Description
A common approach to set transfer prices is via intra-firm negotiation. However, Luft and Libby [Luft, J. L., &
Libby, R. (1997). Profit comparisons, market prices and managers’ judgments about negotiated transfer prices. The
Accounting Review, 72(2), 217–229] found that because of the existence of self-serving biases, negotiating managers have
different expectations regarding what constitutes a ‘fair’ transfer price, leading to a less efficient negotiation process. In
this study, we examine two factors that are expected to affect managers’ transfer price negotiation judgments, namely,
framing as a gain or as a loss and the negotiation partner’s objective
The e?ect of framing and negotiation partner’s objective
on judgments about negotiated transfer prices
Linda Chang, Mandy Cheng, Ken T. Trotman
*
School of Accounting, The University of New South Wales, Sydney 2052, Australia
Abstract
A common approach to set transfer prices is via intra-?rm negotiation. However, Luft and Libby [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] found that because of the existence of self-serving biases, negotiating managers have
di?erent expectations regarding what constitutes a ‘fair’ transfer price, leading to a less e?cient negotiation process. In
this study, we examine two factors that are expected to a?ect managers’ transfer price negotiation judgments, namely,
framing as a gain or as a loss and the negotiation partner’s objective (whether the partner’s objective involves high or
low concern-for-others). We propose that these two factors a?ect managers’ perceptions of the negotiation context, and
thus the way they interpret the economic and social consequences of accounting information. Our results show that a
loss frame (compared to a gain frame) exacerbates managers’ self-serving biases and increases the ‘transfer price expec-
tation gap’ between buyers and sellers. Further, in our experiment where market price is higher than equal-pro?t price,
we ?nd that managers’ transfer price expectations are lower (and deviate more from the prevailing market price) when
they are negotiating with a partner with high concern-for-others than with a partner with low concern-for-others. We
discuss the broader implications of these results for the design of management accounting systems.
Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
Introduction
Negotiation is a common method used by ?rms
to set transfer prices (Ghosh, 2000). Even where an
external market exists, transfer price negotiation is
a potentially useful control mechanism, allowing a
balance between economic considerations and
broader social concerns by interdependent divi-
sions (Kachelmeier & Towry, 2002). These transfer
price negotiations are important to managers as
they in?uence both their own and other divisional
pro?ts. Previous research has shown that these
transfer prices are a?ected by both economic
factors (market prices) and behavioural factors
including fairness (Luft & Libby, 1997).
0361-3682/$ - see front matter Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2008.01.002
*
Corresponding author. Tel.: +61 2 9385 5831; fax: +61 2
9662 4491.
E-mail address: [email protected] (K.T. Trotman).
Available online at www.sciencedirect.com
Accounting, Organizations and Society 33 (2008) 704–717
www.elsevier.com/locate/aos
In the current study, we examine whether the
impact of accounting information on managers’
transfer price expectations are moderated by the
way accounting information is framed (either as
potential gains or potential losses) and the man-
ager’s perception of the other negotiation party’s
objective (whether their partner’s objective
involves high or low concern-for-others). These
expectations are important as they directly a?ect
the costs and outcomes of negotiations (Ghosh,
2000; Luft & Libby, 1997; Trotman, Wright, &
Wright, 2005).
Previous negotiation literature has shown the
importance of ‘fairness’ during negotiation and
that participants’ estimates of a fair price display
a ‘self-serving bias’ (or egocentrism). The self-serv-
ing bias refers to the cognitive bias arising from an
individual’s tendency to view an outcome more
favourable to them as being fairer when resolving
con?icts
1
(Thompson & Loewenstein, 1992). Spe-
ci?cally, where an active external market exists,
and the market price is greater than a price that
would lead to both divisions receiving an equal
pro?t, a seller will generally consider the market
price to be a fairer transfer price as it results in a
higher pro?t for the selling division. The buyer,
however, would view the transfer price that allows
pro?t to be equally shared between the two divi-
sions as a fairer price (Luft & Libby, 1997).
Both Luft and Libby (1997) and Kachelmeier
and Towry (2002) found that where market price
di?ered fromthe equal-pro?t price, managers based
their transfer price judgments on both the market
price and the equal-pro?t price. Furthermore, both
studies found that sellers and buyers placed di?er-
ent weights on these two reference points when for-
mulating judgments. Speci?cally, due to self-
serving biases, sellers’ transfer price expectations
were closer to the market price than that of the buy-
ers, while buyers’ expectations were closer to the
equal-pro?t price. One likely e?ect of a ‘transfer
price expectation gap’ between buyers and sellers
is a prolonged and ine?cient negotiation process.
While this may be avoided by the intervention of
top management to mediate any inter-divisional
dispute, such an approach would undermine the
autonomy of decentralised divisional managers.
Instead, if we have a better understanding of those
factors that in?uence managers’ transfer price judg-
ments, we may be able to overcome managers’
biases by re-designing the negotiation process.
Prior research in psychology suggests that the
key to understanding how managers make negoti-
ation judgments is to examine the way in which
managers de?ne their negotiation context, and
their perception of variables that are critical and
endogenous to the negotiation process (Bazerman,
Curhan, Moore, & Valley, 2000; Ghosh & Boldt,
2004; Kristensen & Garling, 1997; Neale & Bazer-
man, 1992). Neale and Bazerman (1992) in partic-
ular have argued that:
‘‘Rather than focus only on external factors
[to the negotiation process], it may be most
useful to view situations from an interpretive
perspective. It may not be the objective,
external aspects of the situation that directly
a?ect negotiator judgment; instead, it may be
the way that the negotiator perceives these
features and uses those perceptions to inter-
pret and screen information.” (Neale & Bazer-
man, 1992, p. 161, emphasis added).
Two factors that are of particular interest in the
current study are the goal frame adopted by man-
agers, which a?ects the way managers perceive the
negotiation outcome, and the negotiation part-
ner’s objective (also called ‘social concern’) which
a?ects the way managers perceive the negotiation
partner. Both of these variables are found to be
important in the psychology and economics litera-
ture (e.g. Kahneman & Tversky, 1979; Lewicki,
Saunders, & Barry, 2005; Neale & Bazerman,
1992; Roth, 1995), but are generally controlled
for rather than manipulated in prior accounting
studies. For example, both Luft and Libby
(1997) and Kachelmeier and Towry (2002)
adopted a consistent positive goal frame in all their
treatments, and controlled for negotiation part-
ners’ objectives by telling their participants that a
positive relationship existed between negotiators.
1
This is in contrast with ‘self-interest’, which refers to a
negotiator’s motivation to advance their own outcomes. Indi-
viduals with a high level of self-interest do not necessarily have
a biased view of what constitutes a fair outcome; rather, they
are motivated to achieve a favourable outcome for themselves.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 705
We extend these earlier studies by examining
the impact of these variables on managers’ self-
serving biases in a transfer pricing setting. There
are bene?ts in studying these two variables simul-
taneously. We suggest that the reason the loss
frame a?ects negotiation judgments is because it
causes managers to become more concerned about
their own outcome (not to incur any further
losses), exacerbating their self-serving bias. Their
negotiation partner’s objective also is expected to
in?uence the level of concern managers have for
their own outcome. For example, a perception that
the negotiation partner has high concern-for-oth-
ers causes managers to be more willing to give
up some of their divisional pro?t and accept a less
favourable transfer price. By using both cognitive
and social lenses together, we attempt to obtain
a more uni?ed understanding of how the negotia-
tion process works and eventually, how to over-
come barriers to e?ective negotiation.
The study of these variables is important because
of their implications both for transfer pricing and
for their impact more generally on systems and pro-
cesses in decentralised organisations that use man-
agement accounting information. Transfer price
negotiations, in particular, allowbusiness unit man-
agers to utilise and share their local knowledge
(Dikoli & Vaysman, 2006) and maintain inter-divi-
sional coordination while preserving autonomy
(van Helden, van der Meer-Kooistra, & Scapens,
2001). The cost of negotiation, however, is not neg-
ligible, and the negotiation approach to transfer
price determination is only recommended when
the cost of bargaining is relatively low (Dikoli &
Vaysman, 2006).
2
A number of accounting studies
(e.g. Kachelmeier & Towry, 2002; Luft & Libby,
1997) have demonstrated that the self-serving bias
is one factor that can reduce the accuracy of manag-
ers’ transfer price judgments and thus potentially
increase the time and costs of negotiation. An
understanding of framing and negotiation partner’s
objective also has wider implications for the man-
agement accounting literature and these implica-
tions are included in our ‘Discussion’ section.
In summary, our study makes a number of sig-
ni?cant contributions to the accounting literature.
First, we extend the Luft and Libby’s (1997)
results by investigating the in?uence of managers’
perception of the negotiation context on transfer
price judgments. We speci?cally address the role
of framing and the negotiation partner’s objective.
The ?rst factor is directly controllable by manage-
ment accountants. For example, management
accountants can produce reports based on alterna-
tive negotiating reference points to support a
seller–manager involved in a transfer price negoti-
ation. When the market price is used as a reference
point, the management accounting reports are
likely to highlight the potential loss in pro?t as
the negotiated transfer price falls below the market
price (Perera, McKinnon, & Harrison, 2003). This
will lead to the negotiating manager adopting a
loss frame. Alternatively, the reports can use prod-
uct costs as a reference point, focusing on the gains
in pro?t as the negotiated transfer price moves
above the product costs (Colbert & Spicer, 1995).
This is likely to cause the negotiating manager to
adopt a gain frame.
Second, the importance of social considerations
was highlighted by both Luft and Libby (1997) and
Kachelmeier and Towry (2002) when they found
evidence of the e?ect of fairness concerns on trans-
fer price judgments. Building on this research, we
demonstrate, in a situation where market prices
are above equal-pro?t prices, that managers expect
the ?nal transfer price to be lower when they are
dealing with a partner with high concern-for-others
than when negotiating with a partner with low con-
cern-for-others.
3
This is because managers tend to
2
In addition, while the perception of a ‘fairer’ outcome can
result in a more positive feeling at the end of the negotiation
process (Lewicki et al., 2005), this perception may have negative
consequences for ?rms as divisions use ‘pro?t equality’ as an
argument for ‘fairer’ outcomes. As the gap between equal-pro?t
price and market price grows bigger, the pursuit of pro?t
equality may give rise to an ‘internal socialism’ problem where
?rms ine?ciently try to equalise divisional performance (Bolton
& Scharfsein, 1998). The resultant impact is the distortion of
pro?ts as a result of managers’ pursuit of pro?t equality.
3
In this study, we use an example where market price is
above the equal-pro?t price and therefore concern-for-others
results in transfer prices below market prices. We note that
direction of the di?erence between market price and transfer
price would reverse if the market price given in an experiment
was lower than the equal-pro?t price.
706 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
reciprocate the perceived social concerns of their
negotiation partner. The lower price, however, is
also further from the market price which has impli-
cations for production divisions that depend on the
transfer price and may have longer term implica-
tions if managers are put under increased pressure
to reach pro?t targets. Deviations from the market
price also have potential negative implications on
intra-organisational resource allocation (Bolton &
Scharfsein, 1998). As noted by Sprinkle (2003), it
is important to study the extent to what social
motives, and other aspects of a ?rm’s information
systems, interact with the more formal accounting
systems to a?ect managerial behaviour. Our results
imply that these two factors signi?cantly in?uence
the way managers make use of accounting informa-
tion when making transfer price judgments.
Third, our study extends the existing literature
by examining the impact of the above variables
on two dimensions of transfer pricing judgments:
a reservation price and a price premium (i.e. the
di?erence between the reservation price and the
estimated transfer price). Our results show that a
loss frame increases the sellers’ reservation price,
and thus eventually their ?nal transfer price judg-
ment. In contrast, negotiation partner’s objective
did not a?ect reservation price judgment, but
rather, we found that sellers who perceived their
partner to have a high level of concern-for-others
were more willing to accept a lower price premium.
Finally, inter-divisional negotiation (such as
transfer price negotiations) is an important control
mechanism that balances divisional autonomy
with inter-divisional coordination (van Helden
et al., 2001). Our study extends the growing litera-
ture on improving negotiation outcomes in
accounting/auditing situations (Bame-Aldred &
Kida, 2007; Gibbins, McCracken, & Salterio,
2005; Gibbins, Salterio, & Webb, 2001; Ng &
Tan, 2003; Trotman et al., 2005) to the manage-
ment accounting arena, and in doing so, contrib-
utes to our understanding of the challenges faced
by decentralised organisations.
Literature review and hypotheses development
Conventional economic arguments suggest that
transfer price judgments should be based on ‘eco-
nomically rational’ concerns such as the market
price, transaction costs and the division’s cost
structure (e.g. Colbert & Spicer, 1995). However,
prior literature in psychology has demonstrated
that negotiators do not always act ‘rationally’.
Rather, they su?er from a number of judgmental
biases, such as anchoring their decisions on irrele-
vant information, and the escalation of commit-
ment (e.g. Bazerman & Neale, 1992; Neale &
Bazerman, 1992; Northcraft & Neale, 1987).
In the accounting literature, Luft and Libby
(1997) have shown that, during transfer price nego-
tiation, sellers’ estimates of negotiated transfer
prices tend to be signi?cantly higher than those of
buyers, particularly when the market price is higher
than the equal-pro?t price. Luft and Libby (1997)
argue that their ?nding demonstrates the existence
of a ‘self-serving bias’, which causes managers to
overweigh the negotiation outcome that is most
bene?cial to them (Luft & Libby, 1997; Thompson
& Loewenstein, 1992). Thus, where more than one
de?nition of a ‘fair’ transfer price exists (e.g. in the
Luft & Libby, 1997 study, where market price was
higher than the equal-pro?t price),
4
negotiating
managers will interpret fairness in ways that favour
their position, such that the transfer price estimates
by sellers are signi?cantly higher than the transfer
price estimates by buyers. Before developing our
hypotheses we ?rst replicate the baseline condition
established in Luft and Libby (1997) on the di?er-
ence in transfer price judgments between sellers
and buyers resulting from the self-serving bias.
H1: Sellers’ estimated ?nal transfer prices are
higher than buyers’ estimated ?nal transfer
prices.
‘Frames’ are subjective cognitive systems
through which individuals evaluate and make sense
4
The prevailing market price is often perceived as a fair
transfer price because it is the result of ‘impartial’ market forces
of supply and demand. On the other hand, a fair price can also
be de?ned as one that provides equal pro?t to both negotiating
divisions (i.e. the equal-pro?t price). The concept of equal-
pro?t price is likely to be particularly salient in the internal
transfer price negotiation process, where the negotiating man-
agers belong to the same company, and inter-divisional equity
becomes an important concern.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 707
of situations they are in. Di?erent frames adopted
can lead individuals to pursue or avoid subsequent
actions (Lewicki et al., 2005). Traditionally, negoti-
ation literature has focused on the e?ect of ‘risk
preference framing’ (Bottom & Studt, 1993; Kahn-
eman & Tversky, 1979; Neale & Bazerman, 1985;
Thaler, 1992) – a framing e?ect characterised by a
choice between outcome certainty and a more risky
alternative. The underlying decision valence is then
manipulated, such that a loss frame is represented
by uncertainties surrounding a negative conse-
quence, and a gain frame is represented by uncer-
tainty surrounding a positive consequence.
5
For example, Neale and Bazerman (1985) inves-
tigated the risk frame in a management/union
negotiation by either telling the participants that
any concessions made by the company will result
in signi?cant ?nancial losses (loss frame), or any
concessions from the union will result in signi?cant
?nancial gains (gain frame). All impasses were to
be referred to an arbitrator, and as the arbitrator’s
?nal decision was unknown, this presented the
participants with a risk element. Inter alia, they
found that compared to negotiators with gain
frames, negotiators with loss frames were more
likely to have their agreements determined by the
arbitrator (i.e. they chose the riskier option). They
also found that compared to gain framed negotia-
tors, loss framed negotiators were less likely to
make concessions.
In this study, we examine the framing role of
accounting information, and how this a?ects man-
agers’ transfer price judgments. Our focus is on
using accounting information to frame the negoti-
ation goal. We propose that because managers are
more concerned with avoiding losses than increas-
ing gains both buyers and sellers are more likely to
focus on maximising their divisional pro?t when
given a loss frame compared to a gain frame. This
greater concern for achieving their own outcome is
likely to further increase the transfer price judge-
ment gaps between buyers and sellers.
Prior literature on motivated reasoning suggests
that a higher level of motivation to achieve an out-
come can lead people to overestimate the probabil-
ity that a favourable outcome will eventuate
(Brownstein, 2003). Further, motivated reasoning
also distorts people’s perception of others, such
that they tend to expect others to behave in a
way that results in favourable outcomes (Kunda,
1990). In the context of a negotiation, we predict
that the loss framed managers’ greater concern
for maximising their divisional pro?t will cause
them to overestimate the likelihood that their part-
ner will take their view of what constitutes a fair
price, and thus agree on a transfer price more
favourable to them. Speci?cally, sellers (buyers)
with a loss frame are more likely to believe that
their partner will agree on a higher (lower) price
being a fair transfer price, compared to sellers with
a gain frame. As such, we predict that a loss frame
will increase negotiators’ self-serving biases.
In addition, as loss framed managers become
more motivated to achieve a better outcome, they
may be more willing to incur greater bargaining
costs compared to gain framed managers. In the
absence of any information about their partner’s
negotiation frame (and thus the level of their part-
ner’s motivation), the loss framed managers are also
likely to expect their willingness to incur greater
bargaining costs will lead to a more favourable out-
come. As such, we predict that the transfer price
judgement gap between buyers and sellers is greater
under the loss frame than the gain frame condition.
6
5
While the major emphasis in the framing literature has
centred on the standard risky choice framing a?ect introduced
by Kahneman and Tversky (1979) and Tversky and Kahneman
(1981), Levin et al. (1998) developed a typology to distinguish
between risky choice framing, attribute framing and goal
framing. Levin, Schneider, and Gaeth (1998) suggest that the
di?erent operational de?nitions of framing have e?ects that rely
on di?erent psychological processes.
6
As we do not manipulate or provide information about the
negotiation partner’s frame, buyers and sellers may not be
making the same transfer price predictions. For example, the
seller in the loss frame is predicting the price that a loss framed
seller and a buyer with unknown or neutral frame will
negotiate, whereas the buyer in the loss frame is predicting
the price that a loss framed buyer and a seller with an unknown
or neutral frame will negotiate. Even if both buyers and sellers
assume the same (e.g. neutral) frame for their negotiation
partner, the di?erences in transfer price predictions by sellers
and buyers are still not necessarily all self-serving bias and may,
in fact, be partially due to the lack of information about their
negotiation partner’s frame.
708 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
H2: The di?erence in estimated ?nal transfer
price between buyers and sellers is smaller when
information provided to negotiating managers
is framed as gains rather than losses.
A number of prior studies have suggested that
social concerns in?uence transfer price negotiation
judgments (e.g. Kachelmeier & Towry, 2002;
Luft & Libby, 1997). Both Luft and Libby (1997)
and Kachelmeier and Towry (2002) found that
while economic rationality would dictate that
negotiators should expect market-based transfer
prices, negotiators have an aversion to unequal
pro?ts. While they attributed this aversion to
negotiators’ concerns about pro?t sharing and
ensuring both divisions receive satisfactory pro?ts,
the impact of social concern was not directly
tested. In this study, we seek to directly examine
the e?ects social concerns have on transfer price
judgements.
We propose that managers’ concerns about
unequal pro?ts and therefore their transfer price
judgments may be a?ected by the perception they
have of their negotiation partner (i.e. the other
party to the negotiation process). In particular,
during negotiation, managers would try to gauge
their partner’s objective, and then combine this
information with their own negotiation objective
when formulating their transfer price judgments
(e.g. Carroll, Bazerman, & Maury, 1988; Lewicki
et al., 2005).
An established framework used to explain a
negotiator’s objective is the ‘dual concern model’
(e.g. Lewicki et al., 2005; Pruitt, 1983; Sorenson,
Morse, & Savage, 1999). This framework postu-
lates that a negotiator’s objective is in?uenced by
two independent types of concerns: concern for
their own outcomes (‘concern-for-self’) and con-
cern for the other party’s outcomes (‘concern-
for-others’). Our focus in the current study is on
a manager’s perception of their partner’s degree
of concern-for-others. Our manipulation of this
variable is consistent with large variations of con-
cern-for-others in transfer pricing situations; for
example, the level of concern for the pro?ts of
other divisions is likely to vary in organisations
that are quasi-markets compared to quasi-families
(Eccles, 1985, pp. 273–278).
Following the suggestion in previous research
(Kachelmeier & Towry, 2002; Luft & Libby,
1997), in situations where the market price is
higher than the equal-pro?t prices, that this con-
cern-for-others results in transfer prices below
market prices, we predict that when the level of
concern-for-others is stronger, both buyer and
seller will expect the price to be lower. We note
that this prediction only holds for situations where
the market price is higher than the equal-pro?t
price, which is the case in our experiment.
The above prediction particularly applies when
the level of concern is similar for both negotiators
in the pair. While we only manipulate the level of
concern-for-others for the negotiation partner, we
suggest this is likely to result in a similar level of
concern for the negotiating manager for two rea-
sons. First, the psychology literature refers to the
‘reciprocity’ principle as a social norm by which
an individual who acts in a certain way will expect
a similar return action (e.g. Maxwell, Nye, & Max-
well, 2003).
7
The norm of reciprocity therefore
establishes expectations about how one is to
behave in social interactions (Maxwell et al.,
2003). Prior research has consistently found that
negotiators have a tendency to reciprocate negotia-
tion motives of their negotiation partners (Maxwell
et al., 2003). Therefore, negotiating managers who
perceive that their negotiating partner has high
concern-for-others will reciprocate with a similar
objective, showing high concern for pro?t sharing,
In contrast, negotiating managers who perceive
that their partner has low concern-for-others are
expected to reciprocate by showing low concern
for pro?t sharing.
Second, organisations di?er in the types of
employee behaviours that are considered accept-
able. For example, our low concern-for-others
manipulation would be acceptable in some organ-
isations but clearly unacceptable in other organisa-
tions. By informing a participant about the
negotiation partner’s concerns-for-others we also
tell participants something about the culture of
the organisation. Speci?cally, our manipulation
7
The reciprocity principle has also recently been introduced
into the audit literature in audit–client negotiations (Sanchez,
Agoglia, & Hat?eld, 2007; Tan & Trotman, 2007).
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 709
of concern-for-others involves providing partici-
pants with a memo from their negotiation partner.
In the low concern-for-others treatment, the memo
includes references to ‘maximising pro?t of my
division’ which also communicates to the other
party that this is accepted practice in this organisa-
tion. In contrast, in the high concern-for-others
treatment, the culture encourages both divisions
to receive satisfactory pro?ts and therefore both
parties would expect a lower price.
H3: Managers’ estimated transfer prices are
lower when they are negotiating with a partner
with high concern-for-others than when they
are negotiating with a partner with low con-
cern-for-others.
Research methods
Research design
A controlled laboratory experiment was con-
ducted to test the proposed hypotheses, using a
2 Â 2 Â 2 between-subjects design. The three inde-
pendent variables were the negotiating manager’s
role (participants acting as either a buyer or a
seller), goal frame (gain frame or loss frame) and
the negotiation partner’s objective (high or low
concern-for-others).
Experimental task
The experimental task was modi?ed
8
from Luft
and Libby’s (1997) instrument, where participants
assumed the role of a manager who is responsible
for negotiating a transfer price of component
‘Parts’. ‘Parts’ are components sold by the Parts
Division to the Assembly Division, which can then
be processed further by the Assembly Division and
then sold to an external customer. As the two divi-
sions are autonomous, both divisional managers
are free to negotiate a mutually acceptable transfer
price or to trade externally at the prevailing mar-
ket price (which was set at $70 per unit).
9
The cost
structures of the two divisions were designed such
that the equal-pro?t price was $50.
10
Included in
the task was a pro?t schedule illustrating the pro?t
implications of a range of transfer prices for both
parties (between $20 where the pro?t for sellers
was zero, and $80 per unit, where the pro?t for
buyers was zero). Both buyers and sellers were
then asked to predict the ?nal negotiated transfer
price and the sellers’ reservation price.
Independent variables
The negotiation role was manipulated by ran-
domly assigning participants either to the role of
‘Parts Manager’ (i.e. seller) or ‘Assembly Man-
ager’ (i.e. buyer). The goal frame was operationa-
lised by ‘framing’ the instructions provided in the
instrument either as a gain frame or a loss frame.
Speci?cally, instructions provided to Assembly
managers (the buyers) assigned a gain frame were
as follows:
‘‘As you can see from the table, for every $5
decrease in transfer price you stand to gain
$5000 pro?t. For example, by negotiating a
transfer price of $55, your pro?t is $25,000.
But if you negotiate a lower transfer price,
say, $50, your pro?t is $30,000, which means
that you have gained $5000 pro?t. In other
words, as you settle for a lower transfer
price, you stand to gain pro?t for your divi-
sion in $5000 increments.”
8
Two main modi?cations were made to the Luft and Libby
(1997) instrument to accommodate two of our variables of
interest (discussed in more detail later). A pilot test was then
conducted (with 21 undergraduate accounting students) to
ensure that our modi?cations were understood by our partic-
ipants and that the independent manipulated variables had the
intended e?ects. As a result of the pilot test, and discussions
with pilot test participants post-experiment, further minor
modi?cations were made.
9
The current task focuses on a ‘distributive’ (‘?xed pie’)
negotiation task, that is, the managers are negotiating in a win–
lose situation where their goals are in direct con?ict.
10
Speci?cally, based on Luft and Libby’s (1997) scenario, the
value of the shipment of ‘Parts’ to the Assembly Division was
$80 per unit; while the value (or cost) to the Parts Division was
$20 per unit. This means that the pro?t for the Assembly
department was ($80 less negotiated transfer price); while the
pro?t for the Parts Division was (transfer price less $20). Thus,
at a transfer price of $50 per unit, both divisions would obtain a
pro?t of $30 per unit.
710 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
For participants assigned a loss frame, the
description explained how every $5 change in the
transfer price would result in the division losing
$5000. Further, participants were also provided
with a schedule of pro?t framed either as pro?t
increases or pro?t decreases as the transfer price
changed (refer to Fig. 1).
To manipulate the negotiation partner’s objec-
tive, participants were provided with a ?ctitious
memo indicating their negotiation partner’s level
of concern-for-others. Participants assigned to
the ‘high concern-for-others’ conditions were given
a memo stressing their partner’s desire for mutual
concession and maximising pro?t for both divi-
sions. In contrast, participants in the ‘low con-
cern-for-others’ conditions were given a memo
emphasising their partner’s desire to maximise
pro?t for their own division only and their unwill-
ingness to make concessions. For example, the
memo from a partner with low concern-for-others
highlighted their intention to ‘‘. . . achieve the best
pro?t for my division. . . if you are unwilling to
make concessions, I am prepared to trade
externally.”
Dependent variables
We measured the dependent variable, manag-
ers’ estimated negotiation prices, by asking partic-
ipants to predict the ?nal transfer price of the
negotiation process. In addition, participants were
also asked to indicate the expected lowest price
that sellers would likely to be willing to accept
(i.e. seller’s reservation price).
11
Consistent with
Luft and Libby (1997), to minimise the time
required for data collection we did not ask partic-
ipants to estimate buyers’ reservation price.
Participants
One hundred and twenty-eight participants vol-
unteered to participate in this experiment. All par-
ticipants were enrolled in a Master of Commerce
degree or Master of Business Technology at one
Australian university, and each had at least two
years of full time work experience. However, 32
participants failed one or more post-experiment
manipulation tests and were later excluded from
the analysis, resulting in 96 usable responses. The
cell sizes for each of the eight treatment group var-
ied between 11 and 15 (see Table 1).
Manipulation check and post-test measures
After participants completed the negotiation
task, they were given three manipulation checks.
The ?rst asked participants to indicate what role
they played in the negotiation (i.e. whether they
A: Profit schedule for Assembly managers (i.e. buyers)/Gain frame information
Transfer price for Parts 80 75 70 65 60 55 50 45 40 35 30 25 20
PARTS profit ($000) 60 55 50 45 40 35 30 25 20 15 10 5 0
ASSEMBLY profit ($000) 0 5 10 15 20 25 30 35 40 45 50 55 60
B: Profit schedule for Assembly managers (i.e. buyers)/Loss frame information
Transfer price for Parts 20 25 30 35 40 45 50 55 60 65 70 75 80
PARTS profit ($000) 0 5 10 15 20 25 30 35 40 45 50 55 60
ASSEMBLY profit ($000) 60 55 50 45 40 35 30 25 20 15 10 5 0
Fig. 1. Sample pro?t schedules provided to experimental participants.
11
Luft and Libby (1997) pointed out that negotiated transfer
price is a more sensitive measure of potential con?ict, while
reservation price is more sensitive to managers’ mistaken
judgments about their negotiation partner. As we are primarily
interested in the former (i.e. the e?ect of managers’ perception
on their expectations of how the ‘negotiation con?ict’ would be
resolved), our primary analysis will focus on estimated ?nal
transfer prices.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 711
were acting as a Parts manager or an Assembly
manager). The second asked participants to indi-
cate whether their negotiation partners were inter-
ested in maximising both divisions’ pro?ts, or only
their own division’s pro?t. The third asked partic-
ipants to indicate whether the case material stated
that ‘‘for every $5 increase in transfer price you
stand to lose $5000 pro?t”, or ‘‘for every $5
decrease in transfer price you stand to gain
$5000 pro?t”.
12
Results
Hypothesis testing
The descriptive statistics for estimated transfer
price are summarised in Table 1, and a 2 Â 2 Â 2
ANOVA model, with estimated transfer price as
the dependent variable, is presented in Table 2.
As can be seen from Table 1, and consistent with
H1, the average estimated transfer price was
higher for sellers (62.40) than for buyers (56.87).
This di?erence (the main e?ect of role) is statisti-
cally signi?cant (F = 8.71, p = 0.00), thus H1 is
supported.
H2 predicted that the di?erence in estimated
transfer prices between buyers and sellers would
be smaller when potential negotiation outcomes
are framed as gains rather than losses. The descrip-
tive statistics in Table 1 further indicate that the
di?erence in estimated transfer prices between sell-
ers and buyers under the gain frame condition
(60.48 À 58.83 = 1.65) was lower than that under
the loss frame condition (63.84 À 55.00 = 8.84).
This di?erence is shown in Table 2 as a signi?cant
interaction e?ect between role and goal frame
(F = 4.26, p = 0.02), thus H2 is supported.
H3 examined the e?ect of the negotiation part-
ner’s objective on managers’ transfer price judg-
ments. In H3, we expected the negotiation
partner’s objective to have a main e?ect, where
both buyers’ and sellers’ transfer price expecta-
tions would be lower if they were negotiating with
Table 1
Means (standard deviation) of estimated ?nal transfer price ($)
Partner’s objective Frame total Total
High concern-for-others Low concern-for-others
Gain frame Loss frame Total Gain frame Loss frame Total Gain frame Loss frame
Sellers 58.64 61.92 60.42 62.50 65.50 64.30 60.48 63.84 62.40
(7.10) (8.04) (7.65) (9.50) (6.96) (7.89) (8.35) (7.44) (7.93)
n = 11 n = 13 n = 24 n = 10 n = 15 n = 25 n = 21 n = 28 n = 49
Buyers 57.69 52.73 55.42 60.30 56.92 58.39 58.83 55.00 56.87
(8.32) (10.57) (9.55) (7.51) (10.52) (9.29) (7.91) (10.53) (9.44)
n = 13 n = 11 n = 24 n = 10 n = 13 n = 23 n = 23 n = 24 n = 47
Column total 58.13 57.71 57.92 61.40 61.52 61.47 59.61 59.76 59.69
(7.63) (10.21) (8.92) (8.41) (9.56) (9.01) (8.01) (9.95) (9.09)
n = 24 n = 24 n = 48 n = 20 n = 28 n = 48 n = 44 n = 52 n = 96
Table 2
ANOVA model for estimated transfer price – H1 and H2
DF MS F p
Negotiator’s role 1 644.11 8.71 0.00
a
Partner’s objective 1 298.70 4.04 0.02
Frame 1 6.22 0.08 0.39
Role
*
objective 1 0.60 0.01 0.46
Role
*
Frame 1 315.05 4.26 0.02
b
Objective
*
frame 1 2.49 0.03 0.43
Role
*
objective
*
frame 1 5.16 0.07 0.40
Error 88 73.93
Negotiator role – participants acting as either a buyer or a
seller.
Partner’s objective – either high or low concern-for-others.
Frame – accounting information presented in either a gain or a
loss goal frame.
a
The signi?cant main e?ect of negotiator’s role provides
support for H1.
b
The signi?cant interaction e?ect between frame and role
provides support for H2.
12
56% of the manipulation test errors related to the framing
e?ect. All statistical tests were re-run after including partici-
pants who failed the manipulation tests, and all results
remained statistically the same.
712 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
a partner with high concern-for-others than if with
a partner who had low concern-for-others. The
overall ANOVA model as shown in Table 2
con?rmed our expectation (signi?cant main e?ect
for negotiation partner’s objective, F = 4.04, p =
0.02) thus, H3 is supported.
Additional analysis
We also conducted additional analyses to delin-
eate the e?ect of goal framing and the negotiation
partner’s objective on two elements of managers’
transfer price judgments: the reservation price
and the di?erence between the reservation price
and the estimated transfer price (which we refer
to as ‘price premium’). The reservation price (also
known as the resistance point) represents the min-
imal price sellers are willing to accept from the
transaction (e.g. for sellers, this means the minimal
acceptable price – Lewicki et al., 2005). In con-
trast, the price premium re?ects the extent to which
negotiators expect to achieve their desired out-
come. That is, the price premium incorporates
negotiators’ anticipation of the concession they
will make during the o?er–countero?er process in
negotiation. The descriptive statistics of these
two variables are shown in Tables 3 and 4.
Table 3 Panel A indicates that on average sell-
ers’ reservation price was $54.04, which was higher
than the equal-pro?t price of $50.00 (signi?cant
with one-sample t-test, t = 2.464, p = 0.01). This
is consistent with our expectation that sellers in
general would not consider the equal-pro?t price
as a fair outcome of negotiation. Instead, their
minimum acceptable price was signi?cantly higher.
Table 3 (Panel A) also shows that compared to
their gain frame counterparts, sellers in the loss
frame condition reported a higher reservation
price ($56.54 vs. $50.71). The main e?ect for goal
frame in the ANOVA reported in Table 3 Panel
B shows that this di?erence is statistically signi?-
cant (F = 6.33, p = 0.02). Consistent with our ear-
lier argument, this ?nding suggests that a loss
frame focuses individuals on a goal of avoiding
negative consequences, thus increasing their ‘resis-
tance point’ to an unfavourable transfer price,
which results in a higher expected reservation
price. Neither negotiation partner’s objective
(F = 0.01, p = 0.94) or the interaction with fram-
ing (F = 1.06, p = 0.31) are signi?cant for reserva-
tion price.
Table 4 reports the descriptive statistics and
ANOVA results for the sellers’ price premium. A
Table 3
Additional analysis – sellers’ reservation prices
Partner exhibits high concern-for-others Partner exhibits low concern-for-others Total
Panel A: Means (standard deviation) of sellers’ reservation prices
Gain frame 52.27 49.00 50.71
(7.20) (16.13) (12.07)
n = 11 n = 10 n = 21
Loss frame 56.15 56.87 56.54
(8.20) (12.51) (10.55)
n = 13 n = 15 n = 28
Total 54.38 53.72 54.04
(7.85) (14.29) (11.48)
n = 24 n = 25 n = 49
DF MS F p
Panel B: ANOVA model
a
(dependent variable = sellers’ reservation price)
Partner’s objective 1 0.56 0.01 0.94
Frame 1 641.74 6.33 0.02
Partner’s objective
*
frame 1 107.86 1.06 0.31
Error 43 101.40
a
Two outliers were excluded from this analysis. One outlier was excluded because the subject reported a reservation price that was
higher than the expected transfer price. A second outlier was excluded because the reported reservation price was greater than 3
standard deviations away from the mean.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 713
marginally signi?cant main e?ect of the negotia-
tion partner’s objective (F = 3.24, p = 0.08) sug-
gests that sellers who were negotiating with a
high concern-for-others partner expected to give
up a greater share of divisional pro?t during the
negotiation process and thus predicted a lower
price premium compared to those who were nego-
tiating with a partner with low concern-for-others.
Neither framing (F = 2.10, p = 0.16) or the inter-
action (F = 2.53, p = 0.12) are signi?cant.
13
Together, these results show that the goal frame
and the negotiation partner’s objective have di?er-
ent e?ects on di?erent aspects of managers’ trans-
fer price judgments. Speci?cally, as individuals are
more resistant to avoiding losses than increasing
gains, a loss frame increases the sellers’ reservation
price and eventually, their ?nal estimated transfer
price. On the other hand, the negotiation partner’s
concern-for-others provides the sellers with an
indication of the potential o?ers/countero?ers dur-
ing transfer price negotiation, thus in?uencing the
price premium the sellers expect on top of their
reservation price.
Summary and discussion
In this study, we examined whether managers’
perceptions of potential negotiation outcomes
(framed either as potential gains or potential
losses) and of their negotiation partner (exhibiting
high or low concern-for-others) a?ected self-serv-
ing biases and consequently their transfer price
judgments. We found that compared to a gain
frame, a loss frame exacerbates managers’ self-
serving biases and increases the transfer price
expectation gap between buyers and sellers.
Further, we found that the negotiation part-
ner’s objective had a signi?cant impact on sellers’
transfer price judgments. Consistent with the
‘norm of reciprocity’, our results show that, in sit-
uations where market prices are higher than equal-
pro?t prices, managers reciprocated their partner’s
concerns and expected lower transfer prices when
their negotiation partner exhibited high concern-
for-others, and expected higher transfer prices
when their negotiation partner exhibited low con-
cern-for-others. This ?nding is particularly inter-
esting as sellers in our experiment had relatively
strong bargaining power but these sellers did not
exploit their bargaining power by demanding high
transfer prices regardless of their partner’s level of
concern-for-others. Instead, we found that sellers
Table 4
Additional analysis – sellers’ price premium
Partner exhibits high concern-for-others Partner exhibits low concern-for-others Total
Panel A: Descriptive statistics – sellers’ transfer price premium
Gain frame 6.36 13.50 9.76
(6.36) (10.55) (9.15)
n = 11 n = 10 n = 21
Loss frame 5.77 8.63 7.30
(6.41) (8.09) (7.37)
n = 13 n = 15 n = 28
Total 6.04 10.58 8.36
(6.25) (9.27) (8.18)
n = 24 n = 25 n = 49
DF MS F p
Panel B: ANOVA model
a
(dependent variable = price premium)
Partner’s objective 1 166.08 3.24 0.08
Frame 1 107.29 2.10 0.16
Partner’s objective
*
frame 1 129.70 2.53 0.12
Error 43 51.21
a
Price premium = (sellers’ reservation price – transfer price estimate).
13
Due to the relatively small sample size, however, the
statistical inferences of our additional analysis should be
interpreted with care.
714 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
were sensitive to their partner’s objective, and were
more willing to accept a less advantageous out-
come if their partner showed high concern-for-
others.
The additional analysis suggests that goal fram-
ing and the negotiation partner’s objective have
di?erent impacts on managers’ negotiation judg-
ments. When we decompose the transfer price
judgments into two subcomponents: the reserva-
tion price and a price premium, we found that
the loss frame resulted in managers reporting a
higher reservation price. On the other hand, man-
agers’ perception of their partner’s objective had
signi?cant impact on their price premiums.
Our study has important implications for both
researchers and practitioners. Prior research has
shown that negotiating managers su?er from self-
serving biases, which result in a signi?cant di?er-
ence in expected transfer prices between buyers
and sellers. We extended this line of research by
examining how these di?erences in transfer price
expectations are a?ected by managers’ perceptions
of the negotiation context. Understanding manag-
ers’ transfer price expectations is important, as dif-
ferences in expectations between buyers and sellers
can lead to prolonged disputes and thus a costly
negotiation process (Luft & Libby, 1997).
Our ?ndings that the provision of loss framed
information increases the buyer–seller expectation
gap can also have a signi?cant impact on organisa-
tions. We note that systems and processes using
management accounting can either inadvertently
or by design cause managers to adopt di?erent
frames. For example, practitioner literature often
advocates the use of customer pro?tability infor-
mation to support customer negotiation (Kaplan
& Cooper, 1998) and negotiators may be given a
‘‘price menu” listing a range of service levels and
their associated costs (Kaplan & Anderson,
2007). In such circumstances, management
accounting information can be presented in a
way that induces either a gain frame or a loss
frame. Speci?cally, management accounting
reports can either describe the incremental cost
increases with each service level (e.g. incremental
cost of $500 every time a customer requests an
additional sales visit), or the incremental cost sav-
ings (e.g. incremental costs savings of $500 per
sales visit reduced). The former is likely to induce
a loss frame and the latter a gain frame.
14
The practitioner’s ‘self-help’ literature on nego-
tiation often discusses the importance of building
rapport and a?liation at the negotiation table
(Fisher & Shapiro, 2005). Our study provides
empirical support for the importance of communi-
cating a positive objective. Our results show that
managers expect a lower transfer price (closer to
the equal-pro?t price) when they perceive that
their negotiation partners have high concern-for-
others. Our results imply that showing high
concern-for-others (as opposed to showing low
concern-for-others) can be e?ective in persuading
negotiation opponents (especially sellers) to con-
sider their perceptions.
Our additional analysis suggests that managers’
perceptions of the negotiation outcomes and of
their negotiation partners a?ect di?erent aspects
of the negotiation process. This ?nding enhances
our understanding of how to ‘de-bias’ managers’
self-biased transfer price judgments. By framing
the pro?t information di?erently we can encourage
sellers to set a lower reservation price, and at the
same time, organisations can also attempt to pro-
mote greater ‘concern-for-others’ among sellers so
that they are more likely to accept a lower ‘pre-
mium’ on top of their reservation price. For exam-
ple, incentive schemes that focus too much on the
‘stick’ rather than the ‘carrot’ may increase dis-
trust (Fehr & Ga¨chter, 2000), potentially heighten
managers’ concern-for-self relative to their con-
cern-for-others, and thus reduce managers’ will-
ingness to reciprocate positively during
negotiation.
Similar to earlier studies on transfer price nego-
tiation (Luft & Libby, 1997; Kachelmeier &
Towry, 2002), our results demonstrate a strong
desire by participants to take fairness into account
when making transfer price judgments, such that
regardless of their role or the treatment, the resul-
tant transfer price judgment is di?erent from the
external market price. On the other hand, Bolton
and Scharfsein (1998) argue that the pursuit of
14
Framing is also important in capital investment analysis as
?nancial outcomes can be presented in terms of pro?ts or losses
(e.g. Moreno, Kida, & Smith, 2002).
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 715
‘internal socialism’ can be costly, as decentralised
?rms sometimes try to equalise divisional pro?t
at the expense of resource allocation e?ciencies.
Our results show that internal socialism is also
likely to arise from inter-divisional negotiations,
potentially adding to the costs of internal
transactions.
An understanding of framing and negotiation
partner’s objective has wider implications than just
transfer pricing given that other inter-divisional
negotiation is common in decentralised organisa-
tions. For example, a production manager may
need to negotiate inventory management and
delivery policies with the marketing division; and
a research and development (R&D) manager in
one division may need to negotiate with the R&
D manager in another division over resource allo-
cation issues in collaborative projects (Coletti,
Sedatole, & Towry, 2005).
The impact of these variables on negotiations
also has implications for organisation design.
For example, larger self-serving biases result in
bigger errors in judging the outcome a bargaining
partner will accept in the end. These higher self-
serving biases have been shown to have a nega-
tive impact on reaching an agreement and create
more impasses (Babcock & Loewenstein, 1997;
Gelfand et al., 2002). Consequently, as decentra-
lised organisations rely more on negotiation
between peers, if the setting is one where large
self-serving biases are likely to be present, these
decentralised organisations will not work as e?ec-
tively. They will require greater interventions
from headquarters and more hierarchical decision
making, thus making decentralisation more costly
and less e?ective.
Acknowledgment
We gratefully acknowledge a research grant
from the Australian Research Council and the
helpful comments from Joan Luft, Sue Haka,
Kim Lang?eld-Smith, Anne Lillis, Steve Salte-
rio, Jane Baxter, Brian Bur?tt and Habib
Mahama, as well as seminar participants at Uni-
versity of Cincinnati, University of Melbourne,
2006 AFAANZ Conference and 2005 EAA
Conference.
References
Babcock, L., & Loewenstein, G. (1997). Explaining bargaining
impasse: The role of self-serving biases. Journal of Economic
Perspectives, 11(1), 109–126.
Bame-Aldred, C. W., & Kida, T. (2007). A comparison of
auditor and client negotiation decisions. Accounting, Orga-
nizations and Society, 32(6), 497–511.
Bazerman, M., Curhan, J., Moore, D., & Valley, K. (2000).
Negotiation. Annual Review of Psychology, 51, 279–312.
Bazerman, M. H., & Neale, M. A. (1992). Negotiating
rationality. New York: The Free Press.
Bolton, P., & Scharfsein, D. (1998). Corporate ?nance, the
theory of the ?rm, and organizations. Journal of Economic
Perspectives, 12(4), 95–114.
Bottom, W. P., & Studt, A. (1993). Framing e?ects and the
distributive aspect of integrative bargaining. Organizational
Behavior and Human Decision Processes, 56(3), 459–
474.
Brownstein, A. (2003). Biased precision processing. Psycholog-
ical Bulletin, 129(4), 545–568.
Carroll, J. S., Bazerman, M. H., & Maury, R. (1988).
Negotiator cognitions: A descriptive approach to negotia-
tors’ understanding of their opponents. Organizational
Behavior and Human Decision Processes, 41(3), 352–370.
Colbert, G., & Spicer, B. (1995). A multi-case investigation of a
theory of the transfer pricing process. Accounting, Organi-
zations and Society, 20(6), 423–456.
Coletti, A., Sedatole, K., & Towry, K. (2005). The e?ect of
control systems on trust and cooperation in collaborative
environment. The Accounting Review, 80(2), 477–500.
Dikoli, S., & Vaysman, I. (2006). Information technology,
organizational design, and transfer pricing. Journal of
Accounting and Economics, 41, 203–236.
Eccles, R. (1985). The transfer pricing problem. Lexington, MA:
Lexington Books.
Fehr, E., & Ga¨chter, S. (2000). Fairness and retaliation: The
economics of reciprocity. Journal of Economic Perspectives,
14(3), 159–181.
Fisher, R., & Shapiro, D. (2005). Beyond reason: Using
emotions as you negotiate. Random House Business Books.
Gelfand, M., Higgins, M., Nishii, L., Raver, L., Dominguez,
A., Murakami, F., et al. (2002). Culture and egocentric
perceptions of fairness in con?ict and negotiation. Journal
of Applied Psychology, 87(5), 833–845.
Ghosh, D. (2000). Complementary arrangements of organiza-
tional factors and outcomes of negotiated transfer price.
Accounting, Organizations and Society, 25, 661–682.
Ghosh, D., & Boldt, M. (2004). The outcome saliency e?ect on
negotiated transfer prices. Journal of Managerial Issues,
16(3), 305–321.
Gibbins, M., McCracken, S. A., & Salterio, S. E. (2005).
Negotiations over accounting issues: The congruency of
audit partner and chief ?nancial o?cer recalls. Auditing, 24,
171–193.
Gibbins, M., Salterio, S., & Webb, A. (2001). Evidence about
auditor–client management negotiation concerning client’s
716 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
?nancial reporting. Journal of Accounting Research, 39(3),
535–564.
Kachelmeier, S. J., & Towry, K. L. (2002). Negotiated transfer
pricing: Is fairness easier said than done?. Accounting
Review 77, 571–593.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An
analysis of decisions under risk. Econometrica, 47, 263–291.
Kaplan, R., & Anderson, S. (2007). Time-driven activity-based
costing. Harvard Business School Publishing.
Kaplan, R., & Cooper, R. (1998). Cost and e?ect. Harvard
Business School Publishing.
Kristensen, H., & Garling, T. (1997). The e?ects of anchor
points and reference points on negotiation process and
outcome. Organizational Behavior and Human Decision
Processes, 71(1), 85–94.
Kunda, Z. (1990). The case for motivated reasoning. Psycho-
logical Bulletin, 108(3), 480–498.
Levin, I., Schneider, S., & Gaeth, G. (1998). All frames are not
created equal: A typology and critical analysis of framing
e?ects. Organizational Behaviour and Human Decision Pro-
cesses, 76(2), 149–188.
Lewicki, R., Saunders, D., & Barry, B. (2005). Negotiation (5th
ed.). New York: McGraw Hill.
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.
Maxwell, S., Nye, P., & Maxwell, N. (2003). The wrath of the
fairness-primed negotiator when the reciprocity norm is
violated. Journal of Business Research, 56(5), 399–409.
Moreno, K., Kida, T., & Smith, J. (2002). The impact of
a?ective reactions on risky decision making in accounting
contexts. Journal of Accounting Research, 40(5), 1331–1349.
Neale, M. A., & Bazerman, M. H. (1985). The e?ects of framing
and negotiator overcon?dence on bargaining behaviors and
outcomes. Academy of Management Journal, 28(1), 34–49.
Neale, M. A., & Bazerman, M. H. (1992). Negotiator cognition
and rationality: A behavioural decision theory perspective.
Organizational Behavior and Human Decision Processes, 51,
157–175.
Ng, T. B.-P., & Tan, H.-T. (2003). E?ects of authoritative
guidance availability and audit committee e?ectiveness on
auditors’ judgments in an auditor–client negotiation con-
text. The Accounting Review, 78(3), 801–819.
Northcraft, G., & Neale, M. (1987). Experts, amateurs, and real
estate: An anchoring-and-adjustment perspective on prop-
erty pricing decisions. Organizational Behaviour and Human
Decision Processes, 39, 84–97.
Perera, S., McKinnon, J. L., & Harrison, G. L. (2003).
Di?usion of transfer pricing innovation in the context of
commercialization – A longitudinal case study of a govern-
ment trading enterprise. Management Accounting Research,
14(2), 140–164.
Pruitt, D. G. (1983). Strategic choice in negotiation. The
American, 27(2), 167–194.
Roth, A. E. (1995). Bargaining experiments. The handbook of
experimental economics. Princeton: University Press.
Sanchez, M. H., Agoglia, C. P., & Hat?eld, R. C. (2007). The
e?ect of auditors’ use of a reciprocity-based strategy on
auditor–client negotiations. Accounting Review, 82(1),
241–263.
Sorenson, R. L., Morse, E. A., & Savage, G. T. (1999). A test of
the motivations underlying choice of con?ict strategies in
the dual-concern model. International Journal of Con?ict
Management, 10(1), 25–44.
Sprinkle, G. B. (2003). Perspectives on experimental research in
managerial accounting. Accounting, Organizations and Soci-
ety, 28(2-3), 287–318.
Tan, H.-T, & Trotman, K. T. (2007). E?ects of auditors’
concession timing on ?nancial o?cers’ negotiation judg-
ments. Working paper, The University of New South Wales.
Thaler, R. H. (1992). The winner’s curse: Paradoxes and
anomalies of economic life. Princeton, NJ: Princeton Uni-
versity Press.
Thompson, L., & Loewenstein, G. (1992). Egocentric interpre-
tations of fairness and interpersonal con?ict. Organizational
Behavior and Human Decision Processes, 51(2), 176–197.
Trotman, K. T., Wright, A. M., & Wright, S. (2005). Auditor
negotiations: An examination of the e?cacy of intervention
methods. The Accounting Review, 80(1), 349–368.
Tversky, A., & Kahneman, D. (1981). The framing of decisions
and the psychology of choice. Science, 211, 453–463.
van Helden, G., van der Meer-Kooistra, J., & Scapens, R.
(2001). Co-ordination of internal transactions at Hoogovens
Steel: Struggling with the tension between performance-
oriented business units and the concept of an integrated
company. Management Accounting Research, 12, 357–386.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 717
doc_725418399.pdf
A common approach to set transfer prices is via intra-firm negotiation. However, Luft and Libby [Luft, J. L., &
Libby, R. (1997). Profit comparisons, market prices and managers’ judgments about negotiated transfer prices. The
Accounting Review, 72(2), 217–229] found that because of the existence of self-serving biases, negotiating managers have
different expectations regarding what constitutes a ‘fair’ transfer price, leading to a less efficient negotiation process. In
this study, we examine two factors that are expected to affect managers’ transfer price negotiation judgments, namely,
framing as a gain or as a loss and the negotiation partner’s objective
The e?ect of framing and negotiation partner’s objective
on judgments about negotiated transfer prices
Linda Chang, Mandy Cheng, Ken T. Trotman
*
School of Accounting, The University of New South Wales, Sydney 2052, Australia
Abstract
A common approach to set transfer prices is via intra-?rm negotiation. However, Luft and Libby [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] found that because of the existence of self-serving biases, negotiating managers have
di?erent expectations regarding what constitutes a ‘fair’ transfer price, leading to a less e?cient negotiation process. In
this study, we examine two factors that are expected to a?ect managers’ transfer price negotiation judgments, namely,
framing as a gain or as a loss and the negotiation partner’s objective (whether the partner’s objective involves high or
low concern-for-others). We propose that these two factors a?ect managers’ perceptions of the negotiation context, and
thus the way they interpret the economic and social consequences of accounting information. Our results show that a
loss frame (compared to a gain frame) exacerbates managers’ self-serving biases and increases the ‘transfer price expec-
tation gap’ between buyers and sellers. Further, in our experiment where market price is higher than equal-pro?t price,
we ?nd that managers’ transfer price expectations are lower (and deviate more from the prevailing market price) when
they are negotiating with a partner with high concern-for-others than with a partner with low concern-for-others. We
discuss the broader implications of these results for the design of management accounting systems.
Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
Introduction
Negotiation is a common method used by ?rms
to set transfer prices (Ghosh, 2000). Even where an
external market exists, transfer price negotiation is
a potentially useful control mechanism, allowing a
balance between economic considerations and
broader social concerns by interdependent divi-
sions (Kachelmeier & Towry, 2002). These transfer
price negotiations are important to managers as
they in?uence both their own and other divisional
pro?ts. Previous research has shown that these
transfer prices are a?ected by both economic
factors (market prices) and behavioural factors
including fairness (Luft & Libby, 1997).
0361-3682/$ - see front matter Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2008.01.002
*
Corresponding author. Tel.: +61 2 9385 5831; fax: +61 2
9662 4491.
E-mail address: [email protected] (K.T. Trotman).
Available online at www.sciencedirect.com
Accounting, Organizations and Society 33 (2008) 704–717
www.elsevier.com/locate/aos
In the current study, we examine whether the
impact of accounting information on managers’
transfer price expectations are moderated by the
way accounting information is framed (either as
potential gains or potential losses) and the man-
ager’s perception of the other negotiation party’s
objective (whether their partner’s objective
involves high or low concern-for-others). These
expectations are important as they directly a?ect
the costs and outcomes of negotiations (Ghosh,
2000; Luft & Libby, 1997; Trotman, Wright, &
Wright, 2005).
Previous negotiation literature has shown the
importance of ‘fairness’ during negotiation and
that participants’ estimates of a fair price display
a ‘self-serving bias’ (or egocentrism). The self-serv-
ing bias refers to the cognitive bias arising from an
individual’s tendency to view an outcome more
favourable to them as being fairer when resolving
con?icts
1
(Thompson & Loewenstein, 1992). Spe-
ci?cally, where an active external market exists,
and the market price is greater than a price that
would lead to both divisions receiving an equal
pro?t, a seller will generally consider the market
price to be a fairer transfer price as it results in a
higher pro?t for the selling division. The buyer,
however, would view the transfer price that allows
pro?t to be equally shared between the two divi-
sions as a fairer price (Luft & Libby, 1997).
Both Luft and Libby (1997) and Kachelmeier
and Towry (2002) found that where market price
di?ered fromthe equal-pro?t price, managers based
their transfer price judgments on both the market
price and the equal-pro?t price. Furthermore, both
studies found that sellers and buyers placed di?er-
ent weights on these two reference points when for-
mulating judgments. Speci?cally, due to self-
serving biases, sellers’ transfer price expectations
were closer to the market price than that of the buy-
ers, while buyers’ expectations were closer to the
equal-pro?t price. One likely e?ect of a ‘transfer
price expectation gap’ between buyers and sellers
is a prolonged and ine?cient negotiation process.
While this may be avoided by the intervention of
top management to mediate any inter-divisional
dispute, such an approach would undermine the
autonomy of decentralised divisional managers.
Instead, if we have a better understanding of those
factors that in?uence managers’ transfer price judg-
ments, we may be able to overcome managers’
biases by re-designing the negotiation process.
Prior research in psychology suggests that the
key to understanding how managers make negoti-
ation judgments is to examine the way in which
managers de?ne their negotiation context, and
their perception of variables that are critical and
endogenous to the negotiation process (Bazerman,
Curhan, Moore, & Valley, 2000; Ghosh & Boldt,
2004; Kristensen & Garling, 1997; Neale & Bazer-
man, 1992). Neale and Bazerman (1992) in partic-
ular have argued that:
‘‘Rather than focus only on external factors
[to the negotiation process], it may be most
useful to view situations from an interpretive
perspective. It may not be the objective,
external aspects of the situation that directly
a?ect negotiator judgment; instead, it may be
the way that the negotiator perceives these
features and uses those perceptions to inter-
pret and screen information.” (Neale & Bazer-
man, 1992, p. 161, emphasis added).
Two factors that are of particular interest in the
current study are the goal frame adopted by man-
agers, which a?ects the way managers perceive the
negotiation outcome, and the negotiation part-
ner’s objective (also called ‘social concern’) which
a?ects the way managers perceive the negotiation
partner. Both of these variables are found to be
important in the psychology and economics litera-
ture (e.g. Kahneman & Tversky, 1979; Lewicki,
Saunders, & Barry, 2005; Neale & Bazerman,
1992; Roth, 1995), but are generally controlled
for rather than manipulated in prior accounting
studies. For example, both Luft and Libby
(1997) and Kachelmeier and Towry (2002)
adopted a consistent positive goal frame in all their
treatments, and controlled for negotiation part-
ners’ objectives by telling their participants that a
positive relationship existed between negotiators.
1
This is in contrast with ‘self-interest’, which refers to a
negotiator’s motivation to advance their own outcomes. Indi-
viduals with a high level of self-interest do not necessarily have
a biased view of what constitutes a fair outcome; rather, they
are motivated to achieve a favourable outcome for themselves.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 705
We extend these earlier studies by examining
the impact of these variables on managers’ self-
serving biases in a transfer pricing setting. There
are bene?ts in studying these two variables simul-
taneously. We suggest that the reason the loss
frame a?ects negotiation judgments is because it
causes managers to become more concerned about
their own outcome (not to incur any further
losses), exacerbating their self-serving bias. Their
negotiation partner’s objective also is expected to
in?uence the level of concern managers have for
their own outcome. For example, a perception that
the negotiation partner has high concern-for-oth-
ers causes managers to be more willing to give
up some of their divisional pro?t and accept a less
favourable transfer price. By using both cognitive
and social lenses together, we attempt to obtain
a more uni?ed understanding of how the negotia-
tion process works and eventually, how to over-
come barriers to e?ective negotiation.
The study of these variables is important because
of their implications both for transfer pricing and
for their impact more generally on systems and pro-
cesses in decentralised organisations that use man-
agement accounting information. Transfer price
negotiations, in particular, allowbusiness unit man-
agers to utilise and share their local knowledge
(Dikoli & Vaysman, 2006) and maintain inter-divi-
sional coordination while preserving autonomy
(van Helden, van der Meer-Kooistra, & Scapens,
2001). The cost of negotiation, however, is not neg-
ligible, and the negotiation approach to transfer
price determination is only recommended when
the cost of bargaining is relatively low (Dikoli &
Vaysman, 2006).
2
A number of accounting studies
(e.g. Kachelmeier & Towry, 2002; Luft & Libby,
1997) have demonstrated that the self-serving bias
is one factor that can reduce the accuracy of manag-
ers’ transfer price judgments and thus potentially
increase the time and costs of negotiation. An
understanding of framing and negotiation partner’s
objective also has wider implications for the man-
agement accounting literature and these implica-
tions are included in our ‘Discussion’ section.
In summary, our study makes a number of sig-
ni?cant contributions to the accounting literature.
First, we extend the Luft and Libby’s (1997)
results by investigating the in?uence of managers’
perception of the negotiation context on transfer
price judgments. We speci?cally address the role
of framing and the negotiation partner’s objective.
The ?rst factor is directly controllable by manage-
ment accountants. For example, management
accountants can produce reports based on alterna-
tive negotiating reference points to support a
seller–manager involved in a transfer price negoti-
ation. When the market price is used as a reference
point, the management accounting reports are
likely to highlight the potential loss in pro?t as
the negotiated transfer price falls below the market
price (Perera, McKinnon, & Harrison, 2003). This
will lead to the negotiating manager adopting a
loss frame. Alternatively, the reports can use prod-
uct costs as a reference point, focusing on the gains
in pro?t as the negotiated transfer price moves
above the product costs (Colbert & Spicer, 1995).
This is likely to cause the negotiating manager to
adopt a gain frame.
Second, the importance of social considerations
was highlighted by both Luft and Libby (1997) and
Kachelmeier and Towry (2002) when they found
evidence of the e?ect of fairness concerns on trans-
fer price judgments. Building on this research, we
demonstrate, in a situation where market prices
are above equal-pro?t prices, that managers expect
the ?nal transfer price to be lower when they are
dealing with a partner with high concern-for-others
than when negotiating with a partner with low con-
cern-for-others.
3
This is because managers tend to
2
In addition, while the perception of a ‘fairer’ outcome can
result in a more positive feeling at the end of the negotiation
process (Lewicki et al., 2005), this perception may have negative
consequences for ?rms as divisions use ‘pro?t equality’ as an
argument for ‘fairer’ outcomes. As the gap between equal-pro?t
price and market price grows bigger, the pursuit of pro?t
equality may give rise to an ‘internal socialism’ problem where
?rms ine?ciently try to equalise divisional performance (Bolton
& Scharfsein, 1998). The resultant impact is the distortion of
pro?ts as a result of managers’ pursuit of pro?t equality.
3
In this study, we use an example where market price is
above the equal-pro?t price and therefore concern-for-others
results in transfer prices below market prices. We note that
direction of the di?erence between market price and transfer
price would reverse if the market price given in an experiment
was lower than the equal-pro?t price.
706 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
reciprocate the perceived social concerns of their
negotiation partner. The lower price, however, is
also further from the market price which has impli-
cations for production divisions that depend on the
transfer price and may have longer term implica-
tions if managers are put under increased pressure
to reach pro?t targets. Deviations from the market
price also have potential negative implications on
intra-organisational resource allocation (Bolton &
Scharfsein, 1998). As noted by Sprinkle (2003), it
is important to study the extent to what social
motives, and other aspects of a ?rm’s information
systems, interact with the more formal accounting
systems to a?ect managerial behaviour. Our results
imply that these two factors signi?cantly in?uence
the way managers make use of accounting informa-
tion when making transfer price judgments.
Third, our study extends the existing literature
by examining the impact of the above variables
on two dimensions of transfer pricing judgments:
a reservation price and a price premium (i.e. the
di?erence between the reservation price and the
estimated transfer price). Our results show that a
loss frame increases the sellers’ reservation price,
and thus eventually their ?nal transfer price judg-
ment. In contrast, negotiation partner’s objective
did not a?ect reservation price judgment, but
rather, we found that sellers who perceived their
partner to have a high level of concern-for-others
were more willing to accept a lower price premium.
Finally, inter-divisional negotiation (such as
transfer price negotiations) is an important control
mechanism that balances divisional autonomy
with inter-divisional coordination (van Helden
et al., 2001). Our study extends the growing litera-
ture on improving negotiation outcomes in
accounting/auditing situations (Bame-Aldred &
Kida, 2007; Gibbins, McCracken, & Salterio,
2005; Gibbins, Salterio, & Webb, 2001; Ng &
Tan, 2003; Trotman et al., 2005) to the manage-
ment accounting arena, and in doing so, contrib-
utes to our understanding of the challenges faced
by decentralised organisations.
Literature review and hypotheses development
Conventional economic arguments suggest that
transfer price judgments should be based on ‘eco-
nomically rational’ concerns such as the market
price, transaction costs and the division’s cost
structure (e.g. Colbert & Spicer, 1995). However,
prior literature in psychology has demonstrated
that negotiators do not always act ‘rationally’.
Rather, they su?er from a number of judgmental
biases, such as anchoring their decisions on irrele-
vant information, and the escalation of commit-
ment (e.g. Bazerman & Neale, 1992; Neale &
Bazerman, 1992; Northcraft & Neale, 1987).
In the accounting literature, Luft and Libby
(1997) have shown that, during transfer price nego-
tiation, sellers’ estimates of negotiated transfer
prices tend to be signi?cantly higher than those of
buyers, particularly when the market price is higher
than the equal-pro?t price. Luft and Libby (1997)
argue that their ?nding demonstrates the existence
of a ‘self-serving bias’, which causes managers to
overweigh the negotiation outcome that is most
bene?cial to them (Luft & Libby, 1997; Thompson
& Loewenstein, 1992). Thus, where more than one
de?nition of a ‘fair’ transfer price exists (e.g. in the
Luft & Libby, 1997 study, where market price was
higher than the equal-pro?t price),
4
negotiating
managers will interpret fairness in ways that favour
their position, such that the transfer price estimates
by sellers are signi?cantly higher than the transfer
price estimates by buyers. Before developing our
hypotheses we ?rst replicate the baseline condition
established in Luft and Libby (1997) on the di?er-
ence in transfer price judgments between sellers
and buyers resulting from the self-serving bias.
H1: Sellers’ estimated ?nal transfer prices are
higher than buyers’ estimated ?nal transfer
prices.
‘Frames’ are subjective cognitive systems
through which individuals evaluate and make sense
4
The prevailing market price is often perceived as a fair
transfer price because it is the result of ‘impartial’ market forces
of supply and demand. On the other hand, a fair price can also
be de?ned as one that provides equal pro?t to both negotiating
divisions (i.e. the equal-pro?t price). The concept of equal-
pro?t price is likely to be particularly salient in the internal
transfer price negotiation process, where the negotiating man-
agers belong to the same company, and inter-divisional equity
becomes an important concern.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 707
of situations they are in. Di?erent frames adopted
can lead individuals to pursue or avoid subsequent
actions (Lewicki et al., 2005). Traditionally, negoti-
ation literature has focused on the e?ect of ‘risk
preference framing’ (Bottom & Studt, 1993; Kahn-
eman & Tversky, 1979; Neale & Bazerman, 1985;
Thaler, 1992) – a framing e?ect characterised by a
choice between outcome certainty and a more risky
alternative. The underlying decision valence is then
manipulated, such that a loss frame is represented
by uncertainties surrounding a negative conse-
quence, and a gain frame is represented by uncer-
tainty surrounding a positive consequence.
5
For example, Neale and Bazerman (1985) inves-
tigated the risk frame in a management/union
negotiation by either telling the participants that
any concessions made by the company will result
in signi?cant ?nancial losses (loss frame), or any
concessions from the union will result in signi?cant
?nancial gains (gain frame). All impasses were to
be referred to an arbitrator, and as the arbitrator’s
?nal decision was unknown, this presented the
participants with a risk element. Inter alia, they
found that compared to negotiators with gain
frames, negotiators with loss frames were more
likely to have their agreements determined by the
arbitrator (i.e. they chose the riskier option). They
also found that compared to gain framed negotia-
tors, loss framed negotiators were less likely to
make concessions.
In this study, we examine the framing role of
accounting information, and how this a?ects man-
agers’ transfer price judgments. Our focus is on
using accounting information to frame the negoti-
ation goal. We propose that because managers are
more concerned with avoiding losses than increas-
ing gains both buyers and sellers are more likely to
focus on maximising their divisional pro?t when
given a loss frame compared to a gain frame. This
greater concern for achieving their own outcome is
likely to further increase the transfer price judge-
ment gaps between buyers and sellers.
Prior literature on motivated reasoning suggests
that a higher level of motivation to achieve an out-
come can lead people to overestimate the probabil-
ity that a favourable outcome will eventuate
(Brownstein, 2003). Further, motivated reasoning
also distorts people’s perception of others, such
that they tend to expect others to behave in a
way that results in favourable outcomes (Kunda,
1990). In the context of a negotiation, we predict
that the loss framed managers’ greater concern
for maximising their divisional pro?t will cause
them to overestimate the likelihood that their part-
ner will take their view of what constitutes a fair
price, and thus agree on a transfer price more
favourable to them. Speci?cally, sellers (buyers)
with a loss frame are more likely to believe that
their partner will agree on a higher (lower) price
being a fair transfer price, compared to sellers with
a gain frame. As such, we predict that a loss frame
will increase negotiators’ self-serving biases.
In addition, as loss framed managers become
more motivated to achieve a better outcome, they
may be more willing to incur greater bargaining
costs compared to gain framed managers. In the
absence of any information about their partner’s
negotiation frame (and thus the level of their part-
ner’s motivation), the loss framed managers are also
likely to expect their willingness to incur greater
bargaining costs will lead to a more favourable out-
come. As such, we predict that the transfer price
judgement gap between buyers and sellers is greater
under the loss frame than the gain frame condition.
6
5
While the major emphasis in the framing literature has
centred on the standard risky choice framing a?ect introduced
by Kahneman and Tversky (1979) and Tversky and Kahneman
(1981), Levin et al. (1998) developed a typology to distinguish
between risky choice framing, attribute framing and goal
framing. Levin, Schneider, and Gaeth (1998) suggest that the
di?erent operational de?nitions of framing have e?ects that rely
on di?erent psychological processes.
6
As we do not manipulate or provide information about the
negotiation partner’s frame, buyers and sellers may not be
making the same transfer price predictions. For example, the
seller in the loss frame is predicting the price that a loss framed
seller and a buyer with unknown or neutral frame will
negotiate, whereas the buyer in the loss frame is predicting
the price that a loss framed buyer and a seller with an unknown
or neutral frame will negotiate. Even if both buyers and sellers
assume the same (e.g. neutral) frame for their negotiation
partner, the di?erences in transfer price predictions by sellers
and buyers are still not necessarily all self-serving bias and may,
in fact, be partially due to the lack of information about their
negotiation partner’s frame.
708 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
H2: The di?erence in estimated ?nal transfer
price between buyers and sellers is smaller when
information provided to negotiating managers
is framed as gains rather than losses.
A number of prior studies have suggested that
social concerns in?uence transfer price negotiation
judgments (e.g. Kachelmeier & Towry, 2002;
Luft & Libby, 1997). Both Luft and Libby (1997)
and Kachelmeier and Towry (2002) found that
while economic rationality would dictate that
negotiators should expect market-based transfer
prices, negotiators have an aversion to unequal
pro?ts. While they attributed this aversion to
negotiators’ concerns about pro?t sharing and
ensuring both divisions receive satisfactory pro?ts,
the impact of social concern was not directly
tested. In this study, we seek to directly examine
the e?ects social concerns have on transfer price
judgements.
We propose that managers’ concerns about
unequal pro?ts and therefore their transfer price
judgments may be a?ected by the perception they
have of their negotiation partner (i.e. the other
party to the negotiation process). In particular,
during negotiation, managers would try to gauge
their partner’s objective, and then combine this
information with their own negotiation objective
when formulating their transfer price judgments
(e.g. Carroll, Bazerman, & Maury, 1988; Lewicki
et al., 2005).
An established framework used to explain a
negotiator’s objective is the ‘dual concern model’
(e.g. Lewicki et al., 2005; Pruitt, 1983; Sorenson,
Morse, & Savage, 1999). This framework postu-
lates that a negotiator’s objective is in?uenced by
two independent types of concerns: concern for
their own outcomes (‘concern-for-self’) and con-
cern for the other party’s outcomes (‘concern-
for-others’). Our focus in the current study is on
a manager’s perception of their partner’s degree
of concern-for-others. Our manipulation of this
variable is consistent with large variations of con-
cern-for-others in transfer pricing situations; for
example, the level of concern for the pro?ts of
other divisions is likely to vary in organisations
that are quasi-markets compared to quasi-families
(Eccles, 1985, pp. 273–278).
Following the suggestion in previous research
(Kachelmeier & Towry, 2002; Luft & Libby,
1997), in situations where the market price is
higher than the equal-pro?t prices, that this con-
cern-for-others results in transfer prices below
market prices, we predict that when the level of
concern-for-others is stronger, both buyer and
seller will expect the price to be lower. We note
that this prediction only holds for situations where
the market price is higher than the equal-pro?t
price, which is the case in our experiment.
The above prediction particularly applies when
the level of concern is similar for both negotiators
in the pair. While we only manipulate the level of
concern-for-others for the negotiation partner, we
suggest this is likely to result in a similar level of
concern for the negotiating manager for two rea-
sons. First, the psychology literature refers to the
‘reciprocity’ principle as a social norm by which
an individual who acts in a certain way will expect
a similar return action (e.g. Maxwell, Nye, & Max-
well, 2003).
7
The norm of reciprocity therefore
establishes expectations about how one is to
behave in social interactions (Maxwell et al.,
2003). Prior research has consistently found that
negotiators have a tendency to reciprocate negotia-
tion motives of their negotiation partners (Maxwell
et al., 2003). Therefore, negotiating managers who
perceive that their negotiating partner has high
concern-for-others will reciprocate with a similar
objective, showing high concern for pro?t sharing,
In contrast, negotiating managers who perceive
that their partner has low concern-for-others are
expected to reciprocate by showing low concern
for pro?t sharing.
Second, organisations di?er in the types of
employee behaviours that are considered accept-
able. For example, our low concern-for-others
manipulation would be acceptable in some organ-
isations but clearly unacceptable in other organisa-
tions. By informing a participant about the
negotiation partner’s concerns-for-others we also
tell participants something about the culture of
the organisation. Speci?cally, our manipulation
7
The reciprocity principle has also recently been introduced
into the audit literature in audit–client negotiations (Sanchez,
Agoglia, & Hat?eld, 2007; Tan & Trotman, 2007).
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 709
of concern-for-others involves providing partici-
pants with a memo from their negotiation partner.
In the low concern-for-others treatment, the memo
includes references to ‘maximising pro?t of my
division’ which also communicates to the other
party that this is accepted practice in this organisa-
tion. In contrast, in the high concern-for-others
treatment, the culture encourages both divisions
to receive satisfactory pro?ts and therefore both
parties would expect a lower price.
H3: Managers’ estimated transfer prices are
lower when they are negotiating with a partner
with high concern-for-others than when they
are negotiating with a partner with low con-
cern-for-others.
Research methods
Research design
A controlled laboratory experiment was con-
ducted to test the proposed hypotheses, using a
2 Â 2 Â 2 between-subjects design. The three inde-
pendent variables were the negotiating manager’s
role (participants acting as either a buyer or a
seller), goal frame (gain frame or loss frame) and
the negotiation partner’s objective (high or low
concern-for-others).
Experimental task
The experimental task was modi?ed
8
from Luft
and Libby’s (1997) instrument, where participants
assumed the role of a manager who is responsible
for negotiating a transfer price of component
‘Parts’. ‘Parts’ are components sold by the Parts
Division to the Assembly Division, which can then
be processed further by the Assembly Division and
then sold to an external customer. As the two divi-
sions are autonomous, both divisional managers
are free to negotiate a mutually acceptable transfer
price or to trade externally at the prevailing mar-
ket price (which was set at $70 per unit).
9
The cost
structures of the two divisions were designed such
that the equal-pro?t price was $50.
10
Included in
the task was a pro?t schedule illustrating the pro?t
implications of a range of transfer prices for both
parties (between $20 where the pro?t for sellers
was zero, and $80 per unit, where the pro?t for
buyers was zero). Both buyers and sellers were
then asked to predict the ?nal negotiated transfer
price and the sellers’ reservation price.
Independent variables
The negotiation role was manipulated by ran-
domly assigning participants either to the role of
‘Parts Manager’ (i.e. seller) or ‘Assembly Man-
ager’ (i.e. buyer). The goal frame was operationa-
lised by ‘framing’ the instructions provided in the
instrument either as a gain frame or a loss frame.
Speci?cally, instructions provided to Assembly
managers (the buyers) assigned a gain frame were
as follows:
‘‘As you can see from the table, for every $5
decrease in transfer price you stand to gain
$5000 pro?t. For example, by negotiating a
transfer price of $55, your pro?t is $25,000.
But if you negotiate a lower transfer price,
say, $50, your pro?t is $30,000, which means
that you have gained $5000 pro?t. In other
words, as you settle for a lower transfer
price, you stand to gain pro?t for your divi-
sion in $5000 increments.”
8
Two main modi?cations were made to the Luft and Libby
(1997) instrument to accommodate two of our variables of
interest (discussed in more detail later). A pilot test was then
conducted (with 21 undergraduate accounting students) to
ensure that our modi?cations were understood by our partic-
ipants and that the independent manipulated variables had the
intended e?ects. As a result of the pilot test, and discussions
with pilot test participants post-experiment, further minor
modi?cations were made.
9
The current task focuses on a ‘distributive’ (‘?xed pie’)
negotiation task, that is, the managers are negotiating in a win–
lose situation where their goals are in direct con?ict.
10
Speci?cally, based on Luft and Libby’s (1997) scenario, the
value of the shipment of ‘Parts’ to the Assembly Division was
$80 per unit; while the value (or cost) to the Parts Division was
$20 per unit. This means that the pro?t for the Assembly
department was ($80 less negotiated transfer price); while the
pro?t for the Parts Division was (transfer price less $20). Thus,
at a transfer price of $50 per unit, both divisions would obtain a
pro?t of $30 per unit.
710 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
For participants assigned a loss frame, the
description explained how every $5 change in the
transfer price would result in the division losing
$5000. Further, participants were also provided
with a schedule of pro?t framed either as pro?t
increases or pro?t decreases as the transfer price
changed (refer to Fig. 1).
To manipulate the negotiation partner’s objec-
tive, participants were provided with a ?ctitious
memo indicating their negotiation partner’s level
of concern-for-others. Participants assigned to
the ‘high concern-for-others’ conditions were given
a memo stressing their partner’s desire for mutual
concession and maximising pro?t for both divi-
sions. In contrast, participants in the ‘low con-
cern-for-others’ conditions were given a memo
emphasising their partner’s desire to maximise
pro?t for their own division only and their unwill-
ingness to make concessions. For example, the
memo from a partner with low concern-for-others
highlighted their intention to ‘‘. . . achieve the best
pro?t for my division. . . if you are unwilling to
make concessions, I am prepared to trade
externally.”
Dependent variables
We measured the dependent variable, manag-
ers’ estimated negotiation prices, by asking partic-
ipants to predict the ?nal transfer price of the
negotiation process. In addition, participants were
also asked to indicate the expected lowest price
that sellers would likely to be willing to accept
(i.e. seller’s reservation price).
11
Consistent with
Luft and Libby (1997), to minimise the time
required for data collection we did not ask partic-
ipants to estimate buyers’ reservation price.
Participants
One hundred and twenty-eight participants vol-
unteered to participate in this experiment. All par-
ticipants were enrolled in a Master of Commerce
degree or Master of Business Technology at one
Australian university, and each had at least two
years of full time work experience. However, 32
participants failed one or more post-experiment
manipulation tests and were later excluded from
the analysis, resulting in 96 usable responses. The
cell sizes for each of the eight treatment group var-
ied between 11 and 15 (see Table 1).
Manipulation check and post-test measures
After participants completed the negotiation
task, they were given three manipulation checks.
The ?rst asked participants to indicate what role
they played in the negotiation (i.e. whether they
A: Profit schedule for Assembly managers (i.e. buyers)/Gain frame information
Transfer price for Parts 80 75 70 65 60 55 50 45 40 35 30 25 20
PARTS profit ($000) 60 55 50 45 40 35 30 25 20 15 10 5 0
ASSEMBLY profit ($000) 0 5 10 15 20 25 30 35 40 45 50 55 60
B: Profit schedule for Assembly managers (i.e. buyers)/Loss frame information
Transfer price for Parts 20 25 30 35 40 45 50 55 60 65 70 75 80
PARTS profit ($000) 0 5 10 15 20 25 30 35 40 45 50 55 60
ASSEMBLY profit ($000) 60 55 50 45 40 35 30 25 20 15 10 5 0
Fig. 1. Sample pro?t schedules provided to experimental participants.
11
Luft and Libby (1997) pointed out that negotiated transfer
price is a more sensitive measure of potential con?ict, while
reservation price is more sensitive to managers’ mistaken
judgments about their negotiation partner. As we are primarily
interested in the former (i.e. the e?ect of managers’ perception
on their expectations of how the ‘negotiation con?ict’ would be
resolved), our primary analysis will focus on estimated ?nal
transfer prices.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 711
were acting as a Parts manager or an Assembly
manager). The second asked participants to indi-
cate whether their negotiation partners were inter-
ested in maximising both divisions’ pro?ts, or only
their own division’s pro?t. The third asked partic-
ipants to indicate whether the case material stated
that ‘‘for every $5 increase in transfer price you
stand to lose $5000 pro?t”, or ‘‘for every $5
decrease in transfer price you stand to gain
$5000 pro?t”.
12
Results
Hypothesis testing
The descriptive statistics for estimated transfer
price are summarised in Table 1, and a 2 Â 2 Â 2
ANOVA model, with estimated transfer price as
the dependent variable, is presented in Table 2.
As can be seen from Table 1, and consistent with
H1, the average estimated transfer price was
higher for sellers (62.40) than for buyers (56.87).
This di?erence (the main e?ect of role) is statisti-
cally signi?cant (F = 8.71, p = 0.00), thus H1 is
supported.
H2 predicted that the di?erence in estimated
transfer prices between buyers and sellers would
be smaller when potential negotiation outcomes
are framed as gains rather than losses. The descrip-
tive statistics in Table 1 further indicate that the
di?erence in estimated transfer prices between sell-
ers and buyers under the gain frame condition
(60.48 À 58.83 = 1.65) was lower than that under
the loss frame condition (63.84 À 55.00 = 8.84).
This di?erence is shown in Table 2 as a signi?cant
interaction e?ect between role and goal frame
(F = 4.26, p = 0.02), thus H2 is supported.
H3 examined the e?ect of the negotiation part-
ner’s objective on managers’ transfer price judg-
ments. In H3, we expected the negotiation
partner’s objective to have a main e?ect, where
both buyers’ and sellers’ transfer price expecta-
tions would be lower if they were negotiating with
Table 1
Means (standard deviation) of estimated ?nal transfer price ($)
Partner’s objective Frame total Total
High concern-for-others Low concern-for-others
Gain frame Loss frame Total Gain frame Loss frame Total Gain frame Loss frame
Sellers 58.64 61.92 60.42 62.50 65.50 64.30 60.48 63.84 62.40
(7.10) (8.04) (7.65) (9.50) (6.96) (7.89) (8.35) (7.44) (7.93)
n = 11 n = 13 n = 24 n = 10 n = 15 n = 25 n = 21 n = 28 n = 49
Buyers 57.69 52.73 55.42 60.30 56.92 58.39 58.83 55.00 56.87
(8.32) (10.57) (9.55) (7.51) (10.52) (9.29) (7.91) (10.53) (9.44)
n = 13 n = 11 n = 24 n = 10 n = 13 n = 23 n = 23 n = 24 n = 47
Column total 58.13 57.71 57.92 61.40 61.52 61.47 59.61 59.76 59.69
(7.63) (10.21) (8.92) (8.41) (9.56) (9.01) (8.01) (9.95) (9.09)
n = 24 n = 24 n = 48 n = 20 n = 28 n = 48 n = 44 n = 52 n = 96
Table 2
ANOVA model for estimated transfer price – H1 and H2
DF MS F p
Negotiator’s role 1 644.11 8.71 0.00
a
Partner’s objective 1 298.70 4.04 0.02
Frame 1 6.22 0.08 0.39
Role
*
objective 1 0.60 0.01 0.46
Role
*
Frame 1 315.05 4.26 0.02
b
Objective
*
frame 1 2.49 0.03 0.43
Role
*
objective
*
frame 1 5.16 0.07 0.40
Error 88 73.93
Negotiator role – participants acting as either a buyer or a
seller.
Partner’s objective – either high or low concern-for-others.
Frame – accounting information presented in either a gain or a
loss goal frame.
a
The signi?cant main e?ect of negotiator’s role provides
support for H1.
b
The signi?cant interaction e?ect between frame and role
provides support for H2.
12
56% of the manipulation test errors related to the framing
e?ect. All statistical tests were re-run after including partici-
pants who failed the manipulation tests, and all results
remained statistically the same.
712 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
a partner with high concern-for-others than if with
a partner who had low concern-for-others. The
overall ANOVA model as shown in Table 2
con?rmed our expectation (signi?cant main e?ect
for negotiation partner’s objective, F = 4.04, p =
0.02) thus, H3 is supported.
Additional analysis
We also conducted additional analyses to delin-
eate the e?ect of goal framing and the negotiation
partner’s objective on two elements of managers’
transfer price judgments: the reservation price
and the di?erence between the reservation price
and the estimated transfer price (which we refer
to as ‘price premium’). The reservation price (also
known as the resistance point) represents the min-
imal price sellers are willing to accept from the
transaction (e.g. for sellers, this means the minimal
acceptable price – Lewicki et al., 2005). In con-
trast, the price premium re?ects the extent to which
negotiators expect to achieve their desired out-
come. That is, the price premium incorporates
negotiators’ anticipation of the concession they
will make during the o?er–countero?er process in
negotiation. The descriptive statistics of these
two variables are shown in Tables 3 and 4.
Table 3 Panel A indicates that on average sell-
ers’ reservation price was $54.04, which was higher
than the equal-pro?t price of $50.00 (signi?cant
with one-sample t-test, t = 2.464, p = 0.01). This
is consistent with our expectation that sellers in
general would not consider the equal-pro?t price
as a fair outcome of negotiation. Instead, their
minimum acceptable price was signi?cantly higher.
Table 3 (Panel A) also shows that compared to
their gain frame counterparts, sellers in the loss
frame condition reported a higher reservation
price ($56.54 vs. $50.71). The main e?ect for goal
frame in the ANOVA reported in Table 3 Panel
B shows that this di?erence is statistically signi?-
cant (F = 6.33, p = 0.02). Consistent with our ear-
lier argument, this ?nding suggests that a loss
frame focuses individuals on a goal of avoiding
negative consequences, thus increasing their ‘resis-
tance point’ to an unfavourable transfer price,
which results in a higher expected reservation
price. Neither negotiation partner’s objective
(F = 0.01, p = 0.94) or the interaction with fram-
ing (F = 1.06, p = 0.31) are signi?cant for reserva-
tion price.
Table 4 reports the descriptive statistics and
ANOVA results for the sellers’ price premium. A
Table 3
Additional analysis – sellers’ reservation prices
Partner exhibits high concern-for-others Partner exhibits low concern-for-others Total
Panel A: Means (standard deviation) of sellers’ reservation prices
Gain frame 52.27 49.00 50.71
(7.20) (16.13) (12.07)
n = 11 n = 10 n = 21
Loss frame 56.15 56.87 56.54
(8.20) (12.51) (10.55)
n = 13 n = 15 n = 28
Total 54.38 53.72 54.04
(7.85) (14.29) (11.48)
n = 24 n = 25 n = 49
DF MS F p
Panel B: ANOVA model
a
(dependent variable = sellers’ reservation price)
Partner’s objective 1 0.56 0.01 0.94
Frame 1 641.74 6.33 0.02
Partner’s objective
*
frame 1 107.86 1.06 0.31
Error 43 101.40
a
Two outliers were excluded from this analysis. One outlier was excluded because the subject reported a reservation price that was
higher than the expected transfer price. A second outlier was excluded because the reported reservation price was greater than 3
standard deviations away from the mean.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 713
marginally signi?cant main e?ect of the negotia-
tion partner’s objective (F = 3.24, p = 0.08) sug-
gests that sellers who were negotiating with a
high concern-for-others partner expected to give
up a greater share of divisional pro?t during the
negotiation process and thus predicted a lower
price premium compared to those who were nego-
tiating with a partner with low concern-for-others.
Neither framing (F = 2.10, p = 0.16) or the inter-
action (F = 2.53, p = 0.12) are signi?cant.
13
Together, these results show that the goal frame
and the negotiation partner’s objective have di?er-
ent e?ects on di?erent aspects of managers’ trans-
fer price judgments. Speci?cally, as individuals are
more resistant to avoiding losses than increasing
gains, a loss frame increases the sellers’ reservation
price and eventually, their ?nal estimated transfer
price. On the other hand, the negotiation partner’s
concern-for-others provides the sellers with an
indication of the potential o?ers/countero?ers dur-
ing transfer price negotiation, thus in?uencing the
price premium the sellers expect on top of their
reservation price.
Summary and discussion
In this study, we examined whether managers’
perceptions of potential negotiation outcomes
(framed either as potential gains or potential
losses) and of their negotiation partner (exhibiting
high or low concern-for-others) a?ected self-serv-
ing biases and consequently their transfer price
judgments. We found that compared to a gain
frame, a loss frame exacerbates managers’ self-
serving biases and increases the transfer price
expectation gap between buyers and sellers.
Further, we found that the negotiation part-
ner’s objective had a signi?cant impact on sellers’
transfer price judgments. Consistent with the
‘norm of reciprocity’, our results show that, in sit-
uations where market prices are higher than equal-
pro?t prices, managers reciprocated their partner’s
concerns and expected lower transfer prices when
their negotiation partner exhibited high concern-
for-others, and expected higher transfer prices
when their negotiation partner exhibited low con-
cern-for-others. This ?nding is particularly inter-
esting as sellers in our experiment had relatively
strong bargaining power but these sellers did not
exploit their bargaining power by demanding high
transfer prices regardless of their partner’s level of
concern-for-others. Instead, we found that sellers
Table 4
Additional analysis – sellers’ price premium
Partner exhibits high concern-for-others Partner exhibits low concern-for-others Total
Panel A: Descriptive statistics – sellers’ transfer price premium
Gain frame 6.36 13.50 9.76
(6.36) (10.55) (9.15)
n = 11 n = 10 n = 21
Loss frame 5.77 8.63 7.30
(6.41) (8.09) (7.37)
n = 13 n = 15 n = 28
Total 6.04 10.58 8.36
(6.25) (9.27) (8.18)
n = 24 n = 25 n = 49
DF MS F p
Panel B: ANOVA model
a
(dependent variable = price premium)
Partner’s objective 1 166.08 3.24 0.08
Frame 1 107.29 2.10 0.16
Partner’s objective
*
frame 1 129.70 2.53 0.12
Error 43 51.21
a
Price premium = (sellers’ reservation price – transfer price estimate).
13
Due to the relatively small sample size, however, the
statistical inferences of our additional analysis should be
interpreted with care.
714 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
were sensitive to their partner’s objective, and were
more willing to accept a less advantageous out-
come if their partner showed high concern-for-
others.
The additional analysis suggests that goal fram-
ing and the negotiation partner’s objective have
di?erent impacts on managers’ negotiation judg-
ments. When we decompose the transfer price
judgments into two subcomponents: the reserva-
tion price and a price premium, we found that
the loss frame resulted in managers reporting a
higher reservation price. On the other hand, man-
agers’ perception of their partner’s objective had
signi?cant impact on their price premiums.
Our study has important implications for both
researchers and practitioners. Prior research has
shown that negotiating managers su?er from self-
serving biases, which result in a signi?cant di?er-
ence in expected transfer prices between buyers
and sellers. We extended this line of research by
examining how these di?erences in transfer price
expectations are a?ected by managers’ perceptions
of the negotiation context. Understanding manag-
ers’ transfer price expectations is important, as dif-
ferences in expectations between buyers and sellers
can lead to prolonged disputes and thus a costly
negotiation process (Luft & Libby, 1997).
Our ?ndings that the provision of loss framed
information increases the buyer–seller expectation
gap can also have a signi?cant impact on organisa-
tions. We note that systems and processes using
management accounting can either inadvertently
or by design cause managers to adopt di?erent
frames. For example, practitioner literature often
advocates the use of customer pro?tability infor-
mation to support customer negotiation (Kaplan
& Cooper, 1998) and negotiators may be given a
‘‘price menu” listing a range of service levels and
their associated costs (Kaplan & Anderson,
2007). In such circumstances, management
accounting information can be presented in a
way that induces either a gain frame or a loss
frame. Speci?cally, management accounting
reports can either describe the incremental cost
increases with each service level (e.g. incremental
cost of $500 every time a customer requests an
additional sales visit), or the incremental cost sav-
ings (e.g. incremental costs savings of $500 per
sales visit reduced). The former is likely to induce
a loss frame and the latter a gain frame.
14
The practitioner’s ‘self-help’ literature on nego-
tiation often discusses the importance of building
rapport and a?liation at the negotiation table
(Fisher & Shapiro, 2005). Our study provides
empirical support for the importance of communi-
cating a positive objective. Our results show that
managers expect a lower transfer price (closer to
the equal-pro?t price) when they perceive that
their negotiation partners have high concern-for-
others. Our results imply that showing high
concern-for-others (as opposed to showing low
concern-for-others) can be e?ective in persuading
negotiation opponents (especially sellers) to con-
sider their perceptions.
Our additional analysis suggests that managers’
perceptions of the negotiation outcomes and of
their negotiation partners a?ect di?erent aspects
of the negotiation process. This ?nding enhances
our understanding of how to ‘de-bias’ managers’
self-biased transfer price judgments. By framing
the pro?t information di?erently we can encourage
sellers to set a lower reservation price, and at the
same time, organisations can also attempt to pro-
mote greater ‘concern-for-others’ among sellers so
that they are more likely to accept a lower ‘pre-
mium’ on top of their reservation price. For exam-
ple, incentive schemes that focus too much on the
‘stick’ rather than the ‘carrot’ may increase dis-
trust (Fehr & Ga¨chter, 2000), potentially heighten
managers’ concern-for-self relative to their con-
cern-for-others, and thus reduce managers’ will-
ingness to reciprocate positively during
negotiation.
Similar to earlier studies on transfer price nego-
tiation (Luft & Libby, 1997; Kachelmeier &
Towry, 2002), our results demonstrate a strong
desire by participants to take fairness into account
when making transfer price judgments, such that
regardless of their role or the treatment, the resul-
tant transfer price judgment is di?erent from the
external market price. On the other hand, Bolton
and Scharfsein (1998) argue that the pursuit of
14
Framing is also important in capital investment analysis as
?nancial outcomes can be presented in terms of pro?ts or losses
(e.g. Moreno, Kida, & Smith, 2002).
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 715
‘internal socialism’ can be costly, as decentralised
?rms sometimes try to equalise divisional pro?t
at the expense of resource allocation e?ciencies.
Our results show that internal socialism is also
likely to arise from inter-divisional negotiations,
potentially adding to the costs of internal
transactions.
An understanding of framing and negotiation
partner’s objective has wider implications than just
transfer pricing given that other inter-divisional
negotiation is common in decentralised organisa-
tions. For example, a production manager may
need to negotiate inventory management and
delivery policies with the marketing division; and
a research and development (R&D) manager in
one division may need to negotiate with the R&
D manager in another division over resource allo-
cation issues in collaborative projects (Coletti,
Sedatole, & Towry, 2005).
The impact of these variables on negotiations
also has implications for organisation design.
For example, larger self-serving biases result in
bigger errors in judging the outcome a bargaining
partner will accept in the end. These higher self-
serving biases have been shown to have a nega-
tive impact on reaching an agreement and create
more impasses (Babcock & Loewenstein, 1997;
Gelfand et al., 2002). Consequently, as decentra-
lised organisations rely more on negotiation
between peers, if the setting is one where large
self-serving biases are likely to be present, these
decentralised organisations will not work as e?ec-
tively. They will require greater interventions
from headquarters and more hierarchical decision
making, thus making decentralisation more costly
and less e?ective.
Acknowledgment
We gratefully acknowledge a research grant
from the Australian Research Council and the
helpful comments from Joan Luft, Sue Haka,
Kim Lang?eld-Smith, Anne Lillis, Steve Salte-
rio, Jane Baxter, Brian Bur?tt and Habib
Mahama, as well as seminar participants at Uni-
versity of Cincinnati, University of Melbourne,
2006 AFAANZ Conference and 2005 EAA
Conference.
References
Babcock, L., & Loewenstein, G. (1997). Explaining bargaining
impasse: The role of self-serving biases. Journal of Economic
Perspectives, 11(1), 109–126.
Bame-Aldred, C. W., & Kida, T. (2007). A comparison of
auditor and client negotiation decisions. Accounting, Orga-
nizations and Society, 32(6), 497–511.
Bazerman, M., Curhan, J., Moore, D., & Valley, K. (2000).
Negotiation. Annual Review of Psychology, 51, 279–312.
Bazerman, M. H., & Neale, M. A. (1992). Negotiating
rationality. New York: The Free Press.
Bolton, P., & Scharfsein, D. (1998). Corporate ?nance, the
theory of the ?rm, and organizations. Journal of Economic
Perspectives, 12(4), 95–114.
Bottom, W. P., & Studt, A. (1993). Framing e?ects and the
distributive aspect of integrative bargaining. Organizational
Behavior and Human Decision Processes, 56(3), 459–
474.
Brownstein, A. (2003). Biased precision processing. Psycholog-
ical Bulletin, 129(4), 545–568.
Carroll, J. S., Bazerman, M. H., & Maury, R. (1988).
Negotiator cognitions: A descriptive approach to negotia-
tors’ understanding of their opponents. Organizational
Behavior and Human Decision Processes, 41(3), 352–370.
Colbert, G., & Spicer, B. (1995). A multi-case investigation of a
theory of the transfer pricing process. Accounting, Organi-
zations and Society, 20(6), 423–456.
Coletti, A., Sedatole, K., & Towry, K. (2005). The e?ect of
control systems on trust and cooperation in collaborative
environment. The Accounting Review, 80(2), 477–500.
Dikoli, S., & Vaysman, I. (2006). Information technology,
organizational design, and transfer pricing. Journal of
Accounting and Economics, 41, 203–236.
Eccles, R. (1985). The transfer pricing problem. Lexington, MA:
Lexington Books.
Fehr, E., & Ga¨chter, S. (2000). Fairness and retaliation: The
economics of reciprocity. Journal of Economic Perspectives,
14(3), 159–181.
Fisher, R., & Shapiro, D. (2005). Beyond reason: Using
emotions as you negotiate. Random House Business Books.
Gelfand, M., Higgins, M., Nishii, L., Raver, L., Dominguez,
A., Murakami, F., et al. (2002). Culture and egocentric
perceptions of fairness in con?ict and negotiation. Journal
of Applied Psychology, 87(5), 833–845.
Ghosh, D. (2000). Complementary arrangements of organiza-
tional factors and outcomes of negotiated transfer price.
Accounting, Organizations and Society, 25, 661–682.
Ghosh, D., & Boldt, M. (2004). The outcome saliency e?ect on
negotiated transfer prices. Journal of Managerial Issues,
16(3), 305–321.
Gibbins, M., McCracken, S. A., & Salterio, S. E. (2005).
Negotiations over accounting issues: The congruency of
audit partner and chief ?nancial o?cer recalls. Auditing, 24,
171–193.
Gibbins, M., Salterio, S., & Webb, A. (2001). Evidence about
auditor–client management negotiation concerning client’s
716 L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717
?nancial reporting. Journal of Accounting Research, 39(3),
535–564.
Kachelmeier, S. J., & Towry, K. L. (2002). Negotiated transfer
pricing: Is fairness easier said than done?. Accounting
Review 77, 571–593.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An
analysis of decisions under risk. Econometrica, 47, 263–291.
Kaplan, R., & Anderson, S. (2007). Time-driven activity-based
costing. Harvard Business School Publishing.
Kaplan, R., & Cooper, R. (1998). Cost and e?ect. Harvard
Business School Publishing.
Kristensen, H., & Garling, T. (1997). The e?ects of anchor
points and reference points on negotiation process and
outcome. Organizational Behavior and Human Decision
Processes, 71(1), 85–94.
Kunda, Z. (1990). The case for motivated reasoning. Psycho-
logical Bulletin, 108(3), 480–498.
Levin, I., Schneider, S., & Gaeth, G. (1998). All frames are not
created equal: A typology and critical analysis of framing
e?ects. Organizational Behaviour and Human Decision Pro-
cesses, 76(2), 149–188.
Lewicki, R., Saunders, D., & Barry, B. (2005). Negotiation (5th
ed.). New York: McGraw Hill.
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.
Maxwell, S., Nye, P., & Maxwell, N. (2003). The wrath of the
fairness-primed negotiator when the reciprocity norm is
violated. Journal of Business Research, 56(5), 399–409.
Moreno, K., Kida, T., & Smith, J. (2002). The impact of
a?ective reactions on risky decision making in accounting
contexts. Journal of Accounting Research, 40(5), 1331–1349.
Neale, M. A., & Bazerman, M. H. (1985). The e?ects of framing
and negotiator overcon?dence on bargaining behaviors and
outcomes. Academy of Management Journal, 28(1), 34–49.
Neale, M. A., & Bazerman, M. H. (1992). Negotiator cognition
and rationality: A behavioural decision theory perspective.
Organizational Behavior and Human Decision Processes, 51,
157–175.
Ng, T. B.-P., & Tan, H.-T. (2003). E?ects of authoritative
guidance availability and audit committee e?ectiveness on
auditors’ judgments in an auditor–client negotiation con-
text. The Accounting Review, 78(3), 801–819.
Northcraft, G., & Neale, M. (1987). Experts, amateurs, and real
estate: An anchoring-and-adjustment perspective on prop-
erty pricing decisions. Organizational Behaviour and Human
Decision Processes, 39, 84–97.
Perera, S., McKinnon, J. L., & Harrison, G. L. (2003).
Di?usion of transfer pricing innovation in the context of
commercialization – A longitudinal case study of a govern-
ment trading enterprise. Management Accounting Research,
14(2), 140–164.
Pruitt, D. G. (1983). Strategic choice in negotiation. The
American, 27(2), 167–194.
Roth, A. E. (1995). Bargaining experiments. The handbook of
experimental economics. Princeton: University Press.
Sanchez, M. H., Agoglia, C. P., & Hat?eld, R. C. (2007). The
e?ect of auditors’ use of a reciprocity-based strategy on
auditor–client negotiations. Accounting Review, 82(1),
241–263.
Sorenson, R. L., Morse, E. A., & Savage, G. T. (1999). A test of
the motivations underlying choice of con?ict strategies in
the dual-concern model. International Journal of Con?ict
Management, 10(1), 25–44.
Sprinkle, G. B. (2003). Perspectives on experimental research in
managerial accounting. Accounting, Organizations and Soci-
ety, 28(2-3), 287–318.
Tan, H.-T, & Trotman, K. T. (2007). E?ects of auditors’
concession timing on ?nancial o?cers’ negotiation judg-
ments. Working paper, The University of New South Wales.
Thaler, R. H. (1992). The winner’s curse: Paradoxes and
anomalies of economic life. Princeton, NJ: Princeton Uni-
versity Press.
Thompson, L., & Loewenstein, G. (1992). Egocentric interpre-
tations of fairness and interpersonal con?ict. Organizational
Behavior and Human Decision Processes, 51(2), 176–197.
Trotman, K. T., Wright, A. M., & Wright, S. (2005). Auditor
negotiations: An examination of the e?cacy of intervention
methods. The Accounting Review, 80(1), 349–368.
Tversky, A., & Kahneman, D. (1981). The framing of decisions
and the psychology of choice. Science, 211, 453–463.
van Helden, G., van der Meer-Kooistra, J., & Scapens, R.
(2001). Co-ordination of internal transactions at Hoogovens
Steel: Struggling with the tension between performance-
oriented business units and the concept of an integrated
company. Management Accounting Research, 12, 357–386.
L. Chang et al. / Accounting, Organizations and Society 33 (2008) 704–717 717
doc_725418399.pdf