The Recent Controversy in the Theory of Capital

Sraffa’s statement cited above (1960: 6) contains the key not only to a critique of Marx’s labour-value based reasoning, but also of the long-period marginalist concept of capital.

Since for a given system of production the amount of labour is constant irrespective of the level of the rate of profits, also the ratio of the value of the capital goods and the amount of labour employed, or capital-labour ratio, K/L, would tend to fall (rise) with a rise (fall) in the rate of profits:

This is the first claim of marginalist theory. The second claim is that RWEs are also positive. A positive RWE means that with a rise (fall) in the rate of profits - and the corresponding fall (rise) in the wage rate - cost-minimizing producers switch to methods of production that generally exhibit higher (lower) labour intensities, thereby “substitut­ing” for the “factor of production” that has become more expensive - “capital” (labour) - the one that has become less expensive - labour (“capital”). Hence (6) is taken to apply also in the case with a choice of technique. The assumed positivity of the RWE involves the marginalist concept of a demand function for labour (capital) that is inversely related to the real wage rate (the rate of profits).

Careful scrutiny of the marginalist reasoning shows that it cannot generally be sus­tained: there is no presumption that PWEs and RWEs are invariably positive. In fact there is no presumption that techniques can be ordered monotonically with the rate of profits (Sraffa 1960). Reswitching implies that, even if PWEs happen to be posi­tive, RWEs cannot always be positive. As Mas-Colell (1989) stressed, the relationship between K/L and r can have almost any shape whatsoever. The intervals in which K/L is an increasing function of r are spoken of as capital reversal. This implies that, if the neo­classical approach to value and distribution is followed, the “demand for capital” is not decreasing, and therefore the resulting equilibrium, provided there is one, is not stable. Hence the finding that PWEs and RWEs need not be positive challenges the received doctrine of the working of the economic system, as it is portrayed by conventional eco­nomic theory (see Pasinetti 1966; Garegnani 1970; see also Harcourt 1972; Kurz and Salvadori 1995: ch. 14; 1998c).