Knowledge that Transforms

J. L. Dake, Comment: An Empirical Test of Financial Ratio Analysis, The Journal of Financial and Quantitative Analysis, Vol. 7, No. 2, Supplement: Outlook for the Securities Industry (Mar., 1972), pp. 1495-1497

In a previous paper we reviewed the literature on normative stock price models. These models specified the present value of a share of common stock to be equal to the discounted value of dividends accruing to the holder. Using continuous discounting, the present value of a share is (1) where Dt is the dividend rate at time t and k is the cost of equity capital. Presumably k is a function of both the stockholders' time value of money and the perceived risk or uncertainty associated with the future dividend stream.

The interdependence of loan cost and tangible bank activity is an aspect of the cost of bank debt that has not been treated in the literature. Understanding this interdependence is important for banks in pricing their services, especially as banks adopt more flexible pricing policies. This understanding is crucial for a firm in establishing the true cost of bank borrowing, in comparing bank borrowing with other sources of funds, and in evaluating the firm's banks. It is also important for understanding the firm-bank relationship in general and the cost of capital in particular.

This paper will demonstrate that different margin levels are associated with the price behavior differences of certain commodity futures. In 1959, Harry Roberts suggested the methodology of rational subgrouping as a means of testing random versus systematic price changes. His methodological suggestion has had only limited testing in security markets and no direct application to domestic futures markets. This paper uses margin levels as a basis for rational subgrouping of selected commodity futures. Evidence supports the argument that, when a series of price changes is grouped according to margin levels and these levels are analyzed separately, nonrandom characteristics that tend to be offsetting in the aggregate series become evident. The nonrandom behavior observed is consistent with the hypothesis that, in certain periods, margin levels have been set too high to attract a volume of speculative services necessary for the maintenance of market balance.

Consider the following linear programming problem, which is denoted by (S): The various vectors and matrices have the following dimensions: a = an m-vector Aj = (j = 1 … n), an m by njmatrix, bj = (j = 1 … n), an mj-vector, Bj = (j = 1 … n), an mj by nj matrix, Cj = (j = 1 … n), an nj-vector, and xj = (j = 1 … n), a variable nj-vector.Problems with this decomposable structure have been extensively studied in economics and management science literature, because they arise in several economic contexts and situations. Therefore, (S) lends itself to more than one economic interpretation, but the following one is hereby adopted for the purposes of this paper: (S) models a two-level manufacturing organization, consisting of headquarters and n producing divisions. Let this organization be called simply “the company”. The company can produce and sell a number of different products, thereby obtaining fixed unit contributions. These contributions are given by the vectors c1, c2, … cn. The products are produced by the different divisions. The variable vector xj (j = 1 … n) gives the production program for the jth division.

In his famous 1964 article, Paul A. Samuelson [1] presents what he calls “the only sensible definition of depreciation, ” namely, the period change in the present value of future cash flows expected to be generated by an asset.

One of the most vexing areas of business finance is empirical validation of theory. Many problems arise that the investigator cannot avoid. The purpose of this paper is to define a class of problems that can be avoided but that have caused many previous investigators difficulty, even though they may not have realized it. In what follows, the concept of dimensional analysis, borrowed from the physical sciences, is introduced. Then dimensional analysis is extended to the interpretation of estimated regression coefficients.

The dividend decision is a residual corporate decision and yet it is not a residual corporate decision. This theme runs through Professor Higgins' paper. It is based on the one hand on Higgins' belief that other decisions are more important and should take precedence over the dividend decision, and on the other, that the dividend decision is a function of these more important decisions. If dividends were truly residual, they would be explained by the corporate budget identity. In the empirical analysis, however, dividends are a function of other balance sheet decisions. In this sense the decision is neither active nor residual but passive. The reader wonders why the author insists on this role for dividends when interdependency in decision making is the more appealing (and inevitable) framework.

The facts of increased institutional trading on the nation's securities markets are by now well known. On the New York Stock Exchange (NYSE), the six major institutional groups—insurance companies, investment companies, noninsured pension funds, nonprofit institutions, common trusts, and mutual savings banks, now own more than one-fourth of the market value of listed shares compared with less than 16 percent at the end of 1956. But, ownership is merely the tip of the perennial iceberg, since institutional trading of stock has become much more significant than institutional ownership. This fact is pointed up in the recent SEC Study of Institutional Investors. It shows that there has been a relatively slow increase in the share of outstanding stock owned by institutions in all markets, but the institutional share of trading has mushroomed.