Is the triple lock being scrapped? What Burnham's 2030 change means for pensions and care

Sun, 04 Oct 2026 00:00:00 GMT

My verdict: Partly right. Ending the triple lock's ratchet is the right reform, and pensioners stay protected against inflation. But the saving is small until the mid-2030s, and cancelling spending that was never paid for does not fund a new service.

In brief

The triple lock raises the state pension each April by the highest of earnings growth, inflation or 2.5%. The OBR says it has cost about three times what was expected.

From April 2030 the pension would still rise by at least inflation or 2.5% a year, but would only keep pace with earnings over time rather than every year.

The government expects to save £15bn a year by 2039-40. The IFS puts the range at £4bn to £20bn, and expects about £4bn a year in 2034-35.

Free personal care in England would cost about £7.5bn a year on one estimate, far more than the pension change saves in its early years.

I think the pension change is right, but the law should name the earnings level it protects, and the care service needs money of its own.

What Burnham announced

In his first conference speech as Prime Minister, on 29 September, Andy Burnham promised to leave the triple lock unchanged for the rest of this Parliament. From April 2030 he would "adjust it", and use the savings to build a National Care Service in England.

Under the new rule, the state pension would still rise every year by at least inflation or 2.5%. It would also rise by "anything more that is needed" to keep its value relative to average earnings. What goes is the automatic rise with each year's earnings growth.

The Department for Work and Pensions published its estimate of the savings the same day. It expects £15bn a year by 2039-40 and £50bn a year by 2049-50, in cash terms. Downing Street said the care service would be introduced in phases in the next Parliament, as the savings grow. It also said the government would legislate for the pension change in this one.

Writing in the Telegraph on 1 October, Burnham called the change "a better deal for pensioners", The National reported. He accepted that the state pension "will not increase as quickly" as it would under the current lock. On 2 October the Guardian reported that guidance published with the speech commits the government to legislate first, so MPs will vote on the change in this Parliament.

How the triple lock works now

The triple lock was introduced by the coalition government in its first Budget after the 2010 election, and has applied since 2011-12. Each April it raises the basic and new state pension by the highest of earnings growth, CPI inflation or 2.5%. The law only requires a rise at least in line with earnings. The triple lock is a political commitment that goes beyond that.

The rises have been large lately. After the lock was suspended for 2022-23, because pandemic recovery had distorted earnings, pensions rose by 3.1%. They then rose by 10.1% in April 2023 and 8.5% in April 2024. This April they rose by 4.8%, taking the full new state pension to £241.30 a week for over 12 million pensioners.

The lock's cost comes from what the IFS calls the ratchet. In a year when inflation beats earnings, the pension rises with inflation. When earnings catch up, the pension rises with earnings too. Each time, it moves further ahead of wages and never gives the gain back.

That has been expensive. The Office for Budget Responsibility, the official forecaster, says the triple lock will have added £15.5bn a year to spending by 2029-30, compared with rises in line with earnings. That is about three times the £5.2bn expected when it began, because the parts other than earnings were triggered in eight of its first 13 years.

The state pension now costs £154bn a year. The IFS estimates that spending is £16bn a year higher than it would have been without the triple lock.

The lock did what it was designed to do. The basic state pension fell from 26% of average full-time earnings in 1979 to about 16% in the 2000s, after its link to earnings was cut in 1980. The full new state pension is now about 30% of median full-time pay, according to the Pensions Commission. Median pensioner income, after housing costs, is now similar to that of people below pension age. Thirty years ago it was 20% lower.

What changes in 2030

The IFS has set out how the new rule would work. Each April the pension would rise by the highest of CPI inflation, 2.5%, or whatever is needed to match average earnings growth since the new policy began. So it still has three parts, and the government still calls it a triple lock.

The difference shows after a bad patch. Suppose earnings grow by less than inflation for a few years. The pension still rises by inflation or 2.5%, so it moves ahead of wages, as now. When wages recover, the current lock would let the pension rise with them from its new, higher level. The new rule holds it to inflation or 2.5% until wages have caught up, then tracks earnings again.

The gain becomes temporary. The IFS calculates that if the new rule had applied since 2011, spending this year would be £9bn lower, more than half the £16bn extra cost of the current lock. The pension would still have risen by 6% in real terms since 2010.

The IFS called the reform "much improved but not perfect". Its complaint is the 2.5% floor, which it calls arbitrary and potentially costly. The floor does nothing to protect against inflation and can mean many years of higher spending, mostly when inflation is low.

The government has not said which measure of earnings the pension will track, or at what level. The DWP says it will keep its "record high value relative to earnings". Downing Street's example is around a third of average earnings by 2030-31, while Labour has said about 30%.

The rule also needs a change in the law. The current requirement to raise the pension at least in line with earnings each year sits in legislation. In a catch-up year, the new rule would raise the pension by less than earnings.

How much it would save

The government's figures are £15bn a year by 2039-40 and £50bn by 2049-50. In today's prices those are £11bn and £30bn. The DWP counts only the direct saving on pensions, not knock-on effects on tax or debt interest. For scale, £11bn is about 7% of what the state pension costs this year.

Nobody can be confident about any of these numbers, because the saving depends on how volatile inflation and wages turn out to be. The IFS says the 2039-40 saving could be anywhere from £4bn to £20bn in today's money. For 2034-35 it expects about £4bn, and says the reform "could easily save nothing" if earnings grow strongly.

The Resolution Foundation tested the new rule against past decades. In a decade like the 1990s or the 2000s it would have saved nothing at all. In a decade like the 2010s it would have saved about £24bn. Earlier this year the OBR modelled a similar switch and found no significant saving until 2034, according to BBC Verify.

The long-term effect is clearer. In July the OBR projected that, with the triple lock, state pension spending would rise from 5% of national income in 2030-31 to 9% by 2075-76. With rises in line with earnings, it would reach around 7%. It assumed the triple lock would continue beyond this Parliament in the absence of a firm alternative.

Can it pay for the care service?

Care in England is means-tested. People with assets above £23,250, including their home, get no help from their council. The government says about three in four adults over 65 will need care, and one in seven will face costs above £100,000.

The National Care Service would provide free personal care, such as help with eating, bathing and using the toilet, based on need. It would not cover bed and board in a care home, which would stay means-tested. In his speech Burnham promised "no care charges" paid from the basic state pension, and a system that protects homes and savings.

The price depends on the design. The Health Foundation estimates a Scottish-style system of free personal care would cost about £7.5bn a year, and a universal, comprehensive service about £18.5bn extra by 2036. A lifetime cap on care costs of about £86,000 would cost about £4bn. Sir Andrew Dilnot, whose commission first proposed a cap, says an NHS-style system could take decades to build.

Set against those costs, the pension saving of about £4bn in 2034-35 looks small. The IFS makes a more basic point. The pension change prevents unfunded increases in spending from 2030. Cancelling an increase nobody had paid for does not free up money, so tax rises or other cuts will still be needed.

To give a sense of scale, the IFS says raising £10bn a year would take about a one-point rise in VAT, or in every rate of income tax. The government says the service will be "fully funded – and not through borrowing", with Baroness Casey's commission due to recommend how in summer 2027. Before his speech, Burnham declined to rule out tax rises to pay for it.

Paying for care has defeated several governments. As health secretary under Gordon Brown, Burnham proposed a compulsory contribution from people's estates, which opponents called a "death tax". The Health and Social Care Levy, a rise in National Insurance from April 2022, was expected to raise about £13bn a year. It was reversed that November.

What it means for pensioners

Nothing changes before 2030. Average total earnings grew by 3.9% in May to July, the figure that will very likely set next April's rise. The IFS says that would take the full new state pension to £250.70 a week, or about £13,000 a year. Downing Street says the current lock will add over £2,000 to the pension over this Parliament.

That takes the full pension above the personal allowance, the income on which no tax is paid, which is frozen at £12,570 until April 2031. The Treasury says anyone whose only income is the full basic or new state pension will not pay income tax this Parliament, according to the i Paper. Burnham's speech confirmed that low-income pensioners would not be dragged into income tax this Parliament.

After 2030 the pension would rise more slowly than under the current lock in some years. The DWP projects that pensioner poverty after housing costs would fall from around 14% in 2024-25 to around 8% by 2049-50 under the new rule. It has not published the same projection for the current lock, so the figures do not show how much of that fall the change gives up.

The care service would cover England only, but the pension change applies across Great Britain. Scotland has had free personal care for over-65s since 2002. Scotland's First Minister, John Swinney, says an official estimate shows Scottish pensioners losing £4bn a year by 2050, to pay for care in England.

The politics

Burnham has turned the triple lock into an election issue. Because MPs will vote on the change before the election, every party will have to say in its manifesto whether it would keep or repeal it.

The Conservative leader, Kemi Badenoch, says the triple lock is her party's policy and is in its draft manifesto. She says his sums don't add up, and that the real problem is the welfare bill for people of working age who are not working. Jeremy Hunt and Tom Tugendhat have each suggested the current lock is unsustainable.

Reform UK's Nigel Farage accused Burnham of launching an offensive against older people. The Liberal Democrat leader, Sir Ed Davey, said care could not wait for another election, nor be funded "from the pockets of our poorest pensioners". Some Conservative and Liberal Democrat MPs are unhappy with their leaders for defending a policy they think costs too much. On the left, Unite's Sharon Graham called the change "electoral suicide".

The early polling is on Burnham's side. In a YouGov snap poll on 29 September, 48% supported changing the triple lock to help fund a care service, and 28% opposed it. Asked about the change alone, 43% supported it and 30% opposed it.

What to take from it

The triple lock is not being scrapped. From 2030 it would become a rule that protects pensioners against inflation, keeps a 2.5% minimum, and tracks earnings over time rather than every year. That ends the ratchet that has made the triple lock cost three times what was expected.

The government has sold two policies as one. The pension change saves money slowly and uncertainly. Free personal care would cost more than it saves until the mid-2030s at least.

What I'd do instead

The new pension rule is close to right and needs two fixes. The care service needs its own source of money, named before MPs vote on the pension change.

Write the earnings benchmark into the bill

Name the earnings measure and the starting level the pension must keep pace with, and publish the figure every year, so anyone can check the promise is being kept.

How it would work: The bill fixes the full new state pension as a share of the ONS's average weekly earnings, at its April 2030 level. Each year's uprating statement publishes the share and any catch-up still owed.
Who it affects: The more than 12 million people now receiving the state pension, and everyone retiring after 2030, whose pension's long-run value depends on this one number.
What it costs: Nothing beyond the policy as announced. It makes the promise checkable without changing what it pays.
Tried before?: In part. The law already requires the state pension to rise at least in line with earnings. The triple lock is a political commitment on top of that, which this bill would replace with a written rule.
The risk: A fixed share locks in the 2030 level, which the triple lock has pushed to a record. A future government wanting a different level would need a new law.

Drop the 2.5% floor and keep the inflation guarantee

Keep protection against inflation and the long-run link to earnings, but end the 2.5% minimum rise. This is the smoothed earnings link the IFS and the Resolution Foundation have proposed.

How it would work: From April 2030 the pension rises by the higher of CPI inflation or the amount needed to keep pace with earnings since 2030. The 2.5% minimum goes.
Who it affects: Pensioners in years when inflation is low and the pension is already ahead of earnings. Their pension would still never fall in real terms.
What it costs: It saves money in low-inflation years. Neither the government nor the IFS has published a costing against the adjusted lock. The IFS says the floor could mean many years of higher spending.
Tried before?: No UK government has used one. The IFS Pensions Review and the Resolution Foundation, in June 2026, both recommended a smoothed earnings link.
The risk: Opponents would call it a cut, and in some years it would pay less than the government's plan. A run of low inflation would mean small cash rises.

Name what pays for the care service before MPs vote

Publish the cost of each phase of the National Care Service, and the tax or spending choice that covers it, alongside the pension bill. Stop presenting the pension saving as the money.

How it would work: When the pension bill is introduced, the Treasury publishes a costing for each phase of the care service and the revenue that pays for it, for the OBR to assess.
What it costs: Free personal care would cost about £7.5bn a year and a universal service about £18.5bn, on Health Foundation estimates. The IFS expects the pension change to save about £4bn in 2034-35.
Who it affects: Taxpayers, who would fill the gap. The IFS says £10bn is roughly a one-point rise in VAT, or in every rate of income tax.
Tried before?: The Health and Social Care Levy, a 1.25 point National Insurance rise for health and care, started in April 2022 and was reversed that November.
The risk: Naming a tax before an election hands opponents a target, as the 'death tax' row over Burnham's own plan as health secretary showed. Not naming one leaves the promise unfunded.

Two Governments: Social care