The confusing connection between M2 and inflation (2)

This post is a follow-up to a post of mine last month, and to a recent post by Mark Perry questioning whether it is reasonable to expect a pickup in inflation given the big slowdown in M2 growth over the past year. The short answer to his question is yes, it is reasonable to expect inflation to pick up despite the big slowdown in M2 growth. Indeed, an inflation pickup is made more likely by the fact that M2 growth has slowed sharply, after rising sharply in late 2008.

To begin with, the chart above tells us two things: 1) the connection between M2 and inflation is not always consistent, and 2) M2 growth and inflation often (but not always) move in different directions, contrary to what a standard monetarist framework might predict (e.g., faster money growth should lead to more inflation according to standard monetary theory).

For example: Consider the huge pickup in M2 growth from 1970 to 1972, followed by a significant slowdown in inflation from 1972 to 1974. M2 then decelerated sharply in late 1973, only to be followed by a huge pickup in inflation in 1974-75. M2 decelerated in 1978, and was followed by surging inflation in 1979-80. On a more modest scale, M2 decelerated sharply in 2003, and was followed by rising inflation in 2004-5.

The explanation for this, I believe, can be found in the confusing nature of M2. We've all been taught over the years that M2 is arguably the best measure of money supply. I would argue instead that it is one of the best measures of money demand. After all, the Fed only supplies reserves and currency to the world, and currency makes up only 10% of M2. The other 90% of M2 exists because people want to hold money in the form of checking accounts (10%), retail money market funds (10%), small time deposits (13%), and savings deposits (57%).

This distinction—whether M2 represents money demand or money supply—makes a world of difference. When M2 growth slows sharply, I think it usually means that money demand has collapsed; people want less money, so they take it out of their savings account and spend it. That's equivalent to saying that the velocity of money has turned up. Rising money velocity means that a given amount of money can support a greater volume of transactions and/or a higher price level. This follows from the classic equation M*V = P*T.

What we have seen over the past 18 months is this: in Sept. '08 the Lehman bankruptcy sparked a global panic over the imminent collapse of the banking system; this in turn caused money demand to surge as a precautionary measure; rising money demand showed up in a big increase in M2; rising money demand meant a huge decline in money velocity, and that is what caused spending to plunge, economies and industries to shut down, and inflation to fall. Since late March we have been on the positive side of this process: confidence has returned, money demand has declined, M2 growth has slowed to a crawl, the economy has picked up, and inflation has picked up.

I think it is reasonable to expect this process to continue, especially given the 1-2 year lags that we see in this chart. Plus, the inflationary potential of falling money demand is likely to be boosted by the Fed's willingness to keep interest rates very low for an extended period. The evidence of the commodities market and the gold market tells us that people already are actively trying to reduce their money balances in favor of more exposure to commodities and hard assets. And the dollar has fallen very close to its lowest levels ever; dollar demand is clearly weak.

Let me be quick to add that I don't see the seeds of a huge increase in inflation (yet). But I think there's enough monetary fuel out there to push inflation higher over the next year, to 3-4% or more. That might not sound like a lot, but it's above the upper limit of the Fed's 2% target, and it's a lot more than the bond market and the Fed are currently expecting. And that can make for very interesting markets.