After finishing his examination of the production relations among commodity producers (theory of value) and between capitalists and workers (theory of capital), Marx moves on to the analysis of production relations among industrial capitalists in the different branches of production (the theory of production price) in the third volume of Capital. The competition of capitals among different spheres of production leads to the formation of a general, average profit rate and to the sale of commodities at production prices which are equal to costs of production plus average profit and, quantitatively, they do not coincide with the labor-value of commodities. The magnitude of the costs of production and average profit as well as their changes are explained by changes in the productivity of labor and in the labor-value of commodities; this means that the laws of changes in production prices can be understood only if we start with the law of labor value. On the other hand, the average profit rate and the production price, which are regulators of the distribution of capital among various branches of production, indirectly (through the distribution of capitals) regulate the distribution of social labor among the different spheres of production. The capitalist economy is a system of distributed capitals which are in a dynamic equilibrium, but this economy does not cease to be a system of distributed labor which is in a dynamic equilibrium, as is true of any economy based on a division of labor. It is only necessary to see under the visible process of distribution of capital the invisible process of the distribution of social labor. Marx succeeded in showing clearly the relation between these two processes by explaining the concept which serves as the connecting link between them, namely the concept of the organic composition of capital. If we know the distribution of a given capital to constant and variable capital, and the rate of surplus value, we can easily determine the quantity of labor which this capital brings into action, and we can move from the distribution of capital to the distribution of labor.
Thus, if in the third volume of Capital Marx gives the theory of production price as the regulator of the distribution of capital, then this theory is linked to the theory of value in two ways: on one hand, production price is derived from labor-value; on the other hand, the distribution of capital leads to the distribution of social labor. Instead of the schema of a simple commodity economy: productivity of labor-abstract labor-value-distribution of social labor, for a capitalist economy we get a more complex schema: productivity of labor-abstract labor-value-production price-distribution of capital-distribution of social labor. Marx's theory of production price does not contradict the theory of labor-value. It is based on the labor theory of value and includes this theory as one of its components. This is clear if we remember that the labor theory of value analyzes only one type of production relation among people (among commodity producers). However, the theory of production price assumes the existence of all three basic types of production relations among people in the capitalist society (relations among commodity producers, relations among capitalists and workers, relations among individual groups of industrial capitalists). If we limit the capitalist economy to these three types of production relations, then this economy becomes similar to a three-dimensional space in which it is possible to determine a position only in terms of three dimensions or three planes. Since a three-dimensional space cannot be reduced to one plane, so the theory of the capitalist economy cannot be reduced to one theory, the labor theory of value. Just as in three-dimensional space it is necessary to determined the distance of each point from each of three planes, so the theory of the capitalist economy presupposes the theory of production relations among commodity producers, i.e., the labor theory of value. Critics of Marx's theory who see a contradiction between the labor theory of value and the theory of production price do not grasp Marx's method. This method consists of a consistent analysis of various types of production relations among people or, so to speak, of various social dimensions.
As we have seen, Marx analyzed the changes in the value of commodities closely related to the working activity of commodity producers. The exchange of two products of labor at their labor-value means that equilibrium exists between two given branches of production. Changes in the labor-value of a product destroy this labor equilibrium and cause a transfer of labor from one branch of production to another, bringing about a redistribution of productive forces in the social economy. Changes in the productive power of labor cause increases or decreases in the amount of labor needed for the production of given goods, bringing about corresponding increases or decreases in the values of commodities. Changes of value in turn bring about a new distribution of labor between the given productive branch and other branches. The productivity of labor influences the distribution of social labor through the labor-value.
This more or less direct causal relation between the labor-value of products and the distribution of social labor assumes that changes in the labor-value of products directly affect producers, namely the organizers of production, bringing about their transfer from one sphere to another and, consequently, the redistribution of labor. In other words, it is assumed that the organizer of production is a direct producer, a worker, and at the same time the owner of means of production, for example, a craftsman or a peasant. This petty producer tries to direct his labor to those spheres of production where the given quantity of labor yields him a product which is highly valued on the market. The result of the distribution of social labor among different spheres of production is that a determined quantity of labor of equal intensity, qualification, and so on, yields an approximately equal market-value to producers in all the spheres of production. Engaging their living labor in shoe production or in tailoring, the craftsmen at the same time engage past, accumulated labor, i.e., instruments and materials of labor (or means of production in a wide sense of these terms) which are necessary for production in their activity. These means of production are not usually very complicated; their value is relatively insignificant and thus, naturally, they do not lead to significant differences between individual spheres of crafts production. The distribution of labor (living labor) among individual branches of production is accompanied by the distribution of means of production (past labor) among these branches. The distribution of labor, which is regulated by the law of labor-value, has a primary, basic character; the distribution of instruments of labor has a secondary, derived character.
The distribution of labor is completely different in a capitalist economy. Since the organizers of production are in this case industrial capitalists, the expansion or contraction of production, i.e., the distribution of productive forces, depends on them. Capitalists invest their capitals in the sphere of production which is most profitable. The transfer of capital to the given sphere of production creates an increased demand for labor in that branch and consequently an increase of wages. This attracts hands, living labor, to the given branch.[1] The distribution of productive forces among individual spheres of the social economy takes the form of a distribution of capitals among these spheres. This distribution of capitals in turn leads to a corresponding distribution of living labor, or labor-power. If in a given country we observe an increase of capital invested in coal mining, and an increase in the number of workers employed in coal mining, we can ask ourselves which of these events was the cause of the other. Obviously, no one will disagree about the answer: the transfer of capital led to the transfer of labor power, and not inversely. In the capitalist society, the distribution of labor is regulated by the distribution of capital. Thus if our goal (as before) is to analyze the laws of distribution of social labor in the social economy, we must resort to a round-about path and proceed to a preliminary analysis of the laws of distribution of capital.
The simple commodity producer spends his labor in production and tries to get a market value which is proportional to the labor he expends on his product. This market value must be adequate for his own and his family's subsistence, and for the continuation of production at the previous volume, or at a slightly expanded volume. However, the capitalist spends his capital for production. He tries to get a return of capital which is larger than his original capital. Marx formulated this difference in his well-known formulas of the simple commodity economy, C-M-C (commodity-money-commodity) and the capitalist economy, M-C-M + m (money-commodity-increased money). If we split this short formula we will see technical differences (small and large-scale production) and social differences (which social class organizes production) between the simple commodity economy and the capitalist economy. We will see differences in the motives of producers (the craftsman strives to secure his subsistence, the capitalist strives to increase value) as results of the different character of production and the different social position of the producer. "The expansion of value, which is the objective basis or mainspring of the circulation M-C-M, becomes his subjective aim" (C., I, p. 152). The capitalist directs his capital to one or another sphere of production depending on the extent to which the capital invested in the given sphere increases. The distribution of capital among different spheres of production depends on the rate of increase of the capital in them.
The rate of increase of capital is determined by the relation between m, incremental capital, and M, invested capital. In the simple commodity economy, the value of commodities is expressed by the formula: C = c + (v + m).[2] The craftsman subtracts the value of the means of production which he used, namely c, from the value of the finished product, and the rest (v + m), which he added by his labor, is spent partly for his own and his family's subsistence goods (v) and the remainder represents a fund for the expansion of consumption or production (m). The same value of the product has the form C = (c + v) + m for the capitalist. The capitalist subtracts (c + v) = k of invested capital, or the costs of production, from the value of the commodity, whether this is spent on the purchase of means of production (c) or on the labor force (v). He considers the remainder, m, as his profit.[3] Consequently, c + v = k, and m = p. The formula C = (c + v) + m is transformed into the formula C = k + p, i.e., "the value of a commodity = cost-price + profit" (C., III, p. 36). However, the capitalist is not interested in the absolute quantity of profit, but in the relation of the profit to the invested capital, namely in the rate of profit p' = p/k. The rate of profit expresses "the degree of self-expansion of the total capital advanced" (C., III, p. 45). Our earlier statement that the distribution of capital depends on its rate of increase in various spheres of production means that the rate of profit becomes the regulator of the distribution of capital.
The transfer of capital from spheres of production with low rates of profit to spheres of production with higher rates of profit creates a tendency toward the equalization of profit rates in all spheres of production, a tendency toward the establishment of a general profit rate. Obviously this tendency is never realized completely in an unorganized capitalist economy, since in this economy complete equilibrium between the various spheres of production does not exist. But this absence of equilibrium, which is accompanied by differences in rates of profit, leads to the transfer of capital. This transfer tends to equalize profit rates and' to establish equilibrium among the different productive branches. This "incessant equilibration of constant divergences" (C., III, p. 196) provokes the striving of capital for the highest rate of profit. In capitalist production, "it is rather a matter of realizing as much surplus-value, or profit, on capital advanced for production, as any other capital of the same magnitude, or pro rata to its magnitude in whichever line it is applied... In this form capital becomes conscious of itself as a social power in which every capitalist participates proportionally to his share in the total social capital" (C., III, p. 195). In order to establish such a general average rate of profit, the existence of competition among capitalists engaged in different branches of production is necessary. The possibility for the transfer of capital from one branch to another is also necessary, since if this was not the case, various rates of profit could be established in different branches of production. If such competition of capitals is possible, equilibrium among the different productive branches can be theoretically assumed only in case the rates of profit which exist in these branches are approximately equal. Capitalists who work in average, socially necessary conditions in these productive branches will gain the general, average rate of profit.
Capitals of equal value invested in different spheres of production yield the same profit. Capitals which differ in size yield profit in proportion to their size. If capitals K and K1 yield profits P and P1, then
P/K = P1/K1 = p'
where p' is the general, average rate of profit. But where does the capitalist get his profit? From the selling price of his commodity. The profit of the capitalist, p, is the surplus: the selling price of the commodity minus the costs of production. Thus, the selling prices of different commodities have to be set at a level at which capitalists, the producers of these commodities, will receive a surplus from the selling price, a profit, which is proportional to the size of the invested capital, after they reimburse, or pay for, their costs of production. The selling price of goods, which covers the costs of production and yields an average profit on the whole invested capital, is called the production price. In other words, production price is a price of commodities at which capitalists gain an average profit on their invested capital. Since equilibrium in the different branches of production presupposes, as we have seen, that capitalists in all branches of production receive an average profit, equilibrium between the different spheres of production presupposes that the products are sold at production prices. Production price corresponds to the equilibrium of the capitalist economy. This is a theoretically defined, average level of prices at which the transfer of capital from one branch to another no longer takes place. If the labor-value corresponded to the equilibrium of labor among the various spheres of production, then the production price corresponds to the equilibrium of capital invested in the different spheres. "... price of production ... is a prerequisite of supply, of the reproduction of commodities in every individual sphere" (C., Ill, p. 198), i.e., the condition of equilibrium among the different spheres of the capitalist economy.
The production price should not be confused with the market-price, which constantly fluctuates above and below it, sometimes exceeding the production price, sometimes falling below it. The production price is a theoretically defined center of equilibrium, a regulator of the constant fluctuations of market prices. In conditions of a capitalist economy, the production price performs the same social function which the market-price determined by labor expenditures performs in conditions of simple commodity production. The first as well as the second are "equilibrium prices," but labor value corresponds to a state of equilibrium in the distribution of labor among the various spheres of the simple commodity economy, and production price corresponds to the equilibrium state in the distribution of capitals among the different spheres in the capitalist economy. This distribution of capital in turn points to a certain distribution of labor. We can see that competition leads to the establishment of a different price level of commodities in different social forms of economy. As Hilferding said, very much to the point, competition can explain only the "tendency towards the establishment of equality in economic relations" for individual commodity producers. But what does the equality among these economic relations consist of? The equality depends on the objective social structure of the social economy. In one case it will be an equality of labor, in another case an equality of capital.
As we have seen, production price equals costs of production plus the average profit on invested capital. If the average profit rate is given, then it is not difficult to calculate the production price. Let us assume that the invested capital is 100, the average rate of profit 22%. If the advanced capital is amortized during the year, then the production price is equal to the entire capital. The production price equals 100 + 22 = 122. The calculation is more complex if only one part of the fixed invested capital is used up during the year. If the capital of 100 consists of 20 v and 80 c, from which only 50 c are used up during the year, then the costs of production are equal to 50 c + 20 v = 70. To this sum is added 22%. This percentage is not calculated on the basis of the costs of production, 70, but of the entire invested capital, 100. Thus the production price is 70 + 22 = 92 (C., III, pp. 154-155). If from the same constant capital of 80 c, only 30 c were used up during the year, then the costs of production would be 30 c + 20 v = 50. To this sum, as before, is added the profit of 22. The production price of the commodity equals the costs of production plus the average profit on the entire invested capital.
To simplify our computations, we will assume that the entire invested capital is used up during the year, i.e., that the costs of production are equal to the invested capital. If two commodities are produced by means of capitals K and K1 then the production price of the first commodity equals K + p' K, and of the second, K1 + p' K1.[4]The production prices of two commodities are related to each other in the following way:
(K + p'K)/(K1 + p'K1) = (K (1 + p'))/(K1(1 + p')) = K/K1
Production prices of commodities are proportional to the capitals by means of which the commodities are produced. Commodities have the same production price if they are produced with the same capitals. The equalization, on the market, of two commodities which are produced in different branches, means the equality of two capitals.
The market equalization of commodities produced with equal capitals means an equalization of commodities produced with unequal quantities of labor. Equal capitals with different organic compositions put different quantities of labor into action. Let us assume that one capital of 100 consists of 70 c and 30 v. Another capital of 100 consists of 90 c and 10 v. If the rate of surplus value is 100%, the living labor of workers is twice as large as the paid labor expressed by the variable capital (i.e., the wage). Thus 70 units of past labor and 60 units of living labor are expended on the production of the first commodity-a total of 130; 90 units of past labor and 20 units of living labor are expended on the production of the second commodity-a total of 110. Since both commodities were produced by equal capitals, they are equalized with each other on the market regardless of the fact that they were produced by unequal quantities of labor. The equality of capitals means the inequality of labor.
The divergence between the size of the capitals and the amount of labor is also due to differences in the turnover period of the variable part of the capital. We assume that the organic composition of both capitals is equal, namely 80 c + 20 v. However, the variable part of the first capital circulates once a year, and of the second capital, three times, i.e., every third of a year the capitalist pays his workers 20 v. The sum of wages paid to the workers during the year equals 60. It is obvious that the labor expenditures for the first commodity are 80 + 40 = 120, and for the second commodity, 80 + 120 = 200. But since the invested capitals, despite the differences in the turnover period, are 100 in both cases, the commodities are equalized with each other even though they are produced by unequal amounts of labor. It is necessary to mention that "the difference in the period of turnover is in itself of no importance, except so far as it affects the mass of surplus-labor appropriated and realized by the same capital in a given time" (C., III, p. 152}, i.e., if we are dealing with the difference in the turnover period of variable capital. The phenomena mentioned here, namely the differences in the organic composition of capital and in the turnover period, can in the last analysis be reduced to the fact that the size of capital in itself cannot serve as an indicator of the amount of living labor which it activates, since this amount of labor depends on: 1) the size of the variable capital, and 2) the number of its turnovers.
Consequently, we reach a conclusion which at first glance contradicts the labor theory of value. Starting with the basic law of equilibrium of the capitalist economy, namely from equal rates of profit for all spheres of production, from the sale of commodities by production prices which contain equal profit rates, we reach the following results. Equal capitals activate unequal quantities of labor. Equal production prices correspond to unequal labor-values. In the labor theory of value the basic elements of our reasoning were the labor-value of commodities as a function of the productivity of labor, and the distribution of labor among different spheres of production in a state of equilibrium. But the production price does not coincide with the labor value and the distribution of capital does not coincide with the distribution of labor. Does this mean that the basic elements of the labor theory of value are completely superfluous for analyzing the capitalist economy, that we must throw out this unnecessary theoretical ballast and concentrate our attention exclusively on the production price and the distribution of capital? We will try to show that the analysis of production prices and distribution of capital in turn presupposes labor-value, that these central links of the theory of the capitalist economy do not exclude the links of the labor theory of value which were treated above. On the contrary, in our further analysis we will show that production price and distribution of capitals lead to labor-value and distribution of labor and, parallel with them, are included in a general theory of equilibrium of the capitalist economy. We must build a bridge from the distribution of capitals to the distribution of labor, and from production price to labor-value. First of all, we will deal with the first half of this task.
We have seen that the distribution of capitals does not coincide with the distribution of labor, that the equality of capital means an inequality of labor. If a capital of 100, expended in a given sphere of production, is equalized, through the exchange of commodities on the market, with a capital of 100 spent in any other sphere of production, then, if there are differences in the organic composition of these capitals, this will mean that the given quantity of labor expended in the first branch will be equalized with another quantity of labor, expended in the second branch, which is not equal to the first quantity. Now we must still determine precisely what quantities of labor spent in different spheres of production are equalized with each other. Even though the size of the capitals does not coincide quantitatively with the amounts of labor which they activated, this does not mean that there is no close connection between these capitals and the labor. This connection can be observed if we know the organic composition of the capitals. If the first capital consists of 80 c + 20 v, and the second of 70 c + 30 v, and if the rate of surplus value is 100%, then the first capital activates 40 units of living labor and the second 60. At the given rate of surplus value, "a certain quantity of variable capital represents a definite quantity of labor-power set in motion, and therefore a definite quantity of materialized labor" (C., III, p. 144). "The variable capital thus serves here (as is always the case when the wage is given) as an index of the amount of labor set in motion by a definite total capital" (Ibid.). Thus we know that, in the first sphere of production, the total amount of labor expenditure consists of 120 (80 past and 40 living) and in the second of 130 (70 past and 60 living). Starting from a distribution of capitals among given spheres of production (100 each), we have arrived, through the organic composition of capital, to the distribution of social labor between these spheres (120 in the first and 130 in the second). We know that the amount of labor of 120, expended in the first branch, is equalized with a mass of labor of 130 expended in the second sphere. The capitalist economy establishes equilibrium between unequal quantities of labor if they are activated by equal capitals. Through the laws of equilibrium of capitals we have come to the equilibrium in the distribution of labor. Actually, in conditions of simple commodity production, equilibrium is established between equal quantities of labor, and in conditions of a capitalist economy, between unequal quantities. But the task of scientific analysis consists of clearly formulating the laws of equilibrium and distribution of labor no matter what form this formula takes. If we are dealing with a simple schema of distribution of labor which is determined by the labor-value (which in turn depends on the productivity of labor), then we get the formula of equal quantities of labor. If we assume that the distribution of labor is determined by the distribution of capital, which acquires meaning as an intermediate link in the causal chain, then the formula of the distribution of labor depends on the formula of the distribution of capitals: unequal masses of labor which are activated by equal capitals are equalized with each other. The subject of our analysis remains, as before, the equilibrium and the distribution of social labor. In the capitalist economy this distribution is realized through the distribution of capitals. This is why the formula on the equilibrium of labor becomes more complex than for the simple commodity economy; it is derived from the formula for the equilibrium of capitals.
As we have seen, the equalization of things on the market is closely connected with the equalization of labor in a capitalist society as well. If the products of two spheres are equalized on the market, and if they are produced with equal amounts of capital and with the expenditure of unequal masses of labor, this means that in the process of distribution of social labor among the different branches, unequal masses of labor activated by equal capitals are equalized with each other. Marx did not limit himself to pointing out the inequality of the labor-value of two commodities with equal production prices: he gave us a theoretical formula for the deviation of production price from labor-value. Nor did he limit himself to the assertion that in the capitalist economy, unequal masses of labor expended in different spheres are equalized with each other: he gave us a theoretical formula for the deviation of the distribution of labor from the distribution of capitals, i.e., he established a relation between both of these processes through the concept of the organic composition of capital.
To illustrate what we have outlined, we can cite the first half of Marx's table in Volume III of Capital (we have changed some of the headings). "Let us take five different spheres of production, and let the capital in each have a different organic composition" (C., III, p. 155). The total sum of social capital equals 500, and the rate of surplus value is 100%.
| Distribution of Capital | Organic Composition of Capital | Distribution of Labor | |
| I. | 100 | 80c + 20v | 120 |
| II. | 100 | 70c + 30v | 130 |
| III. | 100 | 60c + 40v | 140 |
| IV. | 100 | 85c + 15v | 115 |
| V. | 100 | 95c + 5v | 105 |
We have called the third column "distribution of labor." This column shows the amount of labor expended in each sphere. Marx called this column "Value of products," because the labor value of the total product of each sphere of production is determined by the quantity of labor expended in each sphere. Critics of Marx's theory hold that this title, "Value of the Product," is fictional, artificially constructed, and theoretically superfluous. They do not take into account that this column does not only show the labor value of the different spheres of production, but also the distribution of social labor among the different spheres of production, i.e., a phenomenon which exists objectively and has central significance for economic theory. Rejection of this column is equivalent to the rejection of economic theory, which analyzes the working activity of society. The table clearly shows how Marx bridged the distribution of capital, through the organic composition of capital, with the distribution of social labor.[5] Thus the causal chain of connections becomes more profound and acquires the following form: production price-distribution of capitals-distribution of social labor. Now we must turn to the analysis of the first link of this chain, production price, and to see if this link does not presuppose other, more primary links.
Above we reached the following schema of causal relations: production price-distribution of capitals-distribution of labor. The starting point of this schema is production price. Can we remain with production price in our analysis, or must we take the analysis further? What is production price? Costs of production plus average profit. But what do costs of production consist of? They consist of the value of the constant and variable capital spent in production. Let us take the next step by asking: what is the value of constant and variable capital equal to? It is obviously equal to the value of the commodities which are its components (namely machines, raw materials, subsistence goods, etc.). In this way all our arguments turn in a vicious circle: the value of commodities is explained by production prices, i.e., costs of production or value of capital, and the value of capital, in turn, is reduced to the value of commodities. "Determining the value of commodities by the value of capitals is the same as determining the value of commodities by the value of commodities" (Theorien über den Mehrwert, III, p. 82).
To prevent production price from becoming a vicious circle, we must find those conditions which lead to changes in production prices and in average rates of profit. We will begin with costs of production.
If the average rate of profit remains unchanged, then the production prices of commodities change when the costs of production change. Costs of production of given commodities change in the following instances: 1) when the relative quantities of means of production, and the labor necessary for production, change, namely when the productivity of labor in the given sphere of production changes, given constant prices; 2) when the prices of means of production change; this presupposes changes in the productivity of labor in branches which produce these means of production (if the relative quantity of means of production and labor force are constant). In both cases, costs of production change in relation to changes in the productivity of labor, and, consequently, in relation to changes in labor value. Thus, "the general rate of profit remains unchanged. In this case the price of production of a commodity can change only if its own value has changed. This may be due to more, or less, labor being required to reproduce the commodity in question, either because of a change in the productivity of labor which produces this commodity in its final form, or of the labor which produces those commodities that go into its production. The price of production of cotton yarn may fall, either because raw cotton is produced cheaper than before, or because the labor of spinning has become more productive due to improved machinery" (C., III, p. 206; also see p. 165). It is necessary to note that production prices expressed quantitatively do not exactly coincide with the labor-value of the commodities which are their constituents. "Since the price of production may differ from the value of a commodity, it follows that the cost-price of a commodity containing this price of production of another commodity may also stand above or below that portion of its total value derived from the value of the means of production consumed by it" (C., III, pp. 164-165). We can see that this circumstance, to which Tugan-Baronovskii attached such great significance in his critique of Marx's theory, was well known to Marx himself. Marx even cautioned "that there is always the possibility of an error if the cost-price of a commodity in any particular sphere is identified with the value of the means of production consumed by it" (C., III, p. 165). But this deviation does not in any way conflict with the fact that changes in labor-value which are caused by changes in the productivity of labor bring about changes in costs of production and thus in production prices. This is precisely what had to be proved. The fact that the quantitative expressions of different series of events diverge does not remove the existence of a causal relation among them nor deny that changes in one series depend on changes in the other. Our task is complete if we can only establish the laws of this dependence.
The second part of production price, besides costs of production, is average profit, i.e., the average rate of profit multiplied by the capital. We must now examine in greater detail the formation of average profit, its magnitude, and its changes.
The theory of profit analyzes the interrelations, and the laws of change, of the incomes of individual industrial capitalists and groups of capitalists. But the production relations among individual capitalists and their groups cannot be understood without a preliminary analysis of the basic production relation between the class of capitalists and the class of wage laborers. Thus the theory of profit, which analyzes the interrelations among the incomes of individual capitalists and their groups, is built by Marx on the basis of the theory of surplus value, in which he analyzed the interrelations between the income of the capitalist class and the class of wage laborers.
We know from the theory of surplus value that in capitalist society the value of a product is broken down to the following three components. One part (c) compensates the value of constant capital used up in production-this is a reproduced, and not a newly-produced value. When this value is subtracted from the value of the whole product (C - c), we get the value produced by living labor, "created" by it. This value is a result of the given process of production. It, in turn, is composed of two parts: one (v) reimburses the workers for the value of the subsistence goods, i.e., refunds their wages, or the variable capital. The remainder, m = C - c - v = C - (c + v) = C - k, is the surplus value which belongs to the capitalist and which he spends for the purpose of personal consumption and for the expansion of production (i.e., accumulation). In this way the entire value which is received is divided into a fund for the reproduction of constant capital (c), the subsistence fund of labor or the reproduction of labor power (v), and the fund for the subsistence of the capitalist and for expanded reproduction (m).
Surplus value arises because the labor which is expended by workers in the process of production is larger than the labor necessary for the production of their subsistence fund. This means that surplus value increases to the extent that the labor expended in production increases and the labor necessary for the production of the worker's subsistence fund decreases. Surplus value is determined by the difference between total labor and paid labor, namely by the unpaid or surplus labor. Surplus value is "created" by surplus labor. However, as we explained above, it is erroneous to represent the problem as if the surplus labor, as if the material activity, "created" surplus value as a property of things. Surplus labor "is expressed," "is manifested," "is represented" (sich darstellt) in surplus value. Changes in the magnitude of surplus value depend on changes in the quantity of surplus labor.
The magnitude of surplus labor depends: 1) on its relation to the necessary, paid labor, i.e., on the rate of surplus labor or the rate of surplus value ~; 2) (if we take this rate as given) on the number of workers[6] i.e., on the quantity of living labor which is activated by capital. If the rate of surplus value is given, the total sum of surplus value depends on the total quantity of living labor and, consequently, on the surplus labor. Let us now take two equal capitals, of 100 each, which give equal profit because of the tendency of the profit rate to equalize. If the capitals are spent exclusively to pay for the labor power (v), then they activate equal masses of living labor and, consequently, of surplus labor. Here equal profits correspond to equal capitals and also to equal quantities of surplus labor, so that profit coincides with surplus value. We get the same result if both capitals are allocated in equal proportions to constant and variable capital. The equality of variable capitals means the equality of living labor which this capital activates. But if a capital of 100 in one sphere of production equals 70 c + 30 v, and another capital of 100 in another sphere equals 90 c + 10 v, then the mass of living labor which they activate and consequently the masses of surplus labor are not equal. Nevertheless, these capitals, being equal, yield equal profit, for example 20, because of the competition of capitals among different spheres of production. It is obvious that the profits which these capitals yield do not correspond to the masses of living labor which these capitals activate and consequently, to the masses of surplus labor. The profits are not proportional to the masses of labor. In other words, capitalists get sums of profit which differ from those they would get if profits were proportional to surplus labor or surplus value. Only in this context can we understand Marx's statement that capitalists "do not secure the surplus-value, and consequently the profit, created in their own sphere by the production of these commodities" (C., III, p. 158). Some of Marx's critics understood him to mean that the first of the capitals mentioned above seems to "give" the second capital 10 units of labor activated by the first capital; part of the surplus labor and surplus value "overflow," like liquid, from one sphere of production to another, namely from spheres with a low organic composition of capital to spheres which are distinguished by a high organic composition of capital: "Surplus values which are taken from workers in individual branches of production must flow from one sphere to another until the profit rate is equal and all capitals gain an average rate of profit... However, such an assumption is impossible, since surplus value does not represent an original money price, but only crystallized labor-time. In this form it cannot flow from one sphere to another. And, what is even more important, in reality it is not surplus value that flows, but the capitals themselves that flow from one sphere of production to another until the rates of profit are equalized."[7] It is perfectly obvious, and need not be proved here, that according to Marx the process of equalization of rates of profit takes place through the transfer of capitals, and not of surplus values, from one sphere to another (C., III, pp. 195, 158, 179, 236, and elsewhere). Since the production prices established in different spheres of production contain equal profit rates, the transfer of capital leads to the fact that the profits received by the capitals are not proportional to the quantities of living labor nor the surplus labor activated by the capitals. But if the relationship between the profits of two capitals engaged in different spheres of production does not correspond to the relationship between the living labors engaged by these capitals, it does not follow that a part of surplus labor or surplus value "is transferred," "overflows," from one sphere of production to another. Such a conception, based on a literal interpretation of some of Marx's statements, sometimes steals into the work of some Marxists; it arises from a view of value as a material object which has the characteristics of a liquid. However, if value is not a substance which flows from one man to another, but a social relation among people, fixed, "expressed," "represented" in things, then the conception of the overflow of value from one sphere of production to another does not result from Marx's theory of value but basically contradicts Marx's theory of value as a social phenomenon.
If, in capitalist society, there is no direct dependence between the profit of the capitalist and the quantity of living and thus surplus labor which is activated by capital, does this mean that we should completely give up the search for laws of the formation of average rates of profit and for causes which influence their level? Why is the average rate of profit in a given country 10%, and not 5% or 25%? We do not look to political economy for an exact formula for the calculation of the average profit rate in each case. However, we do look to political economy not to take a given rate of profit as the starting-point for analysis (a starting-point which does not have to be explained), but rather to try to determine the basic causes of the chain of events responsible for increases or decreases in the average rate of profit, i.e., the changes which determine the level of profit. This was Marx's task in his well known tables in Chapter 9 of Volume III of Capital. Since the second and third of Marx's tables take into account the partial consumption of fixed capital, we will take this as the basis of his first table in order not to complicate the computations. We will complete this table in a consistent manner. Marx takes five different spheres of production, with capitals of different organic compositions invested in them. The rate of surplus value is everywhere equal to 100%.
| Capitals | Labor Value of Products | Surplus Value | Average Rate of Profit | Production Price of Products | Deviation of Production Price from Value (and of Profit from Surplus Value) | |
| I. | 80c + 20v | 120 | 20 | 22% | 122 | +2 |
| II. | 70c + 30v | 130 | 30 | 22% | 122 | -8 |
| III. | 60c + 40v | 140 | 40 | 22% | 122 | -18 |
| IV. | 85c + 15v | 115 | 15 | 22% | 122 | +7 |
| V. | 95c + 5v | 105 | 5 | 22% | 122 | +17 |
| 390c + 110v | 610 | 110 | 110 | 610 | 0 | |
| 78c + 22v | --- | 22 | --- | --- | --- |
The total capital of society consists of 500, of which 390 is c and 110 is v. This capital is distributed among five spheres, with 100 in each. The organic composition of capital shows how much living labor, and thus surplus labor, is in each sphere. The total labor-value of the product is 610, and the total surplus value is 110. If the commodities of each sphere would be sold by their labor values, or, which is the same thing, if the profits in each sphere would correspond to the quantities of living labor and thus the surplus labor engaged in each sphere, then the profit rates of the individual spheres of production would be: 20%, 30%, 40%, 15%, and 5%. The spheres with the lowest organic composition of capital would get higher profit, and the spheres with a higher organic composition would get a lower profit. But, as we know, such different rates of profit are not possible in the capitaUst society, since this would cause a transfer of capitals from spheres with low rates of profit to spheres with high rates, until the same rate of profit is established in all spheres. The profit rate in the given case is 22%. Commodities produced by equal capitals of 100 are sold at equal production prices of 122, even though they are produced by unequal quantities of labor. Every capital of 100 receives a profit of 22%, even though equal capitals activate unequal quantities of surplus labor in the different spheres. "Every 100 of an invested capital, whatever its composition, draws as much profit in a year, or any other period of time, as falls to the share of every 100, the n'th part of the total capital, during the same period. So far as profits are concerned, the various capitalists are just so many stockholders in a stock company in which the shares of profit are uniformly divided per 100, so that profits differ in the case of the individual capitalists only in accordance with the amount of capital invested by each in the aggregate enterprise, i.e., according to his investment in social production as a whole, according to the number of his shares" (C., III, p. 158).
However, at which level is the average rate of profit established? Why is this rate equal precisely to 22%? Let us imagine that all the spheres of production are arranged in a decreasing sequence depending on the amount of living labor activated by each 100 units of capital. The variable parts of the capitals (taken in percentage shares) decrease from the top down (or the organic composition of capital increases from the top down). Parallel with this and in the same relation, the rates of profit decrease from the top down. The rate of profit which falls to each capital depends (in this example) on the quantity of living labor which the capital activates, or on the size of its variable capital. But as we know, such a difference in rates of profit is impossible. Competition among capitals would establish an average profit rate for all spheres of production; this average rate would be situated somewhere near the middle of the falling rates of profit. This average rate of profit corresponds to a capital which activates an average quantity of living labor or an average size of variable capital. In other words, the "average rate of profit ... is the percentage of profit in that sphere of average composition in which profit, therefore, coincides with surplus-value" (C., Ill, p. 173). In the given case, the entire social capital of 500 consists of 390 c + 110 v, the average composition of each 100 is 78 c + 22 v; if the rate of surplus value is 100%, every 100 of this capital of average composition gets a 22% rate of surplus value. The magnitude of this surplus value determines the size of the average rate of profit. This rate, consequently, is determined by the relation of the total mass of surplus value (m) produced in the society, to the total social capital (K), or p' = m/k.
Marx reaches the same conclusion in a different way. He uses the method of comparison which he often uses to explain the characteristic properties of the capitalist economy. In the given problem, the question of the average rate of profit, he compares the developed capitalist economy to 1) a simple commodity economy, and 2) an embryonic or hypothetical capitalist economy, which differs from developed capitalism by the absence of competition among capitals in different spheres of production, i.e., each capital is fixed within a given sphere of production.
Thus we can assume first of all a society of simple commodity producers who possess means of production with the value of 390 labor units; the living labor of its members amounts to 220. The productive forces of the society, which makeup 610 units of living and past labor, are distributed among five spheres of production. The combination of living and past labor is different in each sphere, depending on the technical properties of each sphere. Let us assume that the combinations are as follows (the first number represents past labor, the second living): I 80 + 40, II 70 + 60, III 60 + 80, IV 85 + 30, V 95 + 10. Let us assume that the productivity of labor has reached such a level of development that the petty producer reproduces the value of his subsistence goods with half his labor. Then the total value of the production, 610, breaks down into a fund of reproduction of means of production, 390, a fund for the subsistence of the producers, 110, and surplus value, 110. The surplus value remains in the hands of these same petty producers. They can spend it to expand consumption, to expand production (or partly for one and partly for the other). This surplus value of 110 will be proportionally distributed among the different spheres of production and the individual producers in terms of the labor expended. The distribution among the individual spheres will be: 20, 30, 40, 15, 5. Actually, these masses of surplus value are proportional only to the masses of living labor, and not to the past labor allocated to each sphere. If the masses of surplus value are calculated on the whole quantity of labor in each sphere (living and past) they give unequal rates of profit.[8] But in a simple commodity economy, producers are not aware of the category profit. They do not look at means of production as capital which must yield a given rate of profit, but as conditions for the activation of labor which give each commodity producer the possibility to put his labor on equal terms with that of other commodity producers, i.e., on terms or in conditions where equal quantities of living labor yield equal value.
Let us now assume that capitalists, and not petty commodity producers, are dominant in the economy. The other conditions remain unchanged. The value of the entire product, and the value of individual funds into which it breaks down, remain unchanged. The difference is that the fund for expanded consumption and expanded production (or surplus value) of 110 does not remain in the hands of direct producers, but is in the hands of capitalists. The same total social value is distributed in a different way between the social classes. Since the value of the product of individual spheres of production has not changed, the surplus value is distributed in the same proportions as before between individual spheres and individual capitalists. The capitalists in each of the five spheres get: 20, 30, 40, 15, 5. But they calculate these masses of surplus value on the entire invested capital, which is 100 in each sphere. As a result, the rates of profit are different. They can only be different because of the absence of competition between the individual spheres of production.
Finally, let us pass from hypothetical capitalism to actual capitalism, where there is competition of capital between the different spheres of production. Here different rates of profit are impossible, because this would cause a movement of capital from one sphere to another until all spheres had the same rate of profit. In other words, the distribution of the earlier mass of surplus value between different spheres and between individual capitalists will now be different; it will be proportional to the capitals invested in the spheres. The distribution of the surplus value is modified, but the total value of the fund of expanded consumption and expanded reproduction remain unchanged. The earlier mass of surplus value is now distributed among individual capitalists according to the size of their capitals. The average rate of profit is thus derived. It is determined by the relation of the total surplus value to the total social capital.
The comparison of a simple commodity economy, a hypothetical capitalist economy and a developed capitalist economy is not developed by Marx in the form in which we have presented it. Marx speaks of simple commodity production in Chapter 10 of the third volume of Capital. He takes a hypothetical capitalist economy as the basis of his analysis in Chapter 8 and in the tables of Chapter 9, where he assumes the absence of competition among individual spheres, and different profit rates. The comparison of the three different types of economy which we have carried out leads to certain doubts. A simple commodity economy presupposes the dominance of living labor over past labor, and an approximately homogeneous relation between living and past labor in the various branches of production. However, in our schemas this relation is assumed to be different in each sphere. This objection does not have a great deal of significance because different relations between living and past labor (even though they were not characteristic of the simple commodity economy) do not logically contradict that type of economy and may be used as an assumption in a theoretical schema. More serious doubts are aroused by the schema of the embryonic or hypothetical capitalist economy. If the absence of competition among the capitalists of the different spheres of that economy explains why commodities are not sold according to production prices, this absence of competition also makes it impossible to explain the sale of goods according to their labor values. In the simple commodity economy, the sale of goods according to labor-values can be maintained only on the condition that labor can transfer from one sphere to another, i.e., if there is competition among spheres of production. In one passage Marx noted that the sale of goods by their labor values assumes as a necessary condition that no natural or artificial monopoly makes it possible for the contracting sides to sell above value or forces them to sell below value (C., III, p. 178). But if there is no competition among capitals, if each capital is fixed in each sphere, then the state of monopoly results. Sales at prices above labor-values do not bring about a transfer of capital from other spheres. Sales at prices below labor-values do not cause an outflow of capital from the given sphere to others. There is no regularity in the establishment of exchange proportions among commodities in terms of their corresponding labor-values. On what basis does the schema of the embryonic capitalist economy assume that the sale of commodities takes place according to labor values, if competition among capitalists in different spheres is absent?
It is possible to answer this question only if the schema is explained in the form in which we explained it above. Diagram No. 2 is not a picture of an embryonic capitalism which existed in history, but a hypothetical theoretical schema derived from Diagram 1 (simple commodity economy) by means of a methodological procedure which consists of changing only one condition of the schema, all other conditions remaining the same. In schema No. 2, compared to No. 1, only one condition is changed. It is supposed that the economy is not run by petty commodity producers but by capitalists. The other conditions are assumed to be the same as before: the mass of living labor and past labor in each sphere, the value of the total product and the mass of surplus value, and thus the price of products; the selling price of commodities according to labor values is kept at the same level as earlier. The sale of commodities is a theoretical condition transferred to schema 2 from schema 1, and is only possible if there is another, additional theoretical condition, namely if there is no competition among capitalists in different spheres. Therefore, since we change this single condition by moving from schema 2 to schema 3 (developed capitalism), i.e., since we introduce the assumption of competition of capitals, the sale of goods according to their labor-values gives place to the sale of goods according to production prices in which an average rate of profit is realized by capitalists. But in carrying out this transition from schema 2 to schema 3 by the same methodological procedure, by changing one condition, we leave unchanged the other conditions, particularly the earlier mass of surplus value. In this way we reach the conclusion that the formation of a general average rate of profit reflects a redistribution of the earlier total mass of surplus value among capitalists. The share of this surplus value in the total social capital determines the level of the average rate of profit. We repeat that this "redistribution" of surplus value must not, in our view, be understood as a historical process which actually took place and which was preceded by an embryonic capitalist economy with different rates of profit in different spheres.[9] It is a theoretical schema of the distribution of profit in the capitalist economy. This schema is derived from the first schema (simple commodity production) by means of a two-fold change in the conditions. Moving from schema 1 to schema 2 we assumed that the social class which gets the surplus value changed. Moving from schema 2 to schema 3, we assumed that, in the context of the same class of capitalists, a redistribution of capital took place among the different spheres. Both of these transitions in essence represent two logical links of an argument. They are separated for the sake of clarity, even though they do not exist separately. In our opinion, the transformation of the intermediate logical link, schema 2, into a picture of an economy which existed in history as a transition from simple commodity production to developed capitalist production, is erroneous.
Thus, the average rate of profit is quantitatively determined by the relation between the total mass of surplus value and the total social capital. We assume that in Marx's system the magnitude of the average rate of profit is derived from the mass of total surplus value and not from the different profit rates, as it may seem from a first reading of Marx's work. Deriving the average rate of profit from different profit rates provokes objections based on the fact that the existence of different profit rates in different spheres is not logically or historically proved. The existence of different profit rates, according to this view, was brought about by the sale of products of different spheres according to their labor-values. But as we have seen above, different rates of profit in different spheres of production only played the role of a theoretical schema in Marx's work, a schema which explains the formation and magnitude of an average profit rate by means of comparison. Marx himself pointed out that, "The general rate of profit is, therefore, determined by two factors:
"1) The organic composition of the capitals in the different spheres of production, and thus, the different rates of profit in the individual spheres.
"2) The distribution of the total social capital in these different spheres, and thus, the relative magnitude of the capital invested in each particular sphere at the specific rate of profit prevailing in it; i.e., the relative share of the total social capital absorbed by each individual sphere of production" (C., III, p. 163). It is obvious that different rates of profit in individual spheres are used by Marx only as numerical expressions, indicators of the organic composition of capital, i.e., masses of living labor and thus of surplus labor activated by each 100 units of capital in a given sphere. This factor is combined with others; the quantity of surplus labor which belongs to each 100 units of capital in each sphere is multiplied by the size (the number of hundreds) of capital invested in the given sphere. As a result we get the mass of surplus labor and surplus value, first of all in the individual spheres, and then in the whole social economy. Thus the average rate of profit is not determined, in the last analysis, by the different profit rates in different spheres, but rather by the total mass of surplus value and by the relationship of this mass to total social capital,[10] i.e., by magnitudes which are not theoretically suspicious from the standpoint of the labor theory of value. At the same time these magnitudes reflect real facts of the social economy, namely the masses of living social labor and the social capital. The specific character of Marx's theory of production price consists precisely of the fact that the entire question of mutual relations between surplus value and profit is transferred from individual capitals to the total social capital. This is why, in our presentation of Marx's theory, different rates of profit in different spheres do not serve as a necessary intermediate link for a theory of the average rate of profit; this can be briefly summarized in the following way. In the capitalist economy the distribution of capital is not proportional to the distribution of living labor. A different quantity of living labor and thus of surplus labor belongs to each 100 units of capital in the different spheres. (The different rates of profit represent numerical expressions of this mutual relation between surplus labor and capital in each sphere.) This organic composition of capital in the different spheres and the size of the capital in each sphere determine the total mass of surplus labor and surplus value in the individual spheres and in the entire economy. Because of the competition of capitals, equal capitals in different spheres gain equal profits, and thus the profit which the individual capitals gain is not proportional to the quantities of living labor activated by these capitals. Consequently, the profit is not proportional to surplus value but is determined by the average profit rate, i.e., by the relation between the total surplus value and the total social capital.
If a reading of Chapter 8 of the third volume of Capital gives the impression that the differences in profit rates, which arise because of the sale of commodities according to their labor values, play the role of an indispensable link in Marx's constructions, this is explained by the following properties of Marx's exposition. When Marx approaches the decisive places of his system, when he must move from general definitions to more particular explanations, from general concepts to their modifications, from one "determination of form" to another, he resorts to the following method of exposition. With an enormous power of thought, he draws all the logical conclusions from the first definition which he develops, intrepidly developing all the consequences which follow from the concept to their logical end. He shows the reader all the contradictions of these consequences, i.e., their divergence from reality. When the reader's attention has been strained to its limit, when it begins to seem to the reader that the starting definition must be completely rejected because it is contradictory, Marx comes to the reader's help and suggests an exit from the problem, an exit which does not consist of throwing out the first definition, but rather of "modifying," "developing" and completing the first definition. Thus the contradictions are removed. Marx does this in Chapter 4 of the first volume of Capital, when he examines the transition from the value of commodities to the value of labor-power. He draws a conclusion on the impossibility of the formation of surplus value on the basis of an exchange of commodities according to their labor-value, i.e., he reaches a conclusion which openly conflicts with reality. In the further analysis, this conclusion is rejected by the theory of the value of labor power. This is precisely how the eighth chapter of Volume III of Capital is constructed. On the basis of the sale of goods according to labor-values, Marx concludes that different rates of profit exist in different spheres. Developing this conclusion to all its consequences, he ascertains at the end of Chapter 8 that this conclusion conflicts with reality and that this contradiction must be resolved. In Volume I of Capital, Marx had never claimed that the existence of surplus value was impossible; here he does not say that different profit rates are possible. The impossibility of surplus value in Chapter 4 of Volume 1, and the possibility of different profit rates in Chapter 8 of Volume III, do not serve Marx as logically necessary links for his constructions, but as proofs of the opposite. The fact that these conclusions lead to a logical absurdity shows that the analysis is not yet finished and has to be continued further. Marx does not determine the existence of different profit rates, but on the contrary, the inadequacy of any theory which is based on such a premise.
We have reached the conclusion that the average rate of profit is determined by the relation of total surplus value to the total social capital. From this it follows that changes in the average profit rate may result from changes in the rate of surplus value and also from changes in the relation of total surplus value to total social capital. In the first case, the change "can only occur either through a rise, or fall, in the value of labor-power, the one being just as impossible as the other unless there is a change in the productivity of the labor producing means of subsistence, i.e., in the value of commodities consumed by the laborer" (C., III, p. 205). Now we take the second case, when the changes start from capital, namely from an increase or decrease of its constant part. The changed relation of constant capital to labor reflects a change in the productivity of labor. "Thus, there has been a change in the productivity of labor, and there must have occurred a change in the value of certain commodities" (Ibid.). Changes in the average rate of profit, whether they result from the rate of surplus value or from capital, are in both cases brought about, in the last analysis, by changes in the productivity of labor and, consequently, by changes in the value of some goods.
From this it follows that changes in costs of production and changes in average profit rates are caused by changes in the productivity of labor. And since the production price consists of production costs plus average profit, changes in production prices are in the last analysis caused by changes in the productivity of labor and in the labor-value of some goods. If the change in production price is caused by a change in production costs, this means that the productivity of labor in the given sphere of production and the labor-value of the given sphere have changed. "If the price of production of a commodity changes in consequence of a change in the general rate of profit, its own value may have remained unchanged. However, a change must have occurred in the value of other commodities" (Ibid., pp. 205-206), i.e., changes in the productivity of labor in other spheres. In every case, the production price changes in relation to changes in the productivity of labor and corresponding changes in labor-value. Productivity of labor-abstract value-value-costs of production plus average profit-production price: this is the schema of causal relations between production price, on one hand, and the productivity of labor and labor-value, on the other.
Now, finally, we can consider the chain of logical links which complete Marx's theory of production price. The chain consists of the following basic links: productivity of labor-abstract labor-value-production price-distribution of capital-distribution of labor. If we compare this six-element schema to the four-element schema of simple commodity production: productivity of labor-abstract labor-value-distribution of labor, we see that the links of the simple commodity production schema have become components of the schema for the capitalist economy. Consequently, the labor theory of value is a necessary foundation for the theory of production price, and the theory of production price is a necessary development of the labor theory of value.
The publication of the third volume of Capital gave birth to an enormous literature on the so-called "contradictions" between Volume I and Volume III of Capital. Critics held that in Volume III, Marx had in essence repudiated his labor theory of value, and some even assumed that, when he had composed the first volume, he had never dreamed of the difficulties and contradictions into which the labor theory of value would lead him when he had to explain the profit rate. Karl Kautsky's foreword to the third volume of Capital documents that when the first volume of Capital was published, the theory of production price explained in Volume III had already been worked out by Marx in all its details. Already in the first volume, Marx frequently pointed out that in the capitalist society, average market prices deviate from labor-values. The content of the third volume of Theorien liber den Mehrwert also informs us of another important circumstance. All post-Ricardian political economy revolved around the question of the relation between production price and labor-value. The answer to this question was a historical task for economic thought. In Marx's view, the particular merit of his theory of value was that it gave a solution to this problem.
Critics who saw contradictions between the first and third volumes of Capital took as their starting-point a narrow view of the theory of value, seeing it exclusively as a formula of quantitative proportions in the exchange of commodities. From this standpoint the labor theory of value and the theory of production price did not represent two logical stages or degrees of abstraction from the same economic phenomena, but rather two different theories or statements which contradicted each other. The first 'theory holds that commodities are exchanged in proportion to the expenditure of labor necessary for their production. The second theory holds that commodities are not exchanged proportionally to these expenditures. What a strange method of abstraction, said Marx's critics; first it holds one thing, then another which contradicts the first. But these critics did not take into account that the quantitative formula for the exchange of commodities is only the final conclusion of a very complex theory which deals with the social form of the phenomena related to value, the reflection of a determined type of social production relations among people, as well as the content of these phenomena, their role as regulators of the distribution of social labor.
Anarchy in social production; the absence of direct social relations among producers; mutual influence of their working activities through things which are products of their labor; connection between the movement of production relations among people and the movement of things in the process of material production; "reification" of production relations, the transformation of their properties into the properties of "things"-all of these phenomena of commodity fetishism are equally present in every commodity economy, simple as well as capitalist. They characterize labor-value and production price in the same way. But every commodity economy is based on the division of labor, i.e., it represents a system of allocated labor. How is this division of social labor among various spheres of production carried out? It is directed by the mechanism of market prices, which provokes inflows and outflows of labor. Fluctuations of market prices display a certain regularity, oscillating around some average level, around a price "stabilizer," as Oppenheimer appropriately called it.[11]
This price "stabilizer," in turn, changes in relation to the increase of the productivity of labor and serves as a regulator of the distribution of labor. The increase of the productivity of labor influences the distribution of social labor through the mechanism of market price, whose movement is subject to the law of value. This is the simplest abstract mechanism which distributes labor in the commodity economy. This mechanism exists in every commodity economy, including the capitalist economy. There is no mechanism other than the fluctuation of market prices which distributes labor in the capitalist economy. But since the capitalist economy is a complex system of social production relations in which people do not relate to each other only as commodity owners but also as capitalists and wage laborers, the mechanism which distributes labor functions in a more complex manner. Since simple commodity producers spend their own labor in production, the increase of productivity of labor, expressed through the labor-value of products, causes inflows and outflows of labor, i.e., influences the distribution of social labor. In other words, the simple commodity economy is characterized by a direct causal relation between the productivity of labor expressed in the labor-value of products, and the distribution of labor.[12] In the capitalist society this causal relation cannot be direct since the distribution of labor takes place through the distribution of capital. The increase of productivity of labor, expressed in the labor-value of products, cannot influence the distribution of labor any other way than through its influence on the distribution of capital. Such influence on the distribution of capital is in turn possible only if changes in the productivity of labor and labor-value cause changes in costs of production or in the average rate of profit, i.e., influence the production price.
Thus the schema: productivity of labor-abstract labor-value-distribution of labor, represents, so to speak, a theoretical model of direct causal relations between the increase of productivity of labor expressed in labor-value, and the distribution of social labor. The schema: productivity of labor-abstract labor-value-production price-distribution of capital-distribution of labor, represents a theoretical model of the same causal chain, where the productivity of labor does not directly affect the distribution of labor, but rather through an "intermediate link" (an expression which Marx often used in this context): through the production price and the distribution of capital. In both schemas, the first and last terms are the same. The mechanism of causal relations between them is also the same. But in the first schema we assume that the causal connection is more immediate and direct. In the second schema we introduce elements which complicate the situation, namely intermediate links. This is the usual path of abstract analysis, a path which Marx resorted to in all his constructions. The first schema represents a more abstract, more simplified model of the events, but a model which is indispensable for an understanding of the more complex forms of events that take place in capitalist society. If we limited the scope of the analysis to the intermediate links which are visible on the surface of phenomena in the capitalist economy, namely production price and distribution of capital, then our analysis would remain incomplete in both directions, at the beginning and at the end. We would take production price (i.e., production costs plus average profit) as a starting-point. But if production price is explained in terms of costs of production, we simply refer the value of the product to the value of its components, i.e., we do not emerge from a vicious circle. Average profit remains unexplained, as do its volume and its changes. Thus production price can only be explained by changes in productivity of labor or in the labor value of products. On the one hand, we are wrong if we regard the distribution of capital as the final point of our analysis; we have to move on to the distribution of social labor. Thus the theory of production price must without fail be based on the labor theory of value. On the other hand, the labor theory of value must be further developed and completed in the theory of production price. Marx rejected every attempt to construct the theory of the capitalist economy directly from the labor theory of value and to avoid the intermediate links, average profit and production price. He characterized such attempts as "attempts to force and directly fit concrete relations to the elementary relation of value" (Theorien über den Mehrwert, III, p. 145), "attempts which present as existing that which does not exist" (Ibid., p. 97).
Thus the labor theory of value and the theory of production price are not theories of two different types of economy, but theories of one and the same capitalist economy taken on two different levels of scientific abstraction. The labor theory of value is a theory of simple commodity economy, not in the sense that it explains the type of economy that preceded the capitalist economy, but in the sense that it describes only one aspect of the capitalist economy, namely production relations among commodity producers which are characteristic for every commodity economy.
After the publication of the third volume of Capital, opponents of Marx's theory of value, and to some extent its advocates, created the impression that the conclusions of the third volume demonstrated the inapplicability of the law of labor value to the capitalist economy. This is why certain Marxists were prone to construct a so-called "historical" foundation for Marx's theory of value. They held that even though the law of labor value, in the form in which Marx developed it in the first volume of Capital, is not applicable to the capitalist economy, it is nevertheless completely valid for the historical period which precedes the emergence of capitalism and in which petty crafts and peasant economy are dominant. Certain passages which might be interpreted this way can be found in the third volume of Capital. There Marx says that "it is quite appropriate to regard the values of commodities as not only theoretically but also historically prius to the prices of production" (C., III, p. 177). These cursory comments by Marx were developed by Engels in detail in his article published in 1895 in Neue Zeit.[13] Here Engels gave a basis to the idea that Marx's law of value was in force during a historical period which lasted five to seven thousand years, a period which began with the appearance of exchange and ended in the 15th century, when capitalism emerged. Engels' article found ardent supporters, but just as ardent opponents, some of them Marxists. Opponents pointed out that exchange did not encompass the entire social economy before the appearance of capitalism, that it spread first to surpluses which existed after the satisfaction of the requirements of the self-sufficient, natural economic unit, that the mechanism of general equalization of different individual labor expenditures in separate economic units on the market did not exist, and that consequently it was not appropriate to speak of abstract and socially-necessary labor which is the basis of the theory of value. Here we will not be concerned with the historical controversy over whether commodities were exchanged in proportion to the labor expended on their production before the emergence of capitalism. For methodological reasons we are opposed to relating this question to the question of the theoretical significance of the law of labor-value for the explanation of the capitalist economy.
First of all, we turn to Marx's work. Some passages in Volume III of Capital can be used by proponents of a historical explanation of labor value. However, now that other works by Marx are available to us, we know with certainty that Marx himself was strongly opposed to the view that the law of value was in force in the period preceding the development of capitalism. Marx objected to the view of the English economist Torrens, a proponent of a view which one can even find in Adam Smith's work. Torrens held that the full development of a commodity economy, and consequently the full development of the laws which exist in that economy, is possible only in capitalism and not before. "This would mean that the law of labor-value exists in production which is not commodity production (or only partly commodity production), but it does not exist in production which is not based on the existence of products in the form of commodities. This law itself, and the commodity as the general form of products, are abstracted from capitalist production, and now supposedly cannot be applied to it" (Theorien über den Mehrwert, III, p. 80). "It now turns out that the law of value abstracted from capitalist production contradicts its phenomena" (Ibid., p. 78). These ironical notes by Marx clearly show his relation to the view of the theory of value as a law which functions in the pre-capitalist economy, but not in the capitalist economy. But how can we reconcile these statements with some observations in Volume III of Capital? The seeming divergence between them disappears if we return to the "Introduction to the Critique of Political Economy," which gives us a valuable explanation of Marx's abstract method of analysis. Marx emphasizes that the method of moving from abstract to concrete concepts is only a method by which thought grasps the concrete, and not the way the concrete phenomenon actually happened.[14] This means that the transition from labor-value or simple commodity economy to production price or the capitalist economy is a method for grasping the concrete, i.e., the capitalist economy. This is a theoretical abstraction and not a picture of the historical transition from simple commodity economy to capitalist economy. This confirms the view which we formulated earlier that the tables in Chapter 9 of the third volume of Capital, which illustrate the formation of general average rates of profit from different rates of profit, depict a theoretical schema of phenomena, and not the historical development of the phenomena. "The simplest economic category, say exchange value ... can have no other existence except as an abstract one-sided relation of an already given concrete and living aggregate" (Ibid.), i.e., the capitalist economy.
After having explained the theoretical character of abstract categories, Marx asks: "have these simple categories no independent historical or natural existence antedating the more concrete ones?" (Ibid., p. 295). Marx answers that such instances are possible. A simple category (for example value) can exist historically before the concrete category (for example, production price). But in this case the simple category still has a rudimentary, embryonic character which reflects relations of "undeveloped concreteness." "Thus, although the simple category may have existed historically before the more concrete one, it can attain its complete internal and external development only in complex forms of society" (Ibid., p. 297). Applying this conclusion to the question which interests us, we can say: labor-value (or commodity) is a historical "prius" in relation to production price (or capital). It existed in rudimentary form before capitalism, and only the development of the commodity economy prepared the basis for the emergence of the capitalist economy. But labor-value in its developed form exists only in capitalism. The labor theory of value, which develops a logical, complete system of the categories value, abstract labor, socially-necessary labor, etc., expresses the "abstract one-sided relation of an already given concrete and living aggregate," i.e., it expresses the abstraction of the capitalist economy.
The historical question of whether commodities were exchanged in proportion to labor expenditures before the emergence of capitalism must be separated from the question of the theoretical significance of the theory of labor-value. If the first question were answered affirmatively, and if the analysis of the capitalist economy did not require the labor theory of value, we could regard that theory as a historical introduction to political economy, but not in any way as a basic theoretical foundation on which Marx's political economy is built. Inversely, if the historical question were answered negatively, but if the indispensability of the labor theory of value for the theoretical understanding of the complex phenomena of the capitalist economy were proved, this theory would still be the starting-point of economic theory, as it is now. In brief, no matter how the historical question about the influence of the law of labor-value in the period before capitalism were solved, this solution would not in the least free Marxists from their responsibility to accept the challenge of their opponents on the question of the theoretical significance of the law of labor value for an understanding of the capitalist economy. Confusing the theoretical and the historical setting of the theory of value is not only pointless, as we have shown, but also harmful. Such a treatment puts the proportions of exchange into the foreground, and ignores the social form and the social function of value as the regulator of the distribution of labor, a function which value performs to a great extent only in a developed commodity economy, i.e., a capitalist economy. If the analyst finds that primitive tribes, who live in conditions of a natural economy and rarely resort to exchange, are guided by labor expenditures when they establish exchange proportions, he is prone to find here the category of value. Value is transformed into a supra-historical category, into labor expenditures independent of the social form of the organization of labor.[15] The "historical" setting of the problem thus leads to ignoring the historical character of the category value. Other theorists, assuming that "the emergence of exchange value must be sought in a natural economy which developed into a money economy," finally determine value not in terms of the labor which the producer spends on his production, but by the labor which the producer would have to spend in the absence of exchange and of the necessity to make the product by his own labor.[16]
The labor theory of value and the theory of production price differ from each other, not as different theories which function in different historical periods, but as an abstract theory and a concrete fact, as two degrees of abstraction of the same theory of the capitalist economy. The labor theory of value only presupposes production relations among commodity producers. The theory of production price presupposes, in addition, production relations between capitalists and workers, on one hand, and among various groups of industrial capitalists on the other.