In ope-l 4410, Paul C. writes
"The paper to be presented by Allin and I to the
EEA value working group is devoted to applying
the Kliman and Sraffian transformation formulae to
the US economy using fixed capital figures from
the BEA along with the a reduced version of the
1987 commodity use table to provide the circulating
capital. We end up with a 49 sector model, and
evaluate the relative predictive powers of the
different transformation formulae."
I have already --- almost two years ago --- told Paul that the TSS
interpretation of Marx's transformation is simply not intended as a predictor
of actual prices. Therefore it cannot be tested in this manner, and any
alleged test along these lines is meaningless.
One main theoretical result of Marx, obtained in Ch. 5 of Vol. I, is that
value can be redistributed in exchange, if commodities sell for more or less
than they are worth. His transformation account in Ch. 9 of Vol. III confirms
this. Differences between prices and values thus arise due to tendencies to
equalize profit rates, supply and demand imbalances, and a host of other
factors. There is no a priori theoretical reason to expect these differences
to be large or small. Hence, there is no a priori theoretical reason to
expect that values will be good predictors of production prices or of market
prices.
There is also no a prior theoretical reason to expect that production prices
will be good predictors of market prices. I have also pointed out to Paul
that although market prices are said by Marx to fluctuate around production
prices, this does not imply that the deviations are small. Hence, the
correlation coefficient does not test Marx's proposition.
I hope that the paper which he and Allin will present will discuss these
issues and the conclusion to which they lead, that his empirical results are
without implications for the theoretical questions addressed by Marx's
transformation and the TSS interpretation of it.
I say all this even though I recall that interpretation of the transformation
"performed better" than the "Sraffian," according to the estimates reported
two years ago.
Andrew Kliman