Skip to main content

Official Statistics Order 2009

Volume 709: debated on Monday 16 March 2009

Considered in Grand Committee

Moved By

That the Grand Committee do report to the House that it has considered the Official Statistics Order 2009

Relevant Document: 5th Report from the Joint Committee on Statutory Instruments.

All members of the Committee will be aware of the important work being done by the UK Statistics Authority. The body was created by this Government last year with a statutory responsibility to promote and safeguard the production and publication of official statistics.

Two of its main functions are to monitor and report on official statistics, wherever they are produced in the UK, and independently to assess the quality of a core set of key official statistics for formal approval as national statistics.

The order before us today relates to the definition of official statistics; that is, the set of statistics that the authority must monitor and report on. Under the Statistics and Registration Service Act 2007, which created the authority, all statistics produced by the Office for National Statistics, government departments, the devolved Administrations and other Crown bodies, are automatically deemed to be official statistics. This means that numerous bodies are automatically under the oversight of the statistics authority and must follow its code of practice. For example, the Insolvency Service, the Child Maintenance and Enforcement Commission and the DVLA are Crown bodies, and are thus official statistics producers.

However, the Act also allows us to add further statistics by order, so that we can cover bodies that do not fall within this core definition but which clearly produce statistics that we feel it is important for the public to have trust in. Such bodies that do not fall within the core definition but which should produce official statistics include the Training and Development Agency for Schools and the Independent Police Complaints Commission. If a body is included in this order, the public can be assured that the statistics authority can monitor and comment on its statistical work. It also makes it possible for a statistic to be nominated for formal assessment as a national statistic, assuring the public that it has been produced in a way that is fully code compliant.

During debates on a similar order last year—the first such order—the Government accepted that they would need to return to the House of Lords with a new list of bodies. The priority last time was to make sure that all producers of national statistics were covered, but other bodies were included with different departments applying different criteria for nominating bodies for inclusion. We have this year made the criteria clearer and the result, with which I hope the Committee will agree, is a more coherent list. The bodies in the list are producers of significant, national-level statistics which the Government feel are important that the public should trust.

In the Act, the Government are required to consult the statistics authority before laying the order. Officials from the Cabinet Office have worked closely with the authority while drafting this order. The authority has had several opportunities to comment on the drafts. After the formal consultation required by the Act, the authority has said that it is content with the order before us today. We expect to update this list once a year. It is an advantage of the flexible definition in the Act that we can respond to changing needs and refine the list in this way, subject to parliamentary scrutiny.

In summary then, the order extends the number of bodies which are subject to the UK Statistics Authority’s oversight. These bodies will have to work to the new code of practice for official statistics, and their statistics will have the potential to be nominated for formal assessment by the authority to be national statistics. It is a vital part of the Government’s statistical reform programme, allowing greater independent monitoring and assessment of official statistics in order to enhance public trust in statistics. I commend the order to the Committee.

The Minister may or may not be aware that, last year, with the support of the noble Lord, Lord Turnbull, I submitted a paper to the Liaison Committee of the House in which we recommended that the accountability of what was then called the Statistics Board, but which is now the UK Statistics Authority, should be to a Joint Committee of both Houses. I am happy to say that the Liaison Committee accepted that proposal unanimously, and that the former Leader of the House, the noble Baroness, Lady Ashton, wrote to her opposite number in the other place, Harriet Harman, and asked that this should be drawn to the attention of another place with a view to its agreeing to set up a Joint Committee.

I am perfectly certain that the noble Lord, Lord Brett, had nothing whatever to do with this, but the fact is that, after three months, we had no reply at all from another place. In the event, the noble Baroness, Lady Ashton, had to tell me—as she said, “with some regret”; and, of course, the Liaison Committee’s report had been approved unanimously by this House—that the Leader of the House in another place had refused to accept the recommendation of this House.

Therefore, the position now is that the UK Statistics Authority is accountable only to a Select Committee in another place and not to this House. Happily, there are opportunities, and this is one of them, for Members of this House to exercise a form of parliamentary accountability that should, to some extent, make up for the lack of the proper, formal accountability that we should have had. I make no secret of the fact—and I am sure that the noble Lord’s advisers will confirm it—that from the very beginning of the procedures on what became the Statistics and Registration Service Act 2007, I argued that there should be a Joint Committee of both Houses. At every stage of the Bill, I had the same argument. In the end, that was accepted by the Liaison Committee and this House. However, because the other place did not agree—the right honourable Harriet Harman refused to accept that proposition—the noble Lord, Lord Brett, finds himself, through no fault of his own, in the unfortunate position of being subjected to a form of accountability for statistics. This is the only opportunity we have to raise this, apart from going in for a ballot and having a separate debate on a separate question.

The noble Lord produced a perfectly sound argument about why the order is necessary. That is to say, some 54 new bodies whose official statistics are going to be subject to control, review and monitoring by the UK Statistics Authority need to be added. If that authority agrees that the statistics produced by all these bodies comply with the code—and I shall have a word to say about that in a moment—they may be classified not just as official statistics but as national statistics.

What is the purpose of this? The noble Lord very properly said that the purpose is to give assurance to the public that the figures that they are being issued with by these various bodies, including government departments and Ministers, can be trusted. He used the word “trust” twice, and he used the phrase, which is exactly what the Minister said in the other place,

“assuring the public that it has been produced in a way that is fully code compliant”.—[Official Report, Commons, Third Delegated Legislation Committee, 2/3/09; col. 3.]

If the noble Lord looks at his notes, he will see that he said exactly those words.

What is the example that is being set to these 54 bodies to produce statistics in a way that complies with the code? Perhaps we should first look at the code. It has a foreword by the chairman of the authority, Sir Michael Scholar, that sets out a code of practice for official statistics. It is divided into a number of principles. Principle 1 is meeting user needs, Principle 2 is impartiality and objectivity and Principle 3 is integrity. It states:

“At all stages in the production, management and dissemination of official statistics, the public interest should prevail over organisational, political or personal interests”.

That is so obviously right. Indeed, we argued throughout the passage of the Statistics and Registration Service Bill that these were the kind of principles that the Government were seeking to establish in order, as they put it, to build trust or, as we said, restore public trust, in statistics.

What has happened since then? First, an OECD survey showed that, out of the 27 countries in the European Economic Community, this country ranks 27 on the question of public trust in statistics. The EU average for trust is 46 per cent and not trusting is 45 per cent—the balance is “don’t know”. In this country, only one-third of those polled trusted official statistics, and nearly 60 per cent did not trust them. This country ranks 27th out of 27 in the crediting of official statistics. If that is going to be corrected then, at every level, the Government as well as the 54 new bodies being added by this order must comply with the principles of the code of practice. Under Principle 3 about integrity, there is a whole lot of things; I shall not bother to read them all out, because they are all perfectly clear.

What is the most recent example that we have had of a Government issuing statistics in a way that they ought not to have? I put it neutrally. Noble Lords will remember that No. 10 and the Home Office issued press releases apparently quoting statistics about knife crime and admissions to accident and emergency units as a result of knife crime which totally failed to comply with anything in the code of practice. Worse than that, not only did they not comply but a whole series of e-mails have been published by the Select Committee in another place—which has not yet issued its report; I am very much looking forward to that—that passed between No. 10, the Department of Health and the National Statistician on whether those statistics should have been published in that form. At every single level of officialdom, it was said firmly that those statistics should not be published as they had not been checked or verified, had not been subject to any of the normal rules of the dates and time of release of statistics and that it would be an outrage if they were published. The fact of the matter is that that counted for nothing. The demand that the knife crime statistics should be published came from the Prime Minister. There is a minute from 10 December that says:

“Here is the statement that the PM would like us to publish tomorrow as part of the knife crime announcement”.

Then come the details of the statistics that he wanted. There was a very firm view by officials directed to No.10, expressed in another e-mail of 10 December, saying:

“I’ve spoken to a number of people here, including the Head of Profession at the NHS IC and our view is that these provisional data are NOT released.

Our reasons are:

1) As explained already these are provisional data for 2007/08 and 2008/09 and therefore they are potentially inaccurate and may possibly give the wrong impression”.

The second reason is perhaps less creditable. It says:

“2) If we allowed these data to be published, we open ourselves up to provide these provisional data to others who may ask for data which shows different trends”.

That is to say, somebody might ask for data that might show the whole thing up in a much less favourable light. That is the way that this Government now think, apparently.

However, that had no effect. A later e-mail, after a number of futile telephone calls, stated, “I have also been informed that Number 10 are adamant about the need to publish this statistic. As a result, I have been informed that they are likely to publish the data irrespective of the concerns raised”. The representations went right to the top. They went to the Permanent Secretary at Number 10, Jeremy Haywood. The National Statistician, Karen Dunnell, “has spoken to Jeremy this morning about the inclusion of certain unpublished statistics in a statement the PM may be making.” It was all for no purpose at all. Number 10 went right ahead and published the statistics. Why did it do that? It turned out that Mr Brown, the Prime Minister, had an appointment to meet a number of parents of the victims of knife crime that day, and he wanted to have the statistics published to show that everything was getting much better. That is why.

What sort of example does that set to the 54 bodies listed in this schedule? How will that restore public trust in statistics? What sort of temptation may some of these bodies be under to say, “Well if it’s all right for the Prime Minister, it’s all right for us”? We had an apology from Jeremy Haywood who said that perhaps they did not pay enough attention to the views of the professional statisticians. What he meant was that the Prime Minister did not pay enough attention to the views of the professional statisticians. He summed it all up in the word “we”. There was a brief apology on the Floor of the House from the Home Secretary, who said that perhaps some of the statistics should not have been published and that she was sorry. Both apologies were totally inadequate for what was a major breach not only of the code, but of all possible principles that should apply to the issue of statistics.

The Minister has commended this order to the Committee on the grounds that it greatly extends the list of bodies whose statistics the public will be invited to trust. But 27th out of 27? We can only go up. The history of this Government’s misuse of statistics has been utterly deplorable. Ministers must recognise that if they are going to achieve their objective of restoring public trust in statistics, the kind of example set by the Prime Minister last December has simply got to stop.

When it issues its report, I would expect the Select Committee in another place to use more politic language than I have used, but for those of us who took part from the very beginning in the proceedings on the statistics Bill, it is nothing short of disgraceful that here we are, more than two years later, still facing such misbehaviour by Ministers, despite the best advice that they could get from the people who really ought to know, which sets a perfectly deplorable example.

I have now got that off my chest. I hope that the Minister will himself add to the apology given by his colleagues. If he does not like listening to Back-Bench Peers expressing these views, he has a perfectly good remedy. He can go straight to Harriet Harman and tell her that there should be a Joint Committee of both Houses. Then these matters would be dealt with there and not, as they are this afternoon, on the Floor of this Grand Committee.

It is always a pleasure to hear the noble Lord, Lord Jenkin, speak on statistics. We debated these issues during the passage of the Bill and in subsequent debates on the establishment of a Joint Select Committee, when he said that there was a unanimous view in your Lordships’ House that this would be desirable, but the Commons rejected it out of hand.

As the Minister said, the purpose of this exercise is to improve trust in official statistics. If it is the case that 33 per cent of people now trust official statistics, unless I am very much mistaken, that is a significant improvement from the position when the Bill was going through. If my memory serves me correctly I think that only 17 per cent of people trusted statistics. I had a question for the Minister on the extent to which either the Statistics Board or the Government have a regular series of opinion surveys on trust in official statistics. We are at a low level—27th out of 27—when one recalls that Bulgarians, Romanians, Latvians and many others have far greater trust in their statistics than we have in ours. That is dismal, but it would be useful to know whether or not there is a trend in a positive direction.

The noble Lord, Lord Jenkin, referred to knife crime. I am tempted to ask whether he is surprised. The track record of the Prime Minister as Chancellor in this regard was lamentable. He will know better than anybody that the Government, having set their face towards reforming how official statistics and national statistics are managed, then produced a weak Bill that had to be almost completely rewritten by your Lordships’ House to make it stronger. In the end the Government had to capitulate. The Bill was weak because the Government did not want a strong one. There is no effective sanction in cases when the Government misbehave. The Bill allows national statistics to be downgraded to official statistics. When it was going through, Ministers explained that the prospect of having statistics downgraded so appalled Permanent Secretaries and senior officials that there would be no question of misusing information. The sanction would bring such public obloquy that they would bend over backwards to avoid it. The truth is that politicians still override that kind of consideration for the short-term consideration described by the noble Lord, Lord Jenkin. There is no effective sanction against Ministers who misuse statistics in this way.

I read the Explanatory Notes with great interest. Noble Lords will know that we on these Benches have argued that there should be no difference between national and official statistics; all official statistics should be national statistics. For noble Lords who wonder why we take that view I commend them to the simple sentence in paragraph 4 of the Explanatory Notes, which states:

“This was because, once the Act came into force on 1 April 2008, a National statistic had to be an official statistic”.

It is perhaps appropriate that the Act came into force on April Fool’s Day because that statement is complete gobbledegook.

In practice that means that the bodies specified in the order will be expected to follow the code as best as they can. That is pathetic. The principles set out by the noble Lord, Lord Jenkin, about the kind of practices that should be followed in producing statistics would strike the ordinary person as being so blindingly obvious that anybody producing statistics, whether covered by the order or already covered, should be following the code. What is so difficult about that? What circumstances can there be in which the bodies covered by this order would find it impossible to follow the code? If all bodies are being told is that we would quite like them to follow the code as best they can, it is a very weak injunction.

When we discussed the list last year, there was some discussion about where the Bank of England sat in respect of statistics. I believe that it is the case that the Bank of England is not covered by the legislation and that the Statistics Board and the National Statistician does not have oversight of statistics provided by the Bank of England. At that time, there was some discussion about why that might be the case, and I am spurred to raise the question again when I see that the Financial Services Authority is now covered by the schedule and the order. The Office for National Statistics has oversight over two legs of the tripartite authorities involved in statistics—namely, the Treasury and the FSA—but does not have oversight over the Bank of England. Do the Government have any plans to rectify that omission?

I thank the noble Lord, Lord Jenkin, for his remarks, with which I want to associate myself. As a new boy in this place, hearing distinguished Members holding the Government to account over the quality of statistics is very impressive and serious. The charge that was made was extremely serious, and I hope that the Minister will respond to it in his comments and that the Government will adhere to it in going forward.

I make a couple of pleas in considering these orders. The order comes up, we are presented with a list of 54 deemed persons whom we are told are relevant, then we are told in the Explanatory Note on the back that,

“28 are new, and 10 are omitted”,

from the last list. It would probably be helpful for debate if we could see asterisked or sorted in some way which ones are new and which have been dropped from the previous list. The accompanying Explanatory Memorandum gives one example of the ones that have been dropped—the Royal Air Force Museum—and, of course, nobody is going to pick on that. However, I am always slightly suspicious about 10 being omitted from the order last year when only one example is given. It would be helpful to see that information.

I am very struck by, and associate myself with, the remarks made about Bank of England. In the midst of an economic crisis, when the centre of the debate is on the state of public finance, it would seem absolutely appropriate that the Bank of England should be included in that schedule along with many other bodies that are directly associated with that effort. The whole statistical process has resonance; it is not just a debating point about the use or misuse of statistics. These statistics drive the allocation of scarce resources by the Government, so if they do not have veracity and cannot be trusted, we may be wasting money and spending it in the wrong place as well as understating the need in certain other areas. That is something else that I urge the Minister to give care to.

On the guidance that is followed, we are told that we can have confidence in these statistics. However, as has been said, the guidance goes only as far as saying that those bodies specified in the order will be expected to follow the code—that all sounds good—as best they can. Is not being able to do so as best as they can an admissible excuse in this regard?

I am also very interested in the fact that if the statistics comply with the code the board will approve them as national statistics. It would be useful to have a statistic about how many of those statistics have not received such approval and how many have been turned down. To see that level of detail would be useful. That speaks to the point of my noble friend Lord Jenkin, who said that, had there been extra scrutiny in this place and a Joint Committee, that is exactly the type of information which could come before it and help our debate.

The horrendous case mentioned was the misuse of the knife crime statistics. The head of the UK Statistics Authority described the release of the figures as “premature, irregular and selective”. He revealed that statisticians behind the data had tried to prevent the Government publishing the information—again as evidenced by my noble friend Lord Jenkin—and said that the data in their current form were “corrosive of public trust”. That is a bold statement by the person now in charge of the accuracy of our statistics. I ask the Minister to reflect on that, certainly given that it has been raised on a couple of occasions, and hope that we might get a response.

Paragraph 8.2 of the Explanatory Notes, under the heading “Consultation outcome”, states:

“Bodies that have been included in the order have been consulted … those bodies that were in the first order were consulted last year … This has generally been done by the statistical Head of Profession in each Government department contacting those bodies which are sponsored by that department”.

Again, that is the type of information—that is, the extent and nature of the consultation, and the feedback from it—that would be extremely helpful in discussing this order. It is clear that the objective of the statistics order is to build trust in national statistics. We have heard it argued how those statistics have been misused in the past in a grossly irresponsible way and how there is a need for greater scrutiny. When this order was discussed in another place, it took 11 minutes. The fact that we are up to 34 minutes and counting speaks of the importance that we at this end attach to these matters, and I look forward to hearing the Minister’s response.

I rise with some trepidation following the battering that I have received from the noble Lord, Lord Jenkin. I do so also partly protected by my ignorance of the discussions to which he referred in terms of the Joint Committee. However, some of the answers to the questions that he posed are in the questions that he raised.

We are not proud of our position within the OECD in terms of trustworthiness of statistics, but, as the noble Lord, Lord Newby, said, there has been an improvement of some 15 per cent in public perception and trust since this Act came in, which means that we have started to turn the situation around, although there is a lot more to be done. That is why the list has been refined; that is why the code, to which I shall come in a minute, is so important. It is also why the Government have taken more action than any other to ensure the independence of statistics. We have created an independent authority, reporting directly to Parliament—not, I agree, necessarily in the way that the noble Lord wishes—to oversee and assess the production of official statistics. Our setting that authority on a statutory footing built on the framework for national statistics as set in 2000. We are therefore building the statute on what we already had as a practical operation. We have reduced pre-release access to official statistics from five days to 24 hours, and have limited the number of statistics to which it applies and the number of people who get it. We have also taken Ministers out of the governance of the ONS.

I turn to the laboured point on knife crime statistics, which caused such a furore last December. I offer it as proof that the system works, because it led to a furore and the admission that those statistics had been produced and released prematurely. It led to the Home Secretary apologising to the House of Commons. Most importantly, it led to action to put in place measures to ensure that those errors would not be repeated.

How can the Minister possibly argue that the system was working as it should have when the Prime Minister overruled all his officials, including the National Statistician, and insisted upon publishing statistics that he had been firmly told that he ought not to? We had debate after debate on the Floor of the House, in Committee and on Report on the whole question of the misuse of statistics. Here we have had, in the past two or three months, a classic example of Ministers not learning anything about this at all. It is not working properly.

I am afraid that the noble Lord misunderstands me. I am not suggesting that the system was working properly on that date. I am saying that it that caused a furore, and caused Parliament and Ministers to respond. It caused an apology to be received by the House of Commons. That shows that attention is paid to the issue. In that sense, it is working—not that this error was in any way permissible. I was starting to outline how we are trying to ensure that it is not repeated. I know that confession is good for the soul, and that the Opposition are demanding apologies. An apology has been made on this, and I have nothing to add to that.

Of course, in Gus O’Donnell’s and Kevin Brennan’s letters, they said what happened and there had been an apology. It is more important in many ways that we ensure that we take forward action to ensure that it cannot be repeated. On the actors’ role on that occasion and why the National Statistician’s intervention did not prevent the statistics being produced, she phoned No. 10 at 8 am but the press release had been released the previous evening on the basis of a midnight embargo. In that sense, preventing the release of the statistics was not practical.

However, that is history and we know the problems that it caused. Since then, the Government have ensured that the action taken cannot be repeated. They have produced and circulated to all appropriate persons in Whitehall a memorandum on exactly how statistics are to be treated in future to ensure that it does not happen again. Michael Scholar was invited, and spoke, to the Permanent Secretaries’ conference to ensure that it does not happen again. There was a lesson to be learnt, which I believe has been learnt.

A number of other points were made that I will try and respond to, but I—

Before the Minister leaves that point, can he be specific about what has changed so that it would not be allowed to happen again?

That would not be possible for anyone to say. However, the roles of the statisticians, the private offices, official departments, officials and Ministers are now absolutely clear. That has been circulated to all and sundry to ensure that no one is unaware, that the situation does not repeat itself and that everyone understands their role precisely.

Did the Prime Minister apologise to all those that he was now sending this new rule out to? Did he say that it was his fault and that he was very sorry? If he did, it would be a first.

The noble Lord is asking rhetorical questions. This is the circulation of a document to civil servants, from the head of the Civil Service to the appropriate Permanent Secretaries in departments. We are seeking to ensure that officials and special advisers know their role, so there is no question of the political overriding the integrity elements in the code to which the noble Lord has referred.

It is not a rhetorical point. The point was made by my noble friend to say that the Prime Minister specifically said on 10 December that he required this information to be released. In his defence, the Minister has cited the fact that the Home Secretary apologised. It is entirely reasonable to ask why, given that it was actually the Prime Minister, the Prime Minister should not apologise specifically for that.

I am not saying that that is entirely reasonable. What is entirely reasonable is to await the publication of the Select Committee report to see the views of those who examined this in another place. No doubt it will be debated and that will be an appropriate place to debate this issue. What we are seeking to debate here is the operation of the code for the future. I seek to reassure the Committee that we have taken action to prevent this kind of error taking place again. I was going to move on to deal with some of the other points, starting with the premise that—

Is the Minister suggesting that it will be in order for this House to debate the report of a Select Committee made in the other place? Is that what he is suggesting?

The noble Lord is a very old and wise former secretary to a department. He knows precisely that I am not suggesting that. I am suggesting solely that he has raised evidence about a Select Committee report in another place which is dealing with the problem. I am suggesting that we await that report. The Ministers concerned are Ministers in another place. That is the appropriate course of action.

I now turn to some of the points raised. If I do not do justice to them, I am more than happy to reply in writing because all three noble Lords made points in their contributions. I start with the points about the Bank of England, which is not included. The authority and the bank have been in regular contact with each other through 2008. They have agreed not to include the bank in the 2009 order. That is with the consent of the Cabinet Office and the Treasury. The bank has its own code of practice for statistics, which works well for its specific and important role. It passes a lot of data to the ONS and any resulting statistics, including parts of the national accounts and much of the Financial Statistics publication, are national statistics and so are fully compliant with the authority’s code.

The point of greater interest is why we are not including the bank at the present time. Those discussions are continuing. The bank sometimes collects data which it decides not to publish even without announcing that it will not publish them. That would go against the code as it stands—in this case, on the protocol on release practices. As I said, the bank and the authority are working closely to see how best to use the code in a specific area of monetary and financial statistics in a way that does not hinder the bank's ability to perform its duties effectively. That is the current position in respect of the Bank of England.

The other points raised included a sensible point from the noble Lord, Lord Bates, about having a clearer understanding of which bodies were new and which had been dropped. Those dropped from the first order tended to be dropped because they were not producing national statistics or continuing to exist beyond 2009. They include the British Hallmarking Council, the Fleet Air Arm Museum, the Hearing Aid Council, the National Army Museum, the Royal Air Force Museum, the Royal Marines Museum, the Royal Naval Museum and the Royal Navy Submarine Museum. That is because they are not producing significant national-level statistics. Also excluded are the Commission for Healthcare Audit and Inspection, the Commission for Social Care Inspection, the Commission for the New Towns, the Housing Corporation and the Urban Regeneration Agency. That is because, as the Committee will have spotted, they are due for abolition on or before 1 April 2009. The first two bodies in the list will be replaced by the Care Quality Commission and the second two as a result of the Housing and Regeneration Act 2008. None of the statistics has been lost, but will be picked up in respect of the replacement bodies. I hope that that assists with that particular question.

The noble Lord, Lord Newby, asked whether there were surveys. The ONS produces a regular annual survey, which is how it hopes to glean the increasing confidence that it expects to get from the improved performance, notwithstanding the highly rehearsed question about the statistics produced and released prematurely in December.

I will write to the noble Lord, Lord Bates, to answer his question about statistics and the consultation about statistics. That covers most of the points raised by noble Lords. I will examine Hansard tomorrow, and if there are any points that I have not covered, I will write to noble Lords. With that, I beg to move.

Motion agreed.

Committee adjourned at 5.54 pm.