Appointment of the Chair of UK Statistics Authority

Written evidence submitted by Ian Maclean MBE, Chairman, SUC 1989-2004 (UKSA 03)

1 Summary

• Legislation to amend the 2007 Act is essential if official statistics are to meet the needs of a participating democracy in the 21st century.

• Departmental statistics that are in the public domain are an integral part of ´national statistics` and should be under the control of UKSA.

• Pre-release requirements are out of line with other developed democracies and should be strengthened.

• The role and scope of official statistics requires clarification. Does serving the ´public good` just mean making statistics collected for government freely available or is the broader concept of providing a service that considers the needs of society on equal terms with those of the government.

• A marketing based structure should be adopted that starts with identifying the needs of all customers/users and then determines the best way of meeting those needs within approved budgets.

• An effective user consultation mechanism is still lacking. Consultation with users is widely debated, but inadequate in practice. A formal structure is needed.

• Take statistics directly to the public through a packaging programme that will ensure that the statistics that will initiate and inform debate on key policy issues are in the public domain alongside the RPI, National Income and the Balance of Payments.

• The Rayner cutbacks in detailed business statistics should be restored. 

• The relationship between UKSA and the ONS, in particular that of the Chairman and National Statistician, should be reviewed as again their roles and reponsibilities are not clearly defined.

• A vote of thanks is is due to Sir Michael Scholar for his active intervention against the misuse of statistics by Ministers and the threatened cutbacks to Health and other key statistics. He will be a hard act to follow.

2    Legislation  

The 2007 legislation was long overdue as the UK was the only developed country without a stastistics law. The debate on legislation had started as long ago as 1966 with the House of Commons Report on Statistics; been the subject of my paper at the 1978 Statistics Users` Council annual conference; the basis of the 1990 RSS report on ´Counting with Confidence`, included in the Labour Party manifestos of 1992 [ which I helped write] and 1997, leading to the Framework Document for National Statistics in 2000, setting up the Statistics Commission and their report in 2003 on the justification for legislation - which I prepared. Interspersed were several Green and White Papers and Treasury Select Committee hearings. All supported by the 1991 UN listing of The Fundametal Principles for Official Statistics and their 2004 publication providing in detail the statistical laws of over 50 countries. Given this background it would have been reasonable to assume that the legislation when it finally arrived would have set world beating standards for the role and scope of official statistics as a window on the performance of the Government and the basis for informed debate on key public issues by society at large , or in the more emotive French term - the citizen. Not so, the announcement in November 2006 was sudden and unexpected and the Bill was drafted rapidly and rushed through. Only the House of Lords with their strenuous efforts in the committee stage emerge with any credit. The emphasis on ´building trust´ was a distraction from what should have been the real aim of the legislation - modernising the scope and role of official statistics for the participating democracy that we all hope we are living in. The UK is the only developed country where the public mistrust of official statistics is a live political issue, all other statistical legislation is concerned with building an official statistics system to serve democratic debate. Paradoxically unless the position of the Statistics Authority is strengthened it cannot even fully fulfil the building trust function, let alone fully embrace the wider issues of transparent government and ways of engaging in a meaningful dialogue with citizens .

3 Departmental statistics and early release

3.1 These two are linked by the common theme of conflicting views between Ministers and UKSA. The battle lines have already been drawn and while Ministers continue to act virtually as feudal barons, legislation is the only solution if democracy is regarded as preferable to autocracy. Many of the Ministers claims do not stand up to close examination. Is the Treasury handicapped by the fact that it does not control National Income and the other statistics it uses? The question of funding is a real issue, but the answer is to provide a mechanism for the costs to be included in the UKSA budget?

3.2 On early release the question that has to be answered is do we want to be in the backward or leading group of countries?

4 Role and Scope

4.1 There are numerous references to ´the public good`, but while that is not defined in legislation there is the continuing prospect of widely different interpretations. If our ´participating democracy` is to operate effectively the role of official statistics has to be to meet the needs of society at large, not just Government, and scope has to be defined as providing the statistics needed by society even if Government has no need for them. There is a precedent in the 1968/9 revision of business statistic. It is a sad thought that the UK was then in the vanguard, not the rearguard, of official statistics developments.

5   Marketing, User Consultation and Engaging with the Public

5.1  UK official statistics are production orientated. They represent a vast and greatly underused resource that should be much more fully exploited. In the business sector marketing is the answer to meeting customer needs and it applies equally to official statistics. A marketing department was introduced under Bill McLelland, but it was an alien philosophy and did not survive his departure. The basic question posed by marketing is ´what market are we in`? If the answer is society at large, Goverment, business, academia and the public, then the next question is ´do we have the right products to serve this market`?

5.2 The elements of the marketing plan include:

• a product audit to establish who and what use is made of each statistical series or report

• market research to identify the need of each sector of the market

• a product development facility to modify or produce ´new` statistics

• a dissemination programme that concentrates on raising awareness of the value of statistics for decision making and improving accessability.

• Making full use of the internet a potentially powerful tool for the mass dissemination of statistics. It would be interesting to see the impact of the National Statistician´s Twitter and the Chairman´s Facebook.

5.3 The term ´user´ has overtones of just relating to exising outputs and should eventually be replaced by ´customer`, but for the moment the term´ user´ will be assumed to cover both.

Users can be divided into many sectors, but broadly there are two main categories, expert and the rest including the public. The former are well catered for by the Statistics Users` Forum, the successor to the Statistics Users` Council set up by Claus Moser in 1970 as a de facto solution when the 1966 House of Commons Report on Statistics did not lead to legislation. The practice in vitually all statistics legislation is to incorporate a Statistics Council to represent the views of of users, but there is no common format, they vary widely in scope and size. The EU is probably the best model, starting in 1991 with an advisory committee of eventually over 100 members revised down to 25 in the 2009 Statistics Law. The exiting features – for users - of this law apart from the advisory committee are the express references to user requirements and reporting lines.

´The ESAC (European Statistical Advisory Committee) shall assist the European Parliament, the Council and the Commission in ensuring that user requirements and the costs borne by information providers and producers are taken into account in coordinating the strategic objectives and priorities of the Union's statistical information policy 

The EU law was the part of a much wider programme, Democracy Dialogúe and Debate that was launched in 2006/7

5.4Given the existence of the Statistics Authority, incidentally the only body of its kind in the world, it would not be realistic to raise the SUF to the ESAC status at this stage, but there are good grounds for bringing the SUF into a formal reporting relationship with the Authority, including a seat for the SUF Chair on the Board. 5.5This still leaves the difficult task of meeting the needs of a largely statistically inarticulate public. The best solution would be to for UKSA to set up a committee to identify issues of public concern and ensure that existing statistics were made available in an easily accessible form with missing gaps filled where appropriate. Packaging is one of the solutions as relevant statistics are widely scattered and need an experienced hand to be brought together in order to fully inform issues such as child poverty or public sector pensions that are under debate. It is not enough to regard official statistics, in the memorable words of a previous National Statistician, "as a quarry" in which users were free to dig at their leisure.

6 Business Statistics

Given their own heading as they are a special case. Following the reforms of 1969 the UK had the excellent, highly detailed, quartely Business Monitors covering over 5000 product headings. They were widely appreciated, selling over 600,000 reports annually in the 1970s . Sadly they fell victim to the Rayner statistical massacre. It is ironic that the form filling burden argument was refuted by Armstrong Rees Report which, although instructed by Francis Maude to deliver the final blow to the Business Monitors, noted that the average time taken to complete the return was just 2 hours. So a valuable business asset was sacrificed to what the late Jack Hibbert described as political dogma. As manufacturing induastry was increasingly being seen as a dinosaur it didn´t seem to matter too much, but our over dependence on the financial sector was exposed by the credit crunch and we have the example of the German recovery based on manufacturing as an extra reminder that perhaps we should seek to rebuild our manufacturing sector. Better short term statistics would help and could be easily and cheaply collected by using the monthly Intrastat returns. Again the UK is unique among large developed economies in not having a short term detailed product series. Incidentally a Treasury Select Comittee hearing on statistics in 1998 recommended that the ONS should develop such a series.

7 UKSA and the ONS

 7.1 As an outsider it is difficult to know who is responsible for what. UKSA would appear to have the overriding responsibility for ensuring that user views are taken into account, but most of the face to face meetings are with the ONS or Departments and bodies such as PUG - Producer User Group appear virtually unannounced and do not widely publicise either their meetings or the results.

7.2 In terms of governance a casual observer would regard the Authority as the policy making board and the ONS as the production unit , but life is not that simple. The assessment function of the Authority looks remarkably like the quality control department normally associated with the production function and the National Statistician has developed her own Office which looks like a policy unit, so what policies fall within the purvue of each organisation? The relationship between the Chairman of the Authority and the National Statistician is again not clear. The National Statistician appears to be the managing director of ´UK Statistics Ltd` while the Authority, in addition to assesment, is responsible for combating the misuse of statistics by Ministers and ensuring adequate funding. Yet who is actually the public face of National Statistics? The term National Statistican has a ring about it , as Michael Fallon observed in one of the debates on the Bill. " It should become one of the big offices of state. I should like the National Statistican to become a houshold name - a key public figure in leading the profession and championing the public interest in statistics" Perhaps he was influenced by the practice in Australia where the National Statistician is fourth in line. Cetainly with just two days a week for the new chairman you do not need to be clairvoyant to see where the balance will swing.

May 2011