Monday, October 5, 2009

Slow money growth not a concern

M2, arguably the best measure of the broad money supply, has not grown at all since early March. As this chart shows, the 6-mo. annualized rate of growth of M2 is now negative, after being hugely positive early this year. Is this a cause for concern? Hardly. In fact, I think this is a very bullish development.

We see a similar pattern in 2003: strong growth in 2002 and early in 2003, followed by a sharp slowdown in money growth. The main driver of big changes in M2 growth is money demand. Money demand was strong in 2002 and early 2003, and it was strong late last year as well, because of the combination of a) fears of extended economic weakness, and b) very accommodative monetary policy. Faced with uncertainty, people attempted to build up their money balances, and the Fed was willing to accommodate that. When uncertainty subsided in mid-2003 (helped to a significant degree by cuts in marginal tax rates) then money demand fell; money that had been stored up to guard against uncertainty was released into the economy, fueling a strong rebound in economic activity. I think the same thing is happening now. Uncertainty is falling, and money demand is falling, and the economy is expanding once again.

So very low or even negative growth in M2 is actually a positive sign: confidence is returning and money that was accumulated as a hedge is now being released into the economy to fuel growth.


seekingtraceevidence said...

It appears that M2 grows in line with Nominal GDP. Real GDP has been just over 3%. Wouldn't it be better if M2 growth were closer to Real GDP to keep inflation down?

Scott Grannis said...

M2 indeed tends to grow at about the same rate as nominal GDP, but only over very long periods. From year to year, the variations can be huge.

I have studied the relation of M2, GDP and inflation for decades, and have concluded that more often than not they behave in mysterious ways, with the mysterious variable being velocity (i.e., money demand).

Milton Friedman always said that the Fed should limit M2 growth to some ideal (noninflationary) value of nominal GDP growth, but he ignored the impact that changes in money demand can have on things.

I think the best and only way to relate money to inflation is by looking at market based indicators that signal if and when there is an excess of money in the system relative to demand. That would be the dollar, gold, commodities, shape of the yield curve, etc.

seekingtraceevidence said...

With all the data we collect, it seems that the Fed should have at this point discovered a better method to control inflation other than measuring a symptom of inflation once it has started than focusing on a precursor.