April 14, 2003

Random Walk

One of my all-time favorite non-fiction books is a classic by Burton Malkiel called "A Random Walk Down Wall Street". It is basically a mathematician's look at the stock market. Along similar lines is a book by John Allen Paulos called "Innumeracy" which is a mathematician's look at the everyday world.

Anyway, the point of the book is that the only way you make more money in the long-run in the stock market is to take more risk. Aside from insider trading, of course, or any number of the kind of stunts pulled by GW Bush (late of Harken Oil, Arbusto Energy, etc.) and his cronies long ago that make Hillary's cattle futures swindle look like peanuts.

Malkiel says you really can't time the market consistently, and you really should just put your money into an index fund. In a given year, you'll beat at least 80% of the rest of the funds, and in the long term, you'll beat 99% of them. Malkiel also talks about the myth of the successful corporate CEO.

Imagine you have 128 people, and each of them flips a coin. Let's just say for the sake of argument that 64 of them get heads. Take those 64 and have them flip again. Let's say 32 get heads. It will be nearly half every time anyway, so this is a good approximation. Now we keep up with this process until we are down to the two people who have flipped heads 6 times in a row. They flip, and one of them gets heads.

So there is 1 person left out of your original sample who flipped heads 7 times in a row. Would it be accurate to say that this person is a talented coin flipper? Of course not. Now look at the stock prices of companies. They are distributed above and below the market average (so if the market as a whole goes up, about half will be above that average and half will be below the average even if it means the low performers also go up in price).

Over the years, some of these companies beat the averages and some fall behind the averages, and these numbers are similar to the behavior of the coin-flipping experiment. Now, suppose you have a company that beats the market averages seven years running. Would it be fair to say that is an unusually well-run company? Just like the coin-flippers, the answer is "of course not" (keeping in mind we are just tracking changes in the company's price ... the "initial" price is set by the market which factors in the company's quality). Because the distributions look identical statistically.

The moral of the story is don't invest in "hot" companies or sectors of the market. Invest in index funds, with the level of risk inversely proportional to your time horizon. I'm glad I didn't read this 15 years ago. At the time, I had a gift of a few thousand in savings that had built up during my lifetime thanks to my grandparents (who contributed a couple hundred each year on my birthday and gave it all to me when I turned 18).

On the advice of a friend who worked there, and because I was into computers and stuff anyway, I invested about $4k in Intel. Six years later, it had doubled 4 times. I sold it and used it as a hefty down payment on my first house.

Posted by Observer at April 14, 2003 07:16 AM

Comments on entries can only be made in pop-up windows while those entries are still on the main index page. Sorry for the inconvenience this causes, but this blocks about 99.99% of the spam the blog receives.

Yeah, that's what they told us in finance in business school. Barring insider information, investing in specific stocks involves assuming unnecessary risk vs investing in index-based mutual funds.

I've been giving this advice to anyone who asks for years. One friend poo-pooed it because her father is a broker for Paine Webber, and he keeps giving her specific fund advice. I laugh.

Of course, I'm not smart enough to take my own advice, only about 40% of my stock-based retirement funds are in index funds. The rest are in more aggressive funds. (plus my investments in Seattle real estate, a MUCH more profitable return than the market recently.)

The big problem with any fund vs the index fund is the fund has to pay the fund management, but an index fund just invests in the whole market equally, no "Wall Street Whiz" required.

Posted by: Humbaba on April 14, 2003 08:57 AM

I've always thought of it as gambling. And it reminds me of the Simpsons (like ... everything).

Chief Wiggum: What are the odds that I can grab that piece of cheese before the anvil falls?
Officer Lou: About a million to one, Chief.
Wiggum: I like those odds.

Posted by: Polerand on April 14, 2003 11:34 AM

Investing in an broad index mutual fund is basically investing in the US economy as a whole. Over the long run, it's outpaced every other investment.

Posted by: Humbaba on April 14, 2003 12:16 PM