College students studying macroeconomics are likely to encounter a simple scatter plot, putting the supply of money on one axis and rate of inflation on the other. Reliable data collection across countries didn’t really start until after World War II, so the graph starts in the 1950s and plots observations from around the world. The dots line up neatly around along a 45-degree line. More money = higher inflation.
Textbooks attribute this scatter plot to Robert Barro, a monetary economist at Harvard. It illustrates an old idea, commonly called the quantity theory of money. A year ago, Jens Weidmann, head of Germany’s central bank, reminded an audience in Frankfurt that money supply and inflation are treated in Goethe’s Faust, a seminal 18th century German drama. In 1963, Milton Friedman offered an often-quoted and sweeping statement: “Inflation is everywhere and always a monetary phenomenon.” And this month the European Central Bank published a working paper, “Is Quantity Theory Still Alive,” (pdf) that called the idea “one of the most established folk wisdoms in monetary economics.” The paper concludes that, if you correct for a few things, over the long run, quantity theory, yes, lives. More money = higher inflation.
Harald Uhlig of the University of Chicago, one of the paper’s authors, explained over the phone on Tuesday why it was necessary to return to where Goethe, Friedman, and Barro had already trod. For developed economies with competent central banks, he explains, the relationship has not been as clear, particularly over the past 20 years. Take Barro’s scatter plot, says Uhlig, and throw out instances of inflation of more than 12 percent. The dots still indicate a trend but don’t line up as neatly.
It might be tempting to conclude (and some economists have) that quantity theory holds for Zimbabwe and the Weimar Republic but has less value for developed, modern democracies that have been keeping inflation under 5 percent for the past 30 years. Over the past 20 years, monetary economists have been debating whether this simple and ancient relationship—more money = higher inflation—is still of practical value to a competent central banker in a reasonably run, market-based democracy (such as the U.S.).
So Uhlig and Pedro Teles, the paper’s lead author and senior economist at the Bank of Portugal, returned to the scatter plots, looking just at established economies with inflation below 12 percent. First, they corrected for growth in GDP, reasoning, as others have before, that if growth in money matches growth in economic output, it should have little effect on inflation. Second, they corrected for interest rates. Dramatically simplified, low interest rates encourage people to hold their money in cash.
“With correct theoretical bias,” says Uhlig, “things tend to fall in line.” After those two corrections, even for low-inflation economies, “it’s not perfect, but it’s a pretty good-looking relationship.” Even for economies with relatively low inflation, the relationship is still there. More money = higher inflation.
They offer one hitch, though. Restrict the data set to the years from 1990 to 2005, what most of us refer to as “the period of awesome,” and the relationship gets less clear again. Central banks had learned how to hit inflation targets with a variety of tools, and so during that period inflation hovers around the target, regardless of money supply. “We don’t have a particular axe to grind,” says Uhlig, “but I agree, that was a particularly stable period, [and central banks] were all surprisingly successful.”
There are two ways to look at the changes since 1990. One is to say, “We got this.” Central bankers do have more tools at their disposal than just the supply of money. The massive global failures since 2008 are more a consequence of the inadequacy of fiscal politics and financial supervision than they are of monetary policy. The other way is to remember that economists, like all mortals, suffer from generational amnesia and tend to think that the recent past is suggestive of the immediate future. “The warning here,” says Uhlig, “is if you look at longer time periods, the relationship is there. If money grows too quickly, inflation will come back.”
After 2008, advances in macroeconomic modeling tended to ignore the Great Depression as an outlier, preferring the cleaner data sets available since World War II. The painful lesson that modelers made was that the longer the data set, the more reliable the result. This may be true for central bankers as well.
The paper serves as a subtle warning to the Federal Reserve and the Bank of England, which at some point will need to sell what they’ve been adding to their balance sheets over the past four years. Done incautiously, this could expand the money supply, and according to the precepts of Goethe, Barro, and Friedman, lead to inflation. “I think the Fed could do it [successfully],” says Uhlig, but he adds, “One could read the paper as saying you better watch what happens with money in circulation.”
“We conclude,” Teles and Uhlig write, “that quantity theory is still alive.” If you don’t read a lot of central bank working papers, you might miss that that is meant as a bone-dry joke.