The FT has just published an op-ed from one Professor Jeremy Siegel of Wharton. In it, he predicts that the US is going to experience “an extremely inflationary economy in 2021,” where “inflation will run well above the Fed’s 2 per cent target, and will do so for several years.”
This is a contrarian position. The Federal Open Market Committee – the body at the Fed which sets monetary policy – does not project inflation reaching 2% until 2022. Professor Siegel joins the vocal minority warning about inflation.
Professor Siegel bases his argument on an old proposition: the Quantity Theory of Money (QTM). Simplified, it argues that the price level is a function of the quantity of money in an economy. If the quantity of money increases faster than output, all other things being equal, prices will rise.
The QTM has motivated previous inflation warnings. After the Global Financial Crisis, some warned that Quantitative Easing (QE), with its massive expansion of the money supply, would cause runaway inflation. The opposite happened, and the major macroeconomic puzzle of the last decade has been falling, not rising, prices.
Professor Siegel argues this time is different (always a red flag). QE created trillions in reserves, but it turned out banks were happy to sit on them, and not make loans, in part because the Fed began paying interest on excess reserves. Today, thanks to the stimulus checks, grants to local government, and the Paycheck Protection Program, money is “going directly into the bank accounts of individuals and firms.” Where before money wound up in bank vaults, this time it will get into the hands of private sector individuals and be spent.
Is that really the case though?
Siegel’s argument rests upon a distinction between the money held by banks as excess reserves, and the money now in consumer bank accounts; growth of the former had little or no effect on inflation, but growth in the latter will.
This raises the question of why money in a consumer bank account is more likely to be spent than bank reserves? Siegel suggests that liquidity requirements, and the interest the Fed pays on excess reserves, explains the tendency of reserves to accumulate. But if so, why should it be any different with consumers, who, as we all know, also receive interest.
The Fed started paying interest on reserves, and regulators imposed liquidity requirements that could be satisfied with these reserves. The banks easily absorbed the extra reserves created by the Fed and quantitative easing led to only a modest increase in lending.
But, why is money sometimes stashed under the proverbial mattress, and other times not? It seems unlikely the answer is directly related to how many notes are under the pillow. It is not enough to simply increase the money supply, you need people to want to spend it.
In fairness, the QTM does acknowledge this. In addition to the price level, the quantity of money, and a measure of real output, it includes a variable called the “velocity of money,” a measure of how often it changes hands.
High velocity looks like this. I pay you $100 for a bicycle, you use it to (under)pay your neighbor’s son to paint your house, and he spends it taking his girlfriend to a gauche mini-golf bar. In a low velocity world, I buy your bicycle, and you stash the money between the pages of your untouched copy of War and Peace. A healthy economy looks more like the former than the latter.
The problem is, if velocity is subject to unpredictable changes, then there can be no fixed relationship between quantity and the price level. Instead we must look to a third (or fourth or fifth) variable to explain why velocity changes. But, if one wants to posit a stable relationship between quantity and price level, as Professor Siegel does, then velocity must be relatively static in the short to medium term. Otherwise, price is conditional on the quantity of money, which is conditional on velocity, which is conditional on some other unknown. At best we could say the quantity of money is a necessary condition of inflation.
The point is to be skeptical of mechanistic explanations, especially ones first developed in the 18th century. The money supply is broadly correlated with inflation, but that does not mean they are causally related. Even if they were, it would not necessarily tell us the direction or magnitude of the effect.
Let us also look skeptically on theories which avoid falsification tests and rely on verification. QTM theorists can always find historical episodes to verify their arguments. In fact, the allure of the “more money = more inflation” argument is that it is right eventually; at some point there will be inflation, and it is more than likely it will follow an increase in the money supply, if only because the money supply is usually increasing.
To paraphrase Paul Samuelson, inflation hawks have predicted 9 of the last 5 inflationary episodes. Despite this, the opinion pages of major newspapers always seem to find room for them.
It is unclear what would constitute a falsification of their hypothesis. Inflation failed to rise after the GFC, and a series of ad hoc explanations were provided. If inflation fails to increase in 2021, is that a definitive rejection of the QTM, or will more ad hoc explanations be appended to explain it away?
For the second time this week, I am reminded of the Marxists, who, when confronted by the failure of history to match their deterministic predictions (when is the revolution?), could always respond “not yet.”
We remain waiting.