“People calculate too much and think too little.”
This is a quote from Charlie Munger, Warren Buffett’s right hand man, and a world-class investor in his own right.
It is one of my favorite quotes and has rung true throughout my investment career. In my experience, many financial professionals accept numbers too easily without fully understanding assumptions, sensitivity to inputs, and general rules of economics and competition.
We are always looking for ways to better design client investment portfolios. Every year, we are bombarded with new investment approaches, new products and new trading strategies to beat the market. Most new products can be tossed aside immediately, but a few require more detailed investigation.
We necessarily perform calculations to back-test the new idea. That’s 20% of the work, and our first checkpoint to determine if an idea has merit. The remaining 80% of the work is thinking. Here are just a few of the questions we ask:
1. Can the source of enhanced returns be arbitraged away when many investors adopt such a strategy?
Generally, the more assets devoted to, and the more firms employing a strategy, the lower future returns. Some strategies have an enormously large capacity to manage assets before returns begin to suffer, while others, especially those with high turnover, are extremely sensitive to asset levels.
2. What is the risk story? Since expected returns increase with risk level, can we identify a non-diversifiable risk factor that an investment approach is exploiting?
Risk factors provide the basis for a well-designed portfolio because they are much less impacted by the number of investors and amount of money invested to exploit them. Common risk factors include the value and small-cap premiums for stocks, and the term premium for bonds.
3. Do back-tested results rely on constraints that no longer apply?
It’s hard for us to accept back-tested results for a strategy when transaction costs were much higher in the past or when no trading vehicle was available to exploit the strategy. Enhanced performance in the past is most likely a function of these constraints, which of course no longer apply.
4. How is the strategy expected to perform in a bear market or financial crisis when the correlation of all risky securities goes to one?
There are at least a dozen more questions, all of which help us to follow the spirit of Charlie Munger’s advice in designing an investment portfolio for long-term success.