Marx’s Economic Manuscripts of 1861-63

3) Relative Surplus Value

Volume 30, MECW, p. 233-255

[III-125] We call the form of surplus value considered so far absolute surplus value because its very existence, its rate of growth, and its every increase is at the same time an absolute increase of created value (of produced value). It arises, as we have seen, from an extension of the necessary working day beyond its limits, and its absolute magnitude is equal to the magnitude of this extension, whereas its relative magnitude — the proportional surplus value, or the rate of surplus value — is given by the ratio of this extension, this fluxion, to its fluent, [176] necessary labour time. If the necessary labour time is 10 hours, the working day will be extended by 2, 3, 4, 5 hours. As a result, a value of 12-15 hours of labour will be created instead of one of 10. The extension of the normal working day, i.e. the total of necessary labour time + surplus labour time, is here the process by which surplus value grows, is increased.

Let us assume that the overall working day has reached its normal limits. Now there emerges, in the manner peculiar to and characteristic of it, capital’s tendency to posit surplus value, i.e. surplus labour time. Let the normal working day consist of 12 hours, of which 10 are necessary labour time, and 2 surplus labour time. Let an extension beyond this duration, hence a growth in absolute surplus value, be out of the question. It is of course clear that such a barrier — however one may fix it — is bound to assert itself, to be reached. (One may also assume, in order to have the problem present in its purest form, that the sum total of absolute surplus value cannot be raised any further, since the working population is given.) In this case, therefore, where the surplus value cannot be raised any further by lengthening the overall working day, how can it be raised any further at all? By shortening the necessary labour time. Given an overall working day of 12 hours (10 hours necessary labour time, 2 hours surplus labour time) the surplus value or the surplus labour time may e.g. grow by 50%, may grow from 2 hours to 3 — without any extension of the overall working day — lf the necessary labour time is shortened from 10 hours to 9 hours, by 1/10. The quantum of surplus labour time, consequently surplus value, may grow not only through a direct increase of surplus labour time achieved by a simultaneous lengthening of the overall working day, but also through the shortening of the necessary labour time, hence the conversion of labour time from necessary to surplus labour time. The normal working day would not be lengthened, but necessary labour time would be reduced and there would have been a change in the ratio governing the division of the overall working day into labour which replaces wages and labour which creates surplus value.

As we have seen, necessary labour time (as paid labour time) is nothing but the labour time which replaces the labour time contained in the wage, in the purchasing price of labour capacity (which is in reality the labour time required for the production of the wage). It could be reduced by reducing the wage. If the value of the wage is forcibly cut down, so also is the labour time contained in the wage, hence the labour time paid for the reproduction, the replacement, of the wage. As the value fell, so would the equivalent for the value: the equivalent value corresponding to, or rather equal to, this value. This is exactly what happens in practice, of course. The price of labour capacity, like that of every other commodity, does in practice rise and fall above and below its value. But this is of no concern to us, as we proceed from the assumption that the price of the commodity corresponds to its value, or we consider the phenomena on this assumption. The reduction of necessary labour time which is under discussion here has therefore to be analysed under the presupposition that labour capacity is sold at its value, that the worker receives the normal wage, and therefore that no reduction occurs in the amount of the means of subsistence which are required for the normal and traditional reproduction of his labour capacity.

[III-126] //An increase in surplus value achieved by reducing wages below their average level (without increasing the productivity of labour) is an increase in profit achieved by forcing the worker below the level of his normal conditions of life. On the other hand, an increase in wages over their normal average level is, on the part of the worker, a sharing in, an appropriation of, a part of his own surplus labour (similarly assuming the productive power of labour remains constant). In the first case * the capitalist encroaches upon the vital conditions of the workman, and upon the times of labour necessary for its own sustenance.* In the second case * the workman expropriates part of his own surplus labour. In both cases the one loses what the other gains, but the workman loses in life, what the capitalist gains in money, and in the other case the workman gains in enjoyment of life, what the capitalist loses in the rate of appropriating other people’s labour. *//

Any reduction in necessary labour time which takes place on the assumption that the price of labour capacity is equal to its value, hence that wages are not forced down below normal wages, is possible only through an increase in the productivity of labour, or through a higher development of the productive forces of labour, which is the same thing.

We saw when we were considering the commodity[10] that if the productive power of labour increases, the same use value will be produced in a shorter labour time, or a greater quantity of the same use values will be produced in the same labour time (or a shorter time, but this is included in case 2). The use value of the commodity remains the same although its exchange value has fallen, i.e. a smaller quantity of labour time is objectified in it, less labour is required to produce it. The amount of the means of subsistence required for the normal reproduction of labour capacity is not determined by their exchange value but by their use value — qualitatively and quantitatively. It is therefore not determined by the labour time required to produce them, objectified in them, but by the result of this labour time, by the real labour, to the extent that it is present in the product. Hence if the same amount of the means of subsistence can be produced in a shorter working period owing to an increase in the productivity of real labour, the value of labour capacity will fall, and along with that the labour time required for its reproduction, for the production of its equivalent value, the necessary labour time, although labour capacity will continue to be sold at its value. Just as any other commodity continues to be sold at its value if it costs 1/100 less today than before, because 1/100 less labour time is contained in it, although it continues to possess the same use value as before. Here the value of labour capacity falls, and therefore the necessary labour time too, not because the price of labour capacity has fallen below its value but because its value has itself fallen, i.e. because less labour time is objectified in the labour capacity, and therefore less labour time is required for its reproduction. In this case surplus labour time grows because necessary labour time has diminished. A part of the overall working day which was previously reserved for necessary labour is now set free, is annexed to the surplus labour time. A part of the necessary labour time is converted into surplus labour time; hence a portion of the overall value of the product, which previously entered the wage, now enters the surplus value (the capitalist’s gain). I call this form of surplus value relative surplus value.

It is clear from the outset that an increase in the productive power of labour can only lessen the value of its labour capacity or its necessary labour time to the extent that the products of this labour either directly enter into the worker’s consumption, such as means of nourishment and heating, housing, clothing, etc., or go into the constant capital (raw material and — instrument of labour) which is required for the manufacture of those products. The value of the constant capital entering into the product re-appears in the value of the product, and therefore the value of the product clearly falls, not only when there is a fall in the labour time required for its own manufacture but also, and just as much, when there is a fall in the labour time required for the manufacture of its conditions of production; that is to say the value of the raw material and instrument of labour required for the manufacture of the products which enter into the consumption of the worker, in short the value of the constant capital (see Ramsay [177]).

//The distinction between the re-appearance or simple preservation of the value in the product and the reproduction of that value is as follows: In the latter case a new equivalent replaces the exchange value lost through the consumption of the use value in which it was contained. In the first case no new equivalent is put in the place of the original value. For example, the value of the wood which re-appears in the table is not replaced by a newly created equivalent. The value of the wood only re-appears in the table because the wood previously possessed value and the production of its value is a prerequisite for the production of the table’s value.//

But, secondly: Take the worker in the branch of labour in which he himself works. If, owing to a rise in the productive power of labour, a worker in a weaving-mill produces 20 yards of calico in one hour, whereas he previously produced only 1 yard, the 20 yards, after deduction of the increased amount of constant capital contained in them, therefore in so far as they are, in general, value created by the worker himself, [III-127] possess no more value than the 1 yard, did previously. If the productive power of labour remains the same in all the other branches as it was before this transformation in the weaving trade, the worker would be unable to buy more of the means of subsistence than before with his 1 hour, despite the heightened productive power of his labour — i.e. he could only buy commodities in which 1 hour of labour is objectified, just as before. The growth of productive power in his own branch of labour, the increased productivity of his own labour, would therefore only cheapen the reproduction of his own labour capacity and hence only diminish his necessary labour time, in so far as and to the extent to which calico enters into his own consumption as, say, an element in his clothing. Only in this proportion. But this is true of every specific branch of production, hence of every individual capital, taken for itself, in the sphere of its own industrial functioning.

If we take the total capital of society, hence the whole capitalist class vis-à-vis the working class, it is clear that the capitalist class can only increase surplus value without extending the overall working day and without lessening the normal wage in so far as a greater productivity of labour, a higher development of the productive power of labour, makes it possible to maintain the working class as a whole with less labour, to produce the total amount of its means of subsistence more cheaply, and therefore to reduce the amount of labour time in total that the working class requires for the reproduction of its own wages. But this total amount consists simply of the total amount of individual means of subsistence and the total amount of the specific branches of labour; hence of the total amount of the individual branches of labour which produce these means of subsistence; hence of the total amount of the reductions in labour time on account of the increased productive power of labour in each of these individual branches. For the purpose of generalising the presentation, however, we are justified in viewing the process as if the worker lived from the use values he produces himself — and we can only view the process by picturing a particular individual capital with particular workers in a particular sphere. (It is not assumed here that the worker’s need for necessary labour time declines in the same measure as the amount of product he provides in the same period increases, but that his own product, which has now become cheaper, enters into his own consumption in proportion as his necessary labour time declines. This is valid for the whole of society, hence for the sum total of all the individuals, since the social sum of relative surplus labour is nothing but the sum of all the surplus labours of the individual workers in the individual branches of labour. It is only that compensations and adjustments come into the picture, the consideration of which does not belong here, although they hide the true relation.

A reduction in necessary labour time is therefore an increase in surplus labour time. The one grows smaller in proportion as the other grows larger, and inversely. This rise and fall does not, however, affect the overall working day and its magnitude.) The worker himself can in fact only create relative surplus value to the extent that he creates it in the sphere of his own activity, i.e. produces products entering into his own consumption in less time than previously. The political economists therefore always take flight to this assumption, in so far as they go into the nature of relative value at all (see Mill[178]).

In fact, one looks at the usual course of events. If the working day was = 12 hours, surplus labour time = 2 hours, and the capitalist, in consequence of increased productivity of labour, produces e.g. twice as much. Then surplus value can grow — his gain can emerge — only from two sources. Either the product of labour enters in a certain proportion into the reproduction of labour capacity, and labour capacity is cheapened in this proportion, so that the wage, i.e. the value of labour capacity falls in this ratio, hence there is also a fall in the part of the total working day required until then for the reproduction of this part of the value of labour capacity. Or the manufacturer sells the commodity above its value, i.e. as if the productivity of labour had remained the same. Only in the proportion to which he sells it above its value, hence buys all other commodities below their value, cheaper than the ratio between the amount of labour time contained in them and that contained in his commodity would require, does he posit a new surplus value. The worker, however, only receives the same normal wage as before. He therefore obtains a smaller part of the total value of the product, or a smaller part of that value is expended in the purchase of labour capacity than before the increase in the productivity of labour. A smaller part of his whole day is therefore expended in reproducing his wages, a larger part for the capitalist. It is the same thing, in practice, as if his cost of upkeep were lessened as a result of the increased productivity of his labour, or as if he could buy all other means of subsistence cheaper as a result of the greater productivity of his labour in the same proportion as the capitalist receives new value.

[III-128] In any case, we do not need to repeat here that the general presupposition of a sale of commodities above their value negates itself, and competition in fact compensates for sale above the value by sale below the value. What is involved here is the case where an increase in the productivity of labour has not yet become universal in the same branch of business, where the capitalist therefore sells (in a certain proportion at least, for he will always sell cheaper than the other) as if more labour time had been needed for the manufacture of his product than was really necessary. He sells e.g. the product of 3/4 of an hour as the product of 1 hour, because the majority of his competitors still need 1 hour to manufacture this product. If the total working day was 12 hours, he sells, in the given case, as if it had been 15 hours. (12/4 = 3, 12 + 3 = 15.) He has not lengthened the working day. If necessary labour time was = 10, and surplus labour time was = 2, it is still = 2. Actually, however, he sells as if necessary labour time were only 7 hours and surplus labour 5 hours (7 + 5 = 12). In fact, the labour of his own workers relates to that of average workers in such a way that they buy with the value of 7 hours as much as the latter buy with the value of 10 (since value has not fallen in proportion to productivity). With the original ratio, he had to give the workers 10 hours out of the 12, i.e. 5/6 (12/6 = 2; 5× 12/6 = 10).

Now, in consequence of the rise in the productivity of labour, he sells 12 at 15. If an hour of labour were paid for as an hour that stands 1/4 above average labour, the workers, instead of 10 hours, would only have to work 10 — 10/4. If the necessary working day until then was 10 hours, and 2 hours of surplus labour, the workers would now only need to work 10×3/4 hours instead of 10×4/4 hours (since their labour would count for 1/4 more than the average hour of labour), hence they would need to work 71/2 hours instead of 10, and the surplus value, just as before, would come to 1/5 of the necessary labour time (10/5 = 2). It would now be 1/5 of 7 1/2 hours, or 15/2 hours. 1/5 of 15/2 = 15/10 = 1 5/10 = 1 1/2, or 3/2, or 6/4. In fact if 3/4 of an hour of this labour = 1, or 4 /4 of an hour of average labour, 6/4 of the same labour = 8/4, or 2 hours of labour. The working day would thereby be reduced to 7 1/2 + 3/2 = 9 hours. The capitalist has the workers work 12 hours, as before, pays for the necessary labour time with 7 1/2, and therefore pockets 4 1/2 hours.

His gain derives from the fact that the necessary labour time of 10 hours has fallen to 7 1/2, or the worker can buy all his necessary means of subsistence with the product of 7 1/2 hours. It is exactly the same as if he were to produce the whole of his means of subsistence himself and were to be able to produce as much in 3/4 of an hour as previously in 1 hour, owing to the higher productivity of his labour, hence producing in 7 1/2 hours as much as he previously produced in 10. If with the increased productivity of labour the proportion had remained the same, the overall Working day would have become shorter, because necessary labour would have been lessened while the proportion between necessary labour and surplus labour would have remained the same. In practice it comes to exactly the same thing: Whether the value of labour capacity and therefore the necessary labour time is lessened, because the worker’s product enters into his own consumption in a certain proportion and therefore the necessary labour time declines and the surplus labour time, hence also surplus value, increases in this proportion; or whether as a result of the increased productivity of labour this particular branch of labour rises above the level of the socially average worker in the same branch, therefore the value e.g. of the hour of labour rises relative to all other commodities, the capitalist pays this labour as average labour — according to the previous standard — but sells it as labour of a higher than average level. In both cases a smaller number of hours is sufficient to pay the wage, i.e. the [III-129] necessary labour time has been reduced and in both cases the relative surplus value, i.e. surplus value not attained through an absolute prolongation of the working day, results from the decline in the amount of labour time required for the reproduction of the wage consequent on the increased productivity of labour; relative surplus value results in the one case directly, because the worker produces the same quantity of use values in a lesser labour time, although the product continues to be sold at its value. In the other case it results because a smaller quantity of labour time is equated with a greater quantity of average labour time as a result of the rise in productivity, and the worker therefore receives the same quantity of use values with labour time which is smaller but sold at a higher price. In both cases the relative surplus value results from a reduction in the necessary labour time.

The following is in any case clear in itself: If the productivity of labour grows, and the ratio remains the same, the worker would either have to work less labour time to reproduce his wages, say 7 1/2 instead of 10 hours, thereby shortening the working day as a whole; or he would have to receive a greater quantity of means of subsistence, his wages would rise above the [average] level. If neither the one thing nor the other takes place, it is clear that the result of the increased productivity of labour has only been to increase the amount of labour he performs for the capitalist, and to reduce the amount of labour he performs for himself.

The whole difficulty arises from this, that when the individual capitalist raises the productivity of labour, he is not thinking directly of a diminution of necessary labour time, but of its sale above its value — of raising it above average labour time. Of this raised labour time, however, a smaller proportion is needed for the replacement of wages; i.e. the surplus labour time grows, although this growth presents itself in a roundabout way, through sale above value.

The working day as a whole does not grow along with the growth in relative surplus value, hence relative labour time. It therefore follows that there is only a fall in the proportion in which the worker participates in his own working day. There is a fall in relative wages, or the weight of capital rises in relation to labour.

Further: As a result of the growth in the productivity of labour the quantity of products is increased. The same value is present in their total amount (e.g. the total for one working day) as was present previously in a smaller total. The individual product or the individual commodity therefore falls in value, but it is multiplied by a larger factor, which is indicated by the number of products. 6×4 is not more than 12×2. Here, then, we have a growth in the real wealth of use values, without any growth in their exchange value or the labour time contained in them, whereas in the first case — absolute surplus value — the amount of products also grows, but simultaneously with their exchange value, i.e. in proportion to the labour time contained in them.

This is to be understood as follows. If 10 [lbs] of cotton are converted into twist in the same time as previously 1 lb, the 10 lbs have not absorbed more spinning labour than the previous 1 lb. The value added to the 10 lbs is no greater than the value of the 1 ]b. Every 1 lb of twist contains ten times less spinning labour in the first case than in the second. And since they both contain the same amount of cotton, every 1 lb of twist, caeteris paribus, is 1/10 cheaper if the spinning labour amounts to 1/10 of the value. [III-130] If the added day of spinning labour = 10, and the value of 1 lb of cotton = 20 (for the sake of simplicity the instrument is set = 0 in both cases), in the first case 1 lb of twist = 10 + 20 = 30; in the second case 10 lbs of twist = 100 + 10 = 110, making 1 lb of twist = 11, and 10 lbs = 110, whereas in the first case 10 lbs = 300.

Relative surplus value is therefore distinguished from absolute as follows. In both, surplus value = surplus labour, or, the ratio of surplus value is equal to the ratio of surplus labour time to necessary labour time. In the first case the working day is extended beyond its limits and the surplus value grows (or the surplus labour time grows) in proportion as the working day is extended beyond its limit. In the second case the working day is given. Here, the surplus value, or the surplus labour time, is increased owing to the reduction of the portion of the working day that was required, or was necessary, for the reproduction of the wage. In the first case a given level of the productivity of labour is presupposed. In the second case the productivity of labour is raised. In the first case the value of an aliquot part of the total product or a part of the product of the working day remains unchanged; in the second the value of the part of the product changes, but its quantity (number of articles) grows in the same proportion as its value diminishes. The value of the total amount thus remains unchanged, whilst the total amount of products or use values has increased. Further the matter is to be presented simply as follows:

As we saw in our analysis of the commodity[10], the productivity of labour does not increase the value of the product or the commodity in which the labour manifests itself. If we presuppose that the labour time contained in the commodities is, under the given conditions, necessary labour time, socially necessary labour time[70] — and this is always the presupposition we start from once the value of a commodity is reduced to the labour time contained in it — what takes place is rather the following: The value of the product of labour is in an inverse ratio to the productivity of labour. This is in fact an identical proposition. It means nothing more than this: If labour becomes more productive, it can represent a greater quantity of the same use values in the same period, it can embody itself in a greater amount of use values of the same kind. Accordingly, an aliquot part of these use values, e.g. a yard of linen, contains less labour time than previously, has therefore less exchange value and indeed the exchange value of the yard of linen has fallen in the same proportion as the productivity of the labour of weaving has grown. Inversely, if more labour time than previously were required to produce a yard of linen (let us say, because more labour time was required to produce a pound of flax), the yard of linen would now contain more labour time, hence would have a higher exchange value. Its exchange value would have increased in the same proportion as the labour required to produce it had become less productive.

If we therefore take the whole working day — the average normal working day — the value of the sum total of its products remains unchanged, whether the labour becomes more or less productive. For the sum total of use values produced comes to one working day, just as before, it continues to represent the same quantity of socially necessary labour time. If, on the other hand, we take an aliquot part of the daily overall production, or a part of the product, its value rises and falls in inverse ratio to the productivity of the labour contained in it. For example, if 1 quarter or 8 bushels are the product of a month’s labour, let agriculture double its productivity in one case, and halve its productivity in another case. We should then have 3 cases: 8 bushels the product of a month’s labour; 16 bushels the product of the same labour time; 4 bushels the product of the same labour time. The value of the total amount produced during the month, 8, 16 or 4 bushels, would continue to contain respectively the same quantity of necessary labour time. The value of this total amount would therefore have remained unchanged, although in one case the productivity of labour would have doubled, in the other case declined to half its original level. But in the first case 1 bushel would contain 1/8 of a month = 2/16, in the second case 1/4 or 1/8 = 4/16, and in the third case only 1/16. With the productivity of agriculture doubled, the value of the bushels fell by a half; with productivity declining to half its original level, the value doubled. The value of a commodity can therefore never increase as a result of [increases in] the productivity of labour. This would involve a contradiction. Growth in the productivity of labour means that it brings forth the same product (use value) in less time. Growth in the exchange value of the product means that it contains more labour time than previously.

If, therefore, the value of the individual commodity stands in an inverse ratio to the productivity of labour, whilst the value of the total amount of products in which a given labour time is embodied remains untouched, unchanged, by any variation in the productivity of labour, the surplus value in contrast depends on the productivity of labour: if, on the one hand, the commodity is sold at its value, and on the other hand the length of the normal working day is given, the surplus value can only increase as a result of a rise in the productivity of labour. The surplus value is not related to the commodity; it expresses rather a relation between two parts of the overall working day — a relation namely between the part during which the worker works to replace his wage (the value of his labour capacity) and the part during which he works for the capitalist over and above this replacement. Since these two parts together make up the whole of the working day, since they are parts of the same whole, their magnitudes clearly stand in inverse ratio to each other, and the surplus value, i.e. the surplus labour time, rises or falls according to whether the necessary labour time falls or rises. The growth or diminution of the latter, however, stands in inverse ratio to the productivity of labour.

[III-131] But if there were to be a general doubling of the productivity of labour, i.e. in all branches of industry providing directly or indirectly the commodities (use values) required for the reproduction of labour capacity, providing products which enter into the consumption of the workers, the value of labour capacity would fall in proportion as this general productivity of labour uniformly increased, hence the labour time necessary for the replacement of this value would fall, and the part of the day which forms surplus time, which is worked for the capitalist, would increase in the same proportion as the former would decline. However, the development of the productive forces in these different branches of labour is neither uniform nor simultaneous, being subject to uneven, diverse and often mutually opposed motions. If the productivity of labour increases in a branch of industry which enters directly or indirectly into the worker’s consumption, e.g. the industry supplying fabrics for clothes, we cannot say that the value of labour capacity falls in the same proportion as the productivity of this particular industry grows. Ii is only this means of subsistence that is produced more cheaply. The cheapening only affects an aliquot part of the worker’s vital requirements. The increased productivity of labour in this one branch does not lessen the necessary labour time (i.e. the labour time required for the production of the means of subsistence needed by the workers) in proportion to the growth in productivity, but only in proportion as the product of this labour enters, on the average, into the worker’s consumption. No definite calculation of this can be made for each individual branch of industry (excepting perhaps the products of agriculture).

This does not change the general law in any way. It remains correct, just as before, that relative surplus value can only arise and increase in the proportion to which use values (means of subsistence) directly or indirectly entering into the worker’s consumption are cheapened, i.e. not in the proportion to which the productivity of a specific branch of industry has grown, but rather in the proportion to which this increase in its productivity lessens necessary labour time, i.e. produces more cheaply a product which enters into the worker’s consumption. In considering relative surplus value therefore, we not only can but we must always proceed from the presupposition that the development of productive power or the development of the productivity of labour in every particular branch in which capital investment takes place directly reduces the necessary labour time in a definite proportion, i.e. that the product produced by the worker forms a part of his means of subsistence and its cheapening therefore reduces the labour time required for the reproduction of his life in a definite proportion. Since relative surplus value arises only on this condition, we can and must always assume the presence of this condition in considering relative surplus value.

It is clear, further, that the presence and the growth of relative surplus value by no means require as a condition that the worker’s life situation should remain unchanged, i.e. that his average wage should always provide the same quantitatively and qualitatively determined amount of means of subsistence and no more. This is not the case, although relative surplus value can neither arise nor grow without a corresponding fall in the value of labour capacity or the value of wages (average wages). Indeed, relative surplus value might well rise continuously, and the value of labour capacity, hence the value of average wages, fall continuously, yet despite this the range of the worker’s means of subsistence and therefore the pleasures of his life could expand continuously. For this is conditioned by the quality and quantity of the use values (commodities) he can appropriate, not by their exchange value.

Let us assume a doubling of productivity which is universal, covering all branches of production.[179] Assume that before this doubling the normal day was 12 hours, 10 of them necessary labour time, 2 surplus labour time. The total amount of the worker’s daily means of subsistence, which previously cost [10] hours of labour, could now be produced in 5 hours. Instead of needing 10 hours of labour to replace the value (price) of his labour capacity every day, i.e. to provide an equivalent for his daily wages, the worker would now need only [5] hours. The value of his labour capacity would have fallen by a half, for the means of subsistence required for its reproduction would now be the product of 5 hours; instead of 10 as before. If the worker now — after this revolution in the productivity of labour — received a daily wage equivalent to 6 hours, that is to say, if he had in future to work 6 hours [IV-138] [180] a day, his material living situation would have improved in the same proportion as if, under the previous conditions of production, he had worked the whole day of 12 hours for himself (i.e. for the reproduction of his wages) and a labour time of 0 for the capitalist; as if the whole of the working day necessary labour time had been worked, and no surplus labour time at all. For 5:6 = 10:12. (5×12 = 6×10.) Nevertheless, surplus labour time would have increased in this case from 2 hours to 6 hours, and a relative surplus value of 4 hours would have been added to the absolute surplus value of 2 hours. Instead of working as before 10 hours for himself and 2 for the capitalist, hence 10/12 ( = 5 /6), therefore 5/6 of the day, for himself and 1/12 = 1/6 of the day for the capitalist, the worker now works only 6/ 12 or 1/c, of the day for himself and, instead of 1/6 he also works 3 /6, half the day, for the capitalist. Necessary labour time would have fallen from 10 to 6, hence the value of the day’s labour capacity, instead of being 10 hours, would only be 6 hours: 4 hours less, i.e. it would have fallen by 40% (10:4 = 100:40). Surplus value would have increased to 300%, from 2 to 6. //Instead of 1/6 of the day 3/6. 2 2/6 added to 1/6 gives 3/6, therefore a 200% increase. This for the surplus value. On the other hand, from 5 /6 down to 3/6 is a reduction of 2/6, i.e. the increase on the surplus labour [time] side or the side of the capitalist is exactly as much, viewed absolutely, as the reduction on the necessary labour time side or the value of labour capacity side. It amounts to 2 /6 of a day, or 4 hours of labour. (2 /6 = 4/ 12.) But if we look at the increase on one side in proportion to the original surplus labour time, and the decrease on the other side in proportion to the original necessary labour time (or the value of labour capacity), the increase on one side and the decrease on the other are expressed in different proportions, although the absolute magnitude of the time subtracted from one side and added to the other is the same identical magnitude.

Thus in the above case: 10/12 or 5/6 are related to 6/12 or 3/6 or (5-2)/6 as 5:3, as 60% (should be 40%, see the other page[181]), for 5:3 = 100:60 (5×60 = 300 and 3×100 similarly = 300), while 2/12 or 1/6 is related to 6/12 or to (1 + 2)/6 (3 /6) as 1:3, i.e. 100:300, hence as 300%. Therefore, although the absolute increase in surplus labour [time] = the absolute decrease in necessary labour time, which has occurred as a result of the raised productivity of labour, the proportion in which the value of labour capacity declines or the necessary labour time falls is not identical with the proportion in which the surplus labour time or the surplus value rises, but depends rather on the original proportion in which surplus labour time and necessary labour time shared in the normal overall working day, participated in it.//

//It follows from this that in the proportion in which total surplus labour time (both the part of it which arose from the reduction of necessary labour time consequent on the increase in the productivity of labour, and the part which arose from the lengthening of the working day up to its normal limits) already forms a greater part (a more significant portion) of the overall working day, any increase in the productive power of labour and resultant reduction in necessary labour time (or increase in relative surplus value) can only increase the proportional surplus value in a smaller ratio. Or that a reduction of necessary labour time causes an increase in surplus labour time in a proportion which is smaller, the greater the already achieved total magnitude of surplus labour time, and greater, the smaller the achieved total magnitude of the surplus labour time. Therefore (and this must be dealt with in more detail under profit[182]) the more advanced the industry, the smaller the proportional growth of surplus value, if productive power continues to increase in the same degree. Productive power in general, or productive power altogether, to the extent that it influences the reproduction of labour capacity. In other words, the proportion in which an increase in the [IV-139] productive power of labour reduces necessary labour time (hence the value of labour capacity) and raises surplus labour time, hence surplus value, stands in an inverse relation to the proportion in which necessary labour time and surplus labour time originally, i.e. each time before the coming of the new increase in productive power, shared in, or participated in, the overall working day.

Assume that the working day = 12 hours, 10 hours of necessary labour, and 2 hours of surplus labour. Let there be a general doubling of productive power. Now 5 hours would suffice for necessary labour time, surplus labour time would be increased by 5 hours, by the same amount as the decrease in necessary labour time (hence in the value of labour capacity) — i.e. 5 hours. Necessary labour time would have declined from 10 to 5, i.e. by half = 50%. //(If the necessary labour time were to decline from 10 to 6, this would be a reduction of 4 hours. 10:4 = 100:40, therefore by 40%. 1 said 60% before.[181] This is wrong, because I calculated 10:6 = 100:60, whereas in fact what we are concerned with is the ratio of 10 to the remainder of the 10 when 6 is taken away from it, hence the ratio of 10 to 4. After all, the labour time has not been reduced by 6 hours, i.e. not by 60%.) On the other hand surplus labour time has risen from 2 to 7 hours (with the addition of 5 hours of surplus labour time), and 2:7 = 100: 350 (2 × 350 = 700 and 7 × 100 also = 700); hence a rise to 3 5 0 %. I t would have increased to three and a half times its original magnitude.

Let us now. assume that once this proportion has become established, with the overall working day falling into 5 hours of necessary labour, 7 hours of surplus labour, the general productive power of labour is redoubled, i.e. necessary labour time diminishes by 21/2 hours, surplus labour time therefore rising by the same 2 1/2 hours; hence from 7 to 9 1/2 hours. Here the necessary labour time has again fallen by 50%, and surplus labour time risen in the ratio 14 /2 (7) to 19/2 (9 1/2), thus 14:19. 14:19 = 100:x; x = 1900/14, = 135 5/7%. (19×100 = 1,900 and 14×135 5/7 (or 135 10/14) also = 1,900). Therefore, although in both cases the productive power of labour has doubled and the necessary labour time has therefore fallen by a half, by 50%, surplus labour time or surplus value would have risen to 350% in one case and only to 135 5/7% in the other. (The proportion in which the productive power generally increases would always be the same, = the proportion in which the necessary labour time fell as compared with itself, i.e. with its extent before this increase in productive power.) But in the first case the surplus labour time amounted to only 1/6 of the whole working day, 2 hours, = 2/12, before the doubling of the productive power took effect, while in the second case it amounted to 7 hours or 7/12. (The same peculiarity holds for the increase of money, as has been demonstrated by Jacob, for example. It grew more in the 18th century than in the 17th. But the proportional increase was smaller.)

[IV-140] If we now take an actual case, in which productive power is e.g. doubled in one branch, but not in the other branches at the same time, perhaps remaining unaltered in the branches of production which provide constant capital for this one branch, so that the expenditure on raw materials remains the same, i.e. grows along with the increase in productive power, and the expenditure on machinery increases, even if not in the same ratio, it is clear that the profit, i.e. the ratio of the surplus value to the total value of the capital expended, does not increase in the same proportion as the necessary labour falls through the increase in productive power. There are two reasons why this does not happen. Firstly because with the more developed productive power of labour, surplus value does not grow in the same proportion as necessary labour diminishes. Secondly because this surplus value, which has grown in a smaller proportion, is calculated on capital which has increased its value approximately in proportion to the heightening of productive power.//

//One can calculate the diminution in necessary labour time in two ways: 1) in proportion to its own magnitude before the increase in the productive power of labour; 2) in proportion to the whole of the working day. It is clear in the first calculation that — presupposing an overall heightening of productive power — necessary labour time (and therefore the value of labour capacity) declines in the same measure as productive power increases; but the proportional growth of surplus labour time or surplus value depends on the proportion in which the overall working day was originally divided between necessary labour time and surplus labour time. Thus if the working day was 12 hours originally, divided into 10 of necessary and 2 of surplus labour, and if the productive power of labour doubled, necessary labour time would fall from 10 to 5, i.e. by 50%, while productive power doubled. (This proportion is expressed in the case of productive power by a 100% growth, in the case of necessary labour time by a 50% fall. That necessary labour time falls from 10 to 5, i.e. by 50%, means that in 1 hour I can produce as much as I did previously in 2, i.e. twice as much, i.e. the productive power of labour has increased by 100%.) Surplus labour, on the other hand, has grown from 2 to 7, i.e. to 350% (a threefold increase, 2×3, or [6] hours, and a rise by a half, = 2/2 = 1, thus the whole has gone up from 2 to 7), because it originally only amounted to 2 hours out of 12. If it had originally already amounted to 3 hours, and necessary labour only 9 hours, the latter would have fallen by 4 1/2 hours, again by 50%, while surplus labour would have risen from 3 to 7 1/2; i.e. to 250% (for 3:7 %, or 6/2:15/2, or 6:15, = 100:250. 15×100 = 1500 and 6×250 = 1500). If we now consider the whole of the working day, the ratio is not altered. [Necessary] labour time originally amounted to 10 hours or 10/12 of the working day; now it only amounts to 5/12 in the first case. (In the second it was originally 9/12 of the working day, and afterwards came to no more than 4 1/2/12.) It is all the same whether I compare necessary labour time with itself or with the working day as a whole. All that is added is the divisor 12. This Fix has therefore been dealt with. //

Now back to page 138 before the bracket. The worker’s life situation would have improved despite the fall in the value of his labour capacity, the reduction by 4 hours in his necessary labour time and the increase of 4 hours in his surplus labour time for the capitalist, because he himself would have received a share of 1 hour in the time now set free. I.e., the labour time he worked for himself, i.e. for the reproduction of his wages, would not have been reduced to the full extent of the shortening of this necessary labour time resulting from the [increased] product of the labour. He would receive more use values of less value — i.e. containing less labour time than previously. But the degree in which new surplus labour would have been formed in general, in which relative surplus value would have arisen, would correspond completely to the degree in which a part of his necessary labour time had been converted into surplus labour time for the capitalist, or in which the value of his labour capacity had declined. This is enough here. Later on the proportional elements in the matter must in general be put together [183] (see also above) . Thus this in no way alters the nature and the law of relative surplus value — that a greater part of the working day is appropriated by capital as a result of rises in productivity. Hence the preposterousness of wanting to refute this law by statistical demonstrations that the material condition of the worker has improved here or there, in this or that aspect, [IV-141] as a result of the development of the productive power of labour.

//Standard. October 26, 1861. We read here of proceedings taken by the firm of John Bright and Co. against its workers,, before the Rochdale magistrates,

*to prosecute for intimidation the agents of the Carpets’ Weavers’ Trades Unions. Bright’s partners had introduced new machinery which would turn out 240 yards of carpet in the time and with the labour previously required to produce 160 yards. The workmen had no claim whatever to share in the profits made by the investment of their employers’ capital in mechanical improvement. Accordingly, Messrs. Bright proposed to lower the rate of pay from 1 1/2d. per yard to 1d., leaving the earnings of the men exactly the same as before for the same labour. But there was a nominal reduction, of which the operatives, it is asserted, had not had fair warning beforehand.*[184][IV-141]

* * *

[IV-138a] [185] 1) The surplus value capital receives through the development of the productive forces does not flow from an increase in the amount of products or use values created by the same amount of labour, but from a reduction in necessary labour and an increase of the same proportion in surplus labour. The surplus value capital receives through the production process consists in nothing more than the excess of surplus labour over necessary labour.

Surplus value exactly equals surplus labour; an increase in surplus labour is exactly measured by a reduction in necessary labour. With absolute surplus value the reduction in necessary labour is relative, i.e. necessary labour falls relatively because overtime is increased directly. If the necessary labour = 10 hours, and the surplus labour = 2 hours, and if the latter is now increased by 2 hours, i.e. the total working day is lengthened from 12 hours to 14, the necessary labour remains 10 hours, as it was before. But previously its ratio to the surplus labour was 10:2, i.e. 5:1, and now it is 10:4, = 5:2, or, in other words, previously it was equal to 5/6 of the working day, now it is only 5/7. Here, therefore, necessary labour time is reduced relatively, because total labour time, and therefore surplus labour time, has grown absolutely. In contrast to this, if the normal working day is given, and the increase in relative surplus value occurs through an increase in productive forces, the necessary labour time is lessened absolutely and the surplus value is thereby increased both absolutely and relatively without any increase in the value of the product. In the case of absolute surplus value, therefore, there is a relative fall in the value of wages as compared with the absolute growth in surplus value; whereas in the case of relative surplus value there is an absolute fall in the value of wages. Nevertheless, the first case is always worse for the worker. In the first case the price of labour falls absolutely. In the second case the price of labour may rise.

2) The surplus value of the capital is increased not by the multiplier of the productive power but by the fraction of the working day which originally represented the necessary labour time, divided by the multiplier of the productive power.

3) The greater the surplus value prior to the new increase in productive power, i.e. the greater the part of the day worked for no return already is, and the smaller therefore the paid part of the day, the fraction of the day which forms the worker’s equivalent, the less the growth in surplus value which capital obtains from the new increase in productive power. Its surplus value rises, but in an ever smaller proportion to the development of the productive forces. The barrier remains the ratio between the fraction of the day which expresses necessary labour and the whole of the working day. Movement can only take place within these limits. The smaller the fraction allotted to necessary labour, and the greater therefore the surplus labour, the smaller the ratio in which an increase in productive power lessens necessary labour time, since the denominator of the fraction is that much larger. The rate of capital’s self-valorisation therefore grows more slowly in the measure to which it is already valorised. This does not, however, happen because the wage or the worker’s share in the product has risen but because the fraction of the working day which represents necessary labour has already fallen very low in proportion to the working day as a whole. [IV-138a]


[IV-141] A certain development of the productivity of labour is in general presupposed, even for the existence of absolute surplus value, i.e. surplus labour in general, and therefore for the existence of capitalist production, as for all earlier modes of production in which one part of society works not only for itself but also for the other part of society.

*"The very existence of the former (the master-capitalists) as a distinct class is dependent on the productiveness of industry"* (Ramsay, An Essay on the Distribution of Wealth etc., Edinburgh, [London,] 1836, [p.] 206).

* “If each man’s labour were but enough to produce a his own food, there could be no property” * (this word is used here for capital) (Piercy Ravenstone, Thoughts on the Funding System, and its Effects, London, 1824, p. 14).

In any case, the capital-relation develops at a historical stage of the economic formation of society which is already the result of a long series of previous developments. The level of the productivity of labour from which it proceeds is not of natural origin but is something created historically; by that time labour has long emerged from its first raw beginnings. It is clear that if a country possesses soil that is naturally fertile, waters teeming with fish, rich coal deposits (combustible materials in general), metal mines, etc., in comparison with other countries, where these natural conditions for the productivity of labour are present to a lesser degree, less time is required in the first country to produce the necessary means of subsistence, hence a greater quantity of excess labour for others over and above labour for oneself is possible from the outset; and therefore absolute surplus labour time, thus absolute surplus value, is greater here from the outset. Capital (or any other relation of production whereby surplus labour is enforced) is therefore more productive here than under less favourable natural conditions.

The ancients were already aware that natural cheapness of labour capacity, i.e. of its cost of production and reproduction, was a great factor in industrial production. For example, it says in Diodorus’ Historical Library, b. 1, ch. 80, in relation to the Egyptians:

“It is altogether incredible how little trouble and expense the bringing up of their children causes them. They cook for them the first simple food that comes to hand; they also give them the lower part of the papyrus stem to eat, if it can be roasted in the fire, and the roots and stalks of marsh plants, some raw, some boiled, and some roasted. Most of the children go without shoes and unclothed, since the air is so mild. Hence a child, until he is grown up, costs his parents no more than twenty drachmas altogether. This is the main reason why the population of Egypt is so numerous, and, therefore, why it has been possible to undertake so many great works.”

// Once the ratio of surplus value is given, its amount depends on the size of the population; if the size of the population is given, it depends on the ratio of surplus to necessary labour. //

All that follows from this is that, in places where the capital-relation predominates (or a similar relation of production which enforces absolute surplus labour, for this natural fertility only facilitates the prolongation of surplus labour time and its existence; it does not create relative surplus value in our sense), the productivity of capital is at its greatest — i.e. the most surplus labour is available and therefore the most surplus value, or the value of labour capacity is naturally at its lowest, which is the same thing — where the natural conditions of labour, [IV-142] hence in particular the soil, are at their most fruitful. This by no means implies that the most fertile countries are the most well suited to the development, thus also the fruitfulness, of the capital-relation itself. When Ricardo speaks of the fertility of the soil as one of the main conditions for the productivity of labour, he assumes capitalist production, and the proposition is only uttered on this assumption. He is naturally inclined everywhere to presuppose bourgeois relations of production as given. This does not interfere with the development of his argument as he deals exclusively with production in this particular form. The following passage is important, both for the concept of surplus labour in general and for the misunderstanding about the point we have just touched on.

“In different stages of society, the accumulation of capital, or of the means of employing labour* is more or less rapid, and must in all cases depend on the productive powers of labour. The productive powers of labour are generally greatest, where there is an abundance of fertile land (Ricardo).

“If, in the first sentence, the productive powers of labour mean the smallness of that aliquot part of any produce that goes to those whose manual labour produced it, the sentence is nearly identical, because the remaining aliquot part is the fund whence capital can, if the owner pleases, be accumulated. But then this does not generally happen where there is most fertile land. It does in North America, but that is an artificial state of things. It does not in Mexico. It does not in New Holland. The productive powers of labour are, indeed, in another sense, greatest where there is much fertile land, viz. the power of man, if he chooses it, to raise much raw produce in proportion to the whole labour he performs. It is, indeed, a gift of nature, that men can raise more food than the lowest quantity that they could maintain and keep up the existing population on; but ‘surplus produce’ (the term used by Mr. Ricardo p. 93) generally means the excess of the whole price of a thing above that part of it which goes to the labourers who made it; a part, which is settled by human arrangement, and not fixed” (Observations on Certain Verbal Disputes in Political Economy, Particularly Relating to Value, and to Demand and Supply, London, 1821, pp. 74-75).

* It is only in such passages as this that the nature of capital breaks through in Ricardo. So capital is not the means of labour for producing a certain result, but it is “the means for employing labour”, and this involves that the possessor of the means, or those means themselves, employs labour, the means are the power over labour.[186]

This man does not see that in fact “The smallnessor bignessof that aliquot part that goesto the labourer depends on the proportional quantity of raw produce which “the whole labour” of a man can perform daily. He is only right against Ricardo to the extent that he says: Natural fertility brings it about that with one day’s labour I could produce, if I chose, much more than what is absolutely necessary for existence (the lowest quantity to keep the existing population upon). It does not mean that I work a lot, hence produce a lot; still less that the work I do over and above what is necessary forms the fond of capital. This “is settled by human arrangement”. For Ricardo the capital-relation is itself a naturally given relation and is therefore presupposed everywhere.

If capitalist production is presupposed, necessary labour time, i.e. the time required for the reproduction of the worker, will differ in different countries according to how favourable the natural conditions of labour are, and therefore according to its natural level of productivity, and it will stand in inverse relation to the productivity of labour, hence it will be possible for surplus labour time or surplus value to be greater in one country than in the other, in direct relation, even if the same number of hours is worked.

All this concerns* the very existence of absolute surplus labour, and its relative quantity in different countries according to their respective natural facilities for production *. We do not have to deal with this here.

[IV-143] Since it is assumed that the normal working day is already divided into necessary labour and absolute surplus labour, the existence, and indeed a definite level, of the latter is presupposed, hence a definite natural basis for it is also presupposed. The question here is rather the productive power of labour — hence the shortening of necessary labour time, the prolongation of surplus labour time — in so far as it is itself the product of capitalist (in general, of social) production.

The chief forms are: Cooperation, Division of Labour, and Machinery or the application of scientific power, etc.