By Dr. Ken Broda Bahm –
This blog frequently covers recent psychological or communications research bearing on legal persuasion, and an important question is how well results hold up when leaving the laboratory and entering the courtroom. One example is the phenomenon of damage “anchoring,” or the advantage gained when one side offers an ad damnum number as a starting point for jury deliberations. In a long line of studies in laboratory settings, researchers have demonstrated the process of “anchor and adjust,” meaning mock jurors will start with an anchor point, often a number recommended by one party or the other, and adjust up or down in order to arrive at a verdict. Based on the amount of influence exerted by the choice of the starting anchor, the common wisdom is for plaintiffs to ask high to raise the ultimate damages, and for defendants to recommend low in order to lower that number.
In the real world of trials, is the effect of an anchor as simple and automatic? Based on a recent analysis of the deliberation content of video-recorded trials in Arizona courts, the answer is “no, not so simple or automatic.” The research (Diamond et al., 2011) shows while anchoring is a powerful force in deliberations, jurors are also very critical consumers of attorney recommendations, and the supposed biasing effect of damage anchors appears to be overstated. In this post, I take a look at these research findings, and provide practical recommendations for both plaintiffs and defendants arguing damages.
The Problem:
From a legal perspective, the problem with anchoring is it arguably introduces an arbitrary element into deliberations. This potential for a single number to exert a measurable influence has been shown in many studies, sometimes to an absurd degree. For example, in one classic study (Tversky & Kahneman, 1974), research participants were asked to estimate the percentage of African countries that make up the United Nations, after viewing a number arbitrarily selected based on a roulette wheel. Despite clear knowledge the number was arbitrary, participants were still influenced by the number. When the wheel landed on 65 percent, the average estimate by participants was 45 percent, but when the wheel hit 10 percent, the average estimate was just 25 percent.
The idea that a number, often based on attorney choice rather than evidence, could similarly exert an undue influence in deliberations, has led the States of Pennsylvania and New Jersey to bar attorneys from using numbers not derived from evidence, in effect preventing attorneys from offering numbers for pain and suffering, disfigurement, anxiety, loss of consortium, punitives and a number of other general damages categories. Other venues and judges are similarly skeptical of the use of per diem amounts, where attorneys try to concretize an abstract category by suggesting an amount per day that an injury, for example, might be worth.
The Research
Shari Diamond, Mary Rose, Beth Murphy, and John Meixner, in a research paper posted last month on the Social Science Research Network, draw on a unique data set created when a group of Arizona judges allowed fifty actual trials and deliberations to be recorded and subsequently analyzed by scholars. After coding all of the juror deliberations comments based on their recall and use of the anchor amounts provided by both sides, the team was able to document effects of anchoring in a real world context. The result? Anchoring matters. 83 percent of attorney requests were discussed in deliberations, and 86 percent of jurors referred to at least one attorney recommended amount. The other result, however, was jurors are likely to be critical consumers of those amounts. The researchers found no evidence jurors simply accept the amounts that attorneys give them without critically discussing those amounts. Specifically, jurors in the Arizona trials were more skeptical of the anchor amount than the experimental results would suggest.
Specifically, one comment in five was an explicit rejection of the anchor amount (e.g., “that is an outrageous amount”), and the rejection rate was even higher on pain and suffering amounts. Not a single jury in the fifty trials observed awarded the actual amount recommended by the plaintiff for a general damage category like pain and suffering. So, while the process of “anchor and adjust” was observed, the key word, particularly for the less definite categories, is “adjust.”
The Recommendations:
1. Plaintiffs, Use Anchors. It is now common sense among plaintiffs that you should give jurors an anchor number, especially, for the less evidence-grounded general damages categories. Still it is interesting to note in one third of the Arizona cases, plaintiffs did not provide an anchor number. While jurors will not tend to award that number (indeed, the Arizona jurors awarded just 15 percent of plaintiff pain and suffering requests on average), it is still better to begin with a reference point than to leave that to the jurors. Ironically, the fact jurors will often presume that number to be inflated is a reason to give a higher number to begin with.
2. Defendants, You Should Use Anchors Too. Criticizing the plaintiffs numbers is usually not enough. One dissertation (Ellis, 2002) for example found for all cases other than those where liability strongly favors defense, an alternate damage anchor helps to keep defense damages low, without increasing the chances of a liability finding against the defendant. In the 2011 analysis of the Arizona trials, defense anchors were found to be much more influential than plaintiff anchors. Jurors were four times more likely to make comments rejecting a plaintiff anchor than a defense anchor, and four times more likely to end up accepting defense recommended amounts as opposed to plaintiff amounts.
3. Both of You, Be Reasonable and Avoid Extremes. One misunderstanding of anchoring research is the simple conclusion that as far as the initial ask is concerned, “the higher the better” for plaintiffs, and “the lower the better” for defendants. But there is a credibility effect to making a recommendation outside the bounds of what jurors consider reasonable. In the fifty trials examined, there was actually a .77 (that’s high) correlation between plaintiff recommendations and defense recommendations, indicating that both sides were to a large extent basing their recommendations on the “true” extent of damages. Attorneys often focus on the so-called “insurance rule” of a ratio of general damages to special damages of three to one. According to these results, even that is a little overblown. General damages in the Arizona trials averaged only 108 percent of the special categories that could be backed up by evidence such as past medical bills, and the researchers also cite a Texas study showing there it isjust 170 percent. While there are certainly cases where an outsized general damages request would be warranted (e.g., a plaintiff with few medical bills, but lifelong disfigurement), in most cases, the general damages request should not be much more than the total of the specific damages requests.
4. Fight for Your Right to Anchor. As noted earlier, some venues and judges discourage or prohibit attorney ad damnums or per diems in uncertain categories. If those prohibitions and discouragements are based on the belief the numbers exert a prejudicial influence on jurors, then this research (Diamond et al. 2011) contains a compelling refutation. The central conclusion of the study is that jurors are critical consumers of attorney-recommended damage amounts. This is outstanding research: not college students reacting to one-page case summaries, but actual jurors deliberating in fifty actual trials. The next time you are constrained in offering an ad damnum amount, petition, and attach this study to the brief.
______
Related Posts:
- Get the Gist of How Jurors Decide Damage Numbers
- A Mixture of Justice and Revenge: Target Juror Psychology in Awarding Damages
______
Shari Seidman Diamond, Beth Murphy, Mary R. Rose, & John B. Meixner (2011). Damage Anchors on Real Juries Social Science Research Network (July 11)
Photo Credit: Martin A. Hansen (Maasha), Flickr Creative Commons