Exhibit 8-1 contains 13 problems. Take 30 minutes right now to try to solve these problems. When in doubt, make your best guess. As you go along, make a mental note of your thoughts about each problem as you try to solve it. Read the rest of this chapter before you look up the answers in Exhibit 8-9 (at the end of this chapter). As you read the chapter, see whether any insights come to you, and make note of them as they arise.

The most common mental model of negotiation is the haggling model, in which each negotiator tries to obtain the biggest share of the bargaining zone. The haggling model is based upon a fixed-pie perception of negotiation. For example, before setting out for Dungeness crab season each December, fishermen and processors in Newport, Oregon engage in a time-honored tradition of negotiating an opening price for the crabs, known as “parley” in maritime terminology. In another example, a research investigation of haggling revealed that women were initially quoted worse prices than were men in their attempts to get a deal on a car radiator, but when women bothered to haggle, they were successful more often than men who haggled. Specifically, in the investigation, male and female volunteers called auto repair shops asking about the cost to replace a radiator on a 2003 Toyota Camry. With a going accepted rate of $365, the callers stated what they think the radiator should cost—anywhere from $365 to $510 to stating no idea of the cost. When callers mentioned rates for the job in excess $365, they ended up with a quote of $60 more than that, on average. When the men and women expressed no idea of the cost, men on average received a $383 quote but the women were gouged with a quote of $406. But, when callers engaged in “haggling”—that is, asking the shops to lower the quoted price, 35% of women were successful, but only 25% of the men were successful.⁶

Some negotiators think of negotiation as a rational, decision-making model in which they compute a cost-benefit analysis and attempt to maximize their returns. For example, Robert Rubin, former U.S. Treasury secretary, calculates the odds of almost every decision he faces, using both real and mental yellow pads. He once suggested to the board of the American Ballet Theatre, on which he sat, that it enact a cost reduction by cutting 10% of the swans in Swan Lake. ⁷

The chess game model of negotiation elevates negotiation from “fighting in the streets” to a battle of wits between two or more highly intelligent people. In game playing, each person has his or her own interests in mind. For example, California Tortilla’s customers were invited to challenge cashiers to a game of Rock, Paper, and Scissors to get a dollar taken off their bill.⁸ The World RPS (Rock, Paper, Scissors) Society posted the following advice to size up the counterparty: A seemingly intellectually superior counterparty (thus most likely to play paper) can be beaten by playing scissors. Rock being the most aggressive throw, the knuckle-dragging cashier type can be beaten by playing paper, which symbolizes “the victory of modern culture over barbarism.” A contained, clever user of tools can be beaten by playing rock.

A quite different mental model of negotiation is what we call the partnership model, embraced by companies and salespeople who treat their clients as partners. Negotiators who ascribe to the partnership model believe it is important to build rapport to nurture a long-term relationship and in many cases, to make sacrifices in the name of creating long-term goodwill. For example in one model, negotiation is likened to the Brazilian dance of capoeira in which movements and rhythm is interpreted as both fighting and as dancing (collaborating). The belief is that the dance metaphor promotes a more constructive and interests-based approach.⁹

In problem solving, negotiators sit on the same side of the table and attempt to solve a puzzle together. This model focuses on the collaborative or cooperative aspects of the task and involves a great deal of creativity, reframing, and out-of-the-box thinking.

We investigated negotiators’ mental models and how they affect performance.¹⁰ Compared to negotiators who fail to reach win-win outcomes, negotiators who reach win-win outcomes are more accurate in their understanding of the other party’s underlying interests. Negotiators who reach win-win outcomes have mental models that are more similar to one another than do negotiators who fail to reach win-win outcomes. Shared metacognition promotes more cooperative negotiation, greater insight into the opponent’s values and interests, and greater satisfaction with the negotiation outcome.¹¹ To achieve such shared cognition, negotiators should explicitly exchange their understanding with one another. Experience-based training allowed negotiators to develop mental models that resemble expert, win-win models; in contrast, didactic lecturing is uniquely ineffective.

Most negotiation situations appear to contain a single issue. Fractionating negotiation issues into solvable parts and creating several issues from what appear to be single-issue negotiations is probably the most important aspect of creative negotiation.¹³ For example, consider the negotiations between Southwest Airlines and AirTran. The most obvious issue—the Southwest and AirTran merger—took precedent. However, the parties also identified seemingly smaller issues as well. For example, because AirTran featured business class seating and Southwest did not, AirTran raised the seating issue and argued that loyal customers of AirTran might miss their seats.¹⁴

By exchanging offers, negotiators gain insight into possibilities. The timing of offer exchange and the insight it provides is important. During the early part of negotiation, search is exploratory and negotiators are influenced by the value of the offers.¹⁵ However, later search is more focused on refinement and the content of offers. Search by value is more difficult than search by content and negotiators seek value by communicating about content. Similarly, when negotiators make a first offer, it is advisable to do so later rather than earlier in the negotiation. Indeed, negotiators who make first offers later in the negotiation are more likely to discover creative agreements that meet parties’ underlying interests than making early first offers.¹⁶

Consider, for example, the negotiation between architect Daniel Libeskind and developer Larry Silverstein concerning the rebuilding of the World Trade Center site.¹⁷ By identifying additional issues and realigning issues, a negotiated agreement was reached, ending months of heated battles between Silverstein and Libeskind. The impasse centered on how much influence the architect (Libeskind) would have on the design of the first office building to be built at the site—the 1,776-foot tower that defines the rebuilt Trade Center’s presence on the Lower Manhattan skyline. The developer (Silverstein) wanted to involve other architects for the largest tower and make changes to the master plan. An agreement was reached by carving out another issue—namely, the development of commercial design guidelines governing future commercial development on the site. Libeskind won this issue, but Silverstein got to hire another firm to serve as design architect and project manager for the Freedom Tower, the first commercial building constructed on the site.¹⁸

Expanding the pie is an important method by which to create integrative agreements. However, negotiators who labor under the fixed-pie perception may unnecessarily limit their options. Consider for example, how a negotiation between a fire station and an elementary school was transformed from an impasse to a creative arrangement by expanding the pie.¹⁹ Initially, the school wanted to acquire land to expand, but the fire station blocked that move. So the pie was expanded such that each party paid an unrelated third party to acquire a separate parcel of land. In the final deal, the city gave a 3-acre parcel of land to a private owner, who then gave back to the city 7.5 of his own acres (located nearer to the school and fire station). At this point, the fire station and the school successfully expanded the pie of available land to accomplish both of their goals. Thus, by expanding the pie to include another plot of land that could be “swapped,” both parties ultimately achieved what they wanted: The school was able to expand, and the firehouse could make its entrances and exits more convenient and safe.

Oftentimes, it is not possible for negotiators to find a compromise solution, and expanding the pie does not work. Furthermore, perhaps neither party can get what it wants in a trade-off. A bridging solution creates a new alternative that meets parties’ underlying interests. Bridging alerts us to yet another reason to understand the other party’s interests and avoid positional bargaining. If negotiators understand the basic needs of the counterparty, they are more likely to fashion bridging agreements. AMC Networks and DISH Network ended a contentious dispute brought on when the satellite provider blacked out several AMC channels, leaving DISH viewers without access to several popular shows. The standoff ended when a bridging solution was created. AMC agreed to drop a lawsuit stemming from DISH walking away from a distribution agreement with a defunct high-definition channel called Voom. DISH agreed to pay $700 million to AMC, but in turn was granted valuable satellite-grade radio spectrum. The Voom negotiations were not part of the original dispute, but paved the way to an agreement that put AMC back on the DISH network.²⁰

Sometimes, people are reluctant to negotiate because reaching a resolution seems costly to them. Most people are risk seeking when it comes to loss, meaning that they are reluctant to make concessions and may behave irrationally when they believe they will have to make concessions. Cost cutting is a way of making the other party feel whole by reducing that party’s costs. An example of value-added cost cutting occurred in negotiations between The Nature Conservancy and the Great Northern Paper Company.²¹ On the surface, we might expect the interests of these two to be a classic fixed pie—The Nature Conservancy, dedicated to preserving the environment, should hardly want to work with a company that makes its money by cutting down trees for consumer and industrial use. In an unprecedented partnership, The Nature Conservancy assumed $50 million worth of Great Northern’s debts. In return, Great Northern Paper agreed to protect a quarter-million acres from development.²²

Another example of cost cutting occurred in a heated dispute involving Walgreens and CVS drugstores. In 2010, Walgreens threatened to stop filing or renewing prescriptions from competitor CVS Caremark’s massive network because of perceived unpredictable reimbursement rates and prescription plans that favored CVS stores. CVS retaliated by announcing it would pull its network from Walgreens stores within a month. The threats would have blocked millions from filing prescriptions at Walgreens stores. However, a turning point occurred when each company looked at the potential costs. Walgreens’ patients accounted for about 10% of CVS’ Pharmacy Benefit Manager business, and CVS customers accounted for nearly $4.4 billion of Walgreens’ annual sales. The two companies agreed to cut costs via an agreement in which CVS provided financial concessions in return for Walgreens not extracting higher payment for the drugs that CVS Caremark enrollees buy.²³

In a nonspecific compensation agreement, one negotiator receives what he or she wants, and the other is compensated (or paid) by some method that was initially outside the bounds of the negotiation. For example, Phil Jones, managing director of Real Time Consulting, a London-based interactive design consultancy, recalls an instance where he used nonspecific compensation in his negotiations.²⁴ The problem was that his client, a Formula 1 motor-racing team, wanted to launch websites but did not have the budget to pay him. However, in Phil Jones’s eyes, the client was high profile and had creative, challenging projects with which Real Time wanted to get involved. Formula 1 came up with a nonspecific compensation offer to make the deal go through: tickets to some of the major Formula 1 meetings. It worked. Says Phil Jones, “The tickets are like gold dust…and can be used as a pat on the back for staff or as an opportunity to pamper existing clients or woo new ones.”

A major obstacle to reaching negotiated agreements often concerns negotiators’ beliefs about some future event or outcome.²⁵ Impasses often result from conflicting beliefs that are difficult to surmount, especially when each side is confident about the accuracy of his or her prediction and consequently, suspicious of the other side’s forecasts. Often, compromise is not a viable solution, and each party may be reluctant to change his or her point of view. Contingency contracts can provide a way out of the mire. With a contingency (or contingent) contract, differences of opinion among negotiators concerning future events do not have to be bridged; instead, they become the core of the agreement.²⁶ Negotiators can bet on the future rather than argue about it. In some areas of business, contingency contracts are commonplace. For example, some CEOs agree to tie their salary to a company’s stock price.

However, in many business negotiations, contingency contracts are either ignored or rejected for several reasons.²⁷ First, people are unaware of how to construct contingency contracts. Second, contingency contracts are often seen as a form of gambling. Third, no systematic way of thinking about the formulation of such contracts is usually available, meaning that they appear to be a good idea, but how to formalize and act upon them remains an enigma. Fourth, many negotiators have a “getting to yes” bias, meaning they focus on reaching common ground with the other party and are reluctant to accept differences of interest, even when this might create viable options for joint gain.²⁸ The paradoxical view suggested by the contingency contract strategy states that differences are often constructive. With a contingency contract, negotiators can focus on their real mutual interests, not on their speculative disagreements.²⁹ When companies fail to find their way out of differences in beliefs, they often go to court, creating expensive delays, litigation costs, loss of control by both parties, and deteriorating BATNAs. Consider how a contingent contract might have changed the course of one of the most sensitive negotiation issues involving consumers and mobile apps. In the late 1990s, privacy groups such as The Center for Digital Democracy and the Electronic Privacy Information Center, complained that websites often collected sensitive information without their users’ permission or knowledge. In 2000, the Federal Trade Commission began considering the privacy implications raised by consumers’ growing use of mobile devices. For example, a study by the Federal Trade Commission found that only 16% of apps aimed at children disclosed their privacy practice.³⁰ The battle between citizens and app developers became particularly heated and emotional. Investigative reporters in the mid-1990s uncovered a long list of dangerous marketing abuses that negatively leveraged users’ personal information to gain power.³¹

A contingent contract was introduced in 2013 when the National Telecommunications and Information Administration (NTIA), an agency within the Commerce Department, convened a discussion in an effort to implement the White House’s “Privacy Bill of Rights,” which outlines a set of principles for how Internet companies should handle personal information. In particular, John Potter, president of the Application Developers Alliance, worked to craft a code of conduct for companies to inform consumers what sensitive and personal information the apps collect and also if that information is shared. The contingent contract involves testing “notices,” which disclose whether the app is collecting information from a user, including location, browsing history, health records, contacts, financial information, and phone and text logs. Under the contingent contract, privacy groups and consumers test the voluntary code of conduct. The notice program allows apps to compete on privacy and thus gives consumers a tool to pick the most privacy friendly applications.³²

Another advantage of contingency contracts is they provide a nearly perfect lie-detection device. Contingency contracts allow negotiators to test the counterparty’s veracity in a nonconfrontational manner, thereby allowing parties to save face. Contingency contracts also allow parties who are concerned about being cheated to safeguard themselves. This fear of being cheated is precisely what Christopher Columbus was worried about when he negotiated an agreement about the New World with Queen Isabella and King Ferdinand of Spain. Worried he would risk life and opportunity and gain nothing, Christopher Columbus insisted he be offered an opportunity to contribute one-eighth of the costs of future expeditions and be guaranteed one-eighth of all profits. Unfortunately, the crowns reneged on the deal upon his return, and Columbus had to go to court.³³

By the same token, contingency contracts build trust and good faith between negotiators because incentives can be provided for each company to deliver exceptional performance. Therefore, contingency contracts provide a safety net, limiting each company’s losses should an agreement go awry unexpectedly. (For a summary of the benefits of contingency contracts, see Exhibit 8-2.)

Although contingency contracts can be valuable in many kinds of business negotiations, they are not always the right strategy to use. There are three key criteria for assessing the viability and usefulness of contingency contracts in negotiation:³⁴

1. Contingency contracts require some degree of continued interaction between the parties. Because the final terms of the contract will not be determined until sometime after the initial agreement is signed, some amount of future interaction between parties is necessary, thereby allowing them to assess the terms of their agreement. Therefore, if the future seems highly uncertain, or if one of the parties is suspected of preparing to leave the situation permanently, contingency contracts may not be wise.

2. Parties need to think about the enforceability of the contingency contract. Under a contingent contract, one or more of the parties will probably not be correct about the outcome because the contract often functions as a bet. This outcome creates a problem for the “loser” of the bet, who may be reluctant to reimburse the other party when things do not go his or her way. For this reason, the money in question might well be placed in escrow, thereby removing each party’s temptation to defect.

3. Contingency contracts require a high degree of clarity and measurability. If an event is ambiguous, nonmeasurable, or of a subjective nature, overconfidence, egocentric bias, and a variety of other self-serving biases can make the objective appraisal of a contingency contract a matter of some opinion. Parties should agree up front on clear, specific measures concerning how the contract will be evaluated. For this reason, it is often wise to consult a third party.

People’s ability to solve problems in new contexts depends on the accessibility of their relevant knowledge. If a manager is confronted with new business challenges, he or she often consults his or her knowledge base for previous problems in an attempt to see which previous problem solving strategies might be useful in solving the new problem. The inert knowledge problem is the inability to access relevant knowledge when we most need it.³⁵ The information necessary to solve a new problem is part of a manager’s cognitive repertoire but is not accessible at the right time. This unavailability is not due to senility or amnesia but rather, to the peculiar way that our long-term memories are constructed.

A striking dissociation occurs between what is most accessible in our memories and what is most useful in human problem solving and reasoning. People often fail to recall what is ultimately most valuable for solving new problems.³⁶ In one investigation, people studied examples containing principles of probability theory and then attempted to solve problems requiring the use of those principles.³⁷ If the study and test stories were from the same context, people were more likely to be reminded of them than if the stories were from different contexts. In another investigation, participants were given a story to read about a hawk giving feathers to a hunter.³⁸ Participants were then given one of four stories resulting from combining surface and structural similarities (i.e., a story with similar characters and plot, different characters but same plot, similar characters but different plot, or different characters and different plot). People were more than four times more likely to recall the original story when later shown a story with similar characters than when shown a story with different characters, suggesting that people often fail to recall what is ultimately most valuable for solving new problems.³⁹ Upon being informed of the correct approach to a negotiation, management students often express regret: “I knew that. I just did not think to use it.”

Transfer is the ability to apply a strategy or idea learned in one situation to solve a problem in a different, but relevant situation. Surface-level transfer occurs when a person attempts to transfer a solution from one context to a superficially similar context. However, in most situations it is desirable for people to engage in deep-level transfer by applying solutions and strategies that have meaningful similarities, rather than superficial ones. Unfortunately, this task proves to be quite difficult for most managers to do. In general, if two problems have similar surface (or superficial) features, managers are more likely to transfer knowledge from one problem situation to the other. Ideally, managers must transfer solutions to problems with similar deep (or structural) features that possess significantly different superficial features.

As a case in point, consider the “tumor problem” presented in Exhibit 8-3. When presented with this problem, few people successfully solve it; however, if it is preceded by the fortress problem in Exhibit 8-4, the solution rate rises dramatically.⁴⁰ Even though a similar solution can be applied in both problems, because the surface information in each problem is quite different (one deals with a medical situation; the other a political situation), people are often unable to access their knowledge about one of these problems to help them solve the other.

The same problem occurs in negotiation. Studies of MBA students, executives, and consultants acquiring negotiation skills reveal a dramatic inert knowledge problem.⁴¹ Transfer rates are quite low when a key principle needs to be applied to novel negotiation situations that involve different surface features. For example, when people are challenged with a negotiation situation involving a theater company that contains the potential for a contingency contract, they are often unable to employ the principle of contingency contracts even when they have received extensive training on this principle in a negotiation case involving a different context, such as a family-owned farm. This occurs because we tend to use our previous knowledge only when it seems similar to a new problem. People are not able to recognize problems that may benefit from similar problem solving principles and strategies.

The obvious question is: What decreases the inert knowledge problem and increases people’s ability to transfer knowledge they possess when faced with a situation that could potentially benefit from that knowledge? One answer appears to be quite simple and powerful. It involves making an explicit comparison between two or more relevant cases.⁴² To the extent that people mentally compare cases or situations, they are able to create a problem-solving schema that is uncluttered by irrelevant surface information. Thus, problem-solving schemas created through this process of mental comparison are more portable and more likely to be called upon when negotiators are challenged with a novel problem. In the absence of comparison, it is not clear to negotiators which information about a situation is relevant or irrelevant. Furthermore, as helpful as making comparisons can be, recognizing when to make them is not always obvious. For example, in our training of MBA students and executives, we frequently present negotiators with several training cases, usually on the same printed page. Very rarely do negotiators actively compare the cases printed on the same page, even though they contain a similar underlying principle. Thus, the key appears to be making comparisons among experiences, a strategy we elaborate upon later.

The preceding strategy focused on enhancing knowledge transfer by providing negotiators with very similar, specific examples. However, diverse analogical training wherein negotiators compare several different value-creating strategies, may be more effective.⁴³ Negotiators who are prompted to compare (rather than simply read) examples are not only able to profitably transfer their learning to a new negotiation situation, they are able to remember negotiation situations from their own life that illustrate the key learning principle.⁴⁴ In a related investigation, negotiators were able to engage in successful knowledge transfer by focusing on differences in seemingly unrelated tasks.⁴⁵ Specifically, negotiators who were able to think about how others made decisions in versions of the “Monty Hall” game⁴⁶ and the multiparty “Ultimatum” game⁴⁷ were more accurate in analyzing a new problem.⁴⁸

Which is more common: words that start with the letter K—for example, king—or words with K as the third letter—for example, awkward?⁴⁹ In the English language, more than twice as many words have K as the third letter than have K as the first letter. Despite this fact, the majority of people guess incorrectly (they assume that more words have K as the first letter), due to the availability heuristic. According to the availability heuristic, the more prevalent a group or category is judged to be, the easier it is for people to bring instances of this group or category to mind. This heuristic affects the quality of negotiators’ judgments in that they may be biased by the ease with which information can be brought to mind. In another investigation people were presented with a list of 39 names of well-known people.⁵⁰ Nineteen of the people on the list were female; 20 were male. The women happened to be more famous than the men. Afterward, people were asked to judge how many women’s names appeared on the list. People dramatically overestimated the number of female names, presumably because they were easier to recall—another illustration of the availability heuristic.

The availability heuristic is associated with the false consensus effect.⁵¹ This effect refers to the fact that most people think others agree with them more than is actually the case. For example, people who smoke estimate that 51% of others are smokers, but nonsmokers estimate that only 38% of people are smokers.⁵² Furthermore, people overestimate the proportion of people who agree with them about their attitudes concerning drugs, abortion, seatbelt use, politics, and even certain brand-name crackers.⁵³ When a negotiator falls victim to the availability heuristic, the likelihood of employing creative strategies (which are often less available) is severely undermined.

Imagine that you have just met your new boss. She is thin, wears glasses, is soft-spoken, and dresses conservatively. Does your supervisor enjoy poetry or sports? In answering such questions, people make judgments on the basis of a relatively simple rule: The more similar a person is to a group stereotype, the more likely he or she is to also belong to that group. Most people assume the supervisor enjoys poetry. Basically, the more a person looks like the stereotype of a group member, the more we are inclined to stereotype them as belonging to that group. The representativeness heuristic is based on stereotypes of people, which may have a basis in reality but are frequently outdated and wrong. Furthermore, reliance on stereotypical information leads people to overlook other types of information that could potentially be useful in negotiations. The most important type of information is related to base rates. Base rates are the frequency with which some event or pattern occurs in a general population. For example, consider a negotiator interested in purchasing a new car. One source of information concerning the new car is a popular consumer report. This report is based upon thousands of consumer data points and research and therefore is highly reliable. However, in addition to consulting this source, people interested in purchasing a new car often consult their neighbors and friends. Sometimes, a neighbor or friend may have had a personal experience with a car that is quite different from what is reported in the consumer report magazine. Oftentimes however, people who consult their neighbors and friends often discount perfectly valid information (i.e., the base rate information) and choose to rely upon a single, vivid data point. This error is known as the base rate fallacy.

Faulty judgments of probability are associated with what is known as the gambler’s fallacy, the tendency to treat chance events as though they have a built-in, evening-out mechanism. As an example, consider the following problem: Suppose you flip a coin and it comes up heads five times in a row. What do you think the outcome of the next toss will be? Most people feel that the probability is high that the coin will come up tails. Of course, the probability of a heads or tails outcome is always the same (50%) for each flip, regardless of the previous result. However, most people think that some sequences (such as heads, tails, heads, tails) are far more likely to occur than others (such as a string of heads or a string of tails).⁵⁴

Job candidates are often asked by recruiters to state their salary range. The job candidate, wanting to maximize his or her salary but at the same time not remove himself or herself from consideration because of unrealistic demands, faces a quandary. Similarly, the prospective home buyer struggles with what to make as an opening offer. What factors determine how we make such assessments of value?

People use a reference point as an anchor and then adjust that value up or down as deemed appropriate.⁵⁵ For example, a prospective job recruit may have a roommate who just landed a job with a salary of $80,000. The candidate decides to use $80,000 as a starting point. Two fundamental concerns arise with the anchoring-and-adjustment process. First, the anchors we use to make such judgments are often arbitrary.⁵⁶ Oftentimes, anchors are selected on the basis of their temporal proximity, not their relevance to the judgment in question. Second, we tend to make insufficient adjustments away from the anchor; we are weighed down by the anchor. (Remember how people’s estimates of the number of doctors in Manhattan were affected by their Social Security number!) The message for the negotiator is clear: Carefully select anchors, and be wary if the counterparty attempts to anchor you.

Consider the following facts:

• Women living in the San Francisco area have a higher rate of breast cancer.

• Women of lower socioeconomic status are less likely to breast-feed their babies.

• People who marry at a later point in life are less likely to divorce.

Before reading further, attempt to explain each fact. When people are asked to do so, they frequently conclude the following:

• Living in San Francisco causes breast cancer.

• People of lower socioeconomic status are not given postnatal care.

• People become wiser as they grow older.

All of these explanations are reasonable, but they are all unwarranted based upon the information given. The tendency to infer a causal relationship between two events is unwarranted because we do not know the direction of causality (e.g., it is possible that more older women live in the Bay Area). Further, a third variable could be the cause of the event (e.g., people who marry later may be richer or more educated). Maybe women of lower socioeconomic status are younger and less comfortable breast-feeding, more likely to be targeted by formula companies, or less likely to get maternity leave. There is a myriad of possible explanations.

The perseverance effect is the tendency of people to continue to believe that something is true even when it is revealed to be false or has been disproved.⁵⁷ For example, imagine that you have taken an aptitude test and have been told you scored poorly. Later, you learn the exam was misscored. Are you able to erase this experience? Not if you are like most college students, who continue to persevere in their beliefs.⁵⁸ Why is this tendency so prevalent? Once a causal explanation is constructed, it is difficult to change it. If you or your counterparty has an erroneous belief about the other, even when it is proven wrong, the belief may still prevail. The important implication is to carefully examine the beliefs you hold about the counterparty and be cognizant of faulty beliefs they may have about you.

Illusory correlation is the tendency to see invalid correlations between events. For example, people often perceive relationships between distinct pieces of information as a mere consequence of their being presented at the same time.⁵⁹ For example in one investigation, people read diagnoses of mental patients.⁶⁰ Specifically, people were shown pictures allegedly drawn by these patients and then were given the patients’ diagnoses to read. In actuality, there was no correlation between the types of pictures the patients allegedly drew and the nature of their diagnoses (paranoia, schizophrenia). Nevertheless, the people reviewing the evidence believed they saw correlations—for example, between a diagnosis of paranoia and a drawing of a very large eye. Even when people are presented with contradictory or ambiguous evidence, they are extremely reluctant to revise their judgments. As another example, suppose you learn during the course of a negotiation with a business representative from country X that 60% of country X’s male population is uneducated. Suppose the same day you learn 60% of crimes committed in that country are violent. Although no logical relation connects the two statistics, most people assume a correlation; that is, they assume that uneducated men from country X are responsible for violent crimes. In fact, no relationship exists between the two—it is illusory. Such correlations between separate facts are illusory because they lack an objective basis for the relationships. Rather, our implicit theories are constructed so that we interpret relations between temporally proximate events.

Most people believe the world is a fair place: People get out of life what they deserve and deserve what happens to them.⁶¹ This mindset leads to positive evaluations of others who have good things happen to them; for example, most people believe “good” people are likely to win lotteries. Unwarranted negative impressions are produced when others suffer misfortune; for instance, we assume that bad people or ignorant people are victims of crimes.⁶² Blaming-the-victim attributions are defensive attributions because they enable observers to deal with perceived inequities in others’ lives and maintain the belief that the world is just.⁶³ In short, if we believe bad things could easily happen to us (e.g., dying in an airplane crash or losing a limb), the world seems scary and less predictable.

The hindsight bias refers to a pervasive human tendency for people to be remarkably adept at inferring a process once the outcome is known but to be unable to predict outcomes when only the processes and precipitating events are known.⁶⁴ The hindsight bias, or the “I knew it all along” effect, makes integrative solutions to negotiation situations appear obvious when we see them in retrospect, although before they were discovered, the situation appeared to be fixed-sum.

We are frequently called upon to explain the causes of events, such as the demise of an organization or the success of a particular company. We often perceive events that have already occurred as inevitable. Stated another way, once we know the outcome of an event, we perceive the outcome to be an inevitable consequence of the factors leading to the outcome. This creeping determinism ⁶⁵ accounts for the “Monday morning quarterback” or the “I knew it all along” phenomenon. Therefore, once someone knows the outcome, the events leading up to it seem obvious. The hindsight bias also accounts for why negotiators often think integrative agreements are obvious after the fact but fail to see them when encountering a novel negotiation.

Functional fixedness occurs when a problem solver bases a strategy on familiar methods.⁶⁶ The problem with functional fixedness is that previously learned problem-solving strategies hinder the development of effective strategies in new situations. The person fixates on one strategy and cannot readily switch to another method of solving a problem. In other words, experience in one domain produces in-the-box thinking in another domain. Reliance on compromise as a negotiation strategy may produce functional fixedness.

Our past experience can limit problem solving. Consider the tumor problem presented in Exhibit 8-3. The solution rate, when people are given the problem by itself, is 37%; however, when people are shown a diagram of an arrow going through a black dot and then given the problem, the solution rate drops to 9%.⁶⁷ The diagram of the arrow going through the black dot depicted the function of the X-ray as a single line going through the human body; thus, it blocked people’s ability to think of several rays focused on the tumor. Functional fixedness occurs when people have a mental block against using an object in a new way to solve a problem. In another example, people are challenged with the problem of how to mount a candle vertically on a nearby screen to function as a lamp. The only materials they are given are a box of matches, a box of candles, and a box of tacks. The creative solution is to mount the candle on top of the matchbox by melting the wax onto the box and sticking the candle to it, then tacking the box to the screen. This elegant solution is much harder to discover when the people are presented with the boxes filled with tacks (i.e., the way the boxes are normally used), rather than emptied of their contents.⁶⁸

Closely related to the problem of functional fixedness is the set effect, in which prior experience can also have negative effects in new problem-solving situations. Also known as negative transfer, prior experience can limit a manager’s ability to develop strategies of sufficient breadth and generality. Consider the water jug problem presented in Exhibit 8-1. People who had the experience of working on all the water problems typically used a longer, costlier method to solve the problems. People without the experience of solving the problems almost always discovered the short, direct solution. Set effects also plague coalitions. Because war policies can become institutionalized over time, there is a very strong link between coalition shifts and war termination.⁶⁹ Changes in coalitions are often necessary to kick-start an updating process.

In negotiations, we are bombarded with information—the counterparty’s physical appearance, his or her opening remarks, hearsay knowledge, nonverbal behavior, and so on. However, we only perceive about 1% of all information in our visual field.⁷⁰ Thus, we perceive only a tiny fraction of what happens in the negotiation room. How do we know if we are paying attention to the right cues?

The basic function of our sensory information buffers is to parse and code stimulus information into recognizable symbols. Because external stimuli cannot get directly inside our heads, we cognitively represent stimuli as internal symbols and their interrelations as symbol structures. The sensory buffers—visual, auditory, and tactile—maintain the stimulus as an image or icon while its features are extracted. This activity occurs rapidly and below our threshold of awareness. The features extracted from a given stimulus object comprises a coded description of the object. For example, our interaction with a colleague concerning a joint venture is an event that is real, but our minds are not video cameras that record everything; rather, we use a process known as selective attention.

Consider a situation in which you are assessing the probability that a particular company will be successful. Some people might think the probability is quite good; others might think the probability is low; others might make middle-of-the-road assessments. For the decision maker, what matters most is making an assessment that is accurate. How accurate are people in judging probability? How do they make assessments of likelihood, especially when full, objective information is unavailable?

Judgments of likelihood for certain types of events are often more optimistic than is warranted. The overconfidence effect refers to unwarranted levels of confidence in people’s judgment of their abilities and the occurrence of positive events and underestimates of the likelihood of negative events. For example, in negotiations involving third-party dispute resolution, negotiators on each side believe the neutral third party will adjudicate in their favor.⁷¹ Obviously, this outcome cannot happen; the third party cannot adjudicate in favor of both parties. Similarly, in final-offer arbitration, wherein parties each submit their final bid to a third party who then makes a binding decision between the two proposals, negotiators consistently overestimate the probability that the neutral arbitrator will choose their own offer.⁷² Obviously, the probability is only 50% that a final offer will be accepted; nevertheless, typically, both parties’ estimates sum to a number greater than 100%. The message is to be aware of the overconfidence effect. When we find ourselves to be highly confident of a particular outcome occurring (whether it be the counterparty caving in to us, a senior manager supporting our decision, etc.), it is important to examine why.

Perspective taking has been a common treatment to remedy a number of faulty beliefs in negotiation, such as overconfidence and egocentric behavior. Indeed, leading people to consider other peoples’ thoughts reduces self-centered judgments such that people claim it is fair for them to take less from a common pool of resources, yet their behavior actually becomes more selfish!⁷³ Moreover, people seem completely unaware of the fact that their personal beliefs about what is fair do not align with their behavior.

Short-term memory is the part of our mind that holds the information currently in the focus of our attention and conscious processing. Unfortunately, short-term memory has severely limited capacity; only about five to nine symbols or coded items may be currently active. The “seven plus-or-minus two” rule extends to just about everything we try to remember.⁷⁴ Consider, for example, an interaction you might have with the president of a company concerning the details of a consulting engagement. The president tells you many facts about her company; you will recall, on average, five to nine pieces of information. Without deliberate rehearsal, the information in your short-term memory will disappear and be replaced with new information perceived by your sensory registers. Obviously, we perceive much more information than we ultimately store and remember.

Perhaps nothing is more effective in improving the ability of negotiators to reach win-win agreements than training programs. Most typically, training programs involve challenging negotiators to complete a negotiation that has integrative potential that may not be obvious. An important aspect of experience and training is for negotiators to reflect after completing negotiations. For example, negotiators who are prompted to ask questions about the other party’s interests are more likely to craft and reach integrative agreements than negotiators who are not prompted to ask questions, yet could have. Similarly, negotiators who reveal information about their interests and priorities are more likely to reach integrative agreements than those not prompted to reveal information.⁷⁵ And negotiators who are asked to think aloud about the other party’s interests and priorities are more effective than those who are not prompted to think about the other party.⁷⁶ In a study of how performance improves over time, negotiators who engaged in several negotiations had an advantage compared to those who had less experience.⁷⁷

If negotiators can substantially improve their outcomes by training, then perhaps it might behoove negotiators to have more training than their opponent. In other words, might negotiators perform better if they encounter a “naïve” opponent as opposed to an opponent who has similar experience and training in negotiation? One investigation compared situations in which: (a) both buyer and seller had training; (b) neither had training; (c) buyer had training, but seller did not; and (d) seller had training and buyer did not.⁷⁸ Overall, training improved negotiation outcomes. Unilateral training gave sellers an advantage, but not buyers. Why? First, experienced sellers are able to effectively create as much integrative potential as are two trained parties, indicating that one party who has training at the negotiation table goes a long way toward expanding the pie. When buyers are unilaterally trained, the effect seems to backfire due to the framing effect, such that buyers are focused on minimizing losses which results in a greater resistance to making concessions.⁷⁹

If we consider the fact that a strong correlation exists between feedback and performance, doesn’t it make sense to seek feedback on our negotiation ability? The key ingredient for effective negotiation training in the classroom is providing feedback. Feedback improves a negotiator’s ability to negotiate.⁸⁰ Experience accompanied by a debriefing is more effective in improving performance than experience without debriefing.⁸¹

The effectiveness of training programs is that it provides negotiators with feedback about their performance. Indeed, negotiators who are given feedback about the other party’s interests are more effective in reaching integrative agreements than negotiators who do not receive feedback.⁸² Feedback is most helpful when it allows negotiators to understand how important the issues were to the other party. For example, outcome-only feedback is not as effective as outcome feedback combined with insight into the nature of the task and the other party’s preferences and priorities for the issues.⁸³

The type and method of feedback is an important consideration. For example, in one investigation of business managers’ negotiations, negotiators were given one of four types of feedback (allegedly from their counterparty) following a negotiation, ranging from positive to negative, which focused on their abilities or their ethics:⁸⁴

• Positive-ability feedback (“What a skilled negotiator you seem to be.”)

• Negative-ability feedback (“What an unskilled negotiator you seem to be.”)

• Positive-ethicality feedback (“What an ethical negotiator you seem to be.”)

• Negative-ethicality feedback (“What an unethical negotiator you seem to be.”)

The key question was how the feedback would affect the performance of the negotiators in a subsequent negotiation situation. Negotiators who received the negative-ability feedback were the least competitive and achieved the worst individual performance. Negotiators who received the negative-ethicality feedback were the most honest. Negotiators who received the positive-ethicality feedback were the most cooperative.⁸⁵

In addition to the type of feedback negotiators give to one another, we examined the type of feedback a coach might give to a negotiator.⁸⁶ We first measured managers’ baseline performance in an initial negotiation. Then we separated them into one of five different “feedback groups”: no feedback (our “control” condition), traditional lecture-style feedback (also known as “didactic feedback”), information-based feedback (wherein negotiators learned about the other party’s underlying interests), observational feedback (wherein negotiators watched experts-in-action via video for about 15 minutes), and analogical learning (wherein negotiators were given relevant cases that all depicted a key negotiation skill). The results? Nearly everything is better than no feedback at all, and nearly anything is better than traditional, classroom-style, didactic learning (see Exhibit 8-5).⁸⁷

In an in-depth analysis of feedback in negotiation, two types of information were examined: how well negotiators understood the counterparty’s general priorities among the issues under negotiation and how much the counterparty gained for a particular offer. Both types of understanding are important for negotiators to improve: Understanding the counterparty’s interests is not sufficient to reach integrative outcomes; the additional step of assessment of their gains for each offer is key.⁸⁸

Negotiators who want to improve their skills may have a goal of learning or performing. Learning goals focus attention on task strategies rather than outcomes. In contrast, performance goals focus on outcomes. Negotiators who have learning goals are more likely to reach agreement and are viewed as more cooperative than negotiators who have performance goals.⁸⁹ Moreover, negotiators who have learning goals develop greater understanding of their counterparty’s interests and reach more integrative agreements than do negotiators with performance goals.

A key question is how well managers are able to transfer what they learn in the classroom to actual business negotiations. The rates of “positive transfer” (applying knowledge learned in one situation to another) are markedly limited.⁹⁰ Moreover, even the ability to benefit from our own experience is limited. For example, 100% of the respondents who read a negotiation case that contained win-win potential suggested (suboptimal) compromises.⁹¹ When attempting to learn something new (e.g., a key strategy, principle), it is important to have two (or more) cases or examples, rather than just one. The reason is clear: What is essential about any example or case taught in a business school is not the superficial details of the case but rather the underlying idea. The ability of a manager to separate the wheat from the chaff, or the core idea from the idiosyncrasies of the example, is limited if there is only one case. In fact, one case is no more effective than no cases at all.⁹² However, it is not enough to simply be presented with two cases; the manager needs to actively compare the two cases. Moreover, even if the instructor does not provide more than one case, if the manager (or trainee) can think of examples from his or her own experience, it can help significantly. Diverse analogical training, wherein negotiators compared several different value-creating strategies, such as logrolling and contingent contracts is more effective for learning broad, underlying value-creating principles than more narrow training, in which negotiators are only exposed to one type of pie-expanding strategy.⁹³ Negotiators who engage in deep-level comparisons (thinking about how two or more situations are similar to one another at a structural level) are more likely to reach an optimal (win-win) solution in a subsequent situation, and this creates greater trust between the parties, which promotes greater deep-level transfer across situations.⁹⁴

Counterfactual reflection is the process of thinking about the past. Negotiators who reflect on “additive” counterfactuals (e.g., “if only I had …”) learn more than negotiators who reflect on subtractive counterfactuals (e.g., “If only I had not …”).⁹⁵ Generating additive counterfactuals about a previous negotiation led to a distinct advantage for negotiators as compared to subtractive counterfactuals in terms of distributive and creative agreements.

Excellent problem solvers frequently report that after trying to solve a problem and getting nowhere, they put the problem aside for hours, days, even weeks, and upon returning to it, they can see the solution quickly. (For a real-life example of the incubation effect, see Exhibit 8-6.) The incubation phase is usually one step in a process of problem solving detailed in the following sequence:

1. Preparation. During the preparation phase, the problem solver gathers information and makes preliminary attempts to arrive at a solution. The key is to understand and define the problem. As we have noted, finding a good problem is the essence of effective negotiation.

2. Incubation. When initial attempts to solve the problem have failed, problem solvers may put the problem aside to work on other activities or even to sleep. Indeed, negotiators were more likely to reach high-quality, integrative agreements after they took a break in which they were cognitively busy with a distraction task than after a break in which they could reflect upon the negotiation.⁹⁶ Breaks are particularly beneficial for prosocial (versus proself) negotiators: Prosocial negotiators who reflected during a three-minute break were more successful in reaching agreements than those who were distracted during the break or were proself.⁹⁷ Another example: Think about the necklace problem you were challenged with in Exhibit 8-1, in which you are given four chains to make one necklace on a limited budget. Three groups of people worked on this problem.⁹⁸ One group spent 30 minutes trying to solve it, with a solution rate of 55%. A different group spent 30 minutes trying to solve it but was interrupted during the solving period with a 30-minute break. In this group, 64% of participants solved the problem. A third group spent 30 minutes trying to solve the problem but were interrupted in the solving process by a 4-hour break. Of this group, 85% solved the problem. Whereas we do not guarantee that difficult negotiation situations will always be met with illumination after putting the problem aside, it certainly cannot hurt to try.

   Difficult issues in negotiation can stymie negotiations and even threaten agreement. Perhaps facing such obstacles can lead negotiators to step back and look at the big picture, or perhaps if they face such obstacles, they become so myopic they don’t see other possibilities. Negotiators who face obstacles head-on tend to get stuck and are less able to create integrative solutions.⁹⁹ Negotiators with a “distal” (e.g., “10 years from now”) rather than “proximal” (e.g., “next month”) time perspective reached more integrative agreements.

3. Illumination. During the illumination phase, the key to a solution often appears. It often happens when people are doing something completely unrelated to solving the problem.

4. Verification. In the verification phase, problem solvers need to check the solution to make sure it works.

The rational problem-solving model, patterned after Pólya, also describes four steps for solving a problem.¹⁰⁰ However, unlike the incubation method, the rational problem-solving model is deliberate and systematic:

1. Understand the problem. In this step, the negotiator needs to ask himself or herself: What is known? What is unknown? What are the data I am using? What are my assumptions?

2. Devising a plan. During this step, the negotiator may ask himself or herself whether past experience is a profitable means of finding a solution method, engaging in a search for similar problems, or perhaps restating the goal of the problem.

3. Carrying out the plan. In this step, the negotiator carries out the plan and tests it.

4. Looking back. In this step, the negotiator asks himself or herself whether he or she can obtain the result by using another method and looks at how it all fits together. In this step, it is important for the negotiator to ask what is the key takeaway.

The goal of brainstorming is to maximize the quantity and quality of ideas. Paradoxically, quantity is a good predictor of quality: A group is more likely to discover a really good idea if it has a lot of ideas from which to choose. However, brainstorming involves more than mere quantity. Contrary to popular corporate lore that brainstorming sessions are wild and crazy free-for-alls where anything goes, brainstorming has defined rules:¹⁰¹

• Expressiveness: Group members should express any idea that comes to mind, no matter how strange, weird, or fanciful. Group members are encouraged not to be constrained or timid. They should freewheel whenever possible.

• Nonevaluation: Do not criticize ideas. Group members should not evaluate any of the ideas in any way during the generation phase; all ideas should be considered valuable.

• Quantity: Group members should generate as many ideas as possible. Groups should strive for quantity; the more ideas, the better. Quantity of ideas increases the probability of finding excellent solutions.

• Building: Because all of the ideas belong to the group, members should try to modify and extend the ideas suggested by other members whenever possible.

An effective design for promoting creativity in negotiation involves separating the generation of ideas—leaving this task to individual team members—and then evaluating and discussing the ideas as a group.

To be effective at negotiation, negotiators must be good at deductive, as well as inductive, reasoning. Deductive reasoning is the process of drawing logical conclusions. For example, most people have some kind of training in solving logical syllogisms, such as the ones in Exhibit 8-7. The difficulty in solving these syllogisms does not imply that managers are unintelligent; rather, it indicates that formal logic and individual (or psychological) processes are not necessarily the same. However, many people violate rules of logic on a regular basis. Some of the most common violations of the rules of logic are the following:

• Agreement with a conclusion. The desirability of the conclusion often drives people’s appraisal of reality. This behavior, of course, is a form of wishful thinking, as well as an egocentric bias. The tendency is strong for people to judge the conclusions they agree with as valid, and the conclusions they disagree with as invalid.

• Cognitive consistency. People have a tendency to interpret information in a fashion that is consistent with information they already know. The tendency for people to judge conclusions to be true based upon whether the information agrees with what they already know to be true illustrates the need for consistency in one’s belief structure.

• Confirmation bias. People have a strong tendency to seek information that confirms what they already know. A good example of this bias is the card task presented in Exhibit 8-1.

Inductive reasoning is a form of hypothesis testing, or trial and error. In general, people are not especially good at testing hypotheses, and they tend to use confirmatory methods. Another example is the availability heuristic we discussed previously, which states that judgments of frequency tend to be biased by the ease with which information can be called to mind.

For example, people make inaccurate judgments when estimating probabilities. Consider the problem in Exhibit 8-8.¹⁰² When people are asked to answer this question, 22% select the first answer (i.e., the larger hospital), 22% select the second answer (i.e., the smaller hospital), and 56% select the third answer (i.e., both hospitals). They seem to make no compensation for large versus small sample sizes. They believe that an extreme event (e.g., 60% of births being male) is just as likely in a large hospital as in a small one. In fact, it is actually far more likely for an extreme event to occur within a small sample because fewer cases are included in the average. People often fail to take sample size into account when they make an inference.

In summary, managers do not form generalizations (reason inductively) in ways that statistics and logic suggest. When people make inferences about events based on their experience in the real world, they do not behave like statisticians. Rather, they seem to be heavily influenced by salient features that stand out in their memory, and they are swayed by extreme events even when the sample size is small.