[Skip to content]

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

The business of planning

Here is a challenge. Cast your mind back ten years and think of a few client meetings which you attended then. If you had known then what you know now, would your advice have been the same? Think carefully before you answer.
Of course, all good professionals have that self-deprecating streak which prompts after-the-event reflection on how they might have explained a technical subtlety ever so slightly more clearly. While that behaviour is virtuous, the question about hindsight is actually aimed much more at the fundamentals of actuarial work.

Advice in stable conditions
In framing their advice, actuaries have generally:
– built a mathematical model (eg projected unit method);
– made some assumptions (eg about investment return, inflation, mortality);
– performed several intricate calculations;
– produced an answer (eg surplus is £5m, contribution rate is 16%).
They have then explained to the client that if the assumptions are not borne out, then the answer will be different (the actuarial caveat).
Market-based methods have now been adopted by the majority of actuaries. In stable times, market-based methods can be fairly forgiving. A stable environment means that if the assumptions are not borne out, then they probably missed the mark by only a small margin. That means that actual progress is not that far from expected progress, and any corrective action can be accommodated reasonably easily.

Look at the last 40 years
Table 1 shows how inflation has changed over the past 40 years in the UK, US, Germany, and Japan.
Isn’t this convincing evidence that one of the primary drivers of economic conditions can change relatively suddenly and in a quite fundamental way? In 1974, with inflation at 21%, who would have thought that it could come down to around 1.5%, and even be negative in Japan?
Added to that, people are living for longer, but that is not quite such a surprise. The improvement in mortality is more stable, making the corresponding assumptions that much more reliable.

Look at scenarios
Guidance Note 2 was first issued by the Faculty and Institute of Actuaries in its present form in 1996. It is about financial condition reporting for life offices. Its current version includes the following wording:
‘ to assess the ability of an office to withstand changes in both the external economic environment and the particular experience of the office;
‘the combined effect of a change in two or more related assumptions will in many cases be more important than any change in any one of them in isolation.’
Note in particular the specific reference to wholesale differences in assumptions. This is not sensitivity testing to analyse the impact of a few tweaks. Instead, it is about examining a life office in quite different scenarios.
How might this work for a pension plan? The thinking is that, first of all, a valuation would be carried out using a market-based approach in the normal way. That produces an actuarial balance sheet, identifying the surplus or deficit in the context of the methods and assumptions used. Now comes the scenario planning. What would that same pension plan look like in three years if, at that time, the economic conditions were to be as follows?
– Consensus Slower growth and moderate inflation, broadly in line with current economic expectations
– Recession Low growth and low inflation (eg Japan in the early 1990s)
– Boom High growth and moderate inflation (eg US in early to late 1990s)
– Stagflation Low growth and high inflation (eg UK in early to mid-1970s)
The actuarial balance sheet in three years’ time would be prepared using assumptions consistent with each of the four scenarios, with a projection of interim cashflows based on a transition from the present to the future scenario. The four scenarios chosen are merely examples the wholesale differences between each one are what really matters (reference GN2 for the life office actuaries).

Making financial sense of an uncertain future
Actuaries are comfortable with the mathematical concept of uncertainty. The trouble is that clients tend not to be.
There are actuaries who have developed intricate algorithms that combine a probability distribution with thousands of possible scenarios and then carry out the ensuing analysis of likely outcomes. Yet adopting a probability distribution is just another way of making an assumption an assumption which probably will not be borne out in practice.
The beauty of presenting scenarios is that trustees and sponsoring employers can make their own judgements. Their crystal balls for looking into the future are just as cloudy as any actuary’s, but scenarios have a comforting familiarity for them. All businesses plan for the future, ranging from the use of sophisticated techniques by strategy teams, to instinct for smaller enterprises. Irrespective of their size, they take account consciously or subconsciously of likely future trading conditions for their product or service.
They think in scenarios. Why not explain the uncertainty of their pension arrangements in the language they use themselves to plan their own businesses?
Remember Peter Clark’s motto for his Institute presidency: ‘I am an actuary, and I can explain.’ Here is one way of doing just that.

Back to the challenge
If we could have foreseen low inflation, would our advice ten years ago have been different? Probably yes. Could we have foreseen the future? No. What if we had warned clients of the affects of low inflation for example?
I suspect that we could all feel more comfortable if we had at least explained the consequences of each. Beyond that if one or more of the scenarios were assessed as likely we might have wanted to institute some contingency plans within the investment strategy as protection against potential bad times.
In the financial management of pension plans, investment conditions might be good, they might be bad. No one can say which it will be, but actuaries can model and explain the consequences of differing scenarios.