The money supply is falling, and there is a risk the Fed may overdo it and cause total spending growth to fall, which would produce a recession.
Money still matters, and better measures of the money supply exist now than before.
F or the past several months, the Federal Reserve has been sharply raising interest rates to try to curtail record inflation. It’s effectively trying to make up for its earlier actions last year when it pumped too much money into the economy. Now, the money supply is falling, and there is a risk the Fed may overdo it and cause total spending growth to fall, precipitously. This, in turn, would produce an excessively severe recession. In testimony before Congress that may surprise some casual observers, Fed chairman Jerome Powell stated “[money] . . . doesn’t really have important implication for the economic outlook.” They ignore money at our peril.
How could the chairman of the Fed, the body that determines monetary policy, say money doesn’t matter? To grasp why, we first need to understand some history.
In the 1960s and 1970s, many economists thought the money supply was very important for explaining changes in the economy. Perhaps most famously, economists Milton Friedman and Anna Schwartz argued that the Fed caused the Great Depression by allowing the money supply to collapse in the early 1930s. Although the Volcker Fed ended the Great Inflation by bringing down money growth, later research from the 1990s suggested that the relationship between money and other key variables such as output and inflation had broken down by the 1980s. This later research is what informs central banks today.
This contemporary view, however, is based on a faulty premise. Let’s start by clarifying the meaning of money and how we add up different kinds of money. When people talk about the money supply, they are usually talking about the M1 and M2 measures published by the Fed. Historically, M1 mostly consisted of currency in circulation and checking accounts, while M2 mostly consisted of M1 plus savings accounts and small certificates of deposit (CDs). In May 2020, their definitions changed slightly because savings accounts were added to M1.
The Fed’s M1 and M2 measures are called “simple-sum” because they treat their respective components as if they are interchangeable. In reality, however, these components are not the same. Savings accounts are less liquid than cash in your wallet, and CDs are less liquid than savings accounts.
Fortunately, the economist William Barnett figured out a way to address this problem. He derived what are called “Divisia” money-supply measures. Divisia measures give higher weights to more liquid kinds of money and lower weights to less liquid kinds of money. Several economists have found Divisia measures explain changes in the economy well, even if simple-sum measures do not.
Additionally, while M1 and M2 are the most-cited measures, they do not include other assets that serve as money. For example, the Fed used to report M3, which consisted of M2 plus other assets such as repurchase agreements (repos), which are important to the financial sector. In 2006, the Fed stopped publishing M3, arguing that it was not useful for explaining economic activity. However, the real problem was that the Fed added up the components of M3 in a way that did not make sense.
Barnett and his colleagues at the Center for Financial Stability, an independent think tank, publish their own Divisia measures, which are updated monthly. These include Divisia M1 and Divisia M2 — which include the same components as the Fed’s simple-sum M1 and M2 measures, only weighted differently — and other measures including Divisia M3 and an even broader measure, Divisia M4. Divisia M3 consists of the same assets previously included in the Fed’s simple-sum M3, and Divisia M4 includes these assets as well as commercial paper (a form of “IOU”) and short-term Treasury bonds.
That brings us back to where we are today. The graph below shows the year-over-year growth of three money-supply measures: simple-sum M2, Divisia M3, and Divisia M4, from 2000 to the present. Noticeably, the two Divisia series show much steeper contractions in 2009 through 2011 than simple-sum M2. This reflects the Great Recession and the crisis faced by the “shadow banks,” bank-like entities such as subprime-mortgage lenders and hedge funds. In a 2011 book titled Getting It Wrong, Barnett argues that the Fed’s failure to properly measure money led to poor monetary policy during this time.
Looking now at the present era, we see each series spike in 2020 and remain high through most of 2021. Each has since come crashing down. By any of these metrics, money growth is now very close to zero and will likely soon be negative.
What does that mean for inflation? It’s hard to say with too much precision because changes in money do not lead to one-for-one changes in the price level, especially over short time frames. However, if money growth continues to stay very low, then high inflation cannot persist.
But, the Fed should also proceed with caution. Very low money growth may be a welcome development given the explosion in money growth two years ago. However, excessively low money growth can lead to recession if it causes spending growth to fall too quickly. If the Fed’s current policy causes broad Divisia money growth to turn steeply negative as it did during the Great Recession, there is a good possibility we will face a serious economic downturn.