Statistics and Registration Service Bill

My Lords, like all other noble Lords who have spoken so far, I welcome the intent of the Bill overall. Official statistics are vital to decision-making in government and to holding the Government to account. For several decades now, a lack of trust due to a wide perception of political interference has lessened the value of, and undermined, official statistics to an extent not seen elsewhere. That seems to be getting worse in recent years; I refer to the examples of Network Rail and the golden rule. My initial views on the contents of the Bill were positive. I noted with approval that the Bill recognises the need for clear responsibility for independent oversight of the system by a non-ministerial governing board coupled with a clear accountability for the delivery of trustworthy statistics given to the chief statistician. I also welcomed the authority of the board to approve and maintain a code of practice. I then turned to independent expert judgment on the Bill. The Royal Statistical Society overview, which has been quoted by many other noble Lords, is worth reiterating. It said: "““We believe that trust in official statistics and public confidence in the system that produces them is fundamental to the Government’s objectives for the Bill and it should be judged by whether it addresses these””." As drafted, the Bill will not meet this standard. Like the Royal Statistical Society and other noble Lords, I have some major criticisms of the details of the Bill. My main criticisms cover five main areas, all of which have been referred to by other noble Lords. These are the scope of statistics covered by the Bill, the structure of the Statistics Board, the role of the Treasury, pre-release of statistics and, finally, statistical confidentiality. I will deal with each of these in turn. The scope of statistics covered by the Bill is unsatisfactory. All official statistics, such as those on crime, health and education, should be defined as national statistics. The statistical output of the ONS is only a fraction of government statistics. The much wider Government Statistical Service is responsible for many high-profile and important statistics, but is not adequately dealt with in the Bill. To make matters worse, as my noble friend Lord Jenkin has said, the Bill leaves the decision to offer a particular set of official statistics for designation as national statistics to the Minister of the department concerned. Hence the decision whether a department’s statistics will be national, and thereby subject to the Bill, or non-national rests with the Minister. In my view, that is wrong. All official statistics should be defined as national statistics and thereby subject to the code of practice, thus eliminating the current unsatisfactory two-tier system. On the code of practice, first, there is no requirement for those who produce statistics to adhere to the code; and, secondly, it does not apply to official statistics outside the narrow range of national statistics. Those omissions need further scrutiny. Finally, I agree with the RSS that the National Statistician should co-ordinate the statistical system across government departments. I turn to the structure of the new Statistics Board. As already said by many speakers, as drafted, the Bill muddles the roles of the board and the National Statistician. In my view, the National Statistician should be responsible for professional and operational matters and for statistical production. The board should be responsible for holding the National Statistician to account for those responsibilities. It is vital that the board's role of monitoring should be completely separate from the National Statistician's role of statistical production. The board should not produce statistics or maintain systems of classification or be responsible for protecting confidentiality. It must remain independent. The next area of criticism is the proposed role of the Treasury. The Bill assigns residual ministerial authority to the Treasury. The Treasury naturally has a major interest in economic statistics and is currently perceived—by the RSS among others—to exercise influence on the ONS's statistical priorities, particularly with regard to funding. Assigning ministerial authority to the Treasury takes no account of the importance of statistics on crime, health and education over which the Treasury has no authority but on which it may well seek to restrict funding. I agree with the noble Lord, Lord Moser, that if the residual ministerial authority returned to the Cabinet Office, where it began, that would increase public confidence. It would be seen to act as an honest broker across Government should any Minister need to be consulted on an issue. In addition, I agree with our party policy that the independent board should have a similar role to the National Audit Office, reporting to a committee of both Houses and having its resources decided by a parliamentary vote. The Bill gives the board power to decide what Office for National Statistics’ figures are produced and published. It also allows the Chancellor, and the Scottish, Welsh and Northern Irish people, should he or they wish, to direct the board to do something other than what it has determined. Even this proposed board does not have independence. I turn to the pre-release access to statistics for Ministers and policy officials, which has been covered by many other speakers. The RSS and the Statistics Commission both agree with me that there should be no pre-release of statistics to Ministers and policy officials. As I understand it and as others have said, in the UK, the norm is 40 hours or even more for a number of people, and the new rules proposed do not seem much better, whereas in other countries Ministers are given the final statistics only a few hours or even minutes before release. I support the view of the noble Lord, Lord Moser, expressed at a recent meeting that we had with the Financial Secretary to the Treasury, when the noble Lord wished to return to the pre-release regime that existed under Edward Heath's administration of 1970 to 1974, which was very tight. The Statistics Commission makes the recommendation that the board, rather than Ministers, should have the power to determine the period of their pre-release. Like the commission, I believe this is the only way of giving sufficient public reassurance that official statistics are free from government spin. I highlight the RSS's concern about the actual release of official statistics. There is nothing in the Bill at present about that. The current problem is that departmental announcements combine policy and statistics. The RSS recommends, in my view sensibly, that the statistical release should be issued from a separate location under the aegis of the National Statistician and that the Minister's press release should not be permitted until at least 30 minutes after the statistical release. Also, the National Statistician should have the freedom to comment publicly on statistical issues, including erroneous interpretation of official statistics. The final area on which I wish to comment—it has already been covered well by my noble friend Lord Northesk—is statistical confidentiality. Again, the RSS makes a vital point. The public must have confidence that personal data given for statistical purposes will not be made public or used for non-statistical purposes. The Bill does nothing to promote public confidence, as it allows any transfer of information if covered by a ministerial order. We need a much simpler and clearer requirement to hold data in confidence and to use it for statistical purposes only. The same applies to personal data accessed from administrative systems for statistical purposes. Much of the statistical information in connection with health and educational services is derived from such sources. The public should be in no doubt that statistical producers protect confidentiality. It is not too late for the Government to deliver what they first promised in 1995 when the Leader of the other place, Jack Straw, addressed the Royal Statistical Society. He said that the National Statistical Service should be placed at arm’s length to Ministers, on a similar basis to the National Audit Office and should report to a powerful committee of the Commons—as already stated, I believe that should be a Joint Committee. I agree with the Leader of the other place: that is what is needed; that is what the Government promised; and that is what should be delivered in the Bill.