A Comment on 'the Information Content of Earnings and Prices : A Simultaneous Equations Approach' by W.H. Beaver, M.L. Mcanally, and C.H. Stinson (1997)

We comment on a recent paper by Beaver, McAnally and Stinson (1997), drawing attention to the fact that their method ignores some recent developments in time-series econometrics. We apply a bi-variate vector autoregression framework to price and earnings data of listed US companies and the Samp;P 500 index to capture the dynamics of this system of equations. Although some series are I(1), cointegration does not exist and consequently we are unable to estimate a vector error correction model. However, by applying generalised variance decomposition and generalised impulse response analysis, we conclude that evidence favours a linkage from prices to earnings