Ernest Mandel

The Labor Theory of Value and Monopoly Capitalism

(March 1967)

From International Socialist Review, Vol. 28 No. 4, July–August 1967, pp. 29–42.
Transcribed & marked up by Einde O’Callaghan for the Marxists’ Internet Archive.

David Horowitz’s challenge offers a welcome occasion to test the validity of the labor theory of value as an instrument for analyzing and explaining the functioning of contemporary monopoly capitalism. At the same time, it enables us to deepen both our appreciation and our criticism of Baran and Sweezy’s book.

The concept of “surplus” is today widely used by anthropologists and students of primitive societies in its most elementary sense: that part of social production which exceeds the immediate consumption needs of society. Since primitive society, in which “surplus” first appears, is a classless society, consumption by producers (i.e. restoration of the producer’s labor power and reproduction of the given number of producers), and social consumption are largely equivalent. In that sense, “economic surplus” covers the same socio-economic concept as the Marxist concept of “surplus product,” that part of social product over and above “necessary product.”

In all but the most backward of primitive societies, “necessary product” has, however, still another function to fulfill in order to reproduce society’s productive capacities. It also has to guarantee equivalent substitution of all means of production used up in the process of social production. The more a society develops, the more important this second function becomes.

In a capitalist society, the necessary product includes constant plus variable capital (c+v), that is, reproduction of dead and living labor necessary to restart production at the same level as during the previous cycle. This ensures what Marx calls “simple reproduction.” The surplus product represents the difference between the value of the social product, c+v+s, and the value of the necessary product. It is equal to s, surplus value. In fact, surplus value is simply the specific form under which surplus product is appropriated in the capitalist economy.

Baran and Sweezy do not dispute this definition. They actually repeat it on pp. 8–10 of their book. They add that, if they prefer the term “surplus” to the term “surplus value,” it is only because “most people familiar with Marxian economic theory” – contrary to Marx himself – identify surplus value “as equal to the sum of profits plus interest plus rent.” (p. 10) In that sense, they seem to start from identical definitions as Marx, and David Horowitz seems wrong in his assumption that they have abandoned the labor theory of value.

However, as the authors develop their arguments, it becomes more and more apparent that they substantially depart from this initial definition. The impression is created that they have abandoned the labor theory of value. Whether or not this is the intention is for Paul Sweezy himself to clarify.

Depreciation Allowances

In evaluating “surplus”, Baran and Sweezy lay particular emphasis on the question of depreciation allowances. They allege that “excess depreciation allowances” (pp. 99–100 and 372–378) constitute “surplus” and they get entangled in various calculations of this factor. But they do not pose the question the way it should be posed from a Marxist point of view: What is the value of the fixed capital actually used up in the process of production?

Several arguments plead against their and Joseph D. Phillips’ thesis about “excessive depreciation allowances.” The use of a percentage of gross investment similar to that of the Soviet Union is obviously untenable, because the rate of net investment in the Soviet Union is greatly superior to that of the US economy. Excessive depreciation allowances are not the only form of tax evasion. Profits are even better hidden by charging expenses for capital renewal to current operations; this is widely practiced by big business.

And last but not least, in order to have a correct estimate of real fixed capital values used up in current production, one must start by having a correct estimate of real capital value. This is usually even more underestimated than are current profits. And as the accelerated rate of technological expansion, which Baran and Sweezy acknowledge, tends to reduce the lifespan of plant and machinery, the value of annually used up fixed capital is very large indeed, probably larger and not smaller than official depreciation allowances contend.

Consequently, one should subtract, not add, depreciation allowances from gross receipts in order to establish corporate “surplus.” And this calculation strongly reduces Phillips’ statistical demonstration of the “tendency of the surplus to rise.” Without any part of depreciation allowances, the surplus, as it is defined by the authors, declines to 43.3 percent of the GNP in 1929, 49.4 percent in 1949, 49.2 percent in 1959 and 49.8 percent in 1961.

On the other hand, if one defines the “surplus” in the way the authors initially do as “the difference between what a society produces and the costs of producing it” (p. 9) and eliminates interest and rent from the “costs of production,” one is following the labor theory of value: The “surplus,” or “surplus value,” is then the difference between the value of the social product and the value used up (in the form of constant and variable capital) in producing that product

But this classical Marxist definition is inconsistent with the more sloppy definition of “surplus” as “the difference between aggregate net output and the aggregate real wages of productive workers” (p. 125). This definition uses the labor theory of value in the second part but denies it in the first. “Aggregate net output,” as defined by current bourgeois calculations, includes redistribution of surplus value and many incomes which are simply the result of inflation (e.g. payments to armed forces, veterans or state functionaries financed through budgetary deficits, etc.). Our authors thus shift back and forth between value and “aggregate demand” calculations. Horowitz is right in assuming that they try to combine Marx and Keynes. He is wrong in assuming that this contributes to a clearer understanding of the “laws of motion” of monopoly capital.

Horowitz bases his rejection of the labor theory of value on an old article written by Oskar Lange in the thirties. [1] This article contains what amounts in our opinion to several misconceptions both about Marx’s economic theory in general and his labor theory of value in particular. This is not the place to answer Lange’s arguments extensively. But we would like to take up one of his basic points, which has a direct bearing on our critique of Monopoly Capitalism.

Lange’s assumption that Marx’s labor theory of value is “nothing but a static theory of general economic equilibrium” (op. cit., p. 194) seems to us utterly mistaken. One could make this point about the special application of this theory to conditions of simple commodity production. But it is completely wrong to maintain it about the theory of value as applied to capitalism. And it is to this application, and not to the special case of static equilibrium in a pre-capitalist society, that Marx after all devotes nearly all his economic studies, from 1844 until his death.

In order to understand the dynamic nature of the labor theory of value as used by Marx, it is sufficient to understand Marx’s purpose in perfecting Ricardo’s labor theory of value by working out his theory of surplus value. What he wanted to explain was the essentially dynamic problem of capital accumulation: How the exchange of “equal values” between the worker and the capitalist leads to constant enrichment of the capitalist. It is unnecessary to develop the way in which Marx solved this problem at length: the distinction between labor and labor power; the discovery that the worker does not sell his “labor” but his labor power; the distinction between exchange-value of labor power and its use-value for the capitalist (which is precisely to produce more value than its own exchange value) etc.

The labor theory of value thus corrected by Marx introduces two dynamic elements into what Lange mistakenly calls a “theory of general economic equilibrium.” By its very nature, it implies a process of economic growth built into the model. And it indicates the dual processes which provide the rationale of capital accumulation: competition between capitalists; and competition between capitalists and workers. [2]

For the same reason, it is inappropriate to speak of Marx’s model as a model of “general economic equilibrium.” In reality it is a model which presents a dialectical unity of equilibrium and disequilibrium, the one leading necessarily to the other. This is the reason why it is futile to try to “discover” Marx’s theory of crisis in the famous reproduction schemes of Vol. II of Capital, because these schemes actually make abstraction of “competition between capitalists.” And any study of business cycles must necessarily come under that heading, according to Marx himself. [3]

All the “laws of motion” of the capitalist mode of production arise out of the process of capital accumulation, based upon and explained by the labor theory of value as perfected by Marx. This is especially true for the law of centralization and concentration of capital and the law of increasing organic composition of capital, both of which result from competition between capitalists (“the big fish eat the small fish”) and from competition between capital and labor (the necessity to increase the production of relative surplus value, i.e. to increase the productivity of labor).

Indeed, the attempt to divorce the activities of capital accumulation from these two rational explanations offered by Marx, or even to divorce one from the other, must lead to the discovery of some mystic “accumulation urge” beyond the realm of scientific investigation. Authors embarked upon this perilous path generally end up with the kind of tautological explanations like “capitalists accumulate because (!) it is their mission – or function, or role, or goal – to accumulate.” One is reminded of Molière’s immortal definition: “Opium causes sleep because it has dormative qualities.”

Competition between Capitalists

Baran and Sweezy strongly contend that to accumulate capital is still “Moses and the Prophets” for today’s giant corporations. With this we fully agree. But they do not give any exhaustive explanation of the reasons why this is so. On the contrary, they do not incorporate at all the basic competition between capitalists and workers into their analysis; it appears only in the final chapters treating the current displacement of workers by automation. As for competition between capitalists, they waver between erroneous positions: On one hand they identify competition with “price competition”; on the other hand, denying that “price competition” prevails, they seem to say that competition does exist, but in a system which is “radically different” from Marx’s model.

A good deal of clarification is in order. It is true that in Volume III of Capital, when Marx develops his theory of the formation of “production prices” (equalization of the rate of profit as a result of flux and influx of capital between different branches of industry), prices rising or falling are the mechanism through which the profit equalization process takes place. But a moment’s thought shows that this is only a subordinate mechanism, not the crux of the matter. If instead of price cutting, aggressive advertisement is used as the vehicle for appropriating a greater share of the market, the whole reasoning stands exactly as in Volume III of Capital. The important point is that one firm realizes a substantially higher rate of profit, and that this higher rate then attracts capital of other firms (say: other monopolists) to the same field, until equalization occurs. To say that the monopolists try to avoid excessive risks means precisely in this framework that they will avoid excessive deviations from the “normal” monopolistic super-profits, because such deviations would unavoidably attract other capital.

The crucial weakness of Monopoly Capitalism, however, is the authors’ failure to deal with the exploitation of labor by capital and their consequent omission of the capitalists’ need to increase relative surplus value. When speaking about poverty in the United States, Baran and Sweezy correctly point out (p. 286) that the total disappearance of the reserve army of labor during the second world war led to an “improvement of living standards of poor people ... nothing short of dramatic.” This in turn led to an upward pressure on real wages, exemplified in the great postwar strike wave. They continue (p.287) to state that in the fifties “unemployment crept steadily upward, and the character of the new technologies of the postwar period sharply accentuated the disadvantages of unskilled and semi-skilled workers.” It seems to us that the “new technologies of the postwar period” created that upward tendency of unemployment, i.e. that the US economy then entered the most dramatic period of “displacement of labor by machines” in its whole history.

There can be no further doubt that this move was successful beyond all expectations, that for more than 10 years US real wages nearly stagnated as compared with the rapid increases in all other imperialist countries, and that the big increase in profits during that period was a result of the fantastic increases in relative surplus value so produced.

By leaving out’ of their analysis of monopoly capital the continuous struggle of the capitalist class to maintain and increase the rate of exploitation of the working class, Baran and Sweezy put their whole economic concept of the present functioning of the capitalist system outside the realm of contending social forces, i.e. outside the realm of the class struggle. It is not surprising, therefore, that they end by denying any validity to the anti-capitalist potential of the American working class; they imply this negation already in the premises of the argumentation: We are faced with a classical petitio principis.

As for the competition between capitalists, as said before, Baran and Sweezy’s argumentation is vague, to say the least. They recognize the need of the corporations to reduce costs. They recognize the need of the corporations to increase profits for goals of increased capital accumulation. They also recognize the fiercely competitive nature of the “monopolists’ jungle,” not to speak of the fierce competition between the monopolists and the non-monopolized sectors of the economy. Yet they shy away from the obvious conclusion: That the main rational explanation of that accumulation remains competition, exactly as it was in Marx’s model. And this leaves a gaping void at the center of their analysis.

Value Analysis

The reason for this weakness is easy to discover. The labor theory of value implies that, in terms of value, the total mass of surplus value to be distributed every year is a given quantity. It depends on the value of variable capital and on the rate of surplus value. Price competition cannot change that given quantity (except when it influences the division of the newly created income between workers and capitalists, i.e. depresses or increases real wages, and thereby increases or depresses the rate of surplus value). Once this simple basic truth is grasped, one understands that the displacement of free competition by monopolies does not basically alter the problem in value terms. It means that the distribution of the given quantity of surplus value is changed, in favor of the monopolists and at the expense of the non-monopolized sectors. It can mean (but this must then be demonstrated) that the general rate of surplus value is increased. But it does not modify in any sense the basic relationships which explain the creation of surplus value.

By jettisoning the field of value production for the field of monetary aggregate demand, Baran and Sweezy obscure the simple basic relationships. They speak loosely about “the surplus being absorbed” when idle men and machines are put to work. But what has not been produced cannot be absorbed. When machinery is idle, we do not have an “unabsorbed surplus,” i.e. surplus value not spent, or unsold commodities. We have unused capital, which is something quite different. And when “idle men and machinery” are put to work, “surplus” (surplus value) is not being “absorbed” but is being produced, i.e. its amount grows, as a result of an increase in variable capital.

Abandoning the firm ground of value calculation for the slippery field of “aggregate demand,” Baran and Sweezy often show an amazing inability to distinguish between the micro-economic behavior of the firm and the macro-economic result of such generalized behavior. They correctly state that the modern monopolist corporation tends to “maximize profits” at least as much as its competitive ancestor did. But they seem to forget that the average rate of profit is precisely the macro-economic result of such behavior in the individual firms. This follows immediately from the assumption that surplus value, which can be distributed among different firms, is a given limited quantity each year.

If any monopolist corporation succeeds in gaining an excessive share of total surplus value, other corporations quickly move into the same line of business. The examples of aluminum, electronic computers, duplicating machinery, petrochemical products, just to note only a few “growth” industries during the last three decades, clearly confirm that this actually happens. Thus we arrive at the conclusion that under monopoly capitalism exactly as under the “competitive model,” profit maximization of individual firms leads to the tendency toward equalization of the rate of profit. The only distinction one has to make is that under monopoly capitalism, two different average rates tend to evolve: one for the monopolist sector of industries, and another for the competitive sector. [4]

We can therefore conclude that Baran and Sweezy have been unable either to prove that Marx’s model was based on some specific feature linked to price competition, or that capital accumulation under monopoly capitalism unfolds along lines which are qualitatively different from those of “competitive capitalism.” Under monopoly capitalism as under “competitive capitalism” the two basic forces explaining capital accumulation remain competition between capitalists (for appropriating bigger shares of surplus value) and competition between capitalists and workers (for increasing the rate of surplus value).

In Marx’s model, the tendency of the average rate of profit to decline arises from two causes. First, since human labor alone produces surplus value, only one part of capital, variable capital, corresponds to the production of surplus value. If there is a tendency for variable capital to be a smaller part of total capital, there will be a strong pull for the relation s/(c+v) to decline. Second, this pull could be neutralized only if at the same time the rate of surplus value s/v would increase. But historically, it is very unlikely that the increase in the rate of surplus value occurs in the same proportion as the rise in the organic composition of capital. And in the long run, this is impossible. Because, whereas the organic composition of capital can grow infinitely (the limit being complete automation, i.e. complete expulsion of living labor from the process of production), the rate of surplus value cannot grow infinitely, because this would imply that wages of workers actively engaged in production fall toward zero.

Baran and Sweezy contend that the tendency toward a decline of the rate of profit is somehow linked to Marx’s “competitive model” and no longer operates under the reign of monopoly capital. But they do not make the slightest attempt to examine the two basic ratios from which the falling rate of profit results: the organic composition of capital and the rate of surplus value.

In relation to the organic composition of capital, the authors of Monopoly Capitalism do not make any overall assessment. On the one hand, they say that “under monopoly capitalism the rate at which new techniques will supersede old techniques will be slower than traditional economic theory would lead us to suppose ... Technological progress tends to determine the form which investment takes at any given time rather than its amount” (pp. 95, 97). But a few pages further, they write, “the decade 1952-1962 was one of rapid and probably accelerating technological progress” (p. 102). The figures they quote bear out the thesis that investment in fixed capital rises quicker than wages. In 1953, expenses for research and development and outlays on plant and equipment of non-financial corporations amounted to $27.4 billion, while they amounted to $44 billion in 1962 (and have since risen to a figure double that of 1953!). Wages paid to labor engaged by the same corporations have certainly not risen by 100 percent between 1953 and 1966! [5]

Technological Advance

First Baran and Sweezy contend that the only technological revolutions which really caused tremendous spurts in productive investments were related to the steam engine, the railways and automobile. But later (pp. 267–8) they admit that the technological revolution linked with mechanization, automation and cybernation has reduced the number of unskilled workers in the American economy from 13 million in 1950 to less than 4 million in 1962, and that, according to many authorities, this technological revolution is still in its early stages! Surely, a displacement of workers by machines at what Baran and Sweezy call this “fantastic rate” expresses a tendency toward an increase in the organic composition of capital, does it not?

There is no doubt in our mind that starting with the late fifties (i.e. with the upward shift in the unemployment rate) there has been a significant increase in the rate of surplus value, which crystallized in the “profit explosion” of more than 50 percent between 1960 and 1965. But that this increase can continue to displace more and more productive workers, who alone create surplus value, at an equivalent rate with the rise in the organic composition of capital is doubtful. Automation will continue to displace more and more productive workers; the wages of the productive workers may well represent a gradually declining part of the new income generated in industry; but they will certainly not fall rapidly enough to offset the rising organic composition of capital. So there is no reason to assume that the tendency of the rate of profit to decline will be historically reversed.

There is striking proof of this which interestingly enough is quoted by Baran and Sweezy without drawing the necessary conclusions. On pp. 196–7 they indicate that between 1946 and 1963, direct foreign investments of American corporations increased more than five fold, because the rate of return on investment abroad was much higher than in the US. Obviously, the organic composition of capital is lower, and the degree of market control by monopoly capital is less in these foreign countries, than in the US. Isn’t it reasonable then to conclude that the more they become “Americanized,” the more the rate of profit will tend to fall? And in the US new technological progress will result in a new and significant decline of the rate of profit compared to its present level.

Baran and Sweezy’s insistence on a continuous rise of the “surplus” is based upon very simple reasoning: Under monopoly capitalism, costs decline, prices rise together with profits, therefore the surplus must increase (p. 79). But here again price calculations instead of value analysis obscure the macro-economic problems involved.

“Under monopoly capitalism employers can and do pass on higher labor costs in the form of higher prices,” write Baran and Sweezy (p. 77). But a moment’s thought shows that such sloppy statements, useful as they might be in agitation, do not mean very much in terms of real economic relations. For if the employers “pass on” identical higher labor costs in the same way to all consumers, all commodity prices rise in the same proportion, and far from “surplus” having increased, relations between wages and surplus value, and between the parts of total surplus allotted to each firm, remain exactly as they were before. If this “passing on” can be done only by the monopolists, there is a big probability that real wages will actually have risen, and that the biggest gains of the monopolists will have been made at the expense of the non-monopolist sectors of the capitalist class who will have been unable to raise their prices in the same proportion. In that case again, “surplus” has not been increased but only redistributed and probably even slightly reduced at the expenses of one part of the capitalist class. And if prices of consumer goods actually rise more than wages, then there is a decrease in the real wage and indeed a rise of the “surplus” – but not through any special “new” device, but through the age-old method of capital: lowering wages.

The origin of Baran and Sweezy’s theory about the tendency of the “surplus” to rise is easy to determine. It is, on the one hand, an incorrect generalization from a temporary occurrence: the sharp rise in capitalist profits during the late fifties and the first half of the sixties. It is, on the other hand, a result of a tendentious use of the term “surplus,” even to the point of making it synonymous with “aggregate demand.” Such reasoning simply eliminates the problem of inflation and includes a number of cases of counting the same income two or three times.

Here we can see clearly, that contrary to David Horowitz’s contention, Baran and Sweezy’s attempt to combine Marx with Keynes is precisely one of the main reasons they are led astray. Marx makes it crystal clear that on the basis of the labor theory of value, all income generated in capitalist society (except for the income of small owners of the means of production who do not exploit wage labor) can only have two sources: either variable capital or surplus value. When capitalists use their surplus value to buy directly the individual services of housemaids, private teachers, clergymen, etc., they do not create new income. They simple distribute part of surplus value as revenue. It is unimportant how many times this surplus value circulates in a year’s time. It is always the same surplus value which is redistributed. Mayors of small towns, whose industries have disappeared know this through sad experience: Eliminate the original wages and surplus value, and all service income disappears as if by magic! But if you calculate “aggregate demand” in the way in which it is at present defined in the United States, you get the impression that income of all service industries is simply added to profits of industrial firms and you then easily arrive at calculations in which part of the “surplus” (defined in this sloppy way) is two or three times as big as it really is. [6]

Sales Effort

A good example of this is that of the problem of increased sales effort. Sales costs do not add to value produced, but are an example of what Marx called “the expenses of circulation ... paid out of a given quantity of surplus value.” Baran and Sweezy actually quote this passage from Capital on p. 112 of their book. Yet they not only treat increased sales effort as a means of “surplus absorption” (surplus value absorbing surplus value!). They even see therein a means of increasing profit for the capitalists, because part of the initial outlay will be “paid by the workers” through increased consumer prices! They don’t seem to understand that the whole outlay was paid by the capitalists in the first place, and that you can’t add it three times: first as surplus value (capitalist profits); then as advertisement outlays (part of profits used for sales efforts); and finally as additional capitalist profits (part of the sales effort recovered from the workers’ wages).

Here again, the source of Baran and Sweezy’s confusion is easy to discover. For the “sales effort” they speak of (which is not part of the distribution costs treated by Marx) is in reality financed out of capital, and not out of current surplus value. Inasmuch as monopoly capitalism is characterized by huge amounts of surplus capital, “sales efforts” (in the same way as the “service industries”) offer a welcome outlet for this capital. As supplementary workers are employed, and as they buy commodities with their wages and salaries, the “increased sales effort can trigger indirectly increasing “realization” of surplus value, out of an increased capital outlay. But to add this capital (generated from yesteryear’s surplus value) to this year’s surplus value is an evident error in calculation, as far as value calculation is concerned.

The valid and important kernel of truth contained in Baran and Sweezy’s book is their insistence on idle and unused capital. This indeed is a specific feature of monopoly capitalism, arising precisely out of the slowing down of price competition and the concentration of capital in the monopolized sectors. It increases precisely inasmuch as the average rate of profit tends to be higher in the monopolized sectors than in the non-monopolized sectors of the economy. And it poses the crucial question of surplus capital disposition which Baran and Sweezy have elucidated in many important fields. The monopolists receive indeed higher profits – but they are unable to reinvest all of them without endangering this very rate of super profits!

This is, be it said in passing, the main reason which compels monopoly capital to invest more and more capital in armaments and – together with an attempt to counteract the falling rate of profit – one of the main reasons which explains the growing volume of capital exports by US imperialism. Without these two elements added to the analysis, US imperialism’s intervention in both world wars, and its present attempt to “make the world safe for capitalism,” cannot be explained in a sufficiently thorough-going manner, as being inherent in the system.

But adding surplus capital to surplus product doesn’t clarify the issue. Had the authors applied the labor theory of value to this question, they would immediately have noted both the relations and the differences between the two crucial problems aging monopoly capital faces: investment of surplus capital and increasing difficulties in realization of surplus value.

In an essentially underdeveloped economy this difference is negligible. There the social surplus product does not consist of industrial goods which need to be sold; at the same time, the ruling class is not essentially geared to productive capital investment. Social surplus product takes essentially the form of land rent, income of the compradore bourgeoisie, and profits of the foreign trusts, none of which are industrially invested in the country. To lump these incomes together, call them “surplus” and show that the mobilization of this surplus for the goal of productive investment through planning and industrialization would make possible a rapid process of economic growth, is entirely legitimate. That is why the concept of “surplus” is operative when Baran applies it to the problem of underdeveloped countries. [7]

But in an industrialized imperialist country, the situation is entirely different. The social surplus product essentially takes the form of industrial goods which have to be sold before surplus value can be actually realized. This process meets with increasing difficulties. On the other hand, under conditions of monopoly capitalism, there are great reserves of capital on hand – as a result of the past realization of surplus value – which find more and more difficulties for profitable investment, and plants corresponding to invested capital are generally run below the optimum level of capacity. These twin problems of surplus value realization and of surplus capital investment both demonstrate the irrationality of the system. And neither can be lumped together in a new category of “surplus.”

They are even more obscured when one passes from value production and realization analysis to aggregate demand analysis, and thereby adds to surplus value the vast amount of purchasing power of inflationary origin injected into the system since the second world war. Baran and Sweezy themselves state that the post-1945 boom in the US is to be explained by “a second great wave of automobilization and suburbanization, fueled by a tremendous growth of mortgage and consumer debt” (p. 244). If one adds the not less tremendous growth of public debt since 1940, one gets a picture not of an “increased surplus” but of increased difficulties of surplus-value realization, which sooner or later must bring the whole topsy-turvy pyramid down. Surely Sweezy will agree with us that inflationary purchasing power injected into the system can, from a point of view of value production and distribution, only do one of two things in the long run: either redistribute surplus value in favor of certain sections of the capitalist class and at the expense of others, or increase surplus value at the expense of wages. And this second “solution” would only exacerbate the problem of surplus-value realization.

But here we arrive again at the problem of inflation in the US, and its repercussions both on the class struggle inside that country and on the international monetary system. These questions need further elucidation. They are certainly one of the main problems posed by monopoly capitalism, as both bourgeois and Marxist economists know only too well.

March 31, 1967


In my article on Baran and Sweezy’s book which appeared in the January–February 1967 ISR an unfortunate typographical error occurred. On p. 59 it is printed: “Equally to take sales costs en bloc as part of the surplus is to indicate that this notion encompasses something more than surplus value. Evidently, the part of sales cost which is just reproduction of capital invested in the service sector is not part of social capital.” The last sentence should read: “... is part of social capital.”


1. Oscar Lange: Marxian Economics and Modern Economic Theory, Review of Economic Studies, June 1935.

2. Incidentally, in the above named article, Lange completely eliminates competition between capitalists and assumes that technical progress proceeds independently from such competition, thereby introducing an element of built-in evolution. This is a serious misinterpretation of Marxism.

3. In his general plan for Capital, Marx explicitly excludes crisis from the part entitled “capital in general,” and includes it in the part called “different capitals,” i.e. competition.

4. In my Traité d’Economie Marxiste, Vol. II, pp. 46–51, I have tried to offer some statistical proof of this proposition. It is clear that Baran and Sweezy seriously underestimate the amount of competition occurring under monopoly capitalism, both nationally and internationally. When they approvingly quote Galbraith’s list of commodities (p. 74), which in the next generation will still be bought from the same corporations as several decades ago, they have to leave out of this list such important commodities as coal, airplanes, computers, plastics and other petrochemical products, TV sets, office machines and even electrical power or steel, from which the statement is either partially or totally incorrect.

5. At one point of their reasoning, in a very abstract way it is true, Baran and Sweezy seem to imply that the rise in organic composition of capital is impossible. They write on p. 81 that it is “nonsensical” to conceive of capitalist production as implying that “a larger and larger volume of producer goods would have to be turned out for the sole purpose of producing a still larger and larger volume of producer goods in the future. Consumption would be a diminishing proportion of output, and the growth of the capital stock would have no relation to the actual or potential expansion of consumption.” Two words are the source of confusion here: the word “sole purpose” and the word “no relation.”

It seems to us proved that more and more producer goods are being turned out for the purpose of producing still more and more producer goods, although this is of course not their sole purpose. Their purpose is also to produce at cheaper costs more consumer goods. And it seems also proved that consumption is a diminishing proportion of output, although this does not imply there is no relation at all between capital stock and the final output of consumer goods. That producer durables are a growing percentage of current output is born out by US historical statistics. And to deny this possibility is not only to deny a rise of the organic composition of capital under conditions of monopoly capitalism; it means to deny such a rise for capitalism of the 19th century as well!

6. Capital invested in trade and several service industries, as well as in transportation of individuals, does not lead to the creation of additional surplus-value through the hiring of labor; it only participates in the distribution of surplus-value created by labor in the productive sectors of the economy. But in order to calculate the total sum of surplus-value produced, one cannot just add profits of all firms. Some are clearly the result not of distribution, but of re-distribution of surplus-value, e.g. when they sell services in exchange of profits from other firms (to quote only one example: the services of brokerage firms called upon to invest newly realized profits).

7. Cf. Political Economy of Growth, Paul Baran, Monthly Review Press, 1959.

Last updated on 29.12.2011