228 ON MUTUAL ASSURANCE, 



successive subdivisions, until we should come at last to the conclu- 

 sion that no member can be equitably charged with any losses, 

 except those upon his own policy ; which would amount to this, 

 that upon the death of a member there should be paid back to 

 his representatives the exact sums which he paid in, with interest 

 added and expenses deducted. 



Suppose, again, that the members of the society live in two 

 towns, and that experience proves that the rates of mortality are 

 more favourable in one town than in the other. If any account 

 is to be taken of the proportion in which the members have 

 contributed to surplus during any period, it cannot be denied 

 that those who live in the healthier town should be charged with 

 losses according to their own more favourable rates of mortality. 

 On the same principle, the towns should be divided into sections, 

 the members residing in which should be charged with their own 

 losses only. For similar reasons we should go on to subdivide 

 into streets, and then into houses, and finally into single policies, 

 arriving thus at the same conclusion as before ; which is, indeed, 

 the final conclusion to which the assumed principle must 

 necessarily lead, and the only perfectly equitable solution of the 

 whole difficulty. 



I shall now endeavour to show that the fundamental principle 

 which has been so commonly assumed, that any surplus which 

 may accumulate during any period is not really the property of 

 the society, but is merely held in trust to be returned to the 

 members, is inconsistent with the nature of the mutual contract 

 of insurance usually entered upon ; and not only is not, but in 

 many cases could not reasonably or even legally, be adhered to. 



If during any period a surplus is accumulated on account of 

 higher rates of interest or low rates of mortality, or excess of 

 loading, then according to the assumed principle, this is the 

 property of some or all of the members in a certain proportion ; 

 and this proportion, it must be observed, is quite accidental, 

 depending upon a variety of circumstances, which cannot in any 

 case be foreseen. Now, suppose that upon the eve of the day 

 upon which the investigation is to take place, a loss exactly equal 

 in amount to this surplus which has accrued, is incurred through 

 some accidental cause, as the failure of a bank. There could be 



