Postprint version. Published in Economic Inquiry, Volume 27, Issue 4, October 1, 1989, pages 555-586.
Copyright © 1989 Blackwell Publishing. The definitive version is available at http://dx.doi.org/10.1111/j.1465-7295.1989.tb00790.x.
It is widely believed that monetary aggregates have failed to predict economic performance over 1983-87. This paper observes that the traditional definition of money (M1 lessother checkable deposits, or M1A) shows no evidence of structural change, and yields lower prediction errors for both real GNP and inflation over 1983-87Q2 than the errors obtained using M1 or M2. If there is a mystery, it is not why MIA has done so well, but why economists abandoned it for M1 or what was once called M1B (currency, demand deposits, and other checkable deposits).