Bracketing in Mediation: When to Use It and Why It Works

Participants in mediations will be familiar with bracketing — a technique widely used by mediators to bridge seemingly insurmountable gaps between what one side is offering and the other side is demanding. Why bracketing works offers a fascinating glimpse into the psychology of negotiation and mediation.

Let’s assume a plaintiff has sued a defendant for trademark infringement. The litigation has been pending for three years, and there has not yet been any mediation. The plaintiff argues its liability case is very strong, and in direct settlement negotiations with the defendant, demands $25 million. The defendant rejects the demand because it claims it sees things very differently — based on its read of applicable law and the discovery taken to date, it contends that plaintiff’s liability case is very weak. But to spare itself from further defense costs, defendant counters with $1 million.

The midpoint (derived by summing the two numbers and dividing by two) is $13 million, but that number is meaningless since defendant is not going to pay $13 million for a claim it contends is frivolous, and plaintiff is not going to accept $13 million for a case it claims is worth at least $25 million.

At any rate, $24 million is a pretty huge delta. One might wonder how parties looking at the same case could ever be so far apart, and it could very well be that one party knows its case is weak, but is posturing to avoid signaling weakness, as well as to plant seeds of doubt in the mind of its adversary and erode its resolve. On the other hand, as discussed in a prior post, by virtue of their ethical obligation to zealously advocate for clients, lawyers are highly prone to confirmation bias. In other words, they tend to fall in love with their own arguments, and discount the legitimacy of any arguments to the contrary — especially when advanced by an adversary they deem to be stubborn and unreasonable (a phenomenon known as “reactive devaluation”). Thus, each side may genuinely believe that its probability of success is materially higher than 50/50.

Posturing and/or confirmation bias make it difficult for parties to bridge large gaps on their own through direct negotiations. Each side accuses the other of being unreasonable and unrealistic. The problem is, if the parties don’t settle and go to trial, both of them can’t be right on the question of liability. Instead, assuming damages may be close to $25 million if liability is established, the outcome is binary — one side is going to be correct in its assessment of liability and win big, while the other side is going to get crushed.

Enter a mediator to serve as what Stephen Hochman likes to call an “agent of reality” with no stake in the outcome. Assume after extensive caucusing, and a thorough evaluation of the parties’ contentions, the mediator convinces both sides that while their respective arguments have some strengths, there are also glaring weaknesses. If both sides trust the mediator’s evaluation of the merits, they may each tell the mediator they are ready to make meaningful moves to bridge the gap. But neither wants to go first because it signals weakness. Additionally, neither side is interested in a drawn- out negotiation consisting of tiny, tit-for-tat moves that don’t meaningfully advance the dispute towards resolution.

A great option in such circumstances is bracketing. To illustrate, after weighing the strengths and weaknesses of both sides, and the likely range of litigation outcomes (ideally using decision tree analysis to derive an expected value for the case), the mediator might begin by asking the defendant, “if I can persuade the plaintiff to come down to $20 million, would you be willing to come up to $7.5 million?” The midpoint of the mediator’s proposed bracket is $13.75 million, which is not much different than the $13 million midpoint when the parties were engaged in direct negotiations. But now the midpoint is significant because it represents the midpoint of the settlement range within which the mediator thinks the case ought to settle (and thus the parties can negotiate productively). In effect, then, the mediator’s proposed bracket conveys his belief that $13 million represents a realistic settlement amount (or in statistical terms, an “expected value” were the case to be tried one hundred times).

Of course, the defendant might reject the mediator’s opening bracket, and counter with a proposed bracket of, say, $10 million and $2 million, which brings the midpoint down to $6 million. In this way, the defendant signals that while it accepts the mediator’s assessment that the plaintiff’s liability case has some merit, it remains far more confident than the mediator that it would ultimately prevail. The counter bracket thus signals defendant’s belief that $6 million is a more realistic number in terms of where it might (although not necessarily) be prepared to move.

The mediator may need to shuttle a bit back and forth between the parties, but assuming both parties trust the mediator’s judgment, the mediator should be able to pin down a mutually agreeable bracket within which range the parties can continue negotiations and make further moves (consisting of either increasingly tighter brackets or outright offers and counteroffers).

The value of bracketing is that it creates the perception of momentum towards resolution, which generates the optimism necessary to motivate further negotiation. It also allows each party to signal flexibility without signaling weakness because, for example, when the mediator communicates the proposed bracket to defendant he indicates that the settlement range reflects his own personal assessment, and he may need to twist the plaintiff’s arm a bit to make the proposed move (and vice versa when he proposes a bracket to the plaintiff). Finally, as experienced mediators Michael Young and Marc Isserles observe, bracketing takes advantage of the psychological phenomenon known as reciprocation bias. Whereas without bracketing, a party might refrain from a making meaningful move because it is concerned the other side won’t reciprocate, bracketing entails reciprocal moves by both sides simultaneously.

There’s also cognitive dissonance at play as well. Whereas before the mediation started, each side viewed the other as unreasonable and unrealistic, the reciprocal moves implicit within bracketing paint a different picture: because the action of a party agreeing to a bracket is inconsistent with the belief that the party is stubborn, each side starts to view its adversary as flexible and serious about settling the case.

Of course, bracketing is not always appropriate. For example, after the mediator evaluates the strengths and weaknesses of both sides, he may conclude that the defendant’s assessment is correct and the plaintiff’s claim is very weak. The mediator would then lose all credibility with the defendant were he to propose that the defendant entertain a $7.5 million/$20 million bracket (which implies paying $13.75 million to settle a claim with little probability of success). Under such circumstances, the mediator would need to have a frank discussion with the plaintiff, and press it to take a more realistic view of its case.

Timing is also important. Experienced mediators such as Stephen Hochman and Deborah David caution that unrealistic expectations need to be lowered gradually. Therefore, bracketing should not be introduced before the mediator has engaged in what Hochman refers to as “reality testing” with each of the parties concerning the strengths and weaknesses of their case. Such evaluative discussions prime the parties to entertain the bracket proposed by the mediator. For example, as noted, as long as the defendant in our hypothetical views the plaintiff’s claim as frivolous, it won’t accept a bracket with a meaningful midpoint because otherwise it would be conceding the claim has merit. Accordingly, the mediator first needs to persuade the defendant concerning the merits of at least some of the plaintiff’s positions by playing devil’s advocate (for example, by observing that the defendant is taking an  unrealistic view on an issue such as likelihood of confusion).

In sum, bracketing is a valuable technique in the mediator’s toolkit that can break an impasse when deployed appropriately.