+44 (0)20 7799 6677

Honesty and Truth Sidelined In Government Policy-Making Says Think-Tank

The government is accused of sidelining honesty and truth in some of its major policy-making decisions in a new report from independent think-tank Civitas into the way that statistical evidence is collected and deployed.

In Failing to Figure, Mervyn Stone, emeritus professor of statistics at University College, London, demonstrates how ‘a minister sitting at the top of his departmental pyramid’ can put a blanket of confidentiality not only over all the advice he or she gets from policy-making civil servants within the department but also from any advisory committee set up by the minister (p.1). The recommendations of these committees result in allocations of very large sums of public money, and yet we are denied basic information about how their recommendations are arrived at:

  • We do not know who was asked to tender for the contract
  • We do not know why the successful bidder was chosen, nor why the unsuccessful ones were turned down
  • It is therefore impossible to evaluate the policies adopted against alternatives

Defenders of the system cite the independence of the advisory bodies that ministers rely on. However, as Professor Stone shows, they are very far from being independent in the sense in which most people would understand the term. For example, the £74 billion awarded by Primary Care Trusts (PCTs) to NHS providers represents a fifth of all public expenditure. It is carved up between PCTs according to a formula devised by a team at the University of Glasgow, following recommendations from the Advisory Committee on Resource Allocation (ACRA):

‘When the Health Committee began its inquiry, ACRA was made up of six Department of Health employees, five NHS chief executives, five people with GP affiliations, a research council chief executive and two academics-a health economist and a ‘social exclusionist’. Did these busy people have time for more than superficial analysis of the funding formula? Were they even qualified to do that?’ (p.46)

Questions have been raised about the fairness of a funding formula which leaves people in some PCTs with twice the allocation of funds as others (p.9), but it is impossible to debate the merits or demerits of the funding formula because:

‘… confidentiality was, and continues to be, maintained about the proceedings of ACRA and its Technical Advisory Group (TAG). A reporter for the Health Service Journal enquiring into the stop-gap resource allocations for 2008-09 was reminded by DoH that the work of ACRA was “confidential and restricted”‘. (p.46)


Professor Stone gives a number of examples of important policy decisions that were based on calculations that were shown to be not very robust.

When the eight eastern European nations joined the EU in 2004 the government predicted that immigration into Britain from these countries would be between 5,000 and 13,000 per year. Within a year this estimate was shown to be wrong by tens of thousands. When examined, it proved to be based on a number of incorrect assumptions, for example, the assumption that other large economies such as Germany and Italy would give unrestricted access as well as the UK (which could have imposed restrictions).

‘The research that ended with the 5,000-13,000 prediction must have been overseen by intelligent civil servants in the Home Office. Did no-one question the seriousness of the research proposal that got the contract? How many “expressions of interest” had there been? How many competing tenders were submitted-and how serious could they have been if they were judged inferior to the successful tender?’ (p.35)

Another example of ways in which statistical formulae can go haywire is provided by the way in which the Department of Communities and Local Government (DCLG) allocates over £20 billion to 456 local authorities. The formula deployed ‘has an arithmetic complexity whose logic has defied understanding even by those who devised it’ (p.35). Two academics were asked to examine the formula and used a standard statistical test of robustness: they left out one local authority and then ran the calculations again to see how they were affected.

‘When the researchers left out Bolton-one of the large authorities-the proportions of the now somewhat larger cake for the remaining 455 authorities changed only slightly. However, when puny Wokingham-the authority getting the smallest grant of under £8 million-was left out, the consequences were surrealistic. For example, Thames Valley Police lost £14 million while Birmingham gained £38 million! … This is the “unbelievable truth” that must now be acknowledged. The formula is so lacking in robustness that, without exaggeration, one can say that its tail of poorly-funded authorities is wagging the whole of the dog.’ (p.36)


Professor Stone gives a number of examples of important policy decisions that were based on calculations that were shown to be not very robust.

Mervyn Stone asks why there has been such a deafening silence about decisions that have a major impact on so many people. One reason is that the number of people who are qualified to analyse statistical reasoning is small, and most of them are in the universities. However, universities are bidding for the very contracts from government departments that they should be criticising:

‘It does matter for the soft scientific research that is directly funded by government departments using statutory contractual machinery. It matters because that kind of research encroaches on matters of moral and ethical sen-sitivity. Heads of university departments are under constant pressure to win contracts and individual researchers list the grants they have attracted on their personal web-pages… To stay in business, universities have to compete for lucrative government contracts… Too many of us have to serve the financial interests of our institution or department, sometimes by an unhealthily close relationship with government departments, and then remain knowingly silent on the contestable isues that we could help to resolve.’ (pp.40-41)


Mervyn Stone calls for greater transparency in the way in which government departments seek advice from outside bodies. Tenders should be invited from a wide range of interested parties. And all bids should be published. The winning tender should be published, together with the reasons for its acceptance. Professor Stone believes that a radical transformation is required in the quality of policy-making:

‘That change can only come about when Parliament asserts its authority to determine the framework in which the ‘public interest’?which is nothing more than defence of the citizenry against the arrogance of a ruling class-can be guaranteed.’ (p.64)

According to David Green, Director of Civitas:

‘Our system gives the Executive great power. Between elections the only effective check is public criticism, and if the abuse of power is to be controlled – whether it involves gaining the personal advantages revealed by the expenses scandal or forcing through ill-considered measures – then transparency is vital. Professor Stone has performed an invaluable public service by showing how even the most well-informed members of the public are prevented from effectively scrutinising public policies.’

Notes for Editors

Civitas is an independent social policy think-tank. It receives no state funding either directly or indirectly and has no links to any political party.

For more information contact: 020 7799 6677

Failing to Figure: Whitehall’s costly neglect of statistical reasonsing by Mervyn Stone is published by Civitas, at £7.50 inc. pp. ISBN: 978-1-906837-07-05. Buy from Amazon.


Keep up-to-date with all of our latest publications

Sign Up Here