Capital moves through the production sphere and then through the two phases of the circulation sphere, one stage after another in time, as we have seen. The time it spends in the production sphere is its production time. The time it spends in the circulation sphere is its circulation time. So the total time for one full circuit is simply production time plus circulation time.
Production time includes the time the work process takes, but it is bigger than that — the two are not the same thing.
Start with the part of the constant capital that exists in means of labour: things like machinery and buildings that serve the same repeated work processes right up to the end of their working life. When work stops at night, that interrupts these tools' actual functioning — but it doesn't remove them from the workplace. They belong to the production site whether they are working or standing idle.
There's also the stock of raw material and auxiliary materials the capitalist must keep on hand, so that production at a given scale can carry on for shorter or longer stretches without depending on the chance of getting fresh deliveries from the market every day. This stock is only consumed bit by bit, productively, over time. So there is a gap between its production time and the time it actually functions.
So the production time of means of production as a whole covers three things: first, the time they actually function as means of production, serving in the process; second, the pauses when the process — and so the functioning of the means of production built into it — is interrupted; third, the time they lie ready as conditions for the process, already forming part of productive capital, but not yet actually put to work in it.
The difference we've looked at so far was between the time productive capital spends in the production sphere and the time it spends in the actual process of production. But the process of production itself can also cause the work process — and so the working time — to be interrupted: there can be stretches where the object being worked on is left to physical processes acting on it, with no further human labour added. In these stretches the production process, and so the functioning of the means of production, keeps going, even though the work process — and so the functioning of the means of production as tools of labour — has stopped. Take grain that has been sown, or wine fermenting in the cellar, or the raw material in many manufacturing trades — tanning, for instance — undergoing chemical processes. Here production time is greater than labour time. The difference between the two is an excess of production time over labour time. This excess always rests on one of two things: either productive capital sits latent in the production sphere without functioning in the process of production itself, or it functions in the process of production without being part of the work process.
Take the part of latent productive capital that just lies ready as a condition for the process — cotton, coal, in a spinning mill, for example. It doesn't form the product and it doesn't form value. It is capital lying idle — even though that idleness is itself a condition for the production process to flow without interruption. The buildings and equipment needed to hold this productive stock — this latent capital — are themselves conditions of the production process, and so count as part of the productive capital advanced. Their job is to preserve these productive elements while they wait in this preliminary stage. Where labour is needed at this stage, it makes the raw material more expensive, but it is still productive labour and it does create surplus value — because part of it, like all other wage-labour, goes unpaid.
The normal interruptions of the whole production process — the intervals when productive capital simply isn't functioning — create neither value nor surplus value. This is why there is a drive to keep production running at night too. But the interruptions the object of labour must go through during the production process itself are different: they too create neither value nor surplus value, but they do carry the product forward — they are part of what the product's life consists of, a stage it has to pass through. The value of the equipment is transferred to the product in proportion to the whole time it is in use; labour itself is what puts the product into this stage, and using this equipment is just as much a condition of production as the loss of some cotton that never makes it into the product but still passes its value onto it.
The other part of latent capital — buildings, machines, and the like, the means of labour whose functioning is interrupted only by the process's regular pauses (irregular interruptions from cutbacks in production, crises, and so on, are pure losses) — adds value without entering into the making of the product. The total value it adds to the product is fixed by its average lifespan: it loses value because it loses use-value, both while it is functioning and while it is not.
Finally, the value of the part of constant capital that keeps going through the production process even while the work process is interrupted turns up again in the result of that process. Labour itself is what puts the means of production into conditions where they then run through certain natural processes on their own — processes that end in some useful effect, or in a changed form of their use-value. Labour always transfers the value of the means of production onto the product, so long as it genuinely consumes them, for their purpose, as means of production. It makes no difference here whether the labour has to keep acting on the object continuously, through the instruments of labour, to bring about this effect, or whether it only has to set things in motion — putting the means of production into conditions under which, with no further help from labour, they undergo the intended change by themselves, through natural processes.
Whatever the reason for this excess of production time over labour time — whether the means of production are still only latent productive capital, sitting in a preliminary stage before the actual production process; or their own functioning is interrupted by pauses within the production process; or the production process itself brings about interruptions of the work process — in none of these cases do the means of production absorb any labour. And if they absorb no labour, they absorb no surplus labour either. So no valorization of productive capital happens during that part of its production time that exceeds its labour time — however inseparable completing the valorization process may be from these very pauses.
Clearly, the more production time and labour time coincide, the greater the productivity and valorization of a given productive capital over a given period. Hence capitalist production's tendency to shorten, as far as possible, the excess of production time over labour time. But even though a capital's production time may diverge from its labour time, it always contains that labour time within it, and the excess itself is a condition of the production process. Production time, then, is always the time during which capital produces use-values and valorizes itself — that is, functions as productive capital — even though it includes time in which capital is either lying latent, or is producing without valorizing itself.
Within the circulation sphere, capital exists as commodity-capital and as money-capital. Its two circulation processes consist in turning itself from commodity form into money form, and from money form into commodity form. It makes no difference that turning the commodity into money is, at the same time, realizing the surplus-value contained in the commodity, or that turning money into commodities is, at the same time, transforming — or retransforming — the capital-value into the shape of its elements of production. These processes are still, as circulation processes, nothing but processes of simple commodity metamorphosis, the same as any other.
Circulation time and production time exclude one another. While capital is in its circulation time, it is not functioning as productive capital, and so it produces neither commodities nor surplus-value.
Take the circuit in its simplest form, where the whole capital-value passes from one phase to the next all at once: it's obvious that the production process is interrupted — and so is capital's self-valorization — for as long as its circulation time lasts, and the shorter or longer that time is, the faster or slower the production process starts up again. Now take the case where different parts of the capital go round the circuit one after another, so that the circuit of the whole capital-value is completed bit by bit through the circuits of its separate portions. Here it is clear that the longer any given fraction of the capital constantly sits in the circulation sphere, the smaller the fraction that can constantly be functioning in the production sphere. So the expansion or contraction of circulation time acts as a negative limit on the contraction or expansion of production time — on the scale, that is, on which a capital of a given size can function as productive capital. The more the circulation-metamorphoses of capital are merely nominal — the closer circulation time gets to zero — the more capital is functioning, and the greater its productivity and self-valorization. If a capitalist works to order, say, and is paid on delivery of the product, and paid in his own means of production, then circulation time comes close to zero.
So capital's circulation time limits its production time generally, and with that its valorization process — and it limits it in proportion to how long it lasts. But this length can grow or shrink very differently from case to case, and so it can limit capital's production time to very different degrees.
What political economy actually sees, though, is only how this appears — the effect of circulation time on capital's valorization process in general. It mistakes this negative effect for a positive one, because its consequences look positive. It clings to this appearance all the more because that appearance seems to prove that capital has some mystical source of self-valorization, independent of its own production process and so of the exploitation of labour, flowing to it from the circulation sphere. We shall see later that even scientific economics lets itself be taken in by this appearance. And it is reinforced, as we shall also see, by several things. First, the capitalist way of reckoning profit, in which this negative cause figures as a positive one: for capitals in different lines of investment, where only the circulation time differs, a longer circulation time acts as a reason for a higher price — in short, as one of the factors in the equalization of profits. Second, circulation time is only one part of turnover time; but turnover time also includes production time, or reproduction time. What is really owed to production time then looks as if it were owed to circulation time. Third, converting commodities into variable capital, that is wages, depends on first turning them into money. So when capital accumulates, the conversion into additional variable capital happens in the circulation sphere, during circulation time — and the resulting accumulation then looks as if it were owed to circulation time.
Within the circulation sphere, capital passes through two opposite phases — C-M and M-C — whichever order they come in. So its circulation time also splits into two parts: the time it needs to turn itself from commodity into money, and the time it needs to turn itself from money into commodity. We already know from the analysis of simple commodity circulation that C-M, the sale, is the harder part of the metamorphosis, and so, under ordinary conditions, makes up the larger part of the circulation time. As money, value is in its form of constant convertibility. As a commodity, it first has to acquire that shape of immediate exchangeability, and so of readiness for use, by being turned into money.
But in capital's circulation process, in its phase M-C, what matters is turning money into the particular commodities that form the specific elements of productive capital for a given business. The means of production may not be sitting ready on the market at all — they may have to be produced first, or fetched from distant markets, or there may be shortfalls in the usual supply, price changes, and so on. In short, a whole mass of circumstances that don't show up in the bare change of form M-C can still, for this part of the circulation phase too, take up more or less time. And just as C-M and M-C can be separated in time, they can be separated in space too — the buying market and the selling market can be different markets altogether. In factories, for instance, the buyer and the seller are often even different people.
Circulation is just as necessary for commodity production as production itself is, so the agents of circulation are just as necessary as the agents of production. The reproduction process includes both functions of capital, and so it also includes the need for someone to carry them out — whether the capitalist himself, or wage-labourers acting as his agents. But this is no more reason to confuse the agents of circulation with the agents of production than it is reason to confuse the functions of commodity-capital and money-capital with those of productive capital. The agents of circulation have to be paid by the agents of production. When capitalists buy and sell among themselves, that act creates neither products nor value — and that stays true even when the scale of their business lets them, and forces them, to hand this function over to others. In some businesses buyers and sellers are paid a share of the profit. Saying they are paid by the consumers doesn't change anything: consumers can only pay to the extent that they themselves, as agents of production, produce an equivalent in commodities, or obtain one from the agents of production — whether by right, as their partners say, or through personal services.
There is a difference between C-M and M-C that has nothing to do with the difference in form between commodity and money — it comes instead from the capitalist character of production itself. Taken purely in themselves, both C-M and M-C are just translations of a given value from one form into another. But C'-M' is, at the same time, the realization of the surplus-value contained in C'. M-C is not. That is why the sale matters more than the purchase. Under normal conditions M-C is a necessary act for valorizing the value expressed in M, but it does not realize surplus-value; it is the opening move toward producing surplus-value, not something added on afterward.
For the circulation of commodity-capital, C'-M', definite limits are set by the very form in which commodities exist — their being use-values. Commodities are by nature perishable. So if they do not enter productive or individual consumption within a certain period — depending on what they are for — in other words, if they are not sold within a certain time, they spoil, and along with their use-value they lose the ability to carry exchange-value at all. The capital-value contained in them, and whatever surplus-value has grown onto it, is simply lost. Use-values only go on carrying a lasting, self-valorizing capital-value so long as they are constantly renewed and reproduced — replaced by new use-values of the same kind or another. Selling them in their finished commodity form, and so getting them, through that sale, into productive or individual consumption, is the ever-renewing condition for their reproduction. They must change out of their old use-form within a certain time in order to keep existing in a new one. Exchange-value only preserves itself through this constant renewal of its body.
Different commodities' use-values spoil at different speeds, so more or less time can pass between a commodity's production and its consumption; commodities can hold out for a shorter or longer time, without perishing, in the C-M phase of circulation as commodity-capital — they can bear a shorter or longer circulation time. The limit set on commodity-capital's circulation time by the spoiling of the commodity's own body is the absolute limit on this part of circulation time — the most that commodity-capital, as commodity-capital, can ever take.
The more perishable a commodity, the more immediately after its production it must be consumed, and so also sold; the shorter the distance it can travel from where it is produced, the narrower its spatial circulation sphere, the more local its market. So the more perishable a commodity, the greater the absolute limit its physical make-up puts on its circulation time as a commodity, and the less suited it is to be an object of capitalist production at all. Capitalist production can only take it up in densely populated places, or as the development of transport draws distant places closer together. But concentrating the production of an article in a few hands, in one populous place, can still create a relatively large market even for such articles — large breweries or dairies, for instance.
Capital moves through the production sphere and then through the two phases of the circulation sphere, one stage after another in time, as we have seen. The time it spends in the production sphere is its production time. The time it spends in the circulation sphere is its circulation time. So the total time for one full circuit is simply production time plus circulation time.
Production time includes the time the work process takes, but it is bigger than that — the two are not the same thing.
Start with the part of the constant capital that exists in means of labour: things like machinery and buildings that serve the same repeated work processes right up to the end of their working life. When work stops at night, that interrupts these tools' actual functioning — but it doesn't remove them from the workplace. They belong to the production site whether they are working or standing idle.
There's also the stock of raw material and auxiliary materials the capitalist must keep on hand, so that production at a given scale can carry on for shorter or longer stretches without depending on the chance of getting fresh deliveries from the market every day. This stock is only consumed bit by bit, productively, over time. So there is a gap between its production time and the time it actually functions.
So the production time of means of production as a whole covers three things: first, the time they actually function as means of production, serving in the process; second, the pauses when the process — and so the functioning of the means of production built into it — is interrupted; third, the time they lie ready as conditions for the process, already forming part of productive capital, but not yet actually put to work in it.
The difference we've looked at so far was between the time productive capital spends in the production sphere and the time it spends in the actual process of production. But the process of production itself can also cause the work process — and so the working time — to be interrupted: there can be stretches where the object being worked on is left to physical processes acting on it, with no further human labour added. In these stretches the production process, and so the functioning of the means of production, keeps going, even though the work process — and so the functioning of the means of production as tools of labour — has stopped. Take grain that has been sown, or wine fermenting in the cellar, or the raw material in many manufacturing trades — tanning, for instance — undergoing chemical processes. Here production time is greater than labour time. The difference between the two is an excess of production time over labour time. This excess always rests on one of two things: either productive capital sits latent in the production sphere without functioning in the process of production itself, or it functions in the process of production without being part of the work process.
Take the part of latent productive capital that just lies ready as a condition for the process — cotton, coal, in a spinning mill, for example. It doesn't form the product and it doesn't form value. It is capital lying idle — even though that idleness is itself a condition for the production process to flow without interruption. The buildings and equipment needed to hold this productive stock — this latent capital — are themselves conditions of the production process, and so count as part of the productive capital advanced. Their job is to preserve these productive elements while they wait in this preliminary stage. Where labour is needed at this stage, it makes the raw material more expensive, but it is still productive labour and it does create surplus value — because part of it, like all other wage-labour, goes unpaid.
The normal interruptions of the whole production process — the intervals when productive capital simply isn't functioning — create neither value nor surplus value. This is why there is a drive to keep production running at night too. But the interruptions the object of labour must go through during the production process itself are different: they too create neither value nor surplus value, but they do carry the product forward — they are part of what the product's life consists of, a stage it has to pass through. The value of the equipment is transferred to the product in proportion to the whole time it is in use; labour itself is what puts the product into this stage, and using this equipment is just as much a condition of production as the loss of some cotton that never makes it into the product but still passes its value onto it.
The other part of latent capital — buildings, machines, and the like, the means of labour whose functioning is interrupted only by the process's regular pauses (irregular interruptions from cutbacks in production, crises, and so on, are pure losses) — adds value without entering into the making of the product. The total value it adds to the product is fixed by its average lifespan: it loses value because it loses use-value, both while it is functioning and while it is not.
Finally, the value of the part of constant capital that keeps going through the production process even while the work process is interrupted turns up again in the result of that process. Labour itself is what puts the means of production into conditions where they then run through certain natural processes on their own — processes that end in some useful effect, or in a changed form of their use-value. Labour always transfers the value of the means of production onto the product, so long as it genuinely consumes them, for their purpose, as means of production. It makes no difference here whether the labour has to keep acting on the object continuously, through the instruments of labour, to bring about this effect, or whether it only has to set things in motion — putting the means of production into conditions under which, with no further help from labour, they undergo the intended change by themselves, through natural processes.
Whatever the reason for this excess of production time over labour time — whether the means of production are still only latent productive capital, sitting in a preliminary stage before the actual production process; or their own functioning is interrupted by pauses within the production process; or the production process itself brings about interruptions of the work process — in none of these cases do the means of production absorb any labour. And if they absorb no labour, they absorb no surplus labour either. So no valorization of productive capital happens during that part of its production time that exceeds its labour time — however inseparable completing the valorization process may be from these very pauses.
Clearly, the more production time and labour time coincide, the greater the productivity and valorization of a given productive capital over a given period. Hence capitalist production's tendency to shorten, as far as possible, the excess of production time over labour time. But even though a capital's production time may diverge from its labour time, it always contains that labour time within it, and the excess itself is a condition of the production process. Production time, then, is always the time during which capital produces use-values and valorizes itself — that is, functions as productive capital — even though it includes time in which capital is either lying latent, or is producing without valorizing itself.
Within the circulation sphere, capital exists as commodity-capital and as money-capital. Its two circulation processes consist in turning itself from commodity form into money form, and from money form into commodity form. It makes no difference that turning the commodity into money is, at the same time, realizing the surplus-value contained in the commodity, or that turning money into commodities is, at the same time, transforming — or retransforming — the capital-value into the shape of its elements of production. These processes are still, as circulation processes, nothing but processes of simple commodity metamorphosis, the same as any other.
Circulation time and production time exclude one another. While capital is in its circulation time, it is not functioning as productive capital, and so it produces neither commodities nor surplus-value.
Take the circuit in its simplest form, where the whole capital-value passes from one phase to the next all at once: it's obvious that the production process is interrupted — and so is capital's self-valorization — for as long as its circulation time lasts, and the shorter or longer that time is, the faster or slower the production process starts up again. Now take the case where different parts of the capital go round the circuit one after another, so that the circuit of the whole capital-value is completed bit by bit through the circuits of its separate portions. Here it is clear that the longer any given fraction of the capital constantly sits in the circulation sphere, the smaller the fraction that can constantly be functioning in the production sphere. So the expansion or contraction of circulation time acts as a negative limit on the contraction or expansion of production time — on the scale, that is, on which a capital of a given size can function as productive capital. The more the circulation-metamorphoses of capital are merely nominal — the closer circulation time gets to zero — the more capital is functioning, and the greater its productivity and self-valorization. If a capitalist works to order, say, and is paid on delivery of the product, and paid in his own means of production, then circulation time comes close to zero.
So capital's circulation time limits its production time generally, and with that its valorization process — and it limits it in proportion to how long it lasts. But this length can grow or shrink very differently from case to case, and so it can limit capital's production time to very different degrees.
What political economy actually sees, though, is only how this appears — the effect of circulation time on capital's valorization process in general. It mistakes this negative effect for a positive one, because its consequences look positive. It clings to this appearance all the more because that appearance seems to prove that capital has some mystical source of self-valorization, independent of its own production process and so of the exploitation of labour, flowing to it from the circulation sphere. We shall see later that even scientific economics lets itself be taken in by this appearance. And it is reinforced, as we shall also see, by several things. First, the capitalist way of reckoning profit, in which this negative cause figures as a positive one: for capitals in different lines of investment, where only the circulation time differs, a longer circulation time acts as a reason for a higher price — in short, as one of the factors in the equalization of profits. Second, circulation time is only one part of turnover time; but turnover time also includes production time, or reproduction time. What is really owed to production time then looks as if it were owed to circulation time. Third, converting commodities into variable capital, that is wages, depends on first turning them into money. So when capital accumulates, the conversion into additional variable capital happens in the circulation sphere, during circulation time — and the resulting accumulation then looks as if it were owed to circulation time.
Within the circulation sphere, capital passes through two opposite phases — C-M and M-C — whichever order they come in. So its circulation time also splits into two parts: the time it needs to turn itself from commodity into money, and the time it needs to turn itself from money into commodity. We already know from the analysis of simple commodity circulation that C-M, the sale, is the harder part of the metamorphosis, and so, under ordinary conditions, makes up the larger part of the circulation time. As money, value is in its form of constant convertibility. As a commodity, it first has to acquire that shape of immediate exchangeability, and so of readiness for use, by being turned into money.
But in capital's circulation process, in its phase M-C, what matters is turning money into the particular commodities that form the specific elements of productive capital for a given business. The means of production may not be sitting ready on the market at all — they may have to be produced first, or fetched from distant markets, or there may be shortfalls in the usual supply, price changes, and so on. In short, a whole mass of circumstances that don't show up in the bare change of form M-C can still, for this part of the circulation phase too, take up more or less time. And just as C-M and M-C can be separated in time, they can be separated in space too — the buying market and the selling market can be different markets altogether. In factories, for instance, the buyer and the seller are often even different people.
Circulation is just as necessary for commodity production as production itself is, so the agents of circulation are just as necessary as the agents of production. The reproduction process includes both functions of capital, and so it also includes the need for someone to carry them out — whether the capitalist himself, or wage-labourers acting as his agents. But this is no more reason to confuse the agents of circulation with the agents of production than it is reason to confuse the functions of commodity-capital and money-capital with those of productive capital. The agents of circulation have to be paid by the agents of production. When capitalists buy and sell among themselves, that act creates neither products nor value — and that stays true even when the scale of their business lets them, and forces them, to hand this function over to others. In some businesses buyers and sellers are paid a share of the profit. Saying they are paid by the consumers doesn't change anything: consumers can only pay to the extent that they themselves, as agents of production, produce an equivalent in commodities, or obtain one from the agents of production — whether by right, as their partners say, or through personal services.
There is a difference between C-M and M-C that has nothing to do with the difference in form between commodity and money — it comes instead from the capitalist character of production itself. Taken purely in themselves, both C-M and M-C are just translations of a given value from one form into another. But C'-M' is, at the same time, the realization of the surplus-value contained in C'. M-C is not. That is why the sale matters more than the purchase. Under normal conditions M-C is a necessary act for valorizing the value expressed in M, but it does not realize surplus-value; it is the opening move toward producing surplus-value, not something added on afterward.
For the circulation of commodity-capital, C'-M', definite limits are set by the very form in which commodities exist — their being use-values. Commodities are by nature perishable. So if they do not enter productive or individual consumption within a certain period — depending on what they are for — in other words, if they are not sold within a certain time, they spoil, and along with their use-value they lose the ability to carry exchange-value at all. The capital-value contained in them, and whatever surplus-value has grown onto it, is simply lost. Use-values only go on carrying a lasting, self-valorizing capital-value so long as they are constantly renewed and reproduced — replaced by new use-values of the same kind or another. Selling them in their finished commodity form, and so getting them, through that sale, into productive or individual consumption, is the ever-renewing condition for their reproduction. They must change out of their old use-form within a certain time in order to keep existing in a new one. Exchange-value only preserves itself through this constant renewal of its body.
Different commodities' use-values spoil at different speeds, so more or less time can pass between a commodity's production and its consumption; commodities can hold out for a shorter or longer time, without perishing, in the C-M phase of circulation as commodity-capital — they can bear a shorter or longer circulation time. The limit set on commodity-capital's circulation time by the spoiling of the commodity's own body is the absolute limit on this part of circulation time — the most that commodity-capital, as commodity-capital, can ever take.
The more perishable a commodity, the more immediately after its production it must be consumed, and so also sold; the shorter the distance it can travel from where it is produced, the narrower its spatial circulation sphere, the more local its market. So the more perishable a commodity, the greater the absolute limit its physical make-up puts on its circulation time as a commodity, and the less suited it is to be an object of capitalist production at all. Capitalist production can only take it up in densely populated places, or as the development of transport draws distant places closer together. But concentrating the production of an article in a few hands, in one populous place, can still create a relatively large market even for such articles — large breweries or dairies, for instance.