there are unlimited opportunities for investment, unlimited possibilities of increasing the product, merely by adding more instruments of the kind already in use. Then we have the "static" state, and the naked question whether mere increase in the increments of capital (Clark's phrase) or mere extension of the production period (Böhm-Bawerk's phrase) serves to add to the output. To that question the answer, it seems to me, should be in the negative: whereas the question as to what may happen in the dynamic state-when there are “qualitative" improvements or advances in the arts-is not susceptible of such clear-cut answer as both thinkers seem to suppose.

One sort of limitation of the possibilities of using capital must not be overlooked. There are obstacles to the spread and utilization of known improvements which make many of them practically unavailable for great parts of mankind. The use of modern agricultural machinery by the peasants of British India would greatly increase the productiveness of their labor. Were they able to use it well, they could afford to pay handsome interest to those providing it. But lack of intelligence and education, all the rooted conditions of a primitive social state, make this application of capital out of the question. The American traveller in many parts of Europe sees unbounded opportunities for using labor-saving appliances. But, so long as the people are not ready to turn to them, there is here no opening for investment. A change in the intelligence and skill of a great mass of mankind is as much a "dynamic" operation as is the invention of a new mechanical process.

One other aspect of this supposedly far-reaching law of diminishing returns deserves attention. It is sometimes spoken of as if it were but a phase or application

of the general theory of value. Successive increments of any one commodity have diminishing utility and declining exchange value. Diminishing returns on capital are supposed to result directly from the diminishing utility of commodities, the first-named principle being simply the result or manifestation of the second-named.1

These two tendencies, or laws, seem to me entirely different. In the one case we have to deal with the utilities and the values of the several units of quantity: in the other we have to deal with those units of quantity themselves, with physical units. The law of diminishing returns as to land, so often referred to as the type and proof of the wider theorem, neither says nor implies anything as to utilities or as to value. It says only that more and more labor (capital also, if you choose to think of capital as something different from labor) applied to a given area does not remain continuously productive at the same rate; and the productiveness of the several doses is measured in terms of bushels of wheat or tons of hay, not in terms of value. Measured in terms of value, it is by no means necessarily true that there is any tendency to diminishing returns as to land. The smaller quantity of wheat or hay which accrues from the last dose of labor on the land will very likely have not less value, but the same value as preceding products of the same labor. Similarly, the problem as to the increase of return due to added doses of capital is one of quantity. Will more wheat, more cloth, more shoes, be got by making and using more tools or more elaborate tools? The law of diminishing utility, on the other hand, bears on the utilities or satisfactions derived from added units of the same commodity, and so on the relative value of the several

1"'Diminishing returns of indirect agents is a special case of the diminishing utility of goods." Fetter's Principles of Economics, p. 71. Professor Clark does not seem to hold this opinion; for in his recently published Essentials of Economic Theory (p. 56) he refers to the one law as parallel to the other, not identical with it.

products. It thus affects the distribution of labor and capital towards the making of more or less of each product. The one principle has to do with the relative value of different commodities and with the income of satisfactions ("psychic" income) which mankind gets from its exertions. The other has to do with gains in physical quantities, and with the variations in such gains-whether at an increasing, a decreasing, or a stationary rate-from different ways of applying labor.

Returning now to the question as to the law of diminishing returns for successive increase of capital, I may sum up my conclusion by saying that the view which maintains this law seems to me essentially historical, and in that sense unreal. Successive increases of instruments of the same kind lead to no increase of return: they bring mere surplusage. The addition of instruments of a different and better kind-what Professor Clark calls qualitative increase-obeys no law. It is dependent on the progress of invention, and its course is irregular and unpredictable. If we have a steady increase of capital of the first kind,-quantitative only,—the return to capital will soon disappear. If we have a steady increase of the qualitative kind, there is no telling whether the addition to the total output, and so the return in the way of interest, will be at a steady rate, an increasing rate, or a diminishing rate.

The mind of man strives for generalization. It seeks to arrange phenomena in law or regular sequence. To this striving, I suspect, is due the attempt to formulate a universal law of diminishing returns. The attempt is like that to reach sweeping generalizations in history or politics, or to come closer to the sphere of economicsthat to find far-reaching or universal laws of economic development. In fact, the phenomena are not susceptible

of such clear-cut generalization; or at least we do not know enough about them to perceive clearly any underlying general forces. We must content ourselves with learning as much as we can of the irregular forces and the puzzling facts, and with stating our conclusions often in hypothetical terms. If an increase of capital (or a spreading of labor over more time) always brings a greater output, interest will continue, however great the increase of capital. If such an increase always brings a greater output, but at a diminishing rate, interest, while it will continue, will tend to be lower and lower in rate. If an increase of capital brings no greater output at all, and if none the less it takes place regardless of consequences, it will lead infallibly to the complete disappearance of interest. In some such forms as these we can state conclusions with sharpness of outline. But just in what way the increase of capital will in fact take place,-what will be the march of invention and discovery,-on this we are not able to forecast the future or the working of the productive forces in the future.

F. W. TAUSSIG.

ADMINISTRATIVE SUPERVISION OF RAILWAYS UNDER THE TWENTIETH SECTION OF THE ACT TO REGULATE COMMERCE.1

SUMMARY.

The twentieth section in the Act of 1887 was ineffective, 364-365.— Three provisions in the amended section: monthly reports, uniform accounts, special examiners, 365–367.-Operating and capital accounts, and their new supervision, 368.-The four principles underlying the Commission's accounting scheme, 370.-Discount on securities not to be carried as an asset, 370.-Depreciation and betterments accounts regulated, 371.-Surplus to be accurately stated, 372.-Industrial importance of the scheme: greater stability in railway securities, 372376.-Clarifies relation of rates to cost, 376-377.-Wider influence of confidence in railway accounts, 377-379.-Conclusion; administrative supervision, rather than procedure in the courts, initiated by this regulation of accounts, 379–383.

THE twentieth section of the Act to regulate Commerce was regarded as an important feature of the original law passed in 1887, but its significance has been greatly increased by the amendments contained in the Act of June 29, 1906. For students of transportation, and especially for students who regard the problem of transportation as primarily a problem in political and industrial organization, the chief interest in a consideration of this section of the act arises from the fact that in the twentieth section is found a definite, altho an incomplete, expression of the principle of supervisory administration.

For an intelligent understanding of this point of view it will be necessary to describe somewhat in detail the task undertaken by the Interstate Commerce Commission, acting upon the authority conferred by the twentieth section of the Act to regulate Commerce, as amended.

1 Prepared from notes of an address before the Seminary of Economics at Harvard University.