Michael Bromwich has published on this blog before and I’m proud to publish his latest note on the University Superannuation Scheme. This is a long and technical blog which is very topical. Yesterday we heard more news of breakdowns in discussions with the threat of more industrial action. It is important that the scheme is discussed as Michael does, shedding light on its complexities and shining a light on a possible way forward.

In the recent discussions of the USS 2020 valuation the impact of future service (FS) costs (FSCs) and their contributions (FS contributions) have received relatively little attention.  This is surprising as FS contributions are a very high proportion of total yearly pension costs which are made up of deficit recovery amounts, FS contributions, DC contributions together with an allowance for USS’s expenses.

The USS’s Update on the 2020 valuation (3 March 2021) shows the former two in their Table 2 with the length of the recovery period and allowed out-performance for each of three scenarios. Scenario 2 is the one being consulted upon by USS. This note considers whether any flexibility is available to vary these FS contributions to avoid the seemly continual major revisions to the scheme which cannot be easily unwound once enacted.

Table 1: Assumptions and Outcomes

^a Allowable return out-performance above discount rate for the recovery plan. Source: USS Update on the 2020 valuation.

FS contributions dominate deficit recovery contributions and persist for much longer provided that the scheme stays open. They seem to be an enormous percentage of salaries   imposing strain on employers for many years.  They are determined   with an adjustment for prudence of something over 40 percent. In monetary terms they amount to approximately £3 billion per year. However only the   years during the next inter-valuation period require actual payment. The   rest of the annual contributions are ‘ghost’ figures becoming real commitments only as time passes.

The Mechanics

The annual amount of these contributions (calculated by the USS actuary using the Projected Unit Method with a one-year control period) is the present value at the valuation date of one year’s pensionable service expected to be earned by the active members from the valuation date. It assumes that the active member population remains stable over this one-year period with leavers and retirees being replaced by new active members.   The discount period will be shorter for older members than for the predominately younger ones. So the present value of benefits will be higher for those nearing retirement. Converting this present value (the sum of the individual member calculations) into an average percentage of the sum of the pensionable salaries of active members gives the annual contribution rate as a percentage of pensionable salary.

These contributions are a function of a number of assumptions of which the main ones are longevity, the discount rate, the age profile of employees (normally held constant) and the strength of the covenant. The three years of future service accrued during the inter-valuation period will be absorbed in the TPs at the next valuation.

The Joint Expert Panel and others suggested ‘smoothing’ FS contributions over three valuations cycles.  Few other attempts at any analysis have been made. USS rejected the Joint Expert Panel’s suggestion because they asserted that it would increase risk. More recent smoothing suggestions have similarly been rebuffed on the basis that USS intends instead to allow an improved but riskier investment performance and this inhibits the taking of further risk via smoothing. Any improvement here is likely to be small given any consequent adjustment for prudence. Assurance was given to look   at smoothing again while indicating that currently this would increase FS contributions though they seem to smooth valuations rather than FS contributions.

The Regulations 

DB pensions are highly regulated but the focus is on the valuation of the TPs, the statutory funding objective that these should be fully funded and on determining any deficit and the consequent recovery plan.   FS contributions are hardly mentioned in the various Pensions Acts, statutory regulations and the Regulator’s guidance code.

The statutory requirement for a schedule of contributions first appeared in the 1995 Pensions Act. It was replaced by Part 3 of the Pensions Act 2004 and associated regulations. The schedule of contributions is required to show payments and required dates of payments including those where a recovery plan is put in place to repair a deficit. FS contributions form part of the contributions included in the schedule which runs either for five years or for the length of any recovery plan whichever is the longer. It is signed by the scheme actuary who is required to certify where the valuation reveals a deficit that:

The Statement of Funding Principles must, among other things, record the methods and assumptions to be used in calculating the scheme’s technical provisions. In the USS these are determined by the Trustee having received the actuary’s advice and consulted the employers.

The Pensions Act 2004 gives the Pensions Regulator (TPR) considerable powers to modify contributions where Act’s requirements are not met and to impose a schedule of contributions though this requires undertaking a rather complex process possibly involving legal proceedings.  However the focus as usual in    pension legislation is on non-compliance with regard to the TPs and recovery plans.

A number of commentators have taken the view the Regulator’s powers are limited to accrued benefits. This is recognised by the Regulator in their consultation on a new funding code when they say that:

We recognise that Part 3 of the Act does not expressly impose any obligations in respect of future service costs, however that does not eliminate the need for trustees to address the issue….  (Consultation document: defined benefits code of practice, March 2020).

They therefore say there is a need for guidance on best practice which is contained in their existing funding code. This says in section 132:

The cost of future benefit accrual for active members should be assessed on a basis consistent with that being used for the technical provisions. If the employer pays contributions determined on a different basis, this needs to be reflected in the schedule of contributions and any recovery plan required so that at the end of any recovery plan period the additional accrual can be expected to be fully funded on the assumptions made. (Code of practice no. 3. Funding defined benefits, July 2014).

USS and many other schemes therefore use the same assumptions as for the TPs. Reading the whole paragraph together says that a different basis can be used with FS contributions to that used for the TPs.

TPs versus FS Contributions

Table 2 shows some of the assumptions made by USS regarding both the TPs and FS contributions with comments on the differences.

Table 2: Technical Provisions and Future Service Contributions

The table makes it clear that the TPs and FS contributions are very different creatures. TPs are amounts either already accrued or in payment at the time of valuation. These are legal entitlements and cannot be changed in the future without members’ consent other than for indexation when in payment. Any deficit requires a deficit recovery plan to be followed.  Where the employer becomes insolvent the holders of accrued benefits are general creditors but can expect the liability to be taken over by the Pension Protection Fund.

The status of FS contributions is controversial. Unless set out in a legal agreement they disappear with employer insolvency.  They are given no place in the accounting standards dealing with DB pensions. Accounting has difficulty dealing with items accruing in the long-term. In financial reports directors   have to say whether the organisation is believed to a going concern for the foreseeable future but pragmatically auditors require such statements to cover only one year.

With a ‘going concern’ perspective the pension literature sees FS contributions as covering ‘promised benefits’ though these may be defrayed from investment income.  Indeed, the Regulator has said that:

The trustees’ key objective is to pay promised benefits as they fall due. (Code 3 funding defined benefits, para.22).

Such an objective is an aim or aspiration. It is not a guarantee or legal obligation. The trustees have a duty to aim for it but there is no guarantee it will be achieved. It applies to FS benefits but not necessarily using TPs assumptions.

FS benefits are clearly different to accrued benefits. They are unearned at valuation time, restricted to active members, unenforceable by members until they become accrued, and are contingent on work occurring after the valuation date. They can be changed in the future as the recent history of the USS pension scheme suggests-not much of a constructive promise.

The calculation of future benefits cost is not really a valuation in the sense used with TPs. Rather it is a forecast of a category of conditional retirement payments (a scenario) so that FS contributions can be calculated. Such forecasts are subject to an additional risk -that of scheme change.

Possible Changes

Some commentators have argued that FS contributions should not be reported for reasons including their lack of precision, their   contingent nature and generally unearned nature. Internationally several countries including Australia and the United States do not generally require the reporting of FS contributions.

The USS uses the   same assumptions for FS costs as for the TPs.  They do not say what discount rate they use. Given that   FS benefits only apply to active members and are to be accrued over a very long period it would seem reasonable  to assume that all FS contributions will be in growth assets in the period before they come into payment. USS indicate that 90 percent of active member funds will be invested in such assets.

The Pensions Regulator suggests this approach in its funding code consultation for ‘young’ schemes including fully open schemes subject to the possibility of the Regulator setting a maximum return. (Consultation document: defined benefit funding code of practice, March 2020).  Failures to achieve expected FS returns can be recouped by adjustments over the long-life of FS benefits   relative to that of accruals.  Given this FS assets would seem to generate little substantial incremental risk to the scheme.    There is no need to de-risk FS benefits as individuals approach retirement as most of the earned FS benefits will have been moved into the TPs earlier in their life.

There seems to be a consensus in the pension industry that the best estimate of the long-term return on growth assets is gilts+3 percentage points to gilts+5 percentage points. Table 8.4 in the Technical Provisions consultation 2020 valuation has a gilts+5.9 percentage points best estimate with a portfolio consisting of 55 percent growth assets at a time when gilts yields were very low. (A consultation for the 2020 valuation: A consultation with Universities UK on the proposed methodology and assumptions for the Scheme’s Technical Provisions, 28 August 2020). This was reduced to gilts+3.50 percentage points by a considerable adjustment for prudence.

USS   adopts high levels of prudence in determining the discount rate for TPs. As indicated (see above) there is no statutory requirement to use the same discount rate for calculating the FS contributions as for the technical provisions.   With a lower level of prudence or none FS benefits will be picked up at the next valuation, to the extent and in the way in which they are become accrued rights.

Some commentators have suggested that the FS benefits earned during the inter-valuation period will become more costly when they are absorbed into the TPs because of the lower TPs discount rate. However, this criticism is faulty in as much as the previous TPs rate should be revised to reflect the addition of the new liabilities and the asset allocation in respect of them.

With 90% of the new liabilities being funded by growth assets (see above), the single equivalent discount rate (SEDR) of the new awards would be far higher than the rate currently used, but consistent with the assumptions of TPs. If the current allocation is 55% growth (gilts + 5.9%) and 45% gilts (gilts) then the TPs SEDR may be expressed as gilts + 3.245%, and the consistent FS rate with its 90% growth assets allocation would be gilts +5.31%. With this latter discount rate the order of magnitude of the decline in the required contribution would be around 40%. This discount rate would be well within the range of outcomes that the Pensions Regulator has power to accept.

Conclusions

The possible changes suggested above provide some flexibility to reduce FS contributions. Such alterations may help avoid major changes to the scheme due to the short-term problems which cannot be easily undone.  In a time of turbulence this and other flexibilities   may allow delaying the consideration major scheme changes until the situation becomes clearer.