How inflation sets increases to benefits, State Pension and tax thresholds

The State Pension rises each April by the highest of three numbers: the previous September's CPI inflation rate, average earnings growth measured over the summer, or a guaranteed floor of 2.5%¹. That rule, the triple lock, has applied since the 2011/12 financial year, with one break in 2022-23 when it was suspended for a year¹. Working-age benefits such as Universal Credit and Child Benefit follow a different rule: they rise only in line with the previous September's CPI inflation, with no earnings element and no 2.5% floor².

The September inflation figure matters because it is the reference point for both systems. Whichever direction prices have moved by the end of September sets the percentage applied the following April, whether that is to the State Pension, Pension Credit or working-age benefits³. Tax thresholds are the mirror image: when they are frozen instead of rising with inflation, incomes creep above them and tax bills grow, a process known as fiscal drag⁴.

The triple lock: the highest of earnings, inflation or 2.5%

The triple lock is a commitment to uprate the basic and new State Pension every year by the highest of earnings growth, CPI inflation, or 2.5%¹. The House of Commons Library describes it as going beyond a separate statutory requirement to uprate both pensions at least in line with earnings¹. In practice, the three measures are compared each year and whichever is highest is applied to payments from the following April.

The three components are fixed and well established. Which? sets them out as the rate of inflation as of the previous September, average earnings growth as of the previous July, or 2.5%⁸. Full Fact describes the same rule: the State Pension is uprated each year by whichever is the highest of CPI inflation, average earnings or 2.5%². The 2.5% floor is what makes the lock a "triple" rather than a simple link to prices or pay: even in a year when inflation is near zero and wages are flat, the pension still rises by 2.5%⁵.

The lock was announced in the June 2010 Budget and introduced by the Conservative-Liberal Democrat coalition government in 2011⁹. The National Audit Office records that the Government increased spending on the basic State Pension from 2011 by uprating it by the best of average earnings growth, inflation or 2.5%¹⁰. It has been applied every year since, except for the temporary suspension in 2022-23¹.

The effect over time has been substantial. The Office for Budget Responsibility (OBR) notes that the State Pension increased by 18.7% between 2010-11 and 2015-16, considerably more than earnings growth or CPI inflation, thanks to the triple lock¹¹. Which? has found that increases have surpassed inflation in seven of the past 10 years⁸. A Which? policy report from 2019 estimated that the lock leads to the State Pension being uprated annually by 0.36 percentage points more than earnings growth on average¹².

The lock applies to both the basic State Pension, paid to those who reached pension age before 6 April 2016, and the new State Pension paid to those reaching it after that date¹³. The Additional State Pension, part of the old system for people who reached pension age before 6 April 2016, increases in line with CPI instead of the triple lock¹⁴. One further exception is the extra amount earned by deferring a State Pension claim: that boost is excluded from the triple lock and only increases annually in line with inflation⁷.

How the April increase is worked out from September inflation and summer wage growth

The mechanics are straightforward once you know which figures are used. The government uses September's consumer prices index (CPI) measure of inflation, and the three-month average of earnings from July¹⁵. Entitledto summarises the rule as increasing pension age benefits based on the highest of wages, September's CPI or 2.5%¹⁶.

The order of events each year is: earnings growth is measured over May to July, September's CPI figure is published in October, the two are compared with the 2.5% floor, and the resulting percentage is applied the following April⁸. The September reference month is not unique to pensions: it is the same month used to uprate working-age benefits, which is why the two systems move together in years when inflation is the highest measure².

The rises delivered under the lock have varied widely depending on which measure won:

In April 2019 the rise was 2.6%, set by average earnings growth, which came in highest that year¹⁹. In April 2020 it was 3.9%²⁰. In April 2021 the 2.5% floor applied, and the Budget confirmed the State Pension would rise by 2.5% with the triple lock remaining in place²¹. April 2023 saw a 10.1% rise, matching September 2022's CPI inflation²³. April 2024's rise was 8.5%, in line with the earnings growth measure²⁴. April 2025 brought a 4.1% increase to the State Pension and Pension Credit³, and April 2026 a 4.8% rise⁷.

April 2022: when the triple lock was paused

The one break in the lock's history came after the pandemic. Wage statistics were distorted by the end of furlough, as large numbers of people returned to full pay and average earnings growth spiked. The Joseph Rowntree Foundation records that this prompted the Government to pause the triple lock for one year, preventing an 8% rise in the State Pension in April 2022²³. The OBR describes the temporary suspension in 2022-23 as replacing the triple lock with a double lock, the highest of CPI inflation or 2.5%⁹.

The result was a 3.1% uprating in April 2022, at a point when inflation had already reached 9%²². The Resolution Foundation had warned before the decision that following the lock that year would have meant an extra £4 billion being spent on State Pensions due to an unintentional large ratchet effect in the earnings data²⁵. The full triple lock was reinstated for April 2023, when the rise was 10.1%²³.

The episode is the clearest evidence that the lock is a policy commitment rather than an untouchable entitlement. It shows how the choice of reference measure can interact with unusual economic conditions to produce outcomes no one intended, and how governments can and do set the rule aside when the numbers become unpalatable.

What the triple lock costs, and why its future is debated

The lock costs money because it guarantees rises that are, on average, higher than either prices or earnings alone would deliver. The Resolution Foundation calculates that the triple lock has meant State Pension spending is set to be £15.5 billion a year more by 2029-30 than if it had risen in line with earnings growth alone²⁶. Which? reports the same figure as a forecast cost of £15.5bn a year by 2030, three times higher than first expected²⁷. The Resolution Foundation describes this as three times what was initially expected²⁶.

A Which? policy report from 2019 noted that the OBR has predicted that the cost of the triple lock relative to uprating in line with average earnings will eventually reach 1% of GDP¹². Not every year is expensive, though: the OBR found that in 2015-16 the lock actually reduced spending by £0.6 billion relative to uprating by RPI inflation, because CPI inflation had fallen below RPI¹¹.

The long-term question is who pays for it. Analysis by the Institute for Fiscal Studies suggests the State Pension age may need to rise to 74 by the late 2060s to keep the triple lock affordable²⁷. The Pensions Policy Institute (PPI) argued as early as 2017 that the triple lock cannot go on forever and needs to be replaced with a coherent, comprehensive policy clarifying the role of the State Pension in retirement income²⁸. The PPI also observed that what was missing from the debate was any consideration of what the appropriate level for the State Pension is, what it should be linked to, and what it costs²⁸.

There are also distributional arguments. The Resolution Foundation's work on a proposed "triple lock plus" noted that the bulk of that tax cut would benefit richer pensioners, especially those in households with more than one taxpaying pensioner²⁹. Removing the lock would decrease the cost of providing State Pensions, but the PPI notes it would also have implications for pensioner poverty and the amount spent on means-tested benefits such as Housing Benefit¹³. The PLSA, in its Five Steps to Better Pensions report, took the view that the State Pension should protect everyone from poverty and its value should be maintained by keeping the triple lock³⁰.

Where the triple lock stops: living abroad

Whether the lock follows you abroad depends entirely on where you retire. You can claim the State Pension abroad if you have paid enough UK National Insurance contributions to qualify³¹. If you live in the European Economic Area, Switzerland, or a country with a social security agreement with the UK, your State Pension will be uprated every year in line with the triple lock, and Which? reports this continues even when the Brexit transition period ended³². The same source notes the lock applies to those already living in those states, rising each year by whatever is the highest of average earnings, inflation, or 2.5%³².

If you retire in a country without a social security agreement, the position is different: the State Pension is paid but not uprated, so it stays at the level it reached when you left the UK or first claimed it abroad. Which? reports that the State Pension will increase by 4.8% in April 2026 under the triple lock for pensioners in uprated countries⁷, a reminder that the same percentage does not reach everyone with a UK pension.

Two related points are worth checking before a move. If you did not pay National Insurance while abroad, you can check your National Insurance record to see how your State Pension might be affected³³. And the deferred State Pension "extra" boost is excluded from the triple lock and only increases annually in line with inflation, wherever you live⁷.

Working-age benefits follow September inflation, not the triple lock

Working-age benefits are uprated each April in line with September's rate of inflation, and nothing else⁶. There is no earnings element and no 2.5% floor. So in a year when earnings growth or the 2.5% floor is the highest of the three triple lock measures, the State Pension rises faster than Universal Credit, Child Benefit and the other working-age benefits.

The scale of the difference in a single year is clear from April 2025: the State Pension and Pension Credit rose by 4.1% under the triple lock, while working-age benefits rose by 1.7%, pegged to the CPI rate in September 2024³. Entitledto noted that a working-age person on benefits saw their income increase by around £1.50 a week in April 2025³⁴. In April 2024 the gap ran the other way in cash terms only because inflation was high: benefits for working-age people increased by 6.7% from April 2024, in line with September 2023's inflation⁶.

The process by which these rises happen receives little scrutiny. Decisions on uprating are typically announced in the Autumn Statement and then passed through secondary legislation; because that legislation is passed through the negative procedure, there is no requirement for a vote³⁵. The uprating therefore happens largely automatically once the September inflation figure is known.

Benefit freezes and below-inflation rises: the effect on household incomes

The current CPI link for working-age benefits is relatively recent. Between 2016-17 and 2019-20 there was no increase at all in most working-age benefits: the freeze announced in the Summer Budget 2015 by George Osborne, as part of £12 billion of welfare cuts, meant benefits were not increased in April 2016, 2017, 2018 or 2019³⁶. The Resolution Foundation calculates that the four-year freeze delivered a real cut of over 6% to many benefits, and that the freeze as a whole saved the government around £4.4 billion a year by 2019-20, considerably more than the £3.9 billion originally forecast³⁶.

The freeze covered Child Benefit, Universal Credit, non-disability Tax Credits, Housing Benefit limits, Jobseeker's Allowance, Income Support, and Employment and Support Allowance except the support group component³⁷. The human cost was measured by the Resolution Foundation: the average couple with children in the bottom half of the income distribution was £580 poorer in 2019-20 than if benefits had been uprated in line with inflation, and the average poorer single parent was £710 worse off³⁶. The April 2019 freeze year alone meant a £200 hit for an average poorer couple with children and £250 for poorer single parents³⁶.

Even before the freeze, working-age benefits had been held back: they increased by just 1% annually between 2013-14 and 2015-16²⁶. The charity Z2K told a parliamentary committee that this series of freezes and below-inflation increases had pushed the real value of working-age benefits to a 40-year low³⁵. It noted that even after the 10.1% uprating in line with inflation, a single person received £85 a week, and someone under 25 received just £67³⁵. In 2020/21, a single person needed £141 a week after housing costs to be above the poverty threshold³⁵.

The freeze ended in April 2020, when most working-age benefits rose in cash terms for the first time in five years³⁷. Turn2us records that most working-age benefits increased by 1.3% that April³⁸. But the Joseph Rowntree Foundation points out that even though benefits have since been uprated based on inflation, the freeze from 2016 to 2020 means their real value has never caught up²³. Z2K also highlights particular groups, such as unpaid carers and people with no recourse to public funds, who are particularly financially disadvantaged by the current design³⁵.

Frozen tax thresholds: how fiscal drag works

When tax thresholds are frozen rather than raised with inflation, the two systems described on this page pull in opposite directions. The State Pension rises every year under the triple lock, but if the income tax personal allowance stays fixed, more of that pension becomes taxable each year and more pensioners are drawn into paying income tax. This is fiscal drag: incomes are dragged over thresholds that no longer move.

The Resolution Foundation has traced exactly this effect. It noted that the freezes to the personal allowance in place since April 2021 would have meant some pensioners, specifically younger pensioners in receipt of the new State Pension, starting to pay income tax on it from April 2028²⁹. Which? has likewise reported that retirees may need to pay tax on their State Pension from 2027 under threshold freezes³⁹.

The Resolution Foundation has also quantified what a threshold freeze costs in a single year. It calculates that freezing Income Tax and National Insurance thresholds in April 2024, rather than increasing them by 6.7% in line with inflation, raised £8 billion, costing all basic rate employees £270 and pensioners £170⁴. The same analysis found that someone on the median salary of £27,000 would be around £120 worse off, and the biggest loss, in both cash and proportional terms, would fall on someone earning around £13,500⁴. The wider picture is a £40 billion set of tax rises via a six-year freeze in personal and employer tax thresholds⁴.

The interaction with the triple lock is what makes this politically live. The Resolution Foundation examined a proposal known as "triple lock plus", under which the State Pension personal allowance would rise with the triple lock rather than with general thresholds; it estimated around 8 million taxpaying pensioners would see an annual tax cut of around £250 a year by the end of the next Parliament, with the stated aim that no pensioner would ever pay tax on their State Pension²⁹. But it also found the bulk of that cut would benefit richer pensioners, especially households with more than one taxpaying pensioner²⁹. The counterfactual matters too: the Resolution Foundation noted that only people earning between £40,000 and £51,000 would actually be winners from the offsetting changes in April 2024⁴.

Which inflation measure applies, at a glance

The pattern is consistent: September's CPI figure is the reference month for benefits uprating across the board, while the triple lock adds the earnings measure and the 2.5% floor for the State Pension alone. The CPI measure itself, how it is calculated and how it differs from CPIH and RPI, is covered in the guide to CPI and CPIH, and the wider question of what inflation does to your money in what inflation is and how it affects your money.

For the rules on each payment, the guides to benefits in the UK and pensions set out eligibility and amounts, and personal tax in the UK covers how the personal allowance works. If you are trying to work out whether you qualify for means-tested support as prices rise, Pension Credit is worth checking separately: Which? has reported that millions of pensioners are missing out on Pension Credit they are entitled to claim²⁰. Free, impartial help is available from MoneyHelper for pension questions and from the Turn2us benefit checker for entitlements.