The fallout continues from the Federal Reserve Board's recent announcement that they will soon discontinue publication of the M3 monetary aggregate, and many of its components. Tom Iacono -- who must surely need a new name for his blog any day now -- has a good review of recent commentary on the subject. (A tip of the hat to Dave Iverson.) I'm still somewhat surprised by the sentiment that the Board's decision is, at least in part, motivated by the desire to downplay a statistic that appears to be contradictory to the achievement of price stability. I'm surprised because such sentiment seems to imply that the FOMC places significant weight on the behavior of monetary statistics in the first place!
Let's go back in time, to the July 1993 Monetary Policy Report to the Congress. Here is what Chairman Greenspan had to say in his testimony:
... at least for the time being, M2 has been downgraded as a reliable indicator of financial conditions in the economy, and no single variable has yet been identified to take its place.
By July 2000, this footnote appeared in the written report:
At its June meeting, the FOMC did not establish ranges for growth of money and debt in 2000 and 2001. The legal requirement to establish and to announce such ranges had expired, and owing to uncertainties about the behavior of the velocities of debt and money, these ranges for many years have not provided useful benchmarks for the conduct of monetary policy. Nevertheless, the FOMC believes that the behavior of money and credit will continue to have value for gauging economic and financial conditions, and this report discusses recent developments in money and credit in some detail
"Velocity" represents the number of times the stock of money turns over in support of spending, and it is the key piece of information policymakers need to to connect monetary measures to objectives. (You can read a bit more about the theory here.) In short, if you can't predict velocity, you really cannot predict how changes in the money supply will impact the rate of inflation.
With that thought in mind, the following picture of M2 velocity helps to explain why monetary statistics have been "downgraded " in monetary policy deliberations:
The lines represent the trends in velocity -- and the changes in those trends -- identified by formal statistical tests. If you are interested in the formal details of these calculations, you can look them up in a paper by John Calrson, Ben Craig, and Jeff Schwartz that describes the methodology. But you can also just believe your own lyin' eyes, which will tell you, I think, that velocity has not been very predictable over the past decade-and-a-half.
I would have shown you a picture of M3 velocity instead of M2 velocity, but for one problem. One way to think about velocity is that it represents the part of money demand that is sensitive to changes in interest rates. Estimating trends requires some estimate of the return on the monetary measure (so it can be compared to the return on nonmonetary assets -- giving us an idea of the "opportunity cost" of holding money). To the best of my knowledge, no regular estimate of the return to M3 is even available. That alone speaks volumes.
So here's my last word on the subject. If you have a sense that M3 is providing any information at all related to the objectives of monetary policy, you know something I don't know.
For all of you in the States -- or whose hearts are in the States -- Happy Thanksgiving.