Valuing defined benefit pension obligations is seldom easy. For several years, actuaries have been debating for how long people will live – longer and longer, it turns out. But the good news that we are going to live longer than we thought comes with a sting in the tail: our pensions are going to become more expensive, unless we defer our retirement.
And now there is another assumption that is challenging actuaries and accountants alike: the discount rate. There is of course the debate about what this rate should reflect. Some have suggested the risk-free rate would be more appropriate – that is, however, a subject for another time. Indeed the IASB has deferred consideration of this beyond its current programme.
Currently, IAS 19, ‘Employee benefits’, requires defined benefit pension obligations to be discounted by reference to the yield on high-quality corporate bonds (provided there is a deep market in such bonds). Interpreting the words ‘high-quality corporate bonds’ is proving to be a challenge.
Let's start with ‘high quality’. My dictionary tells me that ‘quality’ is "a degree or grade of excellence or worth". So it follows that high quality must mean a high degree of excellence or worth. Typically, this is taken to mean corporate bonds rated AA or above, so we are part of the way there. But the yields on some AA-rated bonds have increased significantly in 2008. In fact, some in the market are forecasting default rates of over 30% on some investment grade bonds. Many are questioning whether the higher yielding bonds in the commonly used indices of AA corporate bonds are really high quality at all. Many of these bonds are, not surprisingly, issued by banks and financial services companies. But which bonds do you exclude? Those on credit watch, or those where only one rating agency has given a rating of AA or better, or those where the yield exceeds the average by more than a certain amount?
Even the word ‘bonds’ is open to interpretation. Going back to my dictionary, we get "a certificate of debt (usually interest-bearing or discounted) that is issued by a government or corporation in order to raise money; the issuer is required to pay a fixed sum annually until maturity and then a fixed sum to repay the principal". In fact, the indices often include convertible bonds, warrants and preference shares. Should these really be included?
Up until now, none of this has mattered a great deal, as yields have been within a fairly narrow range. But yields and spreads have increased over the past year as the credit crunch has changed market perceptions. The broader spread of yields indicates the range of discount rates used in practice is likely to be wider and, generally higher, than last year. This means that although the value of many assets held by pension plans will have fallen, the value of their liabilities may also have fallen. In some cases, the liabilities will have fallen by more than the assets, and a pension deficit could have been transformed into a surplus!
The difficulties of setting discount rates have opened up more questions than answers. Some are saying that IAS 19 is now open to abuse in the same way as some predecessor national standards. In 2007, the IFRIC issued a 22-page interpretation to tell us, among other things, what the word ‘available’ means in the context of pension plan surpluses. I wonder whether they will need to do something similar again in this case.
I would be interested in your thoughts, either by commenting here or by email.
The discrepancy between the insurance industry and pension funds is stark with enormous pressure being put on insurers to acknowledge in setting the discount rate for annuity liabilities (typically backed by corporate bonds) to allow for a siugnificantly increased credit default risk. It is astounding that pension provisions for most companies I have seen are not acknowledging that risk in setting the discount rate. The article notes that default rates are expected to be significantly higher than in the recent past. Many are now comparing possible default rates to those experienced in the great depression. Some are even predicting worse - though there is as usual no right answer. Suffice to say that it is a concern that the accounting profession and its actuarial advisors - who took a long time to get broad consistency on mortality improvements with pension fund accounting only recently catching up - have now got another incredible gap in their accounting. In this instance there has been little comment and it is good our see our article is raising awareness. It is perhaps relevant to note that an already stressed pension fund balance sheet will look even more underfunded under IAS 19 if any credit default adjustment is made. I would be concerned however if the 30% default rate referred to were taken as in any way indicative of a likely prospective outturn!
Posted by: Gordon Ireland | 20 February 2009 at 16:22