The Accumulation of Capital

by Rosa Luxemburg

Yet Rodbertus again contradicts himself in other places, remembering all of a sudden the constant capital and the necessity for its renewal in the reproductive process. Thus, instead of bisecting the aggregate product into v and s, he posits a triple division: c, v, and s. In his third Letter on Social Problems he argues on the forms of reproduction in a slave-economy:

‘Since the master will see to it that part of the slave labour is employed in maintaining or even improving the fields, herds, agricultural and manufacturing tools, there will be “capital replacement”, to use a modern term, in which part of the national economic product is immediately used for the upkeep of the estate, without any mediation by exchange or even by exchange value.’[263] And, passing on to capitalist reproduction, he continues: ‘Now, in terms of value, one portion of the labour product, is used or set aside for the maintenance of the estate, for “capital replacement”, another, for the workers’ subsistence as their money wage; and the owners of the land, of capital, and of the labour product retain the last as their revenue or rent.’[264]

This, then, is an explicit expression of the triple division into constant capital, variable capital, and surplus value. Again, in this third Letter, he formulates the peculiarity of his ‘new’ theory with equal precision: ‘On this theory, then, and under conditions of adequate labour productivity, the portion of the product which remains for wages after the replacement of capital, will be distributed among workers and owners as wages and rent, on the basis of the ownership in land and capital.’[265]

It does seem now as if Rodbertus’ analysis of the value of the aggregate product represents a distinct advance over the classical school. Even Adam Smith’s ‘dogma’ is openly criticised a little further on, and it is really surprising that Rodbertus’ learned admirers, Messrs. Wagner, Dietzel, Diehl & Co. failed to claim their white-headed boy’s ‘priority’ over Marx on such an important point of economic theory. As a matter of fact, in this respect no less than in the general theory of value, Rodbertus’ priority is of a somewhat dubious character. If he seems on occasion to gain true insight, it immediately turns out to be a misunderstanding, or at best a wrong approach. His criticism of Adam Smith’s dogma affords a supreme example of his failure to cope with the triple division of the national product towards which he had groped his way. He says literally:

‘You know that all economists since Adam Smith already divided the value of the product into wage of labour, rent, and capital profit, that it is therefore not a new idea to ground the incomes of the various classes, and especially the various items of the rent, in a division of the product. But the economists at once go off the track. All of them, not even excepting Ricardo’s school, make the mistake, first, not to recognise that the aggregate product, the finished good, the national product as a whole, is an entity in which workers, landowners, and capitalists all share, but conceiving the division of the unfinished product to be of one kind shared among three partners, and that of the manufactured product as of another kind again, shared between only two partners. For these theories both the unfinished product and the manufactured product constitute as such separate items of revenue. Secondly,—though both Sismondi and Ricardo are free from this particular error—they regard the natural fact that labour cannot produce goods without material help, i.e. without the land, as an economic fact, and take the social fact for a primary datum that capital as understood to-day is required by the division of labour. Thus they set up the fiction of a fundamental economic relationship on which they base also for the shares of the various owners, ground rent springing from the contribution of the land lent by the owner to production, capital profits from the contribution of capital employed by the capitalist to this end, and the wages finally from labour’s contribution, seeing that there are separate owners of land, capital, and labour in the society. Say’s school, elaborating on this mistake with much ingenuity, even invented the concept of productive service of land, capital, and labour in conformity with the shares in the product of their respective owners, so as to explain these shares as the result of productive service.—Thirdly, they are caught up in the ultimate folly of deriving the wage of labour and the items of rent from the value of the product, the value of the product in turn being derived from the wage of labour and the items of rent, so that the one is made to depend on the other and vice versa. This absurdity is quite unmistakable when some of these authors attempt to expound “The Influence of Rent Upon Production Prices” and “The Influence of Production Prices Upon Rent” in two consecutive chapters.’[266]

Yet for all these excellent critical comments—the last, particularly acute, actually does to some extent anticipate Marx’s criticism of this point in Capital, volume ii—Rodbertus calmly falls in with the fundamental blunder of the classical school and its vulgar followers: to ignore altogether that part of the value of the aggregate product which is needed to replace the constant capital of the society. This way it was easier for him to keep up the singular fight against the ‘declining wage rate’.

Under capitalist forms of production, the value of the aggregate social product is divided into three parts: one corresponding to the value of the constant capital, the second to the wage total, i.e. the variable capital, and the third to the aggregate surplus value of the capitalist class. In this composition, the portion corresponding to the variable capital is relatively on the decline, and this for two reasons. To begin with, the relation of c to (v + s) within c + v + s changes all the time in the direction of a relative increase of c and a relative decrease of v+s. This is the simple law for a progressive efficiency of human labour, valid for all societies of economic progress, independently of their historical forms, a formula which only states that living labour is increasingly able to convert more means of production into objects for use in an ever shorter time. And if (v + s) decreases as a whole, so must v, as its part, decrease in relation to the total value of the product. To kick against this, to try and stop the decrease, would be tantamount to contending against the general effects of a growing labour productivity. Further, there is within (v + s) as well a change in the direction of a relative decrease in v and a relative increase in s, that is to say, an ever smaller part of the newly created value is spent on wages and an ever greater part is appropriated as surplus value. This is the specifically capitalist formula of progressive labour productivity which, under capitalist conditions of production, is no less valid than the general law. To use the power of the state to prevent a decrease of v as against s would mean that the fundamental commodity of labour power is debarred from this progress which decreases production costs for all commodities; it would mean the exemption of this one commodity from the economic effects of technical progress. More than that: the ‘declining wage rate’ is only another expression of the rising rate of surplus value which forms the most powerful and effective means of checking a decline of the profit rate, and which therefore represents the prime incentive for capitalist production in general, and for technical progress within this system of production in particular. Doing away with the ‘declining wage rate’ by way of legislation would be as much as to do away with the raison d’être of capitalist society, to deal a crippling blow to its entire system. Let us face the facts: the individual capitalist, just like capitalist society as a whole, has no glimmering that the value of the product is made up from the sum total of labour necessary in the society, and this is actually beyond his grasp. Value, as the capitalist understands it, is the derivative form, reversed by competition as production costs. While in truth the value of the product is broken down into the values of its component fragments c, v and s, the capitalist mind conceives of it as the summation of c, v and s. These, in addition, also appear to him from a distorted perspective and in a secondary form, as (1) the wear and tear of his fixed capital, (2) his advances on circulating capital, including workers’ wages, and (3) the current profits, i.e. the average rate of profit on his entire capital. How, then, is the capitalist to be compelled by a law, say of the kind envisaged by Rodbertus, to maintain a ‘fixed wage rate’ in the face of the aggregate value of the product? It would be quite as brilliant to stipulate by law for exactly one-third, no more, no less, of the total price of the product to be payable for the raw materials employed in the manufacture of any commodity. Obviously, Rodbertus’ supreme notion, of which he was so proud, on which he built as if it were a new Archimedean discovery, which was to be the specific for all the ills of capitalist production, is arrant nonsense from all aspects of the capitalist mode of production. It could only result from the muddle in the theory of value which is brought to a head in Rodbertus’ inimitable phrase: that ‘now, in a capitalist society, the product must have value-in-exchange just as it had to have value-in-use in ancient economy’.[267] People in ancient society had to eat bread and meat in order to live, but we of to-day are already satisfied with knowing the price of bread and of meat. The most obvious inference from Rodbertus’ monomania about a ‘fixed wage rate’ is that he is quite incapable of understanding capitalist accumulation.

Previous quotations have already shown that Rodbertus thinks solely of simple commodity production, quite in keeping with his mistaken doctrine that the purpose of capitalist production is the manufacture of consumer goods for the satisfaction of ‘human wants’. For he always talks of ‘capital replacements’, of the need to enable the capitalists to ‘continue their enterprise on the previous scale’. His principal argument, however, is directly opposed to the accumulation of capital. To fix the rate of the surplus value, to prevent its growth, is tantamount to paralysing the accumulation of capital. Both Sismondi and v. Kirchmann had recognised the problem of balancing production and consumption to be indeed a problem of accumulation, that is to say of enlarged capitalist reproduction. Both traced the disturbances in the equilibrium of reproduction to accumulative tendencies denying the possibility of accumulation, with the only difference that the one recommended a damper on the productive forces as a remedy, while the other favoured their increasing employment to produce luxuries, the entire surplus value to be consumed. In this field, too, Rodbertus follows his own solitary path. The others might try with more or less success to comprehend the fact of capitalist accumulation, but Rodbertus prefers to fight the very concept. ‘Economists since Adam Smith have one after the other echoed the principle, setting it up as a universal and absolute truth, that capital could only come about by saving and accumulating.’[268]

Rodbertus is up in arms against this ‘deluded judgment’. Over sixty pages of print he sets out in detail that (a) it is not saving which is the source of capital but labour, that (b) the economists’ ‘delusion’ about ‘saving’ hails from the extravagant view that capital is itself productive, and that (c) this delusion is ultimately due to another: the error that capital is—capital.

v. Kirchmann for his part understood quite well what is at the bottom of capitalist ‘savings’. He had the pretty argument: ‘Everyone knows that the accumulation of capital is not a mere hoarding of reserves, an amassing of metal and monies to remain idle in the owners’ vaults. Those who want to save do it for the sake of re-employing their savings either personally or through the agency of others as capital, in order to yield them revenue. That is only possible if these capitals are used in new enterprises which can produce so as to provide the required interest. One may build a ship, another a barn, a third may reclaim a desolate swamp, a fourth may order a new spinning frame, while a fifth, in order to enlarge his shoe-making business, would buy more leather and employ more hands—and so on. Only if the capital that has been saved is employed in this way, can it yield interest (meaning profit), and the latter is the ultimate object of all saving.’[269]