson: They had a very strong say, actually, because we did have a lot of engagements with Sikh groups and I am sure that there was representation to Parliament at the time that the content of the questionnaire was debated. There was also some research done that showed that the answer to the Sikh category in the religion question was a very good indicator of the answer that would have been given had it been in the ethnicity classification. There is some question of avoiding redundancy as well in making some very tough decisions.

Q14 Chair: Can I ask a bit more about the alternatives to the census? Ms Matheson, you might want to say something about this. Is this address register in fact a national identity register? Is this the identity card by the back door?

Jil Matheson: Shall I say something generally about alternatives? The first point to say, I think, is that what we are talking about is not that the need for this kind of information has gone away-nobody has argued that. It is a question of how you provide that information. It is quite right that we are looking at ways in which that information can be provided in a way that is cost-effective and timely, and that provides the information that users across the country need. In fact, we looked at that after the 2001 census as well and the recommendation was not based on inertia-it was based on an analysis of the available data sources to be able to provide that kind of information. The conclusion was that there was no other single source that would enable that information to be provided, so the decision was made to proceed with the census. However, it is absolutely right that we look again, because there is more information available and out there and because it is getting more difficult across the world to carry out traditional censuses. So, that is what we will be doing-we will be evaluating the alternatives. The address register is simply that: a register of all the addresses in the country, not of the people who live at those addresses-that is different. We do not have a population register, and the address register is simply that-a register of addresses. One of the things that we had to do for the 2011 census was to create an address register, in effect, which is why I think we are all pleased to see that is now going to be taken forward into a national address register.

Q15 Paul Flynn: All your other reports are based on a minute sample of the population. Can you tell us, as a statistician, what the difference in value would have been between the results we are going to get here from doing the whole population-lots of which, with the delay that is likely to take place because of the mass of results that have to be analysed, will be out of date by the time they are reported-and doing one in 10,000 people?

Jil Matheson: The sample surveys certainly have their place, but even now with the sample surveys that we have, the population estimates that come from the census, and are updated intercensally, form a benchmark for those sample surveys. You do not do a sample survey without having some way of being able to calibrate the results. The information that comes for the total population is one way of doing that, and it is not a question of simply being able to do a survey. That is one part of the technical answer. The second is to do with the level of granularity with which information is required. It is granularity in two ways, first of all geographic. It is about being able to provide information for small geographic areas that a sample survey simply cannot do. Secondly, it is granularity in terms of some fairly small population subgroups that will not be represented in random samples of whatever size in a systematic way.

Can I make just one other point on this? Of course, one of the things that we are also looking very hard at is what other countries do. There are countries that have moved away from the traditional biblical form of census taking; there are others that, like us, are looking at the options for the future; and there are some that combine a headcount-a very short census forum-with a sample survey, so you do just the headcount on a populationwide basis and then fill the gaps on a sample-survey basis. All those options are ones that we are looking at and evaluating.

Q16 Robert Halfon: What is the compliance this year compared with previous censuses?

Sir Michael Scholar: In the census that is under way now?

Robert Halfon: Yes.

Sir Michael Scholar: The period after the census during which the field force of 29,000 people went round following up those who had not returned the forms has just ended, and the ONS is cautiously optimistic that more than 90% of the forms that were sent out have been returned or will be returned in the coming few days. It is still very early days; there are still many forms coming in. That leads it to predict that at the end of the process there will be coverage of around 94% across the country as a whole, which in fact is the target that it was seeking to achieve from the outset.

Q17 Robert Halfon: How does that compare with previous censuses?

Sir Michael Scholar: The previous census also had a target of 94%. There was a subsidiary target in this case: no local Authority should be below 80%. The ONS is, again, cautiously optimistic that when all the forms have come back, there will be no local Authority below 80% this time.

Q18 Robert Halfon: Finally, do you, as the Statistics Authority, think it is right that the census is compulsory, given the resentment that it creates among a significant number of individuals?

Sir Michael Scholar: I think the way I would put it is that the Authority thinks it right to recommend to Parliament that there should be a compulsory element in the census in order to achieve the kinds of returns that I have just been talking about. Whether there will be that compulsion in any census in the future is a matter for Parliament, not for us.

Q19 Kelvin Hopkins: In the interest of balance, as most of my colleagues are rather hostile or sceptical about the census, let me say that I am a great supporter of the census and that I enjoy filling in my census form-it gives me a sense of belonging, identity and all of that. Is it not a concern that we would lose some of the time series that goes back to 1801 by abandoning the census? That is just a matter of history and interest. Of more concern to me is that in a constituency such as mine that has 30% or so ethnic minority people who have a very low response level to the census, the process understates that population very substantially. Is not that a worry in all sorts of ways? Can we not do something more to ensure that we get higher returns from those areas?

Sir Michael Scholar: I think that that was the real difference between the 2001 census and the census that is currently in process. The ONS has made an enormous effort to target those areas that are difficult to enumerate and for which there were some low figures in the 2001 census. I think that there were some local authorities in 2001 that achieved no more than 63% coverage, so if we have achieved 80% everywhere, that would be a considerable improvement on what happened in 2001.

Q20 Chair: What needs to be in place for the 2021 census to be definitively cancelled?

Sir Michael Scholar: I will have a stab at it and then I will ask Jil Matheson to put me right. I think that we would certainly need an up-to-date address register.

Q21 Chair: And that is simply an address of every household in the county?

Sir Michael Scholar: Every household in the country. We had to do that. It had to be done by the ONS.

Q22 Chair: But don’t local authorities have that already?

Sir Michael Scholar: No, they do not.

Q23 Chair: Then how do they collect their council taxes?

Sir Michael Scholar: There are three separate sources of addresses: Royal Mail, the Ordnance Survey and local authorities. They considerably overlap, but they had to be put together by the ONS to achieve a comprehensive address list.

Q24 Chair: How inaccurate is the council tax base, then?

Sir Michael Scholar: I cannot answer that question without notice, I am afraid.

Q25 Chair: Are they collecting too much money or not enough money?

Sir Michael Scholar: I am afraid that I cannot answer that.

Q26 Chair: It is a fairly obvious question to ask, isn’t it? If that address list does not satisfy your requirements, what tends to be wrong with it?

Jil Matheson: It does not satisfy our requirements. That was why we had to create it for 2011.

Q27 Chair: But what is wrong with it?

Jil Matheson: Addresses missing, certainly, and lags in getting house conversions, new builds and demolitions on to and off registers. It is not up to date.

Q28 Chair: Why are not local authorities better at keeping their records up to date?

Jil Matheson: Local authorities are one of the sources. They have been part of creating this new address register, and one of the things that some local authorities have been saying to us is that the process of going through this has helped them to update and maintain their address register, which was certainly not fit for statistical purposes.

Sir Michael Scholar: May I complete the answer I was trying to give your question? The first point would be an up-to-date address register. I think the second point would be to achieve a reconciliation of some very large databases that at the moment are not reconciled with one another-certainly not as far as the information technology is concerned. I am thinking of the schools census, GP registers, the national insurance database and HM Revenue and Customs information.

Q29 Chair: And the barriers to this are simply practical, not legislative? There are not legal impediments?

Sir Michael Scholar: Some of them are legislative. Of course, the Statistics and Registration Service Act has allowed us to begin to dismantle some of these barriers, but that is not a straightforward process. There is some understandable hostility towards the proposal that one Department should hand over all its information about the population to another Department, and that a massive database on the whole population should be created. There are political anxieties about that.

Q30 Chair: I think there is also public resentment at having to tell the Government again and again and again where you live.

Sir Michael Scholar: Yes. It cuts both ways, certainly. There is a great deal of work to do before those databases are reconciled with one another and can communicate with one another. For this process to work, that will have to be achieved.

Q31 Robert Halfon: One brief technical question. One of the problems we face, as of course Moser identified 10 years ago, is that 20% of the population are functionally not literate and filling in forms is a challenge for some people. Sometimes hostility to voting or doing anything of that kind is disguising the fact that people are not literate-they are not prepared to admit it, obviously. How do you overcome those problems? Especially, how do you overcome the situation when people will not say, "Look, I cannot read and write; will you help me?" and just say, "I do not fill in forms because I think it is an outrageous intrusion on my personal life," and so on?

Sir Michael Scholar: That is the reason for having 29,000 temporary staff recruited by the ONS for a month or so. It is one of the reasons why the census is expensive, because these people have to be recruited and trained. This time they have been able to act in a much more targeted way than they could in 2001, because in the census this time a system has been in operation so that each form is tracked. If a form has not been returned, there has been management information to tell the ONS that this particular address has not yet replied. These 29,000 staff have been able to visit very specifically the people who have not replied. Of course, some of them will come into the category you have described.

Q32 Chair: When does this address register need to be in place? When will you have sufficient confidence that you can say that we cancel the 2021 census? Perhaps more importantly, when have we got to initiate the 2021 census if it is going to happen? When is it too late to make this decision?

Sir Michael Scholar: Well, the utility of an address register is not simply for statistical purposes.

Q33 Chair: I appreciate that, but I am asking about the timing.

Sir Michael Scholar: "Beyond 2011", which is the ONS’s programme for developing plans for a future census, expects to make recommendations in 2014.

Q34 Chair: And do you feel that the address register is being approached with sufficient urgency to meet that deadline?

Sir Michael Scholar: Well, it wasn’t.

Q35 Chair: Ms Matheson?

Jil Matheson: No, it wasn’t, but it now is.

Sir Michael Scholar: We hope it is now, yes.

Q36 Chair: I think we would like a note on that progress and what we need to look for as a Committee to make sure that progress is being made. Do you think you could furnish us with that, please?

Sir Michael Scholar: Certainly, yes.

Q37 Chair: Thank you very much. Moving on, we next want to ask about measuring well-being. I note that a great deal of resource is being devoted to this. How far have we got on devising a well-being index?

Jil Matheson: Let me just give a bit of the background to this. This is not just about a well-being index; it is about a programme of work that has been going on internationally for a while to develop a suite or a dashboard of measures that will reflect the progress of society, reflect performance and change, and go beyond the traditional measures of GDP. In fact, a lot of this has been developed through the OECD and the EU, and what they call it is "Beyond GDP". In other words, it is a way of looking at how the country as a whole and how different groups in the country are faring economically, how the country is doing environmentally and how it is doing in terms of quality of life, including subjective well-being.

Q38 Chair: When will you unveil your index?

Jil Matheson: We started asking questions on the subjective well-being aspects of this in a survey this year. The results of that will be available from summer next year. It is going to be after the whole-year round.

Q39 Chair: So it is quite a protracted process. What indicators are you using? If someone has got a job, or is living in a house-they have got a home-or is healthy, those are measurable outcomes. Generally, if people do not have these things, they become unhappy. Why do we need to move to more subjective indicators? What indicators are you likely to use?

Jil Matheson: There will be those objective measures of the kind that you describe, many of which are quite well measured in the UK. What they will be supplemented with-as recommended in the report that has generated a lot of interest in this by Stiglitz, Sen and Fitoussi-are subjective measures of how people think they are faring in different aspects of their lives. There is increasingly good evidence that how people feel about things impacts on behaviour and on some outcomes. Of course, there are also people who are interested in that as a measure in its own right.

Q40 Chair: But isn’t the fundamental problem here something that psychologists and psychiatrists refer to as insight? As soon as you are made aware of what you are being asked about, you feel differently about what you are being asked about. Isn’t this an inherently subjective process?

Jil Matheson: Yes, that is part of the academic debate about this. However, I think there is evidence-international, well-based evidence-that you can ask people.

Q41 Chair: But who is producing this evidence? What is the impetus behind this? Isn’t it that some people want different policies and they are looking for new ways to justify them because they have failed to satisfy other measures? For instance, GDP growth in the EU is not very good, so they are looking to produce other measures on the benefit of those policies that are failing to produce better GDP growth.

Sir Michael Scholar: I think that the desire to produce something on well-being goes back quite a long way. There have been statisticians who for decades have hoped to be able to do for well-being what was done for income and wealth in the 1930s and 1940s, when measures of national income and national wealth were developed and agreed internationally so that international comparisons between countries could be made. I think that this is a very ambitious programme and it has a very long way to go. I personally think that the subjective element is just one aspect of it, and it would be very important to have elements that measure health, education, security, liberty and various other things of that kind. Whether they can all be put together to make a single index is, I think, a very open question, but I guess in the 1920s or in the late 19th century it would have seemed inconceivable that you would be able to compare different countries’ incomes and their stock of wealth, but nevertheless that system, for all its imperfections, has been developed by statisticians and economists.

Jil Matheson: May I add one point? The ONS will not stop publishing GDP and all those other measures. This is not a replacement; it is an addition. Nor will other countries, so it is not a replacement.

Q42 Chair: But can you give an example of the sort of things that you might be measuring that might result in different policies and an example of what those different policies would be? I might give an example that divorce makes people unhappy, and if you measure people being unhappy, it might then result in a strengthening of family policy.

Jil Matheson: It would be up to the politicians to decide what the policy is that flows from this.

Q43 Chair: But doesn’t my question demonstrate how subjective and political it all is? Family policy just tears apart the political consensus, doesn’t it?

Jil Matheson: That does not invalidate the need for information to inform your political tearing apart, as you put it.

Q44 Paul Flynn: Does not history show that equality is the main factor in deciding happiness? You mentioned the 1940s, Sir Michael. As a child of the 1930s and 1940s, I remember in those times that the lack of choice and the shared misery-it was equal shared misery-led to a general level of contentment, whereas the inequalities of prosperity and higher GDP made people unhappy. For example, when communism collapsed in Hungary, the suicide rate went up-they had all been in the same boat before then. Equality is the main factor. I am sorry to be getting on to this.

Chair: It is all right, but what is the question?

Paul Flynn: Should not we ask people whether they feel equal in society and whether society is being fair? That would be a far better measure of happiness than GDP.

Sir Michael Scholar: I think it is an enormous and, as I said, ambitious programme. If it is to be successful, its authors will also want it to be an international programme-they will want to be able to make international comparisons of well-being. To secure international agreement on the kinds of points we have been talking about would clearly be a tall order.

Q45 Chair: So do you think that your index would be able to indicate whether the Government should be seeking to make people more free or more equal? Would that be an outcome of this index?

Paul Flynn: America or Scandinavia.

Chair: It is about whether we go towards a Scandinavian system or an American system?

Sir Michael Scholar: I think people would make very different uses of it. They would make all kinds of uses of it, as they do of measures of GDP, measures of earnings and measures of prices and income. They make all kinds of political points based on this information, but they are glad to have the information.

Q46 Chair: Will not different countries try to manipulate their happiness index to show that they are happier than their rivals, rather like East Germany did with their GDP figures?

Sir Michael Scholar: Yes.

Jil Matheson: That is why you have independent statistics offices doing this stuff.

Q47 Alun Cairns : Can I come back to the Chair’s question about where you were seeking to gain a policy consequence out of the information that was gleaned? Is there any one international example that you can come up with whereby, as a result of information that was thrown out of the happiness index, there was a shift in public policy?

Jil Matheson: I think it is very early days. We have not even got our data; we will not have that until next year. In fact, some other countries are at about the same point as we are. I think that that is a good question, but you should ask us a year or two years from now.

Q48 Alun Cairns : But were we not told at the time that this policy was launched that this had been done in Canada for many years? Is that one example or one nation we can look to to see where there has been a shift in public policy?

Jil Matheson: I do not know of shifts in public policy in Canada, I am afraid.

Paul Flynn: Can we go on to prerelease, Chair?

Q49 Chair: Before we leave this well-being stuff, isn’t there something rather authoritarian about the state measuring the happiness of the population and then telling the population how happy they are, when in fact how happy you are is a matter for individuals in their own circumstances? If a whole community is being put out of work, obviously that community will be unhappy. What is this going to tell us that we do not already know, because we are dealing with our fellow human beings?

Sir Michael Scholar: I could turn the question against you if we go back to income and wealth. People could have made precisely that point in 1860: "What is the point of knowing what the national income is? All we are interested in is our own wealth and income, and those of our neighbours." Nevertheless, most people now think it has been useful to have some measure of our national income and growth, and the comparisons with that of neighbouring countries and countries that have other systems.

Q50 Robert Halfon: But it is easy to measure income and wealth, because you are clear what you are measuring, whereas if you are measuring happiness, it is completely subjective.

Sir Michael Scholar: There are considerable problems about measuring income, aren’t there?

Chair: Not really. Mine is published.

Jil Matheson: Being subjective does not mean it is not an issue.

Kelvin Hopkins: But a lot of it can be done with objective measures such as obesity-the number of children who are massively overweight-mental ill health, even among young people, and illnesses of various kinds. There are objective measures that you can build into it rather than saying, "How happy do you feel?" That depends on individual nature as much as anything else. I happen to be a happy person but I know people who are not as happy as me, just because that is their nature.

Chair: I am going to watch this issue very carefully.

Q51 Greg Mulholland: Could I ask you, Sir Michael, if you share the concerns expressed by the RSS about the standing of statistics as a profession in the civil service?

Sir Michael Scholar: I have had real concerns about the standing of statisticians in government, and I believe that the Statistics and Registration Service Act and the creation of the Authority have enabled us to strengthen the position of such statisticians. I think that that has happened over the past three years. I have been concerned about it, and I remain concerned about it, but I think we have made some progress in this respect.

Q52 Greg Mulholland: You recommend giving more weight to the reporting line between the departmental heads of profession and the National Statistician. Do you think that is sufficient in addressing this issue?

Sir Michael Scholar: I think it would make a big difference. As you may know, I wrote to the Prime Minister in May last year and asked him if he would make three changes to strengthen the position of statisticians in Government Departments. One of my proposals was that before any significant changes could be made to the statistical capability of a Department, or any major changes to its statistical output, the Department would be obliged to secure the agreement of the National Statistician. That would be going back to a system that pertained in this country during the time that Claus Moser was head of the Central Statistical Office. I asked the Prime Minister if he would go back to that system, which would be something that he could do through administrative action without any need for legislation or for any additional expenditure. I also asked him if he would accept the proposals we had made on prerelease access.

My third proposal to him was that he should give the Authority a place in the decision making about cuts in statistical capability across the whole Government. Recognising, in the difficult fiscal position that the Government were and are now in, that there were going to be cuts, we felt it was very important that the Statistics Authority, with a view right across the scene of the whole statistical system, should be brought into the process of decision making about where cuts should be made. I put that in a letter to the Prime Minister and if, as I hoped, he accepted those recommendations, he would strengthen the position of statisticians in Departments materially.

Q53 Greg Mulholland: Thirty years ago, senior statisticians in Government Departments were at director level; none of them are now. The RSS has said that they find "this loss of capability alarming". Do you think that that is a result of the lack of priority that Government are giving, or are there other reasons for that happening over the past 30 years?

Sir Michael Scholar: When Governments cut public expenditure, I think that they always try to preserve front-line capability and are very tempted to cut policy analysis. I believe that the statistical service has been cut disproportionately in the past 30 or 40 years, and I think the RSS is quite right to point out that the number of senior statisticians has been greatly reduced. If we look at the number of Fast Stream entrants to the statistical service, that has also been greatly reduced. If we compare the membership of the Government Statistical Service with the Government Economic Service, the statistical service is very much outnumbered by the economic service. Over this period, we have seen a big reduction in statistical capability at the same time as we have seen a big increase in the number of special advisers and political advisers in Departments, some of whom are paid at very significantly high levels.

Q54 Chair: Do you think this is reflected in the politicisation of information?

Sir Michael Scholar: I do. When I saw that this Bill was going through Parliament, I felt very glad about it-it was one of the reasons why I put my name forward to become Chair of the Statistics Authority-because I felt that it gave us an opportunity to reverse some of that direction.

Q55 Paul Flynn: There was a great deal of enthusiasm for the Bill at the time, as you rightly say, but one of the disadvantages of the Bill that Parliament delivered to you was this dual function in which you are the producer of statistics and also responsible for the scrutiny of what you produce yourself and the statistics produced by other Government Departments. Do you think the Chinese wall that you put in-with Lord RoweBeddoe taking a slightly removed position-is working? Does there need to be a change in the air to ensure that there is a division between the two functions?

Sir Michael Scholar: It has been a difficulty that we have wrestled with. The Act did not specifically prescribe two deputy chairmen, but it did not prevent us from seeking the appointment of two deputy chairmen. The decision to do so permitted us some separation-as much separation, in fact, as I think it is possible to achieve. I would say that, in the main, it has worked well. There were of course those who at the outset said that it could not work and that we would never, as a regulatory Authority, criticise ourselves as an Authority as a producer of statistics. In fact, we have done so on a number of occasions, as I think you have probably noticed.

Q56 Paul Flynn: One of the things I recall very strongly from the passage of the Act through Parliament was the main contentious issue of prerelease. I believe that the Lords took the opinion that there should be no prerelease period. The Conservatives’ policy in opposition was that they did not want a prerelease period and the Government defended it. Now the Opposition and the Government have changed scripts. We now have Mr Maude defending prerelease, quite extraordinarily. There really is a strong argument to say that Governments in the past have used the 24 hours prerelease to spin the facts and to give their gloss-they have an advantage over someone else. Other countries do it in a different way with no prerelease period or a three-hour period. What is your view now? Should we get rid of it altogether?

Sir Michael Scholar: My view is that we should get rid of it altogether. Our report recommended that there should be a maximum of three hours and that the norm should be considerably less than that. My personal view is that there should be none. I do not see why we have to have it; I have never understood a single reason why we have to have prerelease access. It is convenient for political advisers and Ministers. I think it is damaging to the credibility of the statistics that are produced by Government. Even if there is never any spinning going on, the public think that because Ministers have these figures, it is likely that there is a manipulation of the figures. We have seen that in a number of cases. When there was the televised debate between the party leaders during the election campaign, the then Prime Minister was aware of a number of economic statistics ahead of those with whom he was debating, which I think was an uncomfortable situation for him as well as for them. Then we have seen in recent weeks a situation in which the Chancellor was aware of the GDP figures that were about to be published by the ONS and, in giving his appraisal of the economic situation to the Cabinet, there were fears and accusations that he had disclosed the figure that had been given to him on a prerelease confidential basis.

Q57 Chair: So are you saying that the Chancellor of the Exchequer cannot disclose prerelease information even to the Cabinet?

Sir Michael Scholar: No, he cannot. Under his own prerelease access orders, he is prohibited from passing on that information.

Q58 Chair: Is it your understanding he did so on this occasion?

Sir Michael Scholar: No, no. I have seen no evidence that he did so. I do not believe that he did so, but there were a number of people who thought that he had, so there was some suspicion about his position that would not have existed if there was no prerelease access.

Q59 Chair: But would it be released to special advisers?

Sir Michael Scholar: Well, it is released to special advisers.

Q60 Chair: On a prerelease basis?

Sir Michael Scholar: Yes, they are on the list.

Q61 Chair: So how much confidence do you have that the rather jejune political community that is special advisers is as disciplined as, say, your former civil service colleagues in the Treasury?

Sir Michael Scholar: Well, the ONS produces the RPI and the CPI, and I must tell you that I am alarmed that-I am told-there are more than 50 people on the ONS prerelease access list in the Treasury, the Bank of England, No. 10 and the Deputy Prime Minister’s office. I do not know why more than 50 people have to see the RPI before it comes out.

Q62 Paul Flynn: That sounds like a step backwards rather than the progressive move that the statistical world had hoped for from the Act-that we would get rid of prerelease and that that would absolve anyone of these accusations of spinning, bias and so on. You have made representations to the Government on this. Do you think that we as a Committee really have to say to Government, "You really must do what you promised in opposition – let’s get rid of this prerelease period because it is not helping to build confidence in national statistics"?

Sir Michael Scholar: I would think that a very appropriate thing for the Committee to decide to do.

Q63 Chair: Are there elements of the civil service that are obstructing this reform? Do you think that your former colleagues in the Treasury are unhappy about losing prerelease, or is it purely a political matter? I am asking you to make quite a controversial statement.

Sir Michael Scholar: I do not think it is simply Ministers and special advisers, no.

Q64 Chair: At the moment, there does not appear to be consensus in the Cabinet on abolishing prerelease or even limiting it.

Sir Michael Scholar: As I understand it, a firm decision was taken in July not to abolish prerelease access-so I was told by Francis Maude.

Chair: Which is contrary to his aspirations as expressed in his own personal statements. Is there anything more on prerelease?

Q65 Paul Flynn: Just a final thing. Do you think that confidence in statistics has been improved since 2007? Are the Act and your Authority really working?

Sir Michael Scholar: It is very hard to say that confidence has increased. It is very difficult to find hard evidence of that.

Chair: Sorry, this is Alun’s question. Do you want to deal with that public confidence issue now?

Alun Cairns : That is fine. I will listen to the answer from Sir Michael first, if that is okay, before I add a supplementary.

Sir Michael Scholar: I think that confidence and trust in the statistical system have multiple sources and I think, if there has been a reduction in public trust in the Government, in authority and in Parliament, and the institutions of government and Parliament, confidence and trust in statistics will suffer. I suspect that that is what has happened in recent years, and I believe on the basis of the surveys we have done that there has been no improvement in trust in official statistics as we have measured it. But against that, I would say that the actions we have taken as an Authority over the past three years have led a number of people who follow these things closely to believe that there has been an improvement in the integrity of the figures and of the statistical releases, and I believe that that indeed is the case. I hope that, in time, that increased confidence will communicate itself to others.

Q66 Alun Cairns : Can you tell me the most important actions you have taken that have made the biggest difference?

Sir Michael Scholar: I think it goes back to December 2008, when the Authority criticised the then Prime Minister’s office and the Home Office for the way in which they published some material on knife crime. If you look back at that episode, you will see that the Prime Minister’s office issued an apology and the Home Secretary made an apology in the House of Commons for what had happened. Very importantly, the Cabinet Secretary sent some very important guidance about statistics around Whitehall at the end of January. That guidance told all officials-not simply press offices, Ministers’ offices and political advisers, but all civil servants-that they had to abide by the terms and the provisions of the Code of Practice for statistics, that they had to consult professional statisticians in their Departments before putting out statistical material, that they had to accept that the word of professional statisticians on these matters was final, and that they must not publish selectively from unpublished datasets numerical or statistical material that was politically convenient. He produced that guidance and I think that it has had a considerable effect in Whitehall.

Q67 Alun Cairns : Can I go back to the period leading up to December 2008? That was quite an unprecedented act by the Statistics Authority. Were there various protestations made within government before you went public and took that bold move?

Sir Michael Scholar: Yes. The head of profession in the NHS Information Centre protested about the use of these statistics before they had been quality assured and before they were ready to be published. They were not due to be published for several months. He protested about the selective use of them.

Q68 Alun Cairns : Excuse me for interrupting, but I am trying to get at not only those statistics specifically, but the trend of the misuse of statistics by Government Ministers. Was the Statistics Authority objecting to or privately complaining about the way that they were being used before you took the bold move to go public and criticise them in the way that you did? What I am trying to lead on to is: are we at any stage following that same path, yet with the new Administration, where you may well be flagging up some early concerns because statistics might not be being used in the way that you would like them to be?

Sir Michael Scholar: Well, I very much hope not, because I think the situation was greatly changed by the guidance that the Cabinet Secretary sent round to which I referred. That guidance is still in force under the coalition Government. Furthermore, it has been strengthened, because at the suggestion of the Statistics Authority-I wrote a letter to the Cabinet Secretary-a paragraph was put in the Ministerial Code of Conduct to enjoin Ministers to abide by the Code of Practice for Official Statistics, so there is now a very clear mechanism that should prevent the kind of situation that took place in December 2008 from happening again-I hope it will not.

Q69 Alun Cairns : What further role could the Statistics Authority play in ensuring greater confidence and developing greater confidence in the statistics that are being shared?

Sir Michael Scholar: If the Government were to accept the three suggestions that I made in my letter to the Prime Minister-giving a greater role to the National Statistician on professional matters; giving a greater role to the Authority on budgetary matters; and abolishing prerelease access-which are the three things that the Authority has consistently been asking for since early last year, the trustworthiness and credibility of statistics would be improved.

Q70 Chair: The only thing I would press you on in that is that aren’t there public expenditure implications to your second recommendation? It is about resourcing the statistical service across the Government more fully than it is currently resourced.

Sir Michael Scholar: We are not asking for special treatment for statistics in the expenditure exercise. We are not defending the statistical service against budgetary cuts; we are not in that situation. The situation we are in is to say that if cuts are being made and the Government wish to cut expenditure-and that is a matter for the Government; it is not a matter for us-please give us the opportunity to comment before it is done. The most powerful reason we have for that is that a cut in one Department’s statistical work can disproportionally affect the work done by another Department, or a statistical series that is not put out by that Department. We have had one example of that in recent weeks, in which the NHS Information Centre proposed to reduce its contribution to the General Lifestyle Survey, which would have the effect of ending several very important series of statistics on smoking, drinking and experience of the National Health Service.

Q71 Chair: Jil Matheson, if the Authority is not consulted, are you consulted about these changes, or do the departmental silos operate in isolation from each other?

Jil Matheson: I do my best, but one of my concerns, exactly as Michael has said, is the fact that departmental silos do operate in that way, of course, because that is where the budgets are held and the decisions are made. I think one of the things that has changed, and one of the impacts that the Statistics Authority has had, is that there is wider recognition of the requirement to consult publicly on changes to statistical work programmes, but nevertheless that is still done on a departmental basis. I certainly try to encourage statisticians across Government to consult each other, but there is no requirement that they should do so.

Q72 Robert Halfon: I have no problem with your criticism of Governments and I think it is probably a good thing, but could I just ask how you made those decisions to come out publicly and criticise Government statistics? Is it decided by you? Do you have a meeting? Is there a set of guidelines that you follow when you do this?

Sir Michael Scholar: We have the Code of Practice for Official Statistics. It has various, very clear clauses. Under "Principle 3: Integrity", it says, "Ensure that those producing statistical reports are protected from any political pressures that might influence the production or presentation of the statistics".

Q73 Robert Halfon: But that is different from making the actual decision to come out publicly. Because you are an independent body-which is right-your word is seen as the law, in essence, so when you come out and criticise a Government of whatever persuasion, it is seen as a massive criticism of that Government. How do you make the decision to come out in that particular way?

Sir Michael Scholar: It is a difficult decision to make. It requires statistical analysis; I need to understand the statistics, what has been said about them and what has been done-the statistical facts, as it were. There is also a political element that must come into my judgment. The Authority has no powers of enforcement under the Act. It has no powers, as other regulatory authorities have, to fine people or to disqualify them from carrying out their business-we have none of that. Our only power is to make a report to Parliament and to the public. So, we can make a report, but if we were to make a report every day or every week, nobody would take the slightest notice of us. We have to judge when an intervention will be effective.

Q74 Robert Halfon: Is it you who makes that decision or is it your board or your colleagues? Is it done in the same way each time?

Chair: Mr Alldritt might like to comment on this, as your principal adviser on the scrutiny of statistics.

Sir Michael Scholar: Yes.

Richard Alldritt: Certainly I am involved in those decisions and, as Sir Michael says, there is a process. There are so many dimensions to this. There are questions such as how sure we are of the facts. If we are talking about a statement that has been made in Parliament or in the media that we think misrepresents the underlying statistics, is there a consensus among the experts that what was said is actually incorrect, and is it sufficiently serious to be worth correcting publicly? There is a whole set of considerations of that kind. There is a process of drawing this information together and setting it before the Chairman and the other relevant members of the Authority and reaching a consensus on whether to proceed. It is also true that on occasion we indicate to a Government Department that we are concerned about a particular use of statistics and that we would like reassurance on particular points, and we get that reassurance and we leave it there.

Sir Michael Scholar: If I may just add to that, we very often act through phone calls and contacts with the Department. A big public letter or interview is not the first thing one does; it is the last thing one does. You try to ensure that there is compliance with the Code of Practice, that people understand the Code and that you bring it to their attention. I think in many cases we have seen that action of that kind has been effective.

Q75 Chair: Sir Michael, I think that this Committee would like to know on every occasion that you, the Authority or Mr Alldritt feels that there has been a breach of the code or that some other legitimate concern has been uncovered about the quality of statistics or the way in which that statistics have been used in Government-rather more like a regulatory Authority. Is that beyond your resources?

Sir Michael Scholar: No it isn’t. What you have just described is in fact our intention and our practice. It is just that I wanted to say to the Committee that we do not go out at every breach with a huge statement and attempt to seek the headlines. That is not our intention at all. We want to be effective, but we very much recognise the obligation to report to Parliament, and that is to this Committee.

Q76 Robert Halfon: But it is you and the board that make that final decision about whether to fire the gun, in essence.

Sir Michael Scholar: Well, if there is time, I would certainly involve the whole board, but sometimes one has to act quickly. Certainly I would say that the actions I have taken have always been endorsed or agreed by the board.

Q77 Robert Halfon: What happens if it is very near an election-I do not mean in the purdah itself, but two or three months before? Your intervention publicly could be incredibly damaging to either the political party or the Government concerned. Is that taken into consideration when you make these decisions?

Sir Michael Scholar: I do not want to engage in party political debate. I do not think it is appropriate for the Authority or for me to do so. Similarly, however, if there is a serious breach in the way in which statistics are being handled and produced, I would not think myself to be prevented from making that public simply because an election was impending. I do not think that would be correct.

Q78 Alun Cairns : Can I press you a little bit further? I think it is right and fair that that judgment should be made on any Government. Do you also think that it is your role to make a similar judgment on the Opposition if they are also misinterpreting your statistics?

Sir Michael Scholar: That is a very vexed question. Our role, as set out in detail in the Act, is to deal with the production and publication of official statistics. That means that most of our regulatory work and public utterances are directed towards the producers of official statistics: the Government. However, we also have the ability under the Act to comment on statistical matters generally. That presents quite a dilemma for us because every day in the newspapers, and on radio and television, statistics are misquoted, and used selectively and in a way that perhaps damages their credibility. If we commented on everything, I think that we would need 10 times as many staff, and I think it would be rather pointless because people would take no notice of us. The kind of test we have set ourselves is to comment on Opposition spokesmen or on media use when the effect of what has been said is to do serious damage, in our judgment, to the integrity of the official statistics that we are there to safeguard. That has meant that we have criticised Opposition voices from time to time, but probably rather less than the Government of the day would have liked.

Q79 Paul Flynn: Could we look at the effect of cuts on jobs? Will you remind me of the number of jobs that were originally expected to be moved from this area to Newport, and the number of jobs that actually were relocated?

Sir Michael Scholar: I have not got those figures at the front of my mind.

Jil Matheson: The plan was to move 700 ONS jobs from London-mainly to Newport, but also a few to the Titchfield office. That relocation programme is now complete. There have not been 700 jobs because of the intervention of efficiency savings, cuts and so on.

Q80 Paul Flynn: How many have there been?

Jil Matheson: It is hard to say, because there have also been changes in the Newport staffing, but it is something like 500.

Q81 Chair: What percentage of the statistics work force is that?

Jil Matheson: Of the total? There is a total of just over 3,000 staff, including 1,000 interviewers, all over the country.

Chair: So it is about 15% of your staff.

Q82 Paul Flynn: Are you expecting any job losses as a result of the 20% cuts?

Jil Matheson: There will be a reduction in staffing in the ONS, and that is partly because of the run-down of census operations, so there is a big buildup around 201112.

Q83 Chair: So it is a cyclical thing.

Jil Matheson: Yes. All those people are on fixed-term appointments, but the total number of jobs will go down.

Q84 Paul Flynn: What kind of percentage?

Jil Matheson: I have not got those numbers, but I can let you have them. There will be a reduction in the total numbers. We think, certainly for Newport, that we certainly ought to be able to manage the reduction as a result of the spending review without compulsory redundancy, and probably without redundancy.

Q85 Paul Flynn: You obviously have a responsibility to defend the jobs of your employees, but do you think you have a responsibility for continuing the policy that was followed by all Governments up until recently of it being desirable to relocate jobs from the overheated south-east of England to places such as Newport, Scotland and so on?

Chair: Newport being Mr Flynn’s constituency.

Paul Flynn: If the cuts are being imposed that way, is it likely to frustrate that regional policy that has, as I say, been pursued and accepted as desirable by all Governments?

Jil Matheson: I do not want to comment on overall policy. For the ONS, our relocation programme is complete. We have now moved all the posts that are there to be moved, so in London we are now down to fewer than 50 staff, which is what we said we were going to do.

Q86 Paul Flynn: I do not want to put words in your mouth, but how successful has it been? How has it gone? Is it something that you think was worth while and beneficial in every way? There was a lot of criticism at the time that this would be very damaging.

Jil Matheson: Yes, and very difficult for a lot of individuals who had to make their choices. However, I think that the ONS ought to be very proud of the fact that it has managed to maintain the production of its key statistics, that it has transferred that work, and that it has been able to recruit and develop, and to provide opportunities for staff who were already in Newport and for new staff.

Chair: Mr Flynn, we do want to get on to the terms and conditions of the new Chairman of the UKSA, but can we just tidy up other areas first? Mr Hopkins, do you want to ask anything more about cuts in the service?

Q87 Kelvin Hopkins : A general point about cuts, but before that, we are very much your Select Committee, so may I say that even though you may think that public confidence in you has not increased, I have great confidence, particularly since Sir Michael has been Chair? I am a statistics anorak; I have studied it and indeed taught it. I think it is important that we say we are your Committee in the same sense as we are the Committee for the ombudsman. From time to time we have had to defend the ombudsman and her reports, and hopefully we can do the same for you as well.

Sir Michael Scholar: Thank you very much.

Kelvin Hopkins : However, there is some concern that we are losing valuable time series through the cuts being imposed by the Government. Can you assure us that we have not lost time series-I know you mentioned one or two, about health in particular-that we should really keep?

Sir Michael Scholar: Well, no, I do not think we can give you that assurance. What we can say to you is that we will report to you on the matter. We have begun a series of Statistical Expenditure Reports. Although the Government have not been willing to bring us into the process of cutting expenditure on statistics, we have decided that we will do what we can through the good offices of Jil Matheson giving as much information as she can to us. Where we have concerns about the discontinuing of a particular series or the dismantling of some statistical capability in some Department, we will write a report and submit it to you for your attention. We did so recently on the Citizenship Survey, where as you perhaps recall I wrote to Mr Pickles and suggested that he think again before cutting that survey. He has not agreed, by the way-he has rejected our request.

Kelvin Hopkins : Perhaps we can take that up, Chair. I wanted to come to some specifics on the RPI and CPI.

Q88 Chair: We will come to that in a moment. Can I just ask you, on cuts, have you been able to help the Government to find reductions in spending among the statistical service?

Sir Michael Scholar: There is a survey called the Place Survey that the Government have decided to discontinue, and we are about to produce a report on that. I think it is likely to say that we think that that is a justified cut, because on a cost-benefit basis, one can see why the Government would want to save money there.

Jil Matheson: I think there are two things. There have been things that have been cut, and some of those decisions have been made on the basis of a good understanding of the impact of that. The other thing I would say is that there is opportunity. Part of this is finding opportunities to do things in different ways, and there are quite a lot of initiatives, for example, in seeing whether there are administrative data that could be used instead of going out and collecting data again from businesses or households. There are therefore innovations in the way in which data that already exist are used as part of the programme, both within the ONS and across Departments.

Q89 Chair: On the question of resources-and, indeed, public confidence-how is the decentralisation of statistics going? In particular, we know that the crime survey has been moved out of the Home Office. Is that a cheaper way of doing things? Is it going to improve public confidence?

Sir Michael Scholar: I do not think that it will be undertaken in order to achieve an economy; I think it will be undertaken in order to improve public confidence in crime statistics. I think that is why the Home Secretary has announced that she wants that to happen. Jil Matheson is at the moment working out, at the Home Secretary’s request, where these statistics should go. I think they are very likely to come to the ONS.

Jil Matheson: Yes.

Q90 Kelvin Hopkins : There has been some criticism that the RPI and CPI indices not only that they are a poor way of measuring inflation, but that they have been poorly constructed. Have you got any thoughts or comments on that?

Sir Michael Scholar: Perhaps I could ask Richard Alldritt to speak on one aspect of this. The Authority recently published its assessment of the RPI and the CPI and made a number of criticisms of these two indices which, as you know, are produced by the ONS. This is another case, to go back to Mr Flynn’s point, of the Authority criticising itself. Richard.

Richard Alldritt: Thank you, Sir Michael. We have published two reports. One is a regular assessment report against the Code of Practice, and the other is what is called a monitoring brief on communicating inflation. I think the main points are first of all that the Authority has said publicly that the consumer prices index needs to be enhanced by information about housing costs. We recognise that both the RPI and CPI are needed; they have different histories. The retail prices index is very much an established index-it is written into legislation and so on and is needed for those purposes-whereas the CPI is the internationally comparable index. We would like to see better explanation in public about the differences between the indices and the reasons for them. We have proposed there should be consultation on issues like the demand for regional indices and indices for particular household types-pensioners and so on-and generally an improvement in communication and explanation. It is not an ideal world to have two indices measuring inflation and showing rather different trends at times, and quite a lot more explanation is needed on that. There are several recommendations and there is a programme of work in hand on that.

Q91 Kelvin Hopkins : With two indices, the Government can opportunistically choose the one that is most convenient to them for uprating purposes, which has caused some political friction from time to time. Should the proceedings of the Consumer Prices Advisory Committee be more open, public and transparent?

Sir Michael Scholar: It is a committee that advises the National Statistician, so I think I would like to leave her to comment on that.

Jil Matheson: Yes is the short answer. It is a committee that brings together lots of experts who advise on the priorities for development and also discuss the technical composition of the indices. I am aware that there has been a request for the committee to operate more publicly. We have already started doing that; we are now publishing papers, minutes and so on, on the ONS website so that we can respond to the interest that there is and encourage people to be interested in what we are doing.

Q92 Kelvin Hopkins : There is another factor in this, of course: international comparisons. It is very important to have a genuine measure of real inflation in each country so that we know we are getting our macro-economic policies right and so on. Do we look at the international comparators?

Jil Matheson: Yes. Indeed the CPI is the one that is determined by international standards. We get reviewed internationally, too, to make sure that we are compliant with the way other countries measure.

Kelvin Hopkins : A flippant comment, if you like. You suggested that the Government economic service had mushroomed and yet it is constantly incompetent. It has made a complete mess of our economy over several decades, in my view, and special advisers have been mischievous as well as incompetent. Why do we not suggest that the Government reduce the size of their economic service and spend a little bit more on the statistical service? I would do that.

Q93 Chair: Can we move on?

Sir Michael, this session has underlined the importance of the work of the Authority and indeed the importance of your role as Chairman. But, to be blunt, you were originally appointed on the basis of three days a week at £150,000 a year. That was subsequently reduced to £100,000 and it is proposed by the Government to reduce the time commitment to two days a week and reduce the salary to £57,000 a year on the basis, as Francis Maude told us, that people do not take public sector jobs like this for the money. Do you have any comment on this?

Sir Michael Scholar: I was concerned when I heard that the advertisement for my successor would be in terms of two days a week and £57,000. I was concerned that it might have an adverse effect on the quality of the people who put themselves forward. I have no idea about that, because I am not part of the process. I was also concerned that it would be seen as an indication of the Government’s regard for the Authority and that it would be taken as a sign that the Government thought that the Authority was not really very important.

Q94 Chair: So what in fact motivated you to take the job? Would you have taken it at £57,000 a year and two days a week?

Sir Michael Scholar: Very hard to say. I was motivated to take the job for the kinds of reasons I explained in reply to earlier questions from Mr Halfon and Mr Flynn. I felt that the Statistics Act gave us the opportunity of turning the tide or moving against developments in government that I personally deplored, and I believe that that has been something that we have been able to do. But I was also aware when I put my name forward that it was going to be a very significant job-a significant commitment of time and energy. It is always hard to know what role money plays in such decisions, but I was concerned to hear that the salary was going to be £57,000.

Q95 Chair: Of course, this is to link the salary to the Prime Minister’s salary. Will Hutton’s fair pay review stated, "The Government should refrain from using the pay of the Prime Minister or other politicians as a benchmark for the remuneration of senior public servants, whose pay should reflect their due desert and be proportional to the weight of their roles and their performance." To either of the professional statisticians sitting either side of you: is the Prime Ministerial benchmark a suitable way of deciding a salary of this nature? Or is that a little bit too hot to handle?

Jil Matheson: I think that is for Government to decide.

Q96 Chair: Mr Alldritt, in your professional opinion?

Richard Alldritt: I agree with the National Statistician on that point.

Chair: But I think we have it in black and white from an independent source.

Paul Flynn: It is clearly a downgrading of the post to reduce it in that way. We are trying to cope with this selfdescribed Maoist egalitarianism of the Government of everyone cut down to the same level, presumably wearing blue boiler suits in the future. We have the evidence from Francis Maude, who told us that money did not matter in jobs, which may not be the case. Those who are the sons of former Cabinet Ministers are in this world where money does not matter, but in the real world money is a major determinant in people’s jobs and accepting jobs, even at this level. To advertise it at £57,000 when previously it was £150,000 suggests that you do not believe the job is that important. With the other evidence coming in about prerelease and so on, the worry must be there that the Government do not have their heart in ensuring that the Statistics Authority does its task in the future-they want to undermine it, perhaps.

Q97 Chair: Would this salary put you below the salary level of any of the special advisers?

Sir Michael Scholar: It is very considerably below the level of a number of special advisers.

Q98 Chair: Prorated at two days a week?

Sir Michael Scholar: Well, I do not accept that this job should be described as a two day a week job; I believe it should be described as a three day a week job.

Chair: Well, can we go on to the time commitment, which is a very important question?

Q99 Alun Cairns : Sir Michael, we have heard what you have just stated, but Francis Maude told the Committee that you had asked for the days to be reduced from three to two. How can I reconcile both statements?

Sir Michael Scholar: When I was appointed, it was envisaged that after a setting-up period-when the work of setting up the Authority had been accomplished-the time commitment might go down from three to two days a week, and it was envisaged that at that point there would be a discussion with the Cabinet Office Minister about whether I should be on three or two days a week. What happened was that before the election, under the previous Government, the then Cabinet Office Minister decided, against my wishes and against my advice, that my time commitment should be reduced from three to two days a week, and my salary reduced commensurately.

Q100 Alun Cairns : Can I be clear on that? You asked the previous Cabinet Office Minister to reduce it?

Sir Michael Scholar: I asked that it should not be reduced.

Alun Cairns : Should not be reduced.

Sir Michael Scholar: Because I had experienced no diminution in the amount of work, time and energy that I had to put into the job over that period of time.

Q101 Alun Cairns : So the statement by the current Cabinet Office Minister is inaccurate. Is that right?

Sir Michael Scholar: If that is what he said, it is inaccurate.

Q102 Alun Cairns : Thank you very much. Can I clarify that you still think that the job should be three days a week? Do you recognise that the job has changed from setting up the Authority now to running the Authority?

Sir Michael Scholar: The job has changed, without any question, but the time commitment has not changed. It is a complex job. These statistical matters require a considerable amount of attention and care. There are lots of bear traps and it is necessary to put a good deal of time into this job.

Q103 Alun Cairns : Finally, then, what are the main challenges for the next three years, and what skills would your successor need to deliver the objectives?

Sir Michael Scholar: I think we have covered some of the important challenges. I think that the whole question of the future of the census is a very important challenge for my successor in doing all that can be done to make progress to a more efficient and a more uptodate record of the population in the way we discussed earlier. I think that would be very important. I think it would also be very important to try to make progress on the three points that I put to the Prime Minister: prerelease access; a stronger central role for the National Statistician; and some strengthening of the Authority’s oversight of the system as a whole.

Q104 Paul Flynn: How do you suggest that we approach the preappointment hearing when we get the candidate along and ask her or him what motivated them to join? How can we reach a decision on whether, if the salary was being paid at the full rate as it was before, it would have produced a broader range of candidates and perhaps more independent candidates, rather than possibly a list of candidates who are secondraters?

Sir Michael Scholar: I can imagine that that will be a significant task for the Committee. I recall the hearing that the Treasury Committee held when I was appointed. To go back to my opening remarks, I think that the Committee was particularly concerned about whether I would act in an independent way. I think it was concerned, because of my previous background as a permanent secretary, that it would all be very cosy. I imagine that when you interview my successor, wherever he or she comes from, that will be a point that will be pretty much in your minds.

Kelvin Hopkins : There is a very strong and admirable sense of public duty and commitment to the public interest in everything you say, your demeanour and so on, but that should not be taken advantage of by paying below the rate for the job. Indeed, if we are fortunate enough to find someone who is your equal to replace you, we will have to pay the rate for the job, I would suggest, and not take advantage of that commitment to public duty. I think it is very important.

Chair: Well, Sir Michael, unless you or either of your fellow witnesses have any further comments, we shall conclude what has been a very full evidence session. I would like to take this opportunity to pay tribute to your work as Chair of the Authority and to thank you for your exemplary public service. I do not think that anyone can possibly say that you have not acted with fearless independence-to the annoyance of some of the establishment. It is always best when members of the establishment defect, in the way that you have, to the antiestablishment, and I think that the nation is in your debt for your tremendous service. As we evaluate your successor in our preappointment hearing, I can assure you that you have set a very, very high benchmark for that person. You have also created a context for our scrutiny of this that is quite exacting for us as well-we appreciate that-so thank you very much indeed and to your two fellow witnesses.

Sir Michael Scholar: Thank you very much.

Prepared 14th June 2011