Experience is the name everyone gives to their mistakes. --Oscar Wilde

In case you have not been keeping score, be informed that the monetary aggregates have been growing at juicy rates in recent months. Fed Chairman Paul Volcker is concerned about this. I am concerned about his concern.

By Oscar Wilde's definition, the Federal Reserve has gained a great deal of experience with monetarism in recent years. Why look for more?

Presumably, the monetary aggregates--the M's--are not goals in themselves, but only instruments for controlling something that matters, like GNP. But GNP is the product of money times velocity, so GNP growth stems from both growth of money and growth of velocity. If velocity drops, the money supply must grow faster, or the economy will stagnate. And it just so happens that velocity has been falling rapidly in recent months.

The monetarist belief in steady money growth presumes that velocity growth is either stable or highly predictable. A few years ago, this doctrine seemed dubious. Now it is ridiculous. Deregulation and rapid financial innovation continue to transform the ways people make payments and store their wealth. Many of these changes affect the demand for one or more of the assets included in the M's, thereby causing velocity to shift.

Let us consider the recent high monetary growth rates in this light.

Since November 1982, M2 has grown at a 16 percent annual rate, a sharp acceleration from the 9 percent rate recorded during the previous 12 months. Is this cause to sound the inflationary alarms? Hardly. In December 1982, a new type of bank account called a money market deposit account (MMDA) was authorized. These accounts proved to be very popular. In less than seven months, balances in MMDAs grew from zero to over $360 billion.

Now it happens that the Fed decided to put MMDAs into M2, which explains why M2 has grown so rapidly. In case you are wondering, a version of M2 that excluded MMDAs would have grown at an annual rate of --17 percent since November 1982.

So which is the "right" growth rate for M2, p16 percent or --17 percent? I don't know, and neither does the Fed. If all the funds that are now in MMDAs came from elsewhere within M2, then 16 percent would be the relevant figure. On the other hand, if all the funds in MMDAs came from assets not included in M2, then --17 percent would be more indicative of monetary growth. Obviously, the truth lies somewhere in between. But where? The sad fact is that no one knows.

The other popular monetary aggregate is M1. During the 12 months ending in June 1982, M1 grew 5.4 percent. Then from June 1982 to June 1983, it grew at a whopping 12.9 percent annual rate, causing much consternation among monetarists.

Here the explanation is less clear, but December 1982 also marked the introduction of Super NOW accounts. These accounts are included in a component of M1 that the Fed calls "Other Checkable Deposits." By no coincidence, the annual growth rate of Other Checkable Deposits from June 1982 to June 1983 was 39 percent. Had these deposits been excluded from M1, the recorded M1 growth rate would have been only 6.6 percent.

On the other hand, had the Fed put the MMDAs into M1, the recorded growth rate of M1 would have been 94 percent. Thus, depending on some subtleties of definition, the Fed could have reported an M1 growth rate anywhere between 6.6 percent and 94 percent.

There is obviously room for fun with numbers here. But I've done enough to illustrate how meaningless monetary growth numbers can be during a period of rapid financial change.

The last quarter of 1981 and the first quarter of 1982 provide a vivid historical example. During these two quarters, M1 grew at a 7 percent rate. Reasonable, right? Wrong, because M1 velocity fell at a 6 percent rate, leaving the annual growth rate of nominal GNP a scant 1 percent. The consequence was a 5 percent rate of decline of real GNP and a terrible recession.

History might have repeated itself a year later had the Fed stubbornly adhered to monetarist dogma. During the fourth quarter of 1982 and the first quarter of 1983, M1 velocity fell at an 8 percent annual rate. Fortunately, Volcker had renounced monetarism --temporarily, he said--in October 1982, and M1 was allowed to grow at a 14 percent rate. So nominal GNP was at least permitted to grow at a mediocre 5.3 percent pace. Real economic performance during these two quarters was not great; but neither was it catastrophic.

The moral of the story is clear: he who targets on the growth rate of money when velocity is behaving erratically is looking for trouble.

Therein lies my worry. Volcker has recently announced that the Fed intends to bring money growth rates back into line with targets. If the Fed returns to M-fetishism, look out. For as long as velocity keeps declining, seemingly high money growth rates are not only appropriate, but actually essential if recession is to be avoided.

In view of our experience with monetarism, perhaps the surgeon general should require that Volcker's cigar wrappers carry a warning: "Monetary targets can be hazardous to the economy's health."