Knowledge that Transforms

SOME of the highest profit rates appear in industries that advertise heavily. These high earnings have been attributed to barriers to entry associated with product differentiation [2, 6]. A possible alternative explanation is that the treatment of long-lived advertising as current expenses leads firms that invest heavily in such intangibles to overstate their rates of return since their equity is understated [1, p. 153; 15, p. 167]. The same practice may result in the understatement of their dollar profits so that they pay less tax than other firms whose investments are all tangible. The purpose of this paper is to work out more precisely the overor under-statement of profit and rate of return involved in the expensing of advertising and to evaluate the mis-statement in practice.' Part I develops conceptually the conditions under which overor under-statements can be expected. Part II recomputes dollar profits and rates of return for a variety of industries, estimates the tax avoidance that results, and examines the relationship between recomputed profit rates and advertising. Part III contains a proposal for policy change.

ECONOMISTS have shown considerable interest in the relationship between productmarket competition and wage rates. Most of the analysis has centered on manufacturing, where the less competitive industries often have larger firms and larger plants. This paper presents evidence that differences in plant size are at least as important as differences in market structure when we try to account for wage differentials among manufacturing industries.

A NUMBER of studies have yielded evidence of significant association between certain characteristics of industry structuresuch as high concentration and substantial entry barriers and variations in industry performance, particularly with respect to profitability.* In general, these studies tend to confirm the expectation that, other things being equal, profits will tend to be higher in industries in which structural conditions depart substantially from those of the competitive model. However, as Stigler [16, p. 145] has noted, the statistical associations found are usually weak, and a substantial amount of performance diversity is left unexplained. Thus, the typical strength and character of the structure-performance associations, and the importance of individual structural factors in the overall pattern, have remained open to question. Many hypotheses have been suggested; only a few are subject to serious empirical investigation; fewer still have actually been examined. This paper presents a summary report on our efforts to test a small number of fairly straightforward structure-performance hypotheses against the most comprehensive collection of relevant data available, the concentration statistics for 1958 and 1963 [18, 19]. These tests have focused on a single performance measure, the percentage price-cost margin, which we take as an indicator of the ability of firms in an industry to obtain prices in excess of direct costs. We have found a significant association between the price-cost margin and the level of four-firm concentration among fourdigit SIC (Standard International Trade Classification) industries; and this association is not eliminated when differences in capital-intensity among industries are taken into account. We have further found: (1) a tendency for the strength of the concentration-margins association to increase over the period 19581963, particularly in industries in which the level of four-firm concentration was stable or increasing; (2) a substantially stronger association between concentration and margins in consumer goods industries, as compared to producer goods industries; and (3) evidence that the principal component of the concentration-margins association in consumer goods industries is a correlation between concentration and margins of the four largest firms alone, in those industries in which these firms have higher margins than their smaller rivals. The first section of this paper establishes the background and framework of our analysis, and the following sections present the evidence of these findings in some detail.

HE objective of this study is to estimate T the magnitudes in which each of several factors influenced interstate migration over the period 1955-1960. Several variables were chosen which might reasonably be expected to explain the movements which occurred, and multiple regression analysis was used on the data. The most unique explanatory variable employed is the stock, i.e., the number of persons born in state i (the origin state) and living in state j (the destination state).' It is shown that the failure to include the migrant stock variable in the estimated relationship causes the true direct effect of most other variables to be obscured.

In recent paper this REVIEW,' Farrar and Glauber (hereafter FG) revisit problem of regression analysis. Viewing problem of as both facet and symptom of poor experimental design, 2 FG propose a three-stage hierarchy of increasing detailed tests presence, location, and pattern, 3 of multicollinearity. The first this series of three tests, on which other two are conditional, is desigped provide useful first measure of presence and severity of multicollinearity 4 sample on hand. Bartlett's well-known statistic testing joint distribution of sample correlations under assumption of vanishing parent correlations between variables is used by FG detecting multicollinearity. Bartlett shows that (natural) logarithm of intercorrelation determinant computed from sample drawn from multivariate, ortho-normal distribution, multiplied by factor k, is approximately distributed as Chi Square with v 1/2 n (n 1) degrees of freedom, where k = -[N 1 1/6 (2n + 5)], N is sample size and n is number of variables considered. If investigator concludes from first stage that exists and that it is severe enough warrant some action, FG propose regress consecutively each explanatory variable on remaining ones. The rp'silting F statistics will test for dependence of particular variables on other members 5 of set of explanatory variables. Finally, patterns of interdependence among independent variables are examined by testing significa-nce of partial correlations of every pair of explanatory variables, all other variables held constant. The main pillar of this three-level test is, of course, Bartlett's test which is properly used making inferences,6 under null hypothesis that all population correlations are zero. Since FG claim, however, that they are not interested drawing inferences from sample population (inferences from sample population . . . are possible . . . however, little importance is attached properties of population from which set of data has been drawn. Attention focuses largely, if not entirely, on sample itself 7), their use of Chi-Square statistic is questionable. Moreover, it is neither practical nor necessary assume orthogonality between parent economic variables, 4f one wishes make such inferences. Here we come heart of problem of multicollinearity. One may agree with FG that it is preferable think of in terms of [its] severity rather than its existence or nonexistence. 8 If one agrees with this approach, natural way proceed is indeed to define terms of departures from hypothesized statistical condition. 9 But what is this hypothesized condition? For FG this condition is the requirement that explanatory variables be truly independent of one another. 10 However, there is no such requirement least-squares solution. On contrary, least squares solu* I share with D. C. Farrar and R. R. Glauber my indebtedness Professor John R. Meyer who introduced us problem, and I am grateful his comments on an earlier draft. I am also grateful Professors J. Johnston, N. Wallace, D. Farrar, and R. Glauber valuable discussions. I am particularly thankful D. Farrar who did not spare his efforts order dig out old forgotten data, which enabled me recompute his regression equations. 1D. C. Farrar and R. R. Glauber, Multicollinearity Regression Analysis: The Problem Revisited, this REvIEw, XLIX (Feb. 1967). 2Ibid., p. 93. 3 Ibid., p. 104. ' Ibid., p. 101. BIbid., p. 104. Bartlett has originally developed this statistic order test number of meaningful components that can be extracted from set of variables. concise statement is given by Bartlett: A Note on Multiplying Factors Various x2 Approximations, Journal of Royal Statistical Society (B), XVI, no. 2 (1954), pp. 296-298. 7D. C. Farrar and R. R. Glauber, op. cit., 100. 8Ibid., p. 106. 9 Ibid., p. 92. 10Ibid., pp. 92 and 100.