Devon Zuegel writes:
Inflation hits some parts of the economy harder and faster than others. It’s obvious once you say it, and yet the way pundits and academics talk about inflation glosses over this reality. As a result, most people who haven’t had direct experience with high inflation have a flawed view of how it affects daily life.
She then adds:
Popular use of the term “inflation” makes it sound like the inflation rate is a single indisputable number that you can plug into a formula to make decisions. [Bold is hers.] A pen pal recently made the following (paraphrased) comment in an email to me:
Inflation isn’t actually a problem, since all it’s doing is changing the measuring stick. Businesses can just price it into what they charge, workers can just price it into their wage negotiations, and banks can just price it into the interest rates on their loans.
Nominal prices go up, but on a relative basis, real prices are essentially the same.
Unfortunately it’s not that simple. That model doesn’t factor in information asymmetry, time delays, or structural differences between economic actors. It can take years for an inflation shock to propagate through the economy and reach a stable equilibrium, and it only does so after wreaking havoc on people’s expectations about money and commerce.
This is from Devon Zuegel, “Inflation propagates unevenly,” January 2, 2022.
Zuegel is right. In fact, economists have a term for the uneven propagation of inflation: Cantillon effects. The term is named after late 17th century and early 18th century Irish economist Richard Cantillon, who studied the phenomenon.
Then Zuegel goes on to give an example, writing:
Current prices in the US illustrate the effect as well. The NYT writes that “prices have soared for physical products but have risen only modestly for services. The cost of gasoline is up 58% in the last year, while health insurance prices have fallen almost 4%. Meat prices are up 13%, dairy 1.6%. Boys’ apparel is up 8.4%, while girls’ apparel is down 0.4%.”
Here’s the problem, and the problem is partly with Zuegel’s analysis and partly with inflation. The problem with Zuegel’s analysis is that she can’t say confidently, as she does above, that those very different rates of price change reflect the uneven propagation of inflation. In an economy with much lower inflation, there would still be changes in relative prices. Are Cantillon effects present in the prices she quotes in the above paragraph? Possibly. But it could also be mainly changes in relative prices that would have occurred without inflation.
The problem with inflation is that we don’t know.
Here’s more on Richard Cantillon, who is pictured above.
READER COMMENTS
JVM
Jan 17 2022 at 8:48pm
Devon is a she, just FYI.
David Henderson
Jan 17 2022 at 10:07pm
Thanks. Correction made.
Dylan
Jan 18 2022 at 7:33pm
One more correction needed in your 2nd blockquote.
David Henderson
Jan 18 2022 at 10:32pm
Thanks.
Thomas Lee Hutcheson
Jan 17 2022 at 9:23pm
I don’t know which academics she has not been listening to. If inflation were uniform, (except for relative price change due to real shocks) then it would be costless. I don’t know of anyone who things it is costless, especially central banks who have to balance the benefits of greater relative price changes, given downward stickiness with the costs of making the formation of good relative price expectations worse. It is true that “pundits” focus mainly on the redistributional effects of inflation and not dead-weight loss, but that just means one should not expect enlightened expositio of economics from “pundits.”
Mark Z
Jan 18 2022 at 9:45am
I’m pretty sure inflation has other costs besides Cantillon effects.
AMW
Jan 17 2022 at 10:19pm
Way back in 2004 I ran lab experiments with a two-good economy and fiat currency. When I gave additional currency to buyers of one of the goods, you could observe its relative price increase. Pretty fun stuff. I got a dissertation out of it.
David Henderson
Jan 17 2022 at 11:54pm
Nice.
Nicholas Decker
Jan 18 2022 at 1:48am
Perhaps I am entirely wrong on this, and my theories simply the consequence of callow youth. However, wouldn’t all of the prices changes that are different from increase in money supply x velocity / output be relative? Can’t one find that result? Could it be that we simply associate times of supply changes with inflation because, given y amount of money to be put in over time, a reduction in supply generates inflation, and so inflation doesn’t actually propagate unevenly – it is simply more common in times when supply is changing or being reduced? Perhaps the publically known change in inflation gives people an opportunity to move sticky prices to where they should be?
I am not doubtful that varying inflation rates has negative consequences. It is sure to reduce certainty and make investing riskier. While theoretically any level would do, it must be positive so as to deal with sticky prices, while not too high as to be a pain in the ass. However, I am very doubtful that inflation actually propagates unevenly, and that we have correctly described what here is cause, and effect.
Matthias
Jan 18 2022 at 6:24am
Careful with that Cantilpon effect. We don’t live in a ‘hydraulic’ economy where changes propagate like pressure in a fluid.
Markets like to anticipate expected price changes. Especially highly developed markets with derivatives and futures, like eg commodity markets or equity markets.
Jon Murphy
Jan 19 2022 at 8:54am
One needn’t have to consider the economy as a hydraulic machine to worry about changes propagating through different channels. One need only recognize that time is a factor. Many economists who refused to model the economy as an engine (eg Hayek, Mises, Smith, Cantillion, etc) all had models where changes propagate at different speeds. Indeed, I’d argue that such effects are a key factor to the Austrian Business Cycle Theory.
Matthias
Jan 26 2022 at 10:58pm
Yes, and that’s why Austrian business cycle theory is mostly nonsense. It assumes that entrepreneurs are idiots whose actions don’t anticipate changes they already expect. (At least in the versions of the theory that I’ve read. I might have seen some simplifications only?)
Yes, changes might reach different parts of the economy at different times. But that has more to do with how quickly people learn about a change, and perhaps with nominal rigidity and transaction costs and menu printing costs.
A theory of how changes propagate through the economy needs to put expectations and information front and centre.
Comments are closed.