Cognitive illusions, popularly known as cognitive biases, affect everyone. In the realm of litigation, two such illusions, egocentric and confirmation biases, act as impediments to settlement. But one might ask, don’t the overwhelming majority of civil cases end in settlement? If they do, then we have little reason to be concerned about the effects of possible egocentric and confirmation bias on settlement negotiations.
Reflecting the conventional wisdom, thelawdictionary.org claims that, “According to the most recently-available statistics, about 95 percent of pending lawsuits end in a pre-trial settlement.” A quick Google search uncovers a substantial number of law firms that agree. A West Virginia law firm asserts, “It’s commonly cited that 95-96% of all civil cases settle out of the courtroom, but the real number may actually be slightly lower than that.” “Slightly lower,” whatever that term might mean seems to suggest a rate in excess of 85 percent. But do the data support the conventional wisdom regarding the high rate of pre-trial?
Lex Machina has classified 1,125,829 civil lawsuits that were resolved in the federal district courts between January 1, 2009 and March 31, 2021. In 16,472, or 1.5 percent, of those cases resolution occurred through a trial. 672,536 of those terminations took the form of stipulated dismissals. In another 144,730, plaintiffs filed voluntary dismissals. For classification purposes Lex Machina groups those two categories into one: likely settlement. Simple division reveals that the 817,266 likely settlements constitute 72.6 percent of the resolutions. Because Lex Machina does not verify that each of the dismissals and voluntary dismissals is the product of a negotiation between the parties, the 72.6 percent figure is an overestimate of the rate of negotiated settlements in the data base. Regardless, while an arguably high percentage, 72.6 percent is a far cry from 95-96 percent. As a result, it is worth considering how cognitive illusions might help explain why the rate of negotiated settlements is not higher.
Negotiation scholars have identified several barriers to negotiated settlements, including, for example, what social psychologists refer to as biases that affect decision making. One such bias is egocentrism. That bias can take the form of a belief that one’s ability as a litigator is superior to that of opposing counsel. Imagine a scenario in which the lawyers for the respective parties have the exact same information about the material facts and the applicable law in the case. If egocentric bias infects both lawyers, in that hypothetical scenario, one or both of them will have an overly optimistic sense of who is more likely to “win” the case, and thus make settlement more difficult. All other things being equal, in the absence of egocentric bias, the respective estimates of how likely each is to “win” the case should be close to one another. Obviously, the presence of egocentric bias will serve as an impediment to settlement.
A form of bias that receives little attention in the literature on barriers to negotiated settlements of civil cases, however, is confirmation bias. The explanation for the bias is rather straightforward and applies to most, maybe all, humans: I develop beliefs about a variety of phenomena. Subsequently, I seek evidence that supports my beliefs, and reject or attempt to rationalize away evidence that conflicts with my belief. Further, one might think that if I am presented with considerable evidence that undermines my belief, there is a good chance I will change my mind. For example, we might expect that people who believe that climate change is not real will change their minds when presented with evidence in the form of data showing near unanimity in scientific community on the matter. That is, if we supply non-believers with facts that they do not have, they will “see the light.” Remarkably, social scientists who have explored the phenomenon have discovered that additional material facts do not cause the believers to alter their beliefs. In fact, in some, albeit rare, instances believers can become more certain of their (mistaken) beliefs, a phenomenon known as the backfire effect. https://link.springer.com/article/10.1007/s11109-010-9112-2?TB%20iframe=true&width=921.6&height=4638.6&error=cookies_not_supported&code=1a71d8d1-e963-432b-851e-17771c195849. More commonly, mistaken beliefs persist because the positive effects of correctives tend to dissipate rather quickly.. https://www.pnas.org/content/118/15/e1912440117.short.
Translate the confirmation bias to mediation, and imagine again the hypothetical scenario of the two opposing lawyers in a civil case who, now, find themselves in a mediation. The plaintiff’s lawyer believes strongly that the plaintiff will win the case. The defendant’s lawyer has an equivalent estimate of the likelihood that the defendant will prevail. At some point in the mediation, the plaintiff’s lawyer asks the mediator to convey some factual information to the defendant’s lawyer. In the mind of the plaintiff’s lawyer that information will serve to cause the defendant’s lawyer to alter her view as to the likelihood that the defendant will win. If the two lawyers are like most people, however, the “new” factual” information will not cause the defendant’s lawyer to become less convinced that the defendant will prevail in the case. Clearly, if both lawyers do not become less convinced of the likelihood of their respective clients’ prevailing in the face of new factual information, the prospects for a negotiated settlement will suffer.
Sadly, law school training cements confirmation bias in aspiring litigators in the way that it teaches legal reasoning. Every law student becomes familiar, even comfortable, with reasoning by analogy. We look for cases that support our side and learn how to develop potentially believable interpretations for cases that on the surface that support our opponent’s side. Thus, a plaintiff’s lawyer will attempt to re-interpret a case that the defendant’s lawyer cites in support of the defendant. If the plaintiff’s lawyer is unable to develop the desired re-interpretation, all is not lost. Instead, the plaintiff’s lawyer will attempt to distinguish the case that supports the defendant. At root, reasoning by analogy in the context of litigation appears, then, to be a manifestation of confirmation bias by another name. If that is true, a deep-seated barrier is likely to be present in the typical settlement negotiation in a civil case. Thus, confirmation bias in legal settings exists in two dimensions: in proof of facts and in its legal arguments.
Imagine a mediator who is aware of how egocentric bias and confirmation bias can negatively affect the litigator’s ability to predict the outcome of litigation if the case does not settle. The mediator could simply commit to not giving up in the effort to effect settlement. That commitment could result in the parties ultimately reaching agreement as the result of a combination of physical and mental exhaustion. At that point settlement terms will appear to be more acceptable than they would have been when everyone involved in the mediation was relatively fresh. Indeed, there is at least anecdotal support for the proposition that a mediator who is willing to keep working at the process can achieve settlement where a mediator who is less committed will not.
But consider another approach. Imagine that the mediator asks the respective lawyers to answer a question: Suppose that you were the lawyer for your client’s opponent. How would you assess the case from that perspective? The rationale behind this question lies in the common saying that where you stand depends on where you sit. Indeed, the wisdom behind this folk saying has been demonstrated to operate for budding lawyers who, compared to experienced litigators, are not as inculcated in either egocentric or confirmation biases. In simple terms, randomly select two groups of law students who were participating in a moot court competition. One group will represent a hypothetical plaintiff. The other group will represent a hypothetical defendant. Provide the exact same information, including facts and applicable law, about the hypothetical legal dispute that is in litigation. After the students have reviewed the information, i.e., worked on the case, ask them to estimate the probability that their “client” will prevail at trial. Legal scholars who ran this experiment demonstrated unequivocally that the lawyers for the plaintiffs assigned a considerably higher probability of a plaintiff win than did the lawyers for the defendant. https://www.journals.uchicago.edu/doi/abs/10.1086/667711. Run the same experiment with law students before they have “worked on the case,” and they display similar role-induced bias. https://www.journals.uchicago.edu/doi/abs/10.1086/710306.
Thus, it seems fair to say that actual lawyers for defendants and actual lawyers for plaintiffs are capable of viewing a case through the eyes of their counterpart. Consequently, having them do so in the context of a mediation should serve to bring their respective estimates of the likelihood of the plaintiff’s prevailing closer together than would otherwise be the case thereby reducing a barrier to settlement. The mediator could achieve this end by asking the lawyers for the respective parties to complete a simple task that relies on what psychologists refer to as theory of mind: “How do you think that opposing counsel views this case? That is, what do you think is the probability that opposing counsel assigns to their client’s winning?”
Works cited:
Eigen, Z. J. & Listokin, Y. (2012). Do Lawyers Really Believe Their Own Hype and Should They?: A Natural Experiment. Journal of Legal Studies. 41(2), 239 -267.
Nyhan, B. (2021). Why the backfire effect does not explain the durability of political misperceptions. Proceedings of the National Academy of Sciences, Colloquium Paper. April 13, 2021, 118(15), e1912440117.
Nyhan, B. and Reifler, J. (2010). When corrections fail: The persistence of political misperceptions. Political Behavior. 32, 303 -330.
Spamann, H. (2020). Extension: Lawyers’ Role-Induced Bias Arises Fast and Persists despite Intervention. Journal of Legal Studies. 48(2), 467-485.