Real Money Balances: An Omitted Variable from the Production Function?

SEVERAL writers have argued that real money balances are a factor of production.1 No one, however, has directly tested the hypothesis that real money balances are a factor input.2 The purpose of our paper is to report the results of such a test. We find that real money balances, regardless of definition, enter significantly in a Cobb-Douglas production function fitted to annual data over the period 1929-1967 for the private domestic sector of the United States economy. Quantity indices of output, capital and labor, published by Christensen and Jorgenson (1970) and adjusted both for quality changes and rates of utilization, are employed to estimate the production function. Data for nominal money balances are taken from Friedman and Schwartz (1970). Our results have important implications for production function analysis, the explanation of total factor productivity, and monetary growth theory. The plan of the paper is as follows. Section II deals with a brief discussion of the rationale for the presence of real balances in the production function. In section III we present the production function used in the study and discuss the data employed. Results are given in section IV. A summary and conclusions follow in the final section.