No CrossRef data available.
Article contents
Some Comments on Short-Run Earnings Fluctuation Bias
Published online by Cambridge University Press: 19 October 2009
Extract
Considerable attention has been given in recent years to studies of the cost of capital and investor preferences for dividends, but empirical analysis of the various theoretical propositions has been complicated by numerous difficulties associated with estimating the parameters in question. In many of the empirical studies, ordinary least squares has been the statistical method applied because, under certain assumptions, ordinary least squares produces unbiased estimates with a minimum of variance. However, the validity of one of the assumptions, namely, that all independent variables contain no measurement error, has been questioned by numerous writers. More specifically, it has been argued that investors tend to ignore temporary or random disturbances in current reported earnings. Thus, values of current earnings that are actually observed for the statistical study may be different from those values which ideally would be observed; i.e., they may contain measurement error, the size of which is reflected by the difference between the current reported earnings and the values on which investors presumably base their decisions. For example, in such least squares regressions as price on dividends and retained earnings, the impact of the measurement error is considered to fall mainly on the retained earnings, since dividends are thought to be relatively stable.
- Type
- Research Article
- Information
- Copyright
- Copyright © School of Business Administration, University of Washington 1970
References
1 Lintner, John, “Distribution of Incomes of Corporations Among Dividends, Retained Earnings, and Taxes,” American Economic Review, 46 (May 1956), pp. 97–113Google Scholar, documents the dividend stabilization practices of corporations, which lends support to the belief that errors of measurement fall mainly on retained earnings. It can be argued that, since dividends reflect long-run earnings expectations because of the stabilization practices of corporate management, the use of current reported earnings as a proxy for expected earnings of a period results in the error falling on retained earnings, retained earnings being equal to earnings minus dividends.
2 Harkavy, Oscar, “The Relation Between Retained Earnings and Common Stock Prices for Large, Listed Corporations,” Journal of Finance, 8 (September 1953), p. 288.Google Scholar
3 Modigliani, Franco and Miller, Merton H., “The Cost of Capital, Corporation Finance, and the Theory of Investment: Reply,” American Economic Review, 49 (September 1959), p. 666.Google Scholar
4 Ibid., p. 667.
5 See, e.g.: Benishay, Haskel, “Variability in Earnings-Price Ratios of Corporate Equities,” American Economic Review, 51 (March 1961), pp. 81–94Google Scholar; Gordon, Myron J., “The Savings, Investment, and Valuation of a Corporation,” Review of Economics and Statistics, 44 (February 1962), pp. 37–49CrossRefGoogle Scholar; Friend, Irwin and Puckett, Marshall, “Dividends and Stock Prices,” American Economic Review, 54 (September 1964), pp. 656–682Google Scholar; Miller, M. H. and Modigliani, Franco, “Some Estimates of the Cost of Capital to the Electric Utility Industry, 1954–57,” American Economic Review, 56 (aJune 1966), pp. 333–391.Google Scholar
6 See, e.g., Benishay, op. cit.
7 Op. cit., esp. pp. 668–669 and 676–677. Friend and Puckett assume that prices and dividends are always “normal” and, hence, that dividend-price ratios are normal but that earnings-price ratios are subject to short-run fluctuations. They estimate normalized values for retained earnings (R) as follows by using a normalized earnings-price ratio (E/P)n. The (E/P)n is a normalized earningsprice ratio found by deriving time-series regressions of the form (E/)it/ (E/)kt = ai + bit for each of i companies in some kth industry group. Normalized earnings-price is then found by 
Normalized retained earnings, 
is found by subtracting dividends from normalized earnings computed by (ai + ait)(E/)kt(P)it.
8 See, e.g., Miller and Modigliani, op. cit., and Brigham, Eugene F. and Gordon, Myron J., “Leverage, Dividend Policy, and the Cost of Capital,” Journal of Finance, 23 (March 1968), pp. 85–103.CrossRefGoogle Scholar
9 The frequently used ratio formulations of price-earnings ratios taken as a function of dividend—earnings ratios are an equivalent form of (1). For example, the model P/E = a + b(D/E) reduces to P = aR + (a + b)D, so that a large “b” in the ratio form is equivalent to having a relatively greater coefficient for dividends in (1). Equal weights on the dividend and retained earnings coefficients in (1) would be equivalent to a “b” coefficient equal to zero in the ratio formulation.
10 See Brigham and Gordon, op. cit., p. 94.
11 Four-and ten–year time periods were also run. Results for these other periods were very close to those obtained with the 7-year period, with the 4– year period generally producing slightly lower weights on R than was the case for the 7- and 10-year periods, which were very close in their final results.
12 incomplete data for several companies resulted in slight variations in the number of companies available for observations in some years shown in Table 5 as compared with those of Tables 2–4.
13 For example, Miller and Modigliani, op. cit., used a more complex approach (a form of two-stage least squares) for handling the short-run earnings fluctuation bias. We have not attempted to perform their computations in this context since serious questions have been raised regarding the validity of their approach [see Crockett, Jean and Friend, Irwin; Gordon, M. J.; and Robichek, A. A., McDonald, J. G., and Higgins, R. C., “Some Estimates of the Cost of Capital to the Electric Utility Industry, 1954–1957: Comment,” American Economic Review, 52 (December 1967), pp. 1258–1288.Google Scholar