Home > Dave's Favourites, General Politics, Marxism > Is the Marxian labour theory of value correct?

Is the Marxian labour theory of value correct?

In responses to my recent postings I have encountered scepticism about the validity or   relevance of Marxian economics today. So I thought it would be worth explaining a few reasons why he has to be taken very seriously as an economist.

It feels a bit like summarizing Proust in 100 words to say this but, the two most important things that Marx wrote about the economy were that labour is the source of exchange value, and that the incomes of the propertied classes derived from the exploitation of labour.

He was not the first to make these points, they were an established  strand of early 19th century thought, but Marx made the points more clearly and with greater cogency than any of his predecessors.  Nowadays, these ideas are absent – not only from the economics curriculum but even from Communist Party policy.

Why? Have they been proven wrong?

Well orthodox economists are pretty confident that the labour theory of value has been proven wrong, but if you follow up the literature, their proofs are of a type peculiar to contemporary economics. In most sciences, hypotheses are evaluated by confronting their predictions with empirical data. In economics proof is rather different. It is mathematical proof of the form: Let us assume the following axioms, and then see what must be true about the economy.

Using proofs of this sort, Samuelson for example claimed to have demonstrated that the labour theory value provided no useful information about prices.

The problem with this is that if the axioms are wrong, the proof is worthless. By a judicious choice of axioms one can prove all sorts of things.

Then in the 1990s Marxian economists started to apply the normal scientific method to the labour theory of value, for example here or  here or  there. What did they find?

They found that the predictions of Marx and before him Ricardo had been spot on. Market prices were actually correlated with labour values to the remarkable degree of 95% or more. That meant that 95% of the variation in the prices of goods is explained by the labour cost of making them.

More strikingly, it was shown that the more capital intensive an industry was, the lower was its rate of profit. This is exactly what one would expect if labour rather than capital was the sole source of value. This explains why railway projects like the Channel Tunnel are almost always unprofitable. They involve a lot of capital but employ little labour on which to make a profit.

Marx had said that : “Very large undertakings, such as railways, on the other hand, which have an unusually high proportion of constant capital, do not yield the average rate of profit, but only a portion of it,”  Those in favour of rail privatisation in other countries take note, they will never be profitable.

Capital itself creates no value.

Once it is realised that the determination of value by labour is a well proven scientific theory, then Marx’s analysis of exploitation follows – with all sorts of disturbing moral consequences for the established order.

It gives moral strength to Unions opposing exploitation and it undermines any claim of capital to a share in the national income.

About these ads
  1. March 19, 2010 at 7:12 pm

    But Paul, Your most important contribution is the way in which You improved Marx’s labor theory of value – its implication at the socialist economic calculation.

  2. S G
    March 20, 2010 at 2:22 pm

    Let me see, you say the axiomatic approach is bad because the axioms can be wrong? That is obvious, but I don’t see you discussing the axioms. By the way, surely economics is not alone in using the axiomatic approach and secondly it is definitely not the only approach being used in economics (have you never seen empirical or experimental work in economics??).

    The fact that market prices and labour value (I guess you mean wages?) are correlated does not mean anything about the direction of the causation. Actually, glancing over the papers you are linking I am surprised to find no regressions, even less panel regressions, and no causality tests!

    Anyway the labour theory of value just makes no sense for obvious reasons: prices in a market depend on supply and demand and thus can be influenced by monopoly power, marketing etc

    If Apple drops the price of an iphone 40% what can the LTV offer as an explanation? If we have a merger in a market and prices rise, how can LTV explain it?

    • Matt
      May 31, 2010 at 9:00 pm

      LTV isn’t supposed to determine the one-time prices of individual commodities, which Marx openly acknowledges are affected by market conditions, scarcity, etc. LTV is a theory that focuses more on the general situation, long-term prices, monopolies, etc. and especially the price at which products tend towards. Having said that, the point you make above about the direction of causation is a serious one. My complaint with LTV is that I haven’t (yet) found a proper attempt to explain why it should be so. Why should the average amount of labour that goes into a given commodity-type determine the price irrespective of how much that labour on the average costs the capitalist? I can quite see why this might be the case when the capitalist is competing with independent workers who own their own means of production, with the latter being the primary factor in setting the price, but why this should persist when the market has no clear means of recognising labour itself (only the price of labour) I am not sure. There may also be a contradiction here with Marx’s assertion that the rate of profit tends to equalize across all industries (something I see at the moment as unproblematic). Given that some industries require more exploitation than others in order to turn over a profit, won’t there be more competition in those sectors where the rate of profit tends to be high due to the high proportion of variable capital employed in production? And won’t this competition be a force which leads the market price of the commodities so produced to have a “natural price” (or equilibrium price) below that of the Value of the commodity?

      I should add that I raise these issues with an expectation of finding a resolution; this has been my experience so far when questioning Marxian thought.

  3. March 20, 2010 at 7:42 pm

    Samuelson’s invalid axiom was that the rate of profit will be the same in all industrial sectors.

    In fact the rate of profit is systematically lower in industries with a high capital labour ratio ( Paper2 table 2, Paper 3 table 8).

    I think your glance at the papers was excessively swift if you found no correlations, see Table 1 and Table 3 of paper 2, Table 1 and 2 of the last paper. The first paper uses a different metric the Mean Absolute Deviation of prices compared to values. It is arguable(http://www.springerlink.com/content/n00531gj73n8v874/) that this is a better measure than correlation. Other economists have argued for the use of different metrics, but whichever metric is used the match between labour values and market prices is very close.

    You say that the labour theory of value makes no sense for the : ‘obvious reason that prices depend on supply and demand’. Well the classical economists were well aware of supply and demand, but argued that these caused short term deviations from the gravitational point set by labour values. The observations in the papers cited are consistent with this theory.

    On the prices of mobile phones, the LTV explains why phones are affordable at all – the reason being that the labour required to make digital components has shrunk by orders of magnitude since the invention of integrated circuits. We dont need to explain the price at which every sale takes place, only the mean price for a class of comparable commodities.

  4. D
    March 20, 2010 at 10:15 pm

    @SG,

    Quite the contrary. A causal mechanism by which a statistical law of value is an emergent phenomena is provided in the last reference, which moreover also uses real employment data and not just labour cost as a proxy.

    The statistical LTV cannot predict an individual price fluctuation but it can explain the underlying real cost structure and relative prices in an integrated capitalist economy far better than any alternative theory.

  5. March 20, 2010 at 11:28 pm

    “Anyway the labour theory of value just makes no sense for obvious reasons: prices in a market depend on supply and demand and thus can be influenced by monopoly power, marketing etc”

    Prices are not values. Prices gravitate around exchange values under conditions of perfect competition.

  6. S G
    March 21, 2010 at 2:49 am

    Simon
    “Prices are not values. Prices gravitate around exchange values under conditions of perfect competition.”

    So what are “values” and why should we care about them as economists?

    Cockshott
    “Samuelson’s invalid axiom was that the rate of profit will be the same in all industrial sectors.”

    You are aware that most of Samuelson’s work is some 40-50 years old?
    The idea that the returns on factors have to be equal across sectors is not an axiom, it is the result of free mobility and profit seeking motives. Now, if you don’t have perfect mobility you can get different results, as has been done in many many papers.

    “your glance at the papers was excessively swift if you found no correlations”

    I said I didn’t see regressions. I am not an expert in econometrics, but my guess is that a panel regression is the least you need to show that wages influence prices over time in a manner that would support the LTV. And the reason is that you want to control for other variables AND show that the correlation is not spurious, not the effect of some common cause for the two variables etc

    “classical economists were well aware of supply and demand, but argued that these caused short term deviations from the gravitational point set by labour values”

    I am sorry for my ignorance, but is there a reason to revert to classical economists on topics where the economic discipline has seen such incredible progress? As far as I am concerned, any economic theory involving market micro-behavior and prices written before 1948 is hopelessly antiquated, since it just hasn’t had the immensely useful benefit of using game theory.

    “We dont need to explain the price at which every sale takes place, only the mean price for a class of comparable commodities.”

    Oh well, I think you do, because modern economic models (for example Industrial Organization models) can explain prices of individual goods pretty well I would think, so why do we need some theory that can’t?

    D
    “A causal mechanism by which a statistical law of value is an emergent phenomena is provided in the last reference, which moreover also uses real employment data and not just labour cost as a proxy.”

    what is the causality test being used? And don’t you mean phenomenon? (sorry, I’m Greek and have a obsessive compulsion with such details)

    “The statistical LTV cannot predict an individual price fluctuation but it can explain the underlying real cost structure and relative prices in an integrated capitalist economy far better than any alternative theory”

    Really surprised then that we don’t see such studies published in some prestigious journal.

    • March 21, 2010 at 5:50 pm

      “You are aware that most of Samuelson’s work is some 40-50 years old?
      The idea that the returns on factors have to be equal across sectors is not an axiom, it is the result of free mobility and profit seeking motives. Now, if you don’t have perfect mobility you can get different results, as has been done in many many papers.”

      Yes but the Samuelson critique of the labour theory of value is still taken to be the authoritative one, I agree that he last wrote on this some 40 years ago, but if his assumption of equal rates of profit is dropped the critique of labour values fails.

      Secondly I dont see how another theory can explain that the differences in profit rate should be such that labour intensive industries have a higher rate of profit. This falls straight out of a simple labour theory on the other hand. More workers to exploit means more profit.

      On panel regressions and tests for alternative causality, these were done in this paper (http://ideas.repec.org/a/oup/cambje/v21y1997i4p545-49.html), and there was a subsequent discussion in the journal over whether the correlations were or were not spurious. The significant result is that other proposed value bases like energy use were shown to be far poorer in predicting price ratios, and to have no additional explanatory effect in panel regressions with labour.

  7. D
    March 21, 2010 at 12:38 pm

    @ S G
    Sorry for the typo.

    “what is the causality test being used?”

    The causal law is that the total (i.e. vertically integrated) labour content of commodities determines their relative prices. I.e. that labour is the source of economic value. The causal mechanism from which this law emerges is the pressing need to meet the wage bill of individual firms integrated in an economy. Why this is so, is shown in one of the references.

    Still there are no alternatives that match the predictive power of the theory.

    “Really surprised then that we don’t see such studies published in some prestigious journal.”

    Well, people were really surprised about the results in the first place because this fundamental theory of classical/Marxian political economy was taken as false on axiomatic grounds of profit rate equalization for more than one hundred years.

    Some of the research was published in the Cambridge Journal of Economics, but as long as the theory is dismissed on non-scientific grounds it cannot be published in neoclassical journals.

  8. Jacob Richter
    March 23, 2010 at 5:09 am

    I wonder if this highly statistical take on the labour value doesn’t need Okishio to deliver a one-two-punch on the exploitation of labour.

    [Perhaps Paul can explain Okishio's math, since mine is a tad rusty.]

  9. March 24, 2010 at 7:59 pm

    Not sure what about Okishio you are refering to. His 1950s work makes the same error as Samuelson

  10. Jacob Richter
    March 25, 2010 at 4:17 am

    The so-called “Fundamental Marxian Theorem”

  11. March 25, 2010 at 9:16 am

    OK, I did not know that Okishio had invented that, I came across it in Morishima.
    In practical terms it is pretty hard to estimate exploitation accurately that way. One can approximate it using the input output tables, but the problem is that the final consumption columns of the tables are not broken down by social class so it is hard to get at the data you need.
    In practice you have to do something like estimate the monetary equivalent of labour time, and then look at the division of value added in key sectors of the economy between labour income and property income. You can then back estimate it in terms of hours if you want, but Marx’s quantity s/v is a dimensionless number so one can derive it directly from monetary aggregates in the national income tables.

  12. Jacob Richter
    March 25, 2010 at 1:30 pm

    Thanks for the info!

  13. Jacob Richter
    April 2, 2010 at 3:17 pm

    I should add belatedly re. Samuelson: if he couldn’t even get his Keynes right in his Economics textbook, how much more could he not get Marx right?

  14. April 2, 2010 at 4:18 pm

    Jacob Richter :

    I should add belatedly re. Samuelson: if he couldn’t even get his Keynes right in his Economics textbook, how much more could he not get Marx right?

    Keep in mind that Samuelson was by no means alone in thinking that KM thought that capital intensive industries would have the same rate of profit as labour intensive ones. He clearly overlooked Marx’s analysis of the railway industry – the pre-eminent capital intensive industry of victorian times.

  15. Jacob Richter
    April 5, 2010 at 9:20 pm

    I meant that as a general statement, since Samuelson is the founder of “Bastard Keynesianism,” according to the Post-Keynesians (Steve Keen, Hyman Minsky, William Mitchell, Joan Robinson, L. Randall Wray, etc.).

  16. Paul Cockshott
    June 1, 2010 at 9:34 am

    Matt :
    LTV isn’t supposed to determine the one-time prices of individual commodities, which Marx openly acknowledges are affected by market conditions, scarcity, etc. LTV is a theory that focuses more on the general situation, long-term prices, monopolies, etc. and especially the price at which products tend towards. Having said that, the point you make above about the direction of causation is a serious one. My complaint with LTV is that I haven’t (yet) found a proper attempt to explain why it should be so. Why should the average amount of labour that goes into a given commodity-type determine the price irrespective of how much that labour on the average costs the capitalist? I can quite see why this might be the case when the capitalist is competing with independent workers who own their own means of production, with the latter being the primary factor in setting the price, but why this should persist when the market has no clear means of recognising labour itself (only the price of labour) I am not sure. There may also be a contradiction here with Marx’s assertion that the rate of profit tends to equalize across all industries (something I see at the moment as unproblematic). Given that some industries require more exploitation than others in order to turn over a profit, won’t there be more competition in those sectors where the rate of profit tends to be high due to the high proportion of variable capital employed in production? And won’t this competition be a force which leads the market price of the commodities so produced to have a “natural price” (or equilibrium price) below that of the Value of the commodity?
    I should add that I raise these issues with an expectation of finding a resolution; this has been my experience so far when questioning Marxian thought.

    You have put your finger on what, for almost a century was a major weakness in Marxian economic theory – what is called the Transformation Problem. From the publication of the 3rd volume of Capital after Marx’s death until the mid 1980s there was no altogether satisfactory answer to the questions you raised.

    Advances since then have given us clear answers.

    The key advance came when the mathematicians Farjoun and Machover applied the methods of statistical mechanics to price formation in their book ‘The Laws of Chaos’, in 1983. They point out, as you do informally above, that individual prices have to be considered as random variables that are subject to all sorts of statistical noise due to multiplicities of complex and contingent circumstances.

    They introduce a means of measuring this randomness. They say let us look at the ‘specific labour content’ of commodities, that is to say for any given commodity the specific labour content is the share of its selling price per £ that was ultimately made up of wages. If I spend £1 on a burger in Macdonalds some of that goes to pay the wages of the cooks at the shop, some of it goes to pay the wages at the bakery that makes the bun, some on wages in the meat processing plant etc. In general the amount that goes on wages will be only a portion of the total, since at each stage going back, some of the price goes as property income. They then ask what are the statistical properties of this random variable ‘specific labour content’.

    The first deduction is that it will have a normal distribution since it is the sum of a large number of independently noisy factors — this follows from what we know about the Gaussian distribution. They then argue that the mean of the distribution will be around 50p per £. This involves some fairly abstruse statistical reasoning, but it does correspond reasonably well with real data, that property income and wage income tend to be roughly the same shares of national income — we do not ever see wages as 95% of national income for example. (Their analysis is a bit more sophisticated than this, but you can check their book for the details).

    They then ask, given the mean, what will be the standard deviation of the specific labour content. They ask what is the probability that a commodity can be sold with a specific labour content of above £1 — ie sold so that the total selling price is insufficient to pay wage bills all the way down the supply chain?
    They suggest that less than 5% of commodities can be sold under these conditions, since the manufacturers would rapidly go insolvent ( since they would have non-wage costs as well). Some such sales will occur, but they will in the main be closing down sales.

    From the properties of the normal distribution they can then compute the standard deviation of the specific labour content and show that in fact the spread of actual prices has to be closely grouped around the mean. But a narrow standard deviation of price to labour content is exactly what the classical labour theory of value had predicted. I and others were able during the 1990s to verify their predictions about the statistical properties of specific prices in several countries. With their work, what had always been a puzzle about the labour theory of value — why it held, became clear: it is an emergent phenomenon analogous to the 2nd law of thermodynamics.

    As to your second point, yes, a consequence of the labour theory of value is that the rate of profit in capital intensive industries should be lower, and yes this is really observed if you examine the I/O tables of all the OECD economies.

    • Matt
      June 1, 2010 at 10:21 am

      Thanks for this. It seems a problem to me that this “specific labour content” variable is actually concerned with wages paid rather than with time. Marx’s original theory was concerned with the latter (although he made allowance for skilled jobs by including training time). This is a problem because the rate of exploitation is different in different industries, meaning that there should be no especially clear relationship between wages paid out and Value added. On the face of it, showing that wages paid out is playing the key role in determining price actually weakens LTV, no?

  17. Paul Cockshott
    June 1, 2010 at 12:13 pm

    It does not weaken it since the way in which labour input registers has to be via wages.

    In one statistical study I worked on, there was a slight improvement in correlation between labour values and prices once one used hours worked rather than wage costs as the labour input surrogate, but I would not place any great significance on this.

    If I recall F&M’s derivation the effect of a wider variation in wage rates between industries is allow a greater variation in mean of the specific labour content. That is the greater the variation in wage rates between industries, the greater the possibility for the wage share to deviate from 50%, but this involves some rather abstruse properties of Gamma distributions.

    • Matt
      June 1, 2010 at 12:40 pm

      The principle that labour input can only register through wages is sensible and was the basis of my original objection to Marx’s LTV. Maybe our disagreement lies in the interpretation of LTV itself? In works such as Capital and Value, Price and Profit, Marx seems to me to be identifying Value with socially average labour time to the exclusion of Value being socially average wage costs – so that whatever the rate of exploitation, the relationship between Value and price tends to remain intact.

      Also, I note that you have responded to points by myself and others by stating that the labour theory of value would predict (relatively stable) different rates of profit in industries depending on the need to employ a higher proportion of fixed capital. But this is from Marx’s Value, Price and Profit:

      “What would be the consequence of this difference in the rates of profit for capitals employed in the different branches of industry? Why, the consequence that generally obtains whenever, from whatever reason, the average rate of profit comes to differ in different spheres of production. Capital and labour would be transferred from the less remunerative to the more remunerative branches; and this process of transfer would go on until the supply in the one department of industry would have risen proportionately to the increased demand, and would have sunk in the other departments according to the decreased demand. This change effected, the general rate of profit would again be equalized in the different branches.”

      I have not read all of Marx’s economic works and haven’t read beyond volume one of Capital. Does Marx later rethink the issue of the equalization of the rate of profit? I know he has a theory of the declining rate of profit due to the increased amount of fixed capital, but I presumed that this applied to the general, global situation, rather than specific industries in isolation from others.

      Thanks again for your time.

  18. Paul Cockshott
    June 1, 2010 at 12:52 pm

    He does deal with the issue at length in Capital III, but you have to realise that this was never a finished work, and it contains some inconsistencies in his assumptions. He says that for industries with a high fixed capital, like railways, there is no equalisation of the rate of profit since the capital is imobile – railways thus earn low rates of profit. But he continues the general argument that the rate of profit should equalise. These are two mutually inconsistent positions, and are an example I think where he was insufficiently critical of Ricardo’s assumption of profit rate equalisation. Imobility of fixed capital is a more general issue than one affecting railways alone.

    • Matt
      June 1, 2010 at 1:04 pm

      To clarify on the issue of LTV and wages, your version of LTV, which I understand you and others have successfully empirically defended, is actually quite different from Marx’s?

      If I understand your version of LTV correctly, the concept of rate of exploitation would have to be abandoned as meaningless, as in your version Value itself can only be measured as the cost of labour, and the rate of exploitation is founded on the difference between the Value added by labour and the cost of labour. As your LTV seems to me to identify wage costs with Value, the concept of rate of exploitation would have to be abandoned. Or is there some other way of thinking through these problems?

      To think of it, there are other examples of where Marx’s thought relies on the idea of the cost of labour to the capitalist being different from the value added. Isn’t this crucial to Marx’s explanation of the possibility of profitability?

      • paul cockshott
        June 1, 2010 at 1:33 pm

        I think that in Capital I there is the assumption in the early chapters of there being a going rate for labour power. if this can be reduced in real terms, the profit share is shown to go up. I dont think we have to abandon this for either individual firms or the economy as a whole.

        At the intermediate level of industries, then the effect of reductions of wages accross whole industries the implications may be different, I am not sure. I would say that if in a whole trade the level of wages are depressed, then the selling price will tend to fall below its value in terms of embodied labour. I am not sure if Marx specifically addresses this, though I have a feeling that he does in his analysis of sweated trades. Without going and reading back through those chapters I am not sure.

  19. Matt
    June 2, 2010 at 11:06 am

    I agree that Marx’s observation on profit and the reduction of wages can still be applied. What seems more difficult though is Marx’s explanation of the very possibility of profit – something which he treats as problematic and to be explained by his theory.

    Perhaps it is more fruitful to focus on less opaque areas to bring out the difference between Marx’s LTV and yours. Obviously I argued above that the “rate of exploitation” is meaningless in your version of LTV – possibly even the concept of exploitation, in the Marxian sense, is meaningless in your version of LTV.

    Another way of introducing the same point would be to look again at Value, Price and Profit. Marx’s argument there clearly depends on wages *not* being the crucial issue. He argues that wage costs are not the determinant of final prices against Citizen Weston’s assertion that they are. He explains that the price of a product is ultimately, in the general case determined by the labour time which goes into making it, and that therefore wage costs can be raised without affecting the price at all, so long as the worker is still being paid less than the value he adds to the product with his labour.

    So your LTV seems to be radically different from Marx’s, and Marx appears to have argued against your “labour theory of value” in one of his most famous works.

    Hopefully I’ve misunderstood you.

  20. Paul Cockshott
    June 3, 2010 at 12:02 am

    I don’t accept that the rate of exploitation is a meaningless concept in this concept. I first started to work on the issue of how accurate the labour theory of value is in order to provide a backup for the idea that one can get estimates of the rate of exploitation using national income data. One can only do that if monetary magnitudes actually correlate well with embodied labour content. We were able to show that this is the case whether one uses money wages as a surrogate for labour input, or whether on corrects for average wage rates per industry and translates money wages into millions of hours worked.

    I dont see that the proposition that labour values make themselves felt via wage expenditures undermines Marx, who, one must remember does not give any particular mechanism by which labour values operate. He merely, like previous economists, takes the labour theory of value as being an obvious empirical given. Since he takes it as an empirical fact, I dont think his argument is undermined if later economists come up with a mechanism to explain this empirical fact.

    To Citizen Weston, what I take Marx to have been doing here is to be arguing against the then prevalent idea of an ‘iron law of wages’, the Malthusian concept that wages could never be raised above the subsistence minimum. Weston was putting forward a slightly different version of this, the argument that any general wage increase will be wiped out by the consequent inflation. This is an argument not about the relative values of commodities, but about the general price level, or the value of money. I can recall former Prime Minister Callaghan putting forward what was essentially the same argument as Weston when he argued that one man’s wage increase is another mans price increase.

    Marx’s response comes down to saying that there is another free variable in the system — the profit share, and that generalised wage increases could reduce the profit share.

    What is the prediction of the Farjoun and Machover theory under these circumstances?

    Suppose the wage share rose from 50% to 75% of national income, then the effect is to compress the dispersion of prices relative to labour values because there is less leeway for firms to sell below labour values without making a loss.
    So a general rise in wages will tend to reduce the profit share, and force market prices into closer correlation with labour values.

    Of course since Marx wrote, the gold standard has been abandoned and with monetary there is now more opportunity for employers to pass on wage increases as price increases than there was in the 19th century.

  21. Karl Engels
    July 24, 2010 at 12:00 pm

    Try this thought experiment. Suppose nobody worked. Suppose no commodities and no exchange. We live on what is provided by nature. Though even picking berries and nuts is work (a universal, nature-imposed necessity). All the same, no-one works in the sense of technologically transforming the earth and its products. Does anything have exchange value ? If the market theory of value is right and desirability creates EXCHANGE value then even in our no-work culture it should exist. This value is not the same as symbolic value. All cultures attribute symbolic value to certain objects. But exchange value ? The complexities of theory mustn’t rob us of our ability to see clearly to the heart of the question. If without work their is no exchange value, then whatever caveats and intricate objections can be raised, it must be true that labour plays a primary role in the creation of exchange value. We need go no further than this to argue that if it is so, then labour should rewarded justly for its contribtion. Further, if capital in and of itself cannot be shown to produce exchange value (ie unless it is combined with labour) then we can justly argue that labour is the primary and capital the secondary source of exchange value. There is a simple, practical test: let everyone who works, except entrepreurs, bankers, currency speculators etc stay at home and do nothing for a month and let’s see what happens to the GDP. You may say, Ah, but of course, capital needs labour. So let’s do the opposite. Everyone continues to work except entrepreneus, bankers, currency speculators etc and let’s see what happens. The complex theory exists only because the practical experiment can’t be tried. If you don’t believe in god theology is meaningless. It is the belief that capitalists create value, that markets magic it out of desire, that keeps the complex theories alive. Just as the belief that the earth is the centre of the universe kept alive complex theories of the rotation of the planets. Copernicus made things much simpler.

  22. Tom the Red Dragon
    August 27, 2010 at 11:39 pm

    I am only curious how Marx and his dear followers would explain the actors and their high wages! They don’t exploit none of the labor of others. They are simply trying to sell a product (their movie, look, acting, etc.). Or someone who is self-employed and after a while is getting very rich with his effort only. The capital that this guy have is his idea only(let’s forget for a while that the idea is not predetermined from the economic relations). Does Marx and his followers would say that the owner of the capital (the idea owner) is exploiting himself? I am not trying to take a side. Just curiosity about this magnificent ideas of Marxism. Thanks!

  23. September 2, 2010 at 10:11 am

    It is a combination of two factors.

    1. The exitence of intellectual property rights legislation which creates artificial monopolies and prevents the operation of a free market. Thus recordings by musicians for example on iTunes sell far above their real cost of production. The presence of the IP law leads to monopoly profits, a part of which can be negotiated by the artist. Proponents of the labour theory of value have, since David Ricardo, recognised that monopoly can give rise to rent incomes.

    2. Bargaining position of the artist is strengthened as compared to an average worker since they are in the public eye. They are thus not freely dispensed with like a shop assistant who can be replaced by another without the customers noticing. Well known artists can then negotiate for a share of the monopoly profit.

  24. JoeHammerman
    September 6, 2010 at 1:51 am

    Hey guys – thanks for all of this. It is fascinating.

    I have two questions:

    1)Is it a consequence of this theory that all profit emerges from arbitrage (technological and legal) if the cost of goods correlates so closely with the average cost of the labor inputs?

    2)The local sushi restaurant charges $8 for lunch and $10 for dinner. Such practices are common; is this explained as a temporary, nightly supply – demand imbalance? It seems that the labor inputs for the same meal would be equivalent.

    • September 6, 2010 at 9:43 am

      No profit is not primarily the result of arbitrage, such profits are zero sum and can not lead to a whole class of people earning profit income. No, profit arises from the legalisation of slavery.
      It is clear enough that the old Southern Plantation owner made their profit from the labour of their slaves. The legal system meant that the slaves had no legal claim to the cotton they had grown, so the value they created went to the plantation owner.
      Slavery comes in many forms, chattel slavery, bond slavery and wage slavery. Whilst chattel and bond slavery still exist in the world today, and indeed there are probably more chattel and bond slaves in the world economy than there were in 1860, the prevalent form of slavery today is wage slavery. What all forms of slavery have in common however is that the slave, unlike the free independent producer, has no right to the value they create. Today propertyless people sell themselves into slavery by the month rather than the years they did under the old indenture system, but they have no more right to what they produce than a dalit bonded to slavery in the brick kilns of India, or a sex slave imprisoned in a Brimingham brothel.

      Profit arises from the value produced by the enslaved portion of the population.

  25. Tom the RED DRAGON
    October 27, 2010 at 7:34 pm

    Profit arises from the value produced by the enslaved portion of the population.

    But, wait minute! Let’s imagine simple situation. We have very few people on the planet, a lot of unowned land and everyone is respecting the rights of the others to have their own property. But, to create property from the unowned land you must “mix your labor” with the unowned resource (J.Locke). And let’s imagine for the sake of argument that every citizen on the planet is doing this and after a while they all have their own capital, their own products, their own property. But, they aren’t self-sufficient. What will they done? Simple (as it happened). They will start exchanging some part of their products for others (don’t forget-everyone is having his own capital, no one is labour “slave”). So, someone will change 3 cows for 2 eggs, others will exchange 4 apples for 1 eggs, and so on. And there would be problems-it would be very hard to find what are you looking for. So, naturally and spontaneously a medium of exchange will develop. So, you will have for example a 1 mg of sugar and every other product will be measured with relation to 1 mg of sugar. And very soon the prices and market will arise. Then the profit will be nothing more and nothing less then the net gains over the net losses at the end of the calculating period. So, you can see that there is nothing that automaticlly connects profit and slave laborers. Again, everyone will produce for them-selves and still the profit will exist.
    p.s. As my previous question that wasn’t answered very accurately, I showed with my example that you can have profit without exploiting anyone (the musicians for example). Although there is big true that IP is creating monopoly, still it isn’t the whole answer. (if you want to see libertarian guys who are fighting against IP with good arguments check Stephan Kinsella). So, the profit doesn’t arise from the value produced by the enslaved portion of the population. It is only a reward for the guy who is making some worthy product to the others. The others are rewarding him with their money, because he is creating something that is more VALUABLE than their money!

  26. Intrigued
    August 28, 2012 at 12:28 am

    Can someone explain how the author is correct by generalising that “capital itself never creates value” when the property rent market, as one example, is incredibly lucrative, especially during bad times such as today. This also begs the question why Marx’s definition of “value” is any more objective than anyone else’s.

    • August 29, 2012 at 10:41 am

      There is nothing special about houses in this respect, you could have cited any commodity whose relative price rises: say silver ov the last two years. But a relative rise in the price of one commodity is a relative fall in the price of others, by itself this is a zero sum game, the gain to holders of one commodity are balanced by the losses of holders of others.

  27. Herbie Destroys The Environment
    August 29, 2012 at 5:59 pm

    Paul,

    Here is my very basic understanding of the LTV. Capitalist buys raw materials and fixed capital. He also employs labour power as wage labour. The total of each is the total value of the product. But he employs the wage labour for hours over and above what it takes to maintain the wage labourer. Say it takes 4 hours to maintain the labourer and the capitalist employs him for 7, the surplus value is created from those 3 additional hours. It is value that doesn’t go to the labourer. Then the question is where is the demand for the surplus to come from, other than reproducing the unproductive elements?

    I have a few of questions:

    Does competition mean that capitalists will take less surplus, i.e. They still employ the wage labour for 7 hours but reduce the selling price to undercut the competition?

    If a capitalist gets the extra hours out of labour won’t it mean that he has to buy extra raw materials, so is this factored into the theory?

  28. August 30, 2012 at 9:31 am

    These are basically perceptive questions you ask, but first let me clear up a little confusion in your initial formulation. In the labour theory of value the value of a product is the total number of hours that society has to work to produce it. It thus includes the labour that went into making the raw materials and the labour actually done directly making the thing. If the thing is made under capitalist relations of production, as opposed to being made by self employed workers, only part of that direct labour is paid for when the employer buys labour power. The ratios you give are reasonably accurate for a developed capitalist country like the UK, 4/7 th is a reasonable estimate of the share that is paid for the purchase of labour power.
    Now, on your question of where the demand comes from for the 3 hours surplus, that I think lies at the heart of successive problems that capitalist economies have faced. As you suggest, one way it can be spent is on reproducing the unproductive elements : allowing them to enjoy a Downton Abbey lifestyle, employing hoards of butlers, cooks, maids etc. This is more or less what happened in the late 19th century.
    In principle, the capitalists can spend it on purchasing new plant and machinery whilst living relatively modestly like early cotton masters like Dale or Arkwright. This is also the main way that a market is provided for the surplus in China. But this option is only really available during industrialisation when there is a big need for new industrial plant and machinery. Once the economy is fully industrialised there tends to be no long term net capital accumulation, just a replacement of plant and equipment with newer plant using depreciation funds.
    This forces the system to find other routes to market the surplus. German firms today today, seek it in export markets, but this is in the end self limiting as the importing countries run into debt. Another big market is selling to the state, either in armaments as in the USA or in big infrastructure projects as in Spain recently. A third is by the wealthy depositing their money with the banks who then lend it out as consumer credit – something familiar in this country. But again this hits limits when working class consumers run up more debt than they can afford.
    So the questions you ask actually point to some of the key causes of current economic problems: the credit crisis and the state debt crisis.

  29. Xhu Siao
    May 10, 2013 at 9:03 am

    Dr. Cockshott,

    I am not sure if you responded to http://www.academia.edu/1067831/Testing_the_labour_theory_of_value_an_exchange

    by Shimshon Bichler and Jonathan Nitzan.

    Can you show me any link, or leave some comment here?

    Thanks.

  30. Xhu Siao
    May 10, 2013 at 9:26 am

    Also, according to my understanding of Farjoun and Machover’s Law of Chaos, they seem to say that the tendency of the rate of profit to fall is not really visible – so are you agreeing with them, and are you saying that the tendency of the rate of profit to fall only occurs in the whole economy when capital accumulation is faster than population (labor force) growth?

    And if that’s true, that seems to be much different from what Professor Kliman and some others say.

    • William Cockshott
      August 8, 2013 at 11:00 am

      That is broadly right. This is somewhat different from what Andrew Kliman says. It is based on Marx’s analysis of over-accumulation of Capital. ________________________________________

  31. Stephan
    October 9, 2013 at 1:59 pm

    Dear Willieam Cockshott,
    maybe it is in your papers, however I am always a bit sceptical when it comes to mathematical or statistical evidence about the LTV. So I tend to reject papers which try that even before I read them. The reason is that Marx states in Capitel Volume 3 Chapter that there is a tendency of the equalisation general rate of profits thus (how I understand it) are always diffrerent from the value. There are different rates of surplus value depending on the composition of capital but the profit rate tends to be everywhere the same, otherwise all capital would only invest in one type of production. However the competition between capitalists will prevent that so that there is an equakisation of the profit rates as the individual capitalist is not as such interested in surplus value but profit.

    Quote: “But it is evident that the balance among spheres of production of different composition must tend to equalize them with the spheres of average composition, be it exactly or only approximately the same as the social average. Between the spheres more or less approximating the average there is again a tendency toward equalization, seeking the ideal average, i.e., an average that does not really exist, i.e., a tendency to take this ideal as a standard. In this way the tendency necessarily prevails to make the prices of production merely converted forms of value, or to turn profits into mere portions of surplus-value. However, these are not distributed in proportion to the surplus-value produced in each special sphere of production, but rather in proportion to the mass of capital employed in each sphere, so that equal masses of capital, whatever their composition, receive equal aliquot shares of the total surplus-value produced by the total social capital.” (http://www.marxists.org/archive/marx/works/1894-c3/)

    So either Marx was wrong here and there is something which conteracts the equalization of profit rates through competition or the approach to show the that LTV is right by looking how it predicts average prices in the long term is just wrong. Is there something I miss here?

    PS: Another Question which is not really this important for that matter but interests me. Isn’t there also the problem to define what is productive and what unproductive labor? I mean, it is not important for the costs of production and thus for profit, however it seems in Marx’s point of view to be important for the value of a commodity, so what do you define as unproductive and what as productive labor when it comes to concrete examples… I could never get my head around that. Is the production manager (or however you call the one who organizes the process of production, a task taken more or less aways from the rank and file worker) adding to the value by his labor or not? Or the one who creates a specific image of the product, I mean the usevalue for one person could not only be that a special type of jacket is warm, but makes her/him be more appreciated because of the image linked to the jacket.

  32. October 9, 2013 at 9:53 pm

    De omnibus dubitandum, Stephan. How do you know that the rate of profit equalises. Marx hypothesises that it does, but any hypothesis must be tested against reality. Marx provides no empirical evidence to back up the assertion.
    As far as I am aware nobody bothered to make this obvious test until me and Allin tried it in the mid 90s. What we found is that the rate of profit is far from being equal in different industries. Actually if you look at industrial profit rates they tend to be more or less inversely related to the organic composition of capital in that industry. Look at Figure 5 in this paper ( http://users.wfu.edu/cottrell/vol3.pdf ).
    Why should the rate of profit be lower in industries with a high organic composition?
    Because prices tend to follow labour values!

    You may be sceptical about statistical evidence, but science depends on statistics to evaluate hypotheses.

  33. November 5, 2013 at 10:15 pm

    Marx does not say that profits are equalised across industries! He says the opposite. He says that profit rates vary across industries, and they have to or otherwise there could be no movement from one area of production (low rate of profit) to other areas of production (high rate of profit), which is the very process by which a general rate of profit is established.

    The rate of profit in each sphere is continually changing, because of technological changes within it, because the value of inputs etc are continually changing, and in addition the general rate of profit is continually changing. But the general rate of profit is only an abstraction.

    • November 12, 2013 at 11:38 pm

      That is fair enough, but in the texts on transforming value into prices of production he does assume a single rate of profit. But of course these were not finished texts, but drafts published after his death so we dont know how satisfied he was with the formulation he had arrived at.

  34. November 17, 2013 at 4:26 pm

    Actually, we do know that he saw those formulations as incomplete. He sets out the historical process by which capital invades an industry and thereby prices of production replace exchange values within it. But, he goes on to discuss the way the prices of production in that industry become the cost prices of the industries that consume its outputs as inputs.

    For example, Capital enters cotton spinning. Production rises, market prices fall below market values in order that demand rises to match supply. The market price/price of production of spun cotton, now becomes the cost price paid by the weaver, whether the weaver is a capitalist enterprise or still a direct producer. The value of the weaver’s production is no longer then determined by exchange value, even as a direct producer, because part of his cost price is based on the price of production, not the exchange value of cotton produced by the spinner.

    As Engels sets out this process by which Exchange Value is increasingly replaced by Price of production proceeds from the 15th century onwards, as Capital invades one industry after another.

    But, Marx states clearly precisely because these output costs are also immediately input costs, the presented tables for the transformation of exchange values into prices of production are incomplete, because it is necessary to transform both input and output prices.

    Moreover, the tables presenting the transformation only provide an indication of what the price of production is. But, much of those texts involve Marx setting out the PROCESS by which this transformation takes place. In fact, in the Preface, Engels discusses all those others who had arrived at the necessary calculation of how Values are transformed into prices as well as Marx. But, he says, what none of them had done was show what the process of that transformation was.

    The whole process of transformation requires that actual profit rates are NOT equal. Part of the problem with the solution provided many years ago by Kliman and McGlone, is that it steps outside the process described by Marx. Marx says that the process requires that the actual market prices must adjust so that low rates of profit rise, and high rates fall. The only means by which that can occur is if there are changes in the level of supply. In other words, where the rate of profit is low, the only way it can rise (apart from changes in productivity etc) is if the market price of these commodities rise. But, that can only happen, as Marx says, if the supply falls relative to the demand. In other words, capital must leave this sphere, reducing supply and causing prices to rise. Similarly, where profits are high, capital will be attracted. Supply will rise, market prices will fall, and the rate of profit will fall.

    But, all of this process requires that the rate of profit is different in each sphere.

  35. January 2, 2014 at 12:37 pm

    What price of production theory fails to explain is why the rate of profit is systematically lower in industries with a high organic composition of capital. Only the labour theory explains that.

  1. October 9, 2013 at 10:01 am
  2. November 18, 2013 at 1:12 am
  3. November 18, 2013 at 1:38 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 124 other followers

%d bloggers like this: