THE AGGREGATE PRODUCTION FUNCTION: A CONSIDERATION OF SOME EXISTENTIAL PROBLEMS

The concept of the aggregate production function, together with it supposedly being a test of the neoclassical marginal product theory of distribution, is now central to much of neoclassical macroeconomic theory. Severe criticisms of this concept that include the Cambridge Capital Theory Controversies and the aggregation problem are now simply ignored. The reason is that estimates of the Cobb-Douglas and the Constant Elasticity of Substitution (CES) production functions, inter alia, give good statistical fits with the estimated output elasticities being often close to their factor shares. However, the reason for this is that in both cases all that is being estimated is merely the mathematical transformation of an accounting identity. As such, it has no economic implications and, in particular, says nothing about “the laws of production”. This analysis considers this criticism of the aggregate production function. In particular, it revisits and extends the criticisms made of Solow’s (1957) model of technical change and the use of the aggregate production function. It is shown that Solow’s (1974; 1987) defence of the aggregate production function is not convincing. The CES production function is also shown to be simply a transformation of the accounting identity and, consequently, also has no economic implications.