In essence, Marx published volume 1 of Capital in German in 1867, but only volume 1 of Capital was published in Marx’s lifetime. The other volumes were edited and published by Engels (for an extended discussion of this, see here). For some reason, Marx refused to publish volumes 2 and 3.
In volume 1, Marx set out a law of value based on the labour theory of value, in which socially necessary labour time was the anchor for the price system in modern capitalism. There is no convincing evidence that (1) Marx regarded this law of value as a totally abstract “simplifying assumption” or that (2) he did not mean to apply it to the advanced industrial capitalism of the 19th century as an empirical explanation of price determination.
But when volume 3 of Capital was edited and published by Engels critics quickly pointed out that the theory of value in volume 3 was radically inconsistent with that of volume 1.
This devastating problem clearly worried Engels, and this can be seen in an article Engels wrote in May 1895 for the Neue Zeit (Marx 1991: 1027, n.), which is available as the “Supplement and Addendum” to Volume 3 of Capital in Marx (1991: 1027–1047).
Right at the beginning of this supplement, Engels notes that people such as Achille Loria had pointed to the devastating contradiction between volume 1 and volume 3 of Capital (Marx 1991: 1027–1028).
Next, Engels mentions that Werner Sombart, in a review of Marx’s work (Sombart 1894), declared that the labour theory of value as presented in volume 1 of Capital could not be empirically supported and was a mere “logical” concept (Marx 1991: 1032).
So, too, Conrad Schmidt in an 1895 review of volume 3 (Schmidt 1895) had also declared that the labour theory of value was a “necessary fiction” (Marx 1991: 1032). Engels describes Schmidt’s criticisms:
“Schmidt, too, has his formal reservations about the law of value. He calls it a scientific hypothesis put forward to explain the actual exchange process, which proves the necessary theoretical point of departure, illuminating and indispensable even for the phenomena of prices under competition, which appear completely to contradict it. Without the law of value, in his opinion too, any theoretical insight into the economic mechanism of capitalist reality is impossible. In a personal letter which he has allowed me to mention, Schmidt declares that the law of value in the capitalist form of production is a fiction, though a theoretically necessary one.” (Marx 1991: 1032).Now it is clear that Engels’ “law of value” here is referring to the idea that commodities tend to exchange at their pure labour values.
Engels was well aware that hostile critics of Marx had declared that volume 3 of Capital utterly contradicted and overthrew the theory of value in volume 1. It seems that Conrad Schmidt was actually one of the first to point out the contradiction between commodities tending to exchange at their labour values and an average rate of profit in his 1889 work Die Durchschnittsprofitrate auf Grundlage des Marxschen Wertgesetzes [The Average Rate of Profit on the basis of Marx’s Law of Value] (Stuttgart, 1889) (see Böhm-Bawerk 1949: 28, with n. 2).
Engels desperately sought a solution and found a passage in volume 3 of Capital where Marx himself was trying to salvage the theory of value in volume 1, which had been overthrown by that in volume 3.
That passage of Marx comes in Chapter 10 of volume 3 and is as follows:
“The exchange of commodities at their values, or approximately at their values, requires, therefore, a much lower stage than their exchange at their prices of production, which requires a relatively high development of capitalist production.So here Marx was saying that the theory of value in volume 1 – that commodities tend to exchange at their pure labour values which are anchors for the price system – was a historically contingent phenomenon existing in the “lower stage … of capitalist production” and before the emergence of a higher stage of capitalism where Ricardo’s prices of production are the anchors for the price system.
Whatever may be the way in which the prices of the various commodities are first fixed or mutually regulated, the law of value always dominates their movements. If the labor time required for the production of these commodities is reduced, prices fall; if it is increased, prices rise, other circumstances remaining the same.
Aside from the fact that prices and their movements are dominated by the law of value, it is quite appropriate, under these circumstances, to regard the value of commodities not only theoretically, but also historically, as existing prior to the prices of production. This applies to conditions, in which the laborer owns his means of production, and this is the condition of the land-owning farmer and of the craftsman in the old world as well as the new. This agrees also with the view formerly expressed by me that the development of product into commodities arises through the exchange between different communes, not through that between the members of the same commune. It applies not only to this primitive condition, but also to subsequent conditions based on slavery or serfdom, and to the guild organisation of handicrafts, so long as the means of production installed in one line of production cannot be transferred to another line except under difficulties, so that the various lines of production maintain, to a certain degree, the same mutual relations as foreign countries or communistic groups.
In order that the prices at which commodities are exchanged with one another may correspond approximately to their values, no other conditions are required but the following: 1) The exchange of the various commodities must no longer be accidental or occasional, 2) So far as the direct exchange of commodities is concerned, these commodities must be produced on both sides in sufficient quantities to meet mutual requirements, a thing easily learned by experience in trading, and therefore a natural outgrowth of continued trading, 3) So far as selling is concerned, there must be no accidental or artificial monopoly which may enable either of the contracting sides to sell commodities above their value or compel others to sell below value. An accidental monopoly is one which a buyer or seller acquires by an accidental proportion of supply to demand.
The assumption that the commodities of the various spheres of production are sold at their value implies, of course, only that their value is the center of gravity around which prices fluctuate, and around which their rise and fall tends to an equilibrium.” (Marx 1909: 208–210).
It is particularly interesting to note how Marx specifically described the theory of value in volume 1 as follows:
“The assumption that the commodities of the various spheres of production are sold at their value implies, of course, only that their value is the center of gravity around which prices fluctuate, and around which their rise and fall tends to an equilibrium.” (Marx 1909: 208–210).This and Marx’s whole discussion around the passage clearly damn and refute all those pathetic Marxist hacks who want to tell us that the law of value in volume 1 – namely, that commodities tend to exchange at their pure labour values which are anchors for the price system – is only a “simplifying assumption” or some highly abstract system never intended to apply to the real world.
Clearly Marx did even in volume 3 of Capital apply it to the capitalist system in an empirical sense, but to those historical periods at a “lower stage … of capitalist production” confined to the older medieval and pre-modern eras. Crucially, this is exactly how Engels interpreted the passage, as we can see below in a quotation from Engels’ supplement to volume 3.
Engels cites the passage I have quoted above from volume 3 of Capital and says this:
“If Marx had been able to go through the third volume again, he would undoubtedly have elaborated this passage significantly. As it stands, it gives only an outline sketch of what needs to be said on the point in question. Let us therefore go into the matter somewhat more closely.The passage in yellow highlighting is crucial: this is how Engels understood the theory of value in volume 1 of Capital at the end of his life.
We all know that at the beginnings of society products are used by the producers themselves, these producers living in indigenous communities that are organized more or less on a communist basis; that the exchange of their surplus products with foreigners, which introduces the transformation of products into commodities, is of later date. It takes place first of all simply between individual communities of different tribes and only later does it come to prevail within the community, where it makes a decisive contribution to the dissolution of this community into larger or smaller family groups. Even after this dissolution, however, the family heads who exchange with one another remain working peasant farmers, who produce almost all their requirements on their own holdings, with the aid of their families, and obtain only a small portion of the items they need from outside, in exchange for their own surplus product. Not only does the family pursue agriculture and stock-raising, it also works up the products of these activities into finished articles of use, still doing its own milling in places with their hand mill, baking bread, spinning, dyeing, weaving flax and wool, curing leather, erecting and repairing wooden buildings, producing tools and equipment, and often doing its own carpentry and metalwork too; so that the family or family group is basically self-sufficient.
Now the little that such a family has to obtain from others by exchange, or buy, consisted right up to the early nineteenth century, in Germany, predominantly of objects of handicraft production, i.e. things whose mode of production was in no way strange to the peasant and which he himself failed to produce only because either the raw material was unavailable or the purchased article was much better or very much cheaper. For the peasant of the Middle Ages, therefore, the labour-time needed to reproduce the objects he obtained in exchange was quite accurately known. The village smith and cartwright were at work under his very eyes; similarly the tailor and shoemaker, who in my own youth still travelled round to our Rhineland peasants in turn, working up materials provided into clothes and shoes. Both the peasant and the people from whom he bought were workers themselves, and the articles exchanged were their own products. What had they applied in the production of these articles? Labour, and labour alone: to replace tools, to produce raw material and work it up, all they spent was their own labour-power; how else then could they exchange these products of theirs with those of other working producers than in proportion to the labour applied to them? The labour-time applied to these products, then, was more than just the most suitable measure for the quantitative determination of the magnitudes to be exchanged; no other measure was possible. Or are we to believe that peasant and village artisan were so stupid that one of them would part with the product of ten hours’ labour for that of a single hour? For the entire period of natural peasant economy, no other exchange is possible except that in which the amounts of commodities exchanged tend more and more to be measured according to the amounts of labour embodied in them. From the moment money penetrates into this economic mode, the tendency of adaptation to the law of value (Marx’s formulation, nota bene!) becomes more explicit, though it is already infringed by the interventions of usurer’s capital and fiscal extortion, so that the periods over which prices approximate on average to values, down to a negligible difference in magnitude, already become more drawn out.
The same applies to exchange between the products of peasants and those of urban artisans. At the beginning, this takes place directly, without the mediation of the merchant, on the town market-days when the peasant sells and makes his purchases. Here, too, the artisan’s conditions of labour are known to the peasant, and the peasant’s to the artisan. He is himself still one part peasant, and not only has his kitchen-garden and orchard but also very often a bit of a field, one or two cows, pigs, fowl, etc. People in the Middle Ages were thus in a position to reckon up each other’s production costs in raw and ancillary materials, and in labour-time, with a fair degree of accuracy – at least as far as articles of general daily use were concerned.
But how could the amount of labour be reckoned, even indirectly and relatively, when this served as the measure of exchange for products that required more prolonged labour, interrupted and at irregular intervals, and uncertain in its results, products like corn or cattle, for instance? And, moreover, with people who were unable to count? Evidently, only by a lengthy process of zig-zag approximation, often groping back and forth in the dark, in which, as in other things, wisdom was attained only by painful accident. But the need for each person to have a rough idea of his own costs helped time and again in the correct direction, and the small number of types of article coming into exchange, as well as the stable mode of their production, often over centuries, made the goal more easily attainable. That it in no way took so long until the relative values of these products were established with a fair degree of accuracy is shown by the simple fact that the commodity in which this seems most difficult on account of the long production time of the individual item, i.e. cattle, was the first fairly generally recognized money commodity. In order to arrive at the value of cattle, its exchange ratio with a whole series of other commodities must already have won established recognition to a relatively unusual degree, it must be unchallenged over an area of several tribes. And the people of that time were certainly clever enough – the cattle-breeders as well as their customers – not to part with the labour-time they had spent without an equivalent in exchange. On the contrary, the closer people stand to the original state of commodity production – e.g. Russians and Orientals – the more time they still spend today in extracting full compensation for the labour-time spent on a product by long and stubborn haggling.
Proceeding from this determination of value by labour-time, commodity production as a whole, and with it the manifold relationships in which the different aspects of the law of value make themselves felt, now develops as presented in Part One of Capital Volume 1; therefore, in particular, the conditions become established under which labour is value-forming. These conditions, moreover, prevail although those involved do not become aware of them, so that they can be abstracted from everyday practice only by tedious theoretical analysis; they operate in the form of a natural law, which as Marx showed followed necessarily from the nature of commodity production. The most important and incisive progress was the transition to metal money, but this had the consequence that the determination of value by labour-time was no longer visibly apparent on the surface of commodity exchange. Money became the decisive measure of value for practical purposes, and all the more so, the more diverse were the commodities coming into trade, the more they originated from distant countries, and the less therefore the labour-time needed for their production could be checked. Even the money itself came mostly from abroad at first; and when it was obtained in a particular country as precious metal, the peasant and artisan were in no position to assess even approximately the labour applied to it, while their own awareness of the value-measuring property of labour was also pretty well obscured by the custom of reckoning in money; money came to represent absolute value in the popular conception.
To sum up, Marx’s law of value applies universally, as much as any economic laws do apply, for the entire period of simple commodity production, i.e. up to the time at which this undergoes a modification by the onset of the capitalist form of production. Up till then, prices gravitate to the values determined by Marx’s law and oscillate around these values, so that the more completely simple commodity production develops, the more do average prices coincide with values for longer periods when not interrupted by external violent disturbances, and with the insignificant variations we mentioned earlier. Thus the Marxian law of value has a universal economic validity for an era lasting from the beginning of the exchange that transforms products into commodities down to the fifteenth century of our epoch. But commodity exchange dates from a time before any written history, going back to at least 3500 B.C. in Egypt, and 4000 B.C. or maybe even 6000 B.C. in Babylon; thus the law of value prevailed for a period of some five to seven millennia. We may now admire the profundity of Mr Loria in calling the value that was generally and directly prevalent throughout this time a value at which commodities never were sold nor could be sold, and which no economist will ever bother himself with if he has a glimmer of healthy common sense!” (Marx 1991: 1034–1038).
This view is that commodities did historically tend to exchange at pure labour values in less developed forms of capitalism up until about the 15th century. That is, it actually happened in the pre-modern “period of simple commodity production” (Marx 1991: 1037).
Then what happened was that the “transition to metal money” obscured exchange at pure labour values:
“The most important and incisive progress was the transition to metal money, but this had the consequence that the determination of value by labour-time was no longer visibly apparent on the surface of commodity exchange. Money became the decisive measure of value for practical purposes, and all the more so, the more diverse were the commodities coming into trade, the more they originated from distant countries, and the less therefore the labour-time needed for their production could be checked. Even the money itself came mostly from abroad at first; and when it was obtained in a particular country as precious metal, the peasant and artisan were in no position to assess even approximately the labour applied to it, while their own awareness of the value-measuring property of labour was also pretty well obscured by the custom of reckoning in money; money came to represent absolute value in the popular conception.” (Marx 1991: 1037).After this point, the advanced form of modern capitalist production developed and prices of production replaced labour values as the anchors for the price system.
This view of Engels is splendidly confirmed in a letter he wrote to Werner Sombart (1863–1941) on March 11, 1895 about the labour theory of value (on which, see here), which was a response to a hostile review of volume 3 of Capital by Sombart (1894).
The crucial passage from this letter of Engels is below:
“When commodity exchange began, when products gradually turned into commodities, they were exchanged approximately according to their value. It was the amount of labour expended on two objects which provided the only standard for their quantitative comparison. Thus value had a direct and real existence at that time. We know that this direct realisation of value in exchange ceased and that now it no longer happens. And I believe that it won’t be particularly difficult for you to trace the intermediate links, at least in general outline, that lead from directly real value to the value of the capitalist mode of production, which is so thoroughly hidden that our economists can calmly deny its existence. A genuinely historical exposition of these processes, which does indeed require thorough research but in return promises amply rewarding results, would be a very valuable supplement to Capital.”Unfortunately, Engels’ attempt to save the law of value in volume 1 – which was undoubtedly a development of Marx’s own desperate attempt to save it as we have seen above – is still a feeble and unconvincing theory.
Letter, Engels to W. Sombart, from London, March 11, 1895
Why? The reason is that Marx, in volume 1, never makes any such qualifications or limitations to the law of value. In fact, in volume 1, Marx states that money prices depend on the labour value embodied in units of gold or silver, so that long-run prices are determined by abstract socially-necessary labour time needed to produce relevant units of the money commodity (Marx 1906: 108, 111). But Marx says nothing about the rise of commodity money overthrowing his law of value in modern capitalist production.
At the same time, Marx thinks that the second mechanism driving prices is the fluctuation of labour values of commodities as against money (Marx 1906: 111). This is succinctly summed up in what Marx calls the “laws of the exchange of commodities” in Chapter 5 of volume 1:
“It is true, commodities may be sold at prices deviating from their values, but these deviations are to be considered as infractions of the laws of the exchange of commodities, which, in its normal state is an exchange of equivalents, consequently, no method for increasing value.” (Marx 1906: 176–177).So either (1) Marx meant to apply this to modern capitalism in its contemporary form or (2) he was so incompetent and useless he never told his readers how the theory had to be strictly limited to pre-modern times. Either way Marx is damned.
Moreover, as I argued in the original version of this post, there is no convincing empirical evidence for Marx’s and Engels’ attempt to salvage the law of value in volume 1 by restricting it to the past. Once we realise this, it is the death blow to the Marxist labour theory of value.
The final view of Engels, then, was that the law of value in volume 1 of Capital had to be restricted to the pre-modern world of commodity exchange, but we have no good reason to accept that the theory properly describes price determination in that era. In volume 3 of Capital as edited by Engels, Classical long-run equilibrium prices of production are the anchors for the price system, not socially necessary labour time values. This is an admission that the law of value in volume 1 is irrelevant to the modern world.
Instead, the Marxism of the volume 3 of Capital, like modern Sraffianism, sees long-run equilibrium prices, based on cost of production and a uniform rate of profit, as the anchors or centres of gravity for the price system around which prices fluctuate.
Unfortunately, even this concept of a tendency to long-run equilibrium prices in modern capitalism has grave difficulties, and in the end such an idea should be regarded as an unrealistic assumption in an overly analytic, abstract model set in logical time (Lee and Jo 2011: 868–869), where ultimately it can only be assumed by definition to be true (on this issue, see here).
Böhm-Bawerk, Eugen von. 1949. “Karl Marx and the Close of His System,” in Paul. M. Sweezy (ed.), Karl Marx and the Close of His System and Böhm-Bawerk’s Criticism of Marx. August M. Kelley, New York. 3–120.
Engels, F. 1895. Letter, Engels to Conrad Schmidt, March 12, 1895
Engels, F. 1895. Supplement to Capital, Volume III
Lee, Frederic S. and Tae-Hee Jo. 2011. “Social Surplus Approach and Heterodox Economics,” Journal of Economic Issues 45.4: 857–875.
Marx, Karl. 1909. Capital. A Critique of Political Economy (vol. 3; trans. Ernst Untermann from 1st German edn.). Charles H. Kerr & Co., Chicago.
Marx, Karl. 1991. Capital. A Critique of Political Economy. Volume Three (trans. David Fernbach). Penguin Books, London.
Schmidt, Conrad. 1889. Die Durchschnittsprofitrate auf Grundlage des Marxschen Wertgesetzes [The Average Rate of Profit on the basis of Marx’s Law of Value]. Stuttgart.
Schmidt, Conrad. 1895. “Der dritte Band des Kapital,” Sozialpolitisches Zentralblatt 22 (25th February): 254–258.
Sombart, Werner. 1894. “Zur Kritik des ökonomischen Systems von Karl Marx” [Toward a Critique of the Economic System of Karl Marx], Archiv für soziale Gesetzgebung und Statistik 7: 555–594.