Chesterton’s Fence and stock prices

John Cochrane has a new post with the following headline:

Gamestop. 1999 déjà vu all over again?

In context, Cochrane seems to be referring to the tech stock “bubble” of 1999, a year when the NASDAQ rose from about 2200 to just over 4000. If you bought a NASDAQ index fund during 1999 at say 3250 and held it until today, it would be valued at over 13,000, even after this week’s drop (and not even including dividend reinvestment.)

So what does 1999 all over again mean? Does it mean that GameStop shares purchased for $325 will be worth $1300 in 22 years, plus dividends? A buy signal? I don’t think that’s what Cochrane means; it seems like he’s rather skeptical of the future prospects of this controversial stock. (Although he wisely refrains from specific investment advice, something I also plan to do.)

I mention this example to remind readers that while 1999 is often viewed as a classic stock price bubble, there’s actually very little evidence that tech stocks were greatly overvalued back in 1999, at least in aggregate.

Bitcoin was about $2 when I first became aware of the topic. All my Bitcoin posts have basically had the same message—it’s probably not a bubble because there is no such thing as bubbles. As Bitcoin became more successful, I also pointed out that in an efficient market, 99% of Bitcoin-type investments should eventually collapse to almost nothing.

That might sound odd, but consider the alternative. Suppose that even 10% of investments such as early Bitcoin, early Amazon, early Tesla, etc., turned out to be wildly successful, rising 1000-fold. The other 90% fell to zero. A diversified portfolio full of that sort of speculative stock would yield a return that is far above the market average, roughly a 100-fold gain.

Investors know that a few of these high risk/high reward stocks will become highly successful. As a result, they all trade at high prices (relative to earnings). Eventually, most of these companies end up doing poorly, which operates via “confirmation bias” to convince most people that they were right about bubbles, even though they were wrong. They remember that they always knew would fail and forget that they refrained from buying Amazon for $7 back in 2001. How could a money-losing seller of books on the internet ever become a huge success?

In the early 2010s, I had no idea what was going on with Bitcoin, and still don’t really understand the investment. But I think I at least understand that I don’t understand it. You say nothing “fundamental” has changed with GameStop in the past week? OK, maybe, but are you sure? Does becoming 10 times more famous and developing a strong emotional connection to many millennials have zero value to a retailer? I don’t even play computer games, so I have absolutely no opinion on this stock. But how confident are you in your opinion?

This is why it’s so hard to test the EMH. The collapse of what looks like speculative bubbles seems like evidence against the EMH, but in fact the theory predicts that the vast majority of speculative “bubbles” will collapse, in order that the expected rate of return on portfolios that include Bitcoin, Amazon and Tesla is consistent with the risk-adjusted rate of return on other portfolios. The statement “speculative stock X is very likely to be lower in a couple years” is not at all equivalent to “speculative stock X is a bad investment.”

This aspect of market behavior is really hard for people to see, and as a result the vast majority of people never see it. Selling people on the EMH is like trying to convince someone that there’s no such thing as objective truth, or free will, or personal identity. If their brains are not wired in such a way as to grasp the idea, there’s no way to make them see it. Vase/profile.

People aren’t even thinking about the EMH question in the right way—they think it can be answered by squinting very hard at some so-called “bubbles” and figuring out whether the market was irrational. The actual question is, “What anti-EMH models are useful to me?”

Don’t bother arguing with me in the comment section; I’m a pragmatist. Do one of these three things:

  1. Short a stock that you think is overvalued.
  2. Go long on a stock you think is undervalued.
  3. Agree with me that markets are efficient.

PS. In 2013, I quoted Tyler Cowen saying the following:

With apologies to Scott Sumner, I say Bitcoin is a bubble.  Outside of war and rebellion, do “normal” new currencies behave this way?

Read my whole 2013 post, and tell me how my claims on everything from Bitcoin to stock prices look today.

PPS. Here’s how to tell if you understood the argument I made in this post. Is the Bitcoin story of the past decade:

A. A single data point against bubble theories?

B. A thousand data points against bubble theories?

The answer is B. It’s an absolutely crushing blow to asset price bubble theories. Devastating. It allows for 999 other asset price collapses in a world totally free of bubbles. NOW how are you going to prove that bubbles exist?




77 Responses to “Chesterton’s Fence and stock prices”

  1. Gravatar of Market Fiscalist Market Fiscalist
    29. January 2021 at 17:47

    If everyone believed that markets were 100% efficient then no one would bother trying to find undervalued or overvalued stock to short or go-long on and (like me) would only invest in index funds. This would make the market inefficient and make it worthwhile to find undervalued or overvalued stock.

  2. Gravatar of Ray Lopez Ray Lopez
    29. January 2021 at 18:35

    In other news, Ssumner thinks MySpace is not a bubble either since he still has an inactive account on it (as do I, unless they’ve purged all accounts nobody has logged into once in the last ten years).

    No, I won’t be taking any investment advice from Ssumner… I rather listen to Nobelist Robert Shiller, who wisely counseled not to buy real estate in cities not landlocked or artificially constrained (Shiller mentioned in his book not to buy in Milwaukee, surrounded by open prairie, which sounds reasonable).

  3. Gravatar of Garrett Garrett
    29. January 2021 at 19:13

    Matt Levine wrote a couple years ago about a court case where some hackers were able to view press releases before they became public. They used these to trade and were able to outperform. At the same time most daytraders fail and most mutual funds can’t outperform. I’d say that’s evidence that the level of efficiency in the markets is somewhere between semi-strong and strong.

    Of course there’s Renaissance but I suspect what they’re doing is really closer to highly leveraged market making. That would explain why they keep the size of the fund under $10bn and seem to do better when volatility is high.

  4. Gravatar of policy_wank policy_wank
    29. January 2021 at 19:57

    I mostly agree with you about the EMH. But c’mon, isn’t Gamestop different? Many of the retail investors are claiming to not be acting rationally, i.e. willing to hold onto the stock they bought indefinitely in order to stick it the hedgers rather than cashing out while they still could turn a profit.

  5. Gravatar of Gene Frenkle Gene Frenkle
    29. January 2021 at 20:04

    I think the 2001-2008 Housing Bubble and Bitcoin are closer to Ponzi schemes than “bubbles”. So I understand why properties in Malibu and Palm Beach and the Hamptons have inherent value that will draw people with money to continue to pay more to live in those areas barring some natural disaster…but why did an average house in suburban Atlanta, Las Vegas, Orlando, etc in 2005 keep getting flipped for more money?? At some point people were going to get left holding the proverbial bag. I don’t see what makes Bitcoin valuable whereas I see what makes property in Malibu valuable.

  6. Gravatar of Gene Frenkle Gene Frenkle
    29. January 2021 at 20:11

    Maybe pyramid scheme is a better term than Ponzi scheme because I don’t believe there is a single person orchestrating Bitcoin like with a classic Ponzi scheme?? So I remember in undergrad in the 1990s a pyramid scheme fever swept through my college and it was a crazy several days. The whole time I was trying to stop people from getting involved but one would be surprised at the number of intelligent individuals caught the fever.

  7. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 20:26

    I’m surprised to see GameStop used as an example here. It closed today more than 14 times the pre-social movement high, which started a bit over two weeks ago.

    Could the company cash in on this extra capital? Yes, though it hasn’t done so yet. Could millennial affection help permanently boost prospects for earnings? Yes, but 14 fold, for a retailer that still mostly depends on buying and selling increasingly obsolete products in many increasingly obsolete retail spaces, like malls?

    And what about all the other brands being propped up this week? AMC theaters… Anyone want to bet on movie theater chains for long-term growth? How about Nokia phones?

    One could credibly argue that this is out of the scope of the EMH theory, as this reflects something other than behavior motivated by pure financial gain. To my knowledge though, such a scope for the theory was never formally claimed.

    Sure, those of us claimimg this is an absurdly rate case of an actual bubble have a burden of proof. We need to profit on our trades in the short run.

    I completely agree that there was no tech bubble circa 2000, and that nearly all bubble claims are wrong. I agree that EMH is highly valid when behavior is purely profit-seeking. That’s not what we have here though, with the Wall Street Bets crowd.

  8. Gravatar of Jonathan Miller Jonathan Miller
    29. January 2021 at 20:27

    Maybe it makes senses to talk about both the local and global efficient market hypotheses. The global EMH, which is I think what you are referring to and what is referred to in general, is almost (or possibly, clearly) true by definition. The definition would be such “There exists some space-time interval during which the price of assets reflect all available information”.

    More interesting, and more applicable, is whether the local EMH is true. The local EMH would be defined as “For all possible spaces of space-time the price of assets reflected all available information”. I think, although we could argue this, that this is on occasion not true. Then we can discuss which spaces of space-time it might not be true (for example, Hertz stock around June 2020 for young online retail investors… we can discuss this). Once we establish (I assume we can) that there are regions of space-time for which the local EMH is not true, then we can talk about ways to predict whether local EMH will be true for some given region of space-time.

    If we establish instances where the local EMH fails, we can also choose to call these instances bubbles.

  9. Gravatar of Philo Philo
    29. January 2021 at 21:24

    I admire this post and, in general, I admire your marvelous insight into economics. John Cochrane is a very smart and impressive guy, but you are right here and he is wrong. I know it means little for a non-economist (me) to gush over your writing, but I wanted to get my A+ grade out there before picking one little nit. You are just wrong to say: “there’s no such thing as objective truth, or free will, or personal identity.” I know from experience that your brain is not wired to grasp this, so I will not belabor the point. The bad philosophy in no way detracts from the wonderful economics (indeed, there is no connection between them, though I think you believe otherwise). So please keep writing on economics (the movie reviews are good, too), and I promise to say no more about the philosophy.

  10. Gravatar of Dale Doback Dale Doback
    29. January 2021 at 21:54

    The GameStop price jives just fine with the EMH for me. The stock is still heavily shorted by hedge funds and widely discussed as a cultural phenomenon on social media. The shorts could certainly be squeezed further and so there is still plenty of upside. Even if the GameStop valuation falls back to earth, a hedge fund or two could be dragged into bankruptcy beforehand and GameStop shareholders would be their primary liabilities. There’s also apparently a strong case against brokers for market manipulation of which GameStop shareholders will be entitled to a class action payout. Obviously this could be wrong and it may end up looking later as an obvious bubble, but to me it is appropriately priced now as there is still serious upside.

  11. Gravatar of BC BC
    29. January 2021 at 22:27

    What seems most notable to me about the GME fiasco is how unscrutinizing people become whenever fed a little populist rhetoric. Suppose, we stripped away the rhetoric about sticking it to hedge funds, Wall St, etc. GME was a stock of a dying company — from spring 2016 to mid 2020, it returned about -80% (just eyeballing chart) — which some people started pumping in an internet chat room. After the online chatter, the stock quickly rose 10-20x in two weeks, trading volume increased 5-10x, and implied volatility rose from 125% to 500-600%. The original pumper of the stock reportedly has a portfolio now worth $30-50 million [].

    If it weren’t for the populist rhetoric, wouldn’t people be at least asking whether this was a classic pump-and-dump scheme? To be fair, I don’t know how much, if any, of his GME holdings “Roaring Kitty” has sold, but I’m pretty sure he bought ahead of everyone else. Also, a lot of people might normally classify a marketer for Mass Mutual Life Insurance Co. as part of “Wall St”. People certainly counted insurance company AIG as part of “Wall St” in 2008. I’m not even denying that Roaring Kitty was sincerely motivated by populism and the 30-50M in profit was just an unintended side effect. Heck, I’m not even sure that intentional pump-and-dump should be illegal: it seems pretty easy for the average person to avoid pump-and-dump schemes just by not believing commentary, rumors, “hot tips”, etc. on the internet. I’m just saying that, were it not for the populist rhetoric wrapper, most people would just view Roaring Kitty as just some guy trying to pump up the stock(s) that he happens to own. Again, I’m pretty sure he bought ahead of the crowd.

  12. Gravatar of BC BC
    29. January 2021 at 22:55

    Bubble vs. EMH seems like the wrong framing. Markets have finite liquidity so, when faced with unusually high trading volumes, prices can swing wildly (up or down) over very short periods. Those markets aren’t completely efficient — lack of liquidity is a type of friction. Yet, “bubble” is also the wrong term as the high prices don’t necessarily reflect some sort of irrational exuberance. Finally, there may not necessarily be easy “arbitrage” opportunities due to liquidity, credit/margin, regulatory, and institutional constraints.

    If the question is whether there is an easy arbitrage around GME, I think the answer is no. “No arbitrage” is not quite the same as EMH. One might be able to predict the future price path of GME without being able to arbitrage that. Back in the early 2000s, futures on the KOSPI Index (the main S. Korean stock index) traded persistently low relative to the prices of the underlying stocks. So, one could reliably predict that the futures price would rise relative to the index. However, one couldn’t arbitrage that easily because one couldn’t (easily) short individual Korean stocks at the time. (I don’t know about now.) Futures prices were low relative to the index precisely because people could short the market only through the futures. So, the market was not completely efficient — inability to short stocks was the inefficiency — prices were predictable contra EMH, but the no arbitrage condition held.

  13. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 23:10


    I agree that the philosophy on this blog is bad, but only in so far as all philosophy is bad. I’m one of those who agreed with many physicists like Hawking and Feynman that philosophy is useless. It’s mental masturbation. Everything is physics and phenomena that emerge from it.

    That’s seen as crude by many, but it’s how I see it.

  14. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 23:15


    You bring up a good point about regulatory risk for these inflated prices on stocks like GME.

    Also, good point about not necessarily being able to trade on predictable price behavior, but I don’t think it applies to this case. Also, I’ve always taken EMH to apply to complete markets in which regulations are not restricting efficiency.

  15. Gravatar of Michael Sandifer Michael Sandifer
    29. January 2021 at 23:19

    Dale Doback,

    It seems like you’re saying that current price behavior for GME and the dozen related stocks are consistent with supply and demand, not EMH. I don’t think EMH applies to price behavior that isn’t driven by the estimated present value of future profits.

  16. Gravatar of anon85 anon85
    30. January 2021 at 00:29

    The problem, as always, is people conflating two different forms of the efficient market hypothesis:

    1. (roughly speaking) You can’t predict if a stock will go up or down.

    2. (roughly speaking) The price of a stock equals time-discounted expected future dividends.

    (1) is essentially true, (2) is not always true. The time-discounted expected future dividends give a lower bound on the price of a stock. But there is nothing preventing people from “irrationally” valuing gold, or valuing fiat currencies, or bitcoin, or collectible baseball cards or something. People can value things that pay no dividends.

    Once you realize this, the mystery disappears: GME stock is valuable because people are willing to pay money for it. Why? Because they find it funny. Does this make it a bubble? Only if we believe that people will stop finding it funny later; if people find it funny forever (a bitcoin situation, perhaps), then GME might never go down. It is also possible for GME to go up.

    I cannot predict where GME will go any better than the market. I agree with Scott that markets are efficient in this sense.

    What I do know is that GameStop the company will not pay future dividends that will ever justify its current stock price. I think it’s ridiculous that Scott is suddenly questioning this, saying maybe GameStop has high earning potential after all. This is a misunderstanding that comes from conflating (1) and (2) above.

  17. Gravatar of Matty Wacksen Matty Wacksen
    30. January 2021 at 01:31

    While I don’t think you’re off with your general point, we should probably consider the WallstreetBets subreddit connection and the fact that there are transaction costs to shorting.

    If GameStop’s price rising sharply is due to the actions of WallstreetBets, and if you read the kind of logic they follow (it’s a public forum, after all), it’s hard to take the efficient market hypothesis seriously in this case. I’d happily short WSB if my broker would let me and it weren’t so expensive. After all, their logic is “more than 100% short relative to float means that every one of us gets to dictate the price to short-sellers” which misses the subtle point that returned shares can (and probably will) be sold on the market at these prices.

  18. Gravatar of Eliezer Yudkowsky Eliezer Yudkowsky
    30. January 2021 at 02:40

    This isn’t what the Gamestop thing is about! It’s a technical issue: the level of outstanding short interest is high enough that WallStreetBets and maybe a number of hedge funds are trying to engineer a short squeeze. I can’t think of any reasonable interpretation of the EMH that says that an engineered short squeeze is *supposed* to have a price that reflects fundamentals. For a longer explanation see:

  19. Gravatar of Tacticus Tacticus
    30. January 2021 at 03:29

    I’ve never liked the Efficient Markets Hypothesis because it seems untestable or unfalsifiable to me. Whenever a price changes we can simply say that the market is reflecting new information, which, okay, so what? I’m not sure what utility the theory adds.

    I also dislike the word ‘bubble’ to describe anything except a fraudulent situation, such as the South Sea Company or Wirecard.

  20. Gravatar of Ray Lopez Ray Lopez
    30. January 2021 at 05:43

    Shoutout to anon85 whose answer upstream is the best in this thread. But his critique of Sumner will be lost in the Memory Hole. Sumner only believes in an Echo Chamber that praises him. Remember, this is his “dark” blog.

  21. Gravatar of Spencer B Hall Spencer B Hall
    30. January 2021 at 06:19

    re: “there’s actually very little evidence that tech stocks were greatly overvalued back in 1999”

    Tripe. I, once again, called the top. It was an obvious overvaluation. Greenspan’s overreaction to Y2K programming made it definitely so.

  22. Gravatar of Spencer B Hall Spencer B Hall
    30. January 2021 at 06:28

    re: “So what does 1999 all over again mean?”

    The only market move I’ve missed was initially QE2 (when stocks decoupled from the economy and inflation fell).

    There are 6 seasonal, endogenous, economic inflection points each year. (they may vary a little from year to year):

    Pivot ↓ #1 3rd week in Jan.
    Pivot ↑ #2 mid Mar.
    Pivot ↓ #3 May 5,
    Pivot ↑ #4 mid Jun.
    Pivot ↓ #5 July 21,
    Pivot ↑ #6 2-3 week in Oct.

    But the downward move starting this week is not just a seasonal swing.

    As I said:
    Salmo Truttra9 January 2021 at 16:33
    We now know that the top in real economic activity is 12/2020. Therefore the approximate top in stocks is January (without a decoupling).

    Money flows:

    12/1/2020,,,,,0.87 top
    01/1/2021,,,,,0.55 stocks fall
    –Michel de Notredame

    Note disclaimer: my time series was eliminated by Chairman Powell. This is a new and untested one.

  23. Gravatar of ssumner ssumner
    30. January 2021 at 07:01

    Ray, You said:

    “No, I won’t be taking any investment advice from Ssumner”

    Probably because I don’t offer any.

    Garrett, Yes, I agree that insider trading works on occasion.

    Policy wonk, Different? Maybe. But don’t forget Chesterton’s fence. Are you shorting Gamestop?

    Gene, House prices are back up to their peak. Is this also a bubble?

    Michael, I’m not saying these stocks won’t crash (I suspect they will). But the same people saying they are obviously a bubble said the same thing about Bitcoin a decade ago.

    Jonathan, Do mutual funds that invest on the basis that bubbles exist outperform other mutual funds? If not, then bubbles are a useless concept.

    Philo, Thanks. You said:

    “I know from experience that your brain is not wired to grasp this,”

    Oh, but it is. I used to believe in those things, until in a flash of inspiration I saw the truth, er . . . what I perceive as the truth.

    Dale, But surely they couldn’t win a payout big enough to justify a $25 billion market cap?

    Michael, Philosophy can be useful if it’s pragmatism. And it can be aesthetically beautiful—i.e. literature.

    anon85, Only your point one is important.

    Notice that I compared it to Bitcoin, which doesn’t pay any dividends. I have no idea what its future dividend prospects are. The EMH typically assumes that people only care about future dividends, but there’s no reason they couldn’t get some other value from GameStop, as they do from Bitcoin.

    Matty, Sure, if short selling is difficult then that might make a stock price less efficient. But people WERE shorting this stock. Presumably now that it’s much further out of whack, they have even more incentive to do so. Right?

    Tacticus, I’ve described its testable implications in other posts. Stock prices follow a random walk, index funds outperform managed funds. Prices respond immediately to new info, etc. There are lots of testable implications.

    Eliezer, Thanks, I’ll take a look.

  24. Gravatar of ssumner ssumner
    30. January 2021 at 07:23

    Eliezer, Yes, that’s much better than my post. I think you are correct.

  25. Gravatar of Michael Sandifer Michael Sandifer
    30. January 2021 at 07:32

    Eliezer Yudkowsky,

    Yes, I think that’s the same me saying one can argue that this social movement to punish the shorts is outside of the scope of EMH. The only thing awkward about it is that I don’t recall such limits on the scope of EMH being specified by the proponents, though that is just a formality. Economics is weird that way and anyway, who would have anticipated this even last month, to say nothing of decades ago?

  26. Gravatar of Mark S Barbieri Mark S Barbieri
    30. January 2021 at 07:48

    I always thought of the EMH as meaning that the market was efficient enough that it was hard to beat it. I definitely think that applies to me, so I just hold the entire market and focus only on tax efficiency and managing risk through asset allocation. It’s dreadfully boring and made me rich fairly slowly.

    Are there bubbles? Don’t know. Don’t care. I just know that I cannot trust my ability to see or predict bubbles enough to profitably trade on that information.

  27. Gravatar of Michael Sandifer Michael Sandifer
    30. January 2021 at 08:17

    Eliezer Yudkowsky,

    The perspective on your blog post is interesting. I hadn’t considered the Wall Street Bets crowd as a possible cartel, as I guess I never considered such a level of cooperation between so many agents, many of whom are anonymous, as likely. It will be interesting to see what level of cooperation is achieved from this perspective. I’m skeptical a cartel like this can work, for many reasons, but on the other hand, potential criminal legal consequences for pump and dump behavior might help keep the cartel together.

    I wish I knew more about the law in this area, as my impression is that any coordinated attempt at manipulating stock prices is illegal, but we’ve perhaps never seen such coordination among so many, and occurring in a public space. There’s perhaps no hidden plan here. Also, I I wonder what constitutional issues this should bring up, ultimately. Why should individuals and firms necessarily be allowed to do squeeze shorts this way, but not cartels of small investors?

    Also interesting, from my ignorant legal perspective, is that the law against pumping and dumping may actually help this cartel succeed in extracting supernormal profits from the shorts. If that’s true, should the law be repealed? Normally, this is an issue with penny stock trading. Additionally interesting, some of the stocks being pumped up by Wall Street Bets are penny stocks. Penny stocks generally can’t be shorted by retail investors. Brokerage firms don’t allow it. But, large investors get special treatment. So, there are lots of potential issues here.

    I also look forward to seeing how the desire to punish these big shorts keeps the cartel together.

    Thanks for adding these much more interesting dimnsions to the discussion.

  28. Gravatar of Michael Sandifer Michael Sandifer
    30. January 2021 at 08:28

    I guess I should be more specific about potential constitutional questions by saying that, to the degree a court was convinced this is a social/political movement, in addition to being profit-seeking, can such behavior be considered speech? If so, how does that differ from considering political donations speech?

  29. Gravatar of Dale Doback Dale Doback
    30. January 2021 at 09:21

    The EMH says that stock prices reflect all available information and therefore it is impossible to know if a (single) stock will go up or down. EMH does not say that the stock price is an accurate reflection of future company profits. The current GameStop valuation is justified because there is a possibility of essentially a large dividend to be paid out by short sellers. The valuation of GameStop at X billion dollars just reflects the huge valuations of the short seller hedge funds on the hook for GameStop stock.

  30. Gravatar of Dale Doback Dale Doback
    30. January 2021 at 09:45

    The coordination of the GameStop squeezers works because there’s no limit to how much they can extract from the short sellers. They all have FOMO if they break from the group and take profits and a new buyer is waiting on the sidelines to get in on the squeeze. Shorting in GameStop went up not down after the initial stock rise after some hedge funds took a loss and were replaced by new ones. The squeezers actions and holdings are all done on public stock exchanges while the hedge funds are secret. The public doesn’t know who the most exposed shorts are and when they can’t be squeezed further. As soon as a hedge fund declares bankruptcy or is revealed to be in distress, then the coordination starts to break down.

  31. Gravatar of Michael Rulle Michael Rulle
    30. January 2021 at 09:47

    My practical view of EMH is investors cannot outperform the market (leave aside the various definitions of market—does not matter) and there is overwhelming evidence this is true. Of course, pure randomness can result in outperformance but not in the long run—which is not that long. Having said that inside information can make that statement false—-that is why Wall Streets’ trading divisions overwhelmingly out perform the various markets. It is legal inside information because their clients tell them what they want to do. And they also buy the bid and sell the offer. I agree with Scott I think (meaning I assume I understand what he is saying).

    Selling naked shorts is pretty dumb—-considering that markets on average have positive returns. If one thinks they are clever, buy the best stock.

    Okay—-what about GME? What is that about? Of course I don’t really know anything except what I read. Assuming what we read on Reddit is relevant, there is a belief that if Shorts are larger than the float, eventually, the short sellers will have to cover. That is clearly not true —-that is like saying if the Government keeps having deficits it will have to eventually pay back the borrowed money—-instead we just keep rolling—-this is also true for shorts. Same thing

    Having said that, the Reddit crowd, regardless of their beliefs have decided to try to force them to buy back their shorts. What is amazing is that the Crowd has gotten so big in just a week or so they actually forced Melvin Capital to cover——they clearly were going under and still may—-but two of the biggest trading firms have chosen to put in 3 billion to support it. Why? Melvin is already out. They may have already lost all their capital. Why invest in a firm with idiot risk management? Don’t know. But we will find out.

    What does this have to do with EMH? I really do not know. If we choose to believe what we read, this spike in GME had only to with the belief that if shorts are bigger than float, it has to be covered. In other words, the Crowd does not believe GME has value—-it just believes it must be covered. It is a technical belief.

    Whatever it is, it is unprecedented. Squeezes are almost always conducted by one major entity against the “Crowd’. This is the opposite. Fascinating

  32. Gravatar of john john
    30. January 2021 at 10:01

    I don’t like the classical definition of the EMH as “prices incorporate all information” because it ignores too many real world frictions around price formation. I would use a practitioners EMH definition of “it’s impossible to make more money without taking more risk.” In that sense GME is pretty close to efficient.

    Currently GME is about 113% short. Earlier this week, it was 144% short. Now there is nothing impossible with a stock being more than 100% short, but you run into a multiple equilibrium situation (sunspot theory), where if the market believes a squeeze will occur then the price shoots to infinity. At that point anyone who is short will be forced to post more margin, and if you can’t you’re forced out of your position.

    Also the cost to borrow GME is 50% APR. Lets say we “know” GME is a bubble that will stop in 2 months, then you have to pay approximately $25/share at current valuations to short it. Not so bad if GME is going to crash $300, but in between, you’re exposed to absurd levels of mark to market risk, because the squeeze equilibrium price is infinity. This is why research has shown stocks with high short interest have low excess returns: they’re hard to borrow so lending fees are high, and they have a ton of idiosyncratic risk. You can make more money shorting them, but only by taking more risk.

    I personally am long some $300 strike puts on GME. Because I’m willing to pay the $25 in implicit borrow fees and I can’t be forced out of my puts. I bought them for about $160 when the GME price was $360. My personal view is that the price will return to its “pre-bubble” price of about $20, so I’ll pocket about $100 on my puts. The main risk is that GME is actually worth > $100 more even minus the squeeze due to fundamentals changing as Scott pointed out (but I don’t think fundamentals have changed). I believe the stock price will collapse when the market psychology is forced away from the squeeze equilibrium. The trigger will probably be exchanges/regulators shutting down trading in GME or GME having less than 100% short interest.

  33. Gravatar of Carl Carl
    30. January 2021 at 10:08

    Does the EMH make any assumptions about property ownership? In addition to pointing out that the GameStop issue points to issues of coordination, Eliezer, I think, highlights problems caused by ambiguous ownership of property. The short sellers are stuck in a situation where multiple people have claims to the same shares. In a situation where it’s not clear how much is owned by whom how can you have an efficient market?

  34. Gravatar of Gene Frenkle Gene Frenkle
    30. January 2021 at 12:53

    sumner, I believe the key number is the overall home ownership rate which peaked in 2004,05,06 at around 69%. So renting makes sense much of the time and in that period you had people that should have been renting trying to make a quick buck by taking out a $300k mortgage!?! Contrast that to 1999 when people were trying to make a quick buck by investing $10k in

  35. Gravatar of Ken P Ken P
    30. January 2021 at 14:27

    @John – Normally, a short squeeze lasts a day or maybe two. This could last longer. I’m guessing your 3rd wk Feb expiration? The $20 price was in the context of a heavily shorted stock. Are you confident there will still be strong short interest on this stock under $50, or will short sellers be wary?

    Personally, I plan to throw my log on the fire on Monday with a single share purchase that I expect to drop to around $50 at some point. I will also ensure that my share can'[t be loaned out. That’s what people don’t get. It’s not about the money. There’s utility value because you are contributing to the cost of hedge funds unwinding their positions. This is the primary motivation for many of the people buying and holding this stock right now and it brings more than $300 in value plus stock losses are tax deductible.

    I have seen hedge fund shorts do extreme damage to biotechs many times so I’m happy to be a part this.

  36. Gravatar of Ken P Ken P
    30. January 2021 at 15:01

    Scott, I usually read your EMH posts and agree with most of your position on the topic. Where I disagree with EMH is the idea that the market contains information but that individual success is strictly luck. Many of those “lucky” individuals IMO are exactly the ones who best understood the reality of the situation.

    As for bubbles, I think prices change when things change (which I think is close to your view). If interest rates go up for example, of course home prices go down. Similarly if mortgage requirements become more restrictive. But that doesn’t mean prices are wrong before those things change. If I think a stock price is going to change next week, I don’t consider THAT PRICE as the correct price for today.

    The price of GME might be perfectly priced under conditions where people are willing to hold at this price and be diligent about blocking their shares from being borrowed, while short sellers are being forced to buy due to margin calls.

  37. Gravatar of Michael Sandifer Michael Sandifer
    30. January 2021 at 17:19

    Ken P,

    Higher interest rates don’t necessarily make for lower home prices. If they reflect higher GDP growth, it could mean there’s actually more demand for housing.

  38. Gravatar of Andrew Andrew
    30. January 2021 at 17:48

    What a relief that Trump is gone. We can back to arguing about EMH and bubble nominalism

  39. Gravatar of anon85 anon85
    30. January 2021 at 17:56

    Eliezer, the problem with the “short squeeze” theory of the current GameStop price is that the short squeeze has no chance of working at this point — or at least no chance of working further. First of all, several hedge funds already closed their short positions, and the price *still* went up afterwards. Second, I don’t see any indication that the market expects the squeeze to work — do we see high interest rates on shorts for GameStop? I’m not following this closely, but I don’t think so.

    Finally, and most importantly, data from Citadel suggests that the retail investors of GameStop are primarily buying GameStop from *other retail investors*. See Matt Levine’s article on this:

    This means that redditors and other retail investors are clearly failing to accumulate GameStop stocks, on aggregate; they own the same total amount of GameStop stocks now as they did last week! The redditors who bought early are simply selling GME to the ones who are buying late.

    And stepping back a bit: forming a cartel out of that many individual redditors was OBVIOUSLY doomed from the start, and in an efficient market people would bet against it succeeding. What’s going on here is not that redditors can all agree not to sell (they are failing miserably at this!). What’s going on is that people are buying GME for the same reason they collect baseball cards.

  40. Gravatar of Postkey Postkey
    31. January 2021 at 02:13


    is an ‘investor’ that regularly ‘outperforms’ the market?

  41. Gravatar of Spencer B Hall Spencer B Hall
    31. January 2021 at 05:33

    The “end game” is clear. The transfer of title to goods, properties, or claims thereto represent leakages in aggregate demand. An unspent balance is one that does not add “value to product”. Financial investment draws off a net volume of funds that would otherwise have been spent on current output.

    Transfer payments are also a leakage from the main income stream because they involve expenditures which do not directly finance current output.

    Prima facie evidence of a gross leakage which collects in the form of unspent balances is the 15 trillion dollars in savings frozen in the payment’s system. From the standpoint of the economy, banks do not loan out existing deposits. They create deposits whenever they lend/invest. All bank-held savings are lost to both consumption and investment.

    Unspent balances, stoppages in the flow of funds derived from the main income stream have a direct and immediate dampening effect on the economy. An expansion of savings deposits in the commercial banks shrinks aggregate demand and therefore produces adverse effects on GDP. The expiration of the FDIC’s unlimited transactions deposit insurance is prima facie evidence. It caused the taper tantrum.

    Likewise, an excess of savings over real investment outlets exerting a contractive economic influence.

  42. Gravatar of Tacticus Tacticus
    31. January 2021 at 06:09

    @ Prof. Sumner, Yeah, I’ve read all the literature on the EMH, but philosophically I still find the theory unhelpful. I don’t think it is necessarily wrong, just unhelpful *shrug*

    The problem with shorting GME is that that takes money – I’ll check tomorrow what the cost is – and I have better things to do with my money. The opportunity cost is too high and all that. There are high transaction costs (as well as apparently societal ones!) which make the risk-reward profile far too skewed for me.

  43. Gravatar of Spencer B Hall Spencer B Hall
    31. January 2021 at 06:44

    GameStop was 140% short. That is illegal. So how did it happen?

  44. Gravatar of Todd Ramsey Todd Ramsey
    31. January 2021 at 07:10

    “1.Short a stock that you think is overvalued.
    2. Go long on a stock you think is undervalued.
    3. Agree with me that markets are efficient.”

    Another possibility: I could be right that GameStop is overvalued, but also believe that the market can be irrational for longer than I can stay solvent.

  45. Gravatar of Tacticus Tacticus
    31. January 2021 at 07:10

    Being 140% short is not illegal.

  46. Gravatar of xu xu
    31. January 2021 at 08:40

    “Short a stock that you think is overvalued.
    Go long on a stock you think is undervalued.
    Agree with me that markets are efficient.”

    1. Bad investment advice, especially since you will almost certainly be squeezed by bigger funds. Short squeezing is not a new concept.

    2. Sure, but be prepared to hold for a long time.

    3. Markets are inefficient & manipulated daily. If they were efficient, then you wouldn’t have booms and busts.

    If you allow 150,000 plebeians to maliciously target a firm out of political hatred, then the clearing house will go bankrupt and the dollar will collapse. Of course, you love the idea of a collapsed dollar so somehow I’m not surprised you’d advocate terrible advice. More plebes in poverty, the more the Sumner patricians can attempt to rig the system via the Federal Reserve.

    End the shorting! End the Fed!

  47. Gravatar of henry henry
    31. January 2021 at 09:07

    As an investor, and not a paper pushing academic, I’ve actually participated in manipulating markets. It’s very easy to do when you are a market maker.

    You really believe prices are at equilibrium?


    Thank you for the laugh.

    For those who want to learn real economics, and not the communist centrally planned economy Sumner dreams about, I recommend reading the book “the mystery of capital” by Hernando de Soto.

    Instead of sitting behind a desk writing garbage, he actually went out into the real world and observed market forces in action!

  48. Gravatar of ssumner ssumner
    31. January 2021 at 09:11

    Andrew, Trump is far from gone; he still controls the GOP. They are terrified of him.

    Todd, I’ve always found that to be a lame excuse. Suppose that a mutual fund was set up to short hundreds of overpriced stocks. Over a period of years it should win most of its bets.

  49. Gravatar of Kevin Dick Kevin Dick
    31. January 2021 at 11:58

    So I invest in startups. My firm specifically invests in the riskiest startups. 2 or 3 rounds before “normal” venture capitalists.

    Now, understanding math, we invest in 100s of these per fund.

    And guess what our returns look like? A few companies doing incredibly well, a modest chunk doing pretty well, and the vast majority doing poorly. In a few cases, our returns on a given position our 100X+. We’ve got a couple that may eventually hit 1000X+.

    If all you did was look at our investments individually, you’d think we were investing in bubbles–and losing most of the time. But you have to look at the larger portfolio and market.

  50. Gravatar of Michael Rulle Michael Rulle
    31. January 2021 at 12:24


    not illegal for market makers

  51. Gravatar of Tacticus Tacticus
    31. January 2021 at 12:27

    Not illegal full stop.

  52. Gravatar of Ken P Ken P
    31. January 2021 at 13:13

    Michael Sandifer,
    You said:
    Higher interest rates don’t necessarily make for lower home prices. If they reflect higher GDP growth, it could mean there’s actually more demand for housing.

    Agree. Didn’t mean to imply that. But all things equal (I know things never are), higher rates result in lower home prices. My point is that if conditions change, and prices change, that doesn’t mean prices were incorrect prior to the conditions changing.

  53. Gravatar of Christian List Christian List
    31. January 2021 at 15:32

    I don’t see how Gamestop is about EMH or how it disproves EMH. EMH is obviously correct, anyone who follows Scott’s blog long enough will recognize this.

    Over a period of years it should win most of its bets.


    I’m not an expert but one problem with shorting might be that losses can be unlimited. So you can win 99 bets and lose just one bet in a crass short squeeze and still make losses. One would have to create 100 different accounts, one for each stock, to prevent such a scenario.

  54. Gravatar of Postkey Postkey
    1. February 2021 at 01:31

    “You really believe prices are at equilibrium?


    Thank you for the laugh.”
    Quite right.
    “The focus on equilibrium and prices is due to the hypothetico-axiomatic method, a.k.a. the deductive methodology. The axioms are postulated that people are individualistic and focus on maximising their own satisfaction (named ‘utility’, in honour of Jeremy Bentham, the first economist to argue for the legalisation of the then banned practice of charging interest; Bentham, 1787). Next, a number of assumptions are made: perfect and symmetric information, complete markets, perfect competition, zero transaction costs, no time constraints, fully flexible and instantaneously adjusting prices. McCloskey (1983) has argued that economics has been using mathematical rhetoric to enhance the impression of operating scientifically. Equilibrium will not obtain, if only one of the axioms and assumptions fails to hold. But their accuracy is not tested. Yet, one can estimate the probability of obtaining equilibrium.
    Despite the claims to rigour, the pervasive equilibrium argument and focus on prices reveal a weak grasp of probability mathematics: Since for partial equilibrium in any market, at least the above eight conditions have to be met, if one generously assumed each condition is more likely to hold than not – corresponding to a probability higher than 50%, for instance, 55% – then the probability of equilibrium equals the joint probability of all conditions, which is 0.55 to the power of 8: less than 1%. As the probability of each of the eight conditions being an accurate representation of reality is likely significantly lower than 55% (most having a probability approaching zero themselves), it is apparent that the probability of partial equilibrium in any one market approaches zero (Werner, 2014b). For equilibrium in all markets, these very low probabilities have to be multiplied by each other many times. So we know a priori that partial, let alone general equilibrium cannot be expected in reality. Equilibrium is a theoretical construct unlikely to be observed in practice. This demonstrates that reality is instead characterised by rationed markets. These are not determined by prices, but quantities: In disequilibrium, the short side principle applies: whichever quantity of supply and demand is smaller can be transacted, and the short side has the power to pick and choose with whom to trade (not rarely abusing this market power by extracting ‘rents’, see Werner, 2005).1
    Without equilibrium, quantities become more important than prices.”

  55. Gravatar of Spencer B HallSp Spencer B HallSp
    1. February 2021 at 05:19

    In Alfred Marshall’s cash-balances equation:

    “If the public considers its real balances excessive or deficient, forces will be set in motion which will alter the value of the cash holdings of the public, but not necessarily in the fashion desire by the public.

    E.g., if the public on balance considers the real worth of its cash balances deficient, this will bring about an increase in the demand for money and a decrease in its supply. The velocity of one will decline, and if prices tend to be sticky, sales, production, employment, and payrolls will fall off. This will lead to reduced bank lending, a decline in the volume of money, and this will not be compensated by an appropriate decline in prices.

    Under these circumstances equilibrium is never reached, and the pubic in seeking to increase its real balances so reduces it effective purchasing power as to create a condition of chronic stagnation.”

    Thus low interest rates may induce people to hold onto their funds and not part with liquidity for such a small price. This will also tend to reduce the supply of funds and their velocity.

  56. Gravatar of Todd Ramsey Todd Ramsey
    1. February 2021 at 06:21

    Any believers in the strong version of EMH willing to back up their beliefs with a bet?

    I believe Gamestop is overvalued at its current price of $325. I am willing to back up this belief with a bet:

    Gamestop will sell for less than $100 per share on February 1, 2022.

    I am giving long odds. For you to win the bet, Gamestop only has to preserve 32% of its current market value.

    We can bet money if you want, but the true prize will be to have your convictions documented in the comments of TheMoneyIllusion.

    Any takers?

  57. Gravatar of Michael Rulle Michael Rulle
    1. February 2021 at 07:38

    @Todd Ramsey

    You probably are just joking. But, you can make that bet already——you know that , right? And if anyone wanted to take the other side, they don’t need you.

    Just pointing out the obvious—-as it struck me as interesting.

  58. Gravatar of J Mann J Mann
    1. February 2021 at 07:42

    It’s worth clarifying exactly what the EMH predicts.

    To Todd’s point, the current price of Gamestop doesn’t predict the price 12 months out. And it doesn’t make a prediction about the future cash flow of Gamestop as a company.

    Are we saying it’s the best predictor of the price tomorrow, or is it something more general?

  59. Gravatar of ssumner ssumner
    1. February 2021 at 09:13

    Kevin, Finally, a voice of reason.

    Todd, There’s already a place for you to bet. It’s called the “stock market”.

    BTW, did you notice that in the post I suggested that GameStop would probably go down? Most highly speculative stocks do, in an efficient market.

  60. Gravatar of Christian List Christian List
    1. February 2021 at 11:14

    I believe Gamestop is overvalued at its current price of $325. I am willing to back up this belief with a bet: Gamestop will sell for less than $100 per share on February 1, 2022.

    Todd R. (not the Toddmeister, I hope),

    How do you define “overvalued” and what would the outcome of your bet really prove? Let’s say your result comes true, then it would only mean that the company was worth this much today and this much in 2022. How would your bet disprove EMH? I just don’t see it, sorry.

  61. Gravatar of Tacticus Tacticus
    1. February 2021 at 12:12

    Kevin, that reminds me of the old joke, what do you call an early stage investor who has lost 95% of his bets? Fabulously wealthy.

    Todd, there are puts expiring 21 January 2022, why don’t you buy one with a strike under $100?

    Of course, the price being under $100 then would not disprove EMH. It would just mean that new information had come about and was reflected in the price.

  62. Gravatar of Christian List Christian List
    1. February 2021 at 12:40

    Interesting background about Gamestop:

    tl;dr version:

    Small investors conspire over the Internet to buy stocks like Gamestop in order to harm certain hedge funds. Hedge funds suffer but are bailed out by rich friends who also happen to run trading platforms. These popular trading platforms then obstruct the purchase of stocks like Gamestop.

    Fair game in the USA.

  63. Gravatar of J Mann J Mann
    1. February 2021 at 13:02

    Todd offers long odds against Gamestop trading above $100 in a year.

    According to, I can buy a put option at $165 a year out for about $120 – is there any way to structure something so my win/loss point is $100, and if so what would the odds be?

  64. Gravatar of Gene Frenkle Gene Frenkle
    1. February 2021 at 13:54

    Christian, GameStop is looking like a fairly typical pyramid scheme, the Redditors can rationalize it any way they want but it’s simply a pyramid scheme powered by the internet. Coincidentally I am watching season 1 of Schitt’s Creek and they made light of a cosmetics pyramid scheme in an episode. So good for the Redditors that made money…but they aren’t the first people to make money off of a pyramid scheme and they won’t be the last.

  65. Gravatar of Peter Schaeffer Peter Schaeffer
    1. February 2021 at 14:59

    Some history may help here. The NASDAQ traded at 1288.54 on 12/20/1996 and 1287.86 on 10/18/2002. In between it went over 5000. It is now well over 5000.

    The strong form of the EMH is laughable. Was the NASDAQ correctly priced when it was at 1288 in 12/20/1996 or 1287 on 10/18/2002 or when it was over 5000 in-between or at present?

    The NASDAQ chart screams “bubble”. Of course, bubbles are nothing new. Tulip bulbs were a famous bubble. Later there was the “South Seas Bubble” and of course, 1929.

    By the way, the weak form of the EMH (there is no strategy for beating the market, any stock could go up or down) is (in my opinion) true.

  66. Gravatar of agrippa postumus agrippa postumus
    1. February 2021 at 17:02

    maybe sumner is right: there are no bubbles, only winners and losers in the aftermath of information evaluation. bubbles exist for those who abstained, and those who were in but got fearful.

  67. Gravatar of Todd Ramsey Todd Ramsey
    2. February 2021 at 06:37

    @Tacticus: The 1/21/22 GME $100 Put ask price is $70 at market open 2/2/21. I actually cannot bet GME will go below $100; it has to go below $30 to get ANY money back. And this is with GME at $140, not the $345 I made the offer at.

    I think that all of you strong-EMHers believe, in your heart of hearts, that GME stock will fall below $100 by 2/1/22.

    I think one can believe in EMH generally, in that it’s not possible to know where the stock market as a whole will be in an hour, a day, a month, or a year, while also acknowledging there might be specific outliers that one could term “bubbles”. For the reasons Eliezer Yudkowsky outlined or for other reasons. That describes me.

  68. Gravatar of Michael Sandifer Michael Sandifer
    2. February 2021 at 11:03

    There’s still a ways to go, but so far I’ve been quite right about these Wall Street Bets stocks. Lot’s of money has been made shorting them and buying puts since the price run-ups.

  69. Gravatar of Tacticus Tacticus
    2. February 2021 at 11:04

    Good news, Todd, the put is down to $66.65! Seems like a steal, no?

    Of course GME will fall below $100 by 21/1/22, but that doesn’t disprove the EMH.

  70. Gravatar of Gene Frenkle Gene Frenkle
    2. February 2021 at 13:45

    Tacticus, pyramid schemes don’t necessarily follow the laws of microeconomics. So microeconomics assumes a population filled with rational actors but everyone knows as PT Barnum said, “a sucker is born every minute.”

  71. Gravatar of Tacticus Tacticus
    2. February 2021 at 14:07

    Gene, that has nothing to do with EMH or GME?

  72. Gravatar of ssumner ssumner
    2. February 2021 at 14:10

    Peter, You asked:

    “Was the NASDAQ correctly priced when it was at 1288 in 12/20/1996 or 1287 on 10/18/2002 or when it was over 5000 in-between or at present?”

    All three, given the information available at the time.

  73. Gravatar of Gene Frenkle Gene Frenkle
    2. February 2021 at 16:25

    Tacticus, correct, because what happened with GameStop had nothing to do with EMH…because what was going was a pyramid scheme. So with respect to Mary Kay the vessel to perpetuate the pyramid scheme is selling cosmetics at parties. The GME pyramid scheme’s vessel was the stock GME and it was pushed through Reddit.

  74. Gravatar of Christian List Christian List
    2. February 2021 at 16:48

    @Gene Frenkle
    I agree with you, it also reminds me of pyramid scheme and it does not really disprove EMH.

    I would never buy this stock. This company’s “new” business plan seems to be that they now want to sell more games over the Internet. Wow, that’s a really disruptive idea that no one has had before. I’m seriously impressed.

    This young Reddit generation seems to have some outlandish ideas about the stock market, short selling and company valuations.

    Nevertheless, it is wrong that many trading platforms suddenly prohibit the purchase of certain stocks.

    And it is also seems to be part of the truth that certain hedge funds have lost a lot of money, and that they are now trying to fight back, by legal means, but perhaps also by very questionable, possibly illegal, corrupt means.

  75. Gravatar of Tacticus Tacticus
    3. February 2021 at 03:48

    I disagree, Gene. What has happened with GME is the EMH at work.

    Christian, why is it wrong for trading platforms to prohibit the purchase of certain stocks?

    A few hedge funds certainly lost a lot of money, but there is no illegal or corrupt activity going on here.

  76. Gravatar of Christian List Christian List
    3. February 2021 at 14:35


    Christian, why is it wrong for trading platforms to prohibit the purchase of certain stocks?

    If you can’t answer that question yourself, then it’s going to be really hard for me. It’s clear from a liberal and libertarian point of view, even extreme opposites like AOC and Ted Cruz agree on it.

    Customers have entered a contract in the strong believe that they can buy certain stocks. And then all of a sudden this trading platform takes stocks off sale on the flimsy grounds that they want to protect the buyers from themselves. It doesn’t get any more anti-liberal and anti-libertarian than that.

    It’s a crass comparison, but there were similar methods among the Nazis from a purely linguistic point of view. Imprisonment for political and other reasons was not called imprisonment, but “Schutzhaft”, in English “protective custody”. It was not a punishment from the Nazi point of view, but the loving protection of the patient from himself.

    Similar things apply to concentration camps, etc., which can be translated as a camp where people learn to “focus”.

  77. Gravatar of Tacticus Tacticus
    4. February 2021 at 01:44

    Of course AOC and Cruz agree; they’re both populists and not as far apart as you imagine.

    Have you not paid any attention to the GameStop saga? Robinhood was forced to pull GameStop because they did not have enough capital with their clearing house. They did not choose to disallow purchases.

Leave a Reply