On Chaos, Butterflies, and Robustness
Robust portfolios that have the greatest probability of performing across various macro outcomes can effectively mitigate and navigate the unpredictable twists and turns of dynamic markets and economies.
In the early 1960s, a meteorologist undertook a mission to predict the weather. Unique to his approach at the time was his extensive use of computers to assist in the predictions. He began running simulations using a fairly simple computer model. Before long, he wanted to run longer simulations, but instead of running the last iteration from its beginning and extending it, he simply started it from the middle. After all, computing power was hard to come by in those days. To simplify matters, he would type in the numbers from the first data set and use that as the initial conditions for the longer prediction. The model was the same, his programming the same, yet the resultant weather patterns diverged wildly.
As is often the case in life, the changes seemed small, but the consequences out into the future were big. His computers stored data to the accuracy of six decimal points. To save space, our meteorologist decided to shorten his typed in conditions to three places. His small simplification, less than a 0.1 percent difference, completely altered the results. Chaos ensued. In 1972, our meteorologist gave a talk regarding his research findings to the American Association for the Advancement of Science titled “Predictability: Does the Flap of a Butterfly’s Wings in Brazil Set Off a Tornado in Texas?” Thus, the “Butterfly Effect” was born.
One complex system we know intimately that is often subject to butterfly effects and seemingly random chaos is the financial markets. What we can learn from our curious meteorologist, Dr. Edward Lorenz, a pioneer of chaos theory, is that the proper approach to managing risks in the market may not lie in ever more complex models and predictions, but something else entirely. The most astute approach often lies in building robust portfolios, or those that have the highest probability of performance across a variety of macro outcomes. When investors make explicit bets on the various stories of the day, whether healthcare reform, tax cuts, central bank action, or economic forecasts, they often find their portfolios in chaos when the proverbial butterfly flaps its wings. Investors are better served by a disciplined process of adding risk exposures when, according to one’s methodology, they are attractively compensated, while decreasing risks when compensation is poor. Fundamental research on individual securities is far more solid a base for modeling a security’s future price than broad political or macro developments.
To be sure, in-depth security analysis incorporates to varying degrees macroeconomic analysis in seeking to identify the best opportunities that may provide robust risk-adjusted returns over a market cycle. The foundation of this rests on two important capabilities: one, a bottom-up, relative value analytical framework and, two, an acute focus on quantifying and assessing risk/reward tradeoffs. Comparing the risk and potential return characteristics of securities across sectors, industries, and even asset classes can help avoid—not always, but perhaps more often than not over the long run—the chaos experienced by many investors. We believe this approach to investing provides an opportunity to generate a robust total return across many macro outcomes while allowing investors to sleep at night.