Thursday, May 7, 2015

Marx on Labour Value and Cost Price in the Grundrisse

Marx’s Grundrisse der Kritik der Politischen Ökonomie (Outlines of the Critique of Political Economy) is an 800 manuscript on political economy that Marx wrote between 1857–1858, but was not published until 1939 (Wheen 2001: 227). Therefore the Grundrisse represents Marx’s thinking on economics in the late 1850s and some 10 years before he published Capital.

In the Grundrisse, there is this very interesting description of the labour value:
The value (the real exchange value) of all commodities (labour included) is determined by their cost of production, in other words by the labour time required to produce them. Their price is this exchange value of theirs, expressed in money. The replacement of metal money (and of paper or fiat money denominated in metal money) by labour money denominated in labour time would therefore equate the real value (exchange value) of commodities with their nominal value, price, money value. Equation of real value and nominal value, of value and price. But such is by no means the case. The value of commodities as determined by labour time is only their average value. This average appears as an external abstraction if it is calculated out as the average figure of an epoch, e.g. 1 lb. of coffee = 1s. if the average price of coffee is taken over 25 years; but it is very real if it is at the same time recognized as the driving force and the moving principle of the oscillations which commodity prices run through during a given epoch. This reality is not merely of theoretical importance: it forms the basis of mercantile speculation, whose calculus of probabilities depends both on the median price averages which figure as the centre of oscillation, and on the average peaks and average troughs of oscillation above or below this centre. The market value is always different, is always below or above this average value of a commodity. Market value equates itself with real value by means of its constant oscillations, never by means of an equation with real value as if the latter were a third party, but rather by means of constant non-equation of itself (as Hegel would say, not by way of abstract identity, but by constant negation of the negation, i.e. of itself as negation of real value). In my pamphlet against Proudhon I showed that real value itself -- independently of its rule over the oscillations of the market price (seen apart from its role as the law of these oscillations) -- in turn negates itself and constantly posits the real value of commodities in contradiction with its own character, that it constantly depreciates or appreciates the real value of already produced commodities; this is not the place to discuss it in greater detail. Price therefore is distinguished from value not only as the nominal from the real; not only by way of the denomination in gold and silver, but because the latter appears as the law of the motions which the former runs through. But the two are constantly different and never balance out, or balance only coincidentally and exceptionally. The price of a commodity constantly stands above or below the value of the commodity, and the value of the commodity itself exists only in this up-and-down movement of commodity prices. Supply and demand constantly determine the prices of commodities; never balance, or only coincidentally; but the cost of production, for its part, determines the oscillations of supply and demand. The gold or silver in which the price of a commodity, its market value, is expressed is itself a certain quantity of accumulated labour, a certain measure of materialized labour time. On the assumption that the production costs of a commodity and the production costs of gold and silver remain constant, the rise or fall of its market price means nothing more than that a commodity, = x labour time, constantly commands > or < x labour time on the market, that it stands above or beneath its average value as determined by labour time. The first basic illusion of the time-chitters consists in this, that by annulling the nominal difference between real value and market value, between exchange value and price -- that is, by expressing value in units of labour time itself instead of in a given objectification of labour time, say gold and silver -- that in so doing they also remove the real difference and contradiction between price and value. Given this illusory assumption it is self-evident that the mere introduction of the time-chit does away with all crises, all faults of bourgeois production. The money price of commodities = their real value; demand = supply; production = consumption; money is simultaneously abolished and preserved; the labour time of which the commodity is the product, which is materialized in the commodity, would need only to be measured in order to create a corresponding mirror-image in the form of a value-symbol, money, time-chits. In this way every commodity would be directly transformed into money; and gold and silver, for their part, would be demoted to the rank of all other commodities.”

Marx, The Grundrisse, October 1857, Chapter on Money
https://www.marxists.org/archive/marx/works/1857/grundrisse/ch02.htm#p136
Some important points:
(1) The initial sentence is crucial to understanding Marx’s thinking:
“The value (the real exchange value) of all commodities (labour included) is determined by their cost of production, in other words by the labour time required to produce them.”
A real “exchange value” or price equal to labour value (“labour time required to produce them”) is the cost of production. This is so obviously the same view that Marx held in his essay Wage-Labor and Capital published in the periodical the Neue Rheinische Zeitung in April 1849:
“The determination of price by cost of production is tantamount to the determination of price by the labortime requisite to the production of a commodity, for the cost of production consists, first, of raw materials and wear and tear of tools, etc., i. e., of industrial products whose production has cost a certain number of work-days, which therefore represent a certain amount of labor-time, and, secondly, of direct labor, which is also measured by its duration.” (Marx 1902: 34).
For more discussion of this point see this post.

(2) Marx thinks that
“The replacement of metal money (and of paper or fiat money denominated in metal money) by labour money denominated in labour time would therefore equate the real value (exchange value) of commodities with their nominal value, price, money value.
Of course real world capitalist economies do not do this, and even if you introduced “labour money” denominated in labour time, then you would have to
(1) demonstrate how to reduce all heterogeneous human wage labour to a common, meaningful homogeneous unit so that, for example, 1 labour money unit equals one unit of homogeneous labour time, and

(2) ensure that in real world capitalism wage labourers are paid exactly in accordance with the socially necessary labour time appropriate for their labour, and

(3) price all non-labour factor inputs in terms of socially necessary labour time required to produce them.
Only this would allow one to say that a cost of production price would be equal to labour value.

Yet, once again, it is obvious that this is not how real-world capitalism works, and the fact that it does not work this way contradicts and negates Marx’s statement in (1) above (that is, “The determination of price by cost of production is tantamount to the determination of price by the labortime ... .”). Marx even recognizes this by saying: “But such is by no means the case.”

(3) After this Marx argues that the “value of commodities as determined by labour time is only their average value,” which appears to be what he would later call the “average price” that equals the cost of production price plus a uniform long-run rate of profit (though admittedly he does directly not refer to profit in the quotation above). We even get some Hegelian claptrap with which Marx dresses this up philosophically (“as Hegel would say, not by way of abstract identity, but by constant negation of the negation, i.e. of itself as negation of real value”).
But as always there is no rational reason to think that the labour theory of value is of any empirical value in the first place, when
(1) it ascribes to human wage labour a special significance that it does not have (that is, if work expended by labour creates a value in commodities, then slaves, animals or the power of nature should be able to create such a value too), and

(2) the problem of reducing (and aggregating) all heterogeneous human labour to a homogenous unit is a problem almost as severe as the idea of aggregating interpersonal subjective utilities, and I see no evidence that Marx or Marxists have ever shown how to do it.
Finally, neither real-world market wages or non-labour factor input prices are determined in the way required by Marx to make a cost of production price equal labour value, even if you could surmount the serious problems in (1) and (2) above.

In short, even here in the early version in Marx’s Grundrisse der Kritik der Politischen Ökonomie of 1857–1858 we have no reason to take the labour theory of value seriously.

BIBLIOGRAPHY
Marx, Karl. 1902. Wage-Labor and Capital. New York Labor News Company, New York.

Wheen, Francis. 2001. Karl Marx: A Life. W. W. Norton & Company, New York and London.

1. Be careful with the Grundrisse. It's a scratch-pad of sorts, never intended for publication. It's useful for working out Marx's thought process at the time, but not so much for learning developed value theory. Plus, there's the "lack of clarity" issue that always exists when other people read things we write just for our own sake.

In this case, this manifests in the not-immediately-obvious point that Marx's whole musings on labor-money here were an extended rant on Proudhonists' proposals for labor-money.

Basically, the Grundrisse is something I would not recommend to someone still sorting out value theory; without that to put it all into context, it'll probably confuse more than it clarifies.

Regarding your (3), I've already pointed out to you that what you here call "average price" (i.e., what Marx calls "price of production") is explicitly and emphatically not the same thing as "average value" except if s accidentally equals p on average. The middle of the passage you quoted is quite clear: "But [value and price] are constantly different and never balance out, or balance only coincidentally and exceptionally."

(Also, if you don't understand what he means by "negation," just say so; most people can't translate LK-ese and will assume "claptrap" means there's something wrong with it.)

Finally, for your two issues at the end:

(1) You continue to miss the distinction between a) value as a social relation and b) the physical output of a process. As I have stated in the past, you're looking for something to make Marx's definition of value somehow logically necessary. This is hardly a demand we place upon other theories. In other fields, we recognize that scientific theories are something to explain the world, and that they don't need to explain themselves; in fact, they can't. As someone who has professed to like Popper, you should already know about his critique of justificationism.

There are, I think, very good explanations for value theory (which I have been sharing at length), but if you don't find them compelling, that's okay because it doesn't matter. Some people don't find theories of anthropogenic climate change palatable, either. But if the theory is logically consistent, which it is, then the next thing to do is evaluate it empirically, not "complain endlessly that you don't find it rhetorically appealing."

(2) You continue to miss the nature of this aggregation. We don't need to do it for the market, because the market itself does it; if it were not done, we would have no market. The classical economists before and during Marx's time did not view this as an assumption, but an empirical datum to use as a starting point — they were saying "by some means, it's clear that the market takes all these many labors and evaluates them all with the same standard quantitative unit." This became the springboard for further inquiry, including (which Marx does address!) why that might be.

Finally, you need to elaborate on your last point ("Neither real-world market wages or non-labour factor input prices...").

1. (1) " You continue to miss the distinction between a) value as a social relation and b) the physical output of a process. As I have stated in the past, you're looking for something to make Marx's definition of value somehow logically necessary. "

I am doing no such thing. This is a figment of your imagination.

The truth of an empirical proposition known a posteiori can only be defended -- at most -- as probabilistically true to a extremely high degree, not as necessarily true. I am demanding strong empirical support and evidence that can justify the LTV as probabilistically true to a high degree, e.g., like the empirical proposition that "cost-based pricing exists."

(2) "You continue to miss the nature of this aggregation. We don't need to do it for the market, because the market itself does it;"

The market does not do this. Yet again this is a bizarre figment of your imagination.

The market mostly creates prices based on (1) cost-based/mark up pricing (in administered price markets) or (2) supply and demand (in flexprice markets).

Show me --- with clear and strong evidence -- even one commodity price which the market has created based on a criterion by which all heterogeneous labour is reduced to common homogeneous unit which be used to measure and price the labour and factor inputs used to produce it.

2. **a posteriori **

3. Also this is just a typo:

"and the fact that it does not work this way contradicts and negatives"

"Negatives" should be "negates". I have fixed this typo in the post now.

4. (1) I am doing no such thing. This is a figment of your imagination.

Oh? Do you deny that you said the following?

"(1) it ascribes to human wage labour a special significance that it does not have (that is, if work expended by labour creates a value in commodities, then slaves, animals or the power of nature should be able to create such a value too)"

Because I can take a screenshot and highlight exactly where it appears in your post.

There are three interpretations I can conjure up for that. Either 1) you were arguing that all of those forms of labor are equally capable of producing output (thereby falling into the physicalist error I indicated), 2) that all those forms of labor are equally capable of producing the objects of subjective utility (i.e., use-values), thereby saying the same thing, plus committing the fallacy of equivocation on the word "value," or 3) you were arguing that a social construct does not require the social relations that so construct it, which is basically arguing for an uncaused cause. Given your aversion to mysticism, this seems unlikely.

I proceeded on the grounds that you had chosen the first of those because it's the most generous to you. Have I missed a fourth possible interpretation?

I am demanding strong empirical support and evidence that can justify the LTV as probabilistically true to a high degree

Strange, one wouldn't know it from the way you've summarily ignored empirical work I presented. I've mentioned this numerous other times, too, and you've ignored all those as well. Hence (a part of) why I suspect you haven't been arguing in good faith.

So now you're finally ready to concede that your problem with Marx is not theoretical but empirical? Please be clear, because I can definitely walk you down that road, too.

(2) The market does not do this. Yet again this is a bizarre figment of your imagination.

Nice of you to neatly trim my statement up to the point where I indicate exactly what sort of proposition we're discussing.

So, based on the above, you've denied outright the existence of a price system. Odd. Should I believe you when you say "the market does not [evaluate various heterogeneous labors in terms of a homogeneous (monetary) numeraire]," or when you follow it with this:

The market mostly creates prices based on...

So which is it? Is there a price system or isn't there? (Be careful, I argue for the former, and you might not want to be seen agreeing with a Marxist.)

Show me --- with clear and strong evidence -- even one commodity price which the market has created based on a criterion by which all heterogeneous labour is reduced to common homogeneous unit which be used to measure and price the labour and factor inputs used to produce it.

Okay. Here you go. You may find this hard to believe, but the remuneration of every one of those workers performing disparate tasks (ranging from engineering to welding) is assessed in the same terms: dollars. "IMPOSSIBLE!" you cry. But alas, it is already done.

It's seriously as simple as that. If you want it to be more complicated, then you'll have to invent a theory where it is and argue against that one instead. But it won't be the theory Marx or his contemporaries/predecessors were using.

I recall Mr. Vienneau tried to explain this to you, too, and he's someone who has concluded that "Marxist political economy should remain a live and exciting field of scholarly research." Funny how all the people who don't think Marx was spouting abject nonsense all interpret his work differently than you do. I wonder if that might mean something?

5. (1) slaves who labour to produce commodities that fetch an exchange value and profit are just as much a "social relation" as wage labour. There is no special status that wage labour has.

(2) I've seen no convincing empirical evidence from you to prove the LTV. Rather, you have been forced to defend it as a mere definition on more than one occasion.

(3) you are now so obviously dishonest that you want to suggest that the existence of a price system in homogeneous money units (which is real) somehow vindicates Marx's idea that all heterogeneous labour can be reduced to a homogeneous unit of SNLT and "labour time" money in Marx's statement above. The way you have distorted my objection is so disingenuous it is comical.

(4) in my challenge to you I obviously meant this: give me an example of one commodity price which the market has created based on Marx's criterion by which all heterogeneous labour is reduced to common abstract homogeneous unit of labour time which is used to measure and price the labour and factor inputs used to produce it in Marx's labour time money.

6. (1) Slaves' labor isn't counted as labor in capitalist accounts; the labor of the people who drive the slaves is. If it's accounted for as an owned machine, then that is the social relation that is ultimately represented.

(2) No, I've been forced to clarify it on multiple occasions against numerous straw men. I've provided empirical work and can do so again. But first, stop stalling and answer my question: Are you prepared to say your theoretical qualms have been answered, and regard the theory empirically from here on? Or are you going to keep shifting the goalposts?

(3) Homogeneous money units are what Marx uses most of the time. If you were less dishonest in your assessment of the material, you might have noticed this. I've also clarified this point, as has Prof. Kliman.

Glad you're amused, but now I need you to be edified.

(4) That is literally all prices. The fact that you think that this calls for some process other than the market process is a deficiency in your grasp of this subject.

Someone once said: "When the facts change, I change my mind. What do you do, sir?" Well, what do YOU do? In this case, it's an empirical proposition that the market puts everything into money terms — a statement of an observed phenomenon. It is through this that heterogeneous labors are brought into commensurability. The time has come to discard whatever preconception compels you to insist there's more to it than that.

And stop this "Marx's labor time money" stuff. I've already explained that the Grundrisse passage is part of a critique of his opponents' labor-money schemes. Playing dumb doesn't help your case.

7. (1) doesn't answer my point.

(2) neither my theoretical nor empirical objections have been answered.

(3) that Marx recognises money units as what goods are priced in doesn't vindicate the LTV.

(4) if you actually think that real world money prices reduce all goods to their SNLT values as defined by Marx you are a joker and a clown.

8. (1) Yes, it does. The "value" one analyzes in the capitalist mode of production depend upon capitalist relations. If you describe different relations, then the value relation is also different.

This is also why your repeated fixation on subjective value misses the point (setting aside for a moment the obvious fact that the existence of one theory does not suffice to rebut another). Marx is analyzing on the basis of abstract, class behavior. Though the concrete, individual values are subjective, the abstract value that one accounts for when one acts in the capacity of a capitalist is not subjective; it's the value one takes into account in the process of capital accumulation, which is objective in the exact same sense that one's profits or losses in a given quarter are objective. Marx chooses to call that value SNLT. You can call it whatever you like; I don't care about names, so long as we're actually discussing the same referent. Whatever nonsense thing you THINK Marx is referring to when he uses "value," throw it away and replace it with the italicized phrase above.

Buying a slave, as with a machine or an animal, will not suffice to facilitate capital accumulation. A machine must be operated, livestock corralled, and slaves brutalized or intimidated into performing their tasks. Further, the act of buying an owning the entity in question is also relevant, since this is an entirely separate dynamic (use and lifetime consumption) than the social control inherent to the wage-form.

If none of the above means anything to you, then maybe you should investigate why that might be, and start your process of learning at whatever more fundamental place to which those ruminations point.

(2) Well that just figures. Alright. Not gonna look a gift horse in the mouth, since at least you answered for once. SO! Please indicate what internal inconsistencies you believe exist with the theory. Let's knock the rest of them out of the park quickly, so we can move on to the more interesting empirical stuff (since, after all, a theory is excluded from possibly being empirically correct in the first place if it's inconsistent).

(3) I'm not trying to vindicate the "LTV" you and I both object to the title "LTV," so let's stop using it. The fact that Marx enumerates value in monetary terms in the majority of cases certainly does answer your concern re: labor aggregation; that's how it's done in the real live market. It's not enough to "vindicate" it against any potential complaint, but this one most certainly is answered.

(4) Seriously? "If you think differently than me you're a poopy head!" Good lord.

Every time you lower the bar I think to myself, this is it — we've hit rock bottom. And then every time, you limbo beneath it again. Tell me, when you were kicked out of debate club as a student, did you tell people that it was because you were "too good at debating"?

I won't beat around the bush: Your reading of Marx is facile garbage, and it has been from the outset. I've given you a great deal of my time to attempt to help you with this — initially in the most friendly, positive and constructive manner possible. These days, I'm just trudging through this hopeless exchange under a pall of ever-expanding disappointment. I was so sure you were better than this, but the scales have fallen from my eyes. Sometimes I wonder why I still bother.

But then I remember that, while you're entitled to your own opinion, you are not entitled to your own facts. If you want to make a million posts saying "Marx sucks, I hate Marx," then more power to you. But if you want to make posts claiming that "Marx's theory holds X" when in fact it holds Y, then someone's got to call you on that. Further, anyone with a shred of integrity would welcome being called on it without resorting to childishness.

9. Hedlund: Thanks for your interesting and informative comments on the TSSI interpretation of Marx's LTV. I know you'd prefer it if we didn't "punch left" in these comments, but how would you respond to the allegations that (a) the TSSI LTV is undetermined/inconsistent/has arbitrary assumptions, as argued in (Mohun & Veneziani 2009; Mongiovi 2002) among other publications, and (2) adopting the TSSI interpretation amounts to nothing more than "sterile nominalism" and a departure from the long-period methodology of the classical political economists and Marx (e.g. in Petri 2012, paragraph 9)?

Mohun, S & Veneziani, R 2009, "The Temporal Single-System Interpretation: Underdetermination and Inconsistency"

Mongiovi, G 2002, "Vulgar economy in Marxian garb: a critique of Temporal Single System Marxism", Review of Radical Political Economy vol. 34, no. 4, pp. 393-416.

Petri, F 2012, "On recent reformulations of the labour theory of value"

10. Heldlund@May 7, 2015 at 3:55 PM

(1) your argument is implying that you cannot accumulate capital without wage labour; this is so obviously false, it is laughable. Did no slave owners in history producing commodities ever accumulate capital?

(2) you already have them above.

(3) "The fact that Marx enumerates value in monetary terms in the majority of cases certainly does answer your concern re: labor aggregation; that's how it's done in the real live market. "

lol.. No, it does not. The market does NOT meaningfully reduce all heterogeneous labour to homogeneous SNLT units which are linked to a given money unit (or units) used to pay wage labourers and price factor inputs strictly in accordance with their SNLT. This is the only way prices could reflect SNLT. The market does not do this.

11. LK:

One of the most consistently informative aspects of our exchanges has been the questions or remarks to which you choose not to respond.

Do you have any thoughts on my point re: a capitalist acting in his capacity as such? (To say nothing of your persistent underhandedness in this discussion?) How come you never responded to my original request for elaboration on the last point in your post? Etc.

Let's get down to brass tacks. Do you deny any of the following:
- that there is a finite amount of social labor available for production?
- that a proportion of this labor goes towards each commodity?
- that the price for a given commodity may be greater or less than the monetary expression of this labor?

If you don't deny any of those things, then you might be a closet Marxist.

On to your... stuff:

(1) No, I'm not implying that at all. Lender's capital and merchant's capital both precede the capitalist system. However, I am stating in no uncertain terms that capitalist accumulation, the *thing that Marx's analysis centers*, is historically distinct from other modes by which wealth has accumulated.

Asking "did no machine owners in history producing commodities ever accumulate capital?" or "did no ranchers in history ever accumulate capital?" misses the point. Nothing prevents a slave owner from accumulating wealth, but to accumulate capital specifically — value in process, advanced with the aim of expansion — calls for particular relations. And when a whole system of reproduction is dominated by capitalist production, there are phenomena, laws, and tendencies that emerge that differ from other systems.

You really need to pay attention to these distinctions. I mean, unless you want to go around being wrong all the time, which so far hasn't seemed to bother you. I dunno, it would definitely drive me nuts, but to each his own.

(2) So, you don't have any theoretical problems; you're just being argumentative. If this is wrong, then I put it to you again: Please indicate what internal inconsistencies you believe exist with the theory.

(In case there's some confusion, this means ones I haven't already laid to rest; the beauty of text communication is that, yes, they're still above, but so are their responses.)

(3) lol.. No, it does not.

So you do deny that the market assesses heterogeneous things in monetary terms, thereby denying the existence of a price system. Fascinating. Is this a Post Keynesian idea, or just something you're working on, yourself?

Repeat after me: "'Abstract' and 'Concrete' are not the same thing." Concrete labor is heterogeneous. Abstract labor (as it is abstracted by capitalists) is.

You may as well be insisting that there's no way to know how many cars are in a parking lot, when some of them are Sonatas and some are Miatas and some are Civics... well, sure, that's true; there's no "car" car, just particular types of cars. But they're all still cars, and can all be tallied via that category. Similarly, all concrete labors belong to the category of "labor" in the abstract, even if there is no "generic" labor one can perform.

used to pay wage labourers and price factor inputs strictly in accordance with their SNLT

Seriously, LK, where the hell are you getting these cockamamie ideas? Just because the price of a screw (input) appears in the value of the chair (output) does not mean that the screw was purchased at its value. Sheesh.

12. Gaffers:

I know you'd prefer it if we didn't "punch left" in these comments

Nah, there's a big difference between what you're doing and the kind of sniping to which I was referring.

Anyway, the papers that you reference are part of a very long conversation, spanning decades. Each of them responds to one or more essays, and each of them has essays responding to them. For example, M&V 2009 was responded to in F&K 2009. Mohun 2002 has been discussed multiple times in numerous sources.

I don't know if Petri has had responses as such, but it seems he's only forwarding a summary of other arguments rather than his own. At a glance, there's much in his summary demanding clarification. For example: "The TSS approach argues ... that labour values depend on the historically performed labour" (p. 18). This is simply wrong, if it means what I think it does. But then again, I'm not sure I'm following his "fluid" analogy. That said, there's extensive argumentation on the long-period question (as well as the broader question of equilibrium) in the TSSI literature, too.

This is still probably the best and most concise single-source reference for the overall explication of the TSSI.

13. "'Abstract' and 'Concrete' are not the same thing." Concrete labor is heterogeneous. Abstract labor (as it is abstracted by capitalists) is."

Don't you mean:

"concrete labor is heterogeneous. Abstract labor (as it is abstracted by capitalists) is **not**."

Marx says clearly in quotations above that abstract labour units are homogeneous quantitative units.

14. "- that there is a finite amount of social labor available for production?
- that a proportion of this labor goes towards each commodity?
- that the price for a given commodity may be greater or less than the monetary expression of this labor?

if you mean:

(1) there is a finite supply of human labour in any specific nation
(2) that human beings' labour is divided between different occupations
(3) that the price of goods can be above or below unit labour costs,

then yes those propositions are true.

Asserting these as true does not vindicate the LTV. You haven't demonstrated how to meaningfully define an abstract SNLT unit and aggregate ALL types of labour. You have not shown how

(1) all labour is remunerated strictly according to its alleged SNLT contribution and

(2) how non labour factor inputs are priced according to the SNLT required to produce them.

And you certainly have not shown how output commodity prices are determined in accordance with (1) and (2) to correspond to aggregate SNLT.

15. Don't you mean: "concrete labor is heterogeneous. Abstract labor (as it is abstracted by capitalists) is **not**."

Yes, thank you. I had rephrased it before posting and forgot to edit that last word in.

(1) Specific nation is fine if you want, though properly we should do all we can to zoom as far out on the capitalist system as we can. But then, if we agree an individual nation is finite, then we can agree that the sum of the finite number of nations would also be.

(2) well, yes, it is also true that said labor is divided into different occupations, but that's not what I was saying. I'm saying that the clock, the chair, the car, and the cat food produced by society all involve a finite but definite portion of the total, such that the same labor can't be spent on multiple things. Labor expended on clocks cannot also be spent again on cat food, just as you can't spend your money twice. Otherwise, we'd get something from nothing.

Asserting these as true does not vindicate the LTV.

Who's trying to do any such thing? Screw Ricardo; I'm defending the propositions underlying Marx's explication of capitalist value. So far, you seem to agree that they're pretty uncontroversial.

(1) (can we please stop repeating numbers in the same comment when we make multiple lists? Maybe use A B C for the second one?) I don't know why I'd have to show "all labour is remunerated strictly according to its alleged SNLT contribution" when it is not a claim of the theory I am defending.

(2) this is also not something the theory claims.

As I keep saying, I seriously think your complaints boil down to simple confusion. Nothing demands that things be priced in line with their value; in fact, it's basically unheard of in any regular or intentional sense.

I think what you're misinterpreting is that output price becomes a component of the value of new output when it is used as an input. So, say a plank of wood has a value of \$5, and sells for a price of \$7. When it's used to make a picnic table, the value the plank adds is \$7 to the table's value instead of \$5, since value is determined by the conditions of reproduction; if I go into business making tables, I still have to acquire wood at \$7 per plank, which means I am investing \$7 worth of capital value on each plank. Accordingly, the plank's contribution to the table's value is not its output labor content as expressed in \$5, but the labor value required to obtain it — \$7 worth.

That is what makes the TSSI "Single System." It's also why I refer to it as a "continuum"; price and value are very different, but related, things — two "moments" in the same process, which condition one another.

2. Before saying that labour value is empirically flawed strenght, maybe you should read the studies that tested the empirical relevance of LTV. And all the results give surprisingly a strong empirical basis...
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2519466
http://hetecon.net/documents/ConferencePapers/2009Non-Refereed/Froehlich.pdf

3. LK, many thanks for your fascinating series on Marx, and the entertaining debates with Hedlund.
Unfortunately my brain grinds to a painful halt whenever I try to understand the issues regarding LTV. Should I persevere?
Does the purported empirical "evidence" referenced by Anonymous above have any policy significance?
Or are all these effusions of intellectual brilliance merely manifestations of a misallocation of academic resources?

1. His purported empirical evidence seems to reduce to the claim that labour costs form a large part of cost-based prices. This does not vindicate the LTV. So, no, I wouldn't bother.

2. KongKing:

I actually agree with LK, here. Empirical studies attempting to argue for a "labor theory of price" are flawed.

Rather, they're certainly entitled to forward the theory, but it's not Marx's.

4. Hi LK,

This is a very interesting discussion. I am on weak ground when it comes to the LTV. I just don't understand it, if I am honest.

Joan Robinson thought that it was labour metaphysics that should be jettisoned by Marxists but that economists should give attention to the kind of questions Marx was asking in Capital concerning the movement of economies in historical time

As for the LTV ,if means of production have past labour time embodied. in them, and if all value added to the raw materials and natural resources (the objects of labour) are due to present and past socially necessary labour, then would not all commodities exchange at their cost of production (in the long run)?

But Marx says in Capital some commodities exchange above and some below their long run 'normal' of 'natural' prices and the day to day prices are determined by supply and demand.

If some commodities long run exchange rate is above the socially necessary labour time and some below, how do Marxists establish that the sum of the deviations of values from long run prices are zero?
Has the transformation problem been solved or not?

What happens if there are long run decreasing returns to scale in some industries, constant long run returns to scale in others and increasing long run returns to scale in yet others? Why would the sets of returns balance each other so that ,on average, labour value would equal cost of production plus a mark up, for say, normal profit?

John Arthur

1. Hi John. Two quick clarifications:

1) The aggregate equality of value and price is the result of exchange, not production. Marx was enamored with double-entry bookkeeping ("Italian" bookkeeping, as he and Engels called it), and this is the basis of it; that everything comes from somewhere and goes somewhere such that all the rows and columns sum to zero. It's just an accounting identity.

2) The transformation problem is the result of particular contrivances in the way people interpret Marx. If you argue that value and prices need to be accounted for using two separate systems rather than a continuum, or that input and output prices are evaluated simultaneously, rather than through time, then you become unable to replicate his results. If you take the other route, you can. So I'd argue that it was something read into him, rather than something "in" the work.

At the very least, we can say that each interpretation corresponds to a different theory, and one of these theories has no transformation problem. Incidentally, that is my favorite one.

5. Hi Hedland,

(2). If "The aggregate equality of value and price is a result of exchange, not production" and is simply a part of double entry book keeping, why do Marxists spend so much time on SNLT? Why not simply go with long run prices?

(2) If some commodities exchange at values above and others below their long run prices, accounting identities would seem to account for the sum of the deviations of values from these prices summing to zero, EX-POST.

(3). Production decisions are made on the expectation that firms will make a profit. In the process of Production M-C-M' value is created including surplus value. In the process of exchange, M-C-M . this surplus value is realised, provided the commodities produced actually sell. So, EX-ANTE, why would the sum of the deviations of values from long run prices
equal zero? Ex-ante, why would long run prices and SNLT embodied in commodities, on average, be equal?

John Arthur

1. Hi John.

Long-run prices are nifty, but the whole point of the analysis is to get at the barest relation that underlies the phenomenon of price within the capitalist system. Remember: at the time Marx was writing, people viewed the entire science differently; the nature of the inquiry centered on how societies become wealthy and reproduce themselves over time, whereas modern economics is much more centered on the process of distribution under scarcity. Whereas every person is unique, the question of reproduction therefore lends itself to a focus on relations that transcend a particular individual. In a given period, someone makes, say, socks, even if it's not the same person as last time. Analyzing roles in this fashion leads to analyzing classes and how they relate, and what their relations tell us about value, etc.

2) right. This stems from the point that value, of the specific sort that accumulates as capital, is not created through exchange. Moreover, anything consumed must have been produced. That is precisely what we seek to account for. In the abstract, society has x total value in a given period. If you produced fraction n of it (=nx) and yet the price you fetch for it enables the consumption of 2nx, then one or more people in the system are left to consume a sum of nx less of the total output. Price alone can't tell us about this.

3) again, because no value is created in circulation (with simple commodity exchange represented C-M-C). I believe you can find the full argumentation in chapter 5, volume 1 of Capital; basically it'd be a fallacy of composition. Plus, as you say, without realization there can be no value or surplus value; inventory that never finds a buyer is no better than the classic "mudpie." So we're still left with a system either way with discrete totals of realized prices and abstract labor time that yielded them.