Technological Change and the Superneutrality of Money: Note

Abstract

The study of the effect of the supply of money on real economic variables has had a long history in economic research. Sidrauski [8] was the first to demonstrate that, in the context of an intertemporal maximizing model with real money balances in the utility function, the rate of monetary expansion did not influence the steady-state capital stock or the real rate of interest. This result, termed the of money, contrasted sharply with previous models, which did not use such an optimizing framework. 1 Subsequent studies have shown that the superneutrality result can be altered if some of the assumptions of the Sisrauski model are modified. If money is a productive input [3, 5] or if the labor supply is endogenous [1], then the real rate of interest, or the capital-labor ratio, is not invariant to the monetary growth.2 However, no one has yet examined the effect of differing money growth when technological change is present. This omission seems especially serious since superneutrality properties apply to the comparative steady-state behavior of the economy.3 4 This note attempts to fill this gap by demonstrating that even the simple forms of technological change will in general nullify the superneutrality result, and that the real rate of interest and the capital-labor intensity will depend on the parameters of the utility function. In particular, it is shown that only if the utility function is of the class exhibiting constant relative risk aversion in all arguments will the superneutrality of money be valid.