With wages and the working day both given, a variable capital — take 100 — stands for a fixed number of workers; it is an index of how many. Say £100 is the wages of 100 workers, for a week. If these 100 workers perform as much necessary labour as surplus labour — that is, they spend as much time each day working for themselves, reproducing their own wages, as they spend working for the capitalist, producing surplus-value — then the total value they produce would be £200, and the surplus-value they create would be £100. The rate of surplus-value, s/v, would be 100 per cent. But as we have seen, this rate of surplus-value would show up as widely different rates of profit, depending on how large the constant capital c is, and with it the total capital C, since the rate of profit is s/C. With the rate of surplus-value at 100 per cent:
If c = 50 and v = 100, then p' = 100/150 = 66⅔ per cent.
If c = 100 and v = 100, then p' = 100/200 = 50 per cent.
If c = 200 and v = 100, then p' = 100/300 = 33⅓ per cent.
If c = 300 and v = 100, then p' = 100/400 = 25 per cent.
If c = 400 and v = 100, then p' = 100/500 = 20 per cent.
So, with the degree of exploitation unchanged, one and the same rate of surplus-value would come out as a falling rate of profit — because the value of the constant capital grows along with its material bulk, and with it the value of the total capital, though not in the same proportion.
Suppose next that capital's composition doesn't shift like this only in a few isolated spheres of production, but more or less across all of them, or at least across the decisive spheres — so that the shift amounts to a change in the average organic composition of society's total capital. Then this steady growth of constant capital relative to variable capital must necessarily produce a gradual fall in the general rate of profit, so long as the rate of surplus-value — the degree to which capital exploits labour — stays the same. And capitalist production has already been shown to obey a law here: as it develops, variable capital declines relative to constant, and so relative to the total capital set in motion. That means only this: the same number of workers, the same quantity of labour-power placed at capital's disposal by a variable capital of a given value, sets in motion, works up and productively consumes in the same time a constantly growing mass of machinery, fixed capital of every kind, means of labour, and raw and auxiliary materials — this being the work of the distinctive methods of production that grow up inside capitalist production. Which is to say that the same workers operate with a constant capital whose value keeps growing.
This progressive decline of variable capital relative to constant capital, and so relative to the total capital as well, is the same thing as a progressively higher organic composition of the social capital on average. It is equally just another way of stating the progressive development of the social productivity of labour — which shows itself precisely in this, that through the growing use of machinery and fixed capital more raw and auxiliary materials are turned into products by the same number of workers in the same time, that is, with less labour. To this growing value of the constant capital there corresponds a product that grows steadily cheaper — although that value only remotely reflects the growth in the real mass of use-values of which the constant capital consists in material terms. Each individual product, taken on its own, contains a smaller sum of labour than it did at lower stages of production, where the capital laid out on labour stood in a far greater ratio to the capital laid out on means of production. So the series we set up hypothetically at the start does express the real tendency of capitalist production. That real tendency, as the variable capital keeps declining relative to the constant, produces an ever higher organic composition of the total capital. The immediate result is that the rate of surplus-value — with the degree of exploitation of labour staying the same or even rising — comes out as a steadily falling general rate of profit. (We shall show later why the fall does not appear in this absolute form, but rather more as a tendency towards a progressive fall.) This progressive tendency of the general rate of profit to fall is thus nothing but the way the capitalist mode of production, and it alone, expresses the steady advance of the social productivity of labour.
This is not to say that the rate of profit may not also fall temporarily for other reasons. But it proves that the nature of the capitalist mode of production itself makes it a self-evident necessity that, as that mode advances, the general average rate of surplus-value must express itself in a falling general rate of profit. Since the mass of living labour applied steadily declines relative to the mass of objectified labour it sets in motion — the means of production it productively consumes — the unpaid part of that living labour, the part that takes shape as surplus-value, must likewise shrink, step by step, in proportion to the value of the whole capital applied. But this ratio between the mass of surplus-value and the value of the total capital applied is what makes up the rate of profit. So the rate of profit must steadily fall.
Simple as this law looks after the arguments so far, no earlier economics managed to discover it, as we shall see later on. Those writers saw the phenomenon, and tormented themselves with one contradictory attempt after another to explain it. Given how much the law matters for capitalist production, one can say it forms the mystery whose solution all of political economy has circled since Adam Smith, and that what separates the various schools since Adam Smith is the different attempts each makes to solve it. But consider the other side: earlier political economy fumbled around the distinction between constant and variable capital without ever managing to state it precisely; it never presented surplus-value as separate from profit, and never presented profit in general in its pure form, apart from the various components that have made themselves independent of one another — industrial profit, commercial profit, interest, ground-rent; and it never thoroughly analysed the differences in the organic composition of capital, and so never the formation of the general rate of profit either. Once you take all that into account, it stops being a puzzle that it never found the answer to this riddle.
We deliberately set out this law before showing how profit falls apart into different categories that have become independent of one another. Presenting it in this order — before profit is split into the various slices that go to various groups of people — proves from the outset that the law, taken in its generality, does not depend on that split, nor on the relations between the categories of profit that grow out of it. The profit we are speaking of here is only another name for surplus-value itself, shown now in relation to the total capital instead of in relation to the variable capital it springs from. The fall in the rate of profit therefore expresses the falling proportion of surplus-value itself to the total capital advanced, and is thus independent of however that surplus-value may be distributed among different categories.
We have seen that when capitalist development has reached a stage at which the composition of capital c:v is 50:100, a rate of surplus-value of 100 per cent comes out as a rate of profit of 66⅔ per cent, and that at a higher stage, where c:v is 400:100, the same rate of surplus-value comes out as a rate of profit of only 20 per cent. What holds for different stages of development that follow one another in a single country also holds for different stages of development that exist side by side, at the same time, in different countries. In the undeveloped country, where the first of those compositions of capital is the average, the general rate of profit would be 66⅔ per cent; in the country at the second, far higher stage of development, it would be 20 per cent.
The difference between the two national rates of profit could disappear, and even be reversed, if labour in the less developed country were less productive — so that more labour went into making less of the same commodity — more exchange-value carried by less use-value. The worker would then have to spend a bigger share of his time reproducing his own means of subsistence, or their value, and a smaller share producing surplus-value; he would supply less surplus labour, so that the rate of surplus-value would be lower. Suppose, for example, that in the less advanced country the worker worked two-thirds of the day for himself and one-third for the capitalist; then, on the assumptions of the example above, the same amount of labour-power would be paid 133⅓ and would yield a surplus of only 66⅔. A constant capital of 50 would correspond to that variable capital of 133⅓. The rate of surplus-value would now come to 133⅓ : 66⅔ = 50 per cent, and the rate of profit to 183⅓ : 66⅔, or roughly 36½ per cent.
We have not yet examined the different components into which profit divides, so for us they do not exist yet. The following is therefore said in advance, only to head off misunderstandings. When countries at different levels of development are compared — especially countries of developed capitalist production with those where labour is not yet formally subsumed under capital — not yet run as a capitalist's own operation — although the worker is in reality exploited by the capitalist (in India, for example, where the ryot, the peasant farmer, works his land as an independent producer, so that his production as such is not yet subsumed under capital, even though the usurer may, in the form of interest, skim off not only his whole surplus labour but — put in capitalist terms — part of his wages as well) — then it would be very wrong to try to measure a country's rate of profit by the level of its rate of interest. Interest of that kind includes the whole profit and more than the profit, whereas in countries of developed capitalist production it expresses only a proportional share of the surplus-value or profit produced. And on the other side, the rate of interest there is predominantly determined by circumstances (the usurers' advances to the big landowners, who draw the ground-rent) which have nothing at all to do with profit, and only show the proportion in which usury appropriates the ground-rent.
In countries at different levels of capitalist development, and therefore with different organic compositions of capital, the rate of surplus-value — one of the factors that determines the rate of profit — can be higher in the country where the normal working day is shorter than in the one where it is longer. First: if England's ten-hour working day, because of its greater intensity, is equal to Austria's fourteen-hour day, then with the working day divided the same way in both, five hours of surplus labour in England can represent a higher value on the world market than seven hours in Austria. Second: in England a larger part of the working day can form surplus labour than in Austria.
The law of the falling rate of profit — the law that expresses the same, or even a rising, rate of surplus-value — says in other words: take any given quantity of average social capital, say a capital of 100. An ever larger part of it is represented by means of labour, and an ever smaller part by living labour. Since the whole mass of living labour added to the means of production keeps falling in proportion to what those means are worth, the unpaid labour too, and the portion of value in which it is represented, falls in relation to the value of the total capital advanced. Or: a smaller and smaller proportional share of the whole capital laid out actually turns into living labour, and so the total capital, in proportion to its size, absorbs less and less surplus labour — even though the ratio of the unpaid part of the labour applied to the paid part may at the same time be growing. The relative decline of variable capital and the relative growth of constant capital, even though both parts grow in absolute terms, is, as we have said, only another expression for the increased productivity of labour.
Say a capital of 100 is made up of 80c + 20v, and that 20v stands for 20 workers. Suppose the rate of surplus-value is 100 per cent: the workers spend half the day working for themselves and half for the capitalist. In a less developed country the capital might be 20c + 80v, and that 80v would stand for 80 workers. But these workers keep two-thirds of the working day for themselves and work for the capitalist for only a third of it.
With everything else taken as equal, the first set of workers turns out a value of 40, the second a value of 120. So the first capital produces 80c + 20v + 20s = 120, a rate of profit of 20 per cent; the second produces 20c + 80v + 40s = 140, a rate of profit of 40 per cent. The rate of profit in the second case is thus twice as large as in the first, even though the rate of surplus-value in the first case is 100 per cent — twice as high as in the second, where it is only 50 per cent. In exchange for that, however, a capital of equal size appropriates the surplus labour of only 20 workers in the first case, and of 80 workers in the second.
The law that the rate of profit falls progressively — that is, that the surplus labour appropriated declines relative to the mass of objectified labour (labour already embodied in the means of production) that living labour sets in motion — in no way rules out that the social capital sets in motion and exploits a growing absolute mass of labour, and so appropriates a growing absolute mass of surplus labour. Nor does it rule out that the capitals under the command of individual capitalists command a growing mass of labour, and so of surplus labour — the latter even when the number of workers they command does not grow.
Suppose the working population is given, two million say. Suppose too that the average working day has a given length and intensity, and that wages are given, and with them the relation between necessary and surplus labour. Then the total labour of these two million — and their surplus labour too, as expressed in surplus-value — always produces the same magnitude of value. But the mass of constant capital — fixed and circulating — that this labour sets in motion grows, and the capital's value grows with its mass, though not in the same proportion. So the ratio of that value to the value of the constant capital falls; and along with that ratio the rate of profit falls, even though the same mass of living labour is commanded as before and the same mass of surplus labour is absorbed by capital.
The ratio changes not because the mass of living labour falls, but because the mass of objectified labour it sets in motion rises. The decrease is a relative one, not an absolute one; in fact it has nothing at all to do with how large the labour and the surplus labour set in motion are in absolute terms. What makes the rate of profit fall is not that the variable part of the total capital shrinks in absolute terms. It is only that this part shrinks relative to the constant part.
What was just true of a fixed quantity of labour and surplus labour also holds when the number of workers grows, and so — under the assumptions just given — when the mass of labour under capital's command in general grows, and in particular its unpaid part, surplus labour. Suppose the working population grows from two million to three million, and suppose the variable capital laid out on labour-power also goes from two million to three million, while constant capital rises from 4 million to 15 million. Then, under the given assumptions — a constant working day and a constant rate of surplus-value — the mass of surplus labour, and so of surplus-value, grows by half, by 50 per cent, from 2 million to 3 million. All the same, despite this 50 per cent growth in the absolute mass of surplus labour and so of surplus-value, the ratio of variable to constant capital would fall from 2 : 4 to 3 : 15, and the ratio of surplus-value to total capital would come out as follows (in millions):
While the mass of surplus-value has risen by half, the rate of profit has halved. But profit is nothing but surplus-value measured against the social capital, and so the mass of profit — its absolute size — is, looked at socially, equal to the absolute size of surplus-value. The absolute size of profit, its total mass, would therefore have grown by 50 per cent, despite the enormous fall in the ratio of that mass of profit to the total capital advanced — that is, despite the enormous fall in the general rate of profit.
So capital can employ more workers, set a larger absolute quantity of labour moving, absorb a larger absolute quantity of surplus labour, produce a larger mass of surplus-value and so a larger absolute mass of profit — and this can keep growing, and keep growing progressively, even while the rate of profit falls progressively. And this is not just something that can happen. Apart from passing fluctuations, on the basis of capitalist production it must happen.
The capitalist production process is essentially also a process of accumulation. We have seen how, as capitalist production advances, the amount of value that has simply to be reproduced and kept going grows along with the productivity of labour, even if the labour-power employed stays the same. But as the social productivity of labour develops, the quantity of use-values produced grows even more, and the means of production are one part of that. And the extra labour that has to be appropriated for this extra wealth to be turned back into capital depends not on the value but on the mass of these means of production (means of subsistence included), because when the worker actually works, what matters to him is not what the means of production are worth but what they can do — their use-value.
Accumulation itself — and the concentration of capital that comes with it — is itself a material means of raising the productivity of labour. But this growth of the means of production includes the growth of the working population: a population comes into being that matches the surplus capital and even, by and large, constantly overflows what it needs, and so there is an over-population of workers.
Suppose that surplus capital momentarily outstrips the working population under its command. That would work in two directions. On the one side, higher wages would gradually add to the number of workers, softening the destructive forces that cut down their children and making it easier to marry. On the other side, the very methods that create relative surplus-value — new machinery and better machinery — would far more quickly produce an artificial, relative over-population, and this in turn is the hotbed of a really rapid growth of population, because in capitalist production misery is what produces population.
So it follows of itself from what capitalist accumulation essentially is — and accumulation is only one aspect of the capitalist production process — that the greater mass of means of production destined to be turned into capital always finds at hand a correspondingly greater, indeed excessive, working population, ready to be exploited. So as production and accumulation move forward, the quantity of surplus labour that can be appropriated — and that is appropriated — must rise, and with it the absolute mass of profit appropriated by the social capital. But those very laws of production and accumulation drive up the value of constant capital together with its mass, and at an ever faster rate, compared with the value of the variable part that is exchanged for living labour. So it is these same laws that give the social capital a mass of profit that grows larger in absolute terms, and a rate of profit that falls.
Here we leave aside entirely the fact that the same amount of value comes to stand for more and more use-values and satisfactions, as capitalist production advances and with it the productivity of social labour develops and the branches of production — and so the products — multiply.
As capitalist production and accumulation develop, labour processes must be carried out on an ever larger scale, and so each individual establishment needs ever larger dimensions and ever larger advances of capital. The concentration of capitals therefore grows — and along with it, though less markedly, the number of capitalists. This growing concentration is both one of the material preconditions of that development and one of the results it itself throws off. And side by side with it, each reacting on the other, goes the progressive expropriation of the more or less immediate producers.
So it stands to reason for the individual capitalists that they command ever bigger armies of workers (however much their variable capital falls relative to constant capital), and that the mass of surplus-value they appropriate, and so of profit, grows — together with the fall in the rate of profit, and in spite of it. The same reasons that concentrate masses of worker armies under the command of individual capitalists are precisely the ones that also swell the mass of fixed capital applied, like that of raw and ancillary materials, in growing proportion as against how much living labour is applied.
One more thing needs mentioning here. When the working population is given, and the rate of surplus-value rises — whether by lengthening or intensifying the working day, or by a fall in the value of wages brought about by the growth in the productivity of labour — then the mass of surplus-value, and with it profit in absolute terms, must grow, despite the relative shrinking of variable capital against constant.
One and the same development of the productivity of social labour, one and the same set of laws, shows itself in the relative fall of variable capital against total capital and in the accumulation this speeds up — while, the other way round, accumulation reacts back and becomes the starting point for productivity to grow further and for variable capital to decline further in relative terms. And this same development is expressed, temporary fluctuations apart, in an ever-rising total of labour-power employed and an ever-growing absolute mass of surplus-value, and so of profit.
In what form, then, must this double-edged law present itself — the law of a falling rate of profit and, at the same time, a rising absolute mass of profit, both springing from the same reasons? A law resting on the fact that, given these conditions, more surplus labour — and so more surplus-value — is appropriated, and that, whether we look at the total capital or at a single capital as a mere piece of the total capital, profit and surplus-value are one and the same magnitude?
Take a fixed fractional share of the capital as the base for reckoning the rate of profit — an aliquot part, say 100. This 100 has the average composition of the total capital, say 80c + 20v. In Part Two of this volume we saw that what fixes the average rate of profit for the different branches of production is not the composition each branch has on its own, but the average social composition.
As the variable part declines relative to the constant part, and so relative to the total capital of 100, the rate of profit falls even when the degree of exploitation of labour stays the same, or even rises. And the relative size of surplus-value falls — that is, its ratio to the value of the total capital of 100 that has been advanced. But it is not just this relative size that drops. The quantity of surplus-value — and so of profit — that the total capital of 100 takes in also drops, and drops absolutely. Take a rate of surplus-value of 100 per cent. Then 60c + 40v turns out a mass of surplus-value, and so of profit, of 40; 70c + 30v turns out 30; 80c + 20v, only 20.
This fall is in the mass of surplus-value, and so of profit. It follows because a total capital of 100 sets less living labour moving altogether, and so, at the same degree of exploitation, also sets less surplus labour moving and produces less surplus-value. Take any fixed fractional share of the social capital — that is, of capital of average social composition — as the unit we measure surplus-value by, and that is what happens in every calculation of profit. On that measure, a relative fall of surplus-value and its absolute fall are simply identical. In the cases above, the rate of profit sinks from 40 per cent to 30 per cent and then 20 per cent, because one and the same capital produces less and less surplus-value — and so less profit — in absolute terms: 40, then 30, then 20. Since the capital against which surplus-value is measured has a fixed value of 100, a drop in surplus-value's proportion to that fixed size can only be another way of saying that surplus-value and profit have shrunk in absolute size. This really is a tautology. But that this decline happens follows, as has been shown, from the nature of the development of the capitalist production process.
But on the other side, the very same reasons that make surplus-value — and so profit — shrink in absolute terms on any given capital, and with it the rate of profit as worked out in percentages, also raise the absolute mass of surplus-value, and so of profit, that the social capital appropriates (that is, all capitalists taken together).
How is this to show itself? How can it show itself at all — or what conditions are contained in this apparent contradiction?
If any aliquot part of the social capital — any equal, measured-out share, say 100, and so any 100 of capital of average social composition — is a fixed magnitude, so that for it a fall in the rate of profit is one and the same thing as a fall in the absolute amount of profit, precisely because the capital it is measured against is a constant magnitude, then the total social capital is a variable magnitude — and so is the capital held by individual capitalists. It has to move in inverse proportion to the shrinking of its variable part, if the assumptions we have made are to hold.
In the earlier example the composition worked out at 60c + 40v; the surplus-value or profit came to 40, and the rate of profit was therefore 40 per cent. Suppose the total capital stood at 1,000,000 at this stage. Then total surplus-value, and so total profit, was 400,000. If the composition later becomes 80c + 20v, then — with the degree of exploitation of labour unchanged — each 100 yields 20 in surplus-value or profit. But since the mass of surplus-value or profit grows, as we showed, despite this falling rate of profit (that is, despite each capital of 100 producing less surplus-value) — say it grows from 400,000 to 440,000 — this only works because the total capital formed together with this new composition has grown to 2,200,000. The total capital now set in motion has grown to 220 per cent of what it was then, while the rate of profit has dropped by 50 per cent.
Had the capital simply doubled, it would at 20 per cent have thrown off exactly the surplus-value and profit that the old 1,000,000 threw off at 40 per cent. If it had grown to less than double, the surplus-value and profit it produced would have been smaller than what the earlier capital of 1,000,000 produced — even though, at its earlier composition, that capital needed to grow only from 1,000,000 to 1,100,000 to lift its surplus-value from 400,000 to 440,000.
Here the law already developed earlier shows itself: with the variable capital declining relatively — that is, as the social productivity of labour develops — more and more total capital is needed to put the same amount of labour-power to work and to take in the same quantity of surplus labour. So in the same measure as capitalist production develops, the possibility of a relatively surplus working population develops too — and the reason is not that the productivity of social labour is falling, but that it is rising; not from any absolute mismatch between labour and the means of subsistence, or the means of producing them, but a disproportion rooted in capital's exploitation of labour — the mismatch between the mounting growth of capital and its relatively falling need for a growing population.
If the rate of profit falls by 50 per cent, it falls by half. So for the mass of profit to stay the same, the capital has to double. For the mass of profit to hold steady while the rate of profit falls, the multiplier showing the growth of total capital must equal the divisor showing the fall in the rate of profit. If the rate of profit drops from 40 per cent to 20 per cent, total capital has to rise in the inverse ratio 20:40 for the result to stay the same. Had the rate of profit fallen from 40 per cent to 8 per cent, capital would have had to grow in the ratio 8:40 — five times over. A capital of 1,000,000 at 40 per cent yields 400,000; a capital of 5,000,000 at 8 per cent yields 400,000 as well. That is the condition for the result to stay the same.
If the result is to grow, on the other hand, capital must grow in a greater proportion than the rate of profit falls. Put another way: for the variable part of the total capital not merely to stay the same in absolute terms but to grow, even though its percentage of the total capital falls, total capital must grow proportionally faster than the percentage of variable capital falls. Its growth must be large enough that, in its new composition, it needs more than the old variable capital for buying labour-power. Take a capital of 100 whose variable part drops from 40 to 20: to use more than 40 in variable capital, total capital must rise above 200.
Suppose the mass of workers being exploited stays the same, and only the length and intensity of the working day grow. Even then, the capital applied would have to grow — because more capital is needed even to set the same amount of labour working under the old conditions of exploitation, once the composition of capital has changed.
So the one and the same development of the social productivity of labour shows itself, as capitalist production advances, on the one hand as a tendency of the rate of profit to fall ever further, and on the other as a steady growth in the absolute amount of surplus-value, or profit, appropriated. Overall, then, as variable capital and profit decline relatively, both increase absolutely. This two-sided effect, as shown, can only present itself as total capital growing at a faster pace than the rate of profit is falling. To put an absolutely larger variable capital to work at a higher composition — or with a relatively steeper rise in constant capital — total capital must grow not merely in proportion to that higher composition, but faster still. It follows from this that the further capitalist production develops, the greater the amount of capital required to employ the same labour-power — and all the more so if that labour-power itself is growing. The rising productivity of labour thus necessarily produces, on a capitalist basis, a permanent and only apparent surplus working population — surplus relative to capital's need for workers, not in itself. If variable capital makes up only a sixth of total capital instead of a half as before, then to employ the same labour-power total capital has to treble; to employ twice the labour-power, it has to grow sixfold.
Earlier economics, unable to explain the law that the rate of profit falls, brings in the rising mass of profit — profit growing in absolute size, whether for a single capitalist or for the whole social capital — as a sort of consolation. But that consolation too rests on mere commonplaces and bare possibilities.
That two things decide the mass of profit — the rate of profit, and how much capital is put to work at that rate — is a bare tautology. So the possibility that the mass of profit grows while the rate of profit falls is just another way of stating this tautology, and it gets us not a single step further — because capital can perfectly well grow while the mass of profit does not grow, and it can even grow while that mass falls. At 25 per cent, 100 gives 25; at 5 per cent, 400 gives only 20.
But if the same reasons that make the rate of profit fall also push accumulation forward — that is, the building of extra capital — and if every bit of extra capital also puts extra labour to work and turns out extra surplus-value; and if, on the other hand, the mere fall in the rate of profit means that constant capital, and with it the whole of the older capital, has grown — then the entire process stops being a mystery. Later we shall see what deliberate falsifications of the accounts are resorted to in order to conjure away the possibility that the mass of profit rises at the same time as the rate of profit falls.
We have seen how the very same reasons that produce a tendency of the general rate of profit to fall also accelerate the accumulation of capital, and so bring about a growth in the absolute size — the total mass — of the surplus labour (surplus-value, profit) that capital appropriates. Everything appears upside down in competition, and so in the minds of competition's agents; and so does this law — I mean this inner, necessary connection between two things that seem to contradict each other.
It is obvious that within the proportions set out above, a capitalist who commands a great deal of capital pockets a bigger mass of profit than a small capitalist who seems to make high profits. And the most superficial look at competition shows, further, that in certain circumstances — when the bigger capitalist wants to carve out room for himself on the market and push the smaller ones out, as in a crisis — he puts this to practical use, that is, he deliberately lowers his rate of profit so as to knock the smaller men out of the field. Commercial capital in particular, on which more later, also shows phenomena that make the falling profit look like an effect of expanding business, and so of larger capital. We give the scientific expression for this false conception later. Similar superficial observations come from comparing the rates of profit made in particular lines of business, depending on whether these fall under free competition or under monopoly.
The whole shallow notion that lives in the heads of competition's agents is there in our Roscher, the textbook economist — namely his assertion that this lowering of the rate of profit is 'more clever and more humane'. Here the decline in the rate of profit appears as an effect of capital growing, and of the calculation of the capitalists that a smaller rate of profit will let them pocket a bigger mass of profit. All of this (Adam Smith aside, on whom more later) rests on a complete failure to grasp what the general rate of profit even is, and on the crude notion that prices really are fixed by adding a more or less arbitrary profit mark-up on top of the actual value of commodities. Crude as these notions are, they arise necessarily from the inverted way the immanent laws of capitalist production — the laws inherent in it — show themselves in competition.
The law that productivity's development brings the rate of profit down while the mass of profit rises also shows itself in this: as the price of the commodities capital produces falls, the masses of profit they contain — the masses realized when they are sold — rise relatively.
The development of productivity, and the higher composition of capital that goes with it, means that more and more means of production are set in motion by less and less labour. So every aliquot portion of the total product — each single commodity, or each particular batch of commodities — takes in less living labour, and also holds less objectified labour, both in the wear and tear of the fixed capital employed and in the raw and ancillary materials used up. Each single commodity therefore holds a smaller total of labour — the labour stored up in means of production plus the labour added fresh in production. The price of a single commodity therefore falls.
The profit contained in a single commodity can still rise, even so, if the rate of surplus-value — absolute or relative — increases. There is less newly added labour in it, yet the part of that labour which goes unpaid grows in relation to the part that is paid. But this holds only within certain definite limits. As production advances, the quantity of living labour freshly added to any single commodity shrinks enormously; and with that, the unpaid labour the commodity contains also declines in absolute terms, however much it may have grown relative to the paid part. The mass of profit on every single commodity will therefore shrink a great deal as the productivity of labour develops — despite the fact that the rate of surplus-value is rising. And this shrinkage is slowed only by the elements of constant capital becoming cheaper, and by the other circumstances set out in Part One of this volume, which raise the rate of profit when the rate of surplus-value is given or even falling — exactly as the fall in the rate of profit is slowed by them.
When the price of every single commodity falls — the commodities that together form capital's total product — that means no more than this: a given quantity of labour is now worked up into a larger mass of commodities, so each single commodity contains less labour than it did before. That holds even if the price of one part of the constant capital, raw material and the like, rises. Leaving aside isolated cases (say, where the productivity of labour makes every element, constant and variable alike, cheaper to the same degree), the rate of profit is going to fall, and to fall even though the rate of surplus-value has gone up. First, because the pile of newly added labour has itself shrunk: even a bigger unpaid share of that smaller pile comes to less than the old, smaller share of the old, bigger pile did. Second, because capital's higher composition shows up inside the single commodity: the piece of its value that carries newly added labour shrinks next to the pieces that carry raw material, ancillary material and the wear of fixed capital.
This shift in the proportions among the parts of the single commodity's price — the part carrying newly added living labour going down, the parts carrying labour objectified earlier going up — is the form in which the fall of variable capital as against constant capital shows up in the price of the single commodity. That fall is absolute for a given amount of capital, 100 for instance, and so it is absolute for every single commodity too, as its proportional share of the capital reproduced. And yet the rate of profit, worked out simply from the price elements of the single commodity, would come out differently from what it really is. And that for the following reason.
You work out the rate of profit on the whole capital you have put to work, but you work it out for a definite stretch of time, and in practice that stretch is a year. Take the surplus-value — or profit — that the year produced and realized, set it against the total capital, turn the comparison into a percentage: that is the rate of profit. So this rate need not be the same as one worked out on how long the capital in question takes to turn over, rather than on the year; the two come together only if that capital turns over exactly once a year.
A year's profit, looked at from another angle, is just the total of the profits on the goods made and sold within that year. If we now work out the profit as a share of the commodities' cost prices — what they cost the capitalist — we get a rate of profit = p/k, where p stands for the profit realized over the year and k for the total of the cost prices of everything produced and sold in that period. Plainly, this rate p/k can coincide with the real rate of profit p/C — p over C, the mass of profit over the total capital — only when k = C, that is, when the capital turns over exactly once in the year.
Take three different states of an industrial capital.
I. Take an industrial capital of £8,000 that turns out 5,000 items a year and sells them at 30 shillings apiece; its annual turnover is therefore £7,500. On every item it makes 10 shillings of profit, which comes to £2,500 over the year. So each item holds 20 shillings of capital advanced plus 10 shillings of profit, which puts the profit rate per item at 10/20 = 50 per cent. Of the £7,500 that turns over, £5,000 is capital advanced and £2,500 is profit, so the rate of profit on the turnover, p/k, is likewise 50 per cent. Worked out on the total capital, though, the rate of profit p/C is 2,500/8,000 = 31¼ per cent.
II. Say the capital rises to £10,000. Suppose that, thanks to a greater productivity of labour, it can now turn out 10,000 items a year at a cost price of 20 shillings each, and sells them at 4 shillings profit — 24 shillings an item. The year's product then sells for £12,000, of which £10,000 is capital advanced and £2,000 is profit. p/k per item is 4/20; on the year's turnover it is 2,000/10,000 — 20 per cent either way. And since the total capital equals the total of the cost prices, £10,000, the real rate of profit p/C is 20 per cent this time too.
III. Say the capital rises to £15,000, with the productivity of labour still growing, and now turns out 30,000 items a year at a cost price of 13 shillings each, sold at 2 shillings profit — 15 shillings an item. The annual turnover is then 30,000 × 15 shillings = £22,500, of which £19,500 is capital advanced and £3,000 is profit. So p/k is 2/13 = 3,000/19,500 = 15 5/13 per cent. p/C, in contrast, is 3,000/15,000 = 20 per cent.
So there is a pattern. Only in the second case — where the capital value that turns over is exactly the whole capital — does the rate on the single item, or on the sum turned over, match the rate worked out on the total capital. In the first case, where less turns over than the total capital, the rate you get from the commodity's cost price is higher. In the third, where the sum turned over is bigger than the total capital, it is lower than the real rate, the one worked out on the total capital. And this holds in general.
In business practice people work out turnover only roughly. The usual assumption is that one turnover is complete the moment the money taken for the goods sold adds up to the capital applied. But a capital has only completed a full circuit when the cost prices of the goods sold add up to the whole capital. — F. E.
Here again it becomes clear how much this matters in capitalist production: the single commodity, or the output of any given stretch of time, is not to be taken on its own, as a mere commodity, but as something the capital advanced has produced, and judged in relation to the total capital that produces it.
Now, the rate of profit has to be worked out by taking the surplus-value that production has turned out and realized, and measuring it not only against the part of the capital that was used up and reappears in the commodities, but against that part plus the part that was not used up, yet is applied and goes on serving in production. Even so, the profit mass can only ever equal the profit or surplus-value that the commodities themselves contain and that their sale will realize.
When industry's productivity rises, the price of each single commodity falls. There is less labour inside it, paid and unpaid alike. Say the same labour now turns out three times the product: then two-thirds less labour goes into each single item. And since profit can only ever be a part of the labour contained in the single commodity, the profit mass on that commodity has to shrink — and within certain limits it shrinks even when the rate of surplus-value is climbing.
But in every case the profit mass on the whole output never drops below the profit mass it started with, as long as the capital puts the same number of workers to work as before, exploited to the same degree. (The same can happen with fewer workers, if they are exploited harder.) For the number of products grows in the same ratio as the profit on each product shrinks. The mass of profit stays the same; it is only spread differently across the commodities as a whole — and that changes nothing either in how the value created by the newly added labour is split between worker and capitalist. With the same quantity of labour employed, the mass of profit can only grow if the unpaid surplus labour grows, or, with the exploitation of labour unchanged, if more workers are employed. Or if both work together.
In all these cases — which, on the premise assumed, presuppose a growth of constant capital relative to variable capital and a growing size of the total capital applied — each single commodity contains less profit and the rate of profit falls, even when it is worked out on the single commodity; a given quantity of extra labour is expressed in a larger quantity of commodities, and the price of the single commodity falls. Taken abstractly, with the price of the single commodity falling because productivity has risen, and the number of these cheaper commodities growing at the same time, the rate of profit could stay the same — for instance if the rise in productivity worked on every component of the commodities evenly and at once, so that the total price fell in the same ratio as the productivity of labour rose, while the relation between the commodity's various price components stayed as it was. It could even go up, if the rate of surplus-value went up at the same time as the value of the elements of constant capital, fixed capital above all, fell sharply. But in reality, as we saw earlier, the rate of profit will fall in the long run.
The price fall of a single commodity, on its own, never allows a conclusion about the rate of profit. Everything depends on how large the total sum of capital is that goes into producing it. Suppose the price of a yard of material drops from 3 shillings to 1⅔ shillings. If you know that before the drop 1⅔ shillings of it went to constant capital — yarn and so on — ⅔ to wages and ⅔ to profit, and that after the drop 1 shilling goes to constant capital, ⅓ to wages and ⅓ to profit, you still do not know whether the rate of profit has stayed the same. That depends on whether the total capital advanced has grown — and by how much — and on how many more yards it turns out in a given time.
The phenomenon that grows out of the nature of the capitalist mode of production — with the productivity of labour rising, the price of the single commodity, or of a given quota of commodities, goes down, the number of commodities goes up, the profit mass on the single commodity shrinks and the rate of profit on the whole collection of commodities shrinks, while the profit mass on the total collection grows — this phenomenon shows itself on the surface only as this: a drop in the profit mass on each single commodity, a drop in its price, a growth in the profit mass on the larger total number of commodities produced by the total capital of society, or by that of a single capitalist.
And it then gets understood this way: the capitalist, of his own free choice, charges less profit on each single commodity but makes up for it on the larger number of items he produces. Behind this view stands the idea of profit upon alienation — profit made in the selling, not in the making — which is itself taken over from the way commercial capital sees things.
We saw earlier, in Parts Four and Seven of Volume 1, that when the productivity of labour rises, the growing quantity of goods and the fact that each single commodity gets cheaper — so far as these goods do not go into fixing the price of labour-power — leave the ratio of paid to unpaid labour inside the single commodity untouched, even though its price is falling.
Since in competition everything shows itself falsely — upside down, in fact — the individual capitalist can imagine: (1) that by cutting the price of the single commodity he is cutting his profit on it, yet earns a bigger profit thanks to the larger mass of commodities he sells; (2) that he fixes the price of the single commodity and then gets the total product's price by multiplying, while the original process runs the other way: it divides, and the multiplication is right only at second hand, on the assumption of that division. In fact, all the vulgar economist does is translate the odd notions of the capitalist caught up in competition into an apparently more theoretical, generalizing language, and wear himself out trying to set those notions up as correct.
In actual fact, when commodity prices fall while the mass of profit grows on the larger quantity of cheapened goods, this is only another way of saying the same law: the rate of profit falling and the mass of profit rising at one and the same time.
How far the rate of profit can fall while prices are rising at the same time is a question that belongs here no more than the point discussed earlier, in Volume 1, in connection with relative surplus-value. The capitalist who uses improved methods of production that are not yet general undersells the market price while staying above his own individual price of production; his rate of profit therefore rises, until competition evens this out. During such a period of adjustment, the second requirement of that earlier discussion is also met — the capital laid out grows. And depending on how far it grows, the capitalist will now be able to employ part of the workers he employed before, perhaps all of them or even more, under the new conditions, and so produce the same mass of profit or a larger one.
With wages and the working day both given, a variable capital — take 100 — stands for a fixed number of workers; it is an index of how many. Say £100 is the wages of 100 workers, for a week. If these 100 workers perform as much necessary labour as surplus labour — that is, they spend as much time each day working for themselves, reproducing their own wages, as they spend working for the capitalist, producing surplus-value — then the total value they produce would be £200, and the surplus-value they create would be £100. The rate of surplus-value, s/v, would be 100 per cent. But as we have seen, this rate of surplus-value would show up as widely different rates of profit, depending on how large the constant capital c is, and with it the total capital C, since the rate of profit is s/C. With the rate of surplus-value at 100 per cent:
If c = 50 and v = 100, then p' = 100/150 = 66⅔ per cent.
If c = 100 and v = 100, then p' = 100/200 = 50 per cent.
If c = 200 and v = 100, then p' = 100/300 = 33⅓ per cent.
If c = 300 and v = 100, then p' = 100/400 = 25 per cent.
If c = 400 and v = 100, then p' = 100/500 = 20 per cent.
So, with the degree of exploitation unchanged, one and the same rate of surplus-value would come out as a falling rate of profit — because the value of the constant capital grows along with its material bulk, and with it the value of the total capital, though not in the same proportion.
Suppose next that capital's composition doesn't shift like this only in a few isolated spheres of production, but more or less across all of them, or at least across the decisive spheres — so that the shift amounts to a change in the average organic composition of society's total capital. Then this steady growth of constant capital relative to variable capital must necessarily produce a gradual fall in the general rate of profit, so long as the rate of surplus-value — the degree to which capital exploits labour — stays the same. And capitalist production has already been shown to obey a law here: as it develops, variable capital declines relative to constant, and so relative to the total capital set in motion. That means only this: the same number of workers, the same quantity of labour-power placed at capital's disposal by a variable capital of a given value, sets in motion, works up and productively consumes in the same time a constantly growing mass of machinery, fixed capital of every kind, means of labour, and raw and auxiliary materials — this being the work of the distinctive methods of production that grow up inside capitalist production. Which is to say that the same workers operate with a constant capital whose value keeps growing.
This progressive decline of variable capital relative to constant capital, and so relative to the total capital as well, is the same thing as a progressively higher organic composition of the social capital on average. It is equally just another way of stating the progressive development of the social productivity of labour — which shows itself precisely in this, that through the growing use of machinery and fixed capital more raw and auxiliary materials are turned into products by the same number of workers in the same time, that is, with less labour. To this growing value of the constant capital there corresponds a product that grows steadily cheaper — although that value only remotely reflects the growth in the real mass of use-values of which the constant capital consists in material terms. Each individual product, taken on its own, contains a smaller sum of labour than it did at lower stages of production, where the capital laid out on labour stood in a far greater ratio to the capital laid out on means of production. So the series we set up hypothetically at the start does express the real tendency of capitalist production. That real tendency, as the variable capital keeps declining relative to the constant, produces an ever higher organic composition of the total capital. The immediate result is that the rate of surplus-value — with the degree of exploitation of labour staying the same or even rising — comes out as a steadily falling general rate of profit. (We shall show later why the fall does not appear in this absolute form, but rather more as a tendency towards a progressive fall.) This progressive tendency of the general rate of profit to fall is thus nothing but the way the capitalist mode of production, and it alone, expresses the steady advance of the social productivity of labour.
This is not to say that the rate of profit may not also fall temporarily for other reasons. But it proves that the nature of the capitalist mode of production itself makes it a self-evident necessity that, as that mode advances, the general average rate of surplus-value must express itself in a falling general rate of profit. Since the mass of living labour applied steadily declines relative to the mass of objectified labour it sets in motion — the means of production it productively consumes — the unpaid part of that living labour, the part that takes shape as surplus-value, must likewise shrink, step by step, in proportion to the value of the whole capital applied. But this ratio between the mass of surplus-value and the value of the total capital applied is what makes up the rate of profit. So the rate of profit must steadily fall.
Simple as this law looks after the arguments so far, no earlier economics managed to discover it, as we shall see later on. Those writers saw the phenomenon, and tormented themselves with one contradictory attempt after another to explain it. Given how much the law matters for capitalist production, one can say it forms the mystery whose solution all of political economy has circled since Adam Smith, and that what separates the various schools since Adam Smith is the different attempts each makes to solve it. But consider the other side: earlier political economy fumbled around the distinction between constant and variable capital without ever managing to state it precisely; it never presented surplus-value as separate from profit, and never presented profit in general in its pure form, apart from the various components that have made themselves independent of one another — industrial profit, commercial profit, interest, ground-rent; and it never thoroughly analysed the differences in the organic composition of capital, and so never the formation of the general rate of profit either. Once you take all that into account, it stops being a puzzle that it never found the answer to this riddle.
We deliberately set out this law before showing how profit falls apart into different categories that have become independent of one another. Presenting it in this order — before profit is split into the various slices that go to various groups of people — proves from the outset that the law, taken in its generality, does not depend on that split, nor on the relations between the categories of profit that grow out of it. The profit we are speaking of here is only another name for surplus-value itself, shown now in relation to the total capital instead of in relation to the variable capital it springs from. The fall in the rate of profit therefore expresses the falling proportion of surplus-value itself to the total capital advanced, and is thus independent of however that surplus-value may be distributed among different categories.
We have seen that when capitalist development has reached a stage at which the composition of capital c:v is 50:100, a rate of surplus-value of 100 per cent comes out as a rate of profit of 66⅔ per cent, and that at a higher stage, where c:v is 400:100, the same rate of surplus-value comes out as a rate of profit of only 20 per cent. What holds for different stages of development that follow one another in a single country also holds for different stages of development that exist side by side, at the same time, in different countries. In the undeveloped country, where the first of those compositions of capital is the average, the general rate of profit would be 66⅔ per cent; in the country at the second, far higher stage of development, it would be 20 per cent.
The difference between the two national rates of profit could disappear, and even be reversed, if labour in the less developed country were less productive — so that more labour went into making less of the same commodity — more exchange-value carried by less use-value. The worker would then have to spend a bigger share of his time reproducing his own means of subsistence, or their value, and a smaller share producing surplus-value; he would supply less surplus labour, so that the rate of surplus-value would be lower. Suppose, for example, that in the less advanced country the worker worked two-thirds of the day for himself and one-third for the capitalist; then, on the assumptions of the example above, the same amount of labour-power would be paid 133⅓ and would yield a surplus of only 66⅔. A constant capital of 50 would correspond to that variable capital of 133⅓. The rate of surplus-value would now come to 133⅓ : 66⅔ = 50 per cent, and the rate of profit to 183⅓ : 66⅔, or roughly 36½ per cent.
We have not yet examined the different components into which profit divides, so for us they do not exist yet. The following is therefore said in advance, only to head off misunderstandings. When countries at different levels of development are compared — especially countries of developed capitalist production with those where labour is not yet formally subsumed under capital — not yet run as a capitalist's own operation — although the worker is in reality exploited by the capitalist (in India, for example, where the ryot, the peasant farmer, works his land as an independent producer, so that his production as such is not yet subsumed under capital, even though the usurer may, in the form of interest, skim off not only his whole surplus labour but — put in capitalist terms — part of his wages as well) — then it would be very wrong to try to measure a country's rate of profit by the level of its rate of interest. Interest of that kind includes the whole profit and more than the profit, whereas in countries of developed capitalist production it expresses only a proportional share of the surplus-value or profit produced. And on the other side, the rate of interest there is predominantly determined by circumstances (the usurers' advances to the big landowners, who draw the ground-rent) which have nothing at all to do with profit, and only show the proportion in which usury appropriates the ground-rent.
In countries at different levels of capitalist development, and therefore with different organic compositions of capital, the rate of surplus-value — one of the factors that determines the rate of profit — can be higher in the country where the normal working day is shorter than in the one where it is longer. First: if England's ten-hour working day, because of its greater intensity, is equal to Austria's fourteen-hour day, then with the working day divided the same way in both, five hours of surplus labour in England can represent a higher value on the world market than seven hours in Austria. Second: in England a larger part of the working day can form surplus labour than in Austria.
The law of the falling rate of profit — the law that expresses the same, or even a rising, rate of surplus-value — says in other words: take any given quantity of average social capital, say a capital of 100. An ever larger part of it is represented by means of labour, and an ever smaller part by living labour. Since the whole mass of living labour added to the means of production keeps falling in proportion to what those means are worth, the unpaid labour too, and the portion of value in which it is represented, falls in relation to the value of the total capital advanced. Or: a smaller and smaller proportional share of the whole capital laid out actually turns into living labour, and so the total capital, in proportion to its size, absorbs less and less surplus labour — even though the ratio of the unpaid part of the labour applied to the paid part may at the same time be growing. The relative decline of variable capital and the relative growth of constant capital, even though both parts grow in absolute terms, is, as we have said, only another expression for the increased productivity of labour.
Say a capital of 100 is made up of 80c + 20v, and that 20v stands for 20 workers. Suppose the rate of surplus-value is 100 per cent: the workers spend half the day working for themselves and half for the capitalist. In a less developed country the capital might be 20c + 80v, and that 80v would stand for 80 workers. But these workers keep two-thirds of the working day for themselves and work for the capitalist for only a third of it.
With everything else taken as equal, the first set of workers turns out a value of 40, the second a value of 120. So the first capital produces 80c + 20v + 20s = 120, a rate of profit of 20 per cent; the second produces 20c + 80v + 40s = 140, a rate of profit of 40 per cent. The rate of profit in the second case is thus twice as large as in the first, even though the rate of surplus-value in the first case is 100 per cent — twice as high as in the second, where it is only 50 per cent. In exchange for that, however, a capital of equal size appropriates the surplus labour of only 20 workers in the first case, and of 80 workers in the second.
The law that the rate of profit falls progressively — that is, that the surplus labour appropriated declines relative to the mass of objectified labour (labour already embodied in the means of production) that living labour sets in motion — in no way rules out that the social capital sets in motion and exploits a growing absolute mass of labour, and so appropriates a growing absolute mass of surplus labour. Nor does it rule out that the capitals under the command of individual capitalists command a growing mass of labour, and so of surplus labour — the latter even when the number of workers they command does not grow.
Suppose the working population is given, two million say. Suppose too that the average working day has a given length and intensity, and that wages are given, and with them the relation between necessary and surplus labour. Then the total labour of these two million — and their surplus labour too, as expressed in surplus-value — always produces the same magnitude of value. But the mass of constant capital — fixed and circulating — that this labour sets in motion grows, and the capital's value grows with its mass, though not in the same proportion. So the ratio of that value to the value of the constant capital falls; and along with that ratio the rate of profit falls, even though the same mass of living labour is commanded as before and the same mass of surplus labour is absorbed by capital.
The ratio changes not because the mass of living labour falls, but because the mass of objectified labour it sets in motion rises. The decrease is a relative one, not an absolute one; in fact it has nothing at all to do with how large the labour and the surplus labour set in motion are in absolute terms. What makes the rate of profit fall is not that the variable part of the total capital shrinks in absolute terms. It is only that this part shrinks relative to the constant part.
What was just true of a fixed quantity of labour and surplus labour also holds when the number of workers grows, and so — under the assumptions just given — when the mass of labour under capital's command in general grows, and in particular its unpaid part, surplus labour. Suppose the working population grows from two million to three million, and suppose the variable capital laid out on labour-power also goes from two million to three million, while constant capital rises from 4 million to 15 million. Then, under the given assumptions — a constant working day and a constant rate of surplus-value — the mass of surplus labour, and so of surplus-value, grows by half, by 50 per cent, from 2 million to 3 million. All the same, despite this 50 per cent growth in the absolute mass of surplus labour and so of surplus-value, the ratio of variable to constant capital would fall from 2 : 4 to 3 : 15, and the ratio of surplus-value to total capital would come out as follows (in millions):
While the mass of surplus-value has risen by half, the rate of profit has halved. But profit is nothing but surplus-value measured against the social capital, and so the mass of profit — its absolute size — is, looked at socially, equal to the absolute size of surplus-value. The absolute size of profit, its total mass, would therefore have grown by 50 per cent, despite the enormous fall in the ratio of that mass of profit to the total capital advanced — that is, despite the enormous fall in the general rate of profit.
So capital can employ more workers, set a larger absolute quantity of labour moving, absorb a larger absolute quantity of surplus labour, produce a larger mass of surplus-value and so a larger absolute mass of profit — and this can keep growing, and keep growing progressively, even while the rate of profit falls progressively. And this is not just something that can happen. Apart from passing fluctuations, on the basis of capitalist production it must happen.
The capitalist production process is essentially also a process of accumulation. We have seen how, as capitalist production advances, the amount of value that has simply to be reproduced and kept going grows along with the productivity of labour, even if the labour-power employed stays the same. But as the social productivity of labour develops, the quantity of use-values produced grows even more, and the means of production are one part of that. And the extra labour that has to be appropriated for this extra wealth to be turned back into capital depends not on the value but on the mass of these means of production (means of subsistence included), because when the worker actually works, what matters to him is not what the means of production are worth but what they can do — their use-value.
Accumulation itself — and the concentration of capital that comes with it — is itself a material means of raising the productivity of labour. But this growth of the means of production includes the growth of the working population: a population comes into being that matches the surplus capital and even, by and large, constantly overflows what it needs, and so there is an over-population of workers.
Suppose that surplus capital momentarily outstrips the working population under its command. That would work in two directions. On the one side, higher wages would gradually add to the number of workers, softening the destructive forces that cut down their children and making it easier to marry. On the other side, the very methods that create relative surplus-value — new machinery and better machinery — would far more quickly produce an artificial, relative over-population, and this in turn is the hotbed of a really rapid growth of population, because in capitalist production misery is what produces population.
So it follows of itself from what capitalist accumulation essentially is — and accumulation is only one aspect of the capitalist production process — that the greater mass of means of production destined to be turned into capital always finds at hand a correspondingly greater, indeed excessive, working population, ready to be exploited. So as production and accumulation move forward, the quantity of surplus labour that can be appropriated — and that is appropriated — must rise, and with it the absolute mass of profit appropriated by the social capital. But those very laws of production and accumulation drive up the value of constant capital together with its mass, and at an ever faster rate, compared with the value of the variable part that is exchanged for living labour. So it is these same laws that give the social capital a mass of profit that grows larger in absolute terms, and a rate of profit that falls.
Here we leave aside entirely the fact that the same amount of value comes to stand for more and more use-values and satisfactions, as capitalist production advances and with it the productivity of social labour develops and the branches of production — and so the products — multiply.
As capitalist production and accumulation develop, labour processes must be carried out on an ever larger scale, and so each individual establishment needs ever larger dimensions and ever larger advances of capital. The concentration of capitals therefore grows — and along with it, though less markedly, the number of capitalists. This growing concentration is both one of the material preconditions of that development and one of the results it itself throws off. And side by side with it, each reacting on the other, goes the progressive expropriation of the more or less immediate producers.
So it stands to reason for the individual capitalists that they command ever bigger armies of workers (however much their variable capital falls relative to constant capital), and that the mass of surplus-value they appropriate, and so of profit, grows — together with the fall in the rate of profit, and in spite of it. The same reasons that concentrate masses of worker armies under the command of individual capitalists are precisely the ones that also swell the mass of fixed capital applied, like that of raw and ancillary materials, in growing proportion as against how much living labour is applied.
One more thing needs mentioning here. When the working population is given, and the rate of surplus-value rises — whether by lengthening or intensifying the working day, or by a fall in the value of wages brought about by the growth in the productivity of labour — then the mass of surplus-value, and with it profit in absolute terms, must grow, despite the relative shrinking of variable capital against constant.
One and the same development of the productivity of social labour, one and the same set of laws, shows itself in the relative fall of variable capital against total capital and in the accumulation this speeds up — while, the other way round, accumulation reacts back and becomes the starting point for productivity to grow further and for variable capital to decline further in relative terms. And this same development is expressed, temporary fluctuations apart, in an ever-rising total of labour-power employed and an ever-growing absolute mass of surplus-value, and so of profit.
In what form, then, must this double-edged law present itself — the law of a falling rate of profit and, at the same time, a rising absolute mass of profit, both springing from the same reasons? A law resting on the fact that, given these conditions, more surplus labour — and so more surplus-value — is appropriated, and that, whether we look at the total capital or at a single capital as a mere piece of the total capital, profit and surplus-value are one and the same magnitude?
Take a fixed fractional share of the capital as the base for reckoning the rate of profit — an aliquot part, say 100. This 100 has the average composition of the total capital, say 80c + 20v. In Part Two of this volume we saw that what fixes the average rate of profit for the different branches of production is not the composition each branch has on its own, but the average social composition.
As the variable part declines relative to the constant part, and so relative to the total capital of 100, the rate of profit falls even when the degree of exploitation of labour stays the same, or even rises. And the relative size of surplus-value falls — that is, its ratio to the value of the total capital of 100 that has been advanced. But it is not just this relative size that drops. The quantity of surplus-value — and so of profit — that the total capital of 100 takes in also drops, and drops absolutely. Take a rate of surplus-value of 100 per cent. Then 60c + 40v turns out a mass of surplus-value, and so of profit, of 40; 70c + 30v turns out 30; 80c + 20v, only 20.
This fall is in the mass of surplus-value, and so of profit. It follows because a total capital of 100 sets less living labour moving altogether, and so, at the same degree of exploitation, also sets less surplus labour moving and produces less surplus-value. Take any fixed fractional share of the social capital — that is, of capital of average social composition — as the unit we measure surplus-value by, and that is what happens in every calculation of profit. On that measure, a relative fall of surplus-value and its absolute fall are simply identical. In the cases above, the rate of profit sinks from 40 per cent to 30 per cent and then 20 per cent, because one and the same capital produces less and less surplus-value — and so less profit — in absolute terms: 40, then 30, then 20. Since the capital against which surplus-value is measured has a fixed value of 100, a drop in surplus-value's proportion to that fixed size can only be another way of saying that surplus-value and profit have shrunk in absolute size. This really is a tautology. But that this decline happens follows, as has been shown, from the nature of the development of the capitalist production process.
But on the other side, the very same reasons that make surplus-value — and so profit — shrink in absolute terms on any given capital, and with it the rate of profit as worked out in percentages, also raise the absolute mass of surplus-value, and so of profit, that the social capital appropriates (that is, all capitalists taken together).
How is this to show itself? How can it show itself at all — or what conditions are contained in this apparent contradiction?
If any aliquot part of the social capital — any equal, measured-out share, say 100, and so any 100 of capital of average social composition — is a fixed magnitude, so that for it a fall in the rate of profit is one and the same thing as a fall in the absolute amount of profit, precisely because the capital it is measured against is a constant magnitude, then the total social capital is a variable magnitude — and so is the capital held by individual capitalists. It has to move in inverse proportion to the shrinking of its variable part, if the assumptions we have made are to hold.
In the earlier example the composition worked out at 60c + 40v; the surplus-value or profit came to 40, and the rate of profit was therefore 40 per cent. Suppose the total capital stood at 1,000,000 at this stage. Then total surplus-value, and so total profit, was 400,000. If the composition later becomes 80c + 20v, then — with the degree of exploitation of labour unchanged — each 100 yields 20 in surplus-value or profit. But since the mass of surplus-value or profit grows, as we showed, despite this falling rate of profit (that is, despite each capital of 100 producing less surplus-value) — say it grows from 400,000 to 440,000 — this only works because the total capital formed together with this new composition has grown to 2,200,000. The total capital now set in motion has grown to 220 per cent of what it was then, while the rate of profit has dropped by 50 per cent.
Had the capital simply doubled, it would at 20 per cent have thrown off exactly the surplus-value and profit that the old 1,000,000 threw off at 40 per cent. If it had grown to less than double, the surplus-value and profit it produced would have been smaller than what the earlier capital of 1,000,000 produced — even though, at its earlier composition, that capital needed to grow only from 1,000,000 to 1,100,000 to lift its surplus-value from 400,000 to 440,000.
Here the law already developed earlier shows itself: with the variable capital declining relatively — that is, as the social productivity of labour develops — more and more total capital is needed to put the same amount of labour-power to work and to take in the same quantity of surplus labour. So in the same measure as capitalist production develops, the possibility of a relatively surplus working population develops too — and the reason is not that the productivity of social labour is falling, but that it is rising; not from any absolute mismatch between labour and the means of subsistence, or the means of producing them, but a disproportion rooted in capital's exploitation of labour — the mismatch between the mounting growth of capital and its relatively falling need for a growing population.
If the rate of profit falls by 50 per cent, it falls by half. So for the mass of profit to stay the same, the capital has to double. For the mass of profit to hold steady while the rate of profit falls, the multiplier showing the growth of total capital must equal the divisor showing the fall in the rate of profit. If the rate of profit drops from 40 per cent to 20 per cent, total capital has to rise in the inverse ratio 20:40 for the result to stay the same. Had the rate of profit fallen from 40 per cent to 8 per cent, capital would have had to grow in the ratio 8:40 — five times over. A capital of 1,000,000 at 40 per cent yields 400,000; a capital of 5,000,000 at 8 per cent yields 400,000 as well. That is the condition for the result to stay the same.
If the result is to grow, on the other hand, capital must grow in a greater proportion than the rate of profit falls. Put another way: for the variable part of the total capital not merely to stay the same in absolute terms but to grow, even though its percentage of the total capital falls, total capital must grow proportionally faster than the percentage of variable capital falls. Its growth must be large enough that, in its new composition, it needs more than the old variable capital for buying labour-power. Take a capital of 100 whose variable part drops from 40 to 20: to use more than 40 in variable capital, total capital must rise above 200.
Suppose the mass of workers being exploited stays the same, and only the length and intensity of the working day grow. Even then, the capital applied would have to grow — because more capital is needed even to set the same amount of labour working under the old conditions of exploitation, once the composition of capital has changed.
So the one and the same development of the social productivity of labour shows itself, as capitalist production advances, on the one hand as a tendency of the rate of profit to fall ever further, and on the other as a steady growth in the absolute amount of surplus-value, or profit, appropriated. Overall, then, as variable capital and profit decline relatively, both increase absolutely. This two-sided effect, as shown, can only present itself as total capital growing at a faster pace than the rate of profit is falling. To put an absolutely larger variable capital to work at a higher composition — or with a relatively steeper rise in constant capital — total capital must grow not merely in proportion to that higher composition, but faster still. It follows from this that the further capitalist production develops, the greater the amount of capital required to employ the same labour-power — and all the more so if that labour-power itself is growing. The rising productivity of labour thus necessarily produces, on a capitalist basis, a permanent and only apparent surplus working population — surplus relative to capital's need for workers, not in itself. If variable capital makes up only a sixth of total capital instead of a half as before, then to employ the same labour-power total capital has to treble; to employ twice the labour-power, it has to grow sixfold.
Earlier economics, unable to explain the law that the rate of profit falls, brings in the rising mass of profit — profit growing in absolute size, whether for a single capitalist or for the whole social capital — as a sort of consolation. But that consolation too rests on mere commonplaces and bare possibilities.
That two things decide the mass of profit — the rate of profit, and how much capital is put to work at that rate — is a bare tautology. So the possibility that the mass of profit grows while the rate of profit falls is just another way of stating this tautology, and it gets us not a single step further — because capital can perfectly well grow while the mass of profit does not grow, and it can even grow while that mass falls. At 25 per cent, 100 gives 25; at 5 per cent, 400 gives only 20.
But if the same reasons that make the rate of profit fall also push accumulation forward — that is, the building of extra capital — and if every bit of extra capital also puts extra labour to work and turns out extra surplus-value; and if, on the other hand, the mere fall in the rate of profit means that constant capital, and with it the whole of the older capital, has grown — then the entire process stops being a mystery. Later we shall see what deliberate falsifications of the accounts are resorted to in order to conjure away the possibility that the mass of profit rises at the same time as the rate of profit falls.
We have seen how the very same reasons that produce a tendency of the general rate of profit to fall also accelerate the accumulation of capital, and so bring about a growth in the absolute size — the total mass — of the surplus labour (surplus-value, profit) that capital appropriates. Everything appears upside down in competition, and so in the minds of competition's agents; and so does this law — I mean this inner, necessary connection between two things that seem to contradict each other.
It is obvious that within the proportions set out above, a capitalist who commands a great deal of capital pockets a bigger mass of profit than a small capitalist who seems to make high profits. And the most superficial look at competition shows, further, that in certain circumstances — when the bigger capitalist wants to carve out room for himself on the market and push the smaller ones out, as in a crisis — he puts this to practical use, that is, he deliberately lowers his rate of profit so as to knock the smaller men out of the field. Commercial capital in particular, on which more later, also shows phenomena that make the falling profit look like an effect of expanding business, and so of larger capital. We give the scientific expression for this false conception later. Similar superficial observations come from comparing the rates of profit made in particular lines of business, depending on whether these fall under free competition or under monopoly.
The whole shallow notion that lives in the heads of competition's agents is there in our Roscher, the textbook economist — namely his assertion that this lowering of the rate of profit is 'more clever and more humane'. Here the decline in the rate of profit appears as an effect of capital growing, and of the calculation of the capitalists that a smaller rate of profit will let them pocket a bigger mass of profit. All of this (Adam Smith aside, on whom more later) rests on a complete failure to grasp what the general rate of profit even is, and on the crude notion that prices really are fixed by adding a more or less arbitrary profit mark-up on top of the actual value of commodities. Crude as these notions are, they arise necessarily from the inverted way the immanent laws of capitalist production — the laws inherent in it — show themselves in competition.
The law that productivity's development brings the rate of profit down while the mass of profit rises also shows itself in this: as the price of the commodities capital produces falls, the masses of profit they contain — the masses realized when they are sold — rise relatively.
The development of productivity, and the higher composition of capital that goes with it, means that more and more means of production are set in motion by less and less labour. So every aliquot portion of the total product — each single commodity, or each particular batch of commodities — takes in less living labour, and also holds less objectified labour, both in the wear and tear of the fixed capital employed and in the raw and ancillary materials used up. Each single commodity therefore holds a smaller total of labour — the labour stored up in means of production plus the labour added fresh in production. The price of a single commodity therefore falls.
The profit contained in a single commodity can still rise, even so, if the rate of surplus-value — absolute or relative — increases. There is less newly added labour in it, yet the part of that labour which goes unpaid grows in relation to the part that is paid. But this holds only within certain definite limits. As production advances, the quantity of living labour freshly added to any single commodity shrinks enormously; and with that, the unpaid labour the commodity contains also declines in absolute terms, however much it may have grown relative to the paid part. The mass of profit on every single commodity will therefore shrink a great deal as the productivity of labour develops — despite the fact that the rate of surplus-value is rising. And this shrinkage is slowed only by the elements of constant capital becoming cheaper, and by the other circumstances set out in Part One of this volume, which raise the rate of profit when the rate of surplus-value is given or even falling — exactly as the fall in the rate of profit is slowed by them.
When the price of every single commodity falls — the commodities that together form capital's total product — that means no more than this: a given quantity of labour is now worked up into a larger mass of commodities, so each single commodity contains less labour than it did before. That holds even if the price of one part of the constant capital, raw material and the like, rises. Leaving aside isolated cases (say, where the productivity of labour makes every element, constant and variable alike, cheaper to the same degree), the rate of profit is going to fall, and to fall even though the rate of surplus-value has gone up. First, because the pile of newly added labour has itself shrunk: even a bigger unpaid share of that smaller pile comes to less than the old, smaller share of the old, bigger pile did. Second, because capital's higher composition shows up inside the single commodity: the piece of its value that carries newly added labour shrinks next to the pieces that carry raw material, ancillary material and the wear of fixed capital.
This shift in the proportions among the parts of the single commodity's price — the part carrying newly added living labour going down, the parts carrying labour objectified earlier going up — is the form in which the fall of variable capital as against constant capital shows up in the price of the single commodity. That fall is absolute for a given amount of capital, 100 for instance, and so it is absolute for every single commodity too, as its proportional share of the capital reproduced. And yet the rate of profit, worked out simply from the price elements of the single commodity, would come out differently from what it really is. And that for the following reason.
You work out the rate of profit on the whole capital you have put to work, but you work it out for a definite stretch of time, and in practice that stretch is a year. Take the surplus-value — or profit — that the year produced and realized, set it against the total capital, turn the comparison into a percentage: that is the rate of profit. So this rate need not be the same as one worked out on how long the capital in question takes to turn over, rather than on the year; the two come together only if that capital turns over exactly once a year.
A year's profit, looked at from another angle, is just the total of the profits on the goods made and sold within that year. If we now work out the profit as a share of the commodities' cost prices — what they cost the capitalist — we get a rate of profit = p/k, where p stands for the profit realized over the year and k for the total of the cost prices of everything produced and sold in that period. Plainly, this rate p/k can coincide with the real rate of profit p/C — p over C, the mass of profit over the total capital — only when k = C, that is, when the capital turns over exactly once in the year.
Take three different states of an industrial capital.
I. Take an industrial capital of £8,000 that turns out 5,000 items a year and sells them at 30 shillings apiece; its annual turnover is therefore £7,500. On every item it makes 10 shillings of profit, which comes to £2,500 over the year. So each item holds 20 shillings of capital advanced plus 10 shillings of profit, which puts the profit rate per item at 10/20 = 50 per cent. Of the £7,500 that turns over, £5,000 is capital advanced and £2,500 is profit, so the rate of profit on the turnover, p/k, is likewise 50 per cent. Worked out on the total capital, though, the rate of profit p/C is 2,500/8,000 = 31¼ per cent.
II. Say the capital rises to £10,000. Suppose that, thanks to a greater productivity of labour, it can now turn out 10,000 items a year at a cost price of 20 shillings each, and sells them at 4 shillings profit — 24 shillings an item. The year's product then sells for £12,000, of which £10,000 is capital advanced and £2,000 is profit. p/k per item is 4/20; on the year's turnover it is 2,000/10,000 — 20 per cent either way. And since the total capital equals the total of the cost prices, £10,000, the real rate of profit p/C is 20 per cent this time too.
III. Say the capital rises to £15,000, with the productivity of labour still growing, and now turns out 30,000 items a year at a cost price of 13 shillings each, sold at 2 shillings profit — 15 shillings an item. The annual turnover is then 30,000 × 15 shillings = £22,500, of which £19,500 is capital advanced and £3,000 is profit. So p/k is 2/13 = 3,000/19,500 = 15 5/13 per cent. p/C, in contrast, is 3,000/15,000 = 20 per cent.
So there is a pattern. Only in the second case — where the capital value that turns over is exactly the whole capital — does the rate on the single item, or on the sum turned over, match the rate worked out on the total capital. In the first case, where less turns over than the total capital, the rate you get from the commodity's cost price is higher. In the third, where the sum turned over is bigger than the total capital, it is lower than the real rate, the one worked out on the total capital. And this holds in general.
In business practice people work out turnover only roughly. The usual assumption is that one turnover is complete the moment the money taken for the goods sold adds up to the capital applied. But a capital has only completed a full circuit when the cost prices of the goods sold add up to the whole capital. — F. E.
Here again it becomes clear how much this matters in capitalist production: the single commodity, or the output of any given stretch of time, is not to be taken on its own, as a mere commodity, but as something the capital advanced has produced, and judged in relation to the total capital that produces it.
Now, the rate of profit has to be worked out by taking the surplus-value that production has turned out and realized, and measuring it not only against the part of the capital that was used up and reappears in the commodities, but against that part plus the part that was not used up, yet is applied and goes on serving in production. Even so, the profit mass can only ever equal the profit or surplus-value that the commodities themselves contain and that their sale will realize.
When industry's productivity rises, the price of each single commodity falls. There is less labour inside it, paid and unpaid alike. Say the same labour now turns out three times the product: then two-thirds less labour goes into each single item. And since profit can only ever be a part of the labour contained in the single commodity, the profit mass on that commodity has to shrink — and within certain limits it shrinks even when the rate of surplus-value is climbing.
But in every case the profit mass on the whole output never drops below the profit mass it started with, as long as the capital puts the same number of workers to work as before, exploited to the same degree. (The same can happen with fewer workers, if they are exploited harder.) For the number of products grows in the same ratio as the profit on each product shrinks. The mass of profit stays the same; it is only spread differently across the commodities as a whole — and that changes nothing either in how the value created by the newly added labour is split between worker and capitalist. With the same quantity of labour employed, the mass of profit can only grow if the unpaid surplus labour grows, or, with the exploitation of labour unchanged, if more workers are employed. Or if both work together.
In all these cases — which, on the premise assumed, presuppose a growth of constant capital relative to variable capital and a growing size of the total capital applied — each single commodity contains less profit and the rate of profit falls, even when it is worked out on the single commodity; a given quantity of extra labour is expressed in a larger quantity of commodities, and the price of the single commodity falls. Taken abstractly, with the price of the single commodity falling because productivity has risen, and the number of these cheaper commodities growing at the same time, the rate of profit could stay the same — for instance if the rise in productivity worked on every component of the commodities evenly and at once, so that the total price fell in the same ratio as the productivity of labour rose, while the relation between the commodity's various price components stayed as it was. It could even go up, if the rate of surplus-value went up at the same time as the value of the elements of constant capital, fixed capital above all, fell sharply. But in reality, as we saw earlier, the rate of profit will fall in the long run.
The price fall of a single commodity, on its own, never allows a conclusion about the rate of profit. Everything depends on how large the total sum of capital is that goes into producing it. Suppose the price of a yard of material drops from 3 shillings to 1⅔ shillings. If you know that before the drop 1⅔ shillings of it went to constant capital — yarn and so on — ⅔ to wages and ⅔ to profit, and that after the drop 1 shilling goes to constant capital, ⅓ to wages and ⅓ to profit, you still do not know whether the rate of profit has stayed the same. That depends on whether the total capital advanced has grown — and by how much — and on how many more yards it turns out in a given time.
The phenomenon that grows out of the nature of the capitalist mode of production — with the productivity of labour rising, the price of the single commodity, or of a given quota of commodities, goes down, the number of commodities goes up, the profit mass on the single commodity shrinks and the rate of profit on the whole collection of commodities shrinks, while the profit mass on the total collection grows — this phenomenon shows itself on the surface only as this: a drop in the profit mass on each single commodity, a drop in its price, a growth in the profit mass on the larger total number of commodities produced by the total capital of society, or by that of a single capitalist.
And it then gets understood this way: the capitalist, of his own free choice, charges less profit on each single commodity but makes up for it on the larger number of items he produces. Behind this view stands the idea of profit upon alienation — profit made in the selling, not in the making — which is itself taken over from the way commercial capital sees things.
We saw earlier, in Parts Four and Seven of Volume 1, that when the productivity of labour rises, the growing quantity of goods and the fact that each single commodity gets cheaper — so far as these goods do not go into fixing the price of labour-power — leave the ratio of paid to unpaid labour inside the single commodity untouched, even though its price is falling.
Since in competition everything shows itself falsely — upside down, in fact — the individual capitalist can imagine: (1) that by cutting the price of the single commodity he is cutting his profit on it, yet earns a bigger profit thanks to the larger mass of commodities he sells; (2) that he fixes the price of the single commodity and then gets the total product's price by multiplying, while the original process runs the other way: it divides, and the multiplication is right only at second hand, on the assumption of that division. In fact, all the vulgar economist does is translate the odd notions of the capitalist caught up in competition into an apparently more theoretical, generalizing language, and wear himself out trying to set those notions up as correct.
In actual fact, when commodity prices fall while the mass of profit grows on the larger quantity of cheapened goods, this is only another way of saying the same law: the rate of profit falling and the mass of profit rising at one and the same time.
How far the rate of profit can fall while prices are rising at the same time is a question that belongs here no more than the point discussed earlier, in Volume 1, in connection with relative surplus-value. The capitalist who uses improved methods of production that are not yet general undersells the market price while staying above his own individual price of production; his rate of profit therefore rises, until competition evens this out. During such a period of adjustment, the second requirement of that earlier discussion is also met — the capital laid out grows. And depending on how far it grows, the capitalist will now be able to employ part of the workers he employed before, perhaps all of them or even more, under the new conditions, and so produce the same mass of profit or a larger one.