Understanding Anchoring

Because of the phenomenon of anchoring, exposure to an initial number can have enduring effects on later decisions. Learn how this aspect of cognition can help or hinder your work as a lawyer.

This piece was first published in the Wisconsin Lawyer magazine.

Who would imagine that sentencing decisions could be affected by rolling dice? Who knew that a judge’s compensatory-damages award might be reduced if the defendant makes a nonserious motion to dismiss? Is the conventional wisdom that the defense should not make the first offer in mediation open to debate? And do counsel err in not suggesting a reasonable damages amount to the jury?

In “Use Brain Science to Make Better Decisions” (Wis. Law., Jan. 2022), I explored how subconscious shortcuts in thinking – such as the affect heuristic and confirmation bias – can lead people astray. Science now teaches that a human being’s cerebral cortex (and its deliberate, logical power) does not solely or separately rule the day. Instead, logic or reason (described as “System 2” thinking) operates alongside and in conjunction with the evolutionary brain and its quick, instinctual impulses (described as “System 1” thinking). To put this construct in everyday terms, System 1 thinking is like driving a car down an open highway while System 2 thinking is like parallel parking on a busy street.

Nobel Laureate Daniel Kahneman, one of the leaders in the developing understanding of human decision-making, explains that System 1 silently “takes over” from System 2 by introducing unconscious mental shortcuts known as “heuristics.”

Fortunately, insights gleaned from decision-making science reveal steps to anticipate and overcome thinking errors. For example, as outlined in the January article, structured interviewing mitigates the affect heuristic in hiring, and embracing authentic humility helps overcome confirmation bias.

In this article, I explore an additional decision-making insight – the concept of anchoring. Anchoring comes into play in trial practice and negotiating, so understanding its effects will add to many lawyers’ toolboxes.

Anchoring: Seeing Numbers Can Affect Unrelated Decisions

The term “anchoring” is used to describe a phenomenon whereby exposure to an initial number can have enduring effects on later decisions.

Random Numbers. Daniel Kahneman provides multiple examples in his book Thinking, Fast and Slow. In one study, participants were asked to estimate the percentage of United Nations member countries that are on the African continent. Before answering, subjects were asked to spin a roulette wheel rigged to stop at either 10 or 65. Those whose wheel stopped at 10 then gave an average estimate that 25% of U.N. countries were from Africa, while those whose wheel stopped at 65 gave an average estimate of 45%.

In a second study, test subjects were shown a bottle of wine and asked to state the most they would pay for the wine. Before answering, the subjects were instructed to write down the last two digits of their Social Security number. Those whose last two digits were 80 and above were willing to pay three times more for the wine than those whose last two digits were 20 or below.

In both studies, the anchoring number had nothing to do with the question, but it nevertheless affected the participants’ thinking as shown in their answers. Another study shows how random anchors can even influence judges (a phenomenon that Kahneman aptly describes as “unsettling”). German judges were asked to specify a jail sentence they would give an individual who had stolen items from a supermarket for the 12th time. The participants first read background materials about the incident, the defendant, an advisory opinion from an expert on psychology, and statements by the defendant and a witness. Next they were asked to fill in a hypothetical sentencing recommendation from the prosecutor by rolling loaded dice. One set was rigged to yield a “3” while the other pair was rigged to yield a “9.” When the judges then made their sentencing decisions, the judges exposed to the high anchor issued a mean sentence of 7.81 months and the judges exposed to the low anchor gave a mean sentence of 5.28 months. The dice roll of “9” increased the sentence by 48%. The study’s authors warned of the many ways judges’ decisions may be influenced by factors that should not come into play, concluding with this thought: “judges may unintentionally play dice with criminal sentences.”

Another judicial decision-making study asked U.S. federal magistrate judges to set compensatory damages in a case with admitted liability. In the hypothetical, the plaintiff (an electrician) was severely injured when the brakes on the defendant’s truck failed. The plaintiff had to spend several months in a hospital and use a wheelchair. The study assessed the anchoring effect via a motion to dismiss for failure to meet the $75,000 jurisdictional threshold. Even though the case description made apparent that the jurisdictional threshold was met and the motion was frivolous, it nonetheless affected the judges’ damages awards. The 66 judges who ruled on damages without getting the dismissal motion gave damages awards in ranges up to $1,925,000 (in 1999), with a mean award of $1,249,000. The 50 judges who ruled on the motion to dismiss, however, gave awards up to only $1 million, with a mean award of $882,000. Exposure to the frivolous jurisdictional-threshold motion that focused attention on $75,000 thus depressed average damages awards by more than 29%.

Nonrandom Numbers. Other studies reveal the power of nonrandom anchors. Researchers asked people two questions: 1) Is the height of the tallest redwood more or less than 1,200 feet; and 2) What is your best guess about the height of the tallest redwood? The first question was modified for one group, replacing “1,200” feet with “180” feet, so the second group’s anchor was 1,020 feet less than the other group’s anchor. As with roulette wheels, Social Security numbers, and dice, the anchor affected the estimated height of the tallest redwood. The group prompted with the number 1,200 gave an average guess that the tallest redwood was 844 feet, while the group prompted with the number 180 made an average height estimate of 282 feet. The lower average was 562 feet lower, or 55% of the spread between the high and low anchors.

Anchoring also has been tested under real-world circumstances. In a study about residential real estate pricing, two groups of agents visited a house that was for sale and studied a booklet of information that included an asking price. The booklets were edited so that half the agents saw a price substantially higher than the listed price, while the other half saw a price that was substantially lower. The agents then were asked to opine about the home’s value and to explain the factors that affected their judgment. Although the agents insisted the price stated in the booklet had no effect on their valuation, once again the anchoring effect kicked in, yielding a 41% spread between the high and low prices.

In another experiment, persons going into a store were asked either 1) how much money they had in their wallet, or 2) how much money they had in their bank account. Those who answered the smaller, amount-in-the-wallet-question spent less in the store than those prompted to focus on the larger, bank account number.

As a final example, consider how car dealers have made use of anchoring in pricing cars. Even though most potential buyers know that the “sticker price” is not real, that price still serves to anchor their thoughts about a car’s value. Thus, when a buyer is told the final, “manager-approved” price, the buyer usually compares it to the price on the car window and self-congratulates on the great purchase (access to more accurate cost information online and new ways to buy cars, such as by using CarMax or Carvana, however, may lead to the demise of window sticker anchoring).

Anticipating and Offsetting Anchoring in Practice

Given the power of the subconscious mental shortcuts outlined above, what are some things lawyers can do to offset or overcome these judgment errors? Research and practical experience provide answers.

Omitting a Damages Number is a Bad Trial Tactic. It is a mistake at trial not to offer jurors a damages number. As I detail in example 16 in my four-part series, “20 Great Ways to Lose a Trial,” www.weberadvising.com/writings, many defense lawyers believe they undercut the strength of a no-liability argument if they suggest any dollar amount as a fair measure of damages. According to this belief, offering thoughts on damages constitutes a concession on liability. But staying silent on damages leaves the plaintiff’s lawyer’s damages number unrebutted in the jury room, making it the only anchor. It is important to provide a counterweight, ideally even in advance of the plaintiff’s lawyer establishing their side’s damages anchor. One way to do this would be for the defense to discuss damages amounts during voir dire, in the opening statement, or both.

Similarly, I have been surprised when plaintiffs’ lawyers did not provide specific amounts when asking for compensatory or punitive damages. From what I knew of the cases and the results, I thought the awards would have been much larger had amounts been suggested (and anchoring research supports that suspicion).

The plaintiff’s demand can even affect a jury’s finding on liability: If a number is perceived as too low when compared to the facts, the claim might seem less credible. Mock jurors were presented with a case in which it was alleged that birth control medication caused a plaintiff’s ovarian cancer. The defendant was an HMO sued for prescribing the medication. When plaintiffs asked for $5 million in damages, the mock jurors were 43.9% confident about causation. When only $20,000 was requested, only 26.4% of the mock jurors were confident about causation. It can work the other way too, with an extremely high damages request undercutting a plaintiff’s credibility.

Putting an Offer on the Table First Isn’t a Disadvantage. Another common belief among defense lawyers is that during negotiations the plaintiff must make the first move by putting a demand on the table. Anchoring research suggests this likewise is a mistake. A study at Northwestern University showed the first offer can materially frame the parties’ valuation and skew the result toward the initial offer or demand. In that study, MBA students played the role of buyers and sellers of a manufacturing plant. All else equal, the final sale price ended up higher when the seller made the first offer and lower when the buyer made the first offer.

The lesson for lawyers engaged in settlement talks or other kinds of negotiation is to try to be the first to present an offer, so as to provide the initial anchor. Mediators no doubt will be surprised when defense lawyers depart from the script and offer a settlement amount at the outset. Lawyers who can’t act first might be able to offset the effect of their opponent’s proposal on their own thinking by focusing on all the reasons the opponent’s number is not credible. This technique counteracts the instinctive response (which plays into the anchoring effect) to look for ways to make sense of the other side’s action, for example, “Their demand is so high. They must have a good reason for that number. What is the reason?”

The consider-the-opposite approach was tested in a study about valuing a used car. Car mechanics and dealers were presented with a 10-year-old car that needed repairs and were asked to estimate the car’s value to see if the repair was worthwhile. The car owner gave an anchor by stating the owner’s value estimate, with one set of subjects getting a suggested value 56% lower than another set. Anchoring was effective; the high anchor led to 25% higher estimates of car value than did the low anchor.

A subset of the subjects, however, were told an additional statement before they gave their value estimates. The additional statement 1) undercut the owner’s suggested value and 2) asked them to consider counterarguments: “A friend of mine mentioned yesterday that he thought this value is too high [or low]. What would you say argues against this price?” The group of experts who got this consider-the-opposite instruction had the anchoring effect reduced by more than half, shrinking the spread in estimates to 11%.

So when the other side makes the first proposal, to counteract its effect as an anchor consider all the reasons that proposal does not make sense. And if you have an evaluative-style mediator, share those reasons with the mediator. This technique triggers System 2 thinking, counteracting the intuitive nature of System 1.

An Expert’s Tip for Countering an Anchor. Kahneman offers a final approach to countering anchoring in Thinking, Fast and Slow, as follows:

“My advice to students when I taught negotiations was that if you think the other side has made an outrageous proposal, you should not come back with an equally outrageous counteroffer, creating a gap that will be difficult to bridge in further negotiations. Instead you should make a scene, storm out or threaten to do so, and make it clear – to yourself as well as to the other side – that you will not continue the negotiation with that number on the table.”

With all due and deep respect to Kahneman, I am skeptical of this approach in any case involving sophisticated lawyers and parties. If the foolish demand or offer is an attempt to anchor a number using System 1’s flaws, activating System 2’s cognition by calmly listing all the reasons your opponent’s proposal is incredible would seem more useful than using energy in an outburst of real or feigned indignation. The study of Northwestern University MBA-student negotiations found a similar approach effective in offsetting anchoring effects. Then, after creating the list, having the mediator convey the reasons to the other side with a counteroffer consciously chosen without regard to the initial offer gives you the opportunity to set an anchor for the other side’s valuation.

Conclusion

A wise mentor once told me, “The person with the most patience gets the best deal.” I would supplement that wisdom with this: “The person with the most patience and a deep appreciation for flaws in human decision-making gets the best deal.”

The reality of human decision-making should prompt discussions about ways to offset errors traceable to intuitive mental processes. I hope this article raises readers’ awareness of anchoring’s surprising power and ways to mitigate its effects.