non-calculating, impulsive man, whose economic reactions are wholly instinctive, is better than one who calculates and carefully compares costs and advantages. The latter is certainly more "human," if by that word we imply the possession of qualities which distinguish us from, rather than ally us with, the loveable brutes. After all, calculating costs and advantages seems to work well, or at least better even from the social point of view, than "going it blindly," or relying upon our instincts and impulses, however admirable these may be. The investor, for example, who calculates carefully whether a certain enterprise will pay or not is really, in the last analysis, calculating whether the utilities resulting from it will be greater or less than those destroyed or wasted in carrying it through. A nation which encourages this kind of calculation on the part of its citizens is more likely in the long run to prosper than the nation which does not.

It is the writer's opinion that "behaviorism " fits into the classical scheme of economics, and fills it out by furnishing detailed analysis and description where they were lacking before. In this respect it performs the same function as that performed a generation ago by the marginal utility theory. This was not revolutionary, tho some ardent souls who did not see where it fitted in thought, for a time, that it had scrapped the whole classical system. On the contrary, it merely furnished an analysis of demand comparable with the analysis of supply which the classical school had worked out. The behavioristic school is furnishing a detailed analysis and description of a group of factors which were taken for granted by the marginal utility and marginal productivity analysis.

In the theory of differential cost, for example, it has long been understood that cost is disinclination, and that the disinclination is not a fixed and unvarying thing. There is no absolute disinclination to work, to risk or to save. Much work is done for many reasons besides the hope of wages, much saving without the hope of interest, much risking without the hope of profits or other objective gain. There is room here for the "instinct of contrivance," for a squirrellike impulse to hoard, for the pleasure of skating over thin

A parallel line of division is found among students of the question of the relative influence of heredity and environment as determining factors in individual success. By analogy, if one were studying jellyfish, one might find them to be the sport of circumstances, the winds, the waves, the tides and the currents. Environment seems to be everything until one asks, why were they jellyfish; then heredity comes in. A human weakling likewise seems to be the sport of circumstances. If they are favorable, he turns out well; if unfavorable, he turns out badly. Again, environment seems to be everything, until one asks, why is he so weak as to be the sport of circumstances. Then heredity has at least to be considered. If one studies sharks, however, one does not find them to be the sport of circumstances, at least not of the same circumstances as those which control the jellyfish. What is the difference? It is partly that one was born a shark and the other a jellyfish. Two men go into a shower bath. One comes out with a glow and the other with a chill. To the one a cold shower bath was favorable, to the other it was unfavorable. The difference is in the men. Two men grow up in a slum environment. One comes out sound, strong and virile; the other diseased, weak and parasitic. Why did they turn out differently?

As to the relative importance of the so-called rational and non-rational factors in human behavior, there is no reason to believe that all men are alike in this or any other particular. It may be that successful men are influenced somewhat more by the rational factors than are unsuccessful men. At any rate, the classical school with its so-called economic man was as truly a behavioristic school as any group of recent students. They were studying a different class of men in the industrial system. Perhaps both schools develop a one-sided theory of human nature because of the fact that each is studying a different class one the successful, the other the unsuccessful class.

This has an important bearing on the question before us. There is no a priori reason for concluding that one kind of man is better than another, certainly not for concluding that a

non-calculating, impulsive man, whose economic reactions are wholly instinctive, is better than one who calculates and carefully compares costs and advantages. The latter is certainly more "human," if by that word we imply the possession of qualities which distinguish us from, rather than ally us with, the loveable brutes. After all, calculating costs and advantages seems to work well, or at least better even from the social point of view, than "going it blindly," or relying upon our instincts and impulses, however admirable these may be. The investor, for example, who calculates carefully whether a certain enterprise will pay or not is really, in the last analysis, calculating whether the utilities resulting from it will be greater or less than those destroyed or wasted in carrying it through. A nation which encourages this kind of calculation on the part of its citizens is more likely in the long run to prosper than the nation which does not.

It is the writer's opinion that "behaviorism" fits into the classical scheme of economics, and fills it out by furnishing detailed analysis and description where they were lacking before. In this respect it performs the same function as that performed a generation ago by the marginal utility theory. This was not revolutionary, tho some ardent souls who did not see where it fitted in thought, for a time, that it had scrapped the whole classical system. On the contrary, it merely furnished an analysis of demand comparable with the analysis of supply which the classical school had worked out. The behavioristic school is furnishing a detailed analysis and description of a group of factors which were taken for granted by the marginal utility and marginal productivity analysis.

In the theory of differential cost, for example, it has long been understood that cost is disinclination, and that the disinclination is not a fixed and unvarying thing. There is no absolute disinclination to work, to risk or to save. Much work is done for many reasons besides the hope of wages, much saving without the hope of interest, much risking without the hope of profits or other objective gain. There is room here for the "instinct of contrivance," for a squirrellike impulse to hoard, for the pleasure of skating over thin

ice, or telling risqué stories. A study of these will add to our knowledge of what may be called "costless" production or production inside the margin. What are all these nonpecuniary motives? There is a large question here; and if the behaviorists can answer it in detail, they will have made a significant contribution to economics. But if they think that they have built up a complete system and can dispense with all that has gone before, they must be placed in the class with men in other fields, such as chemistry, physics, medicine, or zoölogy, who, because of some new observations, hasten to announce that all previous work is of no account.

T. N. CARVER.

HARVARD UNIVERSITY.

THE REVIEW OF ECONOMIC STATISTICS

The Harvard University Committee on Economic Research will begin in January, 1919, the publication of a review devoted to economic statistics. The Committee has had this project under consideration for nearly two years, and in the fall of 1917 engaged Professor W. M. Persons, of Colorado College, to undertake the study of methods of interpreting current economic statistics. Professor Persons is well known to the scientific world through the various papers he has published upon statistical methods and results, and he will have editorial charge of the new publication, which will appear quarterly. The purpose of the committee is to promote the collection, criticism, and interpretation of economic statistics, with a view to making them more accurate and valuable than they are at present for both business and scientific purposes. This, it believes, can be done by more thoro investigation of the sources and accuracy of statistical data, and by developing the application to economic statistics of modern methods of statistical analysis, which have hitherto been utilized more extensively in other sciences than in economics.

It is the hope of the committee that the proposed publication will, in due time, provide a more accurate record of economic phenomena than is now available, and will also supply a method of interpreting current statistics which will make them more useful and significant than they are today. Since many new scientific and technical problems will necessarily be encountered, the 1919 issues will be treated as a preliminary volume, and will not constitute the first volume of a permanent series. The review will be published in connection with a statistical service which the Committee on Economic Research proposes to establish. The cost of the undertaking precludes ordinary publication arrangements, and, therefore, the review will be supplied only in connection with the statistical service which is offered for the year 1919 to all who may desire to become subscribers at the price of $100 per subscription. Besides the review, it is expected that statistical service will include, after April or May, a monthly bulletin supplementing the quarterly publication. Since the purpose of the enterprise is primarily scientific, the quarterly publication will be supplied, without charge, to a limited number of university libraries and to scientific investigators particularly/ interested in the problems with which it will deal.

The Committee on Economic Research consists of Charles J. Bullock, Chairman, Charles F. Adams, Nicholas Biddle, Frederic H. Curtiss, Wallace B. Donham, Edwin F. Gay, Ogden L. Mills, Eugene V. R. Thayer.