David Beckworth writes, referring to 1960-1979,

The Fed’s performance in stabilizing the growth of nominal spending was abysmal

Really, read Beckworth’s post. The graphs are more important than the text. My comments will be in Q&A format. The issue is whether we should think of the Fed as offsetting velocity shocks or we should think of changes in velocity as offsetting monetary shocks.

1. Why did money and velocity move in opposite directions for the most part from 1980 – 2007? Beckworth’s answer is that the Fed did a great job of offsetting shocks to velocity. The answer that I would give (and I suppose James Hamilton is with me, although I do not want to put words in his mouth) is that nominal GDP was doing its own thing, so that when the money supply changed, velocity moved in the opposite direction. I would say that velocity was able to offset monetary shocks.

2. Why did money and velocity show less mirror-image behavior from 1960 through 1979? Beckworth’s answer is that the Fed was doing a lousy job of offsetting velocity shocks. My answer is that velocity was less able to offset monetary shocks.

The reason that velocity was less able to offset monetary shocks in the 1960’s and 1970’s was that the medium of exchange was more clearly defined. (a) There were interest ceilings on checking accounts (for most of this period, checking accounts were not allowed to pay any interest), so there was a definite difference between money and interest-bearing assets. (b) Most people had fewer alternatives to using cash or checks. The idea of pulling out a credit card at a grocery store or fast-food restaurant was unheard-of.

Thus, the institutional setting of the earlier period was one in which the substitutability of the M1 money supply and other financial assets was low. This made it difficult for people to offset monetary shocks with velocity shocks, and it allowed the money supply to affect nominal GDP.

Since then, the substitutability between the M1 money supply and other financial assets has been rising. As a result, velocity shocks tend to offset monetary shocks, and changes in the money supply are less likely to affect nominal GDP.

I do believe, however, that if the U.S. government were really determined to have inflation in the short term, that could be accomplished. If we printed dollars and used them to by foreign assets in massive amounts, that would to the trick.

Such a policy would, in a world of sticky nominal wages, make U.S. production more competitive and reduce unemployment. Given that we have a bad unemployment problem and no inflation problem, this seems like a plausible policy to try, although in a larger sense it risks undermining the de facto dollar standard on which world trade has depended since the second world war.