I apologise for this debate being something of a repeat. Members will think of it, perhaps, as the
BBC Four of parliamentary debates. We were here last November, discussing the twin topics of data and economic performance, and it was clear at that point that the Scottish Government had not altered its position.
With respect to the Minister for Public Finance and Digital Economy, I note that her predecessors, Keith Brown and Derek Mackay, were nothing if not consistent. Whatever it was, they were against it—or, at least, such was their stance regarding changes to pre-release access. I hope that the minister will be more willing to engage with the committee and will consider why we have been pursuing the issue for almost two years now, although there have been times when I have wondered that, too. As the playwright said,
“Try again. Fail again. Fail better.”
Why, then, do we propose a bill addressing the topic? I will outline what we propose to do and why. First, I will address the “why”. Pre-release access, or PRA, is rather a niche topic and we did not envisage that it would occupy the time that it has occupied. What is PRA and why does it matter? It is the practice of making statistics available to ministers and their advisers prior to publication. The Office for National Statistics stopped doing that in July 2017, and the Bank of England followed suit.
In our economic data report, which was published in February 2018, we called for an end to PRA for four sets of statistics that are of national importance, including those on gross domestic product and “Government Expenditure and Revenue Scotland”. That was the majority view of the committee. The minority opinion, although more cautious, still sought a presumption against pre-release. I add that the recommendation was just one of the 29 that we set out in our report, the majority of which were accepted by the Scottish Government and others, including the Scottish Fiscal Commission and the ONS.
Why was our call for an end to PRA not agreed to? The chief statistician suggested that the issue had been overplayed. By some curious logic, it was right for the ONS to end PRA but, in our case, it was “not necessarily straightforward”. We pressed the Scottish Government on the matter in further correspondence and meetings, and the lengthy discourse can possibly be distilled into five words: ministerial benefit versus statistical integrity. I will elaborate on that a little, after which I will come to the matter of there being two views among committee members.
The standard argument for PRA is that it is preferable for ministers to be briefed in advance. Those who are in favour say that that allows ministers to make sensible and informed comments at the time of publication, so the practice has public merit. So far, so plausible. The counter view is that it puts ministers—whichever party is in power—in a privileged position and allows the figures to be framed in a particular way, or even to be spun.
The point is that the Scottish ministers get access before other politicians or commentators do.
It is said that PRA enables ministers to provide a positive slant before anyone else can respond, thereby risking the public’s trust in the veracity of official statistics. The majority view of the committee sides with that side of the argument, on statistical integrity. Some colleagues take a different view, and I shall let them speak for themselves, but I think that it is only fair to say that they, too, would like to see a shift in the approach to PRA.
Such is the context of our bill proposal. I will now outline what the proposed bill would do. There are three strands to that: first, the removal of PRA for two specific categories of economic data; secondly, a phased approach to that removal and review of its impact; and, thirdly, the reduction to one working day of the PRA for those statistics for which five days is currently the norm. I will share the thinking behind each of those strands.
First, the bill would end PRA for two of the four categories of economic data that we identified in our original inquiry—the GDP and retail sales data. Neither of those categories of data is subject to PRA at a United Kingdom level. The Scottish Government would not lose anything that is retained by the UK Government.
The second strand of the bill would stipulate that the removal of PRA be phased: one day would be cut to 12 hours after one year and then removed entirely after two years, and there would be an independent review of the impact after three years. That review would be laid before the Parliament, and ministers would be obliged to respond to its findings.
The third strand of the bill would reduce from the current five days to one day the PRA for other economic statistics. As the Royal Statistical Society remarked of the five-day scenario during our inquiry,
“Scotland is very much an anomaly relative to almost the whole developed world.”—[
Official Report, Economy, Jobs and Fair Work Committee
, 26 September 2017; c 9.]
In fairness to the Scottish Government, it wrote to us, in May, with a compromise suggestion. Mr Mackay said that ministers would seek a period of one day where a period of five days now applies. That sounded promising. However, when pressed, the cabinet secretary told us that he favoured a “pragmatic approach” and did not wish to make what he regarded as “unnecessary amendments to legislation”. We, on the other hand, would rather that a five-days-to-one reform be given legislative underpinning.
At this point, it might be helpful to say what our bill would not do. It would not put the Scottish ministers at a disadvantage compared with their UK counterparts or Whitehall departments, because the statistics that we focus on—the GDP and retail sales data—are in the gift of the ONS and, hence, are not subject to PRA. The bill would not legislate on any data other than the categories of economic statistics that are specified—for example, it would not cover health or education stats.
The bill would not question—nor, indeed, do we question—the integrity of Scottish statisticians. As a representative of the UK Statistics Authority—the guardian of the independence of official statistics—told us,
“They are genuinely highly professional statisticians who do an excellent job. I just think that pre-release access makes their work harder.”—[
Official Report, Economy, Jobs and Fair Work Committee
, 7 November 2017; c 23.]
How is PRA viewed by the wider statistical community? The Royal Statistical Society supports ending it, the ONS supports ending it and the UK Statistics Authority supports ending it. Professor Sir Charles Bean, who led the 2016 independent review of economic data, supports ending it. John Pullinger, the recently retired UK national statistician, supports ending it. The list goes on. His successor, the former principal of the University of Aberdeen, Professor Sir Ian Diamond, supports ending it. In fact, in May 2017, he co-signed a letter to
The Times that described PRA as “outdated and unnecessary” and “detrimental to public trust”. The letter argued that its abolition
“would cost nothing but have the very welcome effects of reducing the opportunities for media spinning, improving the health of our political system and safeguarding public confidence in official statistics.”
There were 114 signatories to that letter, among them senior academics and statisticians as well as the directors of think tanks as diverse as the Institute for Public Policy Research and the Adam Smith Institute.
Nine years earlier, in another letter, the UK Statistics Authority had argued for
“a progressive reduction in the length of time for which privileged access is granted”.
“We would encourage the Scottish Government to adopt statistical policies that promote equal access, the earliest possible publication, and minimise the opportunity to make policy proposals or comments from the advance sight of the unpublished statistics.”
More than a decade on, the direction of travel has moved even further away from PRA, and the Scottish Government finds itself on the wrong side of the argument. It can, of course, always change that.
“People who have what they want are fond of telling people who haven’t what they want that they really don’t want it.”
I can repeat that. [
.] That might be helpful for members who are not fans of Ogden Nash or familiar with his works. He said:
“People who have what they want are fond of telling people who haven’t what they want that they really don’t want it.”
Members can call me an optimist if they want, because, after 19 months of trying and failing, I remain hopeful that we can make some progress today. Our premise is simple: we believe that statistics are a public asset and that they should be an aid to understanding the political and macroeconomic decisions that affect us all. As such, the numbers should be available on an equal and not a privileged basis, which is the purpose of our proposed bill.
That the Parliament agrees under Rule 9.15 to the proposal for a Committee Bill contained in the 7th report (2019) of the Economy, Energy and Fair Work Committee, Pre-release Access - Committee Bill proposal report (SP Paper 553).
As much as I love talking about data and I am looking forward to this afternoon’s debate, it is remarkable that this is, as the convener said, the second debate on this issue. I would have thought that there are far more pressing issues facing the economy right now to consider. There is also the matter of the time and effort that the legislation would require, should the committee introduce it. However, that, of course, is the committee’s choice.
A s the convener has mentioned, the cabinet secretary made an alternative proposal. That pragmatic approach, which the committee referenced, would retain public confidence in economic statistics and provide Scottish and United Kingdom ministers with limited and tightly controlled PRA to statistics.
I will lay out the wider context now, and when I sum up, I will address members’ specific points.
I want to make it clear that there is nothing inappropriate about pre-release access to official statistics. It is in line with legislation. People receive early access only if the person responsible for producing the statistics considers it necessary and legitimate, it is not contrary to the “Code of Practice for Statistics” and it is standard practice that is not limited to ministers. For example, the Scotland Office, local authorities and others receive pre-release access where appropriate.
I have looked at the evidence, which is valuable and valid. I repeat the point that, since the ONS ended PRA to its statistics more than a year ago, only the Bank of England has followed suit. Whitehall departments still operate with 24-hour PRA as per the Pre-release Access to Official Statistics Order 2008.
It is important that members see pre-release access in the context of the whole Scottish official statistics system and the unique role that statistics play in ensuring that we make the most of Scotland’s rich data sources and tap into the valuable information that is contained in our official statistics.
Of much more importance is having the statisticians with the right professional skills, the right processes and—critically—the independence to produce those statistics. If the committee legislates, it will end the independence that the chief statistician currently enjoys. I understand that the UK Government has not legislated on PRA: a decision was taken by the ONS, which is an arm’s-length organisation, to change the way in which it operates.
On ethical and trustworthy government, I am a strong believer in the ethical use and handling of data. This Government has demonstrated such an approach and continues to put that principle at the forefront of our work. In the recent programme for government, we set out clearly our commitment to the ethical use of data.
It is important that we reassure people—the public as well as members of the Scottish Parliament—that we take our responsibilities with data very seriously and that our actions need to be principled and ethical if we are to make the most of data, for the benefit of all. We are clear about our responsibilities on data and statistics.
Our long-standing position is that decision making on and responsibility for statistical matters, including pre-release access, is fundamentally for the professionals—in this case, Scotland’s chief statistician. We are not legally obliged to take that position, but we have made that call. The advice of the professionals who produce official statistics, based on their professional experience, is to have tightly controlled pre-release access, as per the legislation.
In other words, having PRA is an important part of the official statistics system. We recognise that that comes with responsibility.
A myth that I want to dispel, which the convener pursued in his speech, is the idea that pre-release access to GDP statistics, for example, gives the Scottish ministers a first-mover advantage. That is simply not the case, and to believe that is entirely to miss the point that others—the Scotland Office is one example—also get early access to GDP statistics.
We think that people who are in authority or who have responsibility for a policy area that is of national importance should be able to talk about new information with understanding, depth and accuracy when they are asked to do so, as they always will be when statistics are published.
I recognise that that would be the effect of what is proposed. I am trying to dispel the myth that somehow it is only the Scottish ministers who have access and can use—or misuse
—statistics. The Scottish ministers cannot do that, because there is access for others. Jackie Baillie is right to say that if the proposed legislation goes ahead, there will be implications and consequences not just for the Scottish ministers but for the Scotland Office and the Treasury.
Nothing makes us more powerful. However, we are asked to comment, in many cases, the minute that something is published, and I would far rather that comment were informed by the facts, as opposed to being a knee-jerk response to high-level figures.
When I talk about others having access, I am not in any way suggesting that we want to preserve our access but deny access to the Scotland Office, for example. I am making the point, first, that there is an element of accountability in that regard, and secondly, that no one is denying access to statistics for those who legislate or the general public. All will have access.
My point is that PRA enables us to make a far more informed comment when something is published than just a knee-jerk response. Everybody has access to the statistics as soon as they are published.
I am going over my time, so I will draw to a close. Given the whole group of issues that the Scottish economy faces, whether it is a disorganised European Union exit or anything else, the Parliament needs to focus on and devote its energy to doing all that it can to support and protect the people, industries and reputation of Scotland. I look forward to working with the Economy, Energy and Fair Work Committee to do that.
I thank the Economy, Energy and Fair Work Committee clerking team, the Scottish Parliament information centre and the witnesses who contributed to the committee’s valuable inquiry.
I will address the specific issue of pre-release access shortly, but I first want to mention the wider backdrop to the debate, which is the increasing number of concerns about the level of transparency and governance under the Scottish National Party Government. Just this weekend, there were reports that the Scottish Government has refused to answer freedom of information requests to reveal details of how £130 million of taxpayers’ money has been invested under the Scottish growth scheme. That comes on top of the recent widespread concerns over the level and accuracy of information that the Scottish Government has provided in relation to the crisis at Ferguson Marine Engineering.
That should come as no surprise from a Government that, last year, was criticised by its own Scottish Information Commissioner for secretive and biased responses to freedom of information requests from the media and Opposition parties.
Does the member accept that the UK Government does not exactly command total confidence on the issue and that, on the whole, there is an issue for politicians in general and not just a particular group?
Part of the debate, which I will come on to, is about aligning the Scottish Government’s approach to pre-release access with the best practice that is followed by the UK Government. That is the whole point of having the debate.
The SNP Government says that it has
“the most open, far-reaching freedom of information laws in the UK”,
but, as we know, the reality is very different. The concerns about open governance and transparency are reflected in the debate. The Scottish Government continues to insist on having pre-release access to vital economic statistics, which is inconsistent with international best practice, transparent government and democratic accountability and fairness. As Alex Rowley rightly said, the lack of access to information at the same time affects the Parliament’s ability to hold the Government to account.
Pre-release access gives the Scottish Government 24 hours or longer to spin a story around key economic figures, no matter how bad they are. That means that, when the information is released to the public, headlines are already dominated by the Scottish Government spin. For example, yesterday’s news that Scotland’s economy is close to recession was dominated by SNP spin that Brexit was to blame. That is an odd excuse, given that the rest of the UK economy, which is subject to the same Brexit uncertainty, is growing at almost twice the rate of the Scottish economy.
The Scottish Conservatives will support the Economy, Energy and Fair Work Committee’s proposal for a committee bill to equalise access to vital economic statistics. The committee is acting because the Scottish Government has refused to do so. The Government has refused to listen to overwhelming evidence that pre-release access is contrary to the “European Statistics Code of Practice” and the United Nations resolution on the fundamental principles of official statistics.
The member often talks about ways in which the Scottish Government should do more to boost the economy. Does he think that pre-release access is the most pressing issue facing the economy right now?
Parliament’s ability to have equal access to information in order to hold the Scottish Government to account is an important component of what needs to be looked at.
The committee heard compelling evidence from a range of witnesses. I will not repeat everything that the convener set out, but a witness from the UK Statistics Authority made it clear that the issue is important. He said:
“why do we care so much about this? It is because, at the heart of what statistics are about, they are a public asset.”—[
Economy, Jobs and Fair Work Committee
, 7 November 2017; c 23.]
He said that pre-release runs against that principle.
“is very much an anomaly relative to almost the whole developed world.”—[
Economy, Jobs and Fair Work Committee
, 26 September 2017; c 9.]
In written evidence, the committee was told:
“We believe that ... privileged access undermines public trust in official statistics”,
“creates opportunities for figures to be ‘spun’ to the media or ‘buried’ beneath other announcements.”
Despite that evidence, the cabinet secretary refuses to recognise that the Scottish Government is out of line with international best practice. In response to his refusal to listen to that evidence, the Economy, Energy and Fair Work Committee has set forth proposals that are straightforward and represent a compromise.
There are three strands to the proposed bill. First, it proposes the entire removal of pre-release access for two categories of economic data—GDP and the retail sales index—as neither of them are covered by pre-release access at the United Kingdom level. Secondly, it proposes a phased approach to the removal of PRA and an independent review of its impact. Our idea, which we discussed at committee, is that the gradual approach and review will offer accountability in relation to the changes that we propose. Thirdly, we seek reduce to one working day the pre-release access to the economic statistics to which the Scottish Government currently has five working days’ access. Those include statistics on exports, productivity and non-domestic rates. The committee proposals will go some way to bring Scotland in line with the rest of the United Kingdom and the rest of the world.
The minister mentioned that, when giving evidence to the committee, the Scottish Government argued that the ONS approach is an outlier, in that no other UK Government department has ended pre-release access. That misses the point, which is that the ONS is the gatekeeper to key economic statistics, and it gave up the right to grant pre-release access. The scope of the bill is limited to vital economic statistics, and does not make any stipulations on sets of data other than the economic figures that I have mentioned.
The experts are clear that the Scottish Government’s approach to pre-release access is out of step with best practice and the policies that are followed in other Organisation for Economic Co-operation and Development countries. In the interests of transparent government and democratic fairness, it is time to put an end to pre-release access. I support the committee bill to that end.
I thank the Economy, Energy and Fair Work Committee for its investigations into pre-release access. The Scottish Labour Party will support the committee’s bid to introduce a committee bill to address the anomalies regarding pre-release access to statistics.
It is clear that pre-release access puts the Government in a position to spin statistics to their best advantage, which can cause confusion and undermine public trust in the system. Often, on the day that statistics are released, we get a commentary from the Government that puts a gloss on them that is not reflected in the statistics themselves. The following day, after Opposition parties, the press and the rest of Scotland have had a look at the statistics, the garden is not quite so rosy.
Pre-release access happens with regard to statistics on both devolved and reserved issues. PRA on reserved issue statistics is not affected by the bill that the committee proposes, but it is worth noting that they are pre-released a maximum of 24 hours in advance, while some statistics on devolved issues are pre-released five or more days in advance.
For those who receive pre-released statistics, the benefits are clear. They know what is in the report in advance, allowing them time to either bury bad news or spin the findings. It allows them to accentuate good news and mitigate the bad. However, parliamentarians and the press are left behind trying to assess the data while the Government is already setting the scene.
If we follow press coverage of some data releases, we can see that, on the day that data is released, the media highlights what is good about the statistics—what the Government wants to be highlighted—and, the day after, the coverage is much less positive. The minister said that the Government is often asked to comment on statistics very quickly after they are released but, at the moment, the Government has its comment already prepared. That leads to mixed messages and public confusion, and undermines public confidence. How can the public believe what they are told when the same statistics give very different stories on different days? Ending PRA would mean that a more realistic analysis of the statistics was in the public domain. People would clearly see the pros and cons together and be much better informed as a result.
In 2017, the committee expressed a view that the Scottish Government’s pre-release access to economic statistics should end, but the Government made it clear that it would not act on that recommendation. That decision instigated a second committee report, which proposes a committee bill to end PRA.
The bill would not stop all PRA. A modest change is proposed. As stated by others, there would be three strands to the bill. First, PRA would be entirely removed for only two specific categories of statistics. Secondly, that removal would be phased in and be independently reviewed. Thirdly, for economic statistics, five working days’ PRA is currently the maximum; the proposed bill would reduce that to one working day. That would pull Scottish PRA into line with the rest of the UK. The Scottish Government does not dispute that part, but it disputes the fact that we need to legislate for it. I do not understand why that change cannot be enshrined in legislation.
It does not interfere with the independence of the chief statistician. It interferes with the Government’s head start on spin when statistics are released.
The committee does not recommend that we end pre-release access for situations in which the UK Government continues to have it.
That action would bring the Scottish Government in line with EU best practice, which does not agree with PRA. We note that the Bank of England and the Office for National Statistics have ended PRA.
The Royal Statistical Society’s written submission to the committee stated that it believes that
“such privileged access undermines public trust in official statistics as, for example, it creates opportunities for figures to be ‘spun’ to the media or ‘buried’ beneath other announcements.”
It is clear that, by providing some Government officials with an unfair political advantage ahead of statistics being released to the public, PRA creates an uneven playing field. The committee underlines that it is important to build public trust in statistics and to make sure that that trust is retained.
In its report “Pre-Release Access to Official Statistics: A review of the statutory arrangements”, the UK Statistics Authority stated:
“We believe it would be in the public interest if all UK administrations amended their secondary legislation to adopt a maximum period of pre-release access of 3 hours, with a shorter period as the norm. A three hour limit was also recommended by the House of Commons Treasury Committee in 2006.”
That goes back some time. It is clear that the authorities on statistics and their release are in favour of ending pre-release. We support the committee’s position that that should be legislated for, albeit in a modest bill, by ending pre-release in some areas and possibly eventually ending it entirely.
I regret that we are having the debate. The committee’s economic data inquiry focused on a wide range of matters. The only contentious recommendation was around pre-release access to statistics.
It is frustrating, because it is not a new issue. In August 2008, the then Minister for Enterprise, Energy and Tourism, Jim Mather, was in correspondence with the chair of the UK Statistics Authority over the draft Pre-Release Access to Statistics (Scotland) Order 2008. The chair of the UK Statistics Authority was critical of the draft and the policy that lies behind it. I quote from Sir Michael Scholar’s letter of 2008:
“The Statistics Authority would wish to see a commitment both to a progressive reduction in the length of time for which privileged access is granted, as well as in the number of officials and Ministers seeing statistics prior to their publication. We would encourage the Scottish Government to adopt statistical policies that promote equal access, the earliest possible publication, and minimise the opportunity to make policy proposals and comments from advance sight of the unpublished statistics.”
The same statistics authority carried out a review of pre-release in 2010 and it clearly stated that it would be in the public interest, as Rhoda Grant said, if all UK Administrations amended their secondary legislation to adopt a maximum period of pre-release of three hours, with a shorter period as the norm. As we have heard, since then, the ONS has ended all 24-hour pre-release access, as has the Bank of England.
The committee’s inquiry revealed that the outdated 2008 order is still the governing statute for pre-release. Our convener cited some of the expert witnesses who said that pre-release must end, hence our recommendation. Those witnesses included the Royal Statistical Society, the director general of the UKSA and Professor Sir Charles Bean.
However, successive cabinet secretaries have stuck their heads in the sand. First, in response to the committee’s report, Keith Brown, the then cabinet secretary, dodged the recommendation completely in his response, passing the buck to the chief statistician to respond on the question of pre-release. That was spectacularly inappropriate, given that the chief statistician is the person to whom power is given by the 2008 order to authorise pre-release access.
Keith Brown’s successor, Derek Mackay, maintained his distance in a letter dated 10 July 2018, claiming that the question remained under the purview of the chief statistician and that he did not feel that it was appropriate for him to add anything further. In October 2018, Mr Mackay repeated the assertion that
“Fundamentally this is an issue for the Chief Statistician.”
That is misconceived. The issue before us, then and now, is whether it is right or appropriate as a matter of law that the chief statistician be given such powers to authorise pre-release of up to five working days. Our argument with the Government is about what the law should say.
In response to the committee’s direct proposals, Derek Mackay then took an entirely contradictory position in his 20 May 2019 letter to the convener. He maintained that it
“is a matter for the Chief Statistician” but in the next breath, he went on to suggest that he would tell the chief statistician that ministers would
“require PRA of only one working day for those economic statistics”.
If the independence of the chief statistician is so important—the minister for public finance has referred to it on at least two occasions this afternoon—that it prevented ministers from even responding to the committee’s pre-release proposals, what on earth makes the finance secretary suddenly feel able to tell the chief statistician that he requires certain actions to be taken? That is precisely why we need an updated order.
This proposal is not about this Government; it is not about the last Government; it is not about Keith Brown or Derek Mackay or the chief statistician; it is about the statutory framework for pre-release—specifically, should there be pre-release at all? If so, how much and for what purpose? Who should authorise such pre-release and in what circumstances?
Those are precisely the matters that are set out in the 2008 order and they are precisely the matters that the committee has been concerned with. They are precisely the matters that all those giving evidence to the committee have suggested should be dealt with by ending pre-release. They are the matters that this debate is about. They are the matters that ministers should pay far closer attention to and they are the matters that members who are here this afternoon should pay careful attention to.
A government spokesman is quoted in
The Herald newspaper today as follows:
“Pre-release access is consistent with the Code of Practice for Statistics which states that it should be in line with the rules and principles set out in legislation.”
That is a statement of nothing. The Government is basically saying, “We need to abide by the law.” Of course it does. The spokesman is quoted further as saying:
“Indeed, UK Government Departments provide pre-release access to their statistics in a similar way to the Scottish Government.”
That is nonsense—that is rubbish. The UK pre-release order grants PRA for a maximum of 24 hours, not five days. The spokesman continues:
“Pre-release access is a matter for the Chief Statistician and the independence of his role is crucial. Any proposal to curtail access would cut across his ability to ensure the key figures about Scotland are properly communicated and understood.”
That is another meaningless and completely erroneous statement because the minister is telling the chief statistician what to do in response to our report in order to avoid the need to introduce other secondary legislation.
The committee’s proposal does not even go as far as the recommendations of experts. It does not even go as far as the committee’s own recommendations in its economic data report. This is such a modest proposal that no member—far less a Government minister—should have any difficulty supporting it if they believe in good governance and transparency. The way to ensure that is for the minister to introduce a statutory instrument that would amend, in a modest way, the 2008 order. The instrument would fly through the committee. I cannot speak for other committee members but I am pretty sure that if the instrument satisfied the committee’s recommendations, the committee would have no problem with it.
Ministers are fond of coming to the chamber to tell us that this or that policy is world leading, but this one certainly is not. The most sensible thing for the minister to do is to concede that pre-release access is not best practice, follow the advice and recommendations of the country’s leading statistics experts and commit to ending pre-release access.
The committee report of 6 June highlights the three strands that are being proposed for the committee bill—a bill that I am yet to be convinced is necessary or required.
Earlier, we heard from Gordon Lindhurst what the three strands of the bill are, and I will highlight my reasons for why the bill proposal should probably go no further.
I agree that there is no need for this legislation. Does the member agree that the way to resolve this impasse is for the Government to introduce secondary legislation that deals with the committee’s concerns?
I am glad that Andy Wightman agrees that the bill should not go any further and is not necessary. At the last meeting that he appeared at to discuss this matter, the cabinet secretary highlighted what he thought was a way forward that we could all agree with.
Strand 1 is the removal of pre-release access for two specific categories of economic statistics. Much has been made of the decision of the ONS in 2017, and the report states:
“ONS ended all 24-hour PRA for its official statistics.”
However, is that really the case? The Bank of England immediately applied for pre-release access for specific ONS economic statistics, and that was granted. That PRA access has been renewed every year since, including to the end of this year. The letter from the Bank of England to the ONS dated 10 June 2019 asks that the pre-release access period be increased in order that statistics scheduled for release on 18 June be pre-released to it on 14 June—four days early. The letter then requests further brought-forward pre-release dates in September and December 2019.
The committee report points out that
“the ONS approach is unusual in that no other UK Government Departments have followed its example.”
Indeed, a review paper that was produced by the UK Government Cabinet Office at the time of the last review highlighted that
“There was almost universal rejection of the idea of removing pre-release access altogether, and of reducing it to a maximum of three hours. Over 90% of ministerial private offices, over 80% of press offices, over 90% of senior officials and about three quarters of officials who produce briefs had strong objections to the idea of eliminating pre-release access altogether.”
As it pointed out, pre-release access is important for good government, for avoiding misreporting in the media and for helping to spot mistakes. That view is backed up by the fact that, since the ONS removed PRA in 2017, not one of the 30 Government departments or agencies operating across the UK have followed its example—not one of them has removed PRA from its own publications, and they continue to issue in excess of 1,000 statistical releases each year. Further, the UK Government has passed no new legislation in this area since 2008.
I will come back to strand 2 in a moment, but strand 3 suggests that PRA should be reduced from five days to 24 hours for those economic statistics where five working days is currently the maximum. In a letter to the committee dated 20 May, the Cabinet Secretary for Finance, Economy and Fair Work stated, in a spirit of compromise, that he would accept a reduction from five working days to one day
“for those economic statistics where five working days is currently the maximum.”
In a further letter to the committee, on 24 May, Derek Mackay pointed out that the committee would be aware that the chief statistician had
“already taken the decision to restrict pre-release access to a maximum of 24 hours for key economic statistics.”
Strand 2 suggests that there should be a review to assess the impact of the reduction. In his letter to the committee of 24 May, Derek Mackay said:
“The Chief Statistician will review the impact of my proposed changes once they have bedded in. If he is content with how things are operating then he can ensure that changes are permanent.”
One area that the committee report fails to tackle is the independence of the chief statistician. Pre-release access in Scotland comes under the Pre-release Access to Official Statistics (Scotland) Order 2008. The Scottish ministers decided to place the decision making around PRA in the hands of the statisticians and to formalise the framework in which they work, which includes appropriate safeguards to reduce the risks associated with PRA. By proposing a committee bill, are we as politicians introducing an element of political interference in an area that is the preserve of civil servants? I would have thought that Parliament would agree that pre-release access is a matter for the chief statistician and that the independence of his role is crucial.
Scotland is facing a cliff-edge Brexit, which, according to the UK Government’s own papers, will mean increased energy prices, food shortages, price hikes, medicine shortages, impacts on employment, and the potential loss of markets for our fishing industry because of delays at ports, to name but a few areas. I am dismayed that, given the crisis that we face with Brexit, the Economy, Energy and Fair Work Committee wants to tie up parliamentary time legislating in order that a couple of stats are not released 24 hours early. The chief statistician has already limited PRA for all economic statistics to 24 hours, which the committee asked for, and the cabinet secretary has agreed that a review should take place, which the committee asked for. That leaves the remains of a proposed bill that would be so narrow in focus that it would not be a good use of parliamentary time.
As I said at the beginning, I am yet to be convinced that a committee bill is either necessary or required.
I could make my speech very short by just agreeing with what Andy Wightman said. I think that I have never said that before in the Parliament.
To take up the minister’s point about first-mover advantage, I say that information has a time value, and anybody who has information before somebody else simply has an advantage in such a circumstance.
I will go on to the main part of my speech.
As my colleagues have mentioned, pre-release approval is the practice of making official statistics and the written commentary that accompanies them available in advance of publication to specific individuals who have not been involved in their production. Those who issue the statistics may grant pre-release to an eligible person. In most cases, that includes Government ministers and officials who advise them.
In the interests of clarity, it might be helpful to set out what the bill would not do. It would not remove anything from the Scottish Government or its ministers that the UK Government or specific Whitehall departments would retain. It would not make any stipulations on any sets of data other than categories of economic statistics that are specified, and it would not call into question the integrity or professionalism of Scottish Government statisticians or other civil servants who work in the field of economic data or other areas of data.
The reasons for restricting PRA are overwhelmingly laid out in the recent committee bill proposal report. My colleagues have already spoken about those reasons. Among them are a number of preliminary positions, including the adoption of a default position of no PRA except for in exceptional circumstances; removing it for statistics that are deemed to be of particular national significance; and seeking a definitive end to the practice for all Scottish economic statistics.
The remit of the committee in exploring the issue was
“to examine the accuracy, utility and comprehensibility of Scottish economic statistics; to consider what data is required for effective delivery and scrutiny of policy; and to recommend where any improvements might be made.”
In fulfilling its remit, the committee considered a number of arguments both for and against PRA. For example, positions in favour of pre-release access outlined concerns that ministers must be properly briefed ahead of having to make a comment at the time when the statistics are published. That is because ministers are formally accountable for the statistics that have been released, and the practice of pre-release allows ministers ample time to understand the statistics in question and their broader impacts on ministerial portfolios.
For example, “Pre-Release Access to Official Statistics: A review of the statutory arrangements” was published in March 2010 and made the case that there was a “widespread expectation” that ministers should comment immediately when statistics are published. It also commended
“a central principle of good statistical practice—equality of access.”
That includes market-sensitive statistics, for which the Pre-release Access to Official Statistics (Scotland) Order 2008 recommends a PRA maximum of one working day before publication.
The review also recommended that
“it would be in the public interest if all UK administrations amended their secondary legislation to adopt a maximum period of pre-release access of 3 hours, with a shorter period as the norm.”
That position was supported by stakeholders including the Bank of England, the Royal Statistical Society, the UK Statistics Authority board and the Office for National Statistics, which ended all 24-hour pre-release access for its official statistics on 1 July 2017. The national statistician wrote to the chair of the UK Statistics Authority board, stating:
“On the basis of all the information now available to me I consider that the public benefit likely to result from pre-release access to ONS statistics is outweighed by the detriment to public trust in those statistics likely to result from such access”.
The ethical dilemma surrounding that decision demands that we in this chamber care about how statistics are treated, because they are for public consumption as information that enables the public to understand the nature of the world, the nature of policy and the nature of the decisions that are made.
Although there are arguments for and against PRA, I believe, as recommended by the committee, that the practice should end. That is at odds with the SNP’s position. The previous cabinet secretary stated that the current arrangements worked well, with the pre-release access to data allowing ministers to respond quickly to stats at the time of publication in an informed way. However, after an extensive inquiry and evidence sessions, the committee reached its recommendation to end PRA, so I urge the minister to reconsider the recommendation, and the Parliament to agree to the proposal.
I am delighted to speak in the debate. As members might know, I was the deputy convener of the Economy, Energy and Fair Work Committee when we did this work, but I have now been promoted to the Finance and Constitution Committee.
There is a lot more openness throughout society than there used to be. Freedom of information is now much more widespread, and we have added organisations such as housing associations to rules on FOI, and we might add more organisations in the future. Scotland has been good at openness, and we want to be seen as being at the forefront among open and transparent countries.
Therefore, it was not surprising that the committee questioned some of the conventions of the past, whereby certain people had access to certain statistics a considerable time before others got them. Public trust is a key component in this debate. No one is questioning the trustworthiness of statisticians or of the figures that they produce. However, trust in politicians has reduced, and what they might be doing during five days of PRA is the challenge that we face here.
We need to accept that public expectations have changed over the years; people expect more transparency nowadays. That point came to a head with the publication, in 2018, of the committee’s report on economic statistics. Paragraph 230 recommended that PRA should end, with the Government setting out how it would end it. That was the majority view of the committee. I held the minority view, which agreed that
“there should be a presumption against pre-release access”,
but left room for exceptions for specific statistics.
Although I do not want us always to compare ourselves with the neighbours, the reality is that there have been changes down south. The Office for National Statistics ended all PRA in 2017, at the recommendation of the UK Statistics Authority. On the other hand, UK Government departments that produce statistics still give themselves PRA, so there is immediately a problem with comparisons. Should we compare Scottish statistics that are produced here with those from the ONS, and so abolish PRA, or should we compare them with those from UK Government departments, and so keep some PRA? The situation is further complicated by the fact that some figures that are produced by the ONS for England and the rest of the UK are produced by the Scottish Government for Scotland.
I suppose my expectation was that the Government would probably be willing to move a bit in the direction of more openness and less pre-release access, and I suspect the committee might have accepted that, but at that point both sides dug in. The Government refused to budge and the committee considered whether to up the ante by threatening a committee bill. That was the situation on 14 May 2019, when the committee voted by four to two to move towards a bill, with Tom Mason, Angela Constance and me abstaining.
I add that I think this is a good example of the committee system working well. We all looked at the evidence and weighed up the arguments: we did not just vote along party lines.
However, things changed. On 20 May 2019, the Government belatedly agreed to move in the right direction—I see that Derek Mackay has just walked in; that is good timing—and offered to cut the maximum five days PRA to one day. I felt that that was what the committee had been looking for all along. Page 14 of the proposal report notes that I moved that the committee amend the main report to take account of the Government’s changed position, which seemed to be a reasonable course of action, but the committee decided to press ahead with publication by a majority of five to four, so we now have the report.
Overall, I think that the committee was right to use the threat of the bill to encourage the Government to compromise, and that the Government could have compromised earlier. Now that it has compromised, there seems to be much less disagreement between the two sides, so I wonder whether we really need a bill.
I accept that PRA, as others have said, might appear to be not the most urgent issue that we face, in comparison with budgets, life-saving drugs and all the other important issues that we deal with. However, trust in politics and politicians is hugely important, so anything that we can do to improve that trust has to be worth while and of long-term benefit.
The Opposition parties should remember that the rules would apply to them if they were ever to get into government. I accept that it is hugely unlikely that the Lib Dems will ever get into government—they are not even in the chamber—but the rest might think that they have an opportunity, so they should consider that point if they intend to support such a bill.
I know that many colleagues might think that pre-release access to statistics is a boring subject to debate. They would, of course, be entirely wrong, and in my short five minutes, I hope to convince them otherwise.
Before I move to the substance of the proposal, let me tell members that the last time a committee bill was taken forward in the Parliament by a subject committee was in 2003, with the Commissioner for Children and Young People (Scotland) Bill, so it has been 16 years since we had a committee bill. I have no idea why that is, because it is not as if there have not been opportunities and issues on which there has been disagreement with the Government.
I remind members that, at the very start of the Parliament, external commentators considered that having the ability to initiate legislation would give committees, and by extension the Parliament, more teeth. If the Scottish Government really did not want to do something and a committee thought that there was merit in the issue, it could bring that issue forward itself.
I will not rehearse the detail of the bill proposal, as others have done so already. The Economy, Energy and Fair Work Committee has not arrived at this position lightly or quickly. As members heard from John Mason, we have gone backwards and forwards with the cabinet secretary. John Mason himself even tried to find a compromise with the cabinet secretary, without the degree of success that we imagined he would achieve; the suggested changes were not as great as we hoped that they would be. What is before members today is a compromise and a pragmatic approach from the majority of the committee.
Stopping pre-release access to statistics is not a novelty—it really is not anything stunningly new. The Office for National Statistics does it and the Bank of England does it—in fact, they have been doing it for more than two years and the ceiling has not fallen in. The UK Statistics Authority recommends it, and the House of Commons Public Administration and Constitutional Affairs Committee recommends it, too. The Fraser of Allander institute recommends it, in a blog today, and Sir Charles Bean, a former member of the monetary policy committee at the Bank of England, recommends it. The Royal Statistical Society believes that it is absolutely the right thing to do, and would extend it across all departments. In short, the proposal is best practice. It is the gold standard that is expected of statistics. It is about transparency and trust—and facts free of spin.
Ed Humpherson, director general of the UK Statistics Authority, summed it up for me when he talked about statistics being a public asset
“that enables the public to understand the nature of the world, the nature of policy and the nature of decisions that are being made.”—[
Official Report, Economy, Jobs and Fair Work Committee,
7 November 2017; c 23.]
He also pointed to the importance of statistics being “equally available to all” without some having “privileged access”.
Every expert in the field says that we should end pre-release access to economic statistics, but the Scottish Government—
.] Clearly, the Scottish Government is wiser than all those experts in the field and is going to do something different.
The SNP’s arguments are weak. The SNP says that we do not have a problem—what complacency. It says that ministers need time to have an issue explained to them. I think that ministers are not stupid people; I think that they get it. The SNP then says that journalists will not understand it, but I think that the problem might be that journalists understand it all too well. I ask the Scottish Government: if it is such a minor measure, why not just do it?
I know that the SNP loves nation building; it loves talking about Scotland leading the UK and even leading the world. [
.] Despite the cabinet secretary’s heckling from a sedentary position, I like that too. However, in this area, the Government surely cannot be content to be described as secretive or to be second best. In truth, it wants privileged access to the statistics so that it has time to spin them. There is a growing culture of secrecy and a lack of transparency in Government, and it is not only about pre-release access. I am glad that the finance secretary is here, because it is also about the Government’s failure to publish financial information for 10 months and its deliberate delays to FOI responses.
Let us not forget that the SNP has form with statistics—it had its knuckles rapped before by the UK Statistics Authority. I will not dwell on that, because Andy Wightman is right—this is not just about this Government. We must not allow any Government, whatever its political stripe, to weaponise statistics and spin them for its own political agenda.
This is about the fundamental machinery of Government, and the nation deserves better. I ask members to support the committee bill proposal.
I have only just joined the Economy, Energy and Fair Work Committee, but I am wondering whether I have made the right decision.
I welcome the opportunity to contribute to this committee debate on the pre-release access to statistics bill proposal. A key element of producing official statistics is ensuring that they are properly understood by stakeholders and the public. Pre-release access is clearly an integral part of the Scottish statistics system and best supports that aim.
As members know, pre-release access is the practice of making official statistics available in advance of publication to specific individuals who are not involved in their production. That allows ministers and others to make informed comments at the time that the figures are published. So what is the problem? I do not see a problem.
PRA in Scotland comes under the Pre-release Access to Official Statistics (Scotland) Order 2008. The order sets out the rules and principles relating to the granting of pre-release access to official statistics in their final form prior to publication. The order was made using powers in the Statistics and Registration Service Act 2007, which allows the Scottish ministers to set rules on PRA for devolved Scottish statistics.
PRA is a longstanding practice—it has been around since before the 2008 order. The “National statistics—Code of practice and protocol on release practices”, which is non-statutory and was superseded by the order, set a maximum of 40.5 hours for market-sensitive statistics and five working days for non-market-sensitive statistics.
The Economy, Energy and Fair Work Committee carried out an inquiry into economic data and published its findings in a report, “How To Make Data Count: Improving The Quality And Coverage Of Our Economic Statistics”, on 12 February last year. The report set out the prevailing view and recommendation of the committee that PRA should stop completely for four specific areas of Scottish economic statistics.
The Scottish Government accepted a number of recommendations that were made by the committee in the report, but did not agree with the committee’s recommendation for the Government to end
“PRA to economic statistics which are market sensitive—including Scottish GDP, the Retail Sales Index for Scotland (RSIS), Quarterly National Accounts Scotland (QNAS) and Government Expenditure and Revenues (GERS)” and to set out how it would do so.
In spite of on-going and extended correspondence with the convener, Gordon Lindhurst, on behalf of the committee, it was not possible for the Scottish Government and the committee to reach a compromise over that recommendation, and here we are.
The Cabinet Secretary for Finance, Economy and Fair Work wrote again on 20 May, seeking a compromise whereby Scottish ministers would receive 24 hours’ PRA to economic statistics. PRA is still granted for statistics produced by other UK Government departments.
PRA is important for statistical integrity. One of the key policy objectives of PRA is to enable statisticians to manage the release of statistical publications effectively. The PRA period is used by statisticians to ensure that those who need to comment on the statistics at the time that they are released can do so on an informed basis without misinterpretation. It is better for ministers and others to be involved in the orderly release of official statistics than to be commenting on out-of-date or incorrect figures near the publication of the statistical publication, as that would be confusing for the public and could damage confidence in official statistics.
The public, Parliament and the media expect ministers to be able to respond to statistics when they are released, so ministers need to be clued up. I would expect nothing else. We all expect ministers to be aware of what is happening in the public services for which they are ultimately responsible.
Statisticians place considerable value on ensuring that ministers respond to statistics in an informed way, based on a correct understanding. I believe that removing PRA would mean that statisticians would have considerably less opportunity to influence the immediate reaction of ministers to statistics. Alternatively, ministers could end up saying that they are unwilling to comment until such time as they have had a chance to consider the statistics and take advice on the policy implications, which would reduce the scope for discussion and debate. The Opposition would then say that the minister was not up to speed. As far as I am concerned, Opposition members cannae have their cake and eat it. I believe in PRA—it should stay. I do not agree with the Opposition.
I congratulate those in the public gallery who have stayed the course and stuck with it. When I was allocated a speaking slot in the debate, I realised that Gordon Lindhurst would be opening the debate and I would have to listen to his dulcet tones for more than 10 minutes. Naturally enough, my heart sank. However, he exceeded himself and gave one of his wittier performances.
I would not have thought that anyone could get passionate about the subject, but members have done, and Andy Wightman summed it up rather passionately when he said that we may well ask—as someone who is not a member of the committee, I certainly did—what the fuss is all about.
What is it all about? I turned to the very useful report that was produced by the committee, and there it was. John Pullinger, the chair of the UKSA, said that, as the president of the Royal Statistical Society, he has
“always argued that fairness demands that everyone has equal access to statistics”.
That seems reasonable.
The report asks why we should care so much about the issue. Ed Humpherson, the director general of the UKSA, told the committee:
“It is because, at the heart of what statistics are about, they are a public asset.”—[
Economy, Energy and Fair Work Committee
, 7 November 2018; c 23.]
They are a public asset that belongs to us all. The figures do not belong to the Government but are there for public consumption, as information that enables the public to understand the nature of the world, policy and decisions that are being made.
That is all reasonable enough. Then we come to the attitude of the Government, which is probably the attitude of Governments across the world. It wants to know things first, because—as Kate Forbes said—it does not want to give a “knee-jerk” reaction to statistics. We can understand why a Government would say that. However, given that the statistics are a public asset, is it right that the Government should know them before every member of the Scottish Parliament?
I assume, from the quotations that we have heard from Mr Simpson so far, that we will hear a further one from the evidence of the chief statistician of Scotland. However, if members of the Opposition choose to take the bill forward, it will question and—I believe—interfere with the independence of the chief statistician, whose judgment we have trusted to determine what it is appropriate to share with ministers. Why do politicians know better than the chief statistician of Scotland, and will we hear any quotes from him?
I do not need the time back, thank you very much. [
If the cabinet secretary had been here throughout the debate, he would have heard plenty about the chief statistician. Nobody questioned his independence. [
.] Mr Wightman has put forward a perfectly reasonable proposal, which would have been totally unnecessary had the Government engaged more fully and stopped digging in. That is what has been happening, and that is why we have the rare event of a committee wanting to put forward a bill. The issue could be dealt with by regulations—we do not need to be in this position. We are lucky that Mr Mackay has turned up for the debate, but I am afraid that the situation is largely his fault.
I will leave it at that.
Disraeli is quoted as saying:
“There are three kinds of lies: lies, damned lies and statistics.”
When statistics are spun and when politicians throw them back and forward at each other, we see why Disraeli’s comment could seem pertinent and why we have to ensure that that is not the case.
As Rhoda Grant said, in setting out Labour’s support for the motion, the Government is accused of having an advantage that allows it to spin the statistics to suit its specific political narrative. A number of members have raised that concern today. I think that people will be a bit baffled as to why we are here, because it would have been best to try to resolve the issue if that was at all possible.
Andy Wightman highlighted the point that there seemed to be an inconsistency whereby Mr Mackay said that he had no influence over the chief statistician but was able to come back and say that he was prepared to put forward some kind of proposal that the chief statistician would agree to.
Both the minister and the finance secretary have tried to muddy the waters a bit today by making it about people’s confidence in the chief statistician.
Nobody is in any way suggesting that they have anything but confidence in the chief statistician; they are saying that, if statistical information can be made available to the Government and to the Scotland Office, it should be made available to this legislature and to the public. It is really quite straightforward.
My good friend Richard Lyle said that he does not see a problem with the Government’s position. However, as Gordon Lindhurst pointed out, a whole host of distinguished academics, policy wonks and think tanks say that there is a problem. It is as though the Government is doing a Boris Johnson and putting its hands over its ears. It is unwilling to listen to the evidence that has been clearly presented by many academics and others with expertise in the field, who say that a change is needed. I hope that the Government, even at this late stage, will accept that a modest change needs to be made and will proceed to make it.
That is the point. I think that it was John Mason who said that any of the political parties in the Parliament that have aspirations to be in government should be aware that the proposed changes to pre-release access would apply equally to them. The answer is that, if this is the right thing to do, we should do it, regardless of the political colour of the Government. The key point is that we should introduce fairness to the system.
I urge the Government to think again. Let us get this sorted and then, as Kate Forbes said, get on to the big issues that impact on people’s lives.
I add my thanks to our committee clerks for their always-diligent efforts. From the outset of our inquiry into economic data and throughout the committee’s consideration of pre-release access, their input and advice, and that of SPICe, has been immensely valuable.
Our convener, Gordon Lindhurst, provided a comprehensive account of the committee’s work in the area so far. We have looked in significant detail at the whole range of economic figures that are produced in Scotland, and the question of pre-release access has been a common thread across that work.
The committee decided to take a further look into the issue in order to highlight what is to most of us an example of a clear anomaly. Today, we are reporting back to the Parliament with some sensible proposals for change, because the question at the heart of today’s debate is one of fairness and good practice.
In his introduction to the report of the 2010 review of pre-release access, the chair of the UK Statistics Authority commented that equality of access is
“a central principle of good statistical practice”.
At a time when economic statistics often cause significant political ripples, the issue of fairness arises, too. In several cases, pre-release access provides what is obviously an advantage to ministers, allowing them to formulate responses well in advance of release. Where the matters under discussion are controversial, that advantage also acts as a disadvantage for others.
Some have suggested that ministers are in a unique position in that they are expected to give informed comment quickly, but in practice that amounts to saying that ministers—and possibly, in some limited circumstances, the 24-hour news media—will be inconvenienced by a reduction in the scope of pre-release access. I would have more sympathy with that position if those considerations were applied equally to others. In giving evidence in support of a no-change position, Keith Brown commented that pre-release access ensures that,
“when ministers are called upon to respond quickly to stats at the time of publication, they can do so in an informed way”.—[
Official Report, Economy, Energy and Fair Work Committee
, 14 November 2017; c 32.]
I see that that concern does not extend to other parties that are involved, and I do not just mean Opposition parties in this Parliament—I am referring to the wide range of organisations beyond the Parliament that can be greatly affected by such releases.
I will not, at this stage.
Those organisations are also often called on to comment at short notice. It seems a peculiar argument that ministers should be informed while others are not. If ministers cannot prepare informed statements without pre-release access, they must equally accept that it becomes impossible for others to challenge their statements without having a full understanding of the facts.
The committee’s proposals deal with those dual issues of fairness and good practice, which should, in short, be the starting point for the regulation of pre-release access. We have heard from a number of members of the committee, as well as from a number of other members from around the chamber. My colleague Dean Lockhart spoke about the Government’s wider duty to be transparent and accountable to the chamber and to the people we represent. What is proposed would end the unjustified disparity between the Scottish and UK Governments.
We are talking about economic data, but I will come back to that point.
Dean Lockhart spoke at some length about the comparison between the two Governments. I will not reopen his argument, save to say that the Scottish Government appears to find itself in an anomalous position. It is down to this Parliament to take a view on whether that is acceptable.
Bill Bowman went into some depth on the bill proposal document that the committee has published. As I said previously, how we take forward those principles of fairness and good practice will ultimately be a matter for Parliament. He spoke about the merits of the option of having a phased approach and an independent review of the impact of the removal of pre-release access. Although a default position is proposed, it can be adapted to exceptional circumstances. It is not an ill-considered proposal that disregards caution; it is one that addresses legitimate concerns.
I want to briefly cover some other contributions. Graham Simpson highlighted Gordon Lindhurst’s fantastic, witty performance and was disappointed that it was only 11 minutes long, but I thought that he made a good contribution for somebody who was clearly inspired by the subject of the debate.
Andy Wightman spoke with great passion and knowledge. He also offered practical solutions in what I thought was an excellent contribution.
I thought that the minister engaged well with the discussion. It is clear that she is a great fan of all things data—I will leave it to you how you take that comment, minister. She mentioned the Scotland Office and other organisations, but it is our responsibility as MSPs to be able to hold the Government to account—that is why there is so much concern over pre-release.
Jackie Baillie mentioned the Fraser of Allander institute. In relation to today’s debate, it said in its blog:
“this is an important event and covers a crucial aspect of the economic landscape in Scotland.”
I certainly agree. The importance of the debate has been highlighted by the fact that, since it started, the Government front bench has been beefed up by an additional minister and the Cabinet Secretary for Finance, Economy and Fair Work—albeit that he arrived slightly late.
Graham Simpson touched on a quote from Ed Humpherson, the director general of the UK Statistics Authority. He observed that statistics are a “public asset”. The public and civic society have as much of a right to see and comment on the fruits of the public sector’s work as a Government minister. That gets to the heart of the issue and why the committee has taken it forward, and I hope that the Government will support what the committee is proposing.
I happen to agree with the many members who have regretted our use of time to discuss the issue of PRA, not just because more pressing issues face our economy but because I think that the Economy, Energy and Fair Work Committee has a crucial role to play in holding the Government to account. A number of members of that committee regularly push the Government to do more on the economy, and it would be good to see the committee’s time being devoted to improving the economy. I do not see how debating PRA—
The committee’s data inquiry followed a six-month inquiry into the Scottish economy that found that the SNP had failed to meet all seven of its own economic targets. This work is an important part of that analysis. The Fraser of Allander institute’s blog describes this debate as “an important event”. Does the minister not recognise the valuable contribution from stakeholders, such as the Fraser of Allander institute, that recognise the importance of this issue?
This debate will not improve GDP—the very stat that we happen to be discussing—one iota.
The second reason why I regret the debate is that Derek Mackay offered a compromise option, which evidently has been rejected. I say to those who are asking the Government to shift and asking for small changes to be made that those changes were offered and a compromise approach was provided but that was rejected.
Members have made much of the example of the Office for National Statistics. The committee’s position seems to be heavily influenced by evidence heard during its inquiry that the ONS and the Bank of England have ended PRA to their statistics. Of course, that ignores the fact that PRA is still granted to statistics produced by other UK Government departments. Since the ONS ended PRA to its stats on 1 July 2017, only the Bank of England has followed suit. I repeat that Whitehall departments still operate with 24 hours’ PRA as set out in the Pre-release Access to Official Statistics Order 2008. That instrument has not been changed either.
I am interested in two things. First, the minister compares the Scottish Government to Whitehall departments. Is it her sincere view that the Government is equivalent to Whitehall departments? Secondly, she talked a lot about the independence of the chief statistician. The cabinet secretary is sitting next to her. His letter of May 2019 to the committee says:
“PRA is a matter for the Chief Statistician and the independence of his role is crucial ... I inform the responsible statisticians”
—this is his proposal that the minister has just mentioned—
“that I, and all other Ministers, will require PRA of only one working day”.
How does that hold with respecting the independence of the role of the chief statistician?
I do not understand the committee’s position. Committee members are asking us to do something while saying that we cannot because we have talked about the independence of the chief statistician.
On the second matter, of course I would not liken the Scottish Government to Whitehall departments. However, I take issue with the argument that we are the only ones that allow PRA—that is clearly not the case, given what I have said about other Whitehall departments. There is a big distinction between arm’s-length statistical agencies such as the ONS and statisticians working in Government. The evidence of policy and practice that has been presented to the committee is based on the views of individuals working outwith Government.
Pre-release access is a long-standing practice in Scotland and only a small part of maintaining trust in our official statistics, which, incidentally, is something for which we have a strong reputation. The arrangements for PRA to economic statistics in Scotland have been in place for many years, including under Labour, and it was common practice even before the Pre-release Access to Official Statistics (Scotland) Order 2008 was introduced.
We have engaged very positively with the committee’s economic data inquiry and have accepted the majority of its recommendations, but we feel strongly about this matter. There is simply a difference of opinion. The fact that each Administration of the United Kingdom drafted its own order shows the division of opinion on the practice of PRA.
We have always been led by the judgment of professional statisticians. The committee’s proposal disregards the fact that the Scottish Government has a managed and well-functioning process for PRA.
The Pre-release Access to Official Statistics (Scotland) Order 2008—I highlight that it passed through this Parliament with no division—means 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.
We strongly support the existing PRA framework, which strengthens and empowers statisticians to act in a professional manner. There has been no material change that would warrant a change in practice.
This debate has come about because the committee has refused to acknowledge and honour the statistical arrangements that the Government adheres to whereby ministers accept the professional advice about statistical matters of Scotland’s chief statistician. The proposal for a bill disregards the established practice that successfully operates in Scotland.
Presiding Officer, I cannot remember how much time I have left, but I want to draw to a close with some positive words from the Government on this debate.
The point is surely that the statistician operates within the legislative framework that is provided for him, as the statistician will acknowledge. What is at issue here is whether the current legislative framework should continue.
And that is the difference of opinion that I just identified.
Official statistics are crucial and their importance is maintained by statisticians’ work to realise the value that is inherent in the vast amount of data that the Government holds and makes publicly available. Data is a public asset, and the public have access to data. We want to make data publicly available in an ethical and transparent way.
I have enjoyed the exchange of views in today’s debate and look forward to hearing the closing speech.
Thank you, Presiding Officer.
I thank the clerks and SPICe for their kind assistance in providing useful and important background on the topic. As a new member of the committee, I was not part of the build-up to this exciting debate. However, I have had a look at the issues that have arisen since the committee published its report, “How To Make Data Count: Improving The Quality And Coverage Of Our Economic Statistics”, in February 2018.
There are differences of opinion, as we heard today, which makes this debate a wee bit unusual. Committee bill proposals usually emerge from a united front. As I understand it, there have been only seven committee bills to date, and they dealt with regulatory matters or established commissioner posts. The proposal that we consider is quite different.
I commend members on both sides of the debate for their thoughtful and, at times, robust contributions. At the heart of the debate is whether the Government should continue to be afforded pre-release access to certain statistical information.
The committee’s view and proposals—agreed to by majority—are set out in its report of 6 June. The committee proposes that pre-release access to statistics be removed entirely for Scottish GDP and retail sales figures, with a subsequent review of the impact of the change, and that PRA be reduced from five days to one day for statistics for which a five-day PRA arrangement is currently in place.
The justification for the proposed approach is that all statistics are public assets—as many members said during today’s debate—which should be equally available to all people, rather than available to some and not others, which potentially gives first-mover advantage to the people who get them early and, ultimately, risks generating public scepticism about the credibility of the statistics.
The counterarguments are that PRA is a long-established practice, which is determined and controlled independently by the chief statistician; that PRA is an essential part of the support that is afforded to ministers, whatever Government is in power, to enable them to offer informed comment on statistics that are released; and that PRA is enjoyed by and will continue to be available to the Scotland Office and UK Whitehall departments.
The role of the chief statistician in helping to resolve the issue could be crucial. As I understand it, the 2008 order specifies the rules and principles that relate to the granting of PRA, including who gets access and when. It is the responsibility of the independent chief statistician to apply those rules and principles appropriately; he decides who gets advance sight of statistics.
The chief statistician gave his view at a meeting of the committee in November 2017. He said that
“There are much more important issues”,—[
Economy, Jobs and Fair Work Committee
, 14 November 2017; c 33.]
and he gave examples, which included data handling, security and establishing a culture of independence for his function. However, the issues remain, and evidence of concerns was given in today’s debate.
I will give a flavour of one or two of the important contributions that members have made. Our convener, Gordon Lindhurst, opened the debate with a fairly comprehensive summary of how we got to where we are. He regretted that it is a repeat debate, in a sense, but said that it is still important to engage. He said that statistical integrity is crucial and outlined the pros and cons in the debate and, as many members did, the options that the committee has presented.
Like other members, he described Scotland as an “anomaly”. He said that compromise perhaps arrived a little late and argued that there is an external view that PRA should go, that many bodies support an end to it and that abolition costs nothing—I think that that was one of his final remarks. He described himself as an optimist and said that he knows what he wants, using a quotation in that regard.
The Government minister, Kate Forbes, opened with important comments about more pressing economic matters that the committee and other members of Parliament should perhaps face. She said that the cabinet secretary has offered a compromise solution. She made it clear in her opening remarks and in summing up that PRA is the norm and is retained by other UK departments. She pointed out that the critical independence of the chief statistician would end as a result of the committee’s proposal, and that view was shared by some members but not all. She said that there is no higher right of access for ministers and that ministers are expected to respond to the publication of statistics immediately. She said that Parliament needs to focus its energies on more pressing economic matters.
My colleague Dean Lockhart started off by telling us about FOI issues and the culture of secrecy that he feels is prevalent in some Government circles. He argued that PRA is contrary to the principle of equal access for all and said that statistics are a public asset—that is the nature of some of the statistics and the way it should be.
Andy Wightman made a passionate case for ending PRA. He regrets the need to have the debate, which started in 2008, when issues about pre-release and equality of access were raised. He said that successive Administrations passed the buck and that the issue is about what the law should say and whether the chief statistician is effectively independent of Government.
My colleague Gordon MacDonald gave a powerful defence of PRA. He is not convinced that the bill is necessary. He referred to the committee’s proposals and said that the Bank of England, which was mentioned several times, still has PRA, which is renewed every year. He told us that the Cabinet Office has rejected ending PRA and he argued that PRA is important for good Government, that none of the 30 UK departments supports ending it and that there has been no legislation on the issue since 2008.
We can see the ebb and flow of the debate. Jackie Baillie, speaking with her usual passion, said that the most recent committee bill was in 2003, some 16 years ago, and that the committees’ ability to introduce bills gives the Parliament more teeth.
Dick Lyle said that he almost immediately regretted joining the committee, as this has been his first opportunity to speak in the chamber as one of its members. He, too, spoke about the importance of PRA and of allowing Government ministers to do what he considers to be their job in representing their portfolios.
I am running out of time, so I will finish by thanking members and apologising to those whose comments I could not speak about. I thank members for highlighting the issue and for their tenacity in pursuing it since the report was issued.
Statistics offering economic data matter to us a great deal—we can tell that from today’s debate. It is clear that all members are keen that such statistics are handled sensitively, fairly and properly and in a manner that allows the Government to do its job, but does not disadvantage others who are entitled to question the Government and hold it to account.
From what I heard from the Government, it would be content to operate with a maximum of 24 hours’ pre-release access being applied to all statistics, which does not seem too far from the position that the committee set out in June.
There are some people who say that this type of debate is pointless, that the notion of replacing a clash of ideas and visions with a form of policy calculus was always dubious and that anyone still hankering for it should admit that their number is up. A statistician, or even a politician, can have his head in an oven and his feet in ice so that he can say that, on average, he feels fine.
I hope that I have given a fair summary of the committee’s views on the matter and of the important contributions made by its members. I sincerely hope that we can find a solution that will deliver a balanced approach to what is an important issue for the whole Parliament.