The Neoclassical Theory of Technical Progress: Note
Abstract
In his article on technical progress in this Review, Winston Chang presents . . . the asymmetrical results out of Harrod's and Solow's classifications [of bias] in relation to that of Hicks (p. 913) for a two-sector model. He also shows . .. that the usual symmetrical results obtained for aggregate neutrality between the Hicksian and Harrodian schema, ... in general, do not apply to Solow-neutrality (p. 913). This note demonstrates that the asymmetries to which Chang refers result from the manner in which he defines the Solow measure of bias in the capital-goods sector. If this measure is redefined, the asymmetries vanish. Chang defines the Solow parameter, UL], (j= 1, 2), as the proportional rate of reduction in the labor-output ratio in sector j at a constant wage rate (pp. 913, 918): UL] -eLj| =O, where eLj-(Yj-Lj); Yj is the output of sector j, Lj is labor employed in sector j, (^) over a variable represents the relative change in that variable, and w is the wage rate measured in terms of the consumption good.' It follows that the laboroutput ratio under consideration in sector 1 is labor per unit of capital good (L,/Y1) whereas the corresponding ratio in sector 2 is labor per unit of consumption good (L2/Y2). Chang defines the Solow measure of bias in sector j as yj=(i/OKj)ULj, where OKj (O< OK< 1) is the share of capital in sector j.
- Sources
- openalex