The idea that inflation must result if a money supply grows faster than the real economy over a sustained period is accepted as a general rule by most economists.
While this did not originate with Milton Friedman, his articulate evangelism forced central bankers around the world to pay more attention to the growth of money and thus ended most of the high inflations that were so common, especially in the developing world 25 to 50 years ago.
Economic interactions are complex, however, and there are some qualifications that apply to the simple rule of “fast money growth equals inflation.” The experience of the past five years demonstrates that. Since the global financial debacle began to unfold in 2008, the U.S. Federal Reserve and some other central banks have increased the “monetary base” that underlies their nations’ money supplies at unprecedented rates. Yet, inflation is not yet apparent.
As that process accentuated in the fall of 2008, many warned that rapid money supply growth would mean inflation. I wrote several columns warning of that possibility. In the 2012 campaign, several Republicans, notably Rick Perry and Ron Paul, argued that Fed policy under Ben Bernanke made inflation an imminent risk. Yet, inflation remains in check, except to the fringe who think the government cooks the books.
So are Milt and Rick and Ron and I all wet?
The answer is generally not. But the money supply-price level relationship is more complex than we thought. And much of the complexity stems from the fact that both “money” and “the money supply” are inherently less precise in their definition and measurement than one might think.
Just what is “money?” The usual textbook answer is that it is anything generally accepted in buying goods or paying debts. It might be gold or silver or pieces of paper that say “dollar” or “euro.” It might be a check or a money order. It might be a debit card. It might be frequent flier miles. Under certain circumstances, it might be cigarettes, as it was for POWs in World War II, or coarse bread as it was for Solzhenitsyn’s Ivan Denisovich and other prisoners in the Soviet Gulag.
Now what is the “money supply?” In introductory courses, we teach that you start with bills and coins in circulation and add checking accounts to get something called “M1.” Then you add in some other bank accounts and financial instruments and get “M2.” Add even more instruments and, until 2006 when the Fed stopped tabulating it, you got “M3.” For a few decades prior to 1994, the Fed also added a different set and got “L.” And there was also an “M4.”
At some point, the distinctions become highly technical, based on judgments as to how accessible assets such as certificates of deposit and money market mutual funds with “negotiable order of withdrawal” with check-writing privileges really are. Is their function close to cash or not? What about repurchase agreements and T-bills or bankers acceptances?
In the past six years, the Federal Reserve has given up on tabulating anything beyond “M1” or “narrow money” and “M2” or “broad money” plus the “monetary base” of currency plus bank reserves it directly controls. It argued that measures beyond M2 didn’t add anything in terms of understanding what was happening in the economy and even muddied the waters. They were generally right given the evidence at that time.
However, after the economy blew up in the fall of 2008 and the Fed was charged with having let a financial bubble develop, the Fed responded that growth in M2 had not been excessive. Monetary skeptics, who continued to tabulate M3 on their own after the Fed stopped, gleefully pointed out that this indicator did start to show excessive growth in the run-up to the bust and that if the Fed had not spurned it, they would have had some warning of their own errors.
A few specialists, most notably William A. Barnett, questioned the logic of this. Barnett, an aerospace engineer turned economist, devoted years of his Fed career to developing tabulations of the money supply. These included not only currency and bank accounts but also the “near monies” listed above, and yet had better explanatory power about the economy. His work was largely ignored within the broader discipline of economics and by policy makers. At least up to now.
Barnett found that broad measures of the money supply were better than narrow ones, if constructed properly. The different categories should not just be added up. They should be weighted, with items like cash given a high numerical coefficient and those like T-bills or bankers acceptances that were less perfectly money given a lower weight. That process is a statistical technique known as a “Divisia index.” With such an index, the broader measures of the money supply such as the M3 and M4 spurned by the Fed did have explanatory power that was much better than M1 and M2.
The Divisia index measure of the money supply remained an arcane topic for years until the financial crisis brought arguments about the Fed’s policy to a boil. The “monetary base” controlled by the Fed increased 145 percent in the past five years. “Broad money” or M2, the measure preferred by the Fed, increased 40 percent. The CPI has gone up 9.5 percent. But Barnett’s “Divisia M4” has grown by 10.1 percent, well below its trend line for the last 40 years. By that measure, the money supply is not growing fast and the threat of inflation is overblown.
Writers, including Paul Krugman, have mentioned the Divisia monetary aggregates in the past year, but Martin Wolf, writing in the Feb. 12 Financial Times, really lit a fire of interest. Wolf is the most respected economics columnist in the English-speaking world, and his columns are widely read by leading lights in finance and government.
Yes, how to best measure the “money supply” is an arcane topic of interest to few. But it also is turning out to be one of the most important economic policy questions of the day. If “growth in the money supply” directly causes inflation, to what measure of “the money supply” is inflation most directly related?
Expect to hear more about it.