Treasury review of government statistics

At first it seemed a bit like an episode of “The thick of it“. An independent review of an independent body “while fully protecting” that independence. You can almost hear the words coming from the head DoSAC, Nicola Murray. But this is real life, it’s under a Conservative majority government and it’s about something really important – the statistical underpinning of our democracy. Apart from showing wonderfully how the word “independent” has been devalued in politics, this is a stunningly important review that will help to set the course for policy-making and democracy for years to come.  

The Treasury announced on 10 July that there will be “an independent review of the quality, delivery and governance of UK economic statistics”. Everything about this is interesting.

The terms of reference of the review are to:

  • assess the UK’s future statistics needs in particular relating to the challenges of measuring the modern economy;
  • assess the effectiveness of the ONS in delivering those statistics, including the extent to which the ONS makes use of relevant data and emerging data-science techniques
  • while fully protecting the independence of the UK national statistics, consider whether the current governance framework best supports the production of world-class economic statistics

The FT marked the “launch” of the review (and call for evidence) with two articles yesterday (“UK statistics out of date, says Bean” and “Lies, damned lies and the statistical needs of users“) and an editorial today (“Lies, damned lies and a better path for statistics“). The editorial was clear in its message: “UK statistics fall short of what is acceptable for a modern economy”. It seems that this view is now in the mainstream and in the open. I feel less lonely!

But what’s driving this? Perhaps there are three main contributing factors.

The first is a long-standing concern among the so-called key stakeholders – most importantly, the Bank, OBR and Treasury – about the economic statistics coming out of the ONS. These bodies write their annual reports on the ONS and low marks are not uncommon. Mark Carney, the Bank governor, has compared the quality of the ONS outputs unfavourably with his native Canada (here’s one of many articles).

Second, and perhaps accounting for the timing, was the preparation of the Treasury’s major report on productivity, “Fixing the foundations“. Although the announcement of the stats review was only a minuscule part of this (appearing in para 13 of the appendix, page 79), and was justified by the so-called and very real productivity puzzle of recent years (rapid employment growth and relatively modest economic growth), you only have to read the report’s chapter headings to realise how poorly the nation’s existing figures cover the areas of interest to a government looking to understand and modernise an economy.

The third is that perhaps the Treasury did just not see enough awareness of, let alone action on, these points from the UKSA board? This boils down to concerns about the structure of the governance for statistics brought in by the 2007 Act. John Pullinger was appointed as National Statistician a year ago, has been well received and is making progress. But he has so much to do – not least as he follows two National Statisticians with no background in (and perhaps limited understanding of) economic statistics. That was not helped by a passive and far from fully transparent UKSA board that has given far too little attention to the monitoring and assessment side of its activities.

Beyond those themes, all sorts of issues can be improved if the review wants to go looking.

  1. The legislation that set up the “independent” statistics system is awful.
  2. Parliament has shown next to no interest in statistics. PASC tried (vested interest declared, I was an advisor for the committee) but was ignored by UKSA and the rest of the world.
  3. UKSA accountability is hardly any better than what went before. And its vision is unclear. Why had it not already got highly visible work underway to address the ToRs of the latest review?
  4. The code of practice guiding statistical work in government was a great start but needs to be updated.
  5. The appalling web site relaunch in 2011 (damaging credibility) and the lack of urgency to repair/replace it.
  6. The obviously ludicrous emasculating relocation to Newport. (It was said at the time that around 700 of the 800 London-based staff would leave the ONS as a result removing expertise in a stroke.)
  7. The UKSA assessment function is not doing what it ought to – and what virtually all outsiders hoped it would do.
  8. And data? There have been mistakes. The RPI and the review of it is a mess – and remarkably drawn out. The Barker review, published a year ago, led to, err, no action (that I’ve spotted). And lets not mention issues about trade, earnings, productivity and all the other factors that affect the economy, for example migration.
  9. Big data and data science? Pretty much an untouched topic for government statisticians.
  10. Open data. Ditto.

So, what are the possible outcomes from this review? Progress would undoubtedly have been made under Pullinger’s leadership but without this review it was going to be slow and never easy. Inevitably, as you try to sort out deep rooted problems, the news often gets worse before it gets better. A month ago, for example, the ONS published a note on flow of funds work which was not totally well received, at least according to an article in the Telegraph (“BoE faces long wait for Mark Carney’s vital statistics“). With this review might come the support of government that will help progress to be made.

The timing of the review has also led to speculation that the ONS might be lucky enough to be spared the worst of the expenditure cuts in the upcoming spending review. They might even get some new funds for investment.

That said, even if the motives of the review are entirely positive, and a wonderful set of recommendations is forthcoming, there will be a huge question about delivery. Money gets you only so far with statistics, expertise is vital and there not much of it around. On top of that, despite statistics being “independent”, ministers in all the major departments have a huge interest in data, not least because it measures their performance. Getting agreement among them for change will not be easy.

Even so, most of the politicians (and most of the statisticians, no 10 and Treasury officials) who gave us National Statistics and then the Act have gone. So a new government looking at this with a critical but interested eye would not find it hard to advocate change. The potential to come up with a better way of doing things should be welcomed.

There are risks, of course. The prospect of “political interference” does not sound good but it comes in many forms and need not be bad. The clear vision of the new National Statistician coupled with positive political leadership and some money would be a good thing. The worst fears, some sort of “Rayner 2” era of cost-cutting, is surely not where this is going. And without a radical shake-up? Would anyone be really confident that the current UKSA board over-looking the existing Newport set-up would really deliver what’s required?

The review will make interim recommendations to the Chancellor in the autumn, with a final report published by Budget 2016. It looks like money might be involved – justifying spending round cuts to important infrastructure such as statistics would need the support of an “independent report”, as would increases, albeit just a few £millions, at a time when other budgets are being slashed.

Press coverage of the announcement a month ago included the Independent, Times and Telegraph.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s