Have you ever found yourself in a conversation stuck on a thought? If so, you are likely to have been in the midst of a discussion swallowed by anchoring bias. Anchoring is the fixation on a reference point within a continuum of information in decision making and is the latest bias deep dive I am covering for investors. At the end of this series I am going to land with A Theory of Behavioral Finance.

A Helpful Mnemonic Device: LOCHAARM

To my mind the major behavioral biases are:

• Loss aversion
• Overconfidence
• Confirmation
• Herding
• Anchoring
• Availability
• Representativeness
• Mental accounting

A helpful mnemonic device for remembering these biases is LOC HAARM, brain lock that harms investment performance. In that list of eight behavioral biases anchoring is up…NOW.

Anchoring Bias: Origins

To my knowledge, anchoring was first described by psychologists in 1958.^[1] In the paper the psychologists give the example that when people were told that it would be 10 years until the peaceful use of nuclear power was possible, it caused people to adjust their own estimate more toward this anchor. It turns out that no matter what initial number of years was suggested, that the conversation always revolved around that anchor.

Tversky and Kahneman were also avid researchers of anchoring, first writing about this behavioral bias in 1974.^[2] Herein, they describe anchoring as:

“In many situations, people make estimates by starting from an initial value that is adjusted to yield the final answer. The initial value, or starting point, may be suggested by the formulation of the problem, or it may be the result of a partial computation. In either case, adjustments are typically insufficient. That is, different starting points yield different estimates, which are biased toward initial values. We call this phenomenon anchoring.”

Tversky and Kahneman give the example of a wheel of fortune ranging from 1 to 100 being spun in the presence of subjects. These subjects are then asked to estimate the number of nations that are in Africa. Despite the fact that the spin of the wheel is completely arbitrary, they noted that the estimates of the subjects were statistically significantly anchored to the result from the spinning wheel.

It is important to note, that it is logical to start an analysis with an estimate based on the best information available at the time, then adjust that estimate as additional information is obtained. This is known as Bayesian inference, named after the 18^th century mathematician Thomas Bayes. This is a statistically rigorous method for combining the best prior estimate with new information, weighted by the confidence with which each estimate is held.

The problem is that the initial anchor is frequently irrelevant to the task at hand. Say, for example the number 153 pops up randomly in a conversation. Studies have shown that arbitrarily saying a number, like 153, prior to asking a group to estimate the number of jelly beans in a jar drives the group’s estimates toward that number.

If a second group is asked to provide estimates for the same jar and the number 233 is mentioned instead, the resulting estimates will be closer to 233. In fact, studies have shown that the average estimate in the second group will be in the range of 40 points higher than in the first, that is, 50% of the 80 points difference in the initially mentioned numbers. Distressingly, arbitrary anchors easily and frequently distort decisions.

Anchoring Bias: Manifestations

One obvious manifestation of anchoring bias in investing are the discussions of investment committees. As I know from my consulting work embedding with investment teams, they frequently begin their meetings with a stock, bond, or other asset pitch from a member of the committee. Inevitably the discussion revolves/anchors around the correctness of the analyst’s work. Extraordinarily rare is the discussion that rejects the analysis altogether because it is taken as a given that the correct context for the discussion is the one provided by the first person to speak.

Unfortunately, most people, when thinking about anchoring, they think this bias only applies to information provided first in a sequence. However, people can fixate on any reference point amongst a cache of information.

For example, an investor may have a ten-year career during which a deep recession occurred, and during which the investor’s performance suffered. Going forward, the investor’s decision-making may be unduly influenced by this recessionary period. Or it could be that an investor lost money in a waste management company that makes it impossible to evaluate all other waste management businesses objectively.

In investing, anchoring as a bias appears most during trading where there is a tug of war between bid and ask, as well as in the famous obsession that many investors have with the cost basis of a security. This bias you likely also know as the sunk-cost bias where an investor refuses to sell an asset that has lost him money until it climbs back above his initial purchase price. Here the investor has ignored that the future price performance of the asset from this day forward is all that matters. It may be that the probability the asset ever exceeds its cost basis is impossible, and therefore the investor should sell the asset. Note how similar this phenomenon is to the concavity of losses of loss aversion, where people take outsized bets when confronted with losses.

Other examples include recent stock and market highs and lows; 1, 3, 5, and 10-year performance; the cost basis of a security purchased; and the amount invested in a portfolio. We cling to these anchors and it is incredibly difficult, if not impossible, to convince us to abandon them.

Anchoring Bias: Nuances

The anchoring bias nuance that I want you all to focus on is that the distortions in thinking discussed here are all of the same type: an inability to establish objective contexts for discussion. Anchoring bias is not just a numerical phenomenon. Nor is it fixation of the first object in a sequence. Instead it is a lack of awareness about the proper context for a discussion.

In the nuclear power discussions from above the context established for discussion was whether or not it would be a substantial source of electricity within 10 years. In considering purchasing a security, anchoring is sure to occur if the question about whether the discussion is even legitimate never is raised. What is required to overcome anchoring bias is deep self-awareness about the contexts established for a discussion. A second necessary condition for overcoming anchoring is the mental nimbleness to switch contexts to ensure deep thinking.

Conclusion

To me, I consider anchoring to specific narratives or moments in the economic and financial markets, as well as about specific securities, to be rich sources of alpha. Simply put, investors’ inability to let go of their anchors in defiance of reality causes a mispricing in securities, and therefore opportunity.

Importantly, some researchers have found that experts are less susceptible to anchoring than laypeople; this holds out hope for professional investment analysts versus amateurs.^[3]

[1] Sherif, Muzafer, Daniel Taub, and Carl I. Hovland. “Assimilation and Contrast Effects of Anchoring Stimuli on Judgments.” Journal of Experimental Psychology. Vol. 55, No. 2 (1958): 150-155

[2] Tversky, Amos, and Daniel Kahneman. “Judgment under Uncertainty: Heuristics and Biases.” Science, New Series, Vol. 185, No. 4157 (1974): pp. 1124-1131

[3] Wilson, Timothy D., Christopher E. Houston, Kathryn M. Etling, and Nancy Brekke. “A New Look at Anchoring Effects: Basic Anchoring and Its Antecedents.” Journal of Experimental Psychology, Vol. 125, No. 4 (1996): pp. 387-402