On Modelling the Persistence of Profits in the Long Run: A Test of the Standard Model for 156 US Companies, 1950–99

Abstract

‘Persistence of profits’ studies of competitiveness across samples of firms, and for individual firms, have almost always employed a simple first order autoregression model. Reservations over the use and interpretation of the AR1 in this context raise questions both over the reliability of previous results, and for future research strategy. We test the standard model on a common sample of 156 US companies over a 50 year period against a recently proposed, alternative model that adopts structural time series methods. A statistically significant degree of consistency is found between the two approaches in identifying firms persistently above or below the competitive norm in the long run. However, the structural time series method detects a much higher overall incidence of persistence, and appears to offer advantages in cases where the profit dynamics are more complex.