Been doing some reading lately and thought I'd share my impressions of three investing books over the next few days.
The first is Joel Greenblatt's The Little Book that Beats the Market, which has been generating a lot of buzz recently.
If you know anything about investing, much of the "Little Book" is pretty basic, but the meat is really in what Greenblatt calls his "Magic Formula" for picking stocks. The basic idea is that you rank every stock (excluding a couple of types of companies for which the formula doesn't work) based on a combination of two factors: return on capital and earnings yield (i.e. the inverse of P/E). Then you simply buy a basket of 20-30 stocks from the top 10% of the rankings (for diversification) and hold them for a year. Greenblatt has backtested this method and says it yields more than 30% annual returns when averaged over multiple years, although it does have occasional down years. He even has a Web site that screens companies for you.
There's a bit more to the book than that, such as how to time your sales to minimize tax impact and why he thinks the formula will keep working even if everyone uses it (a real pitfall of past winning strategies like "Dogs of the Dow").
Greenblatt's "Magic Formula" is basically a pretty standard value screen that leads you, in theory, to buy good companies that the market has put on sale. Sometimes the market puts companies on sale for good reason, but even bad companies can turn around, and that's usually huge for the share price.
What struck me about the "Little Book" is how much I longed to complicate Greenblatt's method. I mean, it can't be that simple, can it? Suppose you ranked the Magic Formula stocks by market cap -- do small companies tend to do better? Well, as it turns out, the Web site lets you specify a maximum market cap for the stocks it picks, so if you want to concentrate on companies of a particular size it's easy to do, but Greenblatt doesn't address what size of company does better. I wished he had done so at first, and perhaps compared stocks based on their dividend yields as well, and maybe addressed timing, but then I realized: he's already promising north of 30%! What more do you want? More importantly, how much better do you think it's possible to do? Some value investors are suggesting applying some additional due dilligence to the companies the automated picker finds, but I have to wonder how much return on your time investment you're going to get by doing extra work. You're already talking about doubling your money every 2-3 years! And I wouldn't be surprised to discover that second-guessing the formula has risks of its own that lead to a lower return.
Overall, the book and the strategy it describes do make sense to me, enough so that I'm thinking about putting part of my IRA into it next year as an experiment. And I've also ordered his previous book. Thumbs up, in other words.
Posted by kindall at April 12, 2006 01:07 PM