Mankiw’s attempted resurrection of marginal productivity theory

Part 1 of this paper (http://www.paecon.net/PAEReview/issue59/Moseley59.pdf) argued that the marginal productivity theory of distribution has fundamental and insoluble logical problems: (1) the aggregate production function is not a legitimate concept, because capital consists of diverse buildings and equipment that cannot be reasonably aggregated into a total quantity for the economy as a whole; (2) the marginal product of capital (or of labor) is not a legitimate concept, because output in goods-producing industries cannot be increased by adding one unit of capital and holding all other inputs constant, because raw material inputs must also increase if output is to increase; (3) the derivation of the demand for capital (or for labor) is invalid, because it is based on the illegitimate concept of the marginal product of capital; and (4) the return to capital is included in the price of capital, as the “opportunity cost” of the owners of capital, and the opportunity cost is taken as given, like all other costs of firms. Therefore, marginal productivity theory takes as given the main variable that should be explained – the return to capital.