MV=PT. Oh really? Nathan Lewis August 2, 2001

Memo To: Supply-Side Students
From: Nathan Lewis
Re: MV=PT. Oh really?

In our statistics-drenched age, one of the most difficult-to-grasp supply-side notions is that most statistics can be safely ignored. This is particularly true of monetary statistics. Classical economists have always concerned themselves with the value of money, not its volume. Actually, it’s not necessary to know the volume of money at all, measured in whatever arbitrary fashion. The Hodges Genuine Bank Notes of America, 1859, a reference work on the desk of anyone engaged in trade at the time, listed 9,916 notes issued by 1,356 banks, and even so hundreds of legitimate banknotes were omitted. At the time, any bank could legally issue money, and although there are some statistics available from the period, nobody concerned themselves with the total amount of currency in circulation. The only thing they cared about is whether or not the banknotes held their value against gold. As long as each of the 1,356 (or more) banks managed the supply of their banknotes in accordance with the gold standard, the banks didn’t need to know what the other banks were doing. Needless to say, no price or production statistics existed at the time, either. Today’s statistics date from World War II and soon after, the high tide of Keynesian statism.

The monetarist equation MV=PT has confused people for centuries, which is why the better economists dating at least as far back as Adam Smith have always dismissed this antiquarian and fundamentally fallacious device. Milton Friedman didn't invent monetarism. Montesquieu was a monetarist, and I'm sure that there were monetarists in the days of Plato and Confucius too.

Notice that the value of the currency, its single most important element, doesn't appear anywhere in MV=PT, the equation that is supposed to be the accepted axiom of Friedman monetarism. If you decide to go to college and study monetarist economics, MV=PT is the equivalent of the ABC’s: Money times the Velocity of Money equals Prices times the number of Transactions. It was Friedman’s “insight” that because V is constant, controlling “M” would control PT. You could manage the output of the national economy by having the Federal Reserve precisely control the volume of “M.”

Let's look at the components of this equation:

Money: What is it? Base money? M2? M2+CDs? M3? M13? MZM?
Velocity: A totally unmeasurable unknowable residual mystery variable
Prices: What is it? A price index? Whose price index?
Transactions: What is it? GDP? GNP? Measured by whom?

In other words, MV=PT is an uncertainty multiplied by a mystery equivalent to the product of two academic abstractions.

Notions of “Prices” and “Transactions” become even more vague in a situation where the dollar is used all over the world. In fact, most dollar bills are apparently being used overseas, in dollarized countries, in black markets, as a secondary currency, in the international drug trade, by commodities producers, by tourists, and by foreign banks and central banks. These are all dollar transactions. Do the Ms take account of eurodollars, or deposits in Japanese banks? Do price indexes include prices of Middle East arms shipments, or Cambodian heroin or Columbian cocaine, or groceries in dollarized El Salvador? Does GDP take into account dollar-denominated trades in the Russian black market? And what about the dollar-linked regions like Argentina or Hong Kong? Certainly if you’re going to consider bank deposits as “money,” even though they are technically interest-bearing debt securities and not currency, you have to include Argentine pesos and Hong Kong dollars, which are genuine currencies redeemable for U.S. dollars. With a little poking, the seeming certainty of a mathematical equation soon dissolves into a miasma of doubts.

I think Adam Smith sums up monetarism pretty well:

"What is the proportion which the circulating money of any country bears to the whole value of the annual produce circulated by means of it, it is, perhaps, impossible to determine. It has been computed by different authors at a fifth, at a tenth, at a twentieth, and at a thirtieth part of that value."

Monetarism hasn’t progressed much beyond that point in the last 250 years, but causes considerable damage whenever anyone takes it seriously. Please note that Federal Reserve Chairman Alan Greenspan last week told the Senate Banking Committee that there is plenty of liquidity in the banking system, as evidenced by the increased amount of “M” in the banking system. Is it any wonder people are beginning to wonder if Greenspan is all that he has been cracked up to be?