What "efficient market" does and doesn't mean

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
rbaldini
Posts: 902
Joined: Mon Mar 23, 2015 3:20 pm

What "efficient market" does and doesn't mean

Post by rbaldini » Sat Jan 20, 2018 11:05 am

Just yesterday I posted the question of whether Bogleheads thought the crypto market was efficient. There were a variety of answers, but mostly I was struck by how much people disagreed about the meaning of "efficient market". I'm writing to ask about that.

First, investopedia's definition: The efficient market hypothesis (EMH) is an investment theory that states it is impossible to "beat the market" because stock market efficiency causes existing share prices to always incorporate and reflect all relevant information.

What does this imply?

One claim I've heard that makes sense to me is that efficient market price changes tend to be unpredictable (people usually say "random" but I think I prefer "unpredictable"). The reason is simple: if a future price change could be easily predicted, then most participants would anticipate it and drive the price to that point almost immediately by buying/selling, therefore very quickly realizing and erasing the prediction. It follows that in an efficient market, any future price changes that *do* happen, and that you actually have time to take advantage of, ought to be pretty unpredictable.

One reason it wouldn't be *completely* unpredictable is if taking advantage of predictions is hard to do. I believe the housing market, for example, actually has moderately strong price momentum - stronger than in stocks - such that one can predict better than chance how house prices in some location will change over the next few years (statistically speaking). So why don't investors take advantage of this opportunity and therefore destroy that trend? I think it's because buying and selling houses is much more slow and costly than buying and selling stocks/bonds/etc. Participants can't or won't react fast enough to take advantage of those short term opportunities. (This might also explain why there is some quite short-term momentum in stocks - trading on short timescales involves non-trivial transaction costs, not to mention increased taxes.)

Now, here's a claim that I'm skeptical of: some folks said that bubbles and seemingly irrational behavior shouldn't exist in efficient markets. Some representative quotes:
"If markets were perfectly efficient, there would be no bubbles, no panics, just rational responses to events."
"Isn't Bogleheads a non-EMH philosophy? A big reason for Bogleheads to own bonds is to cushion the stock-market-bubble pops. Stock market bubbles wouldn't exist in an efficient market, so Bogleheads is de facto a non-EMH philosophy."

My hunch is this is incorrect. The investopedia definition only states that efficient markets are ones where available information is rapidly disseminated and priced in. It doesn't actually say that the people incorporating the information and pricing the assets are "rational." As long as big price changes and turn-arounds are due to sudden and unpredictable changes in public information, then that would be consistent with an efficient market, no?

Now, you might argue that if efficient markets rapidly price in new information, but possibly do so irrationally, then a rational person should still be able to beat an efficient market. Maybe so, but here are two possible reasons why not.
First, if people are rational, and you are a person, then you are probably irrational. Lots of smart people are working in the market; even if they are highly irrational in many ways, you'd have to believe that you more rational than them.
Second, stock returns are driven not just by underlying "rational" profitability of investments, but to the speculative behavior of others. The latter is a pretty big component of returns, I believe. If the world is irrational and you are rational, then that might actually work *against* you if your goal is to time the vagaries of market returns. Yes, you might be better able to predict early on which companies will actually turn a profit, but that might not be enough.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: What "efficient market" does and doesn't mean

Post by dbr » Sat Jan 20, 2018 11:27 am

I don't think controlling the risk of a portfolio by blending bonds with stocks requires an assessment as to whether or not the market is efficient. How risky the market is comes from a empirical observation rather than from theory. I don't think most Bogleheads concern themselves with what efficient market means and whether or not any of the various forms of the Efficient Market Hypothesis are valid, meaning in accord with the real world.

Aside from that observation I don't know enough about the theoretical concept of the market to comment more.

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: What "efficient market" does and doesn't mean

Post by lack_ey » Sat Jan 20, 2018 11:50 am

EMH is more strictly about informational efficiency. Not directly anything about beating the market, though that is what it is associated with a lot.

As you get to, this doesn't in of itself suggest an equilibrium pricing model, an explanation for how information is incorporated into prices. Usually people who talk about EMH assume that the group behavior is rational, even if individuals are not, and that participants maximizing utility affect price changes. But what that actually means is up for interpretation and needs to be more rigorously defined in order to be evaluated. Does this mean they're pricing in CAPM or something else? Who knows. This needs to be clarified when people invoke EMH in many situations.

The issue is that you can't really evaluate informational efficiency and separate it out from the pricing model—this is the joint hypothesis problem.

If you think about it, if you assume a market is informationally efficient but its participants drive and incorporate prices in completely irrational, unpredictable ways, that would be a kind of empty, useless efficiency. So usually people would just call this "inefficient" as it is probably far enough away from the conception most have of EMH, even if you couldn't prove which part is causing the wacky behavior.
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
One reason it wouldn't be *completely* unpredictable is if taking advantage of predictions is hard to do. I believe the housing market, for example, actually has moderately strong price momentum - stronger than in stocks - such that one can predict better than chance how house prices in some location will change over the next few years (statistically speaking). So why don't investors take advantage of this opportunity and therefore destroy that trend? I think it's because buying and selling houses is much more slow and costly than buying and selling stocks/bonds/etc. Participants can't or won't react fast enough to take advantage of those short term opportunities. (This might also explain why there is some quite short-term momentum in stocks - trading on short timescales involves non-trivial transaction costs, not to mention increased taxes.)
This is one of the important points about how real-world information transmission into prices. Generally if there are fewer transactions in a market, the price discovery will probably be slower in practice. Transaction costs and frictions also reduce the ability of market participants to correct prices by taking advantage of a good deal one way or the other. This is part of the idea behind arguments about limits to arbitrage.
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
Now, here's a claim that I'm skeptical of: some folks said that bubbles and seemingly irrational behavior shouldn't exist in efficient markets. Some representative quotes:
"If markets were perfectly efficient, there would be no bubbles, no panics, just rational responses to events."
"Isn't Bogleheads a non-EMH philosophy? A big reason for Bogleheads to own bonds is to cushion the stock-market-bubble pops. Stock market bubbles wouldn't exist in an efficient market, so Bogleheads is de facto a non-EMH philosophy."

My hunch is this is incorrect. The investopedia definition only states that efficient markets are ones where available information is rapidly disseminated and priced in. It doesn't actually say that the people incorporating the information and pricing the assets are "rational." As long as big price changes and turn-arounds are due to sudden and unpredictable changes in public information, then that would be consistent with an efficient market, no?
Yes, it's possible it's just about different information. Also, nobody said that investor risk preferences and tolerances were constant over time. The discount rate can and should change. It's just that a lot of people would argue that the overall market behavior and swings are probably irrational, at least sometimes.

User avatar
willthrill81
Posts: 5796
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: What "efficient market" does and doesn't mean

Post by willthrill81 » Sat Jan 20, 2018 12:36 pm

The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.

Behavioral finance has demonstrated beyond a reasonable doubt that market participants, from individuals to 'smart money' institutions, suffer from a number of biases that prevent them from acting rationally, particularly in turbulent markets.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

rbaldini
Posts: 902
Joined: Mon Mar 23, 2015 3:20 pm

Re: What "efficient market" does and doesn't mean

Post by rbaldini » Sat Jan 20, 2018 12:42 pm

willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.

User avatar
willthrill81
Posts: 5796
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: What "efficient market" does and doesn't mean

Post by willthrill81 » Sat Jan 20, 2018 12:50 pm

rbaldini wrote:
Sat Jan 20, 2018 12:42 pm
willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.
willthrill81 wrote:
Sat Jan 20, 2018 12:10 pm
Investors, including the likes of Warren Buffett, and researchers have disputed the efficient-market hypothesis both empirically and theoretically. Behavioral economists attribute the imperfections in financial markets to a combination of cognitive biases such as overconfidence, overreaction, representative bias, information bias, and various other predictable human errors in reasoning and information processing. These have been researched by psychologists such as Daniel Kahneman, Amos Tversky, Richard Thaler, and Paul Slovic. These errors in reasoning lead most investors to avoid value stocks and buy growth stocks at expensive prices, which allow those who reason correctly to profit from bargains in neglected value stocks and the overreacted selling of growth stocks. Investors prefer riskier funds in spring and safer funds in autumn.
https://en.wikipedia.org/wiki/Efficient ... hypothesis

The EMH assumes that all of the biases among the actors in a market cancel each other out. However, behavioral finance has demonstrated that a majority of actors in a market can suffer from the same biases, so it is impossible for these biases to balance out.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

staythecourse
Posts: 6011
Joined: Mon Jan 03, 2011 9:40 am

Re: What "efficient market" does and doesn't mean

Post by staythecourse » Sat Jan 20, 2018 1:00 pm

lack_ey wrote:
Sat Jan 20, 2018 11:50 am
EMH is more strictly about informational efficiency.
This is how I perceive EMH. It is matter of ALL the available info. of stock x is already know and incorporated into the current share price so the ONLY way to make more money is to either have info. no one has (insider information which is why it is illegal) OR be lucky guessing which way the price will go next.

The best example of this is to look at the old days. The story of Rothchild (I believe). He used to have courier pigeons report back from the battlefield the results of battles so he could change his investments accordingly. This was useful because he received the same info. much EARLIER then the other market participants.

The problem now is outside of insider info. all financial info. is transmitted in nanoseconds. So, any market inefficiencies are getting smaller and smaller if any.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

User avatar
Oicuryy
Posts: 1246
Joined: Thu Feb 22, 2007 10:29 pm

Re: What "efficient market" does and doesn't mean

Post by Oicuryy » Sat Jan 20, 2018 1:10 pm

Define "rational". Every trade involves both a buyer and a seller. Which one of them is irrational?

Ron
Money is fungible | Abbreviations and Acronyms

bgf
Posts: 609
Joined: Fri Nov 10, 2017 9:35 am

Re: What "efficient market" does and doesn't mean

Post by bgf » Sat Jan 20, 2018 1:15 pm

Oicuryy wrote:
Sat Jan 20, 2018 1:10 pm
Define "rational". Every trade involves both a buyer and a seller. Which one of them is irrational?

Ron
traditional economics defines "rational" as the ability to accurately and immediately calculate probabilities to make the decision that maximizes one's utility even when confronted with uncertainty.

it is immediately obvious to a casual observer that this is not how people work, yet nobel prizes were given for work that included that assumption.

as far as the buyer-seller relationship, given full information, they theoretically met at an equilibrium price, the price at which the odds of winning/losing were the same for both, i.e., a fair game. this was initially an explanation for why stock returns appeared random, random in the sense of Brownian Motion.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

Shallowpockets
Posts: 680
Joined: Fri Nov 20, 2015 10:26 am

Re: What "efficient market" does and doesn't mean

Post by Shallowpockets » Sat Jan 20, 2018 1:25 pm

Oicuryy wrote:
Sat Jan 20, 2018 1:10 pm
Define "rational". Every trade involves both a buyer and a seller. Which one of them is irrational?

Ron

They are both rational. The seller is selling to realize his profit and the buyer is buying on the hopes of continued upward price movement. At the distinct point in time of the sale the most rational would be the seller. His speculation is over, whereas the buyer still has the future to deal with.
The trick in efficiency is to be nimble. You need to bear in mind the prevailing bias. It is like game theory. What will the other guy do with what he thinks, and can I think like that to get ahead of him. And then thinking the same from his perspective. And, how many others are thinking like me, like him, and what are the numbers and timeframes.

petulant
Posts: 326
Joined: Thu Sep 22, 2016 1:09 pm

Re: What "efficient market" does and doesn't mean

Post by petulant » Sat Jan 20, 2018 1:45 pm

One problem that hasn't been discussed is the difference between a stock price being rationally set relative to other stocks and the general price level of the market.

It's very likely that individual stock prices relative to each other are set very rationally because the information relevant to a single company or industry is quickly incorporated into prices. For example, the price of Apple stock relative to Google, Microsoft, and Amazon may be quite efficient. Even if there are irrational or differently motivated investors in a particular security, it is entirely possible for informed investors to drive the price to the efficient level. So even if some people sell Exxon Mobil stock, a hedge fund somewhere will buy it back; or if people are sold that Facebook has a glorious future, bidding the price up, hedge funds can short the stock to a lower level. A sudden change in price isn't the result of inefficiency; it's probably from new information. While there are a variety of EMH ideas, I think this is the core idea. The upshot of this kind of EMH is that a normal retail investor shouldn't be betting that they know something other people don't about Microsoft beating Apple or anything like that; they should buy the market if they want to benefit from equity returns. This is the kind of EMH to which almost every Boglehead is committed, from what I can tell.

It would be another level of EMH to posit that the absolute price levels of all securities are "correct," particularly when absolute price levels incorporate assumptions about discount rates while also being affected by cash flows between asset classes. It may be that there is no "right" price or objective information about interest rates, future economic growth, or the quality of asset classes that allows us to say the S&P 500 Index is perfectly efficient. For example, it may be efficient for Amazon to have a higher P/E than Walmart, so we shouldn't pick between them, but it might not be efficient for Amazon to specifically have a 30 and Walmart a 20 P/E since those specific values are also informed by macro factors.

I think disputes over whether these macro factors are "inefficient" in an actionable way is at the heart of why Bogleheads argue about tilting to certain asset classes and why tactical asset allocation is a topic of discussion for intellectuals like Michael Kitces.

bgf
Posts: 609
Joined: Fri Nov 10, 2017 9:35 am

Re: What "efficient market" does and doesn't mean

Post by bgf » Sat Jan 20, 2018 1:50 pm

petulant wrote:
Sat Jan 20, 2018 1:45 pm
One problem that hasn't been discussed is the difference between a stock price being rationally set relative to other stocks and the general price level of the market.

It's very likely that individual stock prices relative to each other are set very rationally because the information relevant to a single company or industry is quickly incorporated into prices. For example, the price of Apple stock relative to Google, Microsoft, and Amazon may be quite efficient. Even if there are irrational or differently motivated investors in a particular security, it is entirely possible for informed investors to drive the price to the efficient level. So even if some people sell Exxon Mobil stock, a hedge fund somewhere will buy it back; or if people are sold that Facebook has a glorious future, bidding the price up, hedge funds can short the stock to a lower level. A sudden change in price isn't the result of inefficiency; it's probably from new information. While there are a variety of EMH ideas, I think this is the core idea. The upshot of this kind of EMH is that a normal retail investor shouldn't be betting that they know something other people don't about Microsoft beating Apple or anything like that; they should buy the market if they want to benefit from equity returns. This is the kind of EMH to which almost every Boglehead is committed, from what I can tell.

It would be another level of EMH to posit that the absolute price levels of all securities are "correct," particularly when absolute price levels incorporate assumptions about discount rates while also being affected by cash flows between asset classes. It may be that there is no "right" price or objective information about interest rates, future economic growth, or the quality of asset classes that allows us to say the S&P 500 Index is perfectly efficient. For example, it may be efficient for Amazon to have a higher P/E than Walmart, so we shouldn't pick between them, but it might not be efficient for Amazon to specifically have a 30 and Walmart a 20 P/E since those specific values are also informed by macro factors.

I think disputes over whether these macro factors are "inefficient" in an actionable way is at the heart of why Bogleheads argue about tilting to certain asset classes and why tactical asset allocation is a topic of discussion for intellectuals like Michael Kitces.
that's an interesting take, but i don't see any reason why a market would be capable of comparing and relatively valuing two stocks, or two corporate bonds, while at the same time being incapable of comparing a corporate bond and a government bond, or a government bond and a stock.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

magneto
Posts: 951
Joined: Sun Dec 19, 2010 10:57 am
Location: On Chesil Beach

Re: What "efficient market" does and doesn't mean

Post by magneto » Sat Jan 20, 2018 2:15 pm

An intriguing thread.
Confirming, denying or explaining EMH is way above this investor's pay-grade.

What is puzzling is that perhaps while fundamentals may be unchanged, something in the world stirs investors' hopes or fears, and a change in attitude to Stock, Sector or Market may transpire.
These changes may or may not be relevant to the future fundamentals?

There are also occasions when investors are forced to sell good Stocks to meet cash-needs, because the bad Stocks have become unsaleable. Yet for those good Stocks, nothing had really changed :!:

Will be following thread carefully.
'There is a tide in the affairs of men ...', Brutus (Market Timer)

User avatar
willthrill81
Posts: 5796
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: What "efficient market" does and doesn't mean

Post by willthrill81 » Sat Jan 20, 2018 2:23 pm

rbaldini wrote:
Sat Jan 20, 2018 12:42 pm
willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.
Prices cannot reflect all available information unless market participants are acting rationally.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
patrick013
Posts: 2367
Joined: Mon Jul 13, 2015 7:49 pm

Re: What "efficient market" does and doesn't mean

Post by patrick013 » Sat Jan 20, 2018 2:26 pm

When information is imperfect the market acts on guesses, estimates,
forecasts, sometimes disbelief. So quality of information can become
very important. Quality of earnings for example refers to accuracy of
earnings forecasts.

When information is perfect, the actual numbers are posted to the public,
and buyers and sellers act on that, the market is efficient as it quickly moves
to new or adjusted prices and strategies.

The value of perfect information is the difference in market price derived
from the imperfect information compared to the market price derived from
the perfect information.

How that relates to beating the market or having some bias is always in
the above process until the information becomes perfect. New guesses
start and the market changes pricing again shortly thereafter.
age in bonds, buy-and-hold, 10 year business cycle

User avatar
Oicuryy
Posts: 1246
Joined: Thu Feb 22, 2007 10:29 pm

Re: What "efficient market" does and doesn't mean

Post by Oicuryy » Sat Jan 20, 2018 3:24 pm

willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.

Behavioral finance has demonstrated beyond a reasonable doubt that market participants, from individuals to 'smart money' institutions, suffer from a number of biases that prevent them from acting rationally, particularly in turbulent markets.
bgf wrote:
Sat Jan 20, 2018 1:15 pm
traditional economics defines "rational" as the ability to accurately and immediately calculate probabilities to make the decision that maximizes one's utility even when confronted with uncertainty.

it is immediately obvious to a casual observer that this is not how people work, yet nobel prizes were given for work that included that assumption.
Does behavioral finance use that same definition of "rational" that traditional economics uses? That would explain why they found so many irrational investors. As bgf says, that is not how people work.

So securities prices are set by millions of irrational people. If anyone is ever able "to accurately and immediately calculate probabilities to make the decision that maximizes one's utility even when confronted with uncertainty", then she will soon own all the securities.

Ron
Money is fungible | Abbreviations and Acronyms

User avatar
nisiprius
Advisory Board
Posts: 36665
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: What "efficient market" does and doesn't mean

Post by nisiprius » Sat Jan 20, 2018 3:34 pm

1) What is the data that would lead to the conclusion that the efficient market hypothesis has been falsified? Seriously falsified, not just slightly falsified?

2) What are the specific things that are impossible if the EMH is correct, but possible if it is incorrect?

3) How could a Norman Rockwell create an illustration of someone trying to falsify the EMH?

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

bgf
Posts: 609
Joined: Fri Nov 10, 2017 9:35 am

Re: What "efficient market" does and doesn't mean

Post by bgf » Sat Jan 20, 2018 6:28 pm

nisiprius wrote:
Sat Jan 20, 2018 3:34 pm
1) What is the data that would lead to the conclusion that the efficient market hypothesis has been falsified? Seriously falsified, not just slightly falsified?

2) What are the specific things that are impossible if the EMH is correct, but possible if it is incorrect?

3) How could a Norman Rockwell create an illustration of someone trying to falsify the EMH?

Image
people on this site probably aren't going to like my answer, but EMH was falsified by George Soros, Ed Thorp, Warren Buffett, James Simons, J. Doyne Farmer, Didier Sornette, David Shaw, and others. These are individuals that were not only able to pinpoint inefficiencies and other mispricings in the markets, but, most importantly, were able to explain their 'edge' and then execute and exploit for profit, huge profits. Ed Thorp in his most recent book "A Man for All Markets" provides additional examples of public and lasting yet completely illogical mispricings of certain securities.

These are not 'random outliers' that one would expect statistically speaking. They were individuals capable of exploiting the very gaps in markets that EMH postulates do not exist.

If all information were immediately 'priced in' and returns were random, then their careers would have been impossible.

Still, passive investing in an index has its place. it is the optimal strategy for an investor that realizes he is unlike the above and cannot articulate and execute an edge.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

SGM
Posts: 2723
Joined: Wed Mar 23, 2011 4:46 am

Re: What "efficient market" does and doesn't mean

Post by SGM » Sat Jan 20, 2018 7:02 pm

Markets are not price-equals-value efficient , but they are hard-to-beat efficient.

alex_686
Posts: 3851
Joined: Mon Feb 09, 2015 2:39 pm

Re: What "efficient market" does and doesn't mean

Post by alex_686 » Mon Jan 22, 2018 2:13 pm

coming late, starting sort of afresh, lots of comments.

First, you are putting too high a burden on EMH. EMH just states that the markets incorporate all known information into pricing. You need other theories to talk about asset allocation.
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
First, investopedia's definition: The efficient market hypothesis (EMH) is an investment theory that states it is impossible to "beat the market" because stock market efficiency causes existing share prices to always incorporate and reflect all relevant information.
There are a variety of different flavors of EMH, ranging from weak to strong. Only the strong formulation, which is mostly academic in the literal sense, says you can't beat the market. I favor the semi-strong formulation for the stock market, which says that the market is efficient because you have lots of smart people who have squeezed all of the profitable opportunities in the market. i.e., There is no free lunch.
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
One reason it wouldn't be *completely* unpredictable is if taking advantage of predictions is hard to do.
The more common term is "market inefficiencies". Information asymmetry, where information is not perfect. Used call dealers selling cars that they know are lemons. High transactions costs, like the closing cost on houses. There are many known inefficiencies out there, but as you say they may not be actionable.
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
Now, here's a claim that I'm skeptical of: some folks said that bubbles and seemingly irrational behavior shouldn't exist in efficient markets. Some representative quotes:
"If markets were perfectly efficient, there would be no bubbles, no panics, just rational responses to events."
"Isn't Bogleheads a non-EMH philosophy? A big reason for Bogleheads to own bonds is to cushion the stock-market-bubble pops. Stock market bubbles wouldn't exist in an efficient market, so Bogleheads is de facto a non-EMH philosophy."
The theory behind passive investing have very strong foundations in EMH. No EMH, no passive investing. Now EMH is a model of human behavior and a pretty high level one at there. Check out behavioral economics where they dig into some of the quirks.

Once again you may be putting too much on EMH's shoulders. EMH does not say that there are not bubbles or panics. It states that you can't predict them. Here is a thought experiment. The internet has been one of the most profitable and economically transformative events in the economy. If you could step into a time machine, would you go back 20 years and invest all of your savings in dot.com stocks?
rbaldini wrote:
Sat Jan 20, 2018 11:05 am
My hunch is this is incorrect. The investopedia definition only states that efficient markets are ones where available information is rapidly disseminated and priced in. It doesn't actually say that the people incorporating the information and pricing the assets are "rational." As long as big price changes and turn-arounds are due to sudden and unpredictable changes in public information, then that would be consistent with an efficient market, no?
EMH is just about current information. When unpredictable events happen that new information is incorporated into the new prices. People have combed over historical events looking for actionable inefficiencies. They are out there but are hard to exploit.

alex_686
Posts: 3851
Joined: Mon Feb 09, 2015 2:39 pm

Re: What "efficient market" does and doesn't mean

Post by alex_686 » Mon Jan 22, 2018 2:48 pm

Part 1 of my answer responded just to the OP comments, which other commentators commented on. Now to the main part of the OP's question - EMH and currencies.

EHM does not deal with asset allocation, but as I alluded to many the theory behind most asset allocation rests on EMH as a assumption. Modern Portfolio Theory is an example. I prefer Black Litterman.

The OP says "Isn't Bogleheads a non-EMH philosophy? A big reason for Bogleheads to own bonds is to cushion the stock-market-bubble pops." The answer to this is no. Or rather, the above statement is a general rule of thumb, it can't be the basis of a hypothesis. No, bonds are included because it yields the most efficient portfolio - the highest amount of return for the lowest possible risk.

That being said, these theories don't work well for commodities and currencies.

First, the price of an asset is the discounted Future Cash Flow. EMH implies that the information on expected cash flows is distributed across the market and that discount rate - i..e. the cost of risk - is homogeneous. i.e., riskier assets carry higher interest rates / expected returns in a ordinal fashion. The problem here is that commodities don't have a cash flow - just risk. You are now dependent on fickle fashion and fate.

Second, commodities don't have a market cap. Some people use production numbers, others the importance to the economy. Bitcoin has neither. The fact that the number of bitcoins out there does not help. Some are lost. The new futures being issued have increase the number of bitcoins out there from a theory perspective.

Lastly, there are variations of the mean-variant portfolio technique. Input expected returns, risk and volatility, and your goals. The model then spits out a number. The problem here is that Bitcoins price has had exponential growth, so if you input historical numbers you are going to get a 100% Bitcoin portfolio. This often happens in rapidly growing markets where everybody knows the fundamentals are increasing but they have no idea on where the ceiling is.

desafinado
Posts: 116
Joined: Wed Dec 28, 2016 3:14 am

Re: What "efficient market" does and doesn't mean

Post by desafinado » Mon Jan 22, 2018 5:24 pm

bgf wrote:
Sat Jan 20, 2018 6:28 pm
people on this site probably aren't going to like my answer, but EMH was falsified by George Soros, Ed Thorp, Warren Buffett, James Simons, J. Doyne Farmer, Didier Sornette, David Shaw, and others. These are individuals that were not only able to pinpoint inefficiencies and other mispricings in the markets, but, most importantly, were able to explain their 'edge' and then execute and exploit for profit, huge profits. Ed Thorp in his most recent book "A Man for All Markets" provides additional examples of public and lasting yet completely illogical mispricings of certain securities.

These are not 'random outliers' that one would expect statistically speaking. They were individuals capable of exploiting the very gaps in markets that EMH postulates do not exist.
It's also worth thinking about why this might be possible. One answer is that people participate in the markets for different reasons. Some investors are utilitarian investors. Those of us who buy and hold are essentially moving money from the present to the future. Market participants who seek trading profits can make money from utilitarian investors, who tend to not be as well informed about any given security.

InsuranceQuant
Posts: 3
Joined: Mon Jan 22, 2018 6:04 pm

Re: What "efficient market" does and doesn't mean

Post by InsuranceQuant » Mon Jan 22, 2018 7:33 pm

The efficient market hypothesis is a model of human behavior, an abstraction to make it easier to understand. A simpler statement of the hypothesis might be: "If some people figure out a way to make lots of money relatively risk free, other people will try really hard to copy it". Sometimes when its not easy to "copy it" the ways of making money persist. However, in other times, it is easy to "copy it" and doing so works for a while. If this becomes reinforcing (some people are able to copy it successfully, which makes others want to copy it, which causes the first group of copiers to be successful) it can become a bubble.

The link between EMH and bubbles is pretty subtle but profound. I'd argue the mechanism that causes the market to be efficient is the same mechanism that creates bubbles.

JBTX
Posts: 4043
Joined: Wed Jul 26, 2017 12:46 pm

Re: What "efficient market" does and doesn't mean

Post by JBTX » Mon Jan 22, 2018 8:07 pm

willthrill81 wrote:
Sat Jan 20, 2018 2:23 pm
rbaldini wrote:
Sat Jan 20, 2018 12:42 pm
willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.
Prices cannot reflect all available information unless market participants are acting rationally.
My recolllection of EMH is that all information has been reflected in the markets and on average you can’t make money with publicly available information. I am not sure if rational has anything to do with it.

Who defines rational. Was it irrational to think housing prices would never go down when they infact never went down?

The fact that so few can consistently beat the market seems to be evidence that the efficient market hypothesis is correct. The fact that behavioral finance comes up with anomilies is not necessarily evidence of inefficient markets.

User avatar
willthrill81
Posts: 5796
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: What "efficient market" does and doesn't mean

Post by willthrill81 » Mon Jan 22, 2018 8:24 pm

JBTX wrote:
Mon Jan 22, 2018 8:07 pm
willthrill81 wrote:
Sat Jan 20, 2018 2:23 pm
rbaldini wrote:
Sat Jan 20, 2018 12:42 pm
willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.
Prices cannot reflect all available information unless market participants are acting rationally.
My recolllection of EMH is that all information has been reflected in the markets and on average you can’t make money with publicly available information. I am not sure if rational has anything to do with it.

Who defines rational. Was it irrational to think housing prices would never go down when they infact never went down?

The fact that so few can consistently beat the market seems to be evidence that the efficient market hypothesis is correct. The fact that behavioral finance comes up with anomilies is not necessarily evidence of inefficient markets.
The fact that a number of people already cited in this thread have beaten the market over a long-term period is strong evidence that the EMH is not entirely correct. Unless of course you believe that all of those people were just lucky.

A 'rational' view could be one that says, like Buffet has done repeatedly, that company X is truly undervalued by the market, so I will buy it now, expecting that the market will change its valuation of the company later.

Panic can hardly be defined as rational, yet there are times when that describes large swaths of investors.
InsuranceQuant wrote:
Mon Jan 22, 2018 7:33 pm
The efficient market hypothesis is a model of human behavior, an abstraction to make it easier to understand. A simpler statement of the hypothesis might be: "If some people figure out a way to make lots of money relatively risk free, other people will try really hard to copy it". Sometimes when its not easy to "copy it" the ways of making money persist. However, in other times, it is easy to "copy it" and doing so works for a while. If this becomes reinforcing (some people are able to copy it successfully, which makes others want to copy it, which causes the first group of copiers to be successful) it can become a bubble.

The link between EMH and bubbles is pretty subtle but profound. I'd argue the mechanism that causes the market to be efficient is the same mechanism that creates bubbles.
Welcome to the forum!

I think your explanation of the EMH is sound. Put that way, it is not impossible to 'beat' the market, but it's likely to be difficult to do so.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Alex Frakt
Founder
Posts: 10853
Joined: Fri Feb 23, 2007 1:06 pm
Location: Chicago
Contact:

Re: What "efficient market" does and doesn't mean

Post by Alex Frakt » Tue Jan 23, 2018 12:28 am

To understand the role that EMH plays, you need some historical context. Long before EMH, people studying stock or bond price series noticed that they appeared to be random, you simply couldn't predict the next price of a stock or bond or set of stocks and bonds (i.e. stock or bond index) from a review of the previous prices no matter what the timescale. Eugene Fama eventually came up with the EMH as a plausible explanation for why prices should exhibit a random walk. In its simplest form, the EMH hypothesizes that at any given time the price of a stock or bond reflects all available information about that stock or bond. If this is true, then the price will change only when new information becomes available. Since new information must be in a random direction (if it were in a predictable direction, it would already be incorporated into the original price), then prices should exhibit random walk behavior.

As an aside, fitting the hypothesis to the data is backwards by the normal rules of the scientific method. You are supposed to formulate your hypothesis, use it to make predictions and then test the predictions. This may be why Fama's Nobel (Memorial) prize contains no mention of the EMH, instead he shared it with two others "for their empirical analysis of asset prices."

But the EMH was never posited as an absolute. Well, except as an easily demolished strawman. A perfectly efficient market requires instant communication of all information about the stock to all parties, infinite liquidity, instant and unbiased trade execution and zero transaction costs. If any of these are missing, you can profit from being a market maker or strategies like front running, insider trading, algorithmic high frequency trading, or block trading. And there are some markets that are incredibly inefficient: the property market is the one we are most familiar with, private equity is another, collectibles markets, diamonds.

But our interests are as retail investors in stocks and bonds that are traded on major markets. For these instruments, the consequences of EMH do appear to hold. There are simply no demonstrated strategies that consistently beat a representative benchmark on a risk-adjusted basis after costs. Not stock picking, not market timing, not technical analysis, not dogs of the dow, not anything. No one has ever become a billionaire placing market orders on the NYSE.

User avatar
JoMoney
Posts: 5884
Joined: Tue Jul 23, 2013 5:31 am

Re: What "efficient market" does and doesn't mean

Post by JoMoney » Tue Jan 23, 2018 3:06 am

The "efficient market" being about informational efficiency is probably about right.
The idea that the market is "efficient" is wrong in an absolute sense, but may be more useful in the context of "relative efficiency".
Some markets are rife with corruption, insider trading, bucket shops, con-men, etc... Some markets are considerably more "efficient", and the competition among the successful traders who have the information narrows the marginal profitability from their trading against those lesser informed and in lesser positions to act on the information.

Even Warren Buffett suggests it's less easy to garner something extra trading large-cap U.S. stocks relative to other areas.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

JBTX
Posts: 4043
Joined: Wed Jul 26, 2017 12:46 pm

Re: What "efficient market" does and doesn't mean

Post by JBTX » Tue Jan 23, 2018 9:12 am

willthrill81 wrote:
Mon Jan 22, 2018 8:24 pm
JBTX wrote:
Mon Jan 22, 2018 8:07 pm
willthrill81 wrote:
Sat Jan 20, 2018 2:23 pm
rbaldini wrote:
Sat Jan 20, 2018 12:42 pm
willthrill81 wrote:
Sat Jan 20, 2018 12:36 pm
The EMH assumes rational actors. Otherwise, it would be impossible for all information to be reflected in prices.
There's an important distinction between "price quickly comes to reflect available information" and "prices are 'rationally' set by information." The former refers to the speed of information dissemination, the latter refers to what people actually do with it. It seems to me that EMH primarily postulates the former, but not necessarily the latter.

To be fair, if one uses EMH to imply that it's hard to beat the market, then that would seem to imply that participants are at least mostly rational.
Prices cannot reflect all available information unless market participants are acting rationally.
My recolllection of EMH is that all information has been reflected in the markets and on average you can’t make money with publicly available information. I am not sure if rational has anything to do with it.

Who defines rational. Was it irrational to think housing prices would never go down when they infact never went down?

The fact that so few can consistently beat the market seems to be evidence that the efficient market hypothesis is correct. The fact that behavioral finance comes up with anomilies is not necessarily evidence of inefficient markets.
The fact that a number of people already cited in this thread have beaten the market over a long-term period is strong evidence that the EMH is not entirely correct. Unless of course you believe that all of those people were just lucky.

A 'rational' view could be one that says, like Buffet has done repeatedly, that company X is truly undervalued by the market, so I will buy it now, expecting that the market will change its valuation of the company later.

Panic can hardly be defined as rational, yet there are times when that describes large swaths of investors.
InsuranceQuant wrote:
Mon Jan 22, 2018 7:33 pm
The efficient market hypothesis is a model of human behavior, an abstraction to make it easier to understand. A simpler statement of the hypothesis might be: "If some people figure out a way to make lots of money relatively risk free, other people will try really hard to copy it". Sometimes when its not easy to "copy it" the ways of making money persist. However, in other times, it is easy to "copy it" and doing so works for a while. If this becomes reinforcing (some people are able to copy it successfully, which makes others want to copy it, which causes the first group of copiers to be successful) it can become a bubble.

The link between EMH and bubbles is pretty subtle but profound. I'd argue the mechanism that causes the market to be efficient is the same mechanism that creates bubbles.
Welcome to the forum!

I think your explanation of the EMH is sound. Put that way, it is not impossible to 'beat' the market, but it's likely to be difficult to do so.
While i don’t assert warren buffet is a coin flipper, I’m not sure if the occurrence of people
Who beat the markets over time as statistically different than those that keep flipping more heads than tails. If their really were wide and consistent known anomalies they would be taken advantage of. The fact that everybody may be temporarily wrong does not make the market inefficient. The information is reflected in the market, but perhaps not correctly in hindsight. Occasionally anomilies are discovered and more often than not the anomaly goes away.

I’m not saying the market is perfect or 100% efficient but for all practical purposes it is efficient enough that it is near impossible to beat it with statistical significance.

User avatar
JoMoney
Posts: 5884
Joined: Tue Jul 23, 2013 5:31 am

Re: What "efficient market" does and doesn't mean

Post by JoMoney » Tue Jan 23, 2018 10:37 am

JBTX wrote:
Tue Jan 23, 2018 9:12 am
... While i don’t assert warren buffet is a coin flipper, I’m not sure if the occurrence of people Who beat the markets over time as statistically different than those that keep flipping more heads than tails. ...
The distribution of a coin-flip 'trading game' looks amazingly similar to that of mutual funds that manage to "beat the market" and those that go out of business over longer and longer periods of time. An interesting visual demonstration at 4:52 in this video:
https://youtu.be/CsRLVZTYpGo?t=4m52s
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
willthrill81
Posts: 5796
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: What "efficient market" does and doesn't mean

Post by willthrill81 » Tue Jan 23, 2018 11:15 am

JBTX wrote:
Tue Jan 23, 2018 9:12 am
I’m not saying the market is perfect or 100% efficient but for all practical purposes it is efficient enough that it is near impossible to beat it with statistical significance.
Then we're in agreement. :beer

I would say that the market is efficient enough to make passive investing, namely, buying an entire asset class, a generally 'winning' strategy.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Post Reply