Do firm effects drift? Evidence from Washington administrative data

Abstract

We study the time-series properties of firm effects in the two-way fixed effects model popularized by Abowd et al. (1999) (AKM) using two approaches. The first—the rolling AKM approach (R-AKM)—estimates AKM models separately for successive two-year intervals. The second—the time-varying AKM approach (TV-AKM)—is an extension of the original AKM model that allows for unrestricted interactions of year and firm indicators. We apply to both approaches the leave-out methodology of Kline et al. (2020) to correct for biases in the estimated variance components. Using administrative wage records from Washington State, we find, first, that firm effects for hourly wage rates are highly persistent with an autocorrelation coefficient between firm effects in 2002 and 2014 of 0.74. Second, the R-AKM approach reveals cyclicality in firm effects and worker–firm sorting. During the Great Recession the variability in firm effects increased, while the degree of worker–firm sorting decreased. Third, misspecification of standard AKM models resulting from restricting firm effects to be fixed over time appears to be minimal.