Home | Thought Leadership

Thought Leadership

The excepted group life trust: the pension trust you probably already have, but rarely run properly

By Nigel Watson

7 May 2026 6 min read

In a previous piece, we highlighted the healthcare trust hiding in plain sight: common, mature and routinely ignored because it sits outside the annual insurance‑renewal ritual. You can read about it here.

The pensions and risk‑benefit world has its own equivalent: the excepted group life arrangement - typically an excepted group life policy held under a discretionary trust.

It’s the trust most employers already have and the one they most need to work when everything else is going wrong.

Most employers know exactly what they pay for death‑in‑service cover. Far fewer know who the trustees are, when they last met or where the expression‑of‑wish forms are kept.

And that matters, because if you can’t name your trustees, you don’t really have a benefit. What you have instead is an unmanaged liability.

Insurers do not pay “HR”. They pay the trustees. It is the trustees who receive the proceeds and who decide how they are distributed. Trusteeship is not an assumption or a formality; it is a legal role that requires real people, valid appointments and accessible records.

If that chain of authority is not alive and functioning, the benefit you believe you have may not be the one your people can rely on when they most need it.

What an "excepted" arrangement actually is

People use the phrase "excepted trust" loosely. In practice, it is a combination of two things:

an excepted group life policy (i.e. pure protection, not a savings product), and
a discretionary trust that owns that policy, with trustees deciding who receives the proceeds (within a defined class) and in what proportions.

The "excepted" conditions exist for a reason: they keep the arrangement firmly in the risk benefit lane (lump sum on death, typically before age 75, no meaningful surrender value etc.), rather than drifting into something that looks like investment or retirement provision.

The insurer pays the trust. The trustees make - and must evidence - the distribution decision. That discretion is the point of the structure. It is also why governance, appointments and records are not optional extras but core to whether the benefit actually works.

Why employers use excepted arrangements (and why the LSDBA keeps them relevant)

Historically, excepted arrangements were used to sidestep pension tax friction where employers wanted to provide large lump‑sum death benefits for higher earners and the registered pension framework (the Lifetime Allowance) became inefficient or restrictive.

But the abolition of the Lifetime Allowance has not removed that friction. It's moved it.

Since April 2024, the focus has shifted to the new lump sum and death benefit allowance (LSDBA). The question is no longer whether the Lifetime Allowance is breached, but how much tax‑free lump‑sum capacity remains within the registered framework and what happens when employer‑provided death benefits consume it.

For most individuals, the LSDBA is £1,073,100 (absent protection or enhancement) and it caps the total tax‑free lump sums that can arise in respect of an individual, including relevant lump‑sum death benefits.

That creates a real‑world trade‑off. Death‑in‑service benefits provided through registered schemes can now erode, or fully exhaust, the individual’s tax‑free lump‑sum headroom. Excepted arrangements exist precisely to sit outside that constraint. They isolate risk benefits from the registered framework and avoid unintended pressure on LSDBA capacity.

That structural separation is the point. And it is why, despite the end of the Lifetime Allowance, excepted group life arrangements remain highly relevant - provided the trust behind them actually works.

Example: the senior "top slice" problem

Your company provides death-in-service cover at 4x salary. Your CFO earns £350,000, so the benefit is £1.4m. The CFO has substantial pension savings and dies before age 75.

If that entire £1.4 million is delivered through a registered route, it may collide with the individual’s LSDBA headroom. The result is simple but brutal: part of what is meant to be a tax‑free lump sum can tip into income tax for the recipient.

That is not a theoretical edge case. It is a predictable outcome for senior earners.

The standard design response is therefore to split the cover:

registered death benefits up to a sensible cap (often aligned with the wider workforce), and
the “top slice” provided through an excepted group life arrangement held under a discretionary trust.

That is not clever structuring. It is not aggressive planning. It is simply separating risk benefits so that an avoidable tax charge is not imposed on families at the worst possible moment.

The three failure modes we see repeatedly

1) The missing trustee

Trustees named in the deed left years ago. Replacement appointments were never documented. Delegated authorities aren't clear. The insurer goes slow because it can't see who has legal authority to decide and sign.

If you don't know who the trustees are, it can also mean the policy may be held in the wrong names (or not properly settled into trust). Before you debate governance niceties, validate the basics: is the insurer's policy schedule actually aligned to the trust structure you think you have?

2) The stale nomination

The expression-of-wish is missing or out of date - partner vs ex-spouse vs adult children is the classic. Trustees can still exercise discretion, but the absence of a current record creates avoidable delay, conflict and reputational risk.

3) The proceeds that sat in the trust

The insurer pays out to trustees. Distribution stalls because the organisation has no playbook, no minute template and no "decision pack" process. Money lingers in trust longer than it needs to, which is exactly how technical and compliance questions start to appear.

Two technical points worth knowing

1) The “10-year” point is real but manageable

Many excepted group life trusts can be relevant property trusts, so periodic (10-year) and exit charge rules can be in scope in principle. In practice, exposure is often modest if proceeds are distributed promptly, but delays are exactly what create the problem. This is a core reason some employers refresh/re-paper legacy trusts on a sensible cycle, rather than discovering defects when a claim is live.

2) Differentiating benefits can be constrained

Excepted policy conditions include constraints on how far you can run materially different benefits across classes within a single policy. Post-acquisition structures are where this bites: legacy cohorts on different terms, combined into one arrangement. A common market workaround is multiple policies under the same trust, implemented deliberately rather than bolted on at renewal.

FAQ: will the forthcoming IHT changes for pension death benefits affect this?

HMRC policy papers and consultation confirm the direction of travel: from 6 April 2027, most unused pension funds and certain pension death benefits are intended to be brought into scope of IHT, with personal representatives responsible for reporting and paying the IHT (rather than scheme administrators).

Crucially, the same papers confirm that death-in-service benefits payable from a registered pension scheme will remain out of scope of IHT from 6 April 2027.

Two practical takeaways:

Don’t overreact: this is primarily a pension-estate planning change, not an “excepted arrangement is broken” change.
Do treat boundary cases carefully: particularly around how “in service” status is defined, how benefits are documented, and how responsibility is split between personal representatives and schemes.

What good looks like (and it doesn't need to be heavy)

A sensible operating model is simple:

One named internal owner (Reward/HR with Finance and Legal cover).
A light governance calendar (often one annual housekeeping meeting, plus ad hoc meetings for claims).
Trustee appointments and delegations kept current.
A short annual cycle: prompt expression-of-wish refresh, conflicts check and document pack review;
A claims playbook and minute template so decisions are evidenced and repeatable.
A policy schedule sanity check: confirm the policy is actually held under the trust and in the correct names.

One final practical point: don't rely on folklore for compliance. Some life policy trusts can be excluded from Trust Registration Service registration while they only hold a pure protection policy, but the position can change if proceeds are received and retained - which is why having an accountable owner matters.

At Burges Salmon, we help employers turn broker-led set-ups into legal-grade, audit-ready operating models. That means reviewing and updating the trust deed, trustee appointments and delegated authorities; tightening beneficiary class and discretion language; and implementing a short claims playbook and minute template your HR team can run with ease.