[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries

Discounting pension liabilities

In the past year or so, possibly longer, pensions actuaries have been discussing the merits of bonds over equities, sometimes far too acrimoniously for the profession’s good health. We must try to do better and this is a first attempt at getting there. What really worries me is that we are neither talking the same language nor even talking to each other, resulting in the rest of the world picking up the wrong vibes. I suggest that the planning timeframe is crucial and I have what I hope is a tolerable proposal.
Before outlining my approach, I will start with three mantras:
– we do not know what will happen in the future;
– there is no such thing as a free lunch over the long term; and
– we cannot invest the same assets long and short at the same time.

Predicting the future
While others may disagree, I regard it as a fact that no one on earth knows what will happen in the future. Things will probably change, as they always have and, presumably, always will. In particular, we do not know in which direction, or how far, or when, things will move.
Financial economists claim that there is no such thing as a free lunch and I believe that this is fair. On the other hand, one can grab opportunities when one sees them. There is, after all, a real equity risk premium for equities as a class but not necessarily for each individual share. Capitalism would collapse if no one were prepared to take risks for which perceived rewards are demanded. If the equity market were to collapse, then some bond investors would be looking for some replacement equity participation in which case we would be back to the same position. This leads me to continue to believe that there will be a positive equity risk premium, however large or small that might be.
Finally, one cannot invest particular assets for the long term and for the short term at the same time. One must make up one’s mind which end of the time frame one is targeting. This is the crucial element which needs to be factored in to decisions.

Investment decisions
It should always be borne in mind that the assets of a pension fund are held against the liabilities. To my mind, that raises the basic question as to whether the investment decisions should be targeted at either the long term or the short term (OK or somewhere in between!). We simply cannot aim at both ends at the same time. Taking a sporting analogy, one measures a sprinter’s performance over a short distance and a marathon runner’s performance over a long distance. In the case of the investment decision, for distance read time. So, do we actually look at this as carefully as we should when monitoring investment performance? I think not.
So do I have anything to suggest? Yes I do! The first thing is to recognise that there are extremes, namely bonds (or cash) for the short term and equities for the long term. Second, we should recognise that these extremes are bound to continue and we don’t always have to stay at just one end of that continuum. For example, we do not have to regard the bond extreme as the only one which is sensible in all circumstances. That is especially the case if moving from one extreme to the other would lead to losses being crystallised which can never be recovered economically. This would happen, for instance, if moving from equities at their cheapest into bonds at their dearest.

Setting the discount rate
Sticking with the sporting analogy, bonds would be measured over the short term and equities over the long term. Indeed, I submit that taking too much account of equity performance over short-term periods has been highly misleading with adverse economic effects (further details are on www.dvr.org.uk). So let us try out some sort of synthesis based upon the liabilities which will then dictate the timeframe over which we will be planning.
What we could do is aim for the likely buy-out time based upon those liabilities. This time target, say 15 years for argument’s sake, would have to be agreed by the stakeholders as reasonable. So one would assume perhaps that returns would be at bond levels after that target horizon, making allowance for the equity risk premium until the end of the 15 years. At the time of writing, with long gilt yields of around 4.3% pa, with a modest equity risk premium of, say 2.5% pa, this approach would imply a total discount rate of 6.8% pa for 15 years, reducing to 4.3% pa after that, subject to review from time to time.
Since I started drafting this article, EXD51 has been released and I have to say that I am not persuaded that the ‘solvency’ approach contained therein is the best way forward. Given that we are moving towards ‘scheme-specific’ funding approaches, what I have suggested above seems, at least to me, to provide a more adult way forward. Most important, maybe we can now start co-operating and delivering real value to our clients and to the public at large because, at the moment, I fear we are just not doing that.