Models and the Efficient Market Hypothesis

Apparently I’m posturing when I defend the efficient markets hypothesis (EMH). I’d like to posture some more, and at length.

First, I should note that I don’t claim to be an expert in the EMH. But I’ve been following the debate closely, and I don’t think the critics have actually rebutted, and frequently have not even addressed, the basic defenses of the EMH advocates–often, I think, they’ve been attacking strawmen. Also, I do know something about models, and I know that attacking the premises of a model indicates misunderstanding what a model is for, and most of the critiques of the EMH attack the premises.

Fair warning: There’s a lot to say here, so this is a long post, and perhaps a bit technical in parts. I’ll try to avoid jargon and be clear, but I can’t be brief. (I believe it was Einstein who said, “if you can’t explain it simply, you don’t understand it well enough.” I won’t deny that might apply here.)

Assumptions Don’t Always Make an Ass of You and Me
To the layperson, attacking the assumptions, of a model seems fair. If the model has false assumptions, how can it be valid, right? But this is incorrect in two ways.

1: Garbage in is not always garbage out
First, if a model is used to predict/explain how the real world works, what really matters is a model’s accuracy, its output rather than its input. For example, climate models don’t accurately handle rainfall in their assumptions, which has been a big point for critics. But if it turns out that the models accurately predict warming trends, it doesn’t matter if their assumptions about rainfall are accurate, or even if they are totally absent: what matters is that the model works.

One of the most notable statements of this comes from economist Milton Friedman.

In so far as a theory can be said to have “assumptions” at all, and in so far as their “realism” can be judged independently of the validity of predictions, the relation between the significance of a theory and the “realism” of its “assumptions” is almost the opposite of that suggested by the view under criticism. Truly important and significant hypotheses will be found to have “assumptions” that are wildly inaccurate descriptive representations of reality, and, in general, the more significant the theory, the more unrealistic the assumptions (in this sense).

The reason is simple. A hypothesis is important if it “explains” much by little, that is, if it abstracts the common and crucial elements from the mass of complex and detailed circumstances surrounding the phenomena to be explained and permits valid predictions on the basis of them alone. To be important, therefore, a hypothesis must be descriptively false in its assumptions;

Not everyone agrees, naturally, but I do. I always think of some of the assumptions of physics, such as a universe without friction. We all know that objects in motion don’t really stay in motion forever, that they are in fact affected by real-world forces, but nobody says (I hope), “Newton’s first law is sooooooo stupid.”

Of course if the model doesn’t work, then it’s appropriate to critique the assumptions, but not on the grounds that they’re false, just on the grounds that they don’t produce the expected results. So a critique that “the Efficient Markets Hypothesis can’t be right because it assumes people are rational, when in reality they aren’t,” is a bad argument. The critique that “if the Efficient Markets Hypothesis doesn’t make accurate predictions; that might be because it assumes people are rational when in reality they aren’t” is a reasonable (but not necessarily accurate, even if people are in fact not rational) argument.

Presumably, very unrealistic assumptions are more likely to produce bad results than more realistic assumptions. But, odd as it may sound, we can’t really know that a priori for any particular case. So, as Scott Sumner wrote:

The question is not whether the EMH is “true,” how could it be? Almost no economic model is precisely true. The question is whether it is useful.

Or as Noah Smith says of the Dynamic-Stochastic General Equilibrium model,

A DSGE model is just a tool. It’s a gizmo, like a fork lift or a lithium-ion battery.

Now you may think the model isn’t useful, that it’s a bad tool, but Sumner thinks it’s useful. That is, he doesn’t think the anti-EMH models, that would suggest you can reliably beat the market, have ever been proven right.

2: The big “if”–testing assumptions
The second reason attacking the premises is invalid is because premises are often stated on a “if” basis. “If” these premises hold, we would expect this particular result. The model then is accurate insofar as those premises hold, and no more. And then we can examine how much deviation from the model occurs when this or that assumption doesn’t hold.

For example, in game theory the model of the prisoner’s dilemma assumes the only values for players are internal to the game, what they can win or lose by cooperating or defecting. In reality, people hold values about cooperation and defection that are external to the game–they come into the social science laboratory with those values. So the prediction of game theory for one-shot games played with a stranger is 100% defection. It turns out that we don’t get 100% defection–some people defect. (We say they act irrationally, but in fact it means they have a set of values which outweigh the gains they could make by defecting in the game.) But not that many people defect. If I recall correctly, 70% or more of people will defect in one-shot games. So we know the assumption doesn’t (totally) hold, but we can see the degree to which it’s not holding affects the accuracy of the prediction.

Now I hear people thinking loudly, “the premises of the EMH don’t hold!” The point is, that doesn’t disprove the model–it just demonstrates how reality varies from the model when the premises don’t hold.

The Model of Efficient Markets
So what is the Efficient Market Hypothesis? As I noted above, I don’t claim to be an expert in it. It’s a macro theory, and I’ve never made any bones about the fact that I’m not particularly strong in macro. I don’t know all the subtle details and variations. James K and J R could, I am sure, both do a much better job of explaining it. But then explaining it thoroughly would probably be a long post in itself. So I’m going to just explain it briefly.

Eugene Fama describes the hypothesis succinctly:

A market in which prices always “fully reflect” available is called “efficient.”

That’s a bit vague, as Fama immediately notes, but later he gets more specific.

First, it is easy to determine sufficient conditions for capital market efficiency. For example, consider a market in which (i) there are no transaction costs in trading securities, (i) all available information is costlessly available to all market participants, and (iii) all agree on the implications of current information for the current price and distributions of future prices of each security. In such a market, the current price of a security obviously “fully reflects” all available information.

There’s at least one more unstated assumption: rationality. While there are endless debates over how to properly define that concept, for purposes here I think it can fairly be said to mean that people understand how that information can benefit them and will in fact use it in that way.

Now, remember what I said above about unrealistic assumptions like a frictionless universe? Here in (i) and (ii) Fama is talking about a frictionless economy. If it’s useful for studying physics, why not economics? But let’s be sure that we don’t make the mistake of thinking Fama isn’t aware of how unrealistic these assumptions are:

But a frictionless market in which all information is freely available and investors agree on its implication is, of course, not descriptive of markets met in practice. Fortunately, these conditions are sufficient for market efficiency, but not necessary. For example, as long as transactors take account of all available information, even large transaction costs that inhibit the flow of transactions do not in themselves imply that when transactions do take place, prices will not “fully reflect” available information. Similarly (and speaking, as above, somewhat loosely), the market may be efficient if “sufficient numbers” of investors have ready access to available information. And disagreement among investors about the implications of given information does not in itself imply market inefficiency unless there are investors who can consistently make better evaluations of available information than are implicit in market prices.

So even if these assumptions are violated–even if they are unrealistic and don’t hold in the real world–that doesn’t necessarily mean the market won’t be efficient.

But it also does not mean the market will be efficient, and Fama makes a crucially important point here.

But though transaction costs, information that is not freely available to all investors, and disagreement among investors about the implications of given information are not necessarily sources of market inefficiency, they are potential sources. And all three exist to some extent in real world markets. Measuring their effects on the process of price formation is, of course, the major goal of empirical work in this area [emphasis added–JH].

I want to emphasize and clarify this point, keeping in mind what I said above about how we use assumptions in models. The research program of the EMH is not to prove that markets are efficient. Please allow me to say that again, with emphasis.

The research program of the EMH is not to prove that markets are efficient.

So when someone says, “the model’s stupid, people don’t have all the information and markets aren’t efficient,” that is not a refutation, nor even a critique, of the EMH. The research program is to figure out how those real-world diversions from the assumption of the ideal perfect market shape the prices of stocks.

In other words, Fama–the god of the EMH–has argued that real-world market prices are not necessarily “correct” in the way efficient market prices would be.

If you didn’t get that, please stop right now and re-read the last 6 paragraphs. If you didn’t know that the Efficient Markets Hypothesis does not predict that markets will be efficient, then you have not had the hypothesis properly explained to you.

What the EMH Predicts
The EMH, at its core, is a theory about price-formation in markets, not about grand-scale economic outcomes. Do some people use it differently? Oh, sure, you can find examples with ease. But, friends, I’m sticking to Eugene Fama here, the guy who’s most identified with the theory (although its antecedents long predate him) and who won a Nobel prize for it (delightfully, in the same year his critic Robert Shiller also won, for theories very much in contradiction to the EMH).

1: Correct Prices
The EMH does not say market prices will always be “correct” in the sense of reflecting a truly efficient market.

The EMH predicts “incorrect” prices may–not definitively will, but may–occur when the assumptions do not hold. One of those cases occurs when there is private information. The EMH predicts that market prices will reflect this information, but does not say that those prices will be efficient prices.

People often point to the opacity of mortgage-backed securities as a refutation of the EMH. But remember that the pure form of the model assumes perfect information–it’s an assumption, so it cannot be a prediction. That’s just a logical necessity–no model can in any coherent way predict its own assumptions. So what does the opacity–the lack of information–mean about the EMH? It means it would predict that those prices would be inefficent.

And let’s not pretend that kind of opacity is normal. We were all aghast at just how opaque those securities were. They stood out because they were so abnormal, because they don’t actually represent the normal market.

But what about J. P. Morgan profiting off its realization that its MBSs were worthless by selling them as though they were more highly valued? The EMH would predict that the prices would reflect that–as everyone is complaining actually happened–but it would not predict that price to be efficient.

2. Beating the Market
The EMH does predict that if enough information is widespread enough, an investor cannot consistently outperform the market. Remember the old adage that knowledge is power–if everyone has the same knowledge you do, you cannot use that knowledge to gain an edge over them. In the same way, if others have the same knowledge about stocks as you have, you cannot consistently make better choices than they do (unless, perhaps, you are far more rational than nearly all of them).

This prediction has been tested repeatedly, and the evidence suggests it’s accurate. In most studies, throwing darts at stock picks performs as well or better than actually picking stocks. (I’m amused by people I know on the left who scoff at the EMH, but snarkily critique the stock market by pointing out that poo-flinging monkeys can do as well as Wall Street traders. Well, I hate to destroy their illusions, but Fama’s going to happily agree with them.)

What about hedge funds that claim to outperform the market? They emphasize the years they have good returns, but across time they just don’t.

What about Warren Buffet? As a case study, he seems to be the exception that disproves the rule, doesn’t he? But it’s always good to step back from a single-case and look at the world statistically, at least when we have a large enough n. And when it comes to investors, we’ve got a damn large n. And what would we expect to be the distribution, purely from random chance, of such a large n? We’d expect to see most people clustered near the middle, with some outliers at each end. That is, even if all investors were poo-flinging monkeys (and who’s to say they aren’t?), over a lifetime some would wildly underperform the average, and others would “impressively” outperform the average.

John Quiggin has what seems to me a completely unpersuasive reply to this argument.

So supporters of the efficient markets hypothesis have sought a redefinition that would make it invulnerable to refutation. Their central argument is one that has already been discussed – if it is possible to diagnose the existence of a bubble, then it is possible to make arbitrarily large profits betting against it. And if someone like Warren Buffett has in fact done this, that can be put down to luck. Only if everybody can make money betting against the market can the EMH be wrong. But of course, it’s impossible for everyone to bet against the market – the market is just the aggregate of bets.

But, no, that’s just wrong, badly wrong. Wrong in the way I would expect any economist to grasp because they’ve all had methods and stats courses. It doesn’t take “everybody” beating the market to disprove EMH’s claim that you can’t consistently beat the market. It only takes a percentage of them that is larger than is explainable by chance. I’m not sure what that number would be, but operating just off the idea that researchers commonly reject conclusions if there is a greater than 1 in 20 chance of them being errors, I’d suggest that evidence that more than 5% of investors beat the market regularly over time would be a pretty hard blow to the “random walk” theory of the market. Not to say nobody’s tried to measure that, but I’ve yet to see anyone demonstrate what percentage of investors do so, and if it’s greater than 5%, I’d be really interested to see it.

3. Bubbles and Busts
Does the EMH predict there can’t be bubbles or busts? That’s what some critics believe, but it doesn’t really say that. Of course I see lots of “explanations” of the EMH on the web that either get this flatly wrong or don’t make it clear (including, for example, investopedia). But it seems to me that the model incorporates the ideas of bubbles and busts–they’re part of the random walk of the market–it just says that you can’t reliably predict them. If you could, you could always beat the market by buying in early when you predict a bubble, then getting out in a timely manner and shorting stocks in time for the bust.

But didn’t lots of people predict the bust of 2008? Sure (although not nearly as many as predicted it retrospectively). But does that prove anything? Well, not really. First, if anyone could really predict bubbles regularly, they could get mighty damned wealthy doing so. How many people claiming a bust was inevitable in 2008 were so confident in their prediction that they went big on investing against the market? Not a whole lot, so far as I have heard. So while I would agree there was growing uneasiness among a lot of people, I don’t think there were really that many that truly were absolutely confident it was a bubble. And I’m not inclined to believe after-the-fact assertions, what with cognitive biases and the problems of memory.

Second, predicting “a” bubble is no more proof of the ability to predict them generally than is a prediction of an election. There’s a whole cottage industry of academics <wasting their time building models to predict presidential elections. As far as I can tell, all of them can predict–or “predict”–some elections, but none of them can do so repeatedly. A couple of weeks ago my colleage, who’s a fairly big booster of Nate Silver, slyly asked me, “if the Republicans win the Senate, will that prove Nate Silver right?” It was a joke–we both knew the answer. And the same dynamic that holds for predicting an election holds for predicting a bubble: doing it once, even twice, doesn’t actually prove a thing about your ability to do so regularly.

Bubbles, after all, are only truly identifiable retrospectively (if even then). There are booms that look a lot like bubbles but never bust. Mathematician Andrew Odlyzko demonstrates one here. Sumner suggests another likely one here. Sumner also comments that

Just when I was starting to warm up to Shiller’s model, he missed the huge bull market of 2009-11.

. OK, so one miss doesn’t prove EMH right anymore than one hit proves it right. But as Sumner also notes,

Didn’t Shiller “offer the famous “irrational exuberance” phrase to Greenspan in 1996, before the stock bubble occurred? If so, this shows how hard it is to offer useful investment advice, even if you sense the market is overvalued.

OK, that’s only twice Shiller has missed the target, but just how many predictions has he made? For him to be doing better than 1 in 20, and counting him as having predicted the mortgage crisis, he needs to have made at least 20 verifiable predictions, with at least one other correct one. And maybe he has–I honestly don’t know. But it seems doubtful that he actually could have made that many predictions about distinct changes of direction in the market.

But, dammit, bubbles seem so goddamed obvious! And, in retrospect, after we can see the bust they sure do. The rest is pretty easily explained by cognitive biases. Sumner (again, I know! but he writes so damned clearly) discusses this, focusing on pattern-recognition bias-the same thing that makes us see a man (or old woman, or rabbit) in the man.

Now let’s ask why people have this mistaken notion that bubbles are easy to spot, and that Fama is deluded. I believe it is a cognitive illusion. People think they see lots of bubbles. Future price changes seem to confirm their views. This reinforces their perception that they were right all along…

Whatever happens to [a stock] over the next 50 years, the price movements will be interpreted through this congnitive illusion, this confidence that we can see patterns in graphs, even where they don’t exist. The “mountain peaks” will look like bubbles. In fact, you can generate a “stock graph” by just flipping coins repeatedly. And if you show the resulting graph to the average investor he will see patterns. It is all a cognitive illusion.

Here’s a bit more on pattern-recognition bias, that seems to incorporate some other biases that came to my mind.

We look for and see patterns where they don’t exist.

• We give more weight to recent events.
• We pay more attention to highly memorable events.
• Confirmation bias – once we have formulated a theory, we pay more attention to items that support it and ignore evidence that disproves it.

As to the first one, 2008 is still clear in our minds, much more so than Shiller’s missed bubble call in 1996 (and, no, the fact that the market went down sharply 4 years later does not count as a hit). As to the second one, busts that happen are far more memorable than busts that don’t occur–correct calls are going to be remembered, missed calls forgotten, and of course 1929 and 2008, as the two biggest economic crises of the past century are going to be remembered and disproportionately weighted in our thinking. As to the third, confirmation bias is endemic in human thinking about all matters–what would make any reader who’s come this far think it’s any different in this case?

About Those Cognitive Biases…
I know, people aren’t truly rational. We’re bundles of cognitive biases that infect all our thinking. That’s why I despise humanity, because I know that this true to a considerable extent. (To what extent depends on how strict our definition of rationality is, and that’s a contested issue.) Rationality is also a model, although to my on-going chagrin as a user of that model, some of its most fervent adherents forget that and assume its descriptive. It’s a really really useful model–it explains/predicts a lot of behavior very satisfactorily, like why something like 70% of people defect in the one-shot prisoner’s dilemma with a stranger. But it doesn’t explain everything, and it’s a model, not an empirically proved fact.

And I like empiricism, so I’m all gung-ho about cognitive psychologists and behavioral economists and (a small handful of) political scientists running lab experiments to test these things. I hope for a day when this knowledge is so thoroughly incorporated into predictive and functionally useful (e.g., simple enough) models that all of us social scientists adapt them.

But do they really disprove rationality? No, not really (at least not yet). They prove humans aren’t perfectly rational. They just prove that if humans really do have a foundation in rationality it doesn’t mean they’re always very good at it, that they have a variety of cognitive ticks that not infrequently impinge on the effectiveness of whatever rational tendencies they have. We can demonstrate this through game theory. Sure, lots of people don’t behave as predicted, but lots of people do. Some are pretty cold and ruthless about it, like the guy in one of my mentor’s studies on sharing who persuaded everyone in the group that they could trust each other if they all verbally promised to cooperate, and then he–like some of the others in the group–defected. Yet, making promises to groups of strangers does increase the frequency of cooperating, which isn’t strictly rational within the bounds of the game theory paradigm.

So let’s go back to Fama’s description early on. The EMH doesn’t require perfect rationality. Investors don’t have to have totally complete information, or all use it really well all the time, for the loose version of the assumptions to hold, and for markets to potentially–not necessarily, just potentially–be efficient. If the cognitive biases do not too severely–at whatever level “too severely” occurs–screw up investors’ thinking, then they don’t mess up the EMH model.

Or cognitive biases might completely undermine the EMH model, in at least two ways. First, it might be that cognitive biases are so pervasive that market prices rarely–and only randomly–approach efficient prices. I think that’s a reasonable hypothesis. However it’s not at all proven by the events of 2008. Those events are, to be sure, consistent with that hypothesis, but it’s far from any kind of definitive evidence.

And I don’t want to make it sound like I’m surreptitiously poo-pooing the cognitive biases research. I’ve been following it, although not as closely as I ideally would, for about 15 years now, and I’m persuaded that ultimately the social sciences have to incorporate it. Heck, I already do when I talk about political and policy processes. It’s just that saying people have these cognitive biases doesn’t mean their use of information is always wrong. Hell, if it was we couldn’t even talk sensibly about cognitive biases–we probably couldn’t even recognize them in the data.

And with effort, we can be trained to overcome them–probably not in all areas of our lives, but in specialized areas, which is how scientifical folks learn to be all scientificy. They’re critical to our understanding of people’s decision-making, but they aren’t everything. Rationality–non-biased thinking–is also critical to our understanding of people’s decision-making, but it’s not remotely everything, either.

The second way cognitive biases would undermine the EMH would be if people demonstrated that they could improve their rationality by training to overcome their cognitive biases–at least in this one specialized area; probably not in the rest of their lives–and thereby beat the market regularly by exploiting their knowledge of others’ cognitive biases.

Of course there’s the oddity that if that response became widespread, it would tend to support the EMH, if in an unexpected way.

So to be sure I take the issue of cognitive biases as a challenge to the rationality model very seriously, and I think that work is one of the most important areas of social science research today. But 2008 doesn’t prove it right and the EMH wrong, because it’s just one data point that might support that model, and it’s a data point that does not rebut the EMH.

Krugman’s Criticism
I’d like to respond to a series of criticisms here, but as I’ll explain below, I can’t right now. But I do want to address Krugman’s famous criticism of the model in 2009.

Discussion of investor irrationality, of bubbles, of destructive speculation had virtually disappeared from academic discourse. The field was dominated by the “efficient-market hypothesis,” promulgated by Eugene Fama of the University of Chicago, which claims that financial markets price assets precisely at their intrinsic worth given all publicly available information. (The price of a company’s stock, for example, always accurately reflects the company’s value given the information available on the company’s earnings, its business prospects and so on.)

I’m going to take his word that discussion of those things disappeared from academic discourse, because I simply don’t know myself. But notice the elisions in his description, which reads like the flawed investopedia definition. It’s fine as a rough pass, but when you’re critiquing something, it certainly behooves one to be rather less misleading. First, look at the final sentence–“the information available on the company’s earnings, its business prospects, and so on.” Ahem, if businesses mislead investors on that, then investors will be operating on flawed information, leading to flawed markets. The EMH itself says so–it’s not at all a radical model in that respect. And if that information is inaccurate, then, no, the stock will not be valued “precisely at their intrinsic worth.” They will be valued according to the available knowledge, which obscures their intrinsic worth.

Fama said this. It’s public information, but lots of people aren’t making use of it. I’m inclined to say that demonstrates an inefficient academic market.

Now let’s look a bit closer at the mortgage-backed securities. They were opaque, sure, but everyone knew they were ultimately based on averages, and as long as the market was going up the average would continue to go up. That information can rationally outweigh the information that you don’t know the specifics of the mortgages in that tranch.

And what happened when the information about their true fragility became public? Seems to me that investors reponded–quickly. Nothing irrational about their response at all.

Should they have seen it sooner? It’s arguable, but I’m willing to accept the assumption that they should have. So the strong version of EMH–that prices always instantly reflect new information–doesn’t seem to hold. Big deal–I don’t think most people actually took that seriously as a description of the real world, rather than just as the strictest form of the model. Oh, plenty of critics will tell the dumb EMH guys at Chicago all believe Jesus died for the salvation of the strong version, but, dammit, just go back and re-read Fama.

And, look, for those not in the academy–and maybe for someone who is–studying the pure model doesn’t prove one believes it’s truly descriptive of the world. Academics are funny that way; they’ll push a model to its limits and talk about how it functions at those limits, or even just how it would be predicted to function. And they might even find that in some cases it doesn’t actually fail in its pure model…which doesn’t mean they’re claiming it always works.

John Cochrane wrote an interesting, but surely less well known, response to that Krugman piece, in which he addresses Krugman’s thoughts on the EMH.

It’s fun to say we didn’t see the crisis coming, but the central empirical prediction of the efficient markets hypothesis is precisely that nobody can tell where markets are going – neither benevolent government bureaucrats, nor crafty hedge-fund managers, nor
ivory-tower academics. This is probably the best-tested proposition in all the social sciences. Krugman knows this, so all he can do is huff and puff about his dislike for a theory whose central prediction is that nobody can be a reliable soothsayer.

Krugman writes as if the volatility of stock prices alone disproves market efficiency, and efficient marketers just ignored it all these years. This is a canard that Paul knows better than to pass on, no matter how rhetorically convenient. (I can overlook his mixing up the CAPM and Black-Scholes model, but not this.) There is nothing about “efficiency” that promises “stability.” “Stable” growth would in fact be a major violation of efficiency. Efficient markets did not need to wait for “the memory of 1929 … gradually receding,” nor did we fail to read the newspapers in 1987. Data from the great depression has been included in practically all the tests. In fact, the great “equity premium puzzle” is that if efficient, stock markets don’t seem risky enough to deter more people from investing! Gene Fama’s PhD thesis was on “fat tails” in stock returns

It is true and very well documented that asset prices move more than reasonable expectations of future cashflows. This might be because people are prey to bursts of irrational optimism and pessimism. It might also be because people’s willingness to take on risk varies over time, and is lower in bad economic times. As Gene Fama pointed out in 1970, these are observationally equivalent explanations. Unless you are willing to elaborate your theory to the point that it can quantitatively describe how much and when risk premiums, or waves of “optimism” and “pessimism,” can vary, you know nothing. No theory is particularly good at that right now. Crying “bubble” is empty unless you have an operational procedure for identifying bubbles, distinguishing them from rationally low risk premiums, and not crying wolf too many years in a row.

Again we see an economist who holds the EMH view pointing out that the criticisms of the EMH in relation to the crisis fundamentally misinterprets how the EMH supporters actually understand their own model.

Conclusion
I’d like to have added some points addressing some of the research criticizing the EMH. I haven’t for three reasons. First, as I noted, I’m not an expert in this area, and it will take me some more time to re-brush up on that. Second, it’s midnight, I have classes tomorrow, a chapter for my on-line text to write, and about 30 papers to grade, so it’s just not going to happen right now. Third, this post, at least, is already long enough.

If I have the time and energy, I’ll write such a post. Or maybe someone will do first so I don’t have to (J R–the OT accepts guest posts, and you’d surely do it better than I). At this point, though, it’s fair to critique me for not addressing all the critics’ responses to the EMH. I just can’t cover everything at one go.

And if, as I’ve claimed, I’m not an expert, why am I seemingly so confident when I’ve critiqued a fellow OTer for being confident without really demonstrating good knowledge of EMH? The first answer is that I can see the error the critics are making–if they’re making inaccurate statements about how EMH advocates understand their own model, I just can’t have confidence in them. But second, I’m not that confident. It’s an empirical issue–either the model has predictive power, or it doesn’t. Or, it has predictive power to some extent, but only to some extent. And I’ve discussed two ways in which it could be disproven, or at least undermined.

And here’s an example–just an anecdote–that persuades me people don’t always use information well. After 9/11, a friend of mine bought stock in a company that did some kind of airline security business. He was positive this company’s stock was going to skyrocket. His broker, to his credit, tried to talk him out of it, but my friend insisted. The stock tanked. Lots of people bought, not realizing that the increasing price was not evidence of a long term market value for the firm, but a reflection of everyone else’s post 9/11 assumptions. On the other hand, that stock corrected pretty quickly, too, so people didn’t, as a group, demonstrate excessive and long-lasting irrationality any more than they demonstrated any particular degree of rationality at the beginning/

And you might have noticed that I didn’t defend the EMH’s predictive accuracy in its stronger versions. Because I’m not that confident. But the person to whom I’m responding argued it was dead in even it’s weak form, so a defense of that form is relevant.

I want to conclude by coming back to the one crucial point, though. The primary, fundamental, prediction of the EMH is that you can’t reliably predict and beat the market. 2008 did not disprove that. You can’t make–I don’t think–a solid case in logic, one that respects social science standards for evidence, that the financial crisis did disprove that. And if it didn’t disprove that, it didn’t disprove the Efficient Markets Hypothesis. And also, to the extent the model is designed to test how markets respond when the assumptions are violated, then the model itself–in its strict form–has not been violated, because it doesn’t predict efficient markets will necessarily occur when assumptions are violated.

Because I have other writing, and papers to grade, and because I’m excessively irritable due to lack of sleep lately (which I’m stupidly compounding tonight), I probably won’t participate much in any discussion. I’ve said my piece here. I at least have made an effort to explain my position.

easy ways to explain young ladies about method
weight loss tips How to Become a Teen Fashion Model in New York