BOOK REVIEWS

A Review of Economic Theory. By EDWIN CANNAN. London: P. S. King & Son, 1929. Pp. x+448. 16s.

The title, A Review of Eonomic Theory, of this latest publication of Professor Cannan is, in view of the nature of the subject matter and the method of its treatment, probably as appropriate as any that could have been devised. As it is, the title conveys, perhaps, too broad a view. The book could, very properly, have been called A Review of Economic Theory in English Economics. This is, in fact, what it is, though a considerable number of English students of repute receive but scant mention or none at all, while the rich contributions of American, French, German, and other European scholarship since John Stuart Mill occupy a very small space.

A perusal of the book calls to mind the author's earlier and perhaps best-known work, A History of the Theories of Production and Distribution in English Political Economy from 1776 to 1848. This earlier work starts with Adam Smith and ends with John Stuart Mill, and pays very little attention to what went before or came after. The present volume reviews again the history of the theories of production and distribution, but in much briefer form and in entirely new dress, and with much space devoted to the origin of ideas before Smith's time and to the development of theory after Mill's time. The theory of value, which found no place in the earlier study, supplies the longest chapter in the new volume. A critical analysis of the theories of production, value, and distribution comprises eleven of the fourteen chapters of the book. The first two chapters are introductory in character and deal with the origins of economic theory and the evolution of the name of the science. A concluding chapter, called "Aspirations and Tendencies," discusses present social-economic problems and the general tendencies of economic progress.

The limitations of space set on this review permit only the briefest discussion of the results of the author's labors. In a book devoted largely to the development of economic theory in England, one might expect to find mercantilism assuming an important place in a study of origins, but it is not so in this instance. Professor Cannan seems to form his conception of mercantilism from the views of Mun and

Smith. "Mercantile theory,'" as he sees it, "should be the doctrine that states must not leave the balance of trade to take care of itself, but must encourage exportation and discourage importation, and 'Mercantilism' should be ... either this theory or the practice which it recommends" (p. 13). He regards this as a "convenient" use of the term and sees no convenience at all in the use which some German and English writers have made of it as a term to describe a nationalistic economic movement. Yet a view of the whole body of mercantilist literature rather clearly reveals a conception of the interrelations of agriculture, manufactures, commerce, population, and monetary institutions. Mercantilistic writers sensed the unity of the economy of nations and, in some measure, also, the unity of world-economy. Mercantilism seems, therefore, to have some rightful claims, particularly in England, to be regarded as the earliest form of economic theory.

In the fourth section of the chapter on origins, cameralism is brought into the picture, but one is not a little surprised to find this branch of early economic literature represented solely by Antoine de Montchrétien, author of the Traicté de l'Economie politique, 1615, who is characterized as "an extreme example." There is not the slightest reference to any one of a number of seventeenth- and eighteenth-century writers whose names naturally come to mind when cameralism is mentioned. No attempt is made to show the influence of cameralism on the development of economic science and the comment on Montchrétien's book seems to be directed rather more to the softening of the unfavorable judgment which economists have generally passed on it than to a critical estimate of its scientific value.

In tracing the course of events after Mill, Professor Cannan gives much space to the development of economic theory in the hands of Alfred Marshall. This is especially true in the chapters dealing with the theories of value and distribution, in which, in a general way, he traces the evolution of Marshall's thought from the Economics of Industry in 1879 through the eight editions of the Principles. There is abundant evidence of his recognition of the great extent and importance of Marshall's contributions, but he is never quite satisfied with his attainments and finds much to criticize. His attitude toward Marshall's teaching is well illustrated by many passages in the chapter on "The Theory of Value in General." Referring to the value of commodities produced under conditions of diminishing cost, he says (p. 209):

It is . . . . one of the commonest of all economic experiences that enormous numbers of commodities have become cheaper because increase of population or increased means or desires of the same population to buy them has made it possible to produce them "on a larger scale." The phenomenon is much too big to be airily dismissed as "an exception to the general rule that increase of demand tends to raise value." Marshall does not try to treat it as such, but he does not seem quite happy in endeavoring to find it a place in his general theory of supply and demand and value.

In the matter of long-period prices, the author says (p. 210) that Marshall is obliged to follow Ricardo's classification of commodities in accordance with their tendency to obey a law of diminishing or constant or increasing returns. An increase in demand in the first case raises value, leaves it unchanged in the second, and lowers it in the third. This, he concludes, destroys what Marshall calls "the fundamental symmetry of the general relations in which demand and supply stand to value." Then, quoting Marshall's analogy of the scissors, he says (p. 210):

The simile is striking and effective, when applied, as Marshall applies it . . . but is very puzzling when applied to a commodity which will fall in value when more of it is demanded. The two blades of the scissors then seem to be required to move in the same direction, which is contrary to the nature of scissors.

While Marshall's exposition may not be all that could be desired, his critic seems to detect no fundamental error in his reasoning. It is even not impossible for the two blades of the scissors to move in the same direction. The cutting operation would continue, provided one blade is allowed to overtake the other.

Again Professor Cannan thinks (p. 214) that to construct an imaginary "stationary state" in which "economic forces have had time to work out their effects," followed by admission of the fact that production is not so arranged in the real world that equal efforts and sacrifices are equally rewarded, is a bad way to approach the explanation of values and their changes. He says:

...

History tells us how certain values have changed; let theory tell us why such changes took place. . . . . It will be far better to accept the values known to exist. . . . as the result of the physical conditions of the universe and the past history of mankind, and to conceive the theory of the influence of production on values as an orderly discussion of the causes of actual and possible changes. . . .

It would be interesting to have Professor Cannan's theory of value developed in accordance with these suggestions, but he indulges in comparatively little constructive theorizing in this volume.

While the book is neither a systematic exposition of principles nor a complete history of doctrines, it will prove to be an invaluable aid to the future student of the development of economics in England. Like all the books of Professor Cannan, it is written in an attractive style, is orderly and systematic in its presentation of the subject matter, and extremely accurate in matters of fact and in its references to the literature of economics. It could easily have been made much more useful to students by listing the writings of the numerous authors referred to in the text in an orderly manner in a bibliographical note or appendix.

UNIVERSITY OF ILLINOIS

N. A. WESTON

An Approach to Definite Forecasting. By LINCOLN W. HALL. Philadelphia: University of Pennsylvania Press, 1929. Pp. ix+142. $3.00.

The cover page announces that "in this study of business statistics a more powerful, more complete, more easily calculated methodology is set forth for handling all forms of time data in business than has previously been available. An extensive use of constructed series has been made for the purpose of experimentation with the methods described under an approximation of controlled conditions. Definite numerical forecasting, wherein errors can be more accurately checked, may be attempted through the process suggested in this book." The high hopes thus raised in the mind of the reader who is eager for both a time-saving and a more accurate device for forecasting economic series, are hardly fulfilled by the contents of the book.

An Approach to Definite Forecasting is an attempt at projecting time series a year in advance on the basis of the secular, cyclical, and seasonal components of the time series itself. It avoids the usual methods of determining secular trend by fitted curves or moving averages. In their stead the author uses a device called "the cyclical indicator" to determine a trial secular trend line. Deviations from this trial trend are averaged to obtain trial seasonal variations. Each of these three components are then adjusted giving "final" components, and the latter form the basis for projecting or forecasting the time series for a year

haps even because of it-his work is of real value to the men who stay in this field. For its effect is to call attention sharply to certain neglected elements, particularly the sexual, in that background which they must understand if they hope to appreciate either the limits or the possibilities of union action. And it is of even greater importance in forcing the realization that their work, after all, may be only a small part of the story, and that the significance of the problems the unions uncover may be far greater than that of any changes the organizations themselves are likely to propose. For strikes and unions are symptoms of more ills than they alone can cure, and it is the vigor and clarity with which this point is presented that makes "The Casual Laborer" one of the most significant introductions to the wider questions of social science.

UNIVERSITY OF MICHIGAN

CARTER GOODRICH