Money & government · Pensions

Pension wealth is more unequally held than any other kind of wealth this site measures

The top tenth of households hold 64% of all private pension wealth. Women approaching retirement hold roughly half what men the same age hold. And the self-employed have been left almost entirely outside the system that transformed employee saving.

Executive summary

Three separate, independently verified inequalities sit inside "pension wealth", each measured by a different official source. Private pension wealth is more concentrated at the top than total household wealth — ONS's own finding, not an inference. The gender pension gap, at 48%, is roughly three times the size of the gender pay gap, because it compounds a working lifetime of smaller gaps rather than measuring one year’s pay. And self-employed pension saving has collapsed since the 1990s, left out entirely when automatic enrolment transformed saving among employees.

Read this first — what else changes the meaning of the headline

Top decile share of all private pension wealth
64%

April 2018 to March 2020 — the bottom five deciles combined hold under 1%. ONS's own words: more unequal than total household wealth.

ONS, Pension wealth in Great Britain: April 2018 to March 2020

Gender pension gap, age 55-59
48%

£81,000 (women) vs £156,000 (men), 2020 to 2022. Rises to 62% including people with no pension wealth at all.

DWP, Gender Pensions Gap in Private Pensions: 2020 to 2022

Self-employed contributing to a pension
~20%

Since around 2013 — down from 60% in 1998. Auto-enrolment does not apply to the self-employed at all.

Institute for Fiscal Studies, research on self-employed pension saving

Who holds pension wealth

ONS states directly: "there is more inequality in private (non-state) pension wealth than total net wealth" — private pension wealth is a narrower and even more concentrated slice than the total household wealth figures used elsewhere on this site.

Wealth decileMedian private pension wealth
Decile 1 (poorest)£0
Decile 2 (poorest)£0
Decile 3 (poorest)£0
Decile 4£1,200
Decile 5£7,800
Top decile (10th)£637,500

The bottom three deciles have a median private pension wealth of £0 — not a small amount, no private pension at all. The gap between decile 4 (£1,200) and the top decile (£637,500) is over 500-fold.

Vintage note.This decile breakdown is from ONS’s dedicated pension-wealth bulletin, April 2018 to March 2020 — older than the 2020–2022 total-wealth figures used elsewhere on this site, because ONS has not published an equivalent pension-specific decile breakdown since. The Office for Statistics Regulationsuspended the Wealth and Assets Survey’s accredited status on 13 June 2025, and it has not been restored — see the wealth & tax page for the full disclosure, which applies equally to the figures on this page.

The gender pension gap

DWP has published an official measure of this gap since 2023 — The difference between female and male median uncrystallised private pension wealth — pension savings not yet drawn.

£81,000
Median private pension wealth, women 55–59
2020 to 2022
£156,000
Median private pension wealth, men 55–59
2020 to 2022

The headline 48% figure excludes people with zero private pension wealth entirely. Once they are included, the gap widens to 62% — because more women than men have no private pension wealth at all, not only smaller amounts among those who do.

The gender PAY gap for full-time employees was 7.0% in April 2025 (ONS) — the gender PENSION gap is a substantially larger inequality than the pay gap, because pension wealth compounds gaps in pay, career breaks and part-time work over a working lifetime rather than measuring one year's earnings.

The self-employed, left outside the system

Automatic enrolment applies only to employees — the self-employed were deliberately excluded from the policy that took employee participation from 47% to 90%, and no equivalent mechanism has been introduced for them.

60%
Self-employed contributing to a pension
1998
~20%
Self-employed contributing to a pension
Since around 2013
89%
Employees saving into a workplace pension
2024

This figure is repeated consistently across several IFS publications on self-employment and pensions. Attributed to IFS's own analysis, not independently re-derived by this site.

What this page cannot tell you

Real limits on what these three inequality measures can establish:

How much of the gender pension gap comes from pay versus career breaks versus part-time work

DWP's own gap measure is a snapshot of the outcome — it does not decompose how much is attributable to each contributing cause, though all three are well-documented drivers in the wider literature this site has not independently re-verified.

Whether the self-employed pension figures reflect a genuine savings gap or other retirement provision

Some self-employed people hold property, business assets or other savings as their effective retirement provision instead of a pension — this page measures pension contributions specifically, not total retirement preparedness.

What it means

What the data directly shows
Private pension wealth is held even more unequally than total household wealth; women aged 55-59 hold roughly half the private pension wealth men the same age hold; and self-employed pension saving has fallen from 60% to around 20% since automatic enrolment was introduced for employees only.
What can reasonably be inferred
Pension wealth compounds inequalities that exist earlier in working life (pay, career continuity, employment status) rather than creating new ones from nothing — a lifetime of smaller gaps produces a much larger gap by retirement.
What is disputed
Whether extending an auto-enrolment-style mechanism to the self-employed is practical, given the absence of a single 'employer' to make matching contributions, is a genuinely difficult design question, not just a political one.
A political judgement, not a finding
What (if anything) to do specifically about the gender pension gap or self-employed pension gap — beyond the general policy options already covered on the workplace-pensions page — is a live and contested policy area this site does not take a position on.
What the evidence cannot establish
How today's 55-59 gender pension gap will look for younger cohorts once auto-enrolment (introduced in 2012) has had a full working lifetime to take effect — no data yet exists to answer that.

Part of the pensions vertical (overview). Related: workplace pensions for the contribution-adequacy debate this gap sits alongside, and who owns Britain’s wealth for the total-wealth (not pension-specific) picture.