Wednesday, January 28, 2009

Samuelson Jests

I might as well document some jibes Kevin Quinn finds amusing. These are from two different articles from two different periods:
"I have dealt with Karl Marx the economist, not Marx the philosopher of history and revolution. A minor Post-Ricardian, Marx was an autodidact cut off in his lifetime from competent criticism and stimulus."-- Paul A. Samuelson (1957) "Wages and Interest: A Modern Dissection of Marxian Economic Models", American Economic Review, V. 47 (Dec.): pp. 884-912

"The 'transformation algorithm' is precisely of the following form: 'contemplate two alternative and discordant systems. Write down one. Now transform by taking an eraser and rubbing it out. Then fill in the other one. Voila! You have completed your transformation algorithm.'" -- Paul A. Samuelson (1971) "Understanding the Marxian Notion of Exploitation: A Summary of the So-Called Transformation Problem Between Marxian Values and Competitive Prices", Journal of Economic Literature, V. 9, N. 2 (June): pp. 399-431

"The truth has now been laid bare. Stripped of logical complication and confusion, anybody's method of solving the famous transformation problem is seen to involve returning from the unnecessary detour taken in Volume I's analysis of values. As I have cited in my mathematical paper, such a 'transformation' is precisely like that in which an eraser is used to rub out an earlier entry, after which we make a new start to end up with the properly calculated entry." -- Paul A. Samuelson (1971) "Understanding the Marxian Notion of Exploitation: A Summary of the So-Called Transformation Problem Between Marxian Values and Competitive Prices", Journal of Economic Literature, V. 9, N. 2 (June): pp. 399-431
Samuelson refers to:
  • Paul A. Samuelson (1970) "The 'Transformation' from Marxian 'Values' to Competitive 'Prices': A Process of Rejection and Replacement", Proceedings of the National Academy of Sciences, 67(1) (Sept.): pp. 423-425.
As far as I know, no economist responded to Samuelson’s paper in the 1950s. He had reaction in the 1970s. The JEL published replies by Abba Lerner and Michio Morishima; an editorial comment by Mark Perlman; an analysis by Martin Brofenbrenner of submitted but unpublished reactions by Gordon Bjornson, Jean Cartelier, Bruno Jossa, David Laibman, Paul Massick, and Murray Wolfson; a paper on Marx by William Baumol; and responses by Samuelson, including an interchange with Joan Robinson. I suggest the difference in the number of reactions has something to do with events in non-academia. Perhaps the sixties had some impact on the algebra in which some economists were interested.


Ian Wright said...

Hi Robert

Samuelson's conclusions do follow from the math, which is why he can relax and jest, although his argument is simply a modern re-formulation of a truth that Ricardo understood very well: competitive prices depend on the distribution of income yet labor-values -- when defined solely as a function of the technical conditions of the economy -- do not. So there cannot be a conservative transform between these two accounting systems (the transformation problem). And therefore relative values (prices) can vary independently of absolute values (labor-values) due to changes in the distribution of income (which in turn is the root cause of Ricardo's difficulties of finding an invariable measure of value).

The tragic aspect of the debate is that it rests on a false premise: that labor-values *should* be defined solely in terms of the technical conditions of production. They shouldn't be, and don't have to be. The root cause of the classical contradictions are conceptual issues.

I hope you don't mind but I've included a link to one of my papers that expands on the above:

Anonymous said...

Is the transformation problem really a problem? If we take exchange value as a good heuristic analysis device then is there an issue?

By which I mean exchange value can be used to give an insight into how prices are formed rather than the prices as such, so allowing insights into the dynamics of capitalism and how exploitation happens.

After all, Marx did note that in reality prices did not equal labour values:

"The continued oscillations in prices, their rise and fall, compensate each other . . . and carry out their own reductions to an average price, which is their internal regulator . . . the problem of the formation of capital . . . [is]: How can we account for the origin of capital on the supposition that prices are regulated by the average price, i.e., ultimately by the [labour] value of the commodities. I say 'ultimately,' because average prices do not directly coincide with the [labour] values of commodities." [Capital, Vol. 1, p. 269fn]

So, a simplifying assumption helpful to understand the nature and dynamics of capitalism. Although a relatively realistic one, I would add.

I should also add that Steve Keen (in Debunking Economics) notes that the neo-classical marginal productivity theory only works if we assume that all industries have the same amount of capital. Given that differing amounts of capital cause the transformation problem, that similarity between neo-classical theory and what makes Classical theory allegedly non-viable is somewhat ironic.

An Anarchist FAQ

Ian Wright said...

Yes market prices never equal labor values.

Many vulgar critiques of the classical labor theory of value are based on this observation and therefore entirely miss the mark. Labor-values are attractors for market prices, they explain their direction of movement, and the fluctuations of supply and demand are the mechanism by which prices move toward this attractor. But Marx did not claim that this end point is ever reached, especially since technical change alters labor-values.

Marx does assume, however, price-value proportionality in Volume 1 since he is concerned to explain the origin of profit in exploitation rather than unequal exchange. So he assumes, at this level of theoretical abstraction, that the end point has in fact been reached.

The transformation problem arises at this level of abstraction, although there a delicate issues of interpretation.

What is at stake in the transformation problem? The following:
- the logical coherence of a LTV
- the claim that profit is unpaid labor time
- historical materialism
This is hard to get across in a short message. But I think it's not OK to avoid the issue by recourse to the LTV being a "simplifying assumption". Instead, the issue goes right to the heart of the fetish, i.e. whether money represents abstract labor.

I recommend
Lippi, M. (1979) Value an Naturalism
which lucidly explains why the transformation problem, rather than a curiosum, is of singular importance for the materialist conception of history.

I should emphasize that I *completely disagree* with Lippi's conclusions, but that I think he hammers home the theoretical importance of this particular issue.

Robert Vienneau said...

I once engaged in an interchange where I argued the importance of the transformation problem by pointing out that that is where many have staked their claim for the (in)consistency of Marx's economics. My interlocutor was not convinced.

I think Marx provides some worthwhile ideas about economics independently of the validity of the LTV, including some of the Volume 3 invariants. Both Joan Robinson and Paul Samuelson seemed to think Marx's Volume 2 analysis of interest, perhaps better if formulated in terms of price of production.

I'm in agreement with Ian Wright that Samuelson's math is correct. I think effective responses were not available until, maybe, the late 1980s or even more recently. (Caveat: Eatwell and Morishima had interesting things to say in the 1970s.) I don't understand Ian's approach, perhaps because I haven't put enough time into understanding it. And I'll have to add M. Lippi to my reading list.

Anonymous said...

Robert, no comment on Kliman's treatment of the transformation 'problem'? This is a book I know you have said you have read.

Robert Vienneau said...

Hi BenP

When I wrote about "effective responses" becoming available in "the late 1980s" and later, I had in mind, among others, the New Interpretation of, for example, Duncan Foley and the Temporal Single System Interpretation (TSSI). Kliman's work is part of the TSSI.

I don't have a definite opinion on the TSSI. I turn to Roberto Veneziani for interesting critiques.

I think Marx should be read as offering a claim that the source of profits is value added by labor not paid out in wages. I'd like this to be an empirical claim. I worry about some of these approaches, including Morishima's treatment of joint production, turning this claim into an accounting identity or a tautology.

Anonymous said...

You may be interested in this exchange, though I admit it's a minority interest.

This is one of the funniest 'critiques' I have read - if you can get the American cultural references.'s%20Critique%20of%20Mx%20%20TSSI%206.17.07.doc

I was disappointed with Kliman's refutation of Böhm-Bawerk; not that I am claiming I could do better, but am interested in the reasons why anyone (particularly non Marxists) would disagree with TSSI.

Not sure what Morishima has to do with this?

Robert Vienneau said...


I apologize for being slow to respond.

Why anybody would disagree with the TSSI, to mean, is a question of the sociology of knowledge. I think many would disagree because the TSSI is opposed to a century's worth of interpretation of Marx.

Back in the 1970s, some argued about how to extend that interpretation to the case of fixed capital and joint production more generally. Steedman produced a case of joint production in which the rate of exploitation is negative and the rate of profit is positive. Morishima offered an interpretation of Marx in which, following his definition, the rate of exploitation remained positive in Steedman's example. It is another variant of Marx interpretations which saves Marx conclusions, but doesn't perhaps go along with intuition trained in the traditional interpretation.