Statistics and Registration Service Bill

My Lords, when I spoke in the short debate initiated by the noble Lord, Lord Moser, I broadly welcomed the Government’s proposals. Now that we have the Bill as it has emerged from the other place, I still support its aims, although there are at least five issues that the House needs to address. On some, I will argue that the Government should hold their ground against the attempts to change the Bill, but on others I will suggest that they need to rethink their position. The first issue, which has come up many times in this debate, is the governance structure of the Statistics Board. The Bill proposes a unitary board that brings together chair, non-executive directors, the National Statistician, plus two other officials. That has given rise to what is referred to as either the ““judge and jury”” problem or, alternatively, ““the BBC analogy””. How, it is asked, can the board be a producer of statistics and a regulator of their quality? That is a valid question, and, in one way or another, an answer has to be found. In my view, the noble Lord, Lord Burns, was correct in arguing that there was not enough separation between the BBC Trust as regulator and the BBC as broadcaster. The Royal Statistical Society has proposed a number of amendments which seek to enforce a greater separation by making the National Statistician rather than the board responsible for the production of statistics, with the board as an oversight or scrutiny body. The noble Viscount, Lord Eccles, put that case very effectively only minutes ago. However, despite the view that I took in the case of the BBC, I have concluded that I prefer the unitary board in this case. First, I think that the divided accountability would prove less satisfactory for Parliament as it would have two accountables reporting to it rather than one: the board and the National Statistician. Secondly, and most importantly—and this is based on my experience in dealing with the previous National Statistician—I want national statistics produced under the full authority of the board. Never again do I want to see a National Statistician isolated as was the case with Mr Len Cook. When there is pressure from Ministers—for example, on whether or not something is in the public sector—I want a solid unity between board and National Statistician. For example, the then Secretary of State for Health would have found it more difficult to brand as ““absurd”” the serious piece of work done by the ONS on productivity in the health sector if it had been published in the name of the board as a whole. Finally, we should remember that only one-quarter of the series designated as national statistics is produced by the ONS. I agree with the noble Lord, Lord Moser, that it is in the statistics produced by departments—for waiting times, crime and so on—that the greatest difficulties arise. I accept that if we maintain a unitary board we need to find a solution to the ““judge and jury”” problem. The Bill provides a mechanism through the head of assessment, and it is better to build on that and through the internal constitution rather than to reduce the board to pure regulator and weaken the alliance with the National Statistician. The second issue is the concept of national statistics. Here I believe that the Government need to rethink their position. Like the Royal Statistical Society, I do not understand the logic of having a concept of official statistics and then a code of practice that applies only to the subset known as national statistics. That can only give rise to suspicion, even if unwarranted, that the Government want either to tolerate a Ryman league of second-rate statistics not covered by the code or, worse, that Ministers want to keep certain statistics in the lower league so that they can get away with things that are outside the disciplines of the code. The process of promotion and relegation in Clause 13 has no credibility. If the crime statistics do not come up to the standards of the code, they are not going to get demoted from national statistics to mere official statistics, they are going to get improved. My recommendation is to bring all official statistics, whatever they are called, under the code, and to claim credit for this as one of the achievements of the Bill. The third issue is release times. It is not good government to keep Ministers in the dark. If we want them to be accountable we must set up procedures and controls that make that possible. No one expects the President of the United States to appear once a week for questions before Congress; so we have quite different systems and cultures. I doubt whether Parliament really wants to allow Ministers to put forward the excuse that they cannot comment because their officials are still analysing the figures. Nor is it good practice to allow misunderstandings to circulate for hours uncorrected—and the noble Lord, Lord Desai, gave an example of that, although it was rather an old one. The noble Baroness, Lady Thatcher, when Prime Minister, frequently quoted Mark Twain to me—usually when incandescent after listening to the World Service at 5.30 in the morning—saying that a lie, "““can get halfway around the world before the truth even gets its boots on””." However, I am not happy with the rather brusque wording of the Bill in Clause 11, which in effect tells the board to keep its nose out of setting release times and leaving Ministers to help themselves to as much time as they want. I suggest that the Bill should be redrafted to recognise the principle of pre-release, but to place responsibility to decide with the board, subject to a duty to provide Ministers with adequate time to enable them to fulfil their accountability duties. The Government should welcome getting that principle underpinned by statute; the board should be required to consult Ministers before taking any decisions. If under the Bill the board can be entrusted to manage the RPI, subject to procedural safeguards, then it can certainly deal with release times, too. In my experience, 24 hours would normally be plenty, although I would not make releases on Mondays so that figures are not pre-released on a Friday. The fourth issue is that of the residual department. No doubt the noble Lord, Lord Moser, feels nostalgic about the days when the Prime Minister had a deep understanding of statistics. It sounds quite a good idea, put that way. But I think that there are problems. Making the Prime Minister the supervising Minister is in practice unrealistic and the Cabinet Office has little expertise in that area. But if the system is working, that is genuinely a residual role and, on balance, it would be better if it went to the Cabinet Office rather than the Treasury, which itself is a major consumer of statistics. I also welcome the suggestions made in the other place that there could be a statement, perhaps outside the Bill, that either the chair and/or the National Statistician could have access to the Prime Minister should a really serious issue of statistics or their governance arise. The other duty of the residual department is to set the budget. Some prior understandings have been mentioned as a way of ensuring that the board is not unreasonably squeezed, but that too could in the last resort be protected through the assurance of access to the Prime Minister. Finally, if there is to be greater accountability to Parliament, Parliament itself must reconsider how it performs its part of the process. Departmental Select Committees can look at statistics in their area, but what is missing is consideration of statistics across the piece. As the noble Lord, Lord Jenkin, mentioned, that could be a joint enterprise between both Houses.