EMH baffles me. I haven’t absorbed Fama/French, but I have to imagine they have published thorough studies. (They did win some economics prize given in honor of Nobel, eh?) Still, is Efficient Markets a hypothesis in the scientific sense? My casual understanding is that a scientific hypothesis has to be testable. Given that market efficiency is somehow a measure of the degree to which all information is reflected in stock prices, how can one rigorously test any market efficiency hypothesis other than from a position of omniscience? I’m not saying it’s not worth thinking about and studying — I’m only wondering whether any such hypothesis is testable in the scientific sense.

Taking another stab at EMH, from another angle, the hypothesis seems to hold the belief that market actors are net rational. My life is full of watching (and sometimes participating in) irrational microeconomic decisions. Humans make enough irrational decisions to undermine anyone’s hope that stock market phenomena can be free of irrationality. What I’m saying is that even if one’s point of departure is that all private and public information is in the hands of the market actors, whether those actors will always act rationally upon that information is dubious.

One problem I see is, even if the market is efficient, the individual components don’t have to necessarily be efficient. Stocks become “hot” frequently, and well-established corporations fall out of favor because their profits may not increase fast enough.
And, as Ron points out above, if the EMH is real, how did W. Buffet amass so many millions?

@ Steve – “some economics prize given in honor of Nobel” .. haha, zing! No, it would appear that they have not won. http://nobelprize.org/nobel_prizes/economics/laureates/ . In my mind economics is not a science although it tries to be (just like sociology does). The reason is that the hypothesis is essentially untestable.

There’s a bigger problem though. In the world we are essentially dealing with four different systems. Few agents with few degrees of freedom (solar system, most engineering), Many agents with few degrees of freedom (economics, sociology, …), Few agents with many degrees of freedom (biology, group dynamics, …), and Many agents with many degrees of freedom (thermodynamics, statistics).

FF is analytically tractable. MM is statistically tractable. Neither FM or MF can be solved with those methods. They can be computationally treated but the problem is immense.So what people do is that they try to see how much economics diverge from a MM model. That’s the Fama and French study. The hypothesis is not correct, but it is almost correct statistically speaking. I think 90% of the correlation could be explained with the three factor model and somewhat less with a one factor model. Of course if you go into medicine, you are not going to sell any drugs that are only 90% correlated. That would be pretty bad. Medical doctors like to see 95%. Chemists more than 99%. Physicists generally insist on several 9’s e.g. 99.999% in order to say that something is scientifically proven.

One might say that the market actors are rational on average in the long run. But that is true by definition.

@ Ron – Those extended periods are exactly why I’m fearful of index investing (Nikkei 1990-2008 is worse than nowhere). They happen when the large companies in the index are going nowhere while the smaller companies cancel each other out. What is especially bad is when index investing is sold to fund buyers based on projecting 1981-Now returns, around 11%, into the far future. That’s just unethical.

@Adfecto –
A few points

a) The market is efficient insofar that it represents the money weighted average of the smart analysts and the stupid analysts. Sometimes even the smart people are stupid. Some smart people saw the tech bubble as soon as 1998. The wise people left. But some went short and got squeezed out as the folly continued for another two years. Also some smart people went stupid. This was repeated with the CDOs of late.
b) Institutional managers are also people. Career wise it is prudent to not veer to far off of the course of the herd. If a manager is right while the market is wrong, he is taking a chance. If the manager is wrong while the market is wrong, he is wrong along with everyone else. Job secured. It is important to note that many analysts and managers depend more on their paycheck and fees than their management skills.
c) Thus creativity (and market beating returns) are found in the smaller outfits you have never heard about.
c) Highly liquid stocks like blue chips are closely followed. However, stocks which trade a daily value of maybe $50,000 are not (a volume of 5,000 at $10 say). These are not worthwhile for institutions to dabble with since institutions need to move millions each day. This is where individuals and smaller outfits can shine picking up $5,000 here and $5,000 there.

So to reiterate a point I made in the comments above. EMH is an MM (statistical) model of an MF (algorithmic) world. It is a model of reality that is demonstratively an approximation. We can show this with simple deductive reasoning. Still for a herd of a few thousand professional money managers that move big currents of money it is a pretty good model.
Thus individual investors should probably stick to the FF part of Wall Street where institutions rarely go. E.g. individuals should play in the pond and not the ocean. Such companies are not followed by analysts, have little institutional ownership, and have large insider ownership. You can minimize downside risk by picking them up at low P/B values. There are a few other things to look for as well. Maybe I’ll do a post on it right after my favorite pan cake recipe 🙂

Yeah, but I think individuals should stick to their philosophy for the best results. It’s the trend or philosophy chasers that fail in the long term.

I really think you guys would benefit from reading the Bogleheads website as they are the -kings- of index investing.

Even if you disagree their knowlege of topics such as taxation is incredible.

Is it possible to both believe in EMH and yet believe that it misses the bigger picture?

I believe in EMH as I do believe that markets tend to reflect the information that could be accurately known at that time. This does not mean that mistakes are never made. Obviously a big mistake was made during the housing bubble, but all this says is that the markets cannot predict the future with absolute certainty, nor can they predict the exact outcome of the vagaries of easy money and credit expansion! Perhaps another way of looking at it is that the housing bubble wasn’t a bubble until it was, via EMH, but by looking a little bit further, at economic fundamentals and at rising debt levels, one could have at least have seen a dangerous trend forming, even if it would have been impossible to predict just how things would have turned out and when the S would HTF.

Interesting take on EMH and index funds; what do you think about using a value screener like Greenblatt’s Magic Formula? He makes some good points about index funds’ market capitalization weighting systematically playing into buying over and under priced stocks. Another interesting scenario is the EMH and unforeseen events. Take for example Steve Jobs leaving Apple and AAPL shares falling and gaining back $15 billion of market value in a day or two: http://www.bestonlinebrokerage.com/steve-jobs-apple-departure-and-the-efficient-market-hypothesis/

Hi, I don’t think many people, if any at all, think the market is strongly efficient although your comment “Whoa! Hold your frictionless spherical horses right there!” is now a personal favourite comment on this subject.

There is a lot of common sense from Bob with his comments “Diversify your investments across several asset classes…” etc.

David F. Swensen’s model of investing that he published in his ‘Unconventional Success’ books is a role model for me.