In reply to Andrew's OPE-L:5212:
>Duncan: "I'm still not sure what "replacement cost interpretation" means in
>terms of the mathematical and accounting
>conventions."
>
>If P(t) is the vector of input prices of period t, P(t+1) is the vector of
>output prices, A is the vector of consumed means of production (including
>perhaps depreciation), then the pre-production *reproduction cost* of the
>means of production is P(t)A. This is a reproduction cost, because the means
>of production may actually have been acquired at time t-k, and warehoused
>until period t. But because the value of the commodity (not the means of
>production, but the *product* produced by means of them) is determined by the
>cost of REproducing it, it does not matter when the means of production were
>acquired. The value of the commodity is not P(t-k)A + L (where L is living
>labor), but or P(t)A + L. The value transferred is not P(t-k)A but P(t)A,
>
>The *replacement cost* of the means of production is P(t+1)A. But P(t+1)A
>+ L
>is not the amount of labor time needed to reproduce the commodity in
>period t.
> It is a hybrid of the amount of living labor needed in period t, plus what
>the means of production would cost NEXT period.
This clarifies your meaning, but just returns us to the same disagreement
over the treatment of the revaluation of stocks of assets held during the
production period.
Have you thought about this assuming that, while the production period
stays constant, the accounting period becomes smaller and smaller? It's
nice to have the mathematics of an accounting convention invariant to this
type of transformation of the treatment of time.
Duncan
Duncan K. Foley
Department of Economics
Barnard College
New York, NY 10027
(212)-854-3790
fax: (212)-854-8947
e-mail: dkf2@columbia.edu