US Money Supply

Numerous people have commented in recent weeks about the drop in the US money supply over the past few months. The graph below illustrates the concern. “Narrow” money supply (M1) grew more slowly in the end of 2003, but not unusually so. However, the growth in M2 (which includes cash, checking accounts, savings accounts, and money market accounts) took a dramatic tumble toward the end of 2003.

This obviously raises two questions: what caused this fall in M2, and is it a cause for concern? To answer these questions, I talked to a friend who’s an economist in the Federal Reserve System, and is probably among the 50 most knowledgeable people in the world on the subject of the US money supply. (The other 49 are probably also Fed economists.)

First, the causes of the drop in M2. There are three.

1. When homes are refinanced, the old mortgage is paid off with a new one. But most mortgages have been bundled together into a mortgage-backed security, which is then held by a large-scale investor (e.g. a bank or insurance company). Due to the accounting procedures necessary to adjust the mortgage-backed security for the early repayment of one of the mortgages that it contains, the balance refinanced is held for a short time in a special type of account. When there are more refinancings, these accounts swell the size of M2; when refi activity falls, these accounts shrink and M2 falls. During the second half of 2003 mortgage refinancing activity slowed considerably, so the size of those accounts fell, reducing M2. This is a technical side-effect of the slowing of refi activity, and has no economic impact whatsoever.

2. Another cause has to do with the effects of cash-out refinancings. People built up large balances in their checking and savings accounts as they took cash out of their house from refinancing. As refi activity has slowed, less cash has been taken out of houses, so that addition to the money supply has tapered off. Additionally, people have been spending the money that they took out of their houses in 2002 and early 2003. So M2 balances have fallen. So in part, the fall in M2 simply tells us that we’ve seen the end of major refi-related spending.

3. The third cause has to do with the stock market. Since the resolution of the uncertainty surrounding the war in Iraq in the spring of 2003, individuals have poured money at an accelerating rate into the stock market. To do that, people take money out of their savings and money market accounts. The result is a fall in M2.

Now the next question: is it a cause for worry? The short answer is no. First of all, some people (such as my monetary expert at the Fed) make the argument that the size of M2 is irrelevant to anything that we care about. The only possible use it may have is if it contains information about other things in the economy that we do care about. But even a sharp, sustained fall in M2 would have no repercussions for the economy in and of itself.

Second, the causes of the fall in M2 in this case are generally not anything we didn’t already know. We already knew that mortgage refi activity slowed in the end of 2003 (though this confirms that people have largely spent all of their refi cash by now), and we already knew that people have been shifting money into the stock market. So in this case, the fall in M2 doesn’t provide us much new information. And anyway, M2 is starting to grow again as these effects peter out. Within a few more months it will probably be growing at its usual healthy clip.

So, to make a long story short, don’t lose any sleep over the wiggles in M2 growth. As I’ve written about many times before (including yesterday on The American Street), we have plenty of other economic problems to worry about.

Kash