The article looks critically at Piero Sraffa's theory of price and implied theory of money against a backdrop of renewed interest in the great economist's work. The focus is the theory of price in his Production of Commodities by Means of Commodities. It is argued that for all its apparent logical (read mathematical) rigor, Sraffa's explanation of prices (and implied theory of money) is (are) essentially flawed. The flaws are traced to his concern to eliminate the labor theory of value from this explanation. It is this concern that causes Sraffa to make the numerous dubious assumptions and constructs which underpin his explanation of prices and that ultimately brings into question its validity as an explanation of actual price magnitudes and their movement.
In their introduction to a collection of articles on the work of Sraffa presented at a conference to mark the 50th anniversary of the publication of PCMC, Blankenburg, Arena, and Wilkinson (2012; 1270) note that although a number of Sraffians see the primary focus of Sraffa's work to be the study of the “surplus product” and its distribution, they are of the view that the majority see this focus to be the theory of price.
The interested reader is referred to Nicholas (2011, chapter 8) for a critical assessment of these extensions of Sraffa's theory of price.
See also Dobb (1970) and Medio (1977).
For critical reviews of those traditional interpretations and New Interpretations (NIs) of Marx's theory of price which see it as logically flawed (see Nicholas 2011, chapter 4).
Sraffa's unpublished articles suggest that in the early stages of the development of his ideas, before settling on the idea of showing the indeterminate impact of changes in the rate of profit on prices, he even toyed with conceptualizing profit (which he equated with interest) as “necessary” to keep capitalists invested (see Kurz 2012; 1550). This would have allowed him to retain the argument that the magnitudes of relative prices of commodities are determined by the relative magnitudes of value of the relative inputs required for their production. Although it is not clear why he abandoned this line of argumentation, one may surmise that he did so because he realized it would be tantamount to endorsing the Neoclassical explanation of profit.
It is perhaps his recognition of the obvious parallels between his own theory of price and that of Marx's that explains his defense of Marx in respect of Borkiewicz's criticism of Marx's transformation procedure, arguing that Marx's solution was “approximately” correct (see Sraffa 1960).
It warrants noting that Marx denied it is meaningful to layer inputs ad infinitum when computing costs of production since to do so would imply, on the one hand, costs were essentially embodied not required costs and, on the other hand, what was paid for these inputs did not represent the latter, i.e., the required costs (see Marx 1969a, 100–102; 1976, 293–95).
See Nicholas (2011, 146) for an elaboration of this point.
Interestingly, it is a similar interpretation of Marx's theory of price, i.e., as a labor commanded theory, that makes it (Marx's theory) vulnerable to the Sraffian charge of redundancy. Specifically, Sraffians have, with some considerable justification, argued that knowledge of the labor commanded by the commodity inputs, wage, and profit presupposes a prior knowledge of the technical conditions of production (i.e., the required inputs) and the distribution of income (see, e.g., Steedman 1977; Roncaglia 2009; Sinha 2010).
Many Marxists have also questioned Sraffa's and Sraffian contentions that (given the technologically determined magnitude of the surplus product) the magnitude of profit is explained by the distributional struggle between labor and capital over the surplus product, arguing that a characteristic feature of capitalism is the wage is advanced before production, including the production of the surplus product (see Rowthorn [1974] for an early example of this line of Marxist argumentation).
In this context, I am reminded of Tony Lawson's repeated warnings against the excessive emphasis placed in economic analyses on forms of mathematical deductivist reasoning (see Lawson 1997; 2003; 2012).