Cognitive biases are patterns of behavior that can lead us to make bad decisions and suboptimal judgments. A new research report suggests that they influence the investment choices made by asset managers, creating identifiable lifecycles as to when they generate alpha – and, more importantly, when they stop. The bad news for investors is that the most desirable profile is a rare beast; the worse news is that the most common destroys returns.
Chris Woodcock, head of research at Essentia Analytics in London, analyzed 3.5 million data points from more than 40 portfolios around the world that employ a range of different investment styles and have time horizons ranging from a handful of days to several months. Out of that, he derived 12,000 so-called episodes, tracking from when the managers first bought a stake in a company to when they sold their final share.
Woodcock, whose firm analyzes behavioral patterns for investors and traders, came up with four distinct profiles at a conference in London this week. The first, which he dubs “the Linear Accumulator,” is the one we’d all love our pension pots to track.
Although there’s a period of underperformance versus the benchmark at the very start of the investment period, this portfolio manager would generate significant alpha over the entire lifecycle. There’s a problem though. Essentia’s founder, Clare Flynn-Levy, told the conference she has never come across a linear accumulator.
A second investment profile, called “the Coaster,” underlines Woodcock’s underlying thesis that “alpha has a lifecycle that is measurable” and that there’s “a clear overall trend of long-term decay.”
For the first half of the holding period, the investment outperforms its benchmark. For the second half, though, it treads water at best. That still beats the third profile, though, which Woodcock dubs “the Hopeless Romantic.”
A brief period of outperformance is swiftly followed by steady period of decline. The problem identified here isn’t only that portfolio managers fall in love with the stocks they’ve bought as a result of the well-studied endowment effect, which leads us to overvalue something we already own. Rather, they’re besotted by the analytical work they’ve already undertaken during the equity selection process and find it difficult to take on the additional “cognitive load in starting from scratch,” Woodcock says.
The most worrying investment profile, though, is the one that the study found to be by far and away the most common in the data.
The “Round Tripper” investment spends the bulk of its life generating alpha above and beyond its benchmark. When the good times end, though, the decline is precipitous. On average, the study found that investment managers experienced a 400 basis-point decay in peak-to-trough for their asset choices.
Given the huge body of research out there that details our inbuilt biases, why are they so hard to overcome? One problem is known as algorithm aversion, our tendency to accept data that reinforces what we already think, but to be more skeptical of numbers when they go against our intuitions.
Moreover, the kinds of people who are attracted to a career in asset management tend to be have more analytical personality types; unfortunately, those are also the people most susceptible to screening data selectively to fit their existing preconceptions.
When analyzing their processes, fund managers are typically rigorous in scrutinizing the research methodology behind a particular stock pick. But they don’t appear so eager to look at the psychology behind those decisions or, indeed, the choice of which models and data they use in the first place.
This matters. We prefer narrative, stories about how the world works, rather than probabilities. But that same desire may actually harm returns. Managers would be be better off applying more introspection and devoting more attention to the assumptions behind the decisions they are paid to make.
No wonder, then, that active management is having such a hard time proving its worth.
For more Bloomberg Opinion columns, visit http://www.bloomberg.com/opinion.
Mark Gilbert is a Bloomberg Opinion columnist covering asset management. He previously was the London bureau chief for Bloomberg News. He is also the author of “Complicit: How Greed and Collusion Made the Credit Crisis Unstoppable.”