Introduction

On probability much has been written, and we may need to look no further than the correspondence columns of the FT in mid August and letters relating to the Monty Hall question (of goats and cars) to see that there is still plenty of mileage in this one. As to uncertainty, well the saying goes that in this life only two things are certain: death and taxes. Whatever one’s view might be on the second proposition, it takes little effort to see the common acknowledgement that the overwhelming bulk of life must be full of uncertainty.

We have obtained a draft publication issued in July of this year by the General Insurance Reserving Issues Task Force (GRIT) of the Institute of Actuaries which deals with the problems that actuaries experience communicating this feature, and what follows is, in the main, a commentary on that draft paper. We do acknowledge that the paper itself represents work in progress and may be subject to revision, possibly of a significant nature.

We at TGP are by no means unsympathetic to anyone who attempts to do a difficult job, actuaries included, being of the opinion that there is far too big an industry of armchair critics in this country ready to jump on anyone, from business professionals to the military and the police, who actually does something. ‘The man who never made mistakes never made anything’, it is said, and we believe that there is much wisdom in this statement.

The actuarial paper, referred to above, ranges over many technical matters which we do not intend to rehearse here. We would, however, like to look at three issues that arise from the GRIT panel’s work, namely:

- Is the recipient of the information a major part of the problem?

- How is uncertainty handled in other areas of life?

- What is right and wrong in the business of forecasting?

We do this in the hope that it will help to crystallize thinking around some fundamental issues and avoid the danger of seeking solace in technicalities (something that the actuaries wisely appear to be looking to avoid).

The Culpable Recipient

On the first point, about the recipient’s role, let us consider the articulate oaf. We have all met this character who frequently enjoys more than a little success, at least in the short term. Possessing all the confidence that comes from total ignorance, the oaf outlines a glowing scenario of what can be delivered and will often find favour in the selection process over the knowledgeable candidate who, because of a good understanding of the issues, points out downside risks or hedges objectives about with caveats. If appointed the oaf may get lucky, but it is more likely that things will screw up and then where will the fault lie? With those who made the appointment most would say and we would agree; but why did they make it? Because they preferred the certainty of the ignorant one to the caution of the knowledgeable.

So those users of actuarial services who complain that actuaries got things (in this case loss reserves) ‘wrong’ – a subject to which we will return later - maybe only have themselves to blame for seeking certainty where none was to be had.

Uncertainty Outside the Scope of the Actuary

Reading the paper it is hard to avoid the implication of a belief that uncertainty only exists in areas touched by actuarial science. What about medicine, economic forecasting and law, to take just three areas where future outcomes are uncertain? Doctors get diagnoses wrong, it is commonplace to read diametrically opposed forecasts from economists (in the same publication) and lawyers lose cases. It might be worth asking how these professions manage stakeholders’ expectations.

The medical case is an interesting one. A little time ago there was some criticism of oncologists treating lung cancer sufferers for giving their patients unrealistic hopes of being cured. But a doctor also does not wish the patient to become discouraged and give up the will to live. Perhaps there is something to be learned from other professions in this regard.

What is Right and Wrong in Forecasting?

One of our number is fond of observing that the only thing that can be said with certainty about any business plan is that it is wrong. This statement is intended to acknowledge the simple fact that the future cannot be predicted with accuracy. A good business plan (there are other types) is intended to reflect the likely outcome of a trading period in the light of the best information available at the time when it was produced, and it serves to enable management to judge actual performance outturn.

But because every executive knows, or should know, that a business plan cannot be a perfect statement of a future position, it does not mean that the plan can be put together in a cavalier fashion. Major decisions will be based on these projections and huge sums of money may be at stake, so the planning process had better be pretty thoroughgoing.

Next Page