Knowledge that Transforms

Professors Friedman and Becker have recently presented some interesting calculations, embodying six different hypotheses about the determinants of consumption expenditure.1 An examination of the relative efficacy of these alternative functions in predicting income from a known or estimated level of led the authors to the rather startling conclusion that it is a better first approximation to regard real as having no multiplier effects on real consumption whatever and real consumption as simply determined by its own longterm trend, than to regard real as the prime mover and changes in real income as largely mirroring changes in real investment (page 75). Although Professors Friedman and Becker have been concerned solely with the predictive power of various schemes, they suggest that their results indicate a substantive conclusion about the structure of the economy, namely that the simple Keynesian consumption function, in which consumption is largely determined by current income, is not a basically correct, even though admittedly oversimplified, specification of the economic structure. If our analysis is valid, this widespread belief that consumption and income can be regarded as dancing primarily to the tune of unstable is largely the product of a statistical illusion (page 74). The multiplier would appear to be dead. But before we commit the body irrevocably to the grave, I would put forward the following points for consideration. i. It is legitimate and useful to regard the predictive power of a model as a test, but only one test, of the validity of that model. The test, however, is usually made by predicting values for periods outside those from which the model has been estimated. The Friedman-Becker test is concerned with the relative error within the period to which their various consumption functions have been fitted. This may be a crucial factor in determining the conclusions they have reached. During I906-5I, the United States economy has functioned in such a way as to generate a fairly persistent upward trend in real per capita consumption. Thus a trend model such as Ct = I62.0 + 237.oeO2t + Ut

A CCORDING to generally accepted wage 11 theory, the forces affecting the occupational wage structure in the past operated to narrow skill differentials. Statistical evidence supported this theory. This paper presents the view that though the skill differential has narrowed in the past, the current tendency is for a stabilized differential, and the future trend is toward a widening differential. Before analyzing the forces affecting the differential, existing statistical studies of skilledunskilled wage relationships are reviewed briefly in order to describe recent trends in differentials. Throughout the study only both extremes of the occupational wage scale -the skilled and unskilled wageare discussed; a more comprehensive study of changes in the occupational wage structure would entail discussion of movements of all classes within the range, in particular, fluctuations in semi-skilled wages. What follows pertains to the American scene, but in developing the theoretical arguments references will be made to foreign experience.

ONE of the ways of tracing changes in business concentration is to compare the rates of growth of firms of different sizes. If firms that are large at a certain date subsequently grow on the average more rapidly than small firms, business concentration will obviously have increased. It is not however always realized that the obverse proposition does not hold; that is, if the large firms have grown less rapidly than the small firms then -however paradoxical it may seem -concentration does not necessarily decrease. A related paradox -that associated with what is known as jobbing arises if we look at the average rate of growth achieved in the past by firms that are large today, and compare it with that for smaller firms. This differs from the previous example, since we are now looking backward in time instead of forward; as will be seen below, the two points of view are symmetrically related to one another. The matter is more complicated in its logic than appears at first sight, and it is perhaps not surprising that many careful empirical investigations are to be found in the literature which are vitiated by a logical fault in the inferences drawn from them. The errors are generally pointed out subsequently, only to be repeated by the next generation. The present note attempts to give a simplified but systematic exposition of the necessary conditions for changes in concentration; the basic algebra has already been set out from a different point of view in an earlier paper,' but the approach adopted here may be easier to follow. The statistician will recognize all that follows as being no more than the simple theory of in the sense of Galton (not in the modern sense, where regression is often taken as equivalent to the procedure of fitting a line by least-squares). Our understanding of the problem owes much to the review by Hotelling in the Journal of the American Statistical Association, xvIII (933), 463-65 of Secrist's The Triumph of Mediocrity in Business (Chicago, I933), and the subsequent discussion, ibid., XIX (I934), I96-2 00.2 Similar arguments are to be found in other contexts in the writings of Professor Milton Friedman.

PpT HIS note, pertinent to some recent literature on public goods,' presents two propositions which place the question of the optimal expenditure levels for public goods in a somewhat different perspective. By (pure) public goods or, synonymously, (pure) collective consumption goods, we mean those consumer goods having the property that, once produced, their enjoyment by each and every individual does not reduce their availability for the enjoyment of others. Public defense and public health measures may suggest cases in point.

COUNTERCYCLICAL effects of public expenditures have received voluminous treatment in the economic literature of the past thirty years. There is one significant area, however, where improved data and analytical tools may provide new understanding. This is the area of public expenditure effects on employment and industrial output throughout the economy. The question of economy-wide effects was of considerable concern to early analysts of government work and relief programs. Early studies of these effects ordinarily began with a limited number of final materials and laboriously traced their production back through intermediate stages. In contrast to these earlier methods, input-output analysis, as introduced in Professor Leontief's The Structure of the American Economy in I94I, started with a comprehensive picture of the production relationships among all industries and provided a formal technique for tracing out economy-wide production and employment repercussions. The Bureau of Labor Statistics extended Leontief's empirical work into the postwar era, and in I952 published the results of a massive and detailed study of interindustry relationships for the year I947.' The present study uses the same interindustry data to investigate the employment effects of a variety of possible government spending programs. These include several types of public works, defense procurement (195I pattern), and general government spending at both the national and local levels. In most cases these spending patterns were those observed for the year I947, but some material for I954 was also available. Because government policies can also influence expenditures in other parts of the economy, the analysis has been extended to private house construction, all industrial construction, consumer spending (which may be increased through tax cuts, expanded government payrolls, or direct relief payments), and private capital investment. The analysis of these expenditure patterns, however, is of interest to countercyclical policy only to the extent that government policies actually do bring about the type of spending described. Finally, averages for all new construction (1947 pattern), all maintenance construction, and Professor Leontief's earlier results on foreign trade are included for comparison.