Meeting of the Parliament (Hybrid)

Meeting date: Thursday, March 4, 2021

Official Report 1000KB pdf

---

Pre-release Access to Official Statistics (Scotland) Bill: Stage 3

As members will be aware, at this point in the proceedings, I am required under the standing orders to decide whether any provision of the Pre-release Access to Official Statistics (Scotland) Bill relates to a protected subject matter—that is, whether it modifies the electoral system and franchise for Scottish parliamentary elections. In my view no provision of the bill relates to a protected subject matter and therefore the bill does not require a supermajority to be passed at stage 3.

As we know, there are no amendments at stage 3, so we move straight to the debate on motion S5M-24057, in the name of Gordon Lindhurst, on the Pre-release Access to Official Statistics (Scotland) Bill.

15:03  

Presiding Officer,

“They say we’re young and we don’t know”—

or at least they did five years ago. They could be forgiven for feeling that we are all now trapped in a type of groundhog day. We debated pre-release access in November 2018, September 2019 and November 2020 and we are debating it again today. It has taken four years, three cabinet secretaries, two ministers and a change of convener to get here, but it is my hope that we are about to break out of this time loop.

Statistics matter—they not only describe the world, but help to shape it.

Will the member take an intervention?

Not at this stage.

Two hundred years ago, a politician wrote the 21-volume “Statistical Account of Scotland”, an undertaking said to have required the labour of Hercules combined with the patience of Job. The author, Sir John Sinclair, saw it as an inquiry

“for the purposes of ascertaining the quantum of happiness.”

His view of its relevance to the public was that it was

“the means by which their temporal and eternal interests can best be promoted.”

Then, as it is now, data could be a guide to the decisions affecting us, and the Economy, Energy and Fair Work Committee believes that it should be available on an equal and not a privileged basis. That is the premise of the committee’s bill. As the president of the Royal Statistical Society wrote to the First Minister last September,

“Quite simply, allowing a government privileged access to official statistics risks undermining public trust … it creates opportunities for figures to be ‘spun’ to the media or ‘buried’ beneath other announcements.”

The bill would do three things. It would remove pre-release access for two categories of economic data, take a phased approach to that removal and review its impact, and reduce to one working day the pre-release access for statistics where five is now the norm. Let me elaborate.

In the midst of the biggest economic crisis since the second world war, is it not somewhat worrying that the biggest and most pressing issue that the economy committee thinks the Scottish economy is facing is pre-release access to statistics?

That is the same repeated, and, frankly, boring point that has been made previously in these debates. It is worrying that the Scottish Government considers it important to oppose so obvious a solution and spend the Parliament’s time doing that in the course of a pandemic rather than agreeing with it.

On the cabinet secretary’s point, does the member agree that at this time of crisis and emergency, we need facts and figures and that this is about the timing of the release of statistics? No additional effort would be needed; it would just be done a bit more quickly.

Yes, I agree with the member and I will come on to that point.

Let me elaborate on what the bill will do. The first strand would end pre-release access for retail sales and gross domestic product, neither of which is subject to pre-release at the UK level, and the second strand would stipulate that the removal of pre-release access should be phased. One day would be reduced to a half day after one year, then pre-release access would be removed altogether after two years, with an independent review of the impact after three years, the findings of which would be laid before the Parliament. The third strand would bring pre-release access down to one day for economic data where a longer duration presently applies.

However—here we come to the point—why does all that matter? Why the fuss? Why at this stage? John Pullinger, a former United Kingdom national statistician, suggests that, if life can be unpredictable,

“Statistics can help us to assess risk and to stay the right side of foolishness”,

and that they provide

“a balance to our sometimes wayward hearts.”

Perhaps that is relevant to the present Scottish National Party Government in relation to this matter. The trick, Pullinger says, is to encourage statistical thinking. Eight out of 10 cat owners who expressed a preference said that their cats preferred it. However,

“statistical thinking helps us to ask which cats, did they really prefer it, and prefer it to what?”

The risk of not engaging in statistical thinking is highlighted by Daniel Kahneman. The Nobel prize-winning psychologist contends that

“it probably contributes to an explanation of why people litigate, why they start wars, and why they open small businesses.”

Some might not consider limiting pre-release access for economic data to be a headline grabber, but statistics are a public asset. They belong to us all, so they should be available to everyone at the same time. In the language of The Wall Street Journal, there should be no “early peeks”.

We are far from alone in coming to that conclusion. That view is shared by a majority of the statistical community, including the Office for National Statistics; the Royal Statistical Society; the UK Statistics Authority; the Bank of England; Professor Sir Charles Bean, author of the 2016 independent review of economic data; Dame Jil Matheson, former UK national statistician; John Pullinger, whom I already mentioned; Professor Sir Ian Diamond, the current UK national statistician; the Institute for Public Policy Research; the Fraser of Allander institute; the Adam Smith Institute; the House of Commons Public Administration and Constitutional Affairs Committee; Michael Blastland, creator of Radio 4’s “More or Less”; Will Moy, chief executive of Full Fact; and Sir David Spiegelhalter, the statisticians’ statistician.

I will not carry on with the list, because I can see the Presiding Officer looking at me with regard to the time. I will not make any further song and dance about it but, come decision time, I hope that we might add this Parliament to that list.

I move,

That the Parliament agrees that the Pre-release Access to Official Statistics (Scotland) Bill be passed.

Ivan McKee will open for the Government.

15:11  

If we go by some of the debates that we have had on the budget over the past while, it would be fair to say that eight out of 10 Tory MSPs do not understand statistics.

I start by reiterating two key principles that this Government stands by. First, data statistics and evidence are at the heart of our policy decisions, and secondly, we need to understand, explain and be transparent about why decisions have been made. High-quality and relevant official statistics, trusted professional statisticians and well-informed politicians, who understand the data, are vitally important in allowing us to follow those principles.

Faced with unprecedented challenges to physical and mental health, finances and our way of life, the importance of data, evidence and statistics has never been greater. That is why I am disappointed that this Parliament’s focus and energy has been on a bill that aims to challenge a valuable, managed and well-functioning process, rather than on building further trust in the value of our statistics.

Our position remains that we oppose further restrictions on pre-release access and we consider the bill an unhelpful distraction. The quality of our economic statistics—and official statistics more generally—is paramount and the Scottish Government fully complies with the code of practice for statistics.

Will the minister not concede that the bill is not about the time that is taken to prepare statistics? It is purely about their release and who has access to them once they are prepared. I agree with him that it is important that time is taken, but why should the Government have privileged access? Surely, various people’s opinions are equally valid to that of the Government, and that openness is required in order to build trust.

The member is correct that the issue is not about the time that is taken to prepare the statistics, but it is, as I said, about the time that is taken to understand the statistics, because the reality is that the Government uses the statistics to make decisions. Those decisions need to be correct and the Government has to answer to what lies behind those statistics, not just to the headline number. It is easy for people to ask the questions when they have just seen the statistics but, to give a sensible, meaningful answer, we need to understand what lies behind those numbers, why they are what they are and, frankly, what we are going to do about them.

Our current arrangements for pre-release access provide a clear framework for statisticians to manage and communicate the numbers that they understand the best. The reason for our opposition to the bill is not, as has been suggested, that we want to protect our first-mover advantage, but that we believe that the governance and operation of the statistical system in Scotland is best left in the hands of the experts. The experts are the highly skilled professional statisticians, led by the chief statistician, a civil servant who is bound by the civil service code of integrity, honesty, objectivity and impartiality. It is his view that pre-release access is an integral part of the production process for official statistics, and that it operates well and appropriately. He believes that current arrangements strike the correct balance between carefully controlling access and ensuring that responses to questions on public statements are based on a correct understanding of the statistics—that is the key point. The bill will not improve public trust in official figures but will make achieving that balance more challenging.

Doing my job is made easier by the statistics that I see on public finance, economic growth and trade. Being able to work with statisticians helps me to properly interpret the numbers and take decisions that are in the best interests of Scotland. If we are serious about delivering improvements, we need to understand the story behind the statistics rather than rush to comment on numbers that we have just seen. Understanding the why, not just the what, of the numbers is critical to being able to comment from a position of understanding and not just take part in a battle of soundbites. Now more than ever, we should be reducing the risk of misinterpretation or confusion over the figures and the resulting significant and damaging impact on public trust.

I end on a positive note. I am proud to say that the Scottish Government plays a leading role in improving how data and statistics are used to deliver real benefits for Scotland and beyond. Public trust in the Scottish Government to act in the best interests of the country remains significantly higher than it does in the UK Government. To use some statistics, according to the latest Scottish social attitudes survey in 2019, 61 per cent trusted the Scottish Government to work in Scotland’s best interest, compared with 15 per cent for the UK Government.

Regardless of the outcome of the debate, the Government is committed to continuing to build on that success by following the three pillars of the code of practice for statistics, which are invested in the trust, quality and value of our official statistics. We will continue to support the work of highly skilled statisticians to realise the value that is inherent in the vast amount of data that the Government holds, and to make that publicly available in an ethical and transparent way.

15:16  

When last we debated the bill, I pointed out that statistics are not just numbers on a page, but a public asset that is used to inform policy. It is therefore vital that the public has trust in both the statistics themselves and how they are used.

The current model of privileged access in Scotland does not lend itself to maintaining, let alone strengthening, that trust. That is because SNP ministers currently enjoy a level of early access well beyond what is required—a full five days, in some cases. That allows ministers far too much leeway to spin figures or even to try to bury them away. We need only think back to Derek Mackay’s attempts, a few months before he was forced from office, to spin the dropping of employment by 43,000 by deflecting to a 0.3 per cent decline in youth unemployment, or his attempt to spin a £12.6 billion “Government Expenditure and Revenue Scotland” deficit as somehow a sign of strength.

There is a clear need for reform—and experts agree. The UK Statistics Authority has called for the SNP’s excess PRA period to be significantly rolled back, and it is not alone. In evidence to the committee, Martin Weale of the Royal Statistical Society called the lengthy period of pre-access in Scotland

“an anomaly relative to almost the whole developed world.”—[Official Report, Economy, Jobs and Fair Work Committee, 26 September 2017; c 9.]

It is disappointing, then, that the SNP has chosen to defend its privilege and to oppose reform at every opportunity. It rejected the committee’s initial recommendations, which forced the committee to pursue reform through legislation. Every SNP committee member then opposed the bill, and the party refused to vote for it at stage 1. We even had a minister—Ben Macpherson—claiming that the bill was a political attack on the SNP. He said that the

“intention to remove pre-release access, at least somewhat, seems political”.—[Official Report, 12 November 2020; c 87.]

In reality, the bill takes a measured approach to reform that recognises the need for ministers to have a sensible level of early access. In fact, the bill is far more generous than some have been calling for. The UK Statistics Authority wants PRA to be reduced from five days to just three hours, but the bill offers a full 24 hours for certain economic statistics. Even where PRA would be removed—for GDP and retail statistics—there is a phased approach, not a cliff-edge cut. PRA would be initially reduced to 24 hours, then, after a year, to four hours, before being removed entirely. An independent review mechanism will examine the impact on GDP statistics. If access needs to be restored, that will be able to be done without further legislation.

The bill does not seek to intrude beyond the committee’s remit into education, health or any other portfolio area.

I do not know how good the member is with numbers, but how long does he think that it would take him to understand a set of numbers and be able to pass sensible comment on them?

A lot quicker than it would take the minister.

The bill does not question the integrity of Scottish Government statistics. It simply seeks to address valid concerns that have been raised by experts. The bill is entirely reasonable, and I urge members to support it.

15:20  

Before I get to the important substance of the debate, I commend the committee for introducing a committee bill, which is an underused mechanism in the Parliament. I believe that I am correct in saying that the Parliament has passed only seven committee bills since it came into being in 1999, and all but two of those bills were about internal regulation, matters of standards and other such issues, so the committee is to be commended for introducing a bill of substance that will make a difference. I call on committees in the next parliamentary session to seriously consider using the power and capacity that they have. I am recommending that my committee’s legacy paper makes suggestions about future committee bills that could be introduced.

This is a bill that matters and an issue that matters because, as Gordon Lindhurst set out, statistics matter. We live in a world of post-truth politics where we constantly see the undermining of information sources and the questioning of facts. Quite simply, we need to build back trust, because truth matters, experts matter and statistics matter. What destroys trust is the sense that things are being only partially presented, being spun or being presented in a manner that protects particular interests and diminishes others. The concept of framing information is well understood and one that everyone in the chamber understands. The more opportunity we give for things to be framed from a particular vantage point, the more people’s sense of mistrust in facts and statistics will increase.

We have heard from the Scottish Government that it needs time to understand things. I put it gently to the minister that I am sure that he does not need more than 24 hours to understand a set of numbers. I know that he is pretty good with numbers, and I know that, given a statistical release, I do not need more than a single sleep to digest it.

The minister is right that we need to ensure that people understand what numbers are saying, but I politely say to him that the Government’s perspective is not the only valid perspective—it is not even the only important one. It is important that we have equal access so that we have a balanced debate. For as long as people feel that the debate is imbalanced, we run the risk of undermining trust, which we need to combat.

I politely suggest that the Government needs to learn that lesson urgently. In this week of all weeks, the sooner it releases information, the better. Delays in discharging its duty simply undermine public trust.

I also politely suggest to the Government that, right now, it is sitting on information—critically, the Organisation for Economic Co-operation and Development report—that the Parliament has asked for it to release and which it must release. It is not right that the Government can sit on that report for months while it gives suggestions and asks for alterations and updates.

Quite simply, in the 21st century, time matters. There are no secrets in the 21st century; all that there is is openness. The more people delay or seek to delay, the more they build in mistrust. The time periods were introduced at a time of typing pools and paper memos. A century ago, it took time—it might well have been days—to disseminate information, but it now takes literally milliseconds for information to be duplicated and disseminated.

Quite simply, pre-release access is not good practice; best practice demands early release. Best practice is supported by the Royal Statistical Society, the ONS, the UK Statistics Authority and others. I will put it like this: if the Bank of England, whose data sets are among the most sensitive that are produced across these islands, does not enjoy pre-release access, why should the Scottish Government have it?

15:24  

Let me start by picking up on a few things that have already been said. Daniel Johnson talked about data. Data becomes information only when it has been analysed. In other words, providing data is not an immediate provision of information.

We also heard reference, from the minister in response to a Tory member, to the code of practice for the use of statistics. It is worth saying something about that code of practice because, in effect, the Government is bound by it. That is associated, in part, with the proposals before Parliament today, because the code of practice for the use of statistics is not applicable to the political parties that are in opposition.

The code says:

“By complying with the Code, your organisation will show that: ... It is ethical and honest in using any data ... It respects evidence”

and

“It communicates accurately, clearly and impartially.”

Those duties are placed on the Government, and the Government is held accountable for obeying them and the ministerial code. No such obligations are placed on Opposition parties if they receive data without information at the same time as the Government. They can immediately comment and are not held to account should they selectively quote favourable data or communicate it in a way that is not accurate, clear and impartial. However, the Government has to take time to ensure it meets those standards. Therefore, the artificial suggestion that this creates a sense of evenness and balance between Government and those who hold it to account is a false distinction that simply does not bare reasonable analysis.

I am interested in statistics; I am a humble mathematician. My wife is also a mathematician, and she has a statistics qualification in addition to that. I always go to her. She tells me—this is a matter of grave concern to me—that, statistically, I shall be on this planet for another 12 to 14 years. That is not very long, so I take a close interest in that statistic and hope that the actuaries and statisticians who produced it are underestimating the length of time that I now have left.

The bill seeks to provide information to Opposition parties. Giving information to Opposition parties is good; I have been in opposition and know how valuable it is. However, in providing information, the bill provides nothing in the way of controls and responsibilities for the recipients of information who are not in government.

That goes to the heart of the principal flaw in taking the approach that is proposed by the committee. I respect the committee’s work and the reason why it has introduced the bill—those are both to be respected and applauded—but I am afraid that it fails the test of creating a level playing field, which is what advocates for the bill suggest that it does. Unfortunately, it does no such thing.

On a point of order, Presiding Officer. Can you clarify whether all members are under an obligation to speak the truth in the chamber and that misleading Parliament is taken very seriously? I am thinking about Mr Stevenson’s comment that there is no such equivalent obligation for Opposition members.

That point is true: every member has an obligation to speak the truth. I am not entirely sure that Mr Stevenson was not speaking the truth. He was giving a point of view.

I was going to tell Mr Stevenson that his time was up, but after his contribution I do not think that I should use that choice of words.

15:29  

Thank you, Presiding Officer. I hope that my time is not yet up, in the chamber or elsewhere.

World statistics day was last November, and the tagline was

“Connecting the world with data we can trust”.

Although the bill is purely about economic statistics, it is probably fair to say that, over the past 12 months, we have all come to appreciate the importance of data and the excellent work of our statisticians.

The minister’s comment about Tories and statistics is a slightly odd one, given that every member of the cross-party committee, apart from the SNP members, supported the bill. At the same time, even the SNP members, who were a minority on the report, took the following view:

“The Committee considers there should be a presumption against pre-release access and invites the Scottish Government to put forward arguments why pre-release access should be continued for specific statistics.”

Therefore, it was not just the majority and the cross-party consensus but the SNP minority who expressed that view about the current situation.

I think, as other members have said, that we really need to address in Parliament the issues that surround openness, fairness and transparency in these things. I do not think that members of the public who have observed some of what has taken place in Parliament over the past five years would refer to it as a balanced, well-managed and functioning place in every respect. The bill is a small step towards ensuring that the systems that are in place are conducive to having a Scottish Parliament and a Scottish Government that are held to account so that they are in fact balanced, well managed and functioning.

With those words, I close my contribution to the debate.

That concludes the debate on stage 3 of the Pre-release Access to Official Statistics (Scotland) Bill.