The Fed May Be Miscounting the Money Supply

By one metric, the money supply shrank in 2010 instead of expanding
The U.S. Federal Reserve building in Washington Photograph by Brendan Smialowski/Bloomberg

Even a third-grader knows that adding unlike things is problematic. Seven grapes and one watermelon do not equal one grape and seven watermelons, even though both are eight pieces of fruit. But the Federal Reserve persists in adding different kinds of money as if they were identical: assets that are easily spendable and ones that aren’t; ones that earn interest and ones that don’t.

Why does this matter? Because the amount of money could affect economic activity in the short run, and inflation—or deflation—in the long run. Miscounting it could cloud the Fed’s understanding of how the economy is functioning. The Fed’s broad measure of the money supply (M2) grew 4 percent to 6 percent annually during the housing bubble and then 2 percent in 2010 in the aftermath of the burst. But a measure from the New York-based Center for Financial Stability, capturing more kinds of money and using a more technically sound counting method, found that the money supply grew 6 percent to 8 percent during the bubble and then outright shrank in 2010 (chart). Those fluctuations, officially unaccounted for, may have contributed to the boom and bust.

Money to an economist is not just cash but anything that’s as spendable as cash and earns no interest, such as a checking account. A certificate of deposit has less “moneyness” than cash or money in a checking account because it can’t be spent as readily.

The Fed makes two mistakes with this approach. First, its M2 makes no distinction between the moneyness—i.e., spending potency—of cash vs. that of interest-bearing time deposits. (That’s the grapes and watermelons problem.) Second, because the Fed lacks a money measure broader than M2, it doesn’t capture the contribution to liquidity from assets such as commercial paper and Treasury bills, which are at least somewhat spendable.

A crusade to get the Fed to change its money-counting ways is being led by William Barnett, who has a Ph.D. in statistics and is a professor of macroeconomics at the University of Kansas, as well as director of the Center for Financial Stability. Before going into economics, Barnett was an engineer for Rocketdyne, where he helped develop the Saturn V rockets used in the Apollo space program. From 1973 to 1981 he worked for the Fed in Washington, where he began to develop his ideas on monetary statistics. One of his Ph.D. advisees at the University of Texas in the 1980s was Salam Fayyad, now prime minister of the Palestinian National Authority. Barnett co-wrote a book called Inside the Economist’s Mind with the late Nobel economist Paul Samuelson.

All the while he was perfecting his money-counting method, named Divisia after French economist François Divisia (1889-1964), and becoming frustrated with the Fed. “He is a hard-core scientist. He has no patience for people who think they can take shortcuts,” says Peter Ireland, a Boston College economist.

Last year, MIT Press published a book by him called Getting It Wrong: How Faulty Monetary Statistics Undermine the Fed, the Financial System, and the Economy. It earned blurbs from top economists such as James Heckman, a Nobel prize winner at the University of Chicago. Perhaps more important, the Federal Reserve’s measures to stimulate the economy have given banks $1.6 trillion in excess reserves. If they start lending in earnest, the money supply will jump and measuring it will suddenly matter again.

The Bank of England, Bank of Israel, and National Bank of Poland all report Divisia-type money data. The U.S. Bureau of Labor Statistics measures inflation with a Divisia-like method. Even the Fed uses a Divisia-like approach in its industrial production index. So why doesn’t it report the money supply that way? In a written statement, John Driscoll, a senior economist in the Fed’s Division of Monetary Affairs, says, “While Divisia indexes may be a theoretically useful way to think about money demand, they don’t solve certain practical issues that arise in using money as an indicator for policy. And in real time, such indexes can behave in unexpected ways.” In short, Fed officials still say that interest rates are more reliable than money supply as a dashboard indicator of growth and inflation. Would a surge of money creation change their minds?

Before it's here, it's on the Bloomberg Terminal.