Improving Results in Mediation: Should Lawyers Make Extreme Demands To Take Advantage of the Anchoring Effect?

I wrote recently about the impact of the anchoring effect on mediation. Improving Results in Mediation: Should the Defense Request a Pre-Mediation Demand? I wrote that human beings tend to rely too heavily on the first piece of information they receive when making decisions. This is known as the anchoring effect, and that first piece of information is known as the anchor. For example, a high initial sales price for a used car serves as the anchor for the rest of the negotiations, dragging the final sales price higher than a more reasonable initial demand would. Numerous experiments have shown that the anchoring effect exerts a powerful pull, even on experienced professionals. In one study, a simple roll of the dice had an astonishingly strong influence on the sentencing decisions of highly experienced criminal judges. Finally, I explained that because of the anchoring effect, the defendant should not ask the plaintiff for a demand until the defendant sets its own anchors by determining its "walk-away" and "go-for" numbers.

Since I published the article, I've received numerous questions along these lines: "Are you saying that the plaintiff should make a very high demand early in every case? If my demand can anchor the negotiations, shouldn't I make an exceptionally high demand every time, even if the case does not have a high value?"

The answer is no. Despite the impact of anchoring on negotiations, plaintiffs' lawyers should avoid making unreasonably high demands, and the reason is credibility. Credibility is critical in every ongoing relationship, and the prosecution or defense of a law suit is very much an ongoing relationship, one that often lasts for years. This all sounds obvious, I know, but it is very important. While demands can anchor settlement negotiations, extreme numbers damage counsel's credibility, and this loss of credibility can more than offset any gain realized as a result of the anchoring effect.

In order to understand the importance of credibility in negotiation, one first must understand this: Parties derive power in negotiation from their alternatives. "What happens next if I don't get this deal done?" Mediators like to talk about this in terms of a party's best alternative to a negotiated agreement (BATNA) or worst alternative to a negotiated agreement (WATNA). The better a party's alternatives, the more comfortable she will feel walking away, the harder she will press the other side to meet her conditions, and the greater her power at the bargaining table.

Think of a well-informed consumer shopping for a car. He knows that he can get a good price from Dealer A, and this gives him power in his negotiation with Dealer B. He feels comfortable pressing Dealer B to meet his terms, and if Dealer B won't do the deal, he has no problem getting up and walking away. And of course once the dealer knows that the buyer is willing to walk away, he becomes more flexible.

This operates in litigation as well. If a party has a high level of confidence that it is going to win the case, then its alternatives to a negotiated agreement look positive, it feels powerful in the negotiation, and it will have no problem walking away if its conditions of satisfaction are not met.

Back to credibility. Counsel's credibility has a strong impact on how the opposing side sees its alternatives. For example, attorneys frequently tell each other that if a case does not settle, they will litigate aggressively, leave no stone unturned, and take the case to trial. Where opposing counsel has a low level of credibility, the threat of a tough road ahead is minimized, the party's alternatives to a negotiated agreement look better, and the party feels more powerful at the negotiating table. Of course, the threat of hard-fought litigation and trial sounds very different when it comes from an attorney with a high level of credibility.

As another example, plaintiff's counsel in a recent mediation gave me a witness statement to show to the defendants (with the witness's name redacted), and asked me to tell them that she had two more witnesses prepared to give similar testimony. Plaintiff's counsel had credibility with the defense, and she had not damaged that credibility by making outrageous demands in mediation. Although defendants did not take the witness statement or counsel's representation at face value, they did take them into account, and they did increase the settlement value at the end of the day. The witness statement would not have had the same effect if plaintiff's counsel did not have credibility in the other room. 

Plaintiff's attorneys must step carefully in crafting their opening demands. Though they know that their numbers will have an anchoring effect on the negotiation, counsel should consider carefully the effect that overly aggressive demands will have on their credibility -- and the importance of credibility in closing the deal. 

Improving Results in Mediation: Should the Defense Request a Pre-Mediation Demand?

The study of human decision-making holds important lessons for people involved in all types of negotiations, including mediation. Learning the ropes will help the attorneys and their clients achieve better results in mediation.

Employee E sues her former employer, Company C, alleging harassment, discrimination, and retaliation. Company C answers, and the parties conduct discovery and take the depositions of Employee E and other key witnesses. The court orders the parties to mediate, and they select their mediator and set a date.

In order to get a better idea of the plaintiff’s expectations and the likelihood of resolution, defense counsel asks plaintiff’s counsel for a pre-mediation demand. Plaintiff’s counsel says that she will discuss the issue with her client and call defense counsel back if they decide to convey a number. Defense counsel asks again two weeks later, with the same response. The day of mediation arrives, and defense counsel tells the mediator that he still does not have a demand. When Employee E makes her demand an hour into the process, the demand is far more than Company C had anticipated or prepared for. Defense counsel expresses his frustration to the mediator, stating that he has been sandbagged, and that the delay decreases the probability of resolving the case.

It is easy to understand why defense counsel would have wanted a pre-mediation demand, and easy to understand his frustration, but the underlying question is this: Is a pre-mediation demand good for the defendant? The field of behavioral economics, which studies the way that humans make economic decisions, suggests strongly that a pre-mediation demand helps the plaintiff move the final settlement number to his or her favor, and the defendant should avoid a pre-mediation demand at all costs.

Scientists have found that a cognitive bias -- a tendency to think in a certain way -- known as the anchoring effect has a surprisingly strong impact on our decisions. Simply put, human beings tend to rely too heavily on the first piece of information they receive when making decisions. This is known as the anchoring effect, and that first piece of information is known as the anchor. For example, a high initial sales price for a used car anchors the rest of the negotiations, dragging the final sales price higher than a more reasonable initial demand would.

A number of studies demonstrate just how strong the anchoring effect is. In one study, researchers asked two groups of students to estimate how old Mahatma Gandhi was when he died. (Strack and Mussweiler, 1999). The researchers asked the first group whether he was older or younger than nine when he died. Of course, the students knew that he was older than nine. When researches then asked his actual age, the first group, on average, said 50. Researchers asked the second group whether he was older or younger than 144. These students answered correctly and then gave an average answer of 67. The only difference between the two groups was the obviously inaccurate number in the initial question. Yet this seemingly irrelevant difference actually moved the students’ answers by more than 33%.

Back to the question at hand: Should the defendant request a pre-mediation demand? The Gandhi study and many others suggest that obtaining the plaintiff’s demand in advance will anchor the negotiations, dragging the settlement figure higher than it otherwise would be. This obviously is a bad idea for the defense.

“But I have years of experience in litigation and mediation,” you say. “I know the reasonable settlement range and the likely outcomes on summary judgment and at trial. Just getting the plaintiff’s number isn’t going to affect me.”

Yet other studies suggest that the anchoring effect exerts a powerful force even on experienced professionals. In one study, researches asked a number of very experienced judges (averaging more than 15 years on the bench) to read a description of a woman who had been caught shoplifting and then asked them to roll a pair of dice. Yes, a pair of dice. Once they rolled the dice, the judges were asked to sentence the woman. Unbeknownst to the judges, the dice were loaded to land on either three or nine. Shockingly, those who rolled a three gave an average sentence of five months; those who rolled a nine gave an average sentence of eight months. (Kahneman, 2011.) The mere roll of the dice impacted these sentences by more than 50%. Other studies have shown similar results.

So how does a defendant avoid the anchoring effect of the plaintiff’s initial demand? After all, even if the plaintiff does not make the demand in advance, it typically will be the first number on the table at the mediation.

The answer is this: Defendants must create their own anchors before they hear the plaintiff’s opening number. Fortunately for the defense, they frequently do exactly that through early evaluation, reporting to clients and carriers, and the setting of reserves and settlement authority. All of these numbers, developed before the defense hears the plaintiff’s demand, serve to anchor the negotiations for the defense. Armed with their own anchors before they hear the plaintiff’s demand, defendants are in a much better position to weather the storm and keep the final settlement figure closer to their own evaluation of the case’s reasonable settlement value.

In sum, the last thing that a defendant wants to do is to hear the plaintiff’s demand in advance of mediation, allowing that number to anchor the negotiations before the defendant has set its own anchor firmly in the sand.

Mediation and the Science of Decision Making: Conclusion

Daniel Kahneman and Amos Tversky revolutionized economics by changing the focus from hypothetical “rational actors” to real people, who frequently make bad decisions as a result of behavioral patterns and cognitive errors of which they are not aware. Learning to recognize these patterns and errors will help you be more successful in all decision-making processes and in all negotiations. And a mediator who understands these patterns and errors and how to deal with them will do a better job helping the parties resolve their cases.

Here are the links to all eleven installments of this series: 

Part I: Introduction 

Part II: People Act Rationally, Don’t They? 

Part III: People Avoid Risk When They Stand to Gain and Seek Risk When They Stand to Lose 

Part IV: People Evaluate Gains and Losses Relative to Their Reference Points 

Part V: The Greater the Loss or Gain, the Less Any Incremental Change Matters 

Part VI: People Hate Losing More Than They Love Winning. Losses Loom Larger Than Gains. 

Part VII: States of Change -- The Possibility and Certainty Effects 

Part VIII: Set Your Reference Points Before You Begin Negotiating 

Part IX: Think Like A Trader 

Part X: Shift Reference Points And Use The Endowment Effect To Close The Deal. 

Part XI: Do The Math, With Pencil And Paper.

Mediation and the Science of Decision Making Part XI: Do The Math, With Pencil And Paper.

We talked last time about using the endowment effect and shifting reference points to help close deals.  But how do you actually do that?  One good way is by showing the parties the numbers: "What will you gain by resolving the case?  What will you lose if you don't?" 

Some people are very good at math and can run numbers in their heads. Others require a calculator or adding machine. In any case, all of us are subject to the cognitive errors identified by Kahneman, and it can be very difficult to identify and compensate for these errors when called upon to make important decisions. 

One of the best ways to help people avoid these errors is to work through the numbers with them as they make their decisions. And because most people are not great at doing math in their heads, it helps to work through the numbers on paper, or on a white board, which is particularly good for this purpose. Putting the numbers on paper makes them more concrete, increases the likelihood that the parties will adopt them as reference points – even if they differ from their original reference points – and helps the parties evaluate them rationally, rather than emotionally. 

For example, while a party typically will evaluate a $50,000 settlement intuitively as a loss or gain relative to his existing reference points, he typically will not evaluate its expected value relative to a 50% chance of a $100,000 judgment at trial. Explaining and walking through this type of calculation makes the settlement more real, increases the likelihood that the party will adopt it as a reference point, and helps the party accept or reject it on rational, rather than emotional, grounds.  

Mediation and the Science of Decision Making Part X: Shift Reference Points And Use The Endowment Effect To Close The Deal.

Very frequently, the parties to a negotiation will set reference points that do not overlap or allow for easy resolution. The plaintiff may have a bottom line of $200,000, while the defendant has a top dollar of $100,000. The likelihood of mutually exclusive reference points is greater in employment lawsuits, which frequently are zero-sum-games. A zero-sum-game is a situation in which the only issue to be resolved is how to allocate a limited resource, in this case money. (Some employment actions are not zero-sum-games, as where the parties will continue to be involved with each other in the future, and there may ways to “grow the pie,” rather than just dividing it.)  

In a zero-sum-game, where the parties’ reference points do not overlap, resolution is possible only if a party accepts a loss relative to his or her reference point. This "loss" may be the plaintiff accepting less than her bottom line or the defendant paying more than her top dollar. Frequently, both parties must accept such losses in order to get a deal done. As we have seen, people have a very hard time doing this. (See People Hate Losing More Than They Love Winning. Losses Loom Larger Than Gains.)  

A skilled mediator can help the parties move toward resolution by helping them shift their reference points. Shifting reference points involves what behavioral psychologist Richard Thaler dubbed the “endowment effect.” Thaler, a collaborator of Kahneman and Tversky, found that, contrary to rational choice economic theory, people frequently like to keep what they have, even when offered a fair price in exchange. This is particularly true for the things that people hold for their own use or enjoyment, things like nice bottles of wine, tickets to a sold out concert, or a coffee mug with one’s college logo on it.  

In mediation, the endowment effect comes into play when the parties begin to see themselves as already enjoying the benefits of a settlement. For the plaintiff, the benefits of settlement may include the peace of mind of not having to testify at deposition or trial or being able to pay off credit cards used for living expenses after a termination. For the defendant, the benefits may include cutting litigation expenses and being able to focus on running the business, rather than defending the litigation. 

The trick for the mediator is to help the parties see the benefits offered by resolution and then adopt those benefits as their new reference points. Once the parties accept the new reference points, they will be reluctant to give them up, making resolution more likely.  

Next time, another useful tool for closing deals: Do The Math, With Pencil And Paper.

Mediation and the Science of Decision Making Part IX: Think Like A Trader

Last time, we looked at one way to use behavioral economics to help our clients make better decisions in mediation by setting reference points before we begin negotiating.  Today we discuss another way to do better in any negotiation: thinking like a trader.  

Most attorneys have a great deal of experience with negotiation and may even consider themselves to be professional negotiators. (Interestingly, Kahneman and Tversky found that even professionals make the same cognitive errors as the rest of us, but we will put that aside for the moment.) 

Most parties do not have the same level of negotiation experience as their lawyers. And even if they do, the stress of litigation and mediation leaves them less capable of making rational economic decisions. Although there are benefits to long days of mediation, experienced negotiators know that 12 or 14 hours of mediation can reduce one’s ability to think clearly and make good business decisions. Especially at the end of a long day, parties need to understand the impact that cognitive errors have on them.  

Kahneman and Tversky tested ways to help people be less sensitive to these cognitive errors. They found first that professional traders in the financial markets – people who make decisions on gambles for a living – are more tolerant of losses and as a result suffer from fewer cognitive errors when making financial decisions. 

Amazingly, one simple way to help people make better decisions is to tell them to “think like a trader.” Participants in experiments become less risk averse, and their emotional reaction to loss (as measured physiologically) decreases when given this simple advice. In other words, simply asking people to “think like a trader” gives them a better framework for analyzing their choices and helps them make better, more rational decisions.   

A simple yet powerful tool that works well when it's time to close the deal. Next time, another tool for closing deals: Shifting Reference Points And Using The Endowment Effect. 

Mediation and the Science of Decision Making Part VIII: Set Your Reference Points Before You Begin Negotiating

We have seen that behavioral economics holds valuable insights into the way that people make decisions. Specifically, behavioral economics teaches that behavioral patterns and cognitive errors play a tremendously important, if seldom recognized, role in decision-making. 

Now we get to the fun part: learning to use behavioral economics to help our clients and ourselves make better decisions, not just in mediation, but in every negotiation.  

Lesson One: Set Your Reference Points Before You Begin Negotiating. 

Reference points are extremely important in every negotiation. They can be the key to an effective negotiation strategy, but they also can stand as an obstacle to a good settlement. Effective negotiators know how to use them to drive negotiations and to help close deals. 

In Thinking, Fast and Slow, Nobel Prize winner Daniel Kahneman explains that people frequently adopt their goals as reference points. In golf, for example, par is a reference point. A birdie (one shot under par) represents a gain, and a bogey (one shot over par) represents a loss relative to the reference point, par. 

We have seen that people dislike losing more than they like winning, and they work harder to avoid a loss than to make a gain. Economists who have studied golf (yes, there are such people) have found that professional golfers prove this point: they do far better when putting for par (avoiding a bogey, which is a loss relative to par) than when putting for birdie (making a gain over par). Whether the putt is easy or hard, at every distance from the hole, professional golfers are 3.6% more successful when putting for par than when putting for birdie. Over the course of a 72-hole tournament, this easily can be the difference between winning and losing. 

Like golfers, parties frequently set reference points prior to a negotiation. Plaintiffs in mediation frequently know the “bottom line” number below which they will not move, and defendants frequently know their “top dollar.” These “walk-away” numbers become the parties’ reference points. A settlement that improves upon one’s reference point creates a gain; one that does not creates a loss. (It may seem strange to refer to a defendant who pays less than its top dollar as experiencing a gain, but remember that we are speaking of results relative to reference points, and even cold water can feel warm to a hand that has been sitting in a bowl of ice water.) 

Because people work harder to avoid losses than to make gains relative to their reference points, all parties should establish their reference points before they start negotiating. Plaintiffs should know their bottom line, and defendants should know their top dollar before the mediation starts. Surprisingly, many parties walk into mediation with no reference points, thinking only that they want to do “the best they can.” 

Experienced negotiators will set not only the walk-away numbers beyond which they will not move, but also goals that are better than those walk-away numbers. Parties who set “shoot-for” numbers as their reference points typically do better than those who only formulate walk-away numbers. 

For example, a defendant who decides that it will not pay more than $100,000 but has the goal of settling for $80,000 will settle for closer to $80,000 than one who only sets its top dollar number. The same goes for plaintiffs: one who establishes a shoot-for that is higher than her bottom line will recover more than one who does not. The reason for this is simple: once the party sets a shoot-for number as her reference point, she will work very hard to defend it, as anything that falls short will be seen as a loss.  

Setting one's reference points before the negotiations start is a simple strategy that will produce demonstrable results in your next mediation. 

Next time: Think Like A Trader. 

Mediation and the Science of Decision Making Part VII: States of Change -- The Possibility and Certainty Effects

Consider this. You have a chance to win $1,000. Your chances improve by 5% in each step below. Is the news equally good in each case? 

Step A: Going from 0% to 5% 

Step B: Going from 5% to 10% 

Step C: Going from 60% to 65% 

Step D: Going from 95% to 100%

The answer of course is no. Step B (going from 5% to 10%) and Step C (going from 60% to 65%)  are merely quantitative in nature, and we perceive them as representing relatively minor incremental gains. (A 65% chance feels the same as a 60% chance, doesn't it?)  

Step A, the change from 0% (no chance) to 5% (a small chance) is far more significant. It creates a possibility that did not exist before, the possibility of winning $1,000. Kahneman calls this qualitative change the “possibility effect.” He explains that it leads people to give highly unlikely events (like the 5% chance of winning $1,000 or the infinitesimally small chance of winning the lottery) greater weight than their mathematically expected value. In other words, the expected value of a 5% chance of winning $1,000 is $50 (5% x $1,000 = $50), but we experience the value as something substantially higher. Similarly for the lottery. The mathematical value of a lottery ticket is less than the $1 we pay for it, but the possibility of winning a huge sum causes many of us to ignore the mathematical reality and buy the ticket.

Step D, the change from 95% to 100%, also is qualitative. It changes the possibility of gain into a certainty. Not surprisingly, Kahneman calls this the “certainty effect.” Like the possibility effect, the certainty effect has a disproportionate impact on decision-making. People give outcomes that are almost certain (like a 95% chance of winning $1,000) less weight than they should using expected values. Thus, the expected value of a 95% chance of winning $1,000 is $950 (95% x $1,000 = $950), but we experience the value as something substantially lower. 

Don't believe it? Try this. Would you pay $925 or $940 for a 95% chance at winning $1,000? It feels too risky, doesn't it? 

(Those who are not afraid of a little math may consider this: the possibility and certainty effects are mirror images of each other. The possibility effect leads us to over-value the 5% chance of winning the $1,000. Given a 95% chance of winning, the certainty effect leads us to over-value the 5% chance that we will not win. In other words, we over-estimate the negative value of the possibility of losing. But they said there wouldn't be math in law school, so I'll move on.)  

How do the possibility and certainty effects impact parties at mediation? Consider again the examples that I gave in my first post.  

Adam has a very small chance of winning at trial, but if he does win, he could recover a substantial amount. He falls victim to the possibility effect, foregoing a reasonable settlement to chase the possibility -- no matter how small -- of a substantial recovery at trial. 

Company C makes a similar mistake. It faces a high likelihood of a very bad result at trial, but it rejects the opportunity to settle in a reasonable range before trial. Instead, it seizes on the possibility -- small though it may be -- of winning at trial.  

Both Adam and Company C make the mistake of over-valuing small possibilities. Like a person who buys a lottery ticket, These mistakes result from a combination of the possibility effect and loss aversion (discussed here).  

Also consider Denise. She accepts an offer that is relatively small, compared to the very high likelihood that she will obtain more at trial. Like Adam and Company C, she is influenced by loss aversion, but for her it is combined with the certainty effect. She gives the likelihood of a favorable result less weight than she should because it is a mere possibility, rather than the certainty that she craves. At the end of the day, she achieves certainty, but she gives up the very likely probability that she would have done better at trial. 

Now that we have seen some of the cognitive errors that affect people when they make decisions, we can turn to the question at hand: How do we use this knowledge to achieve better results in mediation?  Stay tuned.  

Mediation and the Science of Decision Making Part VI: People Hate Losing More Than They Love Winning. Losses Loom Larger Than Gains.

A third principle of Daniel Kahneman’s work, together with the idea that people evaluate losses and gains relative to their reference points (discussed here) and the idea that people have a diminishing sensitivity to losses and gains (discussed here), is the idea that losses loom larger than gains. Kahneman traces this to the evolutionary need to be alert to danger.

Remember how the body reacts to the idea of financial loss? The idea of financial loss induces the classic fight or flight response because we perceive it as a risk to our well-being. The idea of gain, however, does not induce an equal but opposite visceral reaction.  

To illustrate the idea that losses loom larger than gains, ask yourself whether you would take the following gamble on the toss of a coin: If the coin shows heads, you win $150; if it shows tails, you lose $100. Would you take the bet?  

The simple math shows that this is a good bet. You have a 50% chance of winning $150 and a 50% chance of losing $100, so the expected value of the gamble is $25.  Here are the numbers : (50% x +$150) + (50% x -$100) = $75 - $50 = $25. 

Despite the math, most people would reject this gamble. You probably felt this yourself when you considered the gamble, and it made you feel nervous. In fact, most people require a potential gain of more than two times the potential loss in order to accept the gamble. In other words, I would have to offer you $200 or more to get you to accept a 50% chance of winning $100.  In mixed gambles – that is, situations in which both gains and losses are possible – this imbalance causes people to make extremely risk-averse choices. 

Consider a typical business dispute. Each party is making a claim against the other, so each party faces both the possibility of gain and the possibility of loss. The science suggests that both are likely to be relatively risk averse and make more conservative decisions. 

(As you read this, did you think of a notable business lawsuit in which the parties seemed to show no desire to avoid risk? The Bratz dolls case came to my mind as I wrote it. If you had a similar thought, you likely concluded that the science is wrong on this point about mixed gambles. Your reaction is a function of what Kahneman calls "associative memory," a powerful source of cognitive errors that Kahneman discusses at length, but that will have to wait for another day.) 

In contrast to the typical business law suit, in most employment actions, only the employee has claims against the employer. Further, the law typically allows a successful employer to recover its attorney fees in limited situations, if at all. Because the employee typically does not face this type of potential loss, the typical employment law case is not a mixed gamble, and the employee may feel free to make decisions that entail greater risk. 

However, as we will see next time, even an employee in this type of litigation will be influenced by the risk of loss. 

Mediation and the Science of Decision Making Part V: The Greater the Loss or Gain, the Less Any Incremental Change Matters

In previous posts, I've discussed the theory that individuals act rationally and the efforts of Daniel Kahneman, Amos Tversky, and others to disprove that theory. For example, as I discussed here, Kahneman and Tversky pointed out that people avoid risk when they stand to gain and seek risk when they stand to lose -- hardly rational behavior.  And as I discussed here, people evaluate losses and gains relative to their own reference points, which change over time, rather than to fixed reference points that apply equally to all.  

Another key element of Kahneman’s work is the principle of diminishing sensitivity to loss and gain.  The idea is this: the difference between a gain of $100 and a gain of $200 is $100.  The difference between a gain of $10,100 and a gain of $10,200 also is $100, but the incremental gain in the second example does not have the same impact as the incremental gain in the first.  In other words, for most people the difference between $100 and $200 is far more significant than the difference between $10,100 and $10,200.  

For example, Richard Thaler in 1980 posited that people would go out of their way to save $5 on a $25 purchase, but would not go out of their way to save $5 on a $500 purchase. Others have confirmed this in numerous studies in the years since. Kahneman and Tversky found that 68% of people would drive 20 minutes to save $5 on a $15 calculator (when they also were hypothetically buying a $125 jacket), but only 29% would do so to save $5 on a $125 calculator (and they also were hypothetically buying a $15 jacket). 

The total purchase price is $140 in each scenario, and the total savings is $5, but the human reaction is completely different.  That could not be less rational.  

Diminishing sensitivity to incremental change helps to explain the irrational risk avoiding / risk seeking behavior we discussed earlier.  Consider Denise, the salesperson who took the sure thing, even though she likely could have obtained more by taking more risk.  Part of her reason for accepting the settlement may have been her sense that the settlement offered was a large amount of money, and she would not gain dramatically from any incremental increase over and above that amount.  Her sensitivity to any potential increase diminished as her position improved. 

Diminishing sensitivity to loss also helps explain why Company C would choose to take its chances in court. When people face bad situations – for example, when they face a certain loss in settlement, as compared to a larger loss at trial that is not certain but is merely probable – diminishing sensitivity causes them to ignore the greater risk of going forward. In all likelihood, settlement is Company C's best course of action, but its diminishing sensitivity to the potentially greater loss at trial causes it to ignore that greater risk. In other words, it figures that if it is going to lose anyway, it might as well take a shot at trial, where it could lose zero, even if the odds of that outcome are relatively low.   

What all of this points to is this: the diminishing sensitivity to incremental loss and gain has a tremendous impact on the way that people make decisions. Those of us who work to help people resolve difficult situations must understand these cognitive errors in order to do the best job possible for our clients.  

Next time, we will discuss why the difference between a $10,000 loss and an $11,000 loss is much greater than the difference between $10,000 gain and an $11,000 gain:  Losses Loom Larger Than Gains. 

Mediation and the Science of Decision Making Part IV: People Evaluate Gains and Losses Relative to Their Reference Points

I spoke earlier about the idea that individuals act rationally. In other words, they seek to maximize their satisfaction based on their preferences and the information available to them, in a way that changes little over time or in different contexts. One of the reasons that Daniel Kahneman won the Nobel Prize in 2002 is because he and his colleague, Amos Tversky, challenged this idea of rationality.  

For example, Kahneman and Tversky realized that -- contrary to modern economic theory -- gains and losses are not absolute. They are not perceived in the same for all people, and they are not perceived consistently over time. They change depending on one’s circumstances, or reference points. In Thinking, Fast and Slow, Kahneman illustrates this with a simple example. Hold one hand in a bowl of hot water and the other in a bowl of cold water, then put both in a bowl of room temperature water. One feels warm, the other cold. The starting (reference) point for each is different, and so is the perception of the ending point.  

Reference points play a very important role in financial situations. Consider Jack and Jill. Each has $5 million. Yesterday, Jack had $1 million, and Jill had $9 million. Classical economics holds that the $5 million has the same utility for each, and they are equally happy. But are they? Obviously not. Kahneman and Tversky, who brought human psychology to economics, recognized that Jack and Jill’s happiness depends not only on what they have, but also on what they had – their reference points. Jack is happy with his $5 million; Jill is not at all happy with hers.

This simple idea that reference points are relative and may change over time may seem obvious, but it revolutionized economic theory and it plays an extremely important role in understanding both how people make decisions in reality and how to help them make better decisions. 

In later posts, I will discuss the importance of reference points in mediation and the use of reference points to help seal the deal.  Next time: The Greater the Loss or Gain, the Less Any Additional Incremental Change Matters.

Mediation and the Science of Decision Making Part III: People Avoid Risk When They Stand to Gain and Seek Risk When They Stand to Lose

In my last post, I said that Kahneman and Tversky revolutionized economics by showing that people do not act rationally when making economic decisions. As I explained, classical economic theory predicts that people will analyze risk rationally, regardless of the circumstances.  As it turns out, people don't do this. Instead, they try to avoid risk when they stand to gain, but they seek risk when they stand to lose. 

Try this mental exercise, and you'll see what I mean.  

Which would you choose? (A) a sure gain of $750 or (B) an 80% chance of winning $1,000 and a 20% chance of winning nothing?

The rational choice theory predicts that people will choose Option B, which has the higher expected value. The expected value of an 80% chance of winning $1,000 is $800. That obviously is greater than the value of the $750 sure thing. Any rational person seeking to maximize his financial gain should choose Option B.

But if you are like most people, you chose Option A and you didn't even have to think about it. Your automatic reaction was that you would rather have the sure thing than take the gamble. If you did the math, you likely noticed that the gamble offered only an incremental gain over the sure thing, but it also left you with the possibility of getting nothing. You likely saw this and figured, “Why risk it?”

Now try this one. Which would you choose? (C) a sure loss of $750 or (D) an 80% chance of losing $1,000 and a 20% chance of losing nothing?

Here, the rational actor theory predicts that people will take the sure thing, which is more likely to minimize financial loss.  The gamble here, Option D, has an expected value of negative $800, versus the sure thing’s value of negative $750. Taking the risk here is the worse choice, rationally speaking.

But again, if you are like most people, you chose Option D, and again, you didn't even have to think about it. You had an instant and very negative reaction to the idea of losing $750. The thought actually caused a number of physical stress reactions that you probably did not notice: your pupils dilated, your heart rate increased, the hair on your arms rose slightly, and your sweat glands were activated. The gamble offered you the possibility, even if small, of owing nothing, and thinking about this possibility reduced your stress level. You decided to take your chances on the gamble, even if you realized that it was the worse option, rationally speaking.  

Put these two mental exercises together, and you see what Kahneman and Tversky found: when people stand to gain, they prefer to avoid risk; but when they stand to lose, they prefer to take risks.

If you go back to the examples that I gave in my first post, you can see that this risk-avoiding and risk-seeking behavior helps explain why Adam and Company C fail to take advantage of reasonable settlement opportunities. They have no option that leads to gain – or a sufficiently substantial gain in Adam’s case – and they choose to take the risk instead by taking their cases to trial. They do this even though they recognize that the odds of winning at trial are small, and the likelihood is that they will do worse than in a settlement.

This also helps why Denise takes a relatively low settlement offer when she likely could do better. Being in a position of likely gain, she seeks the certainty of the settlement, rather than risk everything at trial.  

Next week: Reference Points.  

Mediation and the Science of Decision Making Part II: People Act Rationally, Don’t They?

You may not have heard of Daniel Kahneman, but his work has had a tremendous impact on the way that we understand human decision making. Kahneman won the 2002 Nobel Prize in Economic Sciences for his work on this topic with Amos Tversky. (Tversky passed away in 1996, and they do not award the Nobel posthumously.) In his 2011 book, Thinking, Fast and Slow, Kahneman explains many of the factors that lead people to make the decisions that they do.

People act rationally, don’t they?

One of the pillars of modern economic theory is the idea that individuals act rationally: they seek to maximize their satisfaction based on their preferences and the information available to them, in a way that changes little over time or in different contexts. 

The “rational choice theory” holds that a person faced with a decision regarding money will choose the option that maximizes financial gain and minimizes losses. For example, given the choice between a gift of $50 and a gamble that offers a 75% chance of winning $100, a rational person should choose the gamble. The value or “expected utility” of the gamble equals the amount of the potential gain multiplied by its likelihood: in this case, the expected utility of the gamble is 75% of $100, or $75. Because the value of the gamble ($75) exceeds the value of the sure thing ($50), rational choice theory predicts that a rational person will choose the gamble. 

In the settlement context, the rational choice theory predicts that parties will evaluate their cases the same way, based on the likelihood of the anticipated outcomes. Remember Adam from last week? Although he had a weak case, he rejected a reasonable settlement offer, took his case to trial, and -- predictably -- lost. 

Adam should have evaluated his case by looking at the judgment that he was most likely to recover multiplied by the likelihood of achieving that judgment. So if he believed that he had a 10-20% chance of winning a $300,000 judgment, he should have evaluated the case as having a value of $30,000 to $60,000. If the defendant offered him a figure in this range, the rational choice theory predicts that he would accept it. 

But human beings in the real world do not make such rational decisions. Kahneman and Tversky revolutionized economics by showing that a number of behavioral patterns and cognitive errors play an important role in the way that people make decisions. Understanding these cognitive errors will improve your decision-making and your results in litigation and mediation. 

Next week, we will start looking at these cognitive errors.  

Mediation and the Science of Decision Making Part I: Introduction

Very frequently in mediation, as in life generally, we observe people making decisions that that seem irrational, overly emotional, too timid, or unnecessarily risky. Why? Why would intelligent, thoughtful, successful people take courses of action that seem to make no sense?

As it turns out, the field of behavioral economics holds a number of extremely interesting insights into these questions, as well practical lessons that we can use in mediation to help people make better decisions.

Over the next several weeks, I will post a series of short articles addressing these issues.

I'll start with some examples. As you read through them, consider whether you think these people are making the right decisions.

Adam files an action for sexual orientation discrimination and retaliation. Both sides agree that he likely has enough evidence to survive summary judgment, but serious problems with his credibility make success at trial extremely unlikely. If he can succeed, he has a possibility of recovering more than $300,000 in damages. Adam's attorneys tell the mediator in private caucus that Adam is a difficult client, and they do not want to take the case to trial. The defendant offers Adam a reasonable amount that reflects both the potential exposure and the small likelihood of such a result. Adam rejects the offer against the mediator's and his attorneys' advice, takes the case to trial, and loses.

Beth works as an executive for Company C, a mid-sized and growing company. After her termination, she files suit alleging quid pro quo sexual harassment, retaliation, and defamation. Beth has not been able to find a new job, and she has evidence that her former employer’s CEO has defamed her to potential employers. Beth has good evidence, including smoking gun emails that the company attempted to destroy. Three years after her termination, her economic damages are in the mid six figures, her emotional distress is well documented and credible, and the company and its CEO have strong financials, making substantial punitive damages a possibility. Beth's final demand at mediation is at the high end of the reasonable settlement range, with an indication of negotiation flexibility. The defendants realize that they face serious risks at trial, but they decide to take their chances, against their attorneys' advice. The jury brings back a seven-figure verdict, including punitive damages against both defendants, and the Court of Appeal affirms.

Denise is a highly-paid salesperson. She does not make policy decisions or supervise other employees. She is paid a salary plus quarterly bonuses (not commissions), and she works 15 to 20 hours of overtime per week. After she leaves the company, she brings a wage and hour claim for unpaid overtime compensation. Her attorneys calculate her unpaid wages at over $200,000. After Denise wins summary adjudication of the defendant’s exemption defenses, the parties engage in mediation. Late in the day, the defendant makes its “last, best, and final” offer of $50,000. Although Denise's attorneys feel very strongly that she will succeed at trial, and her claim (including penalties, interest, and attorney fees) now exceeds $400,000, Denise feels that the best choice is to take the sure thing and she accepts the offer.

Why would Adam and Company C reject reasonable opportunities to settle and instead take bad bets at trial? Conversely, why would Denise take a relatively low settlement figure, rather than pursuing a strong case at trial? 

Next week, we will start looking at the answers to these questions.  

Steven G. Pearl Joins ADR Services, Inc., as Mediator

Steven G. Pearl has joined the award-winning panel of neutrals at ADR Services, Inc. He is available immediately to resolve matters throughout California.

Mr. Pearl began his litigation career practicing as a defense attorney at Orrick Herrington before founding his own firm in 1994. As Principal of The Pearl Law Firm, Mr. Pearl has litigated hundreds of individual and class action wage & hour, employment, unfair competition, and consumer protection suits on behalf of plaintiffs. His prolific writings on negotiation, employment, wage & hour, and class action law, along with his successful litigation practice, earned him a place on the 2011 “Super Lawyers” list.

Mr. Pearl began mediating in 2007. He received his mediator training at the Straus Institute for Dispute Resolution at Pepperdine Law School. He quickly earned a reputation as a tenacious, creative settlement professional, and his scholarly legal knowledge and compassion and respect for both parties at mediation have allowed him to broker settlements in over 100 disputed matters.

“Steve Pearl has the drive, knowledge, and skill set our firm seeks when recruiting new panelists,” explained Lucie Barron, President and founder of ADR Services, Inc. “He is a great addition to our group of employment mediators, and we strongly believe that putting our firm behind his excellent reputation and substantial litigation and mediation experience will further build his ADR practice.”

"I am honored to be selected to join ADR Services," Mr. Pearl said. "I have always appreciated the professionalism and competency of ADR's leadership, staff, and neutrals. I look forward to working with them and continuing to provide my clients with the highest level of mediation services."

Facebook and The Winklevoss Twins: Ninth Circuit Upholds Contractual Mediation Confidentiality

Over the last several years, California courts have clarified and strengthened mediation confidentiality. In Cassel v. Superior Court (2011) 51 Cal.4th 113, Radford v. Shehorn (2010) 187 Cal.App.4th 852, and Simmons v. Ghaderi (2008) 44 Cal.4th 570, California courts have construed mediation confidentiality very broadly, refusing to allow statements made in the course of mediation into evidence.

But mediation confidentiality in California courts is based on the California Evidence Code. Do the same rules apply in federal court?

The Ninth Circuit addressed this issue in The Facebook, Inc. v. Pacific Northwest Software, Inc., --- F.3d ----, 2011 WL 1346951 (9th Cir. 4/11/11). The case concerns allegations that Mark Zuckerberg stole the idea for facebook.com from Cameron and Tyler Winklevoss, as depicted in the film, The Social Network.

The district court ordered the parties to mediation. Before mediation, they signed a confidentiality agreement, stipulating that all statements made during mediation were privileged, non-discoverable and inadmissible "in any arbitral, judicial, or other proceeding." At mediation, the parties signed a "Term Sheet & Settlement Agreement."

The Winklevosses agreed to give up ConnectU in exchange for cash and a piece of Facebook. The parties stipulated that the Settlement Agreement was “confidential,” “binding” and “may be submitted into evidence to enforce [it].” The Settlement Agreement also purported to end all disputes between the parties.

The settlement fell apart during negotiations over the form of the final deal documents, and Facebook filed a motion with the district court seeking to enforce it. ConnectU argued that the Settlement Agreement was unenforceable because it lacked material terms and had been procured by fraud. The district court found the Settlement Agreement enforceable and ordered the Winklevosses to transfer all ConnectU shares to Facebook. This had the effect of moving ConnectU from the Winklevosses' to Facebook's side of the case.

Slip op. at 1.

The Court held that the settlement agreement was enforceable. First, it contained all necessary material terms, even though it contemplated additional documents to complete the deal.

California allows parties to delegate choices over terms, so long as the delegation is constrained by the rest of the contract and subject to the implied covenant of good faith and fair dealing. Delegation isn't necessary for a contract like the Settlement Agreement to be enforceable, as the court may fill in missing terms by reference to the rest of the contract, extrinsic evidence and industry practice. But the clause quoted above leaves no doubt that the Winklevosses and Facebook meant to bind themselves and each other, even though everyone understood that some material aspects of the deal would be papered later.

Slip op. at 2.

The Court then reached the confidentiality issue.

After signing the Settlement Agreement, Facebook notified the Winklevosses that an internal valuation prepared to comply with Section 409A of the tax code put the value of its common stock at $8.88 per share. The Winklevosses argue that Facebook misled them into believing its shares were worth four times as much. Had they known about the $8.88 valuation during the mediation, they claim, they would never have signed the Settlement Agreement. The Winklevosses charge Facebook with violating Rule 10b–5, and they seek rescission of the Settlement Agreement under Section 29(b) of the Securities Exchange Act of 1934 (the Exchange Act).

Slip op. at 3. The Court held that contractual mediation confidentiality -- not any common law privilege, Federal Rule of Evidence, or Local Rule -- prohibited introduction of evidence of anything said during the mediation.

The district court excluded this evidence under its Alternative Dispute Resolution (ADR) Local Rule 6–11, which it read to create a “privilege” for “evidence regarding the details of the parties' negotiations in their mediation.” But privileges are created by federal common law. See Fed.R.Evid. 501. It's doubtful that a district court can augment the list of privileges by local rule. In any event, the parties used a private mediator rather than a court-appointed one. Their mediation was thus “not subject to the ... ADR Local Rules,” including Local Rule 6–11.

Slip op. at 5.

Nevertheless, the district court was right to exclude the proffered evidence. The Confidentiality Agreement, which everyone signed before commencing the mediation, provides that:

All statements made during the course of the mediation or in mediator follow-up thereafter at any time prior to complete settlement of this matter are privileged settlement discussions ... and are non-discoverable and inadmissible for any purpose including in any legal proceeding.... No aspect of the mediation shall be relied upon or introduced as evidence in any arbitral, judicial, or other proceeding.

This agreement precludes the Winklevosses from introducing in support of their securities claims any evidence of what Facebook said, or did not say, during the mediation. The Winklevosses can't show that Facebook misled them about the value of its shares or that disclosure of the tax valuation would have significantly altered the mix of information available to them during settlement negotiations. Without such evidence, their securities claims must fail.

Slip op. at 5.

The moral of the story for us -- mediators and parties going to mediation -- is to be sure that everyone signs a mediation confidentiality agreement before the mediation starts. And if you reach a settlement agreement at mediation, make sure you get the signatures on the dotted line before anyone leaves.

The opinion is available here.

Cassel v. Superior Court: Cal. Supremes Uphold Mediation Confidentiality

In Cassel v. Superior Court (Wasserman, Comden, Casselman & Pearson, L.L.P.) (1/13/11) 51 Cal.4th 113, the California Supreme Court continued a trend of enforcing strictly California's policy in favor of mediation confidentiality, holding that communications between attorney and client during the course of mediation cannot come into evidence in the client's subsequent malpractice action against the attorney and that all discussions conducted in preparation for a mediation as well as all mediation-related communications that take place during the mediation itself are protected from disclosure, even if these do not occur in the presence of the mediator or other disputants. The Court summarized the case as follows:

Petitioner Michael Cassel agreed in mediation to the settlement of business litigation to which he was a party. He then sued his attorneys for malpractice, breach of fiduciary duty, fraud, and breach of contract. His complaint alleged that by bad advice, deception, and coercion, the attorneys, who had a conflict of interest, induced him to settle for a lower amount than he had told them he would accept, and for less than the case was worth.   

Prior to trial, defendant attorneys moved, under the statutes governing mediation confidentiality, to exclude all evidence of private attorney-client discussions immediately preceding, and during, the mediation concerning mediation settlement strategies and defendants' efforts to persuade petitioner to reach a settlement in the mediation. The trial court [L.A. Superior, Judge McLaughlin] granted the motion, but the Court of Appeal vacated the trial court's order. 

Slip op. at 1. 
The Court began its analysis by stating the purposes of mediation confidentiality: 

As noted above, the purpose of these provisions is to encourage the mediation of disputes by eliminating a concern that things said or written in connection with such a proceeding will later be used against a participant. “Toward that end, ‘the statutory scheme ... unqualifiedly bars disclosure of communications made during mediation absent an express statutory exception.’ “ Judicial construction, and judicially crafted exceptions, are permitted only where due process is implicated, or where literal construction would produce absurd results, thus clearly violating the Legislature's presumed intent. Otherwise, the mediation confidentiality statutes must be applied in strict accordance with their plain terms. Where competing policy concerns are present, it is for the Legislature to resolve them. 

Slip op. at 6 (citations omitted). The Court then reviewed a number of recent decisions upholding mediation confidentiality: Foxgate Homeowners' Assn. v. Bramalea California, Inc. (2001) 26 Cal.4th 1 (barring evidence that party refused to mediate in good faith); Rojas v. Superior Court (2004) 33 Cal.4th 407 (mediation confidentiality applies broadly to all "writings" "prepared for the purpose of, in the course of, or pursuant to, a mediation"); Fair v. Bakhtiari (2006) 40 Cal.4th 189 (written settlement agreement reached in mediation can be disclosed only if it includes, on its face, "a statement that it is ‘enforceable’ or ‘binding,’ or a declaration in other terms with the same meaning"); Simmons v. Ghaderi (2008) 44 Cal.4th 570 (judicial doctrines of equitable estoppel and implied waiver are not valid exceptions to the strict technical requirements set forth in the mediation confidentiality statutes for the disclosure and admissibility of oral settlement agreements reached in mediation). 

Here, as in Foxgate, Rojas, Fair, and Simmons, the plain language of the mediation confidentiality statutes controls our result. Section 1119, subdivision (a) clearly provides that “[n]o evidence of anything said or any admission made for the purpose of, in the course of, or pursuant to, a mediation ... is admissible or subject to discovery....” As we noted in Simmons, section 1119, adopted in 1997, “is more expansive than its predecessor, former section 1152.5. Section 1119, subdivision (a), extends to oral communications made for the purpose of or pursuant to a mediation, not just to oral communications made in the course of the mediation. [Citation.]”   

The obvious purpose of the expanded language is to ensure that the statutory protection extends beyond discussions carried out directly between the opposing parties to the dispute, or with the mediator, during the mediation proceedings themselves. All oral or written communications are covered, if they are made “for the purpose of” or “pursuant to” a mediation. (§ 1119, subds. (a), (b) .) It follows that, absent an express statutory exception, all discussions conducted in preparation for a mediation, as well as all mediation-related communications that take place during the mediation itself, are protected from disclosure. Plainly, such communications include those between a mediation disputant and his or her own counsel, even if these do not occur in the presence of the mediator or other disputants. 

Slip op. at 8-9. 
Further, the Court held that while a represented party may waive the privilege for communications between that party and counsel, "the mediation confidentiality statutes do not create a “privilege” in favor of any particular person." Slip op. at 12. Rather, "all mediation participants involved in a mediation-related communication must agree to its disclosure." Ibid. As a result, the attorney defendants in this case could object to introduction of communications with their client that took place both preceding and during the mediation. 

The opinion is available here. 

Comment from a Satisfied Mediation Client

We recently used Steve Pearl’s mediation services on a very contentious case with excellent results. This was a certified class action with 2,000 class members just two months away from trial.

Steve was able to bring the mediation together and keep both sides motivated and coming back to the table, despite our firm belief that the case would never settle. Steve dealt adroitly with hostile attitudes from some of the participants and corralled the parties into realistic negotiations.

The case would not have settled but for Steve’s follow up and persistence. Steve insisted that the parties come back for a second day of mediation -- even though neither side thought it would resolve the case -- and continued to assist the parties in hammering out the key terms to a settlement that was reached after two full, grueling days of mediation, plus several days of phone calls and emails.

(And to the delight of everyone, we were provided home-baked goodies each day).

I highly recommend Steve as a mediator for any type of case, especially those that require a tough backbone. He can handle the job.

Rene L. Barge
Class Action Litigation Group
Huntington Beach, CA

Cal. Supreme Court Schedules Oral Argument for Mediation Confidentiality Dispute

The California Supreme Court will hear oral argument on Cassel v. Superior Court (Wasserman Comden Casselman & Pearson) on November 2, 2010, at 9:00 a.m., in San Francisco.

See our original post on Cassel here.

Court of Appeal Requires Personal Consent for Enforceable Settlement Agreement

Critzer v. Enos (8/30/10) --- Cal.App.4th ---, emphasizes the need for the consent of all parties -- either in writing or orally before the court -- in order to obtain an enforceable settlement agreement.

In a multi-party action, the parties recited the terms of a settlement agreement in open court, and three of the five parties gave their consent on the record. The parties could not agree on the language of a formal settlement agreement, and one brought a motion to enforce settlement under California Code of Civil Procedure section 664.6, which provides:

If parties to pending litigation stipulate, in a writing signed by the parties outside the presence of the court or orally before the court, for settlement of the case, or part thereof, the court, upon motion, may enter judgment pursuant to the terms of the settlement. If requested by the parties, the court may retain jurisdiction over the parties to enforce the settlement until performance in full of the terms of the settlement.

The trial court granted the motion. The Court of Appeal, noting that the law requires the personal assent of the parties themselves, and not just their counsel, reversed:

We conclude further that because there was neither an oral settlement all parties personally agreed upon, nor a written settlement agreement signed by all of the parties, the court lacked authority under the summary procedure of section 664.6 to enforce any settlement.

Slip op. at 2.

The full text of the opinion is here.