29 July 2008

The difference between predicting the future and fitting a model to past behavior ...

... is that the former is extremely difficult and the latter is almost trivially easy. However some people will be surprised by this Freakonomics post on Jim Collins' "Good To Great" . For those of you who don't remember, Good to Great is based on an analysis of some companies who were once just okay then went to outperforming the market over a long period. It compares these companies to other companies working in the same fields and distills what they did differently to some rules, the Good to Great Principles (more on this below). The Freakonomics blog post looks at the performance of the Good to Great companies:

  • ...It looks like Fannie Mae is going to need to be bailed out by the federal government. If you had bought Fannie Mae stock around the time Good to Great was published, you would have lost over 80 percent of your initial investment... Another one of the “good to great” companies is Circuit City. You would have lost your shirt investing in Circuit City as well, which is also down 80 percent or more. Best Buy has cleaned Circuit City’s clock for the last seven or eight years....Nine of the eleven companies remain more or less intact. Of these, Nucor is the only one that has dramatically outperformed the stock market since the book came out. Abbott Labs and Wells Fargo have done okay. Overall, a portfolio of the “good to great” companies looks like it would have underperformed the S&P 500.
How did this happen? Leonard Mlodinow describes how random variation is routinely mistaken for the results of people's decisions in his book "The Drunkard's Walk: How Randomness Rules Our Lives".

Of course the fact that the management techniques recommended in Good to Great are based on faulty analysis doesn't make those techniques any less (or more) effective. However it will be interesting to see the reaction of those people who promoted those management techniques because they raised the stock prices of the Good to Great companies relative to the market. Will those fans now champion doing the opposite of what the book recommends? A quick look at the Good to Great recommendations shows they are not harmful enough to make companies underperform the market as badly as the Good to Great companies have done.

2.1 Chapter 1: Good is the Enemy of Great
2.2 Chapter 2: Level 5 Leadership
2.3 Chapter 3: First Who, Then What
2.4 Chapter 4: Confront the Brutal Facts (Yet Never Lose Faith)
2.5 Chapter 5: The Hedgehog Concept (Simplicity Within the Three Circles)
2.6 Chapter 6: A Culture of Discipline
2.7 Chapter 7: Technology Accelerators
2.8 Chapter 8: The Flywheel and the Doom Loop
2.9 Chapter 9: From Good to Great to Built to Last
.

This article gives a plausible explanation of how Good to Great works its magic. Here is an excerpt:
  • First of all, the good-to-great principles are true in the same way a horoscope is true. They are fairly generic and thus we all apply them from our own viewpoint to make them true. I believe that some "Good to Great" readers that love the book may be suffering from the Barnum effect. The principles Collins proposes aren't bad ones, but they are ambiguous and open to interpretation, which in effect decreases their usefulness. For instance, Collins says good-to-great companies practice "First Who, Then What," which basically means "hire good people." I'm willing to bet no one read the book and said "Eureka! I've been hiring slimy weasels when I should have been hiring top performers. That is why we aren't a great company."