Journal Detail back to listing
Pension Offsetting - Some Further Thoughts
- Mar 25, 2020
- Latest Journal
Many useful recent publications have been published on putting a value on a pension. The pensions Advisory Group’s July 2019 report (referred to as the “PODE” report) – Actuary article –Hilary Woodward and Rhys Taylor’s “Apples and Pears” paper (2015) - Hay, Hess and Lockett’s “Pensions on Divorce (2018- updated from 2013). Apples and Pears described their offsetting inputs as follows: “The present authors do not presume to be qualified to adjudicate upon the rival views of a congregation of nine experts and the Duxbury proponents. However, they are at odds with one another and it would be a worthwhile exercise to see such positions adjudicated upon by reference to evidence.”
The PODE report appears to have achieved some consensus, but there are still differences of opinion. The following general themes emerging regarding offsetting:
• It is acceptable to use a Cash Equivalent (CE)alone for small/DC cases – but not advisable otherwise
• Possible methods include DCFE (including annuity purchase), Realisable Value, Cash flow with parties’ capacity for risk assessment, “Actuarial value”, where cash flows are discounted for investment return and mortality, and Duxbury tables.
• Critical remarks were made about the CE not being the “True” value (or a similar aphorism)
• Some describe the DCFE as the “true” value or “fair” value, but as the PODE report explains, there is no such thing. Accordingly, all terms should be defined when first introduced.
• There is a fear of “Expert shopping” – pushing an expert whose methodology favours either H or W
In choosing a method, or a combination of methods, my input might be useful. I have worked for some years as a Scheme Actuary – the only actuaries allowed to advise Trustees on and then set the bases for calculating Cash Equivalents – and have also worked on an annuity desk – although that was quite a few years ago.
Methods which are most regularly cited are:
CE (although described in alternative language)
Almost invariably, but not always, the DCFE is greater, sometimes significantly, than the CE.
Two obvious questions then arise:
1 Why is the DCFE so high? and
2 Why is the CE so low?
Regarding 1, annuities are provided commercially by an insurance company. It needs to ensure the price it charges is adequate so that its solvency is not endangered in the future. However, the market is keenly competitive, and rates charged may need to change quickly, especially if market yields change, so the price it can charge is limited. Efficient providers will become market leaders, but must continually watch the market. Despite the competition, annuity rate caution will mean the DCFE is usually at the top end of the range of possibilities.
Regarding 2, the CE is calculated by a Scheme Actuary for a current scheme member. The 2004 Pensions Act transferred the ultimate decisions on pensions funding to Scheme Trustees – however, the Scheme Actuary is responsible for advising them on a suitable range of options. The scheme may not be fully funded on a solvency basis (similar to a DCFE basis for the membership as a whole) – however, THE CE MUST REPRESENT THE COST TO THE SCHEME OF PROVIDING ALL THE BENEFITS ACCRUED, USING ACTUARIAL ASSUMPTIONS – the Pensions Regulator will insist on that. In addition, the Trustees and Scheme Actuary also have a duty to ensure the remaining members are not disadvantaged by the CE being taken. Accordingly, the CE is usually at the bottom end of the range of possibilities.
In other words, the annuity actuary (it is usually an actuary) is keen to prevent new annuitants joining the insurer’s current population at too cheap a cost – so the purchase amount is relatively high.
Correspondingly, the Scheme Actuary (an actuary with certified experience) is keen to prevent leavers exiting the scheme’s current population at too expensive a cost – so the ‘sale’ amount is relatively low.
Some additional points:
• Occasionally, the CE is reduced from the cost to the scheme of providing the benefits, due to underfunding. This must be formalised by a published report, and all CEs reduced by fixed percentages. In this case, both pre and post reduction figures must be published, and special consideration should be given.
• A few well-known schemes (eg, Armed Forces Pension Scheme) offer early payment of accrued benefits, the advantages of which are lost immediately on early leaving (which drives the CE), or gradually, if the member works until later than the option date.
• Both 1 and 2 may suffer from imprecision – some of which is by design. For example, post the Test Achats case, market annuity rates MUST be unisex, whereas CEs generally use sex specific mortality rates, and are more accurate in this respect. On the other hand, CEs may assume a certain proportion married in valuing Spouses’ benefits to construct the CE, rather than consider the current marital status of the individual.
• Duxbury tables contain a large allowance for “husband failure*”, ie, husband being unable to make payments due to sickness, death, or redundancy. Because of this, Duxbury is not thought to be suitable for pension valuation by many practitioners. (* The husband is usually the payer)
• Historically, some schemes have provided relatively low CETVs – primarily government schemes. Under the 2016 guidance
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/755674/basis_for_setting_the_discount_rate_for_cetv.pdf it is stated:
“1.8 Under the transfer value regulations, a CETV should be the amount required within the pension scheme to make provision for the accrued benefits, options and discretionary benefits which would otherwise be provided.”
Our experience is that the CETVS for government schemes have improved in recent years, and such anomalies are rarer. It would be interesting to compare other experts’ recent experience.
Regarding the possible methods above, Realisable Value uses the CE (there is no choice but to do so). Actuarial Value means a PODE replicates in some way what a Scheme Actuary or annuity actuary(or both) is doing – the definition covers what each of these does. Careful thought should be given as to why the PODE believes his Actuarial Value is “better” than that of either actuary.
If the use of the CE is not to be barred, a sensible suggestion seems to be to derive a value between 1 and 2 above – the “high” value, and the “low” value. The two values can also be seen as a “buying price” and a “selling price” – where buying involves absolving oneself of liabilities, and selling means taking them on. That would avoid extremes. To recap, the Scheme will pay no more than the CE, and the pension provider will accept no less.
In addition, it should be noted that neither of the benefits are actually being purchased in the market, nor is the CE actually being taken. Both are hypothetical factors in assessing a current offsetting value.
The above would also appear to be consistent with the aim of making outcomes more predictable and consistent.
Finally, while sole use of the CE would arguably undervalue the benefits on offsetting, which might lead to H favouring offsetting and W pension sharing, sole use of the DCFE would result in the opposite – W would always be tempted to go for offsetting and H pension sharing. Expert shopping would then be inadvertently encouraged.
One way to put all views into context might be to consider a “Pension Offsetting Dashboard” (akin to the much-lauded Pensions Dashboard). In fact, perhaps only a “Pensions Speedometer” would be needed. This might take the following format:
Primary Range of Actuarial Values see figures below.
It would be expected that any figure outside the primary range would need additional explanation.
The PODE report includes the FCA transfer test basis, suggesting this as a starting point for offsetting calculations. Another possible basis is the PPF’s “Section 143” basis, used for schemes entering the PPF. More simply, the mean of the outside limits might be used. The starting point may then be as follows:
The above makes no allowance for tax or other pension illiquidity. It is proposed the experts allow for tax, but leave the Courts to adjust for further liquidity preference.