Almost every time I see an expert interviewed on the macroeconomy, they suggest that a substantial portion of the inflation over the past 5 years has been supply side. That’s wrong; none of it has been supply side. I’d go even further; essentially none of the inflation over the past 50 years has been supply side.
To be clear, I am speaking of the total cumulative increase in prices over 5 years, or over 50 years. It is true that some of the inflation in 1979 was supply side, as well as some of the inflation during 2008, or 2022. There have been individual years where negative supply shocks pushed up prices, but just as many years where positive supply shocks pushed down prices.
Many experts implicitly seem to think there’s some sort of “ratchet effect”, where negative supply shocks push up prices, and then inflation settles back to its average rate. That’s false. When negative supply shocks are not causing inflation to rise above average, positive supply shocks cause it to fall below average.
West Texas crude currently trades at just over $70/barrel. The graph below shows real oil prices over the past 80 years (deflated by the CPI):
Adjusted for inflation, oil prices are about the same as they were in the late 2010s, and about the same as they were in the mid-1970s. That means that the nominal price of oil has risen at roughly the same rate as the overall CPI over the past 5 years, and indeed over the past 50 years. Oil doesn’t explain long run inflation at all.
[To be fair, there was a permanent one-time rise in real oil prices during 1973, when the OPEC moved the industry from being a competitive market to a cartel. Since then, it’s been mostly fluctuations round a real price of about $70/barrel.]
When oil prices rise faster than the CPI, it puts upward pressure on the CPI. Technically, the Fed could prevent this, but due to its dual mandate it generally allows higher oil prices to pass through to higher consumer prices. When oil prices rise slower than the CPI, it puts downward pressure on the CPI. Because oil prices have risen at roughly the same rate as the CPI over the past 5 years, and even over the past 50 years, oil shocks have had essentially no long run impact on the cost of living. None. The same is true of food price shocks, supply chain shocks, and other types of supply shocks. They are a non-factor for long run inflation.
So why do so many experts insist that supply shocks played a big role in the unusual inflation over the past 5 years? They seem to have made the following error. They correctly observed that negative supply shocks pushed consumer prices higher during 2022, but forget to note that positive supply shocks had an equally powerful downward effect on inflation during other recent years. In other words, the supply shock part of the problem really was transitory.
So why wasn’t the overall inflation rate transitory, as many had predicted? The answer is simple. All of the cumulative inflation since 2019 is demand side, and demand side inflation is permanent. PCE inflation over the past 5 years has exceeded the Fed’s 2% target by a total of nearly 8%. NGDP growth has exceeded 4%/year by a total of roughly 10%. That’s the entire problem—supply shocks have nothing to do with it. If anything, we’ve had enough positive supply shocks (mostly immigration) to hold inflation 2% below the level you would expect from the extreme demand stimulus that was provided. The Fed actually got lucky.
READER COMMENTS
Mark Brophy
Sep 19 2024 at 4:06am
Inflation is caused by the government printing money, inflation can even be defined as the government printing money. Right on the note (money), it says, “This note is legal tender for all debts, public and private.” It’s simply a way for the government to borrow and pay interest rather than raise taxes to expand the government. The interest is a hidden tax increase that the citizens find more palatable than the government openly and honestly increasing taxes.
The interest cost is one trillion dollars per year and is increasing. This game will not last indefinitely.
Matthias
Sep 19 2024 at 7:49am
It doesn’t say anything like you what you quoted on my notes here in my part of the world. Yet we sometimes have inflation. So I doubt it’s relevant.
For all your talks about inflation, quoting nominal interest payments is rather silly. (Even more extreme, when nominal interest rates are below inflation, real interest rates are negative. That’s not the case for the US at the moment, but was true for much of the 2010s.)
In any case, be careful not to mix up monetary policy and fiscal policy.
Rand
Sep 20 2024 at 9:44am
Yes. I concur. The rapidly increasing debasement of the dollar (and all other Fiat currencies) due to collective printing of money by central banks— these are the true and nearly unspoken causes of inflation. And they’re all getting ready to crank up the printing again as we enter into refi section of the 4 year biz cycle (see Raoul Pal’s work)Inflation due to above is set to make a massive jump again simply due to continued debasement of US dollar.Watch this video for example of possible scenario:https://youtu.be/AlBOc7Xq9Qc?t=171
Matthias
Sep 19 2024 at 7:52am
Well, it sounds like in the long run supply factors not only don’t cause inflation, they are actually deflationary. Increasing productivity decreases prices (assuming a given level of nominal gdp.)
Scott H.
Sep 19 2024 at 8:58am
Why isn’t demand side inflation more transitory?
My take: Because, while supply side item shocks are bad, a deflationary demand side money shock would be disastrous. And money has a completely elastic supply, so money never needs to be involved in a negative shock like oil or other products. Therefore, the FED just never lets there be any demand side negative shocks. And subsequently, the positive demand shocks never get offset.
dmm
Sep 19 2024 at 10:17am
I detest the term “demand side inflation “. It conceals the source of the extra demand, which is excessive money printing. Just like the original meaning of inflation itself was corrupted from “excessive money printing “ to “general price rise”.
Kurt Schuler
Sep 19 2024 at 10:53pm
Agree. Classifying an increase in the money supply under “demand” is confusing. It is better to divide changes in prices into goods-side and money-side changes. Both the goods side and on the money side have supply and demand. Rising prices can come from rising supply of money, falling demand for money, falling supply of goods, or rising demand for goods. In practice, though, persistent, years-long increases in the general level of prices only come from the supply of money rising faster than the demand for a given price level.
Don Geddis
Sep 19 2024 at 11:27pm
But the supply of money is completely arbitrary, under the control of the central bank, and the US Federal Reserve has an explicit mandate for “stable prices”. So regardless of what happens to goods in the real economy, if the general level of prices rise (too much), then the cause is ALWAYS that the economy has too much money.
It doesn’t actually matter whether the quantity of money rose or not. The quantity of money might have even fallen (presumably while something even more dramatic happened to goods). It is STILL the case that the “problem” of inflation is “too much money”. “Rising supply of money” is NOT the issue. “Too much money” (for the current economic conditions) is the issue.
Bill Conerly
Sep 19 2024 at 4:15pm
Confusion comes because the inflationary process involves supply constraints. In some markets, institutional factors or traditional ways of doing business prevent rapid price adjustment. Some products are then in short supply relative to the new, increased demand.
If an economy operating at full employment & industrial capacity has excessive stimulus, the first steep price increases come for products with high short-run income elasticity of demand and low short-run price elasticity of supply. (Used car prices rising after pandemic stimulus were a sign of excessive stimulus.)
The two factors (slow price changes for some products, and specific prices going up first) will look like supply constraints. And the fact is, if all products had infinite price elasticity of supply, then all changes in nominal GDP would be quantity changes, not price.
Robert EV
Sep 19 2024 at 10:23pm
Cost of living and inflation aren’t synonymous. Cost of living factors in things such as costs to borrow which are not factored into any of the standard inflation metrics that I am aware of.
An oil price shock can result in a person using credit when they otherwise wouldn’t have. A corresponding price shock in the other direction, even if of equal magnitude, will not allow the person to fully pay off the borrowed money at any consumer-available interest rates. The same with any other price shocks. Only substitution works here to keep cost of living in line with inflation. And as we can see with trends in borrowing, this substitution is not happening to the extent necessary to do so.
Steve C
Sep 20 2024 at 4:45am
Quite right. The concepts here may be easier to explain. A supply shock to the economy exists when necessary goods of some type(s) becomes scarce (e.g., energy, food, etc.). If the economy is forced to pay more for A, B, and C, then there is less money available to spend on X, Y, and Z. Downward pressure on X/Y/Z prices counterbalances the inflationary effects of price rises on A/B/C. However, since prices are sticky in nature, it takes time for price equilibrium to be re-established. Meanwhile, transitory inflation is present.
The real issue is the amount of money in circulation, and the quantity of goods and services available for purchase by those dollars. The Fed, in concert with the Treasury, has well-developed tools for dialing up and down the amount of money in circulation. This is a whole separate discussion.
Yes, immigration increases supply of labor, putting downward pressure on wages overall. I’m a bit skeptical that changes in immigration rates alone are the major reason why inflation was quite low for many years, but this may certainly be a factor.
Scott Sumner
Sep 20 2024 at 1:45pm
“I’m a bit skeptical that changes in immigration rates alone are the major reason why inflation was quite low for many years,”
I agree. My claim was that the 2021-23 surge in immigration kept inflation from rising as much as it otherwise would have.
spencer
Sep 20 2024 at 11:16am
re: “Oil doesn’t explain long run inflation at all”
Sumner is right once again, in the minority of overall economic analyses.
Princeton economist Alan Blinder says that “food shocks and OPEC ii (supply shocks) deserve much more blame for the alarming rise in inflation in 1979-1980.”
“Alan Blinder wrote that inflation “still looks transitory” but admits “that doesn’t mean it will be over in a month or two” (“When It Comes to Inflation, I’m Still on Team Transitory,” op-ed, Dec. 30, 2021).”
C-19’s run up in oil prices corresponded to the surge, all-time-high, in the distributed lag effect in the rate-of-change in long-term money flows, the volume and velocity of money, not the war in Ukraine.
Thomas L Hutcheson
Sep 21 2024 at 12:46pm
“When negative supply shocks are not causing inflation to rise above average, positive supply shocks cause it to fall below average.”
What on Earth could it even MEAN to say that a demand or supply shock could affect inflation? Granted shocks can create disequilibria in relative prices to which, given that some nominal prices are sticky downward, the Fed can aid adjustment with temporarily over-target inflation, but that is still the _Fed_ causing the inflation, not the shock.
When will people come to accept that inflation is always and everywhere a monetary phenomenon as Freidman said?
Robert EV
Sep 21 2024 at 2:29pm
Even if the Fed paid people to take money (by setting rates so low that even average consumers could borrow and be paid interest for borrowing), they can’t force people to spend that money on any particular thing.
Now it would be absurd, in such a loose money environment, for producers and retailers to not charge more for their products. But if what their aiming for is a monopoly, or at least an expansion of their market share, then it makes a bit more sense to keep prices the same or even lower them. It would also be absurd to keep doing a job when you can instead get money for free, but maybe enough people like working that they do keep working, even for free.
So theoretically money could be so free that it’s effectively valueless. Yet in such a situation consumers could be throwing their money into Crypto or other gambles that promise an even higher return than just borrowing money. And if people think that the free money era will eventually disappear they would still be incentivized to save instead of consume.
So it’s possible that a Fed loose money policy would not cause inflation. Therefore it has to be proven in some way that these non-inflationary, or even deflationary, scenarios would not happen under a loose monetary policy in order to claim “that inflation is always and everywhere a monetary phenomenon”.
Thomas L Hutcheson
Sep 22 2024 at 10:06am
Many things re possible, but I think experience shows that the Fed can create inflation if it wants to.
spencer
Sep 21 2024 at 4:06pm
Ed Fry’s debit and deposit turnover time series typically rose 3 months after a surge in the money stock ended. That would put the high in May 2022. The high in inflation was in June 2022, CPI (u) 9.1%
Gabriel
Sep 24 2024 at 11:50am
Hi Scott,
Possibly this is not the right post to ask my question but here it is.
I am a long term amateur student of economics (I became a technician instead of going to university, but always kept reading on economics)
Since about 2015, maybe a bit before, I have been convinced by your Market-Monetarist approach, I was a fan of Friedman, and later on Robert Lucas and Bryan Caplan; but I always was only half satisfied with the MacroEconomic theory.
I would back then defined my views during arguments as some form of “Neo-Monetarism”
All this to say that I flund myself almost in perfect agreement with your theory and arguments when I ended up discovering your work.
My question is this, at this point, in 2024, how mainstream do you believe Market Monetarism has become, and do you have a good link/read on that (eviddence that the profession is moving towards or against market monetarism)
Keep on the good work
Best
Gabriel