It has been said that everybody believes in the Law of Errors (i.e. the Normal distribution), the experimenters because they think it is a mathematical theorem and the mathematicians because they think it is an experimental fact. Text-books on statistics often seem to produce the Normal distribution as the conjurer produces a rabbit from a hat and the actuarial student, who should be something of both mathematician and experimenter, is frequently somewhat mystified as to why such an apparently complicated distribution function as

Fig. 1. The Normal distribution

(see Fig. 1) is suddenly introduced and given such prominence.

I have chosen the word ‘Education’ rather than ‘Training’ so that discussion can wander as far afield as possible; for example: school, university, Actuarial Tuition Service, office, post-qualification education and training of various kinds. The experienced actuary is seldom a narrow specialist; the ambition of most young actuaries is to become a manager of some kind or another and this work involves basic disciplines and skills other than those covered by the examination syllabus.

Menzler's book The First Fifty Years will give the reader a good idea of the plight of the actuarial student during the 1920s. There was only one good text-book, that on life contingencies. There were other books that were helpful but they could hardly be classed as very suitable for the beginner. The only official classes were the oral classes in London. A provincial student had either to persuade an official tutor to correct test papers for him, or obtain the services of some other actuary; for example, classes were organized at the Prudential Assurance Company and the tutors were prepared to let other students have the test papers and any notes that were issued.

This paper is not intended to be a fully comprehensive survey of the taxation of insurance business. Rather it is meant to highlight some of the aspects of the taxation of this business which have been affected by the 1965 Finance Act and which should now, I think, be influencing our thinking as regards the transacting of insurance from both the shareholders' and the policyholders' point of view. Despite the splendid efforts of both the Life Offices' and British Insurance Associations during the past year, not all these changes have had the publicity which might have been expected after a normal Finance Act. I have dealt with general insurance and ordinary life assurance separately and, although the paper is written from a proprietary office's point of view, most of it is equally applicable to a mutual office. Neither have I dealt specifically with industrial life business, but I believe that most of the paper applies to this as well.

During the 1930s there were allegations that London bus drivers and conductors had an undue amount of sickness, especially gastric disorders. Dr Bradford Hill (1937) was asked to investigate this allegation, but found that the data were limited. The London Passenger Transport Board determined that in the future adequate data for such investigations should be available. This was not possible until after the war. In 1948 the Central Record of Staff Statistics was set up to compile sickness-absence statistics and for other purposes. This lecture is mainly an account of the uses made of the statistics produced: how they can be used to study the effect of sick pay provisions, the secular trend of sickness absence in occupational groups, and the differences in sickness-absence experience between groups of various kinds. If the experience of these groups is related to the employees' work and the conditions under which it is done, the results are valuable in answering questions on health from management and unions. The data collected can also be used in medical administration and for research into the causation of disease.

H. O. Worger gave a simple rule (J.S.S., 5, 15 and J.S.S., 7, 116) for using tables of Bond Values for finding the equivalent gross redemption yield on a redeemable security where the dividend is subject to income tax but the capital gain is tax-free. He showed that it was unnecessary to go through the process of first finding the net yield by entering the tables at a dividend rate from which tax had been deducted. All that was needed was to reduce the unexpired term in proportion to the rate of tax payable, and enter the table for the full nominal rate of dividend. The purpose of this note is to show that a somewhat similar rule is available if both dividend and capital gain on redemption are subject to the same rate of tax.

Let P be the purchase price of a loan of 1 redeemable at par in n years time and bearing annual dividends at the rate g, and let t be the rate of tax.