Value of a Statistical Life: Relative Position vs. Relative Age

Abstract

This paper examines the influence on estimates of the value of statistical life (VSL) of the worker's relative position in the wage distribution and relative position in the life cycle. Whereas past work on relative position effects in the labor market have been based on illustrative hypothetical examples, this paper develops empirical tests using actual market behavior. To test for the effect of relative wage position, we use two different measures: the individual's wage rank in the state and the wage rank by gender in the state. Using the CPS coupled with constructed BLS fatality risk measures by industry and occupation group, we show that inclusion of relative position variables in a canonical wage equation reduces VSL estimates by 25-33%. This effect is the opposite of what the relative position theorists have hypothesized. In contrast, recognition of the worker's relative position within the life cycle raises VSL estimates by up to 20%, especially for older workers.

JEL Classification: J3, I1