Neil said what???

Over at Social Democracy for the 21st Century, Clonal linked to my recent posts on Friedman's MRTO graphs -- thanks, buddy! -- and he wrote:

Friedman's contention that "Inflation is always and everywhere a monetary phenomenon" is based on a very faulty analysis of data.

Well said, sir.

Neil Wilson replied:

Friedman tried to show that V is stable - which is essentially the ratio between P.T and M

It isn't and anything that relies on that fails with it.

I'm responding to Neil here.

First: V is essentially the ratio between P.T and M

V is Velocity.
M is money.
P is the price level.
T is "the real value of aggregate transactions". (I had to look it up.)

Wikipedia refers to the equation M*V=P*Q (where Q is real output) and then refers to an older form of that equation: M*V=P*T. If you take the older form and divide both sides by M you get V=P*T/M, or "the ratio between P.T and M" as Neil said.

I think Q is "final" transactions, and T is total transactions. I also think this older form is Irving Fisher's equation, but I don't have a reference handy.

I like the form that uses *all* transactions, but economists abandoned that for some reason. Beside the point.

If P is the price level and T (or Q) is the inflation-adjusted value of the transactions we choose not to ignore, then P times T is the total dollar-amount of all the transactions that we don't ignore in a year. (P times Q is the dollar-amount of GDP, GDP at actual prices, which economists call "nominal" output.)

M is how much money there is in the economy. (Which measure of money? Beside the point.) V is how fast that money is being spent. M times V is the total dollar-amount of the transactions we chose not to ignore.

So both sides of the equation are equal to the total dollar-amount of those transactions.

Okay. We have established equality.


Neil says: Friedman tried to show that V is stable, but V isn't stable, and that's why Friedman's MRTO graphs fail.

Well... Neil is right. V isn't stable. Three or four years back, when everybody suddenly decided it was time for deleverage, Velocity dropped like a rock.

But for long stretches of time, V is stable enough that Friedman could use it in the calculations for his graphs, and get the results that he wanted to show.

V is GDP divided by the quantity of money. Friedman's graphs turn that upside-down and divide the quantity of money by GDP. That's all there is to it, except... for... one... little... thing...

Velocity uses GDP at actual prices. Friedman uses GDP at inflation-adjusted prices. In other words, Friedman's calculation brings inflation into his numbers. That is why his graphs look like the price trend, and the velocity graph does not.


Now, the meat. The reason I got out of bed to write this post. Neil says:

Friedman tried to show that V is stable

No, Neil, he didn't.

Friedman did depend on the fact that V tends to change slowly over long periods of time. If it didn't, his graphs would not look like the price-trend.

But he was not trying to show that V is stable. He was trying to convince people that printing money causes inflation. And it worked.

There are an awful lot of people who have a mental block because it worked. Whenever they talk about the economy, all these people can talk about is inflation. It is boring to talk to people like that. But there's a lot of 'em. And it's *not* because Friedman was trying to show that V is stable.