November 15, 2019 | Institutional Perspectives | 7 min read
Battling Bias in Investing
Humans are imperfect beings. We can educate ourselves, gain experience, and even learn from our mistakes but still, our minds can and will play tricks on us. As investors, this is a minefield. With the list of psychological biases a mile long, every time we make a decision we risk stepping on one of those mines and harming our portfolios. Investors need to start by arming themselves – understanding where “most people” go astray and interrogate their own processes to spot weaknesses. This article is a good start.
Deconstructing bias
Biases represent impaired thinking due to emotion, or due to a systematic flaw in our logic or process. Both cause irrationality and can therefore be self-defeating but they go undetected because they are also typically unconscious.
At a high level, investment decisions are made in the context of some very powerful factors that simultaneously increase our susceptibility to biases and, worryingly, amplify their impacts. Time limitations, complexity, fear, peer pressure, risk, loss, noise, and uncertainty have been shown to compound the risk of perpetuating biases. How many of these factors impact your typical day investing? I’d wager most or all.
Another annoying problem with unconscious biases is that even if we identify them and educate ourselves about their triggers and impacts, they can persist as a blind spot.
To illustrate, consider the inner rectangle in Figure 1. Which end, left or right, looks darker? The rectangle’s right side appears darker but is in fact the same shade throughout. Your retinas of course receive the one shade of light, but your brain does some processing before perception. The larger rectangle “background” is lighter on the right, causing your brain to perceive more illumination on the right side than the left so for the inner box to be the same shade across, your brain concludes that the inner box must be lighter on the left, darker on the right. Your brain makes it look that way.
Bucketing biases
With literally hundreds of documented biases, it is helpful to categorize them into four broad categories, describing our tendency to:
- Filter – when we have too much data. We are bombarded daily by market and other data. It’s impossible to process all of it and in practice, for much of it we shouldn’t bother. How good are we at filtering? Are we able to improve it? Often the answers are “poor” and “not much.”
- Fill in gaps – when there is too little data. In many situations, we have insufficient data and we need to “fill in” data (e.g., forecasting or assessing a new company or industry). How good are we at completing the picture? Not great. Do we at least lean on relevant factors? Often not.
- Take action too quickly. Our hunter-gatherer survivalist heritage hardwired us with the bias to act quickly (and emotionally), a tendency which, when combined with powerful rationalization tendencies, can cause more harm than good. Taking your time is valuable.
- Rewrite our memories. As with the optical illusion above, our brains can “pre-process” the facts to make them fit our beliefs. This means we can edit memories rather than merely recall them. This unconscious act can therefore distort decisions.
A few of the most frequent and harmful biases include confirmation bias, representative bias, and anchoring. They are developed on the next page.
Figure 1
It’s easy to see the truth if you cover up the larger box. Remove the covering and the illusion returns. This is the interesting point: knowing what is going on doesn’t remove the effect; your brain is still tricked after understanding it!
This concept applies similarly to biases: understanding them does not in itself remove the effect. You must go further. You have to create and apply tools or processes to neutralize the effect. You have to remove the rectangle.
Confirmation bias is the tendency to seek/accept information that agrees with what we already believe, or conversely to avoid/reject information that challenges our beliefs. Reading media that agrees with your political beliefs and sourcing bearish financial opinions when you expect a recession, are examples.
When we make decisions by asking, “What other information agrees with me?”, we eliminate the important (dissenting) perspective. Questioning assumptions is a more productive approach, but in practice it takes more effort and can sometimes be uncomfortable, so we avoid it.
"In the investing world, confirmation bias contributes to the endemic issue of overconfidence. When confidence exceeds accuracy, investors are prone to taking on too much risk."
Representative bias is the shortcut of using “representative” characteristics when deciding the likelihood of something. For example, if asked if a woman who “is quiet and reads poetry” is more likely to be a librarian than a BMW driver, you’d likely guess librarian because quietness and literature are representatives of that profession. However, given the millions of BMWs sold every year and tens of millions on the road today, the chance is in fact far higher that she drives a BMW.
Representative bias plays a large role in “return chasing,” the act of basing future return expectations (of a market or manager) on recent performance. Skill generates a higher chance of outperformance, but the reverse (“outperformance reflects skill”) is not true even a majority of the time. Manager performance, markets, and styles run in cycles, but many investors still use representativeness as a proxy for likelihood, with predictably poor results.
Anchoring is the use of irrelevant data to make decisions, even when it’s known to be irrelevant. The worst (and most common) offense is analyzing daily price movements to extract meaning. It is basically useless most of the time, but we can’t help but be influenced by it. It anchors us invisibly and skews our view. A big drop in share price suggests doom and we may run for the exits; the reverse and we might dive right in (ironically, often the exact wrong
trades for that moment).
In the investing world , confirmation bias contributes to the endemic issue of overconfidence. When confidence exceeds accuracy, investors are prone to taking on too much risk.
Tools to beat bias
To de-bias you can either “change” the decision maker by educating them, or change their environment by altering how, when and what information is presented, and introduce checklists or other framing tools.
At Leith Wheeler, we employ a number of tools to help manage bias. They include collecting independent, blind forecasts from analysts (interest rates, stock targets), appointing devil’s advocates, engaging in pre-mortems and what-if scenarios, and soliciting “outside views” from external sources. A few of these tactics are detailed below.
Pre-mortems. To perform a pre-mortem, you imagine yourself a few years from now, looking back at your current decision… and it’s all gone wrong. Brainstorm: what could have caused this failure? This approach can identify new risk factors, temper your optimism, and better align your confidence levels to your capacities. The biases you are hitting head-on here are overconfidence, group think and narrow thinking.
Devil’s advocate. You can reduce group think by appointing a formal “detractor” with good knowledge and debating skills to challenge ideas. The advocate might use an “outside view,” which is statistical data that shows how these projects typically go. If you have 80% confidence in the outcome but the devil’s advocate surfaces raw data showing only 10% are completed as expected, it might help better align confidence and expectations.
Decision diaries. Hindsight bias can make you think things were more predictable than they actually were, increasing confidence in your own prediction abilities, and emboldening you to take on more risk. A technique to manage hindsight bias is to write down in a diary the key determinants and expectations from a decision at the time and track performance against those documented views. An online tool we use to record our thinking is Slack. It records our main arguments and thesis, is searchable and can show clearly how our thinking has evolved, sometimes unknowingly.
Reflecting on an unsuccessful management interview, Lewin Capital founder and equity analyst Dan Lewin once said that “This job is hard enough when they’re not lying to you.” Sadly, with investor bias, the liar is you. But take heart. With some awareness and a few of these tools, you can parse some of the illusion and make reasoned, successful decisions. The war is not over yet.
IMPORTANT NOTE: This article is not intended to provide advice, recommendations or offers to buy or sell any product or service. The information provided is compiled from our own research that we believe to be reasonable and accurate at the time of writing, but is subject to change without notice. Forward looking statements are based on our assumptions, results could differ materially.
Reg. T.M., M.K. Leith Wheeler Investment Counsel Ltd.
M.D., M.K. Leith Wheeler Investment Counsel Ltd.
Registered, U.S. Patent and Trademark Office.
