Front Page Titles (by Subject) CHAPTER VIII.: DIFFERENT COMPOSITION OF CAPITALS IN DIFFERENT LINES OF PRODUCTION AND RESULTING DIFFERENCES IN THE RATES OF PROFIT. - Capital: A Critique of Political Economy. Volume III: The Process of Capitalist Production as a Whole
Return to Title Page for Capital: A Critique of Political Economy. Volume III: The Process of Capitalist Production as a Whole
The Online Library of Liberty
A project of Liberty Fund, Inc.
Search this Title:
CHAPTER VIII.: DIFFERENT COMPOSITION OF CAPITALS IN DIFFERENT LINES OF PRODUCTION AND RESULTING DIFFERENCES IN THE RATES OF PROFIT. - Karl Marx, Capital: A Critique of Political Economy. Volume III: The Process of Capitalist Production as a Whole 
Capital: A Critique of Political Economy. Volume III: The Process of Capitalist Production as a Whole, by Karl Marx. Ed. Federick Engels. Trans. from the 1st German edition by Ernest Untermann (Chicago: Charles H. Kerr and Co. Cooperative, 1909).
About Liberty Fund:
Liberty Fund, Inc. is a private, educational foundation established to encourage the study of the ideal of a society of free and responsible individuals.
The text is in the public domain.
Fair use statement:
This material is put online to further the educational goals of Liberty Fund, Inc. Unless otherwise stated in the Copyright Information section above, this material may be used freely for educational and academic purposes. It may not be used in any way for profit.
DIFFERENT COMPOSITION OF CAPITALS IN DIFFERENT LINES OF PRODUCTION AND RESULTING DIFFERENCES IN THE RATES OF PROFIT.
IN the preceding part we demonstrated among other things that the rate of profit may vary, may rise or fall, while the rate of surplus-value remains the same. In the present chapter we assume that the intensity of exploitation, and therefore the rate of surplus-value and the length of the working day, are the same in all spheres of production into which the social labor of a certain country is divided. Adam Smith has already shown explicitly that many differences in the exploitation of labor in different spheres of production balance one another by many actual causes, or causes regarded as such by prevailing prejudices, so that they are mere evanescent distinctions and are of no moment in this calculation. Other differences, for instance those in the scale of wages, rest largely on the difference between simple and complicated labor, mentioned in the beginning of volume I, which do not affect the intensity of exploitation in the different spheres of production, although they render the conditions of the laborers in those spheres very unequal. For instance, if the labor of a goldsmith is paid better than that of a day-laborer, the surplus-labor of the goldsmith produces correspondingly more surplus-value than that of the day-laborer. And while the compensation of wages and working days, and thereby of the rates of surplus-value, between different spheres of production, or even different investments of capital in the same sphere of production, is checked by many local obstacles, it is nevertheless accomplished at an increasing degree with the advance of capitalist production and the subordination of all economic conditions under this mode of production. The study of such frictions, while quite important for any special work on wages, may be dispensed with as being accidental and unessential in a general analysis of capitalist production. In such a general analysis it is always assumed that the actual conditions correspond to the terms used to express them, or, in other words, that actual conditions are represented only to the extent that they are typical of their own case.
The difference in the rates of surplus-value in different countries, and consequently in the degree of national exploitation of labor, is immaterial for our present analysis. For we desire to analyse precisely the way in which a general rate of profit is brought about in a certain country. It is evident, however, that a comparison of the various national rates of profit requires but a collation of previous analyses with that which is to follow. First consider the differences in the national rates of surplus-value, then compare on this basis the differences in the national rates of profit. Those differences which are not due to differences in the national rates of surplus-value, must be due to circumstances in which the surplus-value is assumed to be universally the same, constant, as it is in the analysis of this chapter.
We demonstrated in the preceding chapter that, assuming the rate of surplus-value to be constant, the rate of profit may rise or fall in consequence of circumstances which raise or lower the value of one or the other parts of constant capital, and so affect the proportion between the variable and constant components of capital in general. We observed, furthermore, that circumstances which prolong or reduce the time of turn-over of a certain capital may also influence the rate of profit in a similar manner. Since the mass of profits is identical with the mass of surplus-value, the surplus-value itself, it was also seen that the mass of profits, in distinction from the rate of profits, was not touched by the aforementioned fluctuations of value. These fluctuations modified merely the rate through which a certain surplus-value, and therefore a profit of a given magnitude, express themselves, in other words, they indicate the relative magnitude of surplus-value, or profits, as compared with the magnitude of the advanced capital. To the extent that capital was released or tied up by such fluctuations of value, it was not only the rate of profit, but the profit itself, which could be affected by this indirect route. However, this always applied only to such capital as was already engaged, not to new investments about to be made. Besides, the increase or reduction of profit always depended on the extent to which the same capital could set in motion more or less labor in consequence of such fluctuations of value, in other words, the extent to which the same capital, with the same rate of surplus-value, could obtain a larger or smaller amount of surplus-value. So far from contradicting the general rule, or being an exception from it, this seeming exception was really but a special case in the application of the general rule.
It was seen in the preceding part, that the rate of profit varied, when the degree of exploitation was constant while the value of the component parts of constant capital, and the time of turn-over of capital, changed. The obvious conclusion from this was that the rates of profit of different spheres of production existing simultaneously side by side had to differ, when, other circumstances remaining unchanged, the time of turn-over of the invested capitals differed, or when the proportions of the values of the organic components of these capitals were different in the different lines of production. That which we previously regarded as changes occurring successively in the same capital will now be considered as simultaneous differences of contemporaneous investments of capital in different spheres of production.
Under these circumstances we shall have to analyse: 1) The differences in the organic composition of capitals. 2) The differences in their times of turn-over.
The natural premise in this entire analysis is that, in speaking of the composition, or of the turn-over, of a capital in a certain line of production, we always mean the average normal proportions of the capital invested in this line, or, more generally, of the average of the total capital invested in this sphere, not of the temporary differences of the individual capitals in it.
Since our assumption is, furthermore, that the rate of surplus-value and the working day are constant, and since this assumption implies also the constancy of wages, it follows that a certain quantity of variable capital expresses a definite quantity of exploited labor-power and therefore a definite quantity of materialised labor. In other words, if 100 p.st. represent the weekly wages of 100 laborers, indicating 100 actual labor-powers, then n times 100 p.st. indicates the labor-powers of n times 100 laborers, and 100/n p.st. those of 100/n laborers. The variable capital serves here, as is always the case when the wages are given, as an index of the amount of labor set in motion by a definite total capital. Differences in the magnitude of the employed variable capitals serve, therefore, as indices of the differences in the amount of labor-power set in motion. If 100 p.st. indicate 100 laborers per week, representing 6,000 working hours, if the weekly working time is 60 hours, then 200 p.st. indicate 12,000, and 50 p.st. indicate 3,000 working hours.
By the composition of capital we mean, as we have stated in volume I, the proportions of its active and passive parts, of variable and constant capital. Two proportions require consideration under this heading. They are not equally important, although they may produce the same effects under certain circumstances.
The first proportion rests on a technical basis, and must be considered as existing at a certain stage of development of the productive forces. A definite quantity of labor-power, represented by a definite number of laborers, is required for the purpose of producing a definite quantity of products, for instance in one day, and thereby to consume productively, by setting in motion, a definite quantity of means of production, machinery, raw materials, etc. A definite number of laborers corresponds to a definite quantity of means of production, so that a definite quantity of living labor corresponds to a definite quantity of materialised labor in means of production. This proportion differs a great deal in different spheres of production, and frequently even in different branches of one and the same industry. On the other hand, it may occasionally be entirely or approximately the same in widely separated lines of industry.
This proportion forms the technical composition of capital and is the primary basis of its organic composition.
However, it is possible that this first proportion may be the same in different lines of industry, provided that the variable capital is merely an index of labor-power, and the constant capital merely an index of the mass of means of production set in motion by the labor-power. For instance, certain work in copper and iron may be conditioned on the same proportional composition between labor-power and the mass of means of production. But since copper is more expensive than iron, the proportion of value between variable and constant capital may be different in either case, and then the composition of the value of the total capitals is, of course, likewise different. The difference between the technical composition and the composition of values is manifested by each branch of industry by the fact that the proportion of the values of the two parts of capital may vary while the technical composition is constant, and the proportion of values may remain the same while the technical composition varies. This last eventuality will, of course, be possible only if the change in the proportion of the employed masses of means of production and labor-power is compensated by an opposite change in their values.
The composition of the values of capital, which is determined by, and reflects, its technical composition, is called the organic composition of capital.20
We assume, then, that the variable capital is the index of a definite quantity of laborers, or of labor-power, or a definite quantity of living labor set in motion. We saw in the preceding part that a change in the magnitude of the value of variable capital might eventually indicate nothing but a higher or lower price of the same mass of labor. But here, where the rate of surplus-value and the working day have been assumed to be constant, and the wages for a definite working time are given, this is out of the question. On the other hand, a difference in the magnitude of the constant capital may likewise be an index of a change in the mass of means of production set in motion by a definite quantity of labor-power. Still, it may also be due to a difference in value between the means of production set in motion in one sphere and those of another. Both points of view must be considered here.
Finally, the following essential facts must be taken into account:
Take it that 100 p.st. are the weekly wages of 100 laborers. Take it that the working hours are 60 per week. Take it, furthermore, that the rate of surplus-value is 100%. In that case, the laborers work 30 of the 60 hours for themselves, and 30 hours gratis for the capitalist. In fact, those 100 p.st. of wages represent only 30 working hours of those 100 laborers, or a total of 3,000 working hours, while the other 3,000 hours worked by the laborers are incorporated in the 100 p.st. of surplus-value, or as profit, pocketed by the capitalist. Although the wages of 100 p.st. do not express the value in which the weekly labor of those 100 laborers is materialised, still they indicate (since the length of the working day and the rate of surplus-value are given) that this capital set in motion 100 laborers for 6,000 working hours. The capital of 100 p.st. indicates this, first, because it indicates the number of laborers set in motion, since one pound sterling stands for one laborer per week, and 100 p.st. for 100 laborers per week; and in the second place, because every laborer set in motion performs twice the work for which his wages pay, at the given rate of surplus-value of 100%, so that one pound sterling, his wages, the expression of half a week of labor, actually set in motion one whole week's labor, and in the same way 100 p.st., although they pay only for 50 weeks of labor, set in motion 100 weeks of labor. There is, then, an essential difference between variable capital so far as its value, invested as a wages-capital, represents a certain sum of wages, a definite quantity of materialised labor, and variable capital so far as its value is a mere index of the quantity of living labor set in motion by it. This last-named labor is always greater than that incorporated in the variable capital, and is, therefore, represented by a greater value than that of the variable capital. This greater value is determined on one hand by the number of laborers set in motion by the variable capital, and on the other by the quantity of surplus-labor performed by them.
This mode of looking upon variable capital leads to the following conclusions:
When a capital invested in the sphere of production A expends only 100 in variable capital for each 700 of total capital, leaving 600 for constant capital, while a capital invested in the sphere of production B expends 600 for variable and only 100 for constant capital, then the capital of 700 in A will set in motion only 100 of labor-power, or, in terms of our previous assumption, 100 weeks of labor, or 6,000 hours of living labor, while the same amount of capital in B will set in motion 600 weeks of labor or 36,000 hours of living labor. The capital in A would then appropriate only 50 weeks of labor, or 3,000 hours of surplus-labor, while the same amount of capital in B would appropriate 300 weeks of labor, or 18,000 hours. The variable capital is the index, not only of the labor embodied in it, but also, when the rate of surplus-value is known, of the labor set in motion over and above that embodied in itself, in other words, of the surplus-labor. With the same intensity of exploitation, the profit in the first case would be 100/700, or 1/7, or 14 2/7%, and in the second case 600/700, or 6/7, or 85 5/7%, six times the rate of profit of the first. In this case, the profit itself would actually be six times that of A, 600 in B as against 100 in A, because the same capital set in motion six times the quantity of living labor, which, with the same degree of exploitation, means six times as much surplus-value and thus six times as much profit.
If the capital invested in A were not 700, but 7,000 p.st., while that invested in B were only 700 p.st., and the organic composition of both were to remain the same, then the capital in A would expend 1,000 p.st. of the 7,000 as variable capital, that is to say, it would employ 1,000 laborers per week at 60,000 hours of living labor, of which 30,000 would be surplus-labor. But yet each 700 p.st. of the capital in A would continue to set in motion only one-sixth of the surplus-labor of the capital in B, and produce only one-sixth of the profit of this capital. If we consider the rate of profit, then 1000/7000, or 100/700, or 14 2/7%, would be the rate of the capital in A, compared with 600/700, or 85 5/7%, of the capital in B. Taking equal amounts of capital for comparison, the rates of profit differ here, because the masses of surplus-value, and thus of profits, differ, although the rates of surplus-value are the same, owing to the different masses of living labor set in motion.
The same result follows, if the technical conditions are the same in both spheres of production, while the value of the elements of constant capital is greater or smaller in the one than in the other. Let us assume that both invest 100 p.st. in variable capital and employ 100 laborers per week, which set in motion the same quantity of machinery and raw materials. But let the last-named elements of production be more expensive in B than in A. For instance, let the 100 p.st. of variable capital in A set in motion 200 p.st. of constant capital, and in B 400 p.st. of constant capital. With the same rate of surplus-value, 100%, the surplus-value produced is in either case 100 p.st. Hence the profit is also 100 p.st. But the rate of profit in A is 100/200 c 100 v, or 1/3, or 33 1/3%, while in B it is 100/400 c 100 v, or 1/5, or 20%. In fact, if we select a certain aliquot part of the total capital from either side, we find that every 100 p.st. in B sets aside only 20 p.st., or one-fifth, for variable capital, while every 100 p.st. in A sets aside 33 1/3% p.st., or one-third, for this purpose. B produces less profit to each 100 p.st., because it sets in motion less living labor than A. The difference in the rates of profits resolves itself once more, in this case, into a difference of the masses of surplus-value, and thus masses of profit, produced per each 100 of capital invested.
The difference of this second example from the first is just this: The compensation between A and B, in the second case, would require only a change in the value of the constant capital of either A or B, provided the technical basis remained the same. But in the first case, the technical basis itself is different, and would have to be revolutionised in order to consummate a compensation.
The different organic composition of various capitals, then, is independent of their absolute magnitude. It is always but a question of what part of every 100 is variable and what part constant.
Capitals of different magnitude, calculated in percentages, or, what amounts to the same in this case, capitals of the same magnitude, working with the same working time and the same degree of exploitation, may produce considerably different amounts of surplus-value, and thus of profit, for the reason that a difference in the organic composition of capital in different spheres of production implies a difference in their variable parts, and thus a difference in the quantities of living labor set in motion by them, which implies a difference in the quantities of surplus-labor appropriated by them. And this surplus-labor is the substance of surplus-value and of profit. Equal portions of the total capital in the various spheres of production comprise the sources of unequal portions of surplus-value, and the only source of surplus-value is living labor. With the same degree of labor-exploitation the mass of labor set in motion by a capital of 100, and consequently the mass of surplus-value appropriated by it, depend on the magnitude of its variable component. If a capital, consisting of percentages of 90 c + 10 v, produced as much surplus-value, or profit, with the same degree of exploitation, as a capital consisting of percentages of 10 c + 90 v, then it would be as plain as daylight that the surplus-value, and value in general, must have an entirely different source than labor, and that political economy would then be without a rational basis. If we assume continually that one pound sterling stands for the weekly wages of a laborer working 60 hours, and that the rate of surplus-value is 100%, then it is evident that the total product in values which one laborer can supply in one week, is 2 p.st. Then 10 laborers cannot supply more than 20 p.st. And since 10 p.st. of the 20 reproduce the wages, those 10 laborers cannot produce any more surplus-value than 10 p.st. On the other hand the 90 laborers, whose total product is 180 p.st., and whose wages amount to 90 p.st., produce a surplus-value of 90 p.st. The rate of profit in the one case would be 10%, in the other 90%. If matters were different, then value and surplus-value would be something else than materialised labor. Seeing, then, that capitals in different spheres of production, calculated in percentages—or capitals of equal magnitude—are differently divided into variable and constant capital, so that they set in motion unequal quantities of living labor and produce different surplus-values, and profits, it follows that the rate of profit, which consists precisely of the calculation of the percentage of surplus-value on the total capital, must also differ.
Now, if capitals in different spheres of production, calculated in percentages, in other words, capitals of equal magnitude, produce unequal profits in different spheres of production, in consequence of their different organic composition, then it follows that the profits of unequal capitals in different spheres of production cannot be proportional to the magnitude of their respective capitals, or, in slightly different words, profits in different spheres of production are not proportional to the magnitude of the respective capitals invested in them. For if profits were to grow at the rate of the investment of capital, it would mean that the percentage of profits was the same, so that capitals of equal magnitude in different spheres of production would have equal rates of profit, in spite of their different organic composition. Only within the same sphere of production, in which the organic composition of capital is known, or in different spheres of production with the same organic composition of capitals, do the masses of profits stand in direct ratio to the masses of capitals invested. To say that the profits of capitals of different magnitude are proportional to their magnitudes is only another way of saying that capitals of equal magnitude yield equal profits, or that the rate of profits is the same for all capitals, whatever may be their organic composition and their magnitude.
These statements hold good on the assumption that the commodities are sold at their values. The value of a commodity is equal to the value of the constant capital contained in it, plus the value of the variable capital reproduced in it, plus the increment of this variable capital, which increment is the surplus-value. With the same rate of surplus-value, its mass evidently depends on the mass of the variable capital. The value of the product of a capital of 100 is in the one case 90 c + 10 v + 10 s, or 110, in the other 10 c + 90 v + 90 s, or 190. If the commodities are sold at their values, then the first product is sold at 110, of which 10 represent surplus-value, or unpaid labor; the second product is sold at 190, of which 90 represent surplus-value, or unpaid labor.
This is especially important when international rates of profit are compared with one another. Let us assume that the rate of surplus-value in some European country is 100%, so that the laborer works one-half of the working day for himself and the other half for his employer. Let us assume, furthermore, that the rate of profit in some Asiatic country is 25%, so that the laborer works four-fifths of the working day for himself, and one-fifth for his employer. Let the composition of the national capital in the European country be 84 c + 16 v, that of the national capital of the Asiatic country, where little machinery, etc., is used, and a given quantity of labor-power consumes relatively little raw material productively in a given time, 16 c + 84 v. Then we have the following calculation:
In the European country: Value of product 84 c + 16 v + 16 s, or 116; rate of profit 16/100, or 16%.
In the Asiatic country: Value of product 16 c + 84 v + 21 s, or 121; rate of profit 21/100, or 21%.
The rate of profit in the Asiatic country is higher by more than 25% than in the European country, although the rate of surplus-value is four times smaller in the former than in the latter. Men like Carey, Bastiat, and others, would come to the opposite conclusion.
By the way, different national rates of profit will generally be based on different national rates of surplus-value. But we compare in this chapter unequal rates of profit resting on the same rate of surplus-value.
Aside from differences of organic composition of capitals, which imply different masses of labor, and consequently, other circumstances remaining the same, of surplus-labor, which set in motion capitals of the same magnitude in different spheres of production, there is still another source for the inequality of rates of profit. This is the different length of the time of turn-over of capital in different spheres of production. We have seen in chapter IV that, other circumstances being the same, the rates of profits of capitals of the same organic composition are proportioned inversely as their times of turn-over. We have also seen that the same variable capital, if turned over in different periods of time, produces unequal masses of annual surplus-value. The difference of the times of turn-over, then, is another reason why capitals of the same magnitude in different spheres of production do not produce equal profits in equal times, and why the rates of profit in these different spheres differ.
On the other hand, the proportional composition of capitals as to fixed and circulating capital does not in itself affect the rate of profit. It can affect this rate only in the case that this difference in composition either coincides with a different proportion of the variable and constant parts so that the difference in the rate of profit is due to this difference in organic composition, and not to the different proportions between fixed and circulating capital; or, if the difference in the proportion of fixed and circulating capital is responsible for a difference in the time of turn-over, during which a certain profit is realised. If capitals are divided into fixed and circulating capital in different proportions, it will, of course, always have an influence on the time of turn-over and cause differences in it. But this does not imply that the time of turn-over, in which the same capitals realise certain profits, is different. For instance, A may have to convert the greater part of its product continually into raw materials, etc., while B may use the same machinery, etc., for a longer time, and need less raw material, but both A and B have a part of their capital engaged so long as they are producing; the one in raw materials, that is to say circulating capital, the other in machinery, etc., or fixed capital. The capitalist in A continually converts a portion of his capital from commodities into money, and this into raw materials, while the capitalist in B employs a portion of his capital for a longer time as an instrument of labor without any such conversions. If both of them employ the same amount of labor, they will sell masses of products of unequal value during the year, but both masses of products will contain the same amount of surplus-value, and their rates of profit, calculated on the entire capital invested, will be the same, although their proportional composition of fixed and circulating capital, and their times of turn-over, are different. Both capitals realise equal profits in equal times, although they are turned over in different periods of time.21 The difference in the time of turn-over has in itself no importance except so far as it affects the mass of surplus-value which may be appropriated and realized by the same capital in a certain time. Seeing that a different distribution of the fixed and circulating capital of A and B does not necessarily imply a different time of turn-over, which would in its turn imply a different rate of profit, it is evident, if there is such a difference in the rates of profit of A and B, that it is not due to a difference in the proportions of fixed and circulating capital as such, but rather to the fact that these different proportions indicate an inequality in the times of turn-over affecting the rates of profit.
It follows, then, that a difference in the composition of capitals in various lines of production, referring to their fixed and circulating portions, has in itself no bearing on the rate of profit, since it is the proportion between the constant and variable capital which decides this question, and since the value of the constant capital, and its relative magnitude as compared to that of the variable, is quite independent of the fixed or circulating nature of its components. But it will be found—and this is one of the causes of wrong conclusions—that whenever fixed capital is considerably developed, it is but an expression of the fact that production is carried on at a large scale, so that the constant capital far outweighs the variable, or the living labor-power employed is trifling compared to the mass of the means of production set in motion by it.
We have demonstrated, that different lines of industry may have different rates of profit, corresponding to differences in the organic composition of capitals, and, within the limits indicated, also corresponding to different times of turn-over; the law (as a general tendency) that profits are proportioned as the magnitudes of the capitals, or that capitals of equal magnitude yield equal profits in equal times, applies only to capitals of the same organic composition, with the same rate of surplus-value, and the same time of turn-over. And these statements hold good on the assumption, which has been the basis of all our analyses so far, namely that the commodities are sold at their values. On the other hand there is no doubt that, aside from unessential, accidental, and mutually compensating distinctions, a difference in the average rate of profit of the various lines of industry does not exist in reality, and could not exist without abolishing the entire system of capitalist production. It would seem, then, as though the theory of value were irreconcilable at this point with the actual process, irreconcilable with the real phenomena of production, so that we should have to give up the attempt to understand these phenomena.
It follows from the first part of this volume that the cost-prices are the same for the products of different spheres of production, in which equal portions of capital have been invested for purposes of production, regardless of the organic composition of such capitals. The cost-price does not show the distinction between variable and constant capital to the capitalist. A commodity for which he must advance 100 p.st. in production cost him the same amount, whether he invests 90 c + 10 v, or 10 c + 90 v. He always spends 100 p.st. for it, no more, no less. The cost-prices are the same for investments of the same amounts of capital in different spheres, no matter how much the produced values and surplus-values may differ. The equality of cost-prices is the basis for the competition of the invested capitals, by which an average rate of profit is brought about.
[20.] The above is briefly developed in the third edition of volume I, in the beginning of chapter XXV. Since the two first editions did not contain this passage, it was so much more necessary to repeat it at this place.—F. E.
[21.] It follows from chapter IV that the above statement is correct only in the ease that the capitals of A and B are differently composed so far as their values are concerned, but that the percentages of their variable capitals are proportioned as their times of turn-over, or inversely as their numbers of turn-over. Let capital A have the following percentages of composition: 20 c fixed and 70 c circulating, a total of 90 c, so that the total capital is 90 c + 10 v, or 100. At a rate or surplus value of 100% the 10 v produce in one turn-over 10 s, making the rate of profit for one turn-over 10%. Let capital B have the composition 60 C fixed and 20c circulating, so that we have 80 c + 20 v, or 100. The 20 v produce in one turn-over, at the above rate of surplus-value, 20 s, making the rate of profit for one turn-over 20%, which is double that of A. But if A is turned over twice per year, and B only once, then 2 × 10 also make 20 per year, and the annual rate of profit is the same for both, namely 20%.—F. E.