This paper investigates the behavior of residual income scaled by beginning of period book value, i.e., residual return on equity (ROE), by performing panel unit root tests as well panel regression tests on Swedish data. Results show that residual return on equity does not follow a random walk, nor is it transitory, yet it is much more transitory than what previous studies indicate. The study also proposes a method to design similar studies such that they can use data from privately held firms.