When multiple parties are at the bargaining table, concerns about dividing the pie of resources loom large. Each party advocates in a self-serving fashion for their own interests and multiple definitions of fairness exist.

A key difference between two-party and group negotiations is the potential for two or more parties within a group to form a coalition to pool their resources and exert greater influence on outcomes.⁵ A coalition is a (sub) group of two or more individuals who combine their resources to affect the outcome of a decision in a mixed-motive situation involving at least three parties.⁶ Coalition formation is one way that otherwise weak group members may marshal a greater share of resources. Coalitions involve both cooperation in terms of attracting members and competition in terms of dividing resources.

Group members who form coalitions may engage in private caucusing for strategic purposes. Full-group communication is more time consuming but reduces inequality of group members’ outcomes, increases joint profitability, and minimizes perceptions of competition.⁷ However, when the task structure requires group members to logroll in a reciprocal fashion (as opposed to a circular fashion), restricted communication leads to higher joint outcomes than full communication. Consider one coalition: the Center for Sustainable Shale Development, composed of environmental groups, foundations, and major oil and gas companies with the goal of limiting pollution caused by fracking in the Appalachian region. Drilling companies who voluntarily submit to independent reviews of their operations and abide by stringent measures receive the coalition’s support.⁸

Negotiators may need to communicate with another person in the presence of someone who should not understand the message. For example, consider a couple selling a house engaging in a face-to-face discussion with a potential buyer. Ideally, the couple wants to communicate information to one another in a way that the spouse understands but the buyer does not—better yet, in such a way that the buyer is not even aware that a surreptitious communication is taking place. This is the multiple audience problem.⁹

People are quite skilled at communicating information to the intended recipient without the other party being aware.¹⁰ For example, former U.S. president Ronald Reagan was gifted in his ability to send different messages to different audiences within the same speech. Reagan’s “evil empire” speech of March 8, 1983 to the National Association of Evangelicals (and indirectly, the whole world) is a case in point. In the early sections of this speech, Reagan established identification with the evangelical audience through an ethos that exemplified their ideals, even using their technical vocabulary (e.g., “I believe in intercessionary prayer”). The section of his speech dealing with foreign policy was addressed to a complex array of audiences, global as well as domestic. The “evil empire” phrase had strong resonance not only with evangelicals but also with opponents of the Soviet Union everywhere, including those in Poland and Czechoslovakia. For the benefit of his diplomatic audiences however, Reagan carefully avoided specific references to evil actions of the Soviet Union, personally deleting from early drafts all references to chemical warfare in Afghanistan. And the speech’s attack on the nuclear freeze movement of that time was balanced with a call for “an honest freeze,” a term that created “presence” for his proposal for “extensive prior negotiations on the systems and numbers to be limited and on the measures to ensure effective verification and compliance.”¹¹ To his audience in the international, diplomatic, and arms control communities, including those within the Soviet Union, such praise alluded to extratextual facts that gave this part of the message a pragmatic connotation.¹²

In a multiparty negotiation, integrative trade-offs may be achieved through either circular or reciprocal logrolling.¹³ Circular logrolling involves trade-offs that require each group member to offer another member a concession on one issue while receiving a concession from yet another group member on a different issue. A circular trade-off is typified by the tradition of drawing names from a hat to give holiday gifts to people. People receive a gift from one person and give a gift to yet another person. Ideally, we give gifts that are more appreciated by the recipient than by the giver. In contrast, reciprocal trade-offs are fashioned between two members of a larger group. Reciprocal trade-offs are typified in the more traditional form of exchanging presents. Circular trade-offs are more risky than reciprocal trade-offs because they involve the cooperation of more than two group members.

Groups often simplify the negotiation of multiple issues among multiple parties through voting and decision rules. However, if not used wisely, decision rules can thwart effective negotiation, both in terms of pie expansion and pie slicing. A number of problems are associated with voting and majority rule.¹⁴

Know who will be at the table and understand the interests of the constituencies they represent. When negotiators share a social network, they already have a basis of trust; they are more likely to follow through on commitments, and power differences are not as magnified.²³ When U.S. National Security Adviser Tome Donilon flew to Beijing in May to set up a China-U.S. summit, he didn’t think he needed to schedule a meeting with a relatively unknown diplomat, Liu He. However, upon meeting Chinese leader Xi Jinping, it quickly became clear to Donilon that he had overlooked someone of great importance to the Chinese delegation. Mr. Xi pointed to the aide by his side, and said, “This is Liu He. He is very important to me”. It was only at that moment that Mr. Donilon realized that their diplomatic reconnaissance team had overlooked a key figure who was in charge of crafting a new economic vision to guide China’s economic policy decisions for the next decade. So, the U.S. National Security team scrambled to set up a session with Mr. Liu that very evening, booking a meeting room in Beijing’s opulent Great Hall of the People.²⁴

People experience “information overload” when dealing with multiple parties and multiple issues. It is nearly impossible to keep track of the issues, alternatives, and preferences of each party without some kind of information management device. Negotiators are well advised to develop a matrix that lists each party (along the rows) and each issue (along the columns), and then record each person’s preferences for each issue. To the extent that this information can be publicly displayed, it can greatly enhance the ability of the group to find win-win agreements.

Negotiating groups often severely mismanage their time. For example, negotiating groups begin by engaging in distributive bargaining and then transition to integrative bargaining.²⁵ Groups tend not to make proposals and explore options and alternatives in a systematic fashion. This can lead to tunnel vision, or the tendency for people in group negotiations to underestimate the number of feasible options available. In one of our investigations, we asked people who had just completed a multiparty negotiation how many feasible agreements they thought were possible (the negotiation contained five issues and four to five alternatives within each issue). The modal response was one. On average, people estimated approximately four feasible outcomes for the group (the highest estimate was 12). In fact, there were over 50 feasible outcomes! This example illustrates the tunnel vision (and ensuing desperation) that can overtake a group if members fail to systematize their proposal making. For this reason, negotiators should make several multi-issue proposals and record which proposals have been considered.

Most groups suggest fewer and lower-quality ideas than do individuals thinking independently.²⁶ Brainwriting, or solitary brainstorming, is a strategy whereby group members independently write down ideas for resolving negotiations and then later, when the group meets, they share those ideas. Brainwriting capitalizes on the fact that individuals are better at generating ideas but groups are superior in terms of evaluating ideas.

Multiparty negotiations need a timekeeper, a process manager, and a recorder of information. These roles can be rotated, so as not to give any particular member an advantage or disadvantage.

It is unwise for group members to leave the table when all parties need to reach agreement.²⁷ When groups leave the table, coalitions are more likely to form, which can be detrimental for the group.²⁸

The problem of uneven participation, wherein one or two people do all the talking, thwarts information exchange in groups.

Sometimes group negotiations can get bogged down because it takes longer for parties to reach agreements—even on a single issue. Failure to reach agreement on negotiation issues can make group members feel they are not making progress and that negotiations are stalemated. The more persistent cooperative negotiators are in their use of integrative strategies, the better they do for themselves.²⁹ Avoid reaching agreement just for the sake of reaching settlement but instead, agree on the process of reaching settlement. For example, a group member may suggest something like the following:

A tendency often emerges in group negotiations to divide things equally among the parties involved (see also Chapter 3, on pie-slicing). However, as we saw in Chapter 3, no fair method of allocation is universally acceptable. When parties engage in nonbinding discussions about fairness in a three-party negotiation, they are more likely to divide resources equally than when engaging in nonbinding talk about competitive reasoning.³⁰

The agreement bias occurs when negotiators focus on reaching common ground with the other party and are reluctant to accept differences of interest, even when such acceptance might create viable options for joint gain. Don’t assume everyone wants to “get to yes.” In some negotiation situations, people are paid to break deals and stall agreement.

Groups often use sequential bargaining (i.e., discuss one issue at a time) rather than simultaneous bargaining (where several issues are under consideration at any given time). By independently discussing and voting on each issue, negotiators cannot fashion win-win trade-offs among issues.³¹

Ideally, coalitions should contain the minimum number of people necessary to achieve a desired goal. Coalitions are difficult to maintain because people are tempted by other members to join other coalitions and because agreements are not enforceable.³²

Members of coalitions experience a strong pull to remain intact even when it is not rational to do so.³³ According to the status quo bias, even when a new coalition structure that offers greater gain is possible, members are influenced by a norm of coalitional integrity, such that they stick with their current coalition.³⁴ Negotiators should form coalitions early so as not to be left without coalitional partners. Negotiators should also monitor their negative emotions when forming coalitions. Negotiators form negative impressions of people who use anger and exclude them from coalitions and coalition resources.³⁵ On the rare occasions when people do form coalitions with “angry” partners, they tend to make deep concessions.

The distribution of resources among members of coalitions is complex because a normative method of fair allocation does not exist.³⁶ Novice negotiators often settle for “equal division,” but experienced negotiators never do so.³⁷ Experienced negotiators are much more willing and able to exploit differences in their relative bargaining power. To illustrate this observation, consider the following example: Lindholm, Tepe, and Clauson are three small firms producing specialized products, equipment, and research for the rehabilitation medicine community.³⁸ This area has become a critical, high-growth industry, and each firm is exploring ways to expand and improve its technologies through innovations in the research and development (R&D) divisions. Each firm recently applied for R&D funding from the National Rehabilitation Medicine Research Council (NRMR).

The NRMR is a government agency dedicated to funding research in rehabilitation medicine and treatment. The NRMR is willing to provide funds for the proposed research, but because the firms’ requests are so similar, they will fund only a consortium of two or three firms. The NRMR will not grant funding to Lindholm, Tepe, or Clauson alone.

The largest of the three firms is Lindholm, followed by Tepe, and then Clauson. The NRMR took a variety of factors into consideration when it put caps on funding, as shown in Exhibit 9-4.

The NRMR strictly stipulated that for a consortium of firms to receive funding, the parties in the consortium (either two or three firms) must be in complete agreement concerning the allocation of resources among firms.

If you are Lindholm, what consortium would you consider to be the best for you? Obviously, you want to be in on some consortium, with either Tepe or Clauson or both, to avoid being excluded. But what is the best division of resources within each of those consortiums? Suppose you approach Tepe about a two-way venture, and Tepe proposes receiving half of the $220,000, or $110,000. You argue that you should earn more because you are bigger and bring more synergy to the agreement. You demand $200,000 for yourself, leaving $20,000 for Tepe. Tepe threatens to leave you and approach Clauson. Tepe argues that Tepe and Clauson can command $150,000 as a consortium without you, and each can receive $75,000. At this point, you argue that you can outbid Tepe’s offer to Clauson by proposing $80,000 and keep $110,000 for yourself. Just as Tepe is threatening to overbid you for Clauson, Clauson steps in and tells Tepe that Clauson would want at least $100,000 of the $150,000 pie that Clauson and Tepe together could command. Tepe is frustrated but relents.

You get nervous in your role as Lindholm. You certainly do not want to be excluded. You could attempt to get Clauson or Tepe in a consortium. Then a new thought occurs to you: Maybe all three of you can be in a consortium. After all, all three firms command the greatest amount of funding ($240,000). But how should the $240,000 be divided between the three of you? You are the biggest firm, so you propose that you keep half of the $240,000 (or $120,000), that Tepe get $80,000, and that Clauson get $40,000. This allocation strikes you as fair. At this point, Clauson gets upset and tells you that Clauson and Tepe can team up and get $150,000. Clauson thinks your share is unfair and should be reduced to something less than $90,000. You then remind Clauson that you and Tepe can get $190,000 together, of which you certainly deserve at least half, which is better than the $90,000 offer. Then the three of you are at it again in a vicious circle of coalition formation and demolition.

The negotiation between Lindholm, Tepe, and Clauson illustrates the unstable nature of coalitions. In this example, the excluded party is always able to approach one of the two parties in the coalition and offer a better deal, which can then be beaten by the remaining party, ad infinitum. Furthermore, splitting the pie three ways seems to offer no obvious solution. So, what should the three parties do? Is there a solution? Or are the parties destined to go around in circles forever?

Because of the commitment process, people tend to feel obligated to others with whom they have made explicit or implicit agreements. For this reason, it is important to make contact with key parties early in the process of multiparty negotiation before they become committed to others.

Most people feel obligated to follow through with promises they make to others, even when verbal commitments are not legally binding.⁴⁸

If one or more members of the coalition regard the proposed allocation of resources to be unfair, the coalition will be less stable, and they will be likely to renege. To the extent to which coalitional members feel that the distribution of the pie is fair, they are more likely to resist persuasion from others to leave the coalition.

Using an agent means more parties are claiming a fixed bargaining surplus. A small bargaining zone increases the likelihood of an impasse. (For an example, see Exhibit 9-6.)

It is unwise to trust someone to effectively represent your interests when their incentives are not aligned with yours. Ultimately, an agent’s goal is to broker a deal, and thus agents are motivated to apply pressure to whomever is motivated to reach a deal. Agents have an incentive to make transactions happen. For example, in home buying, a buyer’s agent is really an employee of the selling company. The agent’s preference is the higher price because he or she gets a commission. Furthermore, an agent may give biased information to get an agreement from his or her constituency. If the principal has strong bargaining power (i.e., attractive outside options), the principal has an incentive to understate those options and then renege on promised payments. Conversely, if agents have strong bargaining power, the principal has an incentive to overstate her outside option to capture more surplus.⁵⁷

A key question for agents is whether to align with their principal or to align with the other agent.⁵⁸ Obviously, certain laws and regulations govern disclosure, but the question is which social bond is most important for the final agreement. Agents initially show greater loyalty to their principals, but over time their loyalty to the other agent is greater. Moreover, to the extent that the across-the-table relationship among agents is strong, the likelihood of agreement is greater and settlements occur in the middle of the bargaining zone. Most notably, to the extent that the agents are socially similar (i.e., graduated from the same school, etc.) and familiar with each other, they are more likely to forge a bond. Agents who are “securely” attached to their principals negotiate more effectively and are less pro-self than are agents who are not securely attached.⁵⁹

Because an agent is negotiating in your stead, you are giving up control over the process of negotiation and ultimately, the outcome. Indeed, agents are more active in a negotiation than principals.⁶⁰

Because agents have an incentive to reach agreement, they may fall prey to the getting to yes bias in which agreement becomes more important than the contents of the deal.⁶¹

Do not assume the first agent you meet is uniquely qualified to represent you. Ask the agent how he or she will successfully represent your interests. Ask the agent about what is expected of you. Ask the agent about the nature of your relationship and what obligations if any, you have to one another. For example, when superstar music mogul Jay-Z entered into the sports agency business with his partner, Kimberly Miale, they lured several baseball athletes away from their longtime teams to more lucrative contracts with competing teams. However, Jay-Z was not certified as an official agent by the NFL players association, which has the goal of representing athletes’ interests.⁶² Ask agents about their negotiation training and strategies. (See Exhibit 9-7 for some suggested questions to ask an agent.)

Do your homework before meeting with your agent. Know your own BATNA. Prepare questions to ask your agent that allow you to test the accuracy of your BATNA, but do not reveal your BATNA. For example, a home seller might say, “I would like to find out what average sale prices are for this type of home.”

One of the most challenging tasks for a negotiator is to communicate his or her interests, priorities, and preferences but not reveal his or her BATNA. Help your agent by prioritizing your key interests and what you perceive to be the alternatives for meeting those interests. Anticipate that your agent will ask you about your BATNA. When this question comes up (and it will!), focus on your priorities (e.g., “I am not sure how helpful it is to tell you the most money I am willing to pay for the house you showed me today. However, I am really interested in a home within this school district and a three-car garage. In fact, I would be willing to pay more for those features than I would a master suite and an updated kitchen”).

Good agents have a wealth of expertise. Ask them about their key strategies for targeting opportunities for you and closing deals.

Agents, by virtue of their professional affiliations and networks, have access to a lot of information. If your agent is unwilling or unable to obtain information, interview another agent and see whether he or she can provide the information.

Agents are usually part of larger social networks. A negotiation chain is formed when multiple related negotiations are spread over multiple agents.⁶³ For example, more than five agents, representing three unions, federal mediators, and BART’s board of directors were ultimately involved in the resolution of the Bay Area Rapid Transit (BART) strike of 2013.⁶⁴ Whereas each agent has their own goals and tries to maximize their own interests, their ultimate outcomes are related to how well other agents cooperate. Agents perform better when they understand the larger network and allocate appropriate time for each negotiation.

By nature of the principal-agent relationship, an agent’s authority is limited with respect to making certain concessions or types of agreements (i.e., your agent cannot reduce or increase your offer without explicit direction from you).

Sometimes negotiators make what they regard to be perfectly reasonable proposals that are insulting to the other party. When this situation happens (and if the counterparty is insulted), negotiations may start on a losing course. In an agent-mediated negotiation, you can attempt to salvage damaged egos and relationships by blaming your agent.

Agents can provide an emotional buffer between parties who either may dislike one another or are irrational (see Chapter 5 on bargaining styles). Effective agents will put a positive “spin” on the communications by each party and effectively “tune into” their principal’s needs.

Negotiators are challenged with synchronizing internal with external negotiations. The behind-the-table barrier refers to how negotiators must sell deals to their own internal constituencies as well as the other party.⁶⁷ These barriers are also known as “level-two” barriers, which refer to the formal or informal domestic ratification of traditional diplomatic agreements. One strategy for reaching level-two agreements is when side A negotiates with side B to help B strengthen B’s internal supporters and overcome B’s internal opponents. In this sense, A and B form a tactical coalition to overcome potential blocking coalitions. Another strategy is to reach agreement on the elements of the negotiation process itself, which can send a valuable signal to level 2 players. In one union-management negotiation, the parties realized that reaching agreement too quickly would raise suspicions by the parties’ constituencies. So, the negotiators agreed to lock themselves into a room from mid-afternoon through dawn. Outside the room was heard angry shouts and table pounding. Inside, nice meals were ordered and plenty of alcohol flowed with the parties periodically making loud theatrical noises amid authentic chuckling to dramatize the alleged battle being fought.⁶⁸

The negotiators seated at the bargaining table comprise the primary relationship in negotiation. The relationship parties share with their constituents is the second table.⁶⁹ Constituents do not have to be physically present at the negotiation table for their presence to be strongly felt.⁷⁰ Negotiators who are accountable to their constituents make higher demands and are less willing to compromise than those not accountable to constituents.⁷¹ Gender affects accountability. Women, as compared to men, believe they are more generous when representing a group; conversely, men don’t believe this. When it comes to actual behavior, men are significantly more self-interested when they are responsible for a group as compared to when they are acting only on behalf of themselves; female representatives don’t behave differently whether they are representing a group or just themselves.⁷² Male constituency representatives tend to explain and justify their successes and not their failures; whereas women tend to justify their failures but not their successes.⁷³

The second table has a paradoxical effect on the primary table. Representatives of groups often are not given power to enact agreements; that is, they are not monolithic.⁷⁴ Whereas this would seem to reduce one’s power at the bargaining table, the negotiator whose “hands are tied” is often more effective than the negotiator who has the power to ratify agreements. Anyone who has ever negotiated a deal on a new car has probably experienced the “My hands are tied” or “Let me take it to the boss” ploy, in which the salesperson induces the customer to commit to a price that requires approval before a deal is finalized.

It is important to understand the relationships negotiators share across the bargaining table and the hidden table of constituent relationships.⁸² Teams who report to a “profit-oriented” supervisor claim a greater share of the resources than do teams who report to a “people-oriented” supervisor and teams not accountable to a manager.⁸³ In an investigation of representatives with a proself versus prosocial orientation, prosocial representatives were more willing to sacrifice self-interest to benefit the constituency and the adversary combined.⁸⁴

Representatives need to understand their constituents’ interests, not just their positions. Moreover when constituents feel heard, they are less likely to take extreme action. In many cases, representatives act too early (before they understand their constituency’s real needs) so as to demonstrate their competence. For example, a minority of “hawks” (people who support competitive behavior) is sufficient to induce constituency representatives to act in a competitive way, even when the majority are “doves” (people who support cooperative behavior).⁸⁵ Constituency representatives unconsciously accord more weight to hawkish than to dovish messages.

Constituencies are often composed of individuals and subgroups with different needs and interests.

Constituents, like other people, suffer from egocentric bias and want you to educate the other side with your position. It is important to clearly define your role to your constituents early on in the process. Set realistic expectations. Do not characterize yourself as an “evangelist” for their “crusade.” Share with your constituents all possible outcomes, not just the favorable ones.

Horizon thinking involves making projections about future outcomes. People have a difficult time thinking about future events,⁸⁶ tend to under- or overestimate the duration of future emotional states,⁸⁷ and fail to account for positive or negative circumstances that could arise.⁸⁸ Help your constituents develop a sound BATNA and realistic aspirations by engaging in horizon thinking.

Consider the following criteria for choosing and evaluating teammates:

1. Negotiation expertise. People with good negotiation skills may be able to devise an integrative solution to a complex conflict situation. A negotiation expert can streamline preparation, ensure the team avoids the four major traps of negotiation (see Chapter 1), avoids destructive conflict strategies, and instigates a creative problem-solving process.

2. Technical expertise. It helps to have someone with technical expertise in the domain of interest. For example, when house buying, it is valuable to have someone skilled in architecture, plumbing, electricity, and so on.

3. Interpersonal skills. Negotiation involves many interpersonal skills, such as the ability to establish rapport, communicate effectively, and redirect a power- or rights-based argument to one focusing on interests.¹⁰⁰

Two or three heads can be better than one, but at some point conformity pressures increase with group size, peaking at about five and then leveling off.¹⁰¹ As teams grow in size, coordination problems increase.

Communication, or information pooling, is facilitated if members are acquaintances or share a relationship. For example, when the clues for solving a murder mystery game are distributed among group members, groups of friends are more likely to pool their diverse information than are groups of strangers.¹⁰²

Cohesion is the strength of positive relations within a team,¹⁰³ the sum of pressures acting to keep individuals in a group,¹⁰⁴ and the result of all forces acting on members to remain in a group.¹⁰⁵ Cohesive groups perform better than less cohesive groups.

Different kinds of bonds keep teams together. Common-identity groups are composed of members who are attracted to the group; the individual members may come and go. Common-bond groups are composed of members who are attracted to particular members in the group.¹⁰⁶

Organizational members often negotiate as a team or a group because no single person has the requisite knowledge and expertise required to negotiate effectively. Thus, knowledge is distributed among team members. How effective are teams at utilizing knowledge that is distributed among members?

It is more efficient for each team member to be responsible for a particular piece of information so that each member is not overwhelmed by too much data. However, as storage space is minimized, so are the chances of successfully retrieving the desired information. Furthermore, groups are less likely to consider and discuss information that is shared only by a subset of its members. They suffer from the common information bias.¹⁰⁷

Members of groups are not privy to the same facts and information. People rely on others for information. Teams can be more efficient by dividing the labor among members. However, distributed cognition is risky because information may be lost to the entire group if a team loses one of its members. Thus, groups face a dilemma: Divide responsibility, which increases members’ dependence upon each individual member, or share information, which is clumsy and redundant.

It is imperative that team members have shared goals and interests. A large experiment with 80, 4-person teams revealed that conflict between subgroups exerted a detrimental effect on negotiation performance.¹⁰⁸ The higher the level of team identification, the lower the level of task and relationship conflict in teams.¹⁰⁹ For example, top management teams that have high trust have greater agreement-seeking behavior and collaboration than low-trust-teams.¹¹⁰ When teams aggregate their individual interests, majority members have more influence than do minority members and hawkish members have more influence than dovish members.¹¹¹

Team preparation is so important that we developed a worksheet for effective team preparation (see Exhibit 9-8). Preparing together creates a transactive memory system in which group members understand the information others have and how and when to access it. For example, groups in one investigation were given instructions on how to assemble a transistor radio. Some groups trained together; in other groups, individuals trained individually (or with a different group). When it came to actual performance, groups who had trained together outperformed those who had trained individually or with different groups.¹¹²

Make sure that you schedule breaks into your negotiation to allow team members to meet privately. Admiral Mike Mullen, chairman of the U.S. Joint Chiefs of Staff, and General Ashfaq Parvez Kayani, who heads Pakistan’s Army, differed over strategies for fighting insurgents. However, they came together after U.S. airstrikes on targets in Pakistan and eventually forged a working relationship built on a tense international platform. Mullen casually brought sides together by strategically engineering breaks into the negotiations. During secret meetings in the Indian Ocean aboard an American aircraft carrier, Mullen’s team, worried that Kayani, a chain smoker, would balk at the ship’s no-smoking rule, scheduled regular negotiation breaks and permitted Kayani to smoke on the ship deck. At dusk, the men would gather to watch warplanes take off for missions over Afghanistan, and during these “breaks” several issues were discussed.¹¹³

A word of caution: Many teams spend too much time in private caucus and not enough time at the table. This behavior is ultimately not effective for negotiation.

It is important to assess the extent to which team members are accountable to others outside of the team. For example, when teams are accountable to a supervisor, they are more effective than when they negotiate strictly on their own behalf.¹¹⁴

People identify with many different social groups.¹¹⁶ For example, a student might consider a relevant group to be the other students in his or her study group, the class as a whole, marketing majors in general, or the entire student body. At any given time, one group might be more or less salient to the student: At a football game, students might identify most strongly with the entire student body; in a dining hall, students might identify most strongly with a particular dorm or floor.

Imagine you are in an organization in which marketing and finance are distinct subgroups located on different floors of a building. Contrast that arrangement to a situation in which marketing and finance are not separate functional units but instead, are part of the same product team. What happens in the case in which a marketing manager negotiates with a financial manager? Negotiations among individuals representing different social groups are less mutually beneficial than negotiations among individuals who perceive themselves as belonging to a larger social organization—one that encompasses all those present at the bargaining table.¹¹⁷ When people define their social identity at the level of the organization, they are more likely to make more organizationally beneficial choices than when social identity is defined at an individual or subgroup level. For example, when group members are instructed to consider features they have in common with another group, behavior toward out-groups is much more generous than when they consider features that are distinct.¹¹⁸

Five beliefs propel groups toward conflict: superiority, injustice, vulnerability, distrust, and helplessness.¹¹⁹ These deeply entrenched beliefs can trigger destructive action. Moreover, to the extent that groups receive social support from their fellow in-group members, such beliefs can lead to even greater intergroup conflict.¹²⁰ Bias and discrimination can also thwart effective negotiation. For example, racial in-group affect (i.e., liking one’s racial group) leads to lower joint outcomes and a “chilling and competing” negotiation approach.¹²¹ Group distinctions and social boundaries may be created on the basis of completely arbitrary distinctions.¹²² In one investigation, participants were divided into two groups on the basis of an arbitrary procedure (random draws from a box).¹²³ Then, individuals negotiated with either a member of their “own group” or the “other group.” Even though the information concerning the negotiation situation was identical in both respects, negotiations with members of out-groups were anticipated to be more contentious than negotiations with members of in-groups; further, the mere anticipation of negotiation with an out-group member led to increased in-group bias, positive evaluations of one’s own group relative to the out-group. In another investigation, people allocating money between their group and a competing group took a significantly greater share of the monetary funds than people allocating between themselves and a competing individual.

When we anticipate negotiations with out-group members, we often engage in downward social comparison.¹²⁴ We evaluate the competitor to be less attractive on a number of organizationally relevant dimensions (such as intelligence, competence, and trustworthiness) than members of our group. However, after successful negotiation with out-groups, intergroup relations improve, and downward social comparison virtually disappears.¹²⁵ Negotiation with out-group members is threatening to organizational actors, but to the extent that integrative agreements are feasible, negotiation has remarkable potential for improving intergroup relations. Although our initial expectations may be quite pessimistic, interactions with members of opposing groups often have a beneficial impact on intergroup relations if several key conditions are met, such as mutual dependence for goal attainment.

People of high status, those of low status who have few alternatives, and members of groups who have an opportunity to improve their group are most likely to identify with their own group. Members of groups with lower perceived status display more in-group bias than members of groups with higher perceived status.¹²⁶ However, high-status group members show more in-group bias on group status-related dimensions, whereas low-status group members consider the in-group superior on alternative dimensions.

Groups in conflict do not have an accurate understanding of the views of the other party and exaggerate the position of the other side in a way that promotes the perception of conflict.¹²⁷ Each side views the other as holding more extreme and opposing views than is actually the case. And people perceive more disagreement with rivals about values that are central to their own party’s ideological position than those that are central to their rival’s position.¹²⁸ Moreover, people believe their adversaries are actually motivated by their opposition to their own core values rather than by the promotion of the adversaries’ core values.

Consider the Trayvon Martin case involving the death of a young African American man who was in an altercation with and fatally shot by neighborhood watch patrolee, George Zimmerman, as Martin walked through Zimmerman’s gated community in Sanford, Florida. A trial ultimately led to the acquittal of Zimmerman’s charges of second-degree murder and manslaughter. Many details of the case were ambiguous and controversial, exacerbating the perception of differences in opinion and perceived intent. Partisans on either side of the case greatly overestimated the liberalism of advocates for “Stand Your Ground” laws and the conservatism of opponents of such laws.¹²⁹ The same polarization effect is found for other policy issues, such as abortion and immigration.

Why does this extremism occur? According to the naïve realism principle, people expect others to hold views of the world similar to their own.¹³⁰ When conflict erupts, people are initially inclined to sway the other party with evidence. When this tactic fails to bridge interests, people regard dissenters as extremists who are out of touch with reality.

Many conflicts between groups do not have their roots in resource scarcity but rather, in fundamental differences in values.¹³¹ For example, conflicts are judged as more difficult to resolve when they occur between members of different groups than members of the same group, holding constant the nature of the conflict.¹³² Consider the strong protests made against busing by people whose lives are not affected by it.¹³³ Presumably, people who do not have children or grandchildren are not affected by busing. However, they tend to have strong feelings about it. Busing does not represent an economic issue to them, rather it is a symbolic issue. It is important to understand which issues are symbolic and which are economic. Moreover, adversaries are more optimistic about intergroup negotiation when they are exposed to the actual, rather than assumed, views of their counterparts.

To the extent groups in conflict share a common identity, conflict and competition decrease dramatically.¹³⁴ People in organizations may identify at different levels within their organization (e.g., person, group, department, unit, organization as a whole). In one investigation, groups were told to focus on their group identities. Other groups who were involved in an objectively identical conflict were told to focus on the collective organization. Cooperation increased when groups focused on the collective, rather than their group, identities. Moreover, the stronger the group identification, the more likely it is that groups develop a shared understanding that leads to more integrative outcomes.¹³⁵ Finding common identity may be more efficacious when dealing with out-group members than actually understanding their underlying interests. Negotiations with in-group members are more cooperative when they share information about their underlying self-interests. Conversely, negotiations with out-group members are more cooperative when they don’t share information about their underlying interests.¹³⁶

Suppose three white managers watch a video of a discussion among members of a mixed-race group composed of three African American men and three Caucasian men. After watching the video, the managers are presented with the actual text of the conversation and asked to indicate who said what. They are very accurate at remembering whether an African American or Caucasian person made a particular comment, but their ability to remember which African American male said what is abysmal.¹³⁷ Within-race (or within-group) errors are more prevalent than between-race errors because people categorize members of out-groups not as individuals but simply as “African American.” It is important for people to treat members of out-groups as individuals.

The mere contact strategy is based on the principle that greater contact among members of diverse groups increases cooperation among group members. Unfortunately, contact does not always lead to better intergroup relations, and in some cases it may even exacerbate negative relations among groups. For example, contact between African Americans and Caucasians in desegregated schools does not reduce racial prejudice,¹³⁸ little relationship is noted between interdepartmental contact and conflict in organizations,¹³⁹ and college students studying abroad become increasingly negative toward their host countries the longer they remain in them.¹⁴⁰

Several conditions need to be in place before contact can have its desired effect of reducing prejudice.

• Social and institutional support. For contact to work, a framework of social and institutional support is needed. That is, people in positions of authority should be unambiguous in their endorsement of the goals of the integration policies. This support fosters the development of a social climate in which more tolerant norms can emerge.

• Acquaintance potential. Successful contact must be of sufficient frequency, duration, and closeness to permit the development of meaningful relationships between members of the groups concerned. Infrequent, short, and casual interaction will do little to foster more favorable attitudes and may even make them worse.¹⁴¹ This type of close interaction will lead to the discovery of similarities and disconfirm negative stereotypes.

• Equal status. The third condition necessary for contact to be successful is that participants have equal status. Many stereotypes of out-groups comprise beliefs about the inferior ability of out-group members to perform various tasks. If the contact situation involves an unequal-status relationship between men and women—for example, with women in the subordinate role (e.g., taking notes, acting as secretaries)—stereotypes are likely to be reinforced rather than weakened.¹⁴² If, however, the group members work on equal footing, prejudiced beliefs become hard to sustain in the face of repeated experience of task competence by the out-group member.

• Shared goal. When members of different groups depend on each other for the achievement of a jointly desired objective, they have instrumental reasons to develop better relationships. The importance of a shared group goal is a key determinant of intergroup relations. Sometimes a common enemy is a catalyst for bonding among diverse people and groups. For example, by “waging a war against cancer,” members of different medical groups and laboratories work together.

• Cross-group friendships. Sometimes it is not necessary for groups to have real contact with one another to improve intergroup relations. If group members know that another member of their own group has a friendship or relationship with a member of the out-group or a cross-group friendship, in-group members have less negative attitudes toward the out-group.¹⁴³ It is not necessary that all members of a group have cross-group friendships; merely knowing that one member of the group does can go a long way toward reducing negative out-group attitudes.

Many of these strategies are preventative and can help ward off unhealthy, destructive competition between groups. What steps can a manager take to deal with conflict after it has erupted?

The Graduated and Reciprocal Initiative in Tension Reduction, or GRIT model, is a model of conflict reduction for warring groups. Originally developed as a program for international disarmament negotiations, GRIT also can be used to deescalate intergroup problems on a smaller, domestic scale.¹⁴⁴ The goals of this strategy are to increase communication and reciprocity between groups while reducing mistrust, thereby allowing for de-escalation of hostility and creation of a greater array of possible outcomes. The model prescribes a series of steps that call for specific communication between groups in the hope of establishing the “rules of the game.” Other stages are designed to increase trust between the two groups as the consistency in each group’s responses demonstrates credibility and honesty. Some steps are necessary only in extremely intense conflict situations in which the breakdown of intergroup relations implies a danger for the group members.

Mikhail Gorbachev’s decisions in the period from 1986 to 1989 closely resemble the GRIT model.¹⁴⁵ Gorbachev made a number of unilateral concessions that resulted in serious de-escalation of world tensions in this period. On two occasions, the Soviets stalled resumption of atmospheric nuclear testing despite their inability to extend the prior treaty with the Reagan administration. They then agreed twice to summit meetings, despite the Reagan administration’s refusal to discuss the Star Wars defense system. They then agreed to the Intermediate-Range Nuclear Forces (INF) Treaty (exceeding the United States’ requests for verification) with continued refusal by the United States to bargain about Star Wars. Next came agreements on the Berlin Wall and the unification of Germany. Eventually, even the staunchly anti-Communist/anti-Soviet Reagan-Bush regime had to take notice. These events led to a period of mellowing tensions between these two superpowers (see Exhibit 9-9).

Although the GRIT model may seem overly elaborate and therefore inapplicable to most organizational conflicts, the model clarifies the difficulties inherent in establishing mutual trust between parties that have been involved in prolonged conflict. Although some of the stages are not applicable to all conflicts, the importance of clearly announcing intentions, making promised concessions, and matching reciprocation are relevant to all but the most transitory conflicts.