In the last post I set out that the SEND system isn’t working because it’s a bad system. It’s built on a circular definition of SEND that forces parents, schools and local authorities to fight to determine the support a child gets. And it’s funded in a way that incentivises the opposite to the behaviour we want from schools.
This failure doesn’t have external causes - it is a failure of policy. That’s in some ways more upsetting, as this is a crisis of our own making. But it should also be encouraging. If this is a policy failure then there is a policy solution. Here’s what I think that is.
Ask not what’s ‘wrong’ with this child, but what support they need
The current system is a conflation of a diagnosis-driven model and a need-based model. The definition of SEND is just that a child needs extra support. Yet we say that they need a ‘diagnosis of SEND’[1] in order to access that support. We should shift away entirely into a need-driven model, where the job of the SEND system is simply to give children the help they need as quickly as possible.
This involves changing how we decide on support, how we fund it, and how the system is organised.
What support should a child get?
In the last post I described the lack of evidence around what constitutes great support for children with different needs. This is a big problem. It leads to a trial-by-combat system, where support is determined by what each party is willing and able to argue for. Both sides are also often flying blind, with parents trying to infer what will work from various internet advice and non-specialist EHCP coordinators trying to work out what’s right for children with a myriad of different needs.
We need a NICE for SEND. NICE, the National Institute for Health and Clinical Excellence, set out what good health care looks like. Anyone can go on their website and see what care they’re supposed to get for a condition. It publishes quality standards, that set out specific and actionable things such as:
People attending an emergency department with a head injury have a CT head scan within 1 hour of a risk factor for brain injury being identified.
It provides detailed guidance for primary care professionals on how to assess and diagnose common conditions, and how to treat them once identified.
NICE creates an external reference point for what good care should look like. It empowers patients, who can easily find out what care they should be receiving and advocate for it if necessary. It makes explicit what everyone in the health system should be striving to provide.
We need an equivalent for this in education. We should compile the available evidence and use it to determine practical guidance and quality standards. This would make it clear to everyone how we identify children in need of different types of support, and how to deliver that support effectively.
This guidance could be organised in tiers, with the first tier covering common issues (such as struggling to read at an age appropriate level) and the second tier covering specific conditions. Guidance for the first tier could cover the ‘best bets’ for supporting children who present with this need, with the aim of getting them support as quickly as possible. Just like NICE guidance it would suggest an order for trying out different treatments, with guidance on how to move onto another treatment if a prior one isn’t working. Then if support at this tier isn’t having the desired effect it would suggest further diagnostic activity to identify if there is an underlying condition that is preventing progress.
Having an independent evidence-based arbiter of what support a child should be getting would both raise quality and improve experience.
It would raise quality because all children would be getting evidence-based support. Too often I have seen EHCPs, which remember are legally binding documents, mandating that a child needs to receive an intervention for which there is either no evidence or evidence that it doesn’t work! After all the pain of acquiring an EHCP it’s too common for the end result to be a plan that isn’t the support most likely to help a child. This shouldn’t be acceptable.
It would also improve experience. Rather than determining support by pitching parents and local authorities against each other, the appropriate support would already be set out. Parents and LAs could focus on working together to bring that into reality for their child, rather than on fighting about what is or isn’t good care.
How should we pay for it?
We have a funding system that pays out when inclusion fails, and so encourages the opposite to the behaviour we want from schools. Instead of having a system where funding predominantly follows named pupils, we should fund schools and local areas directly based on the need in their broader area.
In our present system the vast majority of funding is attached to EHCPs. This means that to get funding you need to have children with EHCPs. This penalises schools where inclusion and support succeeds, and rewards those where it doesn’t. It also places huge stakes on the EHCP threshold, as this document is the difference between getting support and not.
There is an excellent 1999 paper that looked at the funding systems across 17 European countries. It describes countries where funding follows pupils as below:
“These countries point at the different forms of strategic behaviour within the educational field (by parents, teachers or other actors). These forms of strategic behaviour may result in less inclusion, more labelling and rising costs. A great deal of money is spent on non-educational matters such as litigation, diagnostic procedures and so on. It is not surprising that these countries can be identified as having relatively higher percentages of pupils with special educational needs in separate settings.”
This could have been written to describe England in 2024.
Although it seems reasonable that funding should follow the pupils who need it, this doesn’t work in practice. There is no objective way of assessing who should and shouldn’t receive funding, and hard thresholds create high stakes around such decisions. And any process that requires independent sign-off sacrifices timeliness, meaning children have to wait longer.
Instead we should shift to a system where schools and local areas are funded to provide a given level of support to all children who need it. Schools should be given the money they need to meet the quality standards set out above, which could increase or decrease based on the demographics of the community they serve and the likely level of need.
Clearly there are cases where individual pupils may need a lot of support, and this could not be funded at the school level as we cannot predict from statistics where these pupils will be studying. To cover this we should give a proportion of funding to local partnerships of schools, who could choose how to distribute it between them based on the needs of pupils they each serve. This partnership could also be responsible for procuring some services at group-level where they might get a better deal than individual schools would on their own.
This funding system shifts the incentives. There is no more reward for non-inclusive behaviour. The quicker you can support a child and the less their need escalates, the more resource you have left over.
How should we organise support?
We have too binary a system at the moment, with a hard cut off between mainstream and special school. This doesn’t work for lots of pupils, many of whom could thrive in a modified mainstream environment.
We should significantly grow the amount of co-located provision, where there is specialist expertise and support within a mainstream school. These special schools within a mainstream school give the flexibility for support to be scaled for a child as needed, and to do so without the delay and disruption of transfers between schools. They also bring specialist expertise into mainstream that can improve the support to all children, including those who never need the specialist end of provision.
I think we should be aiming for a world where every mainstream secondary school has some level of specialist support within it, in proportion to the school’s size. This should not be an unusual part of the system, it should just be how our system works.
What about EHCPs?
The system I’ve outlined above doesn’t mention EHCPs. That’s because I don’t think we need them. In a world of quality standards and funding to meet them, the system doesn’t need EHCPs to work.
Removing EHCPs would significantly grow the amount of money available to support children, as it removes the monumental cost of administering the EHCP funding system. Allocating resource through EHCPs costs local authorities upwards of £500m per year, and one might reasonably assume the cost to schools is at least that. Freeing this up would add 10% to the SEND budget overnight.
We would need an independent process for determining admission to special schools, and this is likely to need to be informed by an independent assessment of need. But this could be run in a much simpler way than the current system of plans.
In summary
The SEND system is broken, because it’s a bad system. It doesn’t tell people what they should be getting, it incentivises the wrong behaviours, and it’s organised to create hard thresholds that serve children badly. All these things are changeable, so let’s change them.
[1] This is not an argument that children should not be diagnosed with conditions when they have them, but a point that we shouldn’t create a hurdle of the ’SEND diagnosis’ in order to get support.