Assuming that the A1924–29 mortality will in the future be generally used in life office valuations, investigation is made as to the margin necessary between the rates of interest realised and anticipated in the valuation in order to maintain a simple or compound bonus at the rate of 30s. per cent.

It is found that the margin necessary is greater than that deemed sufficient under the OM tables.

In the case of a simple bonus, however, it has of course been the practice to value the bonus also at the lower rate, thus making a greater reserve for future bonus than is necessary, while in a compound bonus office the existing bonus has not always been valued at a rate of interest 1½ per cent, lower than that anticipated which would be necessary in order to maintain a future bonus in respect of that existing at the rate of 30s. per cent.

It is shown that under the new tables a margin of ¾ per cent, in the case of a simple and ¼ per cent, in the case of a compound bonus is still sufficient to maintain a bonus of 30s. per cent, subject to certain limitations as regards expenses.

The necessity of making annual valuations is stressed and approximate methods of computing the fluctuations in the Mortality and Loading profits are outlined.

Although the average degree of solvency among friendly societies has been steadily rising in recent years, valuations in a considerable number of cases still disclose a deficiency. Some of the large affiliated orders have methods of their own for assisting branches in difficulties, but in the case of many societies no such help is available, and other means have to be found for improving the financial position. The object of the paper is to consider some of the methods by which the degree of solvency may be raised, more particularly with reference to such aspects of the problem as (1) negative values, (2) the relative shares of the extinguished deficiency borne by members of different ages, and (3) the effect of different age distributions.

This paper deals with various points that may arise in practice, and in preparing it I have used such legal opinions as I have been able to refer to. A list of the books made use of in writing the paper is appended, to enable those interested to pursue the matter further.

The objects of this paper are (1) to set out for members of the profession some account of the further investigations that have been made into the statistics that formed the basis of the A1924–29 Table; (2) to correct certain errors that have been discovered; (3) to explain the possible weaknesses in the statistics; (4) to discuss the variations in mortality shown by individual offices; and (5) to give some idea of the range in mortality above and below the combined result of all the offices.

In an endeavour to fulfil those objects—

• (1) There are made available in the appendices the various enquiries which, subsequent to the publication of the A1924–29 Table, were made of the contributing offices;

• (2) The net alterations in the total deaths as a result of these enquiries are detailed, including the two major errors already known;

• (3) The various factors calling for care and attention are examined;

• (4) The variations in mortality shown by individual offices are set out in tabular form without disclosing their identity;

• (5) Light and heavy mortality tables are produced with certain monetary functions by what are termed light and heavy mortality in comparison with the standard A1924–29, these light and heavy mortality tables being made up from certain combinations of offices at either extreme of the tabulated list.

The comforting conclusion is reached that the Al924–29 Table is a satisfactory representation of the type of mortality revealed by most of the offices, and may be used, for practical purposes by the large majority.

A final word is given in explanation of the census method, having in mind the importance of including all data which should properly be returned, particularly in the case of early deaths.

Thus, it is laid down that a death should not be omitted if the life, had it survived, would have been included amongst the “in force” to which the deaths are related; and there follows the golden rule: “A death must be included at a duration if, had a ‘census’ (or enumeration of the policies in force) been taken the moment before death, the case would have been counted as in force at that duration.”