Skip to main content

Language: English / Gàidhlig

Loading…
Chamber and committees

Finance Committee, 04 Nov 2008

Meeting date: Tuesday, November 4, 2008


Contents


Subordinate Legislation


Pre-release Access to Official Statistics (Scotland) Order 2008 (Draft)

The Convener (Andrew Welsh):

Good afternoon, and welcome to the 25th meeting in 2008 of the Finance Committee. I ask members to turn off any mobile phones and pagers.

Today we have apologies from Jackie Baillie, but Lewis Macdonald is her substitute. I bid you welcome and ask whether you have any relevant interests to declare.

I do not.

The Convener:

Agenda item 1 is consideration of a draft order. The order is subject to the affirmative procedure, which means that Parliament must approve it before it can come into force. The motion in the name of the minister invites the committee to recommend to Parliament that the draft order be approved. Before we debate the motion, under agenda item 2, we will have an evidence session to clarify any technical matters or to allow explanation of details. Officials can answer any questions if need be. I remind the committee that the officials cannot participate in the debate once the motion has been moved.

I welcome to the committee Jim Mather, the Minister for Enterprise, Energy and Tourism. With him are Rob Wishart, chief statistician; Marina Curran, assistant statistician; and Graham Fisher, head of the constitutional and civil law division.

I invite the minister and the chief statistician to make their opening statements, and remind the minister that he should not move the motion at this point.

The Minister for Enterprise, Energy and Tourism (Jim Mather):

Thank you. It is good to be back at the Finance Committee and to see some of my former colleagues.

The order addresses the extent to which ministers and the officials who brief them and others can be given access to official statistics before they are publicly released. There has been limited debate on pre-release access in the Parliament and at Westminster but it is an important question and the right balance is a matter of judgment. I would welcome the committee's thoughts on that.

As presented, the order reflects proposals that the previous Administration outlined in a consultation document in December 2006. It is based on the length of pre-release access that has operated for several years, which the previous Administration indicated its intention to maintain. My ministerial colleagues and I have some reservations about those proposals, given the limited amount of debate about the issue, but we are now happy to bring the order to the committee for consideration for the following reasons.

First, we are persuaded that the proposals are transparent and appropriate. The fact that they will be on a statutory footing in future is an important tightening up of the arrangements. Secondly, the proposals have now been out to consultation twice and there have been no objections from within Scotland or from the users of Scottish statistics. The only objection has come from the chair of the United Kingdom Statistics Authority, which had already publicly committed to a position on pre-release access before considering the Scottish proposals. The third reason—and my prime reason for advocating the proposals—is that the professional Government statisticians in Scotland are strongly of the view that reducing the time for pre-release access would have a negative impact on the integrity of statistics, which could adversely affect decision making and outcomes in Scotland.

We are keen for the committee to discuss and approve the order, but we can consider new issues that arise today. Fundamentally, we do not want the issue to be determined by party politics. The aim should be to agree an order to which any future Scottish Government can adhere. We have therefore suggested that the committee consider the views of the professional statisticians in Scotland, as set out in the note submitted by the chief statistician. He is happy to speak to that for the committee.

Rob Wishart (Scottish Government Corporate Analytical Services Directorate):

Our paper sets out the main reasons why we as professionals regard pre-release access as an important part of the arrangements to maintain integrity. I emphasise that our views are shared by the profession in the Scottish Government.

I will flag up two reasons. The first is that open and clear regulation of what happens in the run-up to release is vital to trust and integrity. Ending or reducing that form of pre-release would not end prior access to statistics, as I have shown in the paper, but it would make it very difficult for anyone to know what is going on, and it would increase the need for ministers and their advisers to speculate on what the statistics will show. Our proposals will mean that the rules that everyone follows in the week before publication are very clear and, critically, that the process is managed by the professional statisticians.

The second reason involves remembering what Government statistics are for: to inform debate on issues of interest to society and to inform the Government. If statisticians cannot explain the real messages behind the numbers to ministers, there is a real risk of misunderstanding and misinterpretation, and that no comment will be made. Ministers could rightly feel that they would need time to digest the information. That would be damaging to the democratic process and it would reduce the value of the information that we produce. It is therefore vital that statisticians have scope and time in the run-up to the statistics being released to ensure that ministers' comments on the statistics reflect a proper understanding and do not misquote or misinterpret the figures.

The paper mentions the various arguments against the proposals, many of which do not have a lot of substance. I would be happy to explore any of them should the committee wish it, but I will highlight two other key issues.

The length of time seems to have been the focus of political debate elsewhere. We suggest that professional statisticians should decide how much time is needed. If the committee accepts that, on balance, there should be pre-release, I suggest that we need the time to do it properly. It is essentially a practical issue, and we propose to continue with the current five days, because that is what works in practice.

Finally, on international principles, I hope that it will come as no surprise to the committee that I want the management of statistics in Scotland to be a model of best international practice, but that is not at all clear; what is available is aimed at independent statistics institutes, not at Government departments. For me, the best solution would be for statisticians to have considerable influence on what ministers say about the statistics in a system that has clear rules and makes it attractive for a range of management information to be produced under official statistics guidelines rather than through some other system. That is what the order seeks to achieve.

Thank you. I now invite questions from members.

Are there any individuals or groups of individuals who might be or have been granted pre-release access under the current regime who would not be granted pre-release access under the proposed one?

Rob Wishart:

In principle, there is no such specific group of individuals. We have changed the order so that it clarifies the purposes for which pre-release access should be given, so that it can be given only when it contributes to the purposes that are highlighted in the order. The statisticians would judge whether any individual would serve that purpose.

Can you give me an example of how the current situation hinders the process that you are describing? Is there a practical example of when you think something would have been better if the situation had been as you propose?

Rob Wishart:

The proposals are very much in line with current practice, although there is a slight tightening up. The proposal is to remain with a time of five days in normal circumstances. As I said, that is what works well from our point of view.

David Whitton:

I remember, in a previous life, having an argument with people in your department about when certain statistics should be released, most notably the unemployment figures. Is that the kind of thing that you are talking about? Special advisers and ministers have access to figures about a day in advance of publication and, if the figures are going the right way, there is always the temptation to do something with them. Is that the kind of thing that you are trying to get rid of?

Rob Wishart:

We use the time period to argue against such things. We would not consider making any attempts to influence whether statistics were published or what they said. It is currently very clear that ministers and their advisers do not get involved in that sort of discussion, and the order does not impinge directly on that sort of thing other than to say that in the period when they have access to the figures, they should not be engaged in that sort of activity because they can engage only in the activity that the order permits.

In case anybody gets overexcited by what I said, I should probably make it clear that officials in your department rightly prevented my enthusiasm from getting the better of me.

Lewis Macdonald:

I, too, have experience of the matter, but from a different perspective. My question is primarily for the minister. He said that the order largely reflects proposals that the previous Administration outlined, which ministers had reservations about, and that those reservations are now behind them. Where does the order diverge in substance—if anywhere—from the proposals that were consulted on? Were ministerial reservations dispelled because of changes that were made or as a result of the consultation?

Jim Mather:

We originally thought that we would have different rules for different types of statistics, but respondents to the first consultation favoured a smaller set of rules to avoid confusion. The order therefore focuses on market-sensitive statistics, which is what it should do. We now have pretty much what has worked in the past, which is the best laboratory for anything, with important statutory backing and extra safeguards.

So, essentially, the aim is to have past practice and what was consulted on, with the simplification of the various rules.

Exactly.

I believe that there have been 10 leaks of statistics since 2002. What procedures are in place to try to stop leaks? Has anything new been considered to enhance the regulations in that area?

Rob Wishart:

We investigate any leak that occurs. I cannot say that we have been hugely successful in identifying the source of leaks, but we have attempted to tighten up arrangements. If procedures have not been followed, we try to ensure that they are. Quite a clear set of rules already exists for staff and recipients of statistics. We try to ensure that we have not missed any angles.

The order will make it a statutory duty to comply with the current arrangements, which are purely voluntary—there is a code. That is the big difference that will occur. We will take a number of actions, if the committee and the Parliament approve the order, to ensure that people are fully aware of the requirements that they must meet. I could go through the specific arrangements relating to materials that are marked "Restricted" or "Confidential". Record keeping of who has the information will be enhanced. Specific arrangements will be in place.

Derek Brownlee:

Any wider political interest in the subject will probably be about instances of leaked information, which James Kelly mentioned. I want to discuss the sanctions that will be available. Let us take a hypothetical example. If a person who has leaked information and breached the conditions of the order is identified, are the available sanctions solely those in paragraph 7 of the schedule to the order? If I have read that paragraph correctly, there is an implicit threat that future access to statistics might not be granted to a person if they breach the rules.

Rob Wishart:

That would be one sanction, but things would depend on the circumstances. If someone leaked material that had been classified as restricted, they would be subject to any rules that apply to such behaviour. The cases that may arise may often involve people who are not fully aware of the rules or do not understand them. Obviously, we would ensure that they were aware of the rules in such cases. As I said, things would depend on the circumstances, but the main sanctions are what is set out in the order and whatever other action is appropriate as a result of leaking restricted material.

Derek Brownlee:

There will be inadvertent leaks and cases when the leak has no negative consequences; it is important not to go overboard.

It is easy to understand that there are other sanctions that you might apply if the leak involved a member of the professional staff, because that is also an employment issue. I am not for a moment suggesting that ministers would leak information, but supposing that, in a hypothetical situation, a minister was minded to leak information, it would not really be realistic to have professional statisticians saying, "Bearing this in mind, I'm not going to give you pre-release access in future." There is an imbalance in the relationship and the power, is there not?

Rob Wishart:

There is. Another factor is that leaks will be made publicly known. Although the arrangements for me to produce an annual report are in a state of flux, I will undoubtedly make an annual report of some sort to the new statistics authority. In addition, under our arrangements I am obliged to report any leak to the UK Statistics Authority, which will make that information publicly available. In the case of the hypothetical situation that you described, in which a minister is found to be responsible for a leak, the leak would be made known publicly.

Derek Brownlee:

That is helpful. I do not want to stray out of the remit of the Finance Committee, but I wonder whether the minister might be able to tell me whether such a leak—again, I stress that it is hypothetical—would be a breach of the ministerial code.

I am not sure. It would mean substantial damage to the reputation of the individual involved and the Government. The sanction is serious.

This is a statutory provision. Which body or which individuals will be responsible for investigating whether there is a breach?

I would assume that the committee system of the Parliament would kick in. The committees would take a lively interest in the matter and would apply sanctions. I look for clarity from Rob Wishart on that.

Rob Wishart:

The UK—

Jeremy Purvis:

I am sorry to interrupt but, other than the Standards, Procedures and Public Appointments Committee, which is responsible for breaches of the members' code, I am not sure which committee would be responsible for investigating whether there is a breach of the ministerial code, or statutory codes such as this one.

Would it not be for the First Minister to become involved in that?

I assumed that we have 100 or so active scrutineers on the benches who would be keen to put the balloon up at any breach of the ministerial code or good practice.

It would be helpful if, after further investigation, the minister would confirm that to the committee.

We would be delighted to do that.

I cut across Mr Wishart a moment ago. I am not sure whether he was going to add anything on that.

Rob Wishart:

I was simply about to say that, if it wishes, the UK Statistics Authority can investigate any issue that arises. We have a commitment to openness with that body. The order refers to the fact that the Freedom of Information (Scotland) Act 2002 would not be used as a reason to limit access to information. Generally, the UK Statistics Authority will report to this Parliament. In so far as the authority raises issues, the committee or the Parliament is at liberty to decide what it wishes to do about those issues. That is quite a powerful element.

Jeremy Purvis:

Further clarification would be helpful. Committees such as this one would have no locus for investigating any alleged breaches by officials. I may be incorrect, but I do not think that the ministerial code covers it either. Mr Brownlee started to ask what read-across there is to the ministerial code. It is not outlined in the draft order.

Apart from highlighting the standard provision, I think that all I can say is that we will have to do some homework on that and report back to the committee separately.

Mr Purvis, you have raised a number of questions that we can seek clarification on.

I would quite like to raise another, if that is possible.

Okay.

Jeremy Purvis:

On conditions of pre-release access, paragraph 5(6) of the schedule refers to

"Any accidental or wrongful disclosure of the statistics, or any broad indication of the contents of the statistics".

That kind of "broad indication" might be a leak of information rather than a leak of the statistics themselves. Under the various Administrations since the Parliament's establishment, there have been cases in which what was arguably a "broad indication" of information was given to the press outwith the formal publication of statistics. The order neither suggests what might constitute a "broad indication" nor indicates whether it would cover a leak of information or, indeed, how that would be determined under the code of conduct for members and the ministerial code. Do you have any more information on that?

Rob Wishart:

The interpretation used would be that of the responsible statistician. That wording has been used because it is very hard to draw a line in such matters; for example, a statement that unemployment is going up is a fairly broad indication of something that anyone might be able to guess. I think that anything that falls between that and a quotation of the actual figures becomes a matter of judgment and an investigation would have to examine whether the comment was made with knowledge of the information.

The order contains a clear duty to report

"Any accidental or wrongful disclosure of statistics … immediately … to the person responsible".

Jeremy Purvis:

I understand that. I am also interested in finding out about the kind of dispute that a responsible statistician and a minister might have over what constitutes a

"broad indication of the contents of statistics".

After all, the matter might well become a policy, rather than a statistical, consideration. What method is set out in the schedule for determining that? Under the legislation, does responsibility for that lie with the responsible statistician rather than with the minister?

Rob Wishart:

Under the terms of the Statistics and Registration Service Act 2005, ministers remain responsible for those matters. As the explanatory note makes very clear, however, ministers effectively leave a lot of these judgments to the responsible statistician.

I am very clear about the procedure that we have followed in the past, and I do not think that the order changes anything in that respect. If I consider that the code has been breached, I make a report on that basis. The UK Statistics Authority, the Parliament and the Permanent Secretary might also make a judgment on the matter; as a professional statistician, I have to judge whether the code has been breached.

Does the Parliament receive information from the UK Statistics Authority when it is published, not before?

Rob Wishart:

There are no specific arrangements for giving Parliament access to that information.

So how would we be able to investigate any potential breach?

Rob Wishart:

The information is made available to the UK Statistics Authority as soon as possible. If a leak has had enough of an impact to make Parliament aware of it, members can take action, as they will with anything that becomes publicly known or emerges as an issue. If there has been a leak, it will be self-evident in many cases, although not in every case. My report to the UK Statistics Authority happens relatively quickly.

Mr Purvis, you have had a fairly good run. Are there any final points?

Jeremy Purvis:

I am concerned by this. It has been indicated that there is a role for Parliament, but Parliament will know about something only if we read it in the paper. We will not know the content of a report to the UK Statistics Authority, and no mechanism exists for presenting to Parliament the information that will be presented to the UK Statistics Authority. The schedule to the draft order concerns statistics in a wholly devolved area. There could therefore be a deficiency in transparency; nothing in the schedule binds the UK Statistics Authority to presenting information to this Parliament. There would be a black hole in the information.

Do you wish to respond, Mr Wishart?

Rob Wishart:

The Statistics and Registration Service Act 2007 requires the authority to be accountable to this Parliament. Arrangements are not yet in place—they are still at an early stage—but the authority's practice is to make everything that it does publicly available. I would therefore expect any report to be publicly available. If this Parliament approached the authority to say that you wished access to reports of any leaks or other breaches, I think that the authority would be obliged to provide you with that information. We can certainly follow that up if you wish us to.

Graham Fisher (Scottish Government Legal Directorate):

Under the Statistics and Registration Service Act 2007, the Statistics Board has to provide the Scottish Parliament with an annual report. That act is the enabling act for the draft order that we are discussing.

Do you wish to pursue the issue, Mr Purvis?

No. If there is a commitment to come back to the committee with more information, that is welcome.

Lewis Macdonald:

Paragraph 5(7) of the schedule says:

"The individual—"

who could be the minister—

"must not seek to change format, content or timing of the publication of the official statistics, except in accordance with sub-paragraph (8)."

The meaning of the terms "timing" and "content" is fairly clear, but "format" is perhaps a wider term. Does paragraph 5(7) mean that a minister whose response to a statistician was, "I don't like the way you've written this news release and I think you should write it differently," would be in breach of the law?

Rob Wishart:

Paragraph 5 says that individuals may "comment" but may not "seek to change". Press releases take two forms, as you may well know: there can be a statistical press release or a ministerial statement. Clearly, the ministerial statement is the prerogative of the minister. The statistical press release would be subject to the order, which says that the minister must not seek to change it.

The term "format" really means "presentation". A long-established part of our procedures is that ministers should be involved in deciding neither the figures nor how they are presented.

Lewis Macdonald:

So, in a hypothetical situation, if a minister felt that a certain set of statistics demonstrated positive progress, but the statistician felt that the statistics showed only a standstill, there could be two Government news releases saying quite opposite things.

Rob Wishart:

Yes. But we would seek to persuade the minister of—

But if the minister sought to persuade you, he would be breaking the law. Is that right?

Yes.

The Convener:

There seem to be no more questions. The questions so far underline the importance of the work that you undertake and the need for absolute integrity.

We move now to the second item on our agenda, and I invite the minister to make an opening statement in the debate on motion S3M-02766, if he wishes, and then to move the motion.

Jim Mather:

What you have seen today is a reinforcement of the primacy of the responsibility of professional statisticians in the delivery of statistics in Scotland and a desire to build on what works, which will be augmented by important statutory backing and extra safeguards.

I move,

That the Finance Committee recommends that the draft Pre-release Access to Official Statistics (Scotland) Order be approved.

I invite contributions from members. Under the standing orders, the debate on the motion cannot last longer than 90 minutes.

Jeremy Purvis:

I will not take up that amount of time, convener. I certainly do not object to the order, but there are a number of areas where clarity would be helpful. I acknowledge that the minister has made a commitment to come back to the committee about that. I will not oppose the motion.

I invite the minister to wind up the debate.

Jim Mather:

I take on board Mr Purvis's point about clarity. We are keen to address the points that need to be clarified as a result of today's exchanges and to build on the relationship with and involvement of the UK Statistics Authority so that Mr Purvis can be confident that there will be comprehensive accountability to Parliament.

Many members will have noted the frequency with which the UK Statistics Authority has come up here and engaged with us, socially and at ministerial level. I expect that to continue. The key point is that we have a proposition that maintains the primacy of the responsibility of statisticians, recognises what is happening in practice and looks to augment that and put it on a firmer footing.

The question is, that motion S3M-02766 be agreed to.

Motion agreed to.

That the Finance Committee recommends that the draft Pre-release Access to Official Statistics (Scotland) Order be approved.

The committee will now communicate its decision formally to the Parliament by way of a short report. Are members content for that to be circulated and agreed by e-mail?

Members indicated agreement.