HAS M2 DEMAND IN THE USA BECOME UNSTABLE - EVIDENCE FROM AN ERROR-CORRECTION MODEL

The purpose of this paper is to estimate an appropriate broad-money demand function for the United States and to examine its stability after 1987, when the Federal Reserve System began using M2 as a policy guide. Special attention is paid to the model specification, its dynamic structure, and to its cointegration properties. The results from various dynamic error-correction models suggest that: (i) the money demand relationship is stable; (ii) most previously estimated models have undoubtedly misspecified the interest rate variable; (iii) interest-rate variability is another important determinant of real money demand; and (iv) in contrast to previous studies, the long-run scale variable is real GDP, whereas real consumer spending is the short-run scale variable. The sample period examined is 1953:1 to 1991:4.