The (In)Efficient Market Hypothesis

The Efficient Market Hypothesis (the EMH) has been the central topic of investment finance textbooks for more than 20 years. It has been the subject of much academic research and the researchers have won a number of Nobel prizes. While it has many useful concepts and modeling tools, a number of the conclusions drawn from the hypothesis and its algorithms have been merely self-serving for commercial gain.

Wikipedia provides a good textbook summary of the EMH. At the risk of being too brief, I'll mention some specifics. The EMH has been used to define portfolio structures based on past performance and correlations between various asset classes. This has been used to justify buy and hold strategies, which work for long periods of time and then fail spectacularly. It has created an entire industry of financial advisors who have little in-depth understanding of investments and business cycles, but have a convenient sales template to push prepackaged investment vehicles on an unsuspecting public. An advisor of limited skill can be a great salesman by saying, "After all, Mr. or Mrs. Investor, how can you go wrong with this collection of funds, trusts, etc. that I have put together following a Nobel prize winning recipe?"

The primary problems with the theory go back to the basic assumptions, which use past returns and correlations between the various asset classes. While many financial advisors and their investor clients recognize that "past performance is no guarantee of future results", they are still willing to base investment decisions on those very same past performances. What many fewer recognize is that correlations between various asset classes encounter periods of time when the future will depart significantly from the past. The author has ongoing research in this area, not yet ready for publication.

Sometimes the correlation between two asset classes can be in a constant state of flux. An example is shown in the following graph (courtesy of Bespoke Investment Group) showing the correlation between the dollar and the S&P 500 over the past 38 years.

Neil Hume wrote yesterday, in the Financial Times, "Soc Gen’s James Montier wants Efficient Markets Hypothesis (EMH) and all its offshoots consigned to the dustbin of history before they inflict any more damage on investors." Read the entire article here.

Gillian Tett, writing on the Curious Capitalist blog, reports that a majority of CFAs (Chartered Financial Analysts), whose training and certfication are centered on EMH, do not believe it is valid. Read what she has to say here.

We must be careful going forward not to discard some valuable modeling tools provided by the EMH as we recognize the severe shortcomings of some of the ways in which they have been applied. For example, the author's own research (again, not ready for publication) shows that recognizing correlation variations with time may provide a basis for dynamic strategies (as opposed to static ones commonly in use). Such dynamic strategies are quite the opposite of buy and hold, and much different from the popular rebalancing strategies that have been employed by many for the past couple of decades.

My own work indicates that a well known toast may be appropriate, with paraphrasing: "EMH is dead! Long live EMH!"

Comments

[john s. gordon:]

EMH might have been a valid tool 20 yrs ago before all the hedge funds and speculators added their distortions to the market.
> jack

Jun 19 08:26 AM

[Clive Corcoran:]

John
Your unpublished material sounds intriguing.
In regard to correlation there is some evidence to suggest that because of the recent prevalence of statistical arbitrage and dispersion trading (alluded to by previous commenter) the historical patterns of correlation are changing and subject to sudden alterations.
The EMH does not allow for the fact that much of today's market activity is what could be called meta-market trading in which decisions about how to allocate funds are based primarily on the price information of the market itself.
Market timing is not supposed to work under the EMH and yet clearly it is (arguably) the main dynamic involved in today's macro trading activities as practiced by hedge funds and prop trading desks.

Jun 19 09:03 AM

[Swashbuckler:]

The EMH originated in academia and should be confined to academia, where it can spur debate. Practical applications of information drawn from EMH should be suspect. The awarding of the Nobel Prizes was like ringing a bell. Stay away. IMO.

Jun 19 10:01 AM

[Swashbuckler:]

To paraphrase Warren Buffet on the EMH----"It's great that the universities are teaching that thinking doesn't count when it comes to investing. It makes my job much easier."

Jun 19 10:13 AM

[LKofScotland:]

Is this the same Warren Buffet whose Berkshire Hathaway stock has been pummeled in this downturn? Wall Street and Buffet hate EMT because it destroys their and all the other market timers' time in the sun. Far from being some "academic excercise" it is the basis for index funds! Is it the sole basis for investing? OBVIOUSLY NOT! I don't think efficient market theory posits "stop thinking and be an idiot." It does however posit one VERY profitible theory, namely, that many an investment manager has made more money "sitting on his butt" than in buying this and buying that. It also fits very nicely into an economy such as the USA's which is BOTH commodity intensive AND capital intensive with EASILY the most efficient and extensive transportation network in the world. In short, WE ARE THE LOW COST LEADER. Buffet made his fortune making large bets on individual BLUE CHIP corporations. (American Express and Coke.) Where r they now? Steamrolled in the great debacle of 2008? Perhaps AmEx. Coke? Sugar water looks pretty good even now. Point being there is a wisdom in index funds that says "stop trying to be the manager of the company when all you know how to do is day trade and you don't even do that well." Capitalism (especially its aweful side) far from being put at risk by Obama and the Dems is now going to run us all into the ground. I can hardly wait for $2000 a barrel oil as Iran's counter-revolution turn violent, expansionist and nuclear.

Jun 19 10:33 AM

[Ferdinand E...:]

Neil Hume and Gillian Tett in the Financial Times...in the Financial Times of all places. Why not Batman Comics. I know, their boss, Martin Wolf writes in that.

And as a disbeliever in the EMH, don't forget the book by Professor Andrew Lo of MIT: 'A Non-random Walk down Wall Street'. I wonder who was dumb enough to pay any attention to that one, and its silly belief that juggling a few regressions will enable you to make millions in the share market.

Burton Malkiel and Eugene Fama accept the EMH, and if anybody in this or any other forum thinks that they know more than Eugene Fama on that subject, please don't make the very bad mistake of finding yourself in a seminar room with him. LKof Scotland above is correct when he says that "many an investment manager" has made more money sitting on his behind than buying this-and-that, and if you transform that slightly to... the average investment manager can make more money...you have an application of the EMH - and that application can be proved.

Jun 19 11:17 AM

[Steven Hansen:]

anyone who believes that the market is not being driven by excessive greed and manipulation at this point has been asleep for the last 10 years. when the market is not acting in a free and non-monopolistic manner, emh is a non-starter.

john, i await your new modeling instructions for emh as i have always been extremely unbalanced in my investing approach. this has served me well in the past, but with all the interventions going on today i worry that any investing approach may get scorched in the short haul.

Jun 19 11:44 AM

[John Lounsbury:]

Commenters - - -

Thanks for weighing in. I especially appreciate the effort and detail put in by the Last King and Prof. Banks.

As I hope my article implied, the EMH has many vauluable features and, although it will be refined over time, it will not go away. However, as has always been the case, there will be investors (a few) that outperform the market averages significantly for decades, just as there are investors (many) who underperform the market over long periods of time. The case that, if you are an average investor, you can not do better, on average, than invest in index funds is an obvious truth. The average of all investors is the average investor. Fortunately, we are not all average investors, and that is what makes a market.

Some are not average investors, Warren Buffet being notable. In spite of underperforming the market for the last 3 and 6 months, he still has an exemplary record. I'll give the approximate period returns for recnt returns: 3 months -14%, 6 months -13%, 1 year +3%, 2 years +20% and 5 years +18%. Berkshire is approximately even for the past five years, while the S&P 500 is down more than 18%.

Buffet has had several previous periods of under performance, which led some to dismiss his previous success as coincidental luck. He has always returned to outperformance, producing an annual return approximately double that of the S&P 500. For example, since 1/1/1990, the annual compounded return of BKA-A has been 14.8%, while the return for the S&P 500 has been 6.3%. I'll place my wager on a return to outperformance over the next few years.

Jun 19 12:13 PM

[rosey99:]

Nice article. I believe that at least one of the guys who founded LTCM was a Nobel Laureate. And that turned out well, right? EMH will probably go by the wayside, only to return with a new name, in a new time of future complacency.

Thanks for the interesting piece, it's good to know that critical thought is not completely dead.

Jun 19 05:01 PM

[TinyTim:]

John-
Looking forward to your "correlations between various asset classes encounter periods of time when the future will depart significantly from the past"

Glad you point out the longstanding irony between EMH and:
"past performance is no guarantee of future results"

Listened to an interesting podcast last night from Justin Fox, who just published a book The Myth of the Rational Market
www.slate.com/id/22207...
I haven't read the book yet and he's a historian rather than an economist (for what that's worth), but his comments sound relevant to your discussion here.

Jun 19 05:04 PM

[contracontr...:]

Professional investment managers didn't outperform the market in recent decades,because they were the market themselves.
If there was a significant player in the market,who is not thinking (investing passively and indiscriminately) like Index-ETFs, relative performance would very likely increase in favor of pro-investors (except always invested long only funds) over time.

Low cost funds save money for the average investor,but there is a price to pay,he's positioning his money on a lower knowledge (and informations) level. In the end,the performance of stock indices doesn't only depend on the performance of the companies involved,but also on the timing and prices of changes (addition and deletions) in the index.
There are many possibilities for pro-investors to make money at the expense of Index-ETF investors,if they become a big enough player in the market and it will be much easier for them,than trying to beat other pros.

IMO people will be disappointed with S&P 500 ETFs 10-15 years from now.

Jun 19 06:08 PM

[dybydx:]

the author demonstrated absolutely no understanding of EMH.

EMH never said buy n hold is a superior investment strategy. In fact semi-strong form of EMH claims that fundamental or technical analysis can NOT generate superior return. although you CAN generate return using buy n hold.

Jun 19 09:06 PM

[Living4Divi...:]

John - Excellent article.

Jun 20 09:01 AM

[donfm:]

Good article --

There are many good theoretical ideas to take fronm EMH. Everybody needs to remember that it is a theory based on alot of unrealistic (real world) assumptions.

I believe that there will be many changes and new theories about market dynamics to come out of this current crisis. One is that, obviuosly, the market is not efficient, but also the phsychology of trading and its reaction to market events (crisis, corrections, bubbles), and the realization that these events happen with much more frequency than EMH predicts. Much of modern market dynamics has changed over the past 20 years. Technology has enabled market players to trade more frequently and speculate more frequently. Greed and fear play a big role market bubbles and corrections.

Jun 20 11:15 AM

[mkreisel:]

Last year was especially rough because correlations among asset classes were either 1 or -1.

Jun 21 03:20 AM