There are a number of issues which get confused in the valuation debate.
So to clarify, the start point is that a DB pension is a fixed claim in the sense that a bond issue or debt is fixed. This means that its cost is fixed at the time of award, albeit that this “fixing” may be parametrically determined. A floating rate note is fixed (at say ¼% over Libor) and similarly DB pensions are fixed though the parameters which set the final pension payments (salary, longevity, indexation) and may allow some significant change over time. This is not a weakness; it is strength of the contract design.
To repeat, the cost of a DB promise is fixed at award, just like a bond. In the absence of changes in the parameters (subsequently experience) that determine pensions this remains unchanged until the liability is discharged. The analogy to a bond is clear. Just as with a bond, we have a solid idea what it will cost us. This is the rate of investment return promised to the scheme member; it is also the cost to the sponsor company. This is the cost underwritten by the scheme sponsor. Most importantly it determines the interim values which should be attributed to the pension promise. This would be the value of a claim in insolvency or voluntary liquidation (where possible and of course confounded by tax considerations).
Now note that members should want this rate to be high – it is the return they are promised on contributions. This is conditioned by their degree of risk aversion; a high rate which endangers the viability of the sponsor is not a good idea. The scheme member preference given the criticality of DB pensions to later life wellbeing should be for the highest certainty equivalent rate.
The rate implicitly promised is slowly moving – for the scheme overall it is the weighted average of many decades of awards. Given the history of financial markets, it should not surprise us that it lies in the range 6% – 8% for many schemes.
The role of assets is to serve as security for the accrued benefits of the scheme member. They also serve a role in defraying or defeasing the obligation the obligation underwritten by the sponsor.
The amount of security provided should be no more than the best estimate of the pension promises calculated using this rate. A balance of probabilities valuation. This is 100% funding.
This is the value of the book of business in force – the pension promises outstanding.
No-one values DB schemes in this way. Largely because legislation specifies methods to be used.
The current valuation methods are actually counterfactuals to this value. Both the gilts/bond and expected return on assets methods examine the question: what would these liabilities be worth if this rate applied. Interesting but they really do not apply. In essence, these valuations are pricing a book of business in force as if it were new business. They mislead.
Our response to the DWP DB Green Paper lists a range alternate measures that throw light on the true position of a scheme. (available here: http://www.longfinance.net/publications.html )
There is a better and most basic calculation which may be done – that is calculation of the rate of return on assets held necessary to fully discharge the liabilities when due. The likelihood of this rate being achieved is a measure of the extent to which the scheme may have to rely upon the sponsor. FABI produces such calculations as incidentals to their indices; at negative real returns, they indicate a system in robust health.
This is the work of Con Keating,