Most of us think of Albert Einstein or Stephen Hawkins when we hear the words “String Theory”. We envision nerdy scientists exploring three, four, or more dimensions of time/space when researching Black Holes, for example. Physicists use algorithms to simplify and interpret data to prove their theories. Interestingly, the use of algorithms has become much more prevalent in the investment business and can dramatically affect your portfolio.
An algorithm is “a set of rules for solving a problem in a finite number of steps, as for finding the greatest common divisor.” For example, when we are looking for a mutual fund or stock to buy, we use a set of our personal decision rules (internal thought processes or “algorithms”) to weed out the list and decide on the best funds for our needs. The good news is that this approach is easy to do, but the bad news is that we humans do not have the mental “bandwidth” to assess the thousands of choices before us. Here is where companies like Morningstar and Fidelity come to the rescue. Using algorithms and high speed computers, they can take our preferences…growth funds with high P/E ratios, for example ….and quickly serve up a selection; this is a great help for the average investor. But, there is a downside. Remember the Flash Crash on May 6th, 2010 when some 9% of all trading on one exchange disappeared into thin air?
In his 2011 TED Talk, “How Algorithms Shape our World” Kevin Slavin (Professor at MIT) estimates that there were 2,000 Physicists on Wall Street who’s job it is to find ways to develop proprietary algorithms to help examine millions of trades and detect, like Morningstar does, the essential information to support an investment strategy. Unlike our leisurely stroll thru the Morningstar site, algorithms with names like the “Boston Shuffler” and “Carnival” operate some 57x faster than we can click a mouse while processing millions of bits of data per second. At the institutional level, such activity makes up 70% or more of all trades executed….all without human intervention. What could go wrong? When two or more competing algorithms draw different conclusions, chaos can happen – this is one theory about what caused the Flash Crash of 2010.
I suspect that we will see more such trading techniques, especially as the global capital markets continue to evolve in their complexity. We are already seeing the growth of so called Robo Advisors who offer similar algorithm-based services to the individual investor. Regulators are increasingly concerned about the risks to the uniformed (see ” ‘Rob Advisor’ Investing Apps Have Risks Too, Regulators Warn”). I am not convinced, however, that such technology alone can be successful in the long-term, for there is value in human intuition and tacit (learned) knowledge. Our Diogenes Fiduciary Management System™ will be the first of its kind to combine algorithms with tacit knowledge to support fiduciaries with their decision making process. It may sound simple to design, but believe me, it is not easy….