612 PROCEEDINOS OF SECTIOX G. 



B. 



It will be seen, therefore, that the j^roblern of setting on foot a 

 pension fund for a service in being is one tliat taxes the skill and 

 experience of the actuary to the utmost. 



Mortality. — He knows that the rate of mortality among con- 

 tributing members of such funds is very light. On this point Mr. 

 Manly, a leading Englisli authority, writes as follows: — " The more 

 experience I gain of these funds the more I am concerned about 

 the future rate of mortality. Some 10 or 15 years ago it used to be 

 considered that the rate of mortality brought out in the experience 

 of these funds was abnormally light, but I am inclined to think 

 that in future it will be much lighter. There has been a remark- 

 able improvement in the mortality in the last ten years, and the 

 rate obtained from an experience extending over 30 or 40 years is 

 certainly not a correct estimate of the mortality likely to prevail 

 during the next 30 or 40 years. 



A table, whicli is probably the most reliable available, has 

 recently been published, giving the mortality experienced among 

 contributing members — officers and workmen — of a large number of 

 British Superannuation Funds. These rates are shown in diagram 

 form compared with : — 



(1) The " O " table — assured lives in British offices. 



(2) 50 per cent, of the " O " table. 



(3) "O" — Endowment assurance in British offices. 



(4) Endowment asurance in Australian offices. 



It has long been known that lives which elect endowment assur- 

 ance policies experience a much lighter mortality than those whicli 

 take whole life policies. Membership of a pension fund is in many 

 ways akin to owning an endowment assurance, so it is not to be 

 wondered at that the mortality in these funds closely approximates 

 that under the endowment assurances. 



As regards the mortality likely to be experienced among pen- 

 sioned employes, I am aware of no Australian published data 

 which would be of any assistance. 



Of British experience some lias been published, but probably 

 owing to the sraallness of the numbers affected the accidental 

 divergences from normal are so relatively large that the various 

 results are inconsistent with each other and puzzling in the 

 extreme. 



The custom is to assume that the rate of mortality practically 

 coincides with that observed among the general population. 



This is by no means a satisfactory state of affairs, and any 

 experience of pensions that is available should, in the interests of 

 all concerned, be examined and made available to the public. 



