Wrong but very useful: the efficient markets hypothesis

Personal finance tip to folks who have some money saved up: Don't throw the baby out with the bath water by completely dismissing the Efficient Markets Hypothesis (EMH). See the end of this post for what I mean in practice.

I cannot resist commenting on an article in today's Globe & Mail (Taking Stock, Brian Milner) entitled, "Sun finally sets on notion that markets are rational". Milner is reviewing a new book, by Justin Fox, whose cover appears in the figure. In summary, Milner writes that EMH, "has finally been buried under an avalanche of unforeseen calamities, uncontrolled greed and other seemingly irrational behaviours it couldn't possibly explain."

What's this all about? Well, the idea behind EMH is that you can't beat the market. Efficient means that the price of an asset that you observe in the market is all you know and you can't predict the future price. Why? Because if a stock or some other asset is under-priced, somebody would take advantage of that and buy it, pushing up the price until there was no longer any profit to be made. So the prices you observe in the market are the "right" prices in some sense. Economists have thought quite deeply and extensively about this idea.

Now, we also know in practice that this hypothesis is wrong. But it is nevertheless an extremely useful hypothesis in the sense of the physicist Wolfgang Pauli, who once famously ridiculed a theory by saying that it, "wasn't even wrong". Let me give you an example of why we know this hypothesis is wrong. As a trader I would often get customers who were not interested in maximizing their profits solely in the financial markets. For example, they would tell me to buy X shares of a stock now. I might tell them if they took a few hours to work their order they could get a better price. They would come back to me and say, well, my staff is busy and they need to work on other important things, so I don't care if you buy X shares and drive up the price somewhat, just fill my order now, please. After I fill the customer's order, I'm 90% certain that the share price is going fall back down, contrary to the efficient market hypothesis.

So why is the EMH wrong but still useful here?

The answer is that anybody else looking at the market for the stock cannot know that the stock is going to rise or fall. They don't know if I'm finished with my order, or if I have a lot more shares to buy. Unless they can deduce something about what I am doing, the market appears to them to be efficient, meaning there is no profit, on average, to made by betting on the direction that the stock price moves. Furthermore, EMH is useful because it implies that you should always have a healthy respect for the market and the prices you observe in a freely trading, liquid market. It leads to the following rules, which I think of as the trader's practical version of the efficient market hypothesis (applicable to liquid markets):

The market usually knows more than you do. You will lose money if you casually think otherwise.

And a corollary,

Unless you do your homework, the best you can do is rely on the prices you observe in the market.

So, what practical advice can I give, especially to all those people who have some money saved up but have better things to do with their lives than worry about their investments? Well, I'm pretty confident (i.e. I would tell my mother to do this) that markets are still efficient enough to say, don't pay somebody to pick particular stocks, bonds or other investments. Invest passively in index funds that have low management fees. Think twice before spending a lot of your own time on stock-picking, market-timing or trading instead of on something else that might be more valuable to you and other people in your life.

"buying opportunity": not just an insensitive statement from Harper, but unjustified

Anybody who knows financial markets would know that unless you really do your homework, you cannot say when there is a 'buying opportunity'. Certainly it would be wrong to say that, based on a casual application of the (economically unjustified) notion that stocks always bounce back after they go down, there was a buying opportunity back on Tuesday. That's what Stephen Harper did. Professionals who have done a lot of trades will tell that, unless you have done your homework and understand why a market is not being efficient, you won't make money betting on any simple strategy like mean reversion.

Now this post is not just about Stephen Harper. It's about commentators like Susan Riley of the Ottawa Citizen, who wrote, "...his logic is impeccable: stock markets go down, but eventually go back up", and Andrew Coyne who wrote of, "...Harper’s perfectly sensible observation that the present panic on the stock markets presents a remarkable buying opportunity ". Experience trading in the markets will tell you that, unless you figure out why somebody sold a stock down too aggressively, you don't have a better than even chance betting on it bouncing back. Right now, I'd say very few people, if any, really know exactly how bad the financial crisis is, and certainly not casual commentators who are mainly focused on Canadian politics.

Don't tell me that if you wait long enough, stocks will bounce back. If you wait, you have inflation to worry about. Look at the inflation adjusted return on stocks in the 1970's. Moreover, you can only wait until you die. In the long run we are all dead, as Keynes observed. For single stocks, consider an extreme example -- look at Lehman Brother's share price. Is it going to bounce back?

No, Stephen Harper's statement was not a sensible or logical statement that was simply insensitive. It was just wrong. As an economist, Harper should know something about market efficiency and how difficult it is to predict whether a stock price will go up or down. But I guess it was just too good an opportunity to try to express to voters his belief that the economy is probably okay.