When bitcoin crashes . . .

. . . .I predict people will say it was a bubble, even though it wasn’t.  The term ‘bubble’ can mean many things, but the sine qua non of definitions includes “rejection of the EMH.”  But the EMH says that bitcoin is very likely to crash.  Why is this so, and why don’t people know this?

1.  We know that market volatility is serially correlated.  Markets that have been highly volatile are likely to remain highly volatile.

2.  Bitcoin prices are super volatile.

3.  The EMH predicts that expected returns are near zero.  Combined with high volatility, this mean the EMH predicts that bitcoin will exhibit large price increases and large price decreases at various times in the future.

When the bitcoin price crash comes, most people will wrong say; “Aha, I told you that it was a bubble.”  But why?

Because they wil have forgotten about their first bubble prediction.  People were calling it a bubble at $2, and again at $30.  Now it’s over $200.  If it plunged to $35 dollars, the bubble predictors will say they were right all along, but they will have been wrong.  They’ve merely remember their bubble predictions, not where bitcoin was when the made the predictions.

[Ha! I wrote this yesterday and delayed posting–that’ll teach me.  It’s $160 tonight]

But the internet never forgets anything.  And I’ll search and search and expose every phony “I told you so.” I can’t predict where bitcoin is going, but I can predict there will be many false “bubble” claims when it eventually crashes—and it will crash.  The only question is whether it will crash from a price so far above the current price, that it’s still a good buy at $200. $160.

And the EMH says that the answer to that question is; “God only knows.”

Memo to commenters:  All comments telling me why bitcoin is a bubble will be ignored, as they’ll show you missed the entire point of the post.

 


Tags:

 
 
 

83 Responses to “When bitcoin crashes . . .”

  1. Gravatar of Bill Ellis Bill Ellis
    10. April 2013 at 19:56

    If bitcoin becomes completely worthless was it a bubble then?

    So Bubbles can only exist if the value of the asset rises and then falls below its starting value ?

    So a house that sold for 2,000 in 1961 near beach here in San Diego and got up to 500 thousand in 1994 then up to one mill in 2004, and then dropped back to 500 thousand in 2008 did not experience a bubble between ’94 and ’08 because it did not fall all the way back to 2000 bucks ? That does not seem right.
    Seems like things can experience bubbles in their value at any point relative to a past price point.

    I guess I missed the point entirely.

  2. Gravatar of Benjamin Cole Benjamin Cole
    10. April 2013 at 20:21

    I would take issue, but first I would have to find out what is bitcoin.

  3. Gravatar of bawld bawld
    10. April 2013 at 20:26

    The point he’s making is that you can’t take cries of “It’s a bubble!” seriously unless someone states what they think the real value should be. If you cry “bubble!” forever on a highly-volatile security, then you will eventually be “right” in the sense that the price will eventually go down from one of the times you claimed it was a bubble.

  4. Gravatar of dkn dkn
    10. April 2013 at 21:30

    @Bill Ellis – you nailed it.

    Reading this reminded me of old fringe socialist doctrine of “State Capitalism” – these folks were revolutionary socialists who didn’t like the USSR so decided that it must be capitalist somehow.

    Here we have fans of Bitcoin who look at a massive collapse in price and say “it wasn’t a bubble.” Sorry, but that is the definition of a bubble. This post is pure sophistry.

    At least fans of bitcoin can take comfort that it will never drop below its true value: absolutely nothing.

  5. Gravatar of polymath polymath
    10. April 2013 at 21:48

    @Bill Ellis, the point is that the EMH is awesome! It relieves us of the need to value any market-priced asset or seriously evaluate any argument that an asset is unfairly valued! Prices go up, prices go down, a hundred people will call a top and one of them will happen to be right, but that person won’t have had any insight, they’ll have been just a lucky penny. And we can know this without even looking at what any of those hundred people say.

    It’s truly brilliant, awesome in the medieval sense of the word.

  6. Gravatar of JG JG
    10. April 2013 at 21:52

    Whether or not it’s a bubble price-wise, the idea that it’s going to be a medium of exchange in the future (however small) is a pretty big bubble. Price stability will be almost impossible to achieve and add in volatility and not many producers of goods will want the added risk to their normal transactions.

  7. Gravatar of Jim Glass Jim Glass
    10. April 2013 at 22:37

    If it plunged to $35 dollars, the bubble predictors will say they were right all along, but they will have been wrong. They’ve merely remember their bubble predictions, not where bitcoin was when the made the predictions.

    Some while ago I saw cable TV finance show celebrating a panel of experts who had called the home market crash for being so smart for having done so.

    One of them was Jim Grant, well known Wall Street financial publisher and commentator, and long one of my favorites as he’s always entertainingly both candid and acerbic, as well as smart and well informed, if highly eclectic and congenitally skeptical about everything. And too much of a hard-money guy for me (Ron Paul named him his choice to replace Bernanke) — the type of guy one should always listen to, if not agree with.

    Anyhow … after everyone else takes a bow for being so smart they get to Grant and he goes “Not me, I have nothing to feel good about, I was flat wrong.” But you predicted the plunge and it happened!

    “After I called the plunge it continued to rise. Every market is going to fall, calling that is nothing. You have to call *when* it is going to fall right or you are wrong. I said it was going to fall and it continued to rise. I was wrong. And the rest of you who did the same thing were wrong too.”

    Cough, ahem, a bit of awkward equivication all around.

    After that he was even more one of my favorites.

  8. Gravatar of Paul Andrews Paul Andrews
    10. April 2013 at 23:04

    Scott,

    You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.

    Would you agree it’s possible there have been positive feedback episodes in the price of bitcoins?

    To refresh your memory, a positive feedback episode occurs as follows:

    Sometimes people buy assets purely because they are increasing in price.

    A positive feedback loop is sometimes created: people buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc.

    Eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out.

  9. Gravatar of Paul Andrews Paul Andrews
    10. April 2013 at 23:51

    Fama himself has admitted that poorly informed investors could theoretically lead the market astray (see http://online.wsj.com/article/0,,SB109804865418747444,00.html ).

    I don’t really see why, then, that the existence of bubbles would disprove the EMH. The EMH is an approximation, and I believe quite a good one at that. One can see bubbles as an example of where the approximation can sometimes diverge from reality.

  10. Gravatar of TravisV TravisV
    11. April 2013 at 00:11

    The Fed did not cause the housing bubble.

    According to new research by Ambrogio Cesa-Bianchi and Alessandro Rebucci, the housing bubble was caused by “regulatory rather than monetary-policy failures”:

    http://www.voxeu.org/article/federal-reserve-breeding-next-financial-crisis

  11. Gravatar of J.V. Dubois J.V. Dubois
    11. April 2013 at 00:13

    Bill: You completely miss the point. It’s like markets saying that if you cast a regular 6-sided dice you have 1/6 probability to get 6 as a result. If you get 1 does it mean that markets were “wrong”?

    So imagine situation like with bitcoin. Here and now there can be a 1% chance that this thing will become a popular and widely used money with a very large price of many thousands of dollars per piece. And there is a 99% chance that for various reasons it will not amass sufficient number of users and its value will be very small – even 0 – in the future. You are bound to see a price of such a thing with such a high inherent uncertainty built into the price mechanism to behave like mad. You cannot use volatility itself to claim that something is a “bubble”

  12. Gravatar of Saturos Saturos
    11. April 2013 at 00:16

    Just to clarify, Scott, what possible future would make you admit that it was a bubble?

  13. Gravatar of Saturos Saturos
    11. April 2013 at 00:31

    They could have done this to you, Scott: http://www.salon.com/2013/04/10/henry_blodgets_horrible_bitcoin_advice/

  14. Gravatar of Saturos Saturos
    11. April 2013 at 00:57

    Petra Geraats from Cambridge on central bank transparency –

    Video: http://www.bloomberg.com/video/geraats-says-central-banks-should-be-more-open-VzGODJiXRfqdISLX0ezKyw.html

    Article: http://www.bloomberg.com/news/2013-04-10/central-bank-transparency-grows-under-geraats-pressure.html

  15. Gravatar of W. Peden W. Peden
    11. April 2013 at 01:09

    Jim Glass,

    There are few things I respect more than a financial pundit who recognises the limitations of human knowledge.

  16. Gravatar of Tom Hannaford Tom Hannaford
    11. April 2013 at 02:10

    I guess the thing that irks me most is the constant rabble about Bitcoins being nothing more than digital trinkets of some variety. I mean, I don’t have any grand imaginative ideas that it’ll become some widely accepted medium of exchange (much as my inner capital-L libertarian wants it to be), but its anonymity and p2p structure make it very useful for transferring funds internationally and in particular could serve as a great way for people to move capital around despite what barriers may exist in their home countries (whether you see that as a good or bad thing is dependent on your own beliefs about such policies, but to each their own that’s a debate for another time). Besides, paying for services in BC is not quite as chaotic as the price fluctuations would lead you to believe (though they certainly make any BC based contracts a pipe-dream); I hesitate to cite the example because of the jeering it has received from atop Mt. Pious here in the States, but having monitored the Silk Road prices for awhile just to see how merchants would adjust to the rise and fall of BC value it seems like prices adjust almost on the fly (not universally of course, but for the more prominent merchants on there it seems to be a constant). IDK, perhaps I’m more wide-eyed than I should be, but I can’t help but feel extremely annoyed by the constant framing of it as nothing more than a “parallel currency” that’s doomed to failure, since this notion ignores a lot of the more unconventional uses of the BC and the areas where it really is far superior to most current currencies (although again, I can’t stress enough how shocked I would be if it ever became a widely accepted medium of exchange)…

    Apologies for the poorly formatted eternal paragraph :p

  17. Gravatar of Tom Hannaford Tom Hannaford
    11. April 2013 at 02:13

    PS- I don’t normally like Salon, but the writer is correct about Blodget: Good chance you’re going to be verrrrry disappointed if you invest in BC like it’s a commodity or normal investment that you expect decent returns on…

  18. Gravatar of Saturos Saturos
    11. April 2013 at 02:16

    More on the BoJ: http://www.reuters.com/article/2013/04/11/us-markets-japan-poll-idUSBRE93A06N20130411

    http://www.ft.com/intl/cms/s/ddd94d14-a27e-11e2-bd45-00144feabdc0,Authorised=false.html?_i_location=http%3A%2F%2Fwww.ft.com%2Fcms%2Fs%2F0%2Fddd94d14-a27e-11e2-bd45-00144feabdc0.html&_i_referer=http%3A%2F%2Ft.co%2FEvrBWeppaO#axzz2Q8tCqnM3

  19. Gravatar of Saturos Saturos
    11. April 2013 at 02:16

    Obama’s 2014 Budget remarks: http://swampland.time.com/2013/04/10/president-obamas-2014-budget-remarks-transcript/

  20. Gravatar of Saturos Saturos
    11. April 2013 at 03:20

    Interesting Twitter discussion on bitcoin value: https://twitter.com/TVHE/status/322192860864724993

  21. Gravatar of Geoff Geoff
    11. April 2013 at 03:33

    Dr. Sumner:

    “The EMH predicts that expected returns are near zero.”

    At what time horizon?

    If the time horizon is all-inclusive, meaning it includes not only the returns being earned in the seconds, minutes, hours, etc, but also the weeks, months, years, then EMH is constantly being refuted by empirical reality, since returns are always non-zero depending on the time horizon considered.

  22. Gravatar of Matt Matt
    11. April 2013 at 03:48

    I think there’s a plausible case that:

    1. The correct value of bitcoins is zero, like the correct value of pet rocks etc., and

    2. Therefore any nonzero price is indicative of a bubble insofar as it “rejects the EMH” by assigning a purely greater-fool-based price to bitcoins.

    That relies on bitcoin not being a viable currency for structural reasons, and on people eventually losing interest in bitcoin speculation if it’s not a workable currency. I agree with you that if bitcoin crashes from $200 to $35 and stays there then this theory is *disproved*, not proved. But I think in these early days it’s still respectable to predict a crash to zero.

    I think there’s a plausible case that a crash to, say, $1 – meaning really quite a small amount of bitcoins in circulation – indicates a crash from greater-fool value to hobbyist-currency value and would also be ex post proof of a bubble. I don’t know what the right number is there.

  23. Gravatar of Justin Irving Justin Irving
    11. April 2013 at 03:55

    Luckily, it is possible to short bitcoins. So we can safely ignore “bubble people” who haven’t become rich on their ‘foreknowledge’.

  24. Gravatar of 123 123
    11. April 2013 at 04:05

    Weak version of EMH is perfectly compatible with bubbles. I have studied the theory of rational bubbles too little so I don’t know if they are compatible with stronger versions of EMH.

  25. Gravatar of Unlearningecon Unlearningecon
    11. April 2013 at 04:05

    This is basically just a bunch of unfalsifiable claims then telling anyone who disagrees with you you’ll ignore them. You say bitcoin is likely to crash, but if it doesn’t crash I’d be willing to bet you’d use that as evidence it’s not a bubble, too. Are you saying that bitcoin not crashing would refute the EMH?

    “3. The EMH predicts that expected returns are near zero. Combined with high volatility, this mean the EMH predicts that bitcoin will exhibit large price increases and large price decreases at various times in the future.”

    Over what time period? Many assets are highly volatile in the short term; bitcoin has been incredibly volatile. I don’t see how the EMH ‘predicts’ this seeing as no new fundamental information is being released, but I’ll ignore that for now.

    “Because they wil have forgotten about their first bubble prediction. People were calling it a bubble at $2, and again at $30. Now it’s over $200. If it plunged to $35 dollars, the bubble predictors will say they were right all along, but they will have been wrong. They’ve merely remember their bubble predictions, not where bitcoin was when the made the predictions.”

    Of course nobody can know exactly when the crash will come. That is, and always has been, the nature of bubbles. Those who say bitcoin is a bubble are talking about trend rates over periods of days and weeks, not the minute-to-minute jumping around we observed last night. Let’s take here:

    http://bitcoinity.org/markets

    If you look at, say, 3 or 7 days, you can see a clear trend: it looks like a classic bubble, one that has been observed throughout history! (eg: http://www.commodityonline.com/uploads/z1.jpg)

    At this stage it looks like bitcoin has just recovered from the bull trap. I’m going to make a falsifiable prediction and say it will crash again in the next week, fully (eg down to where it was in February). Maybe I’m wrong, but you are Not Even Wrong, because there is no event which could make you conclude bitcoin is a bubble.

    Some defenders of the EMH have done a better job than you, btw. They’ve asserted that bitcoin is a tiny market full of idiots and as such the EMH is not really suited to it.

  26. Gravatar of S Cunsolo S Cunsolo
    11. April 2013 at 04:24

    I think the initial point of the post is right, and I’m also a strong supporter of EMH as very-close-to-truth approximation.

    However I don’t see the point of thinking of BTC or of the alleged “BTC bubble” in EMH terms, as I think BTC market is young and small enough to warrant very very cautious application of EMH.

  27. Gravatar of Nick Nick
    11. April 2013 at 04:40

    I agree that the term “bubble” gets tossed around too much. The people calling bitcoin a ‘bubble’ might just be using that term as an epithet for a commodity that they don’t like.

    The real question isn’t whether the bitcoin market is in a bubble, the question is whether the bitcoin market is even a legitimate market. The largest bitcoin exchange (Mt Gox) was, until fairly recently, a place to buy and sell used Pokemon^H^H^H^H^H^H^H Magic: The Gathering cards. The bitcoin exchanges seem to have well-publicized successful hacking attacks every few weeks (and who knows how many not-well-publicized successful hacking attacks).

    I don’t have anything against bitcoins themselves (the technology is a pretty impressive use of crypto), but the infrastructure supporting them is incredibly weak right now. It makes you appreciate just how difficult it is to get a financial system working at all.

  28. Gravatar of W. Peden W. Peden
    11. April 2013 at 04:47

    Unlearningecon,

    The EMH may predict that bitcoin will crash, but nevertheless wouldn’t be refuted if bitcoin doesn’t crash. This is because (a) you have to make additional assumptions to formulate a prediction and (b) the EMH is a probabilistic hypothesis i.e. it concerns what will probably happen, rather than what will certainly happen, so what matters for it will be the balance of probabilities rather than one particular case.

    “Of course nobody can know exactly when the crash will come. That is, and always has been, the nature of bubbles.”

    That’s actually a stronger of efficiency than most defenders of the EMH are willing to assert i.e. they usually say that you can’t reliably know (using analyses of past data for the weak-form hypothesis and publicly available new information for the semi-strong hypothesis) what path prices will take, such as a bubble crashing, rather than saying you can never know exactly when a bubble breaks.

    I think you raise a good point when you say that it isn’t even clear that bitcoin is the type of market that is supposed to be efficient. That bitcoin is the relevant type of market is one of those additional assumptions I mentioned earlier.

  29. Gravatar of Steve Steve
    11. April 2013 at 04:59

    There are about 10mm bitcoins outstanding. At a price of $150 that’s a market value (or a money supply!) of $1.5 billion.

    $1.5 billion is a tad bit more than the market value of AAMRQ, the bankrupt equity claim on American Airlines. Unfortunately, bankrupt AMR stock certificates haven’t caught on as an alternate currency, even though there will never be any more of these created.

  30. Gravatar of ssumner ssumner
    11. April 2013 at 05:31

    Bill, Most people really have trouble understanding this, so don’t feel bad. It’s very counterintuitive. I think the best way to approach this is to consider the following assumptions:

    Imagine a market is highly volatile for legitimate reasons (i.e non-bubble reasons.) That is, suppose the EMH is true.

    Can that market only exhibit volatility on the upside? Obviously not, otherwise it wouldn’t be volatile, it would be continual growth.

    So any highly volatile market, even a non-bubble market, will exhibit price crashes.

    Now consider how you would distinguish between a true bubble market, and the highly volatile market that I just described, that is not a bubble market. It’s much harder than you think.

    If you think long and hard enough you’ll realize that the only way to show there is bubbles is to show the bubble-mongers are usually right, not just that they happen to be right on the one time (or multiplie times) the market crashes. But usually right.

    Obviously they have usually been wrong about bitcoin, so there is no evidence for a bubble. If the price falls to zero then all of them will have been right. That would provide one piece support for bubble theory.

    dkn, Don’t feel bad, this is really difficult stuff for most people to understand. Read my reply to Bill.

    Jim Glass, One of my favorite posts was on an ad from the Economist magazine showing predictions they had made about the housing markets of various countries back in 2003. Then in 2009 in their ad they bragged they were later proved right. But when you actually read the predictions and then look at the data the Economist magazine wasn’t just wrong, they were wildly off base.

    You can see in this comment section how hard it is for people to overcome these cognitive illusions.

    Paul , You said;

    “You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.”

    No I have not. I’ve said there are people who behave that way, not that market prices behave that way.

    Travis, V. That’s right.

    Saturos, I don’t believe bubbles exist. I think what you really mean to ask me is “what sort of bitcoin future would provide evidence that bubbles exist.” That would be future price movements to a place where the majority of bubble predictions were wrong. Right now the price is above the point where most bubble predictions were made, so it’s a point against.

    But even in that case it would simply be one observation. Here’s an analogy. Suppose you ask me what sort of evidence I’d need to support the notion that a coin is unbalanced, and biased toward “heads.” I’d say I’d need a coin flip that came up heads. But that would merely support the theory at P = .50. You like a much higher prob value, and hence lots more coin flips. I’d become convinced of bubbles if in a high percentage of different markets the bubble predictors were beating the markets.

    Back in 2009 I said Robert Shiller had the best anti-EMH pro-bubble model I had ever seen. The one that I respected most. And over the past 4 years his model has done horribly, I mean you couldn’t ask for a more pathetic performance. And that’s the best bubble model!! The bubble people are not winning me over, indeed almost every day I see more and more evidence that it’s really hard to predict markets.

    But yes, my bubble skepticism is refutable, as I’ve explained many times, now lets see the SYSTEMATIC studies that refute it.

  31. Gravatar of Becky Hargrove Becky Hargrove
    11. April 2013 at 05:34

    The fact that bitcoin was given a specific limit makes it seem like an exploratory game, rather than a useful platform. The “bubbles” appear to be a way to gauge what happens in the game. Perhaps if it had no limit, the platform would already have been shut down because it would have proved “too” useful.

  32. Gravatar of Lio Lio
    11. April 2013 at 05:41

    Scott said: “Because they wil have forgotten about their first bubble prediction. People were calling it a bubble at $2, and again at $30. Now it’s over $200. If it plunged to $35 dollars, the bubble predictors will say they were right all along, but they will have been wrong. They’ve merely remember their bubble predictions, not where bitcoin was when the made the predictions.”

    Bad argument: it assumes that the forecaster is able to determine exactly when and at what specific level, the stock market will collapse. Nobody has ever claimed that. If someone believes that the stock market is too high and considers that it is a bubble and, later, the stock market suffered a crash, even if the price remains above the level at which he made this prediction, it does not mean he was completely wrong. Only the timing and level of the stock price at which the stock market fall are inaccurate or wrong, but the forecaster will still have been right about the outcome.

  33. Gravatar of Morgan Warstler Morgan Warstler
    11. April 2013 at 06:04

    The reality is you’d expect bitcoins to climb and fall spasmodically, each time reaching new heights.

    THEN, if you saw such a thing, you’d expect the govt. TRIES to get it…

    And then it fails to do so, you’d continue to expect less drastic peaks and valleys.

    —–

    Someone out there needs to really address the release valve of infinite divisibility.

    Right now there are 2.1 quadrillion bitcoins, people are just trading MASSIVE amounts of them.

    There’s a technical achievement which pushes the decimal point even further to the left.

    Deflation isn’t a real problem, each new person asking to SELL REAL GOODS AND SERVICES to get some of those 2.1 quadrillion BTC, is assuming they are being paid in stock options.

    Note: many many peeps choose to work for equity, they end up owning the the world.

  34. Gravatar of Matt Waters Matt Waters
    11. April 2013 at 06:05

    The post seems to be mainly about semantics. Here’s a simpler way of looking at market prices:

    1. If people are buying something for innate fundamental value, then they’re “rational.”
    2. If people are buying based purely on price appreciation, then they’re “irrational.”

    I put “rational” and “irrational” in quotation marks because acting rational could be irrational and the irrational rational, depending on the structure or incentives and the actor’s discount rate.

    In the case of a pure bubble, where the underlying asset has zero value, it’s a zero-sum game before transaction costs. Any realized profits have to come from somewhere, and since there’s no underlying cash flows and no non-cash utility from the underlying asset, the market as a whole loses money. But the irrational becomes rational if market actors either:

    1. Have a call-option type payoff structure, where they have the upside with the only downside of a “option premium,” such as their salary if they’re a fund manager.
    2. If they have both upside and downside, but their utility curve is inverse and risk-seeking. I.e. they consider it “play money” and value upside more than they’re hurt by downside.

    More typical is case #1, where fund managers have an incentive to increase their option value by increasing volatility, even if it hurts the value of the underlying. Case #2 is more applicable to Bitcoin, where I know somebody who has that exact utility curve.

    But there’s also players in Bitcoins who, rather than rational irrationality, are exhibiting irrational irrationality. Either they believe that the currency has innate value or they think they have a special ability to get out right at the peak of the, yes, bubble.

    In either case, for rational irrationality or irrational irrationality, the fact that market prices do not always reflect true underlying fundamental value is important, for both economic policy and investing. The semi-strong form of EMH would say that Bitcoin’s current market price reflects all publicly available information. Taking the theory literally means that ALL price changes for Bitcoin are due to news concerning fundamental value. That idea is absolutely silly.

    Of course, Bitcoin is a far less liquid market than the market for most stocks, but the issues driving Bitcoin’s volatility do not stop neatly at Bitcoin. What to do about such issues, from either economic policy or investing, is beyond the scope of this post, but it helps to acknoledge that the issues are there.

  35. Gravatar of Morgan Warstler Morgan Warstler
    11. April 2013 at 06:09

    Think about the guys who have lots of bitcoins, who ARE THEY?

    What kind of goods and services do they consume?

    They buy computers, datacenters, bandwidth, coders, and snacks.

    Does anyone who sells those things want to be paid partially in BTC?

    Yes.

  36. Gravatar of ssumner ssumner
    11. April 2013 at 06:10

    Geoff, I should have been more precise—expected risk-adjusted excess returns are near zero. But the point reamins.

    Matt, This post is not about bitcoin, it’s about bubbles.

    unlearningecon. My anti-bubble theory is certainly testable, as I’ve pointed out in numerous posts over the years. I’d encourage you to take a look before dismissing them.

    You said;

    “Over what time period? Many assets are highly volatile in the short term; bitcoin has been incredibly volatile. I don’t see how the EMH ‘predicts’ this seeing as no new fundamental information is being released, but I’ll ignore that for now.”

    No new information? Surely you are kidding. Cyprus? BTW, Not all information is “released.”

    You said;

    “Of course nobody can know exactly when the crash will come.”

    No one is asking them to–please reread what I wrote.

    You said;

    “If you look at, say, 3 or 7 days, you can see a clear trend: it looks like a classic bubble, one that has been observed throughout history! (eg: http://www.commodityonline.com/uploads/z1.jpg)”

    This is just confirmation bias. It’s like you haven’t read anything I’ve said, but simply substituted what you thought I was saying. Don’t be so lazy!!! Read what I actually say.

    Nick, You said;

    “I agree that the term “bubble” gets tossed around too much. The people calling bitcoin a ‘bubble’ might just be using that term as an epithet for a commodity that they don’t like.”

    It’s even worse. They are applying the term “bubble” to any asset that soars in value and then drops sharply. It’s like someone arguing till they are blue in the fact that Paris really is the capital of France, when that’s not an issue even being debated. No one denies that there are sharp rises and falls in asset prices. The debate is over what that means. Most bubble mongers can’t get beyond that, which his why they throw insults. “How can you not see the price soared and collapsed!!!” Umm, I do see that.

  37. Gravatar of ssumner ssumner
    11. April 2013 at 06:13

    Matt, Why do you say there is no “non-cash utility”. Are people using bitcoins to make transactions?

    In any case, the nature of bitcoin has no bearing on this post–people should stick to the topic. The post is not about bitcoin.

  38. Gravatar of Paul Andrews Paul Andrews
    11. April 2013 at 06:20

    Scott,

    I asked on April 4:

    “Do you agree that sometimes people buy assets purely because they are increasing in price?

    If you agree with that, do you agree that a positive feedback loop is sometimes created: people buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc. ?

    If you agree with that, do you agree that eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out?”

    You replied:

    “Paul, I agree with all three statements, but don’t think they imply the existence of bubbles. There are irrational people everywhere.”

    Above I commented:

    “You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.”

    You replied:

    “No I have not. I’ve said there are people who behave that way, not that market prices behave that way.”

    The statements, which you have agreed are correct, refer to the interaction of people and prices, not only to people’s behavior.

    However, let me rephrase my question.

    You’ve previously agreed that there exists a phenomenon in human behavior which I termed a positive feedback episode.

    Would you agree it’s possible there have been positive feedback episodes in people’s behavior with respect to bitcoins?

  39. Gravatar of Paul Andrews Paul Andrews
    11. April 2013 at 06:23

    Scott,

    Saying that a post headlined “When bitcoin crashes” is not about bitcoin doesn’t really help your argument.

  40. Gravatar of J.V. Dubois J.V. Dubois
    11. April 2013 at 07:05

    Paul Andrews: I was also there at the time – identifying bubbles by a fact that the subject of a bubble has value purely on based on how other people perceive the value

    But this view along with a view of “bubble is if good does not have inherent value” has a great flaw. Because it turns into debate of what is the “correct” view of value. Under marginalism a value of something is the amount of utility it provides on the margin, which is assessed in markets.

    Imagine that fur-coat is really highly sought off good. It is a symbol of status and everybody wants it to show-off his power and wealth. And it is so for 100 years until a new generation comes that starts a campaign that fur-coats are immoral and it is no longer viewed as status symbol but as symbol of somebody being arrogant and approving of animal torture. The price of Fur-Coats crashes. Would you claim that there was a bubble in Fur-Coats?

  41. Gravatar of Cthorm Cthorm
    11. April 2013 at 07:19

    Scott,

    I’ve come up with a great trading strategy.

    1) Infect your computer with a trojan horse
    2) Parse the text from any recent documents not found on the blog. Book trades at market open.
    3) Profit

    On a less serious note, here is my narrative for BTC volatility:

    1) Very shallow volume of the BTC FX market
    and
    2) Bimodal expectations of market participants (BTC as blue tulips! vs Dollar collapse!)

    These are the primary factors contributing to current BTC volatility. Overtime the market will grow in depth and there will be some convergence of expectations as it becomes clear that the likelihood of a near term extreme event is very low. Personally, I think the most significant use of BTC will be in tax avoidance/black market/grey market activities; in other words, it will fill the role of paper USD.

  42. Gravatar of Geoff Geoff
    11. April 2013 at 07:19

    Dr. Sumner:

    “Geoff, I should have been more precise””expected risk-adjusted excess returns are near zero. But the point reamins.”

    Expected by who? Everyone? Clearly not everyone. If the baseline is the average market return, then the whole reason investors invest in particular projects, rather than a share of all projects, is because they expect to make profits greater than the market return.

    The “zero excessive profit” is something that you could very well expect, but it doesn’t mean everyone else expects zero excessive profits.

    At any rate, my question remains: Over what time horizon?

    Obviously, empirical profits are always either “excessive” or “inexcessive” relative to some theoretical baseline, by selecting appropriate time horizons in historical data.

  43. Gravatar of Geoff Geoff
    11. April 2013 at 07:21

    Forgot to mention: *risk adjustments* require an assumption of future volatility, and investors also don’t all agree with each other about that either.

  44. Gravatar of Dan Dan
    11. April 2013 at 07:59

    Scott,

    EMH implies that prices reflect all information about fundamental value. This implies that if prices are correct then there is nor free lunch (e.g. you cannot make excess risk-adjusted returns as you’ve mentioned). As Thaler and Barber point out, there reverse is not true: no free lunch does not imply correct prices. There could be a number of reasons for this: noise trader risk, fundamental risk that you cannot hedge, transaction costs, etc. Can I short bitcoin? if I can short it, can I hedge against the risk of a sudden decrease in confidence in fiat currency before bitcoin collapses? , etc.

    The point is, there is no way to test for bubbles (unless a clear arbitrage opportunity exists). The test that you propose is meaningless.

  45. Gravatar of Matt Waters Matt Waters
    11. April 2013 at 08:09

    “Matt, Why do you say there is no “non-cash utility”. Are people using bitcoins to make transactions?”

    The distinction was about why people are buying it. I’m sure some people have bought it in the past for use in black-market transactions, but most of the activity now is to get future price appreciation.

    The issue of “bubbles” is much more clear in the case of pure financial assets such as stocks, where the only fundamental value is future cash flows. If gold doubles in price, who is to say that the demand curve for gold hasn’t shifted outward? Indeed, I had my mind changed about whether there is a bubble in gold after seeing that most new demand was for increased jewelry demand in developing countries. Speculative demand for gold is actually very minimal.

    A better analogy may be homes in the most bubbly housing markets. Most of America’s housing was probably not a bubble, but most of the activity at the peak in the bubble markets was for speculation. Of course the houses still had their fundamental function as residences, but they were bought for future price speculation and not for residences. These markets reverted back to fundamental demand and bottom fell out, even before NGDP in general crashed.

  46. Gravatar of JP Koning JP Koning
    11. April 2013 at 08:09

    Excellent comment by Matt.

    The greater fool theory of assets is that some intrinsically worthless asset x can have a temporarily positive price before falling back to 0.

    If x rise from $1 to $250 and then falls to $130, that $120 fall is no indication that the asset is a greater-fool asset, as Scott points out. It could just be noise. If the asset falls back to a sub $1 amount, then that’s surely an indication that x was an asset driven by the greater-fool mechanism. This would be what Matt calls ex-post proof of a bubble, or a greater fool asset.

    As 123 points out, a greater fool theory of assets can be consistent with the weaker forms of EMH. I prefer these more subtle versions of EMH to the totalitarian form that Scott espouses.

  47. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    11. April 2013 at 08:24

    Unlearningecon is my old buddy Robert Vienneau, who has a fetish for Pierro Sraffa;

    http://unlearningeconomics.wordpress.com/2012/01/01/an-faq-for-mainstreamneoclassical-economists/

    Also he’s long been in denial about QWERTY;

    ‘ Lack of clustering or ‘QWERTY’ effects. It is an obvious observation that firms in particular industries tend to cluster together geographically. Manufacturing requires a continuous stream of inputs, so firms at different stages in the supply chain will group together to minimise transaction costs. Manufacturing often – though not always, to be sure – requires workers with a particular set of skills, so employees and employers who best match together will tend to converge. Services, by their nature, requires face-to-face interaction, as well as even more specialised skills, so they too will group together. In both cases the easy transfer of knowledge around clusters also helps significantly. Clusters become self-enforcing: you set up shop in a cluster because everyone else in your industry is there. QWERTY effects create emergent properties that may suggest a role for government intervention.’

    I’m surprised he hasn’t shown up here before this, but if he thinks I wouldn’t recognize his style and his content then I have to revise my opinion of him even further downward.

  48. Gravatar of Doug M Doug M
    11. April 2013 at 08:29

    Professor,

    Decent job reconciling market volatility with the EMH.

    I am okay with the term bubble. But, I don’t think that bubbles are irrational.

    As far as bitcoins go, I don’t understand them. I don’t understand why they have value, other than that their supply is limited and some people accept them as a medium of exchange. But, I am not buying bitcoins.

  49. Gravatar of Steve Steve
    11. April 2013 at 09:27

    Bitcoin announces capital controls, suspends convertibility, and blames speculators!

    http://www.telegraph.co.uk/finance/economics/9987664/Bitcoin-trading-halted-after-panic-sell-off.html

  50. Gravatar of Nick Nick
    11. April 2013 at 09:35

    I think “bubblemongers” are not simply referring to any asset that soars in value and then drops sharply as a bubble, because every asset soars in value and then drops sharply, for certain values of “soar,” “drop,” and “sharply.” They don’t call everything that goes up or down a bubble. They just call those things that they will profit on by a decline a bubble.

    Even if they don’t have an explicit short position, they tend to put themselves in the position of profiting some way from a decline, even if it is only to be seen as the miraculous prognosticator who foresaw the doom. That’s why they tend to call bubbles in things that are high profile or “sexy” for some reason. When they start calling a bubble in something that nobody cares about, like paper companies, then maybe they mean it.

    Until then, remember that everybody talks their book. Everybody.

  51. Gravatar of 123 123
    11. April 2013 at 09:56

    Wisdom of crowds requires a diversity of opinion. Diversity of opinion means shorting. Low shorting activity before the crash makes me suspicious the crowd was not smart.

  52. Gravatar of Geoff Geoff
    11. April 2013 at 11:13

    Steve:

    That wasn’t Bitcoin, it was the Japanese exchange Mt.Gox. Around 3/4 of Bitcoin transactions go through that exchange.

  53. Gravatar of Mike Sax Mike Sax
    11. April 2013 at 12:41

    Scott I know you believe EMH which says no one can reliably predict when but on this I’m pretty impressed.

    http://abclocal.go.com/wls/story?section=news/technology&id=9061762

    Anyway you’ve missed your calling as a market analyst? LOL

  54. Gravatar of Paul Andrews Paul Andrews
    11. April 2013 at 15:15

    J. V. Dubois:

    You said:

    “Imagine that fur-coat is really highly sought off good. It is a symbol of status and everybody wants it to show-off his power and wealth. And it is so for 100 years until a new generation comes that starts a campaign that fur-coats are immoral and it is no longer viewed as status symbol but as symbol of somebody being arrogant and approving of animal torture. The price of Fur-Coats crashes. Would you claim that there was a bubble in Fur-Coats?”

    No. A bubble occurs when a significant number of people buy purely because the price is going up. This behavior snowballs until the pool of susceptible people runs out. Not every crash is the result of a bubble, of course.

  55. Gravatar of maynardGkeynes maynardGkeynes
    11. April 2013 at 21:06

    Shiller has never claimed that CAPE will predict market swings over as short a period as you cite. He only says that the higher the ratio, the more likely it is that the market is overpriced, and is therefore likely to show below-average returns in the future. The actual time period is not specified, but given that the ratio itself looks back 10 years, 4 years of it being “wrong” would not seem to mean much.

  56. Gravatar of J.V. Dubois J.V. Dubois
    12. April 2013 at 01:03

    Andrew: “No. A bubble occurs when a significant number of people buy purely because the price is going up.”

    This is almost perfect definition of Veblen goods. People will buy these goods just because their price is rising, or because it has reached some level. So can we then claim that all veblen goods are bubble just waiting to be popped?

  57. Gravatar of Paul Andrews Paul Andrews
    12. April 2013 at 01:54

    JV:

    Veblen goods is a different concept. People buy Veblen goods because the price is high, not because it is rising. This pertains mainly to luxury items.

    The phenomenon whereby people buy something because its price is increasing can apply to any durable (non-consumption) asset.

  58. Gravatar of J.V. Dubois J.V. Dubois
    12. April 2013 at 03:27

    Paul: Now you are talking about a lot of stuff. Baseball cards, investment vines, some sports cars, art and many, many other things are being purchased because people think that their price may go up, because their price was rising etc.

    So by your definition they are inevitably all bubbles waiting to be popped. So if the price of Monet paintings was rising for past 100 years – and people are buying these paintings because they expect the price to be higher – by your definition a bubble waiting to be popped. It just somehow didn’t happen during 100 years.

  59. Gravatar of Tom Tom
    12. April 2013 at 03:40

    Like Andrew: “A bubble occurs when a significant number of people buy purely because the price is going up.”

    Because most of these speculators also trust themselves to sell, before a crash, it is not “irrational”. It is overconfidence, which leads to failure to escape the loss. In the case of house flippers, there was a large number who believed that the worst case was an end to price increases.

    Wrong beliefs about a specific subject are not quite the same as irrational — because all rational folk make assumptions first, then act upon them in a rational way.

    I don’t think the EMH requires people to avoid having wrong beliefs.

    It could easily be argued that to discuss market bubbles, in general, by using the very new and market untested example of limited fiat e-money was … not really very rational.

  60. Gravatar of ssumner ssumner
    12. April 2013 at 05:11

    Paul, Sure, anything is possible. But how does that relate to bubbles?

    Cthorm, Good comment.

    Geoff All time horizons.

    Dan, I’m a pragmatist. If the bubble theory has no testable implications then it’s obviously useless. In that case I’d rather stick with the EMH, which has proved very useful to me.

    Matt, I don’t agree, but even if you were right I can’t see how it has any bearing on this post. Your comment suggests that bubbles might exist. So what? How does the impact the argument that I made in the post?

    Nick, Good comment.

    Maynard, Sorry, but if someone has a model of the correct value of the stock market, I want them to be screaming BUY!!!! from the rooftops in March 2009. And he wasn’t. That doesn’t disprove his model but it’s a huge black eye. Someone should dig up all the times in the past 20 years when Shiller thought the market was underpriced and was a buy. I’d probably do a post if someone could put together a list.

  61. Gravatar of Paul Andrews Paul Andrews
    12. April 2013 at 05:40

    J.V.

    You said:

    “So by your definition they are inevitably all bubbles waiting to be popped. So if the price of Monet paintings was rising for past 100 years – and people are buying these paintings because they expect the price to be higher – by your definition a bubble waiting to be popped. It just somehow didn’t happen during 100 years.”

    I’m saying that a bubble can potentially occur in any asset market. I’m not saying they do occur for all assets, or that they occur all the time, or even that they occur often. I think they are more likely to occur when the money supply is expanding.

    Monet paintings don’t qualify because, like antiques, people appreciate their age, which of course increases over time.

  62. Gravatar of Paul Andrews Paul Andrews
    12. April 2013 at 05:51

    Scott,

    I said:

    “Do you agree that sometimes people buy assets purely because they are increasing in price? If you agree with that, do you agree that a positive feedback loop is sometimes created: people buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc. ? If you agree with that, do you agree that eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out?”

    You replied:

    “Paul, I agree with all three statements, but don’t think they imply the existence of bubbles. There are irrational people everywhere.”

    Paul:

    “You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.”

    Scott:

    “No I have not. I’ve said there are people who behave that way, not that market prices behave that way.”

    Paul:

    “The statements, which you have agreed are correct, refer to the interaction of people and prices, not only to people’s behavior. However, let me rephrase my question. You’ve previously agreed that there exists a phenomenon in human behavior which I termed a positive feedback episode. Would you agree it’s possible there have been positive feedback episodes in people’s behavior with respect to bitcoins?”

    Scott:

    “Sure, anything is possible. But how does that relate to bubbles?”

    I haven’t said that it relates to bubbles as you define them.

    Would you agree that positive feedback episodes would tend to last longer for larger populations?

  63. Gravatar of Saturos Saturos
    12. April 2013 at 05:55

    Tyler’s bitcoin defense: http://marginalrevolution.com/marginalrevolution/2013/04/the-devils-defense-of-bitcoin-volatility.html

    Thanks for the comments Scott.

  64. Gravatar of 123 123
    12. April 2013 at 05:58

    Scott: ” Bitcoin prices are super volatile.”

    Here is the weakest point of your argument. Shiller would say Bitcoin fundamentals are much less volatile than bitcoin prices. That’s why there is a bitcoin bubble sometimes, even though we really don’t know when.

  65. Gravatar of David N David N
    12. April 2013 at 09:17

    Was the tulip mania a bubble in 1637? If it wasn’t, then nothing is a bubble. If it was, then Bitcoin is another example. If you thought Bitcoin was a bubble at $2, the fact it briefly crossed $200 doesn’t make you wrong.

  66. Gravatar of myb6 myb6
    12. April 2013 at 09:40

    123, I think Sumner would respond “Cyprus.”

    I don’t think EMH doesn’t say the market is omniscient and perfect, rather that it includes all current information and is thus very hard to beat. Bitcoin is very information-poor right now, hence the volatility when new data is absorbed. Maybe Bitcoin’s fundamentals aren’t volatile, but the market’s understanding of Bitcoin’s fundamentals are. Unfortunately, your (and my) understanding of Bitcoin’s fundamentals is almost certainly worse than the market’s.

    That said, I think Sumner is overselling here. The market is still the sum of its actors, who can have systematic bias at least until enough actors learn and arbitrage away. There’s also just mania/depression, but those events are pretty rare:

    In 3/09 I didn’t have any capital nor income, which sucked both for my own lost once-in-a-generation opportunity and because it limits my proof to emails begging friends/relatives to buy equities and oil futures.

    I could back Sumner 100% on this if he allowed for a bubble every few decades.

  67. Gravatar of myb6 myb6
    12. April 2013 at 09:41

    “I don’t think EMH doesn’t say” -> “I don’t think EMH says”

  68. Gravatar of 123 123
    12. April 2013 at 11:36

    @myb6

    Cyprus does not work. For example, the reaction of Portuguese bonds was inconsistent with the reaction of bitcoin.

  69. Gravatar of Geoff Geoff
    12. April 2013 at 13:14

    Dr. Sumner:

    “Geoff All time horizons.”

    Do you think expecting zero excess risk adjusted profits at all time horizons is a good expectation, considering how there are always excess risk adjusted profits being earned on the basis of stipulating an expected volatility that is not unique and is subject to disagreements?

    Nobody can observe anything other than nominal profits, so while you are expecting zero risk adjusted profits, someone else can expect non-zero risk adjusted profits because they are using a different assumption of systematic or idiosyncratic expected volatility.

  70. Gravatar of Matt Waters Matt Waters
    12. April 2013 at 15:46

    “Matt, I don’t agree, but even if you were right I can’t see how it has any bearing on this post. Your comment suggests that bubbles might exist. So what? How does the impact the argument that I made in the post?”

    This is the issue I see with all the EMH posts. The posts seem to commingle different forms of the EMH and different conclusions. You said:

    “But the internet never forgets anything. And I’ll search and search and expose every phony “I told you so.” I can’t predict where bitcoin is going, but I can predict there will be many false “bubble” claims when it eventually crashes””and it will crash. The only question is whether it will crash from a price so far above the current price, that it’s still a good buy at $200. $160.

    And the EMH says that the answer to that question is; “God only knows.””

    This all makes perfect sense for the WEAK form of the EMH, and I absolutely agree with the weak form of the EMH. The weak form of the EMH allows:

    1. Prices to generally follow random-walk behavior.
    2. A complete inability to predict future price movements based purely on past price movements AND
    3. An asset to exhibit prices out of line with fundamental value, either future cash flows or true non-cash utility.

    Again, the “non-cash utility” part is important for looking at values of non-financial assets. Houses in Las Vegas did not lose their fundamental value as residences, but housing became a bubble as most of the activity was for future price appreciation, not for fundamental utility. Also notice that most of the house price decline in Las Vegas happened before NGDP for the country as a whole plunged.

    http://lasvegas4us.com/wordpress/wp-content/uploads/2010/05/case-shiller-Las-Vegas-home-index1.jpg

    So I couldn’t tell you when exactly LV house prices would peak or when it would crash (based on the weak-form EMH), but looking at the flipping activity and fundamental supply of houses vs. number of residences, the fact that LV house prices would crash soon is clear. This directly shows the semi-strong form of the EMH does not always apply, especially since most of these homes were ultimately bought with private funds through CDO’s.

    The fact the semi-strong EMH does not always apply is important because economic fundamentals and not future price movements determine the efficiency for allocation of capital. Whoever eventually funded the purchases of houses during the LV housing bubble saw a reduction of living standards, and whoever sold at the top saw an increase in living standards without adding value to the economy. Government action to mitigate this poor economic outcome may have more costs than benefits, but these poor economic outcomes have definitely existed and have levied true economic costs.

  71. Gravatar of ssumner ssumner
    12. April 2013 at 19:07

    Paul, No.

    Saturos, I’m afraid that went over my head.

    123, Nobody on the face of the Earth has a very good idea as to precisely what bitcoin “fundamentals” are, much less whether they are volatile or not.

    David, There’s a book out that debunks the tulip mania “bubble.”

    Geoff, Yes.

    Matt, I read a recent post that showed the so-called Wall Street “insiders,” who were experts on real estate, where just as fooled by the housing “bubble” as anyone else. Which makes me think “bubble” is a completely useless idea, with no practical applications.

    In any case, this post is not about the question of whether bitcoin is a bubble.

  72. Gravatar of Russ Anderson Russ Anderson
    12. April 2013 at 20:57

    Bitcoin is just another ponzi scheme. Is a ponzi scheme a bubble or an example of an efficient market? Calling it the former is an insult to true bubbles and calling it the latter is just silly.

  73. Gravatar of Paul Andrews Paul Andrews
    12. April 2013 at 21:24

    Scott,

    I said:

    “Do you agree that sometimes people buy assets purely because they are increasing in price? If you agree with that, do you agree that a positive feedback loop is sometimes created: people buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc. ? If you agree with that, do you agree that eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out?”

    You replied:

    “Paul, I agree with all three statements, but don’t think they imply the existence of bubbles. There are irrational people everywhere.”

    Paul:

    “You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.”

    Scott:

    “No I have not. I’ve said there are people who behave that way, not that market prices behave that way.”

    Paul:

    “The statements, which you have agreed are correct, refer to the interaction of people and prices, not only to people’s behavior. However, let me rephrase my question. You’ve previously agreed that there exists a phenomenon in human behavior which I termed a positive feedback episode. Would you agree it’s possible there have been positive feedback episodes in people’s behavior with respect to bitcoins?”

    Scott:

    “Sure, anything is possible. But how does that relate to bubbles?”

    Paul:

    “I haven’t said that it relates to bubbles as you define them. Would you agree that positive feedback episodes would tend to last longer for larger populations?”

    Scott:

    “No”

    OK, just to clarify – you believe that a market known to 1000 people will have positive feedback episodes that are no longer than those occurring in the same market if it was known to 100 million people. Is that correct?

  74. Gravatar of 123 123
    13. April 2013 at 00:47

    Scott, one of the fundamentals that is actually well correlated with the bitcoin prices is the media coverage of bitcoin, indicating a bubble type information processing.

  75. Gravatar of ssumner ssumner
    13. April 2013 at 04:44

    Paul, If a positive feedback effect exists, it’s not obvious to me why it would depend on the size of the market, except perhaps if the market got really small, say a handful of people.

    123, It’s hard to separate out cause and effect in that correlation.

  76. Gravatar of Paul Andrews Paul Andrews
    13. April 2013 at 17:03

    Scott,

    I said:

    “Do you agree that sometimes people buy assets purely because they are increasing in price? If you agree with that, do you agree that a positive feedback loop is sometimes created: people buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc. ? If you agree with that, do you agree that eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out?”

    You replied:

    “Paul, I agree with all three statements, but don’t think they imply the existence of bubbles. There are irrational people everywhere.”

    Paul:

    “You’ve previously agreed that there exists a phenomenon in asset pricing which I termed a positive feedback episode.”

    Scott:

    “No I have not. I’ve said there are people who behave that way, not that market prices behave that way.”

    Paul:

    “The statements, which you have agreed are correct, refer to the interaction of people and prices, not only to people’s behavior. However, let me rephrase my question. You’ve previously agreed that there exists a phenomenon in human behavior which I termed a positive feedback episode. Would you agree it’s possible there have been positive feedback episodes in people’s behavior with respect to bitcoins?”

    Scott:

    “Sure, anything is possible. But how does that relate to bubbles?”

    Paul:

    “I haven’t said that it relates to bubbles as you define them. Would you agree that positive feedback episodes would tend to last longer for larger populations?”

    Scott:

    “No”

    Paul:

    “OK, just to clarify – you believe that a market known to 1000 people will have positive feedback episodes that are no longer than those occurring in the same market if it was known to 100 million people. Is that correct?”

    Scott:

    “Paul, If a positive feedback effect exists, it’s not obvious to me why it would depend on the size of the market, except perhaps if the market got really small, say a handful of people.”

    I think it’s clear from the description:

    People buy because price is going up -> causes price to increase further -> more people buy because price is still going up -> etc. Eventually the stock of such people who are liable to the described behavior with respect to the particular asset will run out.

    The more people in the market, the longer the last event, the depletion of the stock of susceptible people, will take to occur.

    Can you explain how it could be otherwise?

  77. Gravatar of dtoh dtoh
    13. April 2013 at 18:57

    Paul,
    You said;
    The more people in the market, the longer the last event, the depletion of the stock of susceptible people, will take to occur.

    Can you explain how it could be otherwise?”

    Maybe susceptible people are not binary, i.e. if the price of a good goes up enough susceptible people become skeptical people.

    Also maybe the supply of available goods is proportional to the population size so you price movement is the same as in a smaller population and/or transaction velocity is higher.

  78. Gravatar of Paul Andrews Paul Andrews
    14. April 2013 at 00:27

    dtoh,

    “maybe the supply of available goods is proportional to the population size so you price movement is the same as in a smaller population and/or transaction velocity is higher”

    I am comparing two different populations, with all else equal. I want to examine the effect of a change in population size only. The case where the asset base also changes is not the case I am examining.

    “Maybe susceptible people are not binary, i.e. if the price of a good goes up enough susceptible people become skeptical people.”

    Yes that is a good point – this may place a limit on the length of the feedback event. I would still expect events to be longer in larger populations, but not in linear proportion to the population size. And there may be an upper limit at which higher populations no longer makes a difference to the event length. I think that limit would be far higher than a handful however. Would you agree?

  79. Gravatar of dtoh dtoh
    14. April 2013 at 02:47

    Paul,

    I think it’s possible, but early in my career someone described the market to me as being like the ocean on a warm summer day. A few people venture into the water early on, eventually everyone follows them in and they are all having a great time until someone yells “shark” and they everyone races for the beach. IME, most people are like this, but I think there are also quite a few really greedy, downright stupid people as well.

  80. Gravatar of Jon Jon
    21. April 2013 at 21:03

    Money is the bubble that doesn’t pop. The only question is whether bitcoin becomes money. There’s a non-zero chance it will. And when that happens, what people are calling a bubble now will later be referred to as monetization.

  81. Gravatar of Bitcoin(sanity)? | Rättvist och balanserat Bitcoin(sanity)? | Rättvist och balanserat
    23. April 2013 at 05:21

    […] var offer för sin egen framgÃ¥ng. Andra menade att hela Bitcoin-grejen bara är en bubbla och ytterligare andra att det absolut inte var en bubbla, men att den här typen av stora upp- och nedgÃ¥ngar var och […]

  82. Gravatar of Skepticlawyer » Bubble trouble: all information is not equal Skepticlawyer » Bubble trouble: all information is not equal
    28. May 2013 at 15:33

    […] post is partly prompted by this comment and this paper (pdf) (via) on the US housing price bubbles and busts and (greatly) extends this […]

  83. Gravatar of As Gold Goes So (Doesn't) Go the World | Last Men and OverMen As Gold Goes So (Doesn't) Go the World | Last Men and OverMen
    18. April 2017 at 03:05

    […]       http://www.themoneyillusion.com/?p=20691 […]

Leave a Reply