The Money Multiplier: A Resurrection?

There's a breed of economist that I think of as recovering monetarists. These are the folks that read Hayek and Milton Friedman in their formative years and the skies open and the light showered down upon them. They saw the dream of a free society with minimal government intervention.

Of course the lynchpin of this view is markets delivering efficient outcomes. To make it work, the government does have to guarantee property rights and the rule of law, yadda, yadda....

The government also needs to control the money supply. Actually, some very erudite scholars would argue this point and point to Hayek's competing monies idea, or the Wildcat Banking era in the pre-Civil War U.S., but Milton wasn't sold. His vision was a single government bureaucrat setting the money supply according to some pre-determined growth rate, the k-percent rule. So its government intervention, but only in the most non-invasive, predictable manner.

Such a rule would be stabilizing since the money would move in the steady stream from the Fed, to the banks then onto households and firms via loans, primarily. A key measure of the speed of the stream is the money multiplier m, found in any Money and Banking textbook:

m = change in the money supply/change in the monetary base

The monetary base (MB) is the liabilities of the Fed, which is mostly bank reserves along with cash. So the money multiplier shows how much an extra dollar of bank reserves turns into money. The money multiplier is typically larger than one since that dollar is loaned and deposited repeatedly. (In a Money and Banking class, this gets the label "Multiple Deposit Creation.")

How much bigger than one? Let's go to the data:

OK, there's some issues here, and not just that the series is discontinued. The graph shows the ratio of the levels of M1 and the MB, which is not exactly the formula above. (Let's save any conspiracy theories about FRED for another day.) Using levels makes the multiplier look fairly stable, though the drop in the Great Recession is notable.

Here the graph using the percent change in M and MB over a year:

Doesn't look particularly stable, particularly after the Great Recession. (I used percent changes, since both series trend up. The story is similar with raw changes.) The extended stretches of negative numbers are concerning or intriguing depending on your priors.

If you're trying to defend a k% rule, you've got a problem. The connection between the monetary base and the money supply doesn't seem dependable. You might argue that the Fed is targeting an interest rate, which leads to big fluctuations in MB. That's probably correct, but it should lead to similar fluctuations in M1 if the money multiplier is reliable.

Many aspiring monetarists keep searching for the right definition of money that achieves and stable multiplier (and stable velocity) to rescue Milton. Their arguments are often presented in terms of the quantity equation for money (MV=PY), where velocity is stable and real GDP (Y) stays in the neighborhood of its natural rate, so steady growth of M leads to steady growth of P.

Its a beautiful vision, unfortunately, too beautiful to be true. An unstable money multiplier means unstable velocity and the vision dissolves. 

So should we toss the money multiplier onto the dust heap of history. Maybe not. What is happening when m goes negative? Bank reserves are rising as the money supply falls, so its an indication that banks aren't lending as much as they could. Perhaps we can use this as an indication that the market for loanable funds is not clearing. The zero lower bound on interest rates is one reason for the that to happen, but there may be other factors as well.

For those of us that teach Money and Banking, we should tell students about the money multiplier but make it clear that the world is a bit too complicated to rely on it. As researcher, we need to get past the focus on monetary aggregates and model the flows of funds within the banking system, starting with reserves. In the works of Olivier Blanchard, we need to examine the plumbing of financial markets.

The recent recessions and the associated mess of financial data points up the work to be done.....