This paper develops a "subjectivist" theory of the value that individuals place on risks to their lives. It explains the paradox that although individuals may view their lives as priceless, they still will take small risks for a finite amount of money. Typical public projects that alter risks to life result in small changes in survival probability for a large number of people. Standard tools of benefit cost can therefore be applied, where statistical lives saved are valued at a price equal to the marginal rate of substitution between survival probability and wealth. This "value" is compared to human capital measures of the value of saving a life and is shown under reasonable assumptions to exceed the human capital value.