Apologies for the slight delay in opening the meeting. I welcome the press and the public and ask that all pagers, mobiles and BlackBerrys be switched off. We have received no apologies from members of the committee.
I begin by congratulating Wendy Alexander on her promotion to the post of convener. I look forward to working with her over the coming months.
I will run through how it is envisaged that the bill, as proposed in the legislative consent motion, will work in Scotland.
Thank you. We move straight to questions. I will kick off by probing something that has been touched on, in order to get it on the record. Given the developments in the rest of the United Kingdom, what would be the problem with remaining with the status quo?
It would be seen as unusual if we did not move in response to the changes that have been made down south. The new body that is being set up will not only take on board the ONS work but, more important, oversee quality. Quality is key in all of this—it is not so much about how it is done as about ensuring that we get quality statistics throughout the UK. The announcements and the bill down south meant that we were faced with deciding what we wanted to do in Scotland. After significant debate on the matter—indeed, Jim Mather led a debate on the issues concerned in the Parliament—we came to the view that it was in the best interests of Scotland to join the new statistics board. However, we secured significant concessions to reflect the Scottish situation, namely the powers that ministers will have in various areas and an important commitment that the new organisation will report to this Parliament as well as to the UK Parliament. I suspect that the committee may have a role to play in probing the veracity and quality of the statistics that are produced for Scotland.
We are currently part of the national statistics arrangements, which will in effect be replaced by the new arrangements. As the minister has said, in some respects there is no option but to review what we have got. In a sense, the status quo disappears because of the changes that the UK Parliament has set up.
It occurs to me that at the next formal meeting of the committee we will be probing the veracity of national statistics and the authorised code of practice.
How can we ensure that the bill will improve public trust in Government statistics, given that there is a fair degree of scepticism among the public?
The independence of the new model, the way in which it is constructed, the fact that it is accountable to both Parliaments and that this Parliament will be able to probe the production of statistics and consider their veracity should help in the longer term to improve trust and confidence in statistics. Politicians in both Parliaments will be able to challenge and probe how those statistics have been arrived at. I would hope that the degree of independence that is offered by the new statistics board and the accountability to Parliament will go some way to reassuring people about the quality of the statistics that are being produced.
Is that the statisticians' perspective?
Yes. Although independent scrutiny already exists, the Statistics Commission has not done an awful lot in Scotland. We would hope that the new board will be more active. The line of reporting to the Scottish Parliament will be an important element of that. There has been a fair bit of discussion about whether the complete trust of the public can ever be achieved—perhaps it cannot. Independent and parliamentary scrutiny should provide an assurance to Parliament and the public that issues can be examined.
Minister, will you give me examples of how the arrangements provided for by the bill will improve the quality, relevance and timeliness of Scottish statistics?
It is important that the new body verifies and scrutinises statistics to ensure that statistics in Scotland are produced to international standards—that is the quality mark that has to be met to ensure people's confidence and trust. That will be a key role of the new body. The accountability lines back to the Scottish Parliament as well as to the UK Parliament are also a powerful tool in ensuring that we rebuild trust in statistics. Through an agreement with the Chancellor of the Exchequer, ministers have powers to determine which statistics will be national ones. If we have genuine concerns about the statistics that are being produced, we can do something about that through the powers in the bill. That is a substantial step forward.
As we look forward to this new era, do any statistics come to mind that you believe will have an impact on future policy or adjustments to spending plans? Which statistics will you and your colleagues in the Government use to monitor the effectiveness of governance in Scotland and future policy and spending plan changes?
One of the key arguments for retaining the production of statistics within the departments is that the production of statistics and the interaction between statisticians and economists who work in the departments—both in the Scottish Executive and, I guess, at UK level—are fundamental in the development of policy. The production of statistics determines how we develop policy over the longer term. Of course, policy then inevitably shapes the type of policies that we implement. As we all know, statistics are the currency of debate in Parliament about how well policies are proceeding or otherwise. Despite everyone arguing that they do not trust statistics, by and large the national statistics are quoted back time and again in any debate about how policies should be altered.
I was asking which of the plethora of statistics that are being produced you are using as the north star to guide Government. Which are the key statistics that influence future policy and spending plans?
For every department, we consider the statistics that are relevant to that department. Whether it is waiting times in health or economic factors in terms of economic plans, we use the statistics that are relevant to shape the policy.
Do you agree that the issue of pre-release access affects trust? If pre-release is to go ahead, with the exception perhaps of market-sensitive information, should it be available to more than just ministers? Should the timescale be reduced from five days to a few hours, as the Treasury Select Committee has suggested?
We are still to come to a firm view on that matter. Clearly, there is flexibility within the provision on pre-release access, which will allow us to determine what is appropriate for Scotland. We will come to a view on that in due course.
Do you agree that if there was more even-handedness, trust would improve?
I would agree that access to the information before release is important to ensure that there is a proper response from Government to the statistics that are published. I agree that that is an important issue.
And from Opposition spokespeople?
As I say, I believe that it is important that ministers have access to that information before it is published because they are expected to respond on the day. We will come forward with our proposals in due course.
One of the committee's concerns about the way in which statistics are used is that, when statistical evidence is gathered that demonstrates that the Government has failed to achieve a target for the delivery of a particular service—as happened with several targets under the 2002 spending review—the Government sometimes simply changes the target. How does the minister feel about the Government's use of that technique to avoid facing up to the dispassionate statistical truth that it has failed to achieve a target? Is it legitimate to dump a target and just replace it with another one?
As Mr Swinney will know, the programme for government includes a huge range of commitments and targets that have been set as a result of the partnership agreement between the Liberal Democrats and the Labour Party. I am pleased that, across the vast range of those commitments, the monitoring that is undertaken through the use of statistics and other methods shows that we are meeting virtually every target. I take the point that there have been concerns about one or two targets that have been missed. However, the key issue is that, using good statistics, we have been able to measure and monitor how well we have done across all the partnership agreement commitments and some 96 per cent or 97 per cent of those targets have been met or will be met by the time that we reach the end of the four-year period.
My point is that, in a number of cases when statistical evidence has demonstrated that the Government has failed to meet a particular target, the Government's reaction has been simply to change the target. What is the point of ensuring that people have greater confidence in the statistics if, when those statistics prove the unpalatable truth that the Government has failed to deliver on a particular commitment, the Government simply changes the target? What link should exist between statistical performance and the performance of Government, given that the Government must be accountable when it fails to deliver on particular targets? Will the minister comment, as the committee has done, on the legitimacy of just dumping a target that the Government fails to achieve? Is that the right thing for the Government to do?
That goes back to Mr Mather's question on whether there are areas of policy in which we would change direction if the information showed that the policy was not delivering the desired outcomes. It is valid to say that the Government is entitled to change policy when it believes that it is unable to meet the target that it set. That is a valid role for Government to play. Indeed, we would be criticised if we did not do that. However, in the vast range of targets that we have set in areas such as health, crime and education, we are meeting and, in many cases, going beyond the targets that we set. Across the piece, the Government has set some arguably tough targets for health, crime and education and is meeting or surpassing those targets.
In the section on information sharing, the legislative consent memorandum states:
The problems vary. In practice, the sharing of data involving personal-level information can be extremely difficult because the gateways for such information sharing are unclear. It is appropriate that there should be gateways that are properly understood under the legislation. We have found that it can also be difficult to share even aggregated information. As a statistical service we are not particularly interested in individual-level information, but information from departments such as HM Revenue and Customs or the Department for Work and Pensions can be very valuable in understanding Government policy and programmes. Sometimes it can be difficult to ensure that we have a means of getting that information shared in an appropriate way. There is no suggestion that information should not be shared in an appropriate way.
It strikes me that individual members often receive written answers advising that information is not held centrally. Committees of the Parliament have also had problems in scrutinising budgets especially when the operation of the policy is carried out by another public authority such as a council or health board. As a consequence of the bill, will it be easier for committees of the Parliament to monitor spend and outcomes and outputs in a way that they have not been able to do so far? Will the bill mean that information and monitoring will be easier for parliamentary committees?
We are not certain about the impact of the bill on detailed Scottish information, which is shared reasonably well at present. We are particularly interested in the impact of the bill on the information that we receive from UK Government departments. Such information may also assist members and committees in that, for issues such as the impact of poverty, information from UK departments is important.
So the Executive might be able to get more information from councils on issues that are of interest to the Parliament.
We do that substantially at present. What I am trying to say is that, if Parliament comes to a view that there are significant gaps, it will have more of a locus to express those views.
One snag of being the best small country in the world is that, in any collection of UK statistics, the sample from north of the border is quite small. Will the bill provide sufficient leeway to increase the sampling base in Scotland so that we can have more reliable data?
Clearly, if concerns about the quality of data lead us to believe that wider sampling is required, the provisions in the bill will give us the ability to ask for that to happen. Indeed, the Parliament will be able to raise such concerns when the board reports to the Parliament. However, many of the national statistics are collected by Scottish Executive departments, so we should have good coverage in the information that is gathered. Although it might be more difficult to get breakdowns of information that is collected by organisations such as the DWP, Scottish Executive departments collect a large amount of statistics. In future, those statistics will be validated to international standards by the new statistics board.
I understand that the consultation document that was issued yesterday is split into two parts, with a deadline of early January for responses on the more immediate issues and a deadline of March for responses on the more general issues. Will you elaborate on why you have gone down the route of saying that a legislative consent motion is the best way forward? It strikes me that the process is a wee bit back to front. Would it not have been more sensible to carry out a consultation on a range of options before taking the decision on whether to pursue a legislative consent motion?
There has been quite a lot of discussion on that matter. The driving force for the consultation was the decision on whether we wanted to go down the UK route. Having decided to do that, we now have the opportunity to consult properly on the detail of how that should be implemented. However, we are driven by the Westminster timetable. For that reason, once the decision was taken, it was important that we discussed the LCM with the committee in line with the Westminster timetable to ensure that we can hook into the legislative slot that has been allocated down there.
As you said, we had a debate in Parliament on statistics. I forget exactly when it was, but it was certainly some time before the Queen's speech. I presume that the Executive has been in contact with the Government to discuss the general issue of statistics. Was there some sort of interaction before the Queen's speech?
Yes. It has been an important issue and there has been a lot of discussion about the right approach, as was reflected in the debate that we had in Parliament. It was only right that time was taken to ensure that we examined all the possible options and engaged with Westminster to ensure that the flexibilities that we wanted if we decided to hook into the UK bill would be provided. That took some time, so it proved difficult to go out to consultation, given the Westminster timetable. However, we are now consulting on some of the more detailed implementation measures, which will be important in terms of the amount of pre-release input that ministers and the Parliament will have, as well as in terms of how the accountability procedures might work. Although I am disappointed that we were unable to go out to consultation before, it was right for us to take the time to ensure that we got the right decision and explored all the options properly.
It was not a surprise, presumably, that there was going to be a statistics bill in the Queen's speech, or that the Queen's speech was going to be in November. We are now in a situation where, for some of the fundamental measures in the bill, the consultation period will be less than a month, which includes the Christmas period. I am sure that everyone agrees that that—to put it mildly—is less than ideal. Surely it must have been possible to conduct some sort of formal consultation, either at the time of the Queen's speech, when it was clear that a bill was going to be introduced, or prior to that, when it must have been clear, after discussions with the Government at Westminster, that something was likely to happen in the near future.
It was, but there was quite a lot of discussion as to what the flexibilities would be and how the new statistics board would work, and we wanted assurances on those before coming to a final decision. Although I regret the short consultation period, I believe that it was important for us to flush out some of the Scottish ministers' concerns before final decisions were taken. As you will recall from the debate that we had in Parliament, there was a range of options for the road that we wanted to go down. There was lengthy interaction with Westminster, as well as internal discussion, before a final decision was reached. I regret the fact that the consultation is quite short, but I am sure that those who are interested in the subject will manage to respond and put across their point of view, and we will take those views into consideration.
Was the part of the consultation that is running to the standard timescale, until March, not consulted upon until it was clear what the shorter part of the consultation was going to be? It strikes me that the former is more general and could probably have been consulted on independently.
It was important to come to a decision about the direction that we were going in before the rest of it could unfold. As I said, there was a lot of discussion, internally and with Westminster, before final decisions were taken as to which route we would go down.
The bill provides for the appointment by the Treasury of one person to the statistics board after consultation with the Scottish ministers. Can you tell the committee who that person is likely to be?
No decisions have been taken on that.
Is it likely to be an official such as the chief statistician or is it likely to be some other figure?
I have not been involved in any decisions to date as to whom that individual might be. As soon as we have come to a view, we will be keen to let the Finance Committee know.
What will be the characteristics of the individual whom you will be looking to fulfil that role? Will it be somebody who has a role within Government or somebody outside Government? Can you furnish us with more information in that respect?
As I said, there has been no discussion as to whom the individual might be. Clearly, they will need a close knowledge of how statistics work in Scotland, and it is also important that they are seen to be independent.
The bill refers to
There will be discussions with ministers here in Scotland, and we will come to a view and put forward the person whom we would like to see put on the board to represent Scotland. That is the way that these things work, and I am sure that the Chancellor of the Exchequer will be keen to accept our nomination.
The proposed statistics board will take on the responsibility of the Office for National Statistics, which is a Treasury department, and of the Statistics Commission, which is a non-statutory external commentator on statistics. Can you explain how, in that context, the status of the statistics board as a non-ministerial department that is still part of the Executive will guarantee its independence when it comes to statistical matters?
It is being set up as a non-ministerial body, so it will be free from ministerial interference. Its accountability will be to the UK Parliament and to the Scottish Parliament, but the responsibility for overseeing the independence of the organisation and the quality of the statistics that are produced will rest with the statistics board itself, which will be independent of Government and of ministerial interference. It is right that it should be a standalone body.
Are there any concerns that, although the Statistics Commission is an external body, the statistics board will be a much more internal body, and therefore we could lose an external scrutineer of statistics by replacing it with an internal one?
The Statistics Commission is an independent body but, nevertheless, it is appointed by ministers and its budget is within the Treasury budget. The expectation is that the new board will not only enjoy the same degree of independence but will be a more substantial body than the Statistics Commission. There will be members on the board who have greater weight, so the board will be a more substantial body in that sense.
A press release from the Statistics Commission states:
The board will be funded by the UK Government, so there is no call on the Scottish Executive's finances to meet the cost of it. I understand your concern about long-term commitments to finances, and I understand from discussions with colleagues down south that there will be a commitment to ensuring that on-going budgets will be sufficient to enable the statistics board to do its job properly. There will be guarantees about the longer-term funding, which is important, because we all know that politicians like to intrude on matters when they have concerns about the amount of money that independent bodies are spending. It is often thought that some controls should be put in place, so it is important for the independence of the organisation that there is some guarantee about on-going funding levels, because that could clearly affect performance.
Your guarantee on funding levels is welcome and, indeed, important, as far as independence is concerned. However, my question was specifically about guaranteeing funding levels for operations that relate to the Scottish Executive, not to the Scotland Office.
We would engage to ensure and be confident that the statistics board was being allocated sufficient money. However, as I have already said, Scottish departments themselves tend to produce a substantial amount of statistics, which will be validated by the statistics board.
My understanding is that statistics that are produced by Scottish Executive departments fall within the realm of the Scottish information commissioner and the Scottish public services ombudsman. However, under clauses 37 and 60 of the bill, the statistics board will be under the jurisdiction of the UK information commissioner and the UK parliamentary commissioner. There appears to be no mention of the Scottish information commissioner and the public services ombudsman. Has the Executive consulted on whether the statistics board should fall within their jurisdiction?
I would have expected Scottish statistics to fall within the realm of the Scottish information commissioner and the public services ombudsman. However, if, as you suggest, that is the implication in the bill, and if the concern is raised in the consultation, we will examine it closely.
We have discussed the exact wording of those clauses, but we will need to come back to you on the specifics. The minister is right to suggest that the intention was certainly not to affect the Scottish information commissioner's jurisdiction over Scottish Administration statistics. However, if that is not the case, we will have to reconsider the matter.
Our concern was more about the fact that although the bill specifically mentions the UK commissioners, no mention is made of the precise remit of the Scottish information commissioner and the Scottish public services ombudsman. Any clarification of that would be welcome.
We will see to that.
That would be helpful. As you know, our consultation closes on 8 January and we will consider a draft report on 16 January. It would help if you could write to us before the consultation closes.
We will get something to you quickly.
Do members have any final wrap-up points?
I have a question for the minister and another for the chief statistician. Given the desire to improve public trust, was any consideration given to allowing the chief statistician to be independent of ministers?
Do you mean here in Scotland?
Yes.
A number of models, including a proposal for a separate Scottish body, were discussed and fully considered. We eventually agreed to go down the route that has been outlined to the committee this morning. I assure members that we had wide-ranging discussions on the various alternatives and on what was best for Scotland, and we think that we have made the right choice.
What percentage of Government departments are under what you would classify as comprehensive statistical control, in which performance is predictable and we have the data to make comprehensive comparisons with elsewhere?
I am sorry; I am not quite sure what you mean. Are you talking about the Scottish Administration?
What percentage of departments that are involved in Government activity and in producing statistics to allow us to know where they are are under statistical control to the extent that we can see a predictable pattern of performance and we have the required data to compare performance here with performance elsewhere?
I would not like to make up a figure on the spot. The degree of control varies substantially. We plan such matters annually, so there are always areas where improvements are needed or where new information is required.
Is it likely that the process under discussion will indicate gaps or areas that need to be improved?
You should take some comfort from the fact that about 90 per cent of the statistics that are produced in Scotland are national statistics, and are therefore validated according to international standards. As I said earlier, the key issue is the quality of statistics.
If we are validating the statistics according to international standards, will we also see some international comparisons, to give us a real feel for performance here?
In developing policy, departments regularly make comparisons with other countries to find out whether other approaches might work and how well we are doing. For example, in education, we are in the top three or four European countries with regard to how well our schools are doing.
My final question is for the chief statistician and relates to the key macroeconomic statistics that we use to monitor performance in Scotland. In that respect, I find it easier to follow the example of the corporate world, where, in order to enable it to understand how things are progressing, a board of directors considers aspects such as turnover, net profit, market share and, perhaps, profit by employee. What key macroeconomic statistics should we use to monitor progress in Scotland?
Our economic statistics plan sets out the various priorities and the range of economic statistics that are needed each year.
Is there a subset of, say, four statistics that could be used as a finger on the pulse of the country?
One valid view is that we need to identify a set of key statistics. However, that has not been done because, after all, different people have different views on what constitutes a key statistic. We would have to consider the range of statistics across all areas.
Is it likely that such work will be a by-product of this process?
As the new board will have the power not only to scrutinise in detail national statistics but to comment on the comprehensiveness of the Scottish Administration's statistics, I expect that it will pick up on that issue, particularly if the Parliament raises major concerns about gaps and other issues.
Without wanting to stray into the substance of the debate that the committee will have at the beginning of January, I should point out that the "Government Expenditure and Revenue in Scotland" report that was issued yesterday provides some of the key statistics that might be required to discuss Scotland's future direction.
Does the chief statistician have the professional independence to identify, make a judgment on and recommend to ministers a group of key statistics that could be used to monitor economic and social health in Scotland, or is that matter ultimately decided on by ministers?
Based on our understanding of important issues and on consultation with users and potential providers of information, we indicate to ministers what we consider to be an appropriate range of statistics. Such indications take account of issues such as the partnership agreement, which the minister mentioned earlier and which obviously contains a range of matters on which the Executive will need information. As I told Mr Mather, there is no single set of key statistics for Scotland, although that could be addressed.
A huge range of statistics is produced for Scotland.
Nobody is disputing that. We are trying to find out whether, in the eyes of the chief statistician, a range of limited key performance indicators would enable us to judge the economic and social health of Scotland, which has been the focus of today's questioning. Is it possible to arrive at such indicators as a result of a statistician's professional judgment?
It seems to me that the role of the statistics board is to come up with a wide range of comprehensive indicators. However, the selection of key performance indicators must remain a matter of political choice. Under the most recent Conservative Government, obscure measures—
With the greatest respect—
Let me ask the question.
I asked Mr Wishart whether it is in the professional ambit of a statistician to identify a range of key performance indicators against which we can judge the economic and social health of the nation. I would be obliged if I could get an answer from him before I get a political commentary about the value of my question.
I was simply going to ask the chief statistician to comment on the fact that obscure measures of the monetary supply were incredibly important as KPIs 20 years ago, whereas, more recently, employment has been judged as more significant than it was for the whole of the 1980s. In that context, I am interested to hear whether it would be appropriate for the statistics board to select the KPIs or whether that is a matter of political judgment.
Statisticians throughout the world have made various attempts to identify a set of key indicators, but consensus has not been arrived at. We advise ministers and discuss the range of statistics with users and providers. To take an extreme example, if a Government decided that it did not wish to produce unemployment figures, I would certainly comment accordingly. Beyond a number of key indicators, it becomes much more a matter of political discussion which measures are of real importance. What Parliament wishes to debate—as well as Government's needs—determines what statistics need to be considered. We indicate what we consider to be appropriate and comment specifically on how to measure things, which is a matter of professional judgment.
This is a matter of political debate. However, it is important that a range of statistics is produced right across all the policy areas that inform debate. There will always be a political debate about which indicators are the key performance indicators that indicate whether policies are failing. It is important that quality statistical information is produced across a gamut of areas to allow parliamentarians and the wider public to engage in a political debate about how well—or not so well—policies are being implemented or how well the country is doing.
My final question is on the environment of accountability under which the chief statistician operates. Is the chief statistician free to comment publicly on concerns that he might have about the Government's use of statistics, or does he operate under the same constraints as any other civil servant who is required, in effect, to represent the Government position?
I invite Derek Brownlee to ask his question, before we wrap up the evidence session. The witnesses can then answer both questions.
My question is simple. Mr Wishart mentioned a concordat between Westminster and here. I have heard references to a memorandum of understanding. What is the distinction between the concordat and the memorandum of understanding, and why was the concordat route chosen?
Will I answer Derek Brownlee's question first?
Yes, whatever.
The concordat is an attachment to the memorandum of understanding. The memorandum refers to a small number of specific concordats, including the one on statistics.
Is it common practice for both parts to be made available, so that we can see the ground rules?
They are publicly available.
I will answer John Swinney's question. Currently, the chief statistician does not comment publicly on the use of statistics but, given the provisions on accountability to the Parliament, a view will have to be taken on that.
I thank the witnesses. The evidence deadline for this committee is 8 January. We expect to consider a draft report on 16 January, which we will have an opportunity to revise thereafter.