264 ESSAY ON PROBABILITIES. 



those who first insure their lives are entirely dependent 

 upon the future success of the office. But this capital 

 need not be large : in the present state of things, an 

 engaged capital of one hundred thousand pounds is 

 certainly above the mark, even for an office which is 

 entirely without connection, and starts without one 

 single life insured. If, as very often happens, a tolerably 

 large number of customers has been obtained before the 

 prospectus of the office is announced, then a capital, the 

 interest of which will cover the expenses of management, 

 is sufficient. But here it must be observed that the 

 proprietors of this capital run some risk of losing a 

 portion of their principal, and a still greater one of 

 losing the interest for a limited time. This risk is the 

 greater the smaller the original subscription, and it 

 must be paid for accordingly. At the same time, it must 

 be remembered that the mere existence of the capital 

 diminishes the risk, by making it the interest of every 

 proprietor to procure business for the office. The con- 

 nection thus created is the secret of the successful start 

 which has frequently been made ; and it may be con- 

 sidered as very unlikely that an office will fail, from 

 want of business, which is so well supported in the first 

 instance as is implied when a capital of the preceding 

 amount is announced. 



There is, however, one case in which a larger capital 

 is desirable, and even requisite ; that is, where an office 

 is established which is to insure some new and yet 

 untried risk. Whatever pains may be taken in such a 

 case to procure facts and deduce proper tables, there is 

 always a risk that the experience of the office may be at 

 variance with the facts of the tables. When, for instance, 

 the general conclusions drawn from the mortality of 

 towns were first applied to the insurance of life, it was 

 a risk of unknown amount as to whether the lives of 

 those who would come to insure would be of the same 

 class as those from which the tables were made. They 

 might turn out better, or worse. This risk has been 

 tried, and found to be in favour of the offices; but in 



