Free Schools? Emperor’s New Clothes!

Just because we are Swedish, it doesn't mean we have all the answers

One of the Coalition’s flagship education policies is free schools. Free schools are free from local authority control and derive their funding direct from the Department for Education. They can be set up by any group including parents, as long as they meet the approval of the Secretary of State.  They have been billed as being set up in response to ‘what local people say they want and need in order to improve education for the children in their community’.  They are therefore seen as the epitome of localism, the very essence of local people being empowered to bring about positive change by wresting control from the clammy hand of the local authority and creating schools which really make a difference.  Michael Gove has championed this policy, and despite desultory numbers of applications, still pushes ahead with this strategy.

Gove cites two countries, the US and Sweden, as influencing his Free Schools policy.  The US have charter schools and Sweden has free schools which are run by the private sector for profit. This article focuses on the Swedish free schools.

I was lucky enough to spend some time with Susanne Wiborg from the Institute of Education recently, who is an expert in comparative and international education and has written a report on the Swedish free school model.  Our conversation, and my subsequent reading of some of her work illustrated some very telling points about how the debate (or rather lack of it), has unfolded in this country, and I would like to share some of this here.

I tend to believe people when they talk to me, even politicians, even Michael Gove.  If someone tells me something is good, I tend to believe them because why would they lie to me? So people could tell me Ikea does great stuff for your house, particularly if you are on a budget, and I get down to the store and find this to be true (and knock myself out with a plate of meatballs to boot).  So when Gove started lauding the benefits of the free schools in Sweden I just assumed they were a great success and they had a real coverage and impact, a little like Ikea. The lazy part of my brain ran a nice little riff which went:

‘the Swedes are logical and good at design, they are unsentimental but also have a kind of flinty soul which redeems them, they can take something boring like buying stuff for your house and turn it into a great day out for the family with cheap food; when they turn that kind of intelligence to education they are bound to do something equally transformative, the Tories must have heard of this, fully researched it, and brought the policy back to England for the benefit of all’

It turns out the reality is a great deal different. Firstly Swedish Free Schools are a lot rarer than I had thought. Here are the facts:

In 1991, there were a little over 60 non-public schools in the country and, by 2009/2010, their numbers had reached 709 […] The number of pupils in free schools has increased from 20,247 pupils in 1995/96 to 95,948 pupils in 2009/10.

So there are free schools in Sweden, but they hardly outnumber the public schools, and Susanne explained that when Free Schools were first being brought up as a policy she was surprised as she did not see this as a particularly important feature of the Swedish education system.

Secondly I believed that the Swedish public were in favour of free schools, surely this wonderful new way of providing education was going to capture the hearts and minds of the public and parents. Wrong again! The report explains that free schools have met ‘profound criticism’ and the public seem to have been most exercised by the idea of shares, money and profit rather than any innovative forms of education the schools are providing.  Now the UK model has taken profit out of the system, but the report found little or nothing to suggest that these schools, once free of central or local control, actually did anything meaningful with that freedom.  Free sounds such a good word, who wouldn’t want to be ‘free’, but without substantial and sustained educational innovation, this freedom means nothing.  The Free schools didn’t use their freedom to reinvent education in the same way as Ikea reinvented furniture, they simply set up another beige version of MFI, eroded the workers’ pay and conditions, and were constantly looking over their shoulder at the balance sheet to make sure they made a profit rather than doing anything radical to improve the way they delivered education.

My final belief, soon to go the way of the other two as the hard facts loomed into picture, was that the free schools must have better results and achievement than the public schools. Surely this is where the emperor’s real clothes would shine.  After all, Michael Gove never gives any speech without waving PISA data around like a talisman and berating the English system for falling so far behind international standards. Surely the Swedish free schools, whatever else they were doing, showed much much better results than their public school rivals.  The reality is not the case, based on summaries of Swedish research comparing free schools and public schools. Some studies found no differences at all,  one study did find a short term and small improvement in results in free schools, but:

…the short-term effect is too small to yield any long-term positive effects for young people. In other words, the advantage that children schooled in areas with free schools have by the age of 16 is not translated into greater achievements later in life as they score no better in the final exams in upper secondary education at the age of 18/19. They are also no more likely to participate in higher education than those who were schooled in areas without free schools. The children from highly educated families gain mostly from education in free schools, but the impact on families and immigrants who had received a low level of education is hardly visible. (Wiborg, 2010, p. 14)

“Damn!”, I muttered, wrong 3 times; made a fool of 3 times by my belief that Gove had done his homework (in Latin) and the Swedish Free schools were a) widespread b) popular c) had great results.  Bear in mind the final part of the quotation above which shows there is no visible impact for children of parents with low levels of education or immigrants, and then read these words from a Michael Gove Speech:

In every school year there are 600,000 children. The very poorest are those eligible for free school meals – 80,000 in every year. And out of those 80,000 how many do you think make it to the best universities? Just 45.  More children from one public school – Westminster – make it to the top universities than the entire population of poor boys and girls on this benefit. (Gove, 2010)

I end this post with one simple conclusion. This is an emperor without any clothes, an emperor so naked he would shiver as the cold wind blew from the Thames and tickled his nether regions as he paraded on Parliament Square. Even if we implemented the Swedish free school system on the kind of scale needed to make a difference (say around 60,000 students in free schools in every school year, 10% of the total), Gove’s aim of increasing the intake of Free School Meals students to the top universities would not be met. The importing of the free schools idea from Sweden clearly has nothing to do with proven success in raising educational outcomes and very few commentators to date have picked up on this simple but crucial fact.

The English free school policy is pure ideology on Gove’s part driven by a desire to break the teaching unions and national pay and standards for teachers.  Gove has by now realised that free schools are going to be so rare that they won’t help him achieve this goal. Hence the recent shift to accelerate the academisation of English education.


Gove, M. (2010) Speech to the Conservative Party Conference in Birmingham, 5th October 2010. [online] Available at: Date accessed: 17th June 2011.

Wiborg, S (2010) Swedish Free Schools: Do they work? published by the Centre for Learning and Life Chances in Knowledge Economies and Societies at:

Author: mjp6034

Education consultant specialising in educational technology and change management.

8 thoughts on “Free Schools? Emperor’s New Clothes!”

  1. I couldn’t agree more.

    The schools are the emperor and Gove’s the conniving suit maker. Does anyone really believe that the private sector has a better understanding of how to run schools than an LEA? Private education does, of course, have excellent benefits for those that can afford it, but this is largely due to increased funding, reduced class sizes and pupils from a fairly homogeneous background. I don’t think that academies will be afforded these luxuries.

    There’s a reason that it was announced yesterday that the lowest performing schools will be forced to become academies. National pay and standards aside, the government, in national or local form, is looking to absolve responsibility for poor performers. They can’t lose; if the school performs better, than Gove was right, if it performs less well, then it’s being badly run, but hey, the government didn’t choose the staff / curriculum / budgets.

    My kids attend a faith school, and the local diocese has jumped at the opportunity to take full advantage of self-management and curriculum setting, and it worries the hell out of me. I’m not at all convinced that I want a board of church-appointed directors deciding on what it’s right to teach my kids or what the school’s funding should be spent on.

  2. It’s really interesting to read some more facts on this. I have to admit that I reacted to the announcement about free schools in the opposite way to you: I immediately thought it sounded like a very odd premise and I couldn’t quite get my head around the idea. Something about any group of people being being able to set up a school according to their own rules makes me feel uneasy. I read an article in a local newspaper about plans to set one up in my LA and I still couldn’t really get my head around it or why this model would be a good idea. Your article has just confirmed that more in my mind. Thanks for sharing.

  3. “The English free school policy is pure ideology on Gove’s part driven by a desire to break the teaching unions and national pay and standards for teachers.” It seems to me this notion is the work of conservative governments in more than UK. Our previous government in Australia seemed to take a similar line. The Federal Govt greatly increased funding to private schools with the excuse the state govts were funding public schools so there was a need for equity. While there were funding problems in some private schools, this policy led to some obvious discrepancies. Examples were wealthy private schools receiving $1million+ in federal funding while a public school “down the road” was having difficulties getting the funds to repair the roof. With the extra funding available, more private schools were encouraged to establish.
    That federal govt seemed obsessed with breaking the unions, especially strong teacher unions such the Teachers Federation. They also made membership of student unions in universities voluntary without listening to those telling them student unions were set up to provide welfare and other services to students. They were not trade unions but had the word “union” in their title so needed to be controlled.
    We now have a Labor government in power. Any change? Not really. They seem to be more right than before and promised to maintain funding levels as they are.
    From what I have read, I believe your concern is correct. It’s more about breaking “the teaching unions and national pay and standards for teachers” than any attempt to significantly influence standards. We will see in time.

  4. Matt,

    I welcome anyone to this debate who’s prepared to base their opinion about academies and free schools on research evidence rather than ideological bias, but you need to broaden your reading.

    In Sweden, no less than five separate studies – Bergstrom and Sandstrom, Tegle, Bjorklund, Ahlin and Bohlmark and Lindahl – have found that examination results are better in free schools than municipal schools and that municipal schools in municipalities that contain an above average number of free schools obtain above average exam results. The schools operated by Sweden’s most ubiquitous for-profit free school companies – IES and Kunskapsskolan – consistently get the best test scores throughout the country. And it’s false to claim that the introduction of free schools in Sweden in the early 90s has led to increased social segregation in Sweden’s public school system. The 2009 PISA survey included a table ranking all the school systems in the developed world according to how socially segregated they were, with number one being the most segregated. Sweden was second from bottom, just above Norway.

    The same picture – education reform leads to higher attainment – is emerging in England with regard to academies. I’d refer you to a recent piece of research by Stephen Machin and James Vernoit at the Centre for the Economics of Education at the LSE. After studying all the secondary schools that converted to academies under the last government, as well as their neighbouring schools, they reached the following conclusion: “Our results suggest that moving to a more autonomous school structure through academy conversion generates a … significant improvement in pupil performance and … significant improvements in the performance of pupils enrolled in neighbouring schools.”

    Machin and Vernoit also note that three similar studies examining the impact of charter schools on neighbouring schools in Michigan, Arizona and Texas that have come to the same conclusion. Here’s their summary of that research: “The question of whether a charter school has an impact on the performance of their neighbouring schools has only been addressed by a few studies. Examples of this kind include: Bettinger (2005), which looks at the impact of charter schools in Michigan on neighbouring public schools, Hoxby (2002), which evaluates the impact of charter schools in Michigan and Arizona on their neighbouring regular public schools, and Booker et al. (2007), which looks at the impact of a high concentration of charter schools in Texas on the student achievement in traditional public schools. All three studies find improvements in the traditional public schools that can be attributed to the introduction of charter schools.”

  5. You are right, I did need to broaden my reading, and it’s always welcome to receive these kind of exhortations. In my defence I will say that I read the report by Susanne Wiborg on which the blog posting was based in full. And to cut to the chase, I would that I still feel that her analysis of the impact of Swedish free schools is closer to reality than other representations I have come across, including your own.

    Here I will only look at the evidence surrounding Swedish free schools, as the blog posting made it clear that it was looking only at the Swedish case. But I have gone away and hunted out the sources which you indicated in your comment. You didn’t provide a reference list using an accepted methodology for referencing your sources which would have been helpful, but I have done my best to track down the sources which you indicate.

    Let us look at them one by one.

    Bergstrom and Sandstrom
    I think you mean this paper titled ‘school vouchers in practice: competition won’t hurt you’

    This was an interesting read and in terms of statistical wizardry I think nobody would doubt the efficacy of the authors. My only worry is that their statistics are so in depth they are in grave danger of disappearing up their own chi square. Here is a flavour of what they wrote

    We can formulate an equation explaining the share of independent schools in a municipality in the following way:
    Sj =Rjα+ξj (2)
    where Sj is the share of students attending independent schools in municipality j, Rj is a vector of explanatory variables, α a vector of parameters to be estimated, and ξj is an error term. We use j as an index instead of i to emphasize that, unlike in equation (1), the unit of observation is a municipality and not an individual.

    I urge anyone interested in this debate to read the full paper, and at the end of their reading ask themselves if they understand more or less about whether free schools improve results compared to public schools in Sweden. If you could provide an explanation, for the layman not versed in the niceties of statistics about why these findings are significant, I would be much obliged. In the absence of this, I will have to consider this paper to be of limited use simply because its procedures and analysis are so obscure as to defy comprehension.

    I think the Tegle source you refer to is at
    Tegle, S. (2010) ‘Påverkar förekomst av friskolor betygen i grundskolan? – En statistisk analys av samtliga elever i årskurs 9 år 2006’, Svenskt Näringsliv, Stockholm,retrieved 20 June 2010 (http:// skolans-result_108196.html).

    This source is in Swedish and I could not find a translation. I’d be happy to read the paper if there is a translation available or if you could indicate in a little more detail how this contributes to the debate about whether Swedish free schools raise standards.

    This was the easiest to track down, if indeed I have located the correct source. I believe this to be the book at this address

    I was not able to acquire this book to read before posting this reply, but I did read the ‘product review’ which stated: “The authors find that increased competition did help improve students’ math and language skills, but only slightly, and with no effect on the performance of foreign-born students and those with low-educated parents. ”

    This appears to rather back up my original blog posting, namely free schools appear to improve results slightly, but for foreign and low-educated parents there is no improvement. Of course the product review may be inaccurate, after all, one should not believe everything you read online!, so if the book itself in your possession tells a different story and the product review misrepresents it, then let us know.

    Ahlin and Bohlmark
    I assumed this was the 2008 study for IZA available here

    Once again I have to say on reading the original (rather than Wiborg’s summary paper), I see little to make me change my mind. Here is an extract from the abstract:

    “We find that an increase in the private school share moderately improves short-term educational outcomes such as 9th- grade GPA and the fraction of students who choose an academic high school track. However, we do not find any impact on medium or long-term educational outcomes such as high school GPA, university attainment or years of schooling. We conclude that the first-order short-term effect is too small to yield lasting positive effects.”

    Pretty much what I said in the posting. Of course if you were talking about another paper by Ahlin and Bohlmark, then please provide the full reference, or failing that some kind of clue, as to what it was and where to get it. I am happy to go away and evaluate that source if it exists.

    Nearly done.

    This was problematic as there appears to be nobody called Lindahl writing about free schools or Swedish education reform. There is an academic called Lisbeth Lundahl who works in the department of Applied Educational Science at Umea University. Was it one of her works you meant to reference? If so, then I have read 3 of these papers (I can provide a list if needed, I omitted it here for brevity’s sake) and so far I find her work is largely to do with the organisation of the state reforms to Swedish education and the establishment of the free market. I could find nothing in the material I read which indicated that Free Schools were significantly raising the standard of education in Sweden. Of course I could have simply been reading the wrong papers, so as before, please provide a specific reference to a work and I will be happy to read it and take it into account.

  6. Matt thanks for such a rigorous response to Toby’s Young’s exhortation. I’d also add Toby in response to the recent LSE paper on academies (Machin and Vernoit 2011) the conclusions they are fairly circumspect using words like ‘relatively positive’ to describe their findings. Also they talk about an improvement in the ‘quality’ of Y7 intakes after conversion to academy which is one reason I guess they qualify their fairly cautious findings with words such as ‘relative.’ They also talk about the need for further long term research into the effect of academy conversions as more schools are encouraged down this route. This is completely uncharted territory as this study only looked at the educational climate under the new labour with relatively few academies compared to the high number of conversions we are now witnessing. It seems highly unlikely to be the last word on the issue Toby. Perhaps when we’ve had 5-10 years of large numbers of autonomous schools being established then we might be in a position to make some kind of research informed judgement on the effect such autonomy has had on the educational landscape for all. The readings you suggest to broaden one’s view Toby would not constitute anything near a meta-review of the literature on autonomous schools/academies/free schools. Your conclusions are somewhat premature.

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