Ricardo brings up the distinction between fixed and circulating capital only to point out exceptions to the value rule — cases where the wage rate affects prices. That question only gets taken up later, in Book III.
But the original muddle already shows itself in the way he sets two things side by side, as if they were the same kind of thing:
"This difference in how durable fixed capital is, and this variation in the proportions in which the two kinds of capital may be combined."
So let's ask: what are these two kinds of capital? Here is the answer:
"Also the proportions in which the capital meant to support labour, and the capital laid out in tools, machinery, and buildings, may be variously combined."
So, on Ricardo's reading: fixed capital equals the means of labour, and circulating capital equals the capital laid out on labour. "Capital meant to support labour" is already a tired phrase, taken over from Adam Smith without question. Here, circulating capital gets lumped together, on one hand, with variable capital — the part of productive capital laid out on labour. But on the other hand, because the contrast isn't drawn from the process of valorization — constant versus variable capital — but from the process of circulation instead (the old Smithian confusion), two wrong determinations follow at once.
First: differences in how durable fixed capital is get treated as equivalent to differences in the organic composition of capital — how capital is composed of constant and variable capital. But it's this second difference that decides how much surplus-value gets produced. The first difference, so far as valorization is concerned, only concerns the manner in which a given value passes from the means of production onto the product; so far as circulation is concerned, it only concerns the period over which the capital laid out gets renewed — or, put differently, the length of time for which it has been advanced. If, instead of looking through to the inner workings of the capitalist process of production, one takes the standpoint of the finished, surface phenomena, these two differences do in fact run together. When the social surplus-value gets distributed among the capitals invested in different branches, differences in how long capital is tied up (say, the different lifespans of fixed capital) and different compositions of capital (hence also the different way constant and variable capital circulate) both play an equal part in equalizing the general rate of profit and turning values into prices of production.
From the standpoint of the circulation process, the means of labour stand on one side as fixed capital, while the material of labour together with wages stand on the other side as fluid capital. But from the standpoint of the labour process and valorization, it's different: on one side stand the means of production — both means of labour and material of labour — as constant capital; on the other side stands labour-power as variable capital. For the organic composition of capital, it makes no difference at all whether the same sum of constant capital consists of a lot of means of labour and little material of labour, or the reverse — everything depends only on the ratio between the capital laid out on means of production and the capital laid out on labour-power. The other way round: from the standpoint of the circulation process, of the distinction between fixed and circulating capital, it's just as indifferent in what proportions a given sum of circulating capital splits between material of labour and wages. So from one standpoint, the material of labour is ranked in the same category as the means of labour, against the capital-value laid out on labour-power. From the other standpoint, the part of capital laid out on labour-power is ranked together with the part laid out on material of labour, against the part laid out on means of labour.
That's why, in Ricardo, the part of capital's value laid out on material of labour — raw and auxiliary materials — shows up on neither side. It disappears entirely. It doesn't fit on the side of fixed capital, because the way it circulates coincides exactly with the way the part of capital laid out on labour-power circulates. And it can't be put on the side of circulating capital either, because that would undo the very equation — carried over from Adam Smith and running through silently — between the contrast fixed/circulating and the contrast constant/variable. Ricardo has too much logical instinct not to sense this, so this part of capital simply vanishes for him altogether.
It's worth noting that, in the language of political economy, the capitalist is said to advance the capital laid out in wages at different intervals — depending on whether he pays weekly, monthly, or every three months. In fact the matter runs the other way. It's the worker who advances his labour to the capitalist, for a week, a month, three months, depending on whether he gets paid weekly, monthly, or every three months. If the capitalist bought labour-power outright, instead of paying for it afterward — that is, if he paid the worker's wage in advance, by the day, week, month, or quarter — then one could speak of an advance for that period. But since he actually pays only after the labour has already lasted days, weeks, months, rather than buying it and paying for the stretch it's going to last, the whole thing is a capitalist quid pro quo — the worker's advance of labour gets passed off as the capitalist's advance of money. None of this changes if the capitalist only gets the product itself, or its value, back out of circulation — and only realizes it, together with the surplus-value embodied in it — after shorter or longer intervals, depending on how long production takes or how long circulation takes. What a buyer means to do with a commodity is a matter of complete indifference to the seller. The capitalist doesn't get a machine any cheaper because he has to advance its whole value at once while that same value only flows back to him gradually, in pieces, out of circulation; nor does he pay more for cotton because its value passes entirely into the value of the product made from it, and so is replaced wholly and at once by the sale of that product.
Let's come back to Ricardo.
What's characteristic of variable capital is this: a definite, given — and so, as such, constant — part of capital, a given sum of value (say equal to the value of labour-power, though it doesn't matter here whether the wage is equal to, greater than, or smaller than that value), gets exchanged for a force that valorizes itself, that creates value: labour-power, which doesn't just reproduce the value the capitalist paid for it, but at the same time produces a surplus-value — a value that didn't exist before and that no equivalent was paid for. This characteristic property of the part of capital laid out on wages — the very thing that sets it apart, in every respect, from constant capital, as variable capital — disappears the moment this part of capital is looked at purely from the standpoint of the circulation process, and so appears instead as circulating capital set against the fixed capital laid out on means of labour. This already shows up in the fact that it then gets placed, under one heading — circulating capital — together with a piece of constant capital, the part laid out on material of labour, and set against another piece of constant capital, the part laid out on means of labour. Surplus-value — the very thing that turns the sum of value laid out into capital in the first place — drops out of the reckoning entirely. Just as much left out: the part of value that the capital laid out on wages adds to the product is newly produced, genuinely reproduced, while the part of value that the raw material adds to the product is not newly produced, not genuinely reproduced, but merely preserved, kept intact, in the value of the product — it only reappears as a piece of the product's value. The difference, as it now presents itself from the standpoint of the contrast between fluid and fixed capital, comes down only to this: the value of the means of labour used to produce a commodity enters the value of that commodity only in part, and so gets replaced by the sale of the commodity only in part — bit by bit and gradually. The value of the labour-power and of the objects of labour (raw materials and so on) used to produce a commodity, on the other hand, enters the commodity entirely, and is therefore replaced entirely by its sale. To that extent, seen from the standpoint of circulation, one part of capital presents itself as fixed, the other as fluid or circulating. In both cases what's at issue is simply a transfer of given, advanced values onto the product, and their replacement through the product's sale. The difference now consists only in whether that transfer, and so that replacement, happens bit by bit and gradually, or all at once. With that, the difference that decides everything — between variable and constant capital — is wiped out. And with it goes the whole secret of how surplus-value gets formed, the whole secret of capitalist production, the circumstances that turn certain values, and the things they're embodied in, into capital in the first place. Every part of capital now differs only in how it circulates (and the circulation of a commodity, of course, only ever deals with values that already exist, already given). And one particular manner of circulating turns out to be shared by the capital laid out on wages together with the part of capital laid out on raw materials, semi-finished goods, and auxiliary materials — against the part of capital laid out on means of labour.
One can see, then, why bourgeois political economy instinctively held onto Adam Smith's confusion of the categories "constant and variable capital" with the categories "fixed and circulating capital", and repeated it uncritically, generation after generation, for a century. In its hands, the part of capital laid out on wages no longer differs at all from the part laid out on raw material, and differs from constant capital only in a formal sense — whether it circulates through the product bit by bit or all at once. With that, the basis for understanding how capitalist production, and so capitalist exploitation, really works is buried at a stroke. All that's left at issue is the reappearance of values already advanced.
In Ricardo, taking over Smith's confusion uncritically is more troubling than it is even in the later apologists — for whom this conceptual confusion is, if anything, not troubling at all — and more troubling than in Adam Smith himself. That's because Ricardo, unlike Smith, works out value and surplus-value more consistently and more sharply, in effect asserting the esoteric Adam Smith against the exoteric Adam Smith.
In the Physiocrats there is none of this confusion. The distinction between avances annuelles (annual advances) and avances primitives (original advances) concerns only the different periods over which the different components of capital — especially agricultural capital — get reproduced. Their view of how surplus-value is produced is a separate part of their theory, independent of that distinction, and indeed the very point they turn their theory on. They don't explain the formation of surplus-value from capital as such — they claim it only for one particular sphere where capital is invested: agriculture.
What matters for variable capital — and so for turning any sum of value into capital at all — is this: the capitalist hands over a fixed, given amount of value in exchange for value-creating power. He exchanges a quantity of value for the production of value, for self-expanding value. It makes no difference to this whether he pays the worker in money or directly in food and other means of subsistence. All that changes is the form the value he advances takes — money, with which the worker buys his own means of subsistence on the market, or means of subsistence handed over directly for him to consume. Developed capitalist production does in fact assume payment in money, just as it assumes a production process mediated by circulation, a money economy generally. But the creation of surplus-value — and so the turning of the advanced sum into capital — springs neither from the money-form nor the in-kind form of wages, nor from the money-form or the in-kind form of the capital laid out on buying labour-power. It springs from the exchange of value for value-creating power, from the conversion of a constant magnitude into a variable one. Whether the instruments of labour are more or less fixed depends on how durable they are — a physical property. Depending on their durability, other things being equal, they wear out faster or slower, and so serve longer or shorter as fixed capital. But it is by no means this physical durability alone that makes them function as fixed capital. The raw material in a metal works is just as durable as the machines that work it, and more durable than some of the parts those machines are made of — leather, wood, and so on. Yet the metal used as raw material counts as part of circulating capital, while the working instrument built, perhaps, from that very same metal counts as part of fixed capital. So it isn't the physical stuff, isn't the metal's greater or lesser perishability, that puts the same metal now under fixed and now under circulating capital. That difference comes instead from the role it plays in production: one time as the object being worked on, the other time as the instrument doing the working.
Functioning as an instrument of labour generally requires, on average, that a thing keep serving, again and again, through repeated labour processes, over some longer or shorter period. Its function therefore calls for a material that is more or less durable. But the durability of the material it's made of doesn't by itself make it fixed capital. The same material, used as raw material instead, becomes circulating capital — and for economists who confuse the distinction between commodity-capital and productive capital with the distinction between circulating and fixed capital, the very same material, the very same machine, counts as circulating capital when it's the product, and fixed capital when it's the instrument.
So although it isn't the durable material an instrument is made of that makes it fixed capital, its role as an instrument does require it to be made of relatively durable material. The durability of its material is thus a condition for its functioning as an instrument, and so also the material basis of the way it circulates — the way that makes it capital at all. Other things equal, the less perishable its material, the more strongly it gets stamped with fixity — the more perishable, the less so; its durability is very much bound up with its character as fixed capital.
Now suppose the part of capital laid out on labour-power is looked at purely as circulating capital, in contrast to fixed capital — so that the distinction between constant and variable capital gets lumped together with the distinction between fixed and circulating capital. Then it's natural to make the same move in reverse: just as the material reality of the instrument gets treated as the essential basis of its character as fixed capital, so now, by contrast, the material reality of the capital laid out on labour-power gets treated as the basis of its character as circulating capital — and circulating capital in turn gets defined by the material reality of variable capital.
The real material of the capital laid out on wages is labour itself — active, value-creating labour-power, living labour, which the capitalist exchanges against dead, objectified labour and incorporates into his capital. Only through this does the value in his hands turn into value that expands itself. But this power to expand value is not itself something the capitalist sells. It is always only a component of his productive capital, like his instruments of labour — never a component of his commodity-capital, like the finished product he sells. Inside the production process, as components of productive capital, the instruments of labour do not stand opposed to labour-power as fixed capital, any more than the materials and auxiliary substances coincide with it as circulating capital. Both instruments and materials, as the objective factors, stand opposed to labour-power as the personal factor — that is how it looks from the standpoint of the labour process. But from the standpoint of the valorization process, both instruments and materials stand opposed to labour-power, to variable capital, as constant capital. Or, if there is to be talk here of a material difference that affects the circulation process, it is only this: since value is nothing but objectified labour, and active labour-power is nothing but labour in the process of objectifying itself, it follows that labour-power, for as long as it functions, is constantly creating value and surplus-value — and what appears on its own side as movement, as the creation of value, appears on the side of its product in a settled form, as value already created. Once labour-power has done its work, capital no longer consists of labour-power on one side and means of production on the other. The capital-value that was laid out on labour-power is now value (plus surplus-value) added to the product. To repeat the process, the product has to be sold, and with the money that comes out of it, labour-power has to be bought afresh, again and again, and taken back into the productive capital. This is what gives the part of capital laid out on labour-power — like the part laid out on raw materials and so on — the character of circulating capital, as against the capital that stays fixed in the instruments of labour.
Suppose instead that the secondary feature of circulating capital — a feature it shares with part of constant capital, the raw and auxiliary materials — is made into the essential feature of the capital laid out on labour-power. That feature is this: the value laid out transfers wholly onto the product in whose making it is used up, rather than gradually and piecemeal as with fixed capital, and so must be replaced wholly, all at once, by the sale of the product. Once this becomes the defining test, then the part of capital laid out on wages can no longer, materially speaking, consist of labour-power actually at work. It has to consist instead of the material things the worker buys with his wage — that is, of the part of the social commodity-capital that goes into the worker's own consumption: means of subsistence. Fixed capital then consists of the more slowly perishing instruments of labour, which need replacing more slowly; the capital laid out on labour-power consists of the means of subsistence, which need replacing faster.
But the line between faster and slower perishing blurs.
"The food and clothing the worker consumes, the buildings he works in, the tools that help with his work — these are all perishable. But there's an enormous difference in how long these different kinds of capital last: a steam-engine lasts longer than a ship, a ship longer than a worker's clothing, and a worker's clothing longer again than the food he eats."
Here Ricardo forgets the house the worker lives in, his furniture, and the tools he uses to consume things — knives, forks, dishes and the like — all of which are just as durable as instruments of labour. The very same things, the very same kinds of things, appear here as means of consumption and there as instruments of labour.
The distinction, as Ricardo states it, is this:
"Depending on whether capital perishes quickly and must often be reproduced, or is consumed slowly, it gets classed as circulating or as fixed capital."
To which Ricardo adds this note:
"An unimportant division, and one where, besides, the line can't be drawn precisely." That is Ricardo's own footnote.
So here we are, back happily with the physiocrats, for whom the difference between yearly advances and initial advances was a difference in how long things took to be consumed, and so also in how long the capital applied took to be reproduced. Except that what for them expressed something important about social production as a whole — and was even set out, in connection with the circulation process, in the Tableau économique — turns here into a merely subjective distinction, one which, as Ricardo himself says, is superfluous.
As soon as the capital-part laid out on labour differs from the capital-part laid out on instruments only in how long it takes to reproduce and so circulate — as soon as one part consists of means of subsistence and the other of instruments, the first differing from the second only in perishing a bit faster (since even means of subsistence themselves perish at different rates) — then of course every specific difference between capital laid out on labour-power and capital laid out on means of production has been wiped out.
This flatly contradicts Ricardo's own theory of value, and his theory of profit too, which is really a theory of surplus-value. In fact he only looks at the difference between fixed and circulating capital insofar as different proportions of the two, in equal-sized capitals in different branches of business, affect the law of value — specifically, how far a rise or fall in wages, because of these different proportions, affects prices. But even within this narrow inquiry, by mixing up fixed and circulating capital with constant and variable capital, he makes his biggest mistakes, and starts, in effect, from a wholly false basis. So, first: wherever the value laid out on labour-power gets filed under circulating capital, the very features of circulating capital get wrongly worked out — in particular, the circumstances that put the capital laid out on labour under that heading. Second: there is a confusion between the fact that the capital-part laid out on labour is variable, and the fact that it is circulating as against fixed.
It's clear from the start that classing capital laid out on labour-power as circulating or fluid is a secondary matter, one in which its specific difference within the production process gets wiped out. On one hand, under this classing, capital laid out on labour and capital laid out on raw materials count as the same kind of thing — a heading that lumps part of constant capital together with variable capital has nothing to do with what specifically distinguishes variable from constant capital. On the other hand, the capital laid out on labour and the capital laid out on instruments are indeed set against each other here — but not because they enter into the production of value in completely different ways. They're set against each other only because, in both cases, a given value gets transferred onto the product, just over different lengths of time.
In all these cases, what's at issue is how a given value laid out in the process of producing a commodity — whether it's wages, the price of raw material, or the price of instruments — gets transferred onto the product, circulates along with it, and is brought back to its starting point, replaced, when the product is sold. The only difference lies in the how — in the particular manner in which this value is transferred, and so circulates.
Whether the price of labour-power, fixed in advance by contract in every case, is paid in money or in means of subsistence makes no difference to its character as a determinate, given price. Still, when wages are paid in money, it's obvious that the money itself doesn't enter the production process — not the way the means of production enter it, where not just their value but their actual material substance goes into the process. But if instead the means of subsistence the worker buys with his wage are lumped directly together, as the material form circulating capital takes, with raw materials and the like, and set against instruments of labour, then the matter looks different. If the value of the means of production is transferred onto the product in the labour process, then the value of those other things, the means of subsistence, reappears in the labour-power that consumes them, and through that labour-power's activity gets transferred onto the product too. In every one of these cases, it's equally just a matter of values advanced during production simply reappearing in the product. (The physiocrats took this literally, which is why they denied that industrial labour creates surplus-value.) So too in the passage from Wayland cited earlier:
"It makes no difference in what form the capital reappears... the various kinds of food, clothing and housing that people need for their existence and wellbeing are likewise changed. They get used up over time, and their value reappears, and so on."
The capital-values advanced to production in the shape of means of production and means of subsistence here reappear equally in the value of the product. With that, the capitalist production process has been happily turned into a complete mystery, and the origin of the surplus-value present in the product has been made entirely invisible.
This also completes a fetishism peculiar to bourgeois economics: turning the social, economic character that things acquire within the social process of production into a natural character supposedly springing from their physical make-up. Take the claim that instruments of labour are fixed capital — a scholastic formula that leads straight to contradictions and confusion. Just as with the labour process, shown earlier, everything depends on the particular role the material elements play in a given labour process — on their function, on whether they're acting as instrument, as material, or as product — the same holds here. Instruments of labour are fixed capital only where the production process is a capitalist production process at all, so that the means of production are capital at all, carrying the economic and social character of capital; and, further, they are fixed capital only when they give off their value to the product in a particular way. If they don't, they stay instruments of labour without being fixed capital. Likewise, an auxiliary substance like manure, if it gives off its value in that same particular way most instruments do, becomes fixed capital even though it isn't an instrument of labour at all. None of this is about which definition a thing gets filed under. It's about particular functions, which get expressed through particular categories.
If being capital laid out on wages counts as a property that means of subsistence have in themselves, under any circumstances, then it also becomes a feature of this "circulating" capital that it "supports labour." Were means of subsistence not "capital," on this view, they wouldn't support labour-power at all — whereas in fact it's their character as capital that gives them the property of maintaining capital, through other people's labour.
If means of subsistence are in themselves circulating capital — once turned into wages — then it also follows, on this view, that the size of wages depends on the ratio between the number of workers and the given mass of circulating capital: a favourite proposition among economists. But in fact, the mass of means of subsistence the worker draws off the market, and the mass of means of subsistence the capitalist has available for his own consumption, depend on the ratio between surplus-value and the price of labour.
Ricardo, like Barton, everywhere confuses the ratio of variable to constant capital with the ratio of circulating to fixed capital. We'll see later how this distorts his inquiry into the rate of profit.
Ricardo also treats differences in turnover that arise from quite other causes as if they were the same as the difference between fixed and circulating capital:
"It should also be noted that circulating capital can circulate, or return to the person using it, over very unequal lengths of time. Wheat bought by a farmer for sowing is fixed capital compared with wheat bought by a baker to turn into bread. The farmer leaves it in the ground and can't get a return before a year is up; the baker can have it ground into flour and sell it to his customers as bread, so that within a week he has his capital free again to start the same operation over, or any other."
What's telling here is that wheat, even though it's serving as seed — not as a means of subsistence but as raw material — counts, first, as circulating capital because it is in itself a means of subsistence, and second, as fixed capital because its return takes over a year. But it isn't only a slower or faster return that makes a means of production fixed capital — it's the particular way it gives off its value to the product.
The confusion set going by Adam Smith has led to the following results:
1. The difference between fixed and fluid capital gets confused with the difference between productive capital and commodity-capital. So, for instance, the very same machine counts as circulating capital while it sits on the market as a commodity, and as fixed capital once it's taken up into the production process. Yet there's no seeing, on this basis, why one kind of capital should be any more fixed or any more circulating than the other.
2. All circulating capital gets identified with capital laid out, or to be laid out, on wages. This is what happens with J. S. Mill and others.
3. The difference between variable and constant capital — already confused, in Barton, Ricardo and others, with the difference between circulating and fixed — finally gets reduced entirely to the latter. In Ramsay, for instance, all means of production, raw materials as well as instruments, count as fixed capital, and only the capital laid out on wages counts as circulating capital. But because the reduction happens in this form, the real difference between constant and variable capital never gets grasped at all.
4. Among the most recent English economists, especially the Scottish ones, who view everything from the hopelessly narrow standpoint of a bank clerk — Macleod, Patterson and others — the difference between fixed and circulating capital turns into the difference between money at call and money not at call: deposit money that can be withdrawn without notice, or only after giving notice.
Ricardo brings up the distinction between fixed and circulating capital only to point out exceptions to the value rule — cases where the wage rate affects prices. That question only gets taken up later, in Book III.
But the original muddle already shows itself in the way he sets two things side by side, as if they were the same kind of thing:
"This difference in how durable fixed capital is, and this variation in the proportions in which the two kinds of capital may be combined."
So let's ask: what are these two kinds of capital? Here is the answer:
"Also the proportions in which the capital meant to support labour, and the capital laid out in tools, machinery, and buildings, may be variously combined."
So, on Ricardo's reading: fixed capital equals the means of labour, and circulating capital equals the capital laid out on labour. "Capital meant to support labour" is already a tired phrase, taken over from Adam Smith without question. Here, circulating capital gets lumped together, on one hand, with variable capital — the part of productive capital laid out on labour. But on the other hand, because the contrast isn't drawn from the process of valorization — constant versus variable capital — but from the process of circulation instead (the old Smithian confusion), two wrong determinations follow at once.
First: differences in how durable fixed capital is get treated as equivalent to differences in the organic composition of capital — how capital is composed of constant and variable capital. But it's this second difference that decides how much surplus-value gets produced. The first difference, so far as valorization is concerned, only concerns the manner in which a given value passes from the means of production onto the product; so far as circulation is concerned, it only concerns the period over which the capital laid out gets renewed — or, put differently, the length of time for which it has been advanced. If, instead of looking through to the inner workings of the capitalist process of production, one takes the standpoint of the finished, surface phenomena, these two differences do in fact run together. When the social surplus-value gets distributed among the capitals invested in different branches, differences in how long capital is tied up (say, the different lifespans of fixed capital) and different compositions of capital (hence also the different way constant and variable capital circulate) both play an equal part in equalizing the general rate of profit and turning values into prices of production.
From the standpoint of the circulation process, the means of labour stand on one side as fixed capital, while the material of labour together with wages stand on the other side as fluid capital. But from the standpoint of the labour process and valorization, it's different: on one side stand the means of production — both means of labour and material of labour — as constant capital; on the other side stands labour-power as variable capital. For the organic composition of capital, it makes no difference at all whether the same sum of constant capital consists of a lot of means of labour and little material of labour, or the reverse — everything depends only on the ratio between the capital laid out on means of production and the capital laid out on labour-power. The other way round: from the standpoint of the circulation process, of the distinction between fixed and circulating capital, it's just as indifferent in what proportions a given sum of circulating capital splits between material of labour and wages. So from one standpoint, the material of labour is ranked in the same category as the means of labour, against the capital-value laid out on labour-power. From the other standpoint, the part of capital laid out on labour-power is ranked together with the part laid out on material of labour, against the part laid out on means of labour.
That's why, in Ricardo, the part of capital's value laid out on material of labour — raw and auxiliary materials — shows up on neither side. It disappears entirely. It doesn't fit on the side of fixed capital, because the way it circulates coincides exactly with the way the part of capital laid out on labour-power circulates. And it can't be put on the side of circulating capital either, because that would undo the very equation — carried over from Adam Smith and running through silently — between the contrast fixed/circulating and the contrast constant/variable. Ricardo has too much logical instinct not to sense this, so this part of capital simply vanishes for him altogether.
It's worth noting that, in the language of political economy, the capitalist is said to advance the capital laid out in wages at different intervals — depending on whether he pays weekly, monthly, or every three months. In fact the matter runs the other way. It's the worker who advances his labour to the capitalist, for a week, a month, three months, depending on whether he gets paid weekly, monthly, or every three months. If the capitalist bought labour-power outright, instead of paying for it afterward — that is, if he paid the worker's wage in advance, by the day, week, month, or quarter — then one could speak of an advance for that period. But since he actually pays only after the labour has already lasted days, weeks, months, rather than buying it and paying for the stretch it's going to last, the whole thing is a capitalist quid pro quo — the worker's advance of labour gets passed off as the capitalist's advance of money. None of this changes if the capitalist only gets the product itself, or its value, back out of circulation — and only realizes it, together with the surplus-value embodied in it — after shorter or longer intervals, depending on how long production takes or how long circulation takes. What a buyer means to do with a commodity is a matter of complete indifference to the seller. The capitalist doesn't get a machine any cheaper because he has to advance its whole value at once while that same value only flows back to him gradually, in pieces, out of circulation; nor does he pay more for cotton because its value passes entirely into the value of the product made from it, and so is replaced wholly and at once by the sale of that product.
Let's come back to Ricardo.
What's characteristic of variable capital is this: a definite, given — and so, as such, constant — part of capital, a given sum of value (say equal to the value of labour-power, though it doesn't matter here whether the wage is equal to, greater than, or smaller than that value), gets exchanged for a force that valorizes itself, that creates value: labour-power, which doesn't just reproduce the value the capitalist paid for it, but at the same time produces a surplus-value — a value that didn't exist before and that no equivalent was paid for. This characteristic property of the part of capital laid out on wages — the very thing that sets it apart, in every respect, from constant capital, as variable capital — disappears the moment this part of capital is looked at purely from the standpoint of the circulation process, and so appears instead as circulating capital set against the fixed capital laid out on means of labour. This already shows up in the fact that it then gets placed, under one heading — circulating capital — together with a piece of constant capital, the part laid out on material of labour, and set against another piece of constant capital, the part laid out on means of labour. Surplus-value — the very thing that turns the sum of value laid out into capital in the first place — drops out of the reckoning entirely. Just as much left out: the part of value that the capital laid out on wages adds to the product is newly produced, genuinely reproduced, while the part of value that the raw material adds to the product is not newly produced, not genuinely reproduced, but merely preserved, kept intact, in the value of the product — it only reappears as a piece of the product's value. The difference, as it now presents itself from the standpoint of the contrast between fluid and fixed capital, comes down only to this: the value of the means of labour used to produce a commodity enters the value of that commodity only in part, and so gets replaced by the sale of the commodity only in part — bit by bit and gradually. The value of the labour-power and of the objects of labour (raw materials and so on) used to produce a commodity, on the other hand, enters the commodity entirely, and is therefore replaced entirely by its sale. To that extent, seen from the standpoint of circulation, one part of capital presents itself as fixed, the other as fluid or circulating. In both cases what's at issue is simply a transfer of given, advanced values onto the product, and their replacement through the product's sale. The difference now consists only in whether that transfer, and so that replacement, happens bit by bit and gradually, or all at once. With that, the difference that decides everything — between variable and constant capital — is wiped out. And with it goes the whole secret of how surplus-value gets formed, the whole secret of capitalist production, the circumstances that turn certain values, and the things they're embodied in, into capital in the first place. Every part of capital now differs only in how it circulates (and the circulation of a commodity, of course, only ever deals with values that already exist, already given). And one particular manner of circulating turns out to be shared by the capital laid out on wages together with the part of capital laid out on raw materials, semi-finished goods, and auxiliary materials — against the part of capital laid out on means of labour.
One can see, then, why bourgeois political economy instinctively held onto Adam Smith's confusion of the categories "constant and variable capital" with the categories "fixed and circulating capital", and repeated it uncritically, generation after generation, for a century. In its hands, the part of capital laid out on wages no longer differs at all from the part laid out on raw material, and differs from constant capital only in a formal sense — whether it circulates through the product bit by bit or all at once. With that, the basis for understanding how capitalist production, and so capitalist exploitation, really works is buried at a stroke. All that's left at issue is the reappearance of values already advanced.
In Ricardo, taking over Smith's confusion uncritically is more troubling than it is even in the later apologists — for whom this conceptual confusion is, if anything, not troubling at all — and more troubling than in Adam Smith himself. That's because Ricardo, unlike Smith, works out value and surplus-value more consistently and more sharply, in effect asserting the esoteric Adam Smith against the exoteric Adam Smith.
In the Physiocrats there is none of this confusion. The distinction between avances annuelles (annual advances) and avances primitives (original advances) concerns only the different periods over which the different components of capital — especially agricultural capital — get reproduced. Their view of how surplus-value is produced is a separate part of their theory, independent of that distinction, and indeed the very point they turn their theory on. They don't explain the formation of surplus-value from capital as such — they claim it only for one particular sphere where capital is invested: agriculture.
What matters for variable capital — and so for turning any sum of value into capital at all — is this: the capitalist hands over a fixed, given amount of value in exchange for value-creating power. He exchanges a quantity of value for the production of value, for self-expanding value. It makes no difference to this whether he pays the worker in money or directly in food and other means of subsistence. All that changes is the form the value he advances takes — money, with which the worker buys his own means of subsistence on the market, or means of subsistence handed over directly for him to consume. Developed capitalist production does in fact assume payment in money, just as it assumes a production process mediated by circulation, a money economy generally. But the creation of surplus-value — and so the turning of the advanced sum into capital — springs neither from the money-form nor the in-kind form of wages, nor from the money-form or the in-kind form of the capital laid out on buying labour-power. It springs from the exchange of value for value-creating power, from the conversion of a constant magnitude into a variable one. Whether the instruments of labour are more or less fixed depends on how durable they are — a physical property. Depending on their durability, other things being equal, they wear out faster or slower, and so serve longer or shorter as fixed capital. But it is by no means this physical durability alone that makes them function as fixed capital. The raw material in a metal works is just as durable as the machines that work it, and more durable than some of the parts those machines are made of — leather, wood, and so on. Yet the metal used as raw material counts as part of circulating capital, while the working instrument built, perhaps, from that very same metal counts as part of fixed capital. So it isn't the physical stuff, isn't the metal's greater or lesser perishability, that puts the same metal now under fixed and now under circulating capital. That difference comes instead from the role it plays in production: one time as the object being worked on, the other time as the instrument doing the working.
Functioning as an instrument of labour generally requires, on average, that a thing keep serving, again and again, through repeated labour processes, over some longer or shorter period. Its function therefore calls for a material that is more or less durable. But the durability of the material it's made of doesn't by itself make it fixed capital. The same material, used as raw material instead, becomes circulating capital — and for economists who confuse the distinction between commodity-capital and productive capital with the distinction between circulating and fixed capital, the very same material, the very same machine, counts as circulating capital when it's the product, and fixed capital when it's the instrument.
So although it isn't the durable material an instrument is made of that makes it fixed capital, its role as an instrument does require it to be made of relatively durable material. The durability of its material is thus a condition for its functioning as an instrument, and so also the material basis of the way it circulates — the way that makes it capital at all. Other things equal, the less perishable its material, the more strongly it gets stamped with fixity — the more perishable, the less so; its durability is very much bound up with its character as fixed capital.
Now suppose the part of capital laid out on labour-power is looked at purely as circulating capital, in contrast to fixed capital — so that the distinction between constant and variable capital gets lumped together with the distinction between fixed and circulating capital. Then it's natural to make the same move in reverse: just as the material reality of the instrument gets treated as the essential basis of its character as fixed capital, so now, by contrast, the material reality of the capital laid out on labour-power gets treated as the basis of its character as circulating capital — and circulating capital in turn gets defined by the material reality of variable capital.
The real material of the capital laid out on wages is labour itself — active, value-creating labour-power, living labour, which the capitalist exchanges against dead, objectified labour and incorporates into his capital. Only through this does the value in his hands turn into value that expands itself. But this power to expand value is not itself something the capitalist sells. It is always only a component of his productive capital, like his instruments of labour — never a component of his commodity-capital, like the finished product he sells. Inside the production process, as components of productive capital, the instruments of labour do not stand opposed to labour-power as fixed capital, any more than the materials and auxiliary substances coincide with it as circulating capital. Both instruments and materials, as the objective factors, stand opposed to labour-power as the personal factor — that is how it looks from the standpoint of the labour process. But from the standpoint of the valorization process, both instruments and materials stand opposed to labour-power, to variable capital, as constant capital. Or, if there is to be talk here of a material difference that affects the circulation process, it is only this: since value is nothing but objectified labour, and active labour-power is nothing but labour in the process of objectifying itself, it follows that labour-power, for as long as it functions, is constantly creating value and surplus-value — and what appears on its own side as movement, as the creation of value, appears on the side of its product in a settled form, as value already created. Once labour-power has done its work, capital no longer consists of labour-power on one side and means of production on the other. The capital-value that was laid out on labour-power is now value (plus surplus-value) added to the product. To repeat the process, the product has to be sold, and with the money that comes out of it, labour-power has to be bought afresh, again and again, and taken back into the productive capital. This is what gives the part of capital laid out on labour-power — like the part laid out on raw materials and so on — the character of circulating capital, as against the capital that stays fixed in the instruments of labour.
Suppose instead that the secondary feature of circulating capital — a feature it shares with part of constant capital, the raw and auxiliary materials — is made into the essential feature of the capital laid out on labour-power. That feature is this: the value laid out transfers wholly onto the product in whose making it is used up, rather than gradually and piecemeal as with fixed capital, and so must be replaced wholly, all at once, by the sale of the product. Once this becomes the defining test, then the part of capital laid out on wages can no longer, materially speaking, consist of labour-power actually at work. It has to consist instead of the material things the worker buys with his wage — that is, of the part of the social commodity-capital that goes into the worker's own consumption: means of subsistence. Fixed capital then consists of the more slowly perishing instruments of labour, which need replacing more slowly; the capital laid out on labour-power consists of the means of subsistence, which need replacing faster.
But the line between faster and slower perishing blurs.
"The food and clothing the worker consumes, the buildings he works in, the tools that help with his work — these are all perishable. But there's an enormous difference in how long these different kinds of capital last: a steam-engine lasts longer than a ship, a ship longer than a worker's clothing, and a worker's clothing longer again than the food he eats."
Here Ricardo forgets the house the worker lives in, his furniture, and the tools he uses to consume things — knives, forks, dishes and the like — all of which are just as durable as instruments of labour. The very same things, the very same kinds of things, appear here as means of consumption and there as instruments of labour.
The distinction, as Ricardo states it, is this:
"Depending on whether capital perishes quickly and must often be reproduced, or is consumed slowly, it gets classed as circulating or as fixed capital."
To which Ricardo adds this note:
"An unimportant division, and one where, besides, the line can't be drawn precisely." That is Ricardo's own footnote.
So here we are, back happily with the physiocrats, for whom the difference between yearly advances and initial advances was a difference in how long things took to be consumed, and so also in how long the capital applied took to be reproduced. Except that what for them expressed something important about social production as a whole — and was even set out, in connection with the circulation process, in the Tableau économique — turns here into a merely subjective distinction, one which, as Ricardo himself says, is superfluous.
As soon as the capital-part laid out on labour differs from the capital-part laid out on instruments only in how long it takes to reproduce and so circulate — as soon as one part consists of means of subsistence and the other of instruments, the first differing from the second only in perishing a bit faster (since even means of subsistence themselves perish at different rates) — then of course every specific difference between capital laid out on labour-power and capital laid out on means of production has been wiped out.
This flatly contradicts Ricardo's own theory of value, and his theory of profit too, which is really a theory of surplus-value. In fact he only looks at the difference between fixed and circulating capital insofar as different proportions of the two, in equal-sized capitals in different branches of business, affect the law of value — specifically, how far a rise or fall in wages, because of these different proportions, affects prices. But even within this narrow inquiry, by mixing up fixed and circulating capital with constant and variable capital, he makes his biggest mistakes, and starts, in effect, from a wholly false basis. So, first: wherever the value laid out on labour-power gets filed under circulating capital, the very features of circulating capital get wrongly worked out — in particular, the circumstances that put the capital laid out on labour under that heading. Second: there is a confusion between the fact that the capital-part laid out on labour is variable, and the fact that it is circulating as against fixed.
It's clear from the start that classing capital laid out on labour-power as circulating or fluid is a secondary matter, one in which its specific difference within the production process gets wiped out. On one hand, under this classing, capital laid out on labour and capital laid out on raw materials count as the same kind of thing — a heading that lumps part of constant capital together with variable capital has nothing to do with what specifically distinguishes variable from constant capital. On the other hand, the capital laid out on labour and the capital laid out on instruments are indeed set against each other here — but not because they enter into the production of value in completely different ways. They're set against each other only because, in both cases, a given value gets transferred onto the product, just over different lengths of time.
In all these cases, what's at issue is how a given value laid out in the process of producing a commodity — whether it's wages, the price of raw material, or the price of instruments — gets transferred onto the product, circulates along with it, and is brought back to its starting point, replaced, when the product is sold. The only difference lies in the how — in the particular manner in which this value is transferred, and so circulates.
Whether the price of labour-power, fixed in advance by contract in every case, is paid in money or in means of subsistence makes no difference to its character as a determinate, given price. Still, when wages are paid in money, it's obvious that the money itself doesn't enter the production process — not the way the means of production enter it, where not just their value but their actual material substance goes into the process. But if instead the means of subsistence the worker buys with his wage are lumped directly together, as the material form circulating capital takes, with raw materials and the like, and set against instruments of labour, then the matter looks different. If the value of the means of production is transferred onto the product in the labour process, then the value of those other things, the means of subsistence, reappears in the labour-power that consumes them, and through that labour-power's activity gets transferred onto the product too. In every one of these cases, it's equally just a matter of values advanced during production simply reappearing in the product. (The physiocrats took this literally, which is why they denied that industrial labour creates surplus-value.) So too in the passage from Wayland cited earlier:
"It makes no difference in what form the capital reappears... the various kinds of food, clothing and housing that people need for their existence and wellbeing are likewise changed. They get used up over time, and their value reappears, and so on."
The capital-values advanced to production in the shape of means of production and means of subsistence here reappear equally in the value of the product. With that, the capitalist production process has been happily turned into a complete mystery, and the origin of the surplus-value present in the product has been made entirely invisible.
This also completes a fetishism peculiar to bourgeois economics: turning the social, economic character that things acquire within the social process of production into a natural character supposedly springing from their physical make-up. Take the claim that instruments of labour are fixed capital — a scholastic formula that leads straight to contradictions and confusion. Just as with the labour process, shown earlier, everything depends on the particular role the material elements play in a given labour process — on their function, on whether they're acting as instrument, as material, or as product — the same holds here. Instruments of labour are fixed capital only where the production process is a capitalist production process at all, so that the means of production are capital at all, carrying the economic and social character of capital; and, further, they are fixed capital only when they give off their value to the product in a particular way. If they don't, they stay instruments of labour without being fixed capital. Likewise, an auxiliary substance like manure, if it gives off its value in that same particular way most instruments do, becomes fixed capital even though it isn't an instrument of labour at all. None of this is about which definition a thing gets filed under. It's about particular functions, which get expressed through particular categories.
If being capital laid out on wages counts as a property that means of subsistence have in themselves, under any circumstances, then it also becomes a feature of this "circulating" capital that it "supports labour." Were means of subsistence not "capital," on this view, they wouldn't support labour-power at all — whereas in fact it's their character as capital that gives them the property of maintaining capital, through other people's labour.
If means of subsistence are in themselves circulating capital — once turned into wages — then it also follows, on this view, that the size of wages depends on the ratio between the number of workers and the given mass of circulating capital: a favourite proposition among economists. But in fact, the mass of means of subsistence the worker draws off the market, and the mass of means of subsistence the capitalist has available for his own consumption, depend on the ratio between surplus-value and the price of labour.
Ricardo, like Barton, everywhere confuses the ratio of variable to constant capital with the ratio of circulating to fixed capital. We'll see later how this distorts his inquiry into the rate of profit.
Ricardo also treats differences in turnover that arise from quite other causes as if they were the same as the difference between fixed and circulating capital:
"It should also be noted that circulating capital can circulate, or return to the person using it, over very unequal lengths of time. Wheat bought by a farmer for sowing is fixed capital compared with wheat bought by a baker to turn into bread. The farmer leaves it in the ground and can't get a return before a year is up; the baker can have it ground into flour and sell it to his customers as bread, so that within a week he has his capital free again to start the same operation over, or any other."
What's telling here is that wheat, even though it's serving as seed — not as a means of subsistence but as raw material — counts, first, as circulating capital because it is in itself a means of subsistence, and second, as fixed capital because its return takes over a year. But it isn't only a slower or faster return that makes a means of production fixed capital — it's the particular way it gives off its value to the product.
The confusion set going by Adam Smith has led to the following results:
1. The difference between fixed and fluid capital gets confused with the difference between productive capital and commodity-capital. So, for instance, the very same machine counts as circulating capital while it sits on the market as a commodity, and as fixed capital once it's taken up into the production process. Yet there's no seeing, on this basis, why one kind of capital should be any more fixed or any more circulating than the other.
2. All circulating capital gets identified with capital laid out, or to be laid out, on wages. This is what happens with J. S. Mill and others.
3. The difference between variable and constant capital — already confused, in Barton, Ricardo and others, with the difference between circulating and fixed — finally gets reduced entirely to the latter. In Ramsay, for instance, all means of production, raw materials as well as instruments, count as fixed capital, and only the capital laid out on wages counts as circulating capital. But because the reduction happens in this form, the real difference between constant and variable capital never gets grasped at all.
4. Among the most recent English economists, especially the Scottish ones, who view everything from the hopelessly narrow standpoint of a bank clerk — Macleod, Patterson and others — the difference between fixed and circulating capital turns into the difference between money at call and money not at call: deposit money that can be withdrawn without notice, or only after giving notice.