Fairer Pensions · By Naomi Adeyemi, Head of Research · Published 28 April 2026
Ask most people in the pensions industry about the gender pensions gap and they will quote you a figure. Ask about the ethnicity pensions gap and the room goes quiet. Not because nobody cares, but because the sector has spent decades not collecting the data that would let anyone answer. That silence is a choice, and it has consequences.
What we know, and how thin it is
The evidence that does exist points consistently in one direction: savers from several minority ethnic backgrounds are less likely to be enrolled in a workplace pension, more likely to be in lower-paid or insecure work that falls below the auto-enrolment earnings trigger, and more likely to reach retirement relying heavily on the State Pension. But most of that comes from broad national surveys rather than from scheme-level records, which means it is directional rather than precise.
Precision matters here. A gap you can only describe in general terms is easy to acknowledge and easy to ignore. A gap with a number attached becomes a target.
Why the data is missing
There is no single villain. The absence is the accumulated result of ordinary decisions:
- Ethnicity is rarely collected at enrolment. Payroll passes across name, date of birth, salary and National Insurance number. Nothing else is required, so nothing else arrives.
- Nobody is obliged to report it. Unlike gender pay gap reporting, there is no statutory duty on schemes or employers to publish pension outcomes by ethnicity.
- Legitimate caution about data privacy has hardened into a reason never to start the conversation at all.
- Small sample anxiety. Schemes worry that partial data will be criticised, so they wait for perfect data that never comes.
The case for collecting it properly
We think the sector should move, carefully but deliberately, towards measuring and publishing retirement outcomes by ethnicity. Done properly that means voluntary self-declaration, clear explanation of why it is being asked, strong safeguards, aggregate-only reporting, and a commitment to act on what is found.
The prize is significant:
- It would show whether auto-enrolment reforms help or bypass particular communities.
- It would let employers see their own workforce clearly rather than assuming a national average applies to them.
- It would allow providers to test whether their communications actually reach people, or only reach the people who were already engaged.
- It would move the debate from anecdote to evidence, which is where policy change tends to happen.
Communication is part of the gap
Data collection is only half of it. Pensions are explained in a dialect that assumes a particular educational and cultural background, and that shuts people out. Plain English is not a nice-to-have; it is a distribution mechanism. So are inclusive formats, translated materials where they are genuinely needed, and workplace sessions that meet people where they are rather than at 4pm on a Tuesday in a meeting room.
Our commitment
As a not-for-profit provider with around one in five of the UK's workplace pension savers, we are unusually well placed to contribute evidence, and we have no shareholder to please by staying vague. We will keep publishing research, keep pressing for a common industry standard on measuring ethnicity, and keep stripping jargon out of everything we send to members.
The uncomfortable truth is that a gap nobody measures is a gap nobody has to answer for. Counting it properly is the first honest step towards closing it.