Embarrassed to update your beliefs?

A rational person should not cling to previous beliefs for emotional reasons. They should welcome losing an argument, as that would mean they gained new information. On the other hand, there might be good reason to be embarrassed by the factors that cause you to update your beliefs.

I’m often bemused when I see a news story about someone who begins to question their faith in God after a tragedy strikes someone near and dear to them. Yes, I know this makes me a cold, heartless, bad person, but I can’t help myself. I think to myself “After everything that happened in the 20th century, it’s a loved one getting cancer that makes you think God doesn’t exist?” That’s the new information that leads you to update your priors?

[And don’t lecture me on human nature; I certainly understand that emotion plays a big role in how people view the world. Still, I can’t help noticing the craziness in the way we form beliefs.]

After 2008, the economics profession mostly abandoned its view that the BOJ was incompetent in allowing deflation, and adopted a new view—the idea that monetary policy is mostly ineffective at the zero bound. When I speak with economists, I almost never get a good explanation for this change. They mention things like low interest rate policies and/or QE coinciding with low inflation, as if we didn’t already know that from Herbert Hoover’s policies, or the BOJ policy of the early 2000s. Where is the new information that would lead you to suddenly doubt the efficacy of monetary policy at the zero bound?

Or how about January 6th finally convincing someone that Trump was a lawless authoritarian demagogue? It took that?

People often ask me if some sort of odd market anomaly has led me to re-evaluate my views on the EMH, as if history isn’t full of odd asset price movements accompanied by wild speculative activity. If someone told me that GameStop led them to question their belief in the EMH, I’d wonder if their belief was ever well founded. Had they ever studied financial history?

I’m not embarrassed to re-evaluate my views on various issues; I do so all the time. For instance, I don’t think our long run fiscal situation is as bad as I had thought in the early 2000s. I’ve revised my views on the likely path of real interest rates during the 21st century. On the other hand, I would be embarrassed to re-evaluate my views on an issue on the basis of information that any intelligent person should have known all along.



53 Responses to “Embarrassed to update your beliefs?”

  1. Gravatar of Garrett Garrett
    28. January 2021 at 17:27

    And a rational person shouldn’t be embarrassed to be embarrassed. If a belief was formed on faulty premises, the embarrassment felt should lead one to reconsider the premises of other related held beliefs. Instead, some people turn away from the feeling and dig their heels in the sand.

  2. Gravatar of msgkings msgkings
    28. January 2021 at 18:35


    In your ongoing battle against the word ‘bubble’, don’t you think you should take a step back and consider that the argument is almost a Wittgensteinian (?) battle over a word’s meaning?

    Call it what you will, but the current trading in Gamestop and the other Robinhood names (AMC, Nokia, etc) is not efficient. It’s not market participants aggregating all information about a company. It’s pure gambling in the form of stock trading.

    I am sympathetic to the idea that the entire market might be efficient, and that ‘bubble’ is mostly meaningless as a concept when discussing overall market valuations. But what is going on today in Gamestop is definitely not efficient market behavior.

    So call it a bubble, call it a game, whatever you like…but ‘bubble’ isn’t a bad way to describe it.

  3. Gravatar of ssumner ssumner
    28. January 2021 at 18:45

    msgkings, You said:

    “In your ongoing battle against the word ‘bubble’, don’t you think you should take a step back and consider that the argument is almost a Wittgensteinian (?) battle over a word’s meaning?”

    I prefer the philosophical maxim “That which has no practical importance, has no philosophical importance.” People are free to call something a “bubble” if they choose—I have no objection—as long as they acknowledge that it’s not a useful concept.

    Now if you could reliably assume that shorting a bubble would be profitable in the long run, then you’d have a useful concept.

    My complaint about anti-EMH theories is that proponents seem to believe they have a useful model, and they don’t.

    Regarding Wittgensteinian meaning, there are actually two uses of the term bubble. Some average people just mean a bit surge in the price of an asset. I’ve always agreed that that sort of bubble exists. It’s the Robert Shiller interpretation that I object to. I’m really glad that I didn’t follow his investment advice.

  4. Gravatar of ssumner ssumner
    28. January 2021 at 18:49

    Garrett. Yes. Indeed embarrassment itself is probably a useless emotion. Is there anyone who hasn’t looked back at their youth and thought “There was actually no reason for me to be embarrassed back in grade school over event X. I should have just laughed it off.”

    (Perhaps Trump.)

  5. Gravatar of TRowe TRowe
    28. January 2021 at 19:45

    Great reminder.

  6. Gravatar of Sean Sean
    28. January 2021 at 20:17

    Gme equals qanon. People literally betting their life savings that they can break ken griffin. With wild ass market theories not realizing that exchanges set limits all the
    time. Biden’s no better than trump.

    And you think I’m wrong opposing covid restrictions. They broke the country. Divided us more than ever. Over maybe a .3% pop death. Summer riots. Capital riots. Market riots.

  7. Gravatar of Carl Carl
    28. January 2021 at 20:19

    I wish I could develop your confidence that our fiscal prospects have gotten better. Pandemic aside we were running trillion dollar deficits in peacetime during a recovery while interest rates were low. And, nobody in either party is even talking about trying to curb entitlement spending.

  8. Gravatar of dtoh dtoh
    28. January 2021 at 21:02

    Here’s to one to think about. Suppose inflation is grossly overestimated because we are not accurately measuring product mix (i.e. quality/convenience improvements.) If we hypothesize that actual inflation is in fact minus 3%, how does that impact your views on the path of real interest rates. (Maybe you weren’t wrong.)

  9. Gravatar of dtoh dtoh
    28. January 2021 at 21:05

    If someone’s viewpoint is not Wittgensteinian, does that make them Witless?

  10. Gravatar of Lizard Man Lizard Man
    28. January 2021 at 22:12

    Lower birth rates and less immigration should be a sign that the US’ long run fiscal situation has deteriorated markedly, no?

  11. Gravatar of bb bb
    28. January 2021 at 22:24

    People are simply not wired to admit that they are wrong. Anyone who has managed people knows that most people will jeopardize their career to avoid admitting they were wrong. That’s why science advances one funeral/retirement at a time.
    A few of my own:
    In the 90s I was convinced that welfare reform would be a humanitarian disaster, and it’s success caused me to reassess most of my views. I still think it would have caused immense human suffering if it wasn’t combined with the EIC, but it was a complete paradigm shift for me.
    I used to tell my European friends that a Berlusconi or a Le Pen could never happen in the US. I’m still eating crow on that one.
    Sequestration in 2011 removed any doubt I previously had about monetary offset. Although, in that case I believed in monetary offset, but was not convinced that it was absolute.
    I’ve had numerous other ones in my personal and professional life that have benefited me greatly.
    I’m not willing to criticize folks who finally changed their view of Trump after 1/6. I credit them for being part of a very small minority of people that are self-reflective enough to change their views on Trump. Besides we need these people if we are to save the republic.
    Good post.

  12. Gravatar of Ray Lopez Ray Lopez
    28. January 2021 at 23:12

    Sumner appears to be having a “Come To Jesus” moment. He also claims to be never embarrassed, which is the sure sign of somebody untrustworthy (say psychologists). A trivial breach of the Capital (one involuntary manslaughter from what appears to be an unplanned blow to the head from a fire extinguisher, one intentional homicide when an unarmed white woman was shot by an armed black man for no good reason, and a few heart attacks and one suicide) is deemed by Sumner a big deal, a treasonous insurrection? In Athens, GR I’ve seen much worse, even firebombing, and life goes on (not to mention GR is much safer than the USA, just Google this). As for Sumner’s priors, what will it take to get Sumner to agree money is everywhere and always short-term neutral? The massive failure of monetarism in the last 25 years does not seem to have fazed him one bit.

  13. Gravatar of Cartesian Theatrics Cartesian Theatrics
    29. January 2021 at 01:14

    “In one specific instance, a patient suffering from asomatognosia tried to throw her own left arm into a garbage can because she believed it was her husband’s arm repeatedly falling on her and disrupting her sleep (Keenan, 2004). Even when patients are told that the body part belongs to them, many will deny the reality and remain firm in their belief that it is not a part of them. There is a case in which a patient with severe asomatognosia had the ability to use his right hand to trace his paralyzed left arm to his own left shoulder, but still failed to acknowledge that the left arm belonged to him (Keenan, 2004).” – Wikipedia

    Such is the nature of human belief.

  14. Gravatar of Dubois J.V. Dubois J.V.
    29. January 2021 at 03:28

    If anything my main update is that overall the society is much, much more willing to tolerate authoritarianism overall and that the actual checks and balances are actually only made of paper. I am afraid that the polarization now reached the unstoppable level and is self-fulfilling. On the left the realization to fight real and percieved supremacists everywhere, on the right even more of a feeling of being the really oppressed fighting against increasing authoritarian elites. There are now whole systems set in place to perpetuate the increased tribalism.

    The COVID accelerated the destruction of the veneer of the checks and balances. If threat is sufficiently large and enemy is sufficiently terrifying they can be bent and brushed aside with surprising ease.

    Now I wish I am wrong. As I said before, Democrats now have virtually all the power. Starting with official ones: The House, Senate and Presidency. The Academia, Mainstream Media, the army, three letter agencies and overall the professional state bureaucracy. At least looking at the campaign donations they also have support of companies ranging from Tech Giants to Wall Street. I do not know if there was ever more power in hands of single party like this before – definitely not together with feeling of actually being the oppressed ones, wholly morally and ideologically justified and willing to do whatever it takes to destroy the most vile and dangerous opposition.

    Overall I am with Glenn Greenwald on this one. The victory over Trump is a good thing. But it may have been the worse Pyrhic victory yet.

  15. Gravatar of Egor Duda Egor Duda
    29. January 2021 at 04:15

    Very insightful post!

    However, I’d like to push back somewhat. It’s not exactly related to current GameShop episode, but regarding belief updating in general.

    Could that be some kind of threshold thing? I think being good Bayesian is compatible with saying sometimes “After this or that event I’ve started to consider something”. When choosing a way to act one cannot evaluate all possible alternatives, due to limited mental budget , and have to chose only those than exceed some threshold in their probability.

    So, if after Bayesian update some posterior probability starts to exceed that threshold, one can legitimately say “Now, after this event, I’ve started to view this alternative as being worthy of consideration”

    This doesn’t even have to be conscious thing — Scott Alexander had an excellent post some time ago about Bayesian calculations being implemented in the brain as changing sensitivities for various neurotransmitters.

    So, to return to the emotion of embarrassment — I guess it’s ok to be not embarrassed about updating your beliefs on _new_ information (such as new mode of stock price manipulation), but one does have to be embarrassed about changing your beliefs on some information that you know before, but had compartmentalized it out (such as good people sometimes die young from cancer).

  16. Gravatar of Postkey Postkey
    29. January 2021 at 04:16

    “After 2008, the economics profession mostly abandoned its view that the BOJ was incompetent in allowing deflation, . . . ”

    Quite right. It was competent in allowing deflation.

    ‘ . . . in 1989, Yasushi Mieno said that,
    37:17 “Since the previous policy of monetary easing
    37:19 had caused the land price rise problems,
    37:22 real estate-related lending would now be restricted”.
    37:27 He looked around, looked at the bubble,
    37:29 asset prices rising, the gap between rich and poor getting bigger…
    37:33 Let’s stop it.
    37:35 His name was Mister Mieno, and he was a hero in the press,
    37:40 because he fought against this silly monetary policy.
    37:43 But he was deputy governor during the bubble era,
    37:47 and he was in charge of creating the bubble.
    38:16 All of a sudden land and asset prices stopped rising.
    38:22 In 1990 alone, the stock market dropped by 32%.
    38:31 Then in July 1991, window guidance was abolished.
    38:38 This took the window guidance officers
    38:40 at the Bank of Japan themselves by surprise.
    38:44 Bankers were left almost helpless.
    38:47 They complained that they did not know
    38:48 how to make their lending plans any more.
    38:52 In the past, when a certain branch said they would like to lend more,
    38:57 they would respond
    38:58 that the window guidance quota had been used up.
    39:03 Now they couldn’t do that any more.
    39:08 As banks began to realise
    39:10 that the majority of the ¥99 trillion in bubble loans
    39:14 were likely to turn sour,
    39:16 they became so fearful,
    39:18 that they not only stopped lending to speculators,
    39:22 but also restricted loans to everyone else.’
    “so the Bank of
    10:01 Japan was creating this massive asset
    10:03 bubble and then it was tightening and
    10:06 creating this massive recession why the
    10:08 goal was as they had declared actually
    10:11 before all this happened and that’s you
    10:14 know I used that as sub tablet of the
    10:16 book the structural transformation of
    10:18 the economy which was essentially a US
    10:22 plan and the Bank of Japan was their
    10:24 agent in Japan and to implement this
    10:27 because they the Bank of Japan plan was
    10:29 almost verbatim the same as what the
    10:32 Americans have been demanding in the
    10:33 negotiation with Japan structural
    10:35 impairment initiative Japan has to
    10:36 change”

  17. Gravatar of Student Student
    29. January 2021 at 06:02

    This game stop stuff fascinating. I wouldn’t throw it the EMH over the long run but we seem to have hundreds of thousands or possibly millions of 18-25 year old (my son and his friends all started doing this within the last year via a social media and Reddit fad).

    They are all purposefully coordinating and going into stocks at the same time. I don’t know what the implications are but a shit load of unemployed bar tenders and college kids are pouring their unemployment into the market using robinhood and like applications.

    Got to love crazy 21 year olds. Wreckless and dunning Kruger like with a f you boomer attitude. Should be interesting to watch them all lose their shit basically gambling.

  18. Gravatar of Sean Sean
    29. January 2021 at 06:11


    Good take on GameStop. It is a waste that Biden has to figure out what to do with it right now.

  19. Gravatar of Student Student
    29. January 2021 at 06:15

    Tried to warn them that this volatile gambling… not investing. But reasoning with a 19 or 22 year old testosterone filled gen z’er is like reasoning with a Trumpista… they gonna believe what they want to believe, and learn the hard way. Get your popcorn and what them all learn the hard way. I said, so what u gonna do, but don’t say I didn’t warn you.

  20. Gravatar of Spencer B Hall Spencer B Hall
    29. January 2021 at 06:52

    Ray Lopez re: ” what will it take to get Sumner to agree money is everywhere and always short-term neutral?”

    The pandemic just proved Scott Sumner was unquestionably right. The economy was already decelerating in the 1st qtr. before the crisis. But short-term money flows, proxy for real output for > 100 years, brought the economy out of an abyss.

    The problem with expanding / targeting N-gDp comes from whether AD is driven by money products (Fed’s stimulus), or savings products (velocity). It therefore depends upon the fulcrum of long-term money flows (inflation).

    (FOMC schizophrenia: Do I stop because inflation is increasing? Or do I go because R-gDp is falling?) arises from a sustained increase in the proxy for inflation. And that dilemma typically arises from an outsized increase in the demand for money (liquidity preference) which is not, or cannot be, monetized.

    As Dr. Philip George says: “When interest rates go up, flows into savings and time deposits increase” ( the ratio of M1 to the sum of 12 months savings ).

    Driving existing money (savings) out of circulation into frozen deposits (un-used and un-spent) reduces velocity and therefore AD. All $15 trillion in bank-held savings are lost to both consumption and investment. An increase in bank CDs adds nothing to GDP.

  21. Gravatar of Spencer B Hall Spencer B Hall
    29. January 2021 at 07:06

    Asset inflation today in the U.S. is like Japan in 1989, where stocks and real estate were in bubbles.

    The Japanese save more and keep more of their savings impounded in their payment’s system. They remunerate IBDDs thereby destroying savings’ velocity. They also have unlimited transaction’s deposit insurance (further destroying money velocity).

    All of which induces a 1966 Savings and Loan Association’s like “credit crunch” – where the term originated).

  22. Gravatar of Philo Philo
    29. January 2021 at 07:50

    “I don’t think our long run fiscal situation is as bad as I had thought in the early 2000s.” I would like to read about your new view, along with your reasons for it, if you have expressed in writing. (The government has so much leeway in devising and implementing fiscal policy that it seems to me practically impossible to predict the U.S.’s “long run fiscal situation” with any confidence.)

  23. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 08:59


    Given that our brains are still wired as if we live in relatively small, hunter-gatherer groups, easy embarassment in childhood makes a lot of sense. For most of human existence, the people we’d grow up with were the people we’d know our whole lives, and childhood social events are acutely important, because they set precedences that establish permanent social pecking orders in such a world.

    Of course, all of this looks silly in modern city life, and or considering that so many get to choose their social groups today. Road rage also looks silly, but emotional responses are obviously reflexes designed such that prefrontal cortical control is inversely related to the deemed slight, in terms of negative expectancy violations.

  24. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 09:09

    This Wall Street Bets social activist movement is fascinating to me, and frightening. It’s fascinating in that it violates the EMH. That is, because one tenet of the EMH is that price changes should not be predictable enough to allow for easy profits, yet it seems certain that this social movement will not keep GameStop prices so elevated forever, or even for very long. How long will the activists choose to punish the Wall Street shorts before they are satisfied and want to make some actual money?

    Of course, unless I’m wrong, EMH is not designed with social activist motives in mind. It is still a good approximate model for profit-motivated investors.

    What concerns me about all of this though, and this is my main thought, is how wide, deep, and bipartisan the populist rage revealed here is. I’ve read the comments on social media about this for a few days now, and there is white hot hatred for Wall Street and the politicians and system that bailed out some of the largest financial institutions in 2008/09. Those bailouts have had even more of a toxic spillover effect than I realized.

    This is a huge red flashing light on the dashboard, as if we should have needed another. The Trump experience should have taught us all we need to know, but we have yet more learning to do, and I’m more concerned than ever that an extremist movement on the left could continue to grow to rival the fascists on the right.

    We need deep reforms, in many, many areas, and we don’t seem to have the political system to facilitate them anytime soon.

  25. Gravatar of Brian Donohue Brian Donohue
    29. January 2021 at 09:49

    Good paragraph on monetary policy.

    Thoughts on Jeffrey Epstein? Did the shabby middle finger in your face suicide story cause you to rethink anything?

  26. Gravatar of Mark Z Mark Z
    29. January 2021 at 09:55

    Save for topics of immanent practical importance, like whether the road you’re driving on leads up to a cliff, publicly admitting you’re wrong is very often a strategic blunder. There’s a reason politicians always demure when journalists ask what their biggest mistake was or what they were wrong about, and it’s not just their egos. Same with pundits, otherwise Andrew Sullivan and David French would be the most popular pundits in America (in contrast to a self-styled “always right” NYT blogger who is much more popular). Normal people have even less reason to admit to being wrong. The practical benefit of being publicly right about politics is 0, while there is probably a social cost to it. You probably lose more status among your erstwhile co-ideologues than you gain among your erstwhile opponents.

  27. Gravatar of msgkings msgkings
    29. January 2021 at 10:01


    Agree 100% on Schiller being a bad investment advisor

  28. Gravatar of Dale Doback Dale Doback
    29. January 2021 at 10:40

    I’ll channel my inner Sumner and say that calling the GameStop stock a bubble is fine, but it is just not a useful concept to describe what is going on. Also, I love commenters attributing this to unemployed 20 year olds and not on the short sellers who have proven so far to be the more reckless party in this trade.

  29. Gravatar of ssumner ssumner
    29. January 2021 at 10:44

    Sean, You said:

    “And you think I’m wrong opposing covid restrictions.”

    Did I say that?

    Carl, I should have been clearer. I’m not saying that we still don’t have fiscal problems, rather I now believe that our economy can support bigger deficits than before. But we are also running bigger deficits—too big in my view.

    dtoh, It would change, but then what really matters is the NGDP growth rate relative to the nominal interest rate, and that wouldn’t change even if inflation were mis-measured. The big surprise is zero nominal interest rates when NGDP is growing 3% or 4%.

    Lizard, Other things equal, yes.

    bb, Yes, I should not criticize people who changed their views on Trump. They should be praised. And I was wrong about Trump back in 2016; I didn’t think he’d win.

    Ray, You said:

    “He also claims to be never embarrassed,”

    LOL, you really cannot read. Even at a 4th grade level. I said that about Trump.

    Egor, I guess it depends on the type of belief. There are some theories for which one more example might rationally tip support in that direction.

    Philo, See my reply to Carl, I sort of misspoke.

    Brian, I didn’t follow the Epstein story very closely, so I didn’t have views to re-evaluate. I have no reason to question the press version, but of course it’s possible that it’s wrong.

    I’m 100 times more interested in the vaccine scandal than a sex scandal or a GameStop scandal.

    Mark, I don’t know. People like Tyler Cowen and Scott Alexander sometimes admit they were wrong about something, and it doesn’t seem to hurt their reputation.

    In terms of politics you are probably right. But to be honest, I only care about what people think of me when I respect those people. And on intellectual questions I respect very few people (as a proportion of the US population). Of course I can respect people on other grounds, even if I don’t respect their views on NGDP targeting, etc.

    Xu and his friends think I’m a communist pedophile. I don’t lose sleep over their opinions.

  30. Gravatar of J Mann J Mann
    29. January 2021 at 11:29

    I think people are shifting towards Trump as lawless demagogue because of some combination of:

    1) It’s now lower cost to believe that; and/or

    2) They’ve updated on the amount of damage he can do. Up to now, they were hoping that he would fizzle out, kicking and screaming, but with no real impact (except possibly for revealing a group of right wing populist voters that his would be successors thought they might be able to lead), but the Capitol riot caused them to update as to how dangerous his conduct could be.

    Agreed that both of those are somewhat embarrassing. (As to 2, I mean you should update with new information, but a giant update implies either some embarrassing priors or an embarrassing process.)

  31. Gravatar of Student Student
    29. January 2021 at 11:33

    I don’t see what’s wrong with a short position but I get your point, the big fish do this kind of thing all the time (or so we perceive) and yet they shut out the little guy when he does it.

    My main point was that this is a fad thing to do for 18-25 year olds. That’s new IMO (and they flush with unemployment for someone that age). Also the commissionless trades, right?

    I don’t know what I am talking about in this area but, I am telling you this seems new and unique, go read the Reddit threads.

  32. Gravatar of Todd Ramsey Todd Ramsey
    29. January 2021 at 12:01

    “When Yellen was chair, Powell was no better on monetary policy, maybe a tiny bit worse. But Powell has benefited by observing 4 years of market monetarist critiques of Yellen’s policies.”

    If true, why didn’t Yellen learn from market monetarist critiques of Bernanke’s policies? We don’t know the counterfactual of whether Yellen would have responded boldly in March 2020; but we know Powell did.

    So why downplay Powell’s accomplishments?

    Maybe because you wrote, “Trump’s instincts were apparently to reappoint Yellen, but his aides talked him out of it. This is one of those rare cases where he should have gone with his instincts.”

    Or because you once wrote a story headlined, “Trump wants a non-economist to lead the Fed. That could be dangerous”. (I assume you didn’t write the headline but that it captures the tone of the article).

    And now you don’t want to admit you were wrong about the Powell appointment.

    “A rational person should not cling to previous beliefs for emotional reasons.”

    I’m having a hard time understanding why you don’t come out and say Powell has done a good job and was a good hire. Perhaps this is unfair, but I respect you highly, so I hold you to a high standard.

  33. Gravatar of Christian List Christian List
    29. January 2021 at 13:34


    I hope I am always ready to change my mind. I certainly do this several times a week when new information on a topic arises, or more accurately, when I am ready to process the “new” information. Most of the time, after all, the information isn’t new per se, it’s just “new” to oneself.

    One of my favorite examples in your really nice blog post is your example about God. It made me laugh out loud, because it is so true to life, so real. I have experienced situations like this in real life many times.

    Regarding your blog entries, I have certainly changed my mind about EMH, bubbles, and monetary policy per se, just to name a few examples, by far not all. Your expositions on these topics are very convincing.

    I think the point about Trump for quite some of his supporters was not so much whether he was an authoritarian demagogue or not – that was actually a selling point for some.

    The issue was whether or not he would be reelected, so in other words, whether or not he would continue to hold presidential power. A lame duck and a loser are evaluated differently because they have lost their purpose and their nimbus of invincibility.

  34. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 13:54

    Dale Doback,

    I suspect that GameStop is a rare case in which the word “bubble” applies. Presumably, this social activist desire to punish specific short sellers will be exhausted.

    Should anyone think a long shrinking chain of retail stores that sells physical game copies in the streaming age, and in which a large portion of its stores are in malls, will make a comeback at this point? Should we believe these social activists think this company and others they’re pumping up suddenly have business prospects that are an order of magnitude or more greater in some cases, suddenly?

    No. I read what these activists are writing.

  35. Gravatar of dtoh dtoh
    29. January 2021 at 14:27

    No. Think it through. If we are overestimating inflation then we are also overestimating NGDP. If inflation is actually minus 3%, then NGDP is not growing at 3% or 4%, it’s actually shrinking or flat, which would mean that nominal interest rates are equal to or above the NGDP growth rate.

    So in fact maybe you were not wrong.

  36. Gravatar of Carl Carl
    29. January 2021 at 15:04

    I think this post is related to your recent Will Wilkinson post. They both, in my estimation, boil down to humility. For some reason, we’re often rewarding emotionally brittle behavior and punishing humor and humility. It’s puzzling when you consider how much we individually value the latter and despise the former when we encounter them in our daily lives.

  37. Gravatar of ssumner ssumner
    29. January 2021 at 16:13

    Todd, Yes, I think Powell has done a pretty good job. But I don’t regret favoring Yellen, why should I? Heck, I’d hav epreferred Bernanke, and the Fe ddid poorly when he wa sin charge. I’ve always been consistent in arguing that people vastly overrate the importance of who is Fed chair, who is US president, etc. You can find me making this argument many times in my blog. So I’ve been very consistent.

    You ask why didn’t Yellen learn from Bernanke? OK, but she seems to have learned more than did Powell, who was more hawkish than Yellen during the period when she was Fed chair. We’ll never know if the recent Fed performance is because Powell is better of because the Fed as an institution is gradually improving.

    I don’t recall ever being particularly critical of Powell, and I’ve never in my entire life claimed that someone had to be an certified economist to do any job. If you read the article that way then you misread the article. I probably just meant that the Fed chair needs a good understanding of macro theory.

    dtoh, No, that’s not how NGDP is calculated. The government calculates NGDP from the total sales of final goods and services, adjusted for inventories and net trade. Then it is partitioned into inflation and real growth. If the inflation number gets smaller because of improved techniques in estimation, then it automatically makes the real GDP number bigger, with NGDP unaffected.

    Carl, Good point.

  38. Gravatar of anon anon
    29. January 2021 at 18:27

    Apropos of Galileo here goes https://www.conservativehome.com/thecolumnists/2012/04/andrew-lilico-does-the-establishment-naturally-stifle-new-ideas.html

    Again, seems nothing is what is being bandied about widely in the March of science. Though corroborating this with other such stuides is crucial.

  39. Gravatar of Todd Kreider Todd Kreider
    29. January 2021 at 18:30

    France, Germany and Austria are now ordering its citizens to stop wearing cloth masks since not effective and to instead wear surgical masks that are also not effective. But in Scott-World, mask signaling is what matters. Will he change his beliefs about cloth masks that almost everyone has been wearing for 10 months?

    Of course not.

  40. Gravatar of Ray Lopez Ray Lopez
    29. January 2021 at 18:48

    @ Spencer B Hall – “The pandemic just proved Scott Sumner was unquestionably right. The economy was already decelerating in the 1st qtr. before the crisis. But short-term money flows, proxy for real output for > 100 years, brought the economy out of an abyss. ” – wait: do you realize that Sumner refuses to say that the March 2020 Fed “QE infinity” announcement had anything to do with the stock market recovery or the real economy stabilizing? (the V-shaped and/or K-shaped recovery)? I’ve even quizzed him by email on this (the Great Man does reply to his emails on occasion) and Sumner refuses to say the Fed was behind the recovery after their March ’20 pronouncement, thought it sure seems that way. If the High Priest of the Quantity Theory of Money refuses to give the Fed credit, we have to agree with him, no? Actually the only time in all the years reading this crummy forum (what am I doing here anyway, I’m in the 1% dammit) the only victory lap I’ve seen Sumner run was when he correctly called the de/re-valuation of the Swiss Franc several years ago. Consistent with his theory, the Swiss economy did contract for one quarter, but then it was animal spirits (i.e., Shiller-style non-monetarism) and business as usual. The Indian devaluation completely debunked Sumner’s monetary theories, though Sumner tried lamely to run a victory lap despite the evidence to the contrary.

  41. Gravatar of dtoh dtoh
    30. January 2021 at 03:05


    1. In the post you said you were wrong “on the likely path of real interest rates.

    2. In your comment you said, “what really matters is the NGDP growth rate relative to the nominal interest rate.” Can you explain why this matters. (I ask this from the point of view of trying to understand what motivates economic behavior?)

    Also, if you were wrong, why do you think you were wrong?

  42. Gravatar of Spencer B Hall Spencer B Hall
    30. January 2021 at 05:46

    @Ray Lopez Re: “Teflon Man”

    Sumner generated a lot of discussion around an important monetary policy error. Bernanke destroyed America. The Covid-19 crisis fits his thesis.

    There have been 11 boom/busts in housing since WWII. They have largely been attributed to disintermediation (but not of the banks, of the non-banks). The GFC was not ultimately an exception.

    But Dr. Richard G. Anderson made a mistake saying legal reserves weren’t binding, e-bound vs. non e-bound. Bernanke drained legal reserves (the truistic monetary base), for 29 contiguous months. This turned housing prices upside down and underwater.

    2007 jan ,,,,,,, 43113 ,,,,,,, -0.11
    ,,,,, feb ,,,,,,, 41214 ,,,,,,, -0.09
    ,,,,, mar ,,,,,,, 39159 ,,,,,,, -0.11
    ,,,,, apr ,,,,,,, 41072 ,,,,,,, -0.09
    ,,,,, may ,,,,,,, 42699 ,,,,,,, -0.05
    ,,,,, jun ,,,,,,, 42034 ,,,,,,, -0.05
    ,,,,, jul ,,,,,,, 41164 ,,,,,,, -0.08
    ,,,,, aug ,,,,,,, 39906 ,,,,,,, -0.07
    ,,,,, sep ,,,,,,, 40460 ,,,,,,, -0.07
    ,,,,, oct ,,,,,,, 40161 ,,,,,,, -0.04
    ,,,,, nov ,,,,,,, 40331 ,,,,,,, -0.04
    ,,,,, dec ,,,,,,, 41048 ,,,,,,, -0.04
    2008 jan ,,,,,,, 42398 ,,,,,,, -0.07
    ,,,,, feb ,,,,,,, 41070 ,,,,,,, -0.05
    ,,,,, mar ,,,,,,, 39731 ,,,,,,, -0.04
    ,,,,, apr ,,,,,,, 41642 ,,,,,,, -0.03
    ,,,,, may ,,,,,,, 43062 ,,,,,,, -0.01
    ,,,,, jun ,,,,,,, 41616 ,,,,,,, -0.04
    ,,,,, jul ,,,,,,, 42083 ,,,,,,, -0.03
    ,,,,, aug ,,,,,,, 42055 ,,,,,,, 0.02
    ,,,,, sep ,,,,,,, 42456 ,,,,,,, 0.04
    ,,,,, oct ,,,,,,, 46930 ,,,,,,, 0.17
    ,,,,, nov ,,,,,,, 50363 ,,,,,,, 0.24
    ,,,,, dec ,,,,,,, 53723 ,,,,,,, 0.30

  43. Gravatar of Spencer B Hall Spencer B Hall
    30. January 2021 at 05:50

    re: “Then GDP is partitioned into inflation and real growth”

    You take nominal from 1986 to today and it increased 357%. The deflator increased 104% since then. But the CPI increased 248%.

    How do you reconcile the difference?

  44. Gravatar of ssumner ssumner
    30. January 2021 at 07:33

    dtoh, I was wrong about both “i minus inflation” and “i minus NGDP growth”. The term “real interest rate” usually refers to i minus inflation, but the concept i minus NGDP growth is an analogous and more useful concept. Inflation is an arbitrary concept created by bureaucrats, with no clear definition. Borrowers care about NGDP growth (not inflation), as total nominal income is the funds that borrowers use to repay loans. I’ll do a post on this soon, where Milton Friedman has the same insight.

    I was wrong about both versions of the real interest rate because investment demand slowed more than I expected and saving was stronger than I expected. This may reflect aging demographics, less capital intensive industries, NIMBY policies, etc. Whatever drove rates lower in Japan in the 1990s is now operating worldwide.

    Note that the actual quantity of S&I didn’t change much, but the schedules shifted, producing a lower equilibrium interest rate. “I” shifted left and S shifted right.

  45. Gravatar of dtoh dtoh
    30. January 2021 at 16:51

    I think “borrowers care about NGDP growth (not inflation)” is right and an important concept especially in regard to the monetary transmission mechanism.

    On the other hand, I think lenders care more about inflation or more specifically about the delta between nominal interest rates and inflation. That was the genesis for my original question. Savings is basically a choice between current consumption and future consumption and if you get more future consumption because of deflation then that will impact savings.

    You said, “Whatever drove rates lower in Japan in the 1990s is now operating worldwide.” So that’s a very good question. Human behavior might have changed, but I think the hypothesis that we are mis-measuring inflation is also worth considering.

  46. Gravatar of TAFKAA TAFKAA
    31. January 2021 at 00:41

    Todd, again with the lies. As someone pointed out a few posts ago you seem to derive some kind of sick pleasure from debasing yourself and showing people how ignorant, wrong, and dishonest you are.

    Austria is requiring FFP2 (the European equivalent of N95) masks in stores and on public transit because of the new more contagious strains. They had schools open in fall with no masking whatsoever (foolish, but at least they learned their lesson, if the hard way). When the facts change…

  47. Gravatar of Postkey Postkey
    31. January 2021 at 01:54

    “High growth of real broad money is associated with rising asset prices and healthy balance sheets, and tends to precede above-growth trend in demand and output. In the five months to September US M3 (as measured by Shadow Government Statistics) increased at an annualised rate of 9.2%, in the context of sub-2% inflation. Real money growth was therefore unusually high, the highest in fact since 2007. In the Eurozone also recent money growth has been robust. In the six months to August the annualised rate of growth of M3 was 7.3%, with inflation at about 1%. Since the USA and the Eurozone together account for 40% of world output (in current prices and at current exchange rates), these developments argue against a recession in 2020. They suggest that, on the contrary, the world economy should start 2020 in good form. Admittedly, money growth may slow from here. (Money growth has also picked up in the last couple of months in Japan and the UK, but from low levels, and they of course matter less than the USA and the Eurozone.)” . . . If money growth remains close to recent highs in the USA and the Eurozone, a more positive forecast would be justified. “

  48. Gravatar of ssumner ssumner
    31. January 2021 at 09:13

    dtoh, I think lenders also care about NGDP growth, not inflation. I certainly do, and I’m a lender.

    Again, the real mystery is the difference between i and NGDP growth, and inflation mis-measurement doesn’t solve that mystery.

  49. Gravatar of dtoh dtoh
    31. January 2021 at 14:25

    As a lender I too care about NGDP growth, but I also have an extremely strong preference for a 4% i that is 3% real and 1% inflation to one that is 1% real and 3% inflation.

    Since i is determined by the intersection of the supply and demand curves, a shift in or movement along either curve is sufficient to cause a change in i.

    So unless I’m missing something….for a given a rate of NGDP growth, if there is a change in the rate of inflation, this can shift the supply curve of credit, which can cause a change in i, which results in an increase in the delta between i and the the NGDP growth rate.

    So if inflation has fallen and we have not correctly measured the fall, then we have a possible explanation (which is a step forward from a “mystery”) for the increase in the i to NGDP spread, and in fact you may not have been wrong.

  50. Gravatar of ssumner ssumner
    31. January 2021 at 14:48

    dtoh, Again, I don’t see why it would shift the supply curve for credit, even if one does prefer lower inflation.

    And unless someone can tell me what “inflation” is supposed to be measuring, it’s hard for me to take the concept seriously.

    What’s the inflation rate for PCs? Not a number, but conceptually?

  51. Gravatar of dtoh dtoh
    31. January 2021 at 21:08


    So I’m not sure inflation is so important to the supply curve for credit, but for orthodox economics, it’s axiomatic that increasing the real rate of interest will increase the rate of savings (supply of credit.) So for a given nominal rate of interest, a change in the inflation rate equals a change in the real rate of return, which will increase (or decrease) savings (the supply of credit.) (And BTW it doesn’t make any difference if we frame this as a shift of the curve or a movement along the curve.)

    It follows therefore that if we are mis-measuring the real and nominal components of NGDP, then we are also mis-measuring the real and nominal components of the interest rate, and our expectations regarding the supply of credit will be wrong. And this alone could explain the increased spread between i and the NGDP growth.

    I think we can debate whether or not we are underestimating inflation, but if we ARE underestimating it, then I don’t think there is any logical conclusion other than that this is at least in part what is driving the increase in the delta between i and the rate of NGDP growth.

    I mostly agree with your other comments on inflation, but first I want to show that in fact you may not have been wrong in your views on the path of interest rates.

  52. Gravatar of ssumner ssumner
    1. February 2021 at 09:15

    dtoh, You are reasoning from a price change. Changing interest rates don’t predict any particular change in saving.

  53. Gravatar of dtoh dtoh
    2. February 2021 at 18:14

    No you’re the one who’s reasoning from a price change, well actually the contrapositive of a price change. And BTW, ironic title for this post.

Leave a Reply