Just as I was moving several weeks ago I got a call from my stock broker. He worked for a major brokerage and was leaving to form his own one-man investment advisory service. He wanted me to move my account with him. This scenario is fairly common in the investment world, at least in the so-call high net worth
portfolios. I used to date a girl who worked for an investment advisory company as an assistant. The bulk of her frustrations (at least as she related them to me) were all about advisers who joined the company and their skills with regard to getting their former clients to become clients at the firm for which she worked.
I personally like my broker, but I’ve got a pretty fundamental difference with him about portfolio management. I don’t know to what extent he believes in the underlying theory, but I often got calls from him that I should probably sell or buy a security because it had reached it’s support
or resistance
level. These are terms used by so-called technical analysts
, which should more properly be called trend analysis
according to John Allen Paulos. A large segment of investment advisers out there follow this model. The classic mantra for the stock market is to buy low and sell high. Technical analysis is all about watching the stock price and using some mathematical rules about the stock price to determine when the price is low and when the price is high. Pretty much the only input to these rules is the stock price.
There’s some value to the technique, but in a worst case scenario of a flat but volatile market, a person will lose money due to all the transaction costs associated with buying and selling. More generally, technical analysis may not be much different in its results from investing in a broad-based Standard and Poor’s 500 index fund.
My philosophy is generally to invest in broad-based index funds spread out across a number of asset classes. Then you sit on it for a long time and occasionally re-adjust things to get it back to your overall percentages if some part gains a lot or loses a lot of money (throwing off your percentages). But I also like the idea of watching macro-economic trends, and using some percentage of my portfolio to make bets on those trends. For instance, I’ve watched the explosive growth in emerging markets (a.k.a. developing countries making the transition from the third world to the first world) and years ago decided that a portion of my portfolio needed to be invested there. Or earlier this year seeing the coming housing market meltdown, I put some of that money into a fund that shorts financial companies. Basically, betting that over the next year banks overall will take a lot of losses from bad mortgage investments.
Anyway, to refresh my thoughts regarding whether I should jump to the new firm or not, I decided to pull out A Mathematician Plays the Stock Market and read through it again. Mostly I focused on his chapters on technical analysis, fundamental analysis, and sections on the efficient market hypothesis.
The efficient market hypothesis is a theory that fascinates me. The hypothesis is that the prices on the stock market reflect all information about each stock. To illustrate, say the newspaper reports that the F.D.A. has approved a new drug by GlaxoSmithKline (G.S.K.). That would be good for G.S.K., right? And you’d want to invest in their stock, right? Perhaps not. The hypothesis is that almost immediately investors are going to start paying more for the stock. Meaning that if you go buy it now, the higher price offsets any extra money you’d make because of the drug approval.
Paulos repeats a joke though that shows that a market can’t be 100% efficient. Two efficient market theorists pass a $100 bill laying on the street, but don’t pick it up. Why not? If the $100 was real, someone (the market) would have already picked it up.
The implications of the theory to a common investor such as I am is that (if the theory is true) everything I do in the stock market is essentially gambling. It’s all luck. I am just flipping a coin as to whether I am going to do better than the broad market, or worse, with each individual bet I make. It doesn’t mean I can’t do better. It just means that I might as well not bother with almost any strategy at all. Picking stocks at random will have about the same effect.
Now, there are strategies that can lock people into certain paths. They essentially reduce the betting involved. But making it less random also means that you move in lockstep with other people pretty much.
One thing Paulos also writes about this hypothesis is that if everyone believes it, it can’t be true. In order for it to happen, there has to be a number of people snapping up tidbits of information to move the price of a stock. But if everyone believes the theory, they won’t bother, and the price never moves in accordance with new information.
My one big beef with the book is that a lot of it seems like argument by analogy. Here’s one thing about the stock market that looks a lot like something from mathematics. In other words, correlation. But just because two things are correlated doesn’t mean there’s a connection. In the future that past correlation may fall apart. It’s a point that Paulos brings up with regard to other aspects. But I’m not sure he sees it with regard to his own words. On the other hand, he may know more and know that there is an underlying causal connection, but just isn’t bothering to explain it because this is a book for non-mathematicians. For our purposes, showing the correlation may be sufficient.
Title: A mathematician plays the stock market
Author: John Allen Paulos
Imprint / publisher: Basic Books / Perseus
Format: Hardcover
Length: 216 p. (includes index)
Publication date: 2003
ISBN-10: 0-465-05480-3
Subject: Investments — Psychological aspects
Subject: Stock exchanges — Psychological aspects
Subject: Stock exchanges — Mathematical models
Subject: Investment analysis
Subject: Stocks
LC classification: HG4515.15.P38 2003
