The Wall Street Journal’s report on the reduced increase of the Consumer Price Index is confused. Or so would think an economist who understands the difference between changes in relative prices and a change in the general price level, of all prices together. (See “Milder Inflation Opens Door Wider to September Rate Cut,” July 11, 2024.)
Reading the story, the proverbial Martian landing on Earth would think that, on this strange planet, all prices are the product of inflation (or deflation), that there is no difference between inflation (or deflation) and changes in relative prices, and that the increase in the CPI is inflation. Quoting from the story cited above:
Housing inflation, which measures the cost of renting and accounts for about one-third of the CPI, has kept overall prices high.
If there is such a thing as “housing inflation,” we have a puzzle. What happened to other “product X inflation” in the other two-thirds of goods and services—including, say, bubble gum inflation? Did that part also keep overall prices high? Or did the rest of the CPI show “product deflation” (the opposite of inflation, as we see during economic depressions)? Does the economy witness a continuous struggle between two evils, inflation in some prices and deflation in others? Or could it be that some prices increased or decreased relatively to each other whatever happens to inflation (or deflation)? But then, a contrarian economist would ask—as economists started asking a couple of centuries ago—what causes all prices to rise together besides changes in relative prices? What does it mean then to speak of housing inflation or gasoline deflation?
One easy way to think about relative prices is as follows. Assume that only one price changes in the economy, all other prices remaining constant. (This would of course implies the absence of inflation.) The price that changed has changed relative to all other prices (although in a different proportion of different prices). Arithmetically, if one relative price changes, all other prices also change relatively to that one. All prices are relative prices compared to other prices. Relative prices change all the time, whether there is inflation, deflation, or none of either. The two phenomena obviously need to be distinguished, inflation or deflation on one side, and specific relative price changes on the other.
Ryan Bourne’s recent book The War on Prices (Cato Institute, 2024) contains many discussions of prices besides your humble blogger’s chapter, “A Rising Product Price Doesn’t Cause Inflation.” Over the history of economic analysis, there is much theory and much evidence supporting the hypothesis that inflation—an increase in all prices together, as opposed to relative changes among them—is due to an increase in the money supply over and above what economic agents demand for their transactions. The price of any good or service as observed and included in the CPI is composed of a change in its relative price and, when there is inflation (or deflation), a change in the general level of prices. Reproduced below is a chart published in one of Ryan’s contributions to the book. The correlation between changes in the money supply and estimated inflation supports the hypothesis that inflation is a monetary phenomenon.
We should not be too hard on journalists and the WSJ is not alone in error. Most journalists are just echoing the noises they are hearing, including from many economists. Perhaps some economists are trying so hard to dumb down what they know for journalists and pundits that the latter end up thinking in dumbed-down economics: “inflation is the sum of all price increases,” they seem to think (while inflation, when present, constitutes only part of a price change). Other economists seem happy to forget the economic way of thinking when they leave graduate school and become accountants with large databases and sophisticated statistical software. Still other economists mainly have political points to make, which generally amount to saying that the government has got things under control and is taking good care of its “citizens.”
It is as if, in today’s economic Newspeak, the words to conceive of relative prices were being obliterated. Then, nothing is thinkable but housing inflation, grocery inflation, and bubble gum inflation, which, all added up, produce (total) inflation.
******************************
A Martian puzzled by Earthlings’ theories of bubble gum inflation
READER COMMENTS
steve
Jul 14 2024 at 1:30pm
The egg price “inflation” surely got a lot of coverage. That was depressing.
Steve
Pierre Lemieux
Jul 15 2024 at 2:05pm
Steve: Are you referring to the 2020 price increase of eggs?
Craig
Jul 15 2024 at 6:39pm
Saw the comment on eggs and have a somewhat humorous anecdote. In TN the local barber, there’s only one, charges $8 adult/$6 child. Yes, he’s that cheap and he’s not bad either. He ALSO sold his farm grown eggs. So I’d go for a haircut with my son, 8+6 = 14 + 1.50 for the eggs and I’d give him $20 and call it a day! Nevertheless he caught wind of the egg prices and he now charges $3 for a dozen eggs, but here’s what’s interesting. He didn’t change the price of his haircuts!
Pierre Lemieux
Jul 16 2024 at 12:09am
Craig: Your village barber knows the difference between relative prices and inflation better than the Wall Street Journal or the chairman of the Council of Economic Advisers!
I am sure however that he does not shave all the men of the village who don’t shave themselves.
nobody.really
Jul 17 2024 at 9:52pm
Alfred and Bertrand would be proud.
Jose Pablo
Jul 18 2024 at 12:09pm
Craig, if there is only one, why isn’t he jacking up prices like crazy?
More so taking into account that there are no easy substitutes for (good) haircuts (I tried DIY during Covid with awful results) and even more so if, like in Florida, in TN barbering is a licensed activity.
Craig
Jul 18 2024 at 1:06pm
He should, it should’ve gone up at least a dollar. I think the reason he hasn’t raised his price is because I get the sense he’s the type of person who might undervalue himself and feel funny about raising his price. Personally I won’t pay him less than $15 for a haircut and a dozen eggs. I get the sense that this is his way to charge those who can only give him $8 to give him $8 and if you have a couple of nickels, you pay him more because you can. That’s my Eddie Albert impression of what’s going on in Greenacres — very rural county, NOT affluent by any stretch of the imagination.
Craig
Jul 14 2024 at 7:18pm
“Or could it be that some prices increased or decreased relatively to each other whatever happens to inflation (or deflation)?”
Housing is just a funny animal because many are locked into a cost for a certain time period. My understanding is that OER when inflation was trending higher to a peak of 9.1% was dampening the mathematical calculation and on the way down OER is doing the exact opposite. On top of that rentals tend to be for a year so if you signed a lease in April 2021 and were answering a BLS survey in April 2022 your rent hadn’t gone up yet.
“The BLS uses data from owner-occupied units in the Consumer Expenditure Survey to derive expenditure weights. The expenditure weight in the CPI market basket for OER is based on the following question that the CE Survey asks of consumers who own their primary residence:
“If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?”
The Consumer Expenditure Survey asks the following questions of renter-occupied units to derive the expenditure weight for rent:
“What was your total rental payment for this month for this unit? Include any extra charges for garage or parking facilities, but do not include direct payments by local, state or federal agencies.” — https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-and-rent.htm
With respect to this question: ““If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?””
If somebody were to ask me that question today I really wouldn’t even know how to answer it and I’d likely just answer with my current mortgage payment, which, because I did a refi @ 2.25% went DOWN
And now the backend of inflation I start seeing homeowners go up because the cost to build housing went up and that’s in the renewal quote which took time to filter through.
If inflation is relatively stable at 1-2%, I think the measure if fine, but I think as inflation spikes up OER has some serious drawbacks in measuring inflation in the here and now. I think EVENTUALLY everything will flow through, but if inflation is at 9.1% in April 2022 and you’re asking me the OER question, I’m behind that curve there, its not happening to my housing payment yet. And so OER as a mathematical element in the CPI calculation dampens the overall result?
TMC
Jul 15 2024 at 9:38am
I had no idea how to answer the rent question on my own house, and I have rentals. This prompted me to search Zillow. Rent would be about 60% more than my payment. (Not your point, but) I’ll be paid off in 6 years where rent would be 400% of my payment.
Pierre Lemieux
Jul 15 2024 at 1:49pm
Craid: You may be right and the answer to your closing question may be yes. We would have to see if the statisticians have found a way to compensate for that.
My point, though, was different–as I am sure you know. Suppose the relative price of renting has increased. Someone aware of the housing market would have noticed that, to take illustrative numbers, while inflation is estimated at 9%, apartment rents in his neighborhood or town have been increasing by (say) 20%. Wouldn’t an owner-occupant be aware of that and adjust the estimate of his Owner’s Equivalent Rent by 20%? Isn’t that how the Craigs of this world would find a number to write down for their OERs? Some may overestimate, others underestimate, this figure, but if the error is not systematic, the average may have some information value. This example also illustrates my point that the housing component of the CPI will incorporate both inflation and relative price changes. There is no “housing inflation”: there is general inflation and change in housing relative prices.
As you say, there might still be lags, of course. But don’t lease signing dates vary wildly over the economy?
Craig
Jul 15 2024 at 2:34pm
“Wouldn’t an owner-occupant be aware of that and adjust the estimate of his Owner’s Equivalent Rent by 20%?”
Some will be of course, I will say this that I don’t even ponder what my home could rent for because the home itself isn’t suitable for renting in the sense that if I were to rent the home and you could afford it my first question would be, “Well, why aren’t you just buying a home then?”
I wouldn’t even think about renting it because anybody with the wherewithal to rent it wouldn’t be dumb enough to do something like that! I’m genuinely critical of OER from the get go, I really am.
Thomas L Hutcheson
Jul 17 2024 at 10:20pm
Owners do not estimate their own OER. It’s a BLS calculation.
Craig
Jul 18 2024 at 8:28am
The BLS specifically asks the question, please see my post above I actually do link to the BLS source wherein it states:
“The BLS uses data from owner-occupied units in the Consumer Expenditure Survey to derive expenditure weights. The expenditure weight in the CPI market basket for OER is based on the following question that the CE Survey asks of consumers who own their primary residence:
“If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?”
Thomas L Hutcheson
Jul 15 2024 at 4:13pm
It isn’t particularly helpful, but is it really WRONG to think of inflation, an increase in the average level of prices as made up of the average “inflations” of all of the individual components (including bubble gum)?
Also it makes some amount of sense to wonder aloud if targeting inflation as measured by one aggregation of items is better (higher real incomes in the long run) than another.
Craig
Jul 16 2024 at 10:38am
The way the BLS reports the inflation lends itself to this of course. Indeed one of the more well know inflation stats isn’t just the headline number its the ‘core’ inflation excluding food and energy.
Pierre Lemieux
Jul 16 2024 at 11:52am
Thomas: In response to the question in your first paragraph, I don’t think so. When equilibrium is reached, all prices will have increased by the same percentage (as the increase in the non-demanded money supply). At that point anyway, inflation cannot be conceived as a sum of “inflations”: it is the same as the “inflation” in any price, so you can just pick one, any one (assuming inflation is observable). The fact is that inflation is not observable in prices; what is observable is the total price of a good, which includes both inflation and the change in its relative price.
On your second paragraph, it seems to me that, if the quantity theory of money is valid, the obvious way to target inflation is to look at the money supply. Of course, you need to have a valid measure of the demand of money. With free money-issuing banks, supply would automatically follow demand, but let’s take the current statist status quo for granted: What do you say, then? (Scott would have something to say on that.)
Thomas L Hutcheson
Jul 16 2024 at 12:28pm
This is a pretty important difference in view.
I agree that inflation is produced by central banks [whether by manipulating the money supply or interest rates or whatever], but it has a purpose: to facilitate income maximizing changes in relative prices in response to “shocks” when some prices cannot easily adjust downward. [Often wages are though of as “sticky,” but real estate lease rates are probably even stickier] The increase in the price level would NOT be made up of all prices rising equally. Since “shocks” are constant, it makes sense for a central bank to aim for a constant, non-zero rate of inflation. An extraordinarily large shock would call for temporary over-target inflation.
[This view is conceptually compatible with the political judgement that the central bank at hand is likely to be so incompetent in adjusting inflation to the shocks that it would be better to have free banking with a commodity standard or to use the fiat currency of another country.]
Thomas L Hutcheson
Jul 16 2024 at 12:34pm
Concerning my second paragraph, I do not understand what you are asking me to opine on. Scott, of course does not think central banks should target inflation at all, but NGDP, but I have never really understood exactly why or if in practice it would even make any difference in the way a central bank set its policy instrument.
Comments are closed.