Dude, Where's My Inflation?
From time to time, Austrian economists, goldbugs, and other economists on the saner end of the spectrum warn of the inflationary dangers of modern central bank policies. The most famous case of this was Bob Murphy’s bet with David Henderson that the Federal Reserve’s policies after the financial crisis of 2008 would result in high, i.e., double digit inflation. Unfortunately for Murphy, he lost this bet, and, more generally, the much-dreaded (or in the case of central banks, much-hoped-for) inflation has so far failed to materialize. This is true both when one looks at the official inflation measures and even when one consults alternatives such as shadowstats. What explains this apparent failure of Austrian commentators?
In order to understand the problem, it is necessary to get clear on the basics of the discussion. The Austrian point is that increases in the money supply—the old definition of inflation—must eventually result in a rise in the prices of all goods, what is nowadays meant by inflation. However, most official inflation measures, notably the Consumer Price Index, show only a slight increase. And while alternative measures usually show a higher rate of increase, even these are far from proportional to the increase in the money supply. Now, while the construction of any index measuring the change in consumer prices is arbitrary (as explained by Dr. Karl-Friedrich Israel here), it is surely not completely out of line with everyday experience; while prices on consumer goods do generally go up, it is usually not by more than a few percentage points per year.
Monetary experts counter the claims of those of us who think that increases in the money supply will necessarily result in inflation by arguing that this is not necessarily so; reverting to the equation of exchange, MV=PT, they say that an increase in the money supply need not be inflationary if it simply counters a decline in velocity—indeed, if the increase does not completely make up for the decline in velocity, they will even claim that an expansionary monetary policy increasing the supply of money in existence, is in fact a tight monetary policy since total spending in the economy declines.
We thus have a twofold problem before us: Why is expansionary monetary policy not reflected in increasing price inflation? And is it true to say that money is tight if an increase in M does not completely offset a decline in V? We will answer the second question first.
What Causes a Collapse in Velocity?
Why does the velocity of circulation change and suddenly decline, as it for instance did in the aftermath of the 2008 financial crisis? While the equation of exchange is a poor stand-in for monetary theory (I have criticized it here, and other, more able economists have done so here and here), it does point at a real phenomenon. That is to say, velocity is simply the variable introduced to equate the money supply with total spending. Therefore, if V declines while the money supply stays the same or increases, logically this must mean that total spending in the economy declines. The real question therefore is, why does total spending decline, and is it necessary to offset the fall in nominal spending by an increase in the money supply?
A fall in nominal spending occurs when people decide to spend less of their incomes than previously. That is, a fall in nominal spending is simply the simultaneous consequence of an increase in the demand for money. During crises and recessions, acting individuals experience greater uncertainty and therefore naturally postpone and restrict their purchases. Since they don’t know what the immediate future will bring, they lay more money aside for emergencies, as ready cash, apart from during hyperinflation, is considered the safest asset to possess in the short term. When the crisis has passed, people will gradually draw down their cash balances, as these will then be too large for their requirements.
It is clear that an increase in the demand for money to hold will result in lower prices and is in this sense deflationary. But that does not mean that money is tight or that it is necessary to replace the fall in spending with newly printed money. It is true that having to adjust prices downward will be painful for businesses and workers, especially since the official dogma that wages must not, under any circumstances, fall has been preached for generations. Yet falling prices are precisely what is necessary to get the economy back on track after a depression: the weakest businesses are purged and productive assets are made cheaply available for entrepreneurs, who expect to profit by buying factories, machines, and so on at low prices, hire laborers at lower wages, and restart production in profitable sectors of the economy. There is nothing in this process that requires the injection of new money into the economy; on the contrary, such injections can only frustrate the attempts of capitalists and entrepreneurs to readjust the production structure to the real conditions of the market and the real demands of consumers.
A decline in velocity therefore reflects a crisis in the economy; it does not mean that money is tight but rather that people have stopped spending as much of their incomes as they used to. It also means that even very substantial increases in the money supply do not lead to price inflation, as central banks cannot force people to spend the new money, or to take out new bank loans. Only when people are again willing to increase their spending will increases in the money supply result in higher prices.
What about inflation?
That increases in the money supply during a crisis do not result in price inflation is thus easily explained by an increase in the demand to hold money, but this does not explain the general failure of central banks in generating inflation. The rate of growth in the money supply has almost always outpaced the rate of growth in consumer price inflation. Does this not mean that, whatever other evils we may claim follow from central bank interventions, the dangers of inflation are significantly overstated?
The confusion arises from ignoring the entry point of the new money into the economy. The new money does not instantaneously spread throughout the economy and prices do not immediately adjust. Rather, the new money goes to a few recipients first, who then spend it on goods and services, bidding up prices, and the next recipients then spend it, again bidding up prices, and so on and so forth, until the money has been distributed and the whole price structure has been changed by the infusion. This is what is known as the Cantillon effect. Hayek in a famous image compared it to the pouring of honey on a surface: the honey does not immediately spread out across the surface but forms a mound at the point where it is poured out and only slowly spreads from there.1
Modern central banks, however, don’t pour money on consumers. Rather, they produce new money by injecting it into the financial system. The Federal Reserve buys US Treasurys and other financial assets from banks with new money, and the banks then spend the new money during their normal business operations. What we therefore have is a system of financial inflation: the prices of bonds and other financial instruments are bid up, leading to ever-lower rates of interest on loans, real estate prices are bid up, as it is now cheaper to finance mortgages, and only much later does the new money filter down to normal consumer goods. The rise in CPI is thus only a shadow of the distortive effects brought about by the increase in the money supply.
This does not mean that central banks would be able to cause inflation even were they to give the new money directly to consumers, for they cannot force each individual to spend the money on consumption. We saw an example of this recently when the US government distributed the CARES Act stimulus checks, the so-called Trumpbucks. I previously suggested that this kind of inflationary spending might quickly result in high inflation. However, I was mistaken in assuming that the money would be spent on consumption; most Trumpbucks seem to have been used to either increase savings or reduce debt. Thus the inflationary impact has been redirected or absorbed into financial markets or higher cash balances. (It is even possible that it has been neutralized to the extent that individuals have reduced or paid off their bank loans—if the banks have neglected to offer new loans in order to solidify their own position.)
Does this mean that the Federal Reserve is powerless to cause inflation? Not at all; but if their goal truly is to cause consumer price inflation, most recently stated in the form of “flexible average inflation targeting” (meaning, in effect, higher inflation), they have chosen a very roundabout way to do it. During normal times, “helicopter money” handed out to the population as described by former Fed chief Bernanke is a better way to stoke inflation than is pumping money into financial markets, and an even surer way is to simply finance government expenditures by the printing press. That they don’t do this suggests that their primary objective is not really to increase inflation, but rather to support financial markets and players in the style to which they have become accustomed. This is even official central-bank ideology these days, as they assume the mantle of “market-maker of last resort.”
- 1. Friedrich August von Hayek, "Three Elucidations of the Ricardo Effect," Journal of Political Economy 77, no. 2 (1969), p. 281, https://doi.org/10.1086/259514.