黑洞社区

Headteachers facing agonising decisions about where to make the cuts needed to accommodate the government鈥檚 new school funding formula may have emitted an expletive or two when they read February鈥檚 National Audit Office report on school building. Because as they deal with potential staff cuts and pupils exposed to asbestos in dilapidated schools, the NAO鈥檚 analysis of capital spending showed the government apparently lavishing money on its flagship 拢10bn free schools programme, including on four occasions splashing out more than 拢30m for individual school sites.

鈥淲here鈥檚 the value for the taxpayer?鈥� says Meg Hillier MP, the Labour chair of the influential Public Accounts Committee, which has been analysing the report. 鈥淭he NAO figures speak for themselves.鈥�

At face value, they appear pretty damning. The NAO report found that primary places at free schools cost 33% more than those created by local authorities, and secondary places 51% more 鈥� because of the high land cost. With planned 2015-2020 free school spending of 拢8bn (of a total 拢10bn for the whole free school programme from its start in 2011), that means that 鈥� on paper, at least 鈥� routing the money to local authorities instead could have seen the planned 270,000 places delivered for a saving of up to 拢1.7bn. That鈥檚 enough to pay for another 124 free schools at the programme鈥檚 current average school cost, or another 138,000 local authority school places.

Real life, however, is not always as cut and dried as statistics such as these imply. So, with the Department for Education (DfE) insisting it is simply working to deliver the Conservatives鈥� manifesto commitment for 500 free schools in the most efficient way, is it really true that this much money is being wasted? If so, why, and what more could the department and its school building quango, the Education Funding Agency (EFA), renamed the Education and Skills Funding Agency last week, do to reduce costs?

Ideology over value

Certainly, opposition MPs believe the answer is pretty straightforward. Free schools were introduced by former education secretary Michael Gove in 2010 as a parent-led way of bringing additional choice into the education system, with an eye to disrupting the local authority education establishment he privately referred to as 鈥渢he Blob鈥�. However, the idea evolved quickly from the first experiments in cheap and cheerfully constructed, locally-led schools to become mainstream, with many schools backed by existing academy chains. This ramping-up led to the Conservative Party鈥檚 2015 manifesto pledge for 500 more 鈥� a target that has since gone up by a further 110 new schools, funding for which was indicated in the spring Budget for the next parliament. Now, despite the need to provide an estimated 420,000 new pupil places between 2016 and 2021, the only type of additional new school allowed is a free school (the Priority School 黑洞社区 Programme is only to replace existing schools in poor condition).

Opposition politicians say this is an ideological choice, not a pragmatic one 鈥� a criticism given weight by the NAO鈥檚 findings on their high cost. The report found that not only were free schools more expensive across the programme, they were also getting more and more costly as time went on 鈥� with the most recent secondary schools costing 拢24,600 per place, 76% more than the first wave in 2010. The department spent an average of nearly 拢5m per site for free schools between 2011 and 2016, which is 19% above official RICS 鈥渞ed book鈥� valuations. Twenty sites cost more than 60% above red book value. Labour鈥檚 shadow education secretary, Angela Rayner, said in response to the NAO report that the government was 鈥渨asting millions鈥� on 鈥渢he Tories鈥� pet projects鈥� while failing to deal with major defects in the fabric of state schools. The Public Accounts Committee鈥檚 Hillier tells 黑洞社区 that: 鈥淚f you have a big numerical target in a manifesto, which has been plucked out of the air, then you鈥檙e going to get perverse outcomes in order to meet it. It鈥檚 not the right way to go about governing.鈥�

In evidence to the Public Accounts Committee, teaching unions and local authority groups have pleaded with the government to give them power to commission new schools too, in order to meet the looming squeeze on places. The Association of Teachers and Lecturers said the free schools budget was 鈥渟piralling out of control鈥� and that as local authorities have a statutory duty to provide enough school places 鈥渢hey must also be given the appropriate powers, resources and support from the DfE.鈥� With a fifth of free schools created in places where there is, in fact, no shortage of places, the Association of Teachers and Lecturers also called on the DfE to 鈥減roperly analyse if the contested value of opening free schools outweighs the evident disadvantages of creating spare school places.鈥�

Hillier cites the example of the Olive School from her own Hackney constituency, on which the EFA paid twice book value for a site on which planning permission for a school has now been turned down, leaving students in temporary accommodation for two years (see below, Lesson one: Get the site right). 鈥淚 don鈥檛 think the EFA has the necessary skills to buy land,鈥� she says.

Efficient delivery?

The size of the free schools programme underlines the importance of this buying capability, with the EFA expected to spend up to 拢2.5bn on land between 2016 and 2022, according to the NAO 鈥� the equivalent of a top five UK housebuilder. Defending the department鈥檚 delivery of the programme, DfE permanent secretary Jonathan Slater told the Public Accounts Committee that it was managing the meeting of the free school manifesto commitment in the most efficient way possible.

There is some justification for this statement. Firstly, the construction costs of free schools are the same as under the Priority School 黑洞社区 Programme, and cheaper than those built by local authorities. And though Slater admitted some free school sites have been expensive to buy, this overlooks the fact more than half have been provided by other government agencies on peppercorn rents. All those where prices paid were significantly over book value were checked and approved as value for money by the Treasury, he said.

If you have a big numerical target in a manifesto, you鈥檙e going to get perverse outcomes in order to meet it

Meg Hillier MP

Mike Green, capital director at the EFA, argues that in a rising market where there is tough competition for development land, it is understandable that many sites are achieving higher than a red book valuation, which is based on historical comparisons. He points to the recent creation of specific property arm LocatED as evidence it is taking action to ensure the very best value for land. 鈥淚 genuinely believe we buy sites well in a competitive market. We walk away from deals where the prices are too high. One I can think of we walked away at 拢30m 鈥� it eventually went for 拢56m. The whole point of LocatED is to ensure we have the skills and capacity to buy sites 鈥� that鈥檚 why we did it,鈥� he says.

In the industry outside the Westminster bubble, Green鈥檚 position certainly has some credibility. One consultant working in the schools sector, says: 鈥淪ite buying is problematic for the EFA because often they need a site and the seller knows they need it, but I don鈥檛 think the EFA are seen as particularly bad negotiators.鈥� Likewise, Jonathan Stewart, board director at consultant Gleeds, says: 鈥淪urprise, surprise: often where there is a shortage of places is also where there is high demand for housing, shops and offices. When buying a site, you鈥檙e going to have to compete against those uses.

鈥淏ut there鈥檚 no sense the EFA is being taken advantage of. It鈥檚 been using professional agents 鈥� some of the best in the market.鈥�

Nevertheless, even if the price paid for land is fair, it doesn鈥檛 mean it鈥檚 value for the taxpayer if, were local authorities instead allowed to build, they could use their surplus sites to avoid going to the land market at all. Yet it is unclear whether this would be the case. The NAO figures that show local authorities providing places much more cheaply are based on councils鈥� experience of adding additional places to existing schools 鈥� a very different prospect to finding a site and creating a whole new school from scratch. While historically local authorities have been able to build schools using existing land holdings, it is not at all clear that they would be able to meet the level of demand 鈥� 500 schools 鈥� without going into the land market themselves and thus encountering the same problems as the EFA.

Hence free school advisory charity, the New Schools Network鈥檚 recently appointed director Toby Young says the two figures cannot be fairly compared. 鈥淚鈥檓 not willing to accept the premise that local authorities can create new schools more cheaply. The free schools programme is the most inexpensive way that successive governments have come up with to build new schools.鈥� In addition, Young says the cost argument ignores all the other potential benefits of free schools, which he says are more likely to be Ofsted 鈥渙utstanding鈥�-rated than grant maintained schools, and are 鈥� at this early stage 鈥� outperforming other school types academically at key stages 1 and 5.

More to do

Unsurprisingly, Young鈥檚 point on costs is contested, with others maintaining that if local authorities went into the land market they could get hold of sites more cheaply. The schools consultant says: 鈥淟ocal authorities have land, they control how it is designated and they can do land swaps. They have much more influence in a locality than the EFA, and developers need to have an ongoing relationship with them. Hence I鈥檓 sure the EFA is paying more than they would.鈥� But even if this is true, the government can argue that the system is now set up to get the benefit of this whenever possible 鈥� with councils in areas where there is a need for places encouraged to work hand-in-hand with free school promoters to bring sites forward. Young argues that contributions from developers via Section 106 deals can also play a significant role.

Nevertheless, some say the EFA could still be more innovative in bringing costs down. Part of the rationale for LocatED was the possibility of buying school sites that included space for homes and other uses that might make the public sector some money. But consultants say there has been little evidence of this so far. Marcus Fagent, education sector lead at consultant Arcadis, says: 鈥淪ites in congested cities are getting expensive and the EFA should look to mitigate these costs by driving value from promoting mixed educational/commercial developments. These need not delay delivery of the new school 鈥� the EFA will have the option to develop out the commercial content later.鈥�

I genuinely believe we buy sites well in a competitive market. we walk away from deals where the prices are too high

Mike Green, Education Funding Agency

The EFA鈥檚 Green denies this objective has been abandoned, but admits it will not be a significant part of the programme. 鈥淥ur prime role is to find sites for schools. It鈥檚 the right thing to do to help fund a scheme with other uses and it can allow us to use sites that otherwise wouldn鈥檛 be suitable, but it鈥檚 only ever going to be 10 or 20 sites 鈥� it鈥檚 not the purpose of what we鈥檙e doing.鈥�

In addition, it is not clear that the government鈥檚 different departments and agencies are aiding delivery of the programme by providing land where necessary. This is highlighted by the example of the Olive School, where the previous public sector owner of the site managed to extract a high price from the EFA 鈥� another public body. Green says this issue is a thorny one. 鈥淚t鈥檚 a complicated question. We鈥檇 hope there鈥檇 be a reasonableness in negotiations. But if we felt another body was being stupid over price, we鈥檇 walk away. But in situations where we need schools, by and large councils are co-operating.鈥�

Ultimately, the decision to press ahead with the construction of 500 free schools is a political one. Giving local authorities the responsibility to ensure the need for places are met without the ability to commission new schools themselves clearly creates significant problems. But the question of how much money this directly wastes remains a point of fierce debate.