Mario Cocoy The Falling Rate of Profit and the Theory of Accumulation A Reply to Paul Sweezy Paul Sweezy’s article, ‘‘Some Problems in the Theory of Capital Ac- cumulation,”’' starts from a critique of theses largely held in common by David Yaffe and myself.? The interpretation of the law of the tendential fall of the rate of profit that Sweezy criticizes was not the main point of my article, which was concerned much more with the effects of unproductive expenditure on capital accumulation. However, Sweezy interprets my thinking quite correctly when he counts me among those who regard the law of the tendency of the rate of profit to fall ‘‘as the pivot around which the whole Marxian theory of accumula- tion and crises revolves.’’> Iam therefore glad that Sweezy has emphasized just this aspect, and would like to use the opportunity to state precisely my views on the subject. It is clear that the consequences of different interpretations of this important point are not limited to the theory of accumulation; in my opinion, they involve differences in the evaluation of the laws of motion of capitalism and the role of the state. And this obviously has political consequences, which Sweezy briefly takes up in his last note, with his critique of the ‘‘reformist illusions’’ of the Keynesians. Although these political aspects are of great significance, we must begin by concentrat- ing on the theory of accumulation. I It cannot be claimed that ‘‘the presence or absence of monopoly makes no difference to the accumulation process,’’* a view that Sweezy attri- butes to “‘many of today’s younger Marxist economists.’’ Since the market and competition represent the medium through which the law- Mario Cogoy, “Baise du taux de profit et théorie de l’accumulation,”” pp. 1231-55 in Les Temps Modernes 31: 330 (January 1974). Translated with permission. 54SUMMER 1987 55 fulness of capitalist production imposes itself on the individual capitals, the form of the market has determining importance for the form in which this lawfulness asserts itself. What is at issue is thus only the level of ab- straction at which we can deal with the problem of monopoly. Marx deals with the law of the tendency of the rate of profit to fall and the laws of ac- cumulation at the level of capital in general. On this level neither the distribution of surplus value between different capitalists, nor the divi- sion of surplus value into profit, interest, rent, etc. are of importance, be- cause on this level we examine the relation of total surplus value to total capital. For this reason, the attempt of Baran and Sweezy to replace the concept of ‘‘surplus value’’ by the concept of ‘‘surplus,’’ because of the monopolistic structure of the market, was not very convincing [Wle prefer the concept ‘‘surplus’’ to the traditional Marxian ‘‘surplus value,’’ since the latter is probably identified in the minds of most people familiar with Marxian economic theory as equal to the sum of profits + interest + rent. It is true that Marx demonstrates—in scattered passages of Capital and Theories of Surplus Value—that surplus value also com- prises other items such as the revenues of state and church, the expenses of transforming commodities into money, and the wages of unproductive workers. In general, however, he treated these as secondary factors and excluded them from his basic theoretical schema. It is our contention that under monopoly capitalism this procedure is no longer justified, and we hope that a change in terminology will help to effect the needed shift in theoretical position.* Everyone familiar with Marxian theory knows that for Marx, on the contrary, ‘‘the best points in my book are . . . the treatment of surplus value independently of its particular forms as profit, interest, ground rent, etc. The treatment of the particular forms by classical politi- cal economy, which always mixes them up with the general form, is a regular hash.*’* Baran and Sweezy’s Monopoly Capital was equally *‘a regular hash’’ and it was difficult to understand the reasons for the “needed shift in theoretical position.’’ Now, however, Sweezy assures us that it never occurred to him “‘to call in question, let alone reject the labor theory of value.’’? He adheres, therefore, to the concept of surplus value but argues that the process of monopolization produces crucial changes in the accumulation process. The reason for this is that monopoly leads to a redistribution of total value in favor of capital and so to a tendency for the rate of surplus value to rise. Why had Marx not taken monopoly into account in his theory of accumulation, instead of56 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY leaving this great theoretical leap forward to Sweezy? The answer is characteristic of Sweezy’s method: because in Marx’s time monopoly was the exception, while today it is the rule. It is with this indifference to theory that Baran and Sweezy could claim, in Monopoly Capital, that under competitive capitalism the law of the tendency of the rate of profit to fall applied, while under monopoly capitalism that of the rising surplus applies.* Marx, however, formulated his theory of accumulation independent- ly of monopoly not simply because monopolies were the exception in his day, but on theoretical grounds. For Marx, the production of sur- plus value is the characteristic form of the reproduction of surplus value for developed capital. The production of relative surplus value implies the continuing revolutionization of the methods of production. This has the result that a given quantity of labor time is embodied in an ever growing mass of use values, or-—what amounts to the same thing—that the value of the individual commodity continually falls. Since therefore the value of the commodities consumed by workers falls, the value of the commodity labor power also falls in the course of accumulation. In this way the paid portion of the working day’s labor continuously falls relative to the unpaid portion, although the quantity of use values that the worker can buy with this smaller value (the real wage) may remain unchanged or even increase. An increase and improvement in the use values consumed by work- ers can partially check the long-term tendency of the value of labor power to fall. But the cultural factors that lead to an increase of the real wage do not completely overcome the tendential fall in the value of labor power, because the increase in the real wage is smaller as a rule than the growth in labor productivity. Another factor that can check the fall in the value of labor power is the increase in the intensity of labor. However, this can only operate within certain limits. The fall in the value of labor power can therefore be checked by all these factors, but not overcome; an increasing real wage will be accompanied by a falling value of labor power. Sweezy obviously confuses the real wage and the value of labor power and derives a tendency for the value of labor power to increase from the growth of production costs over time.” He forgets that the value of labor power falls in the course of accumulation along with the values of all other commodities, even though this fall can appear as a rising real wage. The role of monopoly will, therefore, be to accelerate the fall in the value of labor power and not, as Sweezy supposes, to check its rise.SUMMER 1987 57 But if monopoly depresses the value of the commodity labor power, this corresponds precisely to Marx’s hypothesis, basic to his theory of accumulation, that the value of labor power falls as accumulation progresses, since this fall is the expression, under capitalist conditions, of growing labor productivity. Thus Marx did not deal with monopoly, along with all other particular aspects of competition and the market, for methodological and theoretical reasons, and not because monopo- lies were an exception in his time. For this reason it is still correct today to consider the general laws of accumulation independent of monopoly, since these laws are concerned with the general characteris- tics of capital, and it will be correct to do this as long as capitalism exists. This is obviously not to say that monopoly has no significance at all, and that one may always abstract from it. If we wish to analyze the accumulation of individual capital and competition, the distribution of produced surplus value, disturbances in the reproduction process, or the concrete forms of crisis in which the general tendency to crisis asserts itself, the form of competition must be taken into account. The problem of monopoly should be dealt with at this level. However, if we wish to understand the laws that regulate the relation between surplus value and invested capital we must abstract from monopoly. 0 As regards the theory of accumulation, Sweezy sees in my essay a tendency toward the ‘‘fetishization of the falling tendency of the rate of profit.’’!° He believes he sees this fetishization expressed in the espe- cially absurd hypothesis of a ‘runaway organic composition of capital, increasing without assignable limit and much more rapidly than the rate of surplus value.’’!! I should not have to emphasize that there is no mention anywhere in my essay of a ‘‘runaway organic composition,” and that Sweezy cannot cite a single word of mine to substantiate his imputation. I have stressed just the opposite, that is, that the capitalist system is always forced to mobilize counter-tendencies in its effort to keep the organic composition at a level acceptable with regard to profitability. '? I also mentioned the sphere of technology and its possi- ble effects on the cheapening of constant capital,'* although I did not develop this point as it was not the topic of my essay. Although I have never advocated the thesis of a ‘‘runaway organic composition,’’ I do hold that the law of the tendency of the rate of profit to fall is the core of the Marxian theory of accumulation. Sweezy is58 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY convinced that ‘‘nothing follows from the concept of capital except that capital is self-expanding value.’’'* From this concept of capital “‘no deductions or conclusions can be drawn . . . without the introduc- tion of further assumptions,’’ which, from the Marxist point of view, must be historically and empirically based. Thus the fall in the rate of profit is linked to historical, empirical conditions which are not “‘logically dictated by [Marx's] theory of accumulation.’’!5 Why then did Marx formulate the law of the tendency of the rate of profit to fall? For such difficult questions Sweezy always has the same answer ready: because it was so in Marx’s time! *‘For my part, I have no doubt that the reason is that this is precisely what happened during the period of the industrial revolution beginning in the eighteenth cen- tury and still continuing at the time in the 1850s and 1860s when he was working on Capiral.’’'¢ In addition, Sweezy is of the opinion that Marx did not intend to develop a theory of crisis from the law of the tendency of the rate of profit to fall,!7 and that he spoke only of a tendency, which ‘‘like other tendencies, was opposed by various counteracting causes’’!8—as Rosdolsky put it ironically in his critique of Natalie Moskowska, Marx meant to frame ‘‘a law of the falling, or not-falling rate of profit.’’!® This is, indeed, an astonishing interpretation of Cap- ital. According to Marx, one can derive from the concept of capital not only that it is self-expanding value, but practically everything contained in the three volumes of Capital. These volumes are concerned through- out with the concept of capital, that is, what is common to all capitals independent of their empirical and historical forms. Therefore, in Marx’s view, the law of the tendency of the rate of profit to fall also belongs to the general concept of capital, as Marx states explicitly in various places. He says, for example, in the discussion of one of the ‘counteracting factors”’ (the reduction of the wage below the value of labor power), ‘We simply make an empirical reference to this point here, as, like many other things that might be brought in, it has nothing to do with the general analysis of capital, but has its place in an account of competition, which is not dealt with in this work.?° Marx also explicitly associated the capitalist crisis with the law of the tendency of the rate of profit to fall:SUMMER 1987 59 The barriers to the capitalist mode of production show themselves as follows: (1) In the way that the development of labor productivity involves a law, in the form of a falling rate of profit, that at a certain point confronts this development itself in a most hostile way and has constantly to be overcome by way of crises.?! And also: The periodical devaluation of the existing capital, which is a means, immanent to the capitalist mode of production, for delaying the fall in the profit rate and accelerating the accumulation of capital-value by the for- mation of new capital, disturbs the given conditions in which the circula- tion and reproduction process of capital takes place, and is therefore accompanied by sudden stoppages and crises in the production process.?? All these quotations are, however, not decisive, for it is possible that, whatever Marx’s intention, one cannot deduce from the general laws of capitalist production an immanent and necessary tendency of the rate of profit to fall and a tendency toward crisis. Hence we must consider more closely Sweezy’s arguments that the law of the falling tendency of the rate of profit is valid only for a particular phase of capitalist development. There are two very old, and always recurring, arguments against the law of the tendency of the rate of profit to fall. They have been ad- vanced by Sweezy in his Theory of Capitalist Development,?? as well as by Joan Robinson?‘ and Joseph Gillman.?> (J) If the rate of profit is a function of the organic composition and the rate of surplus value, it may be the case that a rising rate of surplus value compensates or overcompensates for the negative effects of an increasing organic composition, so that the tendency of the rate of profit is indeterminate. This argument is supplemented by Sweezy, as well as by Robinson and Gillman, with the criticism that Marx had proved the tendency of the rate of profit to fall only by assuming a constant rate of surplus value (Joan Robinson's famous “‘tautology’’)?6. (2) Technical innovation can be of a labor-saving kind, that is, it can raise productivity by substituting capital for labor. However, since capital, like all other commodities, falls in value, it is possible that the organic composition, as a value relation and not simply a relation between physical quantities of machines and labor, does not rise. Here60 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY again, the tendency of change is indeterminate. The first argument rests on a simple misunderstanding of the concept of the organic composition of capital. What is the ‘‘organic composition’’ of capital? Here Marx is very explicit and gives a clear definition of the concept. He distinguishes between the technical composition, which represents the ratio of means of production to mass of labor employed, and the value composition, which is the ratio of constant to variable capital. It must be noted that the organic composition and the value composition are not always identical. One of the reasons for the difficulties in which Sweezy, Robinson, and Gillman find themselves is that they equate them. After he has introduced the concepts of the technical and the value composi- tions, Marx says: There is a close correlation between the two. To express this, I call the value-composition of capital, in so far as it is determined by its technical composition and mirrors the changes in the latter, the organic composition of capital.?” The organic composition is thus not simply the value composition, but the value composition only insofar as it mirrors the technical com- position. Grossmann has expresssly warned against confusing the or- ganic with the value composition.?* What is the significance of this ‘‘close correlation’’? It implies, for example, that a change in the value composition that is not attributable to a change in the technical compo- sition is not to be regarded as a change in the organic compostion, just as a change in the technical composition that does not lead to a change in the value composition is not to be regarded as a change in the organic composition. Let us take an example. A fall in the value of labor power that is not accompanied by a change in the technical composition changes the ratio of constant to variable capital, because the variable capital is reduced. But since there is no change in the ratio of means of production to mass of labor, this rise in the value composition must not be confused with a rise in the organic composition. Otherwise, every change in the value of labor power, even with no change in the technique of production, would involve a change in the organic composition, although the relation between living and objectified labor would not have changed. Howev- er, it must be remembered that an increase in the organic composition always implies an increase of constant capital for a given mass of employed labor.SUMMER 1987 61 This has direct consequences for any attempt to examine the fall in the rate of profit in mathematical terms. How, for instance, can the organic composition be expressed mathematically? If one begins with a constant rate of surplus value, one can express the organic composition by c/v, because v always represents a given number of workers, a given quantity of labor. An increase in c/v must always mirror a change in the technical composition, i.e. an increase of constant capital in proportion to the mass of labor employed. If, on the other hand, one begins with a rising rate of surplus value, one can no longer express the organic composition by c/v. This is because an increase in this ratio can express not only an increase of constant capital relative to the quantity of labor, but also a simple reduction in the value of labor power, or both at the same time. If we assume a rising rate of surplus value then only the ratio c/v+s is of significance for the organic composition. This is because an increase of this ratio is not affected by a decrease in the value of labor power, but only by an increase of constant capital relative to the quanti- ty of labor. Only c/v+s, and not c/v, expresses changes in the value composition in sofaras it mirrors changes in the technical composition. Marx describes this as a ratio of dead to living labor.?° All the contra- dictions that Sweezy and Robinson® have found in Marx’s treatment of the law of the tendency of the rate of profit to fall are thus easily resolved and are attributable to these authors’ incomprehension of Marx’s method. Marx begins his argument by assuming a constant rate of surplus value and, therefore, treats the organic composition as a relation between constant and variable capital. For under this assump- tion, c/v does express changes in the organic composition. Marx subse- quently abandons this assumption and proves the validity of the law even with a rising rate of surplus value. Robinson and Sweezy wish to examine the law directly under the assumption of a rising rate of surplus value. They forget in doing this that, on this assumption, changes in the organic composition can no longer be expressed by c/v. If one wishes to examine directly the tendency of the rate of profit to fall with a rising rate of surplus value, one must use c/v +s as the expression for the organic composition. For this reason, Sweezy’s formula, s/o+l, where o = c/v,3! is completely useless for the examination of the law in the case of a rising rate of surplus value. Hence Sweezy’s and Robinson’s riddle is easily resolved. We begin with a simple case of capital accumulating with a rising rate of surplus value and a rising organic composition:62 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY c v S Period 0 100 100 100 Period 1 120 90 110 Period 2 144 81 119 Period 3 172.8 72.9 dal We have made the following assumptions in the construction of this example: The number of workers is constant, i.e., v+s always = 200. This assumption has no consequences as far as conclusions about the profit rate are concerned. This is because a growing number of workers would mean multiplying the numerator and the denominator of the formula for the profit rate by the growth rate of the number of workers, and the rate of profit itself would remain unchanged. Further, we have assumed that constant capital increases, in each period, by 20%, and that the value of labor power falls, in each period, by 10%. The organic composition c/v+s therefore increases by 20% while c/v increases by 33%. We can now generalize our example and examine the tendency of the profit rate for any values of the rise in the organic composition and rate of surplus value. We let u be the rate of increase of the organic composition and / the rate of decrease in the value of labor power. If cq is the value of constant capital in period 0, then ¢C, = Co(1+u). ‘We can then write: Cn = Co(1 tu)". Similarly, we can obtain: V, = Vo (1—D". Let A be newly produced value, v+s, which we have assumed constant because we have taken the number of workers as constant. Then, in period 0, the surplus value is: So = A-Vo.SUMMER 1987 63 In period n it is: 5S, = Avo (1—D". We can now write the profit rate as: So = ——— Co + Vo and Sn iad Cr + Vn Thus A-vo(1 — 1)” rn = cal + wy" + voll — DP From this formula it is clear that r, —- Oasn — @. On the other hand, if u = 0, i-e., if the organic composition remains constant, rT, — Alcgasn — ©. If u is negative, then Tr, —~ asn = ©, It must be emphasized that this result holds even when the rate of surplus value tends to infinity. For: Sa A — v1 — D” Vn vol — 1” And if n — 00, then s,/v, ~ ©.64 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY It follows from this analysis that the increase in the rate of surplus value cannot ultimately compensate for the rise in the organic composi- tion. That is, in the long run, the effect of the organic composition will assert itself. The argument developed here is, in fact, already completely con- tained in Marx, although expressed in another form. Marx had pointed out that the organic composition rises without limit while the mass of surplus value is limited by the length of the working day.?? These aspects of the Marxian law have recently been emphasized again and again. The work of Rosdolsky, Meek, and Mattick33 in particular has shown that Sweezy’s, Robinson’s and Gillman’s belief*4 that Marx had proved the law of the tendency of the rate of profit to fall only for the case of a constant rate of surplus value is simply not true. Thus Swee- zy’s question, ‘‘why did Marx feel so sure that the organic composition of capital must rise faster than the rate of surplus value?’’35 is not to be answered with the standard Sweezy answer (because it was so in Marx’s time), but by the reply that it necessarily follows from the laws of capitalist production. We now come to Sweezy’s second argument against the validity of the law for present-day capitalism, that it cannot be decided on the basis of the analysis of the laws of capitalist accumulation whether the organ- ic composition must have a tendency to rise in the course of accumula- tion. ‘‘The only way this question can be decided is through the empiri- cal study of capitalist practice.’’*° Sweezy is right to pose this question which I agree is a very important one for the theory of accumulation one which has not been sufficiently examined, theoretically or empiri- cally. But when Sweezy suggests that empirical studies have demon- strated *‘beyond a reasonable doubt’’>” that the organic composition has had a falling or constant tendency during the last fifty years, one can say ‘‘beyond a reasonable doubt’’ that he is wrong. I believe that a great deal speaks for the correctness of Marx’s analysis on this point also. However, before we analyze more closely Gillman’s research, which Sweezy cites in this context, we must clear up a misunderstanding. We are not dealing here, as Sweezy appears to believe, with a purely empirical problem that can be dealt with sepa- rately from the general analysis of the law of capital accumulation. It is, of course, very easy to refer to empirical data; but Sweezy unfortunate- ly forgets to say which concepts he is using the analyze this data. He finds nothing better than to refer us to Gillman’s study, which in fact shows only that it is impossible to make an empirical analysis of capitalSUMMER 1987 65 accumulation if the concepts employed do not have adequate theoreti- cal explanation. Sweezy himself does not appear to be so certain about the explanatory power of the empirical studies he cites. He first states that ‘‘numerous studies’’ have demonstrated ‘‘beyond reasonable doubt’’ that the organic composition has fallen in the last fifty years. But a page later we learn that these ‘‘numerous studies’ actually make no use of Marxian concepts, so that, since ‘‘organic composi- tion’’ is one of Marx’s concepts, it is not clear what ‘‘numerous studies’’ which do not deal with the organic composition can have to say about the organic composition. However, in order to remove our doubts, Sweezy immediately reassures us that there is at least one study whose findings are ‘‘unequivocal’’: Gillman’s study of the fall of the rate of profit. Later, however we are told that ‘‘Gillman’s statisti- cal procedures have been justifiably criticized.’’ This does not ap- pear to alter Sweezy’s belief that the evolutionary tendency of the organic composition has been reversed in the course of the last fifty years. The problem with Gillman’s research is not so much that his statisti- cal method can be criticized, but that his theory can be criticized in such a way that it becomes very uncertain whether his findings can be of any significance for a theory of capital accumulation. Gillman exam- ines the question whether the long-term trend of the ratios important for the capital structure (i.e., organic composition, rate of surplus value, and rate of profit) corresponds to Marx’s assumption of a rising organic composition and a falling rate of profit. In so doing, he severs the crucial connection in Marx’s theory between the law of the tendency of the rate of profit to fall and the crisis theory. For Marx claimed not that there could be an unbroken long-term trend of the organic compo- sition, but rather the contrary. Precisely because the fall of the rate of profit in the long run is not compatible with the survival of capitalism, crisis is necessary, since it makes possible the resumption of the process of accumulation, by provoking the devaluation and concentration of capital. How then is this conflict to be resolved? How are the relations corre- sponding to a “‘healthy’’ movement of capitalist production to be re- stored? The method of resolution is already implicit in the way in which the conflict is stated. It involves this, that capital should be idle, or even, in part, be destroyed. . . .** Or:66 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY The growing incompatibility between the productive development of soci- ety and its hitherto existing relations of production expresses itself in bitter contradictions, crises, spasms. The violent destruction of capital not by relations external to it, but rather as a condition of its self-preserva- tion, is the most striking form in which advice is given it to be gone and give room to a higher state of social production. . . . These contradic- tions, of course, lead to explosions, crises, in which momentary suspen- sion of all labor and annihilation of a great part of the capital violently lead back to the point where it is enabled [to go on] fully employing its productive powers without committing suicide.* The devaluation of capital in the crisis lowers the organic composi- tion without lowering the productivity of capital, because what is de- stroyed is not the use-value of machines, but their value. In this way the conditions of accumulation are improved, for despite the fall of the organic composition the productivity remains unchanged, or even in- creases as a a result of the concentration of capital. The destruction of capital through crises means the depreciation of values which prevents them from later renewing their reproduction process as capital on the same scale. This is the ruinous effect of the fall in the price of commodities. It does not cause the destruction of any use-values. What one loses, the other gains. Values used as capital are prevented from acting again as capital in the hands of the same person. The old capitalists go bankrupt.*? Marx’s the theory of the tendency of the rate of profit to fall is connect- ed with the cyclical movement of capitalist production. For this reason, Gillman is already on the wrong track in taking the law as a long-term trend independent of the cycle. We cannot here discuss the question whether the devaluation of constant capital effected by the crisis is not today taking new forms, like that of state-induced destruction of part of the constant capital, inflation, monopoly (it seems to me that it would be more interesting to examine the role of monopoly in this context than in that proposed by Sweezy), etc. It can only be pointed out that these new forms cannot overcome the cause of the crisis, which is rooted in the capitalist mode of production, and, at best, can only postpone or modify it. Thus all the new forms of crisis management find their limits in the capitalist mode of production itself, and one cannot expect that the crisis inherent in the tendential fall of the rate of profit can remain latent in the long run. It follows from the connection Marx makes between the theory ofSUMMER 1987 67 accumulation and the theory of crisis that we must distinguish clearly between the counteracting tendencies and the cyclical devaluation. More exactly, we must distinguish the lowering of the organic composi- tion through the power of capital itself (such as capital-saving technol- ogy or lowering the value of constant capital) from the destruction of capital value resulting from crises. If this cannot be done statistically, every ‘‘empirical’’ investigation is condemned to failure. But Gillman confuses precisely what he should keep distinctly apart, and includes monopolization and concentration of capital among the counter-ten- dencies.*! Gillman’s data shows that in the very period where he sees a change in the trend of the organic composition the crises reached an intensity unprecedented in the history of capitalism.*? If we take Gill- man’s own index of the intensity of crises, all crises before 1919 are below 200 (with the exception of the crisis of 1875-1878: 347), while the value of the index for the 1920/22 crisis rises to 387, that for the 1930/31 crisis to 2,512, and that for the 1937/40 crisis to 838. Given this, it is astonishing that Gillman attempts to compute a trend for this period without examining the effects of the crisis on the devaluation of constant capital and without taking into account the fact that it was during this period that the great crises of 1920 and 1929 and the Second World War took place. As a result, Gillman’s thesis that ‘‘new techno- logy’’ was responsible for the change in the trend is not very convinc- ing, since he is never in a position to measure statistically the effects of the ‘tnew technology”’ in isolation from those of the restructuring of the value composition of capital brought about by the crisis. It is significant that when Gillman comes to speak concretely about the “‘new technology”’ he cites only examples that appear to involve every- thing but the saving of constant capital.** Besides this general critique, we can criticize specific concepts that Gillman uses. Gillman attempts to determine statistically the organic composition defined as c/v. We have already indicated above why this expression is not suitable, at the value level of analysis, for compre- hending changes in the organic composition on the assumption of a rising rate of surplus value. If one argues at the level of price, further difficulties are added, namely, that c/v is modified not only by changes in the value of labor power but also by cyclical fluctuations in the price of labor power. Since decisive changes in the intensity of labor (assem- bly line production, etc.) took place during the time discussed, we can assume that an increase in the degree of exploitation of labor appears in Gillman’s statistics as a decrease of the organic composition. It seems68 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY to be fairly obvious that one cannot conclude from this that the organic composition has fallen in Marx’s sense of this concept. For all these reasons, Gillman’s research is completely useless for an empirical test of Marx’s theory of accumulation. This is not because of his statistical method, but because of his theory, which never considers the status of the law of the tendency of the rate of profit to fall in the framework of Marx’s theory of accumulation and crisis, or the relationship between Marxian categories and empirical data. Since Gillman’s data are therefore anything but ‘‘unequivocal,’’ we should ask if data do exist that support the hypothesis of a rising composition of capital. There are some, but they are also not ‘‘unequi- vocal.’’ The capital investment per employed person (in constant prices) rose both in Germany during 1950-70 and in America during 1879-1953. Since constant prices are not constant values these figures are only of limited significance. Bairoch gives the following figures for the value of capital per employed worker in the United States in 1929 prices (dollars).4* Capital per employed worker 1879 1,764 1889 2,702 1899 3,655 1909 5,040 1929 7,530 1948 6,543 1953 7,859 In Germany the capital intensity (gross fixed capital investment per employed person, at constant prices) rose for industry as a whole in the periods 1950-55, 1955-60, 1960-65, and 1965-70 by 1.8%, 21.9%, 36.8%, and 27.1% respectively.*5 And the adjusted capital coefficient (adjusted for cyclical fluctuations)** for German industry as a whole rose between 1960 and 1970 from 1.14 to 1.27.47 Kuznets notes that *‘the constancy of the total/capital product ratios implies that total capital per head increased at the same rate as per capita product.’’48 If this observation is applied to the German data it means that the small increase of the capital coefficient, along with a considerable growth of capital per person employed, implies a similar- ly considerable increase in labor productivity (net volume of produc-SUMMER 1987 oo tion per person employed at constant prices). Thus the labor productiv- ity for West German industry as a whole rose during 1960-65 by 24.8% and during 1965-70 by 29.7%.49 It would be going too far to draw conclusions from these data, but it appears that there is a tendency for capital investment per person employed to increase, even though at this time its negative effects on the profit rate are checked by the increase in labor intensity. Similar tendencies have been established by Mandel in a comparison of the growth rates of the production of production goods and of consumption goods. From the Statistical Abstract of the United States he draws the information that ‘‘between 1919 and 1952 the production of durable goods (mostly belonging to [Department] I) increased 5-fold (growing from index 72 to index 340) whereas that of non-durables only trebled (growing from index 62 to index 190).’’*° Elsewhere Mandel relies on the work of Anne P. Carter’s work to show that ‘“The econo- my behaves as if labor-saving were the goal of technical progress.”’5! It must of course be said at once that all these data are certainly not “unequivocal’’; at best, they may serve to raise a ‘‘reasonable doubt”’ in the minds of those who, like Sweezy, believe that ‘‘on the basis of the data available to us today. . . the law of the falling tendency of the rate of profit to fall is no longer operative.’’5? If the data suggest anything, it is only that the forces behind Marx’s law of the tendency of the rate of profit to fall are apparently still operating in a latent manner, and that they have remained latent until now thanks to the successful phase of capital expansion after the Second World War. There is hardly reason to doubt that the capital expansion of the last thirty years can be attributed to the effects of the war and to the profound restructuring of interna- tional capital during this period. Therefore, Sweezy’s assumption, that the latent nature of these forces has taken on a permanent character due to the changed structure of modern technology, remains completely unfounded. And in the article discussed here, Sweezy supports his position neither theoretically nor by way of a concrete analysis of capitalist reality. So long as he can find no better support for his assertions than a reference to Gillman’s study, it seems to me that everything indicates that the forces analyzed by Marx are still operative and that the following problem arises: What are the factors which have So far kept these forces latent and under what circumstances can they conceivably be kept latent? Unlike Sweezy, I do not believe that capital can succeed in evading the contradictions deriving from the law of the tendency of the rate of profit to fall.70 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY Til Sweezy has understood me very well when he characterizes my posi- tion as the claim that ‘‘Problems of realization and underconsumption (and/or overproduction) are derived from the theory of the falling tendency of the rate of profit and have no independent existence.’’5? Only this must be added: that they have no independent existence as long-term, necessary causes of crisis, because since capitalism is based on the separation between production and consumption, a divergence between production and consumption is obviously always a possibility. In the same way, disproportionality and underconsumption are always possible difficulties for capitalist production. But is is only on the basis of the theory of the tendency of the rate of profit to fall that one can show that a tendency toward overproduction in capitalism is not only possible but necessary. I have not only asserted this, but I have also attempted to demon- strate it by way of a rather detailed critique of Joan Robinson’s theory of underconsumption.** As Sweezy says not a word about my argument I must conclude that he has stil] not refuted it. Thus this new version of his theory of underconsumption, or his theory of the contradiction between production and consumption, is no more convincing than Robinson’s version, or all the other versions of the same theory. Marx shows by means of his reproduction schemas how the contradiction between production and consumption, which always accompanies the capitalist form of production, finds its dynamic form. The reproduc- tion schemas show how it is possible for all commodities to find markets, despite the contradiction between production for consump- tion and production for accumulation.5> Since total demand under capitalism represents accumulation (even the demand for consumption goods representing accumulation—the accumulation of variable cap- ital), the organic composition determines which part of total demand consists of demand for consumption goods and which consists of de- mand for production goods. Thus a decline of demand must stem from the dynamic of capital, and capital reduces its demand only when the rate of profit falls. Logically, therefore, we can only deduce the over- production of commodities from the fall in the rate of profit, and not vice versa. What does Sweezy’s construction of a ‘‘capitalist utopia’? demon- strate? Only that, without consumption, no production is possible, which no one ever disputed because it is simply a tautology. If nothing isSUMMER 1987 71 consumed or produced then there can be neither use-values nor values nor anything. But a justification for Sweezy’s version of undercon- sumption theory can hardly be found in this banal statement. One therefore cannot take seriously Sweezy’s attempt to derive from it a theory of accumulation more fruitful than that based on the tendency of the rate of profit to fall.5° The quotation from Marx does not make Sweezy’s theory more credible, for it is taken from the very chapter in Capital that argues for a tendency of the rate of profit to fall, and only confirms the idea that Marx saw the phenomenon of overproduction as connected with this tendency. The Keynesian origin of Sweezy’s argument is most clearly visible in the fact that he is not able to differentiate his position from Robin- son’s. We learn from his note 14 that the ‘‘radical difference’ between Robinson and himself consists in the Keynesian conception of a ‘‘mal- distribution of income, clearly implying that matters can be put right by a suitable redistribution of income and thus opening the way for all sorts of reformist illusions.’’ In order to substantiate these *‘radical differences,’’ Sweezy quotes a passage from Robinson, which I also cited in my article,5’ in which she speaks of a maldistribution of income. But if Sweezy had troubled to check Robinson’s next page he would have found this passage: Underconsumption theories have been associated with an appeal for re- form rather than revolution—with the view that capitalism might be made to work satisfactorily—and for this reason they are uncongenial to the Marxist creed. . . . But this association is superficial, for the maldistribu- tion of income is quite as deeply embedded in the capitalist system as Marx believed the tendency to falling profits to be, and cannot be elimi- nated without drastic changes in the system.** Robinson thus knows quite as well as Sweezy that unequal distribution cannot be eliminated without the destruction of capitalism, and the “radical difference’ between Sweezy and the Keynesian Left turns out to be a profound entente. It is of course well known that Sweezy holds to different political positions than those of the Keynesian Left, and I do not wish to deny this. What I have tried to show is that such different political positions are based on largely identical analyses of capital- ism—whence the accusation of eclecticism made in my first essay, which Sweezy’s reply seems to confirm. Even the Keynesians know that distribution cannot be changed in capitalism; this is why they call not for a change in distribution but for the state to take over a part of72 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY investment (ignoring here those who have utilized Keynesianism as the ideology for a reformist union strategy). Sweezy evidently wished to know nothing of this confusion. Why? Because he is against capitalism for political reasons. According to him, political theory is distinct from the economic analysis of capital’s laws of motion. His assurance that neither he nor Baran has ever abandoned the theory of value remains an empty assertion so long as Sweezy cannot show how he develops the categories of his analysis of capitalism out of the concept of value. I find no argument in his essay that could not find a place in the Keynesian conceptual apparatus. Thus Sweezy’s political critique must finally remain a moral critique of the inhuman and irrational forms of consumption created by capital- ism in order to overcome the contradiction between production and consumption. For Marx, in contrast, the critique of capitalism is practi- cal, that made by the class struggle sharpened by the difficulties faced by capital in the process of its reproduction and self-expansion. Notes 1, Paul M. Sweezy, “Some Problems in the Theory of Capital Accumulation,”” pp. 25-36 in C.S.E. Bulletin, Autumn 1973 [above, pp. 38-53] 2. David S. Yaffe, ‘*The Marxian Theory of Crisis, Capital and the State,”” in Economy and Society 2:2 (May 1973) and Mario Cogoy, ‘*Les théories néo-marxistes, Marx et l’accumulation du capital,”” pp. 396-427 in Les Temps Modernes 29 (Septem- ber-October 1972) [above, pp. 11-37] 3. Sweezy, op. cit., p. 27 [above, p. 42]. 4. Ibid., p. 26 [above, p. 40] 5. Paul A. Baran and P. M. Sweezy, Monopoly Capital (New York: Monthly Review Press, 1966), p. 10n. 6. Marx to Engels, 24 August 1867 in Karl Marx and Friedrich Engels, Selected Correspondance (Moscow: Foreign Languages Publishing House, n.d.), p. 232. 7. Sweezy, op. cit., p. 25 [above, p. 38] 8. Baran and Sweezy, op. cit., p. 72. 9. Sweezy, op. cit., pp. 34-35 [above p. 51, n.4]. 10. [bid., p. 27 {above, p. 41] 11. Bid., p. 29 [above, p. 44]. 12. Cogoy, op. cit., pp. 407-8 [above, pp. 19-20]. 13. Ibid., p. 427 [above, p. 34]. 14. Sweezy, op. cit., p. 35, note 5 [above, pp. 51-52, n.6]. 15. Ibid., p. 28 (above, p. 43] 16. Ibid. 17. Ibid., p. 30 [above, p. 46] 18. Ibid., p. 31 [above, p. 46] 19. Roman Rosdolsky, The Making of Marx's ‘Capital’ (London: Pluto, 1977), pp- 405-6 n. 19. 20. Marx, Capital, vol. 3 (New York: Vintage, 1981), p. 342. 21. Dbid., p. 367.SUMMER 1987 73 22. Ibid,, p. 358. See also the place in Cogoy, op. cit., p. 413 n. 34 [above, p. 36], where I quote Marx, Grundrisse (New York: Vintage, 1973), pp. 750-51. 23. Sweezy, The Theory of Capitalist Development (New York: Oxford University Press, 1942), chap. 6. 24. Joan Robinson, An Essay on Marxian Economics, 2nd ed. (London: Macmil- lan, 1963), chap. 5. 25. Joseph M. Gillman, The Falling Rate of Profit (New York: Cameron, 1958). 26. Robinson, op. cit., p. 36 27. Marx, Capital, vol. 1 (New York: Vintage, 1977), p. 762. 28. Henryk Grossmann, Das Akkumulations- und Zusammensbruchsgesetz des kapitalistischen Systems (Leipzig: C.L. Hirschfeld, 1929), pp. 326-27 29. Marx, Grundrisse, p. 750. 30. Here what is at question is first of all the alleged contradiction in Marx’s assumption of a rising organic composition together with a constant rate of surplus value. See Robinson, op. cit., p. 36, and Sweezy, Theory, pp. 96-97. 31. Sweezy, ‘‘Some Problems,” pp. 27-28 [above, p. 42]. 32. Marx, Capital, vol. 3, pp. 355-56, and vol. 1, pp. 657-58; Grundrisse, pp. 339-41. 33. Rosdolsky, op. cit., pp. 230ff and 407ff; Ronald Meek, **The Falling Rate of Profit,"’ Science and Society 24:4 (1960); Paul Mattick, Marx and Keynes (Boston: Porter Sargent, 1969), pp. 62-63. 34. Sweezy, Theory, pp. 96-97; Robinson, op. cit., p. 40; Gillman, op. cit., pp. 20, 107. 35. Sweezy, ‘Some Problems,” p. 28 [above, p. 43]. 36. Ibid., p. 29 [above, p. 44]. 37. Ibid. 38. Marx, Capital, vol. 3, p. 362 39. Marx, Grundrisse, pp. 749-50. 40. Marx, Theories of Surplus Value, vol. 1 (Moscow: Progress Publishers, 1968), 42. Ibid., pp. 125-26. 43. Ibid., pp. 74ff. 44. Paul Bairoch, Révolution industrielle et sous-développment (Paris, 1964), p. 63 45. Soziologisches Forschungsinstitut Gottingen, Forschungsberichte: Produktion und Qualifikation (Gottingen, 1973), table 9 46. Ibid., p. 31 47. Ibid. , table 10. 48. Simon Kuznets, Modern Economic Growth, Rate, Structure, and Spread (New Haven: Yale University Press, 1970), p. 78. 49. Soz. Forschungsinstitut, op. cit., table 8. 50. Ernest Mandel, Marxist Economic Theory (London: Merlin, 1968), p. 365n. 51. Anne P. Carter, Structural Change in the American Economy (Cambridge: Harvard University Press, 1970), p. 152 cited in Mandel, Late Capitalism (London: New Left Books, 1975), p. 200. 52. Sweezy, ‘Some Problems,”’ p. 32 [above, p. 49]. 53. Ibid., p. 31 [above, p. 47]. 54, Cogoy, “Les thories,”” pp. 401-5 [above, pp. 13-20]. Ihave also attempted to develop a critique of underconsumption theory in a new essay, **Werttheorie und Staatsausgaben,"’ pp. 129-78 in Claudia von Braunmiihl et al., Probleme einer mater- ialistischen Staatstheorie (Frankfurt: Suhrkamp, 1973) [below, pp. 75-110].74 INTERNATIONAL JOURNAL OF POLITICAL ECONOMY 55. Jérg Glombowski has given a very convincing mathematical proof of this proposition, including the case of a rising organic composition of capital; see Glom- bowski, “‘Gleichgewichtige erwiterte Reproduktion mit variablen Wachstumsraten der Kapitale,”’ in Mehrwert 2 (May 1973). This possibility of equilibrium is naturally not to be understood as an eternal harmony between production and consumption, but as a possibility of equilibrium which can assert itself over the constant disequilibrium. On the methodology of the reproduction schemas, see Rudolf Hickel, ‘*Zur Interpretation der Marxschen Reproduktionsschemata,”’ in Mehrwert 2 (May 1973). 56. Mandel has given a convincing criticism of Sweezy’s version of the undercon- sumption theory in Marxist Economic Theory, pp. 363-64. 57. Robinson, op. cit., p. 71. 58. Ibid., p. 72.