Executive Summary
The UK spent £313 billion on welfare in 2024/25, 24% of all government spending and the largest single departmental budget.1 Public debate about that figure is badly skewed. It focuses on the least powerful recipients (disabled people, unemployed claimants, single parents) and ignores the dominant fact: nearly half the budget goes to pensioners, through a triple lock that costs £12 billion a year more than linking pensions to earnings alone, in a country where 4 million children live in poverty.
This pillar makes four arguments. First, the UK welfare system is skewed, by electoral arithmetic rather than by need, toward the generation that also holds most of the UK's £10.8 trillion in household wealth. Second, the sharp rise in disability and mental health claims is being misdiagnosed as fraud and dependency, when the data shows a public health crisis rooted in NHS waiting lists, insecure work and poor housing. Third, child poverty at 27% is not an act of God but a policy choice, sustained for nine years by a two-child limit that affected 1.75 million children in its final year and cost only about £3 billion a year to abolish. Fourth, Universal Credit, the largest working-age benefit, pays a standard allowance worth just 12.5% of median earnings, among the least generous in comparable economies.
The international comparison is damning but clarifying. The UK does not have an over-generous welfare state; on most working-age measures it is below the OECD average. What it has is a welfare state whose make-up reflects the voting power of its recipients more closely than the spread of need. Until the electoral system that produced this is reformed, welfare will keep drifting from its stated purpose. This pillar and the Political Renewal pillar are the same argument.
Key Proposals
Replace the triple lock with a smoothed earnings link. The state pension rises with earnings, with inflation-year shortfalls made good later — protecting pensioner living standards while ending the asymmetric ratchet that costs £12 billion a year above earnings indexation. Enhanced Pension Credit protects the poorest third throughout the transition.Government: partly adopted · announced 28 October 2026See how Budget 2026 relates to this
Set a statutory child poverty target. A binding trajectory to 10% by 2035, independently monitored, alongside restoring Local Housing Allowance to the 30th percentile of actual local rents.
Introduce a statutory employer return-to-work obligation. Employers fund and manage structured return-to-work for the first twelve weeks of sickness absence, modelled on the Norwegian system.
Build an NHS–DWP waiting list bridge. People on waiting lists for conditions preventing employment get active labour market support while they wait, with NHS treatment prioritised by employment impact.
Raise the Universal Credit standard allowance. To 15% of median earnings within the first parliament, on a pathway to 18% over ten years.
Remove the benefit cap for households with dependent children.
1. The Diagnosis
The UK welfare system isn't failing because it is too generous; on most working-age measures it is less generous than its OECD peers. It is failing because electoral arithmetic, not social need, has shaped what it pays for, and because the political conditions that would allow reform are themselves produced by the system that needs reforming.
The pattern is clear: the system pays out most reliably to those who vote most reliably, and is funded by those who vote least. At the 2024 general election, an estimated 73% of over-65s voted, against 37% of 18–24-year-olds and 41% of 25–34-year-olds.3 The triple lock, the long-universal Winter Fuel Payment, and the protection of pensioners from council tax rises all follow the same arithmetic. None of it was planned as a conspiracy. It grew out of political incentives that reward attention to high-turnout groups and punish attention to low-turnout ones.
1.1 Where the £313 Billion Actually Goes
The public argument about welfare is carried on almost entirely through three stock figures who rarely appear in the spending data: the jobless claimant who could work but won't, the woman having children she can't afford, and the immigrant drawing on a system they haven't paid into. DWP publishes exactly what the money is spent on, benefit by benefit.4 The data supports none of the three.
The largest single item, the State Pension, is covered in full in the Public Spending Overview and in its own pillar, the State Pension Explained. It isn't broken down again here, to avoid giving a third, slightly different figure for the same thing. (DWP's own accounts put it a little below the £146.5bn used on those pages, for the same kind of reason as the £313bn/£383.9bn difference explained above; it is not a new error.) Everything else in the £313bn does need breaking down, because this is where the question of who is actually claiming is settled:
View underlying data as a table
| Category | £bn, 2024/25 | What it actually is |
|---|---|---|
| Universal Credit | £66.7bn | £22.2bn standard allowance, £20.6bn housing costs, £12.6bn child element, £7.3bn health element (paid to claimants formally assessed as unable to work), £1.5bn carer element, £1.5bn disabled child element, £0.6bn childcare |
| Personal Independence Payment | £25.9bn | Disability support, paid whether or not the claimant works: £21.8bn to working-age claimants, £4.1bn to pensioners |
| Legacy Housing Benefit | £14.6bn | Claimants not yet moved to Universal Credit, mostly pensioners |
| Child Benefit | £13.3bn | Universal, paid whether or not parents work |
| Employment and Support Allowance | £12.3bn | Older incapacity benefit, being moved into Universal Credit's health element |
| Attendance Allowance | £7.8bn | Disability support for pensioners only, by definition |
| Disability Living Allowance | £7.7bn | Older disability benefit, mostly children (£4.6bn) and pensioners (£2.4bn) still on the old system |
| Pension Credit | £6.0bn | Means-tested top-up for the poorest pensioners |
| Carer's Allowance | £4.2bn | Paid to 1.3 million people caring for a disabled relative |
| Jobseeker's Allowance | £0.31bn | 68,000 claimants |
These categories add up to £159bn. The State Pension, covered elsewhere, makes up most of the rest of the £313bn, with Northern Ireland's social security system and a long tail of smaller benefits (Winter Fuel Payment, Maternity Allowance, Bereavement Support and others) making up what is left.4 Add Pension Credit and Attendance Allowance, both paid only to pensioners, to the State Pension (£146.5bn + £6.0bn + £7.8bn = £160.3bn), and pensioners receive slightly more than half of the whole £313bn before a single working-age claim is counted. That total is a little above the £150.7bn (48%) headline used earlier, which comes from DWP's own narrower accounting of pensioner spending rather than this page's addition of three line items. Both are legitimate ways of drawing the boundary, and neither is more precise than the other.
The "jobless claimant who could work" costs £0.31bn. Jobseeker's Allowance, paid to people who are unemployed, judged capable of work and required to look for it, has 68,000 claimants and cost £313 million in 2024/25.4 That is one-tenth of one percent of the £313bn total, and it is close to the whole idea the public debate is built around. Universal Credit, which has absorbed almost everyone who would once have claimed Jobseeker's Allowance, serves a quite different population: people already in work receiving a top-up (the £22.2bn of standard allowance goes to a caseload that includes many employed claimants), single parents, carers, and £7.3bn paid specifically to claimants DWP itself has assessed as unable to work or to prepare for work. DWP does publish a finer breakdown of Universal Credit by work-search requirement (how many claimants must look for full-time work, and how many are exempt for health, disability or caring reasons), but it sits behind an interactive query tool (Stat-Xplore) this project could not extract in this pass. It is tracked in the Gaps Register as an open item, because it would sharpen this picture further, not soften it.
The "large family" was 496,000 households. The two-child limit, which restricted the child element of Universal Credit and Child Tax Credit to a family's first two children and was abolished from April 2026, affected 495,990 UC households in its final year, covering 1,752,200 children.5 Of those households, 64% had exactly three children, not the six or eight sometimes implied in political rhetoric; 8% had five, and 4% had six or more.5 Against the roughly 4 million UC households with children, families with three or more children were a minority, and the limit actually restricted 78% of that minority; the rest qualified for an exception (multiple births, adoption, non-consensual conception).5 Ending the limit costs about £3bn a year. The population it targeted was never the mass of claimants the debate assumes.
The immigration claim needs a legal starting point most of the debate skips. Most temporary UK visas, including work, study and family routes under Appendix FM, carry a "No Recourse to Public Funds" condition by default.6 NRPF is not a guideline but a legal bar. It blocks Universal Credit, Pension Credit, PIP, Attendance Allowance, DLA, Carer's Allowance, Housing Benefit, Income Support, income-based ESA and homelessness assistance outright.6 A recent migrant on a standard work or family visa is legally unable to claim almost everything in the table above; only settled status, British citizenship or refugee status lifts the condition. Child Benefit is a specific, narrow exception that NRPF does not bar.6 The one figure DWP does publish is nationality on Universal Credit: in June 2025, 1.3 million claimants, 16.4% of the 7.9 million caseload, were non-UK/Irish nationals who had passed the Habitual Residence Test.7 That definition is narrower than "immigrant" in everyday use. It excludes naturalised British citizens born abroad and includes long-settled EU nationals with British-born children. And it covers only Universal Credit, roughly half of working-age welfare spending; DWP has not published nationality breakdowns for PIP, Housing Benefit, Child Benefit or legacy tax credits. So the 16.4% cannot be applied to the £313bn total, or even to all working-age spending, without inventing data. Any claim that "X% of the welfare bill goes to immigrants" that is not explicitly limited to Universal Credit should be treated as unsourced. The rest of the system, and how the share of foreign-born claimants compares with their share of the population and workforce, is tracked in the Gaps Register.
2. What the State Pension Actually Is: A Misunderstood Transfer
Before looking at the triple lock, there is a more basic point that almost never gets made in public debate, and without it the political argument about pensions rests on a misunderstanding.
This is the basic difference between the state pension and a private or workplace pension fund. A private fund takes contributions, invests them in shares, bonds and property, and pays out the returns as retirement income. The money exists; it has been invested and has grown. When you draw on it, you are drawing on accumulated capital.
The state pension has no such assets. It is pay-as-you-go: a direct transfer from today's workers to today's retirees. "I paid in all my life and I deserve my return", a line heard constantly in debates about the triple lock, describes a private pension, not the state pension. What today's pensioners paid in funded their parents' generation's retirement. What they receive is funded by their children's and grandchildren's earnings.
Norway shows what a funded alternative looks like. It took North Sea oil revenues broadly comparable to the UK's and invested them in the Government Pension Fund Global, now worth over $2.2 trillion (about £1.7 trillion), whose returns help pay for Norwegian public spending. The UK took comparable revenues and spent them, and the state pension has no equivalent asset behind it. This is not a side issue. It is the central story of how the UK has treated different generations in its public finances over the past forty years.
3. The Triple Lock: How It Works
3.1 What It Does
The triple lock raises the state pension each year by whichever is highest of average earnings growth, CPI inflation, or 2.5%. In practice the pension rises faster than wages in most years, and the 2.5% floor guarantees a real-terms rise even when the economy stagnates. The cumulative effect is that state pension spending keeps rising as a share of GDP. The OBR's latest baseline projection has state pension spending rising from about 5% of GDP today to around 9% by 2075-76, driven by an ageing population and the triple lock together; the triple lock accounts for around a third of the rise.8
3.2 The Steel Man
The triple lock has achieved something real. Before 2011, pensioner poverty in the UK was among the highest in Europe. The triple lock and the new flat-rate state pension together reversed this: pensioner poverty fell from about 29% in the late 1990s to 14% in 2024/25, after housing costs. The policy answered a real need, after pensioners' incomes had been eroded by decades of uprating in line with prices only, and it delivered measurable results. Any reform proposal that ignores this history is not serious.
The triple lock costs about £12bn a year more than a smoothed earnings link would have, and the OBR expects its extra cost to reach about £15.5bn a year by 2029-30, three times the original estimate.10 It also helps drive the OBR's projected rise in state pension spending from about 5% of GDP today to around 9% by 2075-76 (around 7% if the pension rose with earnings instead).8 It has also achieved something real: UK pensioner poverty fell from roughly 29% in the late 1990s to 14% in 2024/25. Yet 1.7 million pensioners still live in poverty9, disproportionately women with broken employment histories, ethnic minority pensioners and renters, while the same formula gives its full annual increase to every pensioner regardless of wealth, including the majority who are not poor.
A mechanism that once closed a real poverty gap has kept ratcheting up long after that gap closed for most pensioners. This connecting argument is this project's own; no single source states it. Because the triple lock applies to everyone rather than by need, more and more of its rising cost protects living standards that were never at risk, while the 4 million children in poverty have no equivalent mechanism at all. The same political arithmetic that produced the triple lock's real achievement now produces an increasingly badly targeted transfer, because pensioners vote and children don't.
The causation above fits more than this pillar's proposal. One option is to keep the triple lock exactly as it is, on the Steel Man's grounds that reform puts at risk the vulnerable minority still in poverty. Another is to go further and link the pension to prices alone. A third is the Green Party's narrower reform, removing only the 2.5% floor while keeping the higher-of-earnings-or-inflation guarantee. A fourth is the Labour government's September 2026 plan: from 2030, rises in line with inflation or 2.5%, with extra rises when needed to hold the pension's value against earnings.11 The fifth is this pillar's proposal: a smoothed earnings link, as recommended by the IFS Pensions Review.
Choosing a smoothed earnings link plus a stronger, targeted Pension Credit over full abolition or the status quo rests on a value judgement this pillar makes openly: protecting the poorest pensioners' living standards matters enough to keep a dedicated mechanism for them, but protecting everyone regardless of wealth is no longer justified once its cost is set against unaddressed child poverty in the same budget. A reader who values the stability and simplicity of one universal formula above precise targeting, or who thinks means-testing brings stigma and administrative failures the current universal system avoids, is not wrong on the evidence. They are weighing the same facts against a different view of who should bear the risk of mistargeting.
Replace the triple lock with a smoothed earnings link, under which the pension rises with earnings and any year in which inflation outpaces earnings is made good later, while a stronger Pension Credit protects the poorest third through a ten-year transition monitored by the OBR. The falsification test below sets out what would show that this particular transition, not pension reform in general, is failing the pensioners it is meant to protect.
3.3 The Arithmetic Between Generations
The triple lock is paid for by working-age taxpayers and given to pensioners regardless of wealth. Many of those receiving it saw house prices rise far faster than earnings over their working lives (roughly doubling in real terms since 1990 alone), retired on defined benefit pensions now largely closed to younger workers, and benefited from strong stock market returns. Many of those paying for it face stagnant real wages, a housing market that has put home ownership out of reach in most cities, and the loss of the defined benefit pensions their parents relied on.
4. The Disability and Mental Health Crisis: Misdiagnosed
Working-age spending on health-related benefits rose from 1.3% of GDP in 2019 to 1.7% in 2023/2414, an increase of £12 billion in four years, with a further £15 billion rise forecast by 2028. One in ten working-age adults now claims a sickness or disability benefit. The political response has been consistent: treat it as a problem of fraud and dependency, to be solved with stricter eligibility, tougher assessments and lower payments. The data does not support that diagnosis.
4.1 What the Data Shows
Before the pandemic, in 2019, UK spending on incapacity benefits was 1.3% of GDP. Since then the number of claimants has risen in the UK while staying flat or falling in most comparable countries, according to the IFS, which found increases across every age group and most major conditions. The rise reflects a real deterioration in the health of working-age people. The number of people claiming PIP in England and Wales almost doubled, from 2.05 million in January 2019 to 3.93 million in January 2026, and the share aged 16–29 rose from 14.6% to 16.6%.15 Psychiatric conditions are now the main condition for 43% of working-age PIP claimants in England.16
4.2 The NHS Waiting List Connection
In 2024 there were 7.6 million cases on NHS waiting lists in England, covering about 6.4 million people. People waiting years for mental health treatment, for musculoskeletal surgery or for a cancer diagnosis often cannot work, and claim disability benefits while they wait. The OBR and IFS have both identified clearing waiting lists as one of the most cost-effective long-run welfare savings available, not because it removes entitlement but because it treats the condition stopping people working. No single department can solve this on its own.
4.3 The International Comparison
Countries with strong employer sick-pay obligations and structured return-to-work requirements tend to keep more people with health problems in work, not because they are harsher to disabled people but because they step in earlier, while the person still has a job, before they drift into long-term inactivity. The Netherlands is the clearest case. Employers must pay at least 70% of wages for up to two years of sickness and run a return-to-work plan from the early weeks, and after these reforms in the 2000s new disability claims fell sharply. Norway, the model for this pillar's P3, pays full sickness benefit for up to a year, with the employer covering the first 16 days and required to agree a follow-up plan early in the absence. Norway's overall disability spending is high, though, so the strongest evidence on caseloads is Dutch. The lesson from both is that design matters more than generosity.
5. Child Poverty: A Policy Choice, Not a Natural Condition
In 2024/25, 4.0 million children, 27% of all children in the UK, lived in poverty after housing costs.17 Around seven in ten of them lived in working families. (DWP changed its method in 2026 to use benefit records instead of survey answers; on the old basis the 2023/24 figure was 31%, 4.5 million.) Child poverty is not mainly caused by parents not working. It is caused by wages that are too low, housing costs that are too high, and in-work support that is not enough.
The UK's child poverty rate is among the worst in comparable wealthy economies. On UNICEF's like-for-like measure (children in households below 60% of median income, 2019–21), the UK stands at 20.7%, against 15.5% in Germany, 13.5% in the Netherlands, 10.1% in Finland and 9.9% in Denmark. The UK's rate rose by about a fifth over the decade, ranking it 37th of 39 rich countries on progress.17 (The UK's own after-housing-costs figure is not comparable with other countries' rates, so it is not used here.) These are not small differences. They mean millions of children in comparable countries growing up with enough to eat, stable homes and the resources to develop, while their counterparts in Britain do not.
The two-child limit shows clearly how the welfare system's priorities are set. It was introduced in 2017, kept for nine years, and costs about £3 billion a year to abolish on the government's own figures (campaigners had put it at £3.4 billion).18 That is roughly a quarter of what the triple lock costs above earnings-uprating each year. It was kept through nine years of mounting evidence on child poverty, on the grounds that "taxpayers should not be expected to fund large families", and has now been abolished from April 2026. During those nine years poverty among children in large families rose substantially, as the result of a recorded policy choice, not an inevitability.
6. Universal Credit: Built to Be Inadequate
Universal Credit cost £66.7 billion in 2024/25, making it the largest working-age welfare programme.4 The design idea was sound: merge six older benefits into one payment, with a taper meant to make work pay more reliably than the old system. The implementation and the level of payments are not.
The standard Universal Credit allowance for a single person aged 25 or over was worth 12.5% of median earnings in April 2024, among the least generous working-age rates in comparable economies.19 Support for rent, through the Local Housing Allowance, was frozen from 2020, restored to the 30th percentile of local rents in April 2024, and frozen again from April 2025, so it is falling behind actual rents once more. Claimants make up the difference out of an allowance already set at 12.5% of median earnings.
7. The Electoral Mechanism: Why the System Reproduces Itself
The welfare system's tilt toward older voters is kept in place by incentives, not malice. In 2024, about 73% of over-65s voted, against around 40% of under-35s.3 Under first-past-the-post, where marginal seats are won by a few thousand votes, the arithmetic is simple: protecting the triple lock wins more votes than it costs. Restricting in-work benefits or freezing the Local Housing Allowance costs votes mainly among people who either do not vote or live in safe seats.
8. Social Care: The Bill Hidden Inside Another Bill
This pillar treats social care as part of the welfare system because its funding failure cannot be separated from the welfare state's wider failure to allocate resources. The history of social care reform is the clearest example of the problem this pillar diagnoses.
Tony Blair's government set up the Royal Commission on Long Term Care in 1997. The Dilnot Commission reported in 2011. The Care Act 2014 legislated a version of its reforms. Implementation was delayed in 2015, promised again in 2021, delayed to 2025, and cancelled in July 2024. The means-test threshold has been frozen at £23,250 since 2011; uprated for inflation it would be about £35,000. Care now costs an average of £5,000 a month.
The politics of this failure are simple: social care reform costs money up front, and its benefits are spread across several departments, many years and several elections. In a political system that hands out credit by department and accountability by parliament, that calculation keeps producing inaction.
9. Counter-Arguments
'Pensioners earned their state pension through National Insurance. It is deferred pay, not welfare'
The state pension is not a funded scheme. National Insurance contributions do not sit in a ring-fenced investment fund. Today's pensions are paid from today's workers' contributions and general taxation: it is pay-as-you-go. The contributory framing matters politically but misleads economically. More importantly, it does not explain the triple lock. The legal minimum is earnings-uprating. The triple lock's extra cost, £12 billion a year, is not a deferred entitlement. It is a political choice made every year by governments that depend on the pensioner vote.
'Cutting disability benefits will force sick people into poverty'
Correct, which is why this pillar does not propose cutting disability benefits. It proposes treating the causes rather than managing the symptoms: clearing NHS waiting lists, employer return-to-work duties, active help back into work modelled on the Netherlands and Norway, and investment in community mental health. The difference between reforming the system to help people back into work and cutting their support while leaving the causes untouched is the whole point.
'The UK cannot afford to spend more on welfare'
The UK cannot afford to leave child poverty alone, because it brings long-run costs in health, education, criminal justice and lost productivity that dwarf the short-run cost of adequate support. A fair fiscal argument would apply the same scrutiny to a £150.7 billion universal transfer to pensioners as to a £13.3 billion child benefit budget. The difference is not the size. It is whose needs the political system treats as priorities.
'Not all pensioners are wealthy. Many live in real poverty'
Correct, and this is the most important distinction in the triple lock debate. About 1.7 million pensioners live in poverty, disproportionately women with broken employment histories, ethnic minority pensioners and renters who got none of the housing wealth windfall. The case for reform is not that pensioners are undeserving. It is that raising every pension by the same formula, regardless of wealth, is the wrong way to support those in real need. A smoothed earnings link combined with a stronger Pension Credit for those below a set income would protect the poorest pensioners better, at lower total cost, than the current arrangement.
Cross-Pillar Dependencies
| Pillar | Connection |
|---|---|
| S1_08 State Pension Explained | A companion to this pillar, explaining what the state pension is, what it has been worth over time and compared with other countries, how the pension age has changed relative to life expectancy, and what a funded alternative would look like. This pillar covers the policy on the triple lock and pensioner poverty; S1_08 gives the basic explanation a reader may need first. |
| S1_09 Fiscal Hierarchy | The make-up of the welfare system, with pensioners protected and children in the contestable tier, is the UK's unspoken spending hierarchy made concrete. S1_09 examines what that hierarchy is, why it emerged, whose interests it serves, and how an openly democratic alternative would be designed. The two pillars propose different fixes, left as an open disagreement rather than reconciled. This pillar keeps the state pension's funding priority and reforms only how it is uprated (the smoothed earnings link below); S1_09 §1.2's Options stage instead moves the pension to a lower, conditional Tier Six. The project holds both positions at once; it is not a question of sequencing. See S1_09 §1.2 for the fuller case. |
| Political Renewal | The welfare system's tilt toward older voters is the most direct expression of the electoral arithmetic the Political Renewal pillar addresses. The triple lock, the two-child limit and the Local Housing Allowance freeze all reflect the voting weight of particular age groups under first-past-the-post. Proportional representation, votes at 16 and citizens' assemblies are central to welfare reform, because they are how its democratic distortion gets corrected. |
| Public Office Covenant | Triple lock policy is set by politicians, many of whom hold large pension and property assets: the most direct financial conflict of interest in any policy area. Mandatory disclosure is the precondition for legitimate welfare reform. The Covenant does not assume corruption. It creates the transparency that makes conflicts visible and therefore politically costly. |
| NHS | The disability caseload cannot fall without clearing NHS waiting lists, which hold 7.6 million cases covering about 6.4 million people, many of them claiming disability benefits while they wait for treatment that would let them work. Neither department has the mandate or incentive to take on the other's costs, and fixing this needs joint arrangements that neither currently has. |
| Education | More early years funding is partly undone unless child poverty falls at the same time. The Education pillar's early years investment and this pillar's child poverty strategy must move together; neither delivers fully without the other. The two-child limit directly shaped the early lives of children who then arrived at school already behind. |
| Demographics | This pillar's £3bn a year universal early years commitment, funded from triple lock reform savings, is the project's only currently funded early-years proposal. The Demographics pillar separately raises childcare as part of a broader, currently unfunded idea of recycling inheritance tax revenue (housing, childcare and a citizen's endowment). The two do not conflict, but neither should be read as implying two separately funded childcare programmes. Alaska's Permanent Fund (set up in 1976, saving and investing oil and mineral royalties, paying part of its earnings directly to every resident each year, typically $1,000–2,000, with the state also drawing up to 5% of the Fund a year under a 2018 law to help fund public services) is a different model from Norway's or Singapore's: returns go straight to citizens as cash rather than replacing tax. It is the closest real-world precedent for the Demographics pillar's "citizen's endowment" idea, though it answers a different question: it does not reduce reliance on tax, it shares out a return on state-owned capital.20 |
| Housing | Housing is a major driver of welfare demand. The Local Housing Allowance, restored to the 30th percentile of local rents in 2024 and frozen again from 2025, keeps falling behind rents because of the same supply failure the Housing pillar addresses. Housing insecurity damages mental health, which then shows up in disability claims. |
| Economy | The Economy pillar's analysis of regional divergence and wages describes the economic mechanism behind the child poverty figures here. The two-child limit, inadequate in-work benefits and zero-hours work are welfare tools trying to manage outcomes that need economic reform to fix at source. |
| Criminal Justice | Child poverty is one of the strongest predictors of youth offending. The welfare system's failure to support working families properly creates the conditions in which young people are drawn into criminal exploitation. Welfare reform is crime prevention. |
| Energy | Fuel poverty affects about 6 million households, a welfare cost driven by energy prices and badly insulated homes. The Energy pillar's retrofit and grid investment programme would directly reduce welfare costs. Cold, damp homes are a health cost, a welfare cost and an energy cost at once. |
| Public Debt | Welfare spending, especially the state pension triple lock, is a large and growing day-to-day cost. The Public Debt pillar's balance target cannot be met without explicit decisions here about long-run pension and benefit paths. The two pillars must be read together on the fiscal arithmetic. |
| Defence | Veterans' mental health, insecure housing and insecure work are welfare failures with direct consequences for defence recruitment and retention. Inadequate Universal Credit and insecure housing fall disproportionately on veterans leaving the services. |
11. Proposals for Change
These are the evidence-based proposals of this pillar, put forward for public discussion and challenge. They are not a programme for government.
State Pension and Triple Lock
-
P1 Replace the triple lock with a smoothed earnings link modelled on Australia's system and recommended by the IFS Pensions Review. The state pension rises with earnings. In years when inflation outpaces earnings, the pension rises with inflation and the gap is made good in later years. This protects pensioners' living standards during downturns while ending the one-way ratchet that costs £12 billion a year above earnings indexation.
-
A stronger Pension Credit for the poorest third of pensioners throughout any transition, protecting specifically the 1.7 million pensioners now in poverty who are most exposed to any change in how pensions are uprated.
-
Phase the change in over ten years, with full transparency to Parliament and independent monitoring by the OBR.
Child Poverty
-
P2 Set a statutory child poverty target, defined as the share of children below 60% of median income after housing costs, with a binding path to 10% by 2035, independently monitored.
-
Restore the Local Housing Allowance to the 30th percentile of actual local rents and raise it each year in line with rents.
-
Invest £3 billion a year in universal early years provision, funded from the savings from triple lock reform.
-
Make child poverty impact assessments a legal requirement for all major fiscal decisions.
Disability and Long-Term Sickness
-
P3 Introduce a legal duty on employers to fund and manage a structured return to work during the first twelve weeks of sickness absence, modelled on the Norwegian system.
-
P4 Build an NHS–DWP waiting list bridge: people waiting for treatment for conditions that stop them working get active help toward work while they wait, and NHS treatment is prioritised by its effect on employment.
-
Commission an independent programme to expand the mental health workforce. The PIP caseload for anxiety and depression is a public health crisis needing a clinical response, not a benefits crisis needing tougher assessments.
Universal Credit
-
P5 Raise the UC standard allowance to 15% of median earnings within the first parliament, with a path to 18% over ten years.
-
P6 Remove the benefit cap for households with dependent children.
-
Cut the UC taper rate from 55p to 45p per pound earned, to strengthen the incentive to work.
Social Care
- Implement the Dilnot framework in full: a national cap on lifetime care costs of £86,000 (uprated to 2026 prices), a means-test threshold raised to £100,000 and indexed to asset prices, and a minimum national fee rate. Fund it through a dedicated social care levy raising about £8 billion a year.
The Generational Reset is a non-partisan, public-interest project. It is not affiliated with any political party, does not accept corporate funding, and publishes all its work under open licence for public discussion and adaptation.
For public discussion. Not affiliated with any political party. | generationalreset.org
The Generational Reset | S1_03: Social Welfare | For public discussion. Not affiliated with any political party. | generationalreset.org