INSURANCE. 681 INSURANCE. tion of the theory of probabilities to the results of past experience proceeds on the basis of two assumptions; that all the cases included, whether past or future, are alike in all essential respects, and that the lapse of time brings no change in the factors affecting the degree of probability. Neither of these assumptions corresponds to the fact. As to the first we see that hardly any two lives or pieces of property are alike in all essential respects. In the case of fire insurance, for example, the number of circumstances affect- ing the probability of destruction is very great. It is onlj' by overlooking many minor points and assuming a degree of similarity that does not actually exist, that the application of the sta- tistical method is at all possible. The same con- siderations apply more or less to all other forms of insurance. The common practice is to group risks in classes, according to their more promi- nent elements, and collect data about the differ- ent classes from which to calculate the average. Each class has its own average and its own de- gree of probability. In estimating an individual risk the average of the class to which it belongs is used as a basis from which the particular risk is calculated by making allowance for any special conditions affecting it. The result is at the best a more or less close approximation to reality. It is interesting to consider what the effect of the imperfect classification of risks is, both on the insuring companies and on the insured. To the companies it would make no difference of any kind, if all adopted the same classification and applied it in the same way. A general un- derestimating of risks would of course lead to ruin. But an imperfect classification, as a re- sult of which there were as many risks overesti- mated as there were underestimated, would do them no harm. An average of all kinds of risks would be perfectly safe for them, provided all kinds of risks were charged for according to the average. Thus if there are ten risks, which, when properly estimated, amount to 1, 2, 3. and so on tip to 10, the risk assumed by a company which insured them all would amount to .5.5 ; but the company is equally safe whether it charges each risk according to its proper value, 1. 2, 3, and so on up to 10, or charges each of them according to the average risk, which is 5'-;. Micre there are competing jfompanies, however, and one estimates risks more closely th.in the others, the result of imperfect classifi- cation is an unfavorable selection of risks. If one company assumed all risks from I to 10 at their proper valuation, while another assumed them all at the average valuation of S^^;, the for- mer company would get all the small risks, those from 1 to .5, and the latter the large ones, those from G to 10. We see, then, that where all in- surance companies use the same classification, they have no incentive to perfect their system, but where one company introduces a more ac- curate classification, others are constrained to do the same or suffer from adverse selection. The effect of imperfect classification of risks upon the instired is to cause a disproportionate distribution of the burden of insurance. In an ideal system this burden would be so distributed that each person would pay to the company in proportion (o the risk he brought upon it. This ideal can never be reached, but the more nearly exact the calculation of risks, the more closely is the ideal approximated. The extent to which the attempt is made to estimate the risks ac- curately varies in the different forms of insur- ance. It is probably carried furthest in fire insurance, while in life insurance and other simi- lar kinds comparatively little attention is paid to it. Most life insurance companies recognize only two classes of risks, those which come up to a certain standard and those which fall below it. The latter are rejected, while the former are all accepted at equal rates for equal ages. Yet in many cases the examining physicians would have no hesitation in declaring that some of those accepted would in all probability live long- er than others. The result of this imperfect classification of risks is like that already noted, that in the long run the stronger and healthier lives pay a part of the cost of insurance of those possessing less viability. This injustice is par- tially eliminated through the return to the in- sured of a part of their premiums in the form of dividends, of which the longer lives receive the larger share. But that can mean no more than that the premium rates are scaled so high that the excess paid by the stronger lives is suffi- cient to make up the deficiency on the weaker and leave a remnant to be returned as dividends. The fact that most insurance companies re- fuse to insure lives that do not come up to a certain standard suggests the question, Ahat is an insurable risk? The answer clearly is that any risk — that is, any chance of loss depending on the occurrence of any uncertain event — is in- surable, provided there are sufficient data to enable the degree of probabilitj' to be estimated, and provided it is of such a nature that the in- sured cannot too easily make use of the insur- ance for his own economic advantage. However hazardous the risk, it may be covered by putting the premium high enough. It is sometimes stated that only those risks are insurable which threaten a large number of individuals at the same time. The larger the number of risks, the more closely the degree of probability can be estimated ; but it is possible to make some esti- mate of it on the basis of a very small number of risks, and cover the high degree of uncertainty by a high premium rate. Xo more is it true that the danger must be of such a nature that Ihe loss cannot actually befall a large proportion of the insured at the same time. Here again it is a question of putting the premium rates high enough and accumulating large reserves. If every one of the insured lost at the same time, the company could meet the loss provided it had estimated the risks correctly. Xor. finally, does it make any difference how great the danger of loss may be. It is simply a matter of adjusting premiums to risks. A life insurance company insuring only the lives of consumptives might be on just as sound a financial basis as one which made a specialty of insuring only extra-healthy lives. But while the theory of insurance is equally applicable to all forms of uncertain losses, its actual extension is limited by practical difficul- ties. The chief one of these is the impossibility of inducing people to pay the high rates which would make it safe for a company to assume very hazardous risks. Unless there is consider- able difference between the uncertainty to which an individual is exposed and that to which an insurance company is exposed from the same
