Public Administration and Constitutional Affairs Committee 

Oral evidence: The work of the UK Statistics Authority, HC 524

Tuesday 14 November 2017

Ordered by the House of Commons to be published on 14 November 2017.

Watch the meeting 

Members present: Mr Bernard Jenkin (Chair); Ronnie Cowan; Paul Flynn; Mr Marcus Fysh; Mrs Cheryl Gillan; Dr Rupa Huq; Kelvin Hopkins; Mr David Jones; Sandy Martin; David Morris.

Questions 1 - 101

Witnesses

I: Sir David Norgrove, Chair, UK Statistics Authority and John Pullinger CB, National Statistician and Chief Executive, UK Statistics Authority.

II: Sir David Norgrove, Chair, UK Statistics Authority and Ed Humpherson, Director General for Regulation, UK Statistics Authority.

Examination of witnesses

Witnesses: Sir David Norgrove and John Pullinger.

Q1                Chair: I welcome our two witnesses to this session on the UK Statistics Authority and could I ask each of you to identify yourselves for the record, please.

Sir David Norgrove: I am David Norgrove, Chair of the Authority.

John Pullinger: I am John Pullinger, UK National Statistician.

Q2                Chair: Thank you very much for joining us today. We have divided this session into parts, the first part is going to be about your role as producer of statistics and then, Sir David, you will be joined by Ed Humpherson to discuss your regulatory role. We have very deliberately tried to separate in our minds the two different functions of the UK Statistics Authority.

We have quite a few questions to get through and I will, if I may, encourage my colleagues to ask open questions but short and crisp questions and we will ask of you short and crisp answers if you can manage it and we will get through the business as rapidly as we can.

I will start by asking what has preoccupied you most, Sir David, since you started your job?

Sir David Norgrove: I suppose my first task has been to try to understand what has been going on and what are the key issues in relation to each part of the authority. I have sat in terminal 3 clicking and counting passengers for the international passenger survey. I have been twice to Titchfield, twice to Newport, where I met Mr Flynn, Cardiff, twice to Belfast, Edinburgh and so I think that has been a large part. In terms of preoccupations, of course, there is the normal work of the authority, the board meetings, regulation committee and so on. I do not particularly want to pick out a lot from what has been happening over that time, but I suppose on the production side, the two things that stand out for me are, first of all, supporting John in terms of the ending of pre-release access and the second would be work to clarify and make sure that all the work streams of the transformation fit together and are going to deliver.

Q3                Chair: So it has been administrative and learning but are there any particular issues?

Sir David Norgrove: I mentioned a couple. Of course, there are a huge number of issues that have been coming through about RPI and CPI and I am sure we will come to those.

Q4                Chair: What changes have you made in your first few months?

Sir David Norgrove: I do not think it would be for me to claim credit for any changes. I do not think that is the way that an organisation works best. I have been working with the team to help them to deliver the objectives that have been set out.

Q5                Chair: What changes do you think have been made while you have been the chair, in the first few months?

Sir David Norgrove: We have had redesignation of CPIH, which I think is an important advance. That has been in the works for quite some time. The work on the delivering of census has been important. I think we have a much better understanding now of the way in which the transformation of the census will support the transformation of the organisation as a whole on a continuing basis. Those would be the two main things that I would pick out.

Q6                Chair: You are coping with a 19% reduction in your budget by 2019-20 and you have reported you are on track to achieve this, but how are you achieving this given that you need to increase both your analytical capacity and also the number of your staff in London?

Sir David Norgrove: Of course that is the core budget—and I will hand over to John in a second as the accounting officer—and alongside that we have huge expenditure on the census. One of the main changes this time compared to the previous census is that the investment in delivering the census will give continuing benefit to the organisation. That is over £900 million of expenditure, of course.

John, do you want to add anything on that?

John Pullinger: The financial profile that we have over this spending review period up to 2019-20 includes the 19% reduction in our core business, as David has said, but it also includes three other elements. The census is one, and by far and away the biggest. Unlike many other organisations we were in the fortunate position of being able to make the case to the Treasury that if we were able to make those efficiency savings we could recycle them during the course of this spending review into two other activities.

The first is what we are describing as the reform of our statistics, so putting more money into questions like immigration and crime and productivity. Since the referendum result, trade has been added to that list. So doing better statistics we have some money for, if we can make those savings.

The final element is to transform the business, particularly in relation to use of technology. So we have a 19% efficiency reduction in our delivery of core statistics, extra money for reforming those statistics, extra money for transforming them and the census. I do not feel that I am in a position of having to find savings and not be able to make good use of them. But, as anyone who is responsible for taxpayers’ funding, we need to drive down the costs of our core business whenever we can.

Q7                Paul Flynn: I was very glad to see the new data science hub in Newport but there is still a concern that there is a London-centric magnet operating that has constantly sought to draw jobs back to London. I do not know what your feelings are on this but I hope that decisions—and Government Ministers seem to support this—are balanced with the advantages to an area like Newport and the joy of living in nice surroundings in the green fields of Grange rather than the air-poisoned, congested hell of London. What is your feeling on that?

John Pullinger: Newport is and will remain our headquarters. Most of our investments are going into Newport, as you have seen from your most recent visit. The Data Science Campus is something that is an attraction not just for our office but for Newport and the whole of south-east Wales, and we are determined to make the best use of it.

We are certainly trying to significantly professionalise the skills that we have in Newport so we have increased, in my time, the number of economists from 30 to 100 and the vast majority of that investment has been in Newport. I have no fear at all about the future of Newport and this is a Newport-centric organisation rather than a London-centric one.

Paul Flynn: You quell my fears, I am grateful for that answer and they will be rejoicing on the streets of Newport, I am sure, when they read the minutes of this meeting.

Q8                Chair: Moving to the whole question of pre-release access and your decision to end pre-release access to all ONS official statistics, how significant do you think this change is?

John Pullinger: It is an important change, as an authority and for me personally. Previous to my current role I was President of the Royal Statistical Society and have always argued that fairness demands that everyone has equal access to statistics. Coming into this role, however, I have a very precise statutory duty, which is to balance that argument, which is clear, against arguments that Ministers would wish to make statements at the time a statistic is released or would need to take action when a statistic is released. So as I have approached this in this role, I have tried very carefully to weigh those two things.

Now, during my period in this role I have carried out quite a substantial review that resulted in a reduction in release but having completed that review, looked at the outcome, and weighed the arguments on the one side in relation to the benefits to Ministers of having pre-release against the disbenefits to integrity of statistics, my conclusion in that role was that I should withdraw it, which I have done. As a matter of process it has gone smoothly but I know that in doing it I have made the lives of people who are very much involved in accounting for their actions more difficult—in this place, for example, Ministers are required to respond very quickly to statistics—and my statutory duty is to weigh those. But my conclusion is that it was right for me in relation to ONS statistics to take away pre-release access. I have done it. You asked how it is going; it has gone very smoothly.

Q9                Chair: It was a decision solely for you?

John Pullinger: Yes.

Q10            Chair: Of course, this Committee is absolutely delighted you have made this decision. Since Mr Flynn and Mr Hopkins were on this Committee, this Committee has been calling for an end to pre-release access. Why do you think it has taken so long for the National Statistician to take this decision? It was not you, of course, you are quite fresh in the job, but it begs the question as to whether other people have acted independently.

John Pullinger: I don’t think it does because, as I said earlier, the pre-release access order places a requirement on the responsible persons—in relation to ONS statistics it is me, in relation to other statistics it is the relevant Minister—to weigh these two things. There is a weighing in the scales and it depends on how much you value one against the other. I have given an explanation of the two arguments and at the time I made my decision I concluded it was right now to stop it, which I have done.

Sir David Norgrove: I might add, of course it has also been salami sliced over the last 10 years, so the number of people having pre-release access has steadily gone down and the length of time that they had it for has also been reduced. It is possibly easier to make that final step than it would have been 10 years ago.

Q11            Chair: What has been the reaction in the Government and among the civil servants to this decision?

Sir David Norgrove: My perspective is mostly very positive and I am glad to say that the Bank of England has followed suit and the Scottish Parliament are looking at it in relation to Scottish statistics. Of course, there will be some people who are disadvantaged by it who will not have been so happy about it.

Chair: It is quite something to be ahead of the Scottish Parliament.

Q12            Dr Huq: I wanted to ask about improving the sharing of data. I remember a time when 25 million child benefit records used to get lost on a floppy disc and things like that. There was one Government briefcase, I think, that was lost in a supermarket in the 1990s in my now seat when I was at university. Hopefully none of that ever happens again.

What successes and failures have you had in the last couple of years in improving the sharing of data, particularly administrative data, if any?

John Pullinger: Our first duty is to safeguard the data that we have, both from individuals and from businesses. This is sensitive data and we need first of all to respect that. So a very significant priority for me and for the whole office is to think about our security at all levels and design that into our systems. We need to care about that.

There is huge potential, having put those processes in place, to use data in much more creative ways. The census is the biggest example, but across the range of economic and social statistics, so it was a great result for the statistical system when the Digital Economy Act was passed in April of this year. It enables the statistics authority to have access to public sector and private sector data for the use of statistics. This is data where the individual identities have been masked by aggregation. The opportunities for that are very significant. We are looking to exploit them already.

A consultation has just closed on the codes of practice that will ensure we meet both our legal obligations under the Data Protection Act and in future under the general Data Protection Regulation, but also have processes that are transparent and give the public confidence that we are caring about the kinds of things that you have already mentioned. We are now looking at the responses in that consultation. We are confident that by the beginning of next year, which is when the Act comes into force, we will be able to utilise it very significantly for the public benefit.

Q13            Dr Huq: Any other tangible benefits from the Digital Economy Act?

John Pullinger:  It is a bit early yet because it has not quite come into force but I have given you an idea of some of the things that we have already started to do, working particularly with Government Departments in anticipation of it coming into effect.

The first one relates to migration statistics. The last time I came before this Committee I spoke to you about the potential, particularly of data from the Home Office but also from the Department for Work and Pensions and Her Majesty’s Revenue and Customs. We have systematically made improvements to immigration statistics in the most recently released, using Home Office data to understand patterns of students returning home after their studies, which has been a significant concern of the higher education industry.

Q14            Dr Huq: Do you have an opinion on whether they should be counted in migration targets? A lot of people think it just seems silly to conflate them, that they are not migrants.

John Pullinger: It is entirely a matter for the Government what targets it chooses to set itself. It is our job to provide the numbers to enable a good debate around which will be the most suitable targets or not.

Q15            Dr Huq: What new statistics have been published as a result of access to data?

John Pullinger: I have mentioned that there is this particular information on exit checks data on immigration. Another example I would give where we are currently producing experimental statistics is on the flow of funds around the banking system, where we have a joint project with the Bank of England to understand financial flows and financial risk. Another example, which will be coming next month, is including VAT data in our main estimates of GDP.

Q16            Dr Huq: You said in the opening that the new stuff being done to the census will be transformational. Can you tell us more about that? We have heard that there is a reimagining going on. How is all that going?

Sir David Norgrove: My sense is that it is going well. We still have a way to go. The test that was carried out in the spring was productive and worthwhile and there will be another one in 2019, but of course at this stage it is about, as much as anything else, putting in the foundations in terms of the IT, the processing and the thinking around that, which is the kind of foundation that is also going to carry on being of value to the organisation as a whole after 2021 and in the run up to 2021. It is demanding. I think John, at this Committee in the past, has said that the organisation is skating on thin ice because it is trying to make these major changes at the same time as continuing to deliver so many statistics day by day, month by month, and doing those two things in parallel is demanding. There is a lot to be done but I think it is making good progress.

Q17            Dr Huq: Is it still going to be total enumeration, not a sample, of the entire population?

Sir David Norgrove: Yes.

Q18            Dr Huq: Is Jedi Knight going to be a category?

Q19            Paul Flynn: Absolutely. One of the justifications for this extraordinary decision is that there are major matters to decide in this country on public opinion, on the basis of a fraction of 1% of the population. The ONS seems to be unique in thinking that we need a 100%, biblical view of the census on this, whereas that itself has its own distortions. Couldn’t you relieve the strain on your Department and improve your finances if you decided to have an intelligent representative 1% of the population in the census, rather than continue, for the last time, with 100%?

Sir David Norgrove: I think I gave my view of that when you asked the same question at my consultation hearing. The National Statistician is better equipped to answer that than I am.

Q20            Chair: Mr Pullinger, do you want to add to Mr Flynn’s point?

John Pullinger: I will if you wish me to.

Q21            Chair: Is there anything new to add?

John Pullinger: No.

Q22            Kelvin Hopkins: We have heard on several occasions in this Committee that the census is something that many of us regard as very precious and that the loss, the possible loss, of the historic and valuable time series would be tragic and I would hope that we will preserve those time series so we have a picture of how our society has changed and is changing over time. I think it is extremely valuable that we keep that core value of the census.

Sir David Norgrove: I think I speak for both of us in saying that we agree completely.

Q23            Mr Jones: Mr Pullinger, could you please outline for the Committee the impact of the General Data Protection Regulation on the activity of the GSOs?

John Pullinger: Yes, certainly. The General Data Protection Regulation clearly affects everybody who is a controller of data and we are working through in great detail to ensure we compile fully with the provisions. But at the most strategic level the regulation reflects quite closely what already exists in the UK Data Protection Act, which gives a special place for statistics, where our interest is not in the identity of individuals but is in aggregating the data from those individuals to produce anonymised outputs.

So in that respect the General Data Protection Regulation will not have a major impact on our operations in the way it will on many other organisations because there is a very special set of provisions in there for us. Clearly there are a number of detailed things we need to work through, including the precise roles of the data protection officer and how that will carry forward but we are not expecting it to cause a significant difficulty.

Q24            Mr Jones: Of course, the GDPR will effectively be perpetuated after Britain leaves the European Union?

John Pullinger: Yes.

Q25            Mr Jones: Sir David, about six or seven years ago there was great Government enthusiasm for data. We had open data, data.gov, but this enthusiasm seems to have fallen away of late. What is your take on the Government’s commitment to data at present?

Sir David Norgrove: I do not think I can answer for the Government as a whole. I can really only answer for the authority, ONS and OSR, that our objective is—and it is clear in the code, in both the old version and the new one—that data should be made as freely available as possible. I have seen for myself the work that is going on to achieve that, particularly as we move into greater use of administrative data. However, at the same time, privacy has to be protected and the more detailed the data are that you make available, the more risk that you run that individual people can be identified. A lot of effort goes on to make sure that those individual small cells of data cannot be attacked by, for example, using other sources like Google to identify individuals.

Mr Jones: It is a question of striking a balance, I suppose.

Sir David Norgrove: Yes.

Q26            Mr Jones: You will be aware, of course, that in October the Institute for Government and CIPFA were critical about the Government’s deployment of the data. They said that simply having the data in place is not sufficient, that the Government needs to make good use of it. Would you like to comment on that?

Sir David Norgrove: I do not remember that comment but if they said it, I would be interested to see the evidence. But, clearly, it is our role to encourage the Government to make data available as far as they possibly can.

Q27            Mr Jones: They said that although the data was available, it was still difficult to find basic data on key services, in particular the number of GP consultations conducted across the country.

Sir David Norgrove: This is really coming on to work that is being led by the Office for Statistics Regulation and my suggestion would be that we pick it up then. There is work going on on the ONS site, which perhaps John would like to comment on, but the OSR has been leading a range of collective consultations, so to speak, and discussions about groups of statistics, including in health.

Q28            Mrs Gillan: There was a launch of the CIPFA-Institute for Government performance tracker in October and a former Cabinet Minister, Nicky Morgan, remarked on the fact that Ministers do not often know what is going on as far as what data is and is not available. How can you help to better inform the ministerial cadre?

Sir David Norgrove: Clearly this is role not just for statisticians but also for the other analytical communities within Government. As I know from visiting and talking to various Departments there is, I think, a creative drive happening to create an analytical grouping within each of the measured Departments to bring analysis and data together in a more effective way. It is our role to help support that. John is head of the Government Statistical Service and perhaps he would like to comment from that point of view.

John Pullinger: Yes. Following up on what David has just said, I am head of the Government Statistical Service and I work very closely with my counterparts who are heads of the economics profession and the social research profession and the operations research profession and a number of other similar allied groups to try and do exactly what you are saying, to help the Government improve the evidence base and improve its ability to use it. A very specific initiative that we have been taking forward over the last couple of years is working with the Government policy profession to help them include within their curriculum for all Government officials working on policy a better understanding and appreciation of evidence to help deliver exactly that kind of capability.

Our view is if we can build it up much more strongly across the civil service in all of its forms then there is a better chance that evidence will be coming to the ministerial table.

Q29            Mrs Gillan: How many Ministers have you talked to about this?

John Pullinger: The ones I have spoken to about this have been the Cabinet Office Ministers primarily.

Q30            Mrs Gillan: How many?

John Pullinger: Four.

Q31            Mrs Gillan: Okay, so you are involved in a major exercise across Government to make sure that Ministers and Government Departments use statistics better. I cannot remember how many Ministers there are—considerably more than four—but you have only spoken to four and yet that is the group of people that you are trying to help?

John Pullinger: What I am doing is working through my colleagues in the policy profession and each of the permanent secretaries in the ministries is representing that profession, and they are speaking to their Ministers on a daily basis. What I am doing is a kind of an intermediate role, if you like. I am helping them improve their own capacity and they are the ones who are then using that capacity to support their Ministers.

Q32            Mrs Gillan: Every time you add a layer between the direct customer and the ultimate consumer of that, that is where the problem arises. Surely it would be better to have started with the people that need to be aware of what is available and how to use it?

John Pullinger: I have started with an opportunity, which was that the policy profession were looking at curriculum for all policy advisers coming into Government. Certainly some members of the Committee will recall when I was Librarian of this House, we did a lot of work with members specifically working on statistical literacy and I would be the first to say that the best way of getting any kind of personal skill is to talk directly to the individual. I am just saying I have not done that in this particular case.

Q33            Mrs Gillan: Is it something that you would consider doing and making sure that your team engaged better with the frontline of Ministers on this?

John Pullinger: I would certainly do that but I would want to do that in partnership with my policy colleagues to make the most of Ministers’ very precious time.

Q34            Mrs Gillan: One of the problems is that to get any meaningful data out of the system it crosses so many different areas; it is not just your office. On education you are trying to find out about the passage of a child through the education system and how successful we are; we have everything from the HESA to Ofsted to the HEFCE, the Student Loans Company and so on. That data is quite hard to find and quite hard to pull together. How can you help not just Ministers but people like me as a Back Bench constituency member to understand what is happening in my constituency, say, for example in the success of our education system?

John Pullinger: The first thing to say is I think it is our job to help you understand your constituency and to help Ministers make their decisions, to start from the perspective of the questions you are asking and then drive that back into the system. The main mechanisms that I use for that is my role as head of the Government Statistical Service and my role, as David mentioned, of convening the other analytical professions so that we move from a position where you have to work out effectively which dataset to use to one where we think what are the data that are going to be most salient to the question you have and we bring it together in a form that makes sense to you.

Now, I was very blessed working in this House with the way the House of Commons Library operates; it always instinctively thinks from the Members’ point of view. I am trying to build that culture back into Government Statistical Service. It is a more problematic question because you have so many different Departments and different agencies but I will return to the question of immigration.

I know it has been of great interest to this Committee as I have already spoken about. What do we need to know about different populations coming into this country and leaving it? There is a series of social and economic questions and we are trying to put together a jigsaw of that information. The international passenger survey is one part of that jigsaw. Home Office data—people coming in and going out—is another one. There are tax data on who is paying tax, benefits data on who is getting benefits. We have published a plan that has the elements of that jigsaw and which bits of it are going to be most hard to fill. We are working systematically around the system to try and think where we can get this data.

On the student immigrants question, we have been working with universities to understand their alumni departments, for example. We would not have done that a few years ago. We are just thinking where we can find the knowledge that would help you answer that question. But it is a long call. David mentioned in one of his answers that we are trying to keep the show on the road with the existing statistics, because we must do, at the same time reimagining what a good statistical service looks like. For me it is being helpful; helpful to the individual Member of Parliament, helpful to the Government, helpful to British business and anyone else who relies on our information to make important decisions.

Q35            Mrs Gillan: Do you have any statistics on how many MPs have approached you directly for information?

John Pullinger: I answer many hundreds of parliamentary questions every year. Simply because of my history in this place, I know a number of Members so I would come and talk to Members here and some do approach us directly. We are a service for you and the other Parliaments of the UK, and part of our set-up as the authority is that we report directly to you and therefore are available to you. So I and my staff make sure they are available to Members should they wish it.

Q36            Mrs Gillan: How do you promote yourself to Members of Parliament? For example, has the whole 2017 intake had a communication from you telling them what you can do for them?

John Pullinger: I spent many years in this place involved in the induction of new Members and I am conscious that the time constraints and the way things work means that is quite a challenging task. Typically what I would do is work with my colleagues in the Library who have the relationships with you and they are jointly marketing, if that is the right word, the fact there is a service available through them to us, or directly to us, to make sure you get the information that you need.

My primary vehicle is through that route and we held a joint event in the Attlee Suite a few months ago very specifically to bring together people from a range of perspectives to help you understand where you can go for help.

Q37            David Morris: John, what could the UK Statistics Authority be doing to improve access to data outside the online services and even outside the Government Statistical Service’s website, or even other ways of getting the information out?

John Pullinger: Getting information out? That is one of my top priorities at the moment because I don’t think we are being as creative as we could be. One of the concerns of this Committee three years ago was the ONS website and the fact it was not able to deliver the statistics that you wanted. We have made a step-change improvement in that but we need to make another step-change improvement, particularly to meet the kinds of questions that Mrs Gillan has just talked about, to bring together information from across the whole piece rather than just what ONS does. Once you have that secure foundation, we could be much more creative in exploring particularly social media channels to reach different audiences.

I will give a very specific example where that has worked very well. We have a very challenging evidence base on suicides and we have worked with Samaritans, for example, to reach the kinds of communities where they are anxious about the risk of suicide and the mental health questions that might lead to suicide or else just help to understand what the patterns are, particularly among certain groups of the population.

We are thinking about the kinds of channels that other people have, particularly on social media, and targeting our efforts to reach those but it needs a foundation of us having a good product to start with and so my effort over the last two or three years has been trying to improve the product so we can work with sister organisations. There is quite a lot going on in that space, with Samaritans, Mind, NSPCC, where they have a community and if we could organise our statistics to work alongside them to get the facts at the disposal of those communities, we would have a lot more impact. It is certainly a lot more helpful—to use the word I used to Mrs Gillan’s question—which I think is the core of what our mission really ought to be.

Q38            David Morris: So you think it is possible, through some sort of social media platform, to have a portal that people could access to see a pastiche of what is going on at a glance? Do you think that could be the way ahead in the future?

John Pullinger: That is part of it. A channel that we have introduced in the last couple of years is visual.ONS. That creates a much more tweetable source of information and it brings that data together, in a more readily publicly accessible format. So that achieves what you have just described, I think, as we progress that forward, which we will do, particularly when we start looking at smaller area data and particularly constituency type data as far as Members of this House are concerned. We could make that much richer.

What I am saying is I think we could do better than that if we create partnerships with a range of other organisations that have particular audiences, then we are operating on the platforms that they deal with. Again, I am borrowing from the experience that I had in this House, working with several Committees and the one that I guess was most formative with me was when the Treasury Committee was launching an inquiry into the indebtedness of households. We did a partnership with MoneySavingExpert.com, which had several million subscribers, and to have the Committee Chair and members of the Committee working on that platform enabled the Committee to have a richness of evidence that you would not get if we operated from parliament.uk, which I also was responsible for when I was here.

I am trying to replicate that in the statistical space. You will have bigger reach if you operate from other people’s territory.

Q39            David Morris: What you were just saying about drawing information from other people and exchanges, what conversations have you had with—for instance, you hear all these claims about food bank usage. Do you have an interface with any of the major food bank organisations and, if so, would you be able to verify the figures? How would it work?

John Pullinger: The short answer is, yes. Certainly in relation to food banks. Robert Key, who was Chair of the Information Committee here, was Chair of the Trussell Trust and I have certainly spoken to him on a number of occasions about this. We are looking at how we can integrate information from food banks and others into our official statistics. The question of verification is probably more a question for Ed Humpherson in terms of thinking about what our role is in validating information from beyond official statistics.

I am interested in trying to get good knowledge that people can trust on any subject of public interest, including food banks as one of those.

Q40            Ronnie Cowan: I think I heard that the Trussell Trust is saying that in my consistency food bank usage has gone up 70% since the introduction of Universal Credit. Would I be right to believe that figure?

John Pullinger: At the moment we don’t have a verification system. We have the figures from the Trussell Trust. They are clearly interested and keen to make sure they are trusted so that people can put reliance on them, but we are not there yet. I am saying in response to Mr Morris that we are looking at that because it is something of public interest. We have not done it yet, we are speaking to the relevant people. You are saying you have a number that you do not quite believe yet.

Q41            Ronnie Cowan: No, I do believe it; other people do not believe it.

John Pullinger: People do not believe it? Okay. Then it is in all of our interests to try and work out a way of ensuring that there is trust by the public in these sorts of statistics of public interest.

Q42            Mrs Gillan: How do you go about getting that data? In my experience people have to use a food bank because of all sorts of different reasons and the last thing you want to do is interrogate them at the point they are using a food bank or at the point that they are recommended to use the food bank. How would you go about collecting that data accurately?

John Pullinger: The simplest way of describing the strategy we have for all these difficult areas is thinking of a range of ways of approaching the problem. So the Trussell Trust has the information they have got from their own volunteers and from what is going on in the individual food banks. We have information from our surveys. So the surveys that we are conducting will be talking to people about their experiences, their living conditions and there is also potential for us to gather information that way. We would also be looking at any other opportunities that we have to gather data. This is the potential for the Digital Economy Act, for us to think quite laterally about how you gather the information. I agree with you that asking people is not necessarily the best way, it is intrusive, and you will not necessarily get very good results.

Q43            Mrs Gillan: Is there a danger of a false correlation and is there a danger that if people knew that this sort of statistic was being collected by them it might put them off from using food banks, which is the last thing we would want to do when people are in desperation?

John Pullinger: That would be the last thing we would want to do. I made the comment to Dr Huq about our first duty being to respect the people whose data we are talking about, so we earn their trust, so they are confident that any information we garner about them will not be used in relation to them, it will be used to inform the broader decision. My experience is when we successfully make that case it is often the most vulnerable that feel most empowered because the facts are then being used as Mr Cowan was saying.

Q44            Mrs Gillan: But you are willing to admit that this is a very difficult area to mine accurate information about; a very difficult area—

John Pullinger: Yes, it is.

Q45            Mrs Gillan: —so you cannot truly rely on this at the moment?

John Pullinger: We have to be honest about the limits to our ability to get our knowledge and be open about that. There are whole groups of populations where we really have to be careful. All I can say is that ONS as an organisation has a very good track record of doing very sensitive surveys. When we seek to approach difficult subjects, our staff are very thoroughly trained to ensure that they respect the individual’s rights and make sure that they are not trespassing in any way on their concerns.

Q46            Mrs Gillan: I suppose on a personal level I am worried that it would throw up false things because somebody might be inclined to blame one reason for finding themselves in that situation to hide the real reason so we would not be having the right statistics that would enable us to put the help and support in place that was correct for those people.

John Pullinger: That a fair concern and that is why I talk about trying to look at the subject from different angles so that we can validate the results that we are getting and therefore get the kind of trust that you, Mr Cowan and others are seeking.

Q47            Chair: Do you think it would be helpful for the Committee to have a fuller note on this subject? Would the Committee like that?

Mrs Gillan: I would like that because I think it is a very sensitive area and I am anxious about it in terms of it throwing up maybe false results or disturbing an area that is very important in some people’s lives.

John Pullinger: Of course, and we are at an early stage so the Committee’s views on that would be very helpful. We do want to go into these areas and any guidance on how we approach it sensitively would be well received.

Chair: We may well write to you with a suggestion on how you would respond, in more detail.

Q48            David Morris: John, I digressed a little with other areas, but with us leaving the European Union, how would you gather information and statistics when we leave the European Union or even during the process of leaving European Union? How would you ascertain the data that you need to give out information of any potential benefits or pitfalls of leaving the EU in general? Furthermore, how would it work out in the future when we do leave the EU?

John Pullinger: I would see our role as to give the evidence that enables that particular question to be debated, because I think it is in the eye of the beholder the weight that different individuals would put on different pieces of information. Where I think our duty is, is to focus on those areas where there clearly will be an impact and understand what is happening on the ground. There are many, many of them but there are two that really stand out at the top of the list. The first one is trade and the second one is migration. We need to enrich our knowledge base of trading patterns, both in terms of the goods we are trading and the supply chains of those goods, and of the countries we are trading with, to get that really clear so we can understand where we are now and we can track progress as we go forward. Similarly on migration, which I think I have already covered, both in terms of its social and economic impacts.

Once we have that evidence base on the table, it can be debated in this place and people can come to a conclusion about what has happened but it would not be for me or the Statistics Authority to reach that conclusion. There will never be a definitive answer.

Q49            Kelvin Hopkins: Sir David, the Government has chosen mainly to use one index, namely RPI, when charging members of the public, for example for a student loan, but another index, CPI, is used for paying members of the public, for example state benefits. If RPI is an inferior measure of inflation over CPI, how can it be justified that the Government use the better measure to increase its income but the inferior measure to reduce the cost of state benefits and payments?

Sir David Norgrove: We have made it very clear—and I have made it clear in a letter that has been published that I wrote to Chris Giles of The FT who asked about the RPI—that we did not recommend the use of the RPI for anything except in those areas where it is very difficult to change them at the moment, for example, the use of the RPI in relation to index-linked gilts. I made that clear in my letter, that we deprecate the use in those circumstances.

Q50            Kelvin Hopkins: Do you make it clear also that cumulatively, over time, compounding those changes can make for significant injustices?

Sir David Norgrove: I did not go into the numerical consequences of using RPI, but clearly that would be one of them.

Q51            Kelvin Hopkins: Yes, indeed, and it is time to park the RPI in a backwater for occasional use and to focus on CPI.

Sir David Norgrove: I agree with that, yes.

Q52            Kelvin Hopkins: A question to John. I am sure John will be pleased to know that the Library is still the most superb institution in the Commons, in my view, anyway.

You have acknowledged the need for two measures of inflation, one for macroeconomics-related uses and another currently under development, which today we have referred to as the household costs index, to measure inflation as experienced by households. When will you commit to producing the new household costs index on a monthly basis, alongside the other measures?

John Pullinger: We will produce our first shot at that index next month. We have committed to do it by the end of this year, in December 2017. Once we have produced that, I want to see the response and the feedback so that we can develop it properly. We will make it as frequent as there is a demand to have it, but I want to take it one step at a time. I have committed to producing it the first time around in December 2017, which we will, and then we will develop it from there based on what people want.

Q53            Dr Huq: Just a question about the presentation and clarity of what these crime statistics mean. There is a recent ONS blog that itself says that you need to better inform the debate around crime statistics and that in the next release any key messages will be presented more clearly. How are you going to do that?

John Pullinger: The first thing to say is we do need to raise our game in the presentation of these numbers. We have for a long time produced our quarterly release, which has the information from Crime in England and Wales, our survey, which is directed at victims, alongside the information on what is recorded by the police and really left it to the users to make their own judgment around which is the most useful for answering their questions. What we need to do now is to integrate that. It links with some of the other questions that we have here, to answer the basic question, “What is really going on?” which is what people want to know. It is only by combining these different sources that we can do that.

In our next release, we will make a step forward in bringing together the narrative around what is really happening with crime, rather than presenting—which the last one did—a series of facts on what we did determine from the survey and a series of facts determined from what the police have recorded.

Q54            Dr Huq: Yes, because there are always massive problems with this sort of thing, on reporting and different measures with BCS and then the police’s own figures, and things will always get misinterpreted as well. The predecessor Committee of this one—I was not there at the time, but it is legendary, the “Caught red-handed” report—the Committee was wondering why you do not highlight more prominently the fact that the UK Statistics Authority had to remove that designation for police recorded crime data. It was not found to be robust and independent and it was seen to be pretty lax, the reporting, because of all these targets. Why aren’t you saying that more and how do you control for these things and make them better?

John Pullinger: I make two points on that. First of all, it is clear in the release that it does not have the designation, but more precisely, the release identifies that the changes in police recorded crime are due to a number of factors. They are due to actual changes in crime, people’s willingness to come forward and the police’s ability to record those responses. The report of this Committee identified the variability in the police’s ability to do that. What we are seeing in those numbers now is the consequence of improvements, the work we have done, including at the behest of this Committee, with the inspectorate of constabulary to improve the training and the practices of those people responsible for the recording of crime. We are working very actively with chief constables and others in the forces to raise the standards across the piece. At the moment there is that variability and our release reflects it, but I am saying we need to raise our game in the way we communicate that to people.

Q55            Dr Huq: How do you respond to the suggestion by Simon Jenkins in The Guardian that they should not be by you at all?

John Pullinger: It would be better for us to bring all the data together and explain clearly what the vulnerabilities are of the different elements of it. If we did not do it at all, those numbers would still come out and the confusion for people trying to use the figures would still remain. I think us bringing something together and saying the strengths and weaknesses of each is the better way, so my response to that is we should continue to include them, but be very clear about what you can and cannot say about them.

Q56            Chair: May I just chip in here? I think you are very kind to yourself. If a public authority was publishing statistics that resulted in headlines, which said, “Crime is going up”, as a regulator you might be quite critical of that because the headlines are wrong, the headlines do not reflect what the truth is behind those numbers. You would be writing them a letter, Sir David, and saying, “This is a misuse of statistics” but when you do it yourself and do not explain things properly, you let yourself get away with it.

John Pullinger: Two points here again, if I may, Chair. First of all, the regulator has written to me saying this and Ed has written to me, so I have that letter that is encouraging me to do better. I hope my response to Dr Huq was not implying that I am happy with the situation. We need to raise our game. The line I take inside the office is that our job is not done when we publish the statistics; our job is only done when they are understood and reported properly. In this case they were not, so our job was not done.

Q57            Chair: Why don’t you just let the police publish their own statistics? Why do you need to collate all this together? Why don’t you let the chief police committee publish statistics? They still purport to be official statistics, without the capital letters, when they do not merit that authority.

John Pullinger: I could do that. In some ways, that would make my job easier, but from a public point of view, those statistics would still be produced and then the question would still arise, “What do I make of this, what the police are saying compared to what you are saying?” I think a big part of my job is to integrate these sources and explain.

Q58            Chair: When the headlines are all wrong, you feel responsible?

John Pullinger: Yes.

Q59            Chair: In future, there will be much more done about this?

John Pullinger: There will be more done about it. Whether I can take responsibility for all headline-writers in all newspapers is maybe a bridge too far, but I certainly take responsibility for making sure that our numbers are presented in such a way that a reasonable person can understand and make sense of them properly.

Q60            Chair: But it means explaining things not in little footnotes, it means putting them in a big notice at the top.

John Pullinger: Absolutely. Yes, agreed.

Sir David Norgrove: If I may, it applies in quite a number of areas. The work that is going on, for example on health and housing, to look across the piece at the statistics that are produced and how they fit together and do they cover the key things that people want to know, is a very important part of the work that is happening on both sides of the House.

Q61            Dr Huq: One of the recommendations was to have a more ethical approach after 2014. How do you operationalise that?

John Pullinger: Ethical approach?

Dr Huq: Yes, to the fact that the targets get in the way. People do not report crime in the first place, victims, but also the police want to massage the figures so their targets are being met. One of the report recommendations was that police values have maybe gone wrong and that a more ethical approach is needed. I do wonder how you operationalise such a concept.

John Pullinger: The first response to that is the work we have been doing with the inspectorate of constabulary, where they are, in their audits, working very closely with us to build a culture of accurate reporting into the system. There have been a number of follow-up reports from their audit spelling out how well they are doing with that, and their rolling audit around the system is going on now. This is an opportunity for us to work with others who have a common cause in improving the quality of reporting. We will certainly work with them.

The essence of the change that I want to bring about is to move us from a position where we see our primary job is to publish the statistics as they are, to one that provides a service for people like you and other members who want to understand what is going on. That requires us to bring together all of these sources and explain, as I said before, where they are—it is still problematic—and where they can be relied upon. In this particular case, there are certainly some categories of crime that are not very well reported, sometimes down to trusting the police, sometimes for other reasons.

I will give you two examples relating to recent releases. The first is how we can understand stalking better and looking at how we can bring out what we know and what we do not know about stalking, which is clearly a matter of public concern. The other one is around knife crime, where the police, and the Metropolitan Police in particular, put a huge effort into trying to improve the reporting of knife crime in London and you are now seeing that coming together.

Our job is to look at all of those sources, understand what you can say about them, and in our reports make that much clearer. There is a step change that we need to make there, back to your question about being kind to ourselves. I am maybe being genteel in the confines of this room, but it really is a step change that we need to make in the way we report on this, but also on other questions. If we are really to cut through, particularly in very contentious areas, and get the right headlines to be written, we need to take that responsibility.

Chair: We will move on to the second half of our session with Ed Humpherson now, please, and thank you very much, Mr Pullinger.

Examination of witnesses

Witnesses: Sir David Norgrove and Ed Humpherson.

Q62            Paul Flynn: All sensible people agree with and welcome the words of the Prime Minister last night in drawing attention to fake news and the new invisible influences through botnets, through artificial intelligence, by secretly buying advertising space that is poisoning the information that we get and distorting it, principally from America, but possibly from Russia as well. Don’t you regard it as more important than ever to make sure that we have an authoritative, independent, objective institute that we can trust above all others, and isn’t that ONS?

Sir David Norgrove: It is the authority, with ONS and OSR. If I did not believe that, I would not be in this role.

Q63            Paul Flynn: Indeed, you said this before. There is a controversy going on at the moment about the distortion of statistics and you quite rightly pointed out to the Foreign Secretary that he had repeated a claim about the gross amount that might be coming back from Europe if Brexit takes place; that he has already indicated that it could be spent on the health service. The situation at the moment—I sit on the EFRA Committee as well—is that Michael Gove has promised 40% of that to be paid to the agricultural industry and Boris Johnson has promised 150% of it to the health service or other services, so 190% of that appears to be already committed. Does this make statistical sense to you?

Sir David Norgrove: I have written about the question of the rebate, as my predecessor did, I think four times. The only thing I would say about it is that the public perception of the misuse of statistics by politicians I think is wrong. I have been surprised in this role how little statistics are misused by politicians, not how much. I have had occasion to write three times about misuse of statistics by politicians since I have been in the role. Only one of them, which was the one to the Foreign Secretary, we saw as a misuse of statistics. On the other two, which arose from complaints, we did not find that.

I think the misuse is not the kind of thing that people now see as happening, it is the use of statistics to claim that something is the biggest or the smallest or the least, when it is not a very relevant statistic, “We are spending more on X than ever before” when the number of people needing X has gone up, so it is not a very useful use of the statistic.

Q64            Paul Flynn: When we met in Newport, I showed you a letter that I had from Mrs Thatcher expressing her outrage at the suggestion that statistics could be distorted for political ends. The whole raison d’être of UKSA is to present us with reliable statistics, but if there is no acknowledgement that somebody who holds one of the great offices of state has distorted statistics, what chance is there? I have the two letters in front of me now, the simple one that you sent pointing out that Boris Johnson was wrong, and I have his reply, which is a typical rambling, blustering piece of half-truths and half-deception that we are used to from him. Do you intend to reply to his letter? I sympathise with you in having difficulty replying to a letter of this kind, but the whole truth is that an alarmist, untrue picture presented during the election, on the side of a bus, was repeated by him. There can be no greater abuse of statistics than that.

Sir David Norgrove: We thought it was right to intervene in a very clear and direct way, but I am glad that in his letter the Foreign Secretary acknowledges that he did not intend to suggest that the £350 million would be available for the NHS.

Q65            Paul Flynn: The letter was an extraordinary one. The man is a journalist. He knew perfectly well mentioning again the figure that had so much attention since the referendum was going to excite public opinion and it was going to be interpreted that way. He knows that it is the headline that matters, not the actual context in the letter. I think he is unrepentant. Isn’t this worrying as far as doing the job that you are doing and this Committee has been trying to do and the Statistics Bill is trying to do, which was to make sure there is a body of independent statistics that everyone can trust absolutely? How do we do that if one of the holders of the great offices of state undermines and distorts that figure?

Sir David Norgrove: It is not for me to judge motivations or the Foreign Secretary’s current thinking about this. It would be a matter of concern were that claim to be repeated.

Q66            Paul Flynn: You have recently changed in the Pink Book the way that gross figures and other figures are presented. Would you like to tell us about that? Would it be helpful if this had been done before to make it clear to everyone what was a gross figure and what was a net figure to avoid Mr Johnson making another major mistake?

Sir David Norgrove: I think there was clarity during the course of the referendum campaign about what the true picture is and was. ONS created an interactive graphic very early on that would allow people to understand the difference between gross and net and that has now been updated. As I say, my predecessor wrote four times about it, so I think that there was clarity about the true state of the numbers.

Q67            Paul Flynn: With this wave of fake news coming from invisible sources, people with motives to undermine Europe or bring in a right-wing agenda, this is something that was not there before; it was not there two years ago, it was not there five years ago. It is a new phenomenon. The Electoral Commission tell us they do not have any mechanism for dealing with it. We understand GCHQ might have a mechanism, but we do not have an Intelligence and Security Committee over there. What do you suggest we should do to tackle the new threat to receiving objective truths?

Sir David Norgrove: That is a very broad question. I would prefer to stick to the rather demanding knitting that I have at the moment. The encouraging thing is that people trust ONS’s statistics. In our most recent survey, they trust the independence and objectivity of the statistics, nearly 80% of people. What they do not trust is the use of those statistics by politicians and the media. I think that is more about the trust in those institutions than it is about, as I say, trusting the statistics. All we can do is to carry on trying to ensure that statistics are used appropriately.

Q68            Paul Flynn: We do not need an entirely new mechanism about election spending? There is a huge amount of spend in the referendum campaign and the two general elections we have recently had that was invisible. No one knew anything about it, we did not know what was going on, messages sent around via Facebook and various other influences that is not accounted for. It does mean that it is easier to buy an election now and a referendum result than any time since 1880. You should look to your knitting, I am sure, but you should look to create a tapestry of new policies that would guard us and restore the faith that we should have in the independence of news and facts.

Sir David Norgrove: I wish I had the levers to pull to achieve that. I do not think those are in my control, I am afraid.

Q69            Mr Fysh: Sir David, the Leader of the Liberal Democrats was quoted in The New Statesman on 16 October with a comment on the revision by ONS of the international investment position. The headline was about the missing billions, referring to £490 billion worth of revisions there. What he is quoted as saying was, “News of this massive write-down shows our economy is in real trouble”. Is that true?

Sir David Norgrove: It is a question really that arises more in the first half of the session than the second, but the net investment position of course is the difference between two extraordinarily large numbers. The revision came about as a result of relatively small changes in those numbers that produced an apparently large change in the difference. I would not myself see that as a major indicator of weakness or strength in the performance of the economy.

Q70            Mr Fysh: Yes. I think your Director General in ONS, Jonathan Athow, wrote in City A.M. a couple of weeks later that he thought that it meant the rest of the world is estimated to have slightly more invested in the UK than previously thought. Would that sort of statement by the Leader of the Liberal Democrats be a clear misuse of official statistics?

Sir David Norgrove: I do not remember seeing that statement, so giving an off the cuff remark or comment on it might not be appropriate, but on the face of it, no, I do not think so. It is, by most people’s standards, a very, very large number. It only matters if you look at it in context.

Q71            Mr Fysh: Can I take it from that answer that you did not write a letter to Sir Vince Cable about that?

Sir David Norgrove: No.

Q72            Mr Fysh: What would be the criteria under which you would decide whether to write to a senior politician about a clear misuse of official statistics?

Sir David Norgrove: We have set that out in the letter to the Chair that I think you had last week and it has now been published on our website. It includes things like whether the statistic has been leaked, who is making that statement, whether it is in the form of prepared remarks or whether there is evidence that people are in fact being misled and so on.

Q73            Mr Fysh: Yes, because your overall remit for the statistics service is, is it not, to help Britain make better decisions?

Sir David Norgrove: Yes.

Q74            Mr Fysh: In future, would you look more carefully at such statements as that, which are palpably misleading?

Sir David Norgrove: We look very carefully at every statement that we see that might be misleading. I was not myself aware of that particular statement.

Q75            Mr Jones: Sir David, your letter of 17 September to Mr Johnson was succinct, one might almost say curt. It complained that he had chosen to repeat the figure of £350 million per week that might be available for extra public spending. You pointed out that this confused gross and net contributions. Did you accept therefore that the gross contribution was £350 million per week or thereabouts?

Sir David Norgrove: The gross contribution before rebate, yes, at that time was £350 million a week or thereabouts.

Q76            Mr Jones: Mr Johnson, in what was a detailed letter in response to your very brief letter, explained that he had said that what you had put to him was a distortion of what he had said. He quoted what he had written, “Once we have settled our accounts, we will take back control of roughly £350 million per week. It would be a fine thing, as many of us have pointed out, if a lot of that money went on the NHS”. The point that he was making was that he had asserted that the country, the United Kingdom, would recover control of that £350 million.

Sir David Norgrove: In terms of the ordinary meaning of “control”, we can parse the words that were used in those sentences. The ordinary meaning of control is to have the ability to spend the money in any way you wish and of course we can spend the rebate in any way we wish, because it never goes to the European Union. I am not of course the only person—we are not the only people—who found that misleading. I include the Institute for Fiscal Studies and the Royal Statistical Society on the same side.

The other fact to bear in mind is that even looking at the media, the print media on the Saturday and the Sunday morning, of 11 articles about that article by Mr Johnson, 11 saw it as repeating the original claim, including The Telegraph, in which the article was published.

Q77            Mr Jones: Nevertheless, Mr Johnson did give an explanation of his view on the matter and he suggested that if you had any disagreement with any of his rationale for asserting that £350 million a week would be back under our control, he said that he looked forward to hearing your reasoning. Did you respond to his letter?

Sir David Norgrove: No, I did not. As I said in answer to an earlier question, I am glad he acknowledged that it was not his intention to suggest that the £350 million would be available. I see no point in making a reply.

Q78            Mr Jones: Didn’t you think that it might be a courtesy to respond to the points that he made?

Sir David Norgrove: I felt at the time that it would prolong a discussion that was better closed.

Q79            Mr Jones: One might suggest that, to put it bluntly, you started it. Mr Johnson gave you a very extensive response to the points that you put to him. Don’t you think that it might have been a courtesy simply to respond to those points and frankly settle the matter? Because it seems to me that some of the points that he made are logical. We will recover control of that money; we will be able to spend it whichever way want.

Sir David Norgrove: I do not think there is a need to parse the sentences. As I say, taking back control, control of expenditure, usually means the control of money. It usually means the ability to spend it as you wish. We have that control. As my predecessor said four times, the rebate should not be taken into account in assessing how much will be available to spend on other things.

Q80            Mr Jones: Who alerted you to Mr Johnson’s article?

Sir David Norgrove: I saw it myself.

Q81            Mr Jones: Do you have a system within your office for picking up articles such as this?

Sir David Norgrove: We become aware of these kinds of statements in a variety of ways, as I think I set out in my letter. Some of it is our own noticing of them; others are brought to our attention by other people. Perhaps I could ask Ed Humpherson to expand on that.

Ed Humpherson: The great majority of things that we comment on are brought to us by members of the public or by organisations who raise a question or a concern. But a minority of cases, I would estimate between 10% and 15%, would be resulting from our own review of the production of statistics and their use, often on issues that we have commented on before.

Q82            Mr Jones: Now that Mr Fysh has raised the issue of the Leader of the Liberal Democrats, do you intend to pursue that further?

Sir David Norgrove: We can certainly have a look at it outside this room.

Q83            Mr Jones: Could we ask that you do so?

Sir David Norgrove: Yes.

Paul Flynn: Can I come back, Chair?

Chair: Just a minute. You have had a big go on this one already.

Q84            Mrs Gillan: Sir David, I have read this letter to the Foreign Secretary. Do you think, on reflection, that the way in which you approached the composition of this letter, you could have been slightly more polite?

Sir David Norgrove: I hope it was not impolite. We did not intend it be impolite. The intention was to head off at the pass a restatement of a claim that my predecessor wrote about four times.

Q85            Mrs Gillan: You just end the letter by saying, “It is a clear misuse of official statistics”. You do not even ask him perhaps to explain, which he did in that full letter back, and he made the point that if the BBC coverage was inaccurate that you would write to the BBC. Did you write to the newspaper in question?

Sir David Norgrove: I would have had to have written to every national newspaper and not just the BBC, but as far as I can see, every media organisation reported on the article.

Q86            Mrs Gillan: But surely that is the way to go about this. If you were concerned that the media was repeating inaccuracies and it was casting the wrong impression, I would have expected you to have written to those media outlets. How are they supposed to be alerted to this?

Sir David Norgrove: I think my letter was picked up by those media outlets and I hope achieved the desired outcome.

Q87            Mrs Gillan: I have to say that if I was getting a letter saying, “I look forward to hearing your reasoning” from the Foreign Secretary, no matter in what position I was, I would have thought it was courtesy to just acknowledge it and just say, “Thank you very much for your letter. I acknowledge the points you have made. I feel that that is the end of this correspondence”. Do you not feel that you have poured fuel on the fire here by treating this correspondence in this fashion? I find it quite impolite.

Sir David Norgrove: I am sorry you feel that way.

Q88            Mrs Gillan: I think others do too.

Sir David Norgrove: I think the risk would be that any further raising of this would add fuel to the fire rather than douse it.

Mrs Gillan: I am sure the Foreign Secretary will read the transcript of this and appreciate that you were intending to bring this matter to a close, but I still think that I would review the way in which you correspond with people, because I would have expected a response to that letter and I am very surprised that you did not respond.

Q89            Chair: Can I first say about this, we are very glad you have changed table 9.9 in the Pink Book. It was something that we pressed to be done in a number of communications with your predecessor and with yourself. The fact that the gross figure now no longer appears means it will be impossible for anybody to use the figure in the way that it was used. I think that is a good thing. Why do you think this was not done earlier?

Sir David Norgrove: As I understand it, the detail and the background of this is more John Pullinger. You may want to ask him about that, if you wish. My understanding is that the statisticians concerned originally felt that they were unable to change it because of the international rules around presentation of those numbers. It was only on contacting the authority concerned that they understood that they could change it.

Q90            Chair: It took too long. What we are really concerned about as well is how the authority escalates its interventions because your letter was written on a Sunday. Can I ask who you discussed it with before it was sent?

Sir David Norgrove: I consulted Ed Humpherson and John Pullinger.

Q91            Chair: It is quite a short letter and it refers to another letter. I guess the link would have clicked and he could have looked at the letter that had gone to Norman Lamb and it could have referenced my correspondence with your predecessor, because I also raised this as an issue months and months and months before the referendum and I made that clear. I should also declare my interest, I was a director of Vote Leave at the time, which is why I was concerned about it. What seems to require explanation is how you decide what to do when there is a breach or is there some kind of written protocol—I have asked you about this in private—and are you considering how to formalise how you respond so that there is a bit more predictability about it?

Sir David Norgrove: I hope you have seen in my letter to you, Chair, the protocol, the practices that we follow. I am not sure whether there was anything else you were looking for there.

Q92            Chair: Would you like to say what has changed as a result of looking at it or have you not changed anything?

Sir David Norgrove: My history, as you know, does not go back very far. My reading of the situation is that the broad structure is much as it was in the letter that my predecessor sent to the Committee in 2013, but what we have done is set that out in much more detail.

Ed, you have been around for longer than I have. Perhaps you would like to answer that question.

Ed Humpherson: I inherited these longstanding practices when I started in this role in 2014, because they go right back to the formation of the Authority and were set out before the board in 2011. They have not changed in substance very much. The big change in the last month and a half is David’s letter, which sets them out, and us placing them on our website.

Q93            Chair: Sometimes you make informal contacts with Departments, sometimes you write formally to Permanent Secretaries, sometimes you write to the Permanent Secretary about the conduct of a Minister or whether the Minister has broken the code. In this case, you wrote straight to the Foreign Secretary. Is it now more predictable as to what will happen when under which circumstances?

Sir David Norgrove: I think it is more apparent. From our point of view, it is reasonably predictable internally; I think it is now more apparent externally.

Q94            Paul Flynn: I think it is a great shame that an attempt has been made to undermine you in this Committee this morning. What you did was precisely the job that this Committee approved of when we voted unanimously to approve your appointment. It is the job that was sought to be created by cross-party support for the Statistics Act when it went through. Can you just confirm that there was never any possibility that there was a sum of £350 million a week that could be spent on the health service post-Brexit?

Sir David Norgrove: I think my predecessor and I have made that clear.

Q95            Paul Flynn: The maximum sum that might be available for the health service, and all the other considerations, the farm subsidies, was a maximum of £250 million a week?

Sir David Norgrove: As of 2016, it was £267 million a week.

Paul Flynn: As you and your predecessor have correctly pointed out to politicians out to deceive, they have criticised them from both parties, and I think this is the first time there has ever been a party division on this. We have supported criticisms of Labour Ministers in the past by UKSA and we should continue to do so. It is a shame this morning that an attempt has been to shoot the messenger when the messenger is doing precisely the job that you should be doing.

Q96            Chair: I think Mr Flynn speaks for himself, but I would like to make it clear, as Chair of this Committee, I support your discretion to intervene on any Minister or any public authority whenever you want and I would not want any ambiguity about that. Whatever Mr Flynn feels, that is the view of the entire Committee, I am sure.

Sir David Norgrove: Thank you, Chair.

Q97            Sandy Martin: Sir David, I think ordinary people would tend not to understand how the same body could be responsible for creating something and also adjudicating on it. Can you tell us what progress you have made in putting greater separation between the UK Statistics Authority’s functions?

Sir David Norgrove: Could I just question the premise of the question? I do not feel that the same body is doing both jobs. I feel very clearly, both in the office and outside it, that I am chairing two very separate and different organisations with different characters, and different ways of operating, because they have different objectives. The structures that manage both of those organisations are themselves different. The board of course sits across both, but the work of regulation is run through the regulation committee, not through the board.

The regulation committee is chaired by the deputy chair of the authority, Sir Adrian Smith, and nobody from ONS sits on or even attends that committee. I have attended three meetings of the regulation committee since I have been chair of the authority. A couple of people came along from ONS to explain the background on one issue and then left. They left before the discussion of the issue. That is the only occasion that they have been there. I do not see it as an intermingling of functions.

Q98            Sandy Martin: Can you give us an example of a tangible impact that has been produced by the establishing of the OSR?

Sir David Norgrove: The one that stands out in my time at the authority is CPIH, that it was de-designated by OSR and it took quite some time for OSR to be satisfied and the regulation committee to be satisfied that it could be re-designated. It took a lot longer, I would say, than ONS would have wished or felt comfortable with.

Ed Humpherson: Can I come in on this? For me, there are four dimensions of separateness of the Office for Statistics Regulation that really matter. Can the Office for Statistics Regulation set our strategy? Can we determine our own resources to meet that strategy? Can we determine our own work programme to execute the strategy? With that work programme, when we make decisions on the quality of statistics, their value, their trustworthiness, are we making those decisions separately? I am satisfied that on all four the strategy is our own. It is a separate strategy and it is determined with no influence or discussion with John Pullinger. It is an OSR strategy. I propose my own resources and that is not signed off by anybody in ONS. Our own regulatory programme, we consult on it, but we consult on it publicly. ONS has the opportunity to input into that consultation, as any other producer or any other member of the public can. They have no privileged rights there.

Our decisions are very much our own and often they are very critical of ONS. I am satisfied with that.

I guess there is also a behavioural aspect. My two tests behaviourally are do we treat all producers in the same way? ONS, the Scottish Government, the Welsh Government, the Department of Health, NHS Digital and the 164 other bodies, do we treat them all in the same way? Secondly, do we have a track record of making independent decisions? I feel reasonably comfortable about those. I do not know, David, if you want to comment on those two behavioural aspects.

Sir David Norgrove: No, that is certainly what I have seen.

Q99            Sandy Martin: Can you think of any further moves that you could make to give the public confidence that the OSR operates independently?

Ed Humpherson: Yes. The Office for Statistics Regulation is nearing its first birthday. It was launched on 25 November last year. I would say that we are developing our profile. We do not have a particularly big profile and nowhere near as big a profile as I think that I want or perhaps that we need. We are here for the public, we are here to stand up for the public interest in statistics and I think we need to have a higher profile. That is to do with our communications, the way we communicate, the way we select channels. We have a lot in all of those spaces. We are not a finished article there.

Q100       Sandy Martin: Can you imagine that the OSR might have some public benefit from having a governance structure that is completely independent from the UK Statistics Authority?

Ed Humpherson: To me, the things that really matter are being able to set your strategy, your resources, your work programme and your decisions. If you have that, you have all of the really fundamental underpinnings of independent capacity to act. That is what is important to me. I do not know, David, if you have anything to add on the governance side.

Sir David Norgrove: I would be concerned if Ed came to me ever or I had the feeling that he believed that he was being leant on inappropriately to protect ONS. As I said at my confirmation hearing, I see supporting OSR as always taking priority over the needs and pressures that come from ONS. I would rely on him to tell me if he felt any inappropriate pressure, but he has not.

Q101       Sandy Martin: Sir David, given people’s propensity to shoot the messenger, would it not be helpful to you if, when there was a disagreement about one of the things that you had ruled on, that there was a second opinion that you could rely on or somebody completely independent who could adjudicate on a situation?

Sir David Norgrove: You mean this is where there has been a possible misuse of statistics or information?

Sandy Martin: Or where there is an argument about whether or not there has been a misuse of statistics.

Sir David Norgrove: I suppose you could have some Court of Appeal, but it feels to me that that would be over-egging the pudding, a sort of unnecessary layer of bureaucracy. We already have the OSR itself, which is the repository of expertise in the use of statistics and then the board.

Chair: We are going to write to you about population statistics and population forecasts because we have run out of time, but it has been a very, very interesting session. Thank you very much for your evidence and for John Pullinger’s evidence.