In some branches of production the capital employed is of the mean or average composition — that is, exactly or approximately the same composition as the average capital of society as a whole.
In these branches, the price of production of the commodities produced coincides exactly or approximately with their value expressed in money. If there were no other way of arriving at a mathematical limit, this would be the way. Competition spreads the social capital across the different branches so that the price of production in each branch is built on the same pattern as in these branches of mean composition, that is: k + kp' — cost price plus cost price multiplied by the average rate of profit. That average rate of profit is nothing but the ordinary percentage profit earned in the branches of mean composition — the branches where profit coincides with surplus-value. So the rate of profit ends up the same in every branch, equalized to the rate holding in these branches of mean composition, where the composition of capital matches the average composition of society as a whole.
It follows that the sum of the profits across all the branches must equal the sum of the surplus-values, and the sum of the prices of production of the whole social product must equal the sum of its values. And it is clear that the equalization between branches of differing composition must always tend to adjust them to the branches of mean composition — whether those branches match the social average exactly or only approximately. Even among the ones that only approximate it, to varying degrees, a further tendency toward equalization operates, one that strives toward an ideal middle position — a position that does not actually exist anywhere — tending to settle around it.
In this way a tendency necessarily prevails to turn prices of production into nothing but transformed forms of value, and profits into nothing but portions of surplus-value — portions distributed not in proportion to the surplus-value actually produced in each particular branch, but in proportion to the mass of capital employed in each branch. So equal masses of capital, however they are composed, draw equal shares — equal aliquot parts — of the total surplus-value produced by the whole social capital.
For capitals of mean composition, or approximately mean composition, the price of production coincides exactly or approximately with the value, and the profit with the surplus-value these capitals actually produce. Every other capital, whatever its own composition, strives under the pressure of competition to come level with these. And because capitals of mean composition equal, or approximately equal, the average capital of society, every capital — whatever surplus-value it itself actually produces — strives to realize, through the prices of its commodities, the average profit in place of that surplus-value. In other words, it strives to realize the price of production.
It can also be put this way: wherever an average profit is established — that is, a general rate of profit — however that result comes about, this average profit can only be the profit earned on the average capital of society — a profit whose total sum equals the total sum of surplus-value. And the prices produced by adding this average profit onto cost prices can only be values that have been turned into prices of production.
Nothing would change if capitals in certain branches were, for whatever reason, not drawn into this process of equalization. The average profit would then simply be calculated on the part of the social capital that does enter the equalization process. What is clear is that the average profit can be nothing but the whole mass of surplus-value, shared out over the masses of capital in each branch in proportion to their size. It is the whole of the unpaid labour that has been realized, and this total mass is represented — just as the paid labour is, dead and living — in the total mass of commodities and money that falls to the capitalists.
The genuinely difficult question is this: how does this equalization of profits into a general rate of profit actually come about, given that the general rate is obviously a result and cannot be a starting point?
It is clear, to begin with, that assessing the values of commodities — in money, say — can only be the result of their being exchanged. So if we take such an assessment as given, we have to treat it as the outcome of real exchanges of one commodity's value against another's. But how is this exchange of commodities at their real values supposed to have come about in the first place?
Suppose, first, that all commodities in the different branches of production were sold at their real values. What would follow? Given what has already been worked out, very different rates of profit would prevail from branch to branch. On the face of it, these are two quite different things: whether commodities are sold at their values — that is, exchanged with each other in proportion to the value each contains, at value-prices — or whether they are sold at prices that yield equally large profits on equal masses of capital advanced for producing them.
That capitals setting unequal amounts of living labour in motion produce unequal amounts of surplus-value presupposes, at least to some extent, that the degree of exploitation of labour — the rate of surplus-value — is the same everywhere, or that the differences that do exist count as balanced out by grounds of compensation, real or merely imagined (conventional) ones. This in turn presupposes competition among the workers, and an equalization brought about by their constant movement from one branch of production to another.
Such a general rate of surplus-value — a tendency, like all economic laws — is something we have assumed here as a theoretical simplification. In reality, though, it is an actual presupposition of the capitalist mode of production, even if it is held back, more or less, by practical frictions that produce more or less significant local differences — for instance, the settlement laws for agricultural day-labourers in England. But in theory it is assumed that the laws of the capitalist mode of production develop in a pure form. In reality there is always only approximation — though this approximation is greater the more fully the capitalist mode of production has developed, and the more it has been cleared of contamination and entanglement with the remnants of earlier economic conditions.
The whole difficulty enters because commodities are not exchanged simply as commodities, but as products of capitals that lay claim to shares in the total mass of surplus-value in proportion to their size, and to equal shares where the size is equal. And the total price of the commodities a given capital produces in a given period is supposed to satisfy that claim. But the total price of these commodities is nothing but the sum of the prices of the individual commodities that make up the capital's product.
The salient point comes out most clearly if we take the matter this way: suppose the workers themselves owned their own means of production, and exchanged their commodities directly with each other. These commodities would then not be products of capital. Depending on the technical nature of their work, the value of the tools and materials used in the different trades would differ. And quite apart from that unequal value of the means of production used, different amounts of them would be needed for a given amount of labour, depending on whether a particular commodity can be finished in an hour, or only takes a whole day. Suppose further that these workers, on average, work equal amounts of time, once you allow for the adjustments that come from differences in intensity and the like. Then both workers would, in the commodities that are the product of their day's labour, first replace their outlays — the cost prices of the means of production used up, which would differ according to the technical nature of their trades. Second, both would have created the same amount of new value, namely the working day added to those means of production. This new value would include their wage, plus the surplus-value produced by labour beyond their own necessary needs — a surplus-value whose result would belong to themselves. Putting it in capitalist terms: both would receive the same wage plus the same profit, equal to the value expressed, say, in the product of a ten-hour working day.
But, first, the values of their commodities would be different. Commodity I, say, would contain a larger value-part for the means of production used than commodity II — and, to bring in every possible difference at once, suppose commodity I absorbs more living labour and so needs a longer labour-time to make than commodity II. So the value of I and the value of II are very different, and so are the total sums of value produced by worker I and worker II in a given time. The profit rates, too, would be very different for I and II, if we call the profit rate here the ratio of the surplus-value to the total value of the means of production laid out. The means of subsistence that I and II consume every day during production — and which here take the place of a wage — form, in this case, the part of the advanced means of production that we elsewhere call variable capital.
But the surplus-value produced would be the same for I and II for equal labour-time. Or, more exactly: since I and II each receive the value of the product of one working day, after deducting the value of the advanced 'constant' elements, they receive equal values — part of which can be counted as replacing the means of subsistence consumed in production, and the rest as a surplus-value above that. If I has laid out more, this is made up for by the larger value-part of his commodity that replaces this 'constant' part, so he also has to convert a correspondingly larger part of the total value of his product back into the material elements of that constant part — while II, collecting less for it, has correspondingly less to convert back. Under this supposition, then, the difference in the profit rates would be a matter of no consequence — just as today it is a matter of no consequence to the wage-worker what profit rate the amount of surplus-value squeezed out of him gets expressed in, and just as in international trade the difference in profit rates between nations is a matter of no consequence for their exchange of commodities.
Exchanging commodities at their values, or close to their values, therefore requires a much lower stage of development than exchanging them at prices of production, which needs a certain level of capitalist development to be reached.
However the prices of the different commodities may first be fixed or settled against each other, the law of value governs their movement. Where the labour-time needed to produce them falls, prices fall; where it rises, prices rise — other things being equal.
Setting aside the fact that the law of value governs prices and their movement, it is entirely fitting to regard the values of commodities — not only theoretically but historically too — as coming before prices of production. This holds for conditions where the means of production belong to the worker, a condition found, in both the ancient and the modern world, in the self-employed, land-owning peasant and in the craftsman. It also agrees with what we said earlier, that the development of products into commodities arises out of exchange between different communities, not between the members of one and the same community.
And just as it holds for this original condition, so it holds for the later conditions resting on slavery and serfdom, and for the guild organization of handicrafts — for as long as the means of production tied down in each branch of production can only be transferred from one branch to another with difficulty, so that the different branches of production stand toward each other, within certain limits, like foreign countries or communistic communities.
For the prices at which commodities exchange with each other to correspond approximately to their values, nothing more is needed than this: (1) that the exchange of the different commodities stops being purely accidental or merely occasional; (2) so far as we are looking at direct exchange of commodities, that these commodities are produced on both sides in quantities that correspond roughly to each side's need for the other — something the two sides learn from trading with each other, so that the right proportions settle in as exchange goes on; and (3) so far as we are talking about selling, that no natural or artificial monopoly puts either side to the exchange in a position to sell above value, or forces it to sell below value. By an accidental monopoly we mean the monopoly that falls to a buyer or a seller from the accidental state of supply and demand.
Saying that commodities from different branches of production sell at their values only means this: value is the point their prices keep gravitating around, the point their constant rises and falls balance out to.
Beyond that, there is always a market value — more on this later — that has to be kept separate from the individual value of the particular commodities each producer turns out. Some of these commodities have an individual value below the market value (that is, less labour-time was needed to make them than the market value represents); others have an individual value above it. The market value can be looked at two ways: as the average value of everything produced in a sphere, and as the individual value of the commodities made under that sphere's average conditions — the conditions under which the great bulk of its output is made. Only in unusual combinations of circumstances do the commodities made under the worst conditions, or those made under the most favourable conditions, govern the market value, which in turn forms the centre that market prices fluctuate around — market prices being the same for all commodities of the same kind.
If the supply of commodities at the average value — the middle value of the mass, lying between the two extremes — meets the ordinary demand, then the commodities whose individual value stands below the market value realize an extra surplus-value, a surplus profit, while the commodities whose individual value stands above the market value cannot realize the whole of the surplus-value contained in them.
It doesn't help to say that the sale of commodities made under the worst conditions proves they are needed to meet demand. If the price in the case just assumed were higher than the middle market value, demand would be smaller. At a given price a kind of commodity can take up a certain amount of room in the market; that room stays the same when prices change only if a higher price goes together with a smaller quantity and a lower price with a larger quantity.
If, on the other hand, demand is so strong that it does not shrink when the price is governed by the value of the commodities made under the worst conditions, then those commodities govern the market value. This is only possible if demand exceeds its ordinary level or supply falls below it. Finally, if the mass of commodities produced is larger than can be sold at the middle market value, then the commodities made under the best conditions govern the market value. Those producing under the best conditions can, for instance, sell their commodities wholly or nearly at their individual value, and it can then happen that the commodities made under the worst conditions do not even realize their cost price, while those produced under the middle conditions can realize only part of the surplus-value contained in them.
Whatever is said here about market value holds equally for the price of production, once the price of production has taken the place of market value. The price of production is regulated in every sphere, and regulated too according to its particular circumstances. But the price of production is itself in turn the centre around which daily market prices revolve and to which they are equalized over given periods.
(See Ricardo, on the determination of the price of production by those working under the worst conditions.)
Whatever governs prices, this follows:
(1) The law of value governs their movement: a decrease in the labour-time needed for production makes prices of production fall, and an increase makes them rise. It is in this sense that Ricardo says (he clearly feels that his prices of production diverge from the values of commodities) that "the inquiry to which he wishes to draw the reader's attention, relates to the effect of the variations in the relative value of commodities, and not in their absolute value."
(2) The average profit that determines prices of production must always be roughly equal to the quantity of surplus-value that falls to a given capital as its aliquot part — its proportional share — of the total social capital. Suppose the general rate of profit, and so the average profit, expressed in money, were higher than the real average surplus-value calculated in money. As far as the capitalists themselves are concerned, it makes no difference whether they credit each other with 10 per cent or 15 per cent profit. One percentage covers no more real commodity-value than the other, since the money expression is inflated on both sides equally.
But as for the workers — since it is assumed that they receive their normal wage, so that raising the average profit does not express a real deduction from wages, that is, something quite different from the capitalist's normal surplus-value — any rise in commodity prices produced by raising the average profit must be matched by a rise in the money expression of variable capital. In fact, such a general nominal rise in the rate of profit and the average profit, beyond the rate given by the proportion of the real surplus-value to the total capital advanced, is not possible without also bringing about a rise in wages, and equally a rise in the prices of the commodities that make up constant capital. The same holds in reverse for a fall.
Since the total value of the commodities regulates the total surplus-value, and the total surplus-value in turn regulates the level of the average profit and so of the general rate of profit — as the general law, or as what governs the fluctuations — the law of value regulates prices of production.
What competition achieves, first of all within a single sphere, is the formation of one common market value and market price out of the different individual values of the commodities. Competition among capitals across the different spheres, however, is what first brings about the price of production, which equalizes the rates of profit between the different spheres. The second of these needs a higher development of the capitalist mode of production than the first.
For commodities from the same sphere of production, of the same kind and roughly the same quality, to sell at their values, two things are needed:
(1) First, the different individual values must be equalized into one social value, the market value already described, and this requires competition among the producers of the same kind of commodity, as well as the existence of a market where they all offer their commodities together. For the market price of identical commodities — each produced under somewhat different individual conditions — to match the market value, neither rising above it nor falling below it, the pressure the different sellers put on one another has to be strong enough to throw onto the market the quantity of commodities that social need calls for, that is, the quantity for which society is able to pay the market value. If the mass of commodities produced exceeded this need, the commodities would have to be sold below their market value; conversely, they would sell above their market value if the mass of commodities were not large enough, or, what comes to the same thing, if the pressure of competition among the sellers were not strong enough to force them to bring this mass of commodities to market.
If the market value changed, the conditions under which the whole mass of commodities could be sold would change too. When the market value falls, social need — which here always means need that has money to back it up — expands on average and can, within certain limits, absorb larger masses of the commodity. When the market value rises, social need for the commodity contracts and smaller masses of it are absorbed.
So while demand and supply regulate the market price, or rather the deviations of market prices from the market value, the market value, for its part, regulates the relation between demand and supply, or the centre around which the fluctuations of demand and supply make market prices oscillate.
Looking at the matter more closely, one finds that the conditions holding for the value of the single commodity reproduce themselves here as conditions for the value of the whole sum of a kind of commodity. Capitalist production is from the outset mass production; and even other, less developed modes of production — at least for the main commodities — take what is produced in small quantities by many small individual producers and concentrate it, heap it up, and bring it to market in large masses in the hands of relatively few merchants, as the joint product of a whole branch of production, or of a larger or smaller portion of it.
It may be noted here, quite in passing, that "social need" — that is, what governs the principle of demand — is essentially conditioned by the relation of the different classes to one another and by their respective economic position: firstly by the proportion of total surplus-value to wages, and secondly by the proportion between the different parts into which surplus-value is split (profit, interest, ground-rent, taxes, and so on). So here too it becomes clear that absolutely nothing can be explained from the relation of demand and supply until the basis on which that relation plays out has been worked out.
Even though both commodity and money are unities of exchange-value and use-value, we already saw (in Volume 1) how, in buying and selling, the two split apart at the two ends of the exchange, so that the commodity (the seller) stands for use-value and the money (the buyer) stands for exchange-value. That the commodity should have use-value, that is, should satisfy a social need, was one precondition of the sale. The other was that the quantity of labour contained in the commodity should represent socially necessary labour, so that the individual value of the commodity — which, on this assumption, is the same as its selling price — coincides with its social value.
Let us apply this to the mass of commodities on the market that forms the product of a whole sphere.
The matter is shown most easily if we treat the whole mass of commodities — to begin with, of one branch of production — as a single commodity, and add together the sum of the prices of the many identical commodities into one price. What was said above for the single commodity now holds word for word for the mass of commodities of a given branch found on the market. That the individual value of the commodity should correspond to its social value is now realized, or further specified, as this: the total quantity contains the social labour necessary for its production, and the value of this mass equals its market value.
Now suppose the great bulk of these commodities is produced under roughly the same normal social conditions, so that this value is also the individual value of the particular commodities making up that bulk. Suppose further that a relatively small part is produced below these conditions, and another part above them, so that the individual value of the one part is greater, and that of the other smaller, than the middle value of the great bulk of commodities — but these two extremes balance each other out, so that the average value of the commodities belonging to them equals the value of the commodities belonging to the middle bulk. Then the market value is determined by the value of the commodities produced under the middle conditions.
The value of the whole mass of commodities equals the real sum of the values of all the individual commodities taken together — those produced within the middle conditions as much as those produced below or above them. In this case, the market value, or the social value of the mass of commodities, is the labour-time necessarily contained in them, determined by the value of the great middle bulk.
Suppose instead that the total quantity of the commodity in question brought to market stays the same, but the value of the commodities produced under worse conditions is no longer balanced out by the value of those produced under better conditions, so that the part of the mass produced under worse conditions forms a relatively significant quantity, both against the middle bulk and against the other extreme. Then it is the mass produced under worse conditions that governs the market value, or the social value.
Finally, suppose the mass of commodities produced under better than middle conditions considerably exceeds the mass produced under worse conditions, and itself forms a significant quantity against the mass produced under middle conditions. Then it is the part produced under the best conditions that governs the market value.
Here we leave aside the flooding of the market, where the part produced under the best conditions always governs the market price; what concerns us here is not the market price, insofar as it differs from the market value, but the different ways in which the market value itself is determined.
In fact, taken quite strictly (which of course in reality only ever occurs approximately, modified in a thousand ways), in case I the market value of the whole mass, governed by the average values, equals the sum of the individual values of all the commodities in it — although for the commodities produced at the extremes, this value presents itself as an average value forced upon them. Those producing at the worst extreme then have to sell their commodities below their individual value; those at the best extreme sell them above it.
In case II, the individual masses of value produced at the two extremes do not balance out; instead, the part produced under worse conditions tips the scale. Strictly speaking, the average price or market value of each individual commodity, or of each aliquot part — each proportional part — of the whole mass, would now be determined by the total value of the mass — arrived at by adding together the values of the commodities produced under the different conditions — and by the proportional part of this total value that falls to the single commodity. The market value obtained this way would stand above the individual value not only of the commodities belonging to the favourable extreme but also of those belonging to the middle layer; yet it would still stand lower than the individual value of the commodities produced at the unfavourable extreme. How far it approaches this unfavourable extreme, or finally coincides with it, depends entirely on how much room the mass of commodities produced at the unfavourable extreme takes up in the sphere of commodities in question. If demand only slightly predominates, the individual value of the unfavourably produced commodities governs the market price.
Finally, if, as in case III, the quantity of commodities produced at the favourable extreme takes up more room, not only compared with the other extreme but compared with the middle conditions, then the market value falls below the middle value. The average value, calculated by adding together the sums of value of both extremes and the middle, here stands below the value of the middle and moves closer to or further from it depending on the relative room the favourable extreme takes up. If demand is weak against supply, the favourably placed part, however large it may be, forcibly makes room for itself by drawing its price down to its individual value. The market value can never coincide with this individual value of the commodities produced under the best conditions, except where supply very strongly outweighs demand.
This determination of the market value, presented here in the abstract, is mediated on the actual market by competition among the buyers, provided that demand is just large enough to absorb the mass of commodities at the value thus determined. And here we come to the other point.
(2) That a commodity has a use-value means only that it satisfies some social need. As long as we were dealing with single commodities, we could simply assume the need for that particular commodity was there — its size already folded into the price — without asking how large that need actually was. But once the output of a whole branch of production stands on one side and the social need for it stands on the other, that size becomes essential. We now have to look at the size, the quantity, of this social need.
The account of market value given above assumed that the mass of commodities produced stays the same — a given quantity — and that only the mix of parts within that mass, produced under different conditions, shifts, which is what changes how the market value of that same mass gets regulated. Suppose this mass is the customary quantity supplied, setting aside for now that part of what is produced might be temporarily held off the market.
If the demand for this mass also remains the customary demand, the commodity sells at its market value, whichever of the three cases already examined regulates that value. The mass of commodities then satisfies not just a need, but satisfies it to its full social extent. But if the quantity is smaller or larger than the demand for it, the market price deviates from the market value. And the first such deviation is this: when the quantity is too small, the commodity produced under the worst conditions always regulates the market value, and when it is too large, the commodity produced under the best conditions always regulates it — so that one of the extremes determines the market value, even though the mere proportions of the masses produced under the different conditions would, taken alone, point to a different result. The bigger the gap between demand and the quantity produced, the further the market price deviates from the market value, upward or downward.
This gap between the quantity of commodities produced and the quantity at which they sell at their market value can arise from two different causes. Either the quantity itself changes, becoming too small or too large, so that reproduction has taken place on a different scale than the one that regulated the given market value — here supply has changed while demand stayed the same, and relative overproduction or underproduction has set in. Or reproduction, that is supply, stays the same, but demand has fallen or risen for various reasons — here the absolute size of supply is unchanged, but its size relative to, measured against, the need has changed; the effect is the same as in the first case, only in the opposite direction. Finally, when changes happen on both sides at once — in opposite directions, or in the same direction but not to the same degree, in short, when both sides shift together in a way that changes their earlier proportion — the end result always comes down to one of the two cases just considered.
The real difficulty in giving a general definition of demand and supply is that the definition seems to run in a circle. Let us start with supply: the product on the market, or what can be delivered to it. To avoid needless detail, let us take the mass of the annual output in each particular branch of industry, setting aside how much more or less easily different commodities can be held back from the market and stored for next year's consumption, say. This annual output first of all expresses a definite quantity — a measure or a number, depending on whether that kind of commodity is measured as a continuous amount or counted in discrete units. These are not just use-values that satisfy human needs; these use-values are available on the market on a given scale.
Second, this quantity of commodities has a definite market value, which can be expressed as a multiple of the market value of the single commodity, or of the unit used to measure it. So there is no necessary connection between the quantity of commodities on the market and their market value — some kinds of commodity are worth a great deal by the unit, others very little, so that a given sum of value can show up as a very large quantity of one commodity and a very small quantity of another. The only connection between the quantity of articles on the market and their market value is this: on a given basis of the productive power of labour, producing a definite quantity of articles in any particular sphere of production requires a definite quantity of social labour-time — though this relationship is quite different from one sphere of production to another, and has no inner connection at all with how useful these articles are, or with the particular nature of their use-values. All else being equal: if the quantity a of a kind of commodity costs b of labour-time, the quantity na costs nb of labour-time.
Further: so far as society wants an article produced to satisfy its needs, it has to pay for it. In fact, since commodity production presupposes a division of labour, society buys these articles by devoting part of its available labour-time to producing them — buys them, that is, with a definite quantity of the labour-time it has at its disposal. The part of society which, through the division of labour, falls to producing this particular article must receive an equivalent in the social labour embodied in the articles that satisfy its own needs.
But there is no necessary connection — only an accidental one — between the total quantity of social labour spent on a social article, that is, between the fraction of society's total labour-power, the share of overall production, that society devotes to this article, on the one hand, and, on the other hand, the extent to which society demands satisfaction of the need that this particular article answers. (Only where production stands under society's own real, predetermining control does society itself create the connection between the amount of social labour-time devoted to producing particular articles and the extent of the social need those articles are meant to satisfy.)
Although each single article, or each definite quantity of a kind of commodity, may contain only the social labour required for its production — so that, seen from that side, the market value of this whole kind of commodity represents only the labour socially necessary for it — still, if the commodity has actually been produced in a quantity exceeding the present social need, part of the social labour-time has been wasted, and the mass of commodities then represents on the market a much smaller quantity of social labour than is really contained in it. These commodities must therefore be sold below their market value, and part of them may even prove completely unsaleable. The reverse holds if the quantity of social labour devoted to producing a particular kind of commodity is too small for the extent of the particular social need the product is meant to satisfy.
But if the quantity of social labour devoted to producing a particular article matches the extent of the social need to be satisfied — so that the mass produced corresponds to the customary scale of reproduction, demand being unchanged — then the commodity sells at its market value. Exchanging or selling commodities at their value is the rational thing, the natural law of their equilibrium; it is from this law that the deviations are to be explained, not the law itself from the deviations.
Let us look now at the other side, demand.
Commodities are bought either as means of production or as means of subsistence — it makes no difference that some kinds of commodity can serve both purposes — in order to enter productive or individual consumption. So demand for them comes from the producers (here, capitalists, since we assume the means of production have been turned into capital) and from the consumers. Both seem, at first, to presuppose on the side of demand a given quantity of social needs, matched on the other side by definite quantities of social production in the various branches of production. If the cotton industry is to carry out its annual reproduction again on the same scale, it needs the customary quantity of cotton for that — and, given the annual expansion of reproduction that comes from the accumulation of capital, other things being equal, an additional quantity of cotton besides. The same holds for means of subsistence. The working class must find at least the same quantity of necessary means of subsistence available again — though perhaps distributed a little differently among the different kinds — if it is to go on living in its customary, average way; and, given the annual growth of population, an additional quantity besides; and so on, with more or less modification, for the other classes.
So it looks as though a definite social need of a certain size stands on the demand side, requiring a definite quantity of some article on the market to satisfy it. But how much of that need there is stretches and shifts constantly. Its fixity is mere appearance. If means of subsistence were cheaper, or money-wages higher, the workers would buy more of them, and a greater "social need" for these kinds of commodity would show itself — quite apart from the paupers and so on, whose "demand" stands even below the narrowest limits of their physical need. If, on the other hand, cotton were cheaper, the capitalists' demand for cotton would grow, and more additional capital would be thrown into the cotton industry, and so on.
It must not be forgotten here that the demand for productive consumption is, on our assumption, the demand of the capitalist, and that his real purpose is the production of surplus-value, so that he produces a given kind of commodity only for that purpose. This does not, on the other hand, prevent him, in so far as he stands on the market as a buyer of, say, cotton, from representing the need for cotton — just as it is a matter of indifference to the seller of cotton whether the buyer turns it into shirting or into gun-cotton, or means to stuff his own and the world's ears with it. All the same, this does have a great influence on the manner in which he is a buyer. His need for cotton is essentially modified by the fact that it really only disguises his need to make a profit.
The limits within which the need for commodities represented on the market — demand — differs in quantity from the real social need are naturally very different from one commodity to another; I mean the difference between the quantity of a commodity that is demanded and the quantity that would be demanded at different money-prices for it, or with the buyers in different money and living circumstances.
Nothing is easier than to see the disproportions between demand and supply, and the resulting deviation of market prices from market values. The real difficulty lies in working out what is meant by demand and supply coinciding.
Demand and supply coincide, it is said, when they stand in such a relationship that the mass of commodities from a particular branch of production can be sold at its market value — neither above it nor below it. That is the first thing we are told.
And the second thing we are told: when the commodities can be sold at their market value, demand and supply coincide.
When demand and supply coincide, they stop acting, and that is exactly why the commodity sells at its market value. When two forces act equally and in opposite directions, they cancel each other out — they have no effect at all on the outside — and anything that happens while they are cancelled has to be explained by something other than those two forces acting on it. So when demand and supply cancel each other out, they stop explaining anything at all: they have no effect on the market value, and they leave us completely in the dark as to why the market value comes out to exactly this sum of money and no other.
The real inner laws of capitalist production plainly cannot be explained by the interplay of demand and supply (setting aside a deeper analysis of these two social driving forces, which has no place here), because those laws only show up in their pure form once demand and supply stop acting — that is, once they coincide. In fact demand and supply never actually coincide; and on the rare occasion when they do, it is accidental — so, scientifically, it should be set at zero, treated as if it had not happened.
Yet political economy assumes that they do coincide. Why? First, in order to look at these phenomena in the lawlike shape that corresponds to their concept — that is, to look at them apart from the appearance produced by the movement of demand and supply. Second, in order to find the real tendency of that movement, to pin it down, so to speak. For the disproportions run in opposite directions, and because they keep following one another, they cancel each other out through those opposite directions, through their contradiction with one another.
So even though demand and supply never coincide in any single given case — a deviation in one direction is what produces another deviation in the opposite direction — their inequalities still follow one another in such a way that, once a longer or shorter stretch of time is taken as a whole, supply and demand constantly coincide. But only as the average of the movement that has already happened, and only as the constant movement of their contradiction. In this way market prices, which deviate from market values, balance out to market values when taken as an average figure — the deviations from that value cancelling out as pluses and minuses. And this average figure is by no means only of theoretical importance; it matters practically for capital, whose investment is calculated on the fluctuations and the way they even out over a more or less definite period of time.
The relationship between demand and supply, then, only explains two things: on one side, why market prices drift away from market values, and on the other, the tendency for that drift to be removed — that is, for the effect of the demand-supply relationship to cancel itself out. (Commodities that have a price without having a value are a separate case, not considered here.)
Demand and supply can cancel that effect in very different ways. Say demand falls, so the market price falls too — that can lead capital to withdraw, shrinking the supply. Or it can lead to the market value itself falling, through inventions that shorten the labour-time required, bringing it back into line with the market price.
The reverse also happens: demand rises, and with it the market price rises above the market value. That can draw too much capital into this branch of production, pushing output up until the market price itself falls below the market value; or, on the other hand, it can drive prices up so far that demand itself is pushed back down. In some branches of production it can even happen that the market value itself rises, for shorter or longer stretches, because part of what is demanded now has to be produced under worse conditions during that time.
If demand and supply determine the market price, then, the other way round, the market price — and, looking deeper, the market value — determines demand and supply. This is obvious enough for demand: it moves opposite to the price, rising when the price falls and falling when it rises.
But it holds for supply too. The prices of the means of production that go into a supplied commodity determine the demand for those means of production, and so also the supply of the commodity itself, since supplying that commodity means demanding those means of production. Cotton prices determine the supply of cotton goods.
On top of this confusion — prices determined by demand and supply, and, at the same time, demand and supply determined by prices — comes another: that demand determines supply and supply determines demand, that production determines the market and the market determines production.
Even the ordinary economist can see that, without any change in supply or need brought about by outside circumstances, the relationship between the two can still shift purely because the market value of the commodity has changed. Even he has to admit that, whatever the market value turns out to be, demand and supply must balance out to produce it. In other words, the relationship between demand and supply does not explain the market value — on the contrary, the market value explains the fluctuations in demand and supply. The author of the Observations continues, after the passage cited in the note:
The author of the Observations puts it like this: this proportion between demand and supply — if 'demand' and 'natural price' still mean what they meant when we were discussing Adam Smith — must always be a proportion of equality. That is because the natural price is only actually paid when the supply matches the effectual demand: the demand of buyers who will pay neither more nor less than the natural price. So there can perfectly well be two very different natural prices for the same commodity at different times, and yet the proportion which the supply bears to the demand can be exactly the same both times — the proportion of equality.
So it is conceded: with two different 'natural prices' for the same commodity at different times, demand and supply can and must match up on both occasions, if the commodity is to sell at its 'natural price' each time. Since there is no difference at all in the relationship between demand and supply from the one case to the other, but there is a difference in the size of the 'natural price' itself, then that 'natural price' is clearly fixed independently of demand and supply — and is the last thing they could be said to fix.
For a commodity to sell at its market value — that is, in proportion to the socially necessary labour it contains — the total amount of social labour spent on the whole mass of that kind of commodity has to match the quantity of social need for it: the social need with money to back it up.
Competition — through the fluctuations in market prices that track the fluctuations in the relationship of demand and supply — constantly works to bring the total labour spent on each kind of commodity down to this measure.
In the relationship between demand and supply for commodities, two earlier relationships turn up again. First, the relationship of use-value and exchange-value, of commodity and money, of buyer and seller. Second, the relationship of producer and consumer, even though both may be represented by merchants acting as third parties.
When we looked at buyer and seller on their own, it was enough to set them against each other one by one to work out the relationship. Three people are enough for the whole metamorphosis of a commodity — the whole of buying and selling. A turns his commodity into B's money by selling it to B, and then turns that money back into a commodity by buying from C; the whole process runs between these three.
There is more. When we looked at money, we assumed commodities sell at their value, because there was no reason then to consider prices that departed from value — the question was only the change of form a commodity goes through in becoming money and turning back from money into a commodity. As soon as a commodity is sold at all, and something new bought with the proceeds, the whole metamorphosis is complete before us, and for that metamorphosis, taken on its own, it makes no difference whether the commodity's price stood below or above its value. The commodity's value still matters as the foundation, though, because any rational understanding of money has to start from that foundation, and price, in its general concept, is in the first place just value in the form of money.
That said, when we looked at money as a means of circulation, we were assuming that more than one commodity's metamorphosis was going on — we were looking at how these metamorphoses are socially interwoven. Only that way do we get to the circulation of money and the development of its role as a means of circulation. But important as that connection is for money's move into that role and for the changed shape it takes on because of it, it makes no difference at all to the transaction between the individual buyer and seller.
With supply and demand, on the other hand, supply equals the sum of the sellers or producers of a given kind of commodity, and demand equals the sum of the buyers or consumers — individual or productive — of that same kind of commodity.
These sums act on each other as units, as aggregate forces. Here the individual acts only as part of a social power, as an atom in the mass — and it is in this form that competition asserts the social character of production and consumption.
The side of competition that happens to be weaker at a given moment is also the side where each individual acts independently of the mass of his rivals, and often directly against them — and by doing exactly that, he makes everyone feel how dependent they are on each other. The stronger side, by contrast, always faces its opponent as something closer to a single closed block.
If demand for a particular kind of commodity is greater than supply, then, within certain limits, one buyer outbids another, making the commodity dearer for everyone, above its market value, while on the other side the sellers all seek to sell at a high market price. If, the other way round, supply is greater than demand, one seller starts undercutting, and the others have to follow, while the buyers all work to push the market price as far below the market value as they can.
Belonging to the common side interests anyone only for as long as he gains more with it than he would against it. And that common interest ends the moment that side becomes the weaker one — at which point each person starts trying, on his own account, to wriggle out as well as he can.
Further: if a producer makes something more cheaply and can sell more of it, capturing a bigger share of the market by selling below the going market price or market value, he does exactly that. And this is how the process begins that gradually forces everyone else to introduce the cheaper form of production too, bringing the socially necessary labour down to a new, lower level.
When one side has the upper hand, everyone on it gains, as if they all shared a single monopoly. When a side is the weaker one, each person can try, for himself, to become the stronger party instead — say, by working with lower costs of production — or at least to come out of it as well as possible; and here nobody cares in the least about his neighbour, even though what he does touches not just himself but all his fellows too.
Demand and supply presuppose that value has already been transformed into market value. And so far as they operate on a capitalist basis — so far as the commodities are products of capital — they presuppose capitalist processes of production: relationships far more tangled than the mere buying and selling of commodities. What is at stake with them is not the formal turning of a commodity's value into a price, a mere change of form; it is the definite, quantitative ways market prices depart from market values, and, further, from prices of production.
For simple buying and selling, it is enough to have commodity producers facing each other as such. Demand and supply, looked at more closely, presuppose the existence of the different classes and sections of classes who divide the total revenue of society among themselves and consume it among themselves as revenue — and so form the demand that revenue creates. And to understand the demand and supply formed among the producers themselves, you need, on the other hand, insight into the whole shape of the capitalist process of production.
Capitalist production is not just about throwing commodities into circulation and pulling back an equal mass of value in another form — money, or some other commodity. It is about pulling out, for the capital advanced to production, the same surplus-value or profit that any other capital of the same size would get, whatever branch it happens to be invested in — or, put the other way, a profit in proportion to its size. So it comes down to this: commodities have to sell, at a minimum, at prices that yield the average profit — that is, at prices of production.
It is in this form — the demand that equal capital bring equal profit in any branch — that capital comes to know itself as one social power, and each capitalist holds a share of that power the size of his share of the total social capital.
First: capitalist production, in itself, does not care what particular use-value it produces — it does not care about the specific character of the commodity at all. In any branch, all it wants is to produce surplus-value, to seize a certain quantity of unpaid labour out of the product of labour.
And it is just as much in the nature of wage-labour under capital that it has to be indifferent to the specific character of its own work — it has to remake itself to fit capital's needs, and let itself be thrown from one branch of production into another.
Second: in fact, one branch of production is just as good — and just as bad — as another — each yields the same profit. But each would be pointless if the commodity it produces did not satisfy some kind of social need.
If commodities sold at their values, then, as we have seen, very different rates of profit would arise in the different branches, depending on the differing organic composition of the capital invested in each. But capital pulls out of a branch with a low rate of profit and throws itself into one that yields a higher one. Through this constant migration of capital — through capital distributing itself between the branches, leaving where the rate is falling and entering where it is rising — it so alters the relation of supply to demand that the average profit becomes the same across the different branches of production, and so values turn into prices of production.
This equalization out succeeds for capital more or less completely, depending on how far capitalist production has developed in a given country — that is, how far conditions there have been adapted to the capitalist mode of production. As capitalist production advances, the conditions it needs advance with it, and it brings the whole social setting in which production takes place under its own specific character and under the laws immanent in it.
This constant equalization of inequalities that are constantly renewed happens faster, (1) the more mobile capital is — the easier it is to move from one branch, and from one place, to another — and (2) the faster labour-power can be thrown from one branch into another, and from one local point of production to another.
The first condition assumes complete freedom of trade within society, and the removal of every monopoly except the natural ones — the ones that spring from the capitalist mode of production itself. It also assumes a developed credit system, which draws the otherwise scattered mass of available social capital together into one lump confronting the individual capitalist; and, finally, that the different branches of production are run by capitalists. That last point is already built into the assumption, once we are talking about values turning into prices of production across all the branches capitalism exploits — but the equalization itself meets bigger obstacles where large numbers of branches not run on capitalist lines (small farmers working the land, say) wedge themselves in among capitalist enterprises and get bound up with them. And, finally, a densely settled population.
The second condition assumes the abolition of every law that stops workers moving from one branch of production to another, or from one place where production is carried on to any other. It assumes that the worker does not care what particular kind of work he does. It assumes labour reduced, as far as possible, to simple labour in every branch. It assumes the disappearance of every prejudice of trade and craft among the workers. And, finally and above all, it assumes the worker's subjection to the capitalist mode of production. Carrying this further belongs to a special study of competition.
From all this it follows that every individual capitalist, like all the capitalists of any particular branch together, has a stake in the exploitation of the whole working class by the total capital, and in how far that exploitation goes. And it is not just general class sympathy — it is a direct economic interest. Because, taking everything else as given, including the value of the total constant capital advanced, the average rate of profit depends on how far the total capital exploits the total labour.
The average profit coincides with the average surplus-value that capital produces per 100 — and as far as surplus-value goes, that is obvious from what has already been said. With average profit, the only thing added is the value of the capital advanced, as one of the things that fixes the rate of profit. In fact, the particular interest a capitalist, or the capital of one particular branch, takes in exploiting the workers he directly employs is limited to this: he can make an extra cut, a profit reaching above the average, either through exceptional overwork, or by pushing wages below the average, or through exceptional productivity in the labour applied.
Apart from that, a capitalist who employed no variable capital at all in his branch, and so no workers whatsoever (in fact an exaggerated assumption), would have just as much of a stake in the exploitation of the working class by capital, and would draw his profit just as much from unpaid surplus labour, as a capitalist who — again an exaggerated assumption — employed only variable capital, laying out his whole capital in wages.
But the degree to which labour is exploited depends, given a fixed working day, on the average intensity of the work, and given a fixed intensity, on the length of the working day. The rate of surplus-value depends on that degree of exploitation — and so, given a fixed total variable capital, does the size of the surplus-value, and with it the size of the profit. The special interest that the capital of one branch has, as against the total capital, in exploiting the workers it specifically employs, is the same interest an individual capitalist has, as against his own branch, in exploiting the workers he personally exploits.
On the other hand, every particular branch of capital, and every individual capitalist, shares one and the same interest in the productivity of the social labour that the total capital sets to work. Two things depend on that:
First: the mass of use-values in which the average profit is expressed — and this matters twice over, since that profit serves both as a fund for accumulating new capital and as a fund of revenue to be consumed. Second: the value level of the whole capital advanced, constant and variable together, which, given a fixed amount of surplus-value or profit for the capitalist class as a whole, fixes the rate of profit, or the profit on a given quantity of capital.
The particular productivity of labour in one branch, or in one particular firm within that branch, interests only the capitalists directly involved — and only insofar as it puts that branch in a position to make an extra profit as against the total capital, or that capitalist in a position to make one as against his own branch.
So here we have the mathematically exact proof of why the capitalists, however much they show themselves to be false brothers in their competition with one another, still form a true freemasonry against the working class as a whole.
The price of production includes the average profit. We gave it that name, price of production, but it is in fact the very same thing that Adam Smith calls the natural price, that Ricardo calls the price of production and the cost of production, and that the Physiocrats call the prix nécessaire, the necessary price — though none of them worked out how the price of production differs from value. It gets that name because, in the long run, it is the condition on which the supply, the reproduction, of a commodity in any particular branch of production depends.
This also explains why the very economists who will not accept that the value of commodities is determined by labour-time, by the quantity of labour contained in them, nonetheless keep talking about prices of production as the centres that market prices fluctuate around. They can allow themselves this because the price of production is already a form of commodity-value that has been thoroughly externalized — on the face of it an irrational form, one carrying no trace of its own concept, the form value takes as it appears in competition, that is, in the mind of the vulgar capitalist — the capitalist as competition teaches him to think — and so in that of the vulgar economists too.
We saw that market value — and everything said about it here applies too, with the necessary adjustments, to the price of production — includes a surplus profit for whoever produces under the best conditions in any given branch. Leaving aside cases of crisis and general overproduction, this holds for every market price, however far it departs from market values or market prices of production. What is built into the market price is this: the same price gets paid for commodities of the same kind, even though they may have been produced under very different individual conditions, and so carry very different cost prices. (We are not talking here about surplus profits that come from monopolies in the ordinary sense, whether artificial or natural.)
But a surplus profit can also arise in another way: if certain branches of production are in a position to escape having their commodity-values turned into prices of production, and so escape having their profits reduced to the average.
In the Part on ground-rent we will have to look at how these two forms of surplus profit develop further.
In some branches of production the capital employed is of the mean or average composition — that is, exactly or approximately the same composition as the average capital of society as a whole.
In these branches, the price of production of the commodities produced coincides exactly or approximately with their value expressed in money. If there were no other way of arriving at a mathematical limit, this would be the way. Competition spreads the social capital across the different branches so that the price of production in each branch is built on the same pattern as in these branches of mean composition, that is: k + kp' — cost price plus cost price multiplied by the average rate of profit. That average rate of profit is nothing but the ordinary percentage profit earned in the branches of mean composition — the branches where profit coincides with surplus-value. So the rate of profit ends up the same in every branch, equalized to the rate holding in these branches of mean composition, where the composition of capital matches the average composition of society as a whole.
It follows that the sum of the profits across all the branches must equal the sum of the surplus-values, and the sum of the prices of production of the whole social product must equal the sum of its values. And it is clear that the equalization between branches of differing composition must always tend to adjust them to the branches of mean composition — whether those branches match the social average exactly or only approximately. Even among the ones that only approximate it, to varying degrees, a further tendency toward equalization operates, one that strives toward an ideal middle position — a position that does not actually exist anywhere — tending to settle around it.
In this way a tendency necessarily prevails to turn prices of production into nothing but transformed forms of value, and profits into nothing but portions of surplus-value — portions distributed not in proportion to the surplus-value actually produced in each particular branch, but in proportion to the mass of capital employed in each branch. So equal masses of capital, however they are composed, draw equal shares — equal aliquot parts — of the total surplus-value produced by the whole social capital.
For capitals of mean composition, or approximately mean composition, the price of production coincides exactly or approximately with the value, and the profit with the surplus-value these capitals actually produce. Every other capital, whatever its own composition, strives under the pressure of competition to come level with these. And because capitals of mean composition equal, or approximately equal, the average capital of society, every capital — whatever surplus-value it itself actually produces — strives to realize, through the prices of its commodities, the average profit in place of that surplus-value. In other words, it strives to realize the price of production.
It can also be put this way: wherever an average profit is established — that is, a general rate of profit — however that result comes about, this average profit can only be the profit earned on the average capital of society — a profit whose total sum equals the total sum of surplus-value. And the prices produced by adding this average profit onto cost prices can only be values that have been turned into prices of production.
Nothing would change if capitals in certain branches were, for whatever reason, not drawn into this process of equalization. The average profit would then simply be calculated on the part of the social capital that does enter the equalization process. What is clear is that the average profit can be nothing but the whole mass of surplus-value, shared out over the masses of capital in each branch in proportion to their size. It is the whole of the unpaid labour that has been realized, and this total mass is represented — just as the paid labour is, dead and living — in the total mass of commodities and money that falls to the capitalists.
The genuinely difficult question is this: how does this equalization of profits into a general rate of profit actually come about, given that the general rate is obviously a result and cannot be a starting point?
It is clear, to begin with, that assessing the values of commodities — in money, say — can only be the result of their being exchanged. So if we take such an assessment as given, we have to treat it as the outcome of real exchanges of one commodity's value against another's. But how is this exchange of commodities at their real values supposed to have come about in the first place?
Suppose, first, that all commodities in the different branches of production were sold at their real values. What would follow? Given what has already been worked out, very different rates of profit would prevail from branch to branch. On the face of it, these are two quite different things: whether commodities are sold at their values — that is, exchanged with each other in proportion to the value each contains, at value-prices — or whether they are sold at prices that yield equally large profits on equal masses of capital advanced for producing them.
That capitals setting unequal amounts of living labour in motion produce unequal amounts of surplus-value presupposes, at least to some extent, that the degree of exploitation of labour — the rate of surplus-value — is the same everywhere, or that the differences that do exist count as balanced out by grounds of compensation, real or merely imagined (conventional) ones. This in turn presupposes competition among the workers, and an equalization brought about by their constant movement from one branch of production to another.
Such a general rate of surplus-value — a tendency, like all economic laws — is something we have assumed here as a theoretical simplification. In reality, though, it is an actual presupposition of the capitalist mode of production, even if it is held back, more or less, by practical frictions that produce more or less significant local differences — for instance, the settlement laws for agricultural day-labourers in England. But in theory it is assumed that the laws of the capitalist mode of production develop in a pure form. In reality there is always only approximation — though this approximation is greater the more fully the capitalist mode of production has developed, and the more it has been cleared of contamination and entanglement with the remnants of earlier economic conditions.
The whole difficulty enters because commodities are not exchanged simply as commodities, but as products of capitals that lay claim to shares in the total mass of surplus-value in proportion to their size, and to equal shares where the size is equal. And the total price of the commodities a given capital produces in a given period is supposed to satisfy that claim. But the total price of these commodities is nothing but the sum of the prices of the individual commodities that make up the capital's product.
The salient point comes out most clearly if we take the matter this way: suppose the workers themselves owned their own means of production, and exchanged their commodities directly with each other. These commodities would then not be products of capital. Depending on the technical nature of their work, the value of the tools and materials used in the different trades would differ. And quite apart from that unequal value of the means of production used, different amounts of them would be needed for a given amount of labour, depending on whether a particular commodity can be finished in an hour, or only takes a whole day. Suppose further that these workers, on average, work equal amounts of time, once you allow for the adjustments that come from differences in intensity and the like. Then both workers would, in the commodities that are the product of their day's labour, first replace their outlays — the cost prices of the means of production used up, which would differ according to the technical nature of their trades. Second, both would have created the same amount of new value, namely the working day added to those means of production. This new value would include their wage, plus the surplus-value produced by labour beyond their own necessary needs — a surplus-value whose result would belong to themselves. Putting it in capitalist terms: both would receive the same wage plus the same profit, equal to the value expressed, say, in the product of a ten-hour working day.
But, first, the values of their commodities would be different. Commodity I, say, would contain a larger value-part for the means of production used than commodity II — and, to bring in every possible difference at once, suppose commodity I absorbs more living labour and so needs a longer labour-time to make than commodity II. So the value of I and the value of II are very different, and so are the total sums of value produced by worker I and worker II in a given time. The profit rates, too, would be very different for I and II, if we call the profit rate here the ratio of the surplus-value to the total value of the means of production laid out. The means of subsistence that I and II consume every day during production — and which here take the place of a wage — form, in this case, the part of the advanced means of production that we elsewhere call variable capital.
But the surplus-value produced would be the same for I and II for equal labour-time. Or, more exactly: since I and II each receive the value of the product of one working day, after deducting the value of the advanced 'constant' elements, they receive equal values — part of which can be counted as replacing the means of subsistence consumed in production, and the rest as a surplus-value above that. If I has laid out more, this is made up for by the larger value-part of his commodity that replaces this 'constant' part, so he also has to convert a correspondingly larger part of the total value of his product back into the material elements of that constant part — while II, collecting less for it, has correspondingly less to convert back. Under this supposition, then, the difference in the profit rates would be a matter of no consequence — just as today it is a matter of no consequence to the wage-worker what profit rate the amount of surplus-value squeezed out of him gets expressed in, and just as in international trade the difference in profit rates between nations is a matter of no consequence for their exchange of commodities.
Exchanging commodities at their values, or close to their values, therefore requires a much lower stage of development than exchanging them at prices of production, which needs a certain level of capitalist development to be reached.
However the prices of the different commodities may first be fixed or settled against each other, the law of value governs their movement. Where the labour-time needed to produce them falls, prices fall; where it rises, prices rise — other things being equal.
Setting aside the fact that the law of value governs prices and their movement, it is entirely fitting to regard the values of commodities — not only theoretically but historically too — as coming before prices of production. This holds for conditions where the means of production belong to the worker, a condition found, in both the ancient and the modern world, in the self-employed, land-owning peasant and in the craftsman. It also agrees with what we said earlier, that the development of products into commodities arises out of exchange between different communities, not between the members of one and the same community.
And just as it holds for this original condition, so it holds for the later conditions resting on slavery and serfdom, and for the guild organization of handicrafts — for as long as the means of production tied down in each branch of production can only be transferred from one branch to another with difficulty, so that the different branches of production stand toward each other, within certain limits, like foreign countries or communistic communities.
For the prices at which commodities exchange with each other to correspond approximately to their values, nothing more is needed than this: (1) that the exchange of the different commodities stops being purely accidental or merely occasional; (2) so far as we are looking at direct exchange of commodities, that these commodities are produced on both sides in quantities that correspond roughly to each side's need for the other — something the two sides learn from trading with each other, so that the right proportions settle in as exchange goes on; and (3) so far as we are talking about selling, that no natural or artificial monopoly puts either side to the exchange in a position to sell above value, or forces it to sell below value. By an accidental monopoly we mean the monopoly that falls to a buyer or a seller from the accidental state of supply and demand.
Saying that commodities from different branches of production sell at their values only means this: value is the point their prices keep gravitating around, the point their constant rises and falls balance out to.
Beyond that, there is always a market value — more on this later — that has to be kept separate from the individual value of the particular commodities each producer turns out. Some of these commodities have an individual value below the market value (that is, less labour-time was needed to make them than the market value represents); others have an individual value above it. The market value can be looked at two ways: as the average value of everything produced in a sphere, and as the individual value of the commodities made under that sphere's average conditions — the conditions under which the great bulk of its output is made. Only in unusual combinations of circumstances do the commodities made under the worst conditions, or those made under the most favourable conditions, govern the market value, which in turn forms the centre that market prices fluctuate around — market prices being the same for all commodities of the same kind.
If the supply of commodities at the average value — the middle value of the mass, lying between the two extremes — meets the ordinary demand, then the commodities whose individual value stands below the market value realize an extra surplus-value, a surplus profit, while the commodities whose individual value stands above the market value cannot realize the whole of the surplus-value contained in them.
It doesn't help to say that the sale of commodities made under the worst conditions proves they are needed to meet demand. If the price in the case just assumed were higher than the middle market value, demand would be smaller. At a given price a kind of commodity can take up a certain amount of room in the market; that room stays the same when prices change only if a higher price goes together with a smaller quantity and a lower price with a larger quantity.
If, on the other hand, demand is so strong that it does not shrink when the price is governed by the value of the commodities made under the worst conditions, then those commodities govern the market value. This is only possible if demand exceeds its ordinary level or supply falls below it. Finally, if the mass of commodities produced is larger than can be sold at the middle market value, then the commodities made under the best conditions govern the market value. Those producing under the best conditions can, for instance, sell their commodities wholly or nearly at their individual value, and it can then happen that the commodities made under the worst conditions do not even realize their cost price, while those produced under the middle conditions can realize only part of the surplus-value contained in them.
Whatever is said here about market value holds equally for the price of production, once the price of production has taken the place of market value. The price of production is regulated in every sphere, and regulated too according to its particular circumstances. But the price of production is itself in turn the centre around which daily market prices revolve and to which they are equalized over given periods.
(See Ricardo, on the determination of the price of production by those working under the worst conditions.)
Whatever governs prices, this follows:
(1) The law of value governs their movement: a decrease in the labour-time needed for production makes prices of production fall, and an increase makes them rise. It is in this sense that Ricardo says (he clearly feels that his prices of production diverge from the values of commodities) that "the inquiry to which he wishes to draw the reader's attention, relates to the effect of the variations in the relative value of commodities, and not in their absolute value."
(2) The average profit that determines prices of production must always be roughly equal to the quantity of surplus-value that falls to a given capital as its aliquot part — its proportional share — of the total social capital. Suppose the general rate of profit, and so the average profit, expressed in money, were higher than the real average surplus-value calculated in money. As far as the capitalists themselves are concerned, it makes no difference whether they credit each other with 10 per cent or 15 per cent profit. One percentage covers no more real commodity-value than the other, since the money expression is inflated on both sides equally.
But as for the workers — since it is assumed that they receive their normal wage, so that raising the average profit does not express a real deduction from wages, that is, something quite different from the capitalist's normal surplus-value — any rise in commodity prices produced by raising the average profit must be matched by a rise in the money expression of variable capital. In fact, such a general nominal rise in the rate of profit and the average profit, beyond the rate given by the proportion of the real surplus-value to the total capital advanced, is not possible without also bringing about a rise in wages, and equally a rise in the prices of the commodities that make up constant capital. The same holds in reverse for a fall.
Since the total value of the commodities regulates the total surplus-value, and the total surplus-value in turn regulates the level of the average profit and so of the general rate of profit — as the general law, or as what governs the fluctuations — the law of value regulates prices of production.
What competition achieves, first of all within a single sphere, is the formation of one common market value and market price out of the different individual values of the commodities. Competition among capitals across the different spheres, however, is what first brings about the price of production, which equalizes the rates of profit between the different spheres. The second of these needs a higher development of the capitalist mode of production than the first.
For commodities from the same sphere of production, of the same kind and roughly the same quality, to sell at their values, two things are needed:
(1) First, the different individual values must be equalized into one social value, the market value already described, and this requires competition among the producers of the same kind of commodity, as well as the existence of a market where they all offer their commodities together. For the market price of identical commodities — each produced under somewhat different individual conditions — to match the market value, neither rising above it nor falling below it, the pressure the different sellers put on one another has to be strong enough to throw onto the market the quantity of commodities that social need calls for, that is, the quantity for which society is able to pay the market value. If the mass of commodities produced exceeded this need, the commodities would have to be sold below their market value; conversely, they would sell above their market value if the mass of commodities were not large enough, or, what comes to the same thing, if the pressure of competition among the sellers were not strong enough to force them to bring this mass of commodities to market.
If the market value changed, the conditions under which the whole mass of commodities could be sold would change too. When the market value falls, social need — which here always means need that has money to back it up — expands on average and can, within certain limits, absorb larger masses of the commodity. When the market value rises, social need for the commodity contracts and smaller masses of it are absorbed.
So while demand and supply regulate the market price, or rather the deviations of market prices from the market value, the market value, for its part, regulates the relation between demand and supply, or the centre around which the fluctuations of demand and supply make market prices oscillate.
Looking at the matter more closely, one finds that the conditions holding for the value of the single commodity reproduce themselves here as conditions for the value of the whole sum of a kind of commodity. Capitalist production is from the outset mass production; and even other, less developed modes of production — at least for the main commodities — take what is produced in small quantities by many small individual producers and concentrate it, heap it up, and bring it to market in large masses in the hands of relatively few merchants, as the joint product of a whole branch of production, or of a larger or smaller portion of it.
It may be noted here, quite in passing, that "social need" — that is, what governs the principle of demand — is essentially conditioned by the relation of the different classes to one another and by their respective economic position: firstly by the proportion of total surplus-value to wages, and secondly by the proportion between the different parts into which surplus-value is split (profit, interest, ground-rent, taxes, and so on). So here too it becomes clear that absolutely nothing can be explained from the relation of demand and supply until the basis on which that relation plays out has been worked out.
Even though both commodity and money are unities of exchange-value and use-value, we already saw (in Volume 1) how, in buying and selling, the two split apart at the two ends of the exchange, so that the commodity (the seller) stands for use-value and the money (the buyer) stands for exchange-value. That the commodity should have use-value, that is, should satisfy a social need, was one precondition of the sale. The other was that the quantity of labour contained in the commodity should represent socially necessary labour, so that the individual value of the commodity — which, on this assumption, is the same as its selling price — coincides with its social value.
Let us apply this to the mass of commodities on the market that forms the product of a whole sphere.
The matter is shown most easily if we treat the whole mass of commodities — to begin with, of one branch of production — as a single commodity, and add together the sum of the prices of the many identical commodities into one price. What was said above for the single commodity now holds word for word for the mass of commodities of a given branch found on the market. That the individual value of the commodity should correspond to its social value is now realized, or further specified, as this: the total quantity contains the social labour necessary for its production, and the value of this mass equals its market value.
Now suppose the great bulk of these commodities is produced under roughly the same normal social conditions, so that this value is also the individual value of the particular commodities making up that bulk. Suppose further that a relatively small part is produced below these conditions, and another part above them, so that the individual value of the one part is greater, and that of the other smaller, than the middle value of the great bulk of commodities — but these two extremes balance each other out, so that the average value of the commodities belonging to them equals the value of the commodities belonging to the middle bulk. Then the market value is determined by the value of the commodities produced under the middle conditions.
The value of the whole mass of commodities equals the real sum of the values of all the individual commodities taken together — those produced within the middle conditions as much as those produced below or above them. In this case, the market value, or the social value of the mass of commodities, is the labour-time necessarily contained in them, determined by the value of the great middle bulk.
Suppose instead that the total quantity of the commodity in question brought to market stays the same, but the value of the commodities produced under worse conditions is no longer balanced out by the value of those produced under better conditions, so that the part of the mass produced under worse conditions forms a relatively significant quantity, both against the middle bulk and against the other extreme. Then it is the mass produced under worse conditions that governs the market value, or the social value.
Finally, suppose the mass of commodities produced under better than middle conditions considerably exceeds the mass produced under worse conditions, and itself forms a significant quantity against the mass produced under middle conditions. Then it is the part produced under the best conditions that governs the market value.
Here we leave aside the flooding of the market, where the part produced under the best conditions always governs the market price; what concerns us here is not the market price, insofar as it differs from the market value, but the different ways in which the market value itself is determined.
In fact, taken quite strictly (which of course in reality only ever occurs approximately, modified in a thousand ways), in case I the market value of the whole mass, governed by the average values, equals the sum of the individual values of all the commodities in it — although for the commodities produced at the extremes, this value presents itself as an average value forced upon them. Those producing at the worst extreme then have to sell their commodities below their individual value; those at the best extreme sell them above it.
In case II, the individual masses of value produced at the two extremes do not balance out; instead, the part produced under worse conditions tips the scale. Strictly speaking, the average price or market value of each individual commodity, or of each aliquot part — each proportional part — of the whole mass, would now be determined by the total value of the mass — arrived at by adding together the values of the commodities produced under the different conditions — and by the proportional part of this total value that falls to the single commodity. The market value obtained this way would stand above the individual value not only of the commodities belonging to the favourable extreme but also of those belonging to the middle layer; yet it would still stand lower than the individual value of the commodities produced at the unfavourable extreme. How far it approaches this unfavourable extreme, or finally coincides with it, depends entirely on how much room the mass of commodities produced at the unfavourable extreme takes up in the sphere of commodities in question. If demand only slightly predominates, the individual value of the unfavourably produced commodities governs the market price.
Finally, if, as in case III, the quantity of commodities produced at the favourable extreme takes up more room, not only compared with the other extreme but compared with the middle conditions, then the market value falls below the middle value. The average value, calculated by adding together the sums of value of both extremes and the middle, here stands below the value of the middle and moves closer to or further from it depending on the relative room the favourable extreme takes up. If demand is weak against supply, the favourably placed part, however large it may be, forcibly makes room for itself by drawing its price down to its individual value. The market value can never coincide with this individual value of the commodities produced under the best conditions, except where supply very strongly outweighs demand.
This determination of the market value, presented here in the abstract, is mediated on the actual market by competition among the buyers, provided that demand is just large enough to absorb the mass of commodities at the value thus determined. And here we come to the other point.
(2) That a commodity has a use-value means only that it satisfies some social need. As long as we were dealing with single commodities, we could simply assume the need for that particular commodity was there — its size already folded into the price — without asking how large that need actually was. But once the output of a whole branch of production stands on one side and the social need for it stands on the other, that size becomes essential. We now have to look at the size, the quantity, of this social need.
The account of market value given above assumed that the mass of commodities produced stays the same — a given quantity — and that only the mix of parts within that mass, produced under different conditions, shifts, which is what changes how the market value of that same mass gets regulated. Suppose this mass is the customary quantity supplied, setting aside for now that part of what is produced might be temporarily held off the market.
If the demand for this mass also remains the customary demand, the commodity sells at its market value, whichever of the three cases already examined regulates that value. The mass of commodities then satisfies not just a need, but satisfies it to its full social extent. But if the quantity is smaller or larger than the demand for it, the market price deviates from the market value. And the first such deviation is this: when the quantity is too small, the commodity produced under the worst conditions always regulates the market value, and when it is too large, the commodity produced under the best conditions always regulates it — so that one of the extremes determines the market value, even though the mere proportions of the masses produced under the different conditions would, taken alone, point to a different result. The bigger the gap between demand and the quantity produced, the further the market price deviates from the market value, upward or downward.
This gap between the quantity of commodities produced and the quantity at which they sell at their market value can arise from two different causes. Either the quantity itself changes, becoming too small or too large, so that reproduction has taken place on a different scale than the one that regulated the given market value — here supply has changed while demand stayed the same, and relative overproduction or underproduction has set in. Or reproduction, that is supply, stays the same, but demand has fallen or risen for various reasons — here the absolute size of supply is unchanged, but its size relative to, measured against, the need has changed; the effect is the same as in the first case, only in the opposite direction. Finally, when changes happen on both sides at once — in opposite directions, or in the same direction but not to the same degree, in short, when both sides shift together in a way that changes their earlier proportion — the end result always comes down to one of the two cases just considered.
The real difficulty in giving a general definition of demand and supply is that the definition seems to run in a circle. Let us start with supply: the product on the market, or what can be delivered to it. To avoid needless detail, let us take the mass of the annual output in each particular branch of industry, setting aside how much more or less easily different commodities can be held back from the market and stored for next year's consumption, say. This annual output first of all expresses a definite quantity — a measure or a number, depending on whether that kind of commodity is measured as a continuous amount or counted in discrete units. These are not just use-values that satisfy human needs; these use-values are available on the market on a given scale.
Second, this quantity of commodities has a definite market value, which can be expressed as a multiple of the market value of the single commodity, or of the unit used to measure it. So there is no necessary connection between the quantity of commodities on the market and their market value — some kinds of commodity are worth a great deal by the unit, others very little, so that a given sum of value can show up as a very large quantity of one commodity and a very small quantity of another. The only connection between the quantity of articles on the market and their market value is this: on a given basis of the productive power of labour, producing a definite quantity of articles in any particular sphere of production requires a definite quantity of social labour-time — though this relationship is quite different from one sphere of production to another, and has no inner connection at all with how useful these articles are, or with the particular nature of their use-values. All else being equal: if the quantity a of a kind of commodity costs b of labour-time, the quantity na costs nb of labour-time.
Further: so far as society wants an article produced to satisfy its needs, it has to pay for it. In fact, since commodity production presupposes a division of labour, society buys these articles by devoting part of its available labour-time to producing them — buys them, that is, with a definite quantity of the labour-time it has at its disposal. The part of society which, through the division of labour, falls to producing this particular article must receive an equivalent in the social labour embodied in the articles that satisfy its own needs.
But there is no necessary connection — only an accidental one — between the total quantity of social labour spent on a social article, that is, between the fraction of society's total labour-power, the share of overall production, that society devotes to this article, on the one hand, and, on the other hand, the extent to which society demands satisfaction of the need that this particular article answers. (Only where production stands under society's own real, predetermining control does society itself create the connection between the amount of social labour-time devoted to producing particular articles and the extent of the social need those articles are meant to satisfy.)
Although each single article, or each definite quantity of a kind of commodity, may contain only the social labour required for its production — so that, seen from that side, the market value of this whole kind of commodity represents only the labour socially necessary for it — still, if the commodity has actually been produced in a quantity exceeding the present social need, part of the social labour-time has been wasted, and the mass of commodities then represents on the market a much smaller quantity of social labour than is really contained in it. These commodities must therefore be sold below their market value, and part of them may even prove completely unsaleable. The reverse holds if the quantity of social labour devoted to producing a particular kind of commodity is too small for the extent of the particular social need the product is meant to satisfy.
But if the quantity of social labour devoted to producing a particular article matches the extent of the social need to be satisfied — so that the mass produced corresponds to the customary scale of reproduction, demand being unchanged — then the commodity sells at its market value. Exchanging or selling commodities at their value is the rational thing, the natural law of their equilibrium; it is from this law that the deviations are to be explained, not the law itself from the deviations.
Let us look now at the other side, demand.
Commodities are bought either as means of production or as means of subsistence — it makes no difference that some kinds of commodity can serve both purposes — in order to enter productive or individual consumption. So demand for them comes from the producers (here, capitalists, since we assume the means of production have been turned into capital) and from the consumers. Both seem, at first, to presuppose on the side of demand a given quantity of social needs, matched on the other side by definite quantities of social production in the various branches of production. If the cotton industry is to carry out its annual reproduction again on the same scale, it needs the customary quantity of cotton for that — and, given the annual expansion of reproduction that comes from the accumulation of capital, other things being equal, an additional quantity of cotton besides. The same holds for means of subsistence. The working class must find at least the same quantity of necessary means of subsistence available again — though perhaps distributed a little differently among the different kinds — if it is to go on living in its customary, average way; and, given the annual growth of population, an additional quantity besides; and so on, with more or less modification, for the other classes.
So it looks as though a definite social need of a certain size stands on the demand side, requiring a definite quantity of some article on the market to satisfy it. But how much of that need there is stretches and shifts constantly. Its fixity is mere appearance. If means of subsistence were cheaper, or money-wages higher, the workers would buy more of them, and a greater "social need" for these kinds of commodity would show itself — quite apart from the paupers and so on, whose "demand" stands even below the narrowest limits of their physical need. If, on the other hand, cotton were cheaper, the capitalists' demand for cotton would grow, and more additional capital would be thrown into the cotton industry, and so on.
It must not be forgotten here that the demand for productive consumption is, on our assumption, the demand of the capitalist, and that his real purpose is the production of surplus-value, so that he produces a given kind of commodity only for that purpose. This does not, on the other hand, prevent him, in so far as he stands on the market as a buyer of, say, cotton, from representing the need for cotton — just as it is a matter of indifference to the seller of cotton whether the buyer turns it into shirting or into gun-cotton, or means to stuff his own and the world's ears with it. All the same, this does have a great influence on the manner in which he is a buyer. His need for cotton is essentially modified by the fact that it really only disguises his need to make a profit.
The limits within which the need for commodities represented on the market — demand — differs in quantity from the real social need are naturally very different from one commodity to another; I mean the difference between the quantity of a commodity that is demanded and the quantity that would be demanded at different money-prices for it, or with the buyers in different money and living circumstances.
Nothing is easier than to see the disproportions between demand and supply, and the resulting deviation of market prices from market values. The real difficulty lies in working out what is meant by demand and supply coinciding.
Demand and supply coincide, it is said, when they stand in such a relationship that the mass of commodities from a particular branch of production can be sold at its market value — neither above it nor below it. That is the first thing we are told.
And the second thing we are told: when the commodities can be sold at their market value, demand and supply coincide.
When demand and supply coincide, they stop acting, and that is exactly why the commodity sells at its market value. When two forces act equally and in opposite directions, they cancel each other out — they have no effect at all on the outside — and anything that happens while they are cancelled has to be explained by something other than those two forces acting on it. So when demand and supply cancel each other out, they stop explaining anything at all: they have no effect on the market value, and they leave us completely in the dark as to why the market value comes out to exactly this sum of money and no other.
The real inner laws of capitalist production plainly cannot be explained by the interplay of demand and supply (setting aside a deeper analysis of these two social driving forces, which has no place here), because those laws only show up in their pure form once demand and supply stop acting — that is, once they coincide. In fact demand and supply never actually coincide; and on the rare occasion when they do, it is accidental — so, scientifically, it should be set at zero, treated as if it had not happened.
Yet political economy assumes that they do coincide. Why? First, in order to look at these phenomena in the lawlike shape that corresponds to their concept — that is, to look at them apart from the appearance produced by the movement of demand and supply. Second, in order to find the real tendency of that movement, to pin it down, so to speak. For the disproportions run in opposite directions, and because they keep following one another, they cancel each other out through those opposite directions, through their contradiction with one another.
So even though demand and supply never coincide in any single given case — a deviation in one direction is what produces another deviation in the opposite direction — their inequalities still follow one another in such a way that, once a longer or shorter stretch of time is taken as a whole, supply and demand constantly coincide. But only as the average of the movement that has already happened, and only as the constant movement of their contradiction. In this way market prices, which deviate from market values, balance out to market values when taken as an average figure — the deviations from that value cancelling out as pluses and minuses. And this average figure is by no means only of theoretical importance; it matters practically for capital, whose investment is calculated on the fluctuations and the way they even out over a more or less definite period of time.
The relationship between demand and supply, then, only explains two things: on one side, why market prices drift away from market values, and on the other, the tendency for that drift to be removed — that is, for the effect of the demand-supply relationship to cancel itself out. (Commodities that have a price without having a value are a separate case, not considered here.)
Demand and supply can cancel that effect in very different ways. Say demand falls, so the market price falls too — that can lead capital to withdraw, shrinking the supply. Or it can lead to the market value itself falling, through inventions that shorten the labour-time required, bringing it back into line with the market price.
The reverse also happens: demand rises, and with it the market price rises above the market value. That can draw too much capital into this branch of production, pushing output up until the market price itself falls below the market value; or, on the other hand, it can drive prices up so far that demand itself is pushed back down. In some branches of production it can even happen that the market value itself rises, for shorter or longer stretches, because part of what is demanded now has to be produced under worse conditions during that time.
If demand and supply determine the market price, then, the other way round, the market price — and, looking deeper, the market value — determines demand and supply. This is obvious enough for demand: it moves opposite to the price, rising when the price falls and falling when it rises.
But it holds for supply too. The prices of the means of production that go into a supplied commodity determine the demand for those means of production, and so also the supply of the commodity itself, since supplying that commodity means demanding those means of production. Cotton prices determine the supply of cotton goods.
On top of this confusion — prices determined by demand and supply, and, at the same time, demand and supply determined by prices — comes another: that demand determines supply and supply determines demand, that production determines the market and the market determines production.
Even the ordinary economist can see that, without any change in supply or need brought about by outside circumstances, the relationship between the two can still shift purely because the market value of the commodity has changed. Even he has to admit that, whatever the market value turns out to be, demand and supply must balance out to produce it. In other words, the relationship between demand and supply does not explain the market value — on the contrary, the market value explains the fluctuations in demand and supply. The author of the Observations continues, after the passage cited in the note:
The author of the Observations puts it like this: this proportion between demand and supply — if 'demand' and 'natural price' still mean what they meant when we were discussing Adam Smith — must always be a proportion of equality. That is because the natural price is only actually paid when the supply matches the effectual demand: the demand of buyers who will pay neither more nor less than the natural price. So there can perfectly well be two very different natural prices for the same commodity at different times, and yet the proportion which the supply bears to the demand can be exactly the same both times — the proportion of equality.
So it is conceded: with two different 'natural prices' for the same commodity at different times, demand and supply can and must match up on both occasions, if the commodity is to sell at its 'natural price' each time. Since there is no difference at all in the relationship between demand and supply from the one case to the other, but there is a difference in the size of the 'natural price' itself, then that 'natural price' is clearly fixed independently of demand and supply — and is the last thing they could be said to fix.
For a commodity to sell at its market value — that is, in proportion to the socially necessary labour it contains — the total amount of social labour spent on the whole mass of that kind of commodity has to match the quantity of social need for it: the social need with money to back it up.
Competition — through the fluctuations in market prices that track the fluctuations in the relationship of demand and supply — constantly works to bring the total labour spent on each kind of commodity down to this measure.
In the relationship between demand and supply for commodities, two earlier relationships turn up again. First, the relationship of use-value and exchange-value, of commodity and money, of buyer and seller. Second, the relationship of producer and consumer, even though both may be represented by merchants acting as third parties.
When we looked at buyer and seller on their own, it was enough to set them against each other one by one to work out the relationship. Three people are enough for the whole metamorphosis of a commodity — the whole of buying and selling. A turns his commodity into B's money by selling it to B, and then turns that money back into a commodity by buying from C; the whole process runs between these three.
There is more. When we looked at money, we assumed commodities sell at their value, because there was no reason then to consider prices that departed from value — the question was only the change of form a commodity goes through in becoming money and turning back from money into a commodity. As soon as a commodity is sold at all, and something new bought with the proceeds, the whole metamorphosis is complete before us, and for that metamorphosis, taken on its own, it makes no difference whether the commodity's price stood below or above its value. The commodity's value still matters as the foundation, though, because any rational understanding of money has to start from that foundation, and price, in its general concept, is in the first place just value in the form of money.
That said, when we looked at money as a means of circulation, we were assuming that more than one commodity's metamorphosis was going on — we were looking at how these metamorphoses are socially interwoven. Only that way do we get to the circulation of money and the development of its role as a means of circulation. But important as that connection is for money's move into that role and for the changed shape it takes on because of it, it makes no difference at all to the transaction between the individual buyer and seller.
With supply and demand, on the other hand, supply equals the sum of the sellers or producers of a given kind of commodity, and demand equals the sum of the buyers or consumers — individual or productive — of that same kind of commodity.
These sums act on each other as units, as aggregate forces. Here the individual acts only as part of a social power, as an atom in the mass — and it is in this form that competition asserts the social character of production and consumption.
The side of competition that happens to be weaker at a given moment is also the side where each individual acts independently of the mass of his rivals, and often directly against them — and by doing exactly that, he makes everyone feel how dependent they are on each other. The stronger side, by contrast, always faces its opponent as something closer to a single closed block.
If demand for a particular kind of commodity is greater than supply, then, within certain limits, one buyer outbids another, making the commodity dearer for everyone, above its market value, while on the other side the sellers all seek to sell at a high market price. If, the other way round, supply is greater than demand, one seller starts undercutting, and the others have to follow, while the buyers all work to push the market price as far below the market value as they can.
Belonging to the common side interests anyone only for as long as he gains more with it than he would against it. And that common interest ends the moment that side becomes the weaker one — at which point each person starts trying, on his own account, to wriggle out as well as he can.
Further: if a producer makes something more cheaply and can sell more of it, capturing a bigger share of the market by selling below the going market price or market value, he does exactly that. And this is how the process begins that gradually forces everyone else to introduce the cheaper form of production too, bringing the socially necessary labour down to a new, lower level.
When one side has the upper hand, everyone on it gains, as if they all shared a single monopoly. When a side is the weaker one, each person can try, for himself, to become the stronger party instead — say, by working with lower costs of production — or at least to come out of it as well as possible; and here nobody cares in the least about his neighbour, even though what he does touches not just himself but all his fellows too.
Demand and supply presuppose that value has already been transformed into market value. And so far as they operate on a capitalist basis — so far as the commodities are products of capital — they presuppose capitalist processes of production: relationships far more tangled than the mere buying and selling of commodities. What is at stake with them is not the formal turning of a commodity's value into a price, a mere change of form; it is the definite, quantitative ways market prices depart from market values, and, further, from prices of production.
For simple buying and selling, it is enough to have commodity producers facing each other as such. Demand and supply, looked at more closely, presuppose the existence of the different classes and sections of classes who divide the total revenue of society among themselves and consume it among themselves as revenue — and so form the demand that revenue creates. And to understand the demand and supply formed among the producers themselves, you need, on the other hand, insight into the whole shape of the capitalist process of production.
Capitalist production is not just about throwing commodities into circulation and pulling back an equal mass of value in another form — money, or some other commodity. It is about pulling out, for the capital advanced to production, the same surplus-value or profit that any other capital of the same size would get, whatever branch it happens to be invested in — or, put the other way, a profit in proportion to its size. So it comes down to this: commodities have to sell, at a minimum, at prices that yield the average profit — that is, at prices of production.
It is in this form — the demand that equal capital bring equal profit in any branch — that capital comes to know itself as one social power, and each capitalist holds a share of that power the size of his share of the total social capital.
First: capitalist production, in itself, does not care what particular use-value it produces — it does not care about the specific character of the commodity at all. In any branch, all it wants is to produce surplus-value, to seize a certain quantity of unpaid labour out of the product of labour.
And it is just as much in the nature of wage-labour under capital that it has to be indifferent to the specific character of its own work — it has to remake itself to fit capital's needs, and let itself be thrown from one branch of production into another.
Second: in fact, one branch of production is just as good — and just as bad — as another — each yields the same profit. But each would be pointless if the commodity it produces did not satisfy some kind of social need.
If commodities sold at their values, then, as we have seen, very different rates of profit would arise in the different branches, depending on the differing organic composition of the capital invested in each. But capital pulls out of a branch with a low rate of profit and throws itself into one that yields a higher one. Through this constant migration of capital — through capital distributing itself between the branches, leaving where the rate is falling and entering where it is rising — it so alters the relation of supply to demand that the average profit becomes the same across the different branches of production, and so values turn into prices of production.
This equalization out succeeds for capital more or less completely, depending on how far capitalist production has developed in a given country — that is, how far conditions there have been adapted to the capitalist mode of production. As capitalist production advances, the conditions it needs advance with it, and it brings the whole social setting in which production takes place under its own specific character and under the laws immanent in it.
This constant equalization of inequalities that are constantly renewed happens faster, (1) the more mobile capital is — the easier it is to move from one branch, and from one place, to another — and (2) the faster labour-power can be thrown from one branch into another, and from one local point of production to another.
The first condition assumes complete freedom of trade within society, and the removal of every monopoly except the natural ones — the ones that spring from the capitalist mode of production itself. It also assumes a developed credit system, which draws the otherwise scattered mass of available social capital together into one lump confronting the individual capitalist; and, finally, that the different branches of production are run by capitalists. That last point is already built into the assumption, once we are talking about values turning into prices of production across all the branches capitalism exploits — but the equalization itself meets bigger obstacles where large numbers of branches not run on capitalist lines (small farmers working the land, say) wedge themselves in among capitalist enterprises and get bound up with them. And, finally, a densely settled population.
The second condition assumes the abolition of every law that stops workers moving from one branch of production to another, or from one place where production is carried on to any other. It assumes that the worker does not care what particular kind of work he does. It assumes labour reduced, as far as possible, to simple labour in every branch. It assumes the disappearance of every prejudice of trade and craft among the workers. And, finally and above all, it assumes the worker's subjection to the capitalist mode of production. Carrying this further belongs to a special study of competition.
From all this it follows that every individual capitalist, like all the capitalists of any particular branch together, has a stake in the exploitation of the whole working class by the total capital, and in how far that exploitation goes. And it is not just general class sympathy — it is a direct economic interest. Because, taking everything else as given, including the value of the total constant capital advanced, the average rate of profit depends on how far the total capital exploits the total labour.
The average profit coincides with the average surplus-value that capital produces per 100 — and as far as surplus-value goes, that is obvious from what has already been said. With average profit, the only thing added is the value of the capital advanced, as one of the things that fixes the rate of profit. In fact, the particular interest a capitalist, or the capital of one particular branch, takes in exploiting the workers he directly employs is limited to this: he can make an extra cut, a profit reaching above the average, either through exceptional overwork, or by pushing wages below the average, or through exceptional productivity in the labour applied.
Apart from that, a capitalist who employed no variable capital at all in his branch, and so no workers whatsoever (in fact an exaggerated assumption), would have just as much of a stake in the exploitation of the working class by capital, and would draw his profit just as much from unpaid surplus labour, as a capitalist who — again an exaggerated assumption — employed only variable capital, laying out his whole capital in wages.
But the degree to which labour is exploited depends, given a fixed working day, on the average intensity of the work, and given a fixed intensity, on the length of the working day. The rate of surplus-value depends on that degree of exploitation — and so, given a fixed total variable capital, does the size of the surplus-value, and with it the size of the profit. The special interest that the capital of one branch has, as against the total capital, in exploiting the workers it specifically employs, is the same interest an individual capitalist has, as against his own branch, in exploiting the workers he personally exploits.
On the other hand, every particular branch of capital, and every individual capitalist, shares one and the same interest in the productivity of the social labour that the total capital sets to work. Two things depend on that:
First: the mass of use-values in which the average profit is expressed — and this matters twice over, since that profit serves both as a fund for accumulating new capital and as a fund of revenue to be consumed. Second: the value level of the whole capital advanced, constant and variable together, which, given a fixed amount of surplus-value or profit for the capitalist class as a whole, fixes the rate of profit, or the profit on a given quantity of capital.
The particular productivity of labour in one branch, or in one particular firm within that branch, interests only the capitalists directly involved — and only insofar as it puts that branch in a position to make an extra profit as against the total capital, or that capitalist in a position to make one as against his own branch.
So here we have the mathematically exact proof of why the capitalists, however much they show themselves to be false brothers in their competition with one another, still form a true freemasonry against the working class as a whole.
The price of production includes the average profit. We gave it that name, price of production, but it is in fact the very same thing that Adam Smith calls the natural price, that Ricardo calls the price of production and the cost of production, and that the Physiocrats call the prix nécessaire, the necessary price — though none of them worked out how the price of production differs from value. It gets that name because, in the long run, it is the condition on which the supply, the reproduction, of a commodity in any particular branch of production depends.
This also explains why the very economists who will not accept that the value of commodities is determined by labour-time, by the quantity of labour contained in them, nonetheless keep talking about prices of production as the centres that market prices fluctuate around. They can allow themselves this because the price of production is already a form of commodity-value that has been thoroughly externalized — on the face of it an irrational form, one carrying no trace of its own concept, the form value takes as it appears in competition, that is, in the mind of the vulgar capitalist — the capitalist as competition teaches him to think — and so in that of the vulgar economists too.
We saw that market value — and everything said about it here applies too, with the necessary adjustments, to the price of production — includes a surplus profit for whoever produces under the best conditions in any given branch. Leaving aside cases of crisis and general overproduction, this holds for every market price, however far it departs from market values or market prices of production. What is built into the market price is this: the same price gets paid for commodities of the same kind, even though they may have been produced under very different individual conditions, and so carry very different cost prices. (We are not talking here about surplus profits that come from monopolies in the ordinary sense, whether artificial or natural.)
But a surplus profit can also arise in another way: if certain branches of production are in a position to escape having their commodity-values turned into prices of production, and so escape having their profits reduced to the average.
In the Part on ground-rent we will have to look at how these two forms of surplus profit develop further.