The UK Statistics Authority started life on 1 April, 2008, as an independent body directly accountable to Parliament. The Authority was created by the government with the specific purpose of rebuilding trust in official statistics, which was widely seen to be at a low level: in a recent European Commission survey  of public trust the UK came in 27th out of 27 countries.
Too much time
It has always seemed to me that this low level of trust partly arises from the way in which successive governments publish and communicate statistics. There has been too much time allowed for ministers and political advisers to see statistics ahead of publication, and too great a tendency to leak them selectively.
The present government has itself acted very recently – in December 2008 – to reduce sharply the time available for Ministers and their advisers to see statistics ahead of publication.
This is very welcome, but does not go far enough. The principle should be that the statistics which are going to be published are produced by professionals without any political influence then shown to everyone – government ministers, the opposition, and the media – at the same time. Then the political debate can begin.
New Code of Practice
The Statistics Authority has just published a Code of Practice , against which it will assess the production and handling of official statistics. This Code sets high standards which, if they are observed, ought to do much to rebuild public trust.
This is bound to be a bumpy ride at times. Just before Christmas I had to write to No.10 Downing Street to complain about the way some statistics on knife crime had been handled by the Home Office, under pressure from political advisers in No. 10. My complaint drew an apology from both No. 10 Downing Street and the Home Secretary, and it is my fervent hope that this high-profile incident will make people in government think twice before they ignore professional advice about how official figures should be handled.
The new Code of Practice is not just about the dangers of political manipulation. It sets standards for the producers of statistics right across government and its agencies, with the aim of improving the quality of official statistics, so that they deserve the higher level of public trust we hope to see.
We want to ensure that the right statistics are produced, taking full account of the needs of all users, and in the public interest; that the highest professional standards are maintained in all aspects of statistical production and publication; and that the statistics are communicated to those who need them in ways that lead to better decision-making, and better understanding of our economy and our society.
Quite separately from the regulatory role I have described, the Authority is, under the statute which created it, a major producer of statistics. We are the governing body for the Office for National Statistics, headed by the National Statistician, and so are responsible for the major economic and social statistics, including the National Accounts and the Census.
We have a very big job ahead of us. I do not expect a sudden or early change in the measures we have of public trust. But, with the new Code, the new pre-release rules, and the new arrangements we have made for publishing statistics separately from political commentary, we have made a determined start. I hope we will continue to attract strong public support, because what we are trying to achieve is, I believe, hugely worthwhile