> I'm not sure I understand Jim's references to Marx's
> "silliness" in assuming that prices are proportional
> to (labor) values, or to the "so-called labor theory
> of value". Shaikh has showed, by reference to both
> US and Italian input-output tables, that values and
> prices are indeed very close to proportional, with
> R^2's on the order of 96-98 per cent. (In "Ricardo,
> Marx, Sraffa", E. Mandel, ed., 1984). Petrovic and
> Ochoa (CJE, 1987 and 1989 respectively), confirmed
> this result on Yugoslav and further US data. Paul
> Cockshott and I have recently replicated Shaikh's
> study using the 1984 UK input-output tables: our
> findings were essentially the same. Ricardo and Marx
> were right: the labor theory of value stands up to
> empirical scrutiny as well as just about any theory
> in economics, and better than most!
>
I was initially struck by the results of Shaikh and his students, but
believe they should be taken with several grains of salt; certainly
not as confirmation of the LTV. It just means that as a matter of
empirical fact, the distribution of technical compositions of capital
across sectors is such that values don't diverge much percentage-wise
from prices of production.
Certainly these results don't refute the Steedman/Roemer argument
about the essential superfluity of labor values.
They don't do anything to shore up the (il)logical underpinnings of
value theory, either. Gil [[EMAIL PROTECTED]]
