Abstract
A reexamination is made of the formula for the variance of lifetime income in a stochastic process income distribution model. The formula in the received literature is shown to introduce a significant upward bias in estimates of this variance.