The majority of business risks are not of a kind against which it is possible to insure. Insurance companies confine themselves to risks which are mainly a matter of what we call objective rather than subjective chance, i.e. risks in respect of which knowledge of detailed facts peculiar to the individual case is of minor importance. But such knowledge is of paramount importance in the case of ordinary business risks. If, for example, a new enterprise is to be undertaken, the special knowledge and experience which its promoters possess is a vital factor in determining their estimate of the risk involved. An outsider with no special knowledge would necessarily require to estimate the risk far more highly if we were to form a rational opinion on the basis of _his_ knowledge. So great, indeed, would be the risk to him, that we can lay it down as a sound maxim that people are extremely rash who invest their money in risky undertakings about which they know very little. This subjective aspect of business risk has a significance to which it will be necessary to revert. But, though most business risks are not and cannot be a matter for premiums and policies, the principle, which the practice of insurance illustrates, applies none the less. In the light of their knowledge and experience, the promoters of a new undertaking must weigh up the chances of failure and success, though they will not do so by the precise methods of an actuary. They will require that any chances of serious loss should be balanced by such chances of exceptional gain, as would raise the _expectation_ of profit well above the normal return on secure investments. The more risky the project seems the greater, generally speaking, must be the _expectation_ of profit required to induce people to undertake it. If we suppose business men to calculate reasonably, it follows that the average profits in any industry over a long period of years, reckoning in the losses of the concerns which disappear altogether, are likely to be higher, the more risky is the industry. Such a result will not, of course, occur in every case. Even when the calculations are reasonable, they may be entirely falsified by the event. Moreover, business men may not calculate reasonably on the information which they have. But, unless we suppose their judgment to be subject to a prevailing bias in one direction, i.e. to be unduly optimistic as a general rule, _we_ should expect, and in any case _they_ must expect, profits above the ordinary in a risky industry.