Questions For Sraffians

The central propositions in Steedman’s 1992 paper "Questions for Kaleckians" were (a) that the input-output aspects of production must be taken account of in any account of price setting under capitalism; and (b) that one consequence of this is that the constraints which govern a mathematical representation of input- output analysis--matrix algebra--therefore govern the values that can be taken by the key parameters of the system of production. This paper argues that the first proposition above is, with reservations, correct--and, therefore, that some aspects of "normal" Kaleckian analysis were rightly criticised--but that the second is fallacious, since linear algebra is not an appropriate mathematical tool for the representation of the input-output nature of production. One suitable methodology is discrete-time dynamics, and the application of this to Steedman’s analysis and example shows (a) that his critique of Kaleckian mark-up pricing is largely misplaced, and (b) that his preferred methodology of linear algebra is inapplicable to capitalism. This is because, while it is true to say that "Kaleckian mark-up pricing and distribution theory cannot be done properly if inter-industry relationships are ignored" (Steedman 1992, p. 145), the proper introduction of these issues results in a dynamic, non-equilibrium analysis where, as Sawyer correctly muses, "many of the questions which [Steedman] raises would re-emerge (probably in a more complex form)" (1992, p. 163).