If you spend any time watching "technical analysts"1 on the market watch shows--and who can resist?--you'll notice they spend a fair amount of time talking about "capitulation". This is what they call it when everyone decides the market isn't going up any time soon and sells out in the hopes of sheltering in safer investments. It is supposed to be a good time to buy because once everyone has capitulated, the market starts going up again.
I don't hold much faith in technical analysis. But looking at all the "just reduced!" ads in the Washington DC real estate classifieds, I wonder if homeowners aren't finally throwing the towel and trying to sell at any price.
1 A.k.a. "chartists" a.k.a. cranks
I think it would be a lesser effect in housing. Houses come with bigger fixed costs in transactions. They also come with better incentives to sell high, with the capital gains exemption for family homes. People are more likely to suck it up and just live in it until the market goes up again. Developers probably are more likely to engage in "capitulation".
On the other hand, nobody has to sell their stock because they're new job is in Memphis.
While we're on the subject (sort of) of efficient markets and housing, could you explain how EMH squares with bubbles (e.g. the tech bubble of the early aughts and the housing bubble of the, uh, yesterday)?
If market prices always reflect all known information, how can bubbles arise?
In the case of Adams County, Colorado, the future is already here. The county seat's weekly fishwrapper usually contains 15-30 public notices of foreclosure sale.
You're confusing efficiency with getting the price right. Markets may sometimes overshoot, sometimes undershoot, but that's not evidence of systemic bias; it's evidence of error. The only way you can make money on an economic basis is if there is systemic bias.
Colin-
The trouble is that markets also reflect everyone's (rational or irrational) beliefs and opinions about the market, plus their reactions to what they think other people think, and so on -- a situation where a cascade effect can easily result. One might also argue that bubbles tend to arise when actual information is distorted by signals that aren't reponsive to market info (governmentally-controlled interest rates, mortgage rates, etc.).
I don't hold much faith in technical analysis. But looking at all the "just reduced!" ads in the Washington DC real estate classifieds, I wonder if homeowners aren't finally throwing the towel and trying to sell at any price.
I'm not an expert on existing homes, but as an employee of a DC-area homebuilder I can safely say, yes, we are trying to sell at any price. There are a few communities at which we're selling houses below cost just to get them off our inventory. And at the rest of our communities we're selling at margins far below the threshold we have to greenlight new projects.
Megan,
I am shocked that (so far) there has not been a deluge of "technical analysts" posting to defend their art. From my experience, they can be quite defensive.
I can't help but think though that if the market as a whole is down, then stocks are at least a bit better of a deal than before. Of course timing is everything. The markets may fall some more before making the long grind back up.
Charting is a bit like NYC neighborhoods in a coming snowstorm.
Growing up, the mere report of a storm would cause everyone, including my parents, to hit the store, because, apparently, the storm would prevent you buying food for what? Maybe a day or two. But people loaded up on eggs, milk and bread as though the end was nigh. The next day or two, as you shoveled out, you realized, "Hey look, the main streets are clean and Waldbaums is open". And what the hell do we do with all the extra bread? French toast starts turning up for dessert.
The supply of goods is fine, but an event creates a false shortage based on people's irrationality. Charting works in similar fashion. If enough people with money at their command "believe" in the same price movement theory, then the market behavior will begin to reflect that.
And there are enought fundamental analysts who don't believe in charting, but believe others believe, so they adopt their trading accordingly to reflect the passions of others. They do that because they have learned that being right in terms of actual money is better than being right intellectually.
And you see such irrationality, where an Apple stock can drop from $195 to $130 inside of one month (January), and without major news that should cause such a change.
So you sit there, snow storm coming, and with plenty of bread, but out there the Johnsons are getting into their car to run to the store. You think, "Hmmm, maybe they are right." And off you go, getting the bread you don't need, that is not truly scarce, because the storm is coming. And sure enough, at the store, Wonderbread is scarce, shelves bare.
"Mental note to self," you say, "next storm get to store early". The rational mental note is, "Next storm, don't need bread".
NOTHING about the markets is rational or efficient. (Efficiency being like nirvana, infinity, and world peace).
Technical Analysis is the lottery for rich folks.
"...could you explain how EMH squares with bubbles?"-Colin Fraizer
"You're confusing efficiency with getting the price right."-MM
Colin, Megan is not correct. Eugene Fama defined market efficiency thusly:
But by definition a bubble occurs when the actual price of assets have strayed very far from their intrinsic values. It's not my intention to try to prove that Megan knows nothing about economics. Clearly she knows a fair amount. But she is confused about what market efficiency implies.
So she agrees with both Fama (who believes in market efficiency) and Thaler (who doesn't). I've pointed this contradiction out several times, a have others like MEH, but she goes on blithely repeating her old mantras:
"Markets are efficient" and "Markets can stay irrational longer than you can stay solvent," unwilling or unable to see that these two statements are not compatible.
Fama, unlike McArdle, is consistent. He claims that the price of the Nasdaq in the late 1990's was rational.
Greg Mankiw says this:
Mankiw goes on to explain that other economists have had different views. Keynes, for example, "suggested that asset markets are driven by the 'animal spirits' of investors--irrational waves of optimism and pessimism."
So, again, Colin is right and Megan is wrong. One can believe with Fama that markets are efficient, or with Keynes that they are sometimes quite inefficient, but one cannot believe both (at least not rationally).
Megan's dogmatic insistence on this matter is really not to her credit.
rwe and Colin,
If I understand EMH and bubbles correctly, and I may not, they are entirely compatible. Imagine that there is a new, unkown beetle spreading across the nation, that will destroy all houses in 10 years, but we don't know that yet. That beetle means that all sorts of housing we think will last for many decades, won't, and thus is much less valuable than we think. Thus, our current housing prices are in a "bubble." Nonetheless, because we don't know about this beetle yet, the EMH says that housing prices should be where they would "rationally" be if there were no beetle. But you can't make tons of money by shorting houses to "post-beetle" levels, because you don't know about the beetle yet.
That's basically what happens in real "bubbles." EMH is a guess that prices are about the best we can estimate. Bubbles are declared true after the fact, when uncertainty reduces. (E.g., You can be convinced that home-price-to-rent ratio's have to return to more historical levels, but still be very uncertain about when the market will move, because you don't know how long this meta-stable "group irrationality" will last.) Similarly, as shown by the fact that lots of people don't really know how bad all the debt is in the various CDO's, we still don't know exactly what the true course of the housing economy should be, so there's no "guaranteed money" betting in either direction on the Case-Shiller Index futures. *Someone* will probably make a lot of money on those, but EMH translates into the situation that there is no market position that guarantees you a risk-free bet to make money on the future movement of housing prices, because you can't predict important news that hasn't happened yet.
The point is that EMH *doesn't* mean that that Truth (as it really is out there) is reflected in the price. It is hypothesis that the current price really is about he best estimate of our collective understanding of that Truth. Systemic biases in our collective knowlege are not priced in, but we don't know that because we don't know our system biases (if we did--we'd remove them).
But we all think we *know* that houses were overpriced, and stocks before that in 2000, so we all think that clearly there was a bubble, and thus the price of houses (and stocks before) were clearly too high, and thus EMH must be wrong. I have two ideas why we may be feeling this discrepency.
a) Hindsight bias
b) The "H" in EMH stands for "Hypothesis," not Hard-fast-proven-certainty.
RWE, neither of your quotes exactly proves your case. Both of those quotes make great use of the word "information" whereas you seem to be trying to get to the underlying truth of the situation. (In fact, the EMH is defined as the idea that financial markets are "informationally efficient.") A bubble does not imply that a market was inefficient. It could also mean that the information changed. The EMH does not claim that financial markets are omniscient and can see into the future.
I do, however, agree that the sentence "markets can stay irrational longer than you can stay solvent" does appear to be at variance with the efficient market hypothesis and I'm not at all certain that the two can be reconciled. If markets are irrational, then they are not informationally efficient, pretty much by definition.
As for the truth or falsity of the EMH, the tech bubble and the housing bubble are easily encompassed by the theory; the information clearly did change. The crash of 1987 is much harder to figure out and is a much more serious challenge to the theory.
Though the phrase has a nice rhetorical appeal, "markets can stay irrational longer than you can remain solvent" really has less meaning than it appears. Really, it only means that you can lose your money by being wrong. Of course, such people like to think everyone else was irrational. It helps the self-esteem.
Yancey, the phrase means more than that. It is a quote from Keynes, and he really meant that investors sometimes do foolish things. Keynes believed that they are driven by "animal spirits" as much as rational calculation and that they sometimes become too optimistic, driving prices up to irrational levels.
He was a firm opponent of market efficiency (though of course he came before the development of the modern EMH). Hence he believed there were "bubbles" at times. Fama rejects this notion of "bubbles" arguing, more or less as Tom and Andrew do above that prices that look unreasonable ex post in truth were quite reasonable ex ante.
Megan seems to want to believe with Fama that markets are efficient and with Keynes that there are bubbles driven by irrational behavior. But these two views are not compatible.
Again, it's not my intention to denigrate Megan. When she writes on free trade and health care she does good work explaining the nature of markets to those of her readers who don't fully understand them. But she is badly mistaken about the EMH. And she is unintentionally misleading her readers.
in an efficient market at any point in time the actual price of a security will be a good estimate of its intrinsic value
rwe,
You are misunderstanding this statement.
"A good estimate" doesn't mean "an extimate with less than 10% error", as it does in msot contexts; it means "an estimate that can't be systematically improved upon."
Clarification: my question was just that, a question. I don't feel knowledgeable enough to start an argument about these things.
FWIW, my concept of EMH matches (roughly) the Fama quote, which is what has led to my confusion. If I were industrious, I'd read some recent literature and see what EMH proponents say about the tech bubble. Since I'm not, I was hoping Ms. McA would summarize her knowledge on this blog.
Sam, I'm not misunderstanding it at all. I understand that Fama is arguing that an efficient market is one in which the price cannot be systematically improved upon.
I'm arguing that Megan McArdle's view is not consistent with Fama's view, and indeed that her view is not even internally consistent. She believes that the market is efficient and that there can be bubbles driven by irrationality. Once again, those two views are not compatible.
Fama argues that there is really no such thing as a bubble. Prices can turn out to have been incorrect ex post but were very good estimates ex ante. That's a respectable view (though it isn't my view).
But McArdle concedes there are bubbles. When asked why, given that there are bubbles, one couldn't (systematically) make money by shorting the overpriced assets, she always responds "Markets cans stay irrational longer than you can stay solvent." As I have tried to explain, that is not something Fama would ever say. It is incompatible with the EMH.
It is a quote from Keynes and reflects a radically different understanding of how markets operate--one more in line with modern behavioral economics.
I know it's hard for some people here to accept that McArdle is fallible, but she is. She's wrong about this, just as she was wrong in saying that a decline in the dollar has no effect on oil prices (in dollars).
For Colin or anyone else who is interested, I was trying to post an interesting link, but it keeps getting caught in the spam filter. So just google "Fama Thaler As Two Ecomists Debate Markets The Tide Shifts" and you'll find a Wall Street Journal article that describes the academic debate about this quite engagingly.
What's the difference between technical analysts, chartists and other people who try to predict future prices solely from historical prices? Like anybody who uses Black-Scholes? It seems to me that there /is/ no difference (except the degree of respectability afforded to them.) Is this correct?
How about an exercise to help us laymen. We will be using a commodity market for our exercise, and the commodity will be sealed envelopes with US currency inside - congress develops a sense of humor and decides this is the form of the new stimulus package. Everybody gets one, and is free to trade it. By law, every envelope has exactly the same amount of money inside, but it is unknown to the general public. While some people know, nobody knows who those people are. All envelopes will be opened 1 month after trading begins.
The good has a definite value, but people don't know what it is. What behaviours would indicate efficient or inefficient markets?
All sorts of theories would emerge:
The envelopes hold in aggregate what the stimulus bill would have spent - which bill, the House or Senate? A theory that the envelopes hold almost nothing, but will work as a stimulus due to speculation gets play, and the price drops. A counter-theory that the price drop is a ploy gets credence, and the price jumps. Speculators start trying to figure out what will cause the next irrational price swings and try to plan ahead. Analysts try to statistically predict the actions of the speculators. Etc.
The trading stops. The envelopes are opened. Some people were right and others wrong.
What reflects efficiency and inefficiency?
By the way, Colin, the wikipedia article on the efficient market hypothesis is very good. So you should look there also.
Sam, I'm not misunderstanding it at all. I understand that Fama is arguing that an efficient market is one in which the price cannot be systematically improved upon.
I'm arguing that Megan McArdle's view is not consistent with Fama's view, and indeed that her view is not even internally consistent. She believes that the market is efficient and that there can be bubbles driven by irrationality. Once again, those two views are not compatible.
I believe the reconciliation comes from a belief that bubbles driven by irrationality can not be improved upon. Just consider the housing market. I remember people talking about irrational increases here in the DC area back in the mid 90s. Even with recent declines, those purchases in the mid 90s were decent investments. Because those experts were wrong, there wasn't a bubble and the public wasn't irrational. Because some experts were right in 2003, there was a bubble. What better means of evaluating when experts are right or wrong is there than the market? How do you become more efficient?
"What better means of evaluating when experts are right or wrong is there than the market? How do you become more efficient?"-Njorl
Well these are the right questions to ask, Njorl. Housing is trickier, but in the case of stocks one could use "fundamental anlysis," making estimates of the value of the assets of a company and its future earnings stream and using a discount rate to determine an approximate "intrinsic value." That is Warren Buffett's approach.
Now, those who believe in the EMH (in its strong or semi-strong form) say that Buffett has just been lucky--that he has no superior insight into the values of the securities he purchases.
Buffett contends that he has not just been lucky. he believes that, by studying the financial statements of the Washington Post, for example, and assessing qualitative factors like the quality of management, he can sometimes get a better sense of the true intrinsic value than the market price. The fact that investors sometimes behave irrationally, he argues, creates discrepancies between market price and intrinsic value that shrewd investors like him can exploit.
So that's the debate. I tend to agree with Buffett, but my main point here has been to point out that Megan McArdle's view of this is fundamentally incoherent, not to argue against the EMH.
W. Darlington, Black-Scholes derives futures prices from historical prices? The only thing unobservable is volatility and sometimes volatility is estimated from historical price movements, but I think there's a pretty large gap between saying that it's derived from historical prices. (And volatility is not always estimated from historical price movements.)
As for the actual debate, both sides are probably overstating their case. Markets are pretty clearly weak-form efficient almost all the time, but there probably are some times when irrational money overwhelms the rational money. Take a different kind of market: professional football gambling. During the regular season, the lines are almost always perfectly rational. However, in the Super Bowl, the amount of money gambled by uninformed (and irrational) participants can overwhelm the amount of money that informed participants have access to, leading to lines that almost all educated observers will agree are wrong.
EMH needs basically only "no unexploited profit opportunities" in order to work. There are almost certainly occasional cases when the amount of money in irrational hands simply overwhelms the amount of capital rational participants can access to correct it. The behavioral finance people make too much of this insight of theirs just as the efficient markets people make too much of their insight that markets are normally efficient.
Buffett contends that he has not just been lucky. he believes that, by studying the financial statements of the Washington Post, for example, and assessing qualitative factors like the quality of management, he can sometimes get a better sense of the true intrinsic value than the market price.
Does this contradict EMH? If Buffet were restricting himself to analyzing statements, I'd answer yes. However, if he is evaluating qualitative factors like management, he is using his own personal business acumen to extract information that is not readily available to the public. Only after Buffet acts does the information become public. The purchase by Buffet of a stock is in essence the publication of the datum that a skilled management evaluater has judged the personel to be exceptional, much like a bond rater bumping an offering up to AAA status. Buffet is like the bond rater, without the proscription on trading on his information. He often goes even further, alterring the personel of companies in which he buys a controlling interest. That goes a bit beyond functional analysis.
Maybe I've crossed the line into begging the question here - defining everything that violates EMH as privelidged information.
Njorl, it does contradict the EMH. Buffett contends that he does not acti upon any data that are not publicly available. He has no "inside information" that would give him an edge.
As you say, he does have an advantage, in a sense, in that his superior acumen allows him to make better judgements than the markets about sucurities. But the the EMH (in its usual semi-strong form) maintains that this is impossible. Buffett's sort of "fundamental analysis" cannot yield any insight that is not already contained in the market price.
As I said, I agree with Buffett. I think Andrew is pretty much correct above. And so was Milton Friedman:
And so was Ben Graham (Buffet's mentor):
What they are saying is that markets are generally pretty efficient, and that mispricing is unlikely to persist for a long time, but that mispricing does occur. And sometimes there is a kind of "irrational exuberance" that drives prices in general too high, until more sober analysis prevails and brings things back into line.
And that's my last word on the subject.