Abstract
The usual purpose of negotiations is to explore options and reach an agreement, if possible. We investigated a notable exception to this generalization, where a party negotiates without any intention of reaching an agreement. False negotiation occurs when a party gains more by stalling the negotiations until an external change takes place that improves its position considerably. While false negotiators aim to avoid agreement within the current frame of the negotiations, they also aim to keep the negotiation process alive, since walking away from the negotiation table could endanger their position. We report the results of a study that compared the actions of false and sincere negotiators. The false negotiators used competitive tactics that encumbered the negotiations, yet they concealed their intentions by maintaining a facade of cooperation. Our theoretical discussion is focused on the balancing act involved in false negotiations and the challenges it poses for actors in social, managerial, and political settings. We conclude our analysis with an example from the realm of international negotiations.
Sincerity is everything. If you can fake that, you’ve got it made. — George Burns (American writer and comedian)
The typical purpose of negotiations is to explore options and, if possible, reach an agreement beneficial to all parties (Lax and Sebenius 2006; Mnookin 1993; Raiffa 1982; Raiffa, Richardson, and Metcalfe 2002; Susskind, McKearman, and Thomas-Larmer 1999). This article investigates a notable exception to this generalization, where a party engages in the negotiation process without any intention of reaching an agreement. We call such a party a false negotiator.
Varied circumstances can motivate false negotiations (Anand, Feldman, and Schweitzer 2009; Wallihan 1998). A party may expect to reap greater benefits by delaying the negotiations indefinitely or up to a point at which a significant external change is due to occur that should improve its position considerably. For instance, a country negotiating a bilateral trade treaty with a partner country could stall the current process because it expects the upcoming election (in the partner country) to bring to power a new coalition that would hold softer positions. Or consider a company negotiating with a government regulator. Frustrated with the demands of the present regulator and knowing that the official’s tenure in office is nearing its end, the company hampers the negotiations, hoping that the new appointee will be more amenable to its interests.
There are other reasons for conducting false negotiations. Parties may enter into negotiations to achieve ulterior goals. Thus, a job candidate may negotiate the terms of a position with one employer (in an attempt to solicit an offer) purely for the purpose of improving the terms of a more coveted position with another employer; a businessperson may initiate false negotiations solely to gather intelligence about a competitor; politicians may engage in false negotiations to establish a reputation as either tough or cooperative in an environment where they expect to have future interactions; finally, rival states might enter false negotiations on multilateral issues, such as disarmament or peace accords, to appease concerns of the international community as well as garner international support and avoid potential economic sanctions.
False negotiators, be they individual actors, companies, or states, are motivated to avoid reaching an agreement within the current negotiations. Yet they are also motivated to keep the process alive, since walking away from the negotiation table would worsen their position and expose them to risks and penalties. For example, the company described previously might enter into false negotiations with a government regulator to avoid the government’s threat of initiating harsher legislation or coming to an agreement with one of the company’s competitors. Similarly, a country might carry on lengthy ceasefire negotiations with its neighbors so as to avoid sanctions that might be imposed by the international community. The false negotiator thus elects to engage in false negotiations in the belief that this will best preserve its interests.
In summary, false negotiators assume that their prospects are superior to those they might obtain from sincere negotiations; they also assume that their chances of being exposed are small and that the potential damage to their reputation (if they are exposed) would be limited and well worth the risk. Individuals might be less likely to engage in false negotiations with partners with whom they expect to have a long-term relationship and repeated encounters (e.g., a store owner with merchandise suppliers), though such a possibility should not be excluded.
In this research, we investigate the dynamics that take place when a party enters a negotiation process and, unbeknownst to the other party, intends not to reach an agreement. We start by developing a conceptual framework for understanding false negotiations. We then report the results of a study that allows us to extract the key characteristics of false negotiations and delineate the challenges involved in detecting false negotiators. We conclude with theoretical implications for understanding negotiations in the realm of business, politics, and international relations.
Conceptual Framework
To better understand the nature of false negotiation, let us situate it in the context of negotiation theory. A basic principle of negotiation theory is that an agreement is reached if it provides sufficient benefits to all sides. Such an agreement is said to fall within a theoretical “zone of possible agreement” (Raiffa 1982). An agreement is beneficial to a given party if it is better than the party’s best alternative, that is, better than what the party could achieve without an agreement. The “best alternative to a negotiated agreement,” or the BATNA, might be, for instance, the status quo or a plan that could be enacted should the parties fail to reach an agreement. In theory, the BATNA serves as a salient reference point in negotiations (Fisher and Ury 1981; Thompson, Wang, and Gunia 2010). Negotiators evaluate potential agreements relative to their BATNA and should accept only agreements that are better than their best alternative.
False negotiators believe that their BATNA—which is based on their expectations for the outcome of future events or future negotiations—is superior to any proposal they could realistically receive at present from the other party. In other words, they do not expect to reap greater benefits from an agreement than from remaining in disagreement (Wallihan 1998). Why then do they enter the negotiation process in the first place? The answer is simply that they must go through the negotiation process so as to sustain their BATNA; failing to negotiate could endanger their eventual achievements. For instance, the job candidate described previously may find it necessary to generate a competitive offer from another employer to convince her preferred employer to improve his or her employment terms. The negotiation with one employer is used to produce a BATNA in negotiating with the other. Likewise, the company in the case described previously may be compelled to sustain the negotiation process with the present government regulator so as not to lose the option of negotiating in the future with a new, more forthcoming regulator. A failure to negotiate at present could result in an agreement between the present regulator and a competitor, which would undermine the company’s BATNA. (See a related discussion by Lamb et al. [2005], about how attempts to preserve the status quo could create similar dynamics.) In sum, parties sometimes go through the motions just to create or maintain their BATNA, knowing that failing to do so would endanger their interests (Wallihan 1998; Anand, Feldman, and Schweitzer 2009).
There are many possible variants on this general model. The negotiation process is dynamic and parties could change their mind about their goals midstream. For example, a negotiator could enter the process with sincere intentions and then change her mind upon realizing that stalling the process would serve her interests better. Alternatively, a negotiator could switch from false to sincere negotiations upon receiving an unexpectedly good offer. Notwithstanding these dynamics, our focus here is on false negotiation per se, that is, on the attitudes and actions of negotiators whose intention is not to reach an agreement within a given time frame, as long as they maintain such intentions.
Our theoretical analysis implies that, in order to be successful, false negotiators need to strike a balance between two opposing goals: avoiding an agreement that would endanger their BATNA and maintaining the negotiations so as to preserve their BATNA. This balancing act entails the concurrent use of competitive and cooperative tactics. Successful false negotiators use contentious tactics to stall the negotiations and avoid an unwanted agreement, yet they are also expected to be careful not to use tactics that would appear unacceptable to the other party and would increase the risk of break off and condemnation by the other party. In the interest of keeping the negotiations alive, false negotiators may also make cooperative moves and gestures of good will, signaling their cooperative intentions to the other party, until a point is reached where their desired goals deem feasible and their BATNA seems achievable.
This analysis may lead to an important theoretical conclusion as well as a disturbing practical one. If both false and sincere negotiators use a mix of competitive and cooperative tactics (Lax and Sebenius 1986), then what could be the real difference between them? Moreover, how can we distinguish false negotiators from sincere ones? We show that false and sincere negotiators differ in certain fundamental ways. They have different interests and different internal states—intentions, goals, and cognitions. Sincere negotiators employ mixed-motive tactics in the interest of exploiting opportunities and maximizing their gains. In contrast, false negotiators use mixed-motive tactics to avoid an agreement altogether, while pretending to be aiming for one. As our study reveals, these differences in goals are consequential in that they produce different behaviors. Our study is focused on those aspects and measures of behavior that distinguish false and sincere negotiators.
Before outlining our specific hypotheses and the feature of the study, we note the remarkable fact that false negotiation per se has hardly been investigated in negotiation ethics (see Wallihan 1998; Anand, Feldman, and Schweitzer 2009). To be sure, the role of deception and misrepresentation in negotiations has received much attention. Different types of deceptive behavior and their moral acceptability have been discussed (Lewicki and Stark 1996; Lewicki and Robinson 1998), as well as the effects of deceptive tactics on the negotiation outcomes (Murnighan 1991; Schweitzer, DeChurch, and Gibson 2005), in one-shot and repeated interactions (Boles, Croson, and Murnighan 2000). None of these studies, however, has addressed deception with the ultimate goal of not reaching an agreement; rather, they have been conducted in agreement-seeking contexts. Also, to our knowledge, false negotiation has not yet been studied empirically. The present study targets these gaps.
Hypotheses
Our research hypotheses stem from our theoretical analysis of the motivations underlying false negotiations. We developed an experimental platform that allowed us to test the behavioral differences between false and sincere negotiators. In general, we hypothesized that false negotiators would display two sets of behaviors, some that obstruct the negotiation and some that promote cooperation.
Our first general hypothesis was that false negotiators would adopt delay tactics that stall or prolong the negotiations so as to avoid an agreement within the prescribed time frame, in line with the notion that inaction and delay can cause the negotiations to fail (Pruitt 1983; Lamb et al. 2005). It has been observed that negotiators who increase their use of positional (contentious) strategies over time tend to produce less efficient outcomes, that is, agreements with fewer joint gains (Olekalns, Smith, and Walsh 1996). Thus, false negotiators, who do not seek to benefit their partners, would escalate their positional bargaining as the negotiations progress.
We broke down this general notion into several specific delay measures. Specifically, we predicted that, compared with the sincere negotiators, the false negotiators would take more time to respond to their partners (Hypothesis 1a). We also predicted that they would take more time to offer their opening bids (Hypothesis 1b). In addition, we predicted that, when given an opportunity to shorten the overall negotiation time, false negotiators would choose it more often than sincere negotiators, as it increases the chances of an impasse (Hypothesis 1c). We also explored more subtle tactics of inaction. We predicted that false negotiators would tend to avoid the substantive issues essential to reaching an agreement and discuss fewer of the core issues than sincere negotiators (Hypothesis 1d).
Hypotheses 1e and 1f focused further on the obstructive side effects of well-known competitive tactics. Past research has shown how misrepresentation and escalation can encumber the process, prolong the negotiations, and reduce the chances of agreement (Lewicki et al. 1994; Schweitzer, DeChurch, and Gibson 2005). Accordingly, we predicted that false negotiators would be more likely than sincere ones to invent misrepresentations for justifying their hard positions and their delay behavior (Hypothesis 1e). We also predicted that false negotiators would harden their positions at the final stage of the negotiations, rather than soften them or keep them the same (Hypothesis 1f). This is just the opposite of what one might expect from sincere negotiation partners, who tend to speed up the process when they feel that “the time had come to settle” (Stevens 1963) and make more concessions (Olekalns, Smith, and Walsh 1996; Adair and Brett 2005).
Finally, we explored one more obstructive behavior stemming from research on the role of representatives in negotiation. Specifically, the use of representatives has some well-known advantages, but it can also complicate the negotiation process. Agents can claim to have limited power and thereby delay the negotiations (Rubin and Sander 1988). We predicted that when given an opportunity to use representatives who are not authorized to close a deal, false negotiators would be more likely than sincere ones to opt for this choice (Hypothesis 1g).
In line with our conceptual framework, the second group of hypotheses focuses on cooperative behavior. As part of their balancing act, false negotiators should signal cooperation and make concerted efforts to keep the negotiation on track, so as to avoid the risks and penalties associated with breakdown. Cooperative moves might convince the other party that contentious moves (if they are made) are carried in good faith and issues could nevertheless be resolved in due time. In particular, we predicted that false negotiators would be more likely than sincere negotiators to reiterate their commitment to the process and the importance of cooperation, a cooperative gesture that involves no real commitment (Hypothesis 2a). In addition, we predicted that false negotiators would face more requests to engage in additional interactions beyond the mandatory negotiation sessions and that they would comply with explicit requests to engage in such interactions (Hypothesis 2b) as a sign of good will.
While our main goal was to investigate the behavior of the false negotiators, we were also interested in their self-perceptions; to this end, we administered a postnegotiation questionnaire. We hypothesized that the false negotiators would report more frequent use of tactics such as delaying responses and avoiding discussion of core issues. The postnegotiation questionnaire also probed the partners about their perceptions of their opponents. We wondered whether partners of false negotiators would suspect their opponents’ motives and whether their feelings and attitudes toward their opponents would be different than in sincere negotiations. The data from this questionnaire should be of interest to the extent that our hypotheses (1a through 2b) are borne out by the data, indicating real behavioral differences between false and sincere negotiators. Specifically, we planned to test whether the false negotiators would be rated (by their opponents) as less interested in reaching an agreement than their sincere counterparts (Hypothesis 3a). We also predicted that false negotiators would be rated (by their opponents) more negatively (e.g., as less professional, less forthcoming, more stubborn) than would sincere negotiators (Hypothesis 3b).
Method
Participants
The participants were thirty-eight students majoring in law, business, or cognitive science (eighteen men, twenty women) who volunteered to participate in a negotiation workshop which involved two meetings conducted at the university over the course of a week, during which the experiment was to be conducted.
The Negotiation Scenario
We engaged the participants in a simulation involving two parties—an international high-tech company and a government agency. The two sides were to negotiate the terms of the company’s expansion plan—in particular, its financial terms and employment rules. The negotiations were to be carried out via instant messaging on the Internet, over the span of one week. The study allowed us to compare the negotiators’ behavior under two conditions that varied in terms of the sincerity of the high-tech company. Thus, while the government agency was always sincere, the company role was to be either sincere or false. The goals for the negotiations were explicated in terms of payoff tables. The company negotiators assigned to the sincere condition were given an incentive to reach an agreement with the government agency, while the company negotiators assigned to the false condition were given an incentive not to reach an agreement. The latter group was told that the tenure of the current director of the government agency was nearing its end and the new director was likely to be more forthcoming toward the company’s requirements. While the company preferred not to reach an agreement, it also sought to keep the negotiations on track lest the government break off the negotiations and approach one of the company’s global competitors, thereby producing the worst possible outcome for the company. The payoffs accompanying the role descriptions reflected these interests. We measured the participants’ behaviors during the negotiations and then their attitudes in a postexperimental questionnaire. Importantly, the participants were not instructed or advised to engage in any particular kind of tactic; the instructions merely explained the context and indicated the payoffs for different negotiation outcomes.
Procedure
In the first workshop session, the participants were given a presentation about the negotiation process. Several basic concepts were discussed, including reservation price, zone of potential agreements, interests versus positions, distributive bargaining, and integrative potential. Then participants were presented with the negotiation simulation used in this study, were paired, and assigned roles (government or company) and conditions (company was either false or sincere) at random.
Each participant was provided with background information about the negotiation scenario, a description of the two parties and the main negotiation issues, and payoff tables according to the roles and conditions. The participants were told that the information was confidential and that they should not share it with their negotiation partners or other workshop participants.
The participants were then given guidelines on how to conduct the negotiation. They were told that it was to be conducted through the Internet, using the instant-messaging system available on the Gmail website. They were instructed to conduct three online negotiation sessions of twenty minutes each, over the period of one week. They were further told they could also engage in additional voluntary communication via the instant-messaging and e-mail systems. At the end of the one-week period, they were asked to fill out the postnegotiation questionnaires and subjective ratings and hand in their computer files. A second workshop session was dedicated to debriefing.
There were two main advantages to using Internet communication in this negotiation experiment. First, the computer provided full, reliable records of the exchanges made during the negotiation sessions, their durations, and the interresponse time. Second, the use of Internet communication as a medium allowed the negotiating parties to coordinate and conduct their sessions more easily given their busy schedules; it also allowed them the opportunity to think through their positions and plan their strategies from one session to the next, over the course of the week.
Several disadvantages of negotiations over the Internet (e.g., e-mail) have been pointed out by the literature. A main drawback is the lack of certain verbal cues (such as intonation) as well as nonverbal gestures (i.e., body language). As a result, negotiators may find it harder to build rapport and trust through an electronic medium than in a face-to-face interaction. The lack of rapport might make it harder to dissipate the negative feelings that could develop between remote partners (Drolet and Morris 2000; Nadler 2004a, 2004b). Other researchers have suggested that negotiators tend to be more competitive when using remote media (Paese, Schreiber, and Taylor 2003), thereby increasing the likelihood of an impasse. To address these concerns, we asked the participants to introduce themselves to their partners briefly at the end of the first workshop session, but to refrain from making any reference to the negotiations. We did so because research has shown that allowing a brief encounter between negotiation partners prior to the negotiations helps create a level of rapport between strangers, which reduces the negative feelings that can develop during remote negotiations (Nadler 2004b).
Negotiation Materials
All the participants were given the same scenario involving a negotiation between an international high-tech company and the government ministry of industry, trade, and labor, over the terms of a plan to expand the company’s local manufacturing plant. The company submitted an application to the government asking for a permit for its expansion plan as well as a governmental grant of US$400 million. The company also requested to be exempted from the requirement of recruiting local workers from the area where the plant would be located. The government was interested in the expansion plan, yet would prefer to reduce the financial aid to the minimum. The government was also seeking to boost the local economy outside the big cities, especially in the region where the plant was located, by enacting preferential employment rules. It expected the company to hire all its new workers from the local workforce rather than from other regions. The level of financial support and the employment rules (percentage of workers from the local workforce) were thus the two issues that had to be resolved for the parties to reach an agreement.
The participants were further told that the negotiation process could result in one of the four possible outcomes: (1) an agreement, (2) an impasse (if there was no agreement by the end of the third session), (3) a unilateral decision by the company to break off the negotiations and build the new plant in another country (an unfavorable outcome for the government), and (4) a unilateral decision by the government to break off the negotiations and grant the expansion permit to another global company (an unfavorable outcome for the company).
Table 1 shows the payoffs for the four outcomes. It clearly implies that reaching an agreement would be the most advantageous for both parties in the sincere negotiation condition. It also implies that a unilateral break off of the negotiations would provide a positive value to the party initiating the break off and a zero value to the other party. The payoff schedules allowed for a broad zone in which seven different agreements could be crafted at the start of the negotiations. This zone expanded as negotiation time went by, since the value of the break off alternative decreased with time (from 60 to 0 points, as shown in the Table 1).
The Payoff Schedules for the Company and the Government in the Sincere Negotiation Condition.
Note: Negotiation payoffs ranged from a maximum of 250 to a minimum of −50 points, whereas breaking off the negotiation could yield 60 points at most. While negotiators could not have expected to gain the full 250 points, they could have expected to gain more than 60 points (the value of the exit option). In fact, at the beginning of the negotiation, one should refuse to settle for anything less than sixty, but as the negotiation clock ticks on and the value of the exit option decreases, one might settle for a smaller payoff.
The false condition was created by adding just one piece of information to the background information for the company negotiators and the values attached to the negotiation outcomes. The company negotiators in the false condition were told that the official government negotiator was due to retire and that his or her successor would be more amenable to the company’s requirements. Specifically, the company should be able to reach a more profitable agreement with the successor. Hence, the company (in the false condition) would be better off avoiding an agreement, but maintaining the negotiation process so as to ensure that the government would not break off the negotiations. Should the government negotiators choose to exercise their outside option, that would be by far the worst outcome for the company. The payoffs for the false and sincere company negotiators reflect these preferences (see Table 2). The information for the government negotiators in the false condition was the same as in the sincere condition.
Payoff Schedules for the Company under the Two Negotiations Conditions.
Negotiation Instructions
The negotiators were told to hold three mandatory negotiation sessions (twenty minutes each) over a period of a week, using instant messaging. In addition, they were told they could communicate over the Internet using the instant-messaging and e-mail systems as much as they wanted. The agreement could be signed only during one of the three sessions mandated by the protocol.
As mentioned before, two additional negotiation tactics were made available only to the company negotiators (sincere and false alike). First, they could call off the second session by notifying the government negotiator at the beginning of the session. Second, they could delegate their power (only during the second session) to a representative who was not authorized to close a deal. These tactics clearly have the potential to hamper and even derail the negotiation process. Our goal was simply to measure how often these tactics would be used by the sincere and false negotiators.
Upon completion of the negotiations, all the negotiators were requested to submit the outcome in writing, as well as the full records (computer files) of all Internet sessions, and all additional (e-mail or instant message) communications. They were informed in advance that all the research materials and records would remain confidential. Further, the negotiators were asked to complete a questionnaire about the negotiation process and their attitudes and feelings toward their negotiation partner. They were asked to rate, on a series of 5-point scales (1 = low, 5 = high), their feelings toward their partner (anger, frustration, positivity, pleasantness), their professional assessment of their partner (businesslike, persuasive, professional, communicative), and their view of their partner’s behavior during the negotiation (stubborn, flexible, bureaucratic). Additionally, they were asked to assess to what extent their partners were seeking an agreement. Finally, they were asked to address their own performance in the negotiations by reporting several aspects of their own behavior—whether they initiated extra meetings, delegated responsibility to an unauthorized representative, stalled for time, or avoided core issues. The negotiators were instructed not to reveal their final payoffs to their counterparts at any stage, even after the negotiations were over, so as not to contaminate their partners’ responses to the postexperimental questionnaires.
Results
Negotiation Outcomes
In line with the instructions given to the participants, we predicted more impasses in the false than in the sincere condition. In line with our expectation, only 22 percent of the negotiations in the false condition ended in an agreement (two agreements, five impasses, two break-offs initiated by the government), whereas 90 percent of the negotiations in the sincere condition did so (nine agreements, one break off initiated by the company; Fisher’s exact test, χ2(1, 19) = 8.93, p < .005).
Negotiation Processes: Review of Analyses
The negotiation outcomes in the false and sincere conditions were substantially different, thus setting the stage for the tests of our main research hypotheses regarding false negotiation behavior. Our primary predictions involved behavioral measures. They included the use of various noncooperative tactics intended to hamper and derail the negotiation process, as well as the use of tactics intended to create an impression of cooperation and commitment. The actual goal of these tactics was, in fact, to reduce the risk that the other side would break off the negotiations. The behavioral measures were complemented by postnegotiation questionnaires, which elicited the negotiators’ self-reports as well as their opponents’ opinions and attitudes toward them. While the opponents’ postnegotiation questionnaires were not central to our research, we considered them potentially insightful about what it is like to negotiate with false partners.
All the hypotheses involved directional pairwise comparisons of the false and sincere negotiators. For example, we predicted that false negotiators would exhibit more delays and offer more extreme bids and that they would be more likely to avoid discussion of core negotiation issues—that is, discuss less concrete proposals about financial parameters and employment rules. Given that we had a well-developed, a priori, theoretical basis for each of our directional hypotheses, we were able to use one-tailed t-tests to evaluate the significance of the differences between the false and sincere negotiators. This was done, in particular, to increase the statistical power of our study. We also report the effect sizes in terms of Cohen’s d (Cohen 1992).
Negotiation Processes: Noncooperative Behavior
Response times ( Hypothesis 1a)
We predicted that the false negotiators would delay their responses so as to stall the negotiation. To calculate how long it took negotiators to respond to their partners’ messages, we used the timing (hour and minute) of each response as recorded by the instant-messaging system and calculated the differences. Separate mean response times were obtained for each of the three negotiation sessions. As predicted, false negotiators took longer to respond. Thus, the reaction times (in minutes) of the false negotiators averaged 0.55 minutes (standard deviation [SD] = 0.35), roughly 60 percent longer than the reaction times of sincere negotiators, which averaged 0.33 minutes, SD = 0.16, t (17) = 1.81, p < .05 (one-tailed), effect size d = 0.88.
The false negotiators delayed their responses even more toward the end of the third session. Their response times were more than twice as long as those of the sincere negotiators during the last session, 0.63 versus 0.27 minutes, SDs were 0.31 and 0.19, respectively, t (11) = 2.46, p < .02 (one-tailed), d = 1.48.
The participants seemed aware of their stalling strategy. At the end of the negotiations, the participants were asked to report on a Likert-type scale to what extent they had stalled the negotiation deliberately (1 = low, 5 = high). The false negotiators were more likely than the sincere ones to report that they had done so, 3.86 versus 1.67, SDs 1.68 and 1.12, t (14) = 3.14, p < .004 (one-tailed), d = 1.68.
Timing of the first offer (Hypothesis 1b)
We hypothesized that the false negotiators would delay placing a concrete proposal on the negotiation table as much as possible; an offer was defined as concrete if it referred to at least one of the two core issues. We recorded the first time the company negotiator communicated a concrete proposal for either the grant level or the employment rate, or both. The dependent measure was the time lapse between the onset of the negotiations and the first concrete bid made by the company (based on the timing recorded by the instant-messaging system). The false negotiators placed their first bids later than the sincere ones. The time lapse averaged 27.44 minutes for the false negotiators and 16.56 minutes for the sincere ones; 1 SDs were 16.10 and 9.81, t (17) = 1.77, p < .05 (one-tailed), d = 0.86. Interestingly, in the three cases in which the false negotiators made their initial offer rather early, they quoted illegal values (e.g., one such negotiator demanded a grant of US$500M, which exceeded the highest option listed in Table 1).
Moreover, the false negotiators tended to produce initial offers only in response to an explicit request from the other side. To assess this tendency, the first offers of the company negotiator were coded for whether or not they had been given in response to such an explicit request (see Table 3). Among the false company negotiators, only 33.3 percent introduced their first offer voluntarily, while the rest did so following an explicit request from the other party. Among the sincere company negotiators, 88.9 percent placed their first offer on the table voluntarily, and the rest did so following a request; t (16) = 2.77, p < .01 (one-tailed), d = 1.38.
Examples of Messages Sent by the Company Negotiators.
Canceling a session (Hypothesis 1c)
All company negotiators were informed they could cancel the second session by notifying the government negotiator at the beginning of that session. Specifically, the instructions read, “Notwithstanding the mandatory nature of the sessions, you have the option of canceling the second session. To do so, you must inform your partner at the beginning of the second session that the session has been canceled. This notification cannot be given in advance.” We predicted that the false negotiators would choose to exercise this option more than the sincere negotiators so as make it easier for them to reach an impasse. This prediction was not borne out. Only one false negotiator chose to cancel the second session, whereas one sincere negotiator also chose to cancel the session. It is possible that cancellation seemed like an overly harsh move, and so negotiators in both conditions considered it too risky and feared that it could lead the other party to break off the negotiations.
Content of communications (Hypothesis 1d)
We predicted that the false negotiators would make fewer pragmatic, goal-oriented communications. In particular, we hypothesized that they would refer less often to matter-of-fact issues or the negotiations parameters listed in the payoff schedules. To assess this hypothesis, each message transmitted during the three mandated negotiation sessions (and during the additional e-mail and instant exchanges) was coded independently by two judges as to whether or not it contained relevant data. 2 A message was coded as relevant if it included one (or both) of the following aspects: (1) any reference to or statement on the payoff schedule (e.g., numbers involving the size of the grant or the employment rule), including questions; (2) stipulations invented by the sides that eventually became part of their agreement (e.g., additional guarantees, tax subsidies, implementation of the employment rules). Several detailed examples are listed in Table 3.
The level of agreement between the two coders was 94.9 percent. Disagreements were resolved through discussion. A relevancy index was created for each negotiator, first per negotiation session and then across all sessions. As expected, the false company negotiators produced fewer relevant communications across all sessions compared with the sincere company negotiators, 8.0 versus 16.6 relevant statements; the SDs were 5.05 and 8.58, respectively, t (17) = 2.62, p < .01 (one-tailed), d = 1.31. Furthermore, the relevancy indices were significantly lower for the false negotiators in the first session (p < .05) and in the second one (p < .05). In the third (and last) session, where the largest difference was found, the number of relevant statements averaged 5.71 in the false condition and 11.0 in the sincere condition; SDs were 1.38 and 5.63, respectively, t (12) = 2.41, p < .02 (one-tailed), d = 1.39.
At the end of the negotiations, the participants were asked to indicate the extent to which they deliberately avoided discussion of relevant matters on a 5-point scale (1 = low, 5 = high). Compared with the sincere negotiators, the false negotiators reported more deliberate avoidance of discussion of relevant matters, 3.44 versus 2.0; SDs 1.51 and 1.32, t (16) = 2.16, p < .05 (one-tailed), d = 1.08. The false negotiators thus seemed aware of using this strategy.
Introducing inventive constraints (Hypothesis 1e)
Negotiators who are trying to stall the negotiation process sometimes appeal to external forces that are (allegedly) tying their hands and limiting their ability to reach an agreement (Lewicki and Stark 1996). As predicted, the false negotiators made seven times as many references to external constraints as the sincere negotiators, 3 versus 0.4, on average, per negotiation; SDs were 3.5 and 0.97, respectively, t (17) = 2.26, p < .02 (one-tailed), d = 1.10. In particular, the false negotiators appealed to constraints such as their accountability to their superiors, investors, or shareholders, although the background information did not contain any indication of such constraints. Several examples are given in Table 3.
Escalation toward the end of the negotiations (Hypothesis 1f)
The company negotiators risked the possibility that the government negotiators would break off the negotiations and exercise their outside alternative. The false negotiators, who were especially vulnerable to this risk, had to be cautious not to escalate their demands, at least not until the later stages of the negotiations, when the risk of the government breaking off the negotiations was smallest. Indeed, the false and sincere company negotiators did not present different initial grant requests. However, the bids made toward the end were significantly higher for the false than for the sincere negotiators, in monetary values, 333.33 versus 231.25 million dollars; SDs were 86.60 and 128.0, respectively, t (15) = 1.95, p < .05 (one-tailed), d = 1.01, and in terms of the corresponding point payoffs, 157.22 versus 113.83 points, SDs were 42.58 and 52.67, t (16) = 1.92, p < .05 (one-tailed), d = 0.96.
Notwithstanding this result, the final proposals for the employment rules for the false and sincere negotiators were not statistically different. The false negotiators thus made extreme demands on the more important dimension (grant level), but not on the less important one (employment rules). 3 The false negotiators might have felt that placing high demands on the more important issue was sufficient to stall the agreement without appearing too stubborn or demanding. We suggest that while the false negotiators sought to fend off the possibility of an agreement, they were also careful to avoid the risk that the other party would break off the negotiations.
Use of representatives (Hypothesis 1g)
The company negotiators in both conditions were informed that only during the second mandatory session could they present themselves as representatives with limited power. Specifically, the instructions read, “During the second session you could use the services of a representative (e.g., a lawyer) who would be authorized to discuss all negotiation issues, but who would not be authorized to sign an agreement. Should you choose to exercise this option, you must notify the other party at the beginning of the session that you are playing the role of such a representative.”
We predicted that the false company negotiators would seize the opportunity to present themselves as representatives so as to hamper the negotiation process. As predicted, four of the nine false negotiators chose to use representatives in the second session, whereas none of the sincere negotiators did so; t (17) = 2.68, p < .01 (one-tailed), d = 1.30. Interestingly, several of the false negotiators justified this move to their partners by saying they were obliged to do so, although the instructions explicitly indicated that it was left to their own discretion.
Negotiation Processes: Cooperative Behavior
Cooperative talk (Hypothesis 2a)
Being generally uncooperative, the false negotiators compensated for this by making promises of cooperation, so as to deter suspicion and reduce the chances that their opponents would break off the negotiations and implement the exit alternative. To assess the frequency of such acts, all instant-messaging and e-mail communications were coded independently by two judges. In particular, they coded statements made by the company negotiators that highlighted the importance of cooperation and mutual benefits from an agreement or indicated their intention of reaching an agreement (e.g., “I believe we can overcome our differences,” “I am interested in an agreement,” and “I believe this solution would be good for both of us”). The coders agreed on 96 percent of the cases, and the remaining inconsistencies were resolved via discussion. Table 3 lists some detailed examples.
An index of “cooperative talk” was computed for each negotiator across all mandated sessions and additional electronic exchanges. As predicted, the false negotiators made more cooperative statements overall, 5.44 versus 2.20; SDs were 3.40 and 2.35, t (17) = 2.45, p < .02 (one-tailed), d = 1.19.
Responsiveness to opponents’ demands (Hypothesis 2b)
Recall that the false negotiators try to slow the negotiation process and refrained from discussing the core issues. We predicted that the governmental negotiators would respond to these obstacles by requesting the other side to invest more effort in the negotiations and initiate additional interactions beyond the three mandatory sessions. As predicted, the government negotiators who had been assigned false opponents tended to produce explicit, procedural demands, such as “I want you to send me a concrete offer by Monday” or “We should set up an additional meeting before our next scheduled meeting.” The frequency of requests from the other side to engage in additional e-mail interaction was greater in the false than in the sincere condition, 1 versus 0.1 per negotiation team, respectively; SDs were 1.41 and 0.32, t (17) = 1.97, p < .05 (one-tailed), d = 0.96; similarly, the respective frequencies of chat requests per negotiation team were 0.33 and 0; SDs were 0.5 and 0, t (17) = 2.12, p < .05 (one-tailed), d = 1.03. Interestingly, the false negotiators complied with the requests to communicate via the e-mail system in 100 percent of the cases, but did not comply at all with requests to carry on the interaction via the chat system, a finding that we discuss later.
Postnegotiation Questionnaires: Opponents’ Perceptions
Opponents’ understanding of the false negotiators’ interests (Hypothesis 3a)
Were the government negotiators who faced the false company negotiators aware of their partners’ goal of not reaching an agreement and their attempts to derail the negotiations? We asked the negotiators to indicate (on a series of 5-point scales) the extent to which they thought that their partner wanted “an agreement at any cost,” “an agreement with unrestricted employment rules,” “an agreement with restricted employment rules,” “an agreement with zero grant,” “an agreement with full grant,” “a balanced agreement,” or “no agreement at all.” The item of focal interest was the last one, which was embedded in a sequence to prevent it from being conspicuous. The government negotiators facing false partners and the ones facing sincere partners did not differ significantly in their answers to the focal item (2.22 vs. 2.00, respectively) and the effect size was small; SDs were 1.41 and 0.97, t (16) = 0.39, p > .35 (one-tailed), d = 0.19. In addition, when asked to rate the extent to which their partners stalled for time, the government negotiators who faced false or sincere partners did not differ in their ratings, 3.44 versus 3.00, respectively; SDs were 1.67 and 1.41, t (16) = .6, p > .27 (one-tailed), d = 0.3. In summary, contrary to the hypothesis, the current results do not support the notion that the government negotiators would be able to discern the true motives of the false negotiators.
Opponents’ attitudes to the false negotiators (Hypothesis 3b)
While the previous analysis provided no evidence that the government negotiators were aware of the different motives of the false and sincere company negotiators, we wondered whether the government negotiators felt differently about false and sincere negotiators, and might have discriminated between them on an affective level. To answer this question, we elicited the government negotiators’ feelings and attitudes toward the false and sincere opponents. We asked them to what extent they felt anger, frustration, or pleasantness, or to which they felt their partners to be stubborn, professional, responsive, flexible, and so on. Each item was rated on a 5-point scale. We found that on every scale (i.e., 14 of the 14), the false negotiators were rated (on average) more negatively (i.e., more frustrating, less professional, less forthcoming, less flexible, etc.) than were their sincere counterparts.
This highly consistent data pattern suggests preliminary evidence for implicit discrimination between false and sincere negotiators. 4 While there was no evidence that the government negotiators consciously understood the true motives of the false negotiators, they experienced the interaction with false and sincere negotiators differently at an affective level.
Discussion
Although the occurrence of false negotiations is of theoretical and practical importance, it seems to have escaped the attention of scholars and experimenters (see Anand, Feldman, and Schweitzer 2009). In this article, we have theorized about the motivations underlying false negotiations. Our main research focus has been the behavior of false negotiators. Our theoretical analysis and empirical findings suggest that false negotiators may engage in a balancing act. As they obstruct the process, they also make gestures that seemingly reflect good will to ensure that the other party will not walk away from the negotiating table. Like a virtuoso walking a tightrope, they face danger if they lean too much to one side or the other. While being cooperative could lead to an unintended agreement, showing too little cooperation could lead to termination of the negotiations.
We presented orderly findings about the tactics used by false negotiators and the challenges faced by their opponents. Thus, we predicted that the false negotiators in our study would use subtle tactics to obstruct the negotiations and at the same time would also make overt gestures demonstrating their (alleged) interest in an agreement, to prevent the other party from walking away from the negotiating table.
Consistent with this view, the false negotiators in our study used a range of tactics to stall the negotiations. In particular, they delayed their responses, avoided discussion of core issues, postponed concrete offers, and presented themselves as unauthorized representatives. These measures contributed to derailment of the negotiations. At the same time, the false negotiators cultivated a facade of cooperation. They tried to conceal their true intentions and created a smokescreen by making claims attesting to their cooperative intentions (e.g., “I believe we can overcome our differences,” and “I am interested in an agreement”). They also signaled their sincerity by reacting positively to their partners’ requests to engage in additional e-mail communications beyond the ones mandated by the official protocol. In addition, they were reluctant to cancel a meeting, which might have been perceived as an extreme measure, even though it would have shortened up the negotiation time.
Further insight into the mind-set and actions of the false negotiators can be gained by considering their opponents’ perspectives. To succeed in the negotiations, the opponents need to be able to identify false negotiators and tell them apart from the sincere ones, a task that may prove exceedingly difficult, for two reasons. First, the opponents may believe that their partners are interested in reaching an agreement (“or else why would they be negotiating”). More important, the opponents may have the valid belief that sincere negotiators often have mixed motives in that they display both cooperative behaviors (e.g., trust building and concession making) and competitive ones (e.g., playing tough, presenting high demands, and taking the risk of reaching an impasse; Schelling 1960; Hornstein 1965; Thompson 1990; Bazerman and Neale 1992; Carnevale and Pruitt 1992; Shapiro and Bies 1994; Weingart, Hyder, and Prietula 1996; De Dreu and McCusker 1997; Olekalns and Smith 1999; De Dreu and Carnevale 2003). As the literature has shown, negotiations, by their very nature, involve both competition and cooperation (De Dreu and Carnevale 2003), and both assertiveness and empathy (Mnookin, Peppet, and Tulumello 1996). False negotiators, through their mixed use of competitive and cooperative tactics, disguise their true motives behind the veil of mixed-motive negotiations. Thus, their opponents could readily mistake them for sincere mixed-motive negotiators. They could attribute the (dubious) actions of a false partner to what Schelling (1960) calls brinksmanship, that is, the art of walking on the edge, delaying the process to the last minute and risking an impasse.
Failing to recognize false negotiators for what they are until a late stage in the process could be unfortunate, especially as the sunk costs escalate and the alternative to an agreement is weakened or no longer available. All in all, it is ironic that the mixed-motive negotiation framework—which is considered by theorists (e.g., Mnookin, Peppet, and Tulumello 1996) to be a most effective approach—provides the veil behind which false negotiators manage to disguise their true intents not to reach an agreement at all. That being the case, it should not be surprising that the government negotiators in our study found it difficult to tell a false negotiator masquerading as a tough one from a sincere negotiator who was merely behaving competitively.
Indeed, our governmental negotiators were generally unaware of their partners’ intentions. Most of the governmental negotiators facing false negotiators carried on the negotiation to the end of the third and last mandated session (with only 22 percent choosing to break off the negotiations), even though the break off alternative was available all along. While reaching an agreement was generally superior to the break off alternative (for the government negotiator), choosing the break off alternative should still have been preferable to reaching an impasse at the end of the third session. The fact that most government negotiators did not break off the negotiations before the end is informative, implying that they had difficulty reading their partners’ intentions.
The postnegotiations questionnaires administered at the debriefing stage shed more light on this issue. We investigated the government negotiators’ understanding of their opponents’ goals as well as their feelings toward their opponents. When they were asked explicitly about the possibility that their opponents were not interested in an agreement at all, there was no difference between the government negotiators’ responses in the sincere and false conditions.
We proceeded to query the government negotiators about their attitudes to their opponents. Although we did not expect the government negotiators to fully understand their opponents’ true motives, we did expect them to report more negative affect and fewer favorable experiences with the false negotiators. The data pattern was consistent. The negotiators’ attitudes toward their opponents were less favorable in the false than in the sincere condition. The false negotiators seemed less professional and more stubborn, and also aroused more frustration, anger, and negative feelings. We find this pattern of consistent differences intriguing and worthy of further investigation.
What practical lessons and prescriptive recommendations can we draw from this research? One main lesson is the important recognition that negotiations can be used for other purposes than the presumed ones. By labeling the phenomenon, delineating the preconditions for it, and drawing attention to its consequences, we open an avenue for new research studying the realistic ramifications of false negotiations. As such, this article may serve to redirect researchers’ focus. A second lesson is that the way to protect oneself from falling prey to false negotiators is by recognizing them for what they are. We suggest that false negotiations should be viewed as a “syndrome”—a collection of symptomatic behaviors. While no single alarming behavior might be sufficient to label a negotiator as unequivocally “false,” an accumulation of alarming signs should be indicative. For example, one negotiator might delay his or her response because he or she is ill prepared (and not because the negotiator intends to stall the negotiations); another negotiator might behave stubbornly and harden his or her positions toward the end of the negotiations because he or she wants to extract additional gains; similarly, this negotiator could cancel a meeting due to truly unforeseen circumstances. Put differently, negotiators can be noncooperative for legitimate reasons. Moreover, sincere negotiators occasionally use “dirty tricks.” That sincere parties have such a range of behaviors in their repertoire makes it fairly difficult to distinguish false negotiators from sincere ones. However, the accumulation of alarming behaviors should raise people’s concerns about the intentions of their opponents. Consider a party that delays its responses, refers duties to unauthorized delegates, poses extreme demands, makes petty concessions, preaches abstract principles while avoiding the concrete parameters of the agreement; suppose further that the same party also replaces realistic generous gestures with hollow, high-level talk of the need for cooperation without showing much of it in reality. This list of symptomatic behaviors could be thought of a syndrome of false negotiations.
Our study also reveals some of the signs that could help people distinguish between false and sincere negotiators. Typical diagnostic signs include delaying responses or discussing abstract principles rather than core issues and concrete offers. However, the study also revealed some more subtle signs. First, when the option of delegating the negotiations to an unauthorized representative was made available, the false negotiators were most likely to seize it. Second, when invited by their opponents to engage in additional interactions via the e-mail system (beyond the three instant-messaging sessions mandated), the false negotiators complied, although they did not comply with similar requests to engage in additional interactions via the instant-messaging system. Presumably, the false negotiators were more comfortable with the asynchronous (e-mail) system than with the synchronous (instant-messaging) system. Most likely the asynchronous system, which does not demand online live interaction, allowed the false negotiators to maintain control over the pace and intensity of the interactions (cf. Nadler 2004b). These observations suggest that avoiding live interaction and using unauthorized representatives can be added to the list of alarming signs of the opponents’ true intentions.
In trying to defend themselves against false partners, negotiators can also rely on the practice of “active listening” advocated in the management and negotiations literatures (Gordon 1977). Active listening can help negotiators evaluate their opponents’ true intentions. An anecdote from our experiments serves to illustrate this point. One of the government negotiators completely dominated the interaction with her partner. She therefore failed to take notice of the fact that her partner (who was a false negotiator) had kept quiet and let the time slip by with few interruptions. Had she been less dominant she might have noticed this disturbing aspect of her partner’s behavior. Being attentive to the behavior of one’s partner and asking questions can be informative and can call attention to alarming signs.
Finally, we speculate that negotiators could capitalize on the (plausible) assumption that acting falsely takes a cognitive and emotional toll. Suppose false negotiators are anxious due to their concern about the risk of being exposed; suppose further that they feel uncomfortable about the ethicality of their actions. (Such concerns were raised informally in some of the end-of-study written comments volunteered by the participants in the false condition.) Suspicious parties could take advantage of the emotional toll experienced by their partners. Upon encountering signs of false negotiations, parties could ask pointed questions or mention the emotional, ethical, and reputational implications of misrepresentation, so as to increase the emotional cost of engaging in false behavior. Doing so could dissuade false parties from maintaining their course of action or induce them to give away more signs that are symptomatic of false negotiations. As these symptomatic cues accumulate, breaking off the negotiations becomes more warranted.
Limitations of the Study
Alongside these new findings, our study has some limitations, one of which is the use of electronic media for carrying out the negotiations. There is a well-known view that the medium shapes the message. In our case, negotiations conducted through electronic media differ in several ways from face-to-face negotiations (Drolet and Morris 2000; Paese, Schreiber, and Taylor 2003; Nadler 2004a, 2004b; Sheehy 2006; Thompson, Wang, and Gunia 2010). Extralinguistic cues, including body language, hand gestures, and voice intonation—all of which play a prominent role in interpersonal encounters, including negotiations—were not present in our experiment.
The use of electronic media confers numerous methodological advantages, however. It allows for convenient and reliable collection of rich data that are amenable to content analysis as well as structural analysis of aspects of the negotiations (e.g., response times). Importantly, the use of electronic negotiations also carries a theoretical message, as the Internet has become a major mode of communication in our social, business, and management settings (Sheehy 2006). The conclusions drawn from the research therefore have some ecological validity as they can be generalized to important facets of our present-day environment.
It is of interest and importance to investigate the conduct of false negotiations in face-to-face interactions as well. Here, presumably, false negotiators would need to control their body language so as not to give away their true intentions or their discomfort in misrepresenting their true goals. Such research should connect, for example, with the literature on people’s ability to detect liars (e.g., Frank and Ekman 1997).
Perspectives on Future Research
Finally, this study focused on the tactics and behaviors of false negotiators. We theorize that they engage in a balancing act. False negotiators are said to hide behind the veil of the mixed-motive framework—that is, they capitalize on their opponents’ tendency to view contentious tactics within the framework of mixed-motive negotiations and attribute them to a competitive orientation. This conceptual framework could be adopted, with appropriate modifications, in varied negotiations contexts, including interpersonal, business, political, and international settings.
In particular, it would be instructive to analyze political and international negotiations in terms of this outlook. The political and international spheres—where parties’ interests are complex and dynamic, and decision makers need to contend with external and internal pressures—are a natural setting for false negotiations. Consider various protracted international conflicts that have involved drawn-out, intensive negotiations, and have met limited or no success. Might it be the case that the actors have fallen prey to false negotiations? For example, the negotiations between Iran and the International Atomic Energy Agency (IAEA) and Security Council permanent members on Tehran’s nuclear program have met little success so far. The IAEA has indeed expressed concerns about the pace of the negotiations with Iran and the sincerity of the partners (IAEA Press Office 2013; Amano 2013). The current Iranian president Hassan Rouhani had indeed noted in the past that “while we were talking with the Europeans in Tehran, we were installing equipment in parts of the facility in Isfahan…In fact, by creating a calm environment, we were able to complete the work on Isfahan” (Sciolino 2006, Vick 2012). A closer inspection of the negotiation process indicates some alarming signs, such as stalling for time, maintaining formal rather than substantive dialogue, suggesting (seemingly) inventive constraints, and high-level declarations on cooperation.
The present research has focused on the behavior of parties who engage in false negotiations. In view of the scope and relevance of this phenomenon, we suggest that future research should focus more on the parties’ reactions to false negotiators and the conditions under which they are liable to fall prey to misrepresentation.
Footnotes
Authors’ Note
The Program on Negotiation at Harvard Law School awarded Netta Barak-Corren the 2013 Howard Raiffa Doctoral Student Paper Award for this article. Edy Glozman and Netta Barak-Corren contributed to the project equally. They are former students of the Hebrew University. Ilan Yaniv is on the faculty of the Department of Psychology and a member of the Center for the Study of Rationality at the Hebrew University of Jerusalem, Israel. Please address correspondence to any of the authors. Emails:
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Ilan Yaniv was supported by Grant 327/10 from the Israel Science Foundation.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
