Blidend by Power - Article To Be Reviewed

Group Decis Negot (2017) 26:215–245
DOI 10.1007/s10726-016-9495-5
Blinded by Power: Untangling Mixed Results

Regarding Power and Efficiency in Negotiation
Ricky S. Wong1 · Susan Howard2
Published online: 20 July 2016

© Springer Science+Business Media Dordrecht 2016
Abstract Negotiators are often advised to seek information about their counterparts’
power. However, we know little about how such information affects negotiators’
behaviours and outcomes. Study 1 considered dyadic negotiations in which nego-
tiators have symmetric or asymmetric best alternatives to the negotiated agreement
(BATNAs). It also examined the impacts of (a)symmetry and knowledge of a counter-
part’s BATNA on agreement efficiency (indexed by joint gains), and how knowledge
alters negotiators’ realised power (indexed by percentage of resource claimed) in
BATNA-asymmetric negotiations. Studies 2 and 3 focussed on BATNA-asymmetric
negotiations. Study 2 tested the mechanism by which knowledge affects efficiency.
Study 3 considered the impacts of knowledge on equity concerns, perceived power and
information exchange about preferences. The findings indicate the following: knowl-
edge of BATNA asymmetries (rather than the existence of BATNA asymmetries)
adversely affects agreement efficiency; this knowledge increases strong negotiators’
focus on value claiming, judgement errors about counterparts’ preferences, perceived
power and realised power, but impedes their information-sharing behaviour about pref-
erences. Their focus on value claiming mediates the relationship between knowledge
and judgement errors, whereas judgement errors mediate the relationship between
their focus on value claiming and agreement efficiency. Furthermore, knowledge of
BATNA asymmetries leads to contrasting perceptions of fairness. Strong negotiators
B Ricky S. Wong
[email protected]
Susan Howard
[email protected]
1 Department of Supply Chain and Information Management, Hang Seng Management College,
Room D604-B1, Hang Shin Link, Siu Lek Yuen, Shatin, N.T., Hong Kong
2 The London School of Economics and Political Science, London WC2A 2AE, UK
123
216 R. S. Wong, S. Howard
with knowledge believe that a fair agreement should reflect their power advantage;
weak negotiators generally tend to judge fairness based on equality. Counterintuitively,
knowing one’s own strengths can lead to ‘winning’ a meagre prize and neglecting the
opportunity for value creation by trading-off on negotiated issues.
Keywords BATNA · Agreement efficiency · Power asymmetry · Knowledge ·

Fairness
1 Introduction
Negotiation is the attempt of two or more parties to agree on one or more issues. The
outcome varies according to several factors, including the negotiators’ options beyond
the negotiation. Scholars have shown that having better alternatives gives negotiators
an edge (Fisher and Ury 1981; Kim and Fragale 2005). However, negotiators often
have unequal power or alternatives. A growing body of research has thus begun to study
negotiators’ behaviour and the structure of outcomes in power-asymmetric negotia-
tions (Anderson and Thompson 2004; Coleman et al. 2010; Handgraaf et al. 2008;
Koning et al. 2011; Pietroni et al. 2008; Tost et al. 2012; Van Kleef et al. 2006; Wolfe
and Mcginn 2005). Some studies have examined dyads in which both negotiators were
unaware of their counterparts’ power (Gino and Moore 2008; Magee et al. 2007; Van
Kleef et al. 2006). Others have investigated dyads in which both negotiators were
aware of the power dynamics at play (Belkin et al. 2013; Malhotra and Gino 2011;
Schaerer et al. 2015). Such designs do not allow us to understand how knowledge of
counterparts’ power affects negotiators’ behaviours and outcomes.
Scholars frequently advise negotiators to acquire information about their coun-
terparts’ alternatives before negotiating (Thompson 2012; Young et al. 2012).
However, this raises a question: “If negotiators successfully obtain knowledge of
their counterparts’ power in power-asymmetric negotiations, how will their behav-
iours and outcomes change?” Little research has examined the effects of knowledge
in power-asymmetric negotiations. In this research, we contend that knowledge of
power asymmetries affects several factors, including value creation, communications
between negotiators and negotiators’ perceptions of power and fairness.
Opportunities for value creation are available when negotiators have different pref-
erences across negotiated issues in a multi-issue negotiation (Thompson 2012). For
instance, in a negotiation between a union and an employer, the union may think
working hours more important than transportation subsidies while the employer may
consider subsidies to be more important. Ideally, negotiators would recognise the dif-
ferences in their preferences across issues and give up on less important issues in
exchange for concessions by their counterparts on more important issues. This would
result in more efficient agreements (indexed by high joint gains). However, in reality,
negotiators generally settle for inefficient agreements (Pinkley et al. 1995; Thompson
1990, 1991, 2012; Thompson and Hastie 1990; Thompson et al. 2010).
It remains unclear whether power-asymmetric dyads or power-symmetric dyads
are better at value creation. Mixed findings present an open question: do power
(a)symmetries improve agreement efficiency? (Greer and Kleef 2010; Mannix and
123
Blinded by Power: Untangling Mixed Results Regarding Power. . . 217
Neale 1993; Pinkley et al. 1994; Roloff and Dailey 1987; Wei and Luo 2012; Wolfe and
Mcginn 2005) The current study examines the interplay among power (a)symmetries,
knowledge of counterparts’ power and agreement efficiency.
In terms of theory, our research represents a significant departure from research on
power-asymmetric negotiations. Previous research has emphasised how power affects
behaviours and outcomes (Mannix and Neale 1993; Tost et al. 2012; Van Kleef et al.
2006; Wolfe and Mcginn 2005). However, it has not considered the possibility that
negotiators without knowledge of their counterparts’ power are inattentive to their
own relative power advantage or disadvantage. In the present study, we investigate
the effects of knowledge on judgements of counterparts’ interests and communication
about interests—both of these have been shown to be important in value creation
(Pinkley et al. 1995; Thompson 1991). We also consider how knowledge may affect
negotiators’ focus on value claiming and what makes an agreement fair.
Our research consists of three experimental studies. Study 1 examines whether
knowledge of one’s counterpart’s power affects value creation and value claiming in
both power-symmetric and power-asymmetric negotiations. Studies 2 and 3 focus on
power-asymmetric negotiations. Study 2 proposes and tests the mechanism by which
knowledge of power asymmetries may both help and hinder negotiators’ outcomes.
Study 3 demonstrates how this knowledge changes communications about preferences
between negotiators. Study 3 also tests the effects of knowledge on how negotiators
perceive their own power (relative to their counterparts) and how they determine the
fairness of an agreement. We provide a causal model demonstrating that knowledge
of power asymmetries—despite helping powerful negotiators perform better at value
claiming—hurts agreement efficiency. Our research findings encourage negotiation
scholars to consider new questions about the consequences of learning counterparts’
power in power-asymmetric negotiations. The practical implications of our findings
are also discussed.
1.1 Theoretical Considerations: Power in Negotiation
Power may stem from different sources in the context of negotiation, including nego-
tiators’ dependence on the existing negotiation and their social status (Greer and Kleef
2010; Mannix and Neale 1993; Pinkley 1995). Relational theorists argue that power is
a relational variable, tangible when comparing one’s power status to one’s opponent’s
(Emerson 1962; French and Raven 1959). It is rare for one party to have absolute
power over its opponent.
Power has been shown to have profound impacts on a negotiator’s performance (e.g.,
Kim and Fragale 2005). Although empirically comparing different operationalisations
of power is beyond the scope of our study, it is useful to discuss the three major power
operationalisations in negotiation literature. First, a major source of negotiators’ power
is the “best alternative to a negotiated agreement” (BATNA). A BATNA determines
what a negotiator would walk away with in case of an impasse (Fisher and Ury 1981;
Giebels et al. 2000; Kim et al. 2005; Komorita and Leung 1985; Pinkley 1995). Using
BATNAs to manipulate negotiators’ power is common in experimental studies (Brett
et al. 1996; Gino and Moore 2008; Kim and Fragale 2005; Pinkley 1995; Pinkley et al.
123
1994; Wolfe and Mcginn 2005). Second, power can be operationalised via the number
of alternatives. This refers to the number of alternate negotiators if an agreement
cannot be reached (Mannix and Neale 1993; McAlister et al. 1986; Olekalns and
Smith 2013, 2009). Third, Giebels et al. (2000) manipulated negotiators’ power status
by giving either party or both an exit option. That is, the negotiator could opt to
engage with an alternative negotiator at any moment during negotiation. These three
operationalisations of power create differences in the extent to which negotiators rely
on the existing negotiation.
Kim et al. (2005) integrated different theories of power into a dynamic and concep-
tual model that clarified the relationship between power and performance in distributive
negotiation. Kim et al. (2005) partitioned power into four components: potential power,
perceived power, tactics and realised power. Potential power refers to the extent to
which negotiators may have the capacity to obtain benefits from their negotiated
agreement. Negotiators’ potential power is determined by their dependence on the
negotiation. An attractive BATNA is a source of potential power because this allows
negotiators to secure a reasonably good deal elsewhere.
Power is a relational variable. Perceived power regards how negotiators perceive
their power vis-à-vis their counterparts in the relationship. This requires negotiators’
consideration of their own potential power and that of their counterparts. Central to
the contention in Kim et al. (2005) is the negotiators’ awareness of their counterparts’
BATNAs. In BATNA-asymmetric negotiation, higher potential power (e.g., an attrac-
tive BATNA) does not necessarily lead to higher perceived power. We argue that in
power-asymmetric negotiations, knowledge of others’ power, rather than the existence
of power asymmetries, increases negotiators’ perceived power.
The ‘power-use tactic’ is a way by which negotiators may attempt to leverage
existing power capabilities (i.e., claiming more values). Kim et al. (2005) proposed
that negotiators are more likely to adopt power-use tactics when they perceive that
they are more powerful than their counterparts. Realised power refers to the extent to
which negotiators claim benefits from negotiation; that is, the portion of the ‘pie’ that
they are able to claim indicates their realised power in the negotiation.
Our experimental studies examine whether negotiators’ perceived power and
realised power differ as a function of knowledge of power asymmetries. Although
the studies of Kim and Fragale (2005) and Kim et al. (2005) advance our understand-
ing of power in distributive negotiation, we know little about the role of knowledge
in variable-sum negotiation. Our research extends that of Kim et al. (2005) to include
power-asymmetric negotiations. Before discussing the power implications of this
knowledge, we examine why it warrants attention.
1.2 Why Knowledge of Counterparts’ Power Matters
Scholars emphasise the importance of estimating counterparts’ BATNAs as well as

improving one’s own BATNA prior to negotiation (Neale and Bazerman 1991; Thomp-
son 2012). It has recently become much easier for a negotiator to research counterparts’
power owing to advanced technologies and the increased availability of information;
for example, in the case of a house sale, both the buyer and seller can access infor-
123
mation about the market selling price of a house in similar ways. This explains why
some recent research on power has examined dyads in which actors had knowledge
of their counterparts’ power (although this power was manipulated in different ways)
(Greer and Kleef 2010; Kim and Fragale 2005; Malhotra and Gino 2011; Nelson
et al. 2015; Wolfe and Mcginn 2005; Schaerer et al. 2015). We have learned that
knowledge of counterparts’ power makes a difference. For instance, two studies have
demonstrated that negotiators’ initial offers are shaped by their knowledge of their
counterparts’ BATNAs, or by being primed to concentrate on counterparts’ ability to
obtain attractive alternatives (Buelens and Poucke 2004; Maaravi et al. 2011).
We contend that knowledge of a counterpart’s power may also affect a negotiator’s
approach, behaviour and outcomes, especially in power-asymmetric negotiations. To
this end, we draw on past research that examines how negotiators estimate their coun-
terparts’ positions prior to negotiation. Thompson and Hastie (1990) have shown that
negotiators tend to base their perceptions of others on their own situations. With-
out the knowledge of their counterparts’ potential power, most negotiators tend to
underestimate the power differences between themselves and their counterparts in
power-asymmetric negotiations (Pinkley et al. 1994; Wong 2014). Wong (2014) found
that when negotiators understood the possible range of their counterparts’ BATNAs,
their estimates were still largely anchored to their own BATNAs. We argue that in
BATNA-asymmetric negotiations, negotiators will not necessarily perceive themselves
to be in a position of higher (or lower) power. Thus, negotiators’ knowledge of their
counterparts’ power does matter in power-asymmetric negotiations. Conversely, this
knowledge may matter less in power-symmetric negotiations because it merely con-
firms negotiators’ perceptions of counterparts’ power. Next, we discuss the extant
literature that considered the relationship between power differences and agreement
efficiency.
1.3 Inconsistency in Power Differences and Agreement Efficiency
Empirical research has provided mixed results on whether power (a)symmetries (or
BATNA (a)symmetries) affect agreement efficiency. Past research employing BATNA
manipulation suggests that dyads with asymmetric BATNAs reach more efficient
agreements than those with symmetric BATNAs (Pinkley et al. 1994; Roloff and
Dailey 1987; Wei and Luo 2012). In particular, Pinkley et al. (1994) have shown that
dyads consisting of one party with no specified BATNA and another with an attractive
BATNA (but less attractive than the compromise solution) generate a sufficient imbal-
ance in BATNAs to improve efficiency. However, when there is a relatively smaller
BATNA imbalance between parties (i.e., with one party possessing a BATNA that is
worth more than the compromise solution and another having a BATNA less attractive
than the compromise solution), there are no effects on agreement efficiency.
In contrast, the stream of research using other power manipulations has shown that
dyads of equal power are more likely to reach solutions of higher efficiency than those
with an unequal power balance (Lawler and Yoon 1993; Mannix and Neale 1993).
These studies adopted the ‘number of alternatives’ and the ‘existence of an alternative’
operationalisations of power to create power imbalances between negotiators.
123
There are contradictions within this set of findings. For one, why do dyads with
unequal power sometimes reach more efficient agreements than those with equal power
and sometimes vice versa? Does power asymmetry in fact matter? No a priori theory
suggests that different power manipulations change the relationship between power
asymmetries and agreement efficiency. We speculate that the variability of negotiators’
knowledge of their counterparts’ power may be the key. An assumption of complete
knowledge about counterparts’ power has been made in previous studies, suggesting
that dyads with equal power reach more efficient agreements (Lawler and Yoon 1993;
Mannix and Neale 1993). Giebels et al. (2000), for example, made both negotiators
aware of the exit option even though only one member of the dyad had the power to opt
out. That is, negotiators knew their own and the other’s power status or BATNAs. Wolfe
and Mcginn (2005) did not provide the exact value of the other party’s alternative, but
provided some information. For instance, in the experimental conditions where the
opponents had a low BATNA, the negotiators were told that their opponents were
not happy with their BATNAs. When a high BATNA was assigned to opponents, the
negotiators were told that the other party had a very attractive BATNA that would be
difficult to beat.
However, Pinkley et al. (1994) and Roloff and Dailey (1987), suggesting that
unequal BATNA dyads reached more efficient agreements, did not assume com-
plete knowledge. It is unclear from the descriptions of their experimental design to
what extent negotiators shared information about their BATNAs. Wei and Luo (2012)
created asymmetry with knowledge of counterparts’ BATNAs. They provided knowl-
edge only to negotiators with low BATNAs under a BATNA-asymmetric experimental
condition. Although not specified in their study, we do not expect that low-BATNA
negotiators would reveal their BATNA disadvantage to counterparts. Thus, we do not
expect that the high-BATNA negotiators knew their BATNA advantage in their study.
In contrast, Brett et al. (1996) and Pinkley (1995) found that when negotiators only
knew their own BATNAs and were told not to reveal their BATNAs to counterparts,
BATNA asymmetries did not affect agreement efficiency.1
Study 1 considers both BATNA-asymmetric and BATNA-symmetric negotia-
tions. It explores the possibility that knowledge of counterparts’ BATNAs damages
agreement efficiency in power-asymmetric negotiations (but not in power-symmetric
negotiations). Specifically, power asymmetry per se is necessary but not sufficient to
affect agreement efficiency.
1.4 Knowledge, Negotiation Approach and Judgement in Power-Asymmetric

Negotiation
The present research also addresses the mechanism by which knowledge of power
asymmetries affects agreement efficiency. Here, we confine our attention to BATNA-
asymmetric negotiations. For the sake of brevity, in our studies involving BATNA
1 In Pinkley’s (1995) study, when negotiators only knew their own BATNAs, the impact of BATNA asym-
metries on agreement efficiency disappeared.
123
Strong
Negotiators’ Focus on
Knowledge Judgement Inefficient
the Agreements
of BATNA- Distributive Error
Asymmetries Element
Fig. 1 A proposed theoretical model of strong negotiators’ knowledge of BATNA-asymmetries and agree-
ment efficiency
asymmetries, negotiators with a more attractive BATNA and those with a less attractive
BATNA are referred to as strong negotiators and weak negotiators, respectively.
It is speculated that knowledge of BATNA asymmetries may change negotiators’
negotiation styles and mind-sets, which in turn affects a dyad’s ability to search for
efficient outcomes. This knowledge may signal to negotiators that their counterparts
rely on the negotiation to a relatively greater or lesser extent. As a result, a negotia-
tor’s approach can differ as a function of knowledge of BATNA asymmetries. The
mechanism involves two steps (to be tested in Study 2): first, knowledge of BATNA
asymmetries affects the negotiators’ focus on the distributive element and second,
it affects their judgement accuracy about the others’ preferences across issues (see
Fig. 1).
Before detailing the proposed mechanism, a brief review of relevant research is
useful. Research on social power has shown that the possession of power increases
an individual’s inclination to actively approach rewarding outcomes and to focus on
personal gains (Gruenfeld et al. 2008; Inesi et al. 2012; Keltner et al. 2003). Power
seems to adversely affect an individual’s judgements of other people: high-power
individuals are inaccurate judges of others’ emotions and intentions and have a lower
tendency and motivation to understand the feelings and thoughts of others (Galinsky
et al. 2006; Gonzaga et al. 2008; Inesi et al. 2012). Therefore, awareness of one’s own
power advantage may affect how negotiators approach negotiations and judgement of
a counterpart’s interests.
Studies on power-asymmetric negotiations have found that negotiators in positions
of higher power are likely to expect a resource distribution based on equity rather
than equality2 (Bacharach and Lawler 1981; Komorita 1984; Komorita and Hamilton
1984; Lawler and Yoon 1993; Mannix 1993; Shaw 1981). Kim et al. (2005) also
conjectured that negotiators would be more likely to initiate power-use tactics when
they perceived their own potential power to be higher than that of their counterparts.
Similarly, Olekalns and Smith (2013) contended that power asymmetries invite strong
negotiators to display power and weak negotiators to show submission in negotiation.
Central to these arguments is the negotiator’s awareness of power asymmetries.
Extending these predictions, knowledge of BATNA asymmetries increases strong
negotiators’ perceived power. This knowledge may change strong negotiators’ expec-
tations of resource distribution. It may be that strong negotiators, when informed of
all BATNAs, express their superiority distributively to push for agreements that reflect
2 Note that the definition of power in these studies differs from ours. For example, power was represented
in some studies by the number of alternatives that negotiators have; in others, the difference in power was
manipulated by varying the probabilities of various profits of the alternatives.
123
their BATNA advantage. In other words, informed strong negotiators are likely to
focus on the distributive element of the negotiation and to demonstrate higher realised
power (as reflected by a larger portion of surplus received).
The second step of the mechanism is that focussing on the distributive element
may affect how strong negotiators perceive the structure of negotiations. A large body
of research has shown that negotiators often suffer from fixed-pie bias at the outset
of negotiations (Thompson and Hastie 1990; Thompson 1991; Bazerman and Neale
1983). In light of the pervasiveness of fixed-pie perceptions, strong negotiators’ knowl-
edge of BATNA asymmetries may make this powerful bias more likely to persist during
negotiation.3 This occurs because focussing on the distributive side of the negotiation
may steer their available cognitive capacity away from the creation of values (Lax and
Sebenius 1986) and towards claiming values (Mannix and Neale 1993). Focussing on
the distributive element (i.e., value claiming) makes it difficult for informed strong
negotiators to consider opportunities for value creation. As a result, informed strong
negotiators make more judgement errors about their counterparts’ preferences than
uninformed strong negotiators. We argue that the relationship between knowledge
and judgement errors is to some extent mediated by a negotiator’s focus on the dis-
tributive element.
In contrast, knowledge of this kind may make weak negotiators self-aware that they
are more reliant on the negotiation than their counterparts. Knowledge is thus expected
to reduce weak negotiators’ perceived power. Informed weak negotiators are likely
to be less competitive and focus less on the distributive element than uninformed
weak negotiators. This is because knowing themselves to be in a disadvantageous
position does not justify a larger bargaining surplus than their stronger counterparts.
As a result, when weak negotiators know that they are weaker, they will show less
realised power (as indexed by a lower portion of resource pie claimed). In terms
of judgement errors, we are uncertain at this stage as to whether interacting with
others who focus on value claiming will adversely affect negotiators’ judgements
of the preferences of others. Study 2 will fully examine how knowledge of BATNA
asymmetries affects negotiators’ focus on the distributive element and judgement
errors about their counterparts’ preferences.
1.5 Knowledge and Information-Sharing Behaviour
Knowledge of power asymmetries may also affect communications between negotia-

tors about their interests in variable-sum negotiation. Past studies have demonstrated
that information exchange about preferences across issues is key to the development
of efficient agreements because this improves negotiators’ judgement accuracy about
their counterparts’ preferences (Murnighan et al. 1999; Thompson 1991; Thompson
and Hastie 1990; Walton and McKersie 1965). Information exchange enables nego-
tiators to identify opportunities for value creation.
3 It is important to note that we are not suggesting that only informed strong negotiators suffer from fixed-
pie bias. Instead, it is speculated that informed strong negotiators are more likely to suffer from this bias
than those who lack knowledge of BATNA asymmetries.
123
As argued above, knowledge of BATNA asymmetries increases strong negotiators’

judgement errors. With greater judgement errors, strong negotiators with knowledge
rarely conduct an active search for information that goes against their existing ideas
and attitudes. For instance, they are less likely to search for information about their
counterparts’ preferences across issues. After all, what is the use of learning something
about their weaker counterparts that merely confirms their expectations? Knowledge
may also have a deterrence effect on weak negotiators’ information-sharing behav-
iour. Facing a stronger counterpart who focuses on value claiming makes it harder for
weak negotiators to think creatively, as they need to secure their outcomes. As a result,
informed weak negotiators may be less likely to share information about their prefer-
ences than uninformed ones.4 Study 3 will examine whether knowledge of BATNA
asymmetries blocks communication.
1.6 Knowledge and Perceptions of Fairness in Power-Asymmetric Negotiation
Thus far, we have discussed how knowledge of BATNA asymmetries may shape
outcomes, perceived power, and the way negotiators approach negotiation and
information exchange during negotiation. This research also examines another impor-
tant pre-negotiation parameter, namely, negotiators’ perceptions of fairness, which
helps explain why knowledge of BATNA asymmetries induces different negotiation
approaches.
What seems fair to one party may not be considered fair by another (Bazerman
et al. 2002). We know little about how negotiators determine fairness in BATNA-
asymmetric negotiations. Given the complexity of negotiation structure and power,
we expect that strong and weak negotiators do not necessarily share similar views
on what makes an agreement fair. For example, some may consider an even split of
resources fair, while others take the perspective that a fair agreement should reflect
one’s power advantage or disadvantage. Previous studies have examined self-serving
and egocentric interpretations of fairness in negotiation (Babcock et al. 1995; Gelfand
et al. 2002; Roth and Murnighan 1982; Thompson and Loewenstein 1992). Thompson
and Loewenstein (1992) found that when faced with ambiguity, individuals’ judge-
ments tended to be biased in a manner that favoured themselves. This bias tends to be
stronger when negotiators are in asymmetric roles than when they are in symmetric
roles (Wade-Benzoni et al. 1996).
We posit that knowledge of BATNA asymmetries creates diverse views about what
constitutes a fair agreement from the perspectives of strong and weak negotiators. On
the one hand, one may speculate that informed weak negotiators might think their
stronger counterparts deserve a bigger slice of the resource pie. However, research
concerning egocentric interpretations of fairness suggests the opposite (Epley et al.
2006; Paese and Yonker 2001). Babcock et al. (1995) and Paese and Yonker (2001)
have demonstrated that negotiators will give more weight to the information that
4 Mannix and Neale (1993) conjectured that weak negotiators might be forced to think creatively and come
up with alternative solutions. As more recent studies have shown that power symmetries, rather than power
asymmetries, facilitate value creation (Giebels et al. 2000; Wolfe and Mcginn 2005), we do not expect that
knowledge of power asymmetries increases a negotiator’s ability to create value.
123
favours their side. Even when weak negotiators learn about their power disadvantage,
their concerns about equality are salient. They may determine that an even division of
resources is the foundation of a fair agreement. Similarly, we predict that knowledge
also changes strong negotiators’ assessments of what constitutes a fair agreement in
an egocentric, self-serving manner. Informed strong negotiators would perceive that
a fair agreement should reflect their BATNA advantage.
In summary, Study 1 considers both BATNA-symmetric and -asymmetric negotia-
tions. It examines whether knowledge of counterparts’ BATNAs affects outcomes, in
terms of agreement efficiency and negotiators’ realised power. Studies 2 and 3 only
focus on BATNA-asymmetric negotiations. Study 2 empirically tests how knowledge
of BATNA-asymmetries changes negotiators’ approaches to negotiation and judge-
ments about their counterparts’ preferences. Study 3 considers how knowledge shapes
negotiators’ perceived power, perceived fairness, and information-sharing behaviour
about preferences across issues.
2 Study 1
Study 1 is designed to clarify the relationship among BATNA (a)symmetries, knowl-

edge of counterparts’ BATNAs and agreement efficiency. We argue that the different
patterns of the relationship between power asymmetry and agreement efficiency stem
from negotiators’ varying levels of knowledge about their counterparts’ power. We
posit that knowledge of a counterpart’s BATNA changes agreement efficiency in
BATNA-asymmetric negotiations. Furthermore, when negotiators are not informed of
their counterparts’ BATNAs, there is no difference in agreement efficiency between
dyads with symmetric BATNAs and those with asymmetric BATNAs. Recall that
complete knowledge of another negotiator’s power was assumed in previous studies,
suggesting that equal power leads to more efficient outcomes than unequal power. We
hypothesise that in BATNA-asymmetric negotiations knowledge of another’s BATNAs
hinders a dyad’s ability to reach efficient outcomes. As argued, knowledge of another
negotiator’s power may not affect agreement efficiency in power-symmetric negoti-
ations. This is because this knowledge merely confirms negotiators’ expectations of
others’ power in this situation.
Hypothesis 1 Unequal-BATNA dyads with knowledge of their counterparts’ BAT-

NAs reach less efficient agreements than unequal-BATNA dyads without knowledge
and equal-BATNA dyads with or without knowledge.
We also examine whether knowledge of BATNA asymmetries affects strong and

weak negotiators’ power-use tactics differently, resulting in different levels of realised
power (indexed by the percentage of surplus claimed). As with the assumption in Kim
et al. (2005), in BATNA-asymmetric negotiations when strong (weak) negotiators are
aware of their counterparts’ BATNAs, they are more (less) likely to employ power-use
tactics. Consequently, this knowledge is expected to increase (decrease) the portion of
the resource pie that strong (weak) negotiators will claim. Knowledge of counterparts’
BATNAs increases (decreases) strong negotiators’ (weak negotiators’) value-claiming
ability.
123
Hypothesis 2a When strong negotiators are informed of BATNA asymmetries, they

claim a larger proportion of the bargaining surplus than when they are not.
Hypothesis 2b When weak negotiators are informed of BATNA asymmetries, they

claim a smaller proportion of the bargaining surplus than when they are not.
2.1 Method
2.1.1 Participants
Two hundred and eighty-eight undergraduate and master students at the London School
of Economics (United Kingdom) and the Hang Seng Management College (Hong
Kong) participated in a negotiation experiment. The sample included 152 men and
136 women, ranging from 18 to 43 years of age, with a mean of 23.65 (SD = 4.08)
years. To control for potential cultural differences between the UK and Hong Kong
samples, the ratio of UK subjects to Hong Kong subjects was kept the same in all
experimental conditions (72 Hong Kong subjects in each condition). As an incentive,
subjects were informed that the money that they received at the end of the experiment
was related to the number of points they earned; they received £0.10 (or equivalent)
for every 100 points earned.
2.1.2 Procedure
Subjects were randomly assigned to experimental conditions and roles, and received
a sheet of paper describing the negotiation task before the exercise began. The exper-
imenter provided subjects with specific negotiation instructions, a “payoff” chart,
details about their role and own BATNAs, information about their counterparts’ BAT-
NAs (if applicable) and a short quiz to ensure that subjects understood their BATNAs
and the payoff chart. The instructions, information and quiz were given in writing
on paper. Subjects were tested individually before being paired with another subject
to negotiate. The experimenter checked the answers to every question; subjects with
errors were told to attempt the question again. Most subjects were correct on their
first attempt and all were correct on their second attempt. Subjects negotiated for a
maximum of 30 min. They negotiated face-to-face in private rooms, out of earshot of
other groups. Their interaction was unrestricted, except that subjects were informed
that the payoff schedule was confidential and should not be shared with the other
party.
All participants were asked to complete questionnaires. The pre-negotiation ques-
tionnaire included a number of demographic questions and was given after reading
the initial role materials and receiving details about their own BATNAs. The post-
negotiation questionnaire asked participants to indicate whether they had revealed
their own BATNAs to their counterparts during the negotiation. After participants
completed this questionnaire, they were debriefed about the purpose of the experi-
ment.
123
2.1.3 Negotiation Task
The negotiation simulation used was a variable-sum task. The negotiation involved an
employer and an employee resolving six issues in a job contract, including different
options on the following issues: salary, annual leave, bonus, starting date, medical
coverage and company car. “Appendix” describes all the possible ways participants
could settle this negotiation. There were several alternatives for each issue. Each party
had different preferences for alternatives defined by the points the negotiator would
receive if that alternative was agreed upon.
The task included three types of issues: distributive, compatible and integrative.
Salary was a purely distributive issue: when one party gains, the other party loses in a
direct, fixed-sum fashion. Starting date was one in which both parties have perfectly
compatible interests. In this negotiation task, there were two possible fully integrative
trade-offs; the preferences were inverse meaning that one party has to place a higher
value on one issue and a lower value on another. Negotiators had different priorities
for the annual leave and bonus issues and could logroll these to maximise joint gain
(e.g., the employer giving the employee a higher bonus for less annual leave). Addi-
tionally, they had different priorities for medical coverage and company car issues.
Therefore, this negotiation simulation allowed for variation among the integrative
outcomes.
Negotiators could earn a maximum of 12,800 points or a minimum of 0 points. An
obvious compromise solution would yield each negotiator 6400 points for a joint total
of 12,800 points. A more mutually beneficial agreement was possible if negotiators
logrolled issues and realised the same preference for starting date. The maximum
possible joint outcome was 18,800 points.
2.1.4 Experimental Manipulations and Measures
To fully test the hypotheses developed, we adopted a 3 × 2 [BATNA (A)symmetries:

both high, both low or unequal BATNAs × Knowledge: no knowledge or knowledge]
between-subjects factorial design. This resulted in six experimental conditions: (1)
both negotiators had high BATNAs and neither knew their opponent’s BATNA (Both
Strong & No Knowledge); (2) both had low BATNAs and neither knew their opponent’s
BATNA (Both Weak & No Knowledge); (3) negotiators had unequal BATNAs and
neither knew their opponent’s BATNA (Unequal & No Knowledge); (4) both had high
BATNAs and they knew each other’s BATNA (Both Strong & Knowledge); (5) both
had low BATNAs and they knew each other’s BATNA (Both Weak & Knowledge); and
(6) negotiators had unequal BATNAs and they knew each other’s BATNA (Unequal
& Knowledge).5
In Both Strong (Both Weak) conditions, employers and employees would receive
6000 points (1200 points) if no agreement was reached. In Unequal conditions,
employers would receive 6000 points and employees would receive 1200 points in
the case of an impasse. One might argue that employers (employees) being strong
5 The conditions Both Strong & Knowledge and Both Weak & Knowledge were added to the original design
of Study 1. Hence, the assignment was not fully randomised.
123
(weak) negotiators may have created more than just BATNA differences, and any
differences between strong and weak negotiators may be attributable to their roles
rather than their BATNAs. However, past research suggests that this is unlikely to
be an issue.6 The current study concerns the absolute difference across experimental
conditions. As a result, any difference in role should not interfere with the validity of
the hypotheses. In addition, the employers’ BATNAs were slightly less than the value
of the compromise solution, 6400 points. In most cases, the negotiators had at the
very least an alternative (which may not be attractive), and to strengthen the external
validity, negotiators’ BATNAs were worth 1200 points in Both Weak and Unequal
conditions.
Agreement Efficiency The measure of agreement efficiency for each dyad is joint
outcome. Higher joint outcome indicates more efficient agreements.
Realised Power The measure of the realised power for negotiators is the percentage
of total bargaining surplus claimed. Note that the summed total of the percentages of
the surplus claimed by weak and strong negotiators always equal 100 %.
Manipulation Checks Subjects were asked to specify the number of points they would
receive in the case of an impasse. Less than 2 % of participants gave the wrong answers
in the first trial. All of them were correct on their second attempt. All negotiators who
were informed of their opponent’s BATNAs correctly reported those BATNAs.
As mentioned, subjects were allowed to freely communicate with their opponents
during the negotiation. It is important to consider whether negotiators revealed their
own BATNAs. It was expected that negotiators with a less attractive BATNA would
be unlikely to disclose their BATNAs, particularly when they knew that they were
in the weaker position. However, negotiators with an attractive BATNA in Knowl-
edge & Unequal Condition knew that their BATNAs were better than those of their
counterparts. They might have had an incentive to reveal their BATNA advantage to
their weaker counterparts. As a result, this could contaminate (or at least weaken) the
manipulation of knowledge of BATNA asymmetries.
To check this potential design limitation, subjects were asked whether they had
revealed their BATNAs to their counterparts during the negotiation. Only four (of
288) negotiators reported that they revealed their BATNAs to others. Two of the
four were negotiators with a better BATNA in the Unequal & No Knowledge con-
dition. The other two were the negotiators with a less attractive BATNA in the
Unequal & Knowledge condition. Thus, the revelation of their own BATNAs merely
confirmed the information given to their stronger counterparts. Therefore, there is
evidence that the manipulation of knowledge of counterparts’ BATNAs was robust as
intended.
6 Pinkley (1995) considered the potential effect of role in job contract negotiations but was not found to have
a significant impact on pre-negotiation parameters (i.e., reservation price, aspiration levels) and negotiated
outcomes.
123
Table 1 Effects of knowledge of opponents’ BATNAs and BATNA-differences on joint outcomes
Source SS df MS F p
Knowledge 3,641,736 1 3,641,736 1.409 0.237

BATNA-differences 5,987,917 2 2,993,958 1.158 0.317
Interaction 15,788,472 2 7,894,236 3.054 0.05
Explained 25,418,125 5 5,083,625 1.967 0.087
Residual 356,700,000 138 2,584,972
Total 40,190,000,000 144
Table 2 Means (standard deviations) of joint outcomes in Study 1
Both Strong & Both Weak & Unequal & Both Strong & Both Weak & Unequal &
No Knowledge No Knowledge No knowledge Knowledge Knowledge Knowledge
Joint 16,800a 16,541a 17,017a 16,975a 16,667a 15,763b

Outcomes (1569) (1629) (1270) (1410) (1896) (1786)
Subscripting is based upon comparisons of means using pairwise contrasts; different subscripts indicate
means differ at p < .05 or less. (e.g. joint outcomes in unequal and no knowledge condition are given the
subscript ‘a’ and it is significantly different to that for unequal and knowledge condition given subscript
‘b’)
2.2 Results and Discussion
2.2.1 What Affects Agreement Efficiency (BATNA Differences vs. Knowledge)
An univariate analysis of variance (ANOVA) was used to test the effects of knowledge
and BATNA (a)symmetries on agreement efficiency. No significant main effect of
knowledge was found for joint gain, F(1, 138) = 1.409, p = .237. Furthermore, the
main effect of BATNA (a)symmetries was not significant F(2, 138) = 1.158, p =
.317. Instead, a marginal, significant interaction (knowledge × BATNA differences)
was found on joint outcomes, F(2, 138) = 3.054, p = .05. This interaction suggests
that the effect of the negotiators’ knowledge on joint outcomes differed, depending on
the level of BATNA differences between negotiators (both low, both high or unequal)
(see Table 1 for related statistics).
Pairwise comparisons were conducted to clarify the relationships among BATNA
(a)symmetries, knowledge of counterparts’ BATNAs and agreement efficiency (see
Table 2). Hypothesis 1 predicted that unequal-BATNA dyads with knowledge reached
less efficient agreements than those without knowledge and equal-BATNA dyads
(with or without knowledge). The findings support this hypothesis. Focussing on
the dyads with unequal BATNAs, when negotiators had knowledge of their oppo-
nents’ BATNAs, joint gains were significantly lower than when they did not
(MUnequal & Knowledge = 15,763 vs. MUnequal & No Knowledge = 17,017), t (46) = 2.70,
p = .004, d = .83. Dyads in the Unequal & Knowledge condition (MUnequal & Knowledge
= 15,763) also reached lower joint outcomes than those in Both Strong & Knowledge
123
(MBoth Strong & Knowledge = 16,975), t (46) = 2.61, p = .005, d = .77 and Both Weak
& Knowledge (MBoth Weak & Knowledge = 16,667) conditions, t (46) = 1.95, p = .027,
d = .50. Knowledge of counterparts’ BATNAs hindered dyads from reaching efficient
solutions in BATNA-asymmetric negotiations. These results clarify the relationship
between BATNA (a)symmetries and agreement efficiency. BATNA asymmetries per
se do not necessarily damage joint gains, unless both parties are aware of power
imbalances.
2.2.2 Knowledge and Realised Power
We conducted an independent t test to examine whether knowledge of BATNA

asymmetries significantly increases the percentage of resource pie that strong nego-
tiators claimed, t (46) = 3.75, p < .0005, d = 1.11. The percentage of bargaining
surplus claimed by informed strong negotiators was significantly higher than unin-
formed strong negotiators (MUnequal & Knowledge = 55.2 % vs. MUnequal & No Knowledge =
47.5 %). Because the percentages of the resource pie claimed by members within
the same dyad were summed to 100 %, the effect of knowledge on weak negotia-
tors’ realised power was also significant. The findings support Hypotheses 2a and 2b.
Knowledge of BATNA asymmetries increases strong negotiators’ realised power but
decreases that of weak negotiators.
2.2.3 Effects of Sample Source
No sample differences (United Kingdom vs. Hong Kong) were found in joint gains,
t (142) = −.332, ns and negotiators’ realised power, t (142) = .418, ns.
3 Study 2
Study 1 produced evidence suggesting that when both negotiators have knowledge of
BATNA asymmetries, dyads tend to reach less efficient outcomes than when negotia-
tors possess no knowledge of such asymmetries. Study 2 was designed to replicate this
finding and to examine how knowledge reduces efficiency. Specifically, we tested the
mechanisms by which knowledge of BATNA asymmetries affects the way negotiators
approach negotiations. As stated above, strong negotiators’ knowledge of BATNA
asymmetries triggers a two-step mechanism: first, it induces an increased focus on the
distributive element and second, a greater focus on value claiming means that judge-
ment errors about their counterparts’ preferences are more likely. If true, this helps to
explain why negotiators’ knowledge of BATNA asymmetries hinders a dyad’s ability
to reach an efficient agreement (see Fig. 1). Study 2 also investigates the impact of
knowledge on weak negotiators’ focus and judgement errors.
It is also important to examine whether a strong negotiator’s focus on the distribu-
tive element mediates the relationship between knowledge and judgement errors (see
Fig. 1). This is because, for example, one might argue that an increase in judgement
errors leads to a greater focus on the distributive element than vice versa.
123
Table 3 Items for negotiators’ focus on distributive element
1. I was very concerned if I could outperform the employee (employer)

2. My primary concern in the negotiation task was whether I could claim more surplus on the table than
the opponent did
3. As long as I enjoyed the negotiation, I was not very concerned if the other party earned more than I did
4. I think a fair agreement would be the one that reflected the quality of my BATNA (outside option)
5. The main goal I pursued was to do better than the opponent
Hypothesis 3a When strong negotiators are informed of both BATNAs, they focus
on how to divide the resources to a greater degree than those without this information.
Hypothesis 3b When weak negotiators are informed of both BATNAs, they focus on
how to divide the resources to a lesser extent than those without this information.
Hypothesis 4 Informed negotiators show higher judgement errors about their coun-
terparts’ preferences than uninformed negotiators.
3.1 Method
3.1.1 Participants
A total of 96 undergraduate and master students at the London School of Economics

and University College London participated in this study. The sample included 50 men
and 46 women, with ages ranging from 19 to 43 years and a mean of 24.86 (SD = 3.66)
years. Subjects participated in the experiment as volunteers. They received £0.10 for
every 100 points earned.
3.1.2 Procedure
The instructions and procedures were the same as those used in Study 1, except that
additional post-negotiation questionnaires were used (see Dependent Measures for
details).
3.1.3 Negotiation Task and Independent Variables
The negotiation task was identical to that of Study 1. Study 2 focussed on BATNA-
asymmetric negotiations. There were two experimental conditions: Unequal & No
Knowledge and Unequal & Knowledge.
3.1.4 Dependent Measures
To assess whether negotiators focussed on the distributive element, we asked the par-
ticipants to state the extent of their agreement with the statements illustrated in Table 3.
Negotiators responded on a 7-point Likert scale (1 = strongly disagree, 7 = strongly
123
agree). To avoid response sets, one statement (3) was worded in the reverse direc-
tion. Responses to these five items were summed to obtain a focus score for each
negotiator.
Judgement Errors about Others’ Preferences Following each negotiation, the exper-
imenter provided each negotiator with a blank payoff schedule and the following
instructions:
Below is a blank payoff schedule similar to the one that has been given to you in
this negotiation situation. At the time, we would like you to “fill in the numbers”
to indicate what you think the other negotiator’s payoff schedule looks like. Your
only hint is that the lowest number on their chart is zero and the highest is 4,000.
From this fill-in-the-blank questionnaire, the measures of judgement errors were
computed by examining deviations between negotiators’ estimates and the true val-
ues (Thompson 1990a, b). Judgement errors measured whether negotiators accurately
perceived that four of the issues (e.g., annual leave, bonus, medical coverage and
company car, see “Appendix”) differed in importance to the other party.7 Negotiators
who believe that the other party’s evaluation of the importance of issues is the same
as their own have a larger error score. The accuracy score was computed by summing
the absolute deviations of strong (weak) negotiators’ estimates from weak (strong)
negotiators’ actual values for the four logrolling issues.
3.2 Results
For dependent measures at the individual level, we examined the dyadic structures of
the data to avoid the potential biases of data nonindependence (Turel 2010). We fol-
lowed Alferes and Kenny’s (2009) method and measured the nonindependence for both
the focus score and judgement errors. The correlation coefficients (r ) demonstrated
that focus scores between strong negotiators and weak negotiators and judgement
errors were not correlated, t (46) = .39, p = .70; t (46) = −.12, p = .91.
All dyads were retained for the analyses that follow because all reached an agree-
ment. For focus scores and judgement errors measured at the individual level, we
analysed strong negotiators’ and weak negotiators’ responses in a single model via
hierarchical linear modelling. Both Knowledge and BATNA (strong or weak) were
treated as fixed parameters. We used the restricted maximum likelihood (REML) to
estimate parameters’ standard errors. The model was fitted using a heterogeneous
compound symmetry covariance structure. The results of both models are presented
in Table 4. We followed the procedure used in Raudenbush and Bryk (2002), that is,
the Level 1 equation was the unconditional model (with no independent variables)
and the Level 2 equation was the conditional model (with Knowledge, BATNA and
Knowledge × BATNA).
7 A complete analysis of accuracy in negotiation would also entail assessing negotiators’ perceptions of
the distributive issue (e.g., salary, see “Appendix”). This is not done in the present study because previous
analyses indicate little or no variance on this measure (Thompson 1990a, b).
123
Table 4 Effects of knowledge of BATNA-asymmetries on focus score and judgement errors
Factor Dependent measures at an individual level

Focus score Judgement errors
Knowledge of BATNA-asymmetries
Strong negotiator 6.125† 3862.500∗
Weak negotiator −0.583 485.417
N = 96. All terms are unstandardised regression coefficients. For the analyses predicting to focus scores
and judgement errors, the effects of knowledge were obtained through linear mixed modelling
* p < .05; † p < .001 (all two-tailed tests)
Table 5 Means (standard

Experimental condition
deviations) of focus score and
judgement errors in different Unequal & Unequal &
conditions No Knowledge Knowledge
Focus score
Strong negotiator 17.2 23.3
(6.8) (2.8)
Weak negotiator 18.6 18.0
N = 24 in each condition. (5.5) (4.5)
Maximum focus score = 35.
Higher focus scores indicate a Logrolling accuracy
greater degree to which Strong negotiator 11,900 15,762
negotiators focus on the (4949) (6290)
distributive element. Lower
logrolling accuracy scores Weak negotiator 12,152 12,638
indicate more accurate (6566) (6588)
judgements
3.2.1 Knowledge of BATNA Asymmetries and Agreement Efficiency
Dyads with knowledge of BATNA asymmetries reached less efficient agreements

than those without this knowledge (MUnequal & Knowledge = 15,650 vs. M = 16,800
MUnequal & No Knowledge ), t (46) = 2.24, p < .05, d = −0.66. This is consistent with
the findings from Study 1.
3.2.2 Focus on the Distributive Element
To assess the dimensionality of the scale, we conducted an exploratory factor analy-

sis. An examination of the overall fit measures indicated that the one-factor model
fitted the data, χ2 (5) = 14.5, p = ns. Furthermore, eigenvalues indicated a single
dominant factor with loadings ranging between 0.43 and 0.92. This scale showed an
acceptable level of internal reliability (Cronbach’s alpha = 0.86). Therefore, nego-
tiators’ responses to the five items were summed to produce a focus score for each
negotiator. Table 5 displays the mean focus scores in both experimental conditions.
As predicted by Hypothesis 3a, knowledge of BATNA asymmetries was positively
correlated with strong negotiators’ focus scores β = 6.125, p < .001. Strong nego-
123
β = 480.6***
Focus on the Judgement
Distributive Error
Element
β = 6.13*** β = -0.128**
β = -103.4* / β = -39.6ns
β = 3862.5* / β = 1253.9ns
Strong
Negotiators’
Knowledge Joint Gain
β = -1150.0* / β = -544.4ns
Fig. 2 Study 2: Mediation of knowledge effect by focus score for strong negotiators. * p < .05; ** p < .01;
*** p < .001. Note: the coefficient β is bold when taking the mediator(s) into account, and not bold when
looking at the relationship between two variables alone
tiators with knowledge focused more on the distributive element of the negotiation
(M = 23.3) than those without knowledge (M = 17.2). This finding supports Hypoth-
esis 3a. However, no effect was found for weak negotiators’ focus scores, β = −0.583,
ns. Thus, Hypothesis 3b was not supported by the findings.
3.2.3 Judgement Errors About Others’ Interests
Hypothesis 4 predicted that knowledge would be positively correlated with nego-

tiators’ judgement errors about counterparts’ preferences. The findings partially
supported this hypothesis, β = 3863, p < .05. Informed strong negotiators made
less accurate judgements about others’ preferences (M = 15,762) than uninformed
strong negotiators (M = 11,900) (higher values indicate greater judgement errors).
However, no significant effect was found for weak negotiators’ judgement errors,
β = 485, ns.
3.2.4 Mediation Analyses
We then examined the two-step mediation model to determine whether strong nego-
tiators’ focus on the distributive element mediates the relationship between strong
negotiators’ knowledge and judgement errors, and whether the judgement errors medi-
ate the relationship between strong negotiators’ focus on the distributive element and
agreement efficiency. The results of the analyses are shown in Fig. 2.
We demonstrated that strong negotiators’ knowledge predicted their focus on
the distributive element, β = 3.13, t (46) = 4.10, p < .001. Furthermore, their
focus on the distributive element was associated with higher judgement errors,
β = 425.9, t (45) = 2.84, p < .01. When controlling for the focus on the distribu-
tive element, the effect of knowledge on strong negotiators’ judgement errors was no
longer significant, β = 418.0, t (45) = 0.71, p = .48. To test the significance of the
two-step indirect effect (i.e., the path through focus on the distributive element and
judgement error in serial), we followed a bootstrapping procedure using the PROCESS
technique (Hayes et al. 2011; Preacher and Hayes 2008, 2009). The result of 10,000
123
resamples demonstrated that zero fell outside of the 95 % confidence interval (CI) of
the indirect effect of a focus on the distributive element (95 % CI [−860.8, −97.0]).
That is, the indirect effects of a focus on the distributive element and judgement errors
were likely to be somewhere between −860.8 and −97.0, with the obtained indirect
effect of −334.1 being a sensible point estimate. The indirect effects of a focus on the
distributive element alone and those of judgement errors were probably zero (zero fell
within the 95 % CIs). The results support the two-step mediator model as illustrated
in Fig. 1.
4 Study 3
Study 3 examines whether knowledge of BATNA asymmetries hinders negotiators’

information-sharing behaviour about preferences across issues. This in turn provides
us with a better understanding of why knowledge can hinder agreement efficiency.
Intuitively, knowledge of BATNA asymmetries increases strong negotiators’ perceived
power but decreases that of weak negotiators. However, this has not yet been tested
empirically and will be examined in Study 3.
Study 3 also examines whether knowledge shapes negotiators’ perceptions of
fairness. In BATNA-asymmetric negotiations, we argue that knowledge of others’
BATNAs affects how negotiators determine what agreement point is fair. With no
knowledge of BATNA asymmetries, an obvious way to determine fairness is an equal
division of resources. Negotiators’ perceptions of fairness may change once nego-
tiators learn of their power advantage or disadvantage. Specifically, informed strong
negotiators, compared with uninformed ones, may perceive higher power, as consis-
tent with their higher realised power and focus more on the distributive element as
shown in studies 1 and 2. The knowledge that they themselves hold a more powerful
position justifies their claim to a larger share of the resources to reflect that power. We
will also explore the possibility that perceptions of fairness mediate the relationship
between a focus on the distributive element and knowledge. That is, the belief held
by strong negotiators that their better BATNAs justify a larger surplus explains the
relationship between knowledge and their focus on the distributive element, which
was found in Study 2.
In contrast, knowledge of BATNA asymmetries is expected to lower the agreement
point that weak negotiators consider fair. This is because knowing that they rely on
the negotiation to a greater extent does not justify obtaining higher outcomes than
their stronger counterparts. An equally plausible alternative hypothesis is that when
they know that they are weaker, such negotiators tend to consider an agreement that
favours their position (e.g., equality) to be fair. To test our speculations, we manipulated
knowledge of BATNA asymmetries, as in Study 2.
Hypothesis 5 When both negotiators are aware of BATNA asymmetries, negotiators

are less likely to share information about preferences with their counterparts than when
they are not.
Hypothesis 6a Strong negotiators with knowledge of BATNA asymmetries perceive

themselves as more powerful than those without this knowledge.
123
Hypothesis 6b Weak negotiators with knowledge of BATNA asymmetries perceive

themselves as less powerful than those without this knowledge.
Hypothesis 7a When strong negotiators are informed of BATNA asymmetries, they

perceive that a fair agreement would give them a higher share of the bargaining surplus
than when they are not.
Hypothesis 7b When weak negotiators are informed of BATNA asymmetries, they

perceive that a fair agreement gives them a lower share of the bargaining surplus than
when they are not.
Hypothesis 7c When weak negotiators are informed of BATNA asymmetries, they

indicate that a fair agreement is an equal division of available resources.
4.1 Method
4.1.1 Participants
A total of 104 undergraduates students were recruited from Hang Seng Management
College and Lingnan University, comprising 56 male and 48 female, aged from 18 to 28
years (M = 21.69, SD = 2.30). Subjects voluntarily participated in the experiment.
4.1.2 Procedure
The procedure was similar to that of Study 2, except that additional pre-negotiation
questionnaires were used to elicit participants’ concerns about fairness and perceived
power.
4.1.3 Independent and Dependent Measures
The independent variable was knowledge of BATNA asymmetries, as in Study 2.
Negotiators’ Information-Sharing Behaviour The content of negotiators’ interactions

were transcribed and coded in terms of information-sharing behaviour about pref-
erences. One rater coded all the transcriptions, and a second rater (who was blind
to conditions and hypotheses) coded 58 % of the transcriptions.8 Negotiators were
assigned a score of 0 if they did not share information about preferences across issues
and a score of 1 if the negotiators shared information by asking opponents about their
preferences across issues or telling their opponents about their own preferences across
issues without prompting. This resulted in a dichotomous variable. A higher proportion
indicated that negotiators were more likely to share information about preferences.
8 In the case that disagreements occurred, the code assigned by the first rater was retained to be consistent
with the larger data set. The inter-rater reliability is reported in Sect. 4.2.1.
123
Perceived Power The measure of perceptions of relative power was collected in the
pre-negotiation questionnaire, as in Wolfe and Mcginn (2005). Participants were asked
to rate, on a scale from 0 to 100, “What is your bargaining power in this negotiation?”
On that scale, 0 was labelled “my counterpart has all the power”, 50 was labelled
“equal power” and 100 was labelled “I have all the power”.
Fairness Perception All participants were given a pre-negotiation questionnaire that

assessed their perceptions of what constituted a fair agreement; this was based on
the number of points that it would generate for themselves and their opponents. As
participants’ perceptions of the total bargaining surplus may vary, a percentage of a
fair bargaining surplus for each participant was computed.
4.2 Results
To test the degree of nonindependence in the data regarding negotiators’ information-

sharing behaviours, we followed the procedure used in Alferes and Kenny (2009) for
categorical measures. A Cohen’s κ of 0.601 was obtained, Z = 4.35, p < .0005.
Thus, strong negotiators’ and weak negotiators’ information-sharing behaviours were
moderately correlated.
4.2.1 Information Exchange
To assess the reliability of the coding of the first and second raters, Cohen’s κ was
performed and a reliability coefficient of 0.798 showed an acceptable level of inter-
rater agreement. As information-sharing behaviour was a dichotomous outcome where
strong negotiators’ information-sharing behaviour was somewhat correlated with the
weaker counterparts’, we performed a multilevel logistic regression model to analyse
data in a way that included both the dyad and negotiator as units of analysis (Spain et al.
2012). A generalised estimating equations approach (GEE) procedure was adopted
to estimate the effects of Knowledge and BATNA (strong or weak) and to account
for within-dyad correlations. For the GEE, we assumed the exchangeable compound
symmetry correlation structure and used robust standard errors. Furthermore, GEE
provided unbiased estimates of parameters using the Fisher method. Both Knowledge
and BATNA were treated as fixed parameters (Spain et al. 2012). We fitted a Level 1
model with a random intercept. The Level 2 model included Knowledge, BATNA and
Knowledge × BATNA interaction effects. There was an interaction effect of knowl-
edge by negotiators’ BATNA (strong or weak), on information-sharing behaviour (see
Table 6). Table 7 displays the regression coefficients of the logistic regression model.
The findings lend partial support to Hypothesis 5. For strong negotiators, the likeli-
hood of sharing information about preferences in the Unequal & Knowledge condition
were 0.271 times that in Unequal & No Knowledge, β = −1.307, Wald χ 2 = 4.702,
p < .05. However, no significant effect was found for weak negotiators’ information-
sharing behaviour, β = −0.199, p = ns.
123
Table 6 Logistic regression model summary
Source Wald χ 2 df p
Knowledge 2.120 1 0.145

BATNA (strong or weak negotiators) 2.120 1 0.343
Knowledge × BATNA 4.580 1 0.032
Quasi likelihood under independence model criterion
Goodness of fit 143.164
Table 7 Effects of knowledge of BATNA-asymmetries on information-sharing
Factor Information-sharing behaviour
Knowledge of BATNA-asymmetries
Strong negotiator −1.307†
Weak negotiator 0.199
N = 104. All terms are unstandardised regression coefficients. For the analyses predicting to negotia-
tors’ information-sharing behaviour, the effects of Knowledge were obtained through multilevel logistic
regression model
† p < .05 (all two-tailed tests)
4.2.2 Perceived Power
ANOVAs showed significant differences between the Unequal & No Knowledge condi-
tion and the Unequal & Knowledge condition for strong negotiators’ perceived power,
F(1, 50) = 29.71, p < .0005, η2p = .37, and weak negotiators’ perceived power,
F(1, 50) = 14.03, p < .0005, η2p = .22. With knowledge of BATNA asymmetries,
strong negotiators perceived their position to be more powerful (MUnequal & Knowledge =
76.1) than those without knowledge (MUnequal & No Knowledge = 58.1), t (50) = 5.40,
p < .0005, d = 1.53. In contrast, informed weak negotiators perceived their position
to be less powerful than those without knowledge (MUnequal & Knowledge = 41.6 vs.
MUnequal & No Knowledge = 54.0), t (50) = −3.75, p < .0005, d = −1.06. In support
of Hypotheses 6a and 6b, knowledge affected whether strong and weak negotiators
perceived their positions to be more or less powerful than their counterparts.
4.2.3 Fairness Perception
Knowledge of BATNA asymmetries affected strong negotiators’ perceptions of fair

settlements, F(1, 50) = 22.84, p < .0005, η2p = .31, but not the perceptions of weak
negotiators, F(1, 50) = 1.02, p = .317. Strong negotiators with knowledge thought
that a higher settlement percentage was fair (MUnequal & Knowledge = 62.9 %) compared
with those without knowledge (MUnequal & No Knowledge = 53.0 %), t (50) = 4.78,
p < .0005, d = 0.46. Thus, Hypothesis 7a was supported.
Hypothesis 7b predicted that knowledge of BATNA asymmetries would lower the
agreement point that weak negotiators considered fair. The findings did not sup-
123
Perceived
Fairness
β = 0.36*** / β = 0.28**
β = 9.89***
Strong
Negotiators’ Focus on the
Knowledge Distributive
Element
β = 5.27*** / β = 2.50ns
Fig. 3 Study 3: Mediation of strong negotiators’ knowledge effect by perceived fairness. ** p < .01;
*** p < .001. Note: the coefficient β is bold when taking a third variable into account, and not bold when
looking at the relationship between two variables alone
port this hypothesis. When determining fairness, weak negotiators with knowledge
did not appear to produce a lower settlement point than those without knowledge
(MUnequal & Knowledge = 50.2 % compared with MUnequal & No Knowledge = 52.0 %),
t (50) = −1.01, p = .16). We also conducted two one-sample t tests against a seed
value of 50, examining weak negotiators’ perceptions of fairness in both Knowledge
and No Knowledge conditions. The findings revealed that informed weak negotiators
generally thought that a 50–50 division of the surplus was fair (MUnequal & Knowledge =
50.2 %), t (25) = .19, p = .85; similar results were obtained for uninformed weak
negotiators (MUnequal & No Knowledge = 52.0 %), t (25) = 1.65, p = .11. Thus, these
findings lend support to Hypothesis 7c. The results suggest that even when weak nego-
tiators knew they were weaker, they did not feel entitled to a smaller proportion of the
surplus, but instead constituted fairness via equality.
Figure 3 displays the relationship among knowledge, strong negotiators’ perceived
fairness and a focus on the distributive element. Note that the focus score was com-
puted in the same way as in Study 2. When regressing the focus score on both strong
negotiators’ knowledge and perceived fairness, we found that the effect of knowledge
on the focus score was insignificant, β = 2.50, t (49) = 1.26, p = .21, whereas the
effect of perceived fairness remained significant, β = 0.28, t (49) = −2.88, p < .01.
We then conducted a bootstrapping procedure (Preacher and Hayes 2008, 2009), and
the result of 10,000 resamples demonstrated that zero fell outside the 95 % CI of the
indirect effect of perceived fairness (95 % CI [0.10, 4.70]). Thus, strong negotiators’
perceived fairness mediated the relationship between focus score and knowledge.
However, the regression analysis showed that weak negotiators’ knowledge had no
effect on both perceived fairness, β = −2.33, p = .29 and the focus score, β = −2.25,
p = .13. The bootstrapping result for the indirect effect of perceived fairness with
10,000 resamples showed that zero fell within the 95 % CI (95 % CI [−1.79, 1.52])
5 General Discussion and Conclusion
Although the negotiation literature highlights the importance of accurately estimating

others’ BATNAs, it is not clear how negotiators use this knowledge in power-
asymmetric negotiations. In the present research, we have advanced an alternative
123
view of knowledge of one’s power advantage or disadvantage in variable-sum negoti-

ations. Our three experimental studies have illustrated that such knowledge shapes the
structure of negotiated outcomes. It also has power implications (e.g., perceived power
and realised power), and affects perceptions of fairness and information exchange
about preferences in BATNA-asymmetric negotiations. We argued that variability
in the relationship between power (a)symmetries and agreement efficiency lies in
the level of the negotiators’ knowledge. It was found that dyads with asymmetric
BATNAs and knowledge of counterparts’ power reached less efficient agreements
than dyads with asymmetric BATNAs and no knowledge and equal-BATNA dyads
with or without knowledge. Knowledge of one’s counterparts’ power does matter in
power-asymmetric negotiations but it does not affect efficiency in power-symmetric
negotiations. The same pattern of results was replicated in Study 2.
One of the contributions of our study is that it refines and extends the findings
of existing research on power-asymmetric negotiations. It is not clear from previous
studies whether negotiation dyads with unequal power are more or less able to reach
efficient outcomes (Arunachalam et al. 1998; Komorita and Leung 1985; Pinkley et al.
1994; Roloff and Dailey 1987; Wei and Luo 2012; Wolfe and Mcginn 2005). We are left
to wonder whether power differences between negotiators even matter. The findings
from Study 1 provide insight into the inconsistencies of previous findings regarding
power (a)symmetries and agreement efficiency. Our results have shown that power
asymmetries alone do not affect agreement efficiency.
An important question addressed in Study 2 was why knowledge of BATNA asym-
metries hinders the development of efficient agreements, as suggested in Study 1.
The explanation involved a two-step mechanism. The first step was that informed
strong negotiators would express their superiority distributively, to push for agree-
ments that reflect their BATNA advantage. Furthermore, we know that negotiators
commonly have fixed-pie bias at the outset of a negotiation (Thompson and Hastie
1990; Thompson 1991). The second step was that strong negotiators’ focus on the dis-
tributive element would render this bias more difficult to dislodge, resulting in greater
judgement errors about their counterparts’ preferences. Study 3 also demonstrated
that strong negotiators with knowledge of BATNA asymmetries would be less likely
to exchange information about preferences across issues. It is possible that informed
strong negotiators were less likely to seek information that would contradict their
existing suppositions than uninformed strong negotiators. These findings are consis-
tent with previous studies suggesting that negotiators tend to overestimate information
that is consistent with their expectations (i.e., fixed-pie perception) and underestimate
information that goes against them (Pinkley et al. 1995; Pruitt and Carnevale 1993;
Thompson and DeHarpport 1994; Thompson and Hastie 1990; Thompson 1991).
Additionally, the findings from Study 3 illustrated that knowledge did not affect the
likelihood that weak negotiators would share information about preferences. How
would informed strong negotiators react when their weaker opponents ask them ques-
tions about preferences? It is plausible that informed strong negotiators interpret their
opponents’ actions as a competitive tactic and resist revealing their preferences across
issues.
The positive relationship between knowledge and strong negotiators’ focus on the
distributive element also explains past findings that show strong negotiators, when
123
informed of both BATNAs, tend to outperform their weaker counterparts (Arunacha-

lam et al. 1998; Komorita and Leung 1985; Mannix and Neale 1993; Wolfe and Mcginn
2005). Furthermore, mediation analyses suggested that the relationship between strong
negotiators’ knowledge and efficiency was partly due to their focus on the distributive
element. Strong negotiators’ focus also mediated the effect of knowledge on their
inaccurate judgements.
Conversely, studies 1, 2 and 3 illustrated that knowledge of BATNA asymmetries
did not affect weak negotiators’ focus on the distributive element and judgement errors,
but did lower both their perceived and realised power. Weak negotiators’ perceptions
of fairness may help shed light on why knowledge did not reduce their focus on the
distributive element. Specifically, when weak negotiators knew their power disadvan-
tage, they still indicated that a fair agreement involved the equal division of resources.
This may at first glance contradict the observed negative association between knowl-
edge and their realised power. It is important to note that in multiple-issue negotiations
negotiators cannot easily assess the percentage of resources claimed by themselves and
others. Therefore, informed weak negotiators’ pursuit of fairness may not be reflected
in their outcomes.
Informed strong and weak negotiators appeared to adopt different views of fair-
ness. Informed strong negotiators perceived that a fair settlement should reflect
their BATNA advantage by generating more surplus than their weaker counterparts.
Furthermore, informed strong negotiators’ fairness perceptions explain the positive
relationship between knowledge and their competitiveness. Strong and weak negotia-
tors’ divergent perceptions of fairness may lead to greater hostility or conflict during
negotiations.
An important implication of our findings is that knowing more about one’s coun-
terparts does not necessarily help the performance of negotiators. Wiltermuth and
Neale (2011) share a similar view on the effect of possessing non-diagnostic informa-
tion about opponents on negotiators’ effectiveness and performance. The additional
information considered in our studies is diagnostic because knowledge of BATNA
asymmetries is predictive of negotiated outcomes (at least in terms of how resources
are divided between a relatively powerful negotiator and a weaker counterpart).
Similar to other theorists, the findings from the our research suggest that knowl-
edge could be a source of weakness or strength (Brodt 1994; Schelling 1960). As in
Kim et al. (2005), we also found that informed strong negotiators claimed a larger
slice of the surplus than uninformed strong negotiators. Because variable-sum negoti-
ation is prevalent, the cost of neglecting opportunities for value creation should not be
underestimated. How large or little this cost is depends on the degree to which nego-
tiators have different preferences across issues in negotiations. In negotiations with
an integrative potential, knowing one’s own power advantage could lead to exercising
this power, resulting in receiving lower payoffs than successfully logrolling issues
between negotiators. As a result, we caution negotiators against being too attached to
assumptions about their weaker counterparts and neglecting opportunities for value
creation.
On a practical note, well-prepared negotiators are advised to evaluate each others’
BATNAs before negotiation begins. In some situations, this kind of information is
more accessible today. We can see that studies on power-asymmetric negotiations
123
have considered situations where negotiators have knowledge of their counterparts’

power (Belkin et al. 2013; Giebels et al. 2000; Kim and Fragale 2005; Malhotra and
Gino 2011; Nelson et al. 2015; Schaerer et al. 2015). For instance, a manufacturer of
mobile device displays probably knows that Apple has a stronger BATNA; mortgage
providers are fully aware of alternative deals available to house buyers (e.g., interest
rates); and in a job negotiation, packages offered for similar positions (as stated in
job advertisements) and contacts in the industry or professional network may provide
some information about a counterpart’s BATNAs.
A limitation of the current research is that two of the experimental conditions were
added to Study 1 at a later stage and therefore full randomisation was not achieved.
This could somewhat compromise the internal validity. We reduced its impacts by
drawing samples from populations with similar attributes (in Study 1, both UK and
Hong Kong participants studied at the same universities). Another limitation is that
we are uncertain whether dyads may reach a more efficient agreement when only one
member has access to the knowledge of power asymmetries. In the case of asymmetric
knowledge, when only weak negotiators are informed of both BATNAs, we do not
expect strong negotiators to focus on the distributive element as much as informed
strong negotiators. The negotiation process and negotiators’ behaviour may be largely
different. It is also worth examining whether the same effects may be found using
other operationalisations of negotiators’ power. Our research shows that knowledge
increases strong negotiators’ propensity to focus on capturing more value than their
counterparts. Although knowledge was not found to have an effect on the focus scores
of weak negotiators, knowledge may alter the approaches of weak negotiators in other
ways. For instance, knowledge may place weak negotiators under greater pressure to
claim as much of the value as they can once they learn that their counterparts’ have
stronger BATNAs.
We have revealed several other avenues to be developed in future studies. For
instance, it would be worthwhile to investigate whether knowledge of BATNA asym-
metries leads to impasses in negotiations where the bargaining zone is small. We
adopted a negotiation task with a large bargaining zone in which negotiators were
likely to reach an agreement. As we now know that informed strong and weak nego-
tiators have different perceptions of fairness, it is possible that this knowledge increases
the chance of disagreement when a bargaining zone is small. More research is also
necessary to investigate whether this knowledge reduces negotiators’ willingness to
interact with the same party again in future.
It seems that a greater emphasis should be placed on negotiators’ knowledge about
the power status of others in power-asymmetric negotiations when evaluating the
impact of power asymmetries on the quality of outcomes. In this paper, we have
argued that when a negotiator knows that she is powerful, such knowledge will change
her negotiation tactics in such a way that ultimately hurts her own interests as well
as that of her opponent. To create a fruitful negotiation outcome, strong negotia-
tors may need to shift focus away from exercising their power and toward opening
possibilities.
Acknowledgements This project is partially supported by Hang Seng Management College Research
Fund.
123
Appendix
Pay-off schedules for job negotiation task

Salary Annual leave Bonus Starting date Medical Company car
coverage
Employer pay-off schedule

£24,000 25 days 10 % 1st July Plan A BMW 330i
(0) (0) (0) (1200) (3200) (0)
£23,000 20 days 8% 15th July Plan B VW Golf
(500) (1000) (400) (900) (2400) (200)
£22,000 15 days 6% 1st Aug Plan C Honda
(1000) (2000) (800) (600) (1600) (400)
£21,000 10 days 4% 15th Aug Plan D Ford Focus
(1500) (3000) (1200) (300) (800) (600)
£20,000 5 days 2% 1st Sept Plan E No company car
(2000) (4000) (1600) (0) (0) (800)
Employee pay-off schedule
£24,000 25 days 10 % 1st July Plan A BMW 330i
(2000) (1600) (4000) (1200) (0) (3200)
£23,000 20 days 8% 15th July Plan B VW Golf
(1500) (1200) (3000) (900) (200) (2400)
£22,000 15 days 6 1st Aug Plan C Honda
(1000) (800) (2000) (600) (400) (1600)
£21,000 10 days 4% 15th Aug Plan D Ford Focus
(500) (400) (1000) (300) (600) (800)
£20,000 5 days 2% 1st Sept Plan E No company car
(0) (0) (0) (0) (800) (0)
The possible number of points that negotiators could receive are shown in parentheses
References
Alferes VR, Kenny DA (2009) SPSS programs for the measurement of nonindependence in standard dyadic
designs. Behav Res Methods 41(1):47–54
Anderson C, Thompson LL (2004) Affect from the top down: how powerful individuals’ positive affect
shapes negotiations. Organ Behav Hum Decis Process 95:125–139. doi:10.1016/j.obhdp.2004.1005.
1002
Arunachalam V, Dilla W, Shelley M, Chan C (1998) Market alternatives, third party intervention, and third
party informedness in negotiation. Group Decis Negot 7:81–107
Babcock L, Loewenstein G, Issacharoff S, Camerer C (1995) Biased judgments of fairness in bargaining.
Am Econ Rev 85(5):1337–1343
Bacharach SB, Lawler EJ (1981) Bargaining: power, tactics, and outcomes. Jossey-Bass, San Francisco
Bazerman MH, Neale MA (1983) Heuristics in negotiation: limitations to effective dispute resolution.
Negotiating in organizations. Sage, Beverly Hills
Bazerman MH, Curhan JR, Moore DA, Valley KL (2002) Negotiation. Annu Rev Psychol 51:279–314
Belkin LY, Kurtzberg TR, Naquin CC (2013) Signalling dominance in online negotiations: the role of
affective tone. Negot Confl Manag Res 6(4):285–304
Brett JF, Pinkley RL, Jackofsky EF (1996) Alternatives to having a BATNA in dyadic negotiation: the
influence of goals, self-efficacy, and alternatives on negotiated outcomes. Int J Confl Manag 7(2):121–
138. doi:10.1108/eb022778
Brodt SE (1994) “Inside information” and negotiator decision behavior. Organ Behav Hum Decis Process
58:172–202. doi:10.1006/obhd.1994.1033
Buelens M, Van Poucke D (2004) Determinants of a negotiator’s initial opening offer. J Bus Psychol
19(1):23–35
123
Coleman PT, Kugler K, Mitchinson A, Chung C, Musallam N (2010) The view from above and below: the
effects of power and interdependence asymmetries on conflict dynamics and outcomes in organizations.
Negot Confl Manag Res 3(4):283–311
Emerson RM (1962) Power-dependence relations. Am Sociol Rev 27:31–40
Epley N, Caruso E, Bazerman MH (2006) When perspective taking increases taking: reactive egoism in
social interaction. J Pers Soc Psychol 91(5):872–889
Fisher R, Ury W (1981) Getting to yes: negotiating agreement without giving in. Houghton-Mifflin, Boston
French J, Raven B (1959) The bases of social power. In: Cartwright D (ed) Studies in social power. Institute
for Social Research, Ann Arbor
Galinsky AD, Magee JC, Inesi ME, Gruenfeld DH (2006) Power and perspectives not taken. Psychol Sci
17(12):1068–1074
Gelfand MJ, Higgins M, Nishii LH, Raver JL, Dominguez A, Murakami F, Yamaguchi S (2002) Culture
and egocentric perceptions of fairness in conflict and negotiation. J Appl Psychol 87(5):833–845
Giebels E, De Dreu CKW, Van De Vliert E (2000) Interdependence in negotiation: effects
of exit options and social motive on distributive and integrative negotiation. Eur J
Soc Psychol 30:255–272. doi:10.1002/(SICI)1099-0992(200003/200004)200030:200002<200255:
AID-EJSP200991>200003.200000.CO;200002-200007
Gino F, Moore DA (2008) Why negotiators should reveal their deadlines: disclosing weaknesses can make
you stronger. Negot Confl Manag Res 1(1):77–96
Gonzaga GC, Keltner D, Ward D (2008) Power in mixed-sex stranger interactions. Cogn Emot 22(8):1555–
1568
Greer LL, Van Kleef GA (2010) Equality versus differentiation: the effects of power dispersion on group
interaction. J Appl Psychol 95(6):1032–1044
Gruenfeld DH, Inesi ME, Magee JC, Galinsky AD (2008) Power and the objectification of social targets. J
Pers Soc Psychol 95:111–127
Handgraaf MJJ, van Dijk E, Riel CV, Henk AMW, De Dreu CKW (2008) Less power or powerless?
Egocentric empathy gaps and the irony of having little versus no power in social decision making. J
Pers Soc Psychol 95(5):1136–1149
Hayes AF, Preachers KJ, Myers TA (2011) Mediation and the estimation of indirect effects in political com-
munication research. In: Bucy EP, Lance HR (eds) Sourcebook for political communication research:
methods, measures, and analytical techniques. Routledge, New Work, pp 434–465
Inesi ME, Gruenfeld DH, Galinsky AD (2012) How power corrupts relationships: cynical attributions for
others’ generous acts. J Exp Psychol 48:795–803
Keltner D, Gruenfeld DH, Anderson C (2003) Power, approach, and inhibition. Psychol Rev 110:265–284
Kim PH, Fragale AR (2005) Choosing the path to bargaining power: an empirical comparison of BATNAs
and contributions in negotiation. J Appl Psychol 90(2):373–381
Kim PH, Pinkley RL, Fragale AR (2005) Power dynamics in negotiations. Acad Manag Rev 30(4):799–822.
doi:10.5465/AMR.2005.18378879
Komorita SS (1984) The role of justice and power in reward allocation. Prog Appl Sociol 2:185–206
Komorita SS, Hamilton TP (1984) Power and equity in coalition bargaining, vol. 3. Research in the sociology
of organizations. JAI Press, Connecticut
Komorita SS, Leung K (1985) The effects of alternatives on the salience of reward allocation Norms. J Exp
Soc Psychol 21:229–246. doi:10.1016/0022-1031(1085)90018-90016
Koning L, Steinel W, van Beest I, van Dijk E (2011) Power and deception in ultimatum bargaining. Organ
Behav Hum Decis Process 115(1):35–42
Lawler EJ, Yoon J (1993) Power and the emergence of commitment behavior in negotiated exchange. Am
Sociol Rev 58:465–481. doi:10.2307/2096071
Lax DA, Sebenius JK (1986) The manager as negotiator. The Free Press, New York
Maaravi Y, Pazy A, Ganzach Y (2011) Pay as much as you can afford: counterparts’ ability to pay and first
offers in negotiation. Judgm Decis Mak 6(4):275–282
Magee JC, Galinsky AD, Gruenfeld DH (2007) Power, propensity to negotiate, and moving first in com-
petitive interactions. Pers Soc Psychol Bull 33(2):200–212
Malhotra D, Gino F (2011) The pursuit of power corrupts: how investing in outside options motivates
opportunism in relationships. Adm Sci Q 56(4):559–592
Mannix EA (1993) Organizations as resource dilemmas: the effects of power balance on coalition formation
in small groups. Organ Behav Hum Decis Process 55:1–22
123
Mannix EA, Neale MA (1993) Power imbalance and the pattern of exchange in dyadic negotiation. Group
Decis Negot 2:119–133. doi:10.1007/BF01884767
McAlister L, Bazerman M, Fader P (1986) Power and goal setting in channel negotiations. J Mark Res
23:228–236. doi:10.2307/3151481
Murnighan JK, Babcock L, Thompson L, Pillutla M (1999) The information dilemma in negotiations:
effects of experience, incentives, and integrative potential. Int J Confl Manag 10(4):313–339
Neale MA, Bazerman M (1991) Cognition and rationality in negotiation. Free Press, New York
Nelson B, Bronstein I, Shacham R, Ben-Ari R (2015) The power to oblige: power, gender, negotiation
behaviors, and their consequences. Negot Confl Manag Res 8(1):1–24
Olekalns M, Smith PL (2009) Mutually dependent: power, trust, affect and the use of deception in negoti-
ation. J Bus Ethics 85:347–365
Olekalns M, Smith PL (2013) Dyadic power profiles: Power-contingent strategies for value creation in
negotiation. Hum Commun Res 39:3–20
Paese PW, Yonker RD (2001) Toward a better understanding of egocentric fairness judgments in negotiation.
Int J Confl Manag 12(2):114–131
Pietroni D, Van Kleef GA, De Dreu CKW, Pagliaro S (2008) Emotions as strategic information: effects of
other’s emotional expressions on fixed-pie perception, demands, and integrative behavior in negotia-
tion. J Exp Soc Psychol 44(6):1444–1454. doi:10.1016/j.jesp.2008.1406.1007
Pinkley RL (1995) Impact of knowledge regarding alternatives to settlement in dyadic negotiations: whose
knowledge counts? J Appl Psychol 80(3):403–417. doi:10.1037//0021-9010.1080.1033.1403
Pinkley RL, Neale MA, Bennett RJ (1994) The impact of alternatives to settlement in dyadic negotiation.
Organ Behav Hum Decis Process 57:97–116. doi:10.1006/obhd.1994.1006
Pinkley RL, Griffith TL, Northcraft GB (1995) Fixed pie a la mode—information-availability, information-
processing, and the negotiation of suboptimal agreements. Organ Behav Hum Decis Process
62(1):101–112. doi:10.1006/obhd.1995.1035
Preacher KJ, Hayes AF (2008) Asymptotic and resampling strategies for assessing and comparing indirect
effects in multiple mediator models. Behav Res Methods 40:879–891
Preacher KJ, Hayes AF (2009) SPSS and SAS macro for bootstrapping specific indirect effects in multiple
mediation models. http://www.comm.ohio-state.edu/ahayes/SPSS%20programs/indirect.htm
Pruitt DG, Carnevale PJ (1993) Negotiation in social conflict. Open University Press, Buckingham
Raudenbush SW, Bryk AS (2002) Hierarchical linear models: applications and data analysis methods. Sage,
Thousand Oaks
Roloff ME, Dailey WO (1987) The effects of alternatives to reaching agreement on the development of
integrative solutions to problems: the debilitating side effects of shared BATNA. Paper presented at the
Temple University discourse conference, Conflict interventions perspectives on process, Philadelphia,
PA
Roth A, Murnighan K (1982) The role of information in bargaining: an experimental study. Econometrica
50:1123–1142
Schaerer M, Swaab RI, Galinsky AD (2015) Anchors weigh more than power: why absolute powerlessness
liberates negotiators to achieve better outcomes. Psychol Sci 26(2):170–181
Schelling TC (1960) The strategy of conflict. Harvard University Press, Boston
Shaw ME (1981) Group dynamics: the psychology of small group behaviour. McGraw-Hill, New York
Spain SM, Jackson JJ, Edmonds GW (2012) Extending the actor–partner interdependence model for binary
outcomes: a multilevel logistic approach. Pers Relatsh 19(3):431–444
Thompson LL (1990a) An examination of naive and experienced negotiators. J Per Soc Psychol 59:82–90.
doi:10.1037//0022-3514.1059.1031.1082
Thompson LL (1990b) The Influence of experience on negotiation performance. J Exp Soc Psychol 26:528–
544. doi:10.1016/0022-1031(1090)90054-P
Thompson LL (1990c) Negotiation behavior and outcomes: empirical evidence and theoretical issues.
Psychol Bull 108(3):515–532
Thompson LL (1991) Information exchange in negotiation. J Exp Soc Psychol 27:161–179. doi:10.1016/
0022-1031(1091)90020-90027
Thompson LL (2012) The mind and heart of the negotiator. International edition. Pearson, Upper Saddle
River
Thompson LL, Hastie R (1990) Social perception in negotiation. Organ Behav Hum Decis Process 47:98–
123. doi:10.1016/0749-5978(1090)90048-E
123
Thompson LL, Loewenstein G (1992) Egocentric interpretations of fairness and interpersonal conflict.
Organ Behav Hum Decis Process 51:176–197
Thompson LL, DeHarpport T (1994) Social judgement, feedback, and interpersonal learning in negotiation.
Thompson LL, Wang J, Gunia BC (2010) Negotiation. Annu Rev Psychol 61:491–515. doi:10.1146/annurev.
psych.093008.100458
Tost LP, Gino F, Larrick RP (2012) Power, competitiveness, and advice taking: why the powerful don’t
listen. Organ Behav Hum Decis Process 117:53–65
Turel O (2010) Interdependence issues in analyzing negotiation data. Group Decis Negot 19:111–125.
doi:10.1007/s10726-008-9118-x
Van Kleef GA, De Dreu CKW, Pietroni D, Manstead ASR (2006) Power and emotion in negotiation: power
moderates the interpersonal effects of anger and happiness on concession making. Eur J Soc Psychol
36:557–581. doi:10.1002/ejsp.1320
Wade-Benzoni KA, Tenbrunsel AE, Bazerman MH (1996) Egocentric interpretations of fairness in asym-
metric environmental social dilemmas: explaining harvesting behavior and the role of communication.
Walton RE, McKersie RB (1965) A behavioral theory of labor negotiations: an analysis of a social interaction
system. Cornell University Press
Wei Q, Luo X (2012) The impact of power differential and social motivation on negotiation behavior and
outcome. Public Pers Manag 41(5):47–58
Wiltermuth SS, Neale MA (2011) Too much information: the perils of nondiagnostic information in nego-
tiations. J Appl Psychol 96(1):192–201
Wolfe RJ, Mcginn KL (2005) Perceived relative power and its influence on negotiations. Group Decis Negot
14:3–20. doi:10.1007/s10726-10005-13873-10728
Wong RS (2014) Same power but different goals: How does knowledge of opponents’ power affect nego-
tiators’ aspiration in power-asymmetric negotiations? Glob J Bus Res 8(3):77–90
Young M, Bauman CW, Chen N, Bastardi A (2012) The pursuit of missing information in negotiation.
123

Blidend by Power - Article To Be Reviewed

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Blidend by Power - Article To Be Reviewed

Uploaded by

Copyright:

Available Formats

Group Decis Negot (2017) 26:215–245

Blinded by Power: Untangling Mixed Results

Ricky S. Wong1 · Susan Howard2

Published online: 20 July 2016

Keywords BATNA · Agreement efficiency · Power asymmetry · Knowledge ·

1.1 Theoretical Considerations: Power in Negotiation

1.2 Why Knowledge of Counterparts’ Power Matters

Scholars emphasise the importance of estimating counterparts’ BATNAs as well as

1.3 Inconsistency in Power Differences and Agreement Efficiency

1.4 Knowledge, Negotiation Approach and Judgement in Power-Asymmetric

1.5 Knowledge and Information-Sharing Behaviour

Knowledge of power asymmetries may also affect communications between negotia-

As argued above, knowledge of BATNA asymmetries increases strong negotiators’

1.6 Knowledge and Perceptions of Fairness in Power-Asymmetric Negotiation

Study 1 is designed to clarify the relationship among BATNA (a)symmetries, knowl-

Hypothesis 1 Unequal-BATNA dyads with knowledge of their counterparts’ BAT-

We also examine whether knowledge of BATNA asymmetries affects strong and

Hypothesis 2a When strong negotiators are informed of BATNA asymmetries, they

Hypothesis 2b When weak negotiators are informed of BATNA asymmetries, they

2.1.3 Negotiation Task

2.1.4 Experimental Manipulations and Measures

To fully test the hypotheses developed, we adopted a 3 × 2 [BATNA (A)symmetries:

Table 1 Effects of knowledge of opponents’ BATNAs and BATNA-differences on joint outcomes

Knowledge 3,641,736 1 3,641,736 1.409 0.237

Table 2 Means (standard deviations) of joint outcomes in Study 1

Joint 16,800a 16,541a 17,017a 16,975a 16,667a 15,763b

2.2 Results and Discussion

2.2.1 What Affects Agreement Efficiency (BATNA Differences vs. Knowledge)

2.2.2 Knowledge and Realised Power

We conducted an independent t test to examine whether knowledge of BATNA

2.2.3 Effects of Sample Source

Table 3 Items for negotiators’ focus on distributive element

1. I was very concerned if I could outperform the employee (employer)

A total of 96 undergraduate and master students at the London School of Economics

3.1.3 Negotiation Task and Independent Variables

3.1.4 Dependent Measures

Table 4 Effects of knowledge of BATNA-asymmetries on focus score and judgement errors

Factor Dependent measures at an individual level

Table 5 Means (standard

3.2.1 Knowledge of BATNA Asymmetries and Agreement Efficiency

Dyads with knowledge of BATNA asymmetries reached less efficient agreements

3.2.2 Focus on the Distributive Element

To assess the dimensionality of the scale, we conducted an exploratory factor analy-

3.2.3 Judgement Errors About Others’ Interests

Hypothesis 4 predicted that knowledge would be positively correlated with nego-

3.2.4 Mediation Analyses

Study 3 examines whether knowledge of BATNA asymmetries hinders negotiators’

Hypothesis 5 When both negotiators are aware of BATNA asymmetries, negotiators

Hypothesis 6a Strong negotiators with knowledge of BATNA asymmetries perceive

Hypothesis 6b Weak negotiators with knowledge of BATNA asymmetries perceive

Hypothesis 7a When strong negotiators are informed of BATNA asymmetries, they

Hypothesis 7b When weak negotiators are informed of BATNA asymmetries, they

Hypothesis 7c When weak negotiators are informed of BATNA asymmetries, they

4.1.3 Independent and Dependent Measures

The independent variable was knowledge of BATNA asymmetries, as in Study 2.

Negotiators’ Information-Sharing Behaviour The content of negotiators’ interactions

Fairness Perception All participants were given a pre-negotiation questionnaire that

To test the degree of nonindependence in the data regarding negotiators’ information-

4.2.1 Information Exchange

Table 6 Logistic regression model summary

Knowledge 2.120 1 0.145

Table 7 Effects of knowledge of BATNA-asymmetries on information-sharing

Factor Information-sharing behaviour