There are a lot of situations in which investors find it helpful to use models. By a "model," I mean either a quantitative construct (like a valuation spreadsheet used to estimate the net present value of a company's future cash flows) or a mental model (i.e. a set of heuristics and shortcuts used to quickly come to a conclusion about an investment opportunity). But although they are useful, value investors should use them carefully. Here's why.
A perfectly spherical horse
There is a joke amongst scientists and engineers that goes something like this: An avid gambler comes to his physicist friend and asks him to predict the outcome of a horse race using his knowledge of mechanics and gravity. The physicist comes back with a ranking of how well each horse should do. The gambler goes off and bets on the number one horse, and is disappointed when it loses by a wide margin. When he confronts his physicist friend, the latter responds, "well, my calculations were accurate if you assume that the horse is a perfect sphere in a vacuum."
This (very old) joke illustrates a key problem with the use of models, which is that even the best ones are only approximations of reality, and that overreliance on them can spell disaster, since you're betting the house on the perfectly spherical horse. Although this dated piece of humour might seem a little silly, we can see how the same logic applies to the use of models more generally.
Companies are not perfect spheres in a vacuum. They are complex operations with many moving parts both internally and externally. For example, you might have a detailed spreadsheet that anticipates a transport company's earnings to grow by a steady 5% each year, but fail to account for extrinsic factors that throw a wrench into these assumptions (rising gasoline prices, falling demand for freight, pandemic-induced global lockdown, etc).
So any model will only be a rough approximation of reality. The danger arises when one forgets this simple fact, which happens much more often than you might think. As humans, we place a very high premium on certainty, so it is very tempting to embrace any construct that offers it, however inaccurate it eventually turns out to be.
The marriage between human ingenuity and our increasingly capable computers has only compounded the problem of "mistaking the map for the territory." It has become very easy to create highly complex and detailed models that lull the user into a false sense of security and seduce users into believing them implicitly, so always remember to apply your own "common sense" test to any result you get from a financial model.
Disclosure: The author owns no stocks mentioned.
Read more here:
- Nassim Nicholas Taleb: The Federal Reserve Cannot Prevent All Problems
- Mohnish Pabrai on How He Stumbled Into Owning a 60-Bagger Stock, Part 3
- Mohnish Pabrai on How He Stumbled Into Owning a 60-Bagger Stock, Part 2
Not a Premium Member of GuruFocus? Sign up for a free 7-day trial here.
