The Appropriate Scale Variable in the U.S. Money Demand: An Application of Nonnnested Tests of Consumption versus Income Measures.

The purpose of this article is to employ several nonnested procedures to empirically test the proposition put forward by Mankiw and Summers that consumption rather than income is the appropriate scale measure in the money demand function. Real income, real disposable income, consumer spending on nondurables, personal consumption expenditures, and total private spending are tested as candidates for the scale variable. Tests are carried out for both M1 and M2 concepts of money. The results based on both M1 and M2 support the Mankiw-Summers position, but the results based on M2 are not as strong as those based on M1. Implications for fiscal and monetary policies are discussed.