Independent School Teachers and the Teachers' Pension Scheme

Why this is happening at all

Most teachers encounter this subject the same way: an email lands announcing a staff consultation on "the future of pension provision", and within a fortnight a decision that will shape forty years of retirement income is being made in a series of meetings nobody feels equipped for. The purpose of this guide is to give you the underlying mechanics before you sit in that room, because almost every genuinely consequential detail in this area is a technical one, and almost none of them get properly explained in a consultation deck.

The starting point is a structural asymmetry that surprises a lot of people. The Teachers' Pension Scheme is not a state-sector-only scheme. Independent schools have long been able to participate as accepted establishments, and historically most did — it was a major recruitment advantage, because a teacher could move between a grammar school, a comprehensive and an independent school without their pension arrangements changing at all. That continuity was one of the quiet things that made the teaching labour market unusually fluid.

But participation for an independent school is voluntary in a way it simply is not for a maintained school. A local authority school has no mechanism by which to stop offering the TPS to its teachers. An independent school has always had one. For twenty years that difference was theoretical, because the cost of participation was manageable and the recruitment benefit was obvious. Since 2019 the cost has stopped being manageable for a significant number of schools, and the theoretical difference has become the defining fact of independent sector teaching employment.

The employer contribution rate story

Everything that follows is downstream of one number. The employer contribution rate is the percentage of a teacher's pensionable pay that the employer must pay into the scheme, on top of the teacher's own banded contribution. It is set by periodic actuarial valuations of the scheme, and for most of the 2010s it sat at 16.48% — a figure schools budgeted around without much drama.

From 1 September 2019 it rose to 23.68%, including a 0.08% administration levy. In proportional terms that is an increase of more than 40% in the cost of employing exactly the same teacher on exactly the same salary. Then, from 1 April 2024, it rose again to 28.68%, again including the 0.08% levy. Against a 2018 baseline, the employer cost of a teacher's pension had risen by roughly three-quarters in under five years, with no corresponding change in what the teacher received.

It is worth being precise about what that means in cash for a single teacher, because percentages of pensionable pay abstract the problem in a way that makes it easy to underestimate. Take a teacher on a pensionable salary of £45,000. At 16.48%, the employer's annual pension cost is about £7,420. At 23.68%, it is about £10,660. At 28.68%, it is about £12,910. That is an additional £5,490 a year, per teacher, relative to the pre-2019 position. A modest independent school with sixty teaching staff is therefore looking at something in the region of an extra £330,000 a year in employer pension costs alone, before any of the other cost pressures of the last few years.

Those figures are arithmetic on the published rates rather than any school's actual accounts, and real schools have varied salary distributions, part-time staff and support staff on entirely different arrangements. But the order of magnitude is the point: for a fee-charging institution with no access to additional public funding, a cost increase on that scale has to come from somewhere, and the realistic options are fees, staffing levels, or the pension scheme itself.

The grant independent schools never got

This is the causal hinge of the entire story, and it is routinely left out of summaries. When the employer rate rose in September 2019, the Department for Education did not simply leave schools to absorb it. It created the Teachers' Pension Employer Contribution Grant — the TPECG — which provided financial assistance to maintained schools, maintained nursery schools and high needs settings to cover the increase, with funding allocated from the 2019 financial year and paid to local authorities by the Education and Skills Funding Agency on behalf of the Secretary of State. Academies were funded through the equivalent route. The grant was subsequently continued and eventually rolled into core funding arrangements.

Independent schools received no equivalent. There was no grant, no transitional funding, and no mechanism by which one could have been paid, since the whole architecture of the TPECG is a distribution of public money to publicly funded settings. The only nuance worth noting is that local authorities were allocated funding in respect of pupils with education, health and care plans placed in independent settings, on the basis that those settings' fees would rise — but that is a payment routed to a placing authority to cover fee increases, not a grant to an independent school to cover its own staff pension costs.

That asymmetry is the whole explanation for why the withdrawal wave is an independent sector phenomenon and not a general one. It is not that independent schools value pensions less, or that their governors are less sympathetic. It is that they were handed a very large unfunded cost increase alongside the only exit door in the building. The second rise in April 2024 landed on schools already dealing with VAT on fees, the withdrawal of business rates relief in England and increases in employer National Insurance — a combination the Independent Schools' Bursars Association has publicly identified as stimulating a further wave of pension consultations.

How many schools have actually left

Here we have to be careful, because this is an area where two credible, FOI-sourced datasets circulate widely and do not agree. We are going to set both out rather than pick one, because picking one would create a false impression of precision.

The first comes from an analysis published by Second Sight in July 2025, based on a Freedom of Information request to Teachers' Pensions. It reported that 410 independent schools had left or were in the process of leaving the scheme since 2019 — up from 313 as at January 2025, meaning 97 schools moved in roughly six months. It reported 618 schools remaining in the scheme, 283 schools in phased withdrawal since 2021, and, against the general direction of travel, 80 schools that had joined or rejoined since 2019.

The second comes from a Freedom of Information request obtained by the Telegraph and reported in February 2026. On that basis, 1,066 independent schools were TPS members on 29 July 2024 — the day the introduction of VAT on school fees was confirmed — falling to 880 by January 2026, a decline of 17%, or roughly one in six.

The 80 schools that joined or rejoined since 2019 deserve more attention than they usually get, because they complicate the narrative in a useful way. Withdrawal is not a one-way ratchet. Teachers' Pensions confirms that a school which has left, or which has gone into phased withdrawal, can apply to rejoin at any time — a phased withdrawal school that rejoins fully would then enrol all eligible staff, including those who had been ineligible under phased withdrawal conditions. Rejoining involves providing a bond, which is a real financial commitment, but the door is not bolted. Some schools have clearly concluded that TPS membership is worth more as a recruitment tool than it costs as an employer contribution — which, with the rate falling in 2027, may become a more common calculation.

Phased withdrawal, in detail

Phased withdrawal is the option most independent schools have actually taken, and it is far more consequential in its detail than its name suggests. In England and Wales it was created by regulation 13A of the Teachers' Pensions (Miscellaneous Provisions) (Amendment) Regulations 2021, which came into force on 1 August 2021 and introduced the concept of an "accepted phased withdrawal school". Scotland followed with its own equivalent provisions, made in 2023 and having effect from 16 August 2023.

The basic mechanism is simple. Teachers' Pensions describes phased withdrawal as freezing membership at a particular point in time: no new entrants are allowed to join, except by way of a bulk TUPE transfer from another participating school. Existing members can remain in the scheme until they leave that employment. The school must provide an alternative pension arrangement — it has no choice about that, since auto-enrolment duties apply regardless — and should consult staff and trade unions before making the change.

That is where the simplicity ends. The eligibility rules around what counts as a "recognised break" at a phased withdrawal school are strict, and they are the part that genuinely catches people out.

The opt-out trap

The sharpest rule in the regime is this: if you opt out of the TPS after your school becomes a phased withdrawal school, you cannot opt back in with that employer. Teachers' Pensions states it directly — staff who opt out after the phased withdrawal date will not be able to opt back into the scheme with you, and should be offered the alternative pension scheme instead. Your only route back into the TPS is employment at another participating establishment.

Understand what that means practically. Teachers opt out of pension schemes for entirely ordinary reasons: a period of financial pressure, a mortgage application where maximising net pay for six months looked sensible, a misunderstanding of what the deduction on the payslip was for. In a normal TPS school, opting out is reversible; you rejoin and carry on. At a phased withdrawal school it is not reversible. A decision taken to solve a six-month cash flow problem can permanently end your defined benefit accrual for the remainder of your career at that school.

There is a narrower and slightly kinder rule for teachers whose opt-out was already in place before the phased withdrawal date. Those members can rejoin later — but Teachers' Pensions notes that at the employer's next auto-enrolment staging date they will be brought back into the scheme, and if they choose to opt out at that point, they lose eligibility to return. So even the pre-existing opt-out is only a temporary reprieve, and it ends at a date the teacher probably is not tracking.

Breaks in employment at a phased withdrawal school

The other rule worth knowing is narrower than most people assume. Teachers' Pensions' employer guidance states that a break in pensionable service of at least one day, other than for specifically recognised reasons, will lead to the loss of future TPS membership at a phased withdrawal school. The recognised exceptions are principally non-pensionable sick leave and non-pensionable family leave — extended maternity leave, for example — and even those only preserve eligibility where the member returns within five years.

• You keep accruing if nothing changes. An existing active member at a phased withdrawal school can normally remain in the TPS until they leave that employment. For a teacher who simply carries on in post, phased withdrawal may change nothing at all about their own pension.
• Opting out after the phased withdrawal date is permanent. You cannot rejoin with that employer. The alternative scheme is what you get instead.
• An opt-out that predates the withdrawal date can be reversed once — but the next auto-enrolment staging date is a hard deadline, and opting out at that point ends eligibility.
• Returning from non-pensionable sick or family leave preserves eligibility for up to five years. Teachers' Pensions gives the worked example of a member leaving service on 1 September 2022 and returning on 1 September 2027 as being within five years; returning on 2 September 2027 would not be.
• A new appointment at the school cannot join, with the exception of a compulsory transfer of employment — for instance, a merger of two schools — where the transferring teacher was previously eligible at a participating school.
• New hires must be offered the school's alternative scheme, and Teachers' Pensions warns employers that members will not be able to transfer their existing pension into it unless it is another defined benefit scheme.
• Break in service also affects your cover, not just your accrual. Teachers' Pensions notes that a member who takes a non-ill-health break or opts out is no longer covered for death in service benefits, and no longer covered for enhanced in-service ill-health benefits.

That last bullet is the one that tends to change people's minds. The conversation about withdrawal is almost always framed as being about retirement income, and it gets treated as a problem for future-you. But death in service cover and enhanced ill-health benefits are protections for present-you and your family, and they switch off at the same moment. A teacher in their thirties with young children and a mortgage may reasonably conclude that this, rather than the projected pension at 67, is the part of the decision that actually matters.

What replaces the TPS

A school leaving the scheme has to put something in its place, and in practice that something is almost always a defined contribution arrangement. The best known in this sector is APTIS, the Aviva Pension Trust for Independent Schools, a defined contribution scheme forming part of the Aviva Master Trust and available to member schools of the Independent Schools' Bursars Association. It is not the only option — schools use a range of master trusts and group personal pension arrangements — but it is the one most often named in consultation documents, and encountering the acronym is often a teacher's first sign that a consultation is coming.

The most informative publicly documented example of how these packages get structured is United Learning's 2023 consultation covering its independent schools. As set out on the NASUWT's page for those schools, the consultation opened in November 2023 for changes effective from September 2024, and presented staff with a choice: remain in the TPS but accept a 3.5% salary reduction, or move to APTIS on either a 20% employer contribution, or a 10% employer contribution combined with a 10% salary uplift. NASUWT characterised the proposal as a fundamental change to teachers' pay and rewards and a worsening of terms and conditions, and pointed out that unlike the TPS, APTIS provides no death-in-service or early retirement benefits.

A 20% employer contribution is, in absolute terms, a strong DC offer — far above auto-enrolment minimums and above what most graduate professions provide. It is a mistake to dismiss it. It is an equally serious mistake to treat it as straightforwardly equivalent to a defined benefit accrual, because the two are not the same kind of thing. The genuine analytical difficulty is that the comparison depends on assumptions — future investment returns, future annuity pricing, your own longevity — that nobody can supply honestly. That is a reason to take properly regulated individual advice, not a reason to guess.

The salary sacrifice option deserves its own scrutiny

Where a school offers "keep the TPS, take a pay cut", the arithmetic needs care. A percentage reduction in pensionable salary is not a one-off cost. It reduces the career-average pension you accrue in that year and every subsequent year the reduction is in place, and it lowers the base to which future pay awards are applied, so the effect compounds. It may also reduce your death-in-service benefit and any salary-linked entitlements. Whether the reduction is applied to pensionable pay or structured as a reduction to a non-pensionable element makes a substantial difference, and it is a fair and specific question to put in writing during a consultation.

The five-year rule that costs the most

If you take one thing from this page, take this. It is not about contribution rates, it is not about which scheme your school offers, and it applies with equal force to a teacher moving from the state sector into an independent school and to one moving the other way.

Teachers' Pensions' guidance on breaks in service states that if you have a break in pensionable service of more than five years, the salary link is broken, and the scheme will use your salaries at the time of the break to calculate your final salary benefits. A break of five years or less preserves the link.

To see why this matters so much, consider what the final salary link actually does. If you have service in the pre-2015 final salary arrangements — which, after the McCloud remedy, includes most teachers who started before April 2012 — those benefits are calculated using your final salary at the point you eventually retire, not the salary you happened to be on when you stopped accruing final salary service in 2015. That link is enormously valuable, because it carries two decades of subsequent pay progression and inflation into the calculation of benefits earned long ago.

Break it, and those pre-2015 benefits are calculated on the salary you were on at the moment of the break. A teacher who leaves pensionable service in 2027 on £48,000 and returns in 2033 has their pre-2015 final salary benefits calculated on £48,000 — not on whatever they retire on in 2045, which after a couple of promotions and eighteen years of pay awards could be a great deal more. The pension attached to that tranche of service is permanently smaller as a result, and no amount of subsequent contribution repairs it.

Two related rules are worth knowing alongside it. First, a break in pensionable service of more than five years after 1 April 2015 means you enter the career average scheme when you return to pensionable service. Second, Teachers' Pensions describes a "disqualifying break" for members with a Normal Pension Age of 60 under the final salary arrangements: a break of more than five years ending after 31 December 2007 means that service before the break keeps its age-60 terms, but service taken up after the break and before 1 April 2015 carries a Normal Pension Age of 65 instead. That is a permanent change to when a slice of your benefits becomes payable unreduced.

The practical instruction that falls out of all this is unglamorous but genuinely valuable: if you are contemplating a move that would take you out of pensionable service, ask Teachers' Pensions for your own position in writing before you resign. Your service record, the split between arrangements, and your relevant dates are specific to you, and a general rule is not a substitute for a scheme confirmation of what applies to your record. This is free to ask for and expensive to get wrong.

The 2027 rate cut changes the argument

This is the part of the story that is still moving, and it is the reason any consultation running now deserves a sharper set of questions than one running two years ago would have done.

The employer contribution rate is reset by periodic scheme valuations. The 2024 valuation has now reported, and in June 2026 Teachers' Pensions confirmed the outcome: the employer contribution rate will fall from 28.68% to 17.68% of pensionable pay, including the administration levy, with effect from 1 April 2027. Teachers' Pensions has confirmed that this does not affect the level of benefits members are building up in the scheme — the valuation determines funding, not entitlement.

An eleven percentage point reduction is not a marginal adjustment. Returning to the £45,000 teacher used earlier, the employer's annual pension cost falls from about £12,910 to about £7,960 — below where it stood before the 2024 rise, and within touching distance of the 23.68% era. For a school with sixty teaching staff, that is roughly £297,000 a year of cost relief on the same arithmetic used above. The specific financial pressure that drove the entire withdrawal wave from 2019 onwards is, on the published rate, substantially reversing.

What we cannot tell you is what schools will do about it. Some considerations cut against a quick reversal: schools that have already completed full withdrawal have restructured their reward packages, communicated them to staff and in many cases banked the savings against other cost pressures, and rejoining requires providing a bond. Valuation outcomes also move in both directions over time, and a governing body that has just spent two years on a painful consultation may be reluctant to reopen it on the strength of a rate that is fixed only until the next valuation cycle. Against that, the recruitment advantage of TPS membership is real and becomes more valuable as fewer independent schools offer it.

If your school is consulting on leaving

Consultation on a change to pension provision is a formal process with a real timetable, and the period during which staff input can actually change the outcome is short. The steps below are about extracting the information you need to form a view, in the order that makes that possible.

One final framing point. It is tempting to treat a pension consultation as a negotiation you either win or lose, and to read a decision to withdraw as a defeat. In practice the outcomes that staff bodies have most often secured are not reversals but modifications: phased withdrawal instead of full withdrawal, a higher employer DC contribution than first proposed, guaranteed group life cover to replace the death-in-service benefit, or transitional protection for staff within a defined period of retirement. Those are worth having, and they are considerably more achievable than persuading a governing body to absorb a cost it has concluded it cannot carry. Knowing which of the technical rules above bite hardest on your own circumstances is what lets you argue for the modification that actually helps you.