The UK Has Two Pension Problems — Both Growing
1. STATE PENSION
Pay-as-you-go system funded from NI contributions and general taxation. Currently costs £130bn/year — 5% of GDP. With the triple lock, demographic ageing and rising life expectancy, this is projected to reach 7.7% of GDP by the 2070s. The National Insurance Fund could run out in the mid-2040s.
2. PUBLIC SECTOR PENSIONS
Defined benefit, unfunded schemes for NHS workers, teachers, civil servants, armed forces and police. Total unfunded liability: £1.4 trillion — not counted in official national debt. Annual net cost to taxpayers is currently ~£5bn and rising as the workforce ages.
🚨 Combined, these two pension obligations represent the single largest long-term fiscal pressure on UK public finances — dwarfing even debt interest. Together they are projected to cost over 12% of GDP annually by 2050, equivalent to the entire current NHS budget twice over.
The Triple Lock — How it Works & What it Costs
£12,548
FULL NEW STATE PENSION 2026/27 (£241.30/week × 52)
+4.8%
2026 Uprating
(Earnings won)
(Earnings won)
+30%
Rise since
2022/23
2022/23
£6bn
2026 uprating
total cost
total cost
How the Triple Lock Works Each Year
📈
CPI Inflation
Sep figure
used
used
💰
Earnings Growth
May-Jul avg
used
used
🔒
2.5% Minimum
Floor if others
are lower
are lower
Pension rises by whichever of these three is highest each year
🚨 The ratchet effect: The triple lock is a one-way ratchet — it can only go up, never down. The IFS calculates it now costs around £12bn more per year than if the state pension had simply risen with average earnings since 2011. The OBR projects the annual extra cost will reach £15.5bn by 2030. Over its lifetime, the triple lock has made the state pension about 19% more generous than it would have been under earnings-linking alone.
⚠️ The intergenerational problem: Working-age benefits (Universal Credit, JSA, etc.) rise only with CPI inflation. The triple lock means pensioners — already the wealthiest demographic on average — receive systematically more generous uprating than working-age people. Over 3 million pensioners live in households worth more than £1 million, and the number of older people in households with total wealth above £1m rose 269% from 2010 to 2020.
State Pension Cost — Historical & Projected (% of GDP)
State Pension % GDP (actual)
OBR Projection
If earnings-linked only (no triple lock)
📋 The cost of the state pension has risen from around 2% of UK GDP to 5% currently — equating to £130bn — and is forecast to increase to 7.7% of the economy by the early 2070s. The gap between the red and blue lines is the cost of the triple lock above earnings indexation.
State Pension Value — Projected Future Amounts Under Triple Lock
| Year | Weekly Amount | Annual Amount | Assumed Rise | Total Annual Cost | % GDP |
|---|
📋 Projections assume 2.5% annual triple lock rise (minimum) — actual rises will be higher in years of strong earnings or inflation growth. State pension age rising to 67 by 2028. Recipient numbers grow with demographics despite later retirement age.
Unfunded Public Sector Pension Liability — £1.4 Trillion
£535bn
NHS Workers
£335bn
Teachers
£222bn
Civil Servants
£156bn
Armed Forces
£104bn
Police
£30bn
Other Schemes
Annual Cash Flow — Contributions In vs Payments Out (£bn)
Contributions In
Payments Out
Net Taxpayer Cost
🚨 In 2023/24 public sector pension contributions in were £49.9bn but payments out were £55bn — requiring a £5.1bn annual top-up from taxpayers. This gap grows each year as the workforce ages and more workers retire. By 2040 the annual net cost is projected to reach £20-25bn, adding directly to borrowing requirements.
Demographics — The Ticking Clock Behind Every Pension Number
Workers per Pensioner (ratio)
% of Population Over 65
⚠️ In 1950 there were approximately 5.5 workers for every pensioner. Today there are 3.2. By 2050 there will be fewer than 2.5. This fundamental demographic shift is the root cause of all pension funding pressure. More pensioners, living longer, supported by proportionally fewer workers — that is the mathematical reality behind every projection on this page.
87 yrs
Average Life Expectancy at Birth 2025
22 yrs
Average Retirement Duration (from age 65)
📋 When the state pension was introduced in 1948 at age 65, average life expectancy was 68 — meaning most people collected the pension for only 3 years. Today retirement can last 20+ years. The mismatch between pension system design and modern longevity is structural and worsening.
What Could Be Done — Reform Options & Their Costs/Benefits
❌ Keep Triple Lock as Is
State pension reaches 7.7% of GDP by 2070s. NI Fund exhausted mid-2040s. Intergenerational unfairness continues. Politically very difficult to change.
⚖️ Double Lock (CPI or Earnings)
Remove the 2.5% minimum floor. Saves ~£3-5bn/yr in normal times. Still expensive in high-inflation years. IFS recommended option.
📅 Raise Pension Age Faster
Age 67 by 2028 already legislated. Age 68 by 2046 planned. Each year of delay saves ~£10bn. But regressive — manual workers die earlier than office workers.
🏦 Means-Test Pension
Reduce or eliminate state pension for wealthy retirees. Could save £5-8bn/yr but enormously complex, reduces saving incentives and politically toxic.
🔄 DC Public Sector Pensions
Switch new public sector workers from defined benefit to defined contribution. Caps future liability accumulation but existing £1.4tn remains. New Zealand and Canada model.
📈 Higher Immigration/Growth
More workers paying NI = more revenue to fund pensions. Each 0.5% extra GDP growth reduces the pension cost/GDP ratio meaningfully. But doesn't eliminate the structural problem.
✅ The IFS conclusion: No single reform is sufficient. A combination of gradual pension age increases, removal of the 2.5% triple lock floor, and higher economic growth is the most realistic path. Doing nothing is the most expensive option — but also the most politically comfortable one, which is why successive governments have chosen it.