The pensions trap: how Britain built a retirement system designed to fail working people

Eighteen million working-age adults in Britain, nearly half of everyone in employment between school-leaving age and the state pension age, are not paying a single penny into a private pension. That is not a campaign group’s figure. It is the headline finding of the government’s own Pensions Commission, stated on BBC Morning Live and confirmed in the Department for Work and Pensions’ supporting evidence.

Among the self-employed, the couriers, the cleaners, the cab drivers, the tradesmen, the small builders, the freelancers a decade of casualised labour has produced, only four percent are saving into a private pension at all. One in twenty-five. The other twenty-four are relying, knowingly or not, on the full new state pension of £241.30 a week. That works out to £12,547.60 a year from April. The Pensions and Lifetime Savings Association’s accepted “minimum” retirement living standard for a single pensioner is £13,400. The state pension, on its own, no longer clears the floor that the industry itself defines as the bottom.

This is the system Britain has built. A two-tier retirement, in which the people who could afford to save did, the people who could not did not, and the political class spent two decades insisting that auto-enrolment had fixed the problem. It has not fixed the problem. It has hidden it.

The auto-enrolment promise, and what it actually delivers

Auto-enrolment was sold to the country in 2012 as the answer to the pensions crisis of the New Labour years. Every eligible worker would be opted in by default; eight percent of their qualifying earnings, three from the employee, four from the employer, one from the taxman in relief, would flow into a workplace pension; the cumulative effect, projected over a forty-year career, would deliver a respectable retirement income.

The Resolution Foundation has now put numbers on what that eight percent actually buys. Under current policy, a typical earner saving from their early twenties on the minimum default contribution can expect to replace fifty-one percent of their pre-retirement income in retirement. The original Pensions Commission of 2005, chaired by Adair Turner, set the replacement-rate target at sixty-seven percent. The Foundation calculates that to hit a “Living Pension”, its own benchmark for adequacy, the minimum contribution rate would need to rise from eight percent to 11.2 percent for someone saving for their entire working life, and to 16.1 percent for anyone starting later.

In other words, the floor the country has been told to stand on is two and a half to eight percentage points too low. Doubled for late starters, it is half what is required. The Institute for Fiscal Studies, in its July 2025 Pensions Review final report, reached the same conclusion from a different direction: around four in ten private-sector employees currently saving into a defined contribution pension are undersaving against the Turner sixty-seven percent target, and thirty-two percent of private-sector workers, not those without pensions, but those with them, are projected to fail to clear the PLSA minimum living standard in retirement.

These are the people the system was supposed to be working for.

The self-employed gap

For everyone outside an employer’s PAYE pipeline, auto-enrolment does not exist. The default is no contribution at all. The Pensions Commission’s own data, surfaced again this week, gives the result: ninety-six percent of the self-employed are saving nothing.

The size of the British self-employed workforce in 2026 is somewhere between 4.2 and 4.4 million depending on how the ONS counts platform work. That implies roughly four million people, almost the population of greater Manchester, on track to enter retirement with no private pension at all. Plus the workers in cash-in-hand sectors who never register, whose existence the official figures do not capture, but whose retirements are equally without provision.

The political response to this has been to commission another review. Liz Kendall, the Work and Pensions Secretary, revived the Pensions Commission in July 2025 under Baroness Drake, Sir Ian Cheshire and Professor Nick Pearce. Their terms of reference instruct them to examine “how to improve retirement outcomes, especially for those on the lowest incomes and at the greatest risk of poverty or undersaving”. Their terms of reference do not instruct them to recommend extending mandatory auto-enrolment to self-employed earnings, or to raise the eight percent floor, or to redesign the National Insurance contribution to fund a higher state pension. They are asked to think about it. They are not asked to fix it.

The cost-of-living amputation

While the structural problem has been ignored for two decades, the cyclical one is now eating into it.

Research published by the Pensions Management Institute, conducted across 2024 and 2025, found that thirteen percent of those saving into a pension scheme had reduced their contributions over the previous twelve months, with a further twenty percent considering doing the same in the months ahead. The Financial Services Compensation Scheme, working a wider sample, put the combined figure at twenty-three percent of pension holders having either decreased their contributions or stopped them altogether. Seventeen percent of those old enough to legally do so have withdrawn money from their pension savings to meet short-term needs. More than three-quarters of workers told the same researchers they were worried that the cost-of-living crisis would damage their retirement plans.

These are not the irresponsible. These are not the financially illiterate. These are the people who have been doing what they were told. They are reducing contributions because their rent has gone up by a quarter in two years, because their energy bill doubled and has not come back down, because the supermarket shop that fed a family of four for £80 a week now costs £140, and because the choice between paying the standing charge and paying into the pension is not really a choice when the standing charge is enforced by a debt-collection process and the pension is not.

BBC Morning Live’s personal finance segment also flagged a further figure that ought to alarm anyone responsible for setting policy: almost a third of those reaching the age at which a private pension can be accessed are doing so at the earliest opportunity. The presenter, Laura Pomfret, attributed the trend to debt, mortgage redemption, and family support. Each of those is a euphemism for the same underlying fact. A pension pot was meant to provide an income in retirement. It is instead being drawn down in middle age to keep the rest of household finance from collapsing.

The state’s response to this, announced in the autumn 2025 Budget, was not to raise contribution rates, extend coverage, or shore up the state pension. It was to cap the National Insurance relief on salary-sacrifice pension contributions at £2,000 per year from April 2029. The cap will fall hardest on the middle earners who use salary sacrifice to make their workplace pensions just about work. It will save the Treasury money. It will not save anyone’s retirement.

The renters’ cliff

The number that most clearly reveals where this is heading is the projection on retirement housing tenure.

In the cohort of pensioners aged sixty-six to seventy today, the IFS reports that only three percent are single private renters. That is the legacy of an earlier housing market. Most current pensioners bought a home in the seventies, eighties or nineties, on a mortgage they cleared before retirement, and have either downsized or stayed put. Their housing cost in retirement is, in effect, the council tax and the upkeep. Their pension income, even where modest, goes most of the way.

That cohort is the last of its kind. Home ownership rates among those now aged thirty-five to forty-four are at their lowest level since 1991. The English Housing Survey records a fall from roughly seventy-two percent ownership in that age band in the early 2000s to fifty-six percent today, with the decline concentrated in the bottom half of the income distribution. The people now in their thirties and forties who do not own, in the absence of an enormous policy correction, will not own in their sixties either. They will retire as private renters.

The IFS modelled what that means for retirement adequacy. Almost ninety percent of those not privately renting in retirement are projected to clear the PLSA minimum standard. Among private renters, just under half are projected to clear it. To put it another way: switching the tenure of a retiree from owner-occupier to renter cuts their probability of avoiding statutory minimum-standard poverty by more than forty percentage points. The Joseph Rowntree Foundation has been tracking pensioner poverty as a headline metric for years; it sat at twelve percent across 2008 to 2011, rose to fifteen percent by 2018, and has remained there. The renters’ cliff is not yet visible in the data because the people walking off it have not retired yet. They will, in numbers, over the next twenty years.

This is the part of the crisis nobody in Westminster is willing to name. The pension system, as currently constructed, assumes the retired do not pay market rent. They will. And the gap between the state pension and the rent on a two-bedroom flat in any part of England outside the cheapest postcodes already exceeds the entire weekly state pension payment.

The state pension floor that is not a floor

The full new state pension rises from April 2026, in line with the triple lock, to £241.30 a week, £12,547.60 a year. The full basic state pension, paid to those who reached state pension age before April 2016, is lower still. Anyone with an incomplete National Insurance record, broken career, time abroad, time caring, time self-employed and not contributing voluntarily, receives less than the full amount.

Compare those figures to the PLSA’s published retirement living standards. The “minimum” standard for a single pensioner, defined as covering food, heat, basic clothing, no car, no foreign holiday, is £13,400 a year. The state pension, on its own, leaves an £850 shortfall against the floor of dignity the industry itself has defined. The “moderate” standard is £31,700. The “comfortable” standard is £43,900. The state pension delivers under forty percent of the moderate figure and under thirty percent of the comfortable one.

For the eighteen million working-age adults with no private pension, and the four million self-employed paying in nothing, the state pension is what they will retire on. The state pension does not clear the floor. This is not an obscure technical point. It is the central fact of British retirement policy, and the political class has chosen not to discuss it.

The state pension age is rising from sixty-six to sixty-seven between April 2026 and March 2028. A further review will recommend whether it should rise again. The minimum age at which private pension pots can be accessed is rising from fifty-five to fifty-seven from April 2028. Every one of those changes shifts the burden in the same direction: working longer for less, accessing less for longer, with no corresponding increase in the value of the state pension as a share of average earnings.

Who benefits from the current system

It is worth asking whose interests are served by leaving the eight percent floor in place, the self-employed uncovered, and the state pension below the minimum living standard. The honest answer has three parts.

First, the Treasury. A higher state pension is a higher annual liability on the public finances. A mandatory increase in auto-enrolment contributions raises the cost of employer pension contributions, which the OBR scores as a drag on labour-market participation and on the corporate tax base. The Treasury’s institutional preference, across both Conservative and Labour governments, has been to defer the cost of pensioner inadequacy until it presents as housing benefit, pension credit and council tax support, items that can be cut individually without anyone calling it a state pension cut.

Second, employers. A move from eight percent to twelve percent auto-enrolment contributions would, on the Resolution Foundation’s modelling, cost the average employer roughly two percent of wage bill, depending on the split. The CBI and the Federation of Small Businesses have lobbied successive governments against any such increase, on the grounds of competitiveness. The result is that the cost has instead been shifted onto the future pensioner, who has no lobby.

Third, the private pensions industry. Defined contribution pensions are a fee-bearing product. The longer the pot, the higher the lifetime fees. A national framework that delivered universal adequacy through a higher state pension would shrink the addressable market for private DC providers. A framework that delivers inadequate state coverage, mandates a low default contribution, and forces individual workers to top up if they want to retire on more than the minimum, maximises the market. The biggest names in the British DC industry, Legal & General, Aviva, Standard Life, NEST itself, have built business models around the system as it stands. Their submissions to successive pensions reviews have favoured incremental tweaks. They have not favoured the structural changes that would actually close the gap.

None of these institutions is conspiring. They are doing what their interests dictate. The problem is that those interests do not align with the interests of the working-age adult who will not be able to retire.

What would actually fix it, and what is not on the table

The IFS, the Resolution Foundation, and a long list of independent analysts have between them mapped out what an adequate British pension system would look like. The components are not contested.

Auto-enrolment minimum contributions would rise from eight percent to at least twelve percent, with the increase loaded onto the employer share. Auto-enrolment would be extended to the self-employed, either through the income tax return or via a default NEST contribution on platform earnings. The state pension would be set as a fixed share of median earnings, proposals cluster around thirty-five percent, against the current figure of roughly twenty-eight, and indexed accordingly. Pensioner housing benefit would be uprated to cover the actual private rent of a two-bedroom property, as the IFS specifically recommended. The pension access age would be held, not raised.

None of these reforms is in the terms of reference of the Pensions Commission revived by Liz Kendall in July 2025. None of them is in the working assumptions of the Treasury’s medium-term fiscal plan. None of them was in the autumn 2025 Budget. The cap on salary-sacrifice National Insurance relief was. The headline increase in the state pension was the standard triple-lock uprating, of which Labour was the inheritor, not the architect.

What is on the table is more reviews, longer working lives, smaller relative state pensions, and a private pensions industry trusted to make up the difference for people whose wages will not cover it.

The political claim, and what it actually means

When ministers and shadow ministers talk about the “pensions crisis”, they tend to mean the rising fiscal cost of the triple lock, the state’s pension bill, not the pensioner’s pension income. The two are the opposite of each other. A high state pension bill, on a national accounts spreadsheet, looks like a problem to be managed downward. A low pensioner income, in a flat with rising rent, looks like a person to be left to manage.

Britain is now committed to the second of those problems on a scale it has not seen since the inter-war period. Eighteen million people not saving. Four million self-employed not covered. A third of those who are saving either reducing their contributions or raiding their pots. A state pension below the minimum living standard. A retirement housing tenure shifting from owner-occupation to private renting at exactly the moment housing benefit is being held flat and Local Housing Allowance is failing to cover rent for the majority of recipients who already need it.

This is not a problem about to be caused by demographic change. The demographic change is already underway. The undersaving is already underway. The renters who will retire into the market are already born, already working, already paying rents that prevent them from saving.

The country has between fifteen and twenty-five years to act before the cohort hits state pension age in numbers. The government has, instead, set up a commission to recommend what to do. The commission has been told what to leave out before it begins. The result, on present trajectory, will be a generation of British working people who paid into the system their entire working lives and retire into a poverty the system was supposed to abolish.

That is the pensions trap. It was not built by accident. It is being maintained by choice.