Professor Clark has in some place defined subjective value as the “measure of effective utility.” Menger defined this kind of value as the significance attained by a good in our estimation when we know that some satisfaction of ours is conditioned upon command of this particular good. Value, as an amount, is the measure of the quantity of satisfaction conditioned. The present writer has already expressed his opinion[240] that Menger’s mere definition of value gives the solution to the great riddle of the relation of value to usefulness and satisfaction, and that virtually from the mere proposition contained in this definition a large part of the theory of value can be deduced directly. It is of great interest to note, therefore, that Professor Clark’s definition of subjective value is in entire harmony with Menger’s. The definition of Menger explains value universally, wherever there is value. Clark’s definition, though conceived quite independently by him, is but an extension of the principle in Menger’s definition, but an application of it to a certain special case. This case, though logically a special instance, is however typical of most of the goods we imagine a Crusoe to be producing, consuming and reproducing. This is the case of freely reproducible goods. Here, if a good be destroyed, its value will be revealed by the satisfaction that must be given up because of its destruction, which is the satisfaction finally conditioned upon it. In the end, what Professor Clark points out is merely that this good may be replaced by diverting to its making labor which otherwise would have been employed in producing some other good which Crusoe chooses to resign instead. The satisfaction in effect, or in the end, conditioned by good A is the satisfaction directly conditioned by or afforded by good B, the good given up.[241]

As it appears to the present writer, Clark’s theory of esteem value is to this point so well founded that even the most uncompromising opponent of labor theories can find no ground to deny it. We have here explained a labor measure of esteem value, perfectly justified at least under the conditions of the isolated economy, and Professor Clark’s analysis has disclosed the inner reasons why this measure can be employed. Indeed, while speaking of reproducible goods in the Crusoe economy, it may be affirmed that cost is not only a “measure” of esteem value, but is also a joint regulator of value. To be precise, the costliness of a good acts jointly with the utility of the good in regulating its value. To say that one thing regulates another is, of course, asserting more than that it measures that other. A regulator is a measuring cause, whereas a mere measure is not a cause of the thing estimated.[242]

In what precise sense is costliness here a co-determinant of value? When a good both costs pain or discomfort in its production and affords pleasure in its use, it is common custom to speak of the “cost” of the good as the exact opposite of its “utility.” But these concepts are not direct opposites. Cost consists in the subjective experiences of the producing man, and its precise opposite is pleasure or, specifically in our economic usage, satisfaction of want. But “utility” never is a precise equivalent for satisfaction. On the contrary, utility is another kind of opposite of satisfaction, being always conceived to belong to the good and not to lie within the man, except in the treatises of certain unconscious metaphysicians. In virtue of a certain combination of physical properties a good possesses a power to produce a satisfaction in a man.[243] This power, due to its physical properties, is the best conception of the good’s utility. The precise opposite of utility cannot be designated by cost, but the word “costliness” fits the need. In virtue of their physical properties, or physical and chemical relations with other external things, some goods require a large amount of change of man’s external surroundings to be effected by him in order that they may be produced. We give these goods the attribute of costliness, similar to the attribute of utility, and the relation between costliness and cost is similar to the relation between utility and satisfaction.[244]

The question which now concerns us is whether in the case of freely producible goods the supply of which can be augmented at will by the application of more labor, the costliness of a good takes an equal part with its utility in determining its value. If the marginal utility of a good determines its value, marginal utility is still the mere creature of the supply of the good. The larger the supply, the lower the marginal utility. Putting the matter the way it is often stated in present-day treatises, the supply that will be produced depends upon the cost of the good, and thus at bottom cost determines value. Making a more careful statement we may say that the supply depends upon both the costliness and the utility, since Crusoe will increase the supply—the average yearly supply, we should say if speaking of a crop—until the marginal cost and marginal satisfaction become equivalent. For the normal case, the increase of the number of increments of the good will entail an increase of marginal cost and a decrease of marginal utility. Both the marginal cost and the marginal utility vary when the supply is changed, but the supply tends to rest at the point of equilibrium of these two quantities, and is thus determined by them jointly.[245] If the general costliness of this grain should fall, its value would soon be lowered, for Crusoe would be led to produce more of it in order to reach the point of supply which equilibrates final cost and satisfaction. Thus the marginal utility and value would be lowered. Conversely if the costliness be increased, value will be raised. Under Crusoe conditions we are considering the relations of costliness, utility and value in the simplest of circumstances we can imagine. Even here the question arises, is it correct to affirm that costliness is precisely coördinate with utility in determining value? To me it seems apparent that utility exercises a more direct and intimate control over value than cost, even in the simplified case now before us.[246] In the first place, value is still derived solely from utility.[247] Cost influences value only by way of influencing the value-determining utility itself. Thus cost is more remote from value than is satisfaction or utility. Even in the case of goods valued according to their “effective utility,” that is, according to a foreign utility, there is no violation of the principle that value is derived solely from utility. Cost or costliness is never the source of value. No amount of cost endured for a good without utility will confer value upon it. There can be no discrepancy between utility (here we are speaking of the utility that is the counterpart of the satisfaction conditioned upon the good, i. e., the marginal utility in the cases of goods divisible into increments) and value, whereas there may be between costliness and value. If anything prevents the supply of a good from being increased to the point of marginal equivalence of utility and costliness, then of course the value follows the utility and not the costliness. Costliness is thus not a more fundamental cause of value, but merely a more remote cause than utility, and in any event can influence value only by affecting the utility itself, by helping to determine supply.

7. Turning to social conditions, we meet the new problem of exchange value. It is true, certain value comparisons may be made in the entirely isolated economy which afford a kind of prototype of the exchange value of the market. Crusoe might, perhaps, have occasion to make mental note that ten bushels of his wheat supply possess the same esteem value to him as one cord of his firewood. This comparison involves a ratio between valuable goods, and if Crusoe only had reason to trade with himself we might find the phenomenon of purchasing power—the true conception of exchange value—and ten bushels of wheat would have the purchasing power of one cord of wood. This kind of exchange value—if we dare call it such—would be most simple; for it would be directly determined by, and be in exact proportion to, esteem value. Just as Crusoe will be able to carry the esteem values of reproducible goods in mind most readily in terms of disutility cost, so would he be able to strike value ratios between such goods most easily by comparing the disutility costs of their physical units of measure. Thus, if one cord of wood should exchange for ten bushels of wheat, the reason would be that a bushel of wheat costs one-tenth as much labor as one cord of wood. This thought need not be pursued further. If the exchange value of a good in the social market depended in the same direct manner upon the esteem value of that good to all society or to “society as a unit,” the final theory of exchange value would be much simplified. But, unless the present writer is mistaken, a good cannot possess an esteem value to society as a whole, but can merely have a separate value to each individual member of society.

Though it is questionable whether we can apply the theory of final equivalence of utility and disutility directly and in an unmodified form to all kinds of social value, we still find generally that each individual in society values his personal consumption goods substantially in the same manner as Crusoe. Under the division of labor, the individual does not produce his own consumption goods, but renders certain productive services to society for which he receives remuneration in the shape of a money income. Money income may also be derived from capital acquired at the cost of abstinence on the part of the individual receiving it. But again, it may as well be the result of gift or inheritance, in which case it represents no subjective cost to the recipient. When a dollar costs a man subjective sacrifice to acquire it, and the dollar is spent for a commodity, this commodity thus indirectly costs that sacrifice. The commodity is bought by the consumer because it possesses esteem value. It derives this esteem value from the satisfaction conditioned by it, but this value may well be carried in mind by the consumer in terms of his own subjective cost. Crusoe spends units of disutility, as it were, to obtain from nature certain utilities; the man under social conditions spends dollars, which may represent disutility, to obtain utilities from the social warehouse. Professor Clark’s theory of “effective disutility” as the measure of value may be applied to the individual economy within society as well as to the individual economy in isolation.

Everywhere in society among individuals who earn a part or all of their money incomes,[248] we may expect to find pure “esteem values” being measured by subjective costs. But the relation of subjective cost to exchange value is a different matter. Two parlor tables of the same make and pattern will possess the same exchange values, but the esteem values of the two to their separate owners are not comparable quantities. Each of these owners, for himself, may estimate the esteem value of the table in terms of its indirect cost of acquisition in his own disutility. The dollars cost him disutility and the table cost him dollars. But this is not at all the same as saying that the “exchange value” of a table is measured by the disutility cost of production of that commodity. Exchange value can have no such intimate relation with disutility cost as esteem value. Furthermore, the cost of production of tables is experienced only by makers of tables, and not by their consumers. The exchange values and pain costs of commodities can have no closer relation than one of mere proportionality. It is possible that, if one A has an exchange value of two B, an A has cost twice as much disutility to produce as a B. In this case the exchange values of these commodities, each in terms of the other, are proportional to their subjective costs. It is permissible to predicate equality of subjective cost and esteem value, but to say that the subjective cost of an article equals its exchange value would, of course, convey no meaning. The unqualified classical labor theory asserted that exchange values were determined in proportion to relative labor costs. The aim of this chapter has been, therefore, to prepare to answer this question: Are the exchange values of commodities in the social market in proportion to the subjective costs of production of these commodities? Is there any way of defining, or method of reckoning, the pain cost of a good, which will enable us to show this proportionality?

Before attending directly to these questions, it is best to consider whether it is possible to compare the subjective costs of commodities produced by different persons or groups of persons. A ton of coal may exchange for six bushels of wheat. The subjective cost of the coal consists chiefly in the labor of certain miners; that of the wheat in the labor of certain farmers. To assert that the subjective costs of production of these commodities either are or are not in proportion to their respective exchange values, implies that we are able to compare these costs as quantities. To assert proportionality requires that we be able to say that the disutility experienced by the miners in producing a ton of coal is equal to that experienced by the farmers in producing six bushels of wheat. To assert disproportionality, we must be able to state that these disutilities are unequal. If these disutilities are quite incommensurable, we can assert nothing regarding the relation of these costs to the corresponding exchange values.[249] Men as scholars are accustomed to maintain that the pleasures or pains of different minds cannot be compared as quantities, while in every-day life the same men are equally accustomed to state that John enjoys music more than Paul, or that Primus suffers more or works harder than Secundus. May or may not we affirm that the stoker works harder, or in our own jargon, suffers more disutility, than the dining-room steward? In the hope of settling part of the issues raised in these questions, let us consider the meaning of one of Adam Smith’s statements regarding wages in different employments. I refer to the doctrine that wages tend to be higher than the average in employments where there is a higher degree of disutility. This tendency is operative only under perfect competition, and the existence of numerous non-competing groups occasions a result much changed from that to be expected from this tendency, which is sometimes described in the “evil paradox” that the harder the work, the lower the wages. The question which concerns us here is, how much does either of the above statements imply with respect to the possibility of comparing the pains or pleasures of different persons. It seems to the writer that neither necessitates a direct quantitative comparison of the subjective experiences of different persons. I may, perhaps, say that the persons in occupation A are suffering more disutility and receiving higher wages than those in employment B, but all I can be supposed really to know is that if I were in occupation A, I would suffer more discomfort than if I were in occupation B. If I am a person of “average” (i. e., typical) constitution, I may infer legitimately that this is true also of any average person. While making no affirmation that John suffers more disutility than Paul, either when these persons are in the same or in different employments;[250] we may be able to state that either John or Paul will suffer more in occupation A than in B. The upshot of the matter is that Adam Smith’s proposition implies only our ability to compare the disutility (using this word in the sense it ought to have) of different tasks. A task is objective, consisting in certain objective results to be effected under certain objective conditions. When the objective characteristics of a task necessitate subjective discomfort in the person who performs it, the task or employment possesses disutility, which is thus a concept the opposite of utility. If within a competing group employment A affords a higher wage than employment B, because its disutility is higher, the result is brought about not, in the first instance, through the perception by the workers that certain individual persons work harder than other individual persons, but through the perception that any normal individual for himself would work harder at the task A than at the task B. The possibility of comparison is implied merely between the “pains” of the same person, though there are common forms of expression which imply more. We may conclude, then, that there is a perfectly legitimate sense in which we can compare the subjective costliness of commodities produced in society by entirely different groups of persons. And no one doubts that the day’s product of a coal miner has a higher disutility cost than the day’s product of a farmer.

When we find the statement in an economic treatise that the exchange values of commodities are ultimately regulated by their subjective cost, it is to be assumed that the meaning is the same as that which would be expressed with greater precision by using the word costliness. With the explanations already offered we may henceforth follow common usage and employ the mere word cost. There are two distinct ways of reckoning the pain cost of a commodity, namely, (1) as total cost; (2) as marginal cost. Ricardo reckoned cost according to a hybrid method. The total subjective cost of a good consists in all the discomforts of labor and abstinence actually endured in the past to produce it. Taking the factor of labor alone for illustration, it includes the cost of the labor directly applied to the good, and of the labor indirectly applied by being directly applied to the raw material and machinery which are used up in its making. The machinery, however, has always been made at the combined expense of labor and of using up formerly existing tools and machinery; and the latter tools and machines had a labor cost. A product’s total cost may include, perhaps, one one-hundredth of the labor cost of the first generation of certain machinery used in its production, and as we go back, one one-millionth of the second generation;[251] and the total labor cost of any commodity thus goes back no man knows how far. Therefore, the total labor cost alone of a good (to say nothing of the abstinence element) is an extremely indefinite quantity; and it is impossible to know anything very definite about the comparative total labor costs of different articles. But beyond this, we do know that the existence of differential rents destroys the possibility of proportionality between total labor costs and exchange values.

The concept of the marginal cost of a good appears in the Ricardian theory of rent, and has been involved more or less clearly in the reasonings of most modern economists, but it is almost entirely to J. B. Clark that we owe the consistent development of this idea. The marginal subjective cost of a good may consist either of labor or of abstinence, but not of both combined. In this essay we will arbitrarily set aside the problem of abstinence cost. The marginal labor cost of a good is, of course, determined by ascertaining the marginal product of labor in producing this good. To illustrate in the simplest manner possible, we will follow the time-honored procedure of eliminating capital for the moment, and suppose successive doses of labor to be applied to a given piece of land.[252] Let the labor force applied stand at a certain amount, and suppose the dose to consist of a labor day. Then, if experimentation reveals the fact that the addition of one more dose will increase the whole product by the amount of two bushels, we define these two bushels to be the marginal product of a labor day. Professor Clark frequently refers to this same quantum as the specific product of labor. The land in this case may have consisted of a 100-acre field and the total labor applied may have amounted to, say, 300 labor days. The total produce may have been 3,000 bushels of grain. By hypothesis, capital being eliminated, the total labor cost of these 3,000 bushels is 300 labor days, or ten bushels cost one labor day, or the total cost of a bushel is one-tenth of a day. On the other hand, the marginal cost of a bushel is one-half of a labor day, since two bushels are the marginal product of a day.[253] Ricardo, who so explicitly defined total labor cost as consisting of the labor both directly and indirectly applied to a commodity, also assumed that in one respect value-determining cost is marginal, though he never used the word “marginal.” It was for this reason that not far back we described his method of defining cost as hybrid. His doctrine that rent does not enter into cost was but one way of stating that on land, it is only marginal cost (as he expressed it, the cost of the most costly portion of the supply) which determines value. In real life, products are the result of combining not land and labor alone, but land, labor and capital (in the sense excluding land—our usage at present). When Ricardo was expounding and illustrating the theory of rent which bears his name, he was forced to suppose that the successive doses added to land were composed of capital and labor jointly,[254] which left his marginal quantum the product of both of these agents. This left him with the great interest difficulty with which he occupied himself in his first chapter.[255] It remained for J. B. Clark to point out that the marginal product of labor could be disentangled from the product of capital as well as from that of land.[256] Upon this possibility depends the important productivity theory of wages. In order to explain the process by which the pure marginal product of labor is found by the entrepreneur, Clark adopts what is virtually the business man’s conception of capital, as distinguished from concrete capital goods. The latter alone have been designated capital by most economists in their formal and explicit definitions. Professor Clark prefers to call the two concepts simply capital[257] and capital goods. Capital is a “sum of productive wealth, invested in material things which are perpetually shifting—which come and go continually—although the fund abides.”[258] These material things are the capital goods. Capital as an amount must be measured by its exchange value. A capital of $100,000 may be prepared to employ say 40 men. Should it be rearranged to employ 20 men, its concrete make-up would have to be altered. A less number of machines and tools of better quality would have to compose it. Now as the concrete tissue of a given capital perishes or matures and frees its value for reinvestment in more concrete goods, an entrepreneur has it open to him to alter the concrete constitution of his capital. In this way, in the course of time, an entrepreneur may be able to rearrange his capital so as to augment or decrease the labor force employed with it. In many cases pretty large changes in the labor supply employed with a given capital could be made with little or no alteration of its technical concrete make-up. Somewhat slowly and under this and that frictional difficulty, the experimentation is made which reveals the marginal product of labor. The process which discloses this must always in the end be one in which an increment of labor is added to or removed from the force working with a given capital and an observation made of the resulting addition to or subtraction from the total product. The exposition of this process and the explanation why competition tends to make the wages of labor (of whatever grade) equal to its specific or marginal product, is probably the greatest contribution to economics contained in Clark’s Distribution to Wealth, and occupies a large part of that work.

As was virtually pointed out by Malthus,[259] the presence of rent and interest charges in entrepreneur’s costs is an insuperable obstacle in the way of the theory that a commodity’s total labor cost is proportionate to its exchange value. If, however, an attempt is made to correlate marginal labor cost and exchange value, the difficulties of rent and interest are eliminated. When we say that these difficulties are eliminated, we do not mean that they are arbitrarily set aside, or that we merely run away from them: but the marginal labor cost of a commodity is not affected by the payment of rent and interest. For instance, if wheat is being produced at the same time on land of the best and land of the poorest grade, a large rent will be paid out of the total wheat product on the former soil, and little or no rent may be paid out of the total product on the latter, and yet the cultivation will be carried to the point which makes the marginal product of labor and the marginal labor cost of wheat the same on both grades. The same observations may be applied to rent of capital (or interest, as we call it when it is calculated as a percentage of the value of the rent-bearing agent).[260]

The great difficulty in the way of the theorem that the marginal labor costs of commodities are in proportion[261] to their exchange values, is the problem of skilled labor. The best way to show the effect of skilled labor upon comparative marginal costs is first to eliminate it temporarily from the problem, and show what the relation of marginal labor cost would be to exchange value, if there were only common labor throughout society. If all labor were of a single grade, all commodities which are products of labor would have exchange values in proportion to their respective marginal disutility costs. This would be true whether the products are consumption goods or are merely production goods which are used in making further products. Some valuable goods are not products of labor. Such are bodies of ore lying in their natural state, standing timber, etc. These production goods have no disutility cost, marginal or total, and consequently their exchange values have no relation to cost. Their supplies are determined independently of human agency. Ore at the surface, crushed or smelted ore, are, however, products of labor, and so long as only a part of the known existing ore of mines is removed—a part remaining untouched because of too high cost—the supply of any product resulting from the combination of labor and native ore-bodies, will depend upon marginal labor cost.[262]

The homogeneous labor force (which we have assumed temporarily) will distribute itself among all the various industries in society in proportions determined by the marginal product in each industry. Capital will also distribute itself throughout the system of industries, tending, of course, in the long run, to appear in each industry in such proportions as will, apart from inequalities of risk, produce everywhere an equality of its returns. Assuming the distribution of capital to have reached a condition of equilibrium—it being no part of our present task to follow out a theory of interest—let us try to show that labor will distribute itself over the field of industry in such a manner that exchange values will be proportionate to marginal labor costs. If labor flows from one industry to another, the total output of the first industry will decline and that of the second will increase. The change in the supplies of the respective products of these industries will alter the exchange values of these articles. Different distributions of labor among industries will give rise to different relative supplies of commodities and different exchange values. As the supply of labor in any industry increases, its marginal product decreases. If all occupations possessed the same disutility, the supply of labor would be so distributed that its marginal product would have the same exchange value in all industries. But if some occupations necessitate higher disutility costs than ordinary, the supply of labor obtainable for those industries will decrease until the exchange value of the marginal product is raised till it compensates for the superior disutility.[263] If one commodity is produced at a higher disutility cost (to the labor directly employed upon it) than another, the marginal product of labor in it will have a higher exchange value. If 6 A in one industry and 1 B in another make the marginal product of a labor-day, 6 A will exchange for 1 B, provided the disutility of labor is the same in both employments. But if it costs more disutility to produce 6 A than 1 B, the relative supplies of A’s and B’s would be so adjusted that 6 A will exchange for more than 1 B. Thus a superior disutility cost raises the exchange value of a commodity, in order that this commodity may afford a superior value product to labor. Labor-power is a peculiar production good. Like other production goods of manifold productive uses, its expenditure constitutes potentiality cost; but it is unlike others in that human pain cost is an ever-present incident to its expenditure. The distribution of labor power among different productive uses is not governed solely with reference to its share of value derived from the product, but is governed in part with reference to the pain-cost involved in the production of the product. A higher disutility necessitates a higher share of exchange value. Thus it comes to pass that this most disposable and important of production goods will distribute itself among products in such a manner that these products will have exchange values in proportion to their marginal pain costs. This result is brought about solely by control of the relative supplies of these products, the exchange values of which are all derived from utility solely after the method described in the utility theory.

When we introduce the question of skill into the problem, we find that the supplies of many kinds of labor are limited not with reference to the disutility of the tasks performed, but are limited solely because the requisite brain-power, ingenuity or strength are scarce. The marginal product of such labor is raised by the limitation of its supply. Thus, it is a truism that many occupations of the lowest disutility afford very high wages, and that in the vast majority of cases high wages are not caused by high disutility, but by scarcity of competent persons. Suppose the commodity A is scarce, is of high exchange value, and is the marginal product of a skilled labor day. Article B is the product of a day of the lowest kind of labor. One A may well exchange for three or four B. Yet the marginal labor cost of A is, in the typical case, even less than that of B, for the skilled laborer ordinarily suffers less pain cost per day than the unskilled. Hence, the exchange values of these products are quite out of proportion to their comparative marginal disutility costs. The existence of non-competing groups, first emphasized and named such by Cairnes, is then a fatal obstacle in the way of the adjustment of exchange values to comparative marginal costs.

8. We may now essay a partial summary of the results which have been reached up to this point. The end of the theory of value is primarily to explain exchange value. The only workable definition of this term is purchasing power. The purchasing power of a commodity is measured objectively in terms of the physical units of some other particular good, except when we are speaking of the concept of an article’s general purchasing power. This, its purchasing power over all other commodities,[264] is measured as some kind of mean or average of all its particular purchasing powers. What mean, it is no part of our task to enquire. All goods which possess exchange value also possess that other kind of worth which we termed “esteem value.” Every commodity derives its exchange value solely from its esteem value, or, speaking with precision, from its esteem values. For a commodity has a separate esteem value to each individual person who can utilize it. If society were as one man,[265] the exchange values of goods would be but the exponents of their relative esteem values. In other words, if a physical unit of one commodity exchanged for two units of another, the reason would be merely because the first possessed twice as much esteem value to all society as a unit of the second. But the esteem value of an article is a much more definite thing than a social estimate, i. e., an “average” (or typical) estimate of worth. The esteem value of a good to a person is the measure of the amount of that person’s satisfaction conditioned upon the enjoyment of the good. Goods existing in superfluous abundance give satisfaction but do not condition it, and hence lack esteem value. Taking for granted the amount of an individual’s income, the esteem value which a good has for him determines his price equivalent for that good.[266] The market price, or exchange value, of a good is a resultant from (never in any sense an average of) the individual price equivalents placed upon it by the body of individual consumers.

The exchange value of a good varies inversely with the supply of it presented to the body of consumers. The larger the supply, the lower is the price equivalent which must be reached as the marginal determining point of its market price.[267] A change of supply alters exchange value only because it changes the marginal price equivalent.[268] In the social market, the purchasing powers of all the various products over one another depends upon their relative supplies. So far as cost of production in any form exercises any degree of control over the value of a good, it can act solely by way of influencing the supply of the good. The phenomenon of the apparent regulation of the exchange values of products by their entrepreneur’s costs, is but a part of a large process in which cognate (or “fellow”) products adjust their relative supplies and their exchange values to one another, to the end that the common production goods entering into all of them may produce equal productive contributions or shares of exchange value per unit in all of their productive applications.[269] The relation of the pain costs of products to their exchange values is limited to one of mere proportionality.[270] The pain cost of a product may be calculated in two very distinct ways, giving total pain cost or marginal pain cost. The total pain cost of a good, consisting in all the labor and abstinence ever endured to bring it into existence, is quite an indeterminate quantity,[271] and its influence upon the exchange value of a good is very remote and irregular. The larger part of total labor cost, the part which includes the labor directly applied to commodities, plus the labor indirectly applied by being directly applied to the raw material and machinery immediately used in their production, and so on for the few nearest generations of machines, this being the part which excludes the infinitesimal bits of labor cost expended far in the past, can be shown positively not to be in proportion to their exchange values. For commodities produced at a higher expense of rents of all kinds (as opposed to wages) have exchange values out of proportion to this calculable part of their total costs.[272] We find that the control of marginal cost over value is closer than that of total cost. If it were not for the existence of innumerable grades and classes of skilled labor, the supplies of produced goods would be so adjusted that their exchange values would be in proportion to their respective marginal costs. But on account of skill, we must here again characterize the influence of subjective cost as remote and irregular.

In conclusion, it is true, speaking in very loose and general terms, we may say the exchange value of a good depends both upon its utility and its costliness to mankind. But it would not be proper to say that cost and utility are equal and coördinate regulators of value. Therefore, Professor Marshall’s shears simile is not to be commended. The most noteworthy changes in exchange values have been produced by discoveries which reduced the labor cost of goods. But the amount of the reduction thus produced in the exchange value of a particular commodity could have only the roughest correspondence with the amount by which its relative pain cost was reduced. Also, for reasons already shown, we know that neither before or after these changes was it possible for exchange values to be in proportion to relative pain costs, whether total or marginal costs be taken. Furthermore, all alterations of exchange values produced by cost changes are effected solely by alteration of the value-determining utility itself. Utility has a much more direct and intimate relation with value in either form than cost. Value may exist without cost and cost may be expended without occasioning value. Value never exists without utility and utility (not in the sense of Smith’s “use-value,” but the effectual utility, the utility which measures the satisfaction conditioned by a good) never exists without value. Cost affects value solely by influencing utility itself. From this comes the all-important conclusion that whenever any of the numerous and permanent forces are active which interfere with the influence of cost, value follows the utility and not the cost.