Policy & Research · By Tom Ferguson, Retirement Adequacy Analyst · Published 19 May 2026
Auto-enrolment is the most successful piece of social policy in modern British pensions. It took a country where workplace saving was collapsing and quietly reversed the trend, bringing millions of people into a pension for the first time without asking them to fill in a single form. Nobody serious disputes that. The harder question, and the one the industry has been slow to face, is whether it is delivering an adequate retirement — and for whom.
Participation is not the same as adequacy
The original design solved a participation problem. Inertia was working against savers, so the policy turned inertia around and made saving the default. It worked. But the minimum contribution level was set to be politically and financially tolerable at launch, not to be sufficient for a comfortable retirement. Eight per cent of qualifying earnings, of which only part comes from the employer, was a starting point that has quietly become a destination.
The risk is a generation who did exactly what they were told, never opted out, and still arrive at 67 with less than they need.
Who the current design serves least well
Adequacy is not evenly distributed. Our analysis, and the wider evidence base, consistently identifies the same groups falling short:
- Low earners, for whom contributions are calculated only on earnings above a lower limit rather than from the first pound.
- Multiple job holders, who may not be enrolled anywhere because each individual job sits below the earnings trigger.
- The self-employed, who sit entirely outside auto-enrolment and have no default at all.
- People with interrupted careers, whose contribution history has gaps that compounding never makes up.
- Younger workers under 22, excluded from automatic enrolment during the years when compounding is worth the most.
What adequacy actually requires
There is a reasonable industry consensus that a combined contribution rate in the low-to-mid teens, sustained over a full working life, is closer to what a moderate standard of living in retirement demands. Getting there is not a matter of one dramatic announcement. It needs a sequenced, well-signposted set of changes:
- Contributions from the first pound of earnings, which is the single most progressive change available.
- Lowering the age threshold to 18, capturing the highest-compounding years.
- A timetabled, phased increase in the minimum contribution rate, announced years in advance so employers and households can plan.
- A workable default for the self-employed, most plausibly through the tax system.
- Fixing small pots, so that the savings people do make are not eroded by fragmentation across a dozen forgotten schemes.
Why we argue for this
Higher contributions mean more assets under management, so it is fair to ask whether our interest is self-interested. It is worth being clear: we have no shareholders. Money that flows into the scheme is not a dividend waiting to happen; more than £30 million a year goes back to members through our savings reward and management charge rebate. We argue for adequacy because inadequate retirements are the outcome our organisation was founded eighty years ago to prevent.
What employers can do now
Policy will take time. Employers do not have to wait for it. Matching above the statutory minimum, contributing from the first pound voluntarily, and simply talking about the pension as part of reward rather than a payroll deduction all measurably improve outcomes. So does making it easy for staff to model the effect of a one per cent increase before they commit.
Auto-enrolment answered the question of whether people would save. The next decade has to answer whether they saved enough.