The Pension Regulator’s consultation on the second draft of the defined benefits funding code and the associated Fast Track consultation have now closed. This code and Fast track requirements, following changes to Part 3 of the Pensions Act 2004 and secondary legislation, change the orientation of schemes by requiring them first to have a long-term objective usually of achieving buyout or the equivalent when they reach significant maturity. And second, can pay accrued pensions with a high degree of certainty by attaining low dependency on sponsors, that is a setting where employers are not expected to be called upon to make any further contributions. Separate arrangements are made for schemes not satisfying the qualifications for Fast Track.
While these changes seem to have gone down well with some in the pension industry, severe criticisms exist. In our view there is a large if not enormous elephant in the room. None of the above documents nor commentators mention the cost of these changes.
There is no consideration of the costs of introducing and running a low-dependency model. These may be very high. There are number of costs: a) the cost of the recovery plan for existing low-dependency deficits, b) the cost of de-risking/the cost transition to the long-term objective and c) the opportunity cost of foregoing growth returns. These may be huge but we have not seen any data or analysis. They may lead to impossibly high contributions or very large cuts in benefits or both. Buyout costs give some clue.
The lack of data on costs makes it difficult to give any reasonable estimate of the cost of such a major change. So, we must find data where we can first from a micro approach and then a macro one.
The Universities Superannuation Scheme (USS) have been reporting not only under the pre-consultation requirements but also on a low-dependency basis for some time. This model is based on self-sufficiency (SS), that is complete stoppage of employer contributions, an extreme form of low dependency. It seeks to indicate the requirements if USS is forced to adopt this approach to pay accrued benefits if all contributions ceased.
Their 2020 valuation reported assets of £66.5bn and a Technical Provisions deficit of £14.1bn and a Self Sufficiency requirement of £102bn, resulting in a Self Sufficiency deficit of £35.6bn, which would require a recovery contribution of over 15.6 % of salaries. In addition, USS indicated that transition risk, the deterioration of the funding level due to de-risking, to get to Self Sufficiency was some £5bn-7bn. The opportunity cost of not using growth assets, which is not included in the transition risk estimate, is some £14bn. This opportunity cost is not usually reported in formal accounting documents.
Of course, these results are affected by market conditions. Looking at these items with the additional information and assumptions available in December 2022 gives a much more favourable view. Gilt yields have increased substantially giving a Technical Provision surplus of £5bn and reducing the SS deficit to £4.8bn. Transition risk was £7bn and the opportunity cost of foregone returns not included in transition risk was £3bn.
The USS scheme is idiosyncratic, being a fully open scheme it is difficult to use this to estimate the cost of Self Sufficiency (taken as equivalent to Low-Dependency) for the whole set of DB pension schemes. However, making some cavalier assumptions may give some idea of the scale of required resources.
The full cost of USS adopting Low-Dependency is the Self Sufficiency deficit plus the transition risk, £11.8bn using December numbers. This represents 16.5% of the USS scheme’s assets. The total assets of all DB schemes registered by the Pension Protection Fund in 2022 is £1,409 bn. TPR’s Fast Track consultation document says that 51% of existing schemes satisfy all the conditions for Fast Track. If we assume that all these schemes follow the USS’s journey plan to maturity and are subject to the same constraints as under the USS’s SS regime, this means we can apply the ratio for USS’s assets to the total assets of the qualified schemes. This gives the cost of adopting the Low Dependency approach as £118bn, equal in scale to half of DWP’s yearly budget. This gives only an indication of the sums involved. Schemes not yet fully qualified for fast track may however have a Fast Track objective. Most closed schemes are likely to find themselves needing to transition more rapidly than USS, raising their relative costs. It would seem reasonable to suggest that in favourable markets the cost is likely to be between £120bn-£160bn, and in unfavourable much more.
The USS based illustration is a micro approach and the estimate may be supported also by a macro approach. The cost of the low-dependency approach may be estimated in terms of the required funding level, the present value of liabilities. If we assume that schemes presently have a duration of 16 years and use a discount rate of gilts +1.5%, where the gilt yield is 3.5%, we may compare this with the funding level required when scheme uses a discount rate of gilts +0.5%. The resulting required funding, an estimate of the minimum cost of the introduction of the approach, is 16.5% higher than currently.
Whether reducing most risk, bearing in mind that risk can crop up unexpectedly, justifies these costs is a multi-faceted ultimately political decision but one that needs to be made.
That even in these more favourable conditions the possible costs of adopting a Low Dependency policy are over 15% of assets indicates the need for an authoritative and urgent analysis of the cost of switching to a Low Dependency basis is essential. Until this analysis is done, the introduction of these regulations and the defined benefits funding code should be paused.