The circuit of productive capital runs: P . . . C'-M'-C . . . P. This stands for the periodically renewed function of productive capital — that is, reproduction. It's the production process seen as a reproduction process, with an eye to how value grows: not just production, but the periodic reproduction of surplus-value. Industrial capital in its productive form doesn't function just once — it functions again and again, so that the starting point itself already gives the restart.
Part of C' can go straight back — in certain lines of industrial capital — as means of production into the very labour process it came out of as a commodity. When this happens, the only thing saved is the step of turning its value into actual money or money-tokens; at most it gets an independent expression as money of account. This part of the value never enters circulation at all. So some values enter the production process without ever entering the circulation process. The same holds for the part of C' that the capitalist consumes directly, in kind, out of the surplus product. But this case matters little for capitalist production as a whole — at most it comes up in agriculture.
Two things stand out at once in this form,
First. In the earlier form, M . . . M', the production process — the function of P — interrupted the circulation of money capital, and appeared only as the go-between linking its two phases, M-C and C'-M'. Here it's the other way round: the whole circulation process of industrial capital, its entire movement through the circulation phase, is itself only an interruption, and therefore only the mediation, between productive capital, which opens the circuit as the first extreme and closes it in the very same form, the form of its own restart, as the last extreme. Circulation proper now appears only as the way this periodically renewed, and so continuous, reproduction gets mediated.
Second. The whole circulation now shows up in the opposite form from the one it had in the circuit of money capital. There, leaving the value-determination aside, it was M-C-M (M-C, then C-M). Here, again leaving value-determination aside, it's C-M-C (C-M, then M-C) — the form of simple commodity circulation.
So let's look first at the process C'-M'-C, which runs between the extremes P . . . P within the sphere of circulation.
This circulation starts from the commodity-capital: C' = C + c = P + c. The function of the commodity-capital, C'-M' — realizing both the capital-value contained in it (=P, now existing as the commodity component C) and the surplus-value contained in it (existing as a component of that same mass of commodities, with value c) — was already looked at in the first form of the circuit. But there it was the second phase of an interrupted circulation, and the closing phase of the whole circuit. Here it's the second phase of the circuit, but the first phase of circulation.
The first circuit ended with M', and since M', just like the original M, can open a second circuit afresh as money capital, there was no need yet to ask whether the M and m (the surplus-value) contained in M' go on together or take separate paths. That question would only have mattered if we'd followed the first circuit further, into its renewal. But in the circuit of productive capital this point has to be settled, because how even its own first circuit turns out depends on it — and because C'-M' appears here as the first phase of circulation, to be completed by M-C. Whether the formula represents simple reproduction or reproduction on an expanded scale depends on this decision. So depending on how it's decided, the circuit changes its whole character.
So let's take first the simple reproduction of productive capital — assuming, as in the first chapter, unchanging conditions and that commodities are bought and sold at their value. On this assumption, the whole surplus-value goes into the capitalist's personal consumption. As soon as the commodity-capital C' has been turned into money, the part of that sum of money that represents the capital-value goes on circulating within the circuit of the industrial capital. The other part — the surplus-value now in gilded, money form — enters general commodity circulation. It's money circulation starting from the capitalist, but it runs outside the circulation of his individual capital.
In our example the commodity-capital C' was 10,000 lb of yarn worth £500. Of this, £422 is the value of the productive capital, and this — now in the money-form of 8,440 lb of yarn — carries on the circulation of capital that began with C'. The surplus-value of £78, the money-form of 1,560 lb of yarn, the extra part of the commodity product, steps outside that circulation and takes a separate path within general commodity circulation.
m-c is a series of purchases made with the money the capitalist lays out — whether on actual commodities or on services for himself or his family. These purchases are scattered, made at different times. So for a while the money sits as a stock, a hoard, set aside for ongoing consumption — money that's interrupted in its circulation is, by that very fact, in the form of a hoard. This function as means of circulation, including its passing form as a hoard, never enters the circulation of the capital in its money-form M. The money isn't being advanced here — it's being spent.
We've assumed that the whole capital advanced always passes completely from one of its phases into the next — so here too, that the commodity product of P carries the full value of the productive capital P (£422) plus the surplus-value created during production (£78). In our example, since we're dealing with a countable commodity product, the surplus-value exists in the form of 560 lb of yarn — just as, worked out per pound of yarn, it exists in the form of 2.496 ounces of yarn per pound.
But suppose instead the commodity product were, say, a machine worth £500, with the same value composition. Then one part of that machine's value would still be £78 of surplus-value — but those £78 would exist only within the machine as a whole. The machine can't be split into capital-value and surplus-value without literally breaking it into pieces, which would destroy its use-value and, with it, its value too. So the two value components can only be represented ideally, as parts picked out within the body of the commodity — not as independent elements of the commodity C', the way each pound of yarn is a separable, independent commodity-element among the 10,000 lb. In the machine's case, the whole commodity — the whole commodity-capital, the machine — has to be sold entirely before m can go its own separate way. Whereas if the capitalist sells 8,440 lb of yarn, selling the further 1,560 lb represents a completely separate circulation of the surplus-value in the form c (1,560 lb of yarn) - m (£78) = c (articles of consumption).
Still, the value elements of any single portion of the 10,000 lb of yarn product can be shown in parts of the product just as they can in the whole product. Just as the whole 10,000 lb of yarn can be divided into constant capital-value (c), 7,440 lb of yarn worth £372; variable capital-value (v), 1,000 lb of yarn worth £50; and surplus-value (s), 1,560 lb of yarn worth £78 — so each single pound of yarn divides into c = 11.904 ounces worth 8.928 d., v = 1.600 ounce worth 1.200 d., s = 2.496 ounces worth 1.872 d. The capitalist could even sell the 10,000 lb in stages and consume, stage by stage, the surplus-value elements contained in each successive portion, thereby realizing the sum of c + v just as gradually. But this operation still assumes, in the end, that the whole 10,000 lb gets sold — so that selling 8,440 lb replaces the value of c and v as well. (Volume 1, Chapter VII, 2.)
Whatever the case, through C'-M' both the capital-value contained in C' and the surplus-value acquire a separable existence — the existence of different sums of money. In both cases, M and m are the value's really transformed form, whereas in C' that same value originally had only an ideal expression, as the commodity's price.
c-m-c is simple commodity circulation, whose first phase, c-m, is included within the circulation of the commodity-capital C'-M' — that is, within the circuit of capital — while its complementary phase, m-c, falls outside that circuit, as a separate event within general commodity circulation. So the circulation of C and c — of capital-value and surplus-value — splits apart once C' has been turned into M'. From this it follows:
First: once the commodity-capital is realized through C'-M' = C'-(M+m), the movement of capital-value and surplus-value — up to now joint, carried together by the same mass of commodities — becomes splittable, because both now have independent forms as sums of money.
Second: if this split actually happens — with m spent as the capitalist's revenue, while M, as the functional form of the capital-value, carries on along the path set by the circuit — then that first act, C'-M', taken together with the acts that follow, M-C and m-c, can be represented as two different circulations: C-M-C and c-m-c. Both of these, in their general form, are just instances of ordinary commodity circulation.
In practice, with a continuous body that can't actually be divided, the value components still get isolated in thought. Take the London building trade, run mostly on credit: the builder gets advances as the house passes through its different stages. None of these stages is a house — each is only a really existing piece of a house still being made. So despite being real, each is only an ideal fraction of the whole house — yet real enough to serve as security for a further advance. (See Chapter XII below.)
Third: if the movement of capital-value and surplus-value — still joint in C and M — separates only partly (so that some of the surplus-value doesn't get spent as revenue), or doesn't separate at all, then a change happens in the capital-value itself, still within its own circuit, before that circuit is complete. In our example the value of the productive capital was £422. If it now carries on as M-C at, say, £480 or £500, it travels the remaining stages of the circuit as a value that's grown by £58 or £78 over what it started as. This can also come together with a change in its value composition.
C'-M', the second stage of circulation and the closing stage of circuit I (M . . . M'), is here the second stage of the circuit too, but the first stage of commodity circulation. As far as circulation goes, it still needs completing by M'-C'. But C'-M' has already left behind not just the valorization process (the function of P, the first stage) — its result, the commodity product C', is already realized. So both the capital's valorization process and the realization of the commodity product it's embodied in are already finished by the time we reach C'-M'.
So we've been assuming simple reproduction — that m-c splits off entirely from M-C. Since both circulations, c-m-c and C-M-C alike, belong in their general form to commodity circulation (and so show no value difference between their extremes), it's easy — as vulgar economics does — to picture the capitalist production process as simply the production of commodities, use-values meant for some kind of consumption, which the capitalist only produces in order to replace them with commodities of a different use-value, or exchange them for such — as vulgar economics wrongly puts it.
C appears from the outset as commodity-capital, and the purpose of the whole process — enrichment, valorization — doesn't rule out a level of consumption by the capitalist that grows along with the size of the surplus-value, and so with the capital itself; it positively includes it.
In the circulation of the capitalist's revenue, the produced commodity c (or the ideal fraction of the commodity product C' that corresponds to it) really does serve only to be turned first into money, and from money into a series of other commodities for private consumption. But one small point here must not be overlooked: c is commodity-value that cost the capitalist nothing — it's the embodiment of surplus labour, which is why it first appears on the scene as a component of the commodity-capital C'. So this c, just by existing, is tied to the circuit of the capital-value in process. If that circuit gets stuck, or is disturbed some other way, it's not just that the consumption of c is cut back or stops altogether — the market for the whole series of commodities that replace c shrinks with it. The same happens if C'-M' fails, or if only part of C' can be sold.
We saw that c-m-c, as the circulation of the capitalist's revenue, only enters the capital's circulation as long as c is a value-part of C', of capital in its functional form as commodity-capital. But once it's made independent through m-c — that is, in the full form c-m-c — it no longer enters the movement of the capital the capitalist has advanced, even though it arises out of it. It stays connected to that movement only insofar as the existence of the capital presupposes the existence of the capitalist, and the capitalist's existence is conditioned by consuming surplus-value.
Within general circulation, C' — say, yarn — functions only as a commodity. But as a moment in the circulation of capital, it functions as commodity-capital, a shape that the capital-value takes on and sheds in turn. Once the yarn is sold to the merchant, it's out of the circuit of the capital whose product it was — yet it still goes on existing as a commodity within the sphere of general circulation. The circulation of that same mass of commodities continues, even though it's stopped being a moment in the spinner's own independent circuit of capital. So the real, final metamorphosis of the mass of commodities the capitalist threw into circulation — C-M — its eventual falling out into consumption, can be separated in time and space from the metamorphosis in which that same mass functioned as his commodity-capital. The metamorphosis already completed within the circulation of capital still remains to be completed within the sphere of general circulation.
It changes nothing if the yarn instead goes back into the circuit of some other industrial capital. General circulation covers both things at once: the interweaving of the circuits of the different independent fragments of social capital — that is, the totality of individual capitals — and the circulation of values that aren't thrown onto the market as capital at all, the values that go into individual consumption.
The relation between the capital's circuit, so far as it's part of general circulation, and so far as it forms links of an independent circuit of its own, shows up again if we look at the circulation of M' = M + m. M, as money capital, carries on the circuit of capital. m, spent as revenue (m-c), enters general circulation but flies straight out of the capital's circuit. Only the part that functions as additional money capital re-enters that circuit. In c-m-c money functions only as coin; the purpose of this circulation is the capitalist's individual consumption. It's a mark of the crass confusion of vulgar economics that it passes off this circulation — which never enters the capital's circuit at all, the circulation of the part of the value-product consumed as revenue — as if it were the characteristic circuit of capital itself.
In this second phase, M–C, the capital value shows up again as a sum of money — M — equal to P, the value of the productive capital that opened this circuit. It has shed the surplus-value, so it's back to the same size it had at the very start of the money-capital circuit, M–C. The place in the story is different, but the job this money capital does is the same: it turns into means of production and labour-power, mp and L.
At the same time as c–m, the capital value — now functioning as commodity capital C'–M' — has gone through the phase C–M, and now enters the complementary phase M–C<L/mp>. So its whole circuit is C–M–C<L/mp>.
In Form I (the circuit M...M'), money capital M appeared as the starting form — the value advanced at the outset. Here it's different from the start: it appears as part of the sum of money that the commodity capital turned into during the first phase, C'–M'. In other words, it only exists here because a sale has already happened — the sale of the product converted P, the productive capital, into money form. So this money capital is, from the outset, neither the original form of the capital value nor its final form: only by shedding its money form again can the phase M–C complete the phase C–M that ended in it.
That's why the part of M–C that is also M–L — the purchase of labour-power — no longer appears as a mere money-advance to buy labour-power. It's an advance in which the worker is handed, in money form, the very same 1,000 lb of yarn, worth £50, that make up part of the value his own labour created. The money paid to the worker here is nothing but the converted equivalent-form of a portion of the value of the product he himself produced.
And for exactly this reason, the act M–C, so far as it is M–L, is not simply commodity-in-money-form being swapped for commodity-in-use-form. It brings in other elements, ones that have nothing to do with commodity circulation in general.
M' appears as the converted form of C', which is itself the product of P's past activity — the production process. So the whole sum M' is simply the money-expression of past labour. In our example: 10,000 lb of yarn = £500, the product of the spinning process. Of that, 7,440 lb of yarn = the advanced constant capital, £372; 1,000 lb of yarn = the advanced variable capital, £50; and 1,560 lb of yarn = the surplus-value, £78. If, other things staying the same, only the original £422 of capital is advanced again out of M', then in M–L the worker only gets, advanced to him next week, part of the 10,000 lb of yarn produced this week — the money-value of 1,000 lb. As the result of C–M, money is always an expression of past labour. And so far as the complementary act M–C happens right away on the market — money exchanged for commodities already sitting there — it's again just an exchange of past labour, moving from one form (money) into another (commodity).
But M–C happens at a different time than C–M. They can coincide, as an exception — say when two capitalists hand each other their goods at the same moment, and the money only settles the difference between them. Otherwise the time-gap between carrying out C–M and carrying out M–C can be more or less large. Even though, as the result of C–M, money represents past labour, for the act M–C that same money can represent commodities that aren't on the market at all yet — that will only be there in the future — since M–C only needs to happen after C has been produced anew. Equally, the money can represent commodities being produced at the very same time as the C whose money-expression it is. For example, in the purchase of means of production, coal can be bought before it's even been brought up out of the mine. So far as this money is being accumulated rather than spent as revenue, it can represent cotton that won't be produced until next year — and the same holds when the capitalist spends his revenue.
The same is true of the wage, £50. This money is not only the money-form of the workers' past labour — it is at the same time a draft on labour that is happening now or will happen in the future, labour not yet realized, or still to be realized. The worker might use it to buy a coat that will only be made next week. This holds above all for the great mass of necessary means of subsistence that have to be consumed almost the instant they're produced, on pain of spoiling. So in the money he's paid his wage in, the worker receives the converted form of his own future labour, or of other workers' future labour. With a portion of his past labour, the capitalist hands him a draft on his own future labour: it is his own labour, now or still to come, that forms the stock — not yet existing — with which his past labour gets paid. The whole idea of a stock being laid by beforehand disappears completely here.
Second: in the circulation C–M–C, the same money changes hands twice. The capitalist first gets it as a seller, then gives it away as a buyer. Turning a commodity into money only serves to turn that money back into a commodity. So the money-form of capital — its existence as money capital — is, in this movement, only a vanishing moment. Or rather: so far as this movement flows smoothly, money capital shows up only as a means of circulation, when it serves to buy. It shows up as a genuine means of payment when capitalists buy from one another, so that only the balance of payments has to be settled.
Third: whether money capital works as a mere means of circulation or as a means of payment, its function is only to bring about the replacement of C by L and mp — that is, the replacement of the yarn, the commodity-product the productive capital results in (after subtracting the surplus-value that will be spent as revenue), by its own elements of production. In other words, it turns the capital value back, out of its form as a commodity, into the elements that make up that commodity. So in the end its whole function is just to convert commodity capital back into productive capital.
For the circuit to run its normal course, C' has to be sold at its full value and in its entirety. And C–M–C involves not just replacing one commodity with another, but replacing it under the same value-relations. That's the assumption we're making here. In fact, though, the values of means of production do vary — capitalist production is marked by a constant shifting of value-relations, precisely because the productive power of labour is itself constantly changing. We'll only point to this shifting of the value of the factors of production here; it needs to be discussed later.
The conversion of the elements of production into the commodity-product, P into C', takes place in the sphere of production; the reconversion of C' back into P takes place in the sphere of circulation. That reconversion happens through the simple metamorphosis of commodities. But its content is a moment of the reproduction process taken as a whole. C–M–C, as capital's form of circulation, includes a functionally determined metabolism.
The exchange C–M–C further requires that C equal the elements of production of the quantity of commodities C', and that these elements hold onto their original value-relations to one another. So it assumes not only that commodities are bought at their value, but also that they undergo no change in value during the circuit. If that's not the case, the process can't run its normal course.
In M...M', M is the original form of the capital value — a form it sheds, only to take it up again later. In P...C'–M'–C...P, money is merely a form the process assumes, and one it sheds again within that very process. Here the money-form shows up only as a vanishing, independent value-form of capital. Capital is just as anxious to take on this form as C' as it is to shed it again as M', the moment it has pupated into money-form, so as to turn itself back into productive capital. As long as it stays in its money-shape, it isn't functioning as capital, and so isn't valorizing itself — the capital lies idle. Here money acts as a means of circulation, but specifically as a means of circulating capital.
The appearance of independence that the money-form of the capital value carries in the first form of its circuit — the money-capital circuit — disappears in this second form. This second form is thus the critique of the first: it reduces Form I to a merely particular form. If the second metamorphosis, M–C, runs into obstacles — say the means of production simply aren't to be had on the market — then the circuit, the flow of the reproduction process, is interrupted, just as much as if the capital were stuck fast in commodity-capital form.
But there is a difference. Capital can hold out longer in money-form than in the perishable form of a commodity. It doesn't stop being money just because it isn't functioning as money capital — but it does stop being a commodity, and a use-value at all, if it's held up too long in its function as commodity capital. And second, in money-form capital is able to take on some other form instead of returning to its original productive shape, whereas as C' it can't budge from the spot at all.
C'–M'–C, by its form, only involves circulation-acts that are moments of C''s own reproduction. But the actual reproduction of the C that C' turns into is necessary for C'–M'–C to be carried out at all — and that reproduction is itself conditioned by reproduction processes lying outside the reproduction process of the individual capital represented by C'.
In Form I, M–C only prepares the first conversion of money capital into productive capital. In Form II, it's the reconversion from commodity capital into productive capital — so, so long as the scale of the industrial capital stays the same, it's the reconversion of commodity capital back into the very same elements of production it came out of. So here, as in Form I, it appears as the phase preparing for the production process — but as a return to that process, a renewal of it. It's therefore the forerunner of the reproduction process, and so also of the repetition of the process of valorization.
It's worth noting again: M–L is not a simple exchange of commodities. It's the purchase of a commodity, labour-power, meant to serve the production of surplus-value — just as M–mp is only a step that's materially indispensable for carrying out that same purpose.
Once M–C is carried out, M has been reconverted into productive capital, into P, and the circuit starts over again.
The explicit form of P...C'–M'–C...P is therefore:
Turning money capital into productive capital means buying commodities in order to produce commodities. Only so far as this consumption is productive consumption does it belong within capital's own circuit — and the condition for that is that surplus-value gets made by means of the commodities consumed this way.
That is something very different from production in general, even from commodity production as such, whose purpose is simply the producers' own existence. Replacing one commodity with another under this condition — that it serves the production of surplus-value — is something quite different from what a mere exchange of products, merely mediated by money, is in itself.
But this is exactly how economists take the matter, to prove that overproduction is impossible.
Besides the productive consumption of M as it turns into L and mp, the circuit contains the first leg of M–L — which, from the worker's side, is L–M, that is, C–M. Of the worker's own circulation, L–M–C, which includes his consumption, only the first leg — the result of M–L — falls within capital's circuit. The second act, M–C, does not belong to the circulation of the individual capital, even though it arises out of it. But the continued existence of the working class is necessary for the capitalist class — and so is the worker's consumption that M–C brings about.
The act C'–M' requires, both for the capital value's circuit to continue and for the capitalist's consumption of surplus-value, only that C' has been converted into money — sold. Of course it only gets bought because it's a use-value, fit for some kind of consumption, whether productive or individual. But if C' keeps circulating further — say, in the hands of the merchant who's bought the yarn — that, in the first instance, has no bearing at all on the continuation of the circuit of the individual capital that produced the yarn and sold it to the merchant. The whole process keeps going, and with it the individual consumption of capitalist and worker that depends on it. This is an important point when it comes to crises.
As soon as C' is sold, turned into money, it can be reconverted into the real factors of the labour process, and so of the reproduction process. Whether C' is bought by the final consumer or by a merchant who wants to resell it makes no immediate difference to the matter. The volume of commodity-masses capitalist production turns out is set by the scale of that production and by its need to keep expanding — not by some predetermined circle of supply and demand, of needs waiting to be satisfied. Mass production's immediate buyer can, apart from other industrial capitalists, only be the big merchant. Within certain limits, the reproduction process can go on at the same or an expanded scale even though the commodities it throws out haven't actually entered individual or productive consumption. The consumption of commodities is not included in the circuit of the capital they came from.
Once the yarn, say, is sold, the circuit of the capital value it represents can start over, whatever becomes of the sold yarn afterward. As long as the product is sold, everything runs its regular course from the standpoint of the capitalist producer. The circuit of the capital value he represents is not interrupted. And if this process runs on an expanded scale — which includes expanded productive consumption of means of production — this reproduction of capital can be accompanied by expanded individual consumption, that is, demand, on the part of the workers, since the process itself is set off and mediated by productive consumption. So the production of surplus-value can grow, and with it the capitalist's individual consumption too, and the whole reproduction process can be in its most flourishing state — and yet a large part of the commodities may only appear to have entered consumption, while actually lying unsold in the hands of resellers, in fact still sitting on the market.
Now stream of commodities follows stream of commodities, and it finally comes out that the earlier stream had only seemed to be swallowed up by consumption. The commodity-capitals fight each other for their place on the market. The latecomers, to sell at all, sell under the price. The earlier streams haven't yet been turned into cash, while the deadlines for paying on them fall due. Their holders have to declare themselves insolvent, or sell at any price, just to pay. This kind of selling has absolutely nothing to do with the real state of demand. It has to do only with the demand for payment — the absolute necessity of turning commodities into money. Then the crisis breaks out. It becomes visible not in an immediate fall in demand for consumption, for individual use, but in the fall of the exchange of capital against capital — of capital's own reproduction process.
Suppose mp and L — the commodities M turns into so as to carry out its function as money capital, as capital value destined to reconvert into productive capital — have to be bought or paid for at different dates, so that M–C stands for a whole series of purchases and payments happening one after another. Then part of M carries out the act M–C, while another part stays in its money-state, waiting to serve, at a time set by the conditions of the process itself, for simultaneous or later acts of M–C. This part is only temporarily withdrawn from circulation, so as to spring into action at the fixed moment and do its job. This holding-back is itself, then, a function set by its circulation and for its circulation. Its existence as a fund for buying and paying, the suspension of its movement, this state of interrupted circulation — that is a state in which money is carrying out one of its functions as money capital.
As money capital: because in this case the money temporarily lying still is itself part of the money capital M (from M' − m = M), part of the value of the commodity capital equal to P, the value of the productive capital the circuit starts from. On the other hand, all money withdrawn from circulation takes the form of a hoard. So the hoard-form of money here becomes a function of money capital — just as, in M–C, the function of money as means of purchase or payment becomes a function of money capital — because here the capital value exists in money form, and this money-state is a state the coherence of the circuit prescribes for the industrial capital at one of its stages. And this once again confirms that money capital, within the circuit of industrial capital, performs no functions other than money functions — and that these money functions only take on the significance of capital functions through their connection with the other stages of the circuit.
Representing M' as the ratio of m to M — as a capital-relation — is not directly a function of money capital at all. It's a function of commodity capital, C', which itself, as the ratio of c to C, only expresses the result of the production process: the self-valorization of the capital value that took place within it.
If the circulation process runs into obstacles, so that external circumstances — the state of the market and so on — force M to suspend its function M–C, and it therefore stays in its money-state for a shorter or longer time, that is again a hoard-state of money, the same kind that also occurs in simple commodity circulation whenever the passage from C–M to M–C is interrupted by external circumstances. It is involuntary hoard-formation. In our case, the money thereby takes the form of idle, latent money capital. But we won't go further into this for now.
In both cases, though, money capital staying put in its money-state shows up as the result of an interrupted movement — whether that interruption fits the purpose or works against it, is voluntary or involuntary, in keeping with its function or contrary to it.
The proportions in which production can expand aren't arbitrary — they're fixed by technique. So realized surplus-value, even though it's meant to become capital, often can't reach the size it needs until several cycles have repeated themselves. Until then it must pile up as a hoard — held back because the technical minimum for additional capital has not yet been reached.
So the surplus-value hardens into a hoard, and in this form it is latent money capital. Latent, because as long as it stays in money form, it cannot act as capital.
Treasure-forming shows up here as something that belongs inside the capitalist accumulation process, that goes along with it — but is also essentially different from it. Forming latent money capital does not by itself expand reproduction. The opposite: latent money capital forms here precisely because the capitalist producer cannot yet expand the scale of production directly.
If he sells his surplus product to a gold or silver producer who throws new gold or silver into circulation — or, which comes to the same thing, to a merchant who imports additional gold or silver from abroad in exchange for part of the national surplus product — then his latent money capital adds to the national stock of gold or silver. In every other case, say the £78 that were means of circulation in the buyer's hands simply take on the form of a hoard in the capitalist's hands: only the distribution of the existing national gold or silver stock has changed.
Suppose money functions in our capitalist's dealings as a means of payment — the buyer only has to pay at some later date, near or far. Then the surplus product meant for capitalization doesn't turn into money at all. It turns into debt claims, titles of ownership to an equivalent the buyer may already have, or may only expect to get later.
This claim does not enter the reproduction process of the circuit — no more than money invested in interest-bearing paper does, even though that money can enter the circuit of some other individual industrial capital.
The whole character of capitalist production is set by the valorization of the capital value advanced. That means, first, producing as much surplus-value as possible. But second (see Volume 1, chapter 22) it means producing capital: turning surplus-value into capital.
Accumulation, production on an expanded scale, appears as a means to ever-greater production of surplus-value, and so to the capitalist's own enrichment — his personal purpose, part of capitalist production's general drive. But it becomes something more: as capitalist production develops — the first volume showed this — expansion becomes a necessity for every individual capitalist. Constantly enlarging his capital becomes the condition for keeping it at all. But we need not go back over what was set out earlier.
We began by looking at simple reproduction, where we assumed the whole of the surplus-value gets spent as revenue. In reality, under normal conditions, part of the surplus-value must always be spent as revenue and another part must be capitalized — it makes no difference whether, in any given period, the surplus-value produced is sometimes entirely consumed and sometimes entirely capitalized. Averaged over the whole movement — and the general formula can only represent that average — both things happen together.
Still, to keep the formula simple, it's better to assume the whole surplus-value is accumulated. The formula P . . . C'-M'-C' . . . P' then says: productive capital is reproduced on a larger scale, with a larger value, and as this enlarged productive capital it begins its second circuit — or, the same thing, renews its first circuit.
As soon as this second circuit begins, P is again the starting point — only now P is a larger productive capital than the first P was. In the same way, in the formula M . . . M', if the second circuit begins with M', then M' functions simply as M: an advanced money capital of a given size. It is a larger sum of money capital than the one that opened the first circuit — but every trace of how it grew, through capitalizing surplus-value, has disappeared the moment it appears as advanced money capital opening a circuit. That origin is wiped out in its form as money capital starting a fresh circuit. The same holds for P', the moment it functions as the starting point of a new circuit.
Compare P . . . P' with M . . . M' — or with the first circuit — and they do not mean the same thing at all. M . . . M', taken by itself as an isolated circuit, only says this: that M, money capital (or industrial capital in its circuit-form as money capital), is money that breeds money, value that breeds value — it posits surplus-value.
In the circuit of P, though, the valorization process itself is already completed by the end of the first stage, the production process. And once the second stage — the first stage of circulation, C'-M' — has run its course, capital-value plus surplus-value already exist as realized money capital, as M', which appeared as the last term in the first circuit.
That surplus-value has been produced was already shown, in the earlier form P . . . P, by c-m-c — whose second stage falls outside the circulation of capital and represents the circulation of the surplus-value as revenue. In this form, where the whole movement is represented as P . . . P, so that there is no difference in value between the two endpoints, the valorization of the advanced value — the generation of surplus-value — is expressed just as fully as it is in M . . . M'. Only the act C'-M' appears differently: as the final stage in M . . . M', but as the second stage of the circuit — the first stage of circulation — in P . . . P.
In P . . . P', P' expresses something different: not that surplus-value has been produced, but that the surplus-value produced has been capitalized — that capital has been accumulated. So P', compared with P, consists of the original capital value plus the value of the capital that its movement has accumulated.
M', as the mere conclusion of M . . . M', and C', as it appears within all these circuits, do not by themselves express the movement — they express its result: the valorization of the capital value, realized in commodity-form or money-form. That means capital value as M + m, or as C + c — as the relation of the capital value to its surplus-value, its offshoot. They express this result as different circulation-forms of the valorized capital value.
But neither in the form C' nor in the form M' is the valorization that has taken place itself a function — not of money capital, and not of commodity capital either. As particular, distinct forms of existence, corresponding to the particular functions of industrial capital, money capital can only perform money-functions, and commodity capital only commodity-functions; the difference between them is only the difference between money and a commodity. In the same way, industrial capital in its form as productive capital can only consist of the same elements as any other labour process that produces a product: on one side, objective conditions of labour — means of production; on the other, labour-power activating itself productively, purposefully. Just as industrial capital, within the sphere of production, can only exist in the composition that belongs to the production process as such — and so belongs equally to a non-capitalist production process — so too, in the sphere of circulation, it can only exist in the two forms that belong there: commodity and money.
But just as the sum of the elements of production announces itself as productive capital, from the very start, only because the labour-power is someone else's labour-power, which the capitalist has bought from its own owner — exactly as he bought his means of production from other commodity-owners — and just as the production process itself therefore appears as the productive function of industrial capital, so too money and commodity appear as circulation-forms of that same industrial capital, and their functions appear as its circulation-functions: functions that either open the way to the functions of productive capital, or spring from them. Only through this connection — as functional forms that industrial capital has to go through, in the different stages of its circuit — are the money-function and the commodity-function, at the same time, functions of money capital and of commodity capital.
So it is wrong to try to derive the specific properties and functions that characterize money as money, and the commodity as commodity, from their capital-character. And it is equally wrong, the other way round, to derive the properties of productive capital from its mode of existence in means of production.
Once M' or C' is fixed as M + m, or C + c — that is, as the relation of the capital value to the surplus-value that sprang from it — this relation is expressed in both: once in money-form, once in commodity-form. That changes nothing about the thing itself. So this relation springs neither from properties and functions belonging to money as such, nor from properties belonging to the commodity as such. In both cases, the property that characterizes capital — being value that breeds value — is expressed only as a result.
C' is always the product of the function of P, and M' is always only the transformed, circuit-form of C' within the circuit of industrial capital. So the moment the realized money capital begins its own particular function as money capital again, it stops expressing the capital-relation contained in M' = M + m. Once M . . . M' has run its course and M' begins the circuit afresh, it no longer figures as M' but as M — even if the whole surplus-value contained in M' gets capitalized.
In our example, the second circuit begins with a money capital of £500, instead of the £422 that opened the first. The money capital opening the circuit is £78 larger than before. That difference only shows up when we compare one circuit with the other — it does not exist inside either circuit taken on its own. The £500 advanced as money capital — £78 of which used to exist as surplus-value — plays no different role than £500 with which some other capitalist opens his first circuit. The same holds for the circuit of productive capital: the enlarged P', at the renewed start, appears simply as P — just as P does in simple reproduction, P . . . P.
In the stage M'-C', the increased size is shown only by C' — not separately by L' and mp'. Since C is the sum of L and mp, C' already shows that the sum of L and mp it contains is larger than in the original P.
But calling them L' and mp' would be wrong for a second reason too: we know that as capital grows, its value-composition changes along with it. As this goes on, the value of mp grows while the value of L keeps falling relatively — often even in absolute terms.
Whether m — the surplus-value now in gold, so to speak — can be added straight back to the capital value in process, entering the circuit together with M as the magnitude M', depends on circumstances that have nothing to do with the mere existence of m.
Suppose m is meant to serve as money capital in a second, independent business alongside the first. Then it's obviously only usable once it reaches the minimum size that business requires. Suppose instead it's meant to extend the original business. Then the material makeup of P's factors, and their value-relations, likewise fix a minimum size m must reach. All the means of production working in a business stand to each other not just in a certain kind of relation but in a definite quantitative one — a proportional scale. These material relations, and the value-relations that ride on them, fix the minimum size m must have before it can be converted into additional means of production and labour-power — or just the former — as an increment to productive capital.
So a spinner cannot add to his number of spindles without also buying the matching carding engines and roving frames, quite apart from the extra outlay on cotton and wages that such an expansion requires. To carry this out, the surplus-value already has to amount to a fair sum — roughly £1 per newly acquired spindle is the usual reckoning. As long as m hasn't reached this minimum size, the circuit of capital has to repeat itself several times, until the sum of the m's it successively throws off, together with M, is large enough to function in M'-C'. Even small changes of detail — say, in spinning machinery, where they make it more productive — call for greater outlay on spinning material, more roving machinery, and so on.
In the meantime, then, m piles up — but its piling up is not its own function. It is the result of repeated cycles of P . . . P. Its own function, in the meantime, is simply to sit still in the state of money, until it has received enough addition from outside — from those repeated cycles of valorization — to reach the minimum size its active function requires: the size at which alone it can actually enter, as money capital, into the function of the money capital M already at work, as an accumulated part of it.
In the meantime it is accumulated, and exists only in the form of a hoard that is forming, that is growing. Money accumulation, treasure-forming, appears here, then, as a process that temporarily accompanies the real accumulation — the extension of the scale on which industrial capital works. Temporarily: because as long as the hoard stays in its hoard-state, it does not function as capital. It takes no part in the valorization process. It stays a sum of money that only grows because, with no doing of its own, more money keeps landing in the same chest.
The hoard-form is only the form taken by money that is not in circulation — money whose circulation has been interrupted, and which is therefore kept sitting in its money-form.
As for the process of treasure-forming itself, it is common to all commodity production, and only plays a role as an end in itself in commodity production's undeveloped, pre-capitalist forms. Here, though, the hoard appears as a form of money capital, and treasure-forming appears as a process that temporarily accompanies the accumulation of capital — because, and only insofar as, the money here figures as latent money capital.
That is: treasure-forming, the hoard-state of the surplus-value that exists in money-form, is a preparatory stage for turning that surplus-value into really functioning capital — a stage that happens outside the circuit of capital, fixed in its function by that role. It is latent money capital because of this, its function — which is also why the size it has to reach before it can enter the process is set by whatever the value-composition of the productive capital happens to be at the time.
But as long as it stays in the hoard-state, it does not yet function as money capital. It is still dormant money capital — not, as before, money capital interrupted in a function it already had, but money capital not yet capable of any function at all.
We are taking the accumulation of money here in its original, real form — as an actual hoard of money. It can also exist in the form of mere credit balances: debt claims held by the capitalist who has sold C'.
As for the other forms — where this latent money capital exists in the meantime as money breeding money in its own right, say as an interest-bearing deposit in a bank, or in bills of exchange or securities of some kind — those don't belong here. The surplus-value realized in money is then performing particular capital-functions outside the circuit of the industrial capital it sprang from: functions that, first, have nothing to do with that circuit as such, and second, presuppose capital-functions distinct from those of industrial capital — functions not yet developed at this point.
In the form we've just looked at, the hoard in which the surplus-value exists is a money-accumulation fund — the money-form that capital accumulation temporarily wears, and to that extent itself a condition for that accumulation. But this accumulation fund can also do particular side-jobs: it can enter the circuit process of capital without that process taking the form P . . . P' — that is, without capitalist reproduction actually being expanded.
Suppose the process C'-M' is drawn out beyond its normal length — the commodity capital gets abnormally held up on its way into money-form. Or suppose that conversion does go through, but, say, the price of the means of production the money capital has to be turned into has risen above what it was when the circuit began. In either case, the hoard functioning as an accumulation fund can be used to stand in for the money capital, or part of it. The money-accumulation fund then serves as a reserve fund, evening out disturbances in the circuit.
As a reserve fund of this kind, it is different from the fund of means of purchase or payment we looked at within the circuit P . . . P. That fund is part of the functioning money capital — a form taken by part of the capital value in process generally — whose portions simply come into play one after another, at different points in time. It forms constantly, in the ongoing continuity of the production process: payments come in today that only have to be made again at a later date; larger batches of goods get sold today, and only have to be bought again, in larger batches, on some later day. In these intervals, a part of the circulating capital constantly sits in money-form.
The reserve fund is different. It is not a component of the functioning capital — of the functioning money capital, more precisely — but of capital still caught in an earlier stage of its accumulation: surplus-value that has not yet been turned into active capital. Of course, a capitalist in trouble doesn't stop to ask which particular function some money in his hand was supposed to have — he uses whatever he's got to keep his capital's circuit going.
In our example, M = £422, M' = £500. If part of the £422 exists as a fund of means of payment and purchase — a stock of ready money on hand — it is calculated so that, if circumstances stay the same, the whole of it enters the circuit, and is enough for that purpose. The reserve fund, though, is part of the £78 of surplus-value. It can only enter the circuit process of the £422 capital to the extent that this circuit is carried out under circumstances that are not staying the same — because it is part of the accumulation fund, and figures here without any expansion of the scale of reproduction.
The money-accumulation fund is already the existence of latent money capital — a transformation, that is, of money into money capital.
The general formula for the circuit of productive capital — the one that covers both simple reproduction and reproduction on an expanded scale — is:
If P equals P — the reading rule for the same scale: the second circuit renews unchanged, the surplus m spent as revenue — then M in (2) equals M' minus m. But if P equals P' — if the scale has expanded — then M in (2) is larger than M' minus m. In other words: m has been wholly or partly turned into money capital.
The circuit of productive capital is the form in which classical political economy looks at the circuit process of industrial capital.
The circuit of productive capital runs: P . . . C'-M'-C . . . P. This stands for the periodically renewed function of productive capital — that is, reproduction. It's the production process seen as a reproduction process, with an eye to how value grows: not just production, but the periodic reproduction of surplus-value. Industrial capital in its productive form doesn't function just once — it functions again and again, so that the starting point itself already gives the restart.
Part of C' can go straight back — in certain lines of industrial capital — as means of production into the very labour process it came out of as a commodity. When this happens, the only thing saved is the step of turning its value into actual money or money-tokens; at most it gets an independent expression as money of account. This part of the value never enters circulation at all. So some values enter the production process without ever entering the circulation process. The same holds for the part of C' that the capitalist consumes directly, in kind, out of the surplus product. But this case matters little for capitalist production as a whole — at most it comes up in agriculture.
Two things stand out at once in this form,
First. In the earlier form, M . . . M', the production process — the function of P — interrupted the circulation of money capital, and appeared only as the go-between linking its two phases, M-C and C'-M'. Here it's the other way round: the whole circulation process of industrial capital, its entire movement through the circulation phase, is itself only an interruption, and therefore only the mediation, between productive capital, which opens the circuit as the first extreme and closes it in the very same form, the form of its own restart, as the last extreme. Circulation proper now appears only as the way this periodically renewed, and so continuous, reproduction gets mediated.
Second. The whole circulation now shows up in the opposite form from the one it had in the circuit of money capital. There, leaving the value-determination aside, it was M-C-M (M-C, then C-M). Here, again leaving value-determination aside, it's C-M-C (C-M, then M-C) — the form of simple commodity circulation.
So let's look first at the process C'-M'-C, which runs between the extremes P . . . P within the sphere of circulation.
This circulation starts from the commodity-capital: C' = C + c = P + c. The function of the commodity-capital, C'-M' — realizing both the capital-value contained in it (=P, now existing as the commodity component C) and the surplus-value contained in it (existing as a component of that same mass of commodities, with value c) — was already looked at in the first form of the circuit. But there it was the second phase of an interrupted circulation, and the closing phase of the whole circuit. Here it's the second phase of the circuit, but the first phase of circulation.
The first circuit ended with M', and since M', just like the original M, can open a second circuit afresh as money capital, there was no need yet to ask whether the M and m (the surplus-value) contained in M' go on together or take separate paths. That question would only have mattered if we'd followed the first circuit further, into its renewal. But in the circuit of productive capital this point has to be settled, because how even its own first circuit turns out depends on it — and because C'-M' appears here as the first phase of circulation, to be completed by M-C. Whether the formula represents simple reproduction or reproduction on an expanded scale depends on this decision. So depending on how it's decided, the circuit changes its whole character.
So let's take first the simple reproduction of productive capital — assuming, as in the first chapter, unchanging conditions and that commodities are bought and sold at their value. On this assumption, the whole surplus-value goes into the capitalist's personal consumption. As soon as the commodity-capital C' has been turned into money, the part of that sum of money that represents the capital-value goes on circulating within the circuit of the industrial capital. The other part — the surplus-value now in gilded, money form — enters general commodity circulation. It's money circulation starting from the capitalist, but it runs outside the circulation of his individual capital.
In our example the commodity-capital C' was 10,000 lb of yarn worth £500. Of this, £422 is the value of the productive capital, and this — now in the money-form of 8,440 lb of yarn — carries on the circulation of capital that began with C'. The surplus-value of £78, the money-form of 1,560 lb of yarn, the extra part of the commodity product, steps outside that circulation and takes a separate path within general commodity circulation.
m-c is a series of purchases made with the money the capitalist lays out — whether on actual commodities or on services for himself or his family. These purchases are scattered, made at different times. So for a while the money sits as a stock, a hoard, set aside for ongoing consumption — money that's interrupted in its circulation is, by that very fact, in the form of a hoard. This function as means of circulation, including its passing form as a hoard, never enters the circulation of the capital in its money-form M. The money isn't being advanced here — it's being spent.
We've assumed that the whole capital advanced always passes completely from one of its phases into the next — so here too, that the commodity product of P carries the full value of the productive capital P (£422) plus the surplus-value created during production (£78). In our example, since we're dealing with a countable commodity product, the surplus-value exists in the form of 560 lb of yarn — just as, worked out per pound of yarn, it exists in the form of 2.496 ounces of yarn per pound.
But suppose instead the commodity product were, say, a machine worth £500, with the same value composition. Then one part of that machine's value would still be £78 of surplus-value — but those £78 would exist only within the machine as a whole. The machine can't be split into capital-value and surplus-value without literally breaking it into pieces, which would destroy its use-value and, with it, its value too. So the two value components can only be represented ideally, as parts picked out within the body of the commodity — not as independent elements of the commodity C', the way each pound of yarn is a separable, independent commodity-element among the 10,000 lb. In the machine's case, the whole commodity — the whole commodity-capital, the machine — has to be sold entirely before m can go its own separate way. Whereas if the capitalist sells 8,440 lb of yarn, selling the further 1,560 lb represents a completely separate circulation of the surplus-value in the form c (1,560 lb of yarn) - m (£78) = c (articles of consumption).
Still, the value elements of any single portion of the 10,000 lb of yarn product can be shown in parts of the product just as they can in the whole product. Just as the whole 10,000 lb of yarn can be divided into constant capital-value (c), 7,440 lb of yarn worth £372; variable capital-value (v), 1,000 lb of yarn worth £50; and surplus-value (s), 1,560 lb of yarn worth £78 — so each single pound of yarn divides into c = 11.904 ounces worth 8.928 d., v = 1.600 ounce worth 1.200 d., s = 2.496 ounces worth 1.872 d. The capitalist could even sell the 10,000 lb in stages and consume, stage by stage, the surplus-value elements contained in each successive portion, thereby realizing the sum of c + v just as gradually. But this operation still assumes, in the end, that the whole 10,000 lb gets sold — so that selling 8,440 lb replaces the value of c and v as well. (Volume 1, Chapter VII, 2.)
Whatever the case, through C'-M' both the capital-value contained in C' and the surplus-value acquire a separable existence — the existence of different sums of money. In both cases, M and m are the value's really transformed form, whereas in C' that same value originally had only an ideal expression, as the commodity's price.
c-m-c is simple commodity circulation, whose first phase, c-m, is included within the circulation of the commodity-capital C'-M' — that is, within the circuit of capital — while its complementary phase, m-c, falls outside that circuit, as a separate event within general commodity circulation. So the circulation of C and c — of capital-value and surplus-value — splits apart once C' has been turned into M'. From this it follows:
First: once the commodity-capital is realized through C'-M' = C'-(M+m), the movement of capital-value and surplus-value — up to now joint, carried together by the same mass of commodities — becomes splittable, because both now have independent forms as sums of money.
Second: if this split actually happens — with m spent as the capitalist's revenue, while M, as the functional form of the capital-value, carries on along the path set by the circuit — then that first act, C'-M', taken together with the acts that follow, M-C and m-c, can be represented as two different circulations: C-M-C and c-m-c. Both of these, in their general form, are just instances of ordinary commodity circulation.
In practice, with a continuous body that can't actually be divided, the value components still get isolated in thought. Take the London building trade, run mostly on credit: the builder gets advances as the house passes through its different stages. None of these stages is a house — each is only a really existing piece of a house still being made. So despite being real, each is only an ideal fraction of the whole house — yet real enough to serve as security for a further advance. (See Chapter XII below.)
Third: if the movement of capital-value and surplus-value — still joint in C and M — separates only partly (so that some of the surplus-value doesn't get spent as revenue), or doesn't separate at all, then a change happens in the capital-value itself, still within its own circuit, before that circuit is complete. In our example the value of the productive capital was £422. If it now carries on as M-C at, say, £480 or £500, it travels the remaining stages of the circuit as a value that's grown by £58 or £78 over what it started as. This can also come together with a change in its value composition.
C'-M', the second stage of circulation and the closing stage of circuit I (M . . . M'), is here the second stage of the circuit too, but the first stage of commodity circulation. As far as circulation goes, it still needs completing by M'-C'. But C'-M' has already left behind not just the valorization process (the function of P, the first stage) — its result, the commodity product C', is already realized. So both the capital's valorization process and the realization of the commodity product it's embodied in are already finished by the time we reach C'-M'.
So we've been assuming simple reproduction — that m-c splits off entirely from M-C. Since both circulations, c-m-c and C-M-C alike, belong in their general form to commodity circulation (and so show no value difference between their extremes), it's easy — as vulgar economics does — to picture the capitalist production process as simply the production of commodities, use-values meant for some kind of consumption, which the capitalist only produces in order to replace them with commodities of a different use-value, or exchange them for such — as vulgar economics wrongly puts it.
C appears from the outset as commodity-capital, and the purpose of the whole process — enrichment, valorization — doesn't rule out a level of consumption by the capitalist that grows along with the size of the surplus-value, and so with the capital itself; it positively includes it.
In the circulation of the capitalist's revenue, the produced commodity c (or the ideal fraction of the commodity product C' that corresponds to it) really does serve only to be turned first into money, and from money into a series of other commodities for private consumption. But one small point here must not be overlooked: c is commodity-value that cost the capitalist nothing — it's the embodiment of surplus labour, which is why it first appears on the scene as a component of the commodity-capital C'. So this c, just by existing, is tied to the circuit of the capital-value in process. If that circuit gets stuck, or is disturbed some other way, it's not just that the consumption of c is cut back or stops altogether — the market for the whole series of commodities that replace c shrinks with it. The same happens if C'-M' fails, or if only part of C' can be sold.
We saw that c-m-c, as the circulation of the capitalist's revenue, only enters the capital's circulation as long as c is a value-part of C', of capital in its functional form as commodity-capital. But once it's made independent through m-c — that is, in the full form c-m-c — it no longer enters the movement of the capital the capitalist has advanced, even though it arises out of it. It stays connected to that movement only insofar as the existence of the capital presupposes the existence of the capitalist, and the capitalist's existence is conditioned by consuming surplus-value.
Within general circulation, C' — say, yarn — functions only as a commodity. But as a moment in the circulation of capital, it functions as commodity-capital, a shape that the capital-value takes on and sheds in turn. Once the yarn is sold to the merchant, it's out of the circuit of the capital whose product it was — yet it still goes on existing as a commodity within the sphere of general circulation. The circulation of that same mass of commodities continues, even though it's stopped being a moment in the spinner's own independent circuit of capital. So the real, final metamorphosis of the mass of commodities the capitalist threw into circulation — C-M — its eventual falling out into consumption, can be separated in time and space from the metamorphosis in which that same mass functioned as his commodity-capital. The metamorphosis already completed within the circulation of capital still remains to be completed within the sphere of general circulation.
It changes nothing if the yarn instead goes back into the circuit of some other industrial capital. General circulation covers both things at once: the interweaving of the circuits of the different independent fragments of social capital — that is, the totality of individual capitals — and the circulation of values that aren't thrown onto the market as capital at all, the values that go into individual consumption.
The relation between the capital's circuit, so far as it's part of general circulation, and so far as it forms links of an independent circuit of its own, shows up again if we look at the circulation of M' = M + m. M, as money capital, carries on the circuit of capital. m, spent as revenue (m-c), enters general circulation but flies straight out of the capital's circuit. Only the part that functions as additional money capital re-enters that circuit. In c-m-c money functions only as coin; the purpose of this circulation is the capitalist's individual consumption. It's a mark of the crass confusion of vulgar economics that it passes off this circulation — which never enters the capital's circuit at all, the circulation of the part of the value-product consumed as revenue — as if it were the characteristic circuit of capital itself.
In this second phase, M–C, the capital value shows up again as a sum of money — M — equal to P, the value of the productive capital that opened this circuit. It has shed the surplus-value, so it's back to the same size it had at the very start of the money-capital circuit, M–C. The place in the story is different, but the job this money capital does is the same: it turns into means of production and labour-power, mp and L.
At the same time as c–m, the capital value — now functioning as commodity capital C'–M' — has gone through the phase C–M, and now enters the complementary phase M–C<L/mp>. So its whole circuit is C–M–C<L/mp>.
In Form I (the circuit M...M'), money capital M appeared as the starting form — the value advanced at the outset. Here it's different from the start: it appears as part of the sum of money that the commodity capital turned into during the first phase, C'–M'. In other words, it only exists here because a sale has already happened — the sale of the product converted P, the productive capital, into money form. So this money capital is, from the outset, neither the original form of the capital value nor its final form: only by shedding its money form again can the phase M–C complete the phase C–M that ended in it.
That's why the part of M–C that is also M–L — the purchase of labour-power — no longer appears as a mere money-advance to buy labour-power. It's an advance in which the worker is handed, in money form, the very same 1,000 lb of yarn, worth £50, that make up part of the value his own labour created. The money paid to the worker here is nothing but the converted equivalent-form of a portion of the value of the product he himself produced.
And for exactly this reason, the act M–C, so far as it is M–L, is not simply commodity-in-money-form being swapped for commodity-in-use-form. It brings in other elements, ones that have nothing to do with commodity circulation in general.
M' appears as the converted form of C', which is itself the product of P's past activity — the production process. So the whole sum M' is simply the money-expression of past labour. In our example: 10,000 lb of yarn = £500, the product of the spinning process. Of that, 7,440 lb of yarn = the advanced constant capital, £372; 1,000 lb of yarn = the advanced variable capital, £50; and 1,560 lb of yarn = the surplus-value, £78. If, other things staying the same, only the original £422 of capital is advanced again out of M', then in M–L the worker only gets, advanced to him next week, part of the 10,000 lb of yarn produced this week — the money-value of 1,000 lb. As the result of C–M, money is always an expression of past labour. And so far as the complementary act M–C happens right away on the market — money exchanged for commodities already sitting there — it's again just an exchange of past labour, moving from one form (money) into another (commodity).
But M–C happens at a different time than C–M. They can coincide, as an exception — say when two capitalists hand each other their goods at the same moment, and the money only settles the difference between them. Otherwise the time-gap between carrying out C–M and carrying out M–C can be more or less large. Even though, as the result of C–M, money represents past labour, for the act M–C that same money can represent commodities that aren't on the market at all yet — that will only be there in the future — since M–C only needs to happen after C has been produced anew. Equally, the money can represent commodities being produced at the very same time as the C whose money-expression it is. For example, in the purchase of means of production, coal can be bought before it's even been brought up out of the mine. So far as this money is being accumulated rather than spent as revenue, it can represent cotton that won't be produced until next year — and the same holds when the capitalist spends his revenue.
The same is true of the wage, £50. This money is not only the money-form of the workers' past labour — it is at the same time a draft on labour that is happening now or will happen in the future, labour not yet realized, or still to be realized. The worker might use it to buy a coat that will only be made next week. This holds above all for the great mass of necessary means of subsistence that have to be consumed almost the instant they're produced, on pain of spoiling. So in the money he's paid his wage in, the worker receives the converted form of his own future labour, or of other workers' future labour. With a portion of his past labour, the capitalist hands him a draft on his own future labour: it is his own labour, now or still to come, that forms the stock — not yet existing — with which his past labour gets paid. The whole idea of a stock being laid by beforehand disappears completely here.
Second: in the circulation C–M–C, the same money changes hands twice. The capitalist first gets it as a seller, then gives it away as a buyer. Turning a commodity into money only serves to turn that money back into a commodity. So the money-form of capital — its existence as money capital — is, in this movement, only a vanishing moment. Or rather: so far as this movement flows smoothly, money capital shows up only as a means of circulation, when it serves to buy. It shows up as a genuine means of payment when capitalists buy from one another, so that only the balance of payments has to be settled.
Third: whether money capital works as a mere means of circulation or as a means of payment, its function is only to bring about the replacement of C by L and mp — that is, the replacement of the yarn, the commodity-product the productive capital results in (after subtracting the surplus-value that will be spent as revenue), by its own elements of production. In other words, it turns the capital value back, out of its form as a commodity, into the elements that make up that commodity. So in the end its whole function is just to convert commodity capital back into productive capital.
For the circuit to run its normal course, C' has to be sold at its full value and in its entirety. And C–M–C involves not just replacing one commodity with another, but replacing it under the same value-relations. That's the assumption we're making here. In fact, though, the values of means of production do vary — capitalist production is marked by a constant shifting of value-relations, precisely because the productive power of labour is itself constantly changing. We'll only point to this shifting of the value of the factors of production here; it needs to be discussed later.
The conversion of the elements of production into the commodity-product, P into C', takes place in the sphere of production; the reconversion of C' back into P takes place in the sphere of circulation. That reconversion happens through the simple metamorphosis of commodities. But its content is a moment of the reproduction process taken as a whole. C–M–C, as capital's form of circulation, includes a functionally determined metabolism.
The exchange C–M–C further requires that C equal the elements of production of the quantity of commodities C', and that these elements hold onto their original value-relations to one another. So it assumes not only that commodities are bought at their value, but also that they undergo no change in value during the circuit. If that's not the case, the process can't run its normal course.
In M...M', M is the original form of the capital value — a form it sheds, only to take it up again later. In P...C'–M'–C...P, money is merely a form the process assumes, and one it sheds again within that very process. Here the money-form shows up only as a vanishing, independent value-form of capital. Capital is just as anxious to take on this form as C' as it is to shed it again as M', the moment it has pupated into money-form, so as to turn itself back into productive capital. As long as it stays in its money-shape, it isn't functioning as capital, and so isn't valorizing itself — the capital lies idle. Here money acts as a means of circulation, but specifically as a means of circulating capital.
The appearance of independence that the money-form of the capital value carries in the first form of its circuit — the money-capital circuit — disappears in this second form. This second form is thus the critique of the first: it reduces Form I to a merely particular form. If the second metamorphosis, M–C, runs into obstacles — say the means of production simply aren't to be had on the market — then the circuit, the flow of the reproduction process, is interrupted, just as much as if the capital were stuck fast in commodity-capital form.
But there is a difference. Capital can hold out longer in money-form than in the perishable form of a commodity. It doesn't stop being money just because it isn't functioning as money capital — but it does stop being a commodity, and a use-value at all, if it's held up too long in its function as commodity capital. And second, in money-form capital is able to take on some other form instead of returning to its original productive shape, whereas as C' it can't budge from the spot at all.
C'–M'–C, by its form, only involves circulation-acts that are moments of C''s own reproduction. But the actual reproduction of the C that C' turns into is necessary for C'–M'–C to be carried out at all — and that reproduction is itself conditioned by reproduction processes lying outside the reproduction process of the individual capital represented by C'.
In Form I, M–C only prepares the first conversion of money capital into productive capital. In Form II, it's the reconversion from commodity capital into productive capital — so, so long as the scale of the industrial capital stays the same, it's the reconversion of commodity capital back into the very same elements of production it came out of. So here, as in Form I, it appears as the phase preparing for the production process — but as a return to that process, a renewal of it. It's therefore the forerunner of the reproduction process, and so also of the repetition of the process of valorization.
It's worth noting again: M–L is not a simple exchange of commodities. It's the purchase of a commodity, labour-power, meant to serve the production of surplus-value — just as M–mp is only a step that's materially indispensable for carrying out that same purpose.
Once M–C is carried out, M has been reconverted into productive capital, into P, and the circuit starts over again.
The explicit form of P...C'–M'–C...P is therefore:
Turning money capital into productive capital means buying commodities in order to produce commodities. Only so far as this consumption is productive consumption does it belong within capital's own circuit — and the condition for that is that surplus-value gets made by means of the commodities consumed this way.
That is something very different from production in general, even from commodity production as such, whose purpose is simply the producers' own existence. Replacing one commodity with another under this condition — that it serves the production of surplus-value — is something quite different from what a mere exchange of products, merely mediated by money, is in itself.
But this is exactly how economists take the matter, to prove that overproduction is impossible.
Besides the productive consumption of M as it turns into L and mp, the circuit contains the first leg of M–L — which, from the worker's side, is L–M, that is, C–M. Of the worker's own circulation, L–M–C, which includes his consumption, only the first leg — the result of M–L — falls within capital's circuit. The second act, M–C, does not belong to the circulation of the individual capital, even though it arises out of it. But the continued existence of the working class is necessary for the capitalist class — and so is the worker's consumption that M–C brings about.
The act C'–M' requires, both for the capital value's circuit to continue and for the capitalist's consumption of surplus-value, only that C' has been converted into money — sold. Of course it only gets bought because it's a use-value, fit for some kind of consumption, whether productive or individual. But if C' keeps circulating further — say, in the hands of the merchant who's bought the yarn — that, in the first instance, has no bearing at all on the continuation of the circuit of the individual capital that produced the yarn and sold it to the merchant. The whole process keeps going, and with it the individual consumption of capitalist and worker that depends on it. This is an important point when it comes to crises.
As soon as C' is sold, turned into money, it can be reconverted into the real factors of the labour process, and so of the reproduction process. Whether C' is bought by the final consumer or by a merchant who wants to resell it makes no immediate difference to the matter. The volume of commodity-masses capitalist production turns out is set by the scale of that production and by its need to keep expanding — not by some predetermined circle of supply and demand, of needs waiting to be satisfied. Mass production's immediate buyer can, apart from other industrial capitalists, only be the big merchant. Within certain limits, the reproduction process can go on at the same or an expanded scale even though the commodities it throws out haven't actually entered individual or productive consumption. The consumption of commodities is not included in the circuit of the capital they came from.
Once the yarn, say, is sold, the circuit of the capital value it represents can start over, whatever becomes of the sold yarn afterward. As long as the product is sold, everything runs its regular course from the standpoint of the capitalist producer. The circuit of the capital value he represents is not interrupted. And if this process runs on an expanded scale — which includes expanded productive consumption of means of production — this reproduction of capital can be accompanied by expanded individual consumption, that is, demand, on the part of the workers, since the process itself is set off and mediated by productive consumption. So the production of surplus-value can grow, and with it the capitalist's individual consumption too, and the whole reproduction process can be in its most flourishing state — and yet a large part of the commodities may only appear to have entered consumption, while actually lying unsold in the hands of resellers, in fact still sitting on the market.
Now stream of commodities follows stream of commodities, and it finally comes out that the earlier stream had only seemed to be swallowed up by consumption. The commodity-capitals fight each other for their place on the market. The latecomers, to sell at all, sell under the price. The earlier streams haven't yet been turned into cash, while the deadlines for paying on them fall due. Their holders have to declare themselves insolvent, or sell at any price, just to pay. This kind of selling has absolutely nothing to do with the real state of demand. It has to do only with the demand for payment — the absolute necessity of turning commodities into money. Then the crisis breaks out. It becomes visible not in an immediate fall in demand for consumption, for individual use, but in the fall of the exchange of capital against capital — of capital's own reproduction process.
Suppose mp and L — the commodities M turns into so as to carry out its function as money capital, as capital value destined to reconvert into productive capital — have to be bought or paid for at different dates, so that M–C stands for a whole series of purchases and payments happening one after another. Then part of M carries out the act M–C, while another part stays in its money-state, waiting to serve, at a time set by the conditions of the process itself, for simultaneous or later acts of M–C. This part is only temporarily withdrawn from circulation, so as to spring into action at the fixed moment and do its job. This holding-back is itself, then, a function set by its circulation and for its circulation. Its existence as a fund for buying and paying, the suspension of its movement, this state of interrupted circulation — that is a state in which money is carrying out one of its functions as money capital.
As money capital: because in this case the money temporarily lying still is itself part of the money capital M (from M' − m = M), part of the value of the commodity capital equal to P, the value of the productive capital the circuit starts from. On the other hand, all money withdrawn from circulation takes the form of a hoard. So the hoard-form of money here becomes a function of money capital — just as, in M–C, the function of money as means of purchase or payment becomes a function of money capital — because here the capital value exists in money form, and this money-state is a state the coherence of the circuit prescribes for the industrial capital at one of its stages. And this once again confirms that money capital, within the circuit of industrial capital, performs no functions other than money functions — and that these money functions only take on the significance of capital functions through their connection with the other stages of the circuit.
Representing M' as the ratio of m to M — as a capital-relation — is not directly a function of money capital at all. It's a function of commodity capital, C', which itself, as the ratio of c to C, only expresses the result of the production process: the self-valorization of the capital value that took place within it.
If the circulation process runs into obstacles, so that external circumstances — the state of the market and so on — force M to suspend its function M–C, and it therefore stays in its money-state for a shorter or longer time, that is again a hoard-state of money, the same kind that also occurs in simple commodity circulation whenever the passage from C–M to M–C is interrupted by external circumstances. It is involuntary hoard-formation. In our case, the money thereby takes the form of idle, latent money capital. But we won't go further into this for now.
In both cases, though, money capital staying put in its money-state shows up as the result of an interrupted movement — whether that interruption fits the purpose or works against it, is voluntary or involuntary, in keeping with its function or contrary to it.
The proportions in which production can expand aren't arbitrary — they're fixed by technique. So realized surplus-value, even though it's meant to become capital, often can't reach the size it needs until several cycles have repeated themselves. Until then it must pile up as a hoard — held back because the technical minimum for additional capital has not yet been reached.
So the surplus-value hardens into a hoard, and in this form it is latent money capital. Latent, because as long as it stays in money form, it cannot act as capital.
Treasure-forming shows up here as something that belongs inside the capitalist accumulation process, that goes along with it — but is also essentially different from it. Forming latent money capital does not by itself expand reproduction. The opposite: latent money capital forms here precisely because the capitalist producer cannot yet expand the scale of production directly.
If he sells his surplus product to a gold or silver producer who throws new gold or silver into circulation — or, which comes to the same thing, to a merchant who imports additional gold or silver from abroad in exchange for part of the national surplus product — then his latent money capital adds to the national stock of gold or silver. In every other case, say the £78 that were means of circulation in the buyer's hands simply take on the form of a hoard in the capitalist's hands: only the distribution of the existing national gold or silver stock has changed.
Suppose money functions in our capitalist's dealings as a means of payment — the buyer only has to pay at some later date, near or far. Then the surplus product meant for capitalization doesn't turn into money at all. It turns into debt claims, titles of ownership to an equivalent the buyer may already have, or may only expect to get later.
This claim does not enter the reproduction process of the circuit — no more than money invested in interest-bearing paper does, even though that money can enter the circuit of some other individual industrial capital.
The whole character of capitalist production is set by the valorization of the capital value advanced. That means, first, producing as much surplus-value as possible. But second (see Volume 1, chapter 22) it means producing capital: turning surplus-value into capital.
Accumulation, production on an expanded scale, appears as a means to ever-greater production of surplus-value, and so to the capitalist's own enrichment — his personal purpose, part of capitalist production's general drive. But it becomes something more: as capitalist production develops — the first volume showed this — expansion becomes a necessity for every individual capitalist. Constantly enlarging his capital becomes the condition for keeping it at all. But we need not go back over what was set out earlier.
We began by looking at simple reproduction, where we assumed the whole of the surplus-value gets spent as revenue. In reality, under normal conditions, part of the surplus-value must always be spent as revenue and another part must be capitalized — it makes no difference whether, in any given period, the surplus-value produced is sometimes entirely consumed and sometimes entirely capitalized. Averaged over the whole movement — and the general formula can only represent that average — both things happen together.
Still, to keep the formula simple, it's better to assume the whole surplus-value is accumulated. The formula P . . . C'-M'-C' . . . P' then says: productive capital is reproduced on a larger scale, with a larger value, and as this enlarged productive capital it begins its second circuit — or, the same thing, renews its first circuit.
As soon as this second circuit begins, P is again the starting point — only now P is a larger productive capital than the first P was. In the same way, in the formula M . . . M', if the second circuit begins with M', then M' functions simply as M: an advanced money capital of a given size. It is a larger sum of money capital than the one that opened the first circuit — but every trace of how it grew, through capitalizing surplus-value, has disappeared the moment it appears as advanced money capital opening a circuit. That origin is wiped out in its form as money capital starting a fresh circuit. The same holds for P', the moment it functions as the starting point of a new circuit.
Compare P . . . P' with M . . . M' — or with the first circuit — and they do not mean the same thing at all. M . . . M', taken by itself as an isolated circuit, only says this: that M, money capital (or industrial capital in its circuit-form as money capital), is money that breeds money, value that breeds value — it posits surplus-value.
In the circuit of P, though, the valorization process itself is already completed by the end of the first stage, the production process. And once the second stage — the first stage of circulation, C'-M' — has run its course, capital-value plus surplus-value already exist as realized money capital, as M', which appeared as the last term in the first circuit.
That surplus-value has been produced was already shown, in the earlier form P . . . P, by c-m-c — whose second stage falls outside the circulation of capital and represents the circulation of the surplus-value as revenue. In this form, where the whole movement is represented as P . . . P, so that there is no difference in value between the two endpoints, the valorization of the advanced value — the generation of surplus-value — is expressed just as fully as it is in M . . . M'. Only the act C'-M' appears differently: as the final stage in M . . . M', but as the second stage of the circuit — the first stage of circulation — in P . . . P.
In P . . . P', P' expresses something different: not that surplus-value has been produced, but that the surplus-value produced has been capitalized — that capital has been accumulated. So P', compared with P, consists of the original capital value plus the value of the capital that its movement has accumulated.
M', as the mere conclusion of M . . . M', and C', as it appears within all these circuits, do not by themselves express the movement — they express its result: the valorization of the capital value, realized in commodity-form or money-form. That means capital value as M + m, or as C + c — as the relation of the capital value to its surplus-value, its offshoot. They express this result as different circulation-forms of the valorized capital value.
But neither in the form C' nor in the form M' is the valorization that has taken place itself a function — not of money capital, and not of commodity capital either. As particular, distinct forms of existence, corresponding to the particular functions of industrial capital, money capital can only perform money-functions, and commodity capital only commodity-functions; the difference between them is only the difference between money and a commodity. In the same way, industrial capital in its form as productive capital can only consist of the same elements as any other labour process that produces a product: on one side, objective conditions of labour — means of production; on the other, labour-power activating itself productively, purposefully. Just as industrial capital, within the sphere of production, can only exist in the composition that belongs to the production process as such — and so belongs equally to a non-capitalist production process — so too, in the sphere of circulation, it can only exist in the two forms that belong there: commodity and money.
But just as the sum of the elements of production announces itself as productive capital, from the very start, only because the labour-power is someone else's labour-power, which the capitalist has bought from its own owner — exactly as he bought his means of production from other commodity-owners — and just as the production process itself therefore appears as the productive function of industrial capital, so too money and commodity appear as circulation-forms of that same industrial capital, and their functions appear as its circulation-functions: functions that either open the way to the functions of productive capital, or spring from them. Only through this connection — as functional forms that industrial capital has to go through, in the different stages of its circuit — are the money-function and the commodity-function, at the same time, functions of money capital and of commodity capital.
So it is wrong to try to derive the specific properties and functions that characterize money as money, and the commodity as commodity, from their capital-character. And it is equally wrong, the other way round, to derive the properties of productive capital from its mode of existence in means of production.
Once M' or C' is fixed as M + m, or C + c — that is, as the relation of the capital value to the surplus-value that sprang from it — this relation is expressed in both: once in money-form, once in commodity-form. That changes nothing about the thing itself. So this relation springs neither from properties and functions belonging to money as such, nor from properties belonging to the commodity as such. In both cases, the property that characterizes capital — being value that breeds value — is expressed only as a result.
C' is always the product of the function of P, and M' is always only the transformed, circuit-form of C' within the circuit of industrial capital. So the moment the realized money capital begins its own particular function as money capital again, it stops expressing the capital-relation contained in M' = M + m. Once M . . . M' has run its course and M' begins the circuit afresh, it no longer figures as M' but as M — even if the whole surplus-value contained in M' gets capitalized.
In our example, the second circuit begins with a money capital of £500, instead of the £422 that opened the first. The money capital opening the circuit is £78 larger than before. That difference only shows up when we compare one circuit with the other — it does not exist inside either circuit taken on its own. The £500 advanced as money capital — £78 of which used to exist as surplus-value — plays no different role than £500 with which some other capitalist opens his first circuit. The same holds for the circuit of productive capital: the enlarged P', at the renewed start, appears simply as P — just as P does in simple reproduction, P . . . P.
In the stage M'-C', the increased size is shown only by C' — not separately by L' and mp'. Since C is the sum of L and mp, C' already shows that the sum of L and mp it contains is larger than in the original P.
But calling them L' and mp' would be wrong for a second reason too: we know that as capital grows, its value-composition changes along with it. As this goes on, the value of mp grows while the value of L keeps falling relatively — often even in absolute terms.
Whether m — the surplus-value now in gold, so to speak — can be added straight back to the capital value in process, entering the circuit together with M as the magnitude M', depends on circumstances that have nothing to do with the mere existence of m.
Suppose m is meant to serve as money capital in a second, independent business alongside the first. Then it's obviously only usable once it reaches the minimum size that business requires. Suppose instead it's meant to extend the original business. Then the material makeup of P's factors, and their value-relations, likewise fix a minimum size m must reach. All the means of production working in a business stand to each other not just in a certain kind of relation but in a definite quantitative one — a proportional scale. These material relations, and the value-relations that ride on them, fix the minimum size m must have before it can be converted into additional means of production and labour-power — or just the former — as an increment to productive capital.
So a spinner cannot add to his number of spindles without also buying the matching carding engines and roving frames, quite apart from the extra outlay on cotton and wages that such an expansion requires. To carry this out, the surplus-value already has to amount to a fair sum — roughly £1 per newly acquired spindle is the usual reckoning. As long as m hasn't reached this minimum size, the circuit of capital has to repeat itself several times, until the sum of the m's it successively throws off, together with M, is large enough to function in M'-C'. Even small changes of detail — say, in spinning machinery, where they make it more productive — call for greater outlay on spinning material, more roving machinery, and so on.
In the meantime, then, m piles up — but its piling up is not its own function. It is the result of repeated cycles of P . . . P. Its own function, in the meantime, is simply to sit still in the state of money, until it has received enough addition from outside — from those repeated cycles of valorization — to reach the minimum size its active function requires: the size at which alone it can actually enter, as money capital, into the function of the money capital M already at work, as an accumulated part of it.
In the meantime it is accumulated, and exists only in the form of a hoard that is forming, that is growing. Money accumulation, treasure-forming, appears here, then, as a process that temporarily accompanies the real accumulation — the extension of the scale on which industrial capital works. Temporarily: because as long as the hoard stays in its hoard-state, it does not function as capital. It takes no part in the valorization process. It stays a sum of money that only grows because, with no doing of its own, more money keeps landing in the same chest.
The hoard-form is only the form taken by money that is not in circulation — money whose circulation has been interrupted, and which is therefore kept sitting in its money-form.
As for the process of treasure-forming itself, it is common to all commodity production, and only plays a role as an end in itself in commodity production's undeveloped, pre-capitalist forms. Here, though, the hoard appears as a form of money capital, and treasure-forming appears as a process that temporarily accompanies the accumulation of capital — because, and only insofar as, the money here figures as latent money capital.
That is: treasure-forming, the hoard-state of the surplus-value that exists in money-form, is a preparatory stage for turning that surplus-value into really functioning capital — a stage that happens outside the circuit of capital, fixed in its function by that role. It is latent money capital because of this, its function — which is also why the size it has to reach before it can enter the process is set by whatever the value-composition of the productive capital happens to be at the time.
But as long as it stays in the hoard-state, it does not yet function as money capital. It is still dormant money capital — not, as before, money capital interrupted in a function it already had, but money capital not yet capable of any function at all.
We are taking the accumulation of money here in its original, real form — as an actual hoard of money. It can also exist in the form of mere credit balances: debt claims held by the capitalist who has sold C'.
As for the other forms — where this latent money capital exists in the meantime as money breeding money in its own right, say as an interest-bearing deposit in a bank, or in bills of exchange or securities of some kind — those don't belong here. The surplus-value realized in money is then performing particular capital-functions outside the circuit of the industrial capital it sprang from: functions that, first, have nothing to do with that circuit as such, and second, presuppose capital-functions distinct from those of industrial capital — functions not yet developed at this point.
In the form we've just looked at, the hoard in which the surplus-value exists is a money-accumulation fund — the money-form that capital accumulation temporarily wears, and to that extent itself a condition for that accumulation. But this accumulation fund can also do particular side-jobs: it can enter the circuit process of capital without that process taking the form P . . . P' — that is, without capitalist reproduction actually being expanded.
Suppose the process C'-M' is drawn out beyond its normal length — the commodity capital gets abnormally held up on its way into money-form. Or suppose that conversion does go through, but, say, the price of the means of production the money capital has to be turned into has risen above what it was when the circuit began. In either case, the hoard functioning as an accumulation fund can be used to stand in for the money capital, or part of it. The money-accumulation fund then serves as a reserve fund, evening out disturbances in the circuit.
As a reserve fund of this kind, it is different from the fund of means of purchase or payment we looked at within the circuit P . . . P. That fund is part of the functioning money capital — a form taken by part of the capital value in process generally — whose portions simply come into play one after another, at different points in time. It forms constantly, in the ongoing continuity of the production process: payments come in today that only have to be made again at a later date; larger batches of goods get sold today, and only have to be bought again, in larger batches, on some later day. In these intervals, a part of the circulating capital constantly sits in money-form.
The reserve fund is different. It is not a component of the functioning capital — of the functioning money capital, more precisely — but of capital still caught in an earlier stage of its accumulation: surplus-value that has not yet been turned into active capital. Of course, a capitalist in trouble doesn't stop to ask which particular function some money in his hand was supposed to have — he uses whatever he's got to keep his capital's circuit going.
In our example, M = £422, M' = £500. If part of the £422 exists as a fund of means of payment and purchase — a stock of ready money on hand — it is calculated so that, if circumstances stay the same, the whole of it enters the circuit, and is enough for that purpose. The reserve fund, though, is part of the £78 of surplus-value. It can only enter the circuit process of the £422 capital to the extent that this circuit is carried out under circumstances that are not staying the same — because it is part of the accumulation fund, and figures here without any expansion of the scale of reproduction.
The money-accumulation fund is already the existence of latent money capital — a transformation, that is, of money into money capital.
The general formula for the circuit of productive capital — the one that covers both simple reproduction and reproduction on an expanded scale — is:
If P equals P — the reading rule for the same scale: the second circuit renews unchanged, the surplus m spent as revenue — then M in (2) equals M' minus m. But if P equals P' — if the scale has expanded — then M in (2) is larger than M' minus m. In other words: m has been wholly or partly turned into money capital.
The circuit of productive capital is the form in which classical political economy looks at the circuit process of industrial capital.