1. Now I want to make a point that requires development in several
directions, but which is in itself very simple.
2. Suppose that labor-values and Sraffian prices are, as I have said,
about equally good as predictors of actual prices. Are there then any
grounds for preferring one theory over the other? There are several,
but here is one perspective to start with.
3. Although quite unlike Walrasian General Equilibrium theory in many
ways, Sraffian theory has something in common with the latter. If one
asks, "What determines prices (or price movements over time)?" the
answer in both cases is, more or less, "Everything." (The full set of
preference orderings, technological options and endowments in the case
of Walrasian GE; the full set of technical relationships and the profit-rate
or wage in the Sraffian case.) This answer is strikingly uninformative.
4. But ask the same question of the LTV and you get a clear,
informative answer: the systematic component of both cross-sectional
and time-series relative price variation is primarily governed by the labor
time socially necessary to produce the various commodities. On the
grounds of parsimony in explanation (Occam's razor), the LTV looks
pretty good.
5. Why has the relative price of computing power fallen so dramatically
over the last decade? The (testable) explanatory hypothesis of the LTV
is that technology in the computer industry has progressed in such a way
as greatly to reduce the labour time necessary to make a computer of
given specification (while also, of course, upping the specifications
dramatically). By contrast, the Sraffian answer -- it would presumably
go like this: the input-output structure of the economy has changed in
such a way as to reduce the price for computers consistent with the
computer industry earning the average rate of profit -- seems not to
offer any real explanatory purchase at all.
End of posting the fifth.
Coming up: Does the LTV have a credible mechanism?
==========================
Allin Cottrell
Department of Economics
Wake Forest University
[EMAIL PROTECTED]
(910) 759-5762
==========================
