Description
Prior studies on customer–supplier negotiations (Drake & Haka, 2008; Van den Abbeele,
Roodhooft, & Warlop, 2009) find that negotiators who have access to relevant activitybased
cost information are not always able to use this information to improve joint outcomes.
Our study extends this literature by examining how the type of accountability (process
and outcome accountability) influences the extent to which negotiators can obtain
lower joint costs. We hypothesize and test a model that predicts that the type of accountability
affects negotiated outcomes through its effect on negotiators’ fixed-pie bias revisions
and the negotiation tactics they employ during customer-supplier negotiations.
Results from an experiment show that negotiators held accountable for their negotiation
processes are better able to reduce their fixed-pie biases and achieve lower joint costs compared
to those who are held accountable for their negotiation outcomes.
The effect of outcome and process accountability
on customer–supplier negotiations
Linda J. Chang, Mandy M. Cheng
?
, Ken T. Trotman
School of Accounting, The University of New South Wales, Sydney, NSW 2052, Australia
a b s t r a c t
Prior studies on customer–supplier negotiations (Drake & Haka, 2008; Van den Abbeele,
Roodhooft, & Warlop, 2009) ?nd that negotiators who have access to relevant activity-
based cost information are not always able to use this information to improve joint out-
comes. Our study extends this literature by examining how the type of accountability (pro-
cess and outcome accountability) in?uences the extent to which negotiators can obtain
lower joint costs. We hypothesize and test a model that predicts that the type of account-
ability affects negotiated outcomes through its effect on negotiators’ ?xed-pie bias revi-
sions and the negotiation tactics they employ during customer-supplier negotiations.
Results from an experiment show that negotiators held accountable for their negotiation
processes are better able to reduce their ?xed-pie biases and achieve lower joint costs com-
pared to those who are held accountable for their negotiation outcomes. Using rich data
based on taped negotiations, we demonstrate that the effect of accountability on joint costs
is indirect through its effect on negotiators’ choice of negotiation tactics and the extent to
which negotiators can reduce their ?xed-pie biases.
Ó 2013 Elsevier Ltd. All rights reserved.
Introduction
Negotiations between customers and suppliers form a
crucial part of collaborative supply chain relationships
(Anderson & Dekker, 2005; Drake & Haka, 2008). Activ-
ity-based costing (ABC) techniques, such as customer prof-
itability analysis, enable negotiators to identify activities
that drive supply chain costs, and direct their attention
away from price towards quantifying and managing the to-
tal cost of the customer–supplier relationship (Anderson &
Dekker, 2009a, 2009b; Van den Abbeele, Roodhooft, &
Warlop, 2009). Recent research has established that nego-
tiators who share ABC information are more likely to reach
agreements with higher joint outcomes (e.g., lower supply
chain costs) than those who share volume-based cost
information (Drake & Haka, 2008; Van den Abbeele et al.,
2009). However, these studies also show that negotiators
are reluctant to share ABC information, unless they are
confronted with adverse conditions such as market pres-
sures (Drake & Haka, 2008) or more powerful negotiation
partners (Van den Abbeele et al., 2009). Providing ABC
information to negotiators with bargaining power also
potentially increases their tendency to use confrontational
and competitive negotiation tactics, further reducing joint
outcomes (Van den Abbeele et al., 2009).
The organizational and psychology literatures argue
that a key factor that contributes to negotiators’ unwilling-
ness to work collaboratively is the ‘?xed-pie bias’. The
?xed-pie bias refers to negotiators’ tendencies to believe
that all negotiations are a ?xed pie; that is, the attainment
of one party’s goals precludes the other party from achiev-
ing their goals (Bazerman, 2006; De Dreu, Koole, & Steinel,
2000; Pinkley, Grif?th, & Northcraft, 1995; Thompson &
Hastie, 1990). Such a belief results in negotiators holding
incorrect assumptions about their negotiation partner’s
priorities and preferences. Fixed-pie bias is potentially det-
rimental because it causes negotiators to focus on compet-
ing with each other rather than working collaboratively,
0361-3682/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.aos.2012.12.002
?
Corresponding author. Tel.: +61 2 9385 6343; fax: +61 2 9385 5925.
E-mail addresses: [email protected] (L.J. Chang), m.chen-
[email protected] (M.M. Cheng), [email protected] (K.T. Trotman).
Accounting, Organizations and Society 38 (2013) 93–107
Contents lists available at SciVerse ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er. com/ l ocat e/ aos
thereby lowering the likelihood of negotiators identifying
trade-off opportunities and reaching better joint outcomes
(De Dreu et al., 2000). Despite the central role of the ?xed-
pie bias in understanding negotiation performance, the
accounting literature has not investigated mechanisms
that can help mitigate the negative effects of the ?xed-
pie bias when using accounting data.
In this study, we extend recent negotiation literature by
investigating how one aspect of management control,
namely type of accountability, affects negotiators’ ?xed-
pie bias revisions
1
and performance when they are provided
with ABC information during customer–supplier negotia-
tions. The imposition of accountability underlies the princi-
ple of responsibility accounting, which calls for managers to
be held responsible for their actions and performance out-
comes (Bonner, 2007; Rowe, Birnberg, & Shields, 2008).
Here, we compare two types of accountability, process ver-
sus outcome accountability, in an experiment where cus-
tomer–supplier dyads engage in a single period
negotiation over multiple issues, and have the opportunity
to improve joint outcomes by taking advantage of symmet-
ric trade-off opportunities. In our experimental setting, min-
imizing joint costs (i.e., total costs incurred by the customer
and the supplier in each dyad) is considered desirable; both
negotiating parties within a dyad have equal bargaining
power and are faced with the same accountability pressure
– either outcome or process accountability. Our results show
that dyads which are held accountable for their negotiation
processes achieve lower joint costs compared to dyads
which are held accountable for their negotiation outcomes.
In addition, using rich data from taped face-to-face negotia-
tions, we demonstrate that process accountability is more
effective than outcome accountability in reducing joint costs
through two routes: a cognitive route where process
accountability directly reduces negotiators’ ?xed-pie biases,
and a behavioral route where process accountability encour-
ages negotiators to use more integrative (value-creating)
negotiation tactics, which in turn, reduces their ?xed-pie
biases and ultimately reduces joint costs. Demonstrating
the bene?ts of process over outcome accountability is
important because many performance evaluation systems
implicitly emphasize outcome accountability, such as
requiring managers to justify budget variances (e.g., Rowe
et al., 2008), but do not include the requirement for manag-
ers to justify the processes they undertake to arrive at these
outcomes.
‘Accountability’ refers to ‘‘the implicit or explicit expec-
tation that one may be called to justify one’s beliefs, feel-
ings and actions to others’’ (Lerner & Tetlock, 1999, 255;
emphasis added). Research on accountability contends that
this justi?cation requirement causes individuals to engage
in more effortful processing of information before they
make decisions or judgments (Bonner, 2007; Kennedy,
1993; Lerner & Tetlock, 1999; Peecher, 1996; Tan & Kao,
1999). Accountability literature identi?es two distinct
types of accountability: process accountability and out-
come accountability (Lerner & Tetlock, 1999). Process
accountability requires individuals to justify their decision
making processes (e.g., to justify how a capital investment
was selected or how negotiators reached an agreement),
whereas outcome accountability requires individuals to
justify their decision outcomes (e.g., to justify the suc-
cess/failure of an investment or the ?nal negotiation
agreement).
In the management accounting literature, accountabil-
ity is seen as a key issue in designing management control
systems (e.g., Ahrens, 1996; Evans, Heiman-Hoffman, &
Rau, 1994; Merchant & Otley, 2006; Roberts, 1991). How-
ever, the accountability literature relating to management
control systems design typically does not distinguish be-
tween process accountability and outcome accountability.
Instead, this literature examines how management control
systems and accounting information are used as a conduit
for managers to negotiate and discharge their accountabil-
ities to their organizational constituents. While the related
literature on responsibility accounting focuses on the eval-
uation of managers’ performance based on their responsi-
bilities (Cools & Slagmulder, 2009; Rowe et al., 2008), it
typically does not directly examine the cognitive and
behavioral effects of justi?cation – a key feature of
accountability. Two empirical studies that directly exam-
ine the effect of accountability on managerial judgments
are Libby, Salterio, and Webb (2004) and Kadous and Sedor
(2004). Libby et al. manipulated process accountability and
showed that process accountability improved managerial
judgments compared to no accountability. Kadous and Se-
dor found that the requirement to justify a decision in-
creases the likelihood of consultants recommending the
escalation of a project. Kadous and Sedor’s operationaliza-
tion of accountability includes both outcome and process
elements.
2
Thus, the management accounting literature
has not examined the relative merits of process versus out-
come accountability in terms of managerial judgments.
In the psychology literature, a number of studies have
found that process accountability improves judgments
more than outcome accountability in static judgment tasks
such as escalation of commitment (Simonson & Staw,
1992), calibration and probability judgments (Siegel-Ja-
cobs & Yates, 1996), and consistency in interview judg-
ments (Brtek & Motowidlo, 2002). However, no prior
research has compared the effect of process and outcome
accountability in a negotiation context.
3
Unlike static judg-
ment tasks, negotiations are more dynamic and cognitively
complex. During the negotiation process, negotiators need
to make multiple judgments at various stages of the negoti-
ation about their own interests (i.e., their priorities and pref-
erences for the negotiation issues), their counterpart’s
interests, and ?nd a way to integrate these interests to reach
1
Consistent with prior literature (e.g., De Dreu et al., 2000), we de?ne
?xed-pie bias revisions as the difference between the negotiators’ ?xed-pie
biases before and after the negotiation in each negotiation dyad.
2
Kadous and Sedor (2004) operationalized accountability by telling
participants that ‘‘. . . the Board will require you to justify your recommen-
dations in a convincing manner . . .’’. Thus participants’ justi?cation could
involve an explanation of their decision outcomes or their decision
processes or both.
3
In fact, most studies investigating the impact of accountability on
negotiations do not make a distinction between outcome accountability
and process accountability. See De Dreu et al. (2000) and De Dreu, Beersma,
Stroebe, and Euwenma (2006) for exceptions, both of which study the effect
of process accountability versus no accountability on negotiations.
94 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
an agreement. Furthermore, negotiations involve complex
social interactions; negotiators need to employ appropriate
negotiation tactics and respond to the negotiation tactics
of the other party.
Our study makes a number of important contributions.
First, our primary contribution is to show that the imposi-
tion of process accountability (compared to outcome
accountability) on negotiators improves joint costs in cus-
tomer–supplier negotiations. We extend prior accounting
research (Drake & Haka, 2008; Masschelein, Cardinaels, &
Van den Abbeele, 2012; Van den Abbeele et al., 2009) by
investigating how the choice of management control de-
sign, namely, the imposition of process rather than out-
come accountability via the performance evaluation
process, enables negotiators to use ABC information more
effectively to improve their negotiation performance.
While prior studies show that providing negotiators with
ABC information potentially enables negotiators to identify
cost saving opportunities, they also ?nd that negotiators
are sometimes reluctant to share this information and fail
to reach the maximum level of joint outcomes. Therefore,
to maximize supply chain ef?ciencies, organizations need
to do more than providing negotiators with useful infor-
mation. One approach is to improve an organization’s
management control system (Free, 2007). Our results show
that by explicitly incorporating process rather than out-
come accountability in the performance evaluation pro-
cess, accountants can design management control
systems to encourage negotiators to use integrative nego-
tiation tactics such as providing information to, and ques-
tioning the priorities of, the other negotiation party, and
ultimately, lower supply chain joint costs.
Second, Luft and Shields (2009) argue that individuals’
cognitive biases and information search processes in?u-
ence how they can effectively use management accounting
information to make decisions. Our study highlights the
role of the ?xed-pie bias in customer–supplier negotiations,
a variable that has not previously been examined in the
accounting literature. Importantly, our study considers
the differential effect of process accountability and out-
come accountability on negotiating managers’ revisions of
?xed-pie biases. While the psychology literature has shown
that process accountability can reduce the ?xed-pie bias
compared to no accountability (De Dreu et al., 2000), there
has been no research on the effect of outcome accountabil-
ity on the ?xed-pie bias. A direct comparison between the
two types of accountability is important, as our results
demonstrate that the imposition of accountability is only
bene?cial if organizations speci?cally implement process
accountability rather than outcome accountability.
Third, we develop and test a model that explains how
process accountability enables negotiators to reach more
favorable joint outcomes. Our model shows that process
accountability is more effective in lowering joint costs than
outcome accountability by causing negotiators to revise
their ?xed-pie biases, both directly and indirectly through
negotiators’ greater use of integrative negotiation tactics.
Our study demonstrates that cognitive factors (e.g., the
?xed-pie bias) and behavioral factors (e.g., the use of dif-
ferent tactics during negotiation) are not independent,
even though these factors tend to be examined in isolation
of each other in the negotiation literature (De Dreu & Car-
nevale, 2003).
Hypotheses development
ABC techniques, such as analyzing transactions related
to individual customers or suppliers, are frequently used
by organizations to gain insights about the costs associated
with trading with their customers or suppliers (Kaplan &
Anderson, 2007; Reinartz, Thomas, & Kumar, 2005; Roo-
dhooft, Hiel, Van den Abbeele, & Van Doveren, 2003). How-
ever, providing relevant ABC information to negotiators
does not necessarily mean that they will use this informa-
tion effectively to improve negotiation outcomes. Drake
and Haka (2008) ?nd that negotiators are reluctant to share
ABC information, even though those who do so achieve
higher trade ef?ciencies. This reluctance is somewhat re-
duced by the prospect of potential losses due to market
pressure; nonetheless, even under market pressures, nego-
tiators who received ABC information could only capture
about the same level of available trade ef?ciencies in the
market as those who were provided with volume-based
information. Van den Abbeele et al. (2009) ?nd that while
customers are more willing to share ABC information when
negotiating with a more powerful supplier, they are less
likely to do so when there is no power difference between
negotiators. Furthermore, when both parties have equal
power, customers use more distributive (value-claiming)
negotiation tactics and are less likely to use a problem-solv-
ing approach during negotiations, leading to a lower level of
joint outcomes compared to negotiators who have differen-
tial bargaining power. In addition, Masschelein et al. (2012)
?nd that the provision of ABC information does not facili-
tate collaborative negotiations when buyers use ABC infor-
mation to demonstrate that sellers are responsible for
supply chain inef?ciencies. Together, these studies show
that the provision of ABC information only translates to bet-
ter negotiation outcomes if both parties of the negotiation
work collaboratively and make effective use of the ABC
information to identify mutually bene?cial opportunities
to improve supply chain ef?ciencies.
In the present study, we argue that a cognitive bias that
prevents negotiators from working collaboratively is the
negotiators’ ?xed-pie biases (Pinkley et al., 1995; Thomp-
son & Hastie, 1990). The ?xed-pie bias refers to negotia-
tors’ tendency to project their own values and
perspectives on to their negotiation partners, resulting in
an assumption that their partner’s interests are directly
opposite to their own (Lewicki, Barry, & Saunders, 2010;
Pinkley et al., 1995; Thompson & Hastie, 1990). Although
this assumption is correct in negotiations involving a sin-
gle issue, it is often not appropriate in negotiations involv-
ing multiple issues because negotiators typically derive
different utilities from different issues, such that an overall
increase in joint value becomes possible (i.e., ‘integrative
negotiations’) (Lewicki et al., 2010). Assumptions based
on the ?xed-pie beliefs in integrative negotiations are of-
ten erroneous, and the failure to revise them will prevent
negotiators from making mutually bene?cial trade-offs to
‘expand the pie’. We propose that in integrative
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 95
negotiations, process accountability (as opposed to out-
come accountability) will improve negotiators’ ?xed-pie
bias revisions both directly and indirectly through encour-
aging negotiators to use more integrative tactics. Fixed-pie
bias revisions, in turn, mediate the positive effect of pro-
cess accountability on joint costs. These proposed relation-
ships, summarized in Fig. 1, are discussed in detail in the
following sections.
The prevalence of the ?xed-pie bias and the failure of
negotiators to overcome the bias are well-documented in
the psychology literature (De Dreu, 2003; De Dreu et al.,
2000; Pinkley et al., 1995; Thompson &Hastie, 1990). These
studies showthat for integrative negotiations, the ?xed-pie
bias prevents negotiators from reaching better joint out-
comes, and causes negotiators to distort or ignore informa-
tion provided to them about their negotiation partner’s
interests (including the importance the negotiation partner
places on various options within a negotiation issue and
across different negotiation issues). In addition, the man-
agement accounting literature on transfer pricing ?nds that
negotiators do not have accurate understanding of the
interests of their negotiation partners (Chang, Cheng, &
Trotman, 2008; Kachelmeier & Towry, 2002; Luft & Libby,
1997). In the context of negotiations between a customer
and a supplier, the ?xed-pie bias potentially lowers the
bene?ts that may be realized. Speci?cally, the supply chain
environment is often characterized by negotiators bargain-
ing over multiple issues with varying cost implications for
each party (Dekker, 2003; Free, 2007). For example, a sup-
plier may have a comparatively lower cost of capital than
their customers thus allowing them to negotiate a trade-
off between offering a smaller sales discount, in return for
extending the payment terms (Hurkens, van der Valk, &
Wynstra, 2006). In such situations, negotiators’ ability to
revise their ?xed-pie biases is critical in lowering joint
costs. We therefore propose the following hypothesis:
H1. Joint costs are lower for negotiators with larger ?xed-
pie bias revisions.
Next, we propose that one way organizations can
encourage managers to reduce their ?xed-pie biases is to
impose accountability requirements on negotiators, and
more speci?cally, process (rather than outcome) account-
ability. We argue that accountability has both a direct ef-
fect (see H2a) and an indirect effect via its in?uence on
negotiators’ use of negotiation tactics which in turn in-
creases negotiators’ ?xed-pie bias revisions (refer to H2b
and Fig. 1). First, we consider the direct effect. Accountabil-
ity results in an expectation that individuals may be re-
quired to justify their judgments and decisions (Lerner &
Tetlock, 1999). However, because process and outcome
accountability require different types of justi?cation, they
have different effects on individuals’ ability to revise their
?xed-pie biases. Process accountability, which requires
individuals to justify their decision processes, induces an
accuracy (as opposed to directional) goal and motivates
individuals to hold a correct and accurate view of the
world (De Dreu et al., 2006; Kruglanski, 1989; Kruglanski
& Webster, 1996). In doing so, process accountability
encourages individuals to try to understand the environ-
ment surrounding their decisions rather than merely
attaining better outcomes. In addition, by requiring indi-
viduals to justify their decision processes, process account-
ability focuses their attention on ensuring that their
decision processes are appropriate and defendable. This
leads individuals to make a greater effort and be more sys-
tematic in acquiring and processing information about
their environment (Brtek & Motowidlo, 2002; Lerner & Tet-
lock, 1999; Simonson & Staw, 1992).
De Dreu et al. (2000) ?nd that negotiators who are held
accountable for their negotiation processes are more likely
to revise their ?xed-pie biases and identify trade-off
opportunities than those who are not held accountable.
In a series of experiments, De Dreu et al. (2006) ?nd that
process accountability improves negotiators’ ability to uti-
lize cooperative tactical knowledge to reach higher joint
outcomes. They also ?nd that, in the absence of process
accountability, providing negotiators with cooperative tac-
tics did not enable them to perform any better than those
who are given competitive tactics.
4
Together, these studies
Integrative
Tactics
Employed
Fixed-Pie Bias
Revisions
Joint Costs
Accountability
H2a (+)
H2b (+)
H3 (+)
H1 (-)
H2b (+)
H3 (+)
Fig. 1. Causal link between accountability and joint costs. Figure depicts our theoretical model. H1 predicts that negotiators’ ?xed-pie bias revisions reduce
joint costs. H2a predicts that process accountability (compared to outcome accountability) directly increases negotiators’ ?xed-pie bias revisions, while H2b
predicts an indirect effect of process accountability (compared to outcome accountability) on negotiators’ ?xed-pie bias revisions mediated by their use of
integrative tactics employed. Finally, H3 predicts that ?xed-pie bias revisions mediate the effect of the type of accountability on joint costs.
4
The cooperative tactics provided to participants in their study were
similar to integrative tactics and competitive tactics provided to partici-
pants were similar to distributive tactics (De Dreu et al., 2006).
96 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
indicate that process accountability is effective in encourag-
ing individuals to revise their ?xed-pie biases and reach
higher joint outcomes.
While the effect (but not how the effect occurs) of pro-
cess accountability is predictable from psychology, the sit-
uation for outcome accountability is far less clear. Unlike
process accountability, outcome accountability requires
individuals to justify their decision outcomes. First, a sig-
ni?cant amount of accountability literature in auditing,
while not addressing the ?xed-pie bias or negotiation tac-
tics, does indicate that accountability leads to greater vig-
ilance and increases in various forms of performance
(Asare & Wright, 2003; DeZoort, Harrison, & Taylor, 2006;
Glover, 1997; Hoffman & Patton, 1997; Kennedy, 1993;
Koonce, Anderson, & Marchant, 1995; Nelson & Tan,
2005; Tan, 1995; Tan & Kao, 1999). While this literature
does not use the term ‘outcome accountability’ it does ap-
pear to include a form of outcome accountability, or a com-
bination of outcome and process accountability.
5
In
contrast, De Dreu et al. (2000) suggest that outcome
accountability induces directional motivation, which focuses
individuals’ attention on acquiring and processing informa-
tion that enables them to reach a desired outcome. While
De Dreu et al. (2000) do not test the effect of outcome
accountability, they argue that outcome accountability will
induce negotiators to process information in accordance
with their pre-disposed conclusion (that negotiations are
competitive and represent a ?xed-pie). Research in non-
negotiation judgments in psychology shows that outcome
accountability leads to poorer judgments and decisions
(Brtek & Motowidlo, 2002; Simonson & Staw, 1992). For
example, Simonson and Staw (1992) ?nd that outcome
accountability heightened individuals’ concern for justifying
project outcomes at the expense of applying correct decision
strategies, and is therefore less effective than process
accountability in overcoming individuals’ tendency to esca-
late commitment.
As noted earlier, past psychology studies on outcome
accountability, however, are based on static judgment
tasks, rather than in dynamic decision contexts, such as
negotiations which involve multiple judgments, complex
social interactions and the use of, and response to, negoti-
ation tactics. In negotiations, participants are aware that
obtaining a successful outcome depends on a joint agree-
ment being reached. Thus, the motivation to reach a desir-
able outcome may nonetheless encourage negotiators to
make attempts to ?nd out the other negotiation party’s
interests, and therefore lead to the identi?cation of trade-
off opportunities. However, we posit that despite their ef-
fort to do so, negotiators under outcome accountability
will ?nd it dif?cult to revise their ?xed-pie biases, because
their focus on improving negotiation outcomes for them-
selves will reinforce their existing views that the negotia-
tions represent a competitive situation (Thompson &
Hrebec, 1996).
Based on the preceding discussion, we propose that
process accountability motivates negotiators to process
information about how their negotiation partners’ inter-
ests are different fromtheir own, thereby enabling a reduc-
tion of their ?xed-pie biases. On the other hand, outcome
accountability will cause individuals to focus on meeting
their own interests and therefore does not facilitate
?xed-pie bias revisions. Thus, we predict the following
(see Fig. 1):
H2a. Process accountability (compared to outcome
accountability) directly improves negotiators’ ?xed-pie
bias revisions.
We further propose that accountability also indirectly
in?uences ?xed-pie bias revisions through its effect on
negotiators’ use of negotiation tactics. The negotiation lit-
erature suggests that negotiation outcomes are in?uenced
by both negotiators’ cognitive processes and the social pro-
cesses during the negotiation, in particular, the tactics em-
ployed by different parties (Thompson, Wang, & Gunia,
2010). Negotiation tactics are usually classi?ed into two
basic types: integrative tactics and distributive tactics
(Lewicki et al., 2010). Integrative negotiation tactics are
aimed at addressing the objectives of both parties by gain-
ing a better understanding of the underlying interests of
both parties and trading off less important issues for more
important issues (Weingart, Hyder, & Prietula, 1996; Lew-
icki et al., 2010). Distributive negotiation tactics, on the
other hand, are aimed at obtaining the maximum conces-
sions on each of the issues from the other party, rather
than understanding his/her interests (Hyder, Prietula, &
Weingart, 2000; Trotman, Wright, & Wright, 2005; Wein-
gart et al., 1996). While prior negotiation literature has
established the important role negotiation tactics play in
enhancing joint outcomes (e.g., Lewicki et al., 2010; Ole-
kalns, Smith, & Walsh, 1996), the literature has not exam-
ined how different types of accountability can affect
negotiators’ choice of integrative versus distributive
tactics.
Given that process accountability will motivate negoti-
ators to gain a better understanding of their negotiation
processes and their partner’s priorities, we predict that
managers who are held accountable for their negotiation
processes will employ more integrative tactics (and fewer
distributive tactics) than those who are held accountable
for their negotiation outcomes. Prior literature also ?nds
that negotiators who employ integrative tactics are also
more likely than those who employ distributive tactics to
acquire information from and share information with the
other party (Hyder et al., 2000; Olekalns et al., 1996). This
information acquisition and sharing allow negotiators to
better understand both parties’ interests. Thus, we predict
that the bene?cial effect process accountability has on the
use of integrative tactics (compared to outcome account-
ability) will indirectly increase ?xed-pie bias revisions.
Taken together, we hypothesize:
5
For a review of research on accountability in the auditing literature,
refer to DeZoort et al. (2006). Overall, this literature uses a variety of ways
to operationalize the justi?cation requirement, such as imposing audit
review pressures or requiring auditors to write justi?cations for audit
judgments. Although these manipulations appear more in line with
outcome than process accountability, they could be perceived to incorpo-
rate both, as the review process usually considers both judgments and
processes used to form those judgments (Gibbins & Trotman, 2002).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 97
H2b. Process accountability (compared to outcome
accountability) indirectly improves negotiators’ ?xed-pie
bias revisions mediated by negotiators’ use of integrative
tactics.
Finally, as argued earlier, the ?xed-pie bias is a key
obstacle to reaching a higher level of joint outcomes in
negotiations (Lewicki et al., 2010). Hence, we predict that
the two types of accountability will also indirectly affect
the ability of negotiators to reach agreements resulting in
lower total joint costs. We therefore predict that the effect
of the type of accountability on joint costs is mediated by
?xed-pie bias revisions:
H3. Fixed-pie bias revisions mediate the effect of the type
of accountability on joint costs.
Although prior research shows that the use of more
integrative tactics will lead to higher joint outcomes (Hy-
der et al., 2000; Van den Abbeele et al., 2009; Weingart
et al., 1996), we suggest that this effect can be attributed
to integrative negotiation tactics reducing negotiators’
?xed-pie biases. While the use of integrative negotiation
tactics facilitates the exchange of information, De Dreu
et al. (2000) argue that negotiators are unlikely to reach
agreements with higher joint outcomes without using this
information to revise their ?xed-pie biases. We therefore
do not predict a direct effect from negotiation tactics to
joint costs.
Research methods
Participants
Sixty-six participants (33 negotiation dyads) volun-
teered to participate in this experiment. All participants
were enrolled in either a MBA degree or Master of Com-
merce degree. Approximately half of the participants were
enrolled in a large USA university while the other half was
enrolled in a large Australian university.
6
Three negotiation
dyads failed one or more post experiment manipulation
tests and were later excluded from the analysis, resulting
in usable responses from 30 negotiation dyads. The cell sizes
were 15 dyads for both treatment groups.
7
Research design
Participants were randomly assigned into the role of
either the supplier or the customer, and then were formed
into dyads to negotiate a sales contract. Two types of
accountability were manipulated: outcome accountability
and process accountability. Each dyad was randomly as-
signed to one of these treatments (i.e., both negotiators
within a dyad received the same accountability treatment).
All participants received a ?xed compensation of US$20
and ‘bonus points’ that gave them chances to win a ‘grand
prize’ worth US$100.
8
The bonus points calculation had two
components: ?rst, all participants received bonus points
based on their ?nal negotiation outcomes using a ‘piece rate’
system, where one bonus point was awarded for every
$2000, or part thereof, for costs that fell below the maxi-
mum cost an individual negotiator could incur in the ?nal
agreement. Thus, the lower the cost, the more bonus points
they would earn.
9
We manipulate accountability using the second compo-
nent of the bonus point calculation. Lerner and Tetlock
(1999) de?ne accountability as the expectation that indi-
viduals may be called upon to justify their performance,
action or judgments to others, and that they will be re-
warded (penalized) if they provide satisfactory (unsatisfac-
tory) justi?cations. Based on this de?nition, participants in
the ‘outcome accountability’ condition were required to
write a report to justify their negotiation outcomes, and
were told that they would receive additional bonus points
based on the quality of this report.
10
They were told that
the negotiation outcome report needed to address both the
negotiation result and whether the outcome was considered
acceptable for their company. They were also told that the
negotiation outcome report would be graded by one of the
authors and the participant could receive between 0 and
40 bonus points depending on whether their report was
judged to be unacceptable, needs improvement, competent,
very good or outstanding.
11
For participants in the ‘process
accountability’ condition, they were required to write a re-
port based on their negotiation processes rather than nego-
tiation outcome. In particular, they were told that their
report needed to address the strategies and procedures em-
ployed during their negotiations. Similar to those in the out-
come accountability condition, participants in the process
accountability condition were also told of the method of
grading their report and that they could also receive be-
tween 0 and 40 bonus points depending on the quality of
their reports.
6
In both locations, participants were randomly assigned into one of the
accountability treatments. Nine out of the 15 dyads in the process
accountability treatment and eight out of the 15 dyads in the outcome
accountability treatments were USA participants. To investigate whether
location affects our results, we re-analyzed our data with location (USA
versus Australia) as a co-variate and ?xed-pie bias revisions, proportion of
negotiation tactics, and joint costs as dependent variables. We found that
location was not a signi?cant co-variate and the differences between
process accountability and outcome accountability remained signi?cant for
all three dependent variables after controlling for location.
7
As discussed subsequently, we also included a no accountability control
group. This involves 16 dyads with participants randomly assigned to
complete the same task as the two treatment groups, but without the
accountability requirement.
8
For USA participants the compensation was paid in cash and for
Australian participants, the compensation was in the form of movie
coupons valued at AU$25 (for the ?xed compensation) and a $130 gift card
(for the grand prize). The use of non-cash rewards for Australian partic-
ipants was due to university administrative restrictions. The value of
compensation was approximately matched between the two countries
based on the exchange rate during the experiments.
9
Consistent with prior psychology research on accountability (e.g., De
Dreu et al., 2000), we tie participants’ compensation to their negotiation
outcomes to ensure that participants take the negotiation task seriously.
This component of bonus point calculation does not affect our account-
ability manipulation because it is not linked to a justi?cation requirement.
10
The use of ‘report writing’ to manipulate accountability is consistent
with a number of earlier studies including O’Connor (1997), Gelfand and
Realo (1999) and Libby et al. (2004).
11
The averages of bonus points awarded across the conditions were 34.4
and 33.5 for outcome accountability and process accountability
respectively.
98 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
Experimental task
Participants assume the role of either the sales manager
of Glass Company (the supplier) or the purchasing man-
ager of Furniture Company (the customer). Glass Company
manufactures a wide range of glass-related products,
including tempered glass panels which are used as inputs
for Furniture Company, a company that manufactures a
wide range of household furniture products including glass
top tables. To control for the effect of bargaining power,
participants are told that both companies are similar in
size and that neither of them have an alternative trading
partner.
Participants are informed of instructions from their
respective divisional managers to negotiate a sales con-
tract with the intent of establishing a long-term strategic
relationship with the other party. The sales contract in-
volves the sales of a ?xed number of 2800 tempered glass
panels at a predetermined sales price $250 each (i.e., sales
quantity and price are not negotiable). Similar to previous
studies such as Drake and Haka (2008), the negotiators are
required to negotiate six after-sales services (each with
three levels) to be provided by the supplier to the cus-
tomer. The participants are also told that each of these ser-
vices will have potential additional cost implications for
both parties. For example, the customer has three different
factory locations; the supplier’s costs will increase and the
customer’s costs will decrease with each delivery location
serviced by the supplier. The aim of the negotiation is to
reach an agreement on the level of services associated with
each of the six issues for their respective companies. Par-
ticipants are provided with a table describing the six issues
and their cost implications for their own company but not
for the other company (Table 1 summarizes the suppliers’
and the customers’ cost structures).
Consistent with earlier negotiation studies in the psy-
chology literature on the ?xed-pie bias (e.g., Pinkley
et al., 1995), we include four trade-off issues and two
zero-sum issues. Two examples of trade-off issues are
‘marketing support’ and ‘design support’ (Table 1 shows
that while the customer ?nds ‘marketing support’ more
costly than ‘design support’, the opposite is true for the
supplier). Delivery location and glass cutting are the other
two trade-off issues. The two zero-sum issues are inspec-
tion sample size and waste collection frequency. For exam-
ple, with the zero-sum issue ‘inspection sample size’,
Table 1 shows that the cost function of ‘inspection sample
size’ is exactly the same for both the supplier and the cus-
tomer. This design allows us to test the extent to which
participants can revise their ?xed-pie biases by identifying
trade-off potentials, and thus reach the Pareto-ef?cient
solution.
12
The possible joint costs range from $190,000
Table 1
Customer’s and supplier’s cost structures.
Issues to negotiate Options Supplier’s cost per year ($) Customer’s cost per year ($)
Inspection sample size
a
1% 9000 27,000
2% 18,000 18,000
3% 27,000 9000
Marketing support
b
Low 4500 36,000
Medium 9000 24,000
High 13,500 12,000
Delivery locations
b
1 Factory 15,000 16,500
2 Factories 30,000 11,000
3 Factories 45,000 5500
Waste collection frequency
a
1 per quarter 10,000 30,000
2 per quarter 20,000 20,000
3 per quarter 30,000 10,000
Design support
b
Basic support 12,000 13,500
CAD only 24,000 9000
CAD & CAM 36,000 4500
Glass cutting
b
Standard Cut 5500 45,000
Customized Cut 11,000 30,000
Full Finished Cut 16,500 15,000
Note: Participants only receive cost schedules relative to their roles (e.g., suppliers only receive information about their own costs) and within each issue,
participants are provided with a brief description. For example, in relation to inspection sample size, suppliers are told that ‘‘Your cost increases with larger
sample sizes, i.e. more extensive testing for Furniture Co. ($9000 per%)’’. Customers on the other hand, are told that ‘‘Your cost decreases with larger sample
sizes, i.e. the more extensive testing from Glass Co ($9000 per%).’’
a
Zero-sum issues (i.e., where the cost functions are the same for both the supplier and the customer).
b
Trade-off issues (i.e., where the priorities are different between the supplier and the customer.) The lowest level of joint costs for both parties is
calculated as follows: Zero-sum issues: $36,000 [inspection sample size] + $40,000 [waste collection]; Trade-off issues: $(13,500 + 12,000) [high marketing
support] + $(15,000 + 16,500) [one delivery location] + $(12,000 + 13,500) [basic design support] + $(16,500 + 15,000) [full ?nished glass cutting]. Total joint
costs = $190,000.
12
The trade-off issues are designed to be symmetrical in value, such that
the suppliers’ most costly issues (design support at $12,000 per level and
delivery locations at $15,000 per level) are also the customers’ least costly
issues ($4500 and $5500 per level respectively). Similarly, the customers’
most costly issues (marketing support at $12,000 per level and glass cutting
at $15,000 per level) are the suppliers’ least costly issues ($4500 and $5500
per level respectively) (see Table 1 for more details). Consistent with earlier
studies (e.g., Thompson & Hastie, 1990), we made this design choice to
allow both negotiating parties to have the same opportunities to contribute
to increasing the joint outcomes.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 99
(lowest possible joint costs) to $258,000 (highest possible
joint costs). From the individual negotiator’s perspective,
each negotiator can incur a maximum cost of $168,000
and a minimum cost of $56,000.
Experimental procedures
Upon arrival at the experimental rooms, participants
were seated on opposite sides of a table with a chest-high
barrier that prevented them from seeing each other’s infor-
mation (the barrier, however, did not prevent negotiators
from seeing each other face-to-face). A digital recorder
was placed on top of the barrier to record their negotiation.
The experiment had four stages: introduction, prepara-
tion, face-to-face negotiation and post-negotiation (refer to
Fig. 2). In the introduction stage, participants read through
a package containing background information about their
roles and their cost schedules and were asked two ques-
tions to ensure that they understood the task. They were
asked to indicate the highest and lowest costs on their
schedule and calculate the total cost and the points they
would earn given a hypothetical agreement. Most partici-
pants were correct in their ?rst attempt, and all partici-
pants obtained the correct answers on their second
attempt.
In the preparation stage, in order to elicit participants’
initial level of the ?xed-pie bias (see De Dreu et al.,
2000), participants were ?rst asked to estimate their nego-
tiation partner’s costs by ?lling in a blank cost schedule.
Participants were permitted to refer to their own cost sche-
dule as a point of reference. They were also asked to plan
their negotiation by thinking about and noting down their
expectations about good negotiation outcomes (‘outcome
accountability’ treatment) or good negotiation processes
(‘process accountability’ treatment).
At the face-to-face negotiation stage, participants were
given 20 min to reach an agreement on all six issues relat-
ing to after-sales services. They then completed post-nego-
tiation reports detailing their negotiated agreement. From
this information joint costs were calculated by the
researchers. Finally, in the post-negotiation stage, they
were asked to re-estimate the other party’s cost schedule
(to measure the post-negotiation ?xed-pie bias), and com-
plete manipulation checks and demographic questions.
The average length of time to complete the experiment
was 1 h.
Dependent variables
The dependent variables are joint costs, ?xed-pie bias
revisions, and the proportion of integrative negotiation
tactics used. Joint costs are calculated by summing the
costs incurred by the customer and the supplier within a
negotiation dyad (see also Drake & Haka, 2008; Van den
Abbeele et al., 2009). As noted earlier, the lowest level of
joint costs is $190,000, whereas the highest level of joint
costs is $258,000.
Consistent with prior research in the negotiation litera-
ture (De Dreu, 2003), we measure ?xed-pie bias revisions
by subtracting participants’ ?xed-pie biases before the
negotiation from those immediately after the negotiation.
As the ?xed-pie bias refers to the extent to which a
First stage (Introduction)
• Participants form into negotiation dyads.
• Participants receive background information, and answer two
questions to test their understanding of the task.
Second stage (Preparation)
• Participants estimate their negotiation partner’s cost schedule
(this measures the fixed-pie bias prior to negotiation).
• Participants note down a plan of their negotiation (to reinforce
the manipulation of accountability).
Third stage (Face-to-face negotiation)
• Participants have 20 minutes to reach an agreement.
• Participants complete a report that summarizes their agreement.
• Participants under the accountability treatments complete their
accountability reports.
Fourth stage (Post-negotiation)
• Participants re-estimate their negotiation partner’s cost (the
post-negotiation measure of the fixed-pie bias).
• Participants complete manipulation checks, demographic
questions, and receive payments.
End of experiment
Fig. 2. Stages of experimental procedures.
100 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
negotiator assumes that his/her negotiation partner’s costs
are directly opposite to his/her own costs, the following
formula (see De Dreu et al., 2000) is used to estimate the
?xed-pie bias for both members of each negotiating
dyad
13
:
Fixed-pie bias ¼
X
4
i¼1
X
3
j¼1
ðEstimates of partner
0
s costs
ij
Ànegotiator
0
s own costs
ij
Þ
i refers to the four trade-off issues incorporated in the task,
and j refers to the three levels of cost within each trade-off
issue; the smaller the score, the larger the ?xed-pie bias (a
small score indicates the belief that the negotiation part-
ner’s cost is very similar to the negotiator’s own cost, i.e.,
?xed-pie bias).
14
Therefore, an increase in a dyad’s ?xed-
pie bias score after the negotiation indicates a reduction in
the dyad’s ?xed-pie bias (i.e., a positive ?xed-pie bias
revision).
The third dependent variable is the proportion of inte-
grative tactics used, with the unit of analysis being a nego-
tiation dyad. To compute this variable, we code the types of
negotiation tactics used by the dyads based on the audio
recordings of the negotiations using a well-established
coding scheme developed by Weingart et al. (1996). In
their coding scheme, ?ve different negotiating behaviors
are identi?ed: offers, information provision, questions,
procedural comments and miscellaneous comments. These
behaviors are then classi?ed as either integrative tactics or
distributive tactics. Previous negotiation research argues
that integrative tactics are aimed at addressing the under-
lying interests of the negotiating parties and identifying
mutually bene?cial trade-offs (Pruitt, 1981; Weingart
et al., 1996). For this to happen, negotiators need to under-
stand each other’s priorities and then trade-off those less
important issues for more important issues. Negotiation
behaviors that address multiple issues at a time, and focus
on priorities across issues, enable negotiators to identify
these cost saving trade-offs and are classi?ed as integrative
tactics. Negotiation behaviors are classi?ed as distributive
tactics if they focused on addressing single issues and pref-
erences within those issues. Thus, negotiation behaviors
that address one issue at a time and negotiators’ prefer-
ences within an issue will highlight the con?icting inter-
ests between the two parties. Finally, arguments aimed
at persuading the other party to change his/her attitudes
towards issues are also coded as distributive tactics.
Behaviors that could not be classi?ed as either distributive
or integrative were classi?ed as miscellaneous com-
ments
15
(Weingart et al., 1996).
The audio recordings of each negotiation were ?rst
transcribed to identify each negotiator by role and speak-
ing turns. Each transcript was then coded, using the above
coding scheme, by two research assistants blind to the re-
search hypotheses.
16
Inter-rater reliability, as measured by
Cohen’s (1960) kappa, was computed for each transcript
and ranged from 0.75 to 0.96. Consistent with prior research
(Sullivan, O’Connor, & Burris, 2006; Weingart et al., 1996),
each speaker turn was subdivided into thought units. A
thought unit is a statement or group of statements that fo-
cuses on one idea or thought. Using the above coding
scheme, we coded each thought unit as either a speci?c
negotiation tactic or a miscellaneous comment (e.g., greet-
ings). Negotiation tactics were included in the following for-
mula to determine the proportion of integrative tactics (per
negotiation dyad):
Proportionof integrative tacticsðper dyadÞ
¼
P
integrative tactics coded
P
integrative tactics codedþ
P
distributive tactics coded
Results
Manipulation checks
Participants answered two manipulation check ques-
tions. The ?rst question asked participants to indicate their
role (as a supplier or a customer), and the second question
asked participants to indicate if the reports they wrote re-
lated to the justi?cation of their negotiation outcome or
their negotiation process. As noted earlier, three partici-
pants failed one or more of these questions and conse-
quently, three dyads were excluded from our analysis.
17
We also examined the accountability reports submitted by
each negotiation dyad to check if their reports focused on
justifying the negotiation outcomes or negotiation pro-
cesses. The 30 negotiation dyads included in our ?nal anal-
13
The formula we use to measure the ?xed-pie bias is well established in
the negotiation literature (e.g., De Dreu et al., 2000; Halevy, Chou, &
Murnighan, 2012; Pietroni, Van Kleef, De Dreu, & Pagliaro, 2008). Essen-
tially, the formula takes the absolute difference between the participant’s
estimates of the other party’s costs and his/her own costs across all levels of
the trade-off issues (in our study, these are: marketing support, delivery
locations, design support and glass cutting). An absolute score of zero
indicates that the participant believes that the other party’s cost (prefer-
ence) for the trade-off issues are exactly opposite to his/her own, thus
yielding a zero-sum estimation. As this formula is based on the assumption
that the negotiation partner’s actual cost is not the same as the negotiator’s
own costs, it only applies to the trade-off issues.
14
For example, the issue ‘marketing support’ has three levels of services:
low, medium or high. To calculate the ?xed-pie bias for the supplier, we
sum the absolute difference between the supplier’s own marketing support
costs and his/her estimates of the customer’s marketing support costs for
each of the three levels of service.
15
Miscellaneous comments generally refer to off-task comments that are
not negotiation tactics.
16
Overall, three research assistants were involved in the coding process
but only two coded each transcript. To train the research assistants, each
research assistant was ?rst required to individually code ten randomly
selected transcripts using a detailed coding sheet. The training was
considered complete when the inter-rater reliability among the coders
reached an acceptable level (Cohen’s kappa > 0.75; Cohen, 1960). For the
rest of the coding process, the research assistants were randomly paired to
code ?ve transcripts at a time. After the transcripts were coded, one of the
authors would meet with the coders to discuss the coding and reconcile
disagreements by reviewing the transcripts again. A single set of codes was
then agreed upon for each dyad. The negotiation tactics were then analyzed
using this data.
17
The inclusion of the three dyads reduced the value of the t-statistics for
the impact of accountability on joint costs from t = 2.000 (p = 0.030) to
t = 1.53 (p = 0.068). However, the direction and signi?cance levels of the
other results were not changed by including these three dyads.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 101
ysis all submitted accountability reports that were consis-
tent with our manipulation.
Tests of hypotheses
To test our hypotheses, we employ a partial least square
(PLS) path approach. PLS allows us to simultaneously test
all the relationships in our proposed model, including both
the direct and indirect linkages. As a result of the distribu-
tion-free assumption of PLS, bootstrapping resampling is
used to test the signi?cance of paths (Chin, 1998). We em-
ploy SmartPLS (v2.0) with a bootstrapping resampling pro-
cedure of 500 samples (generated from the original data)
18
to test all hypotheses. The descriptive statistics on joint costs
and ?xed-pie bias revisions are shown in Table 2, and the
path coef?cients and t-statistics of the PLS model are shown
in Fig. 3 and Table 3.
H1 predicts that joint costs are lower for negotiators
with larger ?xed-pie bias revisions. The PLS model pre-
sented in Fig. 3 and Table 3 Panel A show that the path
from ?xed-pie bias revisions to joint costs is negative and
signi?cant (b = À0.310, t = 1.777, p = 0.038),
19
indicating
that larger ?xed-pie revisions result in lower joint costs.
Thus, H1 is supported.
H2a and H2b predict the direct and indirect effects of
process versus outcome accountability on ?xed-pie bias
revisions, respectively. Table 2 Panel B shows ?xed-pie bias
revisions of $176,683 for process accountability and
$71,167 for outcome accountability. We note that while
outcome accountability negotiators had a relatively lower
beginning ?xed-pie bias compared to process accountabil-
ity negotiators ($73,186 versus $56,443), this difference is
not statistically signi?cant (t = 0.710, p = 0.483).
20
Consistent with H2a, the path from accountability to
?xed-pie bias revisions is positive and signi?cant
(b = 0.265, t = 2.003, p = 0.023). In our PLS model, process
accountability is coded as 2 and outcome accountability
is coded as 1; as such, the positive coef?cient shows that
?xed-pie bias revisions are higher for participants under
process accountability than those under outcome account-
ability. Thus, H2a is supported.
H2b predicts an indirect effect from accountability to
?xed-pie bias revisions, mediated by the negotiators’ use
of integrative tactics. Table 4 reports the negotiation tac-
tics used by participants (with the unit of analysis for each
treatment being a dyad). The audio transcripts reveal a to-
tal of 1876 thought units that were coded as negotiation
tactics; of these, 462 thought units were coded as integra-
tive tactics and 1414 thought units were coded as distrib-
utive tactics.
21
The overall proportion of integrative tactics
is 28.22% is consistent with prior negotiation literature
(Weingart et al., 1996), which shows negotiators generally
use a lower portion of integrative tactics compared to dis-
tributive tactics. As predicted by H2b, Table 3 Panel A shows
that accountability is positively associated with the propor-
tion of integrative tactics used by negotiating dyads
(b = 0.340, t = 2.150, p = 0.016),
22
and the proportion of inte-
grative tactics used is positively associated with ?xed-pie
bias revisions (b = 0.538, t = 4.559, p < 0.000). The Sobel’s
test (Table 3 Panel B) reports that this indirect relationship
is signi?cant (b = 0.183, t = 1.984, p = 0.024), showing a com-
plementary mediation effect (Zhao, Lynch, & Chen, 2010)
and providing support for H2b.
Table 2
Means (standard deviation) of joint costs, ?xed-pie bias and ?xed-pie bias revisions.
Process accountability (n = 15) Outcome accountability (n = 15) Overall
Panel A: Joint costs
Joint costs ($) 194,133 (6682) 201,600 (12,826) 197,867 (10,742)
Panel B: Fixed-pie bias
Fixed-pie bias (beginning) $ 56,433 (55,315) 73,186 (72,683) 64,810 (64,032)
Fixed-pie bias (end) $ 233,116 (104,136) 144,353 (91,276) 188,735 (106,278)
Fixed-pie bias revisions $ 176,683 (117,626) 71,167 (99,467) 123,925 (119,729)
Joint costs = combined costs of the supplier and the customer averaged across the 15 dyads in each treatment (Lowest joint costs = $190,000; Maximum
joint costs = $258,000).
Fixed-pie bias is the extent to which a negotiator assumes that his/her negotiation partner’s costs are the same as their own costs.
Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
scosts
ij
Ànegotiator
0
sown costs
ij
Þ
(i refers to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue).
Fixed-pie bias (beginning) = the combined ?xed-pie biases of the two negotiators within each dyad at the beginning of the negotiation, a smaller amount
indicates greater bias.
Fixed-pie bias (end) = the combined ?xed-pie biases of the two negotiators within each dyad after the negotiation has been completed, a smaller amount
indicates greater bias.
Fixed-pie bias revisions = ?xed-pie bias (end) – ?xed-pie bias (beginning); a larger amount indicates greater extent of revision.
18
Consistent with Rosenzweig (2009), we tested the stability of the
coef?cient estimates by repeating the bootstrap resampling procedure
using 250 and 1,000 bootstrap samples with consistent results.
19
All signi?cance tests are one-tailed unless otherwise speci?ed.
20
This relatively higher initial ?xed-pie bias in the process accountability
treatment works against the predictions of H2, as negotiators in the process
accountability treatment have further to go when revising their ?xed-pie
biases.
21
In addition, 1182 thought units are coded as miscellaneous statements,
such as greetings. There is no signi?cant difference in the proportion of
miscellaneous statements between the process accountability treatment
and the outcome accountability treatment (39.85% versus 44.81%, t = 0.995,
p = 0.328 two-tailed).
22
An independent t-test shows that the proportion of integrative tactics
employed by dyads in the process accountability treatment is higher than
the proportion of integrative tactics used by dyads in the outcome
accountability treatment (33.72% versus 22.72%, t = 1.876, p = 0.036, one-
tailed). This provides additional support to the path from accountability to
negotiators’ use of integrative tactics.
102 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
Table 2 Panel A shows that the mean joint costs for the
supplier and the customer combined is $194,133 for pro-
cess accountability and $201,600 for outcome accountabil-
ity (this compares to the lowest level of joint costs of
$190,000 and maximum joint costs of $258,000). H3 pre-
dicts that ?xed-pie bias revisions mediate the effect of
the type of accountability on joint costs. We have previ-
ously shown that a signi?cant path from ?xed-pie bias
revisions to joint costs (H1), and from accountability to
?xed-pie bias revisions both directly (H2a) and indirectly
via the proportion of integrative tactics used by the nego-
tiating dyads (H2b). To examine H3, we conducted two So-
bel’s tests. First, Table 3 Panel B reports that the indirect
relationship from accountability to joint costs via ?xed-
pie bias revisions is marginally signi?cant (b = 0.082,
t = 1.433, p = 0.076). A second Sobel’s test starts with the
Integrative
Tactics
Employed
Fixed-Pie Bias
Revisions
Joint Costs
Accountability
0.265**
-0.310**
0.340**
0.538***
Fig. 3. Path analysis co-ef?cients. Figure only shows path coef?cients signi?cant (one-tailed) at the 5% (
ÃÃ
) and 1% (
ÃÃÃ
) levels. In addition to these signi?cant
paths, our model also includes two other paths: a path from Accountability to Joint Costs (b = À0.141, not signi?cant) and a path from Integrative Tactics
Employed to Joint Costs (b = À0.215, not signi?cant). For indirect effect path co-ef?cients refer to Table 3 Panel B. Accountability refers to process versus
outcome accountability; that is, whether participants are held accountable for their negotiation processes (coded as 2) or negotiation outcomes (coded as
1). Fixed-pie bias revisions refers to the difference between the ?xed-pie bias before and after the negotiation in each negotiation dyad. The following
formula is used to calculate each negotiator’s ?xed-pie bias: Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
s costs
ij
Ànegotiator
0
s own costs
ij
Þj (i refers
to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue). Integrative tactics employed refers to
the proportion of integrative tactics used by a negotiation dyad relative to the total number of negotiation tactics used by that dyad. The following formula
is used to calculate integrative tactics employed (per dyad): Integrative tactics employed = integration tactics Ä (integrative + distributive tactics). Joint
costs refers to the combined costs of the supplier and the customer in each dyad.
Table 3
PLS results.
Variables Path to R
2
Proportion of integrative tactics Fixed-pie bias revisions Joint costs
Panel A: Path coef?cients and t-statistics
Accountability 0.340 (t = 2.150)
**
0.265 (t = 2.003)
**
À0.141 (t = 0.880)
Proportion of integrative tactics 0.538 (t = 4.559)
***
À0.215 (t = 1.274) 0.456
Fixed-pie bias revisions À0.310 (t = 1.777)
**
0.116
Joint costs 0.306
Variables Linkages Path to
Fixed-pie bias revisions Joint costs
Panel B: Indirect effect path coef?cients and t-statistics (Sobel’s test
a
)
Accountability Fixed-pie bias revisions 0.082 (t = 1.433)
*
Proportion of integrative tactics 0.183 (t = 1.984)
**
Proportion of integrative tactics Fixed-pie bias revisions 0.167 (t = 1.691)
**
Accountability = process versus outcome.
Fixed-pie bias revisions = ?xed-pie bias (end) À ?xed-pie bias (beginning); a larger amount indicates greater extent of revision.
Fixed-pie bias is the extent to which a negotiator assumes that his/her negotiation partner’s costs are the same as their own costs.
Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
scosts
ij
Ànegotiator
0
s own costs
ij
Þj
(i refers to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue).
Proportion of integrative tactics = integration tactics Ä (integrative + distributive tactics).
Joint costs = combined costs of the supplier and the customer in each dyad.
a
Sobel’s test is used to test the statistical signi?cance of indirect relationships between an independent variable and a dependent variable through a
mediating variable (Preacher & Leonardelli, 2001).
*
p < 0.10 (one-tailed).
**
p < 0.05 (one-tailed).
***
p < 0.01 (one-tailed).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 103
proportion of integrative tactics (which was directly af-
fected by accountability), and establishes a signi?cant indi-
rect path from proportion of integrative tactics to joint
costs via ?xed-pie bias revisions (b = 0.167, t = 1.691,
p = 0.045). These results, combined with an insigni?cant
direct effect from accountability to joint costs (b = 0.141,
t = 0.880, p = 0.386, two-tailed), show that accountability
has an indirect-only mediation effect on joint costs (Zhao
et al., 2010). Thus, H3 is supported.
23
Although we did not hypothesize a direct link from pro-
portion of integrative tactics to joint costs, we have in-
cluded this link in our ?nal PLS model for completeness.
As expected, the direct path from proportion of integrative
tactics to joint cost is not signi?cant (b = À0.215, t = 1.274,
p = 0.214, two-tailed), indicating that the dyads’ use of
integrative tactics is only effective in reducing joint costs
via their effect on ?xed-pie bias revisions.
Additional analysis
In addition to the above design we included a control
group (no accountability) of 16 dyads. For the control
group, participants were neither required to justify their
negotiation processes nor their negotiation outcomes. In-
stead, their bonus points depended solely on the ?nal cost
to be borne by their respective company at the end of the
negotiation. The inclusion of this control group allows us to
see if either accountability treatments result in different
negotiation performance relative to a baseline condition
where participants are rewarded based only on cost mini-
mization. This control condition thus represents a manage-
ment control system that does not explicitly incorporate
an accountability requirement. We conducted post-hoc
pairwise comparisons between the control group (no
accountability) and each of the two treatment groups.
There were no signi?cant differences between the control
group and the outcome accountability treatment with re-
spect to both ?xed-pie bias revisions ($88,627 versus
$71,167, t = 0.445, p = 0.670, two-tailed) and joint costs
($203,718 versus $201,600, t = 0.399, p = 0.694, two-
tailed). In contrast, compared to participants in the control
group, those in the process accountability treatment have
signi?cantly higher ?xed-pie bias revisions ($176,683 ver-
sus $88,627, t = 2.245, p = 0.030, two-tailed) and signi?-
cantly lower joint costs ($194,133 versus $203,718,
t = 2.130, p = 0.046, two-tailed). The results show that only
process accountability was effective in improving negotia-
tors’ performance beyond the no accountability
alternative.
24
We also conducted further analysis of the tactics used
by the negotiators, by analyzing the four types of negotia-
tion behaviors used by the negotiators across both process
and outcome accountability. Table 4 shows that the rela-
tive proportions of negotiation behaviors adopted by
Table 4
Summary of negotiation tactics used by negotiating dyads.
*
Negotiation behavior Process Accountability (% as a
proportion of total number of
tactic-related thought units coded)
Outcome accountability (% as a
proportion of total number of tactic-
related thought units coded)
Overall (% as a proportion of total
number of tactic-related thought
units coded)
Tactic type Integrative Distributive Total Integrative Distributive Total Integrative Distributive Total
Offers
a
9.40% 10.79% 20.19% 11.73% 15.30% 27.03% 10.57% 13.00% 23.57%
Information provision
b
10.22% 37.05% 47.27% 4.26% 40.70% 44.96% 7.24% 38.92% 46.16%
Questions
c
8.55% 11.59% 20.14% 2.85% 12.65% 15.50% 5.70% 12.12% 17.82%
Procedural comments
d
5.55% 6.85% 12.40% 3.88% 8.63% 12.51% 4.71% 7.74% 12.45%
Total 33.72% 66.28% 100% 22.72% 77.28% 100% 28.22% 71.78% 100%
Total number of thought units
coded as tactics
269 Units 602 Units 871
Units
193 Units 812 Units 1005
Units
462 Units 1414 Units 1876
Units
*
The unit of analysis for negotiation tactics used is a dyad.
a
Integrative offers – offers and counter offers relating to more than one issue at a time. Distributive offers – offers and counter offers relating to only one
issue at a time.
b
Integrative information provision – communicating information about which issues are more important to themselves. Distributive information pro-
vision – communicating information about their preferences within a particular issue, and making arguments for their own position, or make arguments
against the other’s position, in an attempt to persuade the other party to shift their position.
c
Integrative questions – asking questions about which issues are more important to their negotiation partner. Distributive questions – asking questions
about their negotiation partner’s preferences within an issue and questioning the validity of their negotiation partner’s arguments.
d
Integrative procedural comments – suggestions about packaging issues together or discussing sets of issues simultaneously, or when the negotiator
suggests a current concession to be made in exchange for an unidenti?ed future concession. Distributive procedural comments – suggestions to address one
issue at a time or when negotiators discuss one issue at a time or when the negotiators suggest a willingness to concede their position within an issue
(Weingart et al., 1996).
23
We also conduct a 2 Â (2) ANOVA with ‘accountability’ being the
between-subjects factor and ‘time’ (before versus after negotiation) being
the repeated measure factor. The results show a signi?cant interaction
effect between time and accountability (F = 6.069, p = 0.017, two-tailed),
with ?xed-pie bias revisions being higher for participants under process
accountability than those under outcome accountability ($176,683 and
$71,167 respectively). This provides further support that accountability has
?ow-on effects on joint costs.
24
As discussed earlier, this study was conducted in two locations. For the
control group, ?ve dyads are USA participants and 11 dyads are Australian
participants. We re-analyzed our data with location (USA and Australia) as
a covariate, and found that location was not a signi?cant covariate.
However, the inclusion of location as a covariate weakened our results,
such that the difference between the process accountability and no
accountability treatments became marginally signi?cant (?xed-pie bias
revisions: t = 1.786, p = 0.081; and joint costs: t = 1.731, p = 0.091, two-
tailed). The difference between the no accountability and outcome
accountability treatments remained insigni?cant. Therefore, our results
related to the control group need to be interpreted with caution.
104 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
participants differ between the two treatments. Of the four
types of negotiation behaviors, integrative information
provision and integrative questioning behaviors have the
largest differences between participants in the two
accountability treatments. Speci?cally, the proportion of
integrative information provision behavior is 10.22% for
participants in the process accountability treatment, but
only 4.26% for the outcome accountability participants
(t = 2.245, p = 0.033, two-tailed). Similarly, the process
accountability participants also engage in more integrative
questioning behaviors than their outcome accountability
counterparts (8.55% versus 2.85%, t = 2.170, p = 0.039,
two-tailed).
25
Both types of negotiation behaviors are gener-
ally employed as means of exchanging information. This re-
sult is consistent with our expectations that process
accountability participants are motivated to engage in nego-
tiation behaviors that allow them to better understand the
interests of both negotiation parties, in particular, through
active information exchange.
Conclusion and discussion
Customer–supplier negotiation is crucial to unlocking
the value in supply chains (Anderson & Dekker, 2009a,
2009b; Dekker, 2003). However, prior research (e.g., Drake
& Haka, 2008; Van den Abbeele et al., 2009) shows that the
provision of more sophisticated and relevant accounting
information alone is not suf?cient to ensure that negotia-
tors can use this information to identify mutually bene?-
cial opportunities, and to extract maximum values out of
the supply chain relationships. Rather, an organization’s
management control system also plays an important role
in improving supply chain performance (Free, 2007). In
this study, we examine how an important aspect of man-
agement control systems, namely accountability, affects
negotiators’ ability to attain lower joint costs. We ?nd that
negotiators held accountable for their negotiation pro-
cesses have higher ?xed-pie bias revisions and subse-
quently achieve lower joint costs than those held
accountable for their negotiation outcomes. Through our
PLS analysis, we provide evidence on how negotiators can
improve their performance. Our model indicates that pro-
cess accountability lowers joint costs to a greater extent
than outcome accountability through two paths. First, a
cognitive process path, where we ?nd that process
accountability is more effective than outcome accountabil-
ity in directly increasing negotiators’ ?xed-pie bias revi-
sions (thus reducing the ?xed-pie bias), and in doing so,
reduces joint costs. Second, we also ?nd a behavioral path
where process accountability is more effective than out-
come accountability in encouraging negotiators to employ
more integrative tactics. This allows negotiators to reduce
their ?xed-pie biases, and ultimately, lower joint costs. In
addition, we show that compared to the no accountability
control group, only process accountability is effective in
reducing the ?xed-pie bias and improving joint costs. De-
tailed analysis of the negotiation tactics obtained via tape
recordings of the negotiations further supports the ?ndings
that participants in the process accountability treatment
are more likely to exchange information through integra-
tive information provision and integrative questioning
behaviors than participants who are accountable for nego-
tiation outcomes.
The primary contribution of this study is to demon-
strate one way in which a management control system
can be designed to encourage negotiators to use cost infor-
mation to reach better joint outcomes. Prior studies such
as Drake and Haka (2008) and Van den Abbeele et al.
(2009) ?nd that providing negotiators with ABC informa-
tion (compared to volume-based cost information) only
improves negotiation outcomes when negotiators are
faced with external adversities, such as adverse market
conditions or a more powerful other party. In our study,
both negotiation parties received ABC information; our re-
sults show that organizations can actively improve joint
outcomes by imposing an appropriate type of accountabil-
ity (process accountability) on negotiators. We also show
that the ?xed-pie bias inhibits the improvement of negoti-
ation outcomes, and that process accountability can more
effectively mitigate this bias than outcome accountability.
Despite accountability being an important concept in
improving management control, there is little research
that addresses how different types of accountability
requirements affect managers’ decisions and behaviors.
Our study adds to this literature by showing that including
process accountability (instead of outcome accountability)
in the evaluation of negotiators’ performance has bene?ts
in terms of improved supply chain ef?ciencies. This has po-
tential implications for other circumstances such as
responsibility centre managers explaining their perfor-
mance in monthly divisional review meetings (see Rowe
et al., 2008). Our results suggest potential bene?ts of man-
agers justifying their processes rather than outcomes in
these meetings. In addition, contemporary performance
measurement systems (e.g., the balanced scorecard) re?ect
a process-based focus on intermediate initiatives, such as
improving relations with suppliers and developing new
technology capabilities; under process accountability,
managers could be rewarded based on their justi?cations
of how they have taken steps to improve these intermedi-
ate processes, rather than their justi?cations of the ?nal
division outcomes such as divisional revenue or budget
variances.
Our ?ndings build on Pinkley et al.’s (1995) argument
that poor negotiation outcomes are often the result of
negotiators not having accurate information about their
negotiation partner (information availability errors), as
well as negotiators’ tendency to distort or ignore useful
information about their negotiation partner (information
processing errors). We develop and test a path model to
show that process accountability is more effective than
outcome accountability in assisting negotiators to address
both errors: through the employment of more integrative
tactics negotiators can obtain more information about
their negotiation partner’s preferences, and at the same
time, the direct path from accountability to ?xed-pie bias
revisions suggests that process accountability also reduces
25
All other negotiation behaviors were not statistically different between
outcome accountability and process accountability (p > 0.10).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 105
negotiators’ information processing errors, independent of
the negotiation tactics employed.
Our study also contributes to the negotiation literature
by directly comparing the effect of process and outcome
accountability on the ?xed-pie bias and negotiation perfor-
mance. Unlike prior studies such as De Dreu et al. (2000),
which only examined process accountability, or Carnevale,
Pruitt, and Seilheimer (1981) and O’Connor (1997) which
only examined outcome accountability, this study directly
compares the effect of the two different types of account-
ability on negotiation behavior and outcomes. Importantly,
we show that the type of accountability in?uences both
the negotiators’ cognitive processes (?xed-pie bias revi-
sions) and behavior (negotiation tactics). As De Dreu and
Carnevale (2003) have noted, prior negotiation literature
often considers these processes in isolation. Our model
shows that the behavior of negotiators in?uences their
ability to overcome cognitive biases, and in doing so, we
demonstrate the importance of incorporating both factors
concurrently when conducting negotiation research.
Our study has a number of limitations, which also sug-
gest several opportunities for future research. First, the
cost functions of all service-related costs in our study were
linear. There is the opportunity for future research to
examine the role of accountability in situations where
non-linear cost functions make trade-off opportunities
more dif?cult to identify. Second, the negotiations in our
study occur during one period. In some circumstances,
the same parties may negotiate multiple rounds over a
number of years, and this may introduce additional consid-
erations, such as each party’s negotiation reputation (e.g.,
Tinsley, O’Connor, & Sullivan, 2002). Third, the current
study focuses on negotiations between two parties. How-
ever, customer–supplier negotiations sometimes involve
multiple parties (Anderson & Dekker, 2009a). Future re-
search can investigate whether process accountability con-
tinues to have a positive effect on negotiation outcomes in
a more complex negotiation environment. Fourth, another
design feature of our experiments is that the trade-off is-
sues for negotiators were symmetrical in value. However,
in some circumstances one negotiation party may have a
greater ability to contribute to expanding the negotiation
pie, and this may affect negotiation outcomes.
Future research could also examine the wider implica-
tions of our research for the design of management control
systems. This literature has long argued for the importance
of responsibility accounting, which implies holding manag-
ers accountable for their actions in order to improve effort
and performance outcomes (Rowe et al., 2008). While the
existing research on responsibility accounting tends to fo-
cus on the basis upon which managers should be held
accountable for (e.g., cost versus pro?t, individual versus
team outcomes), our study suggests that designers of man-
agement control systems also need to consider the type of
accountability imposed on managers. For example, while
organizations often require managers to justify perfor-
mance outcomes such as budget variances (Rowe et al.,
2008), it may be desirable if these managers are made to ac-
count for the processes leading to these outcomes, such as
the initiatives they had in place to reduce manufacturing
overhead costs. Seen in this light, process accountability
is similar to the notion of a balanced scorecard, which
encourages managers to be held responsible for the ‘causes’
of outcomes (i.e., the lead measures and initiatives). Re-
search could explore whether process accountability in
conjunction with a balanced scorecard systemcan have po-
sitive effects on other types of managerial tasks.
Acknowledgements
We gratefully acknowledge a research grant from the
Australian Research Council and the helpful comments
from the two anonymous reviewers, Shannon Anderson,
Henri Dekker, Andrea Drake, Gary Hecht, Anne Lillis,Habib
Mahama, Karen Sedatole, Michael Shields, and Yee Shih
Phua, as well as seminar participants at Monash Univer-
sity, Macquarie University,conference participants at the
2009 AAA Annual Meeting, the 2009 AFAANZ Conference,
the 2008 New Directions in Management Accounting Con-
ference and the 2008 GMARS Emerging Scholars Forum.
References
Ahrens, T. (1996). Styles of accountability. Accounting, Organizations and
Society, 21(2/3), 139–173.
Anderson, S. W., & Dekker, H. C. (2005). Management control for market
transactions: The relation between transaction characteristics,
incomplete contract design, and subsequent performance.
Management Science, 51(12), 1734–1752.
Anderson, S., & Dekker, H. (2009a). Strategic cost management in supply
chains, part 1: Structural cost management. Accounting Horizons,
23(2), 201–220.
Anderson, S., & Dekker, H. (2009b). Strategic cost management in supply
chains, part 2: Executional cost management. Accounting Horizons,
23(3), 289–305.
Asare, S. K., & Wright, A. M. (2003). A note on the interdependence
between hypothesis generation and information search in conducting
analytical procedures. Contemporary Accounting Research, 20(2),
235–251.
Bazerman, M. H. (2006). Judgment in managerial decision making.
Hoboken, NJ: J. Wiley.
Bonner, S. E. (2007). Judgment and decision making in accounting. New
York: Prentice Hall.
Brtek, M. D., & Motowidlo, S. J. (2002). Effects of procedure and outcome
accountability on interview validity. Journal of Applied Psychology,
87(1), 185–191.
Carnevale, P. J., Pruitt, D. G., & Seilheimer, S. D. (1981). Looking and
competing: Accountability and visual access in integrative bargaining.
Journal of Personality and Social Psychology, 40(1), 111–120.
Chang, L., Cheng, M., & Trotman, K. T. (2008). The effect of framing and
negotiation partner’s objective on judgments about negotiated
transfer prices. Accounting, Organizations and Society, 33(7–8),
704–717.
Chin, W. W. (1998). Issues and opinion on structural equation modeling.
MIS Quarterly, 22(1), vii–xvii.
Cohen, J. (1960). A coef?cient of agreement for nominal scales.
Educational and Psychological Measurement, 20(1), 37–46.
Cools, M., & Slagmulder, R. (2009). Tax-compliant transfer pricing and
responsibility accounting. Journal of Management Accounting Research,
21, 151–178.
De Dreu, C. K. W. (2003). Time pressure and closing of the mind in
negotiation. Organizational Behavior and Human Decision Processes,
91(2), 280–295.
De Dreu, C. K. W., & Carnevale, P. J. (2003). Motivational bases of
information processing and strategy in con?ict and negotiation. In
Advances in experimental social psychology (pp. 235–291). Academic
Press.
De Dreu, C. K. W., Beersma, B., Stroebe, K., & Euwenma, M. C. (2006).
Motivated information processing, strategic choice, and the quality of
negotiated agreement. Journal of Personality and Social Psychology,
90(6), 927–943.
De Dreu, C. K. W., Koole, S. L., & Steinel, W. (2000). Un?xing the ?xed pie:
A motivated information-processing approach to integrative
106 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
negotiation. Journal of Personality and Social Psychology, 79(6),
975–987.
Dekker, H. C. (2003). Value chain analysis in inter?rm relationships: A
?eld study. Management Accounting Research, 14(1), 1–23.
DeZoort, T., Harrison, P., & Taylor, M. (2006). Accountability and auditors’
materiality judgments: The effects of differential pressure strength on
conservatism, variability, and effort. Accounting, Organizations and
Society, 31(4–5), 373–390.
Drake, A. R., & Haka, S. F. (2008). Does ABC information exacerbate hold-
up problems in buyer-supplier negotiations? The Accounting Review,
83(1), 29–60.
Evans, J. H., III, Heiman-Hoffman, V. B. R., & Rau, S. E. (1994). The
accountability demand for information. Journal of Management
Accounting Research, 6(1), 24–42.
Free, C. (2007). Supply-chain accounting practices in the UK retail sector:
Enabling or coercing collaboration? Contemporary Accounting
Research, 24(3), 897–933.
Gelfand, M. J., & Realo, A. (1999). Individualism-collectivism and
accountability in intergroup negotiations. Journal of Applied
Psychology, 84(5), 721–736.
Gibbins, M., & Trotman, K. T. (2002). Audit review: Managers’
interpersonal expectations and conduct of the review. Contemporary
Accounting Research, 19(3), 411–444.
Glover, S. M. (1997). The in?uence of time pressure and accountability on
auditors’ processing of nondiagnostic information. Journal of
Accounting Research, 35(2), 213–226.
Halevy, N., Chou, E. Y., & Murnighan, J. K. (2012). Mind games: The mental
representation of con?ict. Journal of Personality and Social Psychology,
102(1), 132–148.
Hoffman, V. B., & Patton, J. M. (1997). Accountability, the dilution effect,
and conservatism in auditors’ fraud judgments. Journal of Accounting
Research, 35(2), 227–237.
Hurkens, K., van der Valk, W., & Wynstra, F. (2006). Total cost of
ownership in the services sector: A case study. Journal of Supply Chain
Management, 42(1), 27–37.
Hyder, E. B., Prietula, M. J., & Weingart, L. R. (2000). Getting to best:
Ef?ciency versus optimality in negotiation. Cognitive Science, 24(2),
169–204.
Kachelmeier, S. J., & Towry, K. L. (2002). Negotiated transfer pricing: Is
fairness easier said than done? The Accounting Review, 77(3), 571–593.
Kadous, K., & Sedor, L. M. (2004). The ef?cacy of third-party consultation
in preventing managerial escalation of commitment: The role of
mental representations. Contemporary Accounting Research, 21(1),
55–82.
Kaplan, R., & Anderson, S. (2007). Time-driven activity-based costing.
Boston: Harvard Business School Publishing.
Kennedy, J. (1993). Debiasing audit judgment with accountability: A
Framework and experimental results. Journal of Accounting Research,
31(2), 231–245.
Koonce, L., Anderson, U., & Marchant, G. (1995). Justi?cation of decisions
in auditing. Journal of Accounting Research, 33(2), 369–384.
Kruglanski, A. W. (1989). The psychology of being ‘‘right’’: The problem of
accuracy in social perception and cognition. Psychological Bulletin,
106(3), 395–409.
Kruglanski, A. W., & Webster, D. M. (1996). Motivated closing of the mind:
‘‘Seizing’’ and ‘‘freezing’’. Psychological Review, 103(2), 263–283.
Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of
accountability. Psychological Bulletin, 125(2), 255–275.
Lewicki, R. J., Barry, B., & Saunders, D. M. (2010). Negotiation (6th ed.).
New York: McGraw-Hill/Irwin.
Libby, T., Salterio, S. E., & Webb, A. (2004). The balanced scorecard: The
effects of assurance and process accountability on managerial
judgment. The Accounting Review, 79(4), 1075–1094.
Luft, J. L., & Libby, R. (1997). Pro?t comparisons, market prices and
managers’ judgments about negotiated transfer prices. The Accounting
Review, 72(2), 217–229.
Luft, J., & Shields, M. (2009). Psychology models of management.
Accounting Foundations and Trends
Ò
in Accounting, 4(3-4), 199–345.
Masschelein, S., Cardinaels, E., & Van den Abbeele, A. (2012). ABC
information, fairness perceptions, and inter?rm negotiations. The
Accounting Review, 87(3), 951–973.
Merchant, K. A., & Otley, D. T. (2006). A review of the literature on control
and accountability. In A. G. H. Christopher, S. Chapman, & D. S.
Michael (Eds.), Handbooks of management accounting research
(pp. 785–802). Elsevier.
Nelson, M., & Tan, H.-T. (2005). Judgment and decision making research in
auditing: A task, person, and interpersonal interaction perspective.
Auditing, 24, 41–71.
O’Connor, K. M. (1997). Groups and solos in context: The effects of
accountability on team negotiation. Organizational Behavior and
Human Decision Processes, 72(3), 384–407.
Olekalns, M., Smith, P. L., & Walsh, T. (1996). The process of negotiating:
Strategy and timing as predictors of outcomes. Organizational
Behavior and Human Decision Processes, 68(1), 68–77.
Peecher, M. E. (1996). The in?uence of auditors’ justi?cation processes on
their decisions: A cognitive model and experimental Evidence. Journal
of Accounting Research, 34(1), 125–140.
Pietroni, D., Van Kleef, G. A., De Dreu, C. K. W., & Pagliaro, S. (2008).
Emotions as strategic information: Effects of other’s emotional
expressions on ?xed-pie perception, demands, and integrative
behavior in negotiation. Journal of Experimental Social Psychology,
44(6), 1444–1454.
Pinkley, R. L., Grif?th, T. L., & Northcraft, G. B. (1995). ‘‘Fixed pie’’ a la
mode: Information availability, information processing, and the
negotiation of suboptimal agreements. Organizational Behavior and
Human Decision Processes, 62(1), 101–112.
Preacher, K. J., & Leonardelli, G. J. (2001). Calculation for the Sobel test: An
interactive calculation tool for mediation tests (Computer Software).
Accessed
March 2011.
Pruitt, D. G. (1981). In P. Warr (Ed.), Negotiation behavior. New York:
Academic Press.
Reinartz, W., Thomas, J. S., & Kumar, V. (2005). Balancing acquisition and
retention resources to maximize customer pro?tability. Journal of
Marketing, 69(1), 63–79.
Roberts, J. (1991). The possibilities of accountability. Accounting,
Organizations and Society, 16(4), 335–368.
Roodhooft, F., Hiel, G., Van den Abbeele, A., & Van Doveren, B. (2003).
Optimized sourcing strategies using total cost of ownership. Cost
Management, 17(4), 28–35.
Rosenzweig, E. (2009). A contingent view of e-collaboration and
performance in manufacturing. Journal of Operations Management,
27(6), 462–478.
Rowe, C., Birnberg, J. G., & Shields, M. D. (2008). Effects of organizational
process change on responsibility accounting and managers’
revelations of private knowledge. Accounting, Organizations and
Society, 33(2–3), 164–198.
Siegel-Jacobs, K., & Yates, J. F. (1996). Effects of procedural and outcome
accountability on judgment quality. Organizational Behavior and
Human Decision Processes, 65(1), 1–17.
Simonson, I., & Staw, B. M. (1992). Deescalation strategies: A comparison
of techniques for reducing commitment to losing courses of action.
Journal of Applied Psychology, 77(4), 419–426.
Sullivan, B. A., O’Connor, K. M., & Burris, E. R. (2006). Negotiator
con?dence. The impact of self-ef?cacy on tactics and outcomes.
Journal of Experimental Social Psychology, 42(5), 567–581.
Tan, H.-T. (1995). Effects of expectations, prior involvement, and review
awareness on memory for audit evidence and judgment. Journal of
Accounting Research, 33(1), 113–135.
Tan, H.-T., & Kao, A. (1999). Accountability effects on auditors’
performance. The in?uence of knowledge, problem-solving ability,
and task complexity. Journal of Accounting Research, 37(1), 209–223.
Thompson, L. L., & Hastie, R. (1990). Social perception in negotiation.
Organizational Behavior and Human Decision Processes, 47(1), 98–123.
Thompson, L. L., & Hrebec, D. (1996). Lose-lose agreements in
interdependent decision making. Psychological Bulletin, 120(3),
396–409.
Thompson, L. L., Wang, J., & Gunia, B. C. (2010). Negotiation. Annual
Review of Psychology, 61(1), 491–515.
Tinsley, C. H., O’Connor, K., & Sullivan, B. A. (2002). Tough guys ?nish last:
The perils of a distributive reputation. Organizational Behavior and
Human Decision Processes, 88, 621–642.
Trotman, K., Wright, A., & Wright, S. (2005). Auditor negotiations: An
examination of the ef?cacy of intervention methods. The Accounting
Review, 80(1), 349–367.
Van den Abbeele, A., Roodhooft, F., & Warlop, L. (2009). The effect of cost
information on buyer-supplier negotiations in different power
settings. Accounting, Organizations and Society, 34(2), 245–266.
Weingart, L. R., Hyder, E. B., & Prietula, M. J. (1996). Knowledge matters:
The effect of tactical descriptions on negotiation behavior and
outcome. Journal of Personality and Social Psychology, 70(6),
1205–1217.
Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and
Kenny: Myths and truths about mediation analysis. Journal of
Consumer Research, 37(2), 197–206.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 107
doc_584899443.pdf
Prior studies on customer–supplier negotiations (Drake & Haka, 2008; Van den Abbeele,
Roodhooft, & Warlop, 2009) find that negotiators who have access to relevant activitybased
cost information are not always able to use this information to improve joint outcomes.
Our study extends this literature by examining how the type of accountability (process
and outcome accountability) influences the extent to which negotiators can obtain
lower joint costs. We hypothesize and test a model that predicts that the type of accountability
affects negotiated outcomes through its effect on negotiators’ fixed-pie bias revisions
and the negotiation tactics they employ during customer-supplier negotiations.
Results from an experiment show that negotiators held accountable for their negotiation
processes are better able to reduce their fixed-pie biases and achieve lower joint costs compared
to those who are held accountable for their negotiation outcomes.
The effect of outcome and process accountability
on customer–supplier negotiations
Linda J. Chang, Mandy M. Cheng
?
, Ken T. Trotman
School of Accounting, The University of New South Wales, Sydney, NSW 2052, Australia
a b s t r a c t
Prior studies on customer–supplier negotiations (Drake & Haka, 2008; Van den Abbeele,
Roodhooft, & Warlop, 2009) ?nd that negotiators who have access to relevant activity-
based cost information are not always able to use this information to improve joint out-
comes. Our study extends this literature by examining how the type of accountability (pro-
cess and outcome accountability) in?uences the extent to which negotiators can obtain
lower joint costs. We hypothesize and test a model that predicts that the type of account-
ability affects negotiated outcomes through its effect on negotiators’ ?xed-pie bias revi-
sions and the negotiation tactics they employ during customer-supplier negotiations.
Results from an experiment show that negotiators held accountable for their negotiation
processes are better able to reduce their ?xed-pie biases and achieve lower joint costs com-
pared to those who are held accountable for their negotiation outcomes. Using rich data
based on taped negotiations, we demonstrate that the effect of accountability on joint costs
is indirect through its effect on negotiators’ choice of negotiation tactics and the extent to
which negotiators can reduce their ?xed-pie biases.
Ó 2013 Elsevier Ltd. All rights reserved.
Introduction
Negotiations between customers and suppliers form a
crucial part of collaborative supply chain relationships
(Anderson & Dekker, 2005; Drake & Haka, 2008). Activ-
ity-based costing (ABC) techniques, such as customer prof-
itability analysis, enable negotiators to identify activities
that drive supply chain costs, and direct their attention
away from price towards quantifying and managing the to-
tal cost of the customer–supplier relationship (Anderson &
Dekker, 2009a, 2009b; Van den Abbeele, Roodhooft, &
Warlop, 2009). Recent research has established that nego-
tiators who share ABC information are more likely to reach
agreements with higher joint outcomes (e.g., lower supply
chain costs) than those who share volume-based cost
information (Drake & Haka, 2008; Van den Abbeele et al.,
2009). However, these studies also show that negotiators
are reluctant to share ABC information, unless they are
confronted with adverse conditions such as market pres-
sures (Drake & Haka, 2008) or more powerful negotiation
partners (Van den Abbeele et al., 2009). Providing ABC
information to negotiators with bargaining power also
potentially increases their tendency to use confrontational
and competitive negotiation tactics, further reducing joint
outcomes (Van den Abbeele et al., 2009).
The organizational and psychology literatures argue
that a key factor that contributes to negotiators’ unwilling-
ness to work collaboratively is the ‘?xed-pie bias’. The
?xed-pie bias refers to negotiators’ tendencies to believe
that all negotiations are a ?xed pie; that is, the attainment
of one party’s goals precludes the other party from achiev-
ing their goals (Bazerman, 2006; De Dreu, Koole, & Steinel,
2000; Pinkley, Grif?th, & Northcraft, 1995; Thompson &
Hastie, 1990). Such a belief results in negotiators holding
incorrect assumptions about their negotiation partner’s
priorities and preferences. Fixed-pie bias is potentially det-
rimental because it causes negotiators to focus on compet-
ing with each other rather than working collaboratively,
0361-3682/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.aos.2012.12.002
?
Corresponding author. Tel.: +61 2 9385 6343; fax: +61 2 9385 5925.
E-mail addresses: [email protected] (L.J. Chang), m.chen-
[email protected] (M.M. Cheng), [email protected] (K.T. Trotman).
Accounting, Organizations and Society 38 (2013) 93–107
Contents lists available at SciVerse ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er. com/ l ocat e/ aos
thereby lowering the likelihood of negotiators identifying
trade-off opportunities and reaching better joint outcomes
(De Dreu et al., 2000). Despite the central role of the ?xed-
pie bias in understanding negotiation performance, the
accounting literature has not investigated mechanisms
that can help mitigate the negative effects of the ?xed-
pie bias when using accounting data.
In this study, we extend recent negotiation literature by
investigating how one aspect of management control,
namely type of accountability, affects negotiators’ ?xed-
pie bias revisions
1
and performance when they are provided
with ABC information during customer–supplier negotia-
tions. The imposition of accountability underlies the princi-
ple of responsibility accounting, which calls for managers to
be held responsible for their actions and performance out-
comes (Bonner, 2007; Rowe, Birnberg, & Shields, 2008).
Here, we compare two types of accountability, process ver-
sus outcome accountability, in an experiment where cus-
tomer–supplier dyads engage in a single period
negotiation over multiple issues, and have the opportunity
to improve joint outcomes by taking advantage of symmet-
ric trade-off opportunities. In our experimental setting, min-
imizing joint costs (i.e., total costs incurred by the customer
and the supplier in each dyad) is considered desirable; both
negotiating parties within a dyad have equal bargaining
power and are faced with the same accountability pressure
– either outcome or process accountability. Our results show
that dyads which are held accountable for their negotiation
processes achieve lower joint costs compared to dyads
which are held accountable for their negotiation outcomes.
In addition, using rich data from taped face-to-face negotia-
tions, we demonstrate that process accountability is more
effective than outcome accountability in reducing joint costs
through two routes: a cognitive route where process
accountability directly reduces negotiators’ ?xed-pie biases,
and a behavioral route where process accountability encour-
ages negotiators to use more integrative (value-creating)
negotiation tactics, which in turn, reduces their ?xed-pie
biases and ultimately reduces joint costs. Demonstrating
the bene?ts of process over outcome accountability is
important because many performance evaluation systems
implicitly emphasize outcome accountability, such as
requiring managers to justify budget variances (e.g., Rowe
et al., 2008), but do not include the requirement for manag-
ers to justify the processes they undertake to arrive at these
outcomes.
‘Accountability’ refers to ‘‘the implicit or explicit expec-
tation that one may be called to justify one’s beliefs, feel-
ings and actions to others’’ (Lerner & Tetlock, 1999, 255;
emphasis added). Research on accountability contends that
this justi?cation requirement causes individuals to engage
in more effortful processing of information before they
make decisions or judgments (Bonner, 2007; Kennedy,
1993; Lerner & Tetlock, 1999; Peecher, 1996; Tan & Kao,
1999). Accountability literature identi?es two distinct
types of accountability: process accountability and out-
come accountability (Lerner & Tetlock, 1999). Process
accountability requires individuals to justify their decision
making processes (e.g., to justify how a capital investment
was selected or how negotiators reached an agreement),
whereas outcome accountability requires individuals to
justify their decision outcomes (e.g., to justify the suc-
cess/failure of an investment or the ?nal negotiation
agreement).
In the management accounting literature, accountabil-
ity is seen as a key issue in designing management control
systems (e.g., Ahrens, 1996; Evans, Heiman-Hoffman, &
Rau, 1994; Merchant & Otley, 2006; Roberts, 1991). How-
ever, the accountability literature relating to management
control systems design typically does not distinguish be-
tween process accountability and outcome accountability.
Instead, this literature examines how management control
systems and accounting information are used as a conduit
for managers to negotiate and discharge their accountabil-
ities to their organizational constituents. While the related
literature on responsibility accounting focuses on the eval-
uation of managers’ performance based on their responsi-
bilities (Cools & Slagmulder, 2009; Rowe et al., 2008), it
typically does not directly examine the cognitive and
behavioral effects of justi?cation – a key feature of
accountability. Two empirical studies that directly exam-
ine the effect of accountability on managerial judgments
are Libby, Salterio, and Webb (2004) and Kadous and Sedor
(2004). Libby et al. manipulated process accountability and
showed that process accountability improved managerial
judgments compared to no accountability. Kadous and Se-
dor found that the requirement to justify a decision in-
creases the likelihood of consultants recommending the
escalation of a project. Kadous and Sedor’s operationaliza-
tion of accountability includes both outcome and process
elements.
2
Thus, the management accounting literature
has not examined the relative merits of process versus out-
come accountability in terms of managerial judgments.
In the psychology literature, a number of studies have
found that process accountability improves judgments
more than outcome accountability in static judgment tasks
such as escalation of commitment (Simonson & Staw,
1992), calibration and probability judgments (Siegel-Ja-
cobs & Yates, 1996), and consistency in interview judg-
ments (Brtek & Motowidlo, 2002). However, no prior
research has compared the effect of process and outcome
accountability in a negotiation context.
3
Unlike static judg-
ment tasks, negotiations are more dynamic and cognitively
complex. During the negotiation process, negotiators need
to make multiple judgments at various stages of the negoti-
ation about their own interests (i.e., their priorities and pref-
erences for the negotiation issues), their counterpart’s
interests, and ?nd a way to integrate these interests to reach
1
Consistent with prior literature (e.g., De Dreu et al., 2000), we de?ne
?xed-pie bias revisions as the difference between the negotiators’ ?xed-pie
biases before and after the negotiation in each negotiation dyad.
2
Kadous and Sedor (2004) operationalized accountability by telling
participants that ‘‘. . . the Board will require you to justify your recommen-
dations in a convincing manner . . .’’. Thus participants’ justi?cation could
involve an explanation of their decision outcomes or their decision
processes or both.
3
In fact, most studies investigating the impact of accountability on
negotiations do not make a distinction between outcome accountability
and process accountability. See De Dreu et al. (2000) and De Dreu, Beersma,
Stroebe, and Euwenma (2006) for exceptions, both of which study the effect
of process accountability versus no accountability on negotiations.
94 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
an agreement. Furthermore, negotiations involve complex
social interactions; negotiators need to employ appropriate
negotiation tactics and respond to the negotiation tactics
of the other party.
Our study makes a number of important contributions.
First, our primary contribution is to show that the imposi-
tion of process accountability (compared to outcome
accountability) on negotiators improves joint costs in cus-
tomer–supplier negotiations. We extend prior accounting
research (Drake & Haka, 2008; Masschelein, Cardinaels, &
Van den Abbeele, 2012; Van den Abbeele et al., 2009) by
investigating how the choice of management control de-
sign, namely, the imposition of process rather than out-
come accountability via the performance evaluation
process, enables negotiators to use ABC information more
effectively to improve their negotiation performance.
While prior studies show that providing negotiators with
ABC information potentially enables negotiators to identify
cost saving opportunities, they also ?nd that negotiators
are sometimes reluctant to share this information and fail
to reach the maximum level of joint outcomes. Therefore,
to maximize supply chain ef?ciencies, organizations need
to do more than providing negotiators with useful infor-
mation. One approach is to improve an organization’s
management control system (Free, 2007). Our results show
that by explicitly incorporating process rather than out-
come accountability in the performance evaluation pro-
cess, accountants can design management control
systems to encourage negotiators to use integrative nego-
tiation tactics such as providing information to, and ques-
tioning the priorities of, the other negotiation party, and
ultimately, lower supply chain joint costs.
Second, Luft and Shields (2009) argue that individuals’
cognitive biases and information search processes in?u-
ence how they can effectively use management accounting
information to make decisions. Our study highlights the
role of the ?xed-pie bias in customer–supplier negotiations,
a variable that has not previously been examined in the
accounting literature. Importantly, our study considers
the differential effect of process accountability and out-
come accountability on negotiating managers’ revisions of
?xed-pie biases. While the psychology literature has shown
that process accountability can reduce the ?xed-pie bias
compared to no accountability (De Dreu et al., 2000), there
has been no research on the effect of outcome accountabil-
ity on the ?xed-pie bias. A direct comparison between the
two types of accountability is important, as our results
demonstrate that the imposition of accountability is only
bene?cial if organizations speci?cally implement process
accountability rather than outcome accountability.
Third, we develop and test a model that explains how
process accountability enables negotiators to reach more
favorable joint outcomes. Our model shows that process
accountability is more effective in lowering joint costs than
outcome accountability by causing negotiators to revise
their ?xed-pie biases, both directly and indirectly through
negotiators’ greater use of integrative negotiation tactics.
Our study demonstrates that cognitive factors (e.g., the
?xed-pie bias) and behavioral factors (e.g., the use of dif-
ferent tactics during negotiation) are not independent,
even though these factors tend to be examined in isolation
of each other in the negotiation literature (De Dreu & Car-
nevale, 2003).
Hypotheses development
ABC techniques, such as analyzing transactions related
to individual customers or suppliers, are frequently used
by organizations to gain insights about the costs associated
with trading with their customers or suppliers (Kaplan &
Anderson, 2007; Reinartz, Thomas, & Kumar, 2005; Roo-
dhooft, Hiel, Van den Abbeele, & Van Doveren, 2003). How-
ever, providing relevant ABC information to negotiators
does not necessarily mean that they will use this informa-
tion effectively to improve negotiation outcomes. Drake
and Haka (2008) ?nd that negotiators are reluctant to share
ABC information, even though those who do so achieve
higher trade ef?ciencies. This reluctance is somewhat re-
duced by the prospect of potential losses due to market
pressure; nonetheless, even under market pressures, nego-
tiators who received ABC information could only capture
about the same level of available trade ef?ciencies in the
market as those who were provided with volume-based
information. Van den Abbeele et al. (2009) ?nd that while
customers are more willing to share ABC information when
negotiating with a more powerful supplier, they are less
likely to do so when there is no power difference between
negotiators. Furthermore, when both parties have equal
power, customers use more distributive (value-claiming)
negotiation tactics and are less likely to use a problem-solv-
ing approach during negotiations, leading to a lower level of
joint outcomes compared to negotiators who have differen-
tial bargaining power. In addition, Masschelein et al. (2012)
?nd that the provision of ABC information does not facili-
tate collaborative negotiations when buyers use ABC infor-
mation to demonstrate that sellers are responsible for
supply chain inef?ciencies. Together, these studies show
that the provision of ABC information only translates to bet-
ter negotiation outcomes if both parties of the negotiation
work collaboratively and make effective use of the ABC
information to identify mutually bene?cial opportunities
to improve supply chain ef?ciencies.
In the present study, we argue that a cognitive bias that
prevents negotiators from working collaboratively is the
negotiators’ ?xed-pie biases (Pinkley et al., 1995; Thomp-
son & Hastie, 1990). The ?xed-pie bias refers to negotia-
tors’ tendency to project their own values and
perspectives on to their negotiation partners, resulting in
an assumption that their partner’s interests are directly
opposite to their own (Lewicki, Barry, & Saunders, 2010;
Pinkley et al., 1995; Thompson & Hastie, 1990). Although
this assumption is correct in negotiations involving a sin-
gle issue, it is often not appropriate in negotiations involv-
ing multiple issues because negotiators typically derive
different utilities from different issues, such that an overall
increase in joint value becomes possible (i.e., ‘integrative
negotiations’) (Lewicki et al., 2010). Assumptions based
on the ?xed-pie beliefs in integrative negotiations are of-
ten erroneous, and the failure to revise them will prevent
negotiators from making mutually bene?cial trade-offs to
‘expand the pie’. We propose that in integrative
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 95
negotiations, process accountability (as opposed to out-
come accountability) will improve negotiators’ ?xed-pie
bias revisions both directly and indirectly through encour-
aging negotiators to use more integrative tactics. Fixed-pie
bias revisions, in turn, mediate the positive effect of pro-
cess accountability on joint costs. These proposed relation-
ships, summarized in Fig. 1, are discussed in detail in the
following sections.
The prevalence of the ?xed-pie bias and the failure of
negotiators to overcome the bias are well-documented in
the psychology literature (De Dreu, 2003; De Dreu et al.,
2000; Pinkley et al., 1995; Thompson &Hastie, 1990). These
studies showthat for integrative negotiations, the ?xed-pie
bias prevents negotiators from reaching better joint out-
comes, and causes negotiators to distort or ignore informa-
tion provided to them about their negotiation partner’s
interests (including the importance the negotiation partner
places on various options within a negotiation issue and
across different negotiation issues). In addition, the man-
agement accounting literature on transfer pricing ?nds that
negotiators do not have accurate understanding of the
interests of their negotiation partners (Chang, Cheng, &
Trotman, 2008; Kachelmeier & Towry, 2002; Luft & Libby,
1997). In the context of negotiations between a customer
and a supplier, the ?xed-pie bias potentially lowers the
bene?ts that may be realized. Speci?cally, the supply chain
environment is often characterized by negotiators bargain-
ing over multiple issues with varying cost implications for
each party (Dekker, 2003; Free, 2007). For example, a sup-
plier may have a comparatively lower cost of capital than
their customers thus allowing them to negotiate a trade-
off between offering a smaller sales discount, in return for
extending the payment terms (Hurkens, van der Valk, &
Wynstra, 2006). In such situations, negotiators’ ability to
revise their ?xed-pie biases is critical in lowering joint
costs. We therefore propose the following hypothesis:
H1. Joint costs are lower for negotiators with larger ?xed-
pie bias revisions.
Next, we propose that one way organizations can
encourage managers to reduce their ?xed-pie biases is to
impose accountability requirements on negotiators, and
more speci?cally, process (rather than outcome) account-
ability. We argue that accountability has both a direct ef-
fect (see H2a) and an indirect effect via its in?uence on
negotiators’ use of negotiation tactics which in turn in-
creases negotiators’ ?xed-pie bias revisions (refer to H2b
and Fig. 1). First, we consider the direct effect. Accountabil-
ity results in an expectation that individuals may be re-
quired to justify their judgments and decisions (Lerner &
Tetlock, 1999). However, because process and outcome
accountability require different types of justi?cation, they
have different effects on individuals’ ability to revise their
?xed-pie biases. Process accountability, which requires
individuals to justify their decision processes, induces an
accuracy (as opposed to directional) goal and motivates
individuals to hold a correct and accurate view of the
world (De Dreu et al., 2006; Kruglanski, 1989; Kruglanski
& Webster, 1996). In doing so, process accountability
encourages individuals to try to understand the environ-
ment surrounding their decisions rather than merely
attaining better outcomes. In addition, by requiring indi-
viduals to justify their decision processes, process account-
ability focuses their attention on ensuring that their
decision processes are appropriate and defendable. This
leads individuals to make a greater effort and be more sys-
tematic in acquiring and processing information about
their environment (Brtek & Motowidlo, 2002; Lerner & Tet-
lock, 1999; Simonson & Staw, 1992).
De Dreu et al. (2000) ?nd that negotiators who are held
accountable for their negotiation processes are more likely
to revise their ?xed-pie biases and identify trade-off
opportunities than those who are not held accountable.
In a series of experiments, De Dreu et al. (2006) ?nd that
process accountability improves negotiators’ ability to uti-
lize cooperative tactical knowledge to reach higher joint
outcomes. They also ?nd that, in the absence of process
accountability, providing negotiators with cooperative tac-
tics did not enable them to perform any better than those
who are given competitive tactics.
4
Together, these studies
Integrative
Tactics
Employed
Fixed-Pie Bias
Revisions
Joint Costs
Accountability
H2a (+)
H2b (+)
H3 (+)
H1 (-)
H2b (+)
H3 (+)
Fig. 1. Causal link between accountability and joint costs. Figure depicts our theoretical model. H1 predicts that negotiators’ ?xed-pie bias revisions reduce
joint costs. H2a predicts that process accountability (compared to outcome accountability) directly increases negotiators’ ?xed-pie bias revisions, while H2b
predicts an indirect effect of process accountability (compared to outcome accountability) on negotiators’ ?xed-pie bias revisions mediated by their use of
integrative tactics employed. Finally, H3 predicts that ?xed-pie bias revisions mediate the effect of the type of accountability on joint costs.
4
The cooperative tactics provided to participants in their study were
similar to integrative tactics and competitive tactics provided to partici-
pants were similar to distributive tactics (De Dreu et al., 2006).
96 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
indicate that process accountability is effective in encourag-
ing individuals to revise their ?xed-pie biases and reach
higher joint outcomes.
While the effect (but not how the effect occurs) of pro-
cess accountability is predictable from psychology, the sit-
uation for outcome accountability is far less clear. Unlike
process accountability, outcome accountability requires
individuals to justify their decision outcomes. First, a sig-
ni?cant amount of accountability literature in auditing,
while not addressing the ?xed-pie bias or negotiation tac-
tics, does indicate that accountability leads to greater vig-
ilance and increases in various forms of performance
(Asare & Wright, 2003; DeZoort, Harrison, & Taylor, 2006;
Glover, 1997; Hoffman & Patton, 1997; Kennedy, 1993;
Koonce, Anderson, & Marchant, 1995; Nelson & Tan,
2005; Tan, 1995; Tan & Kao, 1999). While this literature
does not use the term ‘outcome accountability’ it does ap-
pear to include a form of outcome accountability, or a com-
bination of outcome and process accountability.
5
In
contrast, De Dreu et al. (2000) suggest that outcome
accountability induces directional motivation, which focuses
individuals’ attention on acquiring and processing informa-
tion that enables them to reach a desired outcome. While
De Dreu et al. (2000) do not test the effect of outcome
accountability, they argue that outcome accountability will
induce negotiators to process information in accordance
with their pre-disposed conclusion (that negotiations are
competitive and represent a ?xed-pie). Research in non-
negotiation judgments in psychology shows that outcome
accountability leads to poorer judgments and decisions
(Brtek & Motowidlo, 2002; Simonson & Staw, 1992). For
example, Simonson and Staw (1992) ?nd that outcome
accountability heightened individuals’ concern for justifying
project outcomes at the expense of applying correct decision
strategies, and is therefore less effective than process
accountability in overcoming individuals’ tendency to esca-
late commitment.
As noted earlier, past psychology studies on outcome
accountability, however, are based on static judgment
tasks, rather than in dynamic decision contexts, such as
negotiations which involve multiple judgments, complex
social interactions and the use of, and response to, negoti-
ation tactics. In negotiations, participants are aware that
obtaining a successful outcome depends on a joint agree-
ment being reached. Thus, the motivation to reach a desir-
able outcome may nonetheless encourage negotiators to
make attempts to ?nd out the other negotiation party’s
interests, and therefore lead to the identi?cation of trade-
off opportunities. However, we posit that despite their ef-
fort to do so, negotiators under outcome accountability
will ?nd it dif?cult to revise their ?xed-pie biases, because
their focus on improving negotiation outcomes for them-
selves will reinforce their existing views that the negotia-
tions represent a competitive situation (Thompson &
Hrebec, 1996).
Based on the preceding discussion, we propose that
process accountability motivates negotiators to process
information about how their negotiation partners’ inter-
ests are different fromtheir own, thereby enabling a reduc-
tion of their ?xed-pie biases. On the other hand, outcome
accountability will cause individuals to focus on meeting
their own interests and therefore does not facilitate
?xed-pie bias revisions. Thus, we predict the following
(see Fig. 1):
H2a. Process accountability (compared to outcome
accountability) directly improves negotiators’ ?xed-pie
bias revisions.
We further propose that accountability also indirectly
in?uences ?xed-pie bias revisions through its effect on
negotiators’ use of negotiation tactics. The negotiation lit-
erature suggests that negotiation outcomes are in?uenced
by both negotiators’ cognitive processes and the social pro-
cesses during the negotiation, in particular, the tactics em-
ployed by different parties (Thompson, Wang, & Gunia,
2010). Negotiation tactics are usually classi?ed into two
basic types: integrative tactics and distributive tactics
(Lewicki et al., 2010). Integrative negotiation tactics are
aimed at addressing the objectives of both parties by gain-
ing a better understanding of the underlying interests of
both parties and trading off less important issues for more
important issues (Weingart, Hyder, & Prietula, 1996; Lew-
icki et al., 2010). Distributive negotiation tactics, on the
other hand, are aimed at obtaining the maximum conces-
sions on each of the issues from the other party, rather
than understanding his/her interests (Hyder, Prietula, &
Weingart, 2000; Trotman, Wright, & Wright, 2005; Wein-
gart et al., 1996). While prior negotiation literature has
established the important role negotiation tactics play in
enhancing joint outcomes (e.g., Lewicki et al., 2010; Ole-
kalns, Smith, & Walsh, 1996), the literature has not exam-
ined how different types of accountability can affect
negotiators’ choice of integrative versus distributive
tactics.
Given that process accountability will motivate negoti-
ators to gain a better understanding of their negotiation
processes and their partner’s priorities, we predict that
managers who are held accountable for their negotiation
processes will employ more integrative tactics (and fewer
distributive tactics) than those who are held accountable
for their negotiation outcomes. Prior literature also ?nds
that negotiators who employ integrative tactics are also
more likely than those who employ distributive tactics to
acquire information from and share information with the
other party (Hyder et al., 2000; Olekalns et al., 1996). This
information acquisition and sharing allow negotiators to
better understand both parties’ interests. Thus, we predict
that the bene?cial effect process accountability has on the
use of integrative tactics (compared to outcome account-
ability) will indirectly increase ?xed-pie bias revisions.
Taken together, we hypothesize:
5
For a review of research on accountability in the auditing literature,
refer to DeZoort et al. (2006). Overall, this literature uses a variety of ways
to operationalize the justi?cation requirement, such as imposing audit
review pressures or requiring auditors to write justi?cations for audit
judgments. Although these manipulations appear more in line with
outcome than process accountability, they could be perceived to incorpo-
rate both, as the review process usually considers both judgments and
processes used to form those judgments (Gibbins & Trotman, 2002).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 97
H2b. Process accountability (compared to outcome
accountability) indirectly improves negotiators’ ?xed-pie
bias revisions mediated by negotiators’ use of integrative
tactics.
Finally, as argued earlier, the ?xed-pie bias is a key
obstacle to reaching a higher level of joint outcomes in
negotiations (Lewicki et al., 2010). Hence, we predict that
the two types of accountability will also indirectly affect
the ability of negotiators to reach agreements resulting in
lower total joint costs. We therefore predict that the effect
of the type of accountability on joint costs is mediated by
?xed-pie bias revisions:
H3. Fixed-pie bias revisions mediate the effect of the type
of accountability on joint costs.
Although prior research shows that the use of more
integrative tactics will lead to higher joint outcomes (Hy-
der et al., 2000; Van den Abbeele et al., 2009; Weingart
et al., 1996), we suggest that this effect can be attributed
to integrative negotiation tactics reducing negotiators’
?xed-pie biases. While the use of integrative negotiation
tactics facilitates the exchange of information, De Dreu
et al. (2000) argue that negotiators are unlikely to reach
agreements with higher joint outcomes without using this
information to revise their ?xed-pie biases. We therefore
do not predict a direct effect from negotiation tactics to
joint costs.
Research methods
Participants
Sixty-six participants (33 negotiation dyads) volun-
teered to participate in this experiment. All participants
were enrolled in either a MBA degree or Master of Com-
merce degree. Approximately half of the participants were
enrolled in a large USA university while the other half was
enrolled in a large Australian university.
6
Three negotiation
dyads failed one or more post experiment manipulation
tests and were later excluded from the analysis, resulting
in usable responses from 30 negotiation dyads. The cell sizes
were 15 dyads for both treatment groups.
7
Research design
Participants were randomly assigned into the role of
either the supplier or the customer, and then were formed
into dyads to negotiate a sales contract. Two types of
accountability were manipulated: outcome accountability
and process accountability. Each dyad was randomly as-
signed to one of these treatments (i.e., both negotiators
within a dyad received the same accountability treatment).
All participants received a ?xed compensation of US$20
and ‘bonus points’ that gave them chances to win a ‘grand
prize’ worth US$100.
8
The bonus points calculation had two
components: ?rst, all participants received bonus points
based on their ?nal negotiation outcomes using a ‘piece rate’
system, where one bonus point was awarded for every
$2000, or part thereof, for costs that fell below the maxi-
mum cost an individual negotiator could incur in the ?nal
agreement. Thus, the lower the cost, the more bonus points
they would earn.
9
We manipulate accountability using the second compo-
nent of the bonus point calculation. Lerner and Tetlock
(1999) de?ne accountability as the expectation that indi-
viduals may be called upon to justify their performance,
action or judgments to others, and that they will be re-
warded (penalized) if they provide satisfactory (unsatisfac-
tory) justi?cations. Based on this de?nition, participants in
the ‘outcome accountability’ condition were required to
write a report to justify their negotiation outcomes, and
were told that they would receive additional bonus points
based on the quality of this report.
10
They were told that
the negotiation outcome report needed to address both the
negotiation result and whether the outcome was considered
acceptable for their company. They were also told that the
negotiation outcome report would be graded by one of the
authors and the participant could receive between 0 and
40 bonus points depending on whether their report was
judged to be unacceptable, needs improvement, competent,
very good or outstanding.
11
For participants in the ‘process
accountability’ condition, they were required to write a re-
port based on their negotiation processes rather than nego-
tiation outcome. In particular, they were told that their
report needed to address the strategies and procedures em-
ployed during their negotiations. Similar to those in the out-
come accountability condition, participants in the process
accountability condition were also told of the method of
grading their report and that they could also receive be-
tween 0 and 40 bonus points depending on the quality of
their reports.
6
In both locations, participants were randomly assigned into one of the
accountability treatments. Nine out of the 15 dyads in the process
accountability treatment and eight out of the 15 dyads in the outcome
accountability treatments were USA participants. To investigate whether
location affects our results, we re-analyzed our data with location (USA
versus Australia) as a co-variate and ?xed-pie bias revisions, proportion of
negotiation tactics, and joint costs as dependent variables. We found that
location was not a signi?cant co-variate and the differences between
process accountability and outcome accountability remained signi?cant for
all three dependent variables after controlling for location.
7
As discussed subsequently, we also included a no accountability control
group. This involves 16 dyads with participants randomly assigned to
complete the same task as the two treatment groups, but without the
accountability requirement.
8
For USA participants the compensation was paid in cash and for
Australian participants, the compensation was in the form of movie
coupons valued at AU$25 (for the ?xed compensation) and a $130 gift card
(for the grand prize). The use of non-cash rewards for Australian partic-
ipants was due to university administrative restrictions. The value of
compensation was approximately matched between the two countries
based on the exchange rate during the experiments.
9
Consistent with prior psychology research on accountability (e.g., De
Dreu et al., 2000), we tie participants’ compensation to their negotiation
outcomes to ensure that participants take the negotiation task seriously.
This component of bonus point calculation does not affect our account-
ability manipulation because it is not linked to a justi?cation requirement.
10
The use of ‘report writing’ to manipulate accountability is consistent
with a number of earlier studies including O’Connor (1997), Gelfand and
Realo (1999) and Libby et al. (2004).
11
The averages of bonus points awarded across the conditions were 34.4
and 33.5 for outcome accountability and process accountability
respectively.
98 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
Experimental task
Participants assume the role of either the sales manager
of Glass Company (the supplier) or the purchasing man-
ager of Furniture Company (the customer). Glass Company
manufactures a wide range of glass-related products,
including tempered glass panels which are used as inputs
for Furniture Company, a company that manufactures a
wide range of household furniture products including glass
top tables. To control for the effect of bargaining power,
participants are told that both companies are similar in
size and that neither of them have an alternative trading
partner.
Participants are informed of instructions from their
respective divisional managers to negotiate a sales con-
tract with the intent of establishing a long-term strategic
relationship with the other party. The sales contract in-
volves the sales of a ?xed number of 2800 tempered glass
panels at a predetermined sales price $250 each (i.e., sales
quantity and price are not negotiable). Similar to previous
studies such as Drake and Haka (2008), the negotiators are
required to negotiate six after-sales services (each with
three levels) to be provided by the supplier to the cus-
tomer. The participants are also told that each of these ser-
vices will have potential additional cost implications for
both parties. For example, the customer has three different
factory locations; the supplier’s costs will increase and the
customer’s costs will decrease with each delivery location
serviced by the supplier. The aim of the negotiation is to
reach an agreement on the level of services associated with
each of the six issues for their respective companies. Par-
ticipants are provided with a table describing the six issues
and their cost implications for their own company but not
for the other company (Table 1 summarizes the suppliers’
and the customers’ cost structures).
Consistent with earlier negotiation studies in the psy-
chology literature on the ?xed-pie bias (e.g., Pinkley
et al., 1995), we include four trade-off issues and two
zero-sum issues. Two examples of trade-off issues are
‘marketing support’ and ‘design support’ (Table 1 shows
that while the customer ?nds ‘marketing support’ more
costly than ‘design support’, the opposite is true for the
supplier). Delivery location and glass cutting are the other
two trade-off issues. The two zero-sum issues are inspec-
tion sample size and waste collection frequency. For exam-
ple, with the zero-sum issue ‘inspection sample size’,
Table 1 shows that the cost function of ‘inspection sample
size’ is exactly the same for both the supplier and the cus-
tomer. This design allows us to test the extent to which
participants can revise their ?xed-pie biases by identifying
trade-off potentials, and thus reach the Pareto-ef?cient
solution.
12
The possible joint costs range from $190,000
Table 1
Customer’s and supplier’s cost structures.
Issues to negotiate Options Supplier’s cost per year ($) Customer’s cost per year ($)
Inspection sample size
a
1% 9000 27,000
2% 18,000 18,000
3% 27,000 9000
Marketing support
b
Low 4500 36,000
Medium 9000 24,000
High 13,500 12,000
Delivery locations
b
1 Factory 15,000 16,500
2 Factories 30,000 11,000
3 Factories 45,000 5500
Waste collection frequency
a
1 per quarter 10,000 30,000
2 per quarter 20,000 20,000
3 per quarter 30,000 10,000
Design support
b
Basic support 12,000 13,500
CAD only 24,000 9000
CAD & CAM 36,000 4500
Glass cutting
b
Standard Cut 5500 45,000
Customized Cut 11,000 30,000
Full Finished Cut 16,500 15,000
Note: Participants only receive cost schedules relative to their roles (e.g., suppliers only receive information about their own costs) and within each issue,
participants are provided with a brief description. For example, in relation to inspection sample size, suppliers are told that ‘‘Your cost increases with larger
sample sizes, i.e. more extensive testing for Furniture Co. ($9000 per%)’’. Customers on the other hand, are told that ‘‘Your cost decreases with larger sample
sizes, i.e. the more extensive testing from Glass Co ($9000 per%).’’
a
Zero-sum issues (i.e., where the cost functions are the same for both the supplier and the customer).
b
Trade-off issues (i.e., where the priorities are different between the supplier and the customer.) The lowest level of joint costs for both parties is
calculated as follows: Zero-sum issues: $36,000 [inspection sample size] + $40,000 [waste collection]; Trade-off issues: $(13,500 + 12,000) [high marketing
support] + $(15,000 + 16,500) [one delivery location] + $(12,000 + 13,500) [basic design support] + $(16,500 + 15,000) [full ?nished glass cutting]. Total joint
costs = $190,000.
12
The trade-off issues are designed to be symmetrical in value, such that
the suppliers’ most costly issues (design support at $12,000 per level and
delivery locations at $15,000 per level) are also the customers’ least costly
issues ($4500 and $5500 per level respectively). Similarly, the customers’
most costly issues (marketing support at $12,000 per level and glass cutting
at $15,000 per level) are the suppliers’ least costly issues ($4500 and $5500
per level respectively) (see Table 1 for more details). Consistent with earlier
studies (e.g., Thompson & Hastie, 1990), we made this design choice to
allow both negotiating parties to have the same opportunities to contribute
to increasing the joint outcomes.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 99
(lowest possible joint costs) to $258,000 (highest possible
joint costs). From the individual negotiator’s perspective,
each negotiator can incur a maximum cost of $168,000
and a minimum cost of $56,000.
Experimental procedures
Upon arrival at the experimental rooms, participants
were seated on opposite sides of a table with a chest-high
barrier that prevented them from seeing each other’s infor-
mation (the barrier, however, did not prevent negotiators
from seeing each other face-to-face). A digital recorder
was placed on top of the barrier to record their negotiation.
The experiment had four stages: introduction, prepara-
tion, face-to-face negotiation and post-negotiation (refer to
Fig. 2). In the introduction stage, participants read through
a package containing background information about their
roles and their cost schedules and were asked two ques-
tions to ensure that they understood the task. They were
asked to indicate the highest and lowest costs on their
schedule and calculate the total cost and the points they
would earn given a hypothetical agreement. Most partici-
pants were correct in their ?rst attempt, and all partici-
pants obtained the correct answers on their second
attempt.
In the preparation stage, in order to elicit participants’
initial level of the ?xed-pie bias (see De Dreu et al.,
2000), participants were ?rst asked to estimate their nego-
tiation partner’s costs by ?lling in a blank cost schedule.
Participants were permitted to refer to their own cost sche-
dule as a point of reference. They were also asked to plan
their negotiation by thinking about and noting down their
expectations about good negotiation outcomes (‘outcome
accountability’ treatment) or good negotiation processes
(‘process accountability’ treatment).
At the face-to-face negotiation stage, participants were
given 20 min to reach an agreement on all six issues relat-
ing to after-sales services. They then completed post-nego-
tiation reports detailing their negotiated agreement. From
this information joint costs were calculated by the
researchers. Finally, in the post-negotiation stage, they
were asked to re-estimate the other party’s cost schedule
(to measure the post-negotiation ?xed-pie bias), and com-
plete manipulation checks and demographic questions.
The average length of time to complete the experiment
was 1 h.
Dependent variables
The dependent variables are joint costs, ?xed-pie bias
revisions, and the proportion of integrative negotiation
tactics used. Joint costs are calculated by summing the
costs incurred by the customer and the supplier within a
negotiation dyad (see also Drake & Haka, 2008; Van den
Abbeele et al., 2009). As noted earlier, the lowest level of
joint costs is $190,000, whereas the highest level of joint
costs is $258,000.
Consistent with prior research in the negotiation litera-
ture (De Dreu, 2003), we measure ?xed-pie bias revisions
by subtracting participants’ ?xed-pie biases before the
negotiation from those immediately after the negotiation.
As the ?xed-pie bias refers to the extent to which a
First stage (Introduction)
• Participants form into negotiation dyads.
• Participants receive background information, and answer two
questions to test their understanding of the task.
Second stage (Preparation)
• Participants estimate their negotiation partner’s cost schedule
(this measures the fixed-pie bias prior to negotiation).
• Participants note down a plan of their negotiation (to reinforce
the manipulation of accountability).
Third stage (Face-to-face negotiation)
• Participants have 20 minutes to reach an agreement.
• Participants complete a report that summarizes their agreement.
• Participants under the accountability treatments complete their
accountability reports.
Fourth stage (Post-negotiation)
• Participants re-estimate their negotiation partner’s cost (the
post-negotiation measure of the fixed-pie bias).
• Participants complete manipulation checks, demographic
questions, and receive payments.
End of experiment
Fig. 2. Stages of experimental procedures.
100 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
negotiator assumes that his/her negotiation partner’s costs
are directly opposite to his/her own costs, the following
formula (see De Dreu et al., 2000) is used to estimate the
?xed-pie bias for both members of each negotiating
dyad
13
:
Fixed-pie bias ¼
X
4
i¼1
X
3
j¼1
ðEstimates of partner
0
s costs
ij
Ànegotiator
0
s own costs
ij
Þ
i refers to the four trade-off issues incorporated in the task,
and j refers to the three levels of cost within each trade-off
issue; the smaller the score, the larger the ?xed-pie bias (a
small score indicates the belief that the negotiation part-
ner’s cost is very similar to the negotiator’s own cost, i.e.,
?xed-pie bias).
14
Therefore, an increase in a dyad’s ?xed-
pie bias score after the negotiation indicates a reduction in
the dyad’s ?xed-pie bias (i.e., a positive ?xed-pie bias
revision).
The third dependent variable is the proportion of inte-
grative tactics used, with the unit of analysis being a nego-
tiation dyad. To compute this variable, we code the types of
negotiation tactics used by the dyads based on the audio
recordings of the negotiations using a well-established
coding scheme developed by Weingart et al. (1996). In
their coding scheme, ?ve different negotiating behaviors
are identi?ed: offers, information provision, questions,
procedural comments and miscellaneous comments. These
behaviors are then classi?ed as either integrative tactics or
distributive tactics. Previous negotiation research argues
that integrative tactics are aimed at addressing the under-
lying interests of the negotiating parties and identifying
mutually bene?cial trade-offs (Pruitt, 1981; Weingart
et al., 1996). For this to happen, negotiators need to under-
stand each other’s priorities and then trade-off those less
important issues for more important issues. Negotiation
behaviors that address multiple issues at a time, and focus
on priorities across issues, enable negotiators to identify
these cost saving trade-offs and are classi?ed as integrative
tactics. Negotiation behaviors are classi?ed as distributive
tactics if they focused on addressing single issues and pref-
erences within those issues. Thus, negotiation behaviors
that address one issue at a time and negotiators’ prefer-
ences within an issue will highlight the con?icting inter-
ests between the two parties. Finally, arguments aimed
at persuading the other party to change his/her attitudes
towards issues are also coded as distributive tactics.
Behaviors that could not be classi?ed as either distributive
or integrative were classi?ed as miscellaneous com-
ments
15
(Weingart et al., 1996).
The audio recordings of each negotiation were ?rst
transcribed to identify each negotiator by role and speak-
ing turns. Each transcript was then coded, using the above
coding scheme, by two research assistants blind to the re-
search hypotheses.
16
Inter-rater reliability, as measured by
Cohen’s (1960) kappa, was computed for each transcript
and ranged from 0.75 to 0.96. Consistent with prior research
(Sullivan, O’Connor, & Burris, 2006; Weingart et al., 1996),
each speaker turn was subdivided into thought units. A
thought unit is a statement or group of statements that fo-
cuses on one idea or thought. Using the above coding
scheme, we coded each thought unit as either a speci?c
negotiation tactic or a miscellaneous comment (e.g., greet-
ings). Negotiation tactics were included in the following for-
mula to determine the proportion of integrative tactics (per
negotiation dyad):
Proportionof integrative tacticsðper dyadÞ
¼
P
integrative tactics coded
P
integrative tactics codedþ
P
distributive tactics coded
Results
Manipulation checks
Participants answered two manipulation check ques-
tions. The ?rst question asked participants to indicate their
role (as a supplier or a customer), and the second question
asked participants to indicate if the reports they wrote re-
lated to the justi?cation of their negotiation outcome or
their negotiation process. As noted earlier, three partici-
pants failed one or more of these questions and conse-
quently, three dyads were excluded from our analysis.
17
We also examined the accountability reports submitted by
each negotiation dyad to check if their reports focused on
justifying the negotiation outcomes or negotiation pro-
cesses. The 30 negotiation dyads included in our ?nal anal-
13
The formula we use to measure the ?xed-pie bias is well established in
the negotiation literature (e.g., De Dreu et al., 2000; Halevy, Chou, &
Murnighan, 2012; Pietroni, Van Kleef, De Dreu, & Pagliaro, 2008). Essen-
tially, the formula takes the absolute difference between the participant’s
estimates of the other party’s costs and his/her own costs across all levels of
the trade-off issues (in our study, these are: marketing support, delivery
locations, design support and glass cutting). An absolute score of zero
indicates that the participant believes that the other party’s cost (prefer-
ence) for the trade-off issues are exactly opposite to his/her own, thus
yielding a zero-sum estimation. As this formula is based on the assumption
that the negotiation partner’s actual cost is not the same as the negotiator’s
own costs, it only applies to the trade-off issues.
14
For example, the issue ‘marketing support’ has three levels of services:
low, medium or high. To calculate the ?xed-pie bias for the supplier, we
sum the absolute difference between the supplier’s own marketing support
costs and his/her estimates of the customer’s marketing support costs for
each of the three levels of service.
15
Miscellaneous comments generally refer to off-task comments that are
not negotiation tactics.
16
Overall, three research assistants were involved in the coding process
but only two coded each transcript. To train the research assistants, each
research assistant was ?rst required to individually code ten randomly
selected transcripts using a detailed coding sheet. The training was
considered complete when the inter-rater reliability among the coders
reached an acceptable level (Cohen’s kappa > 0.75; Cohen, 1960). For the
rest of the coding process, the research assistants were randomly paired to
code ?ve transcripts at a time. After the transcripts were coded, one of the
authors would meet with the coders to discuss the coding and reconcile
disagreements by reviewing the transcripts again. A single set of codes was
then agreed upon for each dyad. The negotiation tactics were then analyzed
using this data.
17
The inclusion of the three dyads reduced the value of the t-statistics for
the impact of accountability on joint costs from t = 2.000 (p = 0.030) to
t = 1.53 (p = 0.068). However, the direction and signi?cance levels of the
other results were not changed by including these three dyads.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 101
ysis all submitted accountability reports that were consis-
tent with our manipulation.
Tests of hypotheses
To test our hypotheses, we employ a partial least square
(PLS) path approach. PLS allows us to simultaneously test
all the relationships in our proposed model, including both
the direct and indirect linkages. As a result of the distribu-
tion-free assumption of PLS, bootstrapping resampling is
used to test the signi?cance of paths (Chin, 1998). We em-
ploy SmartPLS (v2.0) with a bootstrapping resampling pro-
cedure of 500 samples (generated from the original data)
18
to test all hypotheses. The descriptive statistics on joint costs
and ?xed-pie bias revisions are shown in Table 2, and the
path coef?cients and t-statistics of the PLS model are shown
in Fig. 3 and Table 3.
H1 predicts that joint costs are lower for negotiators
with larger ?xed-pie bias revisions. The PLS model pre-
sented in Fig. 3 and Table 3 Panel A show that the path
from ?xed-pie bias revisions to joint costs is negative and
signi?cant (b = À0.310, t = 1.777, p = 0.038),
19
indicating
that larger ?xed-pie revisions result in lower joint costs.
Thus, H1 is supported.
H2a and H2b predict the direct and indirect effects of
process versus outcome accountability on ?xed-pie bias
revisions, respectively. Table 2 Panel B shows ?xed-pie bias
revisions of $176,683 for process accountability and
$71,167 for outcome accountability. We note that while
outcome accountability negotiators had a relatively lower
beginning ?xed-pie bias compared to process accountabil-
ity negotiators ($73,186 versus $56,443), this difference is
not statistically signi?cant (t = 0.710, p = 0.483).
20
Consistent with H2a, the path from accountability to
?xed-pie bias revisions is positive and signi?cant
(b = 0.265, t = 2.003, p = 0.023). In our PLS model, process
accountability is coded as 2 and outcome accountability
is coded as 1; as such, the positive coef?cient shows that
?xed-pie bias revisions are higher for participants under
process accountability than those under outcome account-
ability. Thus, H2a is supported.
H2b predicts an indirect effect from accountability to
?xed-pie bias revisions, mediated by the negotiators’ use
of integrative tactics. Table 4 reports the negotiation tac-
tics used by participants (with the unit of analysis for each
treatment being a dyad). The audio transcripts reveal a to-
tal of 1876 thought units that were coded as negotiation
tactics; of these, 462 thought units were coded as integra-
tive tactics and 1414 thought units were coded as distrib-
utive tactics.
21
The overall proportion of integrative tactics
is 28.22% is consistent with prior negotiation literature
(Weingart et al., 1996), which shows negotiators generally
use a lower portion of integrative tactics compared to dis-
tributive tactics. As predicted by H2b, Table 3 Panel A shows
that accountability is positively associated with the propor-
tion of integrative tactics used by negotiating dyads
(b = 0.340, t = 2.150, p = 0.016),
22
and the proportion of inte-
grative tactics used is positively associated with ?xed-pie
bias revisions (b = 0.538, t = 4.559, p < 0.000). The Sobel’s
test (Table 3 Panel B) reports that this indirect relationship
is signi?cant (b = 0.183, t = 1.984, p = 0.024), showing a com-
plementary mediation effect (Zhao, Lynch, & Chen, 2010)
and providing support for H2b.
Table 2
Means (standard deviation) of joint costs, ?xed-pie bias and ?xed-pie bias revisions.
Process accountability (n = 15) Outcome accountability (n = 15) Overall
Panel A: Joint costs
Joint costs ($) 194,133 (6682) 201,600 (12,826) 197,867 (10,742)
Panel B: Fixed-pie bias
Fixed-pie bias (beginning) $ 56,433 (55,315) 73,186 (72,683) 64,810 (64,032)
Fixed-pie bias (end) $ 233,116 (104,136) 144,353 (91,276) 188,735 (106,278)
Fixed-pie bias revisions $ 176,683 (117,626) 71,167 (99,467) 123,925 (119,729)
Joint costs = combined costs of the supplier and the customer averaged across the 15 dyads in each treatment (Lowest joint costs = $190,000; Maximum
joint costs = $258,000).
Fixed-pie bias is the extent to which a negotiator assumes that his/her negotiation partner’s costs are the same as their own costs.
Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
scosts
ij
Ànegotiator
0
sown costs
ij
Þ
(i refers to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue).
Fixed-pie bias (beginning) = the combined ?xed-pie biases of the two negotiators within each dyad at the beginning of the negotiation, a smaller amount
indicates greater bias.
Fixed-pie bias (end) = the combined ?xed-pie biases of the two negotiators within each dyad after the negotiation has been completed, a smaller amount
indicates greater bias.
Fixed-pie bias revisions = ?xed-pie bias (end) – ?xed-pie bias (beginning); a larger amount indicates greater extent of revision.
18
Consistent with Rosenzweig (2009), we tested the stability of the
coef?cient estimates by repeating the bootstrap resampling procedure
using 250 and 1,000 bootstrap samples with consistent results.
19
All signi?cance tests are one-tailed unless otherwise speci?ed.
20
This relatively higher initial ?xed-pie bias in the process accountability
treatment works against the predictions of H2, as negotiators in the process
accountability treatment have further to go when revising their ?xed-pie
biases.
21
In addition, 1182 thought units are coded as miscellaneous statements,
such as greetings. There is no signi?cant difference in the proportion of
miscellaneous statements between the process accountability treatment
and the outcome accountability treatment (39.85% versus 44.81%, t = 0.995,
p = 0.328 two-tailed).
22
An independent t-test shows that the proportion of integrative tactics
employed by dyads in the process accountability treatment is higher than
the proportion of integrative tactics used by dyads in the outcome
accountability treatment (33.72% versus 22.72%, t = 1.876, p = 0.036, one-
tailed). This provides additional support to the path from accountability to
negotiators’ use of integrative tactics.
102 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
Table 2 Panel A shows that the mean joint costs for the
supplier and the customer combined is $194,133 for pro-
cess accountability and $201,600 for outcome accountabil-
ity (this compares to the lowest level of joint costs of
$190,000 and maximum joint costs of $258,000). H3 pre-
dicts that ?xed-pie bias revisions mediate the effect of
the type of accountability on joint costs. We have previ-
ously shown that a signi?cant path from ?xed-pie bias
revisions to joint costs (H1), and from accountability to
?xed-pie bias revisions both directly (H2a) and indirectly
via the proportion of integrative tactics used by the nego-
tiating dyads (H2b). To examine H3, we conducted two So-
bel’s tests. First, Table 3 Panel B reports that the indirect
relationship from accountability to joint costs via ?xed-
pie bias revisions is marginally signi?cant (b = 0.082,
t = 1.433, p = 0.076). A second Sobel’s test starts with the
Integrative
Tactics
Employed
Fixed-Pie Bias
Revisions
Joint Costs
Accountability
0.265**
-0.310**
0.340**
0.538***
Fig. 3. Path analysis co-ef?cients. Figure only shows path coef?cients signi?cant (one-tailed) at the 5% (
ÃÃ
) and 1% (
ÃÃÃ
) levels. In addition to these signi?cant
paths, our model also includes two other paths: a path from Accountability to Joint Costs (b = À0.141, not signi?cant) and a path from Integrative Tactics
Employed to Joint Costs (b = À0.215, not signi?cant). For indirect effect path co-ef?cients refer to Table 3 Panel B. Accountability refers to process versus
outcome accountability; that is, whether participants are held accountable for their negotiation processes (coded as 2) or negotiation outcomes (coded as
1). Fixed-pie bias revisions refers to the difference between the ?xed-pie bias before and after the negotiation in each negotiation dyad. The following
formula is used to calculate each negotiator’s ?xed-pie bias: Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
s costs
ij
Ànegotiator
0
s own costs
ij
Þj (i refers
to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue). Integrative tactics employed refers to
the proportion of integrative tactics used by a negotiation dyad relative to the total number of negotiation tactics used by that dyad. The following formula
is used to calculate integrative tactics employed (per dyad): Integrative tactics employed = integration tactics Ä (integrative + distributive tactics). Joint
costs refers to the combined costs of the supplier and the customer in each dyad.
Table 3
PLS results.
Variables Path to R
2
Proportion of integrative tactics Fixed-pie bias revisions Joint costs
Panel A: Path coef?cients and t-statistics
Accountability 0.340 (t = 2.150)
**
0.265 (t = 2.003)
**
À0.141 (t = 0.880)
Proportion of integrative tactics 0.538 (t = 4.559)
***
À0.215 (t = 1.274) 0.456
Fixed-pie bias revisions À0.310 (t = 1.777)
**
0.116
Joint costs 0.306
Variables Linkages Path to
Fixed-pie bias revisions Joint costs
Panel B: Indirect effect path coef?cients and t-statistics (Sobel’s test
a
)
Accountability Fixed-pie bias revisions 0.082 (t = 1.433)
*
Proportion of integrative tactics 0.183 (t = 1.984)
**
Proportion of integrative tactics Fixed-pie bias revisions 0.167 (t = 1.691)
**
Accountability = process versus outcome.
Fixed-pie bias revisions = ?xed-pie bias (end) À ?xed-pie bias (beginning); a larger amount indicates greater extent of revision.
Fixed-pie bias is the extent to which a negotiator assumes that his/her negotiation partner’s costs are the same as their own costs.
Fixed-pie bias ¼
P
4
i¼1
j
P
3
j¼1
ðEstimates of partner
0
scosts
ij
Ànegotiator
0
s own costs
ij
Þj
(i refers to the four trade-off issues incorporated in the task and j refers to the three levels of cost within each trade-off issue).
Proportion of integrative tactics = integration tactics Ä (integrative + distributive tactics).
Joint costs = combined costs of the supplier and the customer in each dyad.
a
Sobel’s test is used to test the statistical signi?cance of indirect relationships between an independent variable and a dependent variable through a
mediating variable (Preacher & Leonardelli, 2001).
*
p < 0.10 (one-tailed).
**
p < 0.05 (one-tailed).
***
p < 0.01 (one-tailed).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 103
proportion of integrative tactics (which was directly af-
fected by accountability), and establishes a signi?cant indi-
rect path from proportion of integrative tactics to joint
costs via ?xed-pie bias revisions (b = 0.167, t = 1.691,
p = 0.045). These results, combined with an insigni?cant
direct effect from accountability to joint costs (b = 0.141,
t = 0.880, p = 0.386, two-tailed), show that accountability
has an indirect-only mediation effect on joint costs (Zhao
et al., 2010). Thus, H3 is supported.
23
Although we did not hypothesize a direct link from pro-
portion of integrative tactics to joint costs, we have in-
cluded this link in our ?nal PLS model for completeness.
As expected, the direct path from proportion of integrative
tactics to joint cost is not signi?cant (b = À0.215, t = 1.274,
p = 0.214, two-tailed), indicating that the dyads’ use of
integrative tactics is only effective in reducing joint costs
via their effect on ?xed-pie bias revisions.
Additional analysis
In addition to the above design we included a control
group (no accountability) of 16 dyads. For the control
group, participants were neither required to justify their
negotiation processes nor their negotiation outcomes. In-
stead, their bonus points depended solely on the ?nal cost
to be borne by their respective company at the end of the
negotiation. The inclusion of this control group allows us to
see if either accountability treatments result in different
negotiation performance relative to a baseline condition
where participants are rewarded based only on cost mini-
mization. This control condition thus represents a manage-
ment control system that does not explicitly incorporate
an accountability requirement. We conducted post-hoc
pairwise comparisons between the control group (no
accountability) and each of the two treatment groups.
There were no signi?cant differences between the control
group and the outcome accountability treatment with re-
spect to both ?xed-pie bias revisions ($88,627 versus
$71,167, t = 0.445, p = 0.670, two-tailed) and joint costs
($203,718 versus $201,600, t = 0.399, p = 0.694, two-
tailed). In contrast, compared to participants in the control
group, those in the process accountability treatment have
signi?cantly higher ?xed-pie bias revisions ($176,683 ver-
sus $88,627, t = 2.245, p = 0.030, two-tailed) and signi?-
cantly lower joint costs ($194,133 versus $203,718,
t = 2.130, p = 0.046, two-tailed). The results show that only
process accountability was effective in improving negotia-
tors’ performance beyond the no accountability
alternative.
24
We also conducted further analysis of the tactics used
by the negotiators, by analyzing the four types of negotia-
tion behaviors used by the negotiators across both process
and outcome accountability. Table 4 shows that the rela-
tive proportions of negotiation behaviors adopted by
Table 4
Summary of negotiation tactics used by negotiating dyads.
*
Negotiation behavior Process Accountability (% as a
proportion of total number of
tactic-related thought units coded)
Outcome accountability (% as a
proportion of total number of tactic-
related thought units coded)
Overall (% as a proportion of total
number of tactic-related thought
units coded)
Tactic type Integrative Distributive Total Integrative Distributive Total Integrative Distributive Total
Offers
a
9.40% 10.79% 20.19% 11.73% 15.30% 27.03% 10.57% 13.00% 23.57%
Information provision
b
10.22% 37.05% 47.27% 4.26% 40.70% 44.96% 7.24% 38.92% 46.16%
Questions
c
8.55% 11.59% 20.14% 2.85% 12.65% 15.50% 5.70% 12.12% 17.82%
Procedural comments
d
5.55% 6.85% 12.40% 3.88% 8.63% 12.51% 4.71% 7.74% 12.45%
Total 33.72% 66.28% 100% 22.72% 77.28% 100% 28.22% 71.78% 100%
Total number of thought units
coded as tactics
269 Units 602 Units 871
Units
193 Units 812 Units 1005
Units
462 Units 1414 Units 1876
Units
*
The unit of analysis for negotiation tactics used is a dyad.
a
Integrative offers – offers and counter offers relating to more than one issue at a time. Distributive offers – offers and counter offers relating to only one
issue at a time.
b
Integrative information provision – communicating information about which issues are more important to themselves. Distributive information pro-
vision – communicating information about their preferences within a particular issue, and making arguments for their own position, or make arguments
against the other’s position, in an attempt to persuade the other party to shift their position.
c
Integrative questions – asking questions about which issues are more important to their negotiation partner. Distributive questions – asking questions
about their negotiation partner’s preferences within an issue and questioning the validity of their negotiation partner’s arguments.
d
Integrative procedural comments – suggestions about packaging issues together or discussing sets of issues simultaneously, or when the negotiator
suggests a current concession to be made in exchange for an unidenti?ed future concession. Distributive procedural comments – suggestions to address one
issue at a time or when negotiators discuss one issue at a time or when the negotiators suggest a willingness to concede their position within an issue
(Weingart et al., 1996).
23
We also conduct a 2 Â (2) ANOVA with ‘accountability’ being the
between-subjects factor and ‘time’ (before versus after negotiation) being
the repeated measure factor. The results show a signi?cant interaction
effect between time and accountability (F = 6.069, p = 0.017, two-tailed),
with ?xed-pie bias revisions being higher for participants under process
accountability than those under outcome accountability ($176,683 and
$71,167 respectively). This provides further support that accountability has
?ow-on effects on joint costs.
24
As discussed earlier, this study was conducted in two locations. For the
control group, ?ve dyads are USA participants and 11 dyads are Australian
participants. We re-analyzed our data with location (USA and Australia) as
a covariate, and found that location was not a signi?cant covariate.
However, the inclusion of location as a covariate weakened our results,
such that the difference between the process accountability and no
accountability treatments became marginally signi?cant (?xed-pie bias
revisions: t = 1.786, p = 0.081; and joint costs: t = 1.731, p = 0.091, two-
tailed). The difference between the no accountability and outcome
accountability treatments remained insigni?cant. Therefore, our results
related to the control group need to be interpreted with caution.
104 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
participants differ between the two treatments. Of the four
types of negotiation behaviors, integrative information
provision and integrative questioning behaviors have the
largest differences between participants in the two
accountability treatments. Speci?cally, the proportion of
integrative information provision behavior is 10.22% for
participants in the process accountability treatment, but
only 4.26% for the outcome accountability participants
(t = 2.245, p = 0.033, two-tailed). Similarly, the process
accountability participants also engage in more integrative
questioning behaviors than their outcome accountability
counterparts (8.55% versus 2.85%, t = 2.170, p = 0.039,
two-tailed).
25
Both types of negotiation behaviors are gener-
ally employed as means of exchanging information. This re-
sult is consistent with our expectations that process
accountability participants are motivated to engage in nego-
tiation behaviors that allow them to better understand the
interests of both negotiation parties, in particular, through
active information exchange.
Conclusion and discussion
Customer–supplier negotiation is crucial to unlocking
the value in supply chains (Anderson & Dekker, 2009a,
2009b; Dekker, 2003). However, prior research (e.g., Drake
& Haka, 2008; Van den Abbeele et al., 2009) shows that the
provision of more sophisticated and relevant accounting
information alone is not suf?cient to ensure that negotia-
tors can use this information to identify mutually bene?-
cial opportunities, and to extract maximum values out of
the supply chain relationships. Rather, an organization’s
management control system also plays an important role
in improving supply chain performance (Free, 2007). In
this study, we examine how an important aspect of man-
agement control systems, namely accountability, affects
negotiators’ ability to attain lower joint costs. We ?nd that
negotiators held accountable for their negotiation pro-
cesses have higher ?xed-pie bias revisions and subse-
quently achieve lower joint costs than those held
accountable for their negotiation outcomes. Through our
PLS analysis, we provide evidence on how negotiators can
improve their performance. Our model indicates that pro-
cess accountability lowers joint costs to a greater extent
than outcome accountability through two paths. First, a
cognitive process path, where we ?nd that process
accountability is more effective than outcome accountabil-
ity in directly increasing negotiators’ ?xed-pie bias revi-
sions (thus reducing the ?xed-pie bias), and in doing so,
reduces joint costs. Second, we also ?nd a behavioral path
where process accountability is more effective than out-
come accountability in encouraging negotiators to employ
more integrative tactics. This allows negotiators to reduce
their ?xed-pie biases, and ultimately, lower joint costs. In
addition, we show that compared to the no accountability
control group, only process accountability is effective in
reducing the ?xed-pie bias and improving joint costs. De-
tailed analysis of the negotiation tactics obtained via tape
recordings of the negotiations further supports the ?ndings
that participants in the process accountability treatment
are more likely to exchange information through integra-
tive information provision and integrative questioning
behaviors than participants who are accountable for nego-
tiation outcomes.
The primary contribution of this study is to demon-
strate one way in which a management control system
can be designed to encourage negotiators to use cost infor-
mation to reach better joint outcomes. Prior studies such
as Drake and Haka (2008) and Van den Abbeele et al.
(2009) ?nd that providing negotiators with ABC informa-
tion (compared to volume-based cost information) only
improves negotiation outcomes when negotiators are
faced with external adversities, such as adverse market
conditions or a more powerful other party. In our study,
both negotiation parties received ABC information; our re-
sults show that organizations can actively improve joint
outcomes by imposing an appropriate type of accountabil-
ity (process accountability) on negotiators. We also show
that the ?xed-pie bias inhibits the improvement of negoti-
ation outcomes, and that process accountability can more
effectively mitigate this bias than outcome accountability.
Despite accountability being an important concept in
improving management control, there is little research
that addresses how different types of accountability
requirements affect managers’ decisions and behaviors.
Our study adds to this literature by showing that including
process accountability (instead of outcome accountability)
in the evaluation of negotiators’ performance has bene?ts
in terms of improved supply chain ef?ciencies. This has po-
tential implications for other circumstances such as
responsibility centre managers explaining their perfor-
mance in monthly divisional review meetings (see Rowe
et al., 2008). Our results suggest potential bene?ts of man-
agers justifying their processes rather than outcomes in
these meetings. In addition, contemporary performance
measurement systems (e.g., the balanced scorecard) re?ect
a process-based focus on intermediate initiatives, such as
improving relations with suppliers and developing new
technology capabilities; under process accountability,
managers could be rewarded based on their justi?cations
of how they have taken steps to improve these intermedi-
ate processes, rather than their justi?cations of the ?nal
division outcomes such as divisional revenue or budget
variances.
Our ?ndings build on Pinkley et al.’s (1995) argument
that poor negotiation outcomes are often the result of
negotiators not having accurate information about their
negotiation partner (information availability errors), as
well as negotiators’ tendency to distort or ignore useful
information about their negotiation partner (information
processing errors). We develop and test a path model to
show that process accountability is more effective than
outcome accountability in assisting negotiators to address
both errors: through the employment of more integrative
tactics negotiators can obtain more information about
their negotiation partner’s preferences, and at the same
time, the direct path from accountability to ?xed-pie bias
revisions suggests that process accountability also reduces
25
All other negotiation behaviors were not statistically different between
outcome accountability and process accountability (p > 0.10).
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 105
negotiators’ information processing errors, independent of
the negotiation tactics employed.
Our study also contributes to the negotiation literature
by directly comparing the effect of process and outcome
accountability on the ?xed-pie bias and negotiation perfor-
mance. Unlike prior studies such as De Dreu et al. (2000),
which only examined process accountability, or Carnevale,
Pruitt, and Seilheimer (1981) and O’Connor (1997) which
only examined outcome accountability, this study directly
compares the effect of the two different types of account-
ability on negotiation behavior and outcomes. Importantly,
we show that the type of accountability in?uences both
the negotiators’ cognitive processes (?xed-pie bias revi-
sions) and behavior (negotiation tactics). As De Dreu and
Carnevale (2003) have noted, prior negotiation literature
often considers these processes in isolation. Our model
shows that the behavior of negotiators in?uences their
ability to overcome cognitive biases, and in doing so, we
demonstrate the importance of incorporating both factors
concurrently when conducting negotiation research.
Our study has a number of limitations, which also sug-
gest several opportunities for future research. First, the
cost functions of all service-related costs in our study were
linear. There is the opportunity for future research to
examine the role of accountability in situations where
non-linear cost functions make trade-off opportunities
more dif?cult to identify. Second, the negotiations in our
study occur during one period. In some circumstances,
the same parties may negotiate multiple rounds over a
number of years, and this may introduce additional consid-
erations, such as each party’s negotiation reputation (e.g.,
Tinsley, O’Connor, & Sullivan, 2002). Third, the current
study focuses on negotiations between two parties. How-
ever, customer–supplier negotiations sometimes involve
multiple parties (Anderson & Dekker, 2009a). Future re-
search can investigate whether process accountability con-
tinues to have a positive effect on negotiation outcomes in
a more complex negotiation environment. Fourth, another
design feature of our experiments is that the trade-off is-
sues for negotiators were symmetrical in value. However,
in some circumstances one negotiation party may have a
greater ability to contribute to expanding the negotiation
pie, and this may affect negotiation outcomes.
Future research could also examine the wider implica-
tions of our research for the design of management control
systems. This literature has long argued for the importance
of responsibility accounting, which implies holding manag-
ers accountable for their actions in order to improve effort
and performance outcomes (Rowe et al., 2008). While the
existing research on responsibility accounting tends to fo-
cus on the basis upon which managers should be held
accountable for (e.g., cost versus pro?t, individual versus
team outcomes), our study suggests that designers of man-
agement control systems also need to consider the type of
accountability imposed on managers. For example, while
organizations often require managers to justify perfor-
mance outcomes such as budget variances (Rowe et al.,
2008), it may be desirable if these managers are made to ac-
count for the processes leading to these outcomes, such as
the initiatives they had in place to reduce manufacturing
overhead costs. Seen in this light, process accountability
is similar to the notion of a balanced scorecard, which
encourages managers to be held responsible for the ‘causes’
of outcomes (i.e., the lead measures and initiatives). Re-
search could explore whether process accountability in
conjunction with a balanced scorecard systemcan have po-
sitive effects on other types of managerial tasks.
Acknowledgements
We gratefully acknowledge a research grant from the
Australian Research Council and the helpful comments
from the two anonymous reviewers, Shannon Anderson,
Henri Dekker, Andrea Drake, Gary Hecht, Anne Lillis,Habib
Mahama, Karen Sedatole, Michael Shields, and Yee Shih
Phua, as well as seminar participants at Monash Univer-
sity, Macquarie University,conference participants at the
2009 AAA Annual Meeting, the 2009 AFAANZ Conference,
the 2008 New Directions in Management Accounting Con-
ference and the 2008 GMARS Emerging Scholars Forum.
References
Ahrens, T. (1996). Styles of accountability. Accounting, Organizations and
Society, 21(2/3), 139–173.
Anderson, S. W., & Dekker, H. C. (2005). Management control for market
transactions: The relation between transaction characteristics,
incomplete contract design, and subsequent performance.
Management Science, 51(12), 1734–1752.
Anderson, S., & Dekker, H. (2009a). Strategic cost management in supply
chains, part 1: Structural cost management. Accounting Horizons,
23(2), 201–220.
Anderson, S., & Dekker, H. (2009b). Strategic cost management in supply
chains, part 2: Executional cost management. Accounting Horizons,
23(3), 289–305.
Asare, S. K., & Wright, A. M. (2003). A note on the interdependence
between hypothesis generation and information search in conducting
analytical procedures. Contemporary Accounting Research, 20(2),
235–251.
Bazerman, M. H. (2006). Judgment in managerial decision making.
Hoboken, NJ: J. Wiley.
Bonner, S. E. (2007). Judgment and decision making in accounting. New
York: Prentice Hall.
Brtek, M. D., & Motowidlo, S. J. (2002). Effects of procedure and outcome
accountability on interview validity. Journal of Applied Psychology,
87(1), 185–191.
Carnevale, P. J., Pruitt, D. G., & Seilheimer, S. D. (1981). Looking and
competing: Accountability and visual access in integrative bargaining.
Journal of Personality and Social Psychology, 40(1), 111–120.
Chang, L., Cheng, M., & Trotman, K. T. (2008). The effect of framing and
negotiation partner’s objective on judgments about negotiated
transfer prices. Accounting, Organizations and Society, 33(7–8),
704–717.
Chin, W. W. (1998). Issues and opinion on structural equation modeling.
MIS Quarterly, 22(1), vii–xvii.
Cohen, J. (1960). A coef?cient of agreement for nominal scales.
Educational and Psychological Measurement, 20(1), 37–46.
Cools, M., & Slagmulder, R. (2009). Tax-compliant transfer pricing and
responsibility accounting. Journal of Management Accounting Research,
21, 151–178.
De Dreu, C. K. W. (2003). Time pressure and closing of the mind in
negotiation. Organizational Behavior and Human Decision Processes,
91(2), 280–295.
De Dreu, C. K. W., & Carnevale, P. J. (2003). Motivational bases of
information processing and strategy in con?ict and negotiation. In
Advances in experimental social psychology (pp. 235–291). Academic
Press.
De Dreu, C. K. W., Beersma, B., Stroebe, K., & Euwenma, M. C. (2006).
Motivated information processing, strategic choice, and the quality of
negotiated agreement. Journal of Personality and Social Psychology,
90(6), 927–943.
De Dreu, C. K. W., Koole, S. L., & Steinel, W. (2000). Un?xing the ?xed pie:
A motivated information-processing approach to integrative
106 L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107
negotiation. Journal of Personality and Social Psychology, 79(6),
975–987.
Dekker, H. C. (2003). Value chain analysis in inter?rm relationships: A
?eld study. Management Accounting Research, 14(1), 1–23.
DeZoort, T., Harrison, P., & Taylor, M. (2006). Accountability and auditors’
materiality judgments: The effects of differential pressure strength on
conservatism, variability, and effort. Accounting, Organizations and
Society, 31(4–5), 373–390.
Drake, A. R., & Haka, S. F. (2008). Does ABC information exacerbate hold-
up problems in buyer-supplier negotiations? The Accounting Review,
83(1), 29–60.
Evans, J. H., III, Heiman-Hoffman, V. B. R., & Rau, S. E. (1994). The
accountability demand for information. Journal of Management
Accounting Research, 6(1), 24–42.
Free, C. (2007). Supply-chain accounting practices in the UK retail sector:
Enabling or coercing collaboration? Contemporary Accounting
Research, 24(3), 897–933.
Gelfand, M. J., & Realo, A. (1999). Individualism-collectivism and
accountability in intergroup negotiations. Journal of Applied
Psychology, 84(5), 721–736.
Gibbins, M., & Trotman, K. T. (2002). Audit review: Managers’
interpersonal expectations and conduct of the review. Contemporary
Accounting Research, 19(3), 411–444.
Glover, S. M. (1997). The in?uence of time pressure and accountability on
auditors’ processing of nondiagnostic information. Journal of
Accounting Research, 35(2), 213–226.
Halevy, N., Chou, E. Y., & Murnighan, J. K. (2012). Mind games: The mental
representation of con?ict. Journal of Personality and Social Psychology,
102(1), 132–148.
Hoffman, V. B., & Patton, J. M. (1997). Accountability, the dilution effect,
and conservatism in auditors’ fraud judgments. Journal of Accounting
Research, 35(2), 227–237.
Hurkens, K., van der Valk, W., & Wynstra, F. (2006). Total cost of
ownership in the services sector: A case study. Journal of Supply Chain
Management, 42(1), 27–37.
Hyder, E. B., Prietula, M. J., & Weingart, L. R. (2000). Getting to best:
Ef?ciency versus optimality in negotiation. Cognitive Science, 24(2),
169–204.
Kachelmeier, S. J., & Towry, K. L. (2002). Negotiated transfer pricing: Is
fairness easier said than done? The Accounting Review, 77(3), 571–593.
Kadous, K., & Sedor, L. M. (2004). The ef?cacy of third-party consultation
in preventing managerial escalation of commitment: The role of
mental representations. Contemporary Accounting Research, 21(1),
55–82.
Kaplan, R., & Anderson, S. (2007). Time-driven activity-based costing.
Boston: Harvard Business School Publishing.
Kennedy, J. (1993). Debiasing audit judgment with accountability: A
Framework and experimental results. Journal of Accounting Research,
31(2), 231–245.
Koonce, L., Anderson, U., & Marchant, G. (1995). Justi?cation of decisions
in auditing. Journal of Accounting Research, 33(2), 369–384.
Kruglanski, A. W. (1989). The psychology of being ‘‘right’’: The problem of
accuracy in social perception and cognition. Psychological Bulletin,
106(3), 395–409.
Kruglanski, A. W., & Webster, D. M. (1996). Motivated closing of the mind:
‘‘Seizing’’ and ‘‘freezing’’. Psychological Review, 103(2), 263–283.
Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of
accountability. Psychological Bulletin, 125(2), 255–275.
Lewicki, R. J., Barry, B., & Saunders, D. M. (2010). Negotiation (6th ed.).
New York: McGraw-Hill/Irwin.
Libby, T., Salterio, S. E., & Webb, A. (2004). The balanced scorecard: The
effects of assurance and process accountability on managerial
judgment. The Accounting Review, 79(4), 1075–1094.
Luft, J. L., & Libby, R. (1997). Pro?t comparisons, market prices and
managers’ judgments about negotiated transfer prices. The Accounting
Review, 72(2), 217–229.
Luft, J., & Shields, M. (2009). Psychology models of management.
Accounting Foundations and Trends
Ò
in Accounting, 4(3-4), 199–345.
Masschelein, S., Cardinaels, E., & Van den Abbeele, A. (2012). ABC
information, fairness perceptions, and inter?rm negotiations. The
Accounting Review, 87(3), 951–973.
Merchant, K. A., & Otley, D. T. (2006). A review of the literature on control
and accountability. In A. G. H. Christopher, S. Chapman, & D. S.
Michael (Eds.), Handbooks of management accounting research
(pp. 785–802). Elsevier.
Nelson, M., & Tan, H.-T. (2005). Judgment and decision making research in
auditing: A task, person, and interpersonal interaction perspective.
Auditing, 24, 41–71.
O’Connor, K. M. (1997). Groups and solos in context: The effects of
accountability on team negotiation. Organizational Behavior and
Human Decision Processes, 72(3), 384–407.
Olekalns, M., Smith, P. L., & Walsh, T. (1996). The process of negotiating:
Strategy and timing as predictors of outcomes. Organizational
Behavior and Human Decision Processes, 68(1), 68–77.
Peecher, M. E. (1996). The in?uence of auditors’ justi?cation processes on
their decisions: A cognitive model and experimental Evidence. Journal
of Accounting Research, 34(1), 125–140.
Pietroni, D., Van Kleef, G. A., De Dreu, C. K. W., & Pagliaro, S. (2008).
Emotions as strategic information: Effects of other’s emotional
expressions on ?xed-pie perception, demands, and integrative
behavior in negotiation. Journal of Experimental Social Psychology,
44(6), 1444–1454.
Pinkley, R. L., Grif?th, T. L., & Northcraft, G. B. (1995). ‘‘Fixed pie’’ a la
mode: Information availability, information processing, and the
negotiation of suboptimal agreements. Organizational Behavior and
Human Decision Processes, 62(1), 101–112.
Preacher, K. J., & Leonardelli, G. J. (2001). Calculation for the Sobel test: An
interactive calculation tool for mediation tests (Computer Software).
Accessed
March 2011.
Pruitt, D. G. (1981). In P. Warr (Ed.), Negotiation behavior. New York:
Academic Press.
Reinartz, W., Thomas, J. S., & Kumar, V. (2005). Balancing acquisition and
retention resources to maximize customer pro?tability. Journal of
Marketing, 69(1), 63–79.
Roberts, J. (1991). The possibilities of accountability. Accounting,
Organizations and Society, 16(4), 335–368.
Roodhooft, F., Hiel, G., Van den Abbeele, A., & Van Doveren, B. (2003).
Optimized sourcing strategies using total cost of ownership. Cost
Management, 17(4), 28–35.
Rosenzweig, E. (2009). A contingent view of e-collaboration and
performance in manufacturing. Journal of Operations Management,
27(6), 462–478.
Rowe, C., Birnberg, J. G., & Shields, M. D. (2008). Effects of organizational
process change on responsibility accounting and managers’
revelations of private knowledge. Accounting, Organizations and
Society, 33(2–3), 164–198.
Siegel-Jacobs, K., & Yates, J. F. (1996). Effects of procedural and outcome
accountability on judgment quality. Organizational Behavior and
Human Decision Processes, 65(1), 1–17.
Simonson, I., & Staw, B. M. (1992). Deescalation strategies: A comparison
of techniques for reducing commitment to losing courses of action.
Journal of Applied Psychology, 77(4), 419–426.
Sullivan, B. A., O’Connor, K. M., & Burris, E. R. (2006). Negotiator
con?dence. The impact of self-ef?cacy on tactics and outcomes.
Journal of Experimental Social Psychology, 42(5), 567–581.
Tan, H.-T. (1995). Effects of expectations, prior involvement, and review
awareness on memory for audit evidence and judgment. Journal of
Accounting Research, 33(1), 113–135.
Tan, H.-T., & Kao, A. (1999). Accountability effects on auditors’
performance. The in?uence of knowledge, problem-solving ability,
and task complexity. Journal of Accounting Research, 37(1), 209–223.
Thompson, L. L., & Hastie, R. (1990). Social perception in negotiation.
Organizational Behavior and Human Decision Processes, 47(1), 98–123.
Thompson, L. L., & Hrebec, D. (1996). Lose-lose agreements in
interdependent decision making. Psychological Bulletin, 120(3),
396–409.
Thompson, L. L., Wang, J., & Gunia, B. C. (2010). Negotiation. Annual
Review of Psychology, 61(1), 491–515.
Tinsley, C. H., O’Connor, K., & Sullivan, B. A. (2002). Tough guys ?nish last:
The perils of a distributive reputation. Organizational Behavior and
Human Decision Processes, 88, 621–642.
Trotman, K., Wright, A., & Wright, S. (2005). Auditor negotiations: An
examination of the ef?cacy of intervention methods. The Accounting
Review, 80(1), 349–367.
Van den Abbeele, A., Roodhooft, F., & Warlop, L. (2009). The effect of cost
information on buyer-supplier negotiations in different power
settings. Accounting, Organizations and Society, 34(2), 245–266.
Weingart, L. R., Hyder, E. B., & Prietula, M. J. (1996). Knowledge matters:
The effect of tactical descriptions on negotiation behavior and
outcome. Journal of Personality and Social Psychology, 70(6),
1205–1217.
Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and
Kenny: Myths and truths about mediation analysis. Journal of
Consumer Research, 37(2), 197–206.
L.J. Chang et al. / Accounting, Organizations and Society 38 (2013) 93–107 107
doc_584899443.pdf