turn, is a determinant of money costs. A certain amount of a commodity can frequently be produced at a given unit cost, but to double the production would largely increase the unit cost. Neither of these factors, the amount produced or the cost, is constant with respect to the other. Since the varying factors of cost and production so intimately react, the one upon the other, production influencing cost and cost influencing production, one can not strictly speak of cost of production as one would speak, for example, of the heating units of a grade of coal, or the nutritive value of various foods. One can as well speak of the production of costs (i. e., quantity produced at a specified cost) as of the cost of production. Since there are two interworking factors, therefore, both cost and production, and not one factor constant with respect to the other, the proper object of study may be said to be cost and production rather than strictly cost of production. Production must be studied in its relation to natural resources (domestic and foreign), labor supply, industrial organization and efficiency, foreign trade, social and sanitary conditions, and these related in turn to money costs. Costs under various conditions and extensions of the industry and with respect especially to prices should be investigated and related in turn to the more technical and social factors of production. Data obtained by such inquiries should be employed in deciding question (1) whether the industry under consideration or some portion of its possible development should receive. tariff aid. The determination of this question is largely a matter of policy, but the policy will be enlightened to the degree to which it rests upon ascertained fact. Under the tariff act of 1922 Congress itself decided what articles should be free or dutiable, and, although certain power to change rates was given to the President, he was not authorized to decide what commodities should or should not be dutiable. 2. DIFFERENCES IN FOREIGN AND DOMESTIC COSTS OF PRODUCTION AS A MEASURE FOR A PROTECTIVE TARIFF RATE. This subject may be discussed under the following topics: (a) Reasons for using cost differences as a criterion for tariff rates. (b) Ascertainment of comparable international costs of goods. (c) Averages, marginal cost, and bulk line. (d) Joint costs. (e) Transportation charges. (f) Rent and interest. (a) Reasons for using cost differences as a criterion for tariff rates.-The application of the cost principle is useful in preventing a tariff rate that will permit monopolístic exploitation by a domestic industry. For if duties are much greater than necessary to equalize foreign and domestic costs plus reasonable profits of the home producers, they may enable domestic producers unchecked by foreign competition to combine and sell at a price above that required for the successful maintenance of the industry. To prevent such exploitation, it has been contended that the protective tariff should not be more than sufficient to equalize costs, allowing a reasonable profit. e.. Moreover, even in the absence of a domestic combination, if the local demand exceeds the home supply the price (see) may rise to the full extent Of course, what is a "reasonable" profit to one producer may not be to another. 110572-24- -13 of the duty. Thus if the tariff rate is more than enough to provide equalization, it may give the domestic producers an undue advantage. But if it has been previously decided to maintain a particular domestic industry or a certain portion of such industry by tariff stimulation, it is clear that duties equal to the difference of costs are the lowest that will achieve the object desired. The price of imported articles, it is said, will, in the long run, approximate their foreign cost of production plus transportation costs and customs duties. The customs duties, therefore, for an article whose production through the aid of tariff duties has been decided upon should be sufficient to cover differences in costs of production (possibly taking into consideration transportation), but not greater than this amount. The question may be asked, Why not equalize foreign and domestic prices by means of customs duties? The answer has in part already been indicated. The possibility of domestic monopoly and inflated charges makes unsafe the employment of domestic price as a sole measure of the status of the home industry. Domestic prices used in connection with and as a check of other data, however, are no doubt helpful in showing competitive ability. Foreign prices (i. e., prices of domestic imports) offer a safer measure. If foreign prices are inflated there will be correspondingly less need of protection. If importing prices are below foreign costs, the situation may come under provisions for dumping (see). In any case the import prices measure the competition to be met by the domestic producer at any given time. But the relation of such prices to the foreign cost of production measures more accurately the capacity of the foreign producer to meet and perhaps neutralize any proposed tariff. (b) Ascertainment of comparable international costs. Description of the competitive status of industries in different countries and analysis of ensuing tariff problems require rather definite comparability in the goods whose costs of production are ascertained. Manifestly cost data can not properly be compared when the goods to which they relate are dissimilar in kind or quality. Investigation of international costs demands, therefore, the combined experience of the cost accountant and the expert in the physical and commercial features of the commodity in question. The staff of the United States Tariff Commission includes both classes of experts, who unite to secure the necessary data. Other matters that require consideration in international cost investigations are differences in access to raw materials and in other geographic factors, in industrial organization and technique, and in governmental policies relating to commerce and industry. These differences must be ascertained and, so far as possible, expressed in terms of money. It is essential that all of the money outlay necessary to the production of the commodity be considered in order that comparative costs may be adequately stated. (c) Averages, marginal cost, and bulk line.This problem is one of deciding among the differing production costs of various establishments or lots of goods what costs are to be taken to fix the tariff rate. Investigation shows that the production of any particular kind of article of uniform quality is accomplished by different producers and by the same firms at different times or in different lots at widely varying costs. The following tables indicate this fact. U.S. Tariff Commission, Carpets and Rugs of Wool (T.1.S., K6), pp. 91-95. The elastic provisions of the tariff act of 1922 apply the principle that changes in tariff rates are to be so made as to equalize the difference between the domestic and the foreign cost of production. In the administration of this provision the manner of making cost comparisons obviously requires careful consideration. The statistical methods applicable to such comparisons may be divided into two groups: (1) Averages, involving the derivation of a figure from the cost series by a strictly mathematical process, such as the median, the mode, and the simple or weighted arithmetic averages; and (2) marginal cost and bulk line, entailing the consideration of costs at the points where major, economic decisions are being established. (1) Type of averages. The median is the middle item in a series of factors arranged in order of ascending or descending magnitude. The number of items above the median equals the number below it. This type of average is employed where statistics are available for no more than a few individuals in a group; it is assumed that the median of such individuals corresponds to that of the group itself J The mode is that type of object or quantity occurring most frequently in a class of objects or quantities. It is the most usual and characteristic form or quantity found in a series. In some cases no single and distinct modal type may be found, but where it occurs this type of average is frequently employed. The simple arithmetic average is the sum of a number of separate quantities divided by the total number thereof, the entire series being employed in the determination. Its use is based on the assumption that all the terms of the series are of relatively equal importance. The weighted average of a series of items takes into consideration the relative importance, measured by some established standard, of each item, each item being multiplied by its "weight" thus established. The products of the individual items, multiplied by their respective weights, are totaled. The total, divided by the sum of all the weights, gives the weighted average. In the determination of the weighted average the entire series is employed, but, unlike the simple average, each item here enters into the total on the basis of its relative importance. The use of this form weights in order to avoid an arbitrary figure as the requires the establishment of a rational basis of result. 1 ・(2) Marginal and bulk-line costs. The marginal cost of production takes place at the point where profits disappear; that is, at the point where the cost of producing a commodity is equal to the income derived therefrom. Production may continue to the marginal point, but not beyond it, unless a loss is to be suffered; that is to say, no self-sustaining industry can continuously operate below the margin. The margin, therefore, is the point where decision takes place as to the volume of production, the results of which vitally influence prices and economic conditions generally. The price at which a commodity is sold determines the outlay which a producer can afford to put into its production. For instance, if the market price of cloth is $1 a yard, the outlay which the producer can afford in its production can not be greater than $1 if the industry is to remain selfincome of any producer are equal is the margin, sustaining. The point at which the outgo and the and the undertaker at that point is the marginal producer. Likewise, that portion of any producer's output which is produced at a cost just equal to the income received for it is the marginal product. The use of the marginal principle as applied to the equalization of costs for tariff purposes would require the accurate determination of (1) producers whose costs equal their receipts, and (2) a tariff rate equalizing the costs of these marginal producers in the domestic and in the principal competing foreign country. Because the application of economic forces is focused at the margin, and because decision is there made whether production shall be continued, expanded, or, possibly, discontinued, the marginal cost is sometimes considered a desirable measure for tariff adjustments. It is argued that it is from the marginal point that the economic effects of tariff adjustments radiate and that a comparison of marginal costs will consequently involve consideration of the principal economic factors affected by the tariff. The bulk-line method was applied to price fixing during the war and has since received some attenition in various studies by Prof. Horace Secrist and others. Although it may ultimately be generally employed in a broader field than that of price fixing, this method suffers as yet from the lack of precise definition. Careful study is required to establish definitely its usefulness as a scientific instrument. It has been given some attention by the Tariff Commission in connection with the selection of the type of averages best adapted to making comparisons between the costs of domestic producers and of foreign competitors, with par ticular reference to the flexible provisions embodied in section 315 of the tariff act of 1922. Government experience with prices during and since the European war has renewed interest in the theory which asserts the existence of a more or less clearly revealed difference between industries that operate on a basis of relatively constant or diminishing cost and those that operate on the basis of increasing cost. Prices of the products of the former classes, it is said, tend to equal costs of production of the representative rather than the highest-cost firm; products of the latter class, however, tend to be priced near the cost to the marginal firm. In this use by the Tariff Commission, the aim has been to establish the bulk line at a point where all abnormally high or freak production has been eliminated, while still including all normal production. To that end it is essential to note at what point the rate of production for each added cost interval materially declines. The cost corresponding to this point will include all production that might be considered normal while eliminating all that is unusual or abnormal. Determination of the bulk-line result requires that a cost series be Dr. Philip G. Wright says: "Recent economists, prepared and arranged in sequential order, rising while adhering to the doctrine of marginal cost as from a low-cost production at one end to a high a price determinant in the case of commodities cost at the other. The application of the method subject to the law of diminishing returns, have is aided by the use of a graph having costs arranged been disposed to accept Marshall's theory of the on a "y" or vertical axis and a cumulative quantity representative firm in the case of commodities of production on the "x" or horizontal axis. In subject to the law of constant or decreasing cost. fact, a graph is ordinarily, and should always be, It is not toward the cost of production of the least used in establishing the bulk-line cost. The vol- efficient producer that price gravitates, they say, ume of production at the various individual costs but to that of the well-established, solid business is plotted as a curve, and the selection of the bulk-man-the man doing a conservative, prosperous, line percentage and the corresponding bulk-line but not phenomenally brilliant business. The cost is then made at the point where the plotted high-cost producers, those whose costs are greater curve breaks off from its gradual rise into a sharp than that of the representative firm, are not tendupward swing; for example, an 85 per cent bulk- ing to establish normal price; they are tending to line cost of 10 cents would result if the "breaking effect their own elimination. Every year a sheaf point" in the curve were above that percentage of failures is thrown off, but every year a new sheaf quantity of production on the horizontal axis and of failures appears. The individuals change, the opposite that cost figure on the vertical axis. class is constant." 4 The bulk-line cost is not an average cost. It is employed in lieu of an average cost in an endeavor to secure a significant figure which may serve as a basis of judgment in a particular study. The bulk-line method does not derive a definite figure from the cost series but merely establishes that figure which is the most suggestive in the entire group and upon which consideration must be concentrated. By contrast, an average is a figure obtained from the series by a mathematical process and is, therefore, preferred by some because of its mathematical precision. The argument in favor of the use of the bulk-line cost is that the extremely low-cost producer need be given but little consideration in a cost investigation, particularly for tariff purposes, and that extremely high-cost producers can not be adopted as a normal measure, and that it consequently becomes essential to employ the costs of those producers whose influence upon economic conditions warrants careful consideration. In other words, in the use of the bulkline method the endeavor is made to establish the point at which such attention is usually focused. The bulk line aims at a measurement of costs at the point covering the bulk of the industry after abnormally high-cost and freak producers have been eliminated. In a sense it can be said that prices determine the cost of production, or at least the marginal cost at which an article is produced. For as prices advance, higher-cost producers appear, while a recession in prices causes the withdrawal of the producer up to that time on the margin. It is said that the marginal producer follows prices rather than that prices follow the marginal producer. When, for example, the price of oil advances, the less pro'ductive and more expensively operated wells are put into During the European war the various pricefixing agencies of the United States Government had to deal with the problems under discussion in their attempt to adjust prices to costs of production. Says Prof. F. W. Taussig in summarizing the lessons from this experience: "If the differences in costs rest on physical causes-if they are due to forces in nature, not in man-the normal or long-period price may be expected to conform to * * * service; when the price of oil recedes those wells which produce oil at the highest unit cost are abandoned. Thus the marginal cost the unit cost of the most expensively operated producer-is fixed by existing prices rather than the prices by the marginal producer. Granting that an increase in prices would in most cases lead to an increase in the cost of production at which producers will be willing to, and probably will, operate, the question still remains whether the prices ever would have advanced if increasing production could have been accomplished without the increased unit cost. Suppose that the demand for an article grows. This in itself does not necessarily lead to a greater price. For example, it is commonly said that the Increasing demand for closed automobiles has caused a decrease in their price. Mass output lessens expenses. Only when supply does not keep pace with the growth of demand does price advance. But why does supply sometimes fail to advance commensurately with the advancing demand? Because conditions of production sometimes become increasingly more unfavorable as the quantity of output assumes larger proportions; in other words, because greater production takes place at an increasing cost. Assuming that price is advancing, it may be granted that the margin of production will widen to a more costly basis. But the point is this: We can not assume advancing prices unless we first postulate increasing costs with greater output. True it is that, temporarily, the higher cost production may follow rather than precede higher prices. But none the less true is it that the higher costs are the cause rather than the efect of the higher prices. For the causes and effects of economic processes, as has always been admitted, are not wholly objective, but take place in part in the subjective appraisals and purposes of men. Though the actual increased unit cost follows the higher price, the economic increase in cost has long before taken place in the subjective estimates of the producer. It is because of this greater economic cost-the estimate in the mind of the producer-that prices go higher. Thereupon follows the objective increase in cost by agents farther out on the margin in the scale of efficiency. The Food Administration, the Fuel Administration, and the Price-Fixing Committee of the War Industries Board. Wright, Philip G., in Quarterly Journal of Economics, May, 1919. pp. 560. marginal cost. But if they rest on the differing abilities of men, the normal price may be expected rather to conform to average cost, or at all events to fall below the bulk line. In debates before the Price-Fixing Committee and in its own deliberations, the question was raised whether regard should be paid primarily to average cost or to marginal cost. Assuming that the object to be attained is conformity to fair price in the sense of long-period or normal price ** * the answer would be, marginal cost if the differences in cost arise from causes in nature, but average cost (or something like it) if they arise from differences in managing ability. * * It would not be in accord either with current economic theory or with business experience to give to the least efficient manager (even if the extreme cases or freaks' be disregarded) a dominant price-determining position. The representative firm (wherever in the scale it is to be found) is in that position. "The case would be different if the differences are the ineradicable results of soil, site, or abundance and quality of minerals; then one might be justified in proceeding on the ground that the normal price-determining cost was marginal cost at the hands of a manager of representative ability. In the case represented on the charts [given in the text quoted for lumber, sugar, and coal] the proba bilities are that differences in nature's resources explain the upward trend of the cost curve; and hence that price in the long run would tend to conform to marginal cost, not average cost." But there remains a serious qualification. The doctrine must not be considered as applying to anything but a long-run price. * It would appear that in many manufacturing industries whose costs depend largely upon efficiency, the more efficient producers tend to lower the price below the costs of a less efficient minority, and that this minority can not in many cases be protected even by complete prohibition of imports. It would seem, therefore, that in many manufacturing industries, dependent largely upon efficiency, there is usually a minority of highcost producers who can not be "saved" permanently under any tariff rate. In endeavoring by means of import duties to equalize domestic and foreign costs of production the costs of such minorities might be disregarded. Indeed, if the tariff rate be set high enough to protect the highestcost producer who appears, it will normally be prohibitive. If, in the case of industries operating on the basis of constant or decreasing costs, the high-cost minority be disregarded, the costs of production of the representative producers might be considered a reasonable determinant of the tariff rate. As regards industries whose costs are mostly composed of raw material and subject to the law of increasing costs, a somewhat different position might be taken. The latter, besides being usually subject to wide variations in costs, are likely, as has been explained, to carry prices that cover the expenses of the marginal producers. In such industries the highest-cost producers determine prices. A question arises in extractive industries whose expansion at constant or diminishing cost is precluded by paucity of natural resources. That question is whether the maintenance of the most expensively producing concerns is justifiable. This whole question of the costs of the marginal and the representative dealer requires, however, further investigation before generalizations can be properly made. It is asserted that the efforts to equalize the costs of marginal domestic producers with foreign costs. has significance principally in rather sudden changes in relative conditions of foreign and do mestic production. For in any case the marginal domestic producer tends in the long run to be the one who can withstand existing foreign competition. "Suppose that Congress were to increase the duty on Cuban sugar, say, to 3 cents per pound. The high-cost Cuban producers would be unable to sell at a profit in the American market, and would be forced to find some other market or to go out of business. The decreased supply would tend to raise the price in the United States. The increased price would make it possible for less effective domestic producers to enter the field. Hence, there would be a tendency for the domestic marginal cost of production to be raised, and for the Cuban marginal cost of production to be lowered. Equilibrium would be restored when, allowing for transportation costs, the domestic marginal cost exceeded the Cuban marginal cost by just the amount of the duty; i. e., 3 cents per pound. Since, therefore, the duty determines the difference between the respective marginal costs, it is futile to use marginal costs as a means of determining what the duty should be. Any rate of duty, whatever it is, short of a prohibitory rate, will, after a lapse of years, be found to be just sufficient to protect the domestic marginal producer, and hence in a certain sense may be said to equalize costs of production. What is here said of marginal costs will in the main apply to average costs. "But the process of adjusting the marginal costs to the rate of duty takes several years. Hence, if the rate which has been in force for many years is raised, and a cost investigation is completed immediately after the increase in rate, the new rate is likely to be found too great, on the marginal cost test, by just the amount of increase, and if the rate be lowered, it will be found, on the same test, to be too low." The relation of cost of production to prices is similar in both foreign and domestic industry. In many foreign manufacturing industries prices will therefore approach the costs of the representative firm or "bulk line," while in certain of the extractive industries costs of the highest-cost producer may determine the scale of prices. In any event, the domestic producer is more concerned with the prices offered by the foreign producer than with the latter's costs. From the point of view of international competition the foreigner's costs are important mainly as they give an indication of what he can do with prices. The knowledge of foreign costs makes possible a tariff rate really effective, while a tariff rate based upon the foreign price may be ineffective if that price as compared to cost has been inflated, or it may be excessive if based upon an importing price abnormally low as compared to foreign cost. The object of investigation of foreign costs is mainly to ascertain those costs that form the basis of normal foreign exporting prices. (d) Joint costs.--The production of many articles is a process that can not be separated into its component elements. It is often impossible to determine objectively just what part of the total Wright, Philip G. (Institute of Economics, Washington, D. C.) Cost as a Basis of Tariff Adjustment, in Journal of the American Bankers' Association, May, 1923, p. 720. |