Skip to content
The UK's only independent think tank devoted to higher education.

The Statistics Authority versus the DfE: Round X

  • 11 October 2018
  • By Nick Hillman

At HEPI Towers, we’ve been discussing the latest spat between the UK Statistics Authority and the Department for Education about education spending. In a new exchange of letters, the Department has largely backed down.

But I remain more sympathetic to their original position than, it seems, most people are.

The Department was accused by headteachers who marched on Westminster of cutting public spending on each school pupil. The hard evidence for this comes from the think-tankers think tank, the Institute for Fiscal Studies. They are not immune to errors, and – like everyone else – have not always managed to predict the future perfectly. But they:

  • are well resourced (their income is not far off fifteen times that of HEPI’s);
  • have some great economic models; and
  • are about as good as it gets in think-tank land.

So you take them on at your peril. Plus, if you doubt their numbers on falling spending, teachers and governors will soon put you right about the effects in the classroom. In other words, it’s not worth it.

But the Department didn’t take the IFS on directly. They picked data from another well-respected body, perhaps the most well-respected source of international comparative data on education, the OECD, to tell a different tale.

The headline OECD data were for a different period and showed something different: all education spending rather than the public funding of schools on a per-pupil basis. If it were an A-Level essay, the DfE’s answer would have been too far away from the statement in the original question to bear many marks.

But public policy discussions are more complicated and wide-ranging than that. They should range far beyond a closed essay question. So long as they are not lying, it is the prerogative of every government department to try and set or recalibrate the agenda by talking about their policies in the way they choose, just as it is the prerogative of everyone else to criticise them when they do it in a way that looks tricksy. In other words, the Government’s role when explaining what it does is not just to respond to accusations but also to make the weather.

At worst, the Department can be accused of providing data that might confuse less well-informed observers. That’s bad for the quality of public debate. But they’ve been accused of lying (‘using dreadfully misleading information’, ‘deliberately trying to con the public’ and ‘Failing to face up to the truth’ – see when what they have done cannot bear the accusation of deceit, in my opinion.

It is quite similar to what happens every week at Prime Minister’s Questions (PMQS). The Leader of the Opposition wields one set of facts to claim the Government have harmed something; thePrime Minister then responds saying this is nonsense, typically using different data. Generally, they are both right – and they both know the other is right.

Given their slapping down of the Department, the Statistics Authority clearly disagree with this model of political debate. Their intervention blurs the line between their standard role of querying the misuse of data and stepping into the middle of a party-political dispute: given the Official Opposition stormed into the row to support the heads, it is not just a disagreement between non-partisan headteachers and a Government Minister; it is also a party political row akin to what happens each week at PMQs.

This matters because the spat between the Statistics Authority and the Department for Education is not just standard Westminster village fare. It is serving to conceal some important things. People could be left thinking the OECD data is untrustworthy.

In particular, their numbers have been criticised for including all education spending rather than just public spending on education – meaning the inclusion of private school fees as well as student loans (a little more than half of which are due to paid back to the Government via semi-private loans).

But there two good reasons why the OECD presents the data like this that we should not lose sight of.

  1. First, some private funding displaces public funding. For example, if all private schools were abolished tomorrow, the official education budget would surely need to increase to accommodate them. So there is an intimate relationship between the two and it makes sense to know how much is spent on education in total across the country as a whole rather than just how much the Government spends on it from taxes (and extra borrowing).
  2. Secondly, the OECD used to exclude student loans from much of their data, which gave a warped perception of resources in different countries’ university systems. They, rightly in my opinion, changed this, after complaints from the UK and possibly others. Moreover, the arguments on presenting this information are so finely balanced that the Office of National Statistics is undertaking an international review of accounting standards to determine whether to include some or all student loans as current public spending. If they do this, then the gap between Whitehall’s figures for public spending on education and the OECD data would start to close. Would the Statistics Authority have to change tack?

I think the row has missed something more important, notable and worrying about the Government’s position. Ministers have defended themselves by claiming that spending has increased since 2000, which is true but also a slightly weird thing for them to draw attention to. It implies headteachers are wrong to focus on the decline in public spending per pupil since 2010 and that they should rather look at the increases in spending over the period as a whole since 2000.

Yet, as this chart from the Institute for Fiscal Studies makes clear, that decline took place after the Conservative Party took office in 2010 (as part of a Coalition until 2015 and then on its own) and the increase took place while the Labour Party were in office.

In other words, Conservative Ministers are seeking to take credit for Labour’s spending. This would not generally happen and, if this were a standard political row rather than one that has become confused as a result of the Statistics Authority acting as a referee on a different issue, it would have received far more attention.

And to voters trying to understand the ups and downs of education spending and the priorities of different political parties, it may well be more important.


  1. Hugo Dale-Harris says:

    Nick, as we’ve discussed in ‘HEPI towers’ the crux of the issue is what counts as ‘misleading’. This UK stats authority ruling is making judgment not on the veracity of the OECD statistics but on the way in which they are presented. The OECD data on overall education spending (including tuition fees, private school fees) is a valid, even useful statistic. However good statistics can be used misleadingly. Frequently the Department for Education has used a vague statement along the lines of “the UK spends more on education than ever before”, as a reply to specific IFS school per pupil funding claims, with no attempt to flag the in this quote ‘the UK’ doesn’t mean the UK government, but the whole UK economy and that ‘education’ doesn’t mean state schools but all kinds of education institution including universities and private schools. Your point that this is politics as normal is a little pessimistic. While of course the government can frame the debate in it’s own way – it seems reasonable to expect them present statistics so that they mean what they appear to mean.

  2. albert wright says:

    Thank you for this interesting article.

    The figures say £ per student.

    1 Are the annual £ figures based on historical spending or have they been adjusted for inflation?

    2 In the Higher Education figures, are the numbers based on the supply side and based on what Universities have received by way of income from all sources or just from the Government?

    In the debate on schools you point out that some figures include spending in private education.

    3 In the Higher Education and Further Education figures, does this include all spending on all skills training or just that which is publicly funded?
    Employers fund a mass of skills training, leadership and management training, personal development training, MBAs etc in UK and abroad- is any of this included and if figures are “per student” how do we know the number of “students” involved?

    4 How are employer contributions handled when measuring expenditure on Apprenticeships?

    5 Is spending on Apprenticeship Degrees (where most of the money comes from employers) included in the HE figures?
    Are adjustments made where Universities cross subsidise the cost of delivering Degree Apprenticeships from other income streams?

    6 Where is the funding for Adult Education shown and is student contribution included?

    7 Do the figures include part time students or only full time?

    8 Is expenditure on ESOL included?

    9 Is expenditure funded by the Department of Work and Pensions, ESF and ERDF on training and development for the unemployed and disadvantaged included?

  3. Johnny Rich says:

    Under the same logic, the Government could stop all public sector house building and force the sale of council houses to the private sector while forcing down interest rates. As house prices and rents soar, the Government could face down the criticism about rising homelessness by pointing out the nation is spending more on housing than ever before.

    Politicians trying to ‘make the weather’ is not the problem. Nor are the OECD stats. The problem is using those statistics – deliberately – out of context to face down a criticism about something completely different, namely public spending in schools. Damian Hinds has previous form: I understand this is the fourth time in the past year, his use of statistics has been criticised by ONS. Once might be an error. Twice might be an attempt to change the whether. Three times is deliberating misdirection. But four times is outright Trumpery – hence the justified outrage.

    1. I don’t accept the housing comparison at all, but let’s chat about it in person. In the meantime, we had an interesting further discussion on the education statistics on the Wonkhe podcast, which is out tomorrow…

  4. Ben Verinder says:

    Nick – a rare disagreement from me.
    1. This isn’t a one off but a trend.
    2. Departmental statements/releases carry more weight than a statement at PMQs. They shouldn’t, but they do.
    3. The problem commentators have isn’t with the OECD methodology, but the selection by the DfE of the OECD figures in response to criticisms of Government funding levels.
    4. I think you’re right about the 2010 figures though – that is more important.


Leave a Reply

Your email address will not be published. Required fields are marked *