Estimated Treasury saving from a 40% cut in DB pensions

A 40% cut to defined-benefit (DB) pensions could reduce the Government's annual pension payments by tens of billions of pounds - but the actual saving to taxpayers would be smaller. The crucial distinction is between pensions paid directly by public-sector schemes and pensions paid by private-sector employers.

The latest official figures I found are for the 2024-25 financial year. The Government's Whole of Government Accounts record £60.4 billion in benefits paid by unfunded public-service pension schemes during that year. These include major schemes for civil servants, NHS staff, teachers, police and the armed forces.

A straightforward 40% reduction to that annual payment total would be:

£60.4bn×40%=
£24.16bn

That is the starting point - not yet the net saving to the Treasury.

1. The estimated annual saving

Current annual benefits paid

£60.4bn

40% reduction

£24.2bn

Remaining annual payments

£36.2bn

This is a gross reduction in pension payments, assuming the cut applies immediately to all benefits in those unfunded schemes. It is not a forecast of what a government could legally implement or the final net fiscal saving.

2. What would reduce the saving?

Lost income tax

Pension income is generally taxable, so reducing pensions would also reduce income-tax receipts. The amount depends on each pensioner's other income and tax band.

For illustration, if the tax lost averaged 15-20% of the pension reduction:

Calculation

Annual amount

Gross pension reduction

£24.2bn

Illustrative lost income tax

£3.6bn-£4.8bn

Remainder before other effects

£19.3bn-£20.5bn

That is a modelling assumption, not an HMRC estimate.

Increased means-tested benefits

Some pensioners would have lower incomes and might become eligible for Pension Credit or other support. Pension Credit takes account of other pension income, and eligibility depends on household circumstances.

That additional expenditure would reduce the net saving further. I cannot responsibly put a precise figure on it without modelling pensioners' incomes, savings, household circumstances and the size of their pension cuts.

3. What about private-sector DB pensions?

This is a major distinction. Cutting a pension paid by a private-sector DB scheme would generally reduce the scheme's payments, not directly reduce government expenditure by the same amount. The Treasury might receive less income tax, but the pension payment itself is not normally a government payment.

The figures above therefore concern unfunded public-service schemes, not every DB pension in Britain.

4. What about the Local Government Pension Scheme?

The Local Government Pension Scheme in England and Wales paid £12.4 billion in pensions and annuities in 2024-25, plus £2.4 billion in retirement lump sums. It is funded from scheme assets and contributions.b

A 40% cut to its annual pension and annuity payments would amount to roughly £5.0 billion in lower payments, but that would not translate into £5.0 billion of immediate Treasury savings. The impact would fall principally on the pension funds, their members and sponsoring employers.

5. The overall conclusion

For the unfunded public-service schemes alone, the arithmetic suggests:

£24.2 billion less in gross annual pension payments.

Perhaps £19-21 billion remaining after an illustrative allowance for lost income tax.

A lower net saving after any additional means-tested support and other fiscal effects.

That is a very large potential sum. But it is important not to treat it as a reliable forecast: the £19-21 billion figure is an illustration, not a fully costed policy estimate.

There is also a fundamental distinction between reducing pensions already being paid and changing pension rights earned in the future. A retrospective 40% cut would raise major legal issues and would affect people who made retirement decisions on the basis of the benefits promised to them.