Re: measurement of abstract labor

From: Allin Cottrell (
Date: Mon Jul 19 2004 - 19:33:07 EDT

On Mon, 19 Jul 2004, Rakesh Bhandari wrote:

> A=B
> B=C
> Therefore, A=C
> xdollar equals y of a composite commodity that requires z labor hours
> to produce; hence x dollars is equivalent to z labor hours. From this
> equivalence one can calculate the MELT and the "value of money".

Why would one use such a roundabout and error-prone means of
calculating the MELT?  The obvious approach is Foley's: take the ratio
of the money-value of net output to the aggregate labour time embodied
in that net output.  This is both easier -- given the available
official statistics -- and more accurate.  Your proposal contains an
arbitrary and potentially fluctuating element, namely, the
relationship between the price-to-value ratio for a "small" composite
commodity such as the one Greenspan allegedly uses, and the
price-to-value ratio for social output as a whole.

>> The dollar has for several decades been the "nearest substitute for
>> world money", and the "attendant privileges" have been evident,
>> despite the fact that it has depreciated more or less monotonically
>> against most baskets of commodities (and a fortiori against labour
>> time, since the labour time required to produce most commodities has
>> also been falling more or less monotonically).
> Yet because the dollar is maintained as equivalent to a certain
> quantity of a composite commodity that itself has value it can then
> measure the value of other commodities. Otherwise, one would not know
> how many slips of worthless paper to demand for alienating this or
> that.

I think you are alluding to the "signal extraction" problem discussed
by Milton Friedman and the new classical macroeconomists: if the
aggregate price level is unstable, how can agents distinguish between
absolute and relative price changes?  They can do so only in a
probabilistic manner, and the efficiency of microeconomic calculation
is impaired.

This problem exists at a theoretical level, but as a practical matter
it does not seem to be a big deal so long as inflation is "moderate"
(say, no higher than double-digits).  It takes a severe hyperinflation
before the monetary role of fiat currency is seriously compromised.

Do you think that capitalists calculate how much they ought to be
paying for a commodity by first figuring its labour content then
converting that using a MELT?  That does not seem plausible to me,
except perhaps for services where the labour content is relatively


This archive was generated by hypermail 2.1.5 : Wed Jul 21 2004 - 00:00:01 EDT