Statistics and Registration Service Bill

Part of the debate – in the House of Lords at 4:06 pm on 26th March 2007.

Alert me about debates like this

Photo of Lord Desai Lord Desai Labour 4:06 pm, 26th March 2007

My Lords, it is a privilege to follow the noble Lord, Lord Moser, whom I have known for a long time and who has made distinguished contributions to statistics in this country and around the world. My perspective on the Bill is as a user of statistics when I was a professional economist, as someone who found statistics being debated in a political context when I became active in the Labour Party, and—like all noble Lords—as a reader of statistics when they appear in newspapers.

If they could, Governments would love to manipulate statistics. They are by nature so technical that people somehow trust the numbers. If a Government could somehow distort statistics in their favour, they would do so. The deception does not last long; it is found out quickly, but it offers a brief advantage, somewhat like insider trading. The Government of whom the noble Lord, Lord Jenkin, was a distinguished member changed the definition of unemployment 22 times during their tenure, in trying to get the numbers down. Each time they thought that we would not find out what they had done with the numbers.

What is important here is that we are to have a Statistics Board. One thing the new board ought to do, not only with regard to the statistics under its control but also in making the distinction between national and non-national statistics—a point made by the noble Lord, Lord Moser—is to give its imprimatur of quality to any set of statistics. All statistics should be clearly designated either as those approved by the Statistics Board as national statistics of high quality, or the rest, which may be released by any ministry so long as it is clear that they are not national statistics approved by the board. We will then know how much credence to give them. I say that not because I want to be partisan one way or another, but because the problem we face in statistics is that of educating the people using them. That includes Ministers up to the highest level. The problem is not that Ministers jump and pre-release statistics; it is much more dangerous when they do not understand what they are releasing and therefore get it the wrong way around. Actually, I do not know which is more harmful.

Newspapers too, when they publish statistics, often do not tell us what is or is not relevant, or what is the quality of the statistics. The noble Lord, Lord Moser, will remember how in April 1970, when the balance of trade statistics were released and there was one large payment towards buying some aircraft, the poor Labour Government were found to have mismanaged the economy and caused a balance of payments crisis. There was no such thing, but no one had the patience to find out that it was a freak payment for one month. People concluded that the Government had made a mess of the economy, and the party opposite won the election. No doubt they were grateful for statistics in their favour. That is not a matter of statisticians, however; it is a matter of educating the public about the right way of using statistics.

I hope the board will play its role not only in assuring that good-quality statistics are produced and marked as such, but also in doing something about the education of the general public about what statistics are and how they ought to be used. As the noble Lord, Lord Moser, rightly said, with regard to subjects like migration and crime the most appalling stuff is published all the time, and the people publishing it even carry the title of "Professor" before their name—as if that meant anything. Perhaps we need a publication so we can be told, when these numbers are published, how good they are. As long as that information is publicly available, in this day and age people will quickly pick it up and point out when Governments of whichever party are misusing statistics, especially when socially sensitive topics such as migration or crime are involved.

I welcome the Bill. This is a good opportunity to ensure that the trust in statistics is enhanced; that the high quality of statistics produced in this country is maintained, as it has been for a long time; and that we will make a clear distinction between who has access to statistics before general release and who does not. In a sense, I see the criticism made by the noble Lords, Lord Moser and Lord Jenkin, that the board has no role in the determination of the pre-release code.

On the other hand, I also welcome the fact that Parliament will have the crucial role in deciding that code. If Parliament—by which I mean both Houses—is allowed to play its role in a supervisory capacity in deciding how the code is drafted and how it is implemented, and has a chance to comment when there are deviations from it, it will be much better that Parliament has the supervisory function on this politically sensitive matter. No matter how independent a board is, people will always say that it has been nobbled by the Prime Minister or the Chancellor. Parliament at least can make sure that, given its supervisory power to bring Ministers to account and question them, it will do a better job. I hope that in Committee the Minister will clarify in what ways Parliament will be able to play a positive role in this pre-release matter. It is important that it does so, because the degree to which we have faith in statistics will depend very much on that pre-release code. Almost nothing is as important as that. It is up to the Government to get it absolutely right. The extent to which the board is given some advisory functions in that respect is up to us to decide in Committee deliberations. However, Parliament should have a more important role than the board.

I welcome the fact that explicit provision is made for the board to encourage research. This is a fast-developing field in statistics in terms not only of information technology and speed of dissemination, but of the high-quality theoretical work being done. It is good that the board will have a chance to encourage research and to maintain the quality of national statistics in the UK.

The role of the chairman of the Statistics Board is extremely important. The chairman must not only be sufficiently aware of the technical problems surrounding statistics, and therefore able to make sure that good statistics are produced; they must be aware also that statistics are a socially sensitive matter. When the census comes around, we will find out how many cultural and religious conflicts there will be over questionnaires—whether we ask questions about ethnicity, religion and other various matters. It will be up to the chairman of the board, whoever he or she is, to make quite sure that the quality of statistics is not sacrificed because we cannot counter people's prejudices about what the statistics are about. That will be the case in terms not only of the census, but, as I have pointed out previously, of a variety of other numbers—on crime, on migration and on other matters.

Statistics are a politically and socially sensitive, almost explosive, matter. It will be for the chairman to have both the technical authority to be able to say, "These are good statistics", and the political nous to be able to assure citizens that the statistics collected and produced in their name are not being used by the Government in a manner that is hostile to their interests. Appointing the chairman will be a difficult task. I am sure that we all look forward to the choice being made, and I wish the Government good luck in that.

In most matters, we are too centralised in the UK—we deplore that. In the matter of statistics, we are too decentralised—we deplore that also. About the best that one can say is that if the Statistics Board were to mark for quality the numbers produced by other agencies—it should have moral authority to do that—it would be a sufficiently good first step towards co-ordinating the quality of statistics across different agencies.