1-2 pages Article Summaries

POWER DYNAMICS IN NEGOTIATION PETER H. KIM University of Southern California ROBIN L. PINKLEY Southern Methodist University ALISON R. FRAGALE Stanford University Power is widely acknowledged to affect negotiator performance. Yet few efforts have been made to integrate the most prominent theories of power into a cohesive frame- work that can account for the results from a broad array of negotiation-relevant research. We address this limitation by proposing a dynamic integrative model that decouples power into four components: (1) potential power, (2) perceived power, (3) power tactics, and (4) realized power. Implications, propositions, and future directions are discussed. Negotiation has been defined as “an interper- sonal decision-making process by which two or more people agree how to allocate scarce re- sources” (Thompson, 2000: 2). As such, many view it as a central aspect of organizational life.

The growing complexities of work relationships, the increased reliance on teams to make deci- sions, and the rise of new organizational forms have placed unprecedented pressure on manag- ers to become effective negotiators. Thus, schol- ars and practitioners alike have focused on identifying the mechanisms that can improve negotiator performance (Lewicki, Saunders, & Minton, 1999).

One factor that is widely acknowledged to affect negotiator performance is power (e.g., Bacharach & Lawler, 1981). A negotiator’s power may be critical for the quality of his or her suc- cess, because it can determine the allocation of rewards in an agreement (Kim, 1997; Mannix, 1993a; Pinkley, Neale, & Bennett, 1994). The greater one’s power relative to the power of oth- ers, the more resources one should be able to claim. As a result, the negotiation literature is replete with recommendations to improve one’s power position (e.g., Thompson, 2000).Yet those who wish to improve their power should realize that power may be influenced by a wide range of factors, and they are likely to be overwhelmed by the varied and often inconsis- tent lists of characteristics, properties, strate- gies, and descriptions that have been offered with regard to this topic (Mooney, 1984). Al- though researchers have made many efforts to provide comprehensive, research-based frame- works on power (e.g., Bacharach & Lawler, 1980; Emerson, 1962; Weber, 1947), much work remains if we are to integrate these approaches into a cohesive framework that can account for the re- sults from a broad array of negotiation-relevant research. Such a model should offer a clearly specified conceptual framework that describes what power is, where power comes from, how power is perceived, and how power can be used or changed. In this article we attempt to address these issues by proposing a dynamic model of negotiator power.

Our treatment of negotiator power begins with a review of some prominent theories of organizational power and a discussion of the potential benefits and limitations if we were to apply each to the context of negotiation. Next, we propose a dynamic model of negotiator power that integrates and extends these seem- ingly disparate theories to retain their respec- tive benefits and address their respective limi- tations. This dynamic model decouples power into four distinct components—(1) potential We thank Paul Adler, Jay Kim, Shirli Kopelman, Claus Langfred, Julia Liebeskind, Kyle Mayer, Andrew Molinsky, Steven Postrel, and Jeffrey Sanchez-Burks for their assis- tance with this paper.

Academy of Management Review 2005, Vol. 30, No. 4, 799–822. 799 power (the underlying capacity of negotiators to obtain benefits from their agreement), (2) per- ceived power (negotiators’ assessments of each party’s potential power), (3) power tactics (be- haviors designed to “use” or “change” the power relationship), and (4) realized power (the extent to which negotiators have claimed benefits from the interaction)—to enhance its generalizability to noisy, real-world environments where power- related information may be incomplete, inaccu- rate, or asymmetrical. Finally, throughout this analysis, we propose specific propositions that can be tested in future empirical research.

PROMINENT CONCEPTIONS OF POWER Most conceptions of power are founded on Weber’s (1947) classic definition of power as the probability that a person can carry out his or her own will despite resistance. The majority of the- orists who have written about power would ex- press agreement with this broad definition (Bacharach & Lawler, 1980). However, there are many subtle and not-so-subtle differences among their various perspectives. Arguably, the most commonly referenced appraisals of power are (1) French and Raven’s (1959) five bases of power; (2) Kipnis, Schmidt, and Wilkinson’s (1980) typology of influence tactics; and (3) power- dependence theory, first proposed by Emerson (1962). We begin by briefly reviewing the major characteristics of these frameworks.

French and Raven (1959) suggest that there are five bases of power, which stem from differ- ent aspects of the relationship between an actor and the target of his influence attempts (see Table 1). Specifically, they argue that A’s power over B is determined by (1) A’s ability to provide benefits to B (reward power), (2) A’s ability to punish B if B does not comply with A’s wishes (coercive power), (3) A’s possession of special knowledge or expertise (expert power), (4) A’s legitimate right to prescribe behavior for B (le- gitimate power), and (5) the extent to which B identifies with A (referent power). Raven’s (1974) later versions of this typology add information (i.e., knowledge regarding the inner workings of the organization or the relation of the organiza- tion to the environment) as a sixth base of power. Moreover, other scholars have distin- guished these bases of power, which determine one’s ability to manipulate the behavior of oth- ers, from the sources of power (e.g., personalcharacteristics), which determine how we come to control these power bases (Bacharach & Lawler, 1980).

Kipnis et al. (1980), with subsequent exten- sions by Kipnis and Schmidt (1983) and Yukl and Tracey (1992), in contrast, focus on identifying and categorizing the tactics commonly used by managers when attempting to get others to com- ply with a request. Taken together, these pro- grams of research identify nine dimensions of influence—pressure, legitimation, exchange, co- alition, ingratiation, rational persuasion, inspi- rational appeal, consultation, and personal ap- peal (see Table 2)—and consider how one’s power relationship with others can influence the likelihood that these different influence tactics will be used. Yukl and Tracey (1992), for exam- ple, found that inspirational appeal, ingratia- tion, and pressure were used most in a down- ward direction; that personal appeal, exchange, and legitimation were used most in a lateral direction; that coalition was used most in lateral and upward directions; and that rational per- suasion was used most in an upward direction.

They also found that uses of rational persua- sion, inspirational appeal, and consultation were most effective and that uses of pressure, coalition, and legitimation were least effective for exerting influence in organizations.TABLE 1 French and Raven’s (1959) Typology of Power Bases Power Base Description Reward power A’s power over B is a function of how much B can be rewarded and the extent to which B believes that A controls these rewards.

Coercive power A’s power over B is a function of how much B can be punished by B and the extent to which B believes that this punishment can be avoided if B complies with A’s wishes.

Expert power A’s power over B is a function of B’s perception that A possesses some special knowledge or expertise.

Legitimate power A’s power over B is a function of how much B believes that A has the lawful authority to influence B.

Referent power A’s power over B is a function of how attracted B is to A and, thus, how much A can influence B’s feelings of personal acceptance, approval, and self-esteem. 800October Academy of Management Review Finally, power-dependence theory (Blau, 1964; Emerson, 1962) provides a framework for concep- tualizing relative and total power. More specif- ically, Emerson states, “The power of A over B is equal to and based upon the dependence of B upon A” (1962: 32–33). Dependence, in turn, is based on two dimensions: (1) it is directly pro- portional to the value attributed to the outcome at stake, and (2) it is inversely proportional to the availability of this outcome through alternative sources. Thus, A’s power over B is directly re- lated to the degree to which B is dependent on A (i.e., the extent to which B receives greater ben- efit from the relationship with A than B can get from alternative relationships). Similarly, B’s power over A depends on the degree to which A receives greater benefit from the relationship with B than A can get from alternative relation-ships. Furthermore, because this framework considers power to be non-zero sum in nature, an increase in A’s power does not necessarily decrease B’s power, and vice versa. Each party’s power is independently determined by the oth- er’s dependence. Thus, power dependence dis- tinguishes relative power from the total power (i.e., the sum of each party’s power) in the rela- tionship (Lawler & Bacharach, 1987).

Although each of these frameworks can in- form our study of power, none of them is com- prehensive enough to account for the others or to guide us in the dynamic context of negotiation.

French and Raven (1959), for example, discuss thebasesupon which power may be derived, but they pay little attention to how these power bases determine one’s power relations with oth- ers or the implications of these power relations for the tactics that a powerholder might choose.

Power-dependence theory conceptualizes rela- tive and total power quite clearly, but it pays little attention to how one’svaluationof a rela- tionship or the valuation of one’s alternative relationship is determined, or how one’s power is likely to be used. Finally, Kipnis and col- leagues (1980) discuss power use, but they focus on what people do onlyaftera power relation- ship has been determined, without considering its antecedents or relational determinants.

Moreover, and most important for our present discussion, these theories were not necessarily intended to capture the specific kinds of power dynamics one might find in the negotiation con- text. Thus, none of these theories has ade- quately emphasized the interactive dynamics inherent in negotiation, the role of individuals’ power perceptions, or the ways in which power relationships can change over time, all of which need to be incorporated into any comprehensive model of power in negotiation. As a conse- quence, prior power frameworks provide many important conceptual notions regarding power as a construct but do not fully explicate what power is, where power comes from, how power is perceived, or the ways in which power can be used or changed.

A DYNAMIC MODEL OF NEGOTIATOR POWER Below we build a model that combines potential power, power in use, and realized power in the negotiation context (see Figure 1). In general, we argue that potential power is determined by the TABLE 2 Yukl and Tracey’s (1992) Typology of Influence Tactics Influence Tactic Description Pressure The actor uses demands, threats, or intimidation to increase the target’s compliance.

Legitimation The actor attempts to legitimize a request for compliance or claim the right to request it by referencing existing policy or tradition.

Exchange The actor uses implicit or explicit promises to reciprocate if the target complies.

Coalition The actor obtains the aid of others to help persuade or pressure the target to comply.

Ingratiation The actor attempts to make a favorable impression and improve the target’s mood before requesting compliance.

Rational persuasion The actor uses logical arguments and factual information to support the viability of complying with the request.

Inspirational appeal The actor makes an emotional appeal for compliance by appealing to the target’s values and ideals.

Consultation The actor seeks the target’s participation in the decision- making process and the implementation of the request.

Personal appeal The actor appeals to the target’s sense of loyalty or friendship before requesting compliance. 2005801 Kim, Pinkley, and Fragale FIGURE 1 A Dynamic Model of Negotiator Power 802October Academy of Management Review two-dimensional mutual dependence configura- tion developed in power-dependence theory and that French and Raven’s (1959) typology of power bases can help inform how these dimensions are derived. We further contend that negotiators’ perceptions of these dimensions may diverge from their potential values in a variety of ways owing to imperfect information and negotiators’ bounded rationality. As a result, we propose that power perceptions will drive tactical deci- sions, which can influence negotiators’ mutual dependence and mediate the relationship be- tween potential and realized power. In doing so, we consider a number of power-use and power- change tactics and their implications for the loss and accumulation of power. Finally, we ar- gue that the extent to which negotiators realize power from the focal interaction will affect their potential power in future interactions. Taken to- gether, these arguments emphasize the dynamic nature of power relations and highlight numer- ous implications for those seeking to manage such relations effectively in negotiation.

Potential Power Potential power is defined as the underlying capacity of negotiators to obtain benefits from their agreement. Thus, we follow the guidance of Bierstedt (1950), Wrong (1968), and other re- searchers who distinguish between potential power and its use. Consistent with power- dependence theory, we argue that a negotiator’s potential power is a function of the counterpart’s dependence and that dependence is, in turn, based on two dimensions: (1) the counterpart’s valuation of the negotiation and (2) the value of the counterpart’s best alternative if an agree- ment in the focal negotiation is not reached.

This conceptualization of potential power al- lows us to incorporate prior negotiation re- search, which has typically operationalized power in terms of power-dependence theory’s two dimensions. Negotiation researchers typi- cally have altered the power relationship be- tween negotiators by manipulating either (1) the benefits negotiators can bring to the bargaining table (i.e., their contributions) and, hence, their counterparts’ valuations of the negotiation (Kim, 1997; Mannix, 1993a) or (2) the value of negotia- tors’ best alternatives to the negotiated agree- ment (i.e., their BATNAs; Pinkley et al., 1994).We can, furthermore, draw on French and Raven’s (1959) typology of power bases to under- stand how these valuations of the negotiation and its best alternatives may be formed. The power-dependence framework is based on the principles of social exchange (Blau, 1964), which are, in turn, rooted in economic theory. Thus, the values a negotiator attributes to these two di- mensions of potential power refer to the eco- nomic notion of utility—the general psychologi- cal assessment of the benefit the negotiator would obtain from each of these options (i.e., the negotiation and its best alternative). Although each negotiator is likely to derive these utilities differently, owing to the idiosyncratic nature of their preferences, the French and Raven typol- ogy identifies at least some of the reasons why such utility may be derived (Bacharach & Lawler, 1980).

Specifically, we argue that negotiators’ valu- ations of a negotiation are determined by their assessments of its implications for such factors as rewards, punishments, knowledge, legiti- macy, and identification (e.g., its potential to offer more financial gains, entail fewer legal liabilities, provide more useful information, etc.). Likewise, we contend that negotiators’ val- uations of their alternatives are based on their assessments of whether, and the degree to which, these benefits could be obtained through other means. Thus, rather than having to dis- cuss the implications of these myriad bases of power individually, this conceptualization al- lows us to pursue a higher level of analysis that can encompass these power bases, integrate them into a coherent framework, and thereby present a more parsimonious depiction of power relations.

Limitation.The problem with this conceptual- ization of potential power, however, is that it is ultimately hindered by power-dependence the- ory’s inherent lack of specification. Based on Emerson (1962) and Blau’s (1964) discussions of power dependence, one might infer a variety of relationships between its two dimensions—the value of the relationship and the value of its best alternative (e.g., an additive, multiplica- tive, or more elaborate function). More formally, for any two negotiators A and B, if we use U a(b) to depict A’s valuation of the negotiation with B, U a(alt) to depict A’s valuation of his or her best alternative to that negotiation, U b(a) to depict B’s valuation of the negotiation with A, and U b(alt) to 2005803 Kim, Pinkley, and Fragale depict B’s valuation of his or her best alternative to that negotiation, these discussions of power dependence at best suggest that we can depict each negotiator’s potential power with the fol- lowing equations:

A’s Potential Power B’s Dependence f (Ub(a) ,U b(alt) ) B’s Potential Power A’s Dependence f (Ua(b) ,U a(alt) ) where B’s dependence is positively and mono- tonically related to U b(a) and negatively and monotonically related to U b(alt) , and A’s depen- dence is positively and monotonically related to U a(b) and negatively and monotonically related to U a(alt) .

That is, if B’s Dependence D band A’s Dependence D a then D b U b(a) 0 D b U b(alt) 0 D a U a(b) 0 D a U a(alt) 0 As one might imagine, the enormous range of possible relationships encompassed by this for- mulation severely limits this theory’s predictive capability (i.e., given that it is unclear which of these relationships to use for these calcula- tions). For this reason, at least one subsequent treatment of power-dependence theory that has gained prominence has been more specific, by positing the relationship between its two dimen- sions explicitly as a ratio (Bacharach & Lawler, 1980: 147). Specifically, these researchers con- tend that A’s Potential Power B’s Dependence U b(a) Ub(alt) B’s Potential Power A’s Dependence U a(b) Ua(alt) Our review of the power literature reveals, how- ever, that few efforts have been made to assess this claim, let alone present a competing formu- lation. We next attempt to address these limita- tions to develop a more refined and accurate conceptualization of potential power in negoti- ation.

Power dependence revisited.Power-depen- dence theory suggests that parties will possess the same levels of dependence, and hence po- tential power, when each party’s preference for the relationship relative to its alternative is the same. Thus, Bacharach and Lawler’s (1980) de- piction of dependence as a ratio of its two di- mensions would imply that any two negotiators, A and B, will have equal potential power when A’s Potential Power B’s Dependence U b(a) Ub(alt) U a(b) Ua(alt) A’s Dependence B’s Potential Power The details of such a scenario can be communi- cated through a 2 2 matrix for each pair of actors in a negotiation (see Figure 2a). We can observe that any two negotiators possess the same potential power when each negotiator’s valuation of the relationship and each negotia- tor’s valuation of his or her best alternative are the same (see Figure 2b). However, this formu- lation also suggests that negotiators can pos- sess the same potential power when they differ on their valuations of these two dimensions, so long as the ratio between one’s valuation of the relationship and one’s valuation of alternatives is the same for each negotiator (see Figure 2c).

Thus, by this logic, the two parties depicted in Figure 2c should be equally capable of receiv- ing benefits from the negotiation.

Although this conclusion may sound plausi- ble at first glance, research on the implications of a negotiator’s BATNA suggests that this is an untenable solution (e.g., Pinkley et al., 1994).

This research suggests that negotiators would not accept an agreement that would make them worse off than they would be with their BATNAs.

Yet, if we consider the values depicted in Figure 2c and assume that negotiators’ contributions (i.e., the benefits each negotiator can bring to the negotiation table) may be added to deter- mine the total value of the potential agreement (U total 110 220 330), the equal distribution 804October Academy of Management Review of these benefits (i.e., the distribution that would be implied if both negotiators are viewed as equally powerful) would mean that A’s share would exceed his or her BATNA by 65 points (i.e., 165 100), but B’s share would fall short of his or her BATNA by 35 points (i.e., 165 200). Since B is unlikely to accept an agreement with A that would make him/her worse off than just accept-ing his/her BATNA (Pinkley et al., 1994), this dis- tribution should result in an impasse, despite the fact that U total was large enough to allow both parties to obtain benefits that would have exceeded their BATNAs. Thus, the assumption that the two negotiators depicted in Figure 2c possess equal potential power, and therefore should obtain equal outcomes from the negoti- (a) Generalized Formulation of Potential Power Ratio formulation: A’s Potential Power B’s Dependence U b(a) /Ub(alt) ; B’s Potential Power A’s Dependence U a(b) / U a(alt) ; Total Power U b(a) /Ub(alt) U a(b) /Ua(alt) .

Revised two-stage formulation: A’s Potential Power U a(alt) (U b(a) U b(alt) ); B’s Potential Power U b(alt) (U a(b) U a(alt) ); Total Power U a(alt) (U b(a) U b(alt) ) U b(alt) (U a(b) U a(alt) ).(b) Negotiators with Equal Valuations of Dimensions Ratio formulation: A’s Potential Power B’s Dependence 110/100 1.1; B’s Potential Power A’s Dependence 110/100 1.1; Total Power 1.1 1.1 2.2.

Revised two-stage formulation: A’s Potential Power 100 (110 100) 110; B’s Potential Power 100 (110 100) 110; Total Power 110 110 220. FIGURE 2 Potential Power (c) Negotiators with Unequal Valuations of Dimensions Ratio formulation: A’s Potential Power B’s Dependence 220/200 1.1; B’s Potential Power A’s Dependence 110/100 1.1; Total Power 1.1 1.1 2.2.

Revised two-stage formulation: A’s Potential Power 100 (220 200) 120; B’s Potential Power 200 (110 100) 210; Total Power 120 210 330.(d) Considerations Underlying Each Dimension Q quantity amount of resources; P probability likelihood of obtaining resources; W weight importance of resources. 2005805 Kim, Pinkley, and Fragale ation, is unlikely to hold true. This suggests that an alternative formulation (i.e., something other than the depiction of dependence as a simple ratio of its two dimensions) is needed to deter- mine the potential power of these two negotia- tors. We propose such an alternative formula- tion below.

One conclusion we can draw, given the notion that negotiators must obtain benefits that are at least equivalent to their BATNAs if an agree- ment is to be reached, is that A must obtain at least U a(alt) or 100 points and B must obtain at least U b(alt) or 200 points of utility from their negotiation. However, once the potential bene- fits from the association exceed the value of a negotiator’s BATNA, the threat of departure (i.e., utilizing this BATNA by walking away from the negotiation rather than reaching an agreement) loses its credibility, given that such a departure would now reduce the benefits received by the negotiator. Thus, whereas negotiators’ BATNAs establish the minimum level of benefits they have the potential to obtain, the potential to acquire benefits that remain after this criterion is satisfied (i.e., U total (U a(alt) U b(alt) )or330 (100 200) 30), which we call “surplus utility,” is likely to be determined in other ways. Indeed, research suggests that once the benefits from a negotiation exceed the values of negotiators’ BATNAs, negotiators tend to distribute rewards in direct proportion to their contributions (Kim, 1997; Kim & Fragale, 2005; Mannix, 1993a).

Negotiator contributions affect the potential to acquire this surplus utility via the threat that these contributions might be withdrawn if they are inadequately reciprocated (Blau, 1964). In- deed, evidence indicates that negotiators tend to reciprocate such benefits explicitly and to the degree that they have been provided by their counterpart (Smith, Pruitt, & Carnevale, 1982; Thompson, 1990). Since a negotiator can reduce these benefits up to the point where they barely exceed the counterpart’s BATNA, while still making it worthwhile for that counterpart to re- main in the negotiation, the benefits a negotia- tor can contribute beyond the value of the coun- terpart’s BATNA should affect the negotiator’s potential to obtain this surplus utility. In Figure 2c we can determine the benefits A can contrib- ute to the negotiation beyond the value of B’s BATNA, and hence A’s potential to obtain sur- plus utility, by subtracting the value of B’s BATNA from the utility he or she associates withthe negotiation (i.e., U b(a) Ub (alt) or 220 200 20). Similarly, we can determine the benefits B can contribute to the negotiation beyond the value of A’s BATNA, and hence B’s potential to obtain surplus utility, by subtracting the value of A’s BATNA from the utility he or she associ- ates with the negotiation (i.e., U a(b) U a(alt) or 110 100 10).

Thus, whereas Bacharach and Lawler (1980) depict potential power as a simple ratio of its two dimensions, our analysis suggests that po- tential power may be more accurately assessed through a two-stage process whereby negotia- tors’ BATNAs determine the potential to obtain presurplus utility and negotiators’ contributions that are in excess of their counterparts’ BATNAs determine the potential to acquire the surplus.

We can, therefore, calculate A’s potential power by combining the utility A could obtain from his or BATNA with the surplus utility A could obtain from his or her contribution (i.e., U a(alt) (U b(a) U b(alt) )or100 (220 200) 120). In the same vein, we can calculate B’s potential power by combining the utility B could obtain from his or her BATNA with the surplus utility B could ob- tain from his or her contribution (i.e., U b(alt) (U a(b) U a(alt) )or200 (110 100) 210). These results also indicate that B’s potential power would be substantially larger than A’s, since B would have the potential to obtain 210 units, or 63.6 percent of the total value of the agreement (U total 330), versus A’s 120 units, or 36.4 percent of U total . Our analysis, therefore, offers predic- tions that differ dramatically from Bacharach and Lawler’s (1980) depiction of dependence as a ratio, which suggests that the potential power of A and B would be the same.

Proposition 1: Potential power is de- rived through a two-stage process whereby negotiators’ BATNAs deter- mine their potential to obtain pre- surplus utility and negotiators’ contri- butions that are in excess of their counterparts’ BATNAs determine their potential to obtain surplus utility.

We should emphasize, however, that the real- ization of this potential power is not automatic but, rather, depends on negotiators’ perceptions of this potential (i.e., perceived power), their ef- forts to change these perceptions (i.e., power- change tactics), and both the manner and extent to which they attempt to extract benefits from 806October Academy of Management Review the negotiation (i.e., power-use tactics). Each of these factors, furthermore, should exert a dis- tinct set of influences on the 2 2 dependence matrix depicted in Figure 2a (i.e., affect potential power). The values depicted in this potential power matrix are, therefore, not fixed; they are likely to be altered at multiple stages through- out a negotiation. Thus, we will discuss how these changes can occur.

Perceived Power We define perceived power as negotiators’ as- sessments of a party’s potential power in the relationship. Thus, each negotiator is presumed to develop perceptions of his or her own poten- tial power and the potential power of his or her counterpart. To the extent that a negotiator ob- tains information about the BATNAs and contri- butions of the negotiating parties, his or her potential power should positively influence each party’s perception of the negotiator’s po- tential power in the interaction. This effect of potential power and perceived power is de- picted by the arrows leading from potential power to A and B’s perceptions of potential power in Figure 1.

Proposition 2: A negotiator’s potential power will positively influence each party’s perception of that negotiator’s potential power in the relationship.

However, given substantial evidence that peo- ple actively work to interpret and construe their social environments (e.g., Kim, Diekmann, & Tenbrunsel, 2003; Neale & Bazerman, 1991), ne- gotiators’ perceptions of each party’s BATNA and contribution may also depend on the situa- tion. Although the number of ways in which these perceptions may be influenced is likely to be substantial and their comprehensive por- trayal is beyond the scope of this paper, we can attempt to classify these influences by consider- ing how perceptions of negotiators’ BATNAs and contributions are assessed in greater detail.

Specifically, we argue that the values negotia- tors attribute to their BATNAs and contributions are based on considerations of quantity (i.e., the amount of resources), probability (i.e., the likeli- hood of obtaining those resources), and weight (i.e., the importance of those resources; see Fig- ure 2d), and that inferences regarding each of these criteria can affect the link between po-tential and perceived power, as we describe below. 1 Quantity.In general, the objective amount of resources available from a BATNA or contribu- tion should positively influence a negotiator’s valuation of that dimension. However, there are many occasions where negotiators have incom- plete information about the resources these BATNAs and contributions would provide. Nego- tiators are often instructed not to reveal their BATNAs (Lax & Sebinius, 1986), for example, to limit their counterparts’ knowledge of the re- sources afforded by that dimension (e.g., a job candidate may not tell a potential employer about the salary offered by another employer).

Negotiators may also fail to evaluate their own BATNAs carefully and, hence, may possess only a limited sense of the resources that option would provide (e.g., a job candidate may agree to a salary level with an employer without knowing the salary that could be obtained from a comparable employer). Similarly, negotiators may be relative strangers and, therefore, pos- sess incomplete knowledge of the resources rep- resented by their counterpart’s potential contri- bution (e.g., an employer may have little knowledge of the technical expertise that a job candidate would bring to the job, if hired). This lack of familiarity may also prevent negotiators from conveying details of their own contribution (e.g., a job candidate may not realize that his or her knowledge of foreign languages could be seen as an asset and, hence, may fail to men- tion this competency on his/her application).

Thus, knowledge about these resource quan- tities should strengthen the relationship be- tween the potential and perceived values of these dimensions.

Proposition 3: Knowledge of the quan- tity of resources afforded by a BATNA or contribution will positively influ- ence the link between the potential and perceived values of these dimen- sions.

1These considerations of quantity, probability, and weight correspond to the recognition in the economics liter- ature that value is determined, respectively, by the supply of goods— that is, an objective component; the assessed like- lihood of obtaining these goods—that is, an expected com- ponent; and one’s psychological assessment of these goods—that is, a subjective component (Nicholson, 1989). 2005807 Kim, Pinkley, and Fragale Lack of information about the objective quan- tity of resources offered by these dimensions, in contrast, may cause perceived power to diverge from potential power in systematic ways. Re- search on the false consensus effect, for exam- ple, suggests that people tend to believe that their own views are widely shared and that this effect is particularly likely when there is greater latitude for subjective construal (Gilovich, 1990).

Thus, if a negotiator is unaware of the quantity of resources afforded by the counterpart’s BATNA (e.g., a job candidate does not know who else is being interviewed by the employer) but considers this party’s contribution to be strong (e.g., there would be substantial benefits af- forded by this employment opportunity), the fo- cal negotiator is likely to assume that others will be interested in reaching an agreement with this counterpart and, hence, attribute higher value to the counterpart’s BATNA (e.g., the job candidate is likely to think that the em- ployer has several other qualified applicants interested in the position) than if the counter- part’s contribution were weak.

Along similar lines, research suggests that since items that are difficult to obtain are often better than those that are easy to obtain, people often infer the value of a good based on its scarcity (Cialdini, 1984). Thus, if a negotiator is unaware of the quantity of resources afforded by the counterpart’s contribution (e.g., a job candi- date knows little about the benefits afforded by a given employment opportunity) but knows that the counterpart’s BATNA is strong (e.g., there have been many highly qualified applicants for this position), the focal negotiator is likely to infer that it will be harder to reach an agree- ment with that counterpart (i.e., infer greater scarcity) and, hence, attribute higher value to the counterpart’s contribution (e.g., the job can- didate is likely to assume there will be great employment benefits) than if that counterpart’s BATNA were weak. These considerations sug- gest that, in the absence of perfect information about the quantity of resources provided by a party’s contribution or BATNA, respectively, the perceived value of a negotiator’s BATNA will positively influence the perceived value of that party’s contribution, and vice versa.

Proposition 4: The perceived value of a negotiator’s BATNA (contribution) will positively influence the perceivedvalue of that party’s contribution (BATNA).

Probability.The values of these BATNAs and contributions should be determined, however, not only by the objective quantity of resources they have the potential to afford but also by the likelihood that these resources will be obtained.

This notion underlies the economic principle of “expected value,” which assesses the value of an uncertain outcome by multiplying the value of an outcome by the probability that outcome will occur (Nicholson, 1989). The expected value for a BATNA or contribution that may provide $100,000 but only has a 50 percent chance of doing so, for example, would be $50,000 (i.e., 0.50 100,000).

There are many occasions, however, in which the probabilities for such outcomes are rela- tively unknown. These ambiguities can arise, for example, because of uncertainties in the en- vironment, lack of familiarity, and the difficulty of writing perfect contracts (Williamson, 1975).

For this reason, knowledge about these proba- bilities should strengthen the relationship be- tween the potential and perceived values of these dimensions.

Proposition 5: Knowledge of the likeli- hood resources will be obtained from a BATNA or contribution will posi- tively influence the link between the potential and perceived values of these dimensions.

Lack of information about these probabilities, in contrast, may lead the potential and per- ceived values of these dimensions to diverge in systematic ways. Specifically, we suggest that one important factor that may influence the in- ferences negotiators make about these probabil- ities is the amount of trust they have in their counterparts. Negotiators with low trust in their counterparts should believe that there is a higher likelihood, compared to those with high trust, that their counterparts will fail to contrib- ute the resources that were promised or that their counterparts may have artificially inflated the resources that would be afforded by the counterparts’ BATNAs. For this reason, a lack of trust is likely to lead negotiators to lower their assessments of their counterparts’ contributions or BATNAs, respectively, even in cases where 808October Academy of Management Review these suspicions may be unfounded (e.g., Kim, Ferrin, Cooper, & Dirks, 2004).

For example, a skeptical consumer purchas- ing a used car may question the seller’s claim that the car is reliable and therefore perceive the salesman’s contribution to be relatively low, whereas a more trusting consumer, who fully believes the salesman’s claim, may perceive the salesman’s contribution to be relatively high.

Thus, in the absence of perfect information about the likelihood of obtaining resources from a negotiation or its best alternative, respec- tively, the level of trust negotiators place in their counterparts should positively influence the perceived value of their counterparts’ BATNAs and contributions.

Proposition 6: The level of trust nego- tiators place in their counterparts will positively influence the perceived value of their counterparts’ BATNAs and contributions.

Weight.Finally, the values of negotiators’ BATNAs and contributions should be deter- mined not only by the objective quantity of re- sources they entail and the probability of receiv- ing those resources but also by the subjective importance of those resources for the negotiator.

The negotiations literature distinguishes the quantity of resources from the importance of those resources, in particular, by discussing the difference between negotiator interests and pri- orities (Lax & Sebinius, 1986). Whereas a nego- tiator’s interests concern any resource that would offer a benefit to the negotiator, a nego- tiator’s priorities concern which of these re- sources would provide a larger benefit than oth- ers. Thus, negotiators may offer substantial concessions on an issue and a reliable guaran- tee that those benefits would be provided and yet fail to increase the value of their contribu- tion significantly if this issue is of minor impor- tance to the counterpart (e.g., offering a business owner exclusive access to a proprietary technol- ogy and guaranteeing this offer with an enforce- able contract may do little to improve one’s con- tribution if this technology fails to serve the business owner’s primary strategic goals).

To avoid such mistakes, negotiators should seek to identify the importance each party places on the issues under discussion. However, false assumptions (e.g., the fixed-pie bias), mis- communication, and a variety of other factorsmay prevent the importance each negotiator places on these issues from being completely known. For this reason, knowledge about these weights should strengthen the relationship be- tween the potential and perceived values of these dimensions.

Proposition 7: Knowledge of the im- portance of resources offered by a BATNA or contribution will positively influence the link between the po- tential and perceived values of these dimensions.

Lack of information about these weights, in contrast, may lead the potential and perceived values of these dimensions to diverge in system- atic ways. Specifically, we suggest that one im- portant factor that may affect the importance negotiators place on the resources afforded by these BATNAs and contributions may be their focus of attention. This focus is significant be- cause evidence suggests that our evaluations of a given option can be directly influenced by its salience (e.g., Ensari & Miller, 2002). Thus, any factor that increases the salience of these BATNAs or contributions should increase the as- sessed importance of the resources afforded by these respective dimensions. Negotiators who consider an impasse to be likely, for example, should focus greater attention on their BATNAs (given their expectation that these BATNAs will need to be used) and, hence, weigh the re- sources afforded by these BATNAs more heavily in their subjective appraisals than the resources afforded by their counterparts’ contributions. In contrast, negotiators who expect to reach an agreement should consider the use of their BATNAs to be less likely and, hence, weigh the resources afforded by the counterparts’ contri- bution more heavily in their subjective apprais- als than the resources afforded by their BATNAs.

Research suggests, furthermore, that a key factor that affects the likelihood of reaching an agreement is the amount of surplus utility avail- able in the negotiation (i.e., the amount of re- sources remaining after all parties have claimed resources equivalent to their BATNAs).

The greater the surplus utility, the more oppor- tunities that exist for negotiators to reach an agreement and, thus, the greater the likelihood that the negotiating parties will avoid an im- passe (Lax & Sebinius, 1986). Thus, we expect factors that affect the likelihood of reaching an 2005809 Kim, Pinkley, and Fragale agreement, such as the amount of surplus utility, to exert opposing effects on the perceived val- ues of negotiators’ BATNAs and contributions.

Proposition 8: The amount of surplus utility will positively (negatively) in- fluence the perceived value of negoti- ators’ contributions (BATNAs).

This discussion illustrates how the link be- tween negotiators’ inherent valuations of their BATNAs and contributions (i.e., potential power) and their perceptions of those dimensions (i.e., perceived power) may diverge in a variety of ways, owing to potentially fallible inferences concerning quantity, probability, and weight. In- deed, in subsequent research scholars may identify additional ways in which such consid- erations may lead these potential and perceived values to diverge. These influences suggest that negotiators’ potential power is likely to be trans- formed, as it is being perceived, from Figure 3a to Figure 3b, and that it is ultimately this per- ceived power that will drive negotiators’ tactical decisions.

Power Tactics Power tactics concern negotiators’ efforts to use or change the power relationship. Whereas power-use tactics concern the ways in which negotiators may attempt to leverage existing power capabilities, power-change tactics con- cern the ways in which negotiators may attempt to alter the power relationship, typically to im- prove their own power relative to that of the other party (Lawler, 1992). Since the efforts en- tailed by power-change and power-use tactics inherently differ, we discuss each type of power tactic in turn.

Power-change tactics.One conclusion negotia- tors may draw on perceiving their power relation- ship is that they do not possess enough poten- tial power, relative to that of their counterpart, to obtain desired outcomes. As a result, negotia- tors may attempt to improve their potential power. We expect that negotiators will be more likely to initiate power-change tactics when they perceive their own potential power, relative to that of their counterpart, to be low rather than high. Thus, as indicated in Figure 1, we suggest that negotiators’ perceptions of their potential power will drive their implementation of power- change tactics.Proposition 9: Negotiators will be more likely to initiate power-change tactics when they perceive their own potential power, relative to that of their counterpart, to be low rather than high.

We can furthermore draw on our prior concep- tualization of potential power to identify the kinds of power-change tactics that can be used.

Specifically, in efforts to discuss the implica- tions of power-dependence theory for tactical choice, scholars have identified four basic pow- er-change tactics (Bacharach & Lawler, 1980; Blau, 1964; Emerson, 1962; Lawler & Bacharach, 1976; Pfeffer & Salancik, 1978): in their attempts to alter the power relationship, negotiators may (1) improve the quality of their BATNA (e.g., ob- tain a job offer from another employer), (2) de- crease the quality of the counterpart’s BATNA (e.g., dissuade others from applying for the po- sition), (3) decrease their valuation of the coun- terpart’s contribution (e.g., reduce their interest in the employment opportunity), or (4) increase the counterpart’s valuation of their own contri- bution (e.g., take classes to improve their tech- nical skills).

These tactics involve changes to the two di- mensions of dependence identified in power- dependence theory. Tactics 1 and 3 seek to re- duce a negotiator’s dependence on his or her counterpart and, hence, to alter the power rela- tionship by reducing the counterpart’s power.

Tactics 2 and 4 seek to increase the counter- part’s dependence on the negotiator and, hence, alter the power relationship by increasing the focal negotiator’s power. Thus, power-depen- dence theory not only offers a foundation for evaluating negotiators’ potential power, as dis- cussed earlier in this article, but also provides a basis for understanding how this power rela- tionship can be changed (i.e., by altering these BATNAs or contributions).

The problem with this conceptualization of power-change tactics, however, is that it does not consider whether, or when, any of them should be preferred. Researchers have recog- nized that particular aspects of the bargaining situation can limit the “feasibility” of using some of these tactics to change the power rela- tionship (e.g., Jacobson & Cohen, 1986). Negotia- tors’ abilities to improve their BATNAs may be constrained, for example, by their organization’s 810October Academy of Management Review lines of communication, rules, or traditions (Bacharach & Lawler, 1980), and their abilities to improve their contributions may be constrained by such factors as inadequate talent, opportu- nity, or time. But the primary question of whether or when, beyond these idiosyncratic constraints, certain power-change tactics may be more effective at altering the power relation- ship than others remains largely unexplored (see Kim & Fragale, 2005, for an exception). We examine these issues by drawing on our prior analysis of power perceptions.

In our discussion of perceived power, we sug- gested that (1) the perceived value of negotia- tors’ BATNAs will positively influence the per- ceived value of their contributions, and vice versa (derived from our discussion ofquantity); (2) the level of trust in the relationship will pos-itively influence the perceived value of negoti- ator contributions and BATNAs (derived from our discussion ofprobability); and (3) the amount of surplus utility will positively influ- ence the perceived value of negotiators’ contri- butions and negatively influence the perceived value of their BATNAs (derived from our discus- sion ofweight). We suggest that each of these considerations should affect the relative bene- fits of power-change tactics that target one di- mension of dependence rather than the other.

On the one hand, the potential for the per- ceived value of negotiators’ BATNAs to posi- tively influence the perceived value of their con- tributions, and vice versa (i.e., Proposition 4), should reduce the significance of choosing be- tween power-change tactics that target BATNAs versus contributions to some degree. (a) Potential Power FIGURE 3 Transformations of the Potential Power Matrix (b) Perceived Power PI perceptual influences.

(c) Perceived Power After Power-Change Tactics PC Adjustments due to power-change tactics.(d) Potential Power After Power-Use Tactics PU Adjustments due to power-use tactics. 2005811 Kim, Pinkley, and Fragale Proposition 10: Power-change tactics designed to increase (decrease) the value of negotiator BATNAs will also increase (decrease) the value of nego- tiator contributions, and power- change tactics designed to increase (decrease) the value of negotiator con- tributions will also increase (decrease) the value of negotiator BATNAs.

On the other hand, insofar as these effects are likely to be only partial rather than complete, the choice between power-change tactics that target negotiators’ BATNAs versus contributions should still be significant, and the other percep- tual influences we have identified may illus- trate how their implications may diverge.

To the extent that the perceived values of ne- gotiator contributions and BATNAs are dimin- ished by a lack of trust (i.e., Proposition 6), the perceived value of a given change in each of these dimensions should be smaller when trust is low rather than high. However, beyond this general observation, we might also expect that the susceptibility of these perceptions to the lack of trust would be smaller for BATNAs than contributions, since the implementation of one’s BATNA does not depend on whether this dimen- sion is perceived accurately by the counterpart (e.g., an employee’s ability to take a job offer with another firm is not affected by whether his current employer actually believes that the em- ployee possesses this alternative offer). Increas- ing one’s contribution, in contrast, is of little avail if the counterpart does not believe that these resources will be provided and, hence, discounts the perceived value of this dimension (e.g., an employee’s offer to work longer hours will not improve the employer’s perception of the employee’s contribution if the employer be- lieves the employee is unlikely to follow through with his promise). As a result, we expect that the level of trust will increase the effectiveness of power-change tactics that target negotiator con- tributions to a greater degree than the effective- ness of power-change tactics that target negoti- ator BATNAs.

Proposition 11: The level of trust will increase the effectiveness of power- change tactics that target negotiator contributions to a greater degree than the effectiveness of power-change tac- tics that target negotiator BATNAs.Moreover, we have argued that the amount of surplus utility, which influences the likelihood of reaching an agreement, positively influences the perceived value of negotiators’ contributions and negatively influences the perceived value of negotiators’ BATNAs (i.e., Proposition 8). As a result, a given change in negotiators’ BATNAs will be perceived to be larger when the likeli- hood of reaching an agreement is low rather than high (i.e., surplus utility is small rather than large), and a given change in negotiators’ contributions will be perceived to be larger when the likelihood of reaching an agreement is high rather than low (i.e., surplus utility is large rather than small). Thus, we expect that the amount of surplus utility will positively influ- ence the effectiveness of power-change tactics that target negotiators’ contributions, and nega- tively influence the effectiveness of power- change tactics that target negotiators’ BATNAs, for altering the perceived power relationship.

Indeed, this proposition is supported by recent empirical evidence, which suggests that the benefits of improving one’s contribution in- crease, and the benefits of improving one’s BATNA diminish, as the amount of surplus util- ity grows (Kim & Fragale, 2005).

Proposition 12: The amount of surplus utility will positively (negatively) in- fluence the effectiveness of power- change tactics that target contribu- tions (BATNAs) for altering the perceived power relationship.

We should also note that although we have focused on power-change tactics that target the dimensions of potential power (i.e., BATNAs and contributions) and their implications for negoti- ators’ perceptions of the power relationship (i.e., perceived power), it is also possible for power- change tactics to target these perceptions di- rectly (Ikle, 1964). A negotiator may work to im- prove his or her BATNA, for example, or simply mislead the counterpart into believing that his or her BATNA is better than it actually is. Alter- natively, a negotiator may believe that the coun- terpart’s perception of the counterpart’s BATNA is inflated and, thus, attempt to improve the accuracy of this perception, rather than lower the counterpart’s BATNA itself. However, to the extent that power-change tactics that directly target perceptions are intended to shift these perceptions further from (rather than closer to) 812October Academy of Management Review negotiators’ inherent valuations of these BATNAs and contributions, the implications of such efforts are likely to be unstable and, hence, relatively short-lived (i.e., given Proposition 2).

These power-change tactics, furthermore, are not restricted to a specific moment in a given power relation but, rather, can be initiated on multiple occasions throughout the course of a relationship. Negotiators may implement power- change tactics in anticipation of an upcoming negotiation, they may utilize such tactics during the course of the negotiation (e.g., as they strive to identify issues that may be of worth to the other party), and they may even interrupt the negotiation midcourse to initiate power-change tactics before the negotiation is concluded.

These power-change tactics can also be pur- sued in an iterative fashion, whereby negotia- tors initially conclude that they possess insuffi- cient power to obtain desired outcomes, attempt to increase this power with one or more of these power-change tactics, evaluate the effects of such attempts on their perceived power relation- ship, and then initiate additional power-change tactics if the desired increase in power has not yet been obtained. Indeed, the need to initiate additional power-change tactics may be quite likely, given that each negotiator may have at- tempted to shift the power balance in their favor, thereby canceling out some of the effects of their initial efforts (Diamantopoulos, 1987). Thus, the links among potential power, perceived power, and power-change tactics establish a feedback loop (as depicted by the arrow leading from Ne- gotiators A’s and B’s power-change tactics back to potential power in Figure 1) that may be fre- quently repeated, in order to transform the de- pendence matrix (from Figure 3b to Figure 3c), before a given attempt at power use.

Power-use tactics.Once sufficient power has been achieved in the minds of negotiators, those negotiators should be inclined to use this power to obtain desired outcomes. And, given that per- ceptions of greater potential power, relative to that of one’s counterpart, should increase the likelihood of obtaining these rewards, negotia- tors should be more likely to employ power-use tactics when they perceive their own power, rel- ative to that of their counterpart, to be high than low. Thus, as indicated in Figure 1, we suggest that negotiators’ perceptions of their potential power will drive their implementation of power- use tactics.Proposition 13: Negotiators will be more likely to initiate power-use tac- tics when they perceive their own po- tential power, relative to that of their counterpart, to be high rather than low.

Researchers have identified a variety of power-use tactics negotiators may employ. As previously mentioned, Kipnis et al. (1980), with subsequent extensions by Kipnis and Schmidt (1983) and Yukl and Tracey (1992), identify nine dimensions of influence—pressure, legitima- tion, exchange, coalition, ingratiation, rational persuasion, inspirational appeal, consultation, and personal appeal (see Table 2)—and con- sider how one’s power relationship with others can influence the likelihood that these different tactics will be used. Yet the presentation of this overwhelming number of tactics without the concomitant provision of an organizing frame- work ultimately renders such efforts too cumber- some as a basis for theory.

Lawler (1992) offers a more helpful starting point for a theoretical analysis of power-use tac- tics, making a broad distinction between concil- iatory and hostile power-use tactics and consid- ering when they are likely to be employed.

Whereas conciliatory tactics refer to positive acts, such as communicating a willingness to coordinate or collaborate, hostile tactics refer to negative acts, such as communicating an incli- nation toward competition, intimidation, or re- sistance. By drawing on power-dependence the- ory and research on punitive power, Lawler develops two propositions concerning how the power relationship affects the incidence of these conciliatory and hostile behaviors. First, be- cause the existence of unequal power should cause negotiators to disagree over the legiti- macy of power differences and how such differ- ences should affect the negotiated solution, each negotiator will employ more hostile and fewer conciliatory power-use tactics when their relative poweris unequal than when it is equal.

Second, because highertotal power(i.e., the sum of each negotiator’s potential power) in- creases negotiators’ stakes in reaching a rea- sonable solution, they will employ fewer hostile and more conciliatory power-use tactics when total power is high than when it is low. Each of these propositions has received empirical sup- port (Bacharach & Lawler, 1981; Lawler & 2005813 Kim, Pinkley, and Fragale Bacharach, 1987). Thus, we can observe that this analysis offers a parsimonious and empirically defensible conceptualization of power-use tac- tics that can complement our prior discussions of potential and perceived power.

Yet this analysis also has some important lim- itations. Research on power-use tactics is justi- fied by the notion that the realization of one’s potential power is not automatic. Some efforts to extract benefits can be more effective than oth- ers. However, because Lawler’s (1992) analysis is primarily descriptive in nature, it considers neither the relative effectiveness of hostile ver- sus conciliatory tactics nor the reasons why dif- ferences in the effectiveness of such tactics might occur. As a result, this analysis offers little prescriptive guidance on how to use one’s power effectively in the context of negotiation.

We attempt to address these concerns by con- sidering more closely the implications of efforts to extract benefits from a relationship.

The difficulty of power use arises from the fact that one negotiator’s gain in benefits typically entails some loss in benefits for the other (Lawler, 1992). Indeed, a comparison of the kinds of power-use tactics that have been identified in the literature reveals that their differences arise not from the benefits they would offer the initi- ator, given that the initiator’s benefits are essen- tially kept constant for the purposes of those analyses, but rather from the implications of such actions for the target. Specifically, we can observe that the magnitude of the target’s loss from a power-use tactic can either be exacer- bated or mitigated depending on the type of tactic employed. Whereas conciliatory power- use tactics, such as communicating a willing- ness to coordinate or collaborate, are designed to extract benefits in ways that mitigate the harm to a target, hostile power-use tactics, such as communicating an inclination toward compe- tition, intimidation, or resistance, are designed to extract benefits in ways that actually exacer- bate the target’s harm.

The costs incurred by a target through an ini- tiator’s use of power can be influenced in a variety of ways, which we can illustrate by con- sidering the tactics identified by Kipnis and his colleagues (see Table 2). A negotiator may re- duce these costs (1) by consulting with the target to identify differences in their preferences that would offer opportunities for mutual gain (see Lax & Sebinius, 1986, Chapter 5, for a review), (2)by making inspirational appeals that would le- verage the target’s own preferences so the tar- get would find making these concessions less distasteful, or (3) by using rational persuasion to give the target the satisfaction of believing that the concessions were at least appropriate and fair (see Pruitt & Carnevale, 1993, for a review).

In contrast, a negotiator may increase these costs (1) by pressuring the target, through de- mands, threats, or intimidation, so the target not only loses the benefits of the extracted conces- sions but also suffers the additional costs of negative emotions, the loss of control, and dam- aged self-esteem; (2) by enlisting the aid of oth- ers to magnify this pressure, and thereby in- crease these psychological costs to the target; or (3) by referencing existing policies to impose many of these same kinds of psychological costs without enlisting the explicit aid of others. Fi- nally, power-use tactics such as exchange, in- gratiation, and personal appeal may neither in- crease nor reduce the costs incurred by the target in any significant way and may therefore be considered more neutral.

These differences between the implications of hostile and conciliatory tactics provide the basis for several conclusions. First, we can observe that whereas hostile power-use tactics do not require attention to addressing the specific needs of the target, conciliatory power-use tac- tics require efforts to understand the target’s preferences and satisfy at least some of those interests in order to cause the target less harm.

This suggests that power-use tactics generally will require more effort to implement when they are conciliatory than when they are hostile, be- cause conciliatory tactics require negotiators to understand their counterparts’ needs and wants.

Proposition 14: Power-use tactics will require more effort to implement when they are conciliatory than when they are hostile.

However, to the extent that negotiators seek agreements that would maximize their own ben- efits, targets should exert greater efforts to resist power-use tactics that would cause them greater harm. This resistance should, in turn, reduce the likelihood that such power-use tac- tics will be successful. Thus, given that concil- iatory power-use tactics are designed to inflict less harm on the target than hostile power-use tactics, and hence incite less resistance, we ex- 814October Academy of Management Review pect that power-use tactics will be more suc- cessful for extracting benefits from a negotia- tion when they are conciliatory than when they are hostile.

Proposition 15: Power-use tactics will be more successful for extracting ben- efits from a negotiation when they are conciliatory than when they are hostile.

Yet hostile power-use tactics may offer an ad- vantage over conciliatory power-use tactics with regard to their likelihood of encouraging targets to make preemptive concessions, if the targets expect these power-use tactics to be suc- cessful. For example, in any negotiation there is likely to be a subset of issues that one or more parties consider to be “deal-breakers”—issues that all parties to the negotiation need to agree on if an agreement is to be reached. When en- countering such an issue during the course of a negotiation, a negotiator may realize that resist- ing a power-use tactic is futile, because the ne- gotiator will ultimately have to concede on the issue anyway. In such a case, making preemp- tive concessions to prevent the successful im- plementation of a hostile power-use tactic would be worthwhile, because the cost to the target of making those concessions would only be increased if the hostile tactic were employed (i.e., owing to the additional psychological costs imposed). However, it makes little sense to make preemptive concessions to prevent the success- ful implementation of a conciliatory power-use tactic, because such tactics are designed to de- crease the cost to the target of making those con- cessions. Thus, we predict that when targets ex- pect a power-use tactic to be successful, they will be more likely to make preemptive concessions when that tactic would have been hostile than when that tactic would have been conciliatory.

Proposition 16: When targets expect a power-use tactic to be successful, they will be more likely to make pre- emptive concessions when that tac- tic would have been hostile than when that tactic would have been conciliatory.

These conclusions contribute to research on power use in several ways. First, these insights extend Lawler’s (1992) descriptive analysis of power-use tactics not only by explaining whynegotiators may be inclined to use some power- use tactics more frequently in certain situations than others, but also by offering prescriptive suggestions for those who may want to imple- ment these power-use tactics more effectively.

Second, this discussion of hostile and concilia- tory power-use tactics presents a parsimonious organizing framework that can encompass the myriad tactics that have been identified by Kip- nis and his colleagues (Kipnis & Schmidt, 1983; Kipnis et al., 1980; see also Yukl & Tracey, 1992).

Finally, this analysis helps justify and elaborate Wrong’s (1968) assertion that the compliance of targets is often based simply on their subjective expectation that potential power can and will be used and that this can, in turn, make actual power use unnecessary. Our analysis not only explains why such preemptive compliance may occur but also identifies the kinds of power-use tactics that would be more likely to elicit such compliance than others. By doing so, this anal- ysis helps expand current conceptions of power use from that of a simple one-way relationship, whereby an actor extracts benefits from its tar- get, to that of a bilateral process, whereby each party can initiate such tactics in an offensive or defensive manner to maximize or preserve the benefits, respectively, it might obtain.

Realized Power Realized power refers to the extent to which negotiators claim benefits from their interaction.

Given that this extraction of benefits is achieved through the implementation of power-use tac- tics, the type, frequency, and magnitude of a negotiator’s power-use efforts should directly in- fluence the extent to which that party has real- ized power from the relationship. Thus, as indi- cated in Figure 1, we suggest that negotiators’ power-use tactics will affect their realized power in the negotiation.

Proposition 17: The type, frequency, and magnitude of a negotiator’s pow- er-use tactics will directly influence that party’s realized power.

The constraint on such actions, however, arises from the fact that this realization of power can subsequently alter the power relationship.

Lawler (1992) has observed, for example, that the extraction of benefits from a relationship inflicts some cost on the target, which will reduce the 2005815 Kim, Pinkley, and Fragale target’s valuation of the relationship (see our prior discussion of power-change tactics) and, hence, lower the target’s dependence. This no- tion also provides the basis for the conclusion drawn by many power researchers that extract- ing benefits from a relationship reduces one’s potential power in future interactions (Emerson, 1962; Lawler, 1992), as embodied by the well- known adage “to use power is to lose it.” 2 Yet this conclusion is based on only a narrow assessment of the implications, for the target, of having these benefits extracted, rather than the implications of such actions for the relationship as a whole. We can acknowledge that the ex- traction of benefits from a relationship may re- duce the target’s valuation of the relationship and lower that target’s dependence. However, we can also observe that, by extracting these benefits, the initiator of a given power-use tactic no longer relies on the target for those valued rewards and, hence, becomes less dependent on the target. As a result, the clearest implication of these combined influences is that the extraction of benefits from a negotiation (i.e., realized power) will negatively influence thetotal power (i.e., the sum of each negotiator’s potential power) in the relationship. 3Thus, as indicated in Figure 1, we suggest that negotiators’ realized power will affect their potential power in future negotiations.

Proposition 18: Realized power will negatively influence the total power in the relationship.

This conclusion does not imply that negotia- tors’ relative power is unaffected by power use.

Rather, we expect that the implications of power use for relative power will depend on the type of power-use tactic employed. Specifically, given that conciliatory power-use tactics are designed to extract benefits in ways that reduce costs tothe target, initiators of such tactics may realize power in ways that reduce their own depen- dence on the target (i.e., because of the acquisi- tion of desired rewards) to agreaterdegree than the reduction in the target’s dependence on the initiator (i.e., given the mitigated costs to the target of the foregone benefits).

In contrast, given that hostile power-use tac- tics are designed to extract benefits in ways that magnify costs to the target, initiators of such tactics may extract benefits in ways that reduce the initiator’s dependence on the target (i.e., be- cause of the acquisition of desired rewards) to a lesserdegree than the reduction in the target’s dependence on the initiator (i.e., given the mag- nified costs to the target from having benefits extracted in a hostile way). Thus, we expect that whereas conciliatory power-use tactics will shift relative power in favor of the initiator, hostile power-use tactics will shift relative power in favor of the target.

Proposition 19: Whereas conciliatory power-use tactics will shift relative power in favor of the initiator, hos- tile power-use tactics will shift rela- tive power in favor of the target.

These conclusions offer several insights for an analysis of realized power in the context of ne- gotiation. First, given the loss in total power that power-use tactics entail, negotiators may not necessarily want to realize as much power as they can from a negotiation. Their decisions to do so should depend on their expectations for the relationship, given that such actions will affect their counterpart’s subsequent valuation of the relationship and, hence, the focal negoti- ator’s potential power in future interactions (from Figure 3c to Figure 3d). Thus, our analysis is not limited to one-shot agreements but also has implications for relationships of longer du- ration. In one-shot agreements, negotiators are likely to realize as much power as they can in negotiation, whereas in ongoing relationships, negotiators will be less likely to maximize real- ized power because of the negative effects this will have on their potential power in future negotiations.

Proposition 20: The extent to which ne- gotiators realize power will be nega- tively influenced by expectations of future interaction. 2Although the extraction of benefits may also make clear to the target that the initiator is willing and able to employ a power-use tactic and in this sense increase, to some ex- tent, the perceived level of the initiator’s potential power (March, 1966), this effect is expected to be outweighed by the decrease in the initiator’s perceived power from having re- duced the supply of benefits available from, and hence the target’s valuation of, the relationship.

3This implication holds irrespective of whether the ben- efits from the relationship are fixed or replenishing, “so long as additions to the power supply are independent of the exercise of power” (March, 1966: 64). 816October Academy of Management Review Second, this analysis offers an explanation for why Yukl and Tracey (1992) found some power- use tactics (i.e., rational persuasion, inspiration- al appeal, and consultation) to be more effective than others (i.e., pressure, coalition, and legiti- mation). In our prior discussion of power-use tactics, we actually drew on these examples to illustrate how such tactics can be conciliatory or hostile, respectively, and explained why the former methods would be more successful for extracting benefits than the latter. We now ex- tend these conclusions by illustrating how such efforts can also affect negotiators’ power rela- tionships in future interactions and how one’s relative power can be maintained more effec- tively through long-term relationships by imple- menting power-use tactics that are conciliatory rather than hostile.

Proposition 21: Power-use tactics will be more effective for maintaining one’s relative power through long- term relationships when they are con- ciliatory than when they are hostile.

Finally, this discussion may help explain and extend an emerging body of research on the effects of power on those who possess it (Keltner, Gruenfeld, & Anderson, 2003), which suggests that power may exert a disinhibiting influence on social behavior. According to that perspec- tive, the powerful become less restrained in sat- isfying their desires, less attentive to the inter- ests of others, and more likely to indulge in personal motives than those without power. The present analysis may help explain why this may occur. If negotiators possess greater relative power, they can afford to use hostile power-use tactics without completely eroding their power advantage for the future. Thus, powerful negoti- ators may possess greater liberty to implement a wider range of power-use tactics and, hence, face greater temptations to use those that are hostile (i.e., given the lower effort hostile tactics would require to initiate) than negotiators with less power. This reasoning also suggests that this disinhibition is not merely dichotomous (i.e., negotiators either use hostile tactics or they don’t) but, rather, may be more accurately por- trayed as a continuous relationship whereby ne- gotiators will be more likely to behave in disin- hibited ways (i.e., by increasing the frequency, magnitude, and hostility of their power-use tac- tics) as the size of their power advantage grows.Proposition 22: The size of a negotia- tor’s power advantage will be posi- tively related to the frequency, magni- tude, and hostility of their power-use tactics (i.e., efforts to realize power).

Illustration This range of dynamics that may arise in power relations can be illustrated through an example. Imagine a context in which two par- ties, A and B, are negotiating the terms of a potential joint venture. Because of differences in their preferences and their alternatives to this opportunity, each negotiator’s valuation of the relationship (i.e., due to the counterpart’s poten- tial contribution) and his or her best alternative (i.e., his or her own BATNA) can be represented by Figure 4a. These values provide the basis for calculating each negotiator’s potential power so that, based on our prior analysis, the potential power for A and B would be 190 and 140, respec- tively.

These valuations are likely to be transformed, however, as the negotiators form perceptions of this power relationship. The potential for a ne- gotiator to infer the value of the counterpart’s contribution from the value of the counterpart’s BATNA and vice versa may decrease, to some extent, differences in the negotiators’ valuations of these dimensions. Lack of trust may lead each negotiator to discount the value of the counter- part’s BATNA and contribution. Moreover, the relatively large amount of surplus utility (i.e., 330 90 240) may lead the negotiators to discount the possibility of an impasse and, thus, increase their valuations of the relationship and decrease their valuations of their BATNAs. Such influences should transform the power matrix as it is perceived from Figure 4a to Figure 4b, for example.

These perceptions may, in turn, lead the ne- gotiators to conclude that they possess inade- quate power to obtain desired outcomes and, therefore, to implement power-change tactics.

Negotiator A may choose to improve his or her BATNA (e.g., by identifying an alternative joint venture opportunity), for example, whereas ne- gotiator B may choose to improve his or her contribution (e.g., by investing in skills that would help market the joint venture’s product).

And although each party may achieve the same magnitude change in his or her chosen dimen- 2005817 Kim, Pinkley, and Fragale sion, the implications for the perceived power relationship are likely to differ. Whereas the lack of trust may make A’s efforts to improve his or her BATNA more effective for changing the perceived power relationship than B’s efforts to improve his or her contribution, this advantage may be counteracted to some degree by the abundant surplus utility, which should lead them to discount the value of any change to their BATNAs and magnify the value of any change to their contributions. The nature of these percep- tual influences, the extent to which they operate, and the types, magnitude, and frequency of these power-change efforts should, therefore, transform the negotiators’ perceptions of theirpotential power from Figure 4b to Figure 4c, for example, before any attempt at power use.

The characteristics of these power-use efforts should, furthermore, depend on the negotiators’ assessment of and goals for the relationship. To the extent that the negotiators do not care about the future and also possess a relative power advantage, they may use more hostile tactics and extract the maximum amount of benefits from the agreement. However, to the extent that these negotiators expect a longer-term relation- ship (as would be likely for those who plan a joint venture) and seek to manage this relation- ship effectively, they may use more conciliatory tactics and avoid extracting all possible bene- FIGURE 4 Illustration (a) Potential Power A’s potential power 30 (220 60) 190; B’s potential power 60 (110 30) 140; total power 190 140 330.(b) Perceived Power A’s perceived power 20 (160 40) 140; B’s perceived power 40 (50 20) 70; perceived total power 140 70 210.

(c) Perceived Power After Power-Change Tactics A’s perceived power 60 (160 40) 180; B’s perceived power 40 (80 60) 60; perceived total power 180 60 240.(d) Potential Power After Power-Use Tactics A’s potential power 60 (100 40) 120; B’s potential power 40 (70 60) 50; total power 120 50 170. 818October Academy of Management Review fits so as to retain some potential power for their future interactions, and thereby transform the power matrix from Figure 4c to Figure 4d, for example, which, in turn, provides the basis for the next iteration of this dynamic cycle.

CONCLUSION Our purpose in this article has been to de- velop a comprehensive theoretical model that can integrate the most prominent theories of power, account for the wide range of results found in negotiation research, and offer novel propositions that can be tested in future empir- ical work. Our efforts to do so have led us to decouple power into four distinct components— (1) potential power, (2) perceived power, (3) power tactics, and (4) realized power—and to assess their implications for those who wish to manage power effectively in the context of ne- gotiation. Through these endeavors, we have emphasized the dynamic nature of power rela- tions and examined each phase in this dynamic process with added depth and precision. As a result, this analysis may help address numer- ous limitations that have persisted in the power literature.

By integrating the wide array of approaches researchers have taken in their discussions of power, this analysis helps reconcile the myriad lists of characteristics, properties, strategies, and descriptions that have been offered with regard to this topic and consolidate them into a more coherent, parsimonious, and accessible framework. By depicting power relations as a dynamic process, this analysis helps address the criticisms that have been raised concerning the static nature of existing power theories and the failure of such static conceptions to capture the complexities of power relations that arise in real-world environments (Astley & Sachdeva, 1984; Neale & Northcraft, 1991). Finally, by delv- ing into the components of power that we have differentiated, this inquiry extends conceptual- izations of each of these components.

Our analysis of potential power suggests that prior formulations of this concept may be inad- equate, and it offers a revised formulation that seems better justified by empirical research.

Our discussion of perceived power draws on power theorists’ recognition that assessments of power are inherently subjective in nature (Bacharach & Lawler, 1976), but it then extendsthis notion by discussing how these potential and perceived values may diverge. Our assess- ment of power tactics begins by emphasizing a fundamental, but too often overlooked, distinc- tion between power-change and power-use tac- tics, and then extends prior discussions of such tactics in fundamental ways— considering when some power-change tactics may be more effective for altering the power relationship than others and then consolidating the wide array of power-use tactics that have been identified in the literature into a parsimonious framework that of- fers clear, prescriptive recommendations for their use. Finally, our analysis of realized power draws on the common adage that “to use power is to lose it” but then offers a more precise as- sessment of the implications of power use for total and relative power, and through such con- siderations offers much needed insight into the reasons why negotiators may vary in both the manner and extent to which they attempt to ex- tract benefits from their relationship.

This inquiry, therefore, acknowledges the enormous complexity of power relations, their dynamic transformations, and how power- related behaviors may be influenced by both short- and long-term considerations. We also seek to address these issues in a parsimonious, logically coherent way. However, we recognize this framework’s limitations. Although our anal- ysis of perceived power identifies several per- ceptual influences, others may exist that we have failed to mention. We briefly noted that situational constraints could affect the feasibil- ity of power-related behaviors, but the impor- tance of these constraints should not be dis- counted, given that all behaviors are restricted at least in part by the surrounding social struc- ture (e.g., Brass, 1984; Lawler, 1992). Moreover, although we have implicitly acknowledged the influence of parties outside the focal negotia- tion, given that they ultimately determine nego- tiators’ BATNAs, benefits may be gained through more explicit treatments of the roles such outside parties play in determining the nature of the focal negotiation and the ways in which negotiator BATNAs may be changed (e.g., Komorita & Hamilton, 1984; Mannix, 1993b). Al- though the demands of presenting a clear and concise analysis of the enormous literature on power that would be relevant for negotiation were great, we have done our best to convey the elements of this research that would be most 2005819 Kim, Pinkley, and Fragale important for the reader. However, as these lim- itations should illustrate, there is room for much more work in this area.

It would also be appropriate to place our dy- namic model of power in negotiation within the context of the broader power literature. By re- stricting our analysis to negotiation, we have focused on cases where the focal parties are initiating discrete, strategic acts (e.g., making offers, issuing threats, and selectively disclos- ing information). Yet there are many cases where power operates in a more routine and prosaic fashion—for example, socialization and accreditation processes, technological systems, and insurance and tax regimes (Covaleski, Dirsmith, Heian, & Samuel, 1998; Noble, 1984; Shaiken, 1984; Simon, 1988)—reflecting a mode of power that is more systemic, rather than epi- sodic, in form (Clegg, 1989; Foucault, 1977; Gid- dens, 1984). The interdependent process of nego- tiation, furthermore, is based on the notion that the targets of power are capable of agency or choice (e.g., whether to accept a proposal, make a counterproposal, or abandon the negotiation in favor of one’s BATNA). However, there are many cases where the use of power does not require the target to “do” anything—for exam- ple, physical violence, punishment, and actuar- ial practices (Hearn, 1994; Milgram, 1974; Simon, 1988); in such cases, targets are treated more as objects than as subjects (i.e., actors capable of agency) in the power relationship (Simon, 1988).

These mode (episodic versus systemic) and target (subject versus object) dimensions are useful to consider because their intersection cre- ates four quadrants that provide the basis for a broader typology of power forms (Lawrence, Winn, & Jennings, 2001). Our focus on power in negotiation, and its emphasis on anepisodic form of power that treats targets assubjects, falls within the quadrant that these researchers terminfluence.Thus, one might wonder whether the propositions that have been developed in this quadrant would hold under conditions of force(i.e., when power is episodic and targets are treated as objects),discipline(i.e., when power is systemic and targets are treated as subjects), ordomination(i.e., when power is sys- temic and targets are treated as objects). It is possible, for example, that the episodic use of power, particularly hostile tactics, may not nec- essarily reduce one’s potential power in futureinteractions when a target’s agency is removed (i.e., the conditions of force).

Notwithstanding these limitations, we still feel that our power model extends past work to integrate new research on negotiation. One key difference from earlier work in the area of power is that research on negotiation has a high level of validation and testing (e.g., see Bazerman, Curhan, Moore, & Valley, 2000; Kramer & Mes- sick, 1995; Murnighan, 1993, for reviews). We think our examples and the citations in the ar- ticles above should provide a good foundation for joint empirical testing of many of our propo- sitions. Indeed, if negotiation is as central as it appears to be in building realized power, it may be that the various quadrants of power (influ- ence, force, discipline, and domination) may all rely in some fundamental fashion on the power dynamics of negotiation.

REFERENCES Astley, W. G., & Sachdeva, P. S. 1984. Structural sources of intraorganizational power: A theoretical synthesis.

Academy of Management Review,9: 104 –113.

Bacharach, S. B., & Lawler, E. J. 1976. The perception of power.Social Forces,55: 123–134.

Bacharach, S. B., & Lawler, E. J. 1980.Power and politics in organizations.San Francisco: Jossey-Bass.

Bacharach, S. B., & Lawler, E. J. 1981.Bargaining: Power, tactics, and outcomes.San Francisco: Jossey-Bass.

Bazerman, M. H., Curhan, J. R., Moore, D. A., & Valley, K. L.

2000. Negotiation.Annual Review of Psychology,51: 279 – 314.

Bierstedt, R. 1950. An analysis of social power.American Sociological Review,15: 730 –738.

Blau, P. M. 1964.Exchange and power in social life(5th ed.).

New Brunswick, NJ: Transaction.

Brass, D. J. 1984. Being in the right place: A structural anal- ysis of individual influence in an organization.Admin- istrative Science Quarterly,29: 518 –539.

Cialdini, R. B. 1984.Influence: The psychology of persuasion.

New York: Quill.

Clegg, S. R. 1989.Frameworks of power.London: Sage.

Covaleski, M. A., Dirsmith, M. W., Heian, J. B., & Samuel, S.

1998. The calculated and the avowed: Techniques of discipline and struggles over identity in Big Six public accounting firms.Administrative Science Quarterly,43:

293–327.

Diamantopoulos, A. 1987. Vertical quasi-integration revis- ited: The role of power.Managerial and Decision Eco- nomics,8: 185–194.

Emerson, R. M. 1962. Power-dependence relations.American Sociological Review,27: 31– 40. 820October Academy of Management Review Ensari, N., & Miller, N. 2002. The out-group must not be so bad after all: The effects of disclosure, typicality, and sa- lience on intergroup bias.Journal of Personality and Social Psychology,83: 313–329.

Foucault, M. 1977.Discipline and punish: The birth of the prison.New York: Vintage Books.

French, J. R. P., Jr., & Raven, B. 1959. The bases of social power. In D. Cartwright (Ed.),Studies in social power:

150 –167. Ann Arbor: University of Michigan Press.

Giddens, A. 1984.The constitution of society: Outline of a theory of structuration.Cambridge: Polity Press.

Gilovich, T. 1990. Differential construal and the false consen- sus effect.Journal of Personality and Social Psychology, 59: 623– 634.

Hearn, J. 1994. The organization(s) of violence: Men, gender relations, organizations, and violence(s).Human Rela- tions,47: 731–754.

Ikle, F. C. 1964.How nations negotiate,New York: Harper & Row.

Jacobson, C., & Cohen, A. 1986. The power of social collec- tivities: Towards an integrative conceptualization and operationalization.British Journal of Sociology,37: 106 – 121.

Keltner, D., Gruenfeld, D. H., & Anderson, C. 2003. Power, approach, and inhibition.Psychological Review,110:

265–284.

Kim, P. H. 1997. Strategic timing in group negotiations: The implications of forced entry and forced exit for negotia- tors with unequal power.Organizational Behavior and Human Decision Processes,71: 263–286.

Kim, P. H., Diekmann, K. A., & Tenbrunsel, A. E. 2003. Flattery may get you somewhere: The strategic implications of providing positive vs. negative feedback about ability vs. ethicality in negotiation.Organizational Behavior and Human Decision Processes,90: 225–243.

Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. 2004.

Removing the shadow of suspicion: The effects of apol- ogy vs. denial for repairing ability- vs. integrity-based trust violations.Journal of Applied Psychology,89: 104 – 118.

Kim, P. H., & Fragale, A. R. 2005. Choosing the path to bar- gaining power: An empirical comparison of BATNAs and contributions in negotiation.Journal of Applied Psychol- ogy,90: 373–381.

Kipnis, D., & Schmidt, S. M. 1983. An influence perspective on bargaining within organizations. In M. Bazerman & R. Lewicki (Eds.),Research on negotiation in organiza- tions,vol. 1: 303–319. Greenwich, CT: JAI Press.

Kipnis, D., Schmidt, S. M., & Wilkinson, I. 1980. Intraorgani- zational influence tactics: Explorations in getting one’s way.Journal of Applied Psychology,65: 440 – 452.

Komorita, S. S., & Hamilton, T. P. (Eds.). 1984.Power and equity in coalition bargaining,vol. 3. Greenwich, CT: JAI Press.

Kramer, R. M., & Messick, D. M. 1995.Negotiation as a social process: New trends in theory and research.Thousand Oaks, CA: Sage.Lawler, E. J. 1992. Power processes in bargaining.Sociolog- ical Quarterly,33: 17–34.

Lawler, E. J., & Bacharach, S. B. 1976. Outcome alternatives and value as criteria for multistrategy evaluations.Jour- nal of Personality and Social Psychology,34: 885– 894.

Lawler, E. J., & Bacharach, S. B. 1987. Comparison of depen- dence and punitive forms of power.Social Forces,66:

446 – 462.

Lawrence, T. B., Winn, M. I., & Jennings, P. D. 2001. The temporal dynamics of institutionalization.Academy of Management Review,26: 624 – 644.

Lax, D. A., & Sebinius, J. K. 1986.The manager as negotiator:

Bargaining for cooperation and competitive gain.New York: Free Press.

Lewicki, R. J., Saunders, D. M., & Minton, J. W. 1999.Negoti- ation(3rd ed.). Boston: Irwin-McGraw Hill.

Mannix, E. A. 1993a. The influence of power, distribution norms and task meeting structure on resource allocation in small group negotiation.International Journal of Con- flict Management,4: 5–23.

Mannix, E. A. 1993b. Organizations as resource dilemmas:

The effects of power balance on coalition formation in small groups.Organizational Behavior and Human De- cision Processes,55: 1–22.

March, J. G. 1966. The power of power. In D. Easton (Ed.), Varieties of political theory:39 –70. Englewood Cliffs, NJ:

Prentice-Hall.

Milgram, S. 1974.Obedience to authority: An experimental view.New York: Harper & Row.

Mooney, L. 1984. The social psychology of power.Sociologi- cal Spectrum,4: 31–51.

Murnighan, J. K. 1993.Social psychology in organizations:

Advances in theory and research.Englewood Cliffs, NJ:

Prentice-Hall.

Neale, M. A., & Bazerman, M. H. 1991.Cognition and ratio- nality in negotiation.New York: Free Press.

Neale, M. A., & Northcraft, G. B. 1991. Behavioral negotiation theory: A framework for conceptualizing dyadic bar- gaining.Research in Organizational Behavior,13: 147– 190.

Nicholson, W. 1989.Microeconomic theory(4th ed.). Chicago:

Dryden Press.

Noble, D. 1984.Forces of production: A social history of in- dustrial automation.New York: Knopf.

Pfeffer, J., & Salancik, G. R. 1978.The external control of organizations: A resource dependence perspective.New York: Harper & Row.

Pinkley, R. L., Neale, M. A., & Bennett, R. J. 1994. The impact of alternatives to settlement in dyadic negotiation.Or- ganizational Behavior and Human Decision Processes, 57: 97–116.

Pruitt, D. G., & Carnevale, P. J. 1993.Negotiation in social conflict.Pacific Grove, CA: Brooks/Cole.

Raven, B. H. 1974. A comparative analysis of power and power preference. In J. T. Tedeschi (Ed.),Perspectives on social power:172–198. Chicago: Aldine. 2005821 Kim, Pinkley, and Fragale Shaiken, H. 1984.Automation and labor in the computer age.

New York: Holt, Rinehart & Winston.

Simon, J. 1988. The ideological effects of actuarial practices.

Law and Society Review,22: 771– 800.

Smith, D. L., Pruitt, D. G., & Carnevale, P. J. 1982. Matching and mismatching: The effect of own limit, other’s tough- ness, and time pressure on concession rate in negotia- tion.Journal of Personality and Social Psychology,42:

876 – 883.

Thompson, L. 1990. An examination of naive and experi- enced negotiators.Journal of Personality and Social Psy- chology, 59: 82–90.Thompson, L. 2000.The mind and heart of the negotiator(2nd ed.). Upper Sadde River, NJ: Prentice-Hall.

Weber, M. 1947.The theory of social and economic organiza- tion.New York: Free Press.

Williamson, O. E. 1975.Markets and hierarchies: Analysis and anti-trust implications.New York: Free Press.

Wrong, D. H. 1968. Some problems in defining social power.

American Journal of Sociology,73: 673– 681.

Yukl, G., & Tracey, J. B. 1992. Consequences of influence tactics used with subordinates, peers, and the boss.

Journal of Applied Psychology,77: 525–535.

Peter H. Kimis an associate professor of management and organization at the Uni- versity of Southern California’s Marshall School of Business. He received his Ph.D. in organizational behavior from Northwestern University’s Kellogg Graduate School of Management. His research interests concern the dynamics of interpersonal percep- tions regarding ability, intentionality, and power and their implications for work- groups, negotiations, and dispute resolution.

Robin L. Pinkleyis an associate professor of organizational behavior and business policy at the Southern Methodist University’s Cox School of Business. She received her Ph.D. in social psychology from the University of North Carolina. Her current research interests include the development of a new gain-gain approach to profitable negoti- ation, negotiator power, the use of strategic anchors and bids in negotiation, and fairness.

Alison R. Fragaleis an assistant professor at the University of North Carolina’s Kenan-Flager Business School. She received her Ph.D. in organizational behavior from Stanford University’s Graduate School of Business. Her research interests focus on the determinants and consequences of power, status, and influence in organizations; conflict resolution and negotiation; and verbal and nonverbal communication. 822October Academy of Management Review