Rethinking mediation with behavioral-science data

Using behavioral-science data rather than wishful thinking to make mediation more productive

2017 August

“I will look at any additional evidence to confirm the opinion to which I have already come.”

― Lord Molson, British politician (1903-1991)

Lord Molson was onto something. Behavioral scientists have confirmed as much. Now it’s time for the rest of us to begin using that science to make mediations more productive.

First, the science: A growing body of behavioral research shows how lawyers and clients – indeed all of us – process and filter information, weeding out unwanted input in favor of self-serving affirmations. In other words, we hear what we want to hear and largely disregard the rest. Call it egocentric or self-serving bias.

These patterns are as real for organizations as they are for individuals. Take this as gospel from a litigator turned general counsel turned mediator: Groups often model the very same behavior, particularly when dealing with adversarial or unexpected events. More on this later.

Notably, modern civil mediation practice seems to have taken a contrary course, reducing rather than enhancing everyone’s chances of success. Common practice today includes limited pre-mediation dialogue about the merits, mediation statements that are not shared or mimic trial briefs in tone and temperament, and the absence of joint sessions at the mediation itself.

The goal here is to promote a form of mediation advocacy that embraces the behavioral science and maps a different course. After two decades mediating and prior litigation and general counsel roles where these concepts could be tested, I can tell you they work.

Client perceptions and overconfidence: Tell me what I want to hear, not what I need to hear

A growing number of behavioral studies focus on how clients filter information they receive, holding onto the information that affirms pre-conceived notions much better than the data that casts doubt. (See, e.g., Donna Shestowsky, PhD., Professor of Law at the University of California, Davis, School of Law, The Psychology of Procedural Preference, How Litigants Evaluate Legal Procedures Ex Ante, Iowa Law Review, Vol. 99, No. 2, pp. 637-710 (2014); See also, George Loewenstein, et al., Self-Serving Assessments of Fairness and Pretrial Bargaining, 22 Journal of Legal Studies, pp. 135, 149-53 (1993).)

In one study, litigants involved in various forms of dispute resolution (trial, arbitration, mediation, etc.) were asked to rate the fairness of those different procedures as well as their own chances of success. In addition to confirming that clients prefer dispute resolution processes like mediation where they maintain the most control, this study revealed that 57 percent of litigants believe that they had at least a 90 percent chance of winning, while roughly 24 percent believed they had a 100 percent chance of winning. I confess to having picked law school in part because there was little math involved, but even I know those numbers don’t add up. These findings reveal an egocentric bias, where litigants construe information in a self-serving way, and in turn believe that their case is much stronger than it really is.

Attorney handicapping: the dangers of wishful thinking

Attorneys often fare no better than their clients as to handicapping skills. Multiple behavioral studies reveal that lawyers routinely overestimate their client’s litigation prospects – i.e., the likely outcome at trial – compared to the actual outcome if the case is fully tried. (See, Randall Kiser, Beyond Right and Wrong, The Power of Effective Decision Making for Attorneys and Clients (Springer 2010), pp. 29-48. See also, Jane Goodman-Delahunty, Pär Anders Granhag, Maria Hartwig, and Elizabeth Lofthus, Insightful or Wishful, Lawyers’ Ability to Predict Case Outcomes, Psychology, Public Policy, and Law, 2010, Vol. 16, Nos. 2, pp. 133-157.)

In one set of studies ― repeated over different time periods in both California and New York ― plaintiffs on average erred in their assessments more often than defense counsel. Specifically, plaintiffs often left money on the settlement table – comparing what they turned down in pre-trial settlement offers to the eventual outcome ― reflecting a 60 percent error rate for plaintiffs versus a 25 percent error rate for defense counsel. (Kiser, Id. at 42.)

While this data initially sounds encouraging for defendants, it has a dark side. Specifically, while plaintiff’s average cost of decision error was $73,400, defendants’ average cost of error was over $1,400,000 ― 19 times greater. (Ibid.) Thus, fewer errors, but exponentially costlier when they hit, both in terms of financial losses and client relations.

Making use of the behavioral science data in the mediation process

After two decades of litigating on behalf of plaintiffs and defendants, I started my first general counsel position. There I inherited a large number and variety of pending disputes ― a pattern that repeated itself in two other GC roles. In each position, I began sorting through how we were handling our cases, including how much we really knew with confidence, how much had we shared with the other side, and what alternatives existed to resolve these disputes.

As to many matters, our current course was well-informed and made great sense. As to others, not so much. The litigation path we were on was usually by the book, was requested by the client, and may well have eventually worked in court. But the same questions consistently arose: Did we really know all the key facts? What did the other side see differently? If something was amiss as to our own assessment or theirs, wasn’t it better to sort that out sooner versus later? And did we really need to win or simply to make the dispute go away?

By this time, I had also started mediating at the request of the federal court in San Francisco, and began exploring the behavioral sciences as to how individuals and organizations make decisions about pending or threatened disputes. Then, triggered by these and earlier studies of how people respond to adverse or catastrophic events, we began experimenting with early dispute resolution programs that channeled the findings discussed here. The major steps incorporating these lessons follow, all tested through the practices we employed.

Pre-mediation substantive dialogue

When asked, any litigator will say that they talk to opposing counsel several times before a mediation takes place. Now ask the same litigator how many times they have had two or more substantive pre-mediation discussions of strengths, weaknesses and alternatives ― in person or on the phone (self-serving letters and emails don’t count) ― and you often get a different answer. It may be resistance to sharing too much information; it may be the notion that substantive merits discussions are best left to the mediation itself. Either way, a deep dive into the substance of each side’s position is often delayed until the mediation itself.

The behavioral data argues for the opposite course. Knowing that lawyers and clients view their prospects through rose-colored glasses, the earlier the substantive dialogue starts, the better. Even if the information offered isn’t favorable, the sooner it surfaces, the sooner parties can start revising assumptions and re-examining their position.

This point is even more important as to claims against organizations with many actors in the mix. Absent substantive exchanges with the other side, groups often tend to coalesce around untested assumptions and unrealistic settlement expectations. Turning that ship around takes both time and substantive reasons to change course. Think ocean liner, rather than sail boat.

In both my litigator and general counsel roles, I witnessed the risks of the hermetically-sealed corporate meeting room. Like needed fresh air, contrary ideas and facts can be rare, discounted or discouraged, with bad results down the line when reality finally sets in. To avoid those results, we started requiring multiple substantive conversations between adversary counsel well before any mediation took place. The need for more and better information also trumped any notion of playing hide the ball. Our inquiry was simple: What do you see differently than we do? Obviously, the question needed to be accompanied by a genuine effort to share what we knew or didn’t know. Otherwise, a meaningful exchange was unlikely.

Taking this approach consistently paid off. If our own assessment was thorough and revealed no major weaknesses, the pre-mediation dialogue often led to a negotiated outcome at an appropriate level. If, on the other hand, the pre-mediation dialogue revealed material bad news, we could then update decision-makers and reset appropriate expectations before the mediation. And for all the “grey” matters in between, all sides were better prepared for the mediation session to follow.

Sharing mediation submissions: Briefs?

We don’t have to show you any stinking briefs!

With apologies to “The Treasure of the Sierra Madre,” the failure to share briefs is a wasted opportunity, given the need to overcome ingrained biases and the time often needed to do so.

A well-constructed brief focusing on core facts, key legal issues and damage calculations should preview what a judge, jury or arbitrator will hear. If compelling, it should motivate the other side to set reasonable expectations for the mediation. By contrast, failing to share mediation briefs usually leaves the client with only their own counsel’s brief to rely upon. That only reinforces self-serving biases, making it harder to reset expectations later.

Here, tone and temperament are key. To overcome self-serving biases and convince the other side to reassess, you must first be heard. A mediation brief laced with adjectives, invective and insults will assuredly trigger defensive posturing and counter-attacks on the other side, rather than a real exchange on the core issues. And it won’t impress the mediator either. Believe me.

For what it is worth, the inclination to confuse an aggressive tone with effective advocacy appears to start early on. Maybe it’s the many movies, television shows and books that value domineering behavior and discredit a dispassionate discourse. But it doesn’t work; it’s counterproductive; and it squanders a key opportunity to really be heard by the other side when being heard matters most.

Sharing briefs is arguably more important with multiple actors and constituents on the other side. Organizations with various stake holders, inside and outside counsel and insurers require consensus to set – and time to reset ― settlement parameters. Shared briefs provide a substantive basis for reassessment as well as time to do so before the mediation starts. For anyone who has experienced a mediation session that needs to be halted and resumed later after that session uncovers key information that requires a new round of executive conversations, you know what I mean.

Finally, sharing briefs does not foreclose supplemental letters for the mediator’s eyes only with any content deemed helpful but very sensitive. But the default should be to show more, not less. If truly impactful, it will help reset expectations and prompt the desired result.

Joint sessions: Think conversation, not conflagration

Joint mediation sessions provide the rare opportunity to be heard directly by the other side, to learn what the other side sees differently, and to dispel misimpressions about you and the strengths of your position. Then why have they fallen out of favor?

Discomfort with a potentially volatile dialogue prompts many attorneys to avoid putting adversaries in the room together. Indeed, most experienced litigators have one or more stories about a joint session gone awry ― lawyers behaving badly, clients becoming irate or irrational, and mediators losing control of the room. But lost in these anecdotes is the reality that a properly conducted joint session is a prime opportunity to challenge assumptions and demonstrate that your story (or theirs) may play well before a judge, jury or arbitrator if the dispute does not settle.

Indeed, didn’t we pick litigation as a career because we believed we were effective advocates? If so, we should be able to channel those skills during a direct dialogue with the other side, particularly if we treat the session as a conversation, rather than a conflagration. Invite conversation by explaining your position in the most fact-based, invective-free manner. Then ask, what’s wrong with our picture? The combination of an insult-free presentation and genuine curiosity as to what the other side sees differently is most likely to overcome the biases of both counsel and client on the other side. Doing so should in turn significantly bridge the gap on an acceptable settlement.

Other arguments for avoiding joint sessions include the absence of clients with real control over the settlement – class actions, for example – and the perception that the adversaries are incapable of rational discourse. Here again, our actual experience produced much better results than predicted if we took the steps outlined here to overcome these pre-existing biases and unduly rosy assessments.

In the class-action area, for example, the absence of underlying clients with a significant voice rarely deterred a meaningful mediation if we held substantive pre-mediation conversations, exchanged useful information, and thoroughly and civilly briefed core issues. Indeed, skilled counsel proved very adept at assessing value, potential future sunk costs, and reaching an appropriate settlement with the aid of a capable mediator.

As well, predictions of obstreperous mediation behavior from the other side rarely panned out. Experienced counsel on both sides realize the downside of unruly behavior: It only undermines your credibility with the mediator as well as the prospects of overcoming biases and misimpressions from the other side.

Measuring success

When we began this approach, our primary benchmark was whether it reduced the overall direct cost of legal disputes in terms of legal fees, in-house costs, penalties, settlements. Turns out it did all that, and more. Beyond direct savings, the indirect cost of continuing to litigate in terms of lost client time and opportunities was significantly reduced. So were the number of unpleasant surprises and results from sorting out these problems later. Money saved; time saved; sometimes people saved as well. Remember Lord Molson and give it a try.