NEWSLETTER    |     July 31, 2024

Virgin Media s37 appeal dismissed – time for trustees to act? [Trustees]

Now that the Court of Appeal has ruled on the “s37” validity issue raised by the Virgin Media case, trustees may well want to satisfy themselves that their scheme is not affected (many are not) or to understand the risks and potential impact of s37 concerns around historic benefit changes. Buy-in transactions have been taking place despite the uncertainty and we expect this to continue. It is important to make informed decisions and we encourage trustees to take a sensible step-by-step approach to all validity issues. If copies of self-contained s37 actuarial confirmations are not readily available, that does not prove they were not given. Most likely there is evidence they were given in archived material which is less easy to find but still legally effective. The DWP may yet provide an escape route, for those who do find a material issue.

What is the Virgin Media case about?

The Virgin Media case relates to defined benefit pension schemes which were contracted-out of the state earnings related pension on a salary-related basis between 1997 and 2016. The case revolves around the exact legal requirements to make valid changes to benefit rules for these schemes. The issue is called “the s37 issue” named after the legal section it is derived from.

There has been uncertainty for years about precisely what s37 meant, in particular whether the s37 requirements applied to only past service benefit changes, only future service benefit changes, or both. The Court of Appeal has confirmed the High Court decision that s37 catches both past and future service benefit changes.

In order for a pension scheme to make legally effective changes to benefit rules between 1997 and 2016 the scheme actuary had to confirm in writing to the trustees that the statutory contracting-out test continued to be met after the changes were made.

The High Court confirmed that s37 applied to both improvements and reductions to benefits. It also confirmed the impact of no s37 confirmation being obtained when needed is that the benefit changes are void regardless of whether the scheme was capable of meeting the s37 test. These points were not appealed.

Why is this of concern?

Many schemes do not have a perfect set of documents with s37 confirmations attached.

During the 1990s and 2000s the focus tended to be on the quality of the benefits provided for s37 purposes, not the attendant paperwork. This means that the obvious evidence of compliance with s37 requirements is not as prevalent as one would expect given the legal consequences of no confirmation being given.

The compliance evidence gap, calling into question the validity of benefit changes, is more of an issue where future service benefits were being reduced which was common at the time. Improvements can be corrected retrospectively; reductions cannot.

Closure to accrual is of particular concern as the interaction between s37 and closure to accrual remains unclear.

Why is there still uncertainty?

We now know when a s37 confirmation was required. Whether it was obtained is a matter of evidence. The best evidence is a signed s37 confirmation from the scheme actuary to the trustees, dated before the effective date of the changes appended to the amending document.

However, the best evidence is not always available. Sometimes the amending documents refer to s37 compliance, sometimes not. In our experience it would have been outside market practice to amend rules without actuarial input. The need to provide benefits that met the contracting-out test was well understood; and set out in the actuarial profession’s guidance called “GN28” (as amended over time). Archived files can contain correspondence between the actuary and the trustees which points to compliance being done, even though there is no neat self-contained confirmation. Sometimes nothing survives except recollections of the parties involved that the scheme actuary was actively involved in the scheme affairs including benefit changes.

Amendments are not automatically void because the confirmation cannot be produced. The question for trustees is whether a sufficient body of evidence survives to suggest on balance of probabilities that a s37 confirmation was given despite the fact that the document itself does not survive. If there is, then trustees may well be comfortable to treat changes as valid, recognising that there is some risk of challenge to that view.

Why look at this issue?

Many trustees and sponsors have taken a “wait and see” approach until now. In our view, the dismissal of the appeal may have shifted the balance of risk towards trustees undertaking an initial fact find of what the amending documents show (if this has not already been done). If this shows there is doubt that can be followed by a focussed and sensible levels of investigation on what survives in the archived files for the benefit changes which present the key risks and cannot be easily corrected.

Most schemes will want to take their time before they decide to act on any change in their understanding of the benefit structure. Sometimes “wait and see” has been a good strategy; for example, schemes that acted early on sex equalisation after 1990 often went further than turned out to be required. However it did not work so well with GMP equalisation which, with the benefit of hindsight, would have been less painful if it had been done long ago.

One risk of a wait and see policy is the loss of an opportunity to recover against former advisers, if there is loss caused by breach of duty on their part. Claims can become statute-barred as a result of delay. This kind of claim is rarely open and shut. Few trustees want to sue their advisers, current or former, but we are already starting to see standstill agreements being requested sometimes at the request of sponsors who ultimately foot the bill for past errors.

S37 is not the only cause of invalidity. The implications of this Court of Appeal ruling should be viewed in the context of validity more generally. Experience shows that few schemes can produce a full set of original, fully executed legal documents. Legal risk is no different from investment, funding, or covenant risk in the sense that it is better to acknowledge, manage and mitigate it than to ignore it.

Uncertainty is a fact of life in pensions and not just in relation to member data. The true benefit structure – the benefits to fund, administer or insure – will often be a question of evidence to be decided on the balance of probabilities and in the light of reasonableness and trustee fiduciary duties. But evidence can only be assessed and relied upon if it is gathered and preserved, and doing this protects trustees from criticism.

Will the Government fix the problem?

There is still hope that the DWP will intervene to remove the uncertainty. There is a statutory power that could be used to remove the unintended consequences of the way s37 operates. The purpose of the contracting-out legislation was to ensure that members got good enough benefits in return for the reduction in their state pension. It makes no sense for rule changes to be invalidated if the benefits were in fact good enough to meet that quality test. The DWP has invited evidence about the impact on schemes. We and others across the industry will be giving our views based on what we have seen so far.

The recent Court of Appeal decision highlights the importance of ensuring compliance with s37 requirements in salary-related contracted-out defined benefit pension schemes. If you are a trustee and have concerns about the validity of past benefit changes or the adequacy of your documentation, our team of experts can help you navigate these complexities and assess any potential risks in a sensible way.

Register for our newsletter

Related Newsletters