Re: [OPE-L] Che Guevara and the Sraffian notion of profit

From: Ian Wright (wrighti@ACM.ORG)
Date: Mon May 07 2007 - 13:44:46 EDT

> "the big problem is that Sraffas matrices, unlike Heisenbergs, are not unitary, and as
> such do not express conservation relations."

I think this is a very important point, but the root cause of
non-conservation in PCMC lies elsewhere. The lack of conservation in
Sraffa's matrices arises from his transition from Ch.1 to Ch.2, i.e.
production for subsistence to production with a surplus. This is
because Sraffa tries to combine two approaches:
(i) The concept of a circular flow, in which a network of economic
relations are formalized in terms of simultaneous equations; and
(ii) the concept of an undistributed surplus; a surplus that, by
definition, does not have a corresponding cost.

An undistributed surplus breaks the symmetries between costs and
revenues (whether nominal or real) that obtain in a circular flow.
Hence Sraffa's matrices are non-unitary, and conservation is broken.
There are revenues that have no corresponding costs (i.e. "matter
appears from nowhere").

But Sraffa immediately restores the symmetry in the price system by
specifying a nominal distribution of income. Hence Sraffa's price
equation is conservative: all costs and revenues simultaneously

Yet Sraffa does not restore the symmetry of the real cost system (e.g.
labour-values) by specifying a real distribution of income.

If we do this the resulting matrix is unitary. We get a closed system
of simultaneous equations. But contra Sraffa, this does not imply that
there is no surplus or that the economy is subsistence. It merely
implies that the surplus is distributed, and is now part of the normal
cost structure of the economy.

I think that trying to theorise the consequences of a
symmetry-breaking event that results in an undistributed surplus using
the tools of simultaneous equations is a non-starter. So I agree with
Joan Robinson when she says that Sraffa had "half an equilibrium


This archive was generated by hypermail 2.1.5 : Thu May 31 2007 - 00:00:08 EDT