BROKERAGE AND BROKERING: AN INTEGRATIVE REVIEW AND … Files/Brokerage and... · course by the...

r Academy of Management Annals2019, Vol. 13, No. 1, 215–239.https://doi.org/10.5465/annals.2017.0024

BROKERAGE AND BROKERING: AN INTEGRATIVE REVIEWAND ORGANIZING FRAMEWORK FOR THIRD

PARTY INFLUENCE

NIR HALEVY1

Stanford University

ELIRAN HALALIBar-Ilan University

JULIAN J. ZLATEVHarvard University

Brokerage and brokering are pervasive and consequential organizational phenomena.Prevailing models underscore social structure and focus on the consequences that comefrom brokerage—occupying a bridging position between disconnected others in a network.By contrast, emerging models underscore social interactions and focus on brokering—thebehavioral processes through which organizational actors shape others’ relationships.Our review led us to develop a novel framework as a means to integrate and organizea wide range of theoretical insights and empirical findings on brokerage and brokering.The Changing Others’ Relationships (COR) framework captures the following ideas thatemerged from our review: (a) Different triadic configurations enable different forms ofbrokering, which in turn produce distinct effects on others’ relationships, (b) brokering isamultifaceted social influence process that can take the form of intermediation (connectingdisconnected others) or modification (changing others’ preexisting relationships), (c) com-paring social relations prebrokering versus postbrokering reveals a broker’s impact, (d)brokering can influence others’ relationships positively or negatively, and (e) informationand incentives are two principal means through which individuals change others’ re-lationships. Overall, the current review integrates multiple streams of research relevant tobrokerage and brokering—including those on structural holes, organizational innovation,boundary spanning, social and political skill, workplace gossip, third-party conflict man-agers, and labor relations—and links each of the emergent themes identified in the currentreview to promising directions for future research on brokerage and brokering.

Navigating social interdependence is one of thegreatest challenges of the human existence (Kelley &Thibaut, 1978; Markus & Kitayama, 2010; Rand &Nowak, 2013). Social relations are crucial for our sur-vival, functioning, well-being, and success (Baumeister& Leary, 1995; Coleman, 1988; Eisenberger, Lieberman,&Williams, 2003). Hence, much of our affective, cogni-tive, and behavioral repertoire as humans is gearedtoward initiating, regulating, and managing social

interactions and relationships (Mercier & Sperber, 2011;Schilbach, Eickhoff, Rotarska-Jagiela, Fink, & Vogeley,2008). Critically, as members of an ultrasocial species,we habitually manage and regulate not only our owninteractions and relationships but also other people’sinteractions and relationships. Influencing others’ in-teractions and relationships does not require occupyinga particular social role, formal position, or specific re-lation to others (such as parent, teacher, or manager).Rather, merely being connected to, and interdependentwith, other people provides opportunities, the motiva-tion, and at least some capacity to influence others’ in-teractions and relationships.

Attempts to influence other people’s interactionsand relationships in organizational contexts are bothubiquitousanddiverse.Making face-to-face ande-mailintroductions, engaging in benevolent andmalevolent

We thank Taya Cohen, Amir Goldberg, Sharique Hasan,Jennifer Dannals, Emily Reit, Alisa Yu, Daan Van Knip-penberg, Sharon Parker, two anonymous reviewers, andseminar participants at Stanford University, University ofChicago, and Northwestern University for helpful feedbackon this work. Eliran Halali gratefully acknowledges supportfrom the Israel Science Foundation (grant #1699/17).

1 Corresponding author.

215

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

https://doi.org/10.5465/annals.2017.0024

workplace gossip, and intervening as mediators or ar-bitrators in coworkers’ disputes are just some of themeans through which we routinely influence others’relationships in organizational contexts. The socialinfluence that people exert over others’ relationships isthe essence of brokering and the primary focus of thisarticle.2

The pervasiveness and significance of brokeragepositions and brokering processes in organizationsstimulated a substantial body of scholarly work onthese consequential phenomena. Researchers study-ing social structure have introduced multiple con-structs to address the richness of social networkphenomena at the node level (e.g., centrality andconstraint), the dyadic level (e.g., tie strength andmultiplexity), and the network level (e.g., densityand cohesion) (Burt, 1992; Clifton & Webster, 2017;Freeman, 1978; Krackhardt, 1999;Uzzi&Spiro, 2005;Zhelyazkov, 2018). Indeed, several author teams be-fore us have undertaken the admirable endeavor toreview and summarize the voluminous and dynamicliterature on the consequences of brokerage—that is,the outcomes associated with occupying a bridgingposition between disconnected others in a network(Burt, Kilduff, & Tasselli, 2013; Carpenter, Li, & Jiang,2012; Kilduff & Brass, 2010; Landis, 2016; Stovel &Shaw, 2012).

Here, we aim to review and integrate the thrivingliterature on brokerage with emerging insights andfindings on brokering processes. Multiple researchteams have made important strides in recent yearstoward a deeper understanding of how brokeringhappens (Kaplan, Milde, & Cowan, 2017; Lingo &O’Mahony, 2010; Obstfeld, 2017; Quintane &Carnabuci, 2016). To integrate research on broker-age and brokering, we followed recent recommen-dations (Aguinis, Ramani, & Alabduljader, 2018),and engaged in an iterative process to define thescope of the current review and identify relevantcontent by searching relevant databases, seekingadvice from scholars with pertinent expertise, andconducting back searches for referenced work. Weintentionally took an inclusive approach and choseto cast a wide net in reviewing the relevant litera-tures. This conscious decisionwas based on our goalto advance an integrative conceptualization of bro-kerage and brokering processes in organizations.This conscious effort to integrate and synthesize

a broad body of work in the social and organiza-tional sciences responds to recent calls to betterintegrate structural network phenomena and so-cial psychological phenomena in organizationalresearch (Casciaro, Barsade, Edmondson, Gibson,Krackhardt, & Labianca, 2015; Clifton & Webster,2017) and is manifested in the diversity of theworks we review.

CONCEPTUALIZING BROKERAGEAND BROKERING

Throughout this review, we will use the termbrokerage to denote a particular structural positionin a social network, and the term brokering to denotethe behaviors that individuals pursuewhen acting asbrokers.3 Conceptualizations of both brokerage andbrokering vary in scope. Whereas inclusive concep-tualizations endeavor to reflect the inherent breadthand complexity of these phenomena, exclusive def-initions focus more narrowly on a particular aspect.We find it useful to begin our exploration of thelandscape of conceptualizations of brokerage andbrokering in Simmel’s (1950) classic work on triadsas it provides an early and thought-provoking anal-ysis of the varieties of triadic phenomena. We thenpresent more recent definitions that echo Simmel’sinclusive perspective and contrast them with moreexclusive definitions that focus on a subset of thefunctional forms that brokerage and brokering take.

Simmel’s conceptualization of brokering (albeitusing other terms) is based on his analysis of socialinfluence processes in triads. Simmel (1950: 149)noted that “social life is constantly determined in itscourse by the presence of the third person,” andstipulated a number of distinct ways in which the“third element” fundamentally transforms dyadicinteractions and relationships. These distinct formsof third-party influence include acting as an im-partial mediator or arbitrator “to save the groupunity from the danger of splitting up” (p. 154); the“tertius gaudens”—the third who benefits from us-ing others’ preexisting quarrel or separation; and thesocial divider (“divide et impera”)—the third who

2 Whereas some social influence processes in organiza-tional contexts target others’ identities, motivations, andgoals, brokering processes—our focus here—specificallytarget others’ relationships.

3 Some of the publishedworkwe cite has either used thetwo terms interchangeably or used one of the two terms asa broad umbrella-concept to denote both structural char-acteristics and behavioral processes in networks. Whendirectly quoting others’ work, we naturally maintain theoriginal authors’ choice of terms. In all other instances, weuse brokerage to denote a structural position and brokeringto denote a social process.

216 JanuaryAcademy of Management Annals

“intentionally produces the conflict to gain a domi-nating position” (p. 162).

Simmel explicitly acknowledged that third partiesmay bolster or undermine dyadic relations, noting“. . .among three elements, each one operates as anintermediary between the other two, exhibiting thetwofold function. . .to unite and separate. . . .the in-direct relation does not only strengthen the directrelation. It may also disturb it” (p. 135). Simmelproposed further that brokering may also result in“the intensification of relation by a third element” (p.136). That is, a broker can reinforce a preexistingrelationship between two alters. Brokers may en-hance apositive relationshipbyusing their influenceto promote dyadic intimacy, trust, and cooperation,thereby making a weak tie stronger. By contrast,brokers can aggravate a preexisting negative re-lationship between two alters by using their in-fluence to intensify suspicion, hostility, andcompetition.4 FollowingSimmel,we subscribe to theview that both brokerage and brokering are multi-faceted constructs. Thus, similar to the idea that“qualitatively different roles. . .have equal claim tothe term brokerage” (Gould & Fernandez, 1989: 123,italics in source), we contend that different behav-ioral processes have equal claim to the termbrokering.

Our reading of Simmel’s analysis leads us tohighlight five aspects that we see as particularlynoteworthy and that we suggest can serve as a con-ceptual yardstick when exploring other definitionsof brokerage and brokering. First, Simmel’s analysisfocuses on small groups.5 Second, it defines broker-ing as a social influence process. Third, it proposesthat brokering does not require the absence of pre-existing ties between alters. Thus, it may occur ina closed triad—whenalters have preexisting positiveor negative relations, as well as in an open triad—when the two alters are connected only through thebroker. Accordingly, it conceptualizes brokeringboth as intermediation (instances in which egofunctions as an intermediary who connects two

disconnected alters, either directly or indirectly:Soda, Tortoriello, & Iorio, 2018) and as modification(instances in which ego modifies the nature of thepreexisting relationship between two alters: Obstfeld,Borgatti, & Davis, 2014). Fourth, it explicitly ac-knowledges that third parties can be helpful andharmful in their impact on others’ relationships. Last,Simmel discusses both how brokers influence others’outcomes and how they shape their own outcomes.

Several other conceptualizations of brokerage andbrokering share the first and second elements notedearlier, and similarly emphasize the importance ofstudying interpersonal processes in small groups(Gould & Fernandez, 1989; Krackhardt & Kilduff,1999). For example, Grannovetter (1973: 1360)called attention to the significance of small-scalesocial interactions and the value of learning “whattranspires within the confines of the small group,”and Stovel and Shaw (2012: 139) asserted that “be-cause brokering is built from informal, personal re-lationships, understanding it requires close attentionto micro-level relations and social psychologicalprocesses.” Other definitions of brokering similarlyfocus on the “behavior bywhich an actor influences,manages, or facilitates interactions between otheractors” (Obstfeld et al., 2014: 141).

Definitions of brokerage and brokering divergewith regard to the third aspect noted earlier. Specif-ically, definitions of brokering assert that brokeringmay occur in both closed and open triads (i.e., whenthe two alters have preexisting ties or do not havepreexisting ties, respectively; Obstfeld et al., 2014;Simmel, 1950). By contrast, definitions of brokeragetypically view the absence of a preexisting tie asa prerequisite to brokerage, and, consequently, tendto conceptualize brokerage as occurring only in opentriads. Consistent with the idea that brokerage onlyoccurs in open triads, Gould and Fernandez (1989)proposed that “the whole point of brokering is tocreate an indirect relation where no direct relationexists” (p. 95, italics added), and further emphasizedthe importance of intermediation in brokering byasserting that “brokering in general involves the flowor exchange of resources from one actor to anothervia an intermediary” (p. 123). Similarly, Burt (2000:356) wrote that “brokering is explicitly about actionthat cuts across structural holes in the current socialstructure” (italics added) and in a recent review ofthe literature with colleagues reiterated that “bro-kering is the action of coordinating across the holewith bridge connections betweenpeople onoppositesides of the hole” (Burt et al., 2013: 531). A similaremphasis on the importance of structural holes in

4 Interestingly, Simmel subscribed to the view that “it isusually much easier for the average person to inspire an-other individual with distrust and suspicion towarda third. . . than with confidence and sympathy” (1955: 30).

5 Simmel justified the focus on triads by noting that therelational configurations that exist in triads “are impossi-ble if there are only two elements; and, on the other hand, ifthere are more than three, they are either equally impos-sible or only expand in quantity but do not change theirformal type” (Simmel, 1950: 145).

2019 217Halevy, Halali, and Zlatev

brokerage comes from other definitions (Fleming,Mingo, & Chen, 2007: 443; Stovel, Golub, &Milgrom,2011: 21326).

The aforementioned bifurcationwith regard to thethird aspect we derived from Simmel’s analysisspills over also to the fourth element identifiedabove. Definitions of brokerage as the occupation ofa structural position that enables bridging gaps in thesocial structure tend to focus on the facilitation ofsocial exchange and consider primarily the positiveeffects of brokers. For example, Marsden (1982: 202)defined brokers as “intermediary actors [who] facili-tate transactionsbetweenother actors lacking access toor trust in one another.” Likewise, Stovel and Shaw(2012: 141) emphasized brokers’ ability “to facilitateaccess to valued resources.”Whereas third parties canoften be helpful, we share Simmel’s (1950), Obstfeldet al.’s (2014), and others’ (Posner, Spier, & Vermeule,2010) views that third parties can also impact others’relationships negatively, for example, by spreadingaccurate and inaccurate negative reputational in-formation about others (i.e., gossip: Burt & Knez, 1995;Feinberg, Willer, Stellar, & Kletner, 2012; Wu, Balliet,& Van Lange, 2016).

Integrating structural definitions and process-oriented definitions of brokerage and brokering, wepropose that brokering can take the form of in-termediation, whereby the broker connects (either di-rectly or indirectly) two disconnected alters in theirnetwork, as well as the form ofmodification, wherebythe broker changes the nature of preexisting relation-ships between alters, for better or worse. Figure 1presents this integrative view of brokering processes.

Finally,with regard to the fifth elementwederivedfrom Simmel’s analysis, many treatments of broker-age and brokering acknowledge their multiple con-sequences for the focal actor (i.e., the broker or ego),the targets of brokering (i.e., the alters), and the or-ganizational unit(s) in which the relevant actors areembedded. However, different treatments of broker-age and brokering vary considerably in the extent to

which they emphasize the consequences for ego ver-sus the consequences for alters. The social capitalbranch of the networks literature, in particular, tendsto focus on “how ego gains advantage from the net-work around her” (Burt et al., 2013: 529), and high-lights that “relations with contacts in otherwisedisconnected groups provide a competitive advan-tage in detecting and developing rewarding opportu-nities” (Burt et al., 2013: 531). As noted by Stovel andShaw (2012: 140), “many contemporary scholarsvalorize theeconomicandcontrol benefits thataccrueto brokers as a result of their position,”which includemore positive evaluations of their ideas, higher pay,greater recognition, and faster promotions relative totheir peers (Burt et al., 2013; Stovel & Shaw, 2012).The focusonbrokerage’s consequences for egohas ledsome authors to assert that “althoughmuch is knownabout how brokering positions in social networkshelp individuals improve their own performance, weknow little about the impact of brokers on thosearound them” (Clement, Shipilov, & Galunic, 2018: 1;cf. Galunic, Ertug, & Gargiulo, 2012).

Whereas the social capital branch emphasizes theextrinsic rewards tobrokers,otherperspectivessuggestthat brokering may be an intrinsically rewarding ac-tivity. For instance, recent research suggested that in-dividualsmayengage insome formsof brokerage, suchas matchmaking, because doing so is intrinsically re-warding and promotes the broker’s happiness (Anik &Norton, 2014). In addition, Gould and Fernandez(1989: 91) noted that “an actor who facilitates trans-actions or resource flows” is considered a broker“whether or not the actor attempts to extract direct re-ward”andsuggested thatbrokerage’sconsequences forothersmayvarywith the functional role that thebrokerfulfills, which in turn depends on the subgroup mem-berships of the three actors in a triad. For instance,when all three parties share a common group mem-bership, the broker is viewed as a coordinator; wheneachof the threeparties in the triadbelongs toadistinctsubgroup, the broker is viewed as a liaison; and whenoneparty seeks access to a sub-group that the other twoparties share, the broker is viewed as a gatekeeper(Gould&Fernandez, 1989).Coordinators, liaisons, andgatekeepers influence others’ interactions and re-lationships in qualitatively different ways.

Summary

Although conceptualizations of brokerage andbrokering vary, researchers commonly agree thatnetwork structure sets the stage for brokering activity(Landis, Kilduff, Menges, & Kilduff, 2018; Obstfeld,

FIGURE 1An Integrative View of Brokering as Intermediation

and/or Modification


2017). Our review of the research on brokerage andbrokering identified the need for greater integrationbetween social structure and social process as a keytheme. To facilitate such integration across the dif-ferent literatures we reviewed, we introduce in thenext section an organizing framework that builds oninterdependence theory (IT) (Kelley&Thibaut, 1978;Rusbult & Van Lange, 2003) to delineate distinctbrokerage contexts and brokering processes in orga-nizations. We then use this novel framework to in-tegrate the insights that emerged from our review.

AN ORGANIZING FRAMEWORKFOR BROKERING

The premise of the Changing Others’ Relation-ships (COR) framework is that brokering activitiestarget others’ relationships. The raw material forbrokering is alters’ preexisting relationship, or lackthereof, and the outcome of brokering is alters’ re-lationship postbrokering.6 Thus, the effect of broker-ing is evident from the change to alters’ relationship,and it can be gaugedby comparing alters’ relationshippre-intervention versus postintervention (i.e., beforeversus after the brokering activity has taken place).This conceptualization of brokering activity allowsusto derive different functional forms that brokering inorganizations may take from a consideration of thedifferent ways in which a third party might influencedyadic relationships (which can then be scaled upbeyond the triad). The COR framework’s focus onchanging interpersonal relationships builds on IT,aclassic socialpsychological accountof interpersonaland group phenomena that is uniquely positioned tobridge social structure (i.e., occupying bridging posi-tions) and social processes (i.e., acts of intermediationor modification that shape others’ relationships).Hence, we begin by briefly introducing IT.

IT

IT focusesonoutcome interdependence, theextent towhich individuals’ interests are aligned versus mis-aligned and the extent to which individuals can influ-ence eachother’s outcomes, for better orworse, throughtheir actions. This notion of interdependence should bedistinguished from procedural interdependence, the

extent to which individuals’ work procedures are in-terwoven, and hence requires communication and co-ordination for effective team performance.

IT epitomizes theview that there is somethingbothprimary and unique about dyadic interactions andrelationships (Balliet, Tybur, & Van Lange, 2016;Halevy, Chou, & Murnighan, 2012; Kelley, Holmes,Kerr, Reis, Rusbult, & Van Lange, 2003; Thibaut &Kelley, 1959). IT describes and analyzes inter-personal situations in which two individuals arelinked in ways that enable them to influence theirown and each other’s outcomes and behavior. Theseinterpersonal situations create both the need for co-ordination and the possibility of conflict (Kelleyet al., 2003; Kelley &Thibaut, 1978). IT usesmultipledimensions to analyze dyadic interdependence sit-uations (Balliet et al., 2016; Reis, 2008). The currentanalysis builds on two essential aspects of IT: Thedegree of interdependence and the correspondenceof interests between parties (Kelley et al., 2003: 46).

The degree of interdependence dimension rangesfrom no interdependence (i.e., independence) tocomplete dependence and captures the degree towhich each person is dependent on the other (andconsequently how much power each person has tocontrol the other’s outcomes and influence theirbehavior). The correspondence of interests dimen-sion ranges from perfectly corresponding interests(correlation of 1) to perfectly conflicting interests(correlation of21), and is considered “the singlemostimportant property” of any interdependence situa-tion (Kelley & Thibaut, 1978: 117). Variation alongthis goal compatibility/incompatibility dimensionshapes the extent towhich individuals see each otheras partners versus adversaries and behave coopera-tively versus competitively (Halevy & Katz, 2013;Halevy & Phillips, 2015).

Although IT focuses on dyadic interactions and re-lationships, it explicitly acknowledges the manifoldways in which a third party can either help or harmdyadic interactions and relationships. Kelley et al.(2003) noted that “the large number of possible dyadicsituations. . .is dwarfed by the great variety of possibletriadic situations” (p. 394), a proliferation of inter-dependence patterns that interdependence theoristssolved by focusing on a subset of uniquely importanttriadic patterns. These patterns include, for instance,the formation of coalitions in which two individualsexclude the third; third-party intervention in dyadicconflict; and situations in which the third party’s“presence may threaten the unique intimacy thata dyad has developed and thus result in feelings ofjealousyon thepart of onememberof thepair” (p. 402).

6 Brokering processes can produce multiple outcomes(effects) on alters’ relationships, ego’s relationship witheach alter, ego’s social and material success, and more. Asa first step, we focus here on how brokering by ego in-fluences alters’ relationship.


COR

Building on IT’s focus on dyadic interactions andrelationships, we propose that brokering capturessocial influence processes whereby third parties useinformation (e.g., advice, feedback, and gossip) and/or incentives (e.g., praise and condemnation, mon-etary rewards andpunishments, and social inclusionand ostracism) to shape dyadic interactions and re-lationships. We propose that third parties can influ-ence, through their brokering activities, both of thefoundational dimensions noted earlier (degree of in-terdependence and correspondence of interests). Putdifferently, we propose that two essential aspects ofbrokering involve (a) third parties’ ability to create orterminate relationships between alters (i.e., changethe degree of interdependence) and (b) third parties’ability to change the sign of the relationship betweenalters (from positive to negative or vice versa, i.e.,change the correspondence of interests). Figure 2provides a schematic presentation of our COR orga-nizing framework based on these two aspects.7

The rows inFigure2capturealters’ relationshipbeforethe third party has pursued brokering activity. The col-umns in Figure 2 denote the nature of interdependencebetweenaltersafter thethirdpartyhaspursuedbrokeringactivity. Consistent with IT’s focus on the degree of in-terdependenceandcorrespondenceof interests, theCORframework classifies alters’ relationship pre-brokeringand post-brokering as either negative, neutral/none, orpositive. Crossing the rows and columns in Figure 2 de-lineates distinct functional forms of brokering.

Alters’ preexisting relationship, or lack thereof,provides the structural-relational context for bro-kering. When two alters have no preexisting re-lationship prebrokering, the context for brokering isone of an open triad. By contrast, when the two altershave a preexisting relationship prebrokering, thecontext for brokering is one of a closed triad. Thesestructural-relational conditions delineate the range ofpossible brokering processes. For instance, a thirdparty has an opportunity to introduce disconnectedalters to each other in an open triad but not in a closedtriad.A thirdpartymaynonetheless pursue brokering

activity in a closed triad to transform a weak tie intoa strong tie or vice versa. As another example, a thirdparty has an opportunity to act as a conflict managerwhen the two alters have a negative prebrokering re-lationship, but notwhen the two alters have apositiveprebrokering relationship. Thus, the nature of alters’relationship before the brokering activity determinesthe range of possible effects of brokering.

We label third parties in the three cells in the top rightcorner of the matrix (marked in green) helpful brokers asthey transform negative relationships into neutral or pos-itive relationships (as well as neutral relationships intopositive relationships). In all three cells, alters’ relation-ships benefit from the brokering activity relative to theirstatebefore it tookplace.Welabel thirdparties in the threeredcells in thebottom left cornerof thematrix (marked inred) harmful brokers as they transformpositive or neutralrelationships into negative relationships (as well as posi-tive relationships into neutral relationships). In all threecells,alters’relationshipssuffer fromthebrokeringactivityrelative to their state before it took place. Finally,we labelthirdparties in the threecellsonthediagonal in thematrix(markedinyellow)reinforcersastheymodifytheintensityof alters’ relationships (i.e., along the degree of in-terdependence dimension) rather than change its sign.Thesecellsrepresentthepossibilityofbrokeringprocessesto intensify or galvanize preexisting relations between al-ters noted by Simmel (1950: 135–136; cf. Krackhardt,1999), for instance, by turningweak ties into strong ties, orby turning unidimensional ties into multiplex ties(e.g., from task-related ties only, to task-related and socialties). Brokering activities can make either positive ornegative relationships more intense using information orincentives that reinforce sentiments and behaviors thatpreceded the brokering activity.

As Figure 2 shows, several streams of research thatfocus on third-party influence on interpersonal in-teractions and relationships can be integrated usingthe COR framework. The social networks literaturethat focuses on brokers as cooperation catalysts(Ahuja, 2000; Burt, 2004; Walker, Kogut, & Shan,1997) fits in the green cell inwhichbrokering activitytransforms no relationship (or neutral relationship)pre-intervention into positive relationship post-intervention. The third-party conflict managementliterature that focuses on arbitration, mediation, andother forms of third-party intervention in disputes(Ross & Conlon, 2000; Rubin, 1980; Sheppard, 1984)fits the green cells in which brokering transformsnegative relationships pre-intervention into neutralor positive relationships post-intervention.

The literature on “tertius gaudens” and “divideand rule” phenomena (Case & Maner, 2014; Logan,

7 We acknowledge that additional dimensions may beadded to Figure 2. For instance, the two-dimensional tablecan reasonably be transformed into a three-dimensionalspace that considers alsohowbrokering activity influencesthe relationship between ego and each of the two alters (aninfluence that may be symmetric or asymmetric). How-ever, for the sake of simplicity, we focus on a third party’sinfluence on alters’ relationship, recognizing that variousextensions can be deduced from our general framework.


2006; Posner et al., 2010) fits in the red cells inwhicha third party pursues brokering actions aimed attransforming neutral or positive relationships intonegative relationships. As our review of these liter-atures in the following sections reveals, some cells inthe matrix are populated by numerous empiricalinvestigations, whereas other cells attracted consid-erably less scholarly attention. Specifically, entre-preneurial functional forms of brokering receivedconsiderably more attention than the “dark side” ofbrokering (e.g., malicious workplace gossip). Thesegaps in scholarly knowledge about the less-studiedfunctional forms of brokering in organizations rep-resent opportunities for future research on this topic.Table 1 provides concrete examples from previ-ous research to illustrate helpful, harmful, andreinforcing brokering processes as conceptualized inFigure 2.

In sum, the COR framework8 elucidates behav-ioral, process-oriented definitions of brokering by

integrating concepts from social networks scholar-ship (e.g., open/close triad; weak/strong ties) withpsychological work emanating from IT (e.g., degreeof interdependence and correspondence of out-comes). It also facilitates the identification of distinctfunctional forms of brokering in organizations andprovides a straightforward operationalization of theeffects of brokering activities.

As Table 1 illustrates, here we use the CORframework primarily as a means to conceptually in-tegrate disparate literatures and organize our reviewof various programs of research that are presentlyscattered across the social sciences, thereby enrich-ing our understanding of brokerage and brokeringprocesses and promoting cross-fertilization betweendifferent streams of research.

Cooperation Catalysts: Creating and SupportingPositive Relationships

The vast majority of empirical studies we identi-fied have focused on brokerage positions rather thanon brokering processes. However, there is a growinginterest in how brokers broker (Lingo & O’Mahony,2010; Obstfeld, 2005; Quintane & Carnabuci, 2016),which fuels increased attention to the behavioralstrategies involved in brokering. Obstfeld (2017)distinguished between the brokering activities offormal and routinized brokers―such as those in-volved in the buying and selling of stocks and realestate—who operate in well-structured markets andare commonly recognized as scaffold for these in-teractions, and informal brokers whose continuous

FIGURE 2The COR Organizing Framework: Identifying Distinct Functional Forms of Brokering Based on Alters’

Relationships Prebrokering versus Postbrokering and Corresponding Literatures

8 Although we make simplified and categorical distinc-tions between negative, neutral, and positive relationshipsin Figure 2 for the purposes of illustrating the COR frame-work, we acknowledge that most interactions and re-lationships are mixed-motive in nature (i.e., include bothcooperative and competitive elements: Halevy et al., 2012;Schelling, 1980). In addition, correspondence of outcomesis in fact a continuous variable that can be operationalizedas the correlation between parties’ outcomes across thedifferent cell in apayoffmatrix (Kelley et al., 2003;Kelley&Thibaut, 1978), rather than as a categorical variable. Futurework may build on these observations to further developthe COR framework (cf. Gerpott et al., 2017).


“invisible labor” over coordinating interactions andexchanges among unacquainted parties is oftenoverlookedor downplayed.Henoted that “brokeringrelationships connect the supply chain, in whicheach firm in the chain brokers between its suppliersand downstream buyers by transforming inputs intooutputs, and thereby adding value. In this sense, allcommerce is inherently triadic, from Wal-Mart tothe hot dog vendor. Mergers and acquisitions mayappear to be dyadic pairings of the acquiring and theacquired, but in fact they do not proceed withoutthe intercession of brokers before the deal. . .duringthe deal. . .and after it” (p. 21). Both formal and in-formal brokers engage in activities that transformnonexisting (in open triads) or neutral (in closedtriads) relationships into positive relationships andtherefore fit together in the current section.

Brokering activities aimed at facilitating coordi-nation and cooperation take many different forms.For instance, they may entail promoting direct con-tact between the two alters or serving as the onlybridge between two alters without connecting themdirectly to each other (Soda et al., 2018). Promotingcoordination and cooperation sometimes requireconsiderable effort and even ingenuity by thirdparties who need to identify opportunities fortransferring information or resources from one con-text to another; selectively choose which ideas areworth pursuing and which relationships can poten-tially be fruitfully cultivated; translate, frame, or

otherwise transform the information that has par-ticular meaning and value in context A into usefulinformation in context B; choose whether or not tofacilitate direct contact between the alters; considerhow to present the collaborative pursuit to each ofthe parties; decide whether to charge the brokeredparties for their helpful brokerage; determinewhetherthey need to continuously monitor and take actionsto sustain the relationship; and more (Burt, 1992;Podolny & Baron, 1997).

Recent researchbyQuintane andCarnabuci (2016)integrated actors’ structural position in the networkwith their information exchange patterns to enhanceunderstanding of how brokering processes unfold.These authors analyzed e-mail communications be-tween employees of two organizations―a project-based, digital advertising agency in the Netherlandsand an information technology recruitment firm inAustralia. In both contexts, the extent to which em-ployees’ structural position in the network involveddense ties versus opportunities to bridge acrossstructural holes shaped their brokering activities.Relative to densely embedded employees, actors inbrokerage positions were significantlymore likely toengage in unembedded brokering (i.e., in short-terminteractions with weak ties). In addition, when en-gaging in unembedded brokering, these individualswere more likely to act as intermediaries betweenalters (i.e., pursue a tertius gaudens strategy) than tofacilitate direct contact between alters (i.e., pursue

TABLE 1Examples of Different Functional Forms of Brokering Explored in Previous Research

Function Brokering Behavior Illustrating source

Helpful brokers Integrating knowledge dispersed across many employees and connectingdisconnected employees in the knowledge network to promote inventive output

Grigoriou and Rothaermel (2014)

Facilitating interactions between different role holders (e.g., performers, personalmanagers, songwriters, musicians, and studio executives) to produce music

Lingo and O’Mahony (2010)

Using incentives to motivate cooperative behavior, thereby cultivating enduringnorms of cooperation that outlast the intervention period

Nakashima et al. (2017)

Acting as hostile mediators, who mistreat disputants with rudeness and hostility, tofacilitate a sense of common fate andconsequently collaboration amongdisputants

Zhang et al. (2017)

Harmful brokers Engaging in negative gossip to promote ostracism and compel a particular individualwho does not follow group norms to leave the group

Kniffin and Wilson (2005)

Creating a work environment that undercuts a highly skilled subordinate’sopportunities to communicate and cooperate with other group members

Case and Maner (2014)

Mixing employees with incompatible interests in the same work unit to promotefriction and subgrouping and avoid unionization

Posner et al. (2010)

Promoting fear anddistrustwithin aworkforce andusing threats of targeted layoffs orgeneral downsizing to undermine collective efforts to unionize

Dundon (2002)

Reinforcers Transferring information between alters as a conduit without attempting to changealters’ relationship

Obstfeld (2017)

Supporting ongoing interdisciplinary scientific collaborations usingmethodologicalexpertise that is sought by diverse partners

Kaplan et al. (2017)


a tertius iungens strategy). However, when engagingin embedded brokering involving alters with whomthey had long-term ties, these individuals weremorelikely to facilitate direct contact between alters (ter-tius iungens) rather than to act as intermediaries(tertius gaudens). These findings highlight the im-portance of taking a comprehensive approach thatconsiders both structural positions (brokerage) andinteractive processes (brokering) when studyinghow actors create or reinforce positive relationshipsin organizations.

Additional evidence for the importance of a tertiusiungens strategy, which facilitates direct contactbetween alters as a means to create or reinforcepositive relationships, comes from Obstfeld’s (2005)multimethod study of innovation in the automotiveindustry. Using surveys, interviews, and ethno-graphic observations, Obstfeld (2005) explored therole of third parties who connect alters in drivingorganizational innovation as assessed with self-reports, expert reports, and managers in the firm.The innovations explored included both productinnovations (e.g., clutch interlock defeat) and pro-cess innovations (e.g., creation of a prototype partsmanagement group and process) and spanned engi-neers, designers, and managers across seven unitswithin a large engineering division. Individuals whoscored higher on a self-report measure of tertiusiungens orientation (which included items such as “Iintroduce people to each other who might havea common strategic work interest” and “I see op-portunities for collaboration between people”)showed significantly higher contributions to orga-nizational innovation controlling for a host ofvariables (e.g., social network density/constraint,education and organizational rank, and technicalknowledge). In addition, social knowledge in thefirm, conceptualized as access to informal in-formation about activities and processes in variousunits, but not technical knowledge, emerged asa significant predictor of contributions to in-novations. This highlights the importance of gainingaccess to informal knowledge as a precursor to ef-fective brokering.

Research on brokering in the context of open in-novation communities illustrates important dis-tinctions between bridging activities that occurwithin work units versus across work units. Flemingand Waguespack (2007) analyzed the role that hu-man capital (technical contributions) and socialcapital (brokering and boundary spanning activities)play in ascendance to leadership positions inan open technical innovation community. Using

published proceedings of the Internet EngineeringTask Force, requests for comments publications, andinterviews with community leaders, these authorsfound that although different intermediary activitiesincrease the likelihood that brokers ascend to lead-ership positions, they follow different patterns.Specifically, brokering had a positive effect onleadership ascendance within technological bound-aries and when accompanied by frequent physicalcontact (operationalized as increased conferenceparticipation). Boundary spanning, conceptualizedas facilitating integration across cohesive techno-logical communities (operationalized as contribut-ing to technical publications by multiple workgroups), was a stronger predictor of leadership as-cendance;was not contingent on face-to-face contactin conferences; and had a negative interaction withintra-unit brokering in shaping leadership ascen-dance. These findings suggest that the informationand control advantages that come from occupyingbrokerage positions are contingent on the particularforms that brokering activities take and on the extentto which these activities occur within versus acrossgroup boundaries (cf. Gould & Fernandez, 1989).

Whereas some brokering activities that transformnonexisting or neutral relationships into positiverelationships generalize across organizational con-texts, other brokering activities are context-specific.A recent field study of boundary spanning activitiesin an interdisciplinary nanotechnology researchcenter exemplifies such context-specific bridgingactivities (Kaplan et al., 2017). In this university-based organizational environment, PhD studentsand postdoctoral fellows frommultiple departments(including chemistry, physics, materials science,mechanical engineering, and medicine) served asboundary spanners who facilitated interdisciplinaryresearch collaborations in the face of cognitive andpolitical/economic barriers to such research pro-jects. They did so by developing symbiosis withnovel instruments—atomic force microscopes andscanning tunneling microscopes—and their exper-tise in operating these instruments created newaffordances for research collaborations among alterswhowere disciplinary actors (e.g., faculty in specificdepartments) that lacked these students’ specializa-tionwith the novel instruments (Kaplan et al., 2017).

An interesting effect that emerged from theKaplanet al. (2017: 1397) study is that interdisciplinarypublications were likely to get fewer citations thanintradisciplinary publications. This finding mirrorsfindings by Fleming et al. (2007) in the context ofcollaborative inventions of utility patents. These


authors found that brokering facilitated access toknowledge and ideas and thus created opportunitiesfor novel anduseful recombinations of ideas, therebypromoting collaborative creativity. However, con-sistent with the view that cohesion in networkspromotes trust and mobilization, these authorsfound that “ideas that arise from brokered collabo-ration are less likely to be used in the future” (p. 464).Although brokered collaborations increase genera-tive creativity (operationalized as patents that usetechnologies from previously uncombined pairs ofsubclasses of technologies), “conditional on gener-ating a new idea. . .brokered collaborations decreasethe use of that creativity by other inventors” (p. 464).These findings by Kaplan et al. (2017) and Fleminget al. (2007) emphasize the need to consider not onlythe effects of brokering on idea generation but alsothe longer term consequences of brokering for theutilization and adoption of the products that resultfrom brokered collaborations (i.e., the transitionfrom creativity to innovation; cf. Kauppila, Bizzi, &Obstfeld, 2018).

A common thread that runs through the literatureon brokering concerns the tendency of alters to re-spond to brokering activities with distrust (Fleming&Waguespack, 2007; Podolny&Baron, 1997). Stovelet al. (2011) noted that brokering is an inherentlyfragile process due to three of its essential charac-teristics. First, alters that rely on brokering to facili-tate their exchange typically have social gapsbetween them, which means that they lack thestructural support system that a dense network ofstrong ties to the same actors provides. In the absenceof such shared ties to others who can reinforce therelationship, attest to the broker’s trustworthiness,and potentially also monitor and sanction the bro-ker’s behavior, trust is bound to be lower. Second, asboundary spanners who exist and operate betweendifferent social or organizational worlds, brokers aretrusted less because of the dual nature of their loyaltyand commitment to more than one group. General-ized trust is typically extended within groupboundaries (Yamagishi & Kiyonari, 2000) yet bro-kering activities often take place outside of suchboundaries. Third, brokering activities provideintermediaries with opportunities to exploit alters(e.g., by withholding some information, providingmisinformation, or charging increasingly more fortheir services), especially when their exclusive ac-cess to alters gives them a monopolist brokeragestatus. The possibility of exploitation through thebrokerage position undermines confidence in thebroker (Stovel et al., 2011).

In addition to thedistrust by the very alters that egoseeks to connect, it is important to note that broker-ing activities that promote contact and coordinationamong certain alters (e.g., A1 andA2) simultaneouslyreinforce the separation between these alters andothers in the organization who were excluded fromthe interaction (e.g., between A1 and A2 on the onehand and A3 and A4 on the other hand). As Obstfeld(2017) notes, any effort to coordinate the actions ofa subset of the potentially relevant organizationalactors is inclusive for those invited to takepart (in thedinner party, the group meeting, or the recruitingcommittee), and exclusive for all others, suggestingthat one can conceptualize brokering activities byintermediaries as a form of coalition building. Al-though the COR framework distinguishes differentfunctional forms of brokering activities for analyticalpurposes, we heed Obstfeld’s (2017) argument thatany act of selective inclusion is by definition also anact of selective exclusion, andexplicitly acknowledgethat different brokering activities are interrelated,coexist in organizations, and can simultaneously im-pact different actors’ relationships in different ways.

The merging of divergent, yet complementary,brokering activities is illustrated in Lingo andO’Mahony’s (2010) ethnographic field study of in-dependent country music producers in Nashville.These producers behaved as strategic actors whoused awide repertoire of specific brokering tactics tomanage others’ interactions and relationships: Theykept some parties apart and brought others together,simultaneously facilitating coordination and co-operation between some alters and cultivatingcompetition among others. Producers facilitated in-teractions between performers, their personal man-agers, songwriters, musicians, production studiostaff (e.g., engineers), and labels; coordinated ex-pectations and managed ambiguities concerning thescope of projects, the production process, andparties’ role responsibilities and jurisdictions; andwere responsible for gathering resources, mobilizingconcerted action, and synthesizing inputs from dis-connected others. For example, when working withnew and unfamiliar performers, producers adopteda tertius iungens orientation and facilitated directcontactwith songwriters, in thehopes that apersonalconnection will increase the likelihood that song-writers would give their best material to the newartist. To attract resources for such new artists, pro-ducers created competitions between labels, for in-stance, by bringing “together several label heads ata showcase concert” by their performer (p. 62). Lingoand O’Mahony’s (2010) analysis underscores the


critical role that independent producers play inshaping every stage of the social organizing processthat leads to the final product, from making in-troductions among disconnected parties, to makingsure that different alters do not challenge eachother’s expertise at the studio, to using their uniqueexpertise and position to help production engineersshare their own and others’ vision and inter-pretations during mixing sessions, and to workingwith artists, managers, and label executives to curatethe final product.

Whereas Lingo and O’Mahony (2010) focused onproducers’ brokering activities, Clement et al. (2018)explored the externalities that brokering activities byother actors create for producers of creative projects.These authors theorized that brokering activities byhubs—defined as individuals who connect distinctcommunities within an industry—can create posi-tive and negative externalities for executives whomanage creative projects. They proposed further thatthe extent to which hubs’ brokering activities createa public good (i.e., improving others’ outcomes bycontributing to creative projects’ success) or a publicliability (i.e., hurting others’ outcomes by under-mining creative projects’ success) depends on thefunctional role of the hub’s “neighbors.” Clementet al. (2018) found support for these hypotheses us-ing archival data from the French TV game showproduction industry. Specifically, whereas hubspositively influenced creative directors’ contribu-tions to the success of TV shows, they negativelyinfluencedproducers’ contributions to the success ofTV shows (operationalized as viewership scores).

Hubs’ brokering activities positively influencedcreative directors’ contributions via two processes:(1) They used their cross-community contacts tobenefit the focal project when they were teammembers working alongside the creative director,and (2) they facilitated knowledge spillovers throughcross-community brokering (e.g., contributing novelideas) when they were not team members on the fo-cal creative project. Hubs’ brokering activitiesharmed producers’ contributions to the success ofcreative projects primarily by undermining pro-ducers’ ability to effectively coordinate the con-tributions of different members. Given hubs’involvement with multiple projects across commu-nity boundaries, they often showed lower commit-ment to any given project, causing delays, friction,and conflict. In addition, Hubs’ brokering activitiesalso harmed producers’ contributions to the successof TV shows through a secondary process: Evenwhen hubs were not used on the focal creative

project, the overlap in personnel and high in-terdependence between different production teamsmeant that the delays and conflicts they created oncertain projects had a ripple effect on other creativeprojects in the same community. These findings byClement et al. (2018) illustrate the benefits of con-sidering brokerage positions and brokering activitiestogether, and highlight that the effects of brokeringactivities on the collaborative pursuit of creativeprojects are complex and depend in part on the na-ture of others’ jobs.

THIRD-PARTY CONFLICT MANAGERS:TURNING NEGATIVE RELATIONSHIPS INTONEUTRAL OR POSITIVE RELATIONSHIPS

Conflict in organizations is pervasive, diverse, anddestructive (Anicich, Fast, Halevy, &Galinsky, 2015;Halevy, Cohen, Chou, Katz, & Panter, 2014). Al-though organizational conflicts take different forms,all forms of organizational conflict entail elements ofincompatibility among interdependent parties—discordant interests, claims, perspectives, orvalues—that require management or resolution. Theuse of third-party help to manage and resolve in-terpersonal conflicts atworkhas beenconceptualizedas a distinct conflictmanagement style (i.e., alongsidethe dominating, compromising, accommodating, in-tegrating, and avoiding conflict management styles:Ting-Toomey,Oetzel, &Yee-Jung, 2001) and linked tovarious potential benefits for disputants. For exam-ple, a study in health-care organizations in the Neth-erlands found that high levels of third-party helpduring workplace conflict reduced the harmful con-sequences of conflict-induced stress, including em-ployees’ emotional exhaustion, absenteeism, andturnover intentions. These positive effects of third-party help held controlling for social support fromcoworkers and supervisors (Giebels & Janssen, 2005).These important benefits notwithstanding, the keyquestion of interest here concerns third parties’ pro-pensity and ability to take actions that transform al-ters’ negative relationship into neutral or positiverelationships.

Third-party conflict managers vary on multipledimensions such as whether they have stakes in theconflict,whether theyhavepreexisting relationshipswith the disputants, whether they are supervisors orpeers of the disputants, and more (Karambayya,Brett, & Lytle, 1992; Pinkley, Brittain, Neale, &Northcraft, 1995; Sheppard, 1984). Two cardinaldistinctions that run through the third-party conflictmanagement literature are between formal and


informal third parties (Lewicki, Weiss, & Lewin,1992), and between third parties who possessprocess-control versus outcome-control (Thibaut &Walker, 1975). These fundamental distinctions un-derlie the kinds of brokering activities that thirdparties pursue to change others’ relationships.

Formal third-party conflict managers include me-diators, arbitrators, ombudspersons, and other orga-nizational actors whose role specifically identifiesthem as responsible for (and having authority over)dispute resolution processes in the organization(Morrill &Rudes, 2010). Informal third-party conflictmanagers include supervisors, peers, andotherswhodo not have legal or organizational authority overconflict management processes (or extensive train-ingandexperience indoing so), andare typically freeto choose whether, when, and how they wish to in-tervene inothers’disputes. Informal thirdparties canintervene in others’ conflicts without being invitedto do so by the disputants; are not required to beneutral or external to the conflict; do not have tofollow predefined procedures; and may use theprocess to advance their own personal interests, thedisputants’ interests, and/or organizational interests(Conlon, Carnevale, & Murnighan, 1994; Pinkleyet al., 1995).

Third-party conflict managers who possessprocess-control influence the mechanisms throughwhich disputants manage their disagreement, withmediation being the quintessential process-controlmode of third-party conflict management (Lewicki &Sheppard, 1985; Shapiro & Brett, 1993). By contrast,third-party conflict managers who possess outcome-control have the power to determine how the disputewill be resolved, with arbitration representing thearchetypal outcome-control mode of third-partyconflict management (Lind, Walker, Kurtz, Musante,& Thibaut, 1980; Ross & Conlon, 2000).

Extant research on third-party intervention inothers’ conflicts tends to focus on the factors thatshape either how a third party intervenes or howdisputants react to the third party’s intervention.Examples of the first line of research include studiesthat explored when and why third parties show re-tributive tendencies (Skarlicki & Rupp, 2010), adoptan autocratic style (Karambayya et al., 1992), balancepower by siding with the weaker party in a dispute(Laskewitz, Vliert, & Dreu, 1994), and how organi-zational characteristics and conflict characteristicsinfluence managers’ choices of intervention strate-gies (Kozan, Eegin, &Varoglu, 2007). Examples of thesecond line of research include studies that exploredhow the kind of third party (e.g., peers, bosses, and

ombudspersons) and the mode of third-party in-tervention (e.g., mediation, arbitration, and hybridstrategy) influence disputants’ behavior and theirjudgments of fairness and justice. For instance,Arnold and O’Connor (1999) found that expert thirdparties’ recommendations, but not peers’ recom-mendations, influenced disputants’ offers, andShapiro and Brett (1993) found that U.S. coal minerswhose grievances were assigned to mediation expe-rienced higher levels of outcome control, processcontrol, and third-party fairness than coal minerswhose grievances were assigned to arbitration.

Helpful Brokering Processes by Third-PartyConflict Managers

Research by Karambayya and Brett (1989) foundthat MBA students who role-played managers in-tervening in subordinates’ disputes were more ef-fective in facilitating cooperative solutions whenthey asked questions, requested the parties to submitproposals, and tried to integrate their ideas withdisputants’ ideas, relative towhen they “used threatsand incentives, predicted probable outcomes if thedispute could not be settled at the meeting, andexerted pressure on one or both disputants to en-courage a timely settlement” (p. 686). Whereas thesefindings by Karambayya and Brett (1989) seem tosuggest that brokering activities that use informationmay outperform brokering activities that use in-centives in transforming others’ relationships, sub-sequent research highlighted the benefits of usinghybrid processes, which use both information andincentives, in the conflict management process(Conlon, Moon, & Ng, 2002).

Recently, researchers suggested that third partiescan effectively transform negative relationships intopositive relationships by adding hostility and rude-ness to an already contentious conflict situation(Zhang, Gino, & Norton, 2017). In a series of onlineand in-laboratory experiments, using both deal-making and dispute simulations, the researcherscompared the effectiveness of hostile mediators—who use rude and offensive communications(e.g., “. . .you have sufficiently wasted my time. . .,”“. . .let’s seewhat kind of annoying complaints are onthe table today”)—with that of nice mediators (aswell as neutral mediators, in a subset of thestudies)—who communicated understanding anddisplayed interpersonal warmth. This research con-sistently found that hostile mediators significantlyincreased disputants’ willingness to reach agree-ment relative to nicemediators. The positive effect of


hostilemediators on disputants’willingness to reachagreement was mediated by a sense of common fateamong the disputants, who shared the experience offacing a common enemy and hence experienced re-duced social distance with each other. The re-searchers showed further that only mediators whoare hostile to both disputants transformed competi-tion to cooperation. Mediators who were hostile tojust one of the disputants were not perceived asa common enemy, and therefore did not promotecooperative behavior amongdisputants (Zhang et al.,2017).

Our own researchhas examined both third parties’willingness to use the resources they control to in-centivize cooperative behavior in conflict situationsand disputants’ reactions to the possibility of suchthird-party intervention (Halevy & Halali, 2015;Nakashima, Halali, & Halevy, 2017). To experimen-tally study triadic interactions, we created a groupdecision-making task called the Peacemaker Game,in which two disputants choose whether to co-operate or compete, and a third party chooseswhether or not to introduce side payments that re-ward cooperative behavior and punish competitivebehavior. From disputants’ perspective, these sidepayments (i.e., incentives) effectively transform thesituation they face from a competitive situation (aone-shot Prisoner’s Dilemma game, in which eachparty’s profit-maximizing strategy is to compete re-gardless of what they expect the other party to do)to a cooperative situation (a one-shot MaximizingDifference game, in which each party’s profit-maximizing strategy is to cooperate regardless ofwhat they expect the other party to do). From thethird party’s perspective, the option to intervene inthe dispute presents a risky prospect. Withholdingintervention is a safe strategy that allows the thirdparty to keep their resources (the time, effort,money,or reputation they could invest in the conflict reso-lution process). By contrast, intervening in the con-flict can result either in gains or losses relative to thesmall fixed payoff associated with nonintervention.Third parties in the Peacemaker Game get a positivereturn on their investment if they intervene and ifbothdisputants choose to cooperate, yet they lose theresources invested in the intervention if they in-tervene and if both disputants choose to compete.

In a series of experiments, the mere possibility ofthird-party intervention significantly increased co-operation rates in both interpersonal conflicts (be-tween two individuals) and intergroup conflicts(between two three-person groups); moreover, thehigher the likelihood of third-party intervention, the

greater the disputants’ propensity to cooperate.These experiments showed further that self-interestplays an important role in motivating helpful bro-kering. In one of the experiments, we systematicallymanipulated the possible consequences of third-party intervention for the broker. Third-partyintervention rates were less than 8 percent whenthird-party intervention could not produce a gain forthe broker (i.e., when it was altruistic), at 35 percentwhen nonintervention and intervention resulted inthe same fixed payoffs for the broker, at 39 percentwhen the intervention could result in either gains orlosses to the broker relative to nonintervention, andexceeded 80 percent when the intervention couldnot produce a loss to the broker. The role of self-interest in third-party intervention decisionsreceived further support from two surveys that ex-amined third parties’ motives in real-world in-teractions involving friends and coworkers (Halevy&Halali, 2015). These findings highlight the fact thathelpful brokering does not necessitate altruisticmotives. Rather, consistent with research showingthat self-interest can sometime promote prosocialbehavior (Zlatev & Miller, 2016), self-interested ob-servers of conflict in organizations may choose tointervene to the extent that they believe that doing sowill benefit them.Byusing the resources they controlto reward cooperative behavior and punish com-petitive behavior they can transform interactionsand facilitate win-win-win outcomes for all partiesinvolved.

Reasoning that conflict often spans an extendedperiod of time and repeated interactions betweendisputants and third parties, we subsequently usedamultiround version of the Peacemaker Game to testthree hypotheses. The first hypothesis postulatedthat third parties who observed a history of compe-tition between adversaries would be less likely tointervene in a dispute than third parties who havenot observed such a history of competition. Thesecond hypothesis asserted that third-party in-tervention that changes the nature of outcome in-terdependence between parties will be successful inpromoting cooperation even when it is introducedlate in the repeated interaction, that is, followinga history of competition (cf., Halevy, Weisel, &Bornstein, 2012). Finally, the third hypothesis pro-posed that third parties who intervene early in thecourseof the repeated interactionwouldpromote thedevelopment of sustainable cooperative norms thatwould outlast the intervention period. That is, itpredicted that even after the third party canno longer incentivize cooperation, the disputants


would continue to cooperate at high levels(Nakashima et al., 2017).

To test these hypotheses, we assigned three-person groups to one of two conditions. In the lateintervention condition, participants assigned to playthe role of the third party observed disputants’ de-cisions in rounds 1–30 of the repeated game, andcould only intervene in rounds 31–60 of the repeatedgame (disputants and third parties only learnedabout the possibility of third-party intervention afterround 30). In the early intervention condition, par-ticipants assigned to play the role of the third partycould intervene in rounds 1–30of the repeated game,and could no longer do so in rounds 31–60 (dispu-tants and third parties only learned this after round30). Lending support to the first of the three afore-mentioned hypotheses, third parties’ interventionrates were significantly higher in the early in-tervention condition, in which they had not ob-served a history of conflict between the twoparticipants assigned to play the role of disputants,than in the late intervention condition. Lendingsupport to the second hypothesis, introducing thepossibility of third-party intervention following 30rounds of conflict immediately and powerfully in-creased cooperation rates from around 40 percent toaround 80 percent. Lending support to the third hy-pothesis, cooperation rates in the early interventioncondition did not decrease in rounds 31–60, com-pared with the first 30 rounds, despite the fact thatthird parties could no longer introduce side pay-ments to reward cooperative behavior and punishcompetitive behavior. These findings show thatearly third-party intervention can set in motionmutually beneficial behavioral norms that go beyondmere compliance and persist over time. Whereasprevious research suggested that removing exoge-nous cooperation-promoting mechanisms, such ascontracts, can undermine interpersonal trust (be-cause the parties attribute past cooperation to thepresence of the contract: Malhotra & Murnighan,2002; cf. Chou, Halevy, Galinsky, & Murnighan,2017), the fact that cooperation persisted in the re-peated Peacemaker Game after the third party hadexited the relationship shows that early-stagehelpfulbrokering can facilitate trust development that en-dures in time (Nakashima et al., 2017).

Additional evidence for the endurance of collab-orative norms introduced by helpful brokers comesfrom a unique archival study by Samila, Oettl, andHasan (2016). These authors reasoned that helpfulbrokering can strengthen collaborative ties via bothactive third-party conflict resolution and the

encouragement of cooperative norms. Using data ondyadic research collaborations by immunologistswho lost a third coauthor to unexpected death, thisresearch showed that dyadic scientific collabora-tions who lost a helpful third (assessed via ac-knowledgments of the departed third party in otherpapers) weremore durable than those who lost a lesshelpful third. These findings lend further support tothe idea that early-stage helpful brokering by a thirdparty can set in motion longstanding cooperativenorms between alters.

DIVIDE AND CONQUER: UNDERMININGOTHERS’ RELATIONSHIPS

The previous two sections focused on intermedi-aries and conciliators, whose brokering activitiesspur positive relationships among alters. However,brokers can also serve as “dividers”—actors whoengage in harmful third-party intervention intendedto undermine cooperation and promote friction andconflict between others. We review here empiricalevidence on such harmful brokering from the work-place gossip and labor relations literatures.

Workplace Gossip

Gossip, the communication of “evaluative commentsabout someone who is not present in the conversation”(Foster, 2004: 78), is an extremely common everydaysocial behavior (Giardini, 2012), with some empiricalwork estimating that up to two-thirds of conversationsinclude references to absent parties (Dunbar, 2004;Emler, 1994). Workplace gossip, defined as “informaland evaluative talk in anorganization, usually amongnomore than a few individuals, about another member ofthat organizationwho is not present” (Kurland & Pelled,2000: 429), features prominently in social network re-search, which has considered workplace gossip both asan antecedent and as an outcome of social networkphenomena. For example, a longitudinal study of gossipand friendship networks in a Dutch nonprofit organi-zation (Ellwardt, Steglich, & Wittek, 2012) found thatgossiping was not only reciprocated by friendship nom-inations (rather than the other way around) but also thatexcessive gossiping decreased friendship nominationsover time, suggesting that the social capital account ofgossip has merit but also bounds. As another example,a study in a U.S. manufacturing company (Grosser,Lopez-Kidwell,&Labianca,2010) foundthat themoreanemployeeengages ingossip, themore informal influencetheyhave in theeyesof coworkers, but the lesspositivelytheir work performance is evaluated by their supervisor.


Thus, engaging in gossip influences not only the repu-tation of the object of gossip; rather, but also has conse-quences for the sender and receiver of gossip.

Workplace gossip takes different forms and servesdifferent functions. For example, workplace gossip canbe positive (e.g., praising someone as creative and hard-working, which functions as a form of indirect socialsupport)ornegative (e.g., condemningsomeoneasa free-rider and cheater, which functions as a form of socialundermining). Gossip can serve to strengthen the tie be-tween the sender and the receiver of the gossip (i.e., anaffiliative function), aswell asbetween the receiverof thegossip and the object of the gossip (e.g., recommendinga new colleague as reliable), or weaken the tie betweenthe receiver of the gossip and the object of the gossip(i.e., a social undermining function from the sender’sperspective; Duffy, Ganster, & Pagon, 2002). In line withour focus in this section on divisive behavior by thirdparties, we focus here on negative workplace gossip,9

defined as the communication of unfavorable in-formation about an organizational member in their ab-sence, which serves to hinder relationships between thereceiver and the object of gossip (Ellwardt, Labianca, &Wittek, 2012).

Negative workplace gossip is well documented inthe social networks literature. For instance, Burt andKnez (1995: 275–276) noted: “These are third partiesclose to ego and distant from alter; contacts moreexclusive to ego, confidants on ego’s side viewingalter as a distant contact. . .these exclusive thirdparties. . .are a more willing conduit for negativestories about alter. . .negative stories accumulatewith them....” Hence, consistent with Heider’s cog-nitive balance theory (Heider, 1946), exclusive thirdparties who possess strong positive ties with the re-ceiver of the gossip andweak ties, negative ties, or noties with the object of the gossip communicate in-formation that serves to amplify distrust between thereceiver and the object of gossip.Negativeworkplacegossip can thus function to reduce cooperation be-tween the receiver and object of gossip, and evenfacilitate turnover by employees who develop par-ticularly negative reputations (Burt, 2005; Kniffin &Wilson, 2005).

Negative workplace gossip can serve to warn thereceiver about the object’s untrustworthiness or ex-ploitative behavior (Feinberg et al., 2012), and

promote group goals by curbing opportunistic be-havior (e.g., free-riding, cheating) by groupmemberswho worry about negative reputations, social ex-clusion, and ostracism (Feinberg, Willer, & Schultz,2014; Kniffin & Wilson, 2010; Wu et al., 2016). Con-sistent with this norm reinforcement function, thereis converging evidence that negative gossip tends totake a scapegoating pattern by targeting a relativelysmall number of low-status group members. For in-stance, a study of gossip among the employees ofa Dutch nonprofit organization found that “in thenegative gossip network, centralization was almosttwice as large. . .as in the positive gossip net-work. . .suggesting that negative gossip was centrallystructured around star-like objects (“scapegoats”)”(Ellwardt et al., 2012: 200). Similarly, a study ofgossip in rowing teams in aU.S. university byKniffinandWilson (2005) found that negative gossip tendedto focus on a single target, labeled “the slacker,”wholeft the team following a semester inwhich theywerethe primary target of negative gossip.

Consistent with the COR framework’s focus onoutcome interdependence between organizationalactors, research on workplace gossip suggests thatoutcome interdependence plays a critical role inshaping patterns of gossip. The aforementionedstudy by Ellwardt et al. (2012) postulated that highlevels of task and outcome interdependence withinformal organizational units (e.g., work teams) in-crease the tendency to gossip about in-group mem-bers (i.e., members of the same work unit) relative toout-group members (i.e., members of other workunits in the organization). Indeed, employees tendedto gossipmore (both positively andnegatively) aboutcoworkers from the same work unit than about co-workers from other units, even after controlling forthe higher contact frequency and higher rates offriendships within work units. These authors con-cluded that “interdependence between employees isa predictor of any type of gossip about group mem-bers” (p. 203, italics in source). This assertion is inlinewithKniffin andWilson’s (2005: 279) suggestionthat “the degree of common fate shared by a group’smembers influences the degree to which gossip isused as an instrument of social control.”

In sum, negative workplace gossip can have self-serving, social influence, and group-serving func-tions. These functions are not mutually exclusive;rather, they often go hand-in-hand. Specifically,negative gossip can simultaneously strengthen tiesbetween the sender and receiver of gossip, un-dermine the ties between the receiver and object ofgossip, and enforce group norms.

9 Positive workplace gossip (i.e., communicating favor-able information about others) is often used by cooperationcatalysts as a means to build relationships, reinforce pos-itive relationships, or facilitate constructive conflictmanagement.


Labor Relations

Harmful brokering recurs also in the field of laborrelations, in which labor consultants and other thirdparties sometimes act to exacerbate tensions be-tween management and the workforce or promoteinternal friction among the employees who seek tounionize. The explicit aim of these efforts is to sup-plant unity with discord in the hopes of thwartingcollective action toward unionization. Union bust-ing practices were pervasive throughout most of thesecond half of the 20th century in the United States,despite the illegal status of some of them under theNational Labor Relations Act. Union avoidancecompanies’ list of clients during that time periodspanned multiple sectors, industries, and geo-graphical regions in the United States (Logan,2006: 655).

An insider’s perspective on the practices used inthe so-called “union busting” business comes fromLevitt (1993)who sharedhis firsthandexperiences ina memoire: “The only way to bust a union is to lie,distort, manipulate, threaten, and always, alwaysattack. . . a combined strategy of disinformation andpersonal assaults. . . .the consultants’ attacks are in-tensely personal. . .they invade people’s lives, de-molish their friendships, crush theirwill, and shattertheir families. . . .the enemy was the collective spi-rit. . .I poisoned it, choked it, bludgeoned it. . . I taughtthe supervisors to despise and fear the union. I per-suaded them that a union-organizing drive wasa personal attack on them. . . .Although I took on thesupervisors face to face, my war on the union activ-ists was covert” (pp. 1–2). Levitt explicitly considersthe tactics that he and his colleagues used to fallwithin the realm of divide-and-conquer. Recallingan instance of harmful third-party intervention incoal mines in the U.S. Midwest, he writes: “Twomonths had gone by since a handful of well-dressedstrangers had walked into the. . .Company, bearingpoison and promises. . .Men who had worked likebrothers for years—some were brothers. . .had star-ted to take blows at each other’s heads and sayingnasty things about each other’s wives. Some hadstopped talking altogether. . . .The workers were sodivided, some could not stand next to each other inthe pit without starting a fight” (p. 7).

Scholarly research on the union avoidance in-dustry lends credence to Levitt’s first-person expe-riences and reinforces the notion that many of theharmful brokering activities that third parties some-times pursue in this context are covert rather thanovert (Dundon, 2002; Godard, 2009; Hurd &Uehlein,

1994; Logan, 2006; O’Sullivan & Gunnigle, 2009;Posner et al., 2010). Importantly, this literature alsocorroborates the view that these actors are in factthird-party actors rather than merely an extension ofone of the parties (i.e., management) in a labor dis-pute. For example, Logan (2006: 652) noted: “. . .theunion avoidance industry had developed intoa multimillion-dollar concern that profited frompromoting adversarial labor—management re-lations, and consultants had become important in-dustrial relationsactors in their own right. Theywereno longer simply responding to employer demandsfor their services, but were actively and aggressivelycreating that demand by encouraging managementto fear the allegedly catastrophic consequences ofunionization. . ..” In line with our conceptualizationof this form of harmful brokering as divide-and-conquer, Logan (2006: 659) noted that unionavoidance is “an industry that profits from pro-moting conflict in labor–management relations”(italics added).

Some labor consultants use both information andincentives as means to transform positive or neutralrelationships into negative relationships (Posneret al., 2010). For instance, sabotaging communica-tion channels and spreading rumors involve ma-nipulating information, whereas paying bribes andadministering penalties involve the use of in-centives. It is important to note that, whereas someincentives are designed to suppress unionization(e.g., disciplining pro-union employees), other in-centives are designed to substitute unionization(e.g., creating intra-organizational employee com-mittees as bargaining units; Dundon, 2002). Somenegative brokering processes in the context of laborrelations use both information and incentives. Mostnotably, promises of future rewards (e.g., bonuses)and threats of future penalties (e.g., layoff) involveinformation about the intention to use incentives inthe future. Another example entails spreading ru-mors or releasing personal information about unionactivists, which uses information that may triggersocial sanctions. Finally, some context-specific bro-kering activities that are unique to the domain oflabor relations involve pursuing legal delays, ap-pealing to theNational Labor Relations Board, hiringnew employees to dilute the proportion of unionsupporters in the voting unit before the election, or“intermixing players with dissimilar interests andstakes. . .in the hope that racial antagonisms amongsubgroups would prevent workers as a whole fromconcerting their efforts. . .” (Posner et al., 2010: 434).Collectively, these tactics are designed to change


targets’ relationships from positive (or neutral) tonegative, thereby thwarting unionization (Hurd &Uehlein, 1994).

The negative brokering tactics used in “unionbusting” resemble in many ways the actions of dom-inant leaders. Dominant leaders tend to be selfish,aggressive, and unethical (Cheng, Tracy, Foulsham,Kingstone, & Henrich, 2013; Halevy et al., 2012;Kakkar & Sivanathan, 2017). In a series of experi-mental studies, Case and Maner (2014) showed thatconditionsofanunstablehierarchycreate the“perfectstorm” for dominant leaders, whose selfish, aggres-sive, and unethical tendencies converge to producedivisive behavior. When their power position wasunstable, group leaders who were motivated bydominance chose to limit communication opportu-nities between a highly skilled subordinate and othermembers of the group, create spatial working ar-rangements that physically isolated the highly skilledsubordinate from other group members, minimizedopportunities for a highly skilled subordinate to so-cially bond (and thereby potentially coalesce) withanother group member by choosing a task-orientedover an interpersonally oriented work style, andassigned the highly skilled subordinate to work witha partner with whom they were unlikely to get along.These tactics, which were intentionally designed toreduce the likelihood of cooperation among others,closely resemble those used by alpha-males in groupsof primates (De Waal, 2007), and constitute a mani-festation of negative brokering by a third party.

TAKING STOCK: EMERGENT THEMES ANDPROMISING FUTURE DIRECTIONS

Our review of a diverse set of literatures relevant tobrokerage and brokering has resulted in the identi-fication of six emergent themes: (a) Network struc-ture versus brokering processes, (b) consequencesfor brokers versus alters, (c) helpful versus harmfulforms of brokering, (d) information versus incentivesas instruments of third party influence, (e) situa-tional versus personal antecedents of brokeringprocesses, and (f) brokering processes as in-terpersonal versus intergroup phenomena. Table 2summarizes these emergent themes together withillustrating sources and future directions that emergedirectly from each theme.

Network Structure versus Brokering Processes

The first theme from our review concerns theemerging understanding that network structure

often provides both the social and organizationalcontext for brokering processes and the impetus forbrokering activities. Network structure affords andconstrains different functional forms of brokeringactivities, as captured in the COR framework, yetsocial influence processes are the active ingredientsthat modify alters’ relationships (Obstfeld et al.,2014; Quintane & Carnabuci, 2016). Our reviewhighlights the value of considering the interplay be-tween social structure and social processes.Whereassocial structure creates opportunities and bound-aries for social behavior, failing to consider whatpeople actually do once they occupy a particularposition or find themselves in a particular socialcontext amounts to telling just part of the story. Oc-cupying a bridge position between twodisconnectedalters in one’s networks can give rise to distinctprocesses such as tertius gaudens (keeping altersapart and serving as an intermediary) or tertius iun-gens (introducing and supporting direct collabora-tion between alters). Similarly, individuals whoobserve coworkers in conflict can react in differentways: They may choose to do nothing, attempt to actas mediators in the conflict, or seek to escalate it.Considering brokerage and brokering in tandem canilluminate a wide range of important organizationalphenomena, from creativity and innovation to co-alition formation and dispute resolution. Future re-search is required to explore how both networkstructure and social relations in organizationschange over time as a result of brokering behavioracross both open and closed triads.

Consequences for Brokers versus Alters

The second theme from our review is that distinctliteratures presently focus on the consequences thatoccupying brokerage positions and engaging in bro-kering behaviors produce for brokers versus alters.Whereas the established social capital branch of thenetworks literature emphasizes rewards to brokers(Burt, 2004), researchers increasingly pay more at-tention to the effects of brokering processes onothers’ interactions and relationships (Lingo &O’Mahony, 2010). We encourage researchers toconsider how different functional forms of brokeringshape the outcomes of brokers, alters, andwork unitsas a whole. Importantly, future research may con-sider the possibility that brokering processes mayproduce asymmetric consequences for those in-volved. Specifically, whereas some brokering pro-cesses may benefit the broker and both alters in thetriad (Halevy & Halali, 2015), there are contexts in


which brokeringmay benefit some alterswhile at thesame time harming others (Clement et al., 2018;Laskewitz et al., 1994). In addition, future researchmay explore the possibility that brokers who engagein different brokering activities receive differentreturns on their investments (of time, effort,and other resources expended in the process ofbrokering).

Helpful versus Harmful Forms of Brokering

As captured in Figure 2 and Table 1, a third themethat emerged from our review concerns the distinc-tion between helpful (e.g., Obstfeld, 2005) andharmful (Case & Maner, 2014) brokering processes.We found it extremely useful to borrow tools from IT(Kelley & Thibaut, 1978; Rusbult & Van Lange, 2003)as means to identify distinct functional forms ofbrokering. Our review revealed that brokering caninvolve creating relationships (e.g., via social in-troductions), reinforcing relationships (e.g., via gos-sip), changing the sign of relationships (fromnegative to positive or frompositive to negative), andterminating relationships. Thus, brokers impact so-cial and organizational relationships by modifyingboth the degree of outcome interdependence andthe correspondence of interests between alters.

Delineating different brokering processes can po-tentially enhance our ability to explain variance inimportant social and organizational phenomena(e.g., innovation, merger failures, and turnover).

Our review identified many more studies of help-ful brokering as compared with harmful brokering(although we provide a similar number of examplesto illustrate these in Table 1). Future research is re-quired to enhance our understanding of when andwhy individuals engage in harmful brokering, aswell as the effects of harmful brokering on the func-tioning of performance groups (e.g., sport teams andtheatre groups) and other organizational units. No-tably, because helpful and harmful brokering pro-cesses may co-occur across different parts of thesame social or organizational network, future re-search may benefit from exploring multiple kinds ofbrokering processes simultaneously to determinetheir combined effects.

Information versus Incentives as Instruments ofThird-Party Influence

The fourth theme that emerged from our reviewconcerns the levers that brokers use to influenceothers’ interactions and relationships. These toolstypically fall into one of two categories: Incentives

TABLE 2Emergent Themes from the Review and Future Directions for Research on Brokerage and Brokering

Emergent Themes Illustrating Sources Future Directions

Structure versus Process Grigoriou and Rothaermel (2014) Understanding how network structure and socialrelations change over time by studying brokeringbehavior in both open and closed triadslongitudinally

Occupying brokerage positions in open triadsversus brokering in open and closed triads

Obstfeld et al. (2014)Quintane and Carnabuci (2016)Simmel (1950)

Consequences for Brokers versus Alters Burt (2004) Consideringmultiple functional forms of brokeringsimultaneously when exploring brokering’sconsequences for brokers, alters, and work units

Emphasizing consequences of brokerage andbrokering for brokers’ success versus alters’ andwork units’ success

Clement et al. (2018)Lingo and O’Mahony (2010)Soda et al. (2018)

Helpful versus Harmful Forms of Brokering Case and Maner (2014) Focusing greater attention on harmful brokering toadvance knowledge on how divisive behaviorsundermine cooperation and innovation

Focusing on the positive versus negative effects ofoccupying brokerage positions and engaging inbrokering behaviors

Halevy and Halali (2015)Obstfeld (2005)Posner et al. (2010)

Information versus Incentives as Instruments Conlon et al. (1994) Exploring alters’ reactions to brokering behaviorsthat use information (e.g., advice), incentives(e.g., promises of rewards) or both as means ofinfluence

Using privileged access to information versuscontrol over valued resources to influence others’interactions and relationships

Feinberg et al. (2012)Fleming et al. (2007)Shapiro and Brett (1993)

Situational versus Personal Antecedents Kleinbaum et al. (2015) Understanding which situational and personalcharacteristics matter most, as well as whenand why individual differences matter moreversus less

Brokerage positions as “strong situations” thatuniformly shape outcomes versus individualdifferences in motivation and ability to broker

Landis et al. (2018)Sasovova et al. (2010)Ward, Stovel, and Sacks (2011)

Interpersonal versus Intergroup Phenomena FlemingandWaguespack (2007) Exploring the extent towhich the intragroup versusintergroup context of brokering moderatesbrokering’s consequences for brokers and alters

Brokering as an interpersonal processwithin groupsversus a social influence process that cuts acrossgroup boundaries

Stovel et al. (2011)Hogg et al. (2012)Kaplan et al. (2017)


(that follow from control over valuable resources)and information (that results from privileged accessto knowledge). Although both incentives and in-formation can be effective as ameans of influence, atleast some of the research we reviewed suggests al-ters respond more favorably to information than toincentives (e.g., disputants perceive the process ofmediation as fairer than arbitration: Shapiro & Brett,1993). Future research may explore the trade-offsassociated with the use of different instruments tochange others’ interactions and relationships. Forinstance, some modes of third-party interventionthat produce strong and immediate effects in theshort term may be costly and hence unsustainablelonger term, whereas other modes of third-party in-tervention may take longer to materialize yet pro-duce enduring effects (e.g., compliance versusinternalization: Kelman, 1961, 2006).

Situational versus Personal Antecedents ofBrokering Processes

The fifth theme that emerged from our review con-cerns the extent to which powerful situations versusindividual differences (e.g., in needs, abilities, andpersonality traits) shape brokering processes in orga-nizations. The strong situation hypothesis asserts thatstrong situations mute the impact of personality andresult in largely uniform behavior across individuals(Cooper &Withey, 2009;Mischel, 1977). Hence, to theextent that structural properties of networks producestrong psychological situations, individual differenceshould be uncorrelated with brokering behavior andoutcomes. The research we reviewed, however, sug-gests that personality traits, such as self-monitoring,play an important role in shaping brokering processesin organizations (Kilduff & Brass, 2010; Kilduff & Day,1994; Mehra, Kilduff, & Brass, 2001; Oh & Kilduff,2008; Sasovova, Mehra, Borgatti, & Schippers, 2010).Related research has considered how individual dif-ferences in sense of power (Landis et al., 2018), traitssuch as openness to experience (Baer, 2010), and theability and motivation to share information (Reinholt,Pedersen, & Foss, 2011) influence brokering processesand outcomes. Recent research has also consideredinteractions between different individual characteris-tics, documenting, for example, that self-reported dif-ferences in self-monitoring interact with perceivedempathy (as reported by peers) in shaping changes inMBAstudents’networks (Kleinbaum, Jordan,&Audia,2015).

Future research may integrate the study of situa-tional characteristics and individual characteristics

to explain organizational phenomena (Goldberg,Srivastava, Manian, Monroe, & Potts, 2016). For ex-ample future research may explore how differentaspects of psychological situations (e.g., adversity,sociality, and power: Gerpott, Balliet, Columbus,Molho, & de Vries, 2017; Parrigon, Woo, Tay, &Wang, 2017; Rauthmann et al., 2014) interact withsocial skill (defined as “the ability to induce co-operation”: Obstfeld, 2017: 49, cf. Fligstein &McAdam, 2012) and political skill (defined as “theability to effectively understand others at work, andto use such knowledge to influence others to act inways that enhance one’s personal and/or organiza-tional objectives”: Ferris, Treadway, Perrewe,Brouer, Douglas, & Lux, 2007: 291; cf. Pfeffer, 1992)to shape brokering behavior in organizationalcontexts.

Brokering Processes as Interpersonal versusIntergroup Phenomena

The sixth theme that emerged from our review isthat brokerage and brokering can be conceptual-ized as interpersonal phenomena (Obstfeld, 2017;Simmel, 1950) and intergroup phenomena (Gould &Fernandez, 1989; Hogg, Van Knippenberg, & Rast,2012;Mehra, Kilduff, &Brass, 1998). Future researchmay integrate concepts from the intergroup relationsliterature, such as in-group bias (Hewstone, Rubin, &Willis, 2002) and the contact hypothesis (Pettigrew&Tropp, 2006), to explore the extent to which bro-kering within group boundaries and across groupboundaries have similar versus distinct antecedentsand consequences. Addressing these questionswould enhance our understanding of both brokeringprocesses and intergroup relations in organizationalcontexts.

CONCLUSION

Brokerage and brokering play pivotal roles inshaping important social and organizational phe-nomena, including patterns of cooperation andcompetition, trust and suspicion, status conferraland the accumulation of power. The current reviewused an inclusive approach to brokerage and bro-kering, casting a wide net in search of literaturespertinent to third-party influence. Such an inclusiveapproach enables different ideas to have conceptualplaydates with their theoretical siblings, cousins,and classmates, a necessary process in the healthydevelopment of novel theories and frameworks(Higgins, 2017). In the present article, this effort led


us to develop the integrative COR framework asa means to map distinct social influence processesthrough which organizational actors shape others’relationships, for better or worse. FollowingGranovetter (1973) and others, we subscribe to theview that “the analysis of processes in interpersonalnetworks provides the most fruitful micro-macrobridge” (p. 1360). Thus, we see the greatest potentialfor future research in this field in studies that willdeepen our understanding of the behavioral pro-cesses through which organizational actors shapeothers’ relationships. We hope that the current re-view will disrupt (in the most positive, Silicon Val-ley sense) the current conversation about brokerageand brokering in organizations, by making it theo-retically richer, more inclusive (i.e., broader), andmore process-oriented.

REFERENCES

Aguinis, H., Ramani, R. S., & Alabduljader, N. 2018. Whatyou see is what you get? Enhancing methodologicaltransparency in management research. Academy ofManagement Annals, 12(1): 83–110.

Ahuja, G. 2000. Collaboration networks, structural holes,and innovation: A longitudinal study.AdministrativeScience Quarterly, 45: 425–455.

Anicich, E. M., Fast, N. J., Halevy, N., & Galinsky, A. D.2015. When the bases of social hierarchy collide:Power without status drives interpersonal conflict.Organization Science, 27(1): 123–140.

Anik, L., & Norton, M. I. 2014. Matchmaking promoteshappiness. Social Psychological and PersonalityScience, 5: 644–652.

Arnold, J. A., & O’Connor, K. M. 1999. Ombudspersons orpeers? The effect of third-party expertise and recom-mendations on negotiation. Journal of Applied Psy-chology, 84: 776–785.

Baer, M. 2010. The strength-of-weak-ties perspective oncreativity: a comprehensive examination and exten-sion. Journal of Applied Psychology, 95: 592–601.

Balliet, D., Tybur, J. M., & Van Lange, P. A. 2016. Func-tional interdependence theory: An evolutionary ac-count of social situations. Personality and SocialPsychology Review, 21: 361–388.

Baumeister, R. F., & Leary, M. R. 1995. The need to belong:desire for interpersonal attachments as a fundamentalhumanmotivation.PsychologicalBulletin, 117: 497–529.

Burt. R. S. 1992. Structural holes: The social structure ofcompetition. Cambridge,MA:HarvardUniversityPress.

Burt, R. S. 2000. The network structure of social capital.Research in Organizational Behavior, 22: 345–423.

Burt, R. S. 2004. Structural holes and good ideas. Ameri-can Journal of Sociology, 110: 349–399.

Burt, R. S. 2005. Brokerage and closure: An introductionto social capital. New York: Oxford University Press.

Burt, R. S., Kilduff, M., & Tasselli, S. 2013. Social networkanalysis: Foundations and frontiers on advantage.Annual Review of Psychology, 64: 527–547.

Burt, R. S., & Knez, M. 1995. Kinds of third-party effects ontrust. Rationality and Society, 7: 255–292.

Carpenter, M. A., Li, M., & Jiang, H. 2012. Social networkresearch in organizational contexts: A systematic re-view of methodological issues and choices. Journalof Management, 38(4): 1328–1361.

Casciaro, T., Barsade, S. G., Edmondson, A. C., Gibson,C. B., Krackhardt, D., & Labianca, G. 2015. The in-tegration of psychological and network perspectivesin organizational scholarship.Organization Science,26(4): 1162–1176.

Case, C. R., &Maner, J. K. 2014. Divide and conquer:Whenand why leaders undermine the cohesive fabric oftheir group. Journal of Personality and Social Psy-chology, 107: 1033–1050.

Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., &Henrich, J. 2013. Two ways to the top: Evidence thatdominance and prestige are distinct yet viable ave-nues to social rank and influence. Journal of Person-ality and Social Psychology, 104(1): 103–125.

Chou, E. Y., Halevy, N., Galinsky, A. D., & Murnighan,J. K. 2017. The Goldilocks contract: The synergis-tic benefits of combining structure and autonomyfor persistence, creativity, and cooperation. Jour-nal of Personality and Social Psychology, 113(3):393–412.

Clement, J., Shipilov, A., & Galunic, D. C. 2018. Brokerageas a public good: The externalities of network hubsfor different formal roles in creative organizations.Administrative Science Quarterly, 63(2): 251–286.

Clifton,A., &Webster, G.D. 2017.An introduction to socialnetwork analysis for personality and social psychol-ogists. Social Psychological and Personality Sci-ence, 8(4): 442–453.

Coleman, J. S. 1988. Social capital in the creation of hu-man capital. American Journal of Sociology, 94:S95–S120.

Conlon, D. E., Carnevale, P. J., & Murnighan, J. K. 1994.Intravention: Third-party intervention with clout.Organizational Behavior and Human DecisionProcesses, 57: 387–410.

Conlon, D. E., Moon, H., & Ng, K. Y. 2002. Putting the cartbefore the horse: the benefits of arbitrating beforemediating. Journal of Applied Psychology, 87(5):978–984.


Cooper, W. H., & Withey, M. J. 2009. The strong situationhypothesis. Personality and Social Psychology Re-view, 13(1): 62–72.

DeWaal, F. B. 2007.Chimpanzee politics: Power and sexamong apes. Baltimore: JHU Press.

Duffy, M. K., Ganster, D. C., & Pagon, M. 2002. Socialundermining in the workplace. Academy of Man-agement Journal, 45(2): 331–351.

Dunbar, R. I. 2004. Gossip in evolutionary perspective.Review of General Psychology, 8(2): 100–110.

Dundon, T. 2002. Employer opposition and union avoid-ance in the UK. Industrial Relations Journal, 33(3):234–245.

Eisenberger, N. I., Lieberman, M. D., & Williams, K. D.2003. Does rejection hurt? An fMRI study of socialexclusion. Science, 302(5643): 290–292.

Ellwardt, L., Labianca, G. J., &Wittek, R. 2012.Who are theobjects of positive and negative gossip at work?: Asocial network perspective on workplace gossip. So-cial Networks, 34(2): 193–205.

Ellwardt, L., Steglich, C., & Wittek, R. 2012. The co-evolution of gossip and friendship inworkplace socialnetworks. Social Networks, 34(4), 623–633.

Emler, N. 1994. Gossip, reputation, and social adapta-tion. In R. F. Goodman & A. Ben-Ze’ev (Eds.), Goodgossip: 117–138. Lawrence, KS: University Press ofKansas.

Feinberg, M., Willer, R., & Schultz, M. 2014. Gossip andostracism promote cooperation in groups. Psycho-logical Science, 25(3): 656–664.

Feinberg, M., Willer, R., Stellar, J., & Keltner, D. 2012. Thevirtues of gossip: reputational information sharing asprosocial behavior. Journal of Personality andSocialPsychology, 102: 1015–1030.

Ferris, G. R., Treadway, D. C., Perrewe, P. L., Brouer, R. L.,Douglas, C., & Lux, S. 2007. Political skill in organi-zations. Journal of Management, 33: 290–320.

Fleming, L., Mingo, S., & Chen, D. 2007. Collabora-tive brokerage, generative creativity, and creativesuccess. Administrative Science Quarterly, 52:443–475.

Fleming, L., & Waguespack, D. M. 2007. Brokerage,boundary spanning, and leadership in open in-novation communities. Organization Science, 18:165–180.

Fligstein N., & McAdam, D. 2012. Rethinking power inorganizations, institutions, and markets. Research inthe Sociology of Organizations, 34: 287–316.

Foster, E. K. 2004. Research on gossip: Taxonomy,methods, and future directions. Review of GeneralPsychology, 8(2): 78–99.

Freeman, L. C. 1978. Centrality in social networks con-ceptual clarification. Social Networks, 1(3): 215–239.

Galunic, C., Ertug, G., & Gargiulo, M. 2012. The positiveexternalities of social capital: Benefiting from seniorbrokers. Academy of Management Journal, 55(5):1213–1231.

Gerpott, F. H., Balliet, D., Columbus, S.,Molho, C., & deVries,R. E. 2017.Howdopeople think about interdependence?A multidimensional model of subjective outcome in-terdependence. Journal of Personality and Social Psy-chology, 115(4): 716–742.

Giardini, F. 2012. Deterrence and transmission as mecha-nisms ensuring reliability of gossip. Cognitive Pro-cessing, 13(2): 465–475.

Giebels, E., & Janssen, O. 2005. Conflict stress and reducedwell-being at work: The buffering effect of third-partyhelp.European Journal ofWork andOrganizationalPsychology, 14(2): 137–155.

Godard, J. 2009. The exceptional decline of the Americanlabor movement. ILR Review, 63(1): 82–108.

Goldberg, A., Srivastava, S. B., Manian, V. G., Monroe, W.,& Potts, C. 2016. Fitting in or standing out? Thetradeoffs of structural and cultural embeddedness.American Sociological Review, 81(6): 1190–1222.

Gould, R. V., & Fernandez, R. M. 1989. Structures of me-diation: A formal approach to brokering in transactionnetworks. Sociological Methodology, 19: 89–126.

Granovetter, M. S. 1973. The strength of weak ties. Amer-ican Journal of Sociology, 78(6): 1360–1380.

Grigoriou, K., & Rothaermel, F. T. 2014. Structural micro-foundations of innovation: The role of relational stars.Journal of Management, 40(2): 586–615.

Grosser, T. J., Lopez-Kidwell, V., & Labianca, G. 2010. Asocial network analysis of positive andnegative gossipin organizational life. Group & Organization Man-agement, 35(2): 177–212.

Halevy, N., Chou, E. Y., Cohen, T. R., & Livingston, R. W.2012. Status conferral in intergroup social dilemmas:behavioral antecedents and consequences of prestigeand dominance. Journal of Personality and SocialPsychology, 102(2): 351–366.

Halevy, N., Chou, E. Y., & Murnighan, J. K. 2012. Mindgames: The mental representation of conflict.Journal of Personality and Social Psychology,102: 132–148.

Halevy, N., Cohen, T. R., Chou, E. Y., Katz, J. J., & Panter,A. T. 2014. Mental models at work: Cognitive causesand consequences of conflict in organizations. Per-sonality and Social Psychology Bulletin, 40(1):92–110.

Halevy, N., & Halali, E. 2015. Selfish third parties act aspeacemakers by transforming conflicts and promoting


cooperation. Proceedings of the National Academyof Sciences, 112: 6937–6942.

Halevy, N., & Katz, J. J. 2013. Conflict templates: Thinkingthrough interdependence. Current Directions inPsychological Science, 22(3): 217–224.

Halevy, N., & Phillips, L. T. 2015. Conflict templates innegotiations, disputes, joint decisions, and tourna-ments. Social Psychological and Personality Sci-ence, 6(1): 13–22.

Halevy, N., Weisel, O., & Bornstein, G. 2012. “In‐grouplove” and “out‐group hate” in repeated interactionbetween groups. Journal of Behavioral DecisionMaking, 25(2): 188–195.

Heider, F. 1946. Attitudes and cognitive organization.Journal of Psychology, 21: 107–112.

Hewstone, M., Rubin, M., & Willis, H. 2002. Intergroupbias. Annual Review of Psychology, 53(1): 575–604.

Higgins, E. T. 2017. Useful theories in social psychology.Presented at the (2017) annual meeting of the societyfor experimental social psychology, Boston, MA.

Hogg, M. A., Van Knippenberg, D., & Rast III, D. E. 2012.Intergroup leadership in organizations: Leadingacross group and organizational boundaries. Acad-emy of Management Review, 37(2): 232–255.

Hurd, R. W., & Uehlein, J. B. 1994. Patterned responses toorganizing: Case studies of the union-busting con-vention. In S. Friedman, R. W. Hurd, R. A. Oswald &R.L. Seeber (Eds.)Restoring thepromiseofAmericanlabor law: 61–74. Ithaca, NY: ILR Press.

Kakkar, H., & Sivanathan, N. 2017. When the appeal ofa dominant leader is greater than a prestige leader.Proceedings of the National Academy of Sciences,114: 6734–6739.

Kaplan, S., Milde, J., & Cowan, R. S. 2017. Symbiontpractices in boundary spanning: Bridging the cogni-tive and political divides in interdisciplinary re-search. Academy of Management Journal, 60(4):1387–1414.

Karambayya, R., & Brett, J. M. 1989. Managers handlingdisputes: Third-party roles andperceptions of fair-ness. Academy of Management Journal, 32(4):687–704.

Karambayya, R., Brett, J. M., & Lytle, A. 1992. Effects offormal authority and experience on third-party roles,outcomes, and perceptions of fairness. Academy ofManagement Journal, 35(2): 426–438.

Kauppila, O. P., Bizzi, L., & Obstfeld, D. 2018. Connectingand creating: tertius iungens, individual creativity,and strategic decision processes. Strategic Manage-ment Journal, 39(3): 697–719.

Kelley, H. H., Holmes, J. G., Kerr, N. L., Reis, H. T., Rusbult,C. E., & Van Lange, P. A. M. 2003. An atlas of

interpersonal situations. New York: Cambridge Uni-versity Press.

Kelley, H. H., & Thibaut, J. W. 1978. Interpersonal re-lations: A theory of interdependence. New York:Wiley.

Kelman, H. C. 1961. Three processes of social influence.Public Opinion Quarterly, 25: 57–78.

Kelman, H. C. 2006. Interests, relationships, identities:Three central issues for individuals and groups innegotiating their social environment.Annual Reviewof Psychology, 57: 1–26.

Kilduff, M., & Brass, D. J. 2010. Organizational social net-work research: Core ideas and key debates. Theacademy of management annals, 4(1): 317–357.

Kilduff, M., & Day, D. V. 1994. Do chameleons get ahead?The effects of self-monitoring on managerial careers.AcademyofManagement Journal, 37(4): 1047–1060.

Kleinbaum, A. M., Jordan, A. H., & Audia, P. G. 2015. Analtercentric perspective on the origins of brokering insocial networks: How perceived empathy moderatesthe self-monitoring effect. Organization Science,26(4): 1226–1242.

Kniffin, K. M., & Wilson, D. S. 2005. Utilities of gossipacross organizational levels. Human Nature, 16(3):278–292.

Kniffin, K. M., & Wilson, D. S. 2010. Evolutionary per-spectives on workplace gossip: Why and how gossipcan serve groups. Group & Organization Manage-ment, 35(2): 150–176.

Kozan, M., Ergin, C., & Varoglu, D. 2007. Third party in-tervention strategies of managers in subordinates’conflicts in Turkey. International Journal of ConflictManagement, 18(2): 128–147.

Krackhardt, D., & Kilduff, M. 1999. Whether close or far:Social distance effects on perceived balance infriendship networks. Journal of Personality and So-cial Psychology, 76(5): 770–782.

Krackhardt, D. 1999. The ties that torture: Simmelian tieanalysis in organizations. Research in the Sociologyof Organizations, 16(1): 183–210.

Kurland, N. B., & Pelled, L. H. 2000. Passing the word:Toward a model of gossip and power in the work-place. Academy of Management Review, 25(2):428–438.

Landis, B. 2016. Personality and social networks in orga-nizations: A review and future directions. Journal ofOrganizational Behavior, 37(S1): 107–121.

Landis, B., Kilduff, M., Menges, J. I., & Kilduff, G. J. 2018.The paradox of agency: Feeling powerful reducesbrokerage opportunity recognition yet increases will-ingness to broker. Journal of Applied Psychology,103(8): 929–938.


Laskewitz, P., Vliert, E., & Dreu, C. K. 1994. Organizationalmediators siding with or against the powerful party?.Journal of Applied Social Psychology, 24(2): 176–188.

Levitt, M. J. 1993. Confessions of a union buster. NewYork: Random House.

Lewicki, R. J., & Sheppard, B. H. 1985. Choosing howto intervene: Factors affecting the use of processand outcome control in third party dispute resolu-tion. Journal of Organizational Behavior, 6(1):49–64.

Lewicki, R. J., Weiss, S. E., & Lewin, D. 1992. Models ofconflict, negotiation and third party intervention: Areview and synthesis. Journal of Organizational Be-havior, 13: 209–252.

Lind, E. A., Walker, L., Kurtz, S., Musante, L., & Thibaut,J.W. 1980. Procedure andoutcomeeffects on reactionsto adjudicated resolution of conflicts of interest.Journal of Personality and Social Psychology, 39(4):643–653.

Lingo, E. L., & O’Mahony, S. 2010. Nexus work: Brokeringon creative projects. Administrative Science Quar-terly, 55: 47–81.

Logan, J. 2006. The union avoidance industry in theUnitedStates. British Journal of Industrial Relations, 44(4):651–675.

Malhotra, D., & Murnighan, J. K. 2002. The effects of con-tracts on interpersonal trust. Administrative ScienceQuarterly, 47(3): 534–559.

Markus, H. R., & Kitayama, S. 2010. Cultures and selves: Acycle of mutual constitution. Perspectives on Psy-chological Science, 5(4): 420–430.

Marsden, P. V. 1982. Brokering behavior in restricted ex-change networks. in P. V. Marsden & N. Lin (Eds.)Social structure and network analysis: 201–218.Beverly Hills, CA: Sage.

Mehra,A., Kilduff,M., &Brass, D. J. 1998.At themargins:Adistinctiveness approach to the social identity andsocial networks of underrepresented groups. Acad-emy of Management Journal, 41(4): 441–452.

Mehra, A., Kilduff, M., & Brass, D. J. 2001. The social net-works of high and low self-monitors: Implications forworkplace performance. Administrative ScienceQuarterly, 46(1): 121–146.

Mercier, H., & Sperber, D. 2011. Why do humans reason?Arguments for an argumentative theory. Behavioraland Brain Sciences, 34(2): 57–74.

Mischel,W. 1977. The interaction of person and situation.In D. Magnusson & N. S. Endler (Eds.), Personalityat the crossroads: Current issues in interac-tional psychology: 333–352. Hillsdale, NJ: LawrenceErlbaum.

Morrill, C., & Rudes, D. S. 2010. Conflict resolution in or-ganizations. Annual Review of Law and Social Sci-ence, 6: 627–651.

Nakashima, N. A., Halali, E., & Halevy, N. 2017. Thirdparties promote cooperative norms in repeated in-teractions. Journal of Experimental Social Psychol-ogy, 68: 212–223.

O’Sullivan, M., & Gunnigle, P. 2009. “Bearing all the hall-marks of oppression”: Union avoidance in Europe’slargest low-cost airline. Labor Studies Journal, 34(2):252–270.

Obstfeld, D., Borgatti, S. P., & Davis, J. 2014. Brokering asa process: Decoupling third party action from socialnetwork structure. Contemporary Perspectives onOrganizational Social Networks, 40: 135–159.

Obstfeld, D. 2005. Social networks, the tertius iungensorientation, and involvement in innovation. Admin-istrative Science Quarterly, 50: 100–130.

Obstfeld, D. 2017. Getting new things done: Networks,brokerage, and the assembly of innovative action.Stanford, CA: Stanford University Press.

Oh, H., & Kilduff, M. 2008. The ripple effect of personalityon social structure: self-monitoring origins of networkbrokerage. Journal of Applied Psychology, 93:1155–1164.

Parrigon, S., Woo, S. E., Tay, L., & Wang, T. 2017.CAPTION-ing the situation: A lexically- derived tax-onomy of psychological situation characteristics.Journal of Personality and Social Psychology,112(4): 642–681.

Pettigrew, T. F., & Tropp, L. R. 2006. A meta-analytic testof intergroup contact theory. Journal of Personalityand Social Psychology, 90(5): 751–783.

Pfeffer, J. 1992. Managing with power: Politics and in-fluence in organizations. Harvard Business Press.

Pinkley, R. L., Brittain, J., Neale, M. A., & Northcraft,G. B. 1995. Managerial third-party dispute inter-vention:An inductive analysis of intervenor strategyselection. Journal of Applied Psychology, 80(3):386–402.

Podolny, J. M., & Baron, J. N. 1997. Resources and re-lationships: Social networks and mobility in theworkplace. American Sociological Review, 62: 673–693.

Posner, E. A., Spier, K. E., &Vermeule, A. 2010. Divide andconquer. Journal of Legal Analysis, 2: 417–471.

Quintane, E., & Carnabuci, G. 2016. How do brokers bro-ker? Tertius gaudens, tertius iungens, and the tempo-rality of structural holes. Organization Science, 27:1343–1360.

Rand, D. G., & Nowak, M. A. 2013. Human cooperation.Trends in Cognitive Sciences, 17(8): 413–425.


Rauthmann, J. F., Gallardo-Pujol, D., Guillaume, E. M.,Todd, E., Nave, C. S., Sherman, R. A., Ziegler, M.,Jones, A. B., Funder, D. C., & Funder, D. C. 2014. Thesituational eight DIAMONDS: A taxonomy of ma-jor dimensions of situation characteristics. Jour-nal of Personality and Social Psychology, 107(4):677.

Reinholt, M. I. A., Pedersen, T., & Foss, N. J. 2011. Whya central network position isn’t enough: The role ofmotivation and ability for knowledge sharing in em-ployee networks.Academy of Management Journal,54(6): 1277–1297.

Reis, H. T. 2008. Reinvigorating the concept of situation insocial psychology. Personality and Social Psychol-ogy Review, 12(4): 311–329.

Ross, W. H., & Conlon, D. E. 2000. Hybrid forms of third-party dispute resolution: Theoretical implications ofcombining mediation and arbitration. Academy ofManagement Review, 25: 416–427.

Rubin, J. Z. 1980. Experimental research on third-partyintervention in conflict: Toward some generaliza-tions. Psychological Bulletin, 87(2): 379–391.

Rusbult, C. E., & Van Lange, P. A. 2003. Interdependence,interaction, and relationships. Annual Review ofPsychology, 54(1): 351–375.

Samila, S., Oettl, A., & Hasan, S. 2016. Helpfulthirds and the durability of collaborative ties.Available at SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id52601338. Accessed June 7,2018.

Sasovova, Z., Mehra, A., Borgatti, S. P., & Schippers, M. C.2010. Network churn: The effects of self-monitoringpersonality on brokering dynamics. AdministrativeScience Quarterly, 55(4): 639–670.

Schelling, T. C. 1980. The strategy of conflict. Cambridge,MA: Harvard University Press.

Schilbach, L., Eickhoff, S. B., Rotarska-Jagiela, A., Fink,G. R., & Vogeley, K. 2008. Minds at rest? Socialcognition as the default mode of cognizing andits putative relationship to the “default system” ofthe brain. Consciousness and Cognition, 17(2):457–467.

Shapiro, D. L., & Brett, J. M. 1993. Comparing three pro-cesses underlying judgments of procedural justice:A field study of mediation and arbitration. Journalof Personality and Social Psychology, 65(6): 1167–1177.

Sheppard, B. 1984. Third party conflict intervention: Aprocedural framework. Research in OrganizationalBehavior, 6: 141–190.

Simmel, G. 1950. In K. Wolf (Ed.), The sociology of GeorgSimmel. New York: Simon and Schuster.

Simmel, G. 1955.Conflict & thewebof group-affiliations.New York: Free Press.

Skarlicki, D. P., & Rupp, D. E. 2010. Dual processing andorganizational justice: the role of rational versusexperiential processing in third-party reactions toworkplace mistreatment. Journal of Applied Psy-chology, 95(5): 944–952.

Soda, G. B., Tortoriello, M., & Iorio, A. 2018. Harvestingvalue frombrokerage: individual strategic orientation,structural holes, and performance. Academy ofManagement Journal, 61(3): 896–918.

Stovel, K., Golub, B., & Milgrom, E. M. M. 2011. Sta-bilizing brokerage. Proceedings of the NationalAcademy of Sciences, 108(Supplement 4): 21326–21332.

Stovel, K., & Shaw, L. 2012. Brokerage. Annual Review ofSociology, 38: 139–158.

Thibaut, J.W., &Kelley, H. H. 1959.The social psychologyof groups. New York: Wiley.

Thibaut, J. W., & Walker, L. 1975. Procedural justice: Apsychological analysis. Mahwah, NJ: L. ErlbaumAssociates.

Ting‐Toomey, S., Oetzel, J. G., & Yee‐Jung, K. 2001. Self‐construal types and conflict management styles.Communication Reports, 14(2): 87–104.

Uzzi, B., & Spiro, J. 2005. Collaboration and creativity: Thesmall world problem. American Journal of Sociol-ogy, 111(2): 447–504.

Walker, G., Kogut, B., & Shan, W. 1997. Social capital,structural holes and the formation of an industry net-work. Organization Science, 8(2): 109–125.

Ward,M.D., Stovel, K., & Sacks, A. 2011. Network analysisand political science. Annual Review of PoliticalScience, 14: 245–264.

Wu, J., Balliet, D., & Van Lange, P. A. 2016. Reputationmanagement: Why and how gossip enhances gener-osity. Evolution and Human Behavior, 37: 193–201.

Yamagishi, T., & Kiyonari, T. 2000. The group as the con-tainer of generalized reciprocity. Social PsychologyQuarterly, 63 (2): 116–132.

Zhang, T., Gino, F., & Norton, M. I. 2017. The surprisingeffectiveness of hostile mediators. Management Sci-ence, 63: 1972–1992.

Zhelyazkov, P. I. 2018. Interactions and interests: Collab-oration outcomes, competitive concerns, and thelimits to triadic closure. Administrative ScienceQuarterly, 63: 210–247.

Zlatev, J. J., & Miller, D. T. 2016. Selfishly benevolent orbenevolently selfish: When self-interest underminesversus promotes prosocial behavior. OrganizationalBehavior and Human Decision Processes, 137:112–122.


https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2601338



Nir Halevy ([email protected]): Nir Halevy is an As-sociate Professor of Organizational Behavior at StanfordUniversity’s Graduate School of Business. His researchfocuses on cooperation and competition between in-dividuals and groups, interdependent decision-making,group processes, and intergroup relations.

Eliran Halali ([email protected]): Eliran Halali isa Lecturer (Assistant Professor) of Social and Organiza-tional Psychology at Bar-Ilan University’s Department of

Psychology. His research focuses on social preferences,cooperation and competition between individuals andgroups, unethical behavior, and cognitive control.

Julian Zlatev ([email protected]): Julian Zlatev is an Assis-tant Professor in the Negotiation, Organizations &MarketsUnit at Harvard Business School. He received his PhD inOrganizational Behavior at Stanford Graduate School ofBusiness. His research interests include ethics and mo-rality, decision making, and prosocial behavior.


mailto:[email protected]



BROKERAGE AND BROKERING: AN INTEGRATIVE REVIEW AND … Files/Brokerage and... · course by the...

Documents

Transcript of BROKERAGE AND BROKERING: AN INTEGRATIVE REVIEW AND … Files/Brokerage and... · course by the...