I expected the pension saving to fund the pay rise. It didn't, and that is the real story.

Everyone is reacting to the 3.5%. I understand why, but I think the number that actually matters is sitting quietly underneath it.

Here's my honest take When I saw the pension change coming, I assumed the government would do the strategic thing with it. From April 2027 the employer contribution to the Teachers' Pension Scheme drops from 28.6% to 17.6%. That is a substantial saving, and I fully expected them to point it straight at the pay award. Fund the 3.5%, offset by whatever was needed to cover the pension implications, leave schools roughly whole. Joined up. The kind of design you'd build if the genuine aim was to make a pay rise land without destabilising trust budgets.

That is not what happened.

The money was already there The £3.3bn the pension change saves nationally has not been recycled into the pay award. For most trusts it's effectively been taken back out as a cut to core school funding of a broadly similar size. And it's worth sitting with the scale of that, because the saving is larger than the new pay funding the government actually provided:

-£3.3bn saved on the pension side -£700m rising to £1.1bn offered towards the pay award -Schools still asked to find the first 1% of the rise from their own budgets

The money to fund this properly was arguably already sitting in the pension line. Instead of using it, the government banked the saving, cut funding by a comparable amount, and passed a share of the bill back to schools anyway.

Why this is the bit to get right A lower TPS rate flows straight into every staffing model and every three year forecast, and read on its own it looks like freed up cash. Net it against the funding reduction and the headroom largely disappears. In some trusts, once timing and pupil numbers are layered in, it tips slightly the wrong way. This is exactly the sort of item that reads as good news on a one page summary and quietly evaporates in the detail. Catching that when you're producing your budgets is the job.

The headline, in context Teachers and leaders receive 3.5% from September 2026, with a further 3% the year after. Support staff have been offered 3.3%, backdated to April. It's presented as certainty. The catch is that schools carry the first 1% of each year's rise themselves, so the figure you actually plan against is not 3.5%, it's 3.5% with roughly a third of the cost on you. On a £20m teaching salary bill, that unfunded 1% is around £200,000 a year to find internally, recurring, and growing as the base rises. One further change lands squarely on trusts. Executive pay above £174,000 now needs official approval before a role can be advertised. Whatever your view of it, it's a governance and remuneration question that belongs in your pay policy now rather than mid recruitment.

The wider picture keeps it live The IFS puts most teachers' pay around 7% below its 2010/11 level in real terms by 2027/28 Inflation was 2.8% in May, but Bank of England modelling has it drifting towards 3.7% by year end The NEU has already signalled a strike ballot in October if the offer isn't fully funded

The timing is the sting Here's the part that makes this harder than it needs to be. Your budget runs September to August, and the pay award lands cleanly at the start of it in September 2026. The pension change doesn't. It arrives in April 2027, part way through a budget you've already set and signed off. So the saving everyone will talk about doesn't turn up as a clean line at the top of your planning cycle. It appears in the middle of the year, tangled up with an offsetting funding cut, in a budget that was built before you knew its exact shape. That is precisely how a change this size ends up mismodelled.

What I keep coming back to The pay rise and the pension change were the two biggest levers the government had this year, and they've been pulled in opposite directions. One puts money into schools. The other quietly takes a comparable amount back out, on a different timetable, in a way most people won't connect. Handled together, they could have delivered the thing schools actually need right now, which is stability and a settlement you can plan around with confidence. Instead they've been designed as if they were separate decisions, and the job of reconciling them has landed, as it usually does, on finance.

That's the frustrating part, and I don't think it's worth pretending otherwise. A genuinely strategic settlement would have used the pension headroom to fund the pay award and left trusts on steady ground. This one doesn't, and the gap between what was announced and what was designed is the whole story.

If you're modelling the September 2026 pay award and the April 2027 pension change together rather than separately, you're already ahead of most. Compare your assumptions with peers doing the same, because that's where the real value sits.