Nick Rowe's misconceptions about Sraffians I

Nick Rowe continues his very welcome discussion of the issues related to the capital debates in a recent post (see also the reply by Unlearning Economics). However, there are several misconceptions in his post that are worth clarifying. I'm going to deal with his first four assumptions (1 and 2 in this post, and 3 and 4 in a subsequent one), which are the more substantial from a theoretical point of view, namely:

1. Some economists in Cambridge UK wanted to explain prices without talking about preferences. I don't know why they didn't want to talk about preferences;

2. They made some special assumptions that helped them explain prices from technology alone, without talking about preferences. Like: all labour is identical; all technology is linear; prices never change over time;

3. But they still couldn't explain the rate of interest. Because it's hard to explain the rate of interest if you don't want to talk about time preferences. And all the other prices depend on the rate of interest, as well as on technology. So they assumed the rate of interest was exogenous;

4. Some economists in Cambridge US made a very special assumption that let them explain the rate of interest without talking about time preferences. They assumed that there was only one good, and it could be converted back and forth between the consumption good and the capital good by waving a wand.

The Cambridge economists are, of course, Sraffa and his followers. First, one of the most frequent confusions about Sraffa was that he assumed that demand, and, as a result, preferences were irrelevant. Before I tackle the issue per se, it is worth quoting this phrase, brought to my attention in Robert Vienneau’s blog:

"I am sorry to have kept your MS so long - and with so little result. The fact is that your opening sentence is for me an obstacle which I am unable to get over. You write: 'It is a basic proposition of the Sraffa theory that prices are determined exclusively by the physical requirements of production and the social wage-profit division with consumers demand playing a purely passive role.' Never have I said this: certainly not in the two places to which you refer in your note 2. Nothing, in my view, could be more suicidal than to make such a statement. You are asking me to put my head on the block so that the first fool who comes along can cut it off neatly. Whatever you do, please do not represent me as saying such a thing." -- Piero Sraffa (1964). Letter to Arun Bose (italics added).

Clearly Sraffa says that demand plays a role. However, the role is not the same as in marginalist theory. One has to understand what role demand played in the surplus approach in the determination of relative prices to get what Sraffa is saying.

Classical authors, in particular Adam Smith and David Ricardo (and certainly not Marx), did not think in terms of individual utility. When they talk about utility they are referring to social utility. Hence, commodities to be produced must be socially useful, otherwise they would not be produced, since nobody would buy them, but their price, their exchange value, is not based or connected with their use value. For example, Smith in his discussion of the diamond/water paradox (Wealth of Nations, Book I, ch. IV) argues that things with a high use value often have little or no exchange value, since things that are not costly to produce will command no price, even if they are useful. It is only with Thomas De Quincey, after Ricardo and the demise of classical economics, that the notion that utility (and use value) had a functional relation to exchange value becomes entrenched in economics, an idea that was picked up by Stuart Mill, and through him by Marshall (marginalism or neoclassical economics), as it is well documented by Krishna Bharadwaj (subscription required).

As Bharadwaj says of the Quincey/Mill/Marshall notion:

"This was a different notion of use-value than that accepted by Smith and Ricardo, for whom use-value was a necessary condition for a commodity to possess in order to be an object of exchange, but referred to the physical properties socially known to belong to a commodity, and not dependent upon the individual's estimation of its capacity to gratify subjective inclinations, measured in quantitative terms. In fact, use-value and exchange-value were incomparable in so far as the former covered the qualitative aspect and the latter was a quantitative notion. In De Quincey and Mill, the two notions had become quantitatively comparable (one acting as an extreme limit upon another) and this was only a step towards the later resolution of the paradox in terms of 'total' and 'marginal' utility."

In other words, social utility not individual estimation is behind the classical notion of preferences. The point then is that social utility and, as a result, demand considerations are essential for the determination of long term prices. However, these preferences are not the subjective preferences of individuals, about which nothing scientific can be said, since there are no regularities and they can change for irrational and circumstantial reasons.

The utility that society attaches to a particular good, say a car, however, can be taken as given at a particular point in time. The reasons are not only directly connected to objective characteristics, like the fact that a car is a means of transportation or negatively that they worsen environmental conditions, but also that cars socially may be a source of status, as Veblen later suggested. In other words, preferences (and demand) are socially determined and taken as given for the determination of relative prices, but there is a role for historical and institutional analysis in understanding why and how demand and preferences change over time. The idea was, also, that social preferences are relatively slow to change, and for that reason one can take them as given.

Thus, in the surplus approach there is a role for historical/institutional analysis (not just social preferences, but income distribution too, e.g. the discussion of the determination of the exogenous real wage), and a different role for theoretical analysis (the determination of exchange value). Note that the assumption of given social preferences, as correctly noted by Unlearning Economics, can be seen as a ceteris paribus clause. So Nick is right that there is no reason not to talk about preferences. However, it is far from clear that knowledge has been advanced by the marginalist treatment of individual subjective preferences. As I noted in my comments to his post, the reasons for convex, homothetic preferences is not dictated by knowledge about a regularity about people’s behavior, but simply by the teleological need of finding a solution to the maximization problem. I would say rather that there is no reason to talk about individual preferences. Social utility is fine.

Point 2 is just a poorly built straw man of the Sraffian/surplus approach model. There is no assumption of a linear technology. In fact, that’s typical of the neoclassical aggregative models. The only situation in which a linear relation between wages and profits in the Sraffian model is in the case of the standard commodity [I still owe you all a post on that topic; didn't forget], if this is what Nick means by linearity, which would be the only relevant case (unless he is against input-output models). The input-output framework of any system in which production is done by using commodities to produce commodities (a feature of the real world, by the way), does not imply that there is no technical change either. Technical input-output coefficients can change.

Note that any theory has to say something about the determination of relative prices for a given technology anyway. By the way, in this case Sraffa assumes, like the classical authors, a given level of output (again is a ceteris paribus condition, and a different theory of the determination of output is needed; we know through Garegnani, Sraffa’s disciple, that effective demand is what Sraffa had in mind, and not some version of Say’s Law). That does not imply, hence, constant returns to scale, since that would require an increase in output proportional to the increase in inputs, something that cannot happen when output is given. Remember that output is given for the theory of long term prices only.

Finally, there is enough literature showing that one can reduce, theoretically speaking, labor of different qualities to a uniform type. It is ironic that a neoclassical author would complain about this, since the production function, which is beset with unfathomable problems, also assumes identical labor, and in fact it also assumes a unique capital good (not many means of production) and is fundamentally a world of only one commodity (on this and point 4 by Nick, more on the following post).

TO BE CONTINUED