Capital Volume II

### Chapter 21: Accumulation and Reproduction on an Extended Scale

#### Part 1

It has been shown in Book I how accumulation works in the case of the individual capitalist. By the conversion of the commodity-capital into money the surplus-product, in which the surplus-value is represented, is also turned into money. The capitalist reconverts the so metamorphosed surplus-value into additional natural elements of his productive capital. In the next cycle of production the increased capital furnishes an increased product. But what happens in the case of the individual capital must also show in the annual reproduction as a whole, just as we have seen it happen on analysing simple reproduction, namely, that the successive precipitation – in the case of individual capital – of its used-up fixed component parts in money which is being hoarded, also finds expression in the annual reproduction of society. If a certain individual capital is equal to 400 c + l00 v, and the annual surplus-value is equal to 100, then the commodity-product amounts to 400 c + 100 v + 100 s. These 600 are converted into money. Of this money, again, 400 c are converted into the natural form of constant capital, 100 v into labour-power, and – provided the entire surplus-value is being accumulated – 100 s are converted besides into additional constant capital by transformation into natural elements of the productive capital. It is assumed in this case: 1) that this amount is sufficient under the given technical conditions either to expand the functioning constant capital or to establish a new industrial business. But it may also happen that surplus-value must be converted into money and this money hoarded for a much longer time before this process, i.e., before real accumulation, expansion of production, can take place; 2) that production on an extended scale has actually been in process previously. For in order that the money (the surplus-value hoarded in money-form) may be converted into elements of productive capital, one must be able to buy these elements on the market as commodities. It makes no difference if they are not bought as finished products but made to order. They are not paid for until they are in existence and at any rate not until actual reproduction on an extended scale, an expansion of hitherto normal production, has taken place so far as they are concerned. They had to exist potentially, i.e., in their elements, as it requires only the impulse of an order, that is, the purchase of commodities before they actually exist and their anticipated sale, for their production really to take place. The money on the one side then calls forth extended reproduction on the other, because the possibility of it exists without money. For money in itself is not an element of real reproduction.

For instance capitalist A, who sells during one year or during a number of years certain quantities of commodities successively produced by him, thereby converts into money also that portion of the commodities which is the vehicle of surplus-value – the surplus-product – or in other words the very surplus-value produced by him in commodity-form, accumulates it gradually, and thus forms for himself new potential money-capital – potential because of its capacity and mission to be converted into elements of productive capital. But in actual fact he only engages in simple hoarding, which is not an element of actual reproduction. His activity at first consists only in successively withdrawing circulating money out of the circulation. Of course it is not impossible that the circulating money thus kept under lock and key by him was itself, before it entered into circulation, a portion of some other hoard. This hoard of A, which is potentially new money-capital, is not additional social wealth, any more than it would be if it were spent in articles of consumption. But money withdrawn from circulation, which therefore previously existed in circulation, may have been stored up at some prior time as a component part of a hoard, may have been the money-form of wages, may have converted means of production or other commodities into money or may have circulated portions of constant capital or the revenue of some capitalist. It is no more new wealth than money, considered from the standpoint of the simple circulation of commodities, is the vehicle not only of its actual value but also of its ten-fold value, because it was turned over ten times a day, realised ten different commodity-values. The commodities exist without it, and it itself remains what it is (or becomes even less by depreciation) whether in one turnover or in ten. Only in the production of gold – inasmuch as the gold product contains a surplus-product, a depository of surplus-value – is new wealth (potential money) created, and it increases the money material of new potential money-capitals only so far as the entire money-product enters into circulation.

Although this surplus-value hoarded in the form of money is not additional new social wealth, it represents new potential money-capital, on account of the function for which it is hoarded. (We shall see later that new money-capital may arise also in a way other than the gradual conversion of surplus-value into money.)

Money is withdrawn from circulation and stored up as a hoard by selling commodities without subsequent buying. If this operation is therefore conceived as a general process, it seems inexplicable where the buyers are to come from, since in that process everybody would want to sell in order to hoard, and none would want to buy. And it must be conceived generally, since every individual capital may be in the process of accumulation.

If we were to conceive the process of circulation between the various parts of the annual reproduction as taking place in a straight line-which would be wrong as it always consists with a few exceptions of mutually opposite movements-then we should have to start from the producer of gold (or silver) who buys without selling, and to assume that all others sell to him. In that case the entire yearly social surplus-product (the bearer of the entire surplus-value) would pass into his hands, and all the other capitalists would distribute among themselves pro rata his surplus-product, which naturally exists in the form of money, the natural embodiment in gold of his surplus-value. For that portion of the product of the gold producer which has to make good his active capital is already tied up and disposed of. The surplus-value of the gold producer, created in the form of gold, would then be the sole fund from which all other capitalists would draw the material for the conversion of their annual surplus-product into money. The magnitude of its value would then have to be equal to the entire annual surplus-value of society, which must first assume the guise of a hoard. Absurd as these assumptions would be, they would do nothing more than explain the possibility of a universal simultaneous formation of a hoard, and would not get reproduction itself one step further, except on the part of the gold producer.

Before we resolve this seeming difficulty we must distinguish between the accumulation in department I (production of means of production) and in department II (production of articles of consumption). We shall start with I.

#### 1. The Formation of a Hoard

It is evident that both the investments of capital in the numerous lines of industry constituting class I and the different individual investments of capital within each of these lines of industry, according to their age, i.e., the space of time during which they already have functioned, quite aside from their volumes, technical conditions, market conditions, etc., are in different stages of the process of successive transformation from surplus-value into potential money-capital, whether this money-capital is to serve for the expansion of the active capital or for the establishment of new industrial enterprises – the two forms of expansion of production. One part of the capitalists is continually converting its potential money-capital, grown to an appropriate size, into productive capital, i.e., with the money hoarded by the conversion of surplus-value into money they buy means of production, additional elements of constant capital. Another part of the capitalists is meanwhile still engaged in hoarding its potential money-capital. Capitalists belonging to these two categories confront each other: some as buyers, the others as sellers, and each one of the two exclusively in one of these roles.

For instance, let A sell 600 (equal to 400 c + 100 v + 100 s ) to B (who may represent more than one buyer). A sells 600 in commodities for 600 in money, of which 100 are surplus-value which he withdraws from circulation and hoards in the form of money. But these 100 in money are but the money-form of the surplus-product, which was the bearer of a value of 100. The formation of a hoard is no production at all, hence not an increment of production, either. The action of the capitalist consists here merely in withdrawing from circulation the 100 in money he grabbed by the sale of his surplus-product, holding on to it and impounding it. This operation is carried on not alone by A, but at numerous points along the periphery of circulation by other capitalists, A', A'', A''', all of them working with equal zeal at this sort of hoard formation. These numerous points at which money is withdrawn from circulation and accumulated in numerous individual hoards or potential money-capitals appear as so many obstacles to circulation, because they immobilise the money and deprive it of its capacity to circulate for a certain length of time. But it must be borne in mind that hoarding takes place in the simple circulation of commodities long before this is based on capitalist commodity production. The quantity of money existing in society is always greater than the part of it in actual circulation, although this swells or subsides according to circumstances. We find here again the same hoards, and the same formation of hoards, but now as an element immanent in the capitalist process of production.

One can understand the pleasure experienced when all these potential capitals within the credit system, by their concentration in the hands of banks, etc., become disposable, "loanable capital," money-capital, which indeed is no longer passive and music of the future, but active capital growing rank.

However, A accomplishes the formation of a hoard only to the extent that he acts only as a seller, so far as his surplus-product is concerned, and not afterward as a buyer. His successive production of surplus-products, the vehicles of his surplus-value to be converted into money, is therefore the premise of his forming a hoard. In the present case, where we are examining only the circulation within category I, the bodily form of the surplus-product, as that of the total product of which it is a part, is the bodily form of an element of constant capital I, that is to say, it belongs in the category of means of production creating means of production. We shall see presently what becomes of it, what function it performs, in the hands of buyers B, B', B'', etc.

It must be noted at this point first and foremost that although withdrawing money to the amount of his surplus-value from circulation and hoarding it, A on the other hand throws commodities into it without withdrawing other commodities in return. The capitalists B, B', B'', etc., are thereby enabled to throw money into circulation and withdraw only commodities from it. In the present case these commodities, according to their bodily form and their destination, enter into the constant capital of B, B', etc., as fixed or circulating element. We shall hear more about this anon when we deal with the buyer of the surplus-product, with B, B', etc.

Let us note by the way: Once more we find here, as we did in the case of simple reproduction, that the exchange of the various component parts of the annual product, i.e., their circulation (which must comprise at the same time the reproduction of the capital, and indeed its restoration in its various determinations, such as constant, variable, fixed, circulating, money- and commodity-capital) does not by any means presuppose mere purchase of commodities supplemented by a subsequent sale, or a sale supplemented by a subsequent purchase, so that there would actually be a bare exchange of commodity for commodity, as Political Economy assumes, especially the free-trade school since the physiocrats and Adam Smith. We know that the fixed capital, once the expenditure for it is made, is not replaced during the entire period of its function, but continues to act in its old form, while its value is gradually precipitated in the form of money. Now we have seen that the periodical renewal of fixed capital II (the entire capital-value II being converted into elements worth I(v+s) ) presupposes on the one hand the mere purchase of the fixed part of II c, reconverted from the form of money into its bodily form, to which corresponds the mere sale of I s ; and presupposes on the other hand the mere sale on the part of II c, the sale of its fixed (depreciation) part of the value precipitated in money, to which corresponds the mere purchase of I,. In order that the exchange may take place normally in this case, it must be assumed that the mere purchase on the part of II c is equal in magnitude of value to the mere sale on the part of II c, and that in the same way the mere sale of I s to II c, section 1, is equal to its mere purchase from II c, section 2. Otherwise simple reproduction is disturbed. Mere purchase here must be offset by a mere sale there. It must likewise be assumed in this case that the mere sale of that portion of I s which forms the hoards of A, A', A'' is balanced by the mere purchase of that portion of I s which converts the hoards of B, B', and B'' into elements of additional productive capital.

So far as the balance is restored by the fact that the buyer acts later on as a seller to the same amount of value, and vice versa, the money returns to the side that advanced it on purchasing, and which sold before it bought again. But the actual balance, so far as the exchange of commodities itself, the exchange of the various portions of the annual product is concerned, demands that the values of the commodities exchanged for one another be equal.

But inasmuch as only one-sided exchanges are made, a number of mere purchases on the one hand, a number of mere sales on the other – and we have seen that the normal exchange of the annual product on the basis of capitalism necessitates such one-sided metamorphoses – the balance can be maintained only on the assumption that in amount the value of the one-sided purchases and that of the one-sided sales tally. The fact that the production of commodities is the general form of capitalist production implies the role which money is playing in it not only as a medium of circulation, but also as money-capital, and engenders certain conditions of normal exchange peculiar to this mode of production and therefore of the normal course of reproduction, whether it be on a simple or on an extended scale – conditions which change into so many conditions of abnormal movement, into so many possibilities of crises, since a balance is itself an accident owing to the spontaneous nature of this production.

We have also seen that in the exchange of I v for a corresponding amount of value of II c, there takes place in the end, precisely for II c, a replacement of commodities II by an equivalent commodity-value I, that therefore on the part of aggregate capitalist II the sale of his own commodities is subsequently supplemented by the purchase of commodities from I of the same amount of value. This replacement takes place. But what does not take place is an exchange between capitalists I and II of their respective goods. II sells its commodities to working-class I. The latter confronts it one-sidedly, as a buyer of commodities, and it confronts that class one-sidedly as a seller of commodities. With the money proceeds so obtained II c confronts aggregate capitalist I one-sidedly as a buyer of commodities, and aggregate capitalist I confronts it one-sidedly as a seller of commodities up to the amount of I v. It is only by means of this sale of commodities that I finally reproduces its variable capital in the form of money-capital. If capital I faces that of II one-sidedly as a seller of commodities to the amount of I v, it faces working-class I as a buyer of commodities purchasing their labour-power. And if working-class I faces capitalist II one-sidedly as a buyer of commodities (namely, as a buyer of means of subsistence), it faces capitalist I one-sidedly as a seller of commodities, namely, as a seller of its labour-power.

The constant supply of labour-power on the part of working-class I, the reconversion of a portion of commodity-capital I into the money-form of variable capital, the replacement of a portion of commodity-capital II by natural elements of constant capital II c – all these necessary premises demand one another, but they are brought about by a very complicated process, including three processes of circulation which occur independently of one another but intermingle. This process is so complicated that it offers ever so many occasions for running abnormally.

#### 2. The Additional Constant Capital

The surplus-product, the bearer of surplus-value, does not cost its appropriators, capitalists I, anything. They are by no manner of means obliged to advance any money or commodities in order to obtain it. Even among the physiocrats an advance was the general form of value embodied in elements of productive capital. Hence what capitalists I advance is nothing but their constant and variable capital. The labourer not only preserves by his labour their constant capital; he not only replaces the value of their variable capital by a corresponding newly created portion of value in the form of commodities; by his surplus-labour he supplies them with a surplus-value existing in the form of surplus-product. By the successive sale of this surplus-product they form a hoard, additional potential money-capital. In the case under consideration, this surplus-product consists from the outset of means of production of means of production. It is only when it reaches the hands of B, B', B'', etc. (I) that this surplus-product functions as additional constant capital. But it is this virtualiter even before it is sold, even in the hands of the accumulators of hoards, A, A', A'' (I). If we consider merely the amount of value of the reproduction on the part of I, we are still moving within the bounds of simple reproduction, for no additional capital has been set in motion to create this virtualiter additional constant capital (the surplus-product), nor has any greater amount of surplus-labour been expended than that on the basis of simple reproduction. The difference is here only in the form of the surplus-labour performed, in the concrete nature of its particular useful character. It has been expended in means of production for I c instead of II c, in means of production of means of production instead of means of production of articles of consumption. In the case of simple reproduction it was assumed that the entire surplus-value I is spent as revenue, hence in commodities II. Hence the surplus-value consisted only of such means of production as have to replace constant capital II in its bodily form. In order that the transition from simple to extended reproduction may take place, production in department I must be in a position to fabricate fewer elements of constant capital for II and so many the more for I. This transition, which does not always take place without difficulties, is facilitated by the fact that some of the products of I may serve as means of production in either department.

It follows, then, that, considering the matter merely from the angle of volume of values, the material substratum of extended reproduction is produced within simple reproduction. It is simply surplus-labour of working-class I expended directly in the production of means of production, in the creation of virtual additional capital I. The formation of virtual additional money-capital on the part of A, A' and A'' (1) – by the successive sale of their surplus-product which was formed without any capitalist expenditure of money – is therefore simply the money-form of additionally produced means of production 1.

Consequently production of virtual additional capital expresses in our case (we shall see that it may also be formed in a quite different way) nothing but a phenomenon of the process of production itself, production, in a particular form, of elements of productive capital.

The production of additional virtual money-capital on a large scale, at numerous points of the periphery of circulation, is therefore but a result and expression of multifarious production of virtually additional productive capital, whose rise does not itself require additional expenditure of money on the part of the industrial capitalist. The successive transformation of this virtually additional productive capital into virtual money-capital (hoard) on the part of A, A', A'', etc. (I), occasioned by the successive sale of their surplus-product – hence by repeated one-sided sale of commodities without a supplementing purchase – is accomplished by a repeated withdrawal of money from circulation and a corresponding formation of a hoard. Except in the case where the buyer is a gold producer, this hoarding does not in any way imply additional wealth in precious metals, but only a change in the function of money previously circulating. A while ago it functioned as a medium of circulation, now it functions as a hoard, as virtually new money-capital in the process of formation. Thus the formation of additional money-capital and the quantity of the precious metals existing in a country are not in any causal relation to each other.

Hence it follows furthermore: The greater the productive capital already functioning in a country (including the labour-power, the producer of the surplus-product, incorporated in it), the more developed the productive power of labour and thereby also the technical means for the rapid expansion of the production of means of production – the greater therefore the quantity of the surplus-product both as to its value and as to the quantity of use-values in which it is represented – so much the greater is

1) the virtually additional productive capital in the form of surplus-product in the hands of A, A', A'', etc., and

2) the quantity of this surplus-product transformed into money, and hence that of the virtually additional money-capital in the hands of A, A', A''. The fact that Fullarton for instance does not want to hear of over-production in the ordinary sense but only of the over-production of capital, meaning money-capital, again shows how extremely little of the mechanism of their own system even the best bourgeois economists understand.

Whereas the surplus-product, directly produced and appropriated by the capitalists A, A', A'' (I), is the real basis of the accumulation of capital, i.e., of extended reproduction, although it does not actually function in this capacity until it reaches the hands of B, B', B'', etc. (I), it is on the contrary absolutely unproductive in its chrysalis stage of money – as a hoard and virtual money-capital in process of gradual formation – runs parallel with the process of production in this form, but lies outside of it. It is a dead weight of capitalist production. The eagerness to utilise this surplus-value accumulating as virtual money-capital for the purpose of deriving profits or revenue from it finds its object accomplished in the credit system and "papers." Money-capital thereby gains in another form an enormous influence on the course and the stupendous development of the capitalist system of production. The surplus-product converted into virtual money-capital will grow so much more in volume, the greater was the total amount of already functioning capital whose functioning brought it into being. With the absolute increase of the volume of the annually reproduced virtual money-capital its segmentation also becomes easier, so that it is more rapidly invested in any particular business, either in the hands of the same capitalist or in those of others (for instance members of the family, in the case of a partition of inherited property, etc.). By segmentation of money-capital is meant here that it is wholly detached from the parent stock in order to be invested as a new money-capital in a new and independent business.

While the sellers of the surplus-product, A, A', A'', etc. (I), have obtained it as a direct outcome of the process of production, which does not envisage any additional acts of circulation except the advance of constant and variable capital required also in simple reproduction; and while they thereby construct the real basis for reproduction on an extended scale, and in actual fact manufacture virtually additional capital, the attitude of B, B', B'', etc. (I), is different. 1) Not until it reaches the hands of B, B', B'', etc. (I), will the surplus-product of A, A', A'', etc., actually function as additional constant capital (we leave out of consideration for the present the other element of productive capital, the additional labour-power, in other words, the additional variable capital).

2) In order that that surplus-product may reach their hands an act of circulation is wanted – they must buy it.

In regard to point 1 it should be noted here that a large portion of the surplus-product (virtually additional constant capital), although produced by A, A', A'' (I) in a given year, may not function as industrial capital in the hands of B, B', B'' (I) until the following year or still later. With reference to point 2, the question arises: Whence comes the money needed for the process of circulation?

Since the products created by B, B', B'', etc. (I), re-enter in kind into their own process, it goes without saying that pro tanto a portion of their own surplus-product is transferred directly (without any intervention of circulation) to their productive capital and becomes an additional element of constant capital. And pro tanto they do not effect the conversion of the surplus-product of A, A', etc. (I), into money. Aside from this, where does the money come from? We know that B, B', B'', etc. (I) have formed their hoard in the same way as A, A', etc., by the sale of their respective surplus-products. Now they have arrived at the point where their hoarded, only virtual, money-capital is to function effectively as additional money-capital. But this is merely going round in circles. The question still remains: Where does the money come from which the B's (I) before withdrew from circulation and accumulated? We know from the analysis of simple reproduction that capitalists I and II must have a certain amount of money at hand in order to be able to exchange their surplus-product. In that case the money which served only as revenue to be spent for articles of consumption returned to the capitalists in the same measure in which they had advanced it for the exchange of their respective commodities. Here the same money re-appears, but performing a different function. The A's and B's (I) supply one another alternately with the money for converting surplus-product into additional virtual money-capital, and throw the newly formed money-capital alternately back into circulation as a means of purchase. The only assumption made in this case is that the amount of money in the country in question (the velocity of circulation, etc., being constant) should suffice for both the active circulation and the reserve hoard. As we have seen this is the same assumption as had to be made in the case of the simple circulation of commodities. Only the function of the hoards is different in the present case. Furthermore, the available amount of money must be larger, first, because under capitalist production all the products (with the exception of newly produced precious metals and the few products consumed by the producer himself) are created as commodities and must therefore pass through the pupation stage of money; secondly, because on a capitalist basis the quantity of the commodity-capital and the magnitude of its value is not only absolutely greater but also grows with incomparably greater rapidity; thirdly, because an ever expanding variable capital must always be converted into money-capital; fourthly, because the formation of new money-capitals keeps pace with the extension of production, so that the material for corresponding hoard formation must be available.

This is generally true of the first phase of capitalist production, in which even the credit system is mostly accompanied by metallic circulation, and it applies to the most developed phase of the credit system as well, to the extent that metallic circulation remains its basis. On the one hand an additional production of precious metals, being alternately abundant or scarce, may here exert a disturbing influence on the prices of commodities not only at long, but also at very short intervals. On the other hand the entire credit mechanism is continually occupied in reducing the actual metallic circulation to a relatively more and more decreasing minimum by means of sundry operations, methods, and technical devices. The artificiality of the entire machinery and the possibility of disturbing its normal course increase to the same extent.

The different B's, B''s, B'''s, etc. (I), whose virtual new money-capital enters upon its function as active capital, may have to buy their products (portions of their surplus-product) from one another, or to sell them to one another. Pro tanto the money advanced by them for the circulation of their surplus-product flows back under normal conditions to the different B's in the same proportion in which they had advanced it for the circulation of their respective commodities. If the money circulates as a means of payment, then only balances are to be squared so far as the mutual purchases and sales do not cover one another. But it is important first and foremost to assume here, as everywhere, metallic circulation in its simplest, most primitive form, because then the flux and reflux, the squaring of balances, in short all elements appearing under the credit system as consciously regulated processes present themselves as existing independently of the credit system, and the matter appears in primitive form instead of the later, reflected form.

#### 3. The Additional Variable Capital

Hitherto we have been dealing only with additional constant capital. Now we must direct our attention to a consideration of the additional variable capital.

We have explained at great length in Book I that labour-power is always available under the capitalist system of production, and that more labour can be rendered fluent, if necessary, without increasing the number of labourers or the quantity of labour-power employed. We therefore need not go into this any further, but shall rather assume that the portion of the newly created money-capital capable of being converted into variable capital will always find at hand the labour-power into which it is to transform itself. It has also been explained in Book I that a given capital may expand its volume of production within certain limits without any accumulation. But here we are dealing with the accumulation of capital in its specific meaning, so that the expansion of production implies the conversion of surplus-value into additional capital, and thus also an expansion of the capital forming the basis of production.

The gold producer can accumulate a portion of his golden surplus- value as virtual money-capital. As soon as it becomes sufficient in amount, he can transform it directly into new variable capital, without first having to sell his surplus-product. He can likewise convert it into elements of the constant capital. But in the latter case he must find at hand the material elements of his constant capital. It is immaterial whether, as was assumed in our presentation hitherto, each producer works to stock up and then brings his finished product to the market or fills orders. The actual expansion of production, i.e., the surplus-product, is assumed in either case, in the one case as actually available, in the other as virtually available, capable of delivery.

##### II. Accumulation in Department II

We have hitherto assumed that A, A', A'' (I) sell their surplus-product to B, B', B'', etc., who belong to the same department I. But supposing A (I) converts his surplus-product into money by selling it to one B in department II. This can be done only by A (I) selling means of production to B (II) without subsequently buying articles of consumption, i.e., only by a one-sided sale on A's part. Now whereas II c cannot be converted from the commodity-capital form into the bodily form of productive constant capital unless not only I but also at least a portion of I, is exchanged for a portion of II c, which II exists in the form of articles of consumption; but now A converts his I into money by not making this exchange but rather withdrawing from circulation the money obtained from II on the sale of his I s instead of exchanging it in the purchase of articles of consumption II c – then what we have on the part of A (I) is indeed a formation of additional virtual money-capital, but on the other hand a portion of the constant capital of B (II) of equal magnitude of value is tied up in the form of commodity-capital, unable to transform itself into the bodily form of productive, constant capital. In other words, a portion of the commodities of B (II), and indeed primo facie a portion without the sale of which he cannot reconvert his constant capital entirely into its productive form, has become unsaleable. As far as this portion is concerned there is therefore an over-production, which, likewise as far as the same portion is concerned, clogs reproduction, even on the same scale.

In this case the additional virtual money-capital on the side of A (I) is indeed a moneyed form of surplus-product (surplus-value), but the surplus-product (surplus-value) considered as such is here a phenomenon of simple reproduction, not yet of reproduction on an extended scale. I(v+s) for which this is true at all events of one portion of s, must ultimately be exchanged for II c, in order that the reproduction of II c may take place on the same scale. By the sale of his surplus-product to B(II), A(I) has supplied to the latter a corresponding portion of the value of constant capital in its bodily form. But at the same time he has rendered an equivalent portion of the commodities of B (II) unsaleable by withdrawing the money from circulation – by failing to complement his sale through subsequent purchase. Hence, if we survey the entire social reproduction, which comprises the capitalists of both I and II, the conversion of the surplus-product of A (I) into virtual money-capital expresses the impossibility of reconverting commodity-capital of B (II) representing an equal amount of value into productive (constant) capital; hence not virtual production on an extended scale but an obstruction of simple reproduction, and so a deficit in simple reproduction. As the formation and sale of the surplus-product of A (I) are normal phenomena of simple reproduction, we have here even on the basis of simple reproduction the following interdependent phenomena: Formation of virtual additional money-capital in class I (hence under-consumption from the view-point of II); piling up of commodity-supplies in class II which cannot be reconverted into productive capital (hence relative over-production in II); surplus of money-capital in I and reproduction deficit in II.

Without pausing any longer at this point, we simply remark that we had assumed in the analysis of simple reproduction that the entire surplus-value of I and II is spent as revenue. As a matter of fact however one portion of the surplus-value is spent as revenue, and the other is converted into capital. Actual accumulation can take place only on this assumption. That accumulation should take place at the expense of consumption is, couched in such general terms, an illusion contradicting the nature of capitalist production. For it takes for granted that the aim and compelling motive of capitalist production is consumption, and not the snatching of surplus-value and its capitalisation, i.e., accumulation.

Let us now take a closer look at the accumulation in department II.

The first difficulty with reference to II c, i.e., its reconversion from a component part of commodity-capital II into the bodily form of constant capital II, concerns simple reproduction. Let us take the former scheme: (1,000 v + 1,000 s) I are exchanged for 2,000 II c.

Now, if for instance one half of the surplus-product of I, hence 1,000/2 s or 500 I s is reincorporated in department I as constant capital, then this portion of the surplus-product, being detained in I, cannot replace any part of II c. Instead of being converted into articles of consumption (and here in this section of the circulation between I and II the exchange is actually mutual, that is, there is a double change of position of the commodities, unlike the replacement of 1,000 II c by 1,000 I v effected by the labourers of I), it is made to serve as an additional means of production in I itself. It cannot perform this function simultaneously in I and II. The capitalist cannot spend the value of his surplus-product for articles of consumption and at the same time consume the surplus-product itself productively, i.e., incorporate it in his productive capital. Instead of 2,000 I(v+s), only 1,500, namely (1,000 v + 500 s) I, are therefore exchangeable for 2,000 II c ; 500 II c cannot be reconverted from the commodity-form into productive (constant) capital

II. Hence there would be an over-production in II, exactly equal in volume to the expansion of production in I. This over-production in II might react to such an extent on I that even the reflux of the 1,000 spent by the labourers of I for articles of consumption of II might take place but partially, so that these 1,000 would not return to the hands of capitalists I in the form of variable money-capital. These capitalists would thus find themselves hampered even in reproduction on an unchanging scale, and this by the bare attempt to expand it. And in this connection it must be taken into consideration that in I only simple reproduction had actually taken place and that its elements, as represented in our scheme, are only differently grouped with a view to expansion in the future, say, next year.

One might attempt to circumvent this difficulty in the following way: Far from being over-production, the 500 II c which are kept in stock by the capitalists and cannot be immediately converted into productive capital represent, on the contrary, a necessary element of reproduction, which we have so far neglected. We have seen that a money-supply must be accumulated at many points, hence money must be withdrawn from circulation, partly for the purpose of making it possible to form new money-capital in I, and partly to hold fast temporarily the value of the gradually depreciating fixed capital in the form of money. But since we placed all money and commodities from the very start exclusively into the hands of capitalists I and II when we drew up our scheme and since neither merchants, nor money-changers, nor bankers, nor merely consuming and not directly producing classes exist here, it follows that the constant formation of commodity stores in the hands of their respective producers is here indispensable to keep the machinery of reproduction going. The 500 II c held in stock by capitalists II therefore represent the commodity-supply of articles of consumption which ensures the continuity of the process of consumption implied in reproduction, here meaning the passage of one year to the next. The consumption-fund, which is as yet in the hands of its sellers who are at the same time its producers, cannot fall one year to the point of zero in order to begin the next with zero, any more than such a thing can take place in the transition from today to tomorrow. Since such supplies of commodities must constantly be built up anew, though varying in volume, our capitalist producers II must have a reserve money-capital, which enables them to continue their process of production although one portion of their productive capital is temporarily tied up in the shape of commodities. Our assumption is that they combine the whole business of trading with that of producing. Hence they must also have at their disposal the additional money-capital, which is in the hands of the merchants when the individual functions in the process of reproduction are separated and distributed among the various kinds of capitalists.

To this one may object:

1) That the forming of such supplies and the necessity of doing so applies to all capitalists, those of I as well as of II. Considered as mere sellers of commodities, they differ only in that they sell different kinds of commodities. A supply of commodities II implies a previous supply of commodities I. If we neglect this supply on one side, we must also do so on the other. But if we take them into account on both sides, the problem is not altered in any way.

2) Just as a certain year closes on the part of II with a supply of commodities for the following year, so it was opened with a supply of commodities on the same part, taken over from the preceding year. In an analysis of annual reproduction, reduced to its most abstract form, we must therefore strike it out in both cases. If we leave to the given year its entire production, including the commodity-supply to be yielded up for next year, and simultaneously take from it the supply of commodities transferred to it from the preceding year, we have before us the actual aggregate product of an average year as the subject of our analysis.

3) The simple circumstance that in the analysis of simple reproduction we did not stumble across the difficulty which is now to be surmounted proves that we are confronted by a specific phenomenon due solely to the different grouping (with reference to reproduction) of elements I, a changed grouping without which reproduction on an extended scale cannot take place at all.