100 new free schools a year: achievable or not?

Will Theresa May’s target of “building” 100 free schools annually for the rest of this Parliament be achieved?

The short answer, looking at the available statistics so far, is that it is likely to be a very challenging target. In addition, the setting of it must raise the obvious question: will quantity be put above quality in the acceptance of free schools projects?

The promise(s)

The 100-frees-a-year target originated in the Conservatives’ ill-fated 2017 general election manifesto, the party pledging there that it would “continue with our programme of free schools, building at least a hundred new free schools a year”. With the manifesto applying for the length of the Parliament, this would take us to 2022.

Then, last month in her also-ill-fated-but-arguably-for-different-reasons party conference speech, May re-iterated the pledge, saying her government would be “building 100 new free schools in every year of this Parliament. Not because our ideology says so… but because free schools work [Hmm…sadly, that’s a bit sweeping for me, as the statistics below may demonstrate…]. And it’s the right thing to do.”

It’s possible to debate in what way the commitment “building 100 new free schools in every year of this Parliament” might differ, in terms of a pledge, from “building at least a hundred new free schools a year”. However, the thrust of the manifesto commitment was clearly retained in the speech.

The data so far

But what does the history of the free schools policy tell us about the achievability of this target?

Well, if we use the past as any guide to the future, it’s going to be very challenging. In the seven academic years in which free schools have been opened since the first ones did so in September 2011, in only two cases has the 100-plus schools opening figure been achieved. Since 2013-14, the numbers opened have actually been falling year-on-year. And even the two years when the 100-mark was reached, these figures take the broadest possible definitions of “free schools”, including the troubled studio schools and university technical colleges categories, which the government has sometimes seemed reluctant to include in its statistics for the policy. http://bit.ly/2woCuRs

So, as you can see, the graph, based on official DfE data, shows the number of frees opening rising annually in the first three years, hitting a peak of 119 in 2013-14 and then falling each year since. So far in 2017-18, only 54 have opened. (Yes, it’s true we’re only in the second month of this academic year, but previous years’ data show that the overwhelming majority of frees have opened, logically enough though sometimes the policy may not feel very logical, at the start of the academic year in September. So that number of 54 is likely to be close to the total for this year).

And, as mentioned, these figures actually incorporate the sub-categories of studio schools and university technical colleges. (The DfE data includes these two sub-categories in a broader categorisation called “free schools”, alongside a sub-category also called “free schools”. Are you keeping up?) If we concentrate on the narrower/sub-category definition of “free schools”, the annual totals so far are lower again: only in three years of the seven have numbers opened in the narrower category of “free schools” even topped 50.

Looking to the future

The distinctions between the different categories of free schools become important when we look at what data there is on new projects already in what the DfE describes as the “pipeline”.

 DfE data show a total of 302 free school projects as in development. That looks to be quite a healthy figure in terms of hitting the Prime Minister’s target, given that perhaps we are now only four years away from the end of the Parliament in terms of complete academic years in which free schools could be built.

However, there is only one university technical college on this pipeline list, and no studio schools at all. This suggests that the DfE is going to have to open far more conventional free schools annually than has been the case so far. Can it do it? Well, I have to hold my hands up and say I haven’t done any analysis as to how long it takes pipeline projects to see reality, and how many do not, so we are probably in the realms of educated guesswork. But, to repeat, these targets look challenging.

Free schools which have closed: the data

A final – or near-final - aspect of this analysis is to consider the attrition rate of these schools. For, although they have been opened in the numbers set out above, quite a high proportion of them have been closing, too. Indeed, among those which were opened in the first four years of the policy, the attrition rate up to now (seven years in) is running at 11 per cent overall, or one in nine of the 323 schools which were opened in the period 2011-2014/15 having closed already.

That is quite a failure rate. And it has been driven largely by the closures of studio schools – where a staggering 44 per cent of those opened in these first four years of the policy have since closed – and UTCs, where the comparable closure rate is 24 per cent. (The rate for the more narrowly-defined free schools set up over this period is a still-not-low 4.7 per cent, although a couple of these “closed” schools have actually simply relaunched as new schools, it seems from the DfE spreadsheets). These statistics perhaps demonstrate why the number of studios and UTCs in the “pipeline” is so small.

The closure rate among those institituions – free schools of all types - which have opened since 2014/15 has plummeted dramatically, with only a single example of closure – and that a technical one – among the 183 openers. On one level, this is unsurprising: these institutions are only just over two years old at most, and it would be extraordinary to see any school closing after a year – to think of the 2016 openers – and unthinkable – almost-unthinkable- in relation to the 54 which only opened last month. http://bit.ly/2yxxMpw

Sometimes, however, advocates of the free schools policy have suggested that quality control has improved in recent years. This might suggest that this sudden dip in closures could prove to be the start of a longer-lasting trend. Let’s hope so, for the children attending them. Perhaps the fall in closure numbers in recent years adds to this sense of allowing only the most solid proposals to go through, although it would be best to reserve judgement about that, I think, given the history of the policy.

But…(you may be able to see where this is going…) will the DfE have to relax on quality as it seeks to drive up quantity? We don’t know, and developments on the free schools scene such as the advent of the site acquisition service LocatEd and the willingness to…whisper it…allow greater involvement of local authorities in the planning of free schools may allow the department some grounds for hope that projects can be developed at greater scale without a greater risk of failure.

But the history of free schools, and specifically the failure rates and some of the more chaotic stories which this website is going to chronicle, suggest grounds for great caution.

-Postscript: all of the above assumes that when ministers say the government will “build 100 free schools per year” they actually mean that they will open 100 frees a year. Actually building 100 schools, rather than opening some in old buildings, might be more ambitious still than just having new schools open. There may be further cost implications of actually doing the building, and cost implications more generally in relation to the 2017-22 free schools targets, which we may explore in a future post.