Value Theory (2): TSS Paradigm (Corrected)

Les Schaffer schaffer at
Wed Mar 7 12:33:41 MST 2001

[ bounced from unsubbed Rakesh Narpat Bhandari <rakeshb at Stanford.EDU> ]

Dear Sid,

I agree with much of what you say here.  Just wanted to clarify that
the TSS demonstration only shows the possibility of rising
productivity resulting in a general crisis from a falling rate of
profit; it does not show that such a outcome is plausible or
probable. I have been putting standard objections to Marx's theory of
the falling rate of profit (many of which were dealt with by
Grossmann); I surely don't think the tendency is as weak as George S
seems to think.

Just want all the reasons for skepticism to be on the table. Only one
set of objections has been provided by simultaneist valuation--which
seems to be nothing more than the methodology of comparative statics
with its characteristic neglect of transitional processes and eschewal
of a true causal explanation of the movement of a system over time.

For someone not trained in the intricacies of Paul Samuelson and John
Hicks Thought, the simultaneism of comparative statics seems the least
interesting of the objections to Marx's value theoretic explanation of
the tendency for the fall in the rate of profit. But this is not to
say that the critique of the methodology of bourgeois economics by
Ernst, Freeman, Carchedi, Giusanni and others is not a major advance
in getting our interpretation of Marx's own theory correct.

I am also hesitant to put Sweezy and Shaikh in the equilibrium camp.
That the former seems to think that there should be an equilibrium
vector for prices of production in the case of simple reproduction
does not mean that he thought the economy could ever settle down into
simple reproduction (though this is an error implicit in some radical
theories of captialist stagnation or unemploymet equilibrium). One can
only conclude that Sweezy thought that the transformation procedure
should be generalizable to all cases, the easiest of which to handle
in mathematical terms is an economy in simple reproduction.  But this
does not make him an equilibrium thinker.

I think we agree that the error made by Bortkiewicz and Sweezy is the
assumption that the inputs are in the form of commodity values. It is
claimed that in the first of the the two transformation tables in
Capital 3, ch 9 we find the circuit C-C' while in the second table we
supposedly have C-M' since Marx has only transformed the outputs from
the form of commodity values into prices of production.

In the case of simple reproduction, where both the input and output
prices of production are equal, the complete transformation will
decouple the price of production scheme from the so called value
scheme. This is the basic claim of Marx's critics.

In Marx's transformation, the total value and total surplus value in
the first table equal (or actually determine) total price of
production and total profit in the price of production scheme,
respectively.  For Marx this demonstrates that the average rate of
profit not only does not negate the law of value, the average rate of
profit itself becomes the form in which the law of value asserts

But this demonstration collapses--it is claimed--if not only the
outputs but also the inputs are transformed from commodity value (C)
form to the price of production form (M'). It is no longer possible to
show that value or labor time relations determine price phenomena.
Once the inputs are included in the transformation procedure either
total price will not be equal to total value or total profit will not
be equal to total surplus value. Price phenomena, the outward
appearance, are no longer explainable in terms of value relations, the
inner essence.

>From Bortkiewicz to Samuelson, the "transformation problem" has become
the major criticism by bourgeois economics of Marx's theory of
value. In fact, the Bortkiewicz-Sweezy transformation calculations are
claimed to be as destructive for the postulation of value as the
Michelson Morley experiment was for the hypothesis of ether.

And this verdict has been given the imprimatur of the Anglo American
world's leading Marxists--Sweezy, Meek and Dobb. The last two came to
show sympathy for an economics without value theory at all, and so
Marx came to an end within the academy (except an enclave here or
there) with the approval of the leading academic Marxists. But this
outcome can all be traced back to Paul Sweezy, as a Harvard professor,
giving his stamp of approval to the statistician Bortkiewicz's
misinterpretation of Marxian theory (the Bohm Bawerkian subjectivist
critiques of Marx's value theory had been successfully fended off by
Hilferding, Boudin, Bukharin and Wm J Blake). When Dobb as a Cambridge
don also threw the towel in, the game was lost. But it took the
alchemy of Harvard and Cambridge credentials to spirit Marx's own
value theory away from the academy.

So the critics of this misinterpation--Paul Mattick Sr and Jr, the TSS
school, Fred Moseley, Geoffrey Pilling, Moishe Postone--face an uphill
battle.  Following Mattick Jr, Moseley and Alejandro Ramos, we agree
that the inputs in Marx's first table are not in the form of commodity
values. They rather represent the money sums which have been laid out
as constant and variable capital. The inputs thus don't have to be
transformed into the money price form.

Unlike Alejandro and Fred, I do think however that there is an error
in Marx's transformation tables, as Marx himself underlines in Capital
3, p. 265 (Vintage) But it is not that the inputs are in the form of
commodity values. It is rather that Marx had initially assumed that
the value transferred from the machine to the final product is exactly
matched by the visible flow price of the machine.  What I think Marx
is saying is once we understand why value and price of production
diverge, it's clear that the value of the consumed means of production
had diverged from their price, so he was wrong to assume that the
value transferred from the machine had been the same as the flow price
of the machine. But my interpretation is novel, and it has no
destructive implications for Marx's value theory.

I also have a lot to say about Shaikh's iterative solution. I do
believe if surplus value is defined as Marx himself defined it, an
iterative solution can be shown to respect both equalities. That is,
even if one assumes that the inputs are in the form of commodity
values and have to be transformed into the same unit prices of
production as the outputs, one can maintain both equalities (though
Shaikh and Jacques Gouverneur do not argue this themselves).  But I
have yet to write this up as publishable scientific paper.

As for the lynchpin of the TSS (temporal single system) theory, it is
undoubtedly correct: in the study of the formation of prices of
production and techncial change, one should dispense with the
assumption of constant values or input prices=output prices.

Which means that I agree that Marx's prices of production are not
centers of gravity in the sense of long term equilibrium prices. As
Marx argues in Capital 3, ch 10 what becomes more effective over time
is the capitalists' sense of their social power--their claim on
society of an average rate of profit on their investments. As capital
and labor become more mobile, they are able to collectively exert this
specific power over society. But this says nothing about market prices
converging on new long term equilibrium prices. The tendency towards
the equalisation of profit rates (and Marx himself underlines that it
works in conjunction with ever renewed inequalities in the profit
rate) is logically independent of any tendency towards the formation
of long term equilibrium prices.

Prices of production are counterfactual prices--what the prices would
have been in any one period had the profit rate equalized and demand
equalled supply.

In each period prices will oscillate around what prices would have
been had the profit rate equalized and demand equalled supply--that
is, prices of production.

But there are many other factors working on prices, and following Karl
Korsch, I don't think Marx had any intention in descending into the
world of actual market prices.

Some seem to justify the use of stationary prices because they serve
as real centers of gravity over the long term--an ontological
justification, to use terms as David Laibman has; while others jsutify
the use of stationary prices because they allow for the simplest
models--an epistemological justification. Marx does indeed assume
constant values on epistemological grounds in vol 2 (in both the study
of the circuits of capital and reproduction).

But in the former case, Marx is assuming constant values just so he
can lay out analytically the moments in the circuit (he of course
himself underlines that revolutions in value do actually lead to grave
disturbances in the circuit).  In the case of reproduction, Marx seems
to be arguing much like in equilibrium theory that under specified
conditions which help to render the problem cognitively tractable it
is possible for expanded reproduction not to founder on a permanent
consumption deficit.

Marx may be wrong on his own terms, as Rosa L argued; or as with
equilibrium theory, the conditions may be so restrictive (annual
turnover of fixed capital, uniform OCC, constant values, etc) that the
demonstration may not even have heuristic value and leap into a
Platonic, noumenal beyond.

Of course this may be Marx's point: there are so many possible moments
for disturbance in expanded reproduction even assuming constant values
and annual turnover of fixed capital that it is unimaginable that
capitalist development could ever take the form of equilibrium growth,
though Rudolf Hilferding wrongly read vol II as a demonstration of
this very possibility.  Demonstrating that vol II is, if anything, a
proof of endemic disequilibrium is one of the real strengths of
Meghnad Desai's Marxian Economics, though the argument was first
developed in Henryk Grossman's 1941 book about dynamics.

TSS seems to give good reasons why however in the study of technical
change, which is an inherently dynamic phenomenon, the epistemological
assumption of constant value or stationary prices simply makes no
sense. It doesn't make the problem cognitively tractable; it simply
wipes away the problem of study. You have indeed provided a
brilliantly lucid demonstration of this.

Yours, Rakesh

More information about the Marxism mailing list