The Gone-Away Problem No One Is Solving

UK pension schemes are losing track of millions of members. The infrastructure to find them has not kept pace with how people actually live and work.

Published 6 May 2026, PensionHunter, for pension scheme trustees, administrators, and consultants

The number that should keep trustees awake

According to the Pensions Policy Institute, £31.1 billion sits unclaimed across 3.3 million forgotten UK pension pots (Source: PPI BN138, 2024). The Association of British Insurers puts the comparable figure even higher when private pensions are added in. The Australian Tax Office reports AUD 18.9 billion in lost and ATO-held superannuation (Source: ATO, 2025). Taken together with the other systems that publish a number, there is at least $74 billion in verified unclaimed pensions worldwide, a floor based only on the 4 of 47 systems that publish a number. (See our methodology and data sources page for the full per-country attribution.)

Most of that money is not lost. It is simply unreachable. The pension fund knows the entitlement exists. The fund cannot reach the member to discharge it.

This is the gone-away problem. And in 2026, it is a problem the UK pension industry has not solved.

What "reasonable efforts" actually looks like in most schemes

The Trustee Act and The Pensions Regulator's guidance on member tracing both expect schemes to make reasonable efforts to maintain contact with their members. In practice, what most schemes do amounts to:

An annual benefit statement posted to the last-known address. If returned undelivered, the member is flagged "gone-away" in the scheme administrator's system. A periodic credit-bureau search may be commissioned through Tracesmart, LexisNexis, Capita or a similar tracing provider. Hit rates of 30 to 60 per cent on UK domestic addresses. The members not found at this stage typically remain in the gone-away column indefinitely, sometimes for decades.

For most schemes, this is the full extent of the tracing programme. Annual letters and a domestic agency search every few years.

We do not say this to criticise individual trustees. We say it because the infrastructure of UK pension tracing has not been rebuilt for the way people actually live and work. The model assumes members stay in one country, change addresses occasionally, and remain reachable through credit-bureau and electoral-roll data. None of those assumptions hold for the members most schemes have lost.

Why the hardest cases are systematically ignored

Tracesmart, LexisNexis, Capita and the rest of the UK tracing industry are competent at what they do. Within the United Kingdom, with a member who has remained UK-resident, who appears in credit bureau records, who has been on the electoral roll, and who has a current address findable through public records, the existing infrastructure works.

The members most schemes cannot find are precisely the ones outside that profile. Members who emigrated. Members who returned to a country of origin. Members who took international postings and never updated the scheme. Members who married and changed names abroad. Members whose UK credit footprint stopped years ago because they no longer live in the UK at all.

These are the members in the long tail. They are the hardest to find. They are also frequently the members with the largest unclaimed entitlements, because cross-border careers tend to coincide with senior earnings periods. And they are the members the existing infrastructure cannot reach. Domestic tracing agencies do not operate in 41 countries. Credit-bureau data does not extend across borders. Electoral roll searches do not work in Singapore.

The result is that pension schemes in the UK and elsewhere maintain gone-away registers populated almost entirely with members who have moved abroad. The scheme knows the entitlement exists. The scheme has tried, in good faith, the tools available. The tools were never designed for the problem they now face.

What changed about how pensions actually accumulate

Three structural shifts in the labour market over the past three decades have made the gone-away problem worse, not better.

The first shift is international mobility. According to the United Nations, more than 280 million people now live outside their country of birth. Many of them have accumulated pension entitlements in countries they no longer live in. A British engineer who worked in the Netherlands for six years has Dutch pension entitlements. A Filipino domestic worker who spent a decade in Hong Kong has MPF contributions. An Australian academic who held a UK university post has USS membership. Each of those entitlements lives in a system that was built for residents.

The second shift is auto-enrolment. The 2012 introduction of UK auto-enrolment created millions of new pension records, many of them for short-tenure workers, hospitality, retail, gig economy, students taking summer jobs. Those small pots accumulate in NEST and other workplace schemes. Many of the holders never thought of themselves as having a pension. They will not write to their old employer to update an address. The scheme does not know who has moved.

The third shift is the dissolution of long-term employer relationships. Members who left employers twenty or thirty years ago have no current contact channel. The employer may have been acquired, dissolved, restructured. The scheme's last-known address for the member is from 1998. The member may not even remember the scheme exists.

Each of these shifts compounds the gone-away problem in a different way. None of them are reversible. The number of unreachable members will continue to rise unless tracing infrastructure is rebuilt to match the reality of how careers and lives now span borders.

What good would actually look like

A serious approach to member tracing in 2026 should look fundamentally different from an annual letter and a credit bureau check. The components are not exotic, but they are rare:

Cross-border tracing methodology must be genuinely worldwide. Pension members who moved are not limited to any published country list. They are in Brazil, Vietnam, Kenya, Saudi Arabia, the Philippines, anywhere a career took them. Country-by-country knowledge of how to locate a person in jurisdictions where credit-bureau data does not exist. Public registries, employer records, immigration patterns, alumni networks, professional registers. Different in every country, but learnable and codifiable. This work has not been done at scale by any UK tracing agency we are aware of.

Member-led trust signals on outbound contact. A member who receives a letter from their old pension scheme at an address they have not used in fifteen years assumes it is a scam, particularly if the member is now living overseas. Recognised consumer brands generate dramatically higher response rates than cold scheme letters. The infrastructure for trustee-instructed outbound communication via consumer-trusted channels does not currently exist in the pension industry.

Multi-factor identity verification. Schemes cannot pay benefits to someone they cannot verify. The verification process for an emigrated member needs to combine date of birth, scheme reference, security questions, document upload, and where appropriate video confirmation. Most schemes do not have this capability internally and cannot demand it of an unverified contact at the other end of an email.

Audit-grade per-case documentation. Every step of the tracing and verification process needs to be logged for trustee compliance review. Most existing tracing agencies deliver a spreadsheet of "found" or "not found" against each name. That is not evidence. That is an outcome list.

Throughput at scale. Schemes have thousands of gone-aways. Internal teams process tens per month. The mathematics of internal tracing teams completing a meaningful sweep of a scheme's gone-away book within any reasonable timeframe do not work. Industrialised, structured tracing, with consistent rigour applied to every record, is the only realistic path.

These five components together describe a service category that is barely served by the existing pension tracing infrastructure. Building it is hard. Not building it leaves billions of dollars parked in schemes that cannot be returned to the people entitled to them.

Why this matters beyond the regulatory angle

There is a regulatory case for action, TPR's reasonable-efforts duty, member outcomes principles in trustee fiduciary obligations, growing political and media interest in unclaimed pension assets. We will not labour the regulatory case here. Trustees and scheme managers know it.

There is a member case that is harder to argue with. Every gone-away record represents a person, usually a working-class or middle-class person, often someone who emigrated for economic reasons, whose savings are sitting in a scheme they have lost contact with. They are entitled to that money. The scheme knows they are entitled to it. The infrastructure to return it has not been built.

A pension scheme that genuinely commits to finding and reconnecting with its gone-away members is doing two things at once. It is discharging a regulatory duty. It is also returning savings to people who deserve to have them.

The question for any trustee board is whether the scheme's current member tracing programme would survive scrutiny on either dimension. Most do not.

The honest question to ask at your next trustee meeting

If you do not currently know what percentage of your scheme members you cannot trace, that is the question your next trustee meeting should answer. The answer is almost certainly larger than the trustee board assumes. The answer for most schemes also reflects a tracing programme that has not been updated for the way members actually live in 2026.

The infrastructure exists, in 2026, to do this work much better than most schemes currently do. The cross-border tracing methodology, the verification flows, the audit-grade documentation, the consumer-brand outbound contact, these are buildable now. They are not science fiction. They are simply not yet standard.

The schemes that take this seriously over the next 24 months will discharge significant unclaimed liability, return material savings to members who deserve them, and put their compliance position substantially ahead of where the rest of the industry sits. The schemes that do not will, eventually, find that the regulator's view of "reasonable efforts" has shifted underneath them.