The trouble with using bond yields to calculate pension fund liabilities
Defined-benefit pension funds calculate their liabilities using gilt yields as an indicator of the long-term return they are likely to get. But that could expose them to a whole lot of new problems, says Merryn Somerset Webb.
An interesting piece in the FT today. We have written time and time again about the problems with the way defined-benefit pension funds calculate their liabilities the total amount of money they are going to need to pay out all the pensions they owe to all the people in their schemes.
Most funds use gilt yields as an indicator of the likely return they can get over the long term. So the more gilt yields fall (as they have been doing for years now) the more money their spreadsheets tell them they need to have right now in order to meet future obligations. That's why we keep being told the nation's defined-benefit pension funds have vast deficits and why companies keep being forced to shovel more cash into those funds.
The question we keep asking is why use only bond yields? Why not use the low end of the kind of yield an average multi-asset manager can get from a diversified portfolio of assets?
That would be both much higher and much more realistic. It would also make the pension fund deficits look much lower and the problem much smaller (this is after all more a technical than a real problem). And finally it would take the pressure off the UK's companies, allowing them to use their profits to do the things we really want them to do increase productivity and raise wages.
Ex-pensions minister Ros Altmann agreed with us on this and so, it seems, does the ex-chair of the Pensions Regulator, Michael O'Higgins. He reckons that there are "better ways" of assessing liabilities than just using bond yields, says the FT. But even more interestingly, he reckons that the current method (which is not compulsory, by the way) puts companies at risk of making a "significant misallocation" of cash from their operations into their pension schemes while also "inhibiting M&A" across the UK (who wants to buy a company that comes with huge and uncertain liabilities?).
So far, so interesting. But here's a further thought for you particularly if you are pension-fund trustee. Might it be the case that by using bond yields to figure out their liabilities and then acting on that number, might trustees be opening themselves up to legal action at some point in the future? Perhaps from the shareholders of the companies that have compromised long term growth in their efforts to satisfy what O'Higgins calls the "economically nonsensical" demands of their pension funds?
Otherwise it is worth thinking about the other consequence of using gilt yields to value pension liabilities the stunningly large transfer values being offered to those prepared to leave the schemes. It won't be long before these aren't available any more. But what if the whopping amounts of cash being paid out to those who ask for a quote at the right time leave the fund short in the future? What if it looks like their gain is everyone else's loss? Who'll pay then? And who will sue who?