Scottish National Statistics
The final item of business is a members' business debate on motion S2M-4171, in the name of Jim Mather, on ensuring the independence of Scottish national statistics. The debate will be concluded without any question being put.
Motion debated,
That the Parliament, in light of the proposals from HM Treasury for the establishment of an independent UK Office for National Statistics, notes that no proposals were made for Scotland; further notes that such proposals for Scotland are the responsibility of the Scottish Executive, and considers that, in light of previous statistical issues such as the loss of Objective 1 funding for the Highlands and Islands through statistical error, the Executive should take similar measures to place the production of Scottish statistics on a statutory basis, incorporating the National Statistics Code of Practice, and establish an independent Office of Scottish Statistics with independent governance answerable directly to the Scottish Parliament.
First and foremost, I thank those members—not all of whom are here—who gave me cross-party support to ensure that this debate could take place. Despite the turnout of members, the debate is important, because it is about trust in official statistics. That issue continues to grow as demands for evidence-based policies, international comparability and transparency increase.
Statistics drive decisions that affect everyone, including decisions on the allocation of public money to local government and the national health service. Operational decisions, policy interventions, policy evaluations and assessments of public service performance depend on statistics. The electorate obviously has a right to know as part of the democratic process.
Trust in statistics is essential for effective administration and the delivery of public services. If members of the public do not trust the figures, they will not trust decision makers and they will not believe Government statements. Survey results show that two thirds of the United Kingdom public believe that figures are changed to support arguments and that information about mistakes is frequently suppressed. I am among that 66 per cent. I am also among the 50 per cent-plus who believe that there is political interference in the production of statistics.
The June 2000 "Framework for National Statistics", which was signed by Scottish ministers and members of other Administrations, was a move in the right direction. It established a more formal code of practice and set up the Statistics Commission. Now, in "Independence for statistics: A consultation document", the Treasury proposes to entrench UK statistical independence in legislation by
"introducing direct reporting and accountability to Parliament, rather than through Ministers"
and
"placing a statutory responsibility on a new independent governing board to assess and approve all National Statistics against the code of practice, also backed by statute".
Those are further steps in the right direction, but what about Scotland?
In such a climate, the Scottish Parliament and the Executive have several options. Option 1 is that we do nothing and revert to pre-2000 arrangements, with no Statistics Commission and no statutory code. That would fail to meet any of the objectives of a statistics service. There would be no independent oversight, no guarantee of impartiality, no openness in selecting statistics to produce and no consistent handling and release of statistics. That would go completely against the UK Government's moves and recent legislation, such as freedom of information legislation. If that happened, the independence that has been proposed for the Office for National Statistics in London would be sacrificed, as Government statisticians there would use Scottish figures that would undermine their independence and the quality of their data, which would be disastrous.
Option 2 is that we replicate the current arrangements and set up a statistics commission for Scotland. The Statistics Commission for the UK has consistently worked to have itself replaced as a statutory body, so that would get us off to a poor start. It would, essentially, take us nowhere and give us no momentum. It would create no opportunity for Parliament to be the arbiter of the quality of our statistics.
Options 1 and 2 are seriously flawed.
That brings us to options 3 and 4. Option 3 is that we create an independent statistics office for Scotland that is modelled, in part, on the proposal for the Office for National Statistics. Option 4 is that we replicate the Treasury's proposals for the oversight of statistics at Whitehall and create a statutory compliance and audit body. Either of those proposals would be a significant improvement. They would offer greater independence for the Scottish system and ensure that the current UK system was not compromised by a lack of Scottish independence and accountability. That would be in the best interests of everyone and would offer a high-integrity direction of travel. Under current constitutional arrangements, there is sense in mirroring the proposed structure of the Office for National Statistics so that it and the Scottish body can work together without compromising the work of either.
Scottish legislation should be passed to enshrine that objective. We need a world-class system. Scotland is one of the few countries in the world that does not have statistics legislation. We need to correct that anomaly as soon as possible. No country can now join the European Union without such legislation, so we should not operate under a lesser standard. We should aim to meet all the statistical standards of the European Union, the Organisation for Economic Co-operation and Development and the United Nations.
Ideally, the Scottish body should report annually to the Scottish Parliament, for example to the Finance Committee or the Audit Committee. With the current constitutional arrangement, the Scottish body should have statutory representation in the UK body to ensure that the UK figures relating to Scotland comply with the same standard of independence. At no point should there be ministerial control over any element of the statistics system; it would be better for ministers to work on the statistics that come out of the process.
I echo the words of the 1997 Labour Party manifesto:
"Unnecessary secrecy in government leads to arrogance in government and defective policy decisions. The Scott Report on arms to Iraq revealed Conservative abuses of power. We are pledged to a Freedom of Information Act, leading to more open government, and an independent National Statistical Service."
That is exactly what we need.
My guru, W Edwards Deming, who transformed modern manufacturing, said that no improvement is possible in any system until that system is under objective, accurate statistical control. We should be satisfied with nothing less, especially when we are talking about the statistics of our country.
I congratulate Jim Mather on securing the debate and on not quite managing to clear the public gallery, as perhaps might have been expected given his chosen subject. I also congratulate him on mentioning Mr Deming once again. That seems to be a recurring feature of Jim Mather's speeches, which shows that persistence is one of his virtues.
We all agree that it is important that we can have confidence in statistics and that we have an objective and impartial statistics service. We can agree on that as a general aim; it is how we get to that that is more difficult to agree on. I am not saying that the current statistics system is not working at all; I am just acknowledging that it could be improved and that good statistics are important not just for good government, but to enable businesses, voluntary organisations and a host of other groups in society to plan ahead and to evaluate what work is being undertaken. The worst thing that we could do would be to have statistics that had all the reliability of a bar chart on a Lib Dem leaflet—that would not be a helpful manoeuvre. An independent statistics service would increase confidence.
Jim Mather is right to learn some lessons from what has happened south of the border. We should not be afraid to learn lessons from Westminster on this issue. The Treasury is only now, in 2006, consulting on an independent statistics service, despite having promised one back in 1997 in the manifesto to which Jim Mather referred. That has handily allowed the Chancellor of the Exchequer to play fast and loose with a number of statistics, not least those on the economic cycle. It has conveniently allowed him to pretend that he has managed to keep to his golden rule rather than breaking it. I notice that such is the interest of Labour members in statistics and the record of the chancellor that none of them is here to defend it.
The current Government at Westminster created the Statistics Commission in 1998. Of course, its objective then was
"to advise on quality assurance and statistical integrity across the statistical system"
if we are to believe the HM Treasury consultation document. If the proposals in the consultation are enacted, the Statistics Commission will be abolished.
The Statistics Commission has played a valuable role, to some extent. In its 2004 report, it revealed that the Government had been using "unreliable figures" for "political reasons". That is very unlike this Government; no doubt the minister will tell us that it is purely a Westminster problem. In the same report, the Statistics Commission also identified several instances where the Scottish Executive and Scottish ministers had breached the statistics code of practice, so we should not be complacent.
Jim Mather outlined four options that we are faced with. The idea that we can go back to what existed before 2000 is nonsensical; I agree with him on that. If we are to have confidence in the statistics that we use in Scotland, we need to have a reliable set of statistics that everyone—Government, Opposition and everyone else—can deal with. An independent statistics service that is underpinned by legislation and accountable to the Parliament is a healthy proposal and one that members across all parties could support. I hope that the minister will give us a firm steer in his remarks tonight about the route that the Executive is likely to take. As I said earlier, I do not think that the route of doing nothing is one that the Executive can credibly take if it is to inspire confidence in the statistics on which we all depend.
I join Derek Brownlee in thanking Jim Mather for bringing to the chamber this debate on a very important subject. Given the way in which the issue of waiting times is bandied about—there is a statistical rammy about waiting times at just about every First Minister's question time—I am disappointed that there are not more members in the debate. We use statistics as the bedrock of our political discussion; where they come from and how they are assessed should be much more seriously dealt with. The responsibility of Parliament is to hold the Executive to account. The key element of that is the Parliament's ability to measure what the Executive is doing and to assess the success of Executive programmes. All of that means that we have to have reliable statistics.
I have an A level in statistics, so I am well aware that they can be presented in many different ways. However, whatever the presentation, the underlying statistics should be acceptable, reliable and agreed by everyone. Jim Mather made the entirely valid point that ministerial control would be inappropriate in any kind of office of national statistics. Although I am sure that George Lyon is in every way beyond reproach in how he deals with statistics, it is important that ministers are not only beyond reproach but clearly seen to be beyond reproach. The source of the statistics that we chuck around the Parliament like confetti must be seen to be independent.
That is why I agree very strongly with Jim Mather and Derek Brownlee that we need an independent office to act as a creature of the Parliament. It should be reliant not on the Executive but on the Parliament. My understanding is that that would be quite similar to the situation in the United States of America, where there is an independent congressional office of statistics that fulfils that purpose. As we discuss the potential creation of another commissioner, ombudsman or creature of this Parliament, it is worth bearing in mind that, under the current constitutional set-up, that is the only viable way we have of making certain that the information and activities that are undertaken by an office of statistics, a commissioner or another institution are genuinely independent of the Executive.
I welcome Jim Mather's efforts in securing the debate. I hope that Scotland can move forward in a way that is similar to the way in which the Treasury proposes that England will move forward, and that we will be able to create an office of national statistics as a creature of this Parliament, accountable to Parliament, and not to ministers.
I congratulate Jim Mather on securing a debate on this important subject. It is disappointing that there appears to be so little interest in the subject in the chamber.
As the speakers have pointed out, the issue is important. Statistics affect the decisions that we make about the spending of huge amounts of Government finance and Government-raised taxation and underpin the political discourse that takes place in the Parliament. A great deal centres on statistics, their reliability and what they have to say about whether policies are working.
The debate gives the Parliament an opportunity to feed into the on-going ministerial discussions about how we are going to respond to Whitehall's proposals.
Ministers have a strong commitment to the integrity of official statistics. We recognise that reliable and objective statistics are vital not only to informed decision making by Government but to public and parliamentary debate. In particular, we recognise that the professional independence of statisticians from ministers is a key feature of statistical standards across most nations. Our current arrangements reflect that. Ministers are not involved in decisions on the format, content or timing of statistics releases and our compliance with standards on those matters and with all other standards is open to constant scrutiny by the independent Statistics Commission.
We will, of course, be considering whether any changes are appropriate in the light of changes that are being pursued by the UK Government and I do not wish to pre-empt that matter, although I was interested to hear the views that were expressed by members on the subject.
I would like to correct some of the impressions that might have been given during the debate. First, it is entirely wrong to suggest that the statistical error that is referred to in the motion would have been prevented had there been a separate office of Scottish statistics. Indeed, the errors were in statistics produced by the UK Office for National Statistics, which, even then, was largely independent of Government. The member might wish to reflect on the fact that it was Scottish Executive statisticians who identified the error and that their understanding of the policy implications—stemming from the fact that they were working within Government—was a key factor in their ability to identify the mistake. We shall endeavour to ensure that they are able to continue to exercise that function under the new UK set-up.
It is also important to understand the type of statistics that the UK Government will be transferring to the new independent body. A range of economic and labour market statistics that are produced by the UK Government rather than the Executive will be transferred, as will statistics such as neighbourhood statistics and registration statistics, which raise few concerns. The UK Government will not be transferring most of the types of statistics that the Executive produces. For example, it will not be transferring education statistics, crime statistics, or homelessness statistics.
We will also wish to weigh up proposals to increase organisational independence against the considerable benefits that are to be gained from having statisticians work as part of an integrated approach to the analysis that underpins our policy making and from ensuring that they have, in the words of the UK consultation document,
"good working links to policy makers, allowing them key insights into developments and needs".
The "National Statistics Code of Practice" ensures that that is done in a way that does not impinge on the independence of statistical decisions and that statistics are made publicly available to everyone at the same time. I am sure that the member will agree that decisions that we take on independent production need to be right for Scotland and should not simply follow what might be right for the UK Government. Devolution, of course, allows us to ensure that that is the case.
It is also important to recognise that a substantial part of the UK proposals—and, indeed, the part that is most relevant to Scottish Executive statistics—is about independent scrutiny of the statistics that are produced by Government rather than about independent production. Those are two separate matters. At present, we are part of the scrutiny arrangements that apply by mutual agreement throughout the UK, so we will want to consider our position on the matter as the UK Government's arrangements start to emerge.
As I said at the outset, I welcome the contribution of today's debate—and of the issues that are raised in the UK Government's consultation document—to deciding the best way forward on the matter. The key point is that we must get the right balance between ensuring that Government statistics are robust, reliable and subject to independent scrutiny and ensuring that they provide the proper information at the right time to shape decisions that involve large sums of public expenditure and many issues that affect the lives of the people of Scotland.
Tonight's debate has been helpful in teasing out some of the concerns that exist and allowing us to consider a range of opinions before we make our final decisions. I congratulate Jim Mather on bringing the debate to the Parliament.
Meeting closed at 17:30.