Statistics and Registration Service Bill

My Lords, it is with some trepidation that I intervene in the Second Reading of this Bill on the Government’s statistical services. I am no statistician, unlike others in this House with much greater qualifications to be heard, prominent among them my noble friend Lord Moser, whose views on this Bill I broadly share. Statistics matter to us all. They are an essential foundation for policy-making, act as an early warning when damaging trends are beginning to develop and are at the heart of handling such hugely complex issues as global warming and climate change. We need to take statistics seriously, even if our proficiency in understanding them is limited and our capacity to be misled by them considerable. The capacity for ordinary people to be misled by statistical information is at the heart of the current legislative proposals. The misuse of statistics, not just by politicians but also by pressure groups or any group of people with an axe to grind, is a phenomenon of fairly long standing. It is not something that has just arisen in the past few years; here I separate from some of those who have preceded me in this debate in attributing the decline in trust to recent times. It was a former Member of this House—the Earl of Beaconsfield, better known as Disraeli—who gave us the epigram that there are, "““lies, damned lies and statistics””," and he was no slouch at manipulation. The problem seems to be getting worse, perhaps largely because so much more statistical material is made available. The result is that the credibility of that material, and public trust in it, seems to be dropping quite sharply. The Government are to be congratulated on bringing forward legislation designed to strengthen the autonomy of their statistical services and thus, one would hope, to enhance public confidence in the material they provide. The overall thrust of this legislation is admirable, but the delivery of real outcomes is half-hearted. The detail of the Bill seems to fall well short of its broad objectives. In some cases, the proposals seem seriously flawed. As so often happens, the other place appears to have done little to confront those flaws, so I hope that it will be possible to remedy some of these as the legislation is scrutinised in your Lordships’ House. The first area of doubt arises over the structure chosen for overseeing the activities of the Government’s Office for National Statistics and accepting ultimate responsibility for them. Noble Lords preceding me have all mentioned this point. It is surely right to consign that task to a depoliticised body and to remove it from ministerial interference; but the idea of having a single body in which non-executive appointees and executive professional statisticians are mingled, and which has an uneasy mix of supervisory responsibilities and everyday executive ones, does not at first sight seem to be the best way to proceed. It is exactly the opposite of what has been chosen for the BBC, where the supervisory and executive functions have been separated. It is the opposite of the way we handle the governance of our higher education institutions. It will be interesting to hear how the Minister justifies the proposed structure, but it could well be necessary to rethink the approach when we come to Committee. Then there is the question of resources to be made available to the Office for National Statistics to do its work. This is not something that normally figures in legislation and this Bill is no exception, but in introducing it the Minister said something about that aspect, repeating what was said in the other place. How real is the autonomy being granted to the office if the purse strings are still ultimately held by Her Majesty’s Treasury? How much confidence would we have in the autonomy of the Bank of England when it comes to setting interest rates if it relied totally on the Treasury for the resources it needed to carry out its research? Not much, I suggest. There is a case for underpinning the autonomy of the Office for National Statistics in some way analogous to that applied to the resource requirements of the National Audit Office. Indeed, the more general case for giving Parliament a direct role in guaranteeing the autonomy of the Office for National Statistics would seem hard to refute. I turn now to the sensitive and vexed question of the arrangements for the pre-release—in advance of publication—to departments and Ministers of the result of statistical inquiries. This is at the heart of the issue of public confidence. It is the belief that statistics are spun and their presentation manipulated for political purposes that is undermining the credibility of our national and departmental statistics. The arrangements proposed are complex, Byzantine and almost certainly capable of being operated in such a way as to provide no substantial change from the present unsatisfactory arrangements. It is worth pointing out that if pre-release was simply banned, there would be no need to provide for any such complex arrangements in the Bill at all. I believe I am right in saying that Norway operates a ban on pre-release; why can we not do so too? I have little doubt that the Minister will say that statistics are often highly sensitive and it is only right for Ministers and their officials to have time to prepare a public reaction to them, but does the argument really hold water? Perhaps the Minister can say whether there are any arrangements for the pre-release of decisions on interest rates by the Monetary Policy Committee of the Bank of England. Are they not published as soon as they are taken? Surely nothing in the economic field is much more sensitive than those decisions on interest rates, so why can the publication of statistics not follow a similar course? It is no secret that the tabling of this legislation has been greeted with less than wholehearted enthusiasm by those parts of the press with some specialist knowledge; the rest of the press has probably not noticed that it exists at all. The Economist magazine was particularly scathing. While that does not prove that the legislation is at fault, it does underline a real risk; namely, that the detailed provisions of the Bill will undermine its overall target of improving the objectivity of our national statistics and of increasing public support for and confidence in them. If that were to happen, it would be a major opportunity lost, and it will not be one that will recur any time soon. Quite rightly, legislation like this is designed to last for at least a generation, perhaps two, so it is worth getting it right.