Previous Section Index Home Page

In a number of instances, as we have acknowledged throughout the debate in Parliament, the Bill represents a step in the right direction—a step that we believe is too timid, but, nevertheless, one that will in the main help to enhance the credibility of Government statistics. However, in this context, we believe that the Bill involves a step backwards. It removes one of the limited existing safeguards in relation to the Government statistical system. The Bill proposes to merge the Statistics
13 Mar 2007 : Column 162
Commission and the ONS. Under the Bill, the commission’s function of scrutinising statistics and the ONS’s function of producing them will both be carried out by the new statistics board. The Opposition believe that losing a fearless and independent watchdog and combining scrutiny and production functions will significantly undermine the effectiveness of the Government’s proposals. Amendments Nos. 8 to 15 seek to take the board out of direct involvement with statistical production. They remove key production functions from the board and vest them in the National Statistician, leaving the board tasked with scrutinising her activities and those of the ONS, and to a lesser extent those of the departmental statistical services.

Alun Michael (Cardiff, South and Penarth) (Lab/Co-op): I am following the points about independence and objectivity with great interest. What does the hon. Lady think is the purpose of statistics?

Mrs. Villiers: Statistics have many purposes, one of the most crucial of which is informing decision making in government, in local government—I know that that is a matter of importance to the right hon. Gentleman—and in business. Statistics allow international organisations to assess the performance of the UK economy. Official statistics have a range of vital functions.

Let me turn to the objective of our amendments. There is considerable confusion in the statistical community about the way in which the Bill allocates functions and responsibilities to different entities, especially regarding the allocation of functions between the board and the National Statistician. In short, it is not easy to say exactly who will do what and who will be accountable to whom. When I asked the Financial Secretary to give the Public Bill Committee an example of a similar institutional structure that was in operation in the UK or internationally, he was not able to provide one and said that there were no direct comparators. He effectively acknowledged to the Committee that the proposal in the Bill represents more or less uncharted territory.

On Second Reading, the Minister was asked a question about the person to whom queries about the way in which the census deals with matters relating to the Sikh community should be addressed under the new structure. He said that such a technical query relating to methodology would be dealt with by the National Statistician at present, and that that would be unlikely to change under the proposed structure. However, the Bill expressly provides for the board to have a role in considering that type of methodological question. It is true that the Government have tried to provide for a separation of the functions of the board by creating a new statutory head of assessment. They claim that that will mean that production and scrutiny functions will be kept separate during day-to-day operations. However, such a separation is inconsistent with clause 29(1), which makes the National Statistician the chief executive of the board. That measure leads one to expect that she will have an executive remit on all the board’s activities, whether they are assessment or production. We have tabled amendments to address that situation.

The problem of the blurring of the functions of the board and the National Statistician is exacerbated by clause 29(4), which gives the board the right to substitute its decision for that of the National Statistician. I
13 Mar 2007 : Column 163
believe that that problem is highlighted by Lib Dem amendments Nos. 46 and 47. The Government’s attempt to provide for an internal split of responsibilities is insufficient to deal with the concerns expressed by many during the consultation about the conflict of interest to which the structure proposed in the Bill will give rise. As the North East Regional Information Partnership has pointed out, the board “will not be detached”. In short, the board will be judge and jury in its own case.

The problem is best illustrated by considering what would happen if a complaint about a decision were made by a member of the public. At present, someone making a complaint about a decision made by the ONS can go to the Statistics Commission, if the ONS refuses to respond to his concern. Under the new structure proposed in the Bill, a complainant will take his concerns to the board. However, the board would have been responsible for making the decision about which the complaint had been made. In a real sense, the board would have been the entity that made the decision, given that the Bill merges the ONS into the board. The board would thus be asked in such a situation to rule on whether its own decision was correct. The chairman of the board could regularly find himself in the embarrassing position of having to issue a rebuke or correction to himself. If the board decided that the approach was correct, a complainant could well feel that the failure to impose a sanction resulted from not an impartial assessment of the correctness of the approach taken, but the board’s desire not to admit fault in its statistical production. No doubt the statistical production kitemark is being produced with the board’s imprimatur. The Royal Statistical Society has pointed out that it is difficult to envisage how the board will be able to command public confidence in its impartiality in such a situation.

The practical problems with such arrangements have been debated at length in the House in the context of the BBC. Many people, including the Burns panel, have stated that arrangements whereby the BBC regulates its own output are unsustainable. The conflict that we are discussing is even more stark than the conflict in the case of the BBC, as the BBC’s regulatory function is confined to its own output, whereas the board’s remit is wider. It would be difficult for the Statistics Commission to hold the statistical system to account as effectively as it has done if it had responsibility for running the Office for National Statistics, too.

The concern was expressed repeatedly during the consultation. To take just two examples, Jill Tuffnell, head of research for Cambridgeshire county council, said on behalf of the Central and Local Government Information Partnership that a much clearer divide was needed between scrutiny and operational functions. The statistics users forum stated that it did not believe that it was good governance, or conducive to restoring or maintaining public trust in the system, for the same body to be responsible for the delivery of statistics and for ensuring quality and adherence to standards. It said that the proposed removal of the Statistics Commission would eliminate a check on the system and replace it with a system that at least appears to be weaker.

We hope that the Government will listen to the concerns of the experts who use official statistics every day, and we can see small signs of movement in Government amendment No. 48. In Committee, the Government had the embarrassment of having to vote down an
13 Mar 2007 : Column 164
amendment tabled by my hon. Friend the Member for Sevenoaks (Mr. Fallon) about the purpose of the board. The amendment said that the board should seek to ensure that the statistical system served the public good. I was surprised that the Government sought to vote down an amendment of that sort, but they have tried to remedy the problem in amendment No. 48. However, although that amendment might give us more of a clue about the objectives of the board, it does nothing to deal with the structural problems that I outlined.

Rob Marris (Wolverhampton, South-West) (Lab): I might have got lost somewhere in the hon. Lady’s amendments, but I understand the separation that she seeks to achieve through them. I might have missed something, but having read her amendments, I do not think that they would remove the National Statistician from the board. If I am right about that, and her amendments would leave the National Statistician on the board, is that not a contradiction in her argument? I do not think that she has tabled an amendment to remove the National Statistician from the board; why is that?

Mrs. Villiers: I did not table such an amendment. To be honest, I puzzled over the issue in Committee, but on balance I think that there is an argument for keeping the National Statistician on the board. I acknowledge that if one took a purist approach, perhaps one would want complete separation, and that would involve taking the National Statistician off the board. I concluded that if I were to propose such a complete and radical rewriting of the Bill, it would not get anywhere, so with compromise in mind, I tried to target the functions that cause the most problems, but I acknowledge that that is not an ideal approach.

John Bercow (Buckingham) (Con): I heard my hon. Friend’s explanation of her thinking. Does she think that there is a qualitative difference between the National Statistician being a member of the board, and her being its chief executive, which, I understand, is what the Government propose?

Mrs. Villiers: Yes, certainly. My amendments reflect my particular concerns about clause 29, which makes the National Statistician chief executive of the board. The clause makes explicit her executive role in relation to the board’s activities, and I think that that is the issue of greater concern. That is why I tabled an amendment to address that point, but did not go as far as tabling an amendment to take the National Statistician off the board altogether.

I shall now turn to the second of the three points to which I referred at the start of my remarks. It relates to new clause 3(1)(b) and (c), and the Liberal Democrats’ amendment No. 39. The Opposition believe that it is crucial that the National Statistician is given the explicit remit of co-ordinating the Government’s statistical system. The plea for co-ordination has been made strongly by a number of groups, including the Royal Statistical Society and the statistics users forum. The decentralised system has a number of important strengths, but it comes with in-built disadvantages, too. In cases in which critically important statistics are produced by different Departments, there is a self-evident risk of inconsistency, duplication and inefficiency. In the 1990s, for example, I am told
13 Mar 2007 : Column 165
that it took an enormous effort to get the neighbourhood statistics project off the ground because it pulled in data from so many Government sources.

The new clause provides a strong co-ordinating role for the National Statistician, which would help to minimise inefficiency and duplication. It would provide momentum and political drive for co-ordinated projects such as neighbourhood statistics, and it could provide critical direction and coherence for the diverse work of the Government’s statistical systems.

4 pm

We believe that that co-ordinating role should extend to the promotion of the consistency of statistics across the UK. Concern about the fragmentation of data has been expressed by many organisations, including the Society of Business Economists. The problem did not start with devolution, but devolution has certainly heightened concerns about the inconsistency of statistical data across the UK. There is an increasing pull from devolved assemblies to fragment statistics, but there is an insufficient counter-balancing pull from the centre to promote consistency, which led to serious problems in the 2001 census, according to John Pullinger, who was heavily involved in that census and is now chief of the Royal Statistical Society national statistics working party. According to the statistics editor of the Financial Times, Simon Briscoe, the census left the Office for National Statistics enfeebled by the pressure to fragment data across the UK. The RSS told the Treasury Select Committee that the problem was serious and worsening.

It should be a matter of concern to the House that academics such as Dr. Kadhem Jallab of Tyne and Wear Research and Information have pointed out that differences in the index of deprivation have made it impossible to compare levels of poverty in Newcastle and Glasgow. Alison Macfarlane, professor of perinatal health at City university, told the Treasury Committee that she had to source information from 18 different data sets to compile what she described as a very basic set of maternity indicators. May I make it clear that the Opposition do not seek to impose a one-size-fits-all model on the UK? Of course, the devolved areas will wish to produce statistics tailored to their particular needs. Different regions and local areas may well wish to do the same, but if statistics are collected and compiled on similar topics, consistency should be encouraged wherever possible. If wise and informed decisions are to be made on the impact of health, education, housing and social deprivation on the whole of the UK, we need a core of statistical indicators common to the entire country so that we have empirical evidence on which to base those decisions.

Rob Marris: Amendment No. 9 would leave out clause 7(1), but how would new clause 3 improve clause 7(2)—I do not think that the hon. Lady tabled an amendment to remove that provision. The explanatory notes are as helpful as ever, and paragraph 47 on page 8 refers to “timeliness and comparability”. Surely, her proposal is already in the Bill?

Mrs. Villiers: We wished to amend clause 7 to provide space for new clause 2, which was not selected. The
13 Mar 2007 : Column 166
amendment is probably no longer relevant, because it is linked to new clause 2. I hope that the rationale is clear to the hon. Gentleman.

Rob Marris: Perhaps I did not make myself clear. Amendment No. 9 would remove clause 7(1). The objectives that the hon. Lady seeks to achieve in new clause 3 appear to be addressed by clause 7(2), which she is not seeking to remove. I suggest that the new clause is, in fact, redundant, because the objectives that it seeks to achieve are encompassed in subsection (2).

Mrs. Villiers: I am grateful for the hon. Gentleman’s clarification. Clause 7 refers only to the board. The change that we would like to see is the National Statistician having a remit to drive forward consistency and co-ordination of statistics. I apologise if I did not understand the hon. Gentleman’s point initially. The Bill provides for the board to perform certain functions, but we need a clearer steer that the National Statistician should be involved with them. That is the purpose of this group of amendments.

The Opposition’s new clauses and amendments highlight a number of important duties for the National Statistician to carry out, but there can be few of greater importance than the third and last point that I shall raise in relation to this group. New clause 3(3) and Liberal Democrat amendment No. 45 propose that the National Statistician should be acknowledged in the Bill as the Government’s chief adviser on statistics and should provide professional leadership to all in the Government’s statistical service.

New clause 1 would give the National Statistician direct access to the Prime Minister. If the reform is to succeed, the National Statistician’s writ must run throughout the Government’s statistical services, not just in the Office for National Statistics. In a decentralised system, we need a strong figure to promote good practice and ensure that Departments do not slip below the appropriate ethical and professional standards in the production and release of statistics.

Ensuring that the National Statistician has the ear of the Prime Minister and access to the Prime Minister, and ensuring that Departments know that she has the personal backing of the Prime Minister, is key to giving her authority to maintain high standards of integrity and impartiality in relation to departmental statistics, and to help her eliminate the mistreatment and manipulation of statistics, which the Bill is designed to stop.

The National Statistician’s leadership role is crucial. Departmental statisticians should account to her, as well as to their direct bosses. They should look to her for a lead on technical matters, but more importantly, they need to be able to look to the National Statistician for support in maintaining the highest standards of statistical quality and integrity. When they are under pressure from their line managers in Departments to compromise on those standards, they should be able to refer to the National Statistician for guidance and backing.

It is critical that the power and the status of the National Statistician should act as a counter to the political pressure on departmental statisticians that might be exerted by Ministers and policy officials. That is one of a number of reasons why the Opposition
13 Mar 2007 : Column 167
amendments in this group could do so much to strengthen the reform and make it work to protect Government statisticians from political interference.

John Bercow: Will my hon. Friend give way?

Mrs. Villiers: Certainly. I was about to conclude, but I give way to my hon. Friend.

John Bercow: I am grateful to my hon. Friend for generously giving way, particularly in the circumstances of her approaching her denouement. Clearly, there is an important distinction between the technical quality of statistics and the question whether they are duly representative. I did not have the privilege and the salivating experience of serving on the Standing Committee, so I hope that my hon. Friend will understand my ignorance on this point. What if, to put it in simple terms, the National Statistician has a particular view about what ought to make up the basket of items that constitute the basis of the retail prices index, and the Chancellor or the Prime Minister takes a different view? Will we get to know that the National Statistician has a view? Will it be in the public domain? Can we debate the issue and hold Ministers to account on the matter?

Mrs. Villiers: One of the changes introduced by the Bill is to pass responsibility for the RPI from the Chancellor of the Exchequer to the board. I hope that the National Statistician would be involved in exactly the kind of technical decision to which my hon. Friend refers. It is important, as he says, that those decisions are made transparently and that we can debate them. It is a positive feature of the Bill that it will transfer powers over the RPI and such decisions to the board. It is important that those decisions should be taken, in effect, by the National Statistician. My hon. Friend’s question illustrates the problem that we are dealing with, in that the Bill provides for the board to take the decision, whereas the National Statistician has the technical expertise to do that.

I hope that the Minister and the House will consider the amendments, which would considerably strengthen and clarify the institutional structures set out in the Bill and improve our chances of restoring trust in official statistics.

Alun Michael: Anybody who believes in an evidence-based approach to public policy must welcome the Bill and our debate. I am sure that the hon. Member for Buckingham (John Bercow) was disappointed to miss the opportunity of serving on the Committee, which was one of the best that I have had the pleasure of taking part in. The Minister made good responses, took away and reflected on some of the important issues that were raised, and has responded positively.

In that context, I particularly welcome amendment No. 48, in the name of the Chancellor of the Exchequer, which gives a clear sense of direction and purpose for the board and for the chief statistician. Statistics are now clearly defined as being there to “serve the public good”, which is an important principle to have in the Bill. The definition of “the public good” is particularly useful in its reference to


13 Mar 2007 : Column 168

That means that it is about informing general public policy debate, and I have no doubt the standard of that debate will thereby be improved. It also gives the board the remit of ensuring that statistics are used in


Next Section Index Home Page