Recent projections by the Local Government Association suggest that the population bulge that has hit schools at the Primary level is now reaching Secondary. It is now estimated that the country may need over 500,000 additional Secondary places by 2024. Meeting this basic need is becoming one of the most preoccupying challenges facing school commissioners and local authorities alike. Meanwhile, the rate of free school starts appears to be slowing and academy sponsors appear put off by mixed messages from the government and its agencies about growth and the lack of incentive to do so. Against this background, it’s easy to lose sight of why new school start-ups might be important in and of themselves, quite apart from their usefulness to meeting basic need.
On the evening of 25th May members of CMRE’s executives’ policy forum and a number of guests from the world of education gathered to discuss strategies for leveraging new supply and the importance of introducing new ‘challenger schools’ to the local schools landscape.
The evening began with a welcome from CMRE’s Executive Director and Forum Chair, James Croft, followed by a brief presentation from Nick Timothy, Director of the New Schools Network. An informal discussion followed, under Chatham House rules, which this post attempts to summarise.
The opening
Nick Timothy opened by stating that the interrelated questions of why the government should be opening new schools when they are not – in inverted commas – “needed”, when they demonstrably do not consistently relate to basic need, and when the opening of these schools often makes the viability of existing schools even more doubtful, go the nub of the rationale for the free schools programme.
Breaking down the numbers, Nick acknowledged the basic need case for new schools. A rising school age population means we are projected to need more than 130,000 new school places by the end of the decade (60,000 for primary schools and 70,000 for secondary schools). But, he said, there is little point in making sure that there are enough school places overall if a significant number of those places are not good enough. According to Ofsted, 620,000 children go to primary schools and 650,000 children go to secondary schools that are inadequate or require improvement. And that ignores the fact that there are many schools that are below the national averages when it comes to attainment and progress, but which are nevertheless judged to be ‘good’ or even ‘outstanding’ by Ofsted.
The free schools policy was never intended to be a basic need programme. But with the school age population projected to grow at such a fast rate, and the fiscal squeeze applying not just to the Department for Education but to local authorities too, it was inevitable that the programme would become dominated by this agenda.
Nevertheless, the original logic of the free schools policy was that new schools should be set up in the communities served by failing schools: this would improve standards, give parents more choice, and allow new schools to innovate. While this has been eclipsed to some degree by more pressing political considerations, the government’s ambition to open 500 new schools over the course of this parliament indicates that the free school policy isn’t just about meeting basic need, as some now claim. According to DfE analysis, basic need alone does not merit 500 new schools.*
The question, Nick maintained, is whether the free school process as designed can effectively accommodate both aims. Put more directly, couldn’t basic need be addressed more directly?
The effects of using the free school process for this purpose are clear. Are free schools really free schools if many of them are new schools opened by the big multi-academy trusts or by existing schools? Nick professed to having no problem with MATs using the free school process to set up schools that do something different – perhaps with an external partner to give the new school a unique identity – or as a route to reaching deprived communities and those served by bad schools with a proposition clearly focused on improving outcomes – but very often its just basic capability to run schools that authorities are after.
This being the case it might behove the government to transition to a two-track framework for new school starts. Nick spoke of his experience that many local authorities have been suspicious of the free schools programme and have often sought to obstruct their opening – using the planning system to impose unhelpful conditions on start-ups; denying the need for them to both Regional Schools Commissioners and the Department; engineering the outcomes to get a provider with whom they’re comfortable; or trying to force new schools to sign up to deals that restrict the school’s freedom. In hindsight, clarity about the fundamental nature of each proposal, the main priority, might have served to diffuse some of this opposition. There’s a case to be made now, he maintained, for developing two separate programmes: a basic need new schools programme, on which the school commissioner’s office and the Education Funding Agency would work closely with local authorities to identify the local need and find a suitable provider to meet it; and a free schools programme, through which the DfE would allow new schools to open in areas or poor provision or where parents have very little choice.
Nick then went on to discuss the free school apparatus, its fitness for the purpose for which the programme was designed, and supporting market reforms that are still needed. The first issue he raised was the very limited amount of information available to people and the poor reliability of the information that does exist. Given that the government is effectively trying to create a market in the schools system, which depends on effective information provision, parents know very little about what they are choosing. They rely on word of mouth, reputation, and what Ofsted says. Given that the validity and reliability of Ofsted ratings has been so much a subject of debate, this doesn’t leave parents in an especially good situation.
To address this issue, Nick suggested tthe creation of a street-level ‘school maps’ website, which would include a mixture of qualitative and quantitative information about schools’ performance, approach and ethos. Importantly, it should help people not just to compare local schools but also similar schools in other parts of their city, county, or indeed the country. If you want parents to be a force in agitating for change and improving standards, they need to know not just what they’re getting but what they’re not getting too.
But, Nick pointed out, it isn’t just parents of course who lack information. If you’re a potential school provider – whether a first-time community group or a tried-and-tested school or trust – you don’t know where there is basic need or demand. First, because to be fair, it is very difficult to estimate; second, because the government relies on local authority estimates; and third, because even the estimates that do exist are not published in a transparent manner. But it’s not just that providers do not know where the basic need exists: neither do they know where there is a need caused by poor school standards. These things should be published openly and online, he maintained – both the original data and navigable maps. There also needs to be an attempt made to publish information about land ownership and use, he said, so identifying sites for new schools can be a straightforward and efficient process – rather than free school proposers cycling around town looking for disused buildings.
Nick also felt that it was time to change the criteria that a free school proposer group needs to meet. At the moment, an application needs to prove basic need – which, as he had already demonstrated, can be difficult to identify – or educational need. In practice, this means you cannot propose more new school places than already exist in local schools that Ofsted rate as less than good. This is too restrictive. The application form was changed at the start of the year to encourage mass applications from trusted MATs, and this had worked well judging by the number of applications made to the DfE in March. Nick felt that it was now time for the form to be changed to encourage more applications from new providers. The criteria should include basic need, with better information made available; educational need, defined more liberally; social need, to address particular social problems like the segregation of children in cities along religious or racial lines; a need for innovation; or a need for diversity, if for example all the schools in an area were religious in character and people wanted a non-religious alternative, or vice versa. Also – notwithstanding what an official in Sanctuary Buildings thinks about local community need – if you can prove parental demand for your proposed school, then subject to all the other quality checks, you should be able to open it. If local parents are saying they need a new school, he said, that should be more powerful than the contents of a spreadsheet.
Nick went on to argue that existing government policy unnecessarily restricts the potential supply of great new schools. It is possible to open a faith-designated free school, under existing rules, but not in effect if you’re a Roman Catholic. This is because faith-designated free schools need to sign up to an ‘inclusivity rule’, which means that if you are over-subscribed, you cannot reserve more than fifty per cent of the places at your school for children of your faith. The Roman Catholic Church believes it is against canon law for a Catholic bishop to designate a school a Catholic school but then turn away Catholic pupils on the basis that they are Catholic. So there are no Catholic free schools. This is a shame because not only are Catholic schools successful and popular with parents, but their pupils are also more likely to be from Black or Minority Ethnic (BME) backgrounds and more likely to live in deprived communities, while a third of Catholic school pupils are not themselves Catholics. Faith-designated free schools from other minority faiths, however, are not diverse despite the fifty per cent rule. So the rule fails according to its own objective, while depriving the country of successful schools – at a time when the Catholic population, and therefore demand for a Catholic education, is growing.
Nick offered for discussion a number of other ideas for how to increase capacity – not just via free schools but in the academy sector more broadly. As university tuition fees increase, should we make it a condition on Russell Group universities that they should set up or sponsor schools in their name in return for charging maximum fees? Could we allow the conditional expansion of existing selective schools? In return for opening a new grammar school, its foundation or trust would have to open a non-selective school also? Should the government do more to encourage the best people in local authorities to spin out to form their own MATs? Should the government seed-fund the establishment of specialised school chains – for example, with specific focus on maths or science or foreign languages? Given the likely future of 16-19 provision, specialist sixth forms might be the only way to guarantee some important A-level subjects.
In conclusion, Nick stressed that though he’d talked pretty much exclusively about free schools, the challenges facing the programme reflected those facing the system as a whole. The government is trying to create a market in the education system. This, he said, is the right track for reform, but at the moment there’s a risk that we’re building in the potential for market failures too. A functioning market needs enough genuinely new entrants to challenge existing providers, enough capacity for competition to be meaningful, enough information for providers and users alike, ways of breaking up failing or monopolistic providers, and exit points for providers that aren’t doing a good enough job. The direction of travel is the right one, but there’s a lot that still needs to be done.
* The target number of 500 new schools to be opened over the course of this Parliament includes schools procured by local authorities and not just free schools. The figure is supposed to represent the expected number of new schools required to meet basic need after accounting for the number of places that can be provided through expansion of existing facilities. However the figure also includes schools that were due to open before the 2015 election and so the real figure could be more like 350.
The discussion
The discussion began with a question about the government’s preference for ‘mum and pop’ schools. Rather than opening the market up properly – to more sustainable for-profit propositions – the preference had been for not-for-profit parent-led schools and then propositions put by community stakeholder or teacher-led groups. This was seen as unnecessarily limiting. It was recognised that public opinion would probably be against it and that politically a more pragmatic and incremental approach was needed. It may be possible to make a case for ‘for profit’ providers being given an opportunity as a ‘last resort’ where schools had been persistently under-achieving over a very long period of time.
But if the purpose is to open up new supply, then it might be more feasible to propose letting trusted existing providers have more freedom to acquire and develop sites by taking on debt. It was reiterated that many good Catholic schools were interested to convert to free schools, and expand capacity, but inclusivity rules currently prevent this. Similarly many good independent schools carried spare capacity and the prospects for a revival or modification of the Assisted Places scheme were discussed. Some independent schools appeared fearful of converting to free schools because of the perception that it might lose them their existing customers. It was pointed out that this fear may be groundless, after all the most over-subscribed free school in the country converted from being an independent. Nevertheless it was acknowledged that the offer of a per pupil subsidy might be a more flexible way of unlocking spare places in fee charging schools. The guiding principle should be ‘what works, works’.
In respect of site identification, acquisition and development, it was suggested that the current process was over-complicated. The government ought to enter into standardised contracts with a small number of providers to save money.
One participant explained that the DfE had shied away from standardization, because of the likelihood that such an approach would be criticised for lack of scope for innovation. The strategy had instead been to offer baseline designs and the opportunity to better them.
It was felt that better guidance was required in relation to the placement of new schools. It was noted that GIS mapping was being looked into for this purpose. It was suggested new schools could be established in the buildings of independent schools that have closed or have extra capacity.
The discussion turned to the priority of ensuring quality and Ofsted’s role in that. According to one participant, the ‘disruptive’ opportunity presented by free schools carries a responsibility to offer a proposition that is in some tangible way ‘better’. At the moment the sector seemed mainly concerned about numbers and meeting basic need.
Agreeing that the development of free schools and meeting basic need were fundamentally different things, another speaker drew a parallel with the diverse housing associations that were in existence before the sale of many council houses in the 1980s. Housing associations lost their distinctiveness, he said. The story might be different for schools, but not if all new schools were driven by central government. Free schools should be more interested in the quality of places they are going to provide, not the number.
Turning to Ofsted’s role in quality assurance, it was suggested by one participant that there are still serious questions about the reliability of its rating systems, with areas in the North having no ‘requires improvement’ or ‘inadequate’ primary schools, but many secondary schools in those categories, for example. More attention should be paid to the quality of information Ofsted is generating as there is great concern about it in the in the education sector.
One participant asked if enough was known about the quality of schools (beyond GCSE results) to decide what makes a good and sustainable free school. For example, do we know enough about MATs and why the ones that are successful are successful? Do we know when to intervene or if and when to allow schools to fail?
Another participant remarked that assessing quality was particularly difficult in the case of new schools. It takes five years to show what a secondary school can do (measured by GCSEs) and seven years for a primary school to do the same (measured by Key Stage 2). He asked how free schools make their case in relation to education quality in the absence of robust evidence of impact. In response it was stressed that, contrary to popular research belief, consumers are more sophisticated than the measurements on offer, and need a mix of qualitative and quantitative data. We can surely extrapolate something from the fact that free schools are, on average, undersubscribed in their first year, oversubscribed by their third year and are, overall, the most over-subscribed school type?
On quality, another asked how we get innovation in the system that allows free schools to thrive and enable quality in a way that goes beyond exam results, and that would deliver efficiency gains also? Ultimately, until levels of efficiency are understood, you can’t open a dialogue about ‘for profits’ and their level of expected surpluses.
Turning to competition effects, and the importance of attracting high quality sponsors, further consideration was given to the speaker’s proposal to open up free school/academy supply by grammar schools through offering to remove restrictions on the expansion of selective entry provision in the sponsoring school in return. Birmingham and Bromley were cited as good examples of places where schools had raised their game in response to grammar schools competition.
In response, one participant expressed skepticism about the reliability of this mechanism. The example of Bristol was discussed, where secondary school head-teachers had got together to take it in turns to bid for free schools, effectively colluding to thwart competition, and in the process making the notion of market reform redundant. In reply, another held that while Bristol’s education landscape had improved thanks to Cabot and Oasis academies driving up standards, other providers are now required to provide challenge and spur improvement. Concerns were expressed that there was not sufficient existing or potential new provider capacity to enable consistent implementation of this strategy across the system. In respect of whether we should be looking to local authorities to take forward new school development, the same contributor felt this to be unrealistic without a standard setting and monitoring role. He suggested that the logical thing to do now would be to transfer the responsibility for place planning to the school commissioners’ office, and met via an expanded regional structure with greater local reach.
It was widely acknowledged that the state of teacher recruitment was of more fundamental concern. However many more new teachers were being recruited, the increase was not keeping abreast of the increasing number of vacancies left by the departure of more experienced teachers to overseas markets, to other employment, and to retirement, and arising in response to basic need demand.
There was wide agreement that the system asks teachers to do the impossible at the moment (and indeed that policymakers are likewise over-optimistic about what the average governor can do). While MATs can alleviate this to a degree, there aren’t multiple Dan Moynihans to overcome this problem.
Speaking to concerns about retention rates, one participant expressed the view that the best way to combat this was to give more teachers more of a personal stake in the system. A co-operative shareholder model should be offered. This would entail more collaborative ways of working than generally characterized the corporate CEO-led model.
Concluding remarks
The Chair thanked Nick Timothy for his candid reflections and contributors for their remarks and questions.
Nick had talked about restoring choice and challenge to a central place in the brief for those who propose free schools, the need for which CMRE had repeatedly emphasised. He had made a special point of highlighting the lack of information for parents, necessary to support the ‘choice challenge’ proposition. The Chair noted some promising proposals in the recent White Paper in this regard, such as a ‘parent portal’ to provide more accessible comparative information on quality.
The speaker’s thoughts on how the process of developing and implementing plans for challenger schools could be made more efficient were also very useful, he said. The weakness of local authority estimation of basic need and the lack of provider access to this data was noted. Information on land ownership and usage could be made more widely available and useful.
The main focus of Nick’s thinking had been the question whether we oughtn’t to have a two-track system for originating new schools, with one addressed to meeting basic need and the other to more context-specific propositions meeting other kinds of need, including the need for a challenge to the local status quo. This reorganisation would open the way for further efficiency gains on the basic need side, such as the recent decision to allow MATs to submit bulk applications in relation to basic need. In respect of the second track, however, the Chair raised the question of how priorities should be weighted.
The speaker said that the current process of approving schools is unclear, a curious mixture of procurement, market-initiated proposals and a panel of people at the DfE making the final decisions according to criteria that were, to say the least, less than clear. Indeed no one knows the ultimate criteria for deciding on competing proposals for different types of schools. This would become more obvious if a range of different needs were to be validated as legitimate grounds for application.
The New Schools Network had put much effort recently into engaging a broader range of providers. This was clearly helping to reveal what inhibits potential sponsors from getting involved, and where the process of application and approval could be more efficient and reduce associated costs.
Limited, conditional expansion of grammar schools on the model proposed (whereby this might be permitted on the basis of non-selective free school sponsorship) was an idea that demonstrated a welcome pragmatism, on the proviso that that expansion was in proportion to an overall and overriding commitment to mainstream provision. However, while selection by ability probably has a place in education to an extent, it should not be the principal allocation device at primary and lower-secondary level. Little rigorous evidence suggests that school-level ability selection has strong direct effects on achievement. Moreover, to a certain extent, selection within schools (i.e. setting by ability) could in fact fulfil a function similar to the one that proponents of school-level selection envisage, without undermining the incentives at work on schools to compete (as argued by CMRE Research Director, Gabriel Heller Sahlgren, in his essay on selection by choice for the Civitas publication,
The Ins and Outs of Selective Secondary Schools). It should not be supposed that governance and leadership expertise from grammar schools would be readily transferable to free schools. This latter point should be subject to further investigation.
The argument for reconsideration of the ‘inclusivity rule’ was well put. According to this rule, in the event of over-subscription, you cannot reserve more than fifty per cent of the places at your school for children of your faith. Catholic schools tend to be academically successful, and as a group are more diverse in ethnic composition, and better at reaching the disadvantage – and they are in demand. The case for a measure of discretion to be introduced around such proposals should be examined.
The notion that universities should be taking a lead in sponsoring free schools has been received positively in policymaking circles – though whether as a means of demonstrating subject expertise, or as a means to an end in terms of extending their reach among disadvantaged pupils. The Chair questioned both the assumption that Universities had expertise to bring to school governance and leadership, or contextually transferable and relevant skills, and expressed scepticism about whether Russell Group universities would sponsor free schools in numbers, or should be pressed to do so, in relation to the equalities brief. What impact would this be likely to have in real terms for improving equity of higher education access? From an incentive point of view would the brand lift among schools be sufficient to justify the commitment from Universities?
Beyond this, if the priority of increasing choice and competition was to be re-established, the Chair said that if local authorities statutory responsibility for place planning were to be transferred to the commissioner’s office, this would open up possibilities for local authority constituted but now independent service providers to get involved in sponsoring free schools and academies, and this should be given serious consideration. Companies such as One Education and Essex Educational Services would undoubtedly be interested in creating their own MATs.
But, he said, we cannot avoid the conclusion that whether we are looking for basic need capacity alone, or beyond this to the potential of regional or system-wide lift from the increase of choice and competition, ensuring a steady pipeline of new sponsors and viable propositions would be much more feasible were there a more direct route for businesses to get involved. On the basis of what works, works, further exploration of this option should not be ruled out.
CMRE messages for policymakers
- Acknowledging the importance of basic need, there is little point in making sure that there are enough school places overall if a significant number of those places are not good enough.
- It is important that allowance is made for a ‘critical mass’ of challenger schools in the plans for the programme’s expansion.
- In that the requirements of each are quite different, the merits of developing two distinct programmes – one oriented to basic need, the other to free schools – should be considered.
- A number of measures are required to support the proper functioning of the schools market, including information for both parents and providers.
- Plans for a new ‘parent portal’, including a mixture of qualitative and quantitative information about schools’ performance, approach and ethos, should go ahead. A facility should be included for comparing not just local, but similar, schools across the country.
- Any and all information collected or held by LAs relating to basic need or demand should be released, transparently and in an accessible format, into the public domain.
- In order to encourage new provision, the application criteria need to be less restrictive. Applications primarily addressed to social need, such as the effects of alleviating ethnic/religious segregation, should be better accommodated.
- Local parental demand, subject to meeting the quality criteria, should be sufficient to open a new school.
- The offer of a per pupil subsidy for non-selective entry fee-charging schools, in order to unlock spare places, should be considered.
- Limited, conditional expansion of grammar schools should be considered also, on the proviso that that expansion was in proportion to an overall and overriding commitment to expanding choice in the context of mainstream provision. However, it should not be supposed that governance and leadership expertise from grammar schools would be readily transferable to free schools. This latter point should be subject to further investigation.
- The terms of the ‘inclusivity rule’ – whereby you cannot reserve more than fifty per cent of the places at your school for children of your faith – should be reconsidered. Catholic schools tend to be academically successful, and as a group are more diverse in ethnic composition, and better at reaching the disadvantage, than other types – and they are in demand. The case for a measure of discretion to be introduced around such proposals should be examined.
- The potential of Universities to sponsor free schools in numbers is dubious, for a number of reasons outlined in the Chair's concluding remarks above.
- There is appetite among former local authority-constituted, now-independent service providers to get involved in sponsoring free schools and academies. For this to happen, to avoid obvious conflicts of interest, local authorities’ statutory responsibilities for place planning would have to be transferred to the commissioner’s office.
- A governance and leadership deficit was acknowledged. The government should be firm in its intention to drop the requirement for local governors and give priority to the modelling of rationalised MAT governance and leadership structures, with clear schemes of delegation to optimise effective decision-taking.
"Incentivising excellence marks a significant achievement in evidence-based policy analysis. Sahlgren rigorously analyses the existing literature and offers policymakers a pragmatic programme of options for improving school quality."