Investment Process and predictable cognitave errors

I meet with two colleagues each Wednesday. We needed a name for our committee, back when we started meeting and so we named it the Investment Committee (portfolio performance is our business imperative). Each person brings a unique something to the table. Most committees don't seem to benefit from the “sum of the parts” axiom, which, in most other spheres, theorizes that the sum of the parts is worth more than the whole. This committee is different. (You may recall my sailboat experiment. So far I've sold the mast, bow pulpit, and the trailer. The kick-up rudder is, at this moment, on Craigslist. But I digress.

We don't tell each other what we want to hear. Dissent is actually encouraged. We get to ‘go to school’ on each other’s mistakes. My ‘school,’ at 32 years, is the longest running, including 12 years as a stock broker (where I learned from Sir John Templeton, that outperforming the majority of investors requires doing what they are not doing, and to buy when pessimism is at its maximum) and 20 years as Registered Investment Advisor. Brian, our Senior Portfolio Manager, has 14 years’ worth. Justin has 3 (but they’re in dog years). He’s actually studied for the CFA Institute’s Chartered Financial Analyst designation) the last three years while putting his money where his mouth is. He knows mistakes are just an inevitable part of the investment process, “if we don't make too many bad mistakes, we'll do just fine,” he says. Now there’s a man who understands the wisdom of Yogi Berra.

I’m not sure what I did to deserve the privilege of working with not one but two highly calibrated decision makers, who focus constantly on knowing the 60/40 end of any investment proposition; not only do they know the 60 end of the stick from the 40 end, they operate with integrity whether the chips are up or down.

Cognitive errors are predictable, this we know, and they fall chiefly into four buckets: a) investor overconfidence, b) loss aversion, c) narrow framing, and d) overweighting samples of short term data. We’re every bit as capable of making the cognitive errors as the clients we serve, but the difference is, we’re trained to not act on those errors.