It’s all demand side

Here’s The Economist:

The economists who had warned that excessive stimulus and overheating demand, rather than production snarls, would make inflation a more serious problem seemed prescient. In the shorthand of the day, it looked as if “Team Persistent” had defeated “Team Transitory”.

Fast-forward to the present, and something strange has happened. The Fed, along with most other major central banks, has acted as if Team Persistent was right. It jacked up short-term interest rates from a floor of 0% to more than 5% in the space of 14 months. Sure enough, inflation has slowed sharply. But here is the odd thing: the opposite side of the debate is now celebrating. “We in Team Transitory can rightly claim victory,” declared Joseph Stiglitz, a Nobel prizewinner, in a recent essay.

It’s hard to believe that some prominent economists believe our recent inflation has been due to supply side factors. In fact, the inflation experienced over the past 4 years has been all demand side. The total amount of supply-side inflation has been zero, none, nada.

Over the past 4 years, the PCE price index is up 16.7%. Under FAIT it should have risen by 8.2% (i.e., 2%/year). Thus we’ve had roughly 8.5% excess inflation (a bit less due to compounding.)

Aggregate demand (NGDP) is up by 27.6%. Under FAIT targeting (which is similar to NGDPLT) it should have been up by about 17% (i.e., 4%/year). So we’ve had a bit less than 10.6% extra demand growth. That explains all of the extra inflation.

Supply shocks do explain inflation over shorter periods of time. But people who point to supply shocks tend to forget that they are transitory. For every negative supply shock like 2022, there’s a positive supply shock like 2023. Over any extend period of time the supply shocks cancel out, leaving virtually all excess inflation due to demand shocks.

It’s always been this way. The PCE price index rose at a 6.1% annual rate between 1965 and 1981. NGDP rose by 9.6% annual rate. That means the “Great Inflation” was 100% due to excessive growth in demand—supply shocks had nothing to do with it. Our textbooks are wrong. (Of course supply shocks do explain some of the inflation in individual years such as 1974 and 1980, but they do not contribute at all to the longer run trend in inflation during 1965-81.

As a first approximation, all supply side inflation is transitory, and all demand side inflation is permanent. Those who predicted transitory inflation (including for a brief period me) were completely wrong. In 2021, I naively believed the Fed was serious about FAIT.)

Some people seem to believe that transitory means something like the following inflation path:

2% 2% 4% 4% 4% 2% 2% 2% 2%

In fact, it would mean something like:

2% 2% 4% 4% 4% 0% 0% 0% 2%

If it’s truly transitory, then inflation should average 2% in the long run.



31 Responses to “It’s all demand side”

  1. Gravatar of Solon of the East Solon of the East
    6. March 2024 at 15:50

    This post is probably correct but does overlook the housing problem. Chronically tightening housing markets can be measured as inflation, but are actually a supply side problem.

  2. Gravatar of Student Student
    6. March 2024 at 16:12

    Can please elaborate on this a bit… I don’t see how this means 100%. “The PCE price index rose at a 6.1% annual rate between 1965 and 1981. NGDP rose by 9.6% annual rate. That means the “Great Inflation” was 100% due to excessive growth in demand—supply shocks had nothing to do with it.”

    Also Solon makes a good point. Housing is a supply side issue caused by dumb zoning rules… even you agree with that, right?

  3. Gravatar of bobster bobster
    6. March 2024 at 18:20

    Bingo. People struggle with the concept that with supply side inflation, the price level fully returns to its previous level.

    Transitory folks shifted to arguing that transitory means returning closer to 2% (and they’re even wrong there, we’re stuck at 3%).

  4. Gravatar of Viennacapitalist Viennacapitalist
    7. March 2024 at 00:54


    “…In 2021, I naively believed the Fed was serious about FAIT…”

    do you believe Fed actions show a Fed whose prime focus is NOT inflation control, but some other overriding goal (Public finances, stock market and wealth effect, etc.)?

    Can the Fed still considered to be independent, given its actions in the recent past?

  5. Gravatar of Sara Sara
    7. March 2024 at 02:25

    The efficient market hypothesis is wrong, and therefore Sumner is wrong. Subjective perceptions shape economics outcomes. Market participants can be influenced by psychological factors, emotions, and imperfect information, leading to the formation and bursting of bubbles

    Additionally, nominal GDP is a terrible indicator of economic health. If you spend 1M on a line of credit, then it will be included in nominal GDP. It will appear that demand has increased. Companies might boast of higher than expected returns, and traders might jump into a particular stock or asset, seeking to profit from the companies rise without ever reading the financial statements.
    But Nominal GDP says nothing about the health of an economy. Heavy borrowing doesn’t necessarily improve your economic position.
    Once the line of credit is reduced or eliminated, it’s more likely than not that the economy will contract as companies are unable to pay their debt. Banks may burst, government might be forced to print to meet FDIC deposits.
    In such a scenerio, people might flock to bitcoin or gold, and other assets, pushing up their prices.
    The Fed may artificially increase or decrease interest rates, intervening in the marketplace in such a way that distorts the true nature of reality, thus creating bubbles.
    And it’s not just Fed debt, btw, which is so bloated that it funds drag queen parties for the state department. Most small companies are living off their credit cards, borrowing into the future. And larger companies, particularly tech start-ups, are still losing money.

    All of that is unsustainable.

  6. Gravatar of Rajat Rajat
    7. March 2024 at 03:23

    I don’t disagree with your specific point about the current inflation in the US, but are all supply shocks necessarily transitory? You refer to the period from 1965 to 1981 in the US. Over that period, the 2024-dollars price of oil rose from about $28 per barrel up to $150 in 1980 and had fallen to $120 in 1981. It then didn’t really return to 1965 levels except for a brief period in 1998 and for a couple of months when Covid-19 started. See here:
    From an Australian small-country perspective, because many of our consumer and building products are imported and their prices have mostly not reversed since early 2020, what may have started as a demand-side shock elsewhere has become a sustained negative supply shock here. That’s not to say we haven’t had plenty of excess demand here ourselves, but I wonder if the fact that the US seems to have had the strongest NGDP growth since 2020 means that it has effectively ‘exported’ inflation, at least until PPP takes effect.

  7. Gravatar of spencer spencer
    7. March 2024 at 04:36

    AD = M*Vt. If AD stays the same, and oil prices increase, then other prices decrease. Contrary to Alan Blinder, the FED overcompensated for OPEC’s price rises.

    Housing prices and interest rates fell when Bernanke conducted the most restrictive monetary policy since the GD. Housing prices rose along with interest rates when Powell conducted the most expansionary monetary policy in U.S. history.

  8. Gravatar of bill bill
    7. March 2024 at 05:02

    I’m seeing a large percentage of Team Transitory claiming 2,2,4,4,4,3,3… is the new Transitory.

  9. Gravatar of Trina Halppe Trina Halppe
    7. March 2024 at 05:02

    When people say “higher inflation is transitory” I think it is commonly understood to refer to one of the periodically reported rates which are the focus of media attention and consequently public attention; not the average rate of multiple reporting periods that span multiple years. The average of multiple reporting periods that span multiple years never seems to be the focus of media attention and consequently not of public attention.

    The foregoing means I don’t agree with commenter bobster that Team Transitory shifted the goal posts. IIRC, Team Transitory seemed to be commenting on the same things that the media and the public were paying attention to.

    If and when people (other than our blog host) talk about a multi-year average, or any average at all, it seems they don’t say what the inputs are for the calculation so that makes it easier to declare victory someday.

  10. Gravatar of ssumner ssumner
    7. March 2024 at 08:59

    Student, Inflation isn’t high prices (housing or otherwise), it’s rising prices. If the Fed had kept NGDP growth at 5% during 1965-81, there would have been no Great Inflation, as RGDP growth averaged 3.2%.

    Viennacapitalist, I don’t think it’s the items you mentioned, I think they put too much weight on jobs, and had the wrong Phillips Curve model. Hopefully they learned their lesson. And their FAIT was asymmetrical, which was a big mistake.

    Sara, You packed a lot of errors into one short comment!

    Rajat, Yes, but during this period the US also experienced some huge positive AS shocks, notably a big surge in the labor force.

    In net terms, RGDP growth was fine, 3.2%/year. The high inflation was because the Fed didn’t hold NGDP growth to around 5%.

    Trina, No, the transitory people were claiming that inflation could be brought down without a tight money policy restricting AD. They thought it was just supply side increases.

    Everyone agrees that inflation would eventually come down—that’s not what the debate was about. Now the transitory people are claiming that it wasn’t necessary to sharply raise interest rates, which is absurd.

  11. Gravatar of Robert Simmons Robert Simmons
    7. March 2024 at 20:12

    Isn’t this pattern what transitory -should- mean, with level targeting, but that’s not actually the Fed’s approach?
    “2% 2% 4% 4% 4% 0% 0% 0% 2%”

  12. Gravatar of BC BC
    7. March 2024 at 20:58

    If the whole point of the transitory vs. persistent debate wasn’t over whether the Fed should tighten, then what was it about? Team Transitory thought that inflation in 2022 was due to transitory supply factors so that, if the Fed were to tighten, then as transitory factors eased the excessive Fed tightening would throw the economy into recession. Obviously, that didn’t happen.

    If Team Transitory didn’t/doesn’t believe that Fed tightening in response to transitory supply factors produces adverse effects, then why even debate transitory vs. persistent? The Fed could tighten in either case. So, either Team Transitory was wrong or, the next time they claim that inflation is transitory, the Fed can go ahead and tighten anyways as if Team Persistent is right. Team Transitory can’t have it both ways. They can’t both (1) deny that the fact that aggressive Fed tightening didn’t throw the economy into recession proves them wrong and (2) say that the Fed shouldn’t tighten in response to transitory supply shocks.

  13. Gravatar of Solon of the East Solon of the East
    8. March 2024 at 03:53

    Does the Fed need to shrink its balance sheet?

    The Bank of Japan has a gigantic balance sheet and inflation is not a problem there….

  14. Gravatar of William Peden William Peden
    8. March 2024 at 05:29


    It’s interesting to look back through what Stiglitz has said.

    By the second half of 2023, he was already covering his posterior by saying that maybe Biden was saving the day:

    Note how what he says in that interview is evidence-proof: if there was a recession, he could have said that he predicted it based on the “bad economics” of the Fed’s tightening policy; if there wasn’t a recession, he could have said that he predicted it based on the great economics of the Inflation Recovery Act.

    Therefore, I suggest that Stiglitz’s views and credibility should receive zero support from anything that has happened since at least September 2023. The Oracle of Delphi could learn a lot from his sophistry.

    Still, we can go back further than what Stiglitz said in 2023. We find more sophistry in 2022:

    For example, note how he uses consumption data, IN REAL TERMS, as his key evidence that aggregate demand has not been excessive. This seems appealing to many people on the surface, providing an argument for e.g. a US Senator wanting to criticise the Fed for tightening. However, that real consumption is less cyclical than aggregate nominal spending has been known since at least the days of Kuznets and the Cowles Commission. It’s like using city-only data to argue for the stability of global temperatures.

    I’ll let people guess how much he talks about nominal GDP or nominal incomes…

    Finally, we see what Stiglitz was really saying back in 2022: “Most importantly, such increases in interest rates will not substantially lower inflation unless they induce a major contraction in the economy, which is a cure worse than the disease… To the contrary, the paper explains several reasons why large and rapid increases in interest rates, beyond normalizing them, may be counterproductive. For instance, they could impede investments that might alleviate some of the supply shortages.”

    Of all the famous New Keynesians, Stiglitz seems to have gone back furthest to the views that might have been found in Cambridge UK around 1970. He has been wrong because that isn’t how economies actually work. He is now using his formidable polemical skills and powers of bluster to insulate his own credibility and those of his views from what he actually said.

    Team Transitory, which includes some smart and honest people, should want nothing to do with Stiglitz.

  15. Gravatar of William Peden William Peden
    8. March 2024 at 05:33

    To add to that: notice the persistence of the lightswitch model of demand-driven inflation that comes from implicitly (or explicitly) assuming a flat SRAS curve. I see that sort of thinking in historical texts from the Great Inflation (and often imbibed by historians writing about that period) all the time. And I see it today, including among professional economists (at seminars) who don’t know what an SRAS curve is…

  16. Gravatar of spencer spencer
    8. March 2024 at 06:23

    re: “The Bank of Japan has a gigantic balance sheet and inflation is not a problem there….”

    All things are not the same. It’s incontrovertible that from the standpoint of the system, macro-economics, banks don’t lend deposits. But all bank-held savings originate within the payment’s system.

    Secular stagnation, as predicted in “Should Commercial Banks Accept Savings Deposits?” Conference on Savings and Residential Financing 1961 Proceedings, United States Savings and loan league, Chicago, 1961, 42, 43., is due to the impoundment of savings in the banks.

    Dr. Philip George’s “The Riddle of Money Finally Solved” confirms this. He ended up saying the same thing Dr. Leland Pritchard said in IMTRAC at least 40 years earlier.

    The BOJ has unlimited transaction deposit insurance, the Japanese save more, and keep more of their savings in their banks.

    “Japanese households have 52% of their money in currency & deposits, vs 35% for people in the Eurozone and 14% for the US.”

  17. Gravatar of spencer spencer
    8. March 2024 at 06:37

    Mon 1/22/2024 10:12 AM
    Thank you for your question. Net assets of retail money funds are a component of the M2 money stock, while net assets of institutional money funds are not. For more information on the money stock measures published by the Federal Reserve, please see the H.6 statistical release ( If an investor transfers funds from a deposit account at their bank and places the funds into a retail money fund, the funds are accounted for in the retail money fund component of the M2 money stock rather than in the deposit components of M2. If a money fund is holding funds with a bank in the form of a transaction account (e.g., demand deposit or other liquid deposit) or small-time deposit (time deposit held in amounts less than $100,000), then those funds would be counted in the money stock but those amounts are likely to be rather small.
    Board staff

    It’s just like MSBs balances between 1913 and 1980.

    Leland J. Pritchard
    First published: March 1954

  18. Gravatar of ssumner ssumner
    8. March 2024 at 09:46

    Robert, Yes, but average inflation targeting should produce a similar pattern, if it were symmetrical.

    BC and William, Good comments.

  19. Gravatar of Robert Simmons Robert Simmons
    8. March 2024 at 10:36

    What’s the practical difference between average inflation targeting and level targeting? Don’t they end up in exactly the same place?

  20. Gravatar of Michael Sandifer Michael Sandifer
    8. March 2024 at 23:02

    This is just another example of your advantage over even Nobel winners. You’re less likely to see what you expect to see when you look at data, and you’re more fundamentally sound at arithmetic. Hence, you’re less clumsy when looking at simple data.

  21. Gravatar of TravisV TravisV
    9. March 2024 at 07:26

    Prof. Sumner,

    This is a golden opportunity to disagree with Larry Summers:

    Larry warns that we are on the “foothills of bubbles.”

  22. Gravatar of spencer spencer
    9. March 2024 at 12:32

    >5% N-gDp. Powell should have and could have stopped the madness long ago. Managing the economy through interest rates is ill considered. Money and liquid assets are worlds apart.

    Pritchard: “No asset has the “monetary store of purchasing power” quality unless there can be a net conversion of that asset into money. It must be possible to affect this conversion without necessitating that any present money holder reduce/liquidate his holdings”

    You manage the economy through base money, as Sumner says. You leave interest rates to the free market. A contraction in base money has no lag effect.

    An example. Some people think Feb 27, 2007, started “across the ocean”.

    “On Feb. 28, Bernanke told the House Budget Committee he could see no single factor that caused the market’s pullback a day earlier”.

    In fact, it was home grown. It was the seventh biggest one-day point drop ever for the Dow. On a percentage basis, the Dow lost about 3.3 percent – its biggest one-day percentage loss since March 2003. It was the result of a drop in base money.

    ME -flow5 (2/26/07; 14:34:35MT – msg#: 152672)
    Suckers Rally. If gold doesn’t fall, then there’s a new paradigm.

  23. Gravatar of dtoh dtoh
    11. March 2024 at 19:34

    How do you know the target is not 2.44%, in which case

    2% 2% 4% 4% 4% 2% 2% 2% 2%

    would be transitory.

  24. Gravatar of dtoh dtoh
    11. March 2024 at 19:47

    It still amazes me how most economists still don’t understand how monetary economics work.

  25. Gravatar of Doug M Doug M
    12. March 2024 at 03:38

    “In 2021, I naively believed the Fed was serious about FAIT.”

    I warned you. Don’t judge the Fed based on what they say, judge them on what they do.

  26. Gravatar of ssumner ssumner
    13. March 2024 at 07:56

    dtoh, Because Jay says so.

  27. Gravatar of dtoh dtoh
    13. March 2024 at 20:40

    Maybe Jay’s not telling the truth. Maybe he’s just trying to manipulate expectations…. or keep politicians off his back.

  28. Gravatar of spencer spencer
    14. March 2024 at 05:05

    Economists operate under the delusion that banks lend deposits. No, deposits are the result of lending. It’s the difference between micro-economics and macro-economics. Then it becomes stock vs. flow.

    C-19 reversed the bull market in bonds. The volume and percentage of gated deposits has fallen, so velocity is higher. I.e., GDP is higher as a result.

    Dr. Lester V. Chandler’s theoretical explanation in 1961 was: (1) that monetary policy has as an objective a certain level of spending for gDp, and that a growth in time (savings) deposits involves a decrease in the demand for money balances, and that this shift will be reflected in an offsetting increase in the velocity of demand deposits, DDs.

    1961: Dr. Corwin D. Edwards: “It seems to be quite obvious that over time the “demand for money” cannot continue to shift to the left as people buildup their savings deposits; if it did, the time would come when there would be no demand for money at all”

  29. Gravatar of spencer spencer
    14. March 2024 at 05:07

    This is confirmed by Dr. Philip George: “The velocity of money is a function of interest rates”

  30. Gravatar of Jeff Jeff
    14. March 2024 at 05:32

    > How do you know the target is not 2.44%

    There was an interview with Evans a year or two ago where he said that he thought long term 2.5% PCE was consistent with the 2% target. Maybe because it “rounds to 2”?? Once again, a case of seriously murky verbiage and obscurantism.


    Another question I had recently was this: I learned on this blog two years ago that monetary policy operates largely through housing. I understood this to mean that the interest rate either incentivizes or deters people from using their savings to buy or build housing.

    Doesn’t this suggest that the relevant inflation rate for determining whether policy is stimulating or restrictive is the housing CPI, not the overall CPI? If housing is inflating at 6%, then a 5.5% policy rate doesn’t restrict at all, rather it stimulates people to plow their cash into housing. So why do FOMC members keep talking about how restrictive the policy rate is??

  31. Gravatar of spencer spencer
    14. March 2024 at 06:13

    Powell should be fired. PPI up 6% mo-to-mo

    Contrary to the FED’s technical staff, retail MMMFs are nonbanks.

    In my 1958 Money and Banking text. “Purchases and sales between the Reserve banks and non-bank investors directly affect both bank reserves (outside money) and the money stock (inside money).”

Leave a Reply