The developments of the past few years have, however, cast serious doubt upon the desirability of postponing comprehensive study of any important economic institution in the United States until such number of detailed monographs dealing with specific aspects of the subject have been completed as will permit full exposition and safe generalization. Experience seems to suggest that the prime usefulness of such intensive studies, very great though it be, is educational and local, and that variety of approach, distinctness of treatment, change of environment are grave qualifications, under existing conditions, of the value, and certainly of the economy, of large reliance upon this monographic method of economic investigation.

It may thus fairly be claimed that the time has now arrived when-without any cessation of the historical and local studies referred to above the economic investigator in the United States, if he is to attain his highest scientific possibility and to fulfill his greatest practical usefulness, must adopt a larger mode of inquiry, a mode that may perhaps be described as extensive or inductive, rather than intensive or institutional.

He must derive his subject-matter not from past history alone, nor from the present experience of restricted localities; but he must observe and collate the phenomena under consideration from an area practically co-extensive with their manifestation, and he must interpret each group of facts in the light of the conditions prevailing in that particular place. If he is attempting safe and useful generalizations, he must consider, for example, the taxation of corporations, not by one State, but by every State in the Union; he must study the question of trades-unions and the restriction of output, not with respect to a handful of labor organizations and a few convenient cities, but in the light of the policy, declared and actual, of every important national labor union, as displayed in many representative localities. In a word, the basis of economic induction in the United States must henceforth be, to a much greater degree than heretofore, qualitative data, at least approximating in comprehensiveness the quantitative material which the public statistician makes available with increasing efficiency.

In order that economic investigation may proceed along the "extensive" course thus outlined, at least four indispensable requisites must be present: (a) books and documents; (b) means of publications; (c) scientific leisure; (d) material resources. In the first place, the investigator must be able to command, in addition to ordinary bibliographical apparatus, all primary documentary material relevant to his inquiry, whether it be as ephemeral as municipal reports and trade-union journals, or as unobtainable by formal request as trade agreements and corporation records. Secondly, he must be able to publish the results of his investigations in the precise form which scientific fidelity or practical usefulness demands, without regard to their commercial attractiveness or to the limited publication resources of existing scientific agencies. Thirdly, his time and his energy, if not entirely available for scientific inquiry, must certainly not be unduly absorbed by the routine engagements of the student or the teacher. Finally, he must be in command of funds sufficient to enable him to visit, and upon certain occasions temporarily to reside in representative localities for the purpose of gathering additional evidence and of testing or verifying tentative conclusions. A more liberal policy of library administration, and a more intelligent appreciation of the proper relation of publication to investigation in the social sciences, have notably improved conditions in the past few years with respect to the first two requisites. The same can not be said with regard to the third and fourth essentials, and the need here is urgent.

To the extent that the economic investigator is still a student or teacher in academic attendance, opportunity for extensive inquiry can only come with greater prominence of field-work and laboratory exercise in economic instruction. Economic teaching can properly harken to the message of the physical sciences, that the ideal of student training is less the accumulation of detail than the development of a mode of thought. An association of courses, a reduction of lecture attendance and a unification of "seminaries" will ordinarily effect an economy of time, making possible that amount of experimental field-work demanded both by student development and scientific progress.

Similarly, the well-equipped department of political economy in the American university may be expected in the future to command such material resources as will relieve the properly qualified student, teacher or affiliated investigator from the present necessity of devoting himself in the main to historical, institutional or local studies. The immediate environment will be utilized as an economic laboratory for the development of scientific spirit in economic study and sound method in economic research, and as the field from which the bases of working hypotheses may be derived. Thereafter the investigator will be encouraged to extend the range of his inquiry by visits to, and even residence in, representative localities with a view to collecting wider and more varied data and to testing tentative conclusions, and he will be supplied with the means actually necessary for the purpose. To some extent such funds can be made available by a modification of the fellowship system-the original purpose of which, the attraction of students to post-graduate study, has ceased to be necessary, and the further extension of which along existing lines threatens serious evils. University authorities will recognize that economic "investigation funds" are as essential to scientific activity as physical apparatus and medical clinics. Beyond this, and doubtless to a far greater degree, aid may be anticipated from coöperation with governmental agencies and with endowed institutions of research. Less and less will lack of material resources operate as a handicap. As long as the method be sound and truth light the way, economic investigation will probably receive as generous an equipment as the economic investigator deserves. JACOB H. HOLLANDER.

Johns Hopkins University.

INCREASING AND DIMINISHING COSTS IN INTER

NATIONAL TRADE.

INTERNATIONAL trade presents problems of great inter

est in both practical and theoretical economics, but it is a familiar fact that the solutions of the practical statesmen have not shown any general agreement with the conclusions of the theoretical economists. In recent years some progress has been made in the scientific discussion of the question, and the latest results have not always supported the conclusions drawn by the original exponents of the theory. Important modifications of these conclusions have followed from a more extensive analysis of the subject and a more exact or "quantitative" treatment of values. It is proposed here to examine and illustrate certain particular problems, namely, some of those affected by increasing and diminishing costs.

Almost from the beginning the facts of increasing and diminishing costs were recognized as of some importance in determining the advantage of international trade,1 but only recently have they received something like a general appreciation. Economists are agreed that, roughly speaking, increasing amounts of manufactures are produced at diminishing costs per unit, in labor and capital, while on the other hand agricultural commodities are produced at increasing costs. This, at least, is likely to be the case where manufactures involve a large use of machinery, kept "up-to-date," and where agriculture has passed the first stages of development, i. e. extensive cultivation.2 If a country, therefore, engaged chiefly in agriculture, exchanges its products for those of a manufacturing country, as their trade increases in extent, the former will find itself in a relatively unfavorable condition, and the latter in a relatively favorable condition. The significance of this fact in practical economics has been strongly

1 Cournot, Recherches sur les Principes Mathématiques de la theorie des richesses, 1838.

I have been told that John Rae laid great emphasis on these facts, but I have not seen his work. Rae, Political Economy, Boston, 1834.

Cf. Marshall, Principles, 1st ed., pp. 374-380.

It

insisted on by Professor Patten in his well-known treatise.1 is perhaps unnecessary to state that the case is of some real significance in the commerce of the world, and not a mere curiosity of economic theory. It is a situation which frequently confronts new countries in their dealings with those of long established industries. What is proposed here is to determine, in a particular case, the gains and losses of two countries under conditions of non-intercourse, free-trade, or certain impediments and encouragements thereto.

In theory, according to the familiar principle of comparative cost, it might happen that the agricultural country could produce both sorts of commodities more cheaply than the manufacturing country, and yet the trade, if left to private interest, would lead to the production of only one of them, and the importation of the other. The question would still remain whether this was a desirable result for the country under consideration. It is freely admitted nowadays that it is not a very exceptional thing to find that the gain of individuals in trade is inconsistent with the advantage of society as a whole, and the question here is whether such a trade is always beneficial, whether we consider one country by itself, or jointly with the community with which it is trading. In the particular example selected the calculation will show that the country possessing superiority in both agriculture and manufactures, under the initiative of private traders seeking for individual profit," will tend to produce and export agricultural products, and to diminish its production of manufactures, not only to the prejudice of the whole country as respects the utility of its total consumption, etc., but also to the prejudice of both countries taken together; i. e., although the second country will gain by the trade, its gain will not equal the first country's loss.

The possibility of such a result in international trade is discussed by some writers, notably by Professor Edgeworth in his

1 Patten, Economic Bases of Protection, Philadelphia, 1890.

? "Comparative cost means the cost under conditions of isolation."-Bastable, Economic Journal, June, 1901, p. 228.

For a good statement of this motive see Mangoldt, Grundriss der Volkswirthschaftlehre, Stuttgart, n. d.