Rules-Based Decision Making and What Wealth Advisors Can Learn from Robo-Advisory Solutions
A lot is being made of robo-advisors these days. Custodians and brokerage firms like Charles Schwab, TD Ameritrade, JPMorgan, and Goldman Sachs have joined the likes of Betterment and Wealthfront in offering low-cost, automated solutions to rebalancing and tax loss harvesting. The growth in robo-advisory solutions and the AUM in them is impressive and makes sense–as the millennial demographic enters the wealth accumulation phase of their careers and lives, their familiarity with and preference for technology-based solutions points to this trend continuing.
For advisors, the ability to leverage these tools to automate transactions and processes will lend scale and efficiency to their investment management operations.
Following the Investment Algorithm
One common thread among robo-advising solutions is the usage of rules-based decision making, or algorithms in computer science terms. An algorithm is a process or set of rules to be followed in calculations or other problem-solving operations. So, given a certain level of risk tolerance, robo-advisors set an asset allocation, rebalance over time to the asset class targets determined in the previous step, and harvest tax losses–all based on a set of pre-determined rules and inputs. They all have simple, highly-repeatable investment processes. And a repeatable process is exactly what you want from your investment manager.
This approach may sound perfectly reasonable at face value, but it begs the question as to why. The answer lies in the field of behavioral science where decades of research into probabilistic reasoning have exposed the extent to which people routinely base their forecasts and judgements on flawed rules of thumb as opposed to careful examination of evidence. In other words, everyone’s brains are subject to what are now commonly referred to as cognitive biases. What a repeatable investment process helps to do is “de-bias” some of these flaws and improve the accuracy of judgements and forecasts, sharpening one’s decision making.
Overcoming Cognitive Bias in Real Life
Long before receiving the Nobel Prize in Economics, Daniel Kahneman was a psychology officer in the Israeli Defense Force, where some might argue he made his first big discovery. In this role, he and his team were charged with the task of predicting those soldiers who would make the best officers in the Defense Force, as well as which recruits were best suited to the different branches of the military. After tracking the intuitive predictions of himself and his team, he found them to be worthless, so he tried something different.
Kahneman went on to create a “personality test,” listing the traits to be evaluated, with specific questions to ask for each trait, and a predetermined system for rating each characteristic one at a time. He made his best attempt to create an algorithm; a structured process for forecasting a future outcome. The results of this approach? Despite the strong distaste from those doing the interviewing, the predicted likelihood of success in any branch of the military was increased. In fact, this approach has proven so successful that the Israeli military continues to use it today, with only minor adjustments1.
Additionally, these results, where algorithms (rules-based decisions) outperform raw human intuition and judgement, have been replicated many times over in a variety of fields. Studies have shown that more structured approaches outperform in hiring employees2 for your business, for example. In fact, research on this phenomenon goes all the way back to the 1950s, where Paul Meehl’s book Clinical versus Statistical Prediction showed 20 cases of algorithms outperforming unaided expert human judgement.
Combining the Strengths of Human and Algorithm
By building and adhering to a repeatable investment process, an advisor, robo or human, will produce better results with higher consistency over the long term. This does not mean that robo-advisors and rigid algorithms are a silver bullet solution to investment management problems. Understanding and measuring causal factors, finding historical data, and assessing how well the current state of the world resembles the past are still the job of human advisors and investment managers. The lesson is to take one’s accumulated expertise and ensure that it is systematically applied in a uniform way over time.
So while the future of robo-advisors and their impact on the industry continues to be a popular topic in the financial media, the underappreciated aspect of their approach is the integration of highly repeatable processes. Traditional human advisors would be well-served to apply these lessons throughout more segments of their investment efforts. Combining the strengths of both human and algorithm is the way ahead for forward-thinking advisors.
Here at Lake Street, we continue to emphasize investing client capital according to a continuously improving, repeatable investment process grounded in evidence-based decision making. This involves a bias toward cost and tax efficient rules-based passive allocations, a continuous search for highly process-oriented active managers in markets with clear opportunities for attractive returns, and systematic portfolio rebalancing/loss harvesting. For more on our approach, visit us at lakestreet.wpengine.com or contact us directly.