I am having troble deciding between two theories of value: scarcity need and labor cost of production. I found by accident a long, long thread on this list about the labor theory of value where B. DeLong said "a commodity has (exchange) value if it's scarce regardless of whether there is wage labor. M Perelman replied "a commodity has value if it embodies labor even if it is not scarce".
"Labor cost of production" theory is most coherently understood as a (very) special case of a scarcity-based theory of value. One way to see this is to note three significant caveats to Michael P's reply to Brad DeLong. First, as Marx himself insists, a necessary condition for a commodity to have value is that it have use-value (that is, people "need" it) [Capital V.I p. 131,]. So as Marx says, even if a commodity contains labor, it doesn't have value if there isn't demand for it. [And if you think about it, this statement would need to be further qualified to make any sense, i.e.: there would need to be demand for the good *at its value-determined cost*, since *categorical* lack of demand and zero quantity demanded *at a given cost-price* lead to the same outcome.]
Second and more generally, Michael's response obscures the point that the labor embodied in a commodity is itself a reflection of scarcity, that is, of expenditure of a resource that has alternative economic uses. Marx makes just this point in his 11 July 1868 letter to Dr. Kugelmann defending his formulation of value theory in _Capital_: "Every child knows too that the mass of products corresponding to the different needs [of a country] require different and quantitatively determined masses of the total labor of society. That this necessity of distributing social labor in definite proportions cannot be done away with by the particular form of social production, but can only change the form it assumes is self evident." Thus, to withdraw a portion of the "total labor of society" away from some line of production in order to devote it to another is deny some form of "needs" being met. Labor allocation matters because labor is economically scarce, and couldn't possibly matter if labor weren't ultimately scarce in this way.
Marx's statement above gives away quite a lot to a scarcity-based conception of value, but even so it is based on some very restrictive assumptions, in particular 1) that the "total labor of society" is given (lump of labor fallacy, anyone?) and thus invariant to relative prices; 2) that the "different needs" of a society are invariant (and thus invariant to relative prices); and (3) that meeting a given configuration of social "needs" for consumption goods implies a specific allocation of social labor "in definite proportions," denying the possibility of varying embodied labor requirements through changes in production technique. It's at best not obvious that any of these conditions hold in general. And if they don't, the "labor cost of production" theory of value becomes problematic at best, and typically incoherent.
More specifically, as Jon's comment affirms, labor value theory is a *production*-based theory of value, i.e. it abstracts from *levels* of commodity demand. However, a *necessary* condition for commodity values--let alone prices--to be invariant to the pattern of demand is that the stringent conditions of some form of "nonsubstitution theorem" obtain (see. e.g., the entry on this term in the New Palgrave Dictionary of Economics). This type of theorem requires in general three conditions: (a) the rate of profit determined outside the system that determines commodity prices and wages; (b) constant returns to scale; and (c) only one scarce (N.B.!) primary input (e.g. labor). Assumption (a) is challenged by Marx himself in Capital V. I Ch. 25 on the general law of capital accumulation, where he allows that relative demand for labor affects the wage and *and thus* the profit rate. Concerning (b), absent constant returns to scale at industry levels, the level of prices will in general depend on the pattern of demand; for example, if an industry's long-run supply curve is upward sloping, then the relative price of the commodity it produces will increase with market demand for that commodity, other things equal.
Note in addition that the conditions of the nonsubstitution theorem, stringent as they are, are *weaker* than those required for the specifically *labor*-based theory of value, since output prices may be invariant to the pattern of demand, and yet *embodied labor values* might vary depending on the cost-minimizing choice of production technique, implying that commodity values are dependent on commodity prices rather than vice-versa. John Roemer makes this point in detail for the case of "von Neumann" production conditions [a generalization of constant-returns technology] in Ch. 5 of his _General Theory of Exploitation and Class_.
Finally, condition (c) relates to my third caveat re Perelman's response to DeLong, i.e.: not only is it problematic on its own terms, per the above, but it's also not a response to what DeLong said, which is that goods can have exchange values even if they have no embodied labor time. Take for instance the case of lots of unimproved land, which often have positive exchange values in modern economies, even though no labor has been expended in creating them. [This possibility also serves as a rebuttal by counter-example to Marx's argument for the labor theory of value in Capital V. I, pp. 127-29: given that lots of unimproved land can and do exchange with products of labor, it is manifestly not the case that exchange values "express something equal" in the sense Marx's argument requires.] If we allow further that "land" is, along with labor, an input in commodity production--which of course it often, if not typically, is--then condition (c) falls, and commodity prices, and perforce values, will be dependent on the pattern of demand.
Bottom line: to the extent that commodity labor values matter, they are an index of economic scarcity, but as such they are incomplete and inessential under general and arguably "real-world" conditions. Thus, the "labor theory of production" is at best understood as a very special case of a scarcity-based theory of value.
FWIW,
Gil
But marginalists since Jevons have claimed that the labor affects the value of a commodity only to the extent that the commodity becomes scarce - a commodity that can be produced in 5 days will be more scarce that one produced in 1 hour. Neoclassical economic theory claims to discover universal economic rules and thus diregards whether the costs of production are born by labor or some natural force. What for Marxism only appears as a relationship between things is really a relationship between things for the Marginalist because natural scarcity not a social relation. What is Marxism's objection to this characterization of value? (Here's what Lionel Robbins says economics is about:
"Human behavior as a relationship between ends and scarce means that have alternative uses". cit in Sweezy, Theoy of Capitalist Development.)
Could anyone suggest works (preferably from Marxian perspective) that deal with this issue very plainly? (I'm reading the hilferding/bohm-bawerk debate - difficult for the amateur. Where does Marx deal with this in Capital? )
Thank You,
Jon
Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+ countries) for 2ยข/min or less.
