Accounting Earnings, Permanent Cash Flow and the Distribution of the Earnings to Price Ratio

This paper explores the suggestion that users of financial statements seek a ‘normalized’ version of earnings from which to estimate equity value. We argue that this leads to a theory of earnings to price ratios which is rooted in thecertainty-based ‘economic income’ models of Hicks. Methods laid down in previous research are then used in conjunction with a neoclassical model of capital accumulation to generalize the Hicksian income model to uncertainty. Our analysis shows that in ‘steady state’, Pearson's Type IV distribution ought to be a good candidate for the statistical process which describes the evolution of the earnings to price ratio. This, in turn, implies that the probability density for the earnings to price ratio will be skewed, with the possibility of ‘thin’ or ‘fat’ tails relative to the normal distribution. Also, some of the lower moments may not exist. Our empirical evidence is in broad agreement with these predictions.