How do local authorities manage high needs funding pressures?

It is a spring afternoon in 2026 and you are looking at a high needs forecast that is still climbing, on top of a deficit that has only had 90% of its historic portion promised away. The override now ticks to March 2028, Safety Valve is closing, and the Stability Grant will not land until your reform plan is approved. The real question is not what the high needs block is. It is which levers are left before 2028, and which ones cut demand rather than just move the deficit around.

The 2026 reset: what changed and what didn't

On 9 February 2026 the government committed to absorb 90% of each council's accumulated DSG deficit as it stood at March 2026, around £6bn collectively, paid through a new High Needs Stability Grant from autumn 2026-27 (Schools Week, 2026). The grant comes as a Section 31 grant, which matters because it puts the money on the council's books cleanly rather than as another line inside the ringfenced schools grant (DfE explanatory note, 2026).

Read the small print before you plan around it. The other 10%, roughly £600m across England, stays with councils to repay by March 2028. There is no commitment to absorb any deficit you run up between April 2026 and March 2028. So the write-off clears the past; it does nothing for the gap you are about to open this year and next.

Two dates carry the rest of this. The DSG statutory override, the rule that keeps these deficits off your usable reserves and stops a deficit alone from forcing a section 114 notice, now runs to the end of March 2028. That extension was confirmed in June 2025; it is not the old March 2026 sunset that some cached guidance still quotes (Tes, 2025). Combined DSG deficits stood at around £6.6bn at the end of March 2026, and the Office for Budget Responsibility warned the cumulative figure could reach about £14bn by March 2028 if nothing changed (OBR, cited Feb 2026).

The Safety Valve programme, the roughly £1bn intervention route that traded extra funding for a deficit-recovery agreement, is ending. The new settlement is more generous than most Safety Valve deals would have been, so it has overtaken them. Existing agreements run on, some with payment profiles stretching to 2031-32, for example Devon's £95m profile, so if you are mid-deal you keep delivering it (NEU, 2026). For the mechanics, see what the Safety Valve programme is and how the DSG statutory override works.

Why the block keeps outrunning the grant

The block keeps rising because demand keeps rising, and a write-off does not touch demand. EHC plans reached 638,700 in January 2025, up 10.8% on the year before, with 97,700 new plans issued in 2024, up 15.8% (DfE, 2025). The stock is growing and so is the pipeline behind it: 154,500 requests for an EHC needs assessment in 2024, up 11.8%. You are not managing a one-off bulge.

Over the longer run the shift is structural. The share of pupils with an EHC plan roughly doubled between 2016 and 2025, and high needs spend rose around 66%, from about £7.5bn in 2016 to at least £12bn in 2025 (IFS, 2025). One driver sits inside that. Autism-related plans roughly tripled between 2015 and 2025 and account for around 40% of the total increase, which is specific enough to forecast and plan around rather than treat as general growth.

This is the trap finance teams fall into after good news. The deficit drops, the pressure eases for a quarter, and the demand-side work that actually changes the trajectory gets deprioritised because it is slow and the balance sheet looks better. The honest planning frame is that the Stability Grant bought you runway, not a fix. For the modelling side, how councils forecast high needs demand is the companion to this article, and what a high needs block deficit is sets out the underlying accounting.

Demand-side levers: inclusion that holds

The single biggest cost driver councils underweight is avoidable independent and non-maintained special placements, and the structural answer is mainstream capacity strong enough that fewer children need to move out of it. From 2026-27 the School and Early Years Finance (England) Regulations let you spend from the high needs budget to support mainstream inclusion, which legitimises funding prevention rather than only paying for placements (DfE operational guide, 2026).

That flexibility only helps if you spend it on things that work. The cheapest credible evidence base here is the Education Endowment Foundation. Its guidance on making best use of teaching assistants is blunt: deploy TAs to add to the teacher, not to replace them, and use them to build a child's independence through scaffolding and least-to-most prompting, rather than sitting next to one child all day (EEF, 2026). The pattern it warns against, where a TA becomes a permanent prop and the child never works unsupported, is the one most likely to be written into casework as a reason a mainstream placement “broke down”.

The Teaching and Learning Toolkit puts numbers on it. Structured, TA-delivered interventions add roughly four months of progress, an effect size of around 0.2 to 0.3, but only where there is training, a structured programme and protected time. Without those, the same spend can do nothing or slightly harm progress (EEF, 2026). For schools forum and for the DfE officials reviewing your reform plan, that is a far more defensible line than another round of top-up inflation, because you can show what the money buys.

Two practical companions sit underneath this for your schools: building pupil independence rather than TA dependence and how councils build inclusive mainstream capacity. The cost lever they feed is the one on reducing independent special school placement costs.

The statutory and governance levers you still control

You still hold several levers that do not depend on national money, and the schools forum is the one most under-used as a strategic tool rather than a sign-off. Consulting the forum on high needs arrangements is a statutory requirement, and place numbers, top-up rates, independent placement budgets and central support all run through it (School and Early Years Finance (England) Regulations). A forum that understands the demand forecast, not just this year's allocations, is part of your recovery plan; a forum kept at arm's length is a future dispute.

Block transfer flexibility, moving a capped percentage from the schools block into high needs, still exists, but be honest about its limits. It needs schools forum or Secretary of State approval, the cap has been tightened over recent years, and it moves money between pots rather than creating any. Treat it as a marginal adjustment, not a recovery strategy.

Governance is where the threads tie together. Area SEND inspections under the Ofsted and CQC framework judge the whole local SEND system, not your finances in isolation. In practice a worsening deficit and a poor inspection are the same governance failure seen from two angles, because both come from a system that is not meeting need early enough. Planning for inspection and planning for recovery should be one piece of work; see how councils prepare for an Area SEND inspection.

Two further levers cut cost by cutting conflict. Co-production with your parent carer forum and joint working with health through the integrated care board are statutory expectations, and done well they reduce the tribunal and disagreement-resolution costs that quietly eat the block. Tribunal volume is itself a demand-side number you can move; see how councils reduce SEND tribunal losses.

The lever that ties the rest together is the Local SEND reform plan, which is now a funding condition. The High Needs Stability Grant is only paid once that plan is submitted to and approved by the DfE (DfE explanatory note, 2026). So the reform plan cannot sit in a drawer next to the financial recovery plan. It has to be the spine of it. The cleanest way to think about it: your recovery plan is your reform plan with a money column.

Planning to 2035: White Paper and the new Bill

The medium-term picture changed on 23 February 2026, when the Schools White Paper “Every Child Achieving and Thriving” was published, to be legislated through the Education for All Bill announced in the 2026 King's Speech (DfE Education Hub, 2026). For a finance lead, three points in it shape the next decade.

• Every school, nursery and college would gain a legal duty to create an Individual Support Plan (ISP) for every child with SEND, a digital record sitting below EHC plan level.
• EHC plans would be kept and protected for the children with the most complex needs, narrowing towards that group by around 2035, with no statutory change before September 2030.
• The model is backed by £7bn more for SEND in 2028-29 than in 2025-26, but the money is tied to the new approach embedding, and the DfE expects plan numbers to keep growing short term before stabilising or reducing by around 2035.

The planning implication is a hard window, and it is yours to carry. The override runs to March 2028. The new statutory model lands no earlier than September 2030. So for roughly two to four years you hold residual deficit risk under the old rules while needing to build ISP-level mainstream capacity ahead of a legal duty that has not started yet. Councils that wait for the Bill before building inclusion capacity will meet the new duty cold, with the deficit still live.

What a credible recovery plan looks like

A credible plan pairs a financial recovery trajectory with a demand forecast, not a flat overspend projection. The forecast has to combine three things: EHC plan stock, the assessment pipeline behind it, and growth by need type, because autism alone is driving around 40% of the increase and behaves differently from, say, social, emotional and mental health need.

The sequence that holds together is straightforward to say and hard to do:

1. Secure the Stability Grant by getting the reform plan approved. Nothing else is funded until this is done.
2. Stop the residual 10% deficit rebuilding, which means acting on intake and placement mix now, not after the grant lands.
3. Invest the inclusion-funding flexibility in evidenced mainstream provision, using EEF-aligned TA deployment and structured intervention as the measurable inclusion metric in the reform plan. That metric is defensible to the forum, to the DfE and to Ofsted and CQC at once.

Then watch the cliff edges, because they are dated and unforgiving. The residual 10% is due by March 2028. Where local government reorganisation creates successor councils, those successors must meet any remaining deficit from their own resources in 2028-29 (DfE explanatory note, 2026). Neither of those moves if you look away.

If you take one idea from this into your next finance and SEND strategy meeting, take this. The override extension is runway, not reprieve. The balance sheet has been reset for you once; the demand curve has not, and the demand curve is the real liability. Build the mainstream capacity now, while the national money is buying you time to do it, because the version of 2028 where the grant landed and the demand work did not is the version where you are back where you started with fewer options.

This article is general information for SEND and finance professionals, not legal or financial advice on your authority's position. Figures and dates here reflect the position as understood in spring 2026 and the rules are moving quickly; check the live DfE operational guidance and your own Section 251 returns before acting on any specific number.