Good Things Come to Those Who Wait: Late First Offers Facilitate Creative Agreements in Negotiation

Abstract

Although previous research has shown that making the first offer leads to a distributive advantage in negotiations, the current research explored how the timing of first offers affects the creativity of negotiation agreements. We hypothesized that making the first offer later rather than earlier in the negotiation would facilitate the discovery of creative agreements that better meet the parties’ underlying interests. Experiment 1 demonstrated that compared to early first offers, late first offers facilitated creative agreements that better met the parties’ underlying interests. Experiments 2a-2b controlled for the duration of the negotiation and conceptually replicated this effect. The last two studies also demonstrated that the beneficial effect of late first offers was mediated by greater information exchange. Thus, negotiators need to consider the timing of first offers to fully capitalize on the first offer advantage. Implications for our understanding of creativity, motivated information exchange, and timing in negotiations are discussed.

Full text

http://psp.sagepub.com/
Bulletin
Personality and Social Psychology
http://psp.sagepub.com/content/39/6/814
The online version of this article can be found at:
DOI: 10.1177/0146167213483319
2013 39: 814Pers Soc Psychol Bull
Marwan Sinaceur, William W. Maddux, Dimitri Vasiljevic, Ricardo Perez Nückel and Adam D. Galinsky
Good Things Come to Those Who Wait : Late First Offers Facilitate Creative Agreements in Negotiation
Published by:
http://www.sagepublications.com
On behalf of:
Society for Personality and Social Psychology
can be found at:Personality and Social Psychology BulletinAdditional services and information for
http://psp.sagepub.com/cgi/alertsEmail Alerts:
http://psp.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- May 21, 2013Version of Record >>
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Personality and Social
Psychology Bulletin
39(6) 814 –825
© 2013 by the Society for Personality
and Social Psychology, Inc
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0146167213483319
pspb.sagepub.com
Article
The 2000 Camp David meeting between the Israelis and the
Palestinians was a critical negotiation aimed at putting an end
to the conflict between the two parties. However, experts
agree that the talks ended in failure, yielding disastrous con-
sequences. According to diplomats, this failed negotiation
meeting eventually led to the collapse of the peace process
(e.g., Miller, 2008). Although there are many factors in this
complex political situation, one critical reason for the failure
of the 2000 Camp David meeting might have had to do with
the timing of the first offers in this negotiation. As a close
observer recalled, the parties made their initial offers early in
the negotiation process, well before “neither the Israelis nor
the Palestinians had been prepared to fully own up to the fears
and needs of the other” (Malley, 2010, p. 3). Specifically, the
initial offers came before the parties fully understood each
other’s underlying interests, leaving some fundamental inter-
ests unattended (especially critical issues such as the status of
Jerusalem, security, and the status of the refugees; Miller,
2008; New York Times, 7/26/01). Putting early offers on the
table appears to have backfired and prevented creative options
from being explored during the Camp David meeting (Miller,
2008). Tellingly, however, in another negotiation in January
2001, Israelis and Palestinians were more successful at the
Taba Summit meeting, potentially because the parties better
understood each other’s underlying interests and included
additional issues that were not fully discussed before (Israeli–
Palestinian Joint Statement, 2001; Miller, 2008).
Although prior research has shown that making the first
offer provides a critical advantage in “distributive” or zero-
sum negotiations (Galinsky & Mussweiler, 2001; Magee,
Galinsky, & Gruenfeld, 2007), the 2000 Camp David nego-
tiation example seems to suggest that beyond the simple
question of who makes the first offer, the actual timing of
first offers may critically affect negotiations, especially ones
that require exploring the parties’ underlying interests in cre-
ative ways. By making a first offer later rather than earlier in
the negotiation process, negotiators may increase the infor-
mation exchanged, hence facilitate the uncovering of cre-
ative solutions that better address the underlying interests of
the parties. Thus, late first offers may lead to more efficient
conflict resolution.
483319PSP
XXX10.1177/0146167213483319Personality and Social Psychology BulletinSinaceur et al.
research-article2013
1
INSEAD, Fontainebleau, France
2
Université de Paris Ouest, Nanterre La Défense, France
3
Université de Paris 1, Panthéon Sorbonne, France
4
Columbia University, NY, USA
Corresponding Author:
Marwan Sinaceur, Organizational Behavior Department, INSEAD,
Boulevard de Constance, 77300 Fontainebleau, France.
Email: marwan.sinaceur@insead.edu
Good Things Come to Those Who
Wait: Late First Offers Facilitate
Creative Agreements in Negotiation
Marwan Sinaceur
1
, William W. Maddux
1
, Dimitri Vasiljevic
2
,
Ricardo Perez Nückel
3
, and Adam D. Galinsky
4
Abstract
Although previous research has shown that making the first offer leads to a distributive advantage in negotiations, the current
research explored how the timing of first offers affects the creativity of negotiation agreements. We hypothesized that making
the first offer later rather than earlier in the negotiation would facilitate the discovery of creative agreements that better
meet the parties’ underlying interests. Experiment 1 demonstrated that compared with early first offers, late first offers
facilitated creative agreements that better met the parties’ underlying interests. Experiments 2a and 2b controlled for the
duration of the negotiation and conceptually replicated this effect. The last two studies also demonstrated that the beneficial
effect of late first offers was mediated by greater information exchange. Thus, negotiators need to consider the timing of first
offers to fully capitalize on the first offer advantage. Implications for our understanding of creativity, motivated information
exchange, and timing in negotiations are discussed.
Keywords
creativity, negotiation, underlying interests, anchoring, motivated information exchange
Received August 4, 2012; revision accepted January 26, 2013
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 815
First Offers in Different Negotiation
Contexts
Negotiations are crucial to resolving conflicts between
nations, organizations, and individuals (Bazerman & Neale,
1992; Pruitt & Rubin, 1986). In solving conflicts or making
deals, negotiators often take a “distributive” (i.e., zero-sum)
perspective and start the negotiation by asserting their posi-
tions (Pruitt, 1981; Pruitt & Rubin, 1986). Prior research has
shown that making the first offer provides a potent bargain-
ing advantage in distributive or zero-sum negotiations
(Galinsky, Leonardelli, Okhuysen, & Mussweiler, 2005;
Galinsky & Mussweiler, 2001; Galinsky, Mussweiler, &
Medvec, 2002; Kray, Thompson, & Galinsky, 2001; Magee
et al., 2007; Moran & Ritov, 2002; Ritov, 1996; see Maaravi,
Ganzach, & Pazy, 2011). In particular, when people negoti-
ate over a fixed set of issues, particularly those that are quan-
tifiable such as price, more aggressive first offers lead to
better individual outcomes, with the extremity of the first
offer explaining up to 50% of the variance in individual out-
comes (Galinsky et al., 2002; Galinsky & Mussweiler, 2001;
Magee et al., 2007). First offers influence final individual
outcomes through anchoring, thus setting standards that
affect the entire postoffer interaction (Galinsky & Mussweiler,
2001; Magee et al., 2007). The impact of first offers on final
outcomes in negotiation is so important that first offers have
been recently deemed “key concepts in the study of negotia-
tion” (Maaravi et al., 2011, p. 245).
However, it is important to note that most of the research
that has documented the advantage of first offers has been in
the context of negotiations in which the issues were fixed—
typically, distributive, single-issue negotiations where it was
only possible to negotiate over price. Yet, it is unclear
whether such benefits would hold in different types of nego-
tiations, in particular those where agreements can be achieved
by uncovering the parties’ underlying interests and creatively
adding issues to the table. Although anchoring early in a
negotiation may enable negotiators to claim a greater share
of existing sets of issues (Galinsky & Mussweiler, 2001),
anchoring early might also prove detrimental if the first offer
anchors thinking on a fixed set of issues and prevents nego-
tiators from considering additional issues or frameworks that
would better meet the parties’ underlying interests. That is,
anchoring may limit negotiators’ focus on the issues explic-
itly contained in the offer, making negotiators less likely to
consider creative options outside the issues explicitly con-
tained in the initial offer. In contexts where discovering cre-
ative yet hidden solutions is necessary to achieve a good deal
(e.g., Kray, Galinsky, & Markman, 2009; Loewenstein &
Howell, 2010; Maddux & Galinsky, 2009), negotiators may
miss critical opportunities to meet their fundamental, under-
lying interests.
Taking the time to uncover the parties’ underlying inter-
ests and explore creative options may allow negotiators to
add novel perspectives and issues to the table that can
increase the value of the deal for both sides (Bazerman &
Neale, 1992; Fisher & Ury, 1981). Indeed, negotiation
scholars have suggested that thinking creatively beyond
existing issues to uncover underlying interests is a key stan-
dard of negotiators’ efficiency (Bazerman & Neale, 1992;
Fisher & Ury, 1981), but one that has been relatively under-
studied (Giacomantonio, De Dreu, & Mannetti, 2010;
Loewenstein & Howell, 2010). As Giacomantonio and col-
leagues (2010) recently proposed, researchers know little
about the psychological mechanisms that help negotiators
explore underlying interests. It is important to note that, to
date, no prior research examined how making offers might
affect negotiators’ ability to explore underlying interests and
craft creative solutions.
The Importance of Timing: Early
Versus Late
In the current article, we propose that the timing of first offers
is a key factor that determines the ability of negotiators to
discover creative solutions that better meet the parties’ under-
lying interests. Thus, we move beyond the simple question of
who makes the first offer and the resulting impact on distribu-
tive outcomes (the focus of most previous research). We sug-
gest that making a first offer relatively late in the negotiation
(rather than relatively early) provides an additional benefit
over simply making the first offer by leading to more creative
agreements. Our prediction draws from research on the differ-
ent stages of negotiation process (for reviews, see Morris &
Keltner, 2000; Pruitt, 1981; Pruitt & Rubin, 1986). In particu-
lar, problem solving, which is characterized by free discus-
sion and generation of new options through information
exchange about priorities, is a stage that comes later rather
than earlier in the negotiation process. Early in the negotia-
tion, negotiators typically engage in positional, aggressive
argumentation; they are not open to exchange information
about their interests, and do not engage in creative problem
solving. In contrast, later in the negotiation, negotiators are
more likely to move away from existing positions, and
become more open to exchanging information about their
interests, and to engaging in collaborative thinking.
Given these stages in the negotiation, making a late first
offer may be more efficient because late offers will tend to
be made at a time when the parties are more likely to focus
on exchanging information about their interests and priori-
ties. In turn, greater information exchange is likely to affect
the creativity of final agreements as negotiators will be able
to make a more informed judgment about how to satisfy
both parties’ underlying interests. When first offers come
later, they will not have anchored negotiators on extant
issues, allowing negotiators to consider novel issues and
think more freely beyond extant issues. Thus, we propose
that late first offers versus early first offers will be more
likely to lead to agreements that better meet the parties’
underlying interests.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

816 Personality and Social Psychology Bulletin 39(6)
Overview
Three experiments examined whether the timing of first
offers affects the discovery of creative solutions in conflict
and negotiation. Specifically, we investigated whether
negotiators would be more likely to meet the parties’ under-
lying interests in final agreements when making first offers
later rather than earlier in the negotiation. We experimen-
tally manipulated the timing of first offers in actual, face-
to-face interactions. In all studies, the set of issues to be
negotiated was not fixed, but instead was left open to allow
negotiators the leeway to craft creative agreements. Thus,
participants could include any additional issues they wanted
in their first offers and any final deals. Experiment 1 pro-
vided initial evidence for the idea that late first offers facili-
tate the discovery of creative agreements that satisfied the
parties’ underlying interests compared with early first
offers. Experiments 2a and 2b controlled for the duration of
the negotiation and conceptually replicated this effect using
a different task. Experiments 2a and 2b further investigated
whether the beneficial effect of late first offers on uncover-
ing underlying interests would be mediated by greater
information exchange.
Experiment 1
Method
Participants and Design. Participants were 108 undergraduate
business students (62 women, 44 men, 2 unreported) enrolled
in a negotiations course and the experiment took place at the
beginning of the course. Dyads were randomly assigned to
one of three conditions: a late-first-offer condition in which
one participant was instructed to make a late first offer and
the other was a control (first offers were not mentioned to the
control participant); an early-first-offer condition in which
one participant was instructed to make an early first offer and
the other was a control; or a control condition in which both
participants were controls (first offers were not mentioned to
either participant).
Negotiation Task. Participants were randomly assigned to
either one of two roles in a negotiation between two organi-
zations that were considering buying and selling red bananas.
One party was the seller who produced the bananas (Red
Bananas Producer) and the other was a potential buyer who
needed the bananas to produce drugs in response to the out-
break of a worldwide epidemic (Universal Lab, a pharma-
ceutical company). In this exercise, both parties had
underlying interests that were comparatively more important
than the issue of sale price. The seller/bananas producer
wanted to foster a long-term business relationship beyond
the current sale transaction while also needing to avoid the
epidemic spreading to their country (i.e., they wanted the
buyer to commit to something more than merely buying
bananas). The buyer/pharmaceutical company had also
fundamental interests that overshadowed the issue of sale
price: They needed to produce the drug out of the bananas as
efficiently as possible to maximize the number of people
cured. Thus, both parties’ underlying interests were actually
compatible.
There was a large bargaining zone for the sale price of the
bananas, consistent with previous research on first offers
(e.g., Galinsky & Mussweiler, 2001). The bargaining zone
was US$19.90: For the seller/bananas producer, any sale sig-
nificantly above the local consumption price of US$0.10
would represent a profit, and for the pharmaceutical com-
pany, any price up to US$20 would represent a profit. Thus,
adding issues was not necessary for the participants to reach
an agreement. Yet, because both parties had underlying inter-
ests that were comparatively more important than the issue
of sale price, agreements based solely on sale price would
satisfy their interests less than agreements that included
additional issues.
Given the underlying interests of the buyer/pharmaceuti-
cal company, additional issues could include quality checks
of the bananas or guarantees of careful conditioning and
transportation to avoid spoiling of the bananas. Given the
underlying interests of the seller/bananas producer, addi-
tional issues could include assurances for access to the drug
if the epidemic spread to their own country, or ways to entice
the pharmaceutical company to invest in the producer’s
country. Additional issues logical for both parties might
involve an agreement that the pharmaceutical company
would sell back the drug to the bananas producer if the epi-
demic spread. Although there were no specific numbers
attached to such issues, and thus their exact value was not
possible to quantify, the parties gained explicit benefits from
their inclusion, and thus were important to the final value of
the deal.
Thus, the exercise made it possible to observe variance
along (a) the number of added creative issues and (b) the
extent to which these would meet the parties’ underlying
interests. Importantly, across experimental conditions, both
parties were instructed to get the best deal and meet their
underlying interests. Irrespective of experimental conditions,
both parties’ instructions emphasized that the agreement
could include issues other than the agreed-upon price (and
quantity) of bananas. It is important to note that, although the
instructions clearly allowed for such possibilities, no explicit
hints were given as to what such additional issues might be.
Thus, participants had to discover creative solutions sponta-
neously during the course of the negotiation for them to meet
the parties’ underlying interests.
Experimental Manipulation. Dyads were randomly assigned
to one of three conditions: early-first-offer, late-first-offer, or
control. Along with their task instructions, participants were
given a recommendation about what strategy to adopt in the
forthcoming negotiation. Because of concerns with adequate
power, and because prior research suggests that making a
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 817
first offer (e.g., Galinsky & Mussweiler, 2001; Moran &
Ritov, 2002) and anchoring (e.g., Galinsky et al., 2002;
Huber & Neale, 1986) is equally effective for buyers and
sellers, we only instructed the buyer to make a first offer.
Thus, buyers were told to make either (a) a first offer late in
the negotiation, (b) a first offer early in the negotiation, or (c)
were given no specific recommendation (control).
In the late-first-offer condition, instructions indicated that
experts recommended participants to make a first offer late
in the negotiation, after the first 15 min. Instructions indi-
cated that participants could, of course, raise and discuss any
issue that seemed relevant, but it was important to make a
first offer only after the first 15 min in the negotiation. In the
early-first-offer condition, instructions were similar except
that experts recommended participants to make a first offer
early in the negotiation, within the first minute. Instructions
again indicated that participants could, of course, raise and
discuss any issue that seemed relevant, but it was important
to make a first offer within the first minute in the negotiation.
In the Control condition, participants were given standard
negotiations advice (e.g., Maddux, Mullen, & Galinsky,
2008; Swaab, Maddux, & Sinaceur, 2011) about preparing in
depth for the negotiation, quantitatively and qualitatively, in
addition to allowing them to raise and discuss any issue that
seemed relevant. Note that all counterparts (sellers) were
given these same control instructions. Thus, across experi-
mental conditions, participants were explicitly told that
although they had to include price in their first offer, they
could also include any additional issues they wanted in the
negotiations and any final deals.
Procedure. Participants had 20 min to read their instructions
and prepare for the negotiation. Then, they were given 20
min to negotiate face to face in dyads. After negotiating, they
indicated the terms of the agreement they reached, if any.
Measures
Sale price. The sale price for the bananas was recorded
from dyads’ agreements.
Number of added creative issues. Dyads’ agreements were
first coded by counting the number of added creative issues
beyond the extant issue of price, such as reusing the parts of
bananas not used for the drug, putting the bananas producer
logo on the drug, but also quality checks of the bananas
and health-related plans for the bananas producer’s popu-
lation. A measure for the number of added creative issues
was obtained from averaging results by two coders who
were blind to the conditions. To obtain a reliable count of
the number of added creative issues, the two coders first
coded about 25% of the agreements. Then, all the differ-
ences between coders were resolved by discussion. Finally,
the two coders coded independently the remaining 75% of
the agreements. The reliability of the resulting measure was
high (Cronbach’s α = .99).
Meeting underlying interests. Dyads’ agreements were also
coded for the extent to which they met parties’ underlying
interests based on the nature of the issues added beyond the
extant issue of price (using a 0-7 scale; 0 = not at all, 7 = a
lot). Based on both parties’ instructions, the following added
issues were deemed to have met the parties’ underlying inter-
ests to a great extent: (a) quality checks of the bananas, (b)
health-related plans for the bananas producer’s population
(including the pharmaceutical company selling back the drug
to the bananas producer if the epidemic spread and/or health
campaign for the drug in the bananas producer’s country),
(c) the pharmaceutical company’s making additional invest-
ments in the bananas producer’s country, and/or (d) care-
ful conditioning and transportation to avoid spoiling of the
bananas. Specifically, these added issues helped the seller
and the buyer meet their underlying interests by allowing the
seller to extend the transaction to a longer term business rela-
tionship or protect themselves against the epidemic, and by
allowing the buyer to produce the drug more efficiently out
of the bananas or potentially cure more people. Based on the
parties’ instructions, added issues not deemed to have met
the parties’ underlying interests included reusing the parts of
bananas not used for the drug or putting the bananas producer
logo on the drug. A measure for meeting underlying inter-
ests was obtained from averaging results by two indepen-
dent coders who were blind to conditions. The same coding
procedure as above was used. The reliability of the resulting
measure was high (Cronbach’s α = .99). As expected, added
creative issues and meeting underlying interests correlated
highly (r = .69, p < .001).
Manipulation check. After the negotiation, participants
reported when the first offer was made in the negotiation
(“When was the first offer made, that is, after how many
minutes into the negotiation?”). The two parties’ responses
were averaged into a single index about when the first offer
was made (Cronbach’s α = .97).
Results
Manipulation Check. There was a significant effect of condi-
tion on when the first offer was made, F(2, 52) = 27.11, p <
.0001, η
P
2
= .52. Mean comparisons indicated that in the late-
first-offer condition, the first offer was made later in the
negotiation (M = 10.95 min, SD = 4.86 min) than in the
early-first-offer condition (M = 2.06 min, SD = 1.10 min),
t(50) = 7.35, p < .0001, d = 2.52, and later than in the control
condition (M = 6.34 min, SD = 3.70 min), t(50) = 3.74, p <
.0001, d = 1.07. Furthermore, in the early-first-offer condi-
tion, the first offer was made earlier in the negotiation than in
the control condition, t(50) = 3.35, p < .005, d = 1.57.
1
Thus,
the timing of first offer manipulation was successful.
Sale Price. The timing of the first offer had a significant
effect on sale price, the distributive issue in the negotiation,
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

818 Personality and Social Psychology Bulletin 39(6)
F(2, 53) = 4.44, p < .02, η
P
2
= .15. There was no difference in
sale price between the early- and late-first-offer conditions,
t(51) < .75, p > .45. However, in the early-first-offer condi-
tion (M = 11.88, SD = 4.54) and the late-first-offer condition
(M = 10.76, SD = 5.19), buyers received a better price (paid
less) than in the control condition (M = 15.15, SD = 3.76),
both ts(51) > 2.05, ps < .05, ds > .78. Thus, we replicated the
first offer advantage on the distributive issue, although nota-
bly the timing of first offers did not affect the first offer
advantage on the distributive issue. That is, making the first
offer but not the timing of the first offer affected sale price.
The one who made the first offer got a distributive advantage
regardless of timing.
Number of Added Creative Issues. The timing of the first offer
affected the number of added creative issues in the final
agreements, F(2, 53) = 3.50, p < .04, η
P
2
= .12. In the late-
first-offer condition, the final agreements contained a greater
number of added creative issues beyond the extant issue of
price (M = 1.59, SD = 1.81) than in the early-first-offer con-
dition (M = .71, SD = .99), t(51) = 2.03, p < .05, d = .60, and
more than in the control condition (M = .53, SD = .80), t(51) =
2.44, p < .02, d = .76. There was no difference between the
early-first-offer condition and the control condition, t(51) <
.40, p > .65. Thus, supporting our argument, a late first offer
facilitated the discovery of creative agreements.
Underlying Interests. The timing of the first offer affected the
extent to which the final agreements actually met the parties’
underlying interests, F(2, 53) = 8.47, p < .001, η
P
2
= .25. In
the late-first-offer condition, the final agreements met under-
lying interests to a greater extent (M = 2.84, SD = 2.17) than
in the early-first-offer condition (M = 1.21, SD = 1.24), t(51)
= 3.07, p < .005, d = .92, and more than in the control condi-
tion (M = .79, SD = 1.08), t(51) = 3.85, p < .001, d = 1.20.
There was no difference between the early-first-offer condi-
tion and the control condition, t(51) < .75, p > .45. Thus,
supporting our argument, a late first offer facilitated the dis-
covery of creative agreements that better satisfied the parties’
underlying, fundamental interests.
Experiment 2a
We conducted a second experiment to replicate and extend
the beneficial effect of late first offers on underlying inter-
ests. First, Experiment 2a used a different negotiation task
with different underlying interests. Specifically, we used a
task in which adding creative issues is critical to even reach-
ing an agreement in the first place, rather than simply increas-
ing the quality of an already existing agreement (e.g.,
Galinsky, Maddux, Gilin, & White, 2008; Kray et al., 2009;
Kray & Haselhuhn, 2007; Maddux & Galinsky, 2009).
We also wanted to rule out a potential alternative explana-
tion for our results, namely, that we did not control for the
amount of time participants took to complete the negotiation.
This alternative explanation is important because the dura-
tion of negotiations has been shown to generally affect the
integrativeness of agreements (De Dreu, 2003); hence, nego-
tiations in which the first offer was made late might have
simply lasted longer than negotiations in which the first offer
was made early, potentially giving negotiators more time to
generate creative solutions (see Bechtoldt, De Dreu, Nijstad,
& Choi, 2010). Therefore, in Experiment 2a, we measured
the actual duration of the negotiation and controlled for it in
our analyses.
Finally, although the results from the first study support
the idea that a late (vs. early) first offer has a beneficial effect
on constructing creative agreements, they did not test the
proposed causal mechanism behind this effect. Therefore, in
Experiment 2a, we tested whether greater information
exchange between the parties mediated the timing effect of
first offers on the extent to which creative agreements were
constructed.
Method
Participants and Design. Participants were 178 undergraduate
students (72 women, 106 men). Dyads were randomly
assigned to one of two conditions: a late-first-offer condition
in which one participant was instructed to make a late first
offer and the other was a control (the control participant was
not told about the timing of the first offer); or an early-first-
offer condition in which one participant was instructed to
make an early first offer and the other was a control.
Negotiation Task. Participants were randomly assigned to
either one of two roles in negotiating over the sale of a res-
taurant (see Galinsky et al., 2008; Kray et al., 2009; Kray &
Haselhuhn, 2007; Maddux & Galinsky, 2009). In this exer-
cise, a deal based solely on sale price was impossible. Spe-
cifically, the buyer’s reservation price (the maximum he or
she was authorized to pay) was lower than the seller’s reser-
vation price (the minimum he or she was authorized to
accept), resulting in a negative bargaining zone for sale
price. However, the parties’ underlying interests were com-
patible: The buyer wanted to hire managers to run the res-
taurant, and the seller needed sufficient funds to finance a
2-year sailboat trip while also needing employment for after
the trip. Thus, the parties could agree to a sale price below
the seller’s reservation price, but with a stipulation of future
employment, with the value of a future job allowing the
seller to satisfy his or her interests despite going below their
stated reservation price.
In this exercise, dyads often reach impasses because they
tend to focus only on the sale price of the restaurant.
Specifically, and contrasting to the task in the prior experi-
ment, participants could not achieve a deal via sale price
alone. Thus, participants had to discover creative solutions
spontaneously during the course of the negotiation by going
beyond the sale price and adding novel issues such as future
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 819
employment. Importantly, such creative solutions were not
explicitly suggested in the materials: The sale price of the
restaurant was presented as the only issue up for negotiation.
As in the prior experiment, no hint was given to participants
as to how to meet underlying interests, although these inter-
ests were laid out in the role instructions.
Experimental Manipulation. Participants were randomly assigned
to roles and to experimental conditions. As in Experiment 1,
we instructed buyers to make the first offer. The experimental
manipulations of timing were the same as in Experiment 1.
Unlike Experiment 1, the control instructions given to sellers
told them to not make the first offer.
The timing for making first offers was the same as in
Experiment 1. In the late-first-offer condition, buyers were
instructed to make the first offer after the first 15 min in the
negotiation. In the early-first-offer condition, buyers were
instructed to make the first offer within the first minute in the
negotiation.
Procedure. The procedure was the same as in Experiment 1.
Participants had 20 min to read their instructions and prepare
for the negotiation. Then, they were given 20 min to negotiate
face to face in dyads. After negotiating, they indicated the
terms of the agreement they reached, if any. Finally, they
completed a brief postnegotiation questionnaire containing
checks and a measurement of dyadic information exchange.
Measures
Coding of agreements. On the basis of a previously estab-
lished coding scheme (e.g., Galinsky et al., 2008; Kray et al.,
2009; Kray & Haselhuhn, 2007; Maddux & Galinsky, 2009),
creative performance of the dyad was measured as a dichoto-
mous variable. Specifically, an outcome was considered an
appropriate and creative agreement if the terms involved (a)
a sale price not greater than the buyer’s reservation price or
less than the seller’s reservation price and (b) the addition of
extra issue(s), such as future employment, where the value
of the issue(s) helped the seller and the buyer meet their res-
ervation prices and also their underlying interests. Outcomes
were considered unsuccessful if they (a) involved only the
sale price of the restaurant (which violated the parties’ res-
ervation prices) or (b) if parties reached an impasse. Note
that, contrasting to the task in the prior experiment, adding
creative issues was necessary to even get a deal.
Information exchange. Information exchange was the
hypothesized mediator for the timing effect of first offers on
creative agreements. It was the average of five items (“Was
much information exchanged between the parties during the
negotiation?” “Did you gather important new information
during the negotiation?” “Did the counterpart want to learn
more about the issues you were facing?” “Did the counter-
part want to understand your reasons for acting while in the
negotiation?” “Did the counterpart share information that
provided you with insights about the negotiation?”). Impor-
tantly, these items were conceptually consistent with prior
empirical research (e.g., De Dreu, Koole, & Steinel, 2000;
Sinaceur, 2010; Ten Velden, Beersma, & De Dreu, 2010) and
theorizing (e.g., De Dreu, Nijstad, & Van Knippenberg, 2008;
Pruitt, 1981; Thompson, 1990) on information exchange in
negotiation. We first averaged each party’s ratings to obtain
parties’ ratings on information exchange (Cronbach’s αs
> .70). We then averaged the two parties’ ratings to obtain
an overall measure of information exchange at the dyadic
level (Cronbach’s α = .71). As can be seen, parties’ ratings
on information exchange included not only participants’ rat-
ing their own perceptions but also participants’ rating their
counterpart’s behavior to ensure greater objectivity for the
resulting measure at the dyadic level. A confirmatory factor
analysis verified that, for each party’s rating, all items loaded
on one factor (for each party’s rating, it yielded only one fac-
tor with an eigenvalue greater than 1, that is, 2.28 or more,
which explained 45.69% or more of the total variance). Note
that the reliability of the resulting measure at the dyadic level
suggests that, indeed, participants’ ratings about each other
were aligned.
Manipulation checks. After the negotiation, participants
reported when the first offer was made in the negotiation
(i.e., after how many minutes; same item as in Experiment 1).
The two parties’ responses were averaged into a single index
(Cronbach’s α = .99). In addition, participants indicated who
made the first offer (“Did you make the first offer in the
negotiation?” either “Yes, I made it” or “No, my counterpart
made it”).
Duration of the negotiation. Participants were asked how
long their negotiation lasted. The two parties’ responses were
averaged into a single index (Cronbach’s α = .96).
Results
Manipulation Checks. In the late-first-offer condition, the first
offer was made later in the negotiation (M = 13.28 min, SD =
3.37) than in the early-first-offer condition (M = 1.94 min,
SD = 2.22); F(1, 88) = 363.42, p < .0001; η
P
2
= .81. Thus, the
timing of the first-offer manipulation was successful.
We also examined whether the buyer made the first offer.
This was the case for 83 out of the 89 dyads (93.3%). We
report below results with all dyads because a closer examina-
tion showed that the 6 dyads (6.7%) in which the counterpart
(i.e., the seller) preempted the buyer were cases in which the
buyer had acted consistent with his or her condition.
Specifically, in these 6 dyads, buyers were in the late condi-
tion and first offers were always made after 15 min (signifi-
cantly after the time observed in the early condition as
indicated by t tests, ps < .0001). Importantly, excluding these
6 dyads did not change any of the results below (neither in
pattern nor in significance).
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

820 Personality and Social Psychology Bulletin 39(6)
Creative Agreements. We examined whether a late versus
early first offer predicted whether a creative agreement was
constructed. We conducted a binary logistic regression on the
dyads’ creative agreements with timing of first offers as a
predictor. There was a significant effect of timing, such that
dyads in the late-first-offer condition were more likely to
construct a creative agreement than dyads in the early-first-
offer condition (B = 1.26, SE = .59; Wald(1) = 4.57, Exp(B)
= 3.54, p < .04). Specifically, 28.2% of dyads constructed a
creative agreement when first offers were late, whereas
10.0% did so when first offers were early.
The timing effect of first offers on creative agreements (B
= 1.73, SE = .69; Wald(1) = 6.28, Exp(B) = 5.62, p < .02)
held even when adding the duration of the negotiation as a
covariate (ns) in a hierarchical binary logistic regression.
Thus, a late first offer increased the likelihood of reaching a
creative agreement and this effect was independent of how
long the negotiation lasted. Note that, although timing of first
offers also affected duration such that a late first offer did
increase the duration of the negotiation (M = 19.62, SD =
2.58) compared with an early first offer (M = 17.47, SD =
4.65); F(1, 88) = 6.67, p = .01, η
P
2
= .07, duration did not
affect creative agreements (p > .27) and therefore could not
be a mediator. In addition, duration was not related to infor-
mation exchange (p > .66). We return to the issue of duration
in the “General Discussion” section.
Information Exchange. Dyads were more likely to exchange
information when first offers were late (M = 6.34, SD = .92)
rather than early (M = 5.67, SD = 1.11); F(1, 88) = 8.55, p <
.005, η
P
2
= .09, controlling for duration. This effect was simi-
lar whether controlling for duration or not.
Mediation by Information Exchange. We examined whether
information exchange mediated the effect of a late versus
early first offer on constructing creative agreements. Figure 1
presents the results of the regression analyses. Importantly,
we controlled for the duration of the negotiation in all
regression analyses (controlling for it or not did not affect
any of the results). As can be seen, a regression on creative
agreements was conducted with timing of first offers and
information exchange as simultaneous predictors. In this
regression, the timing effect of first offers was reduced
(B = 1.43, SE = .76; Wald(1) = 3.60, Exp(B) = 4.20, p =
.06), whereas the effect of information exchange was sig-
nificant (B = 1.33, SE = .46; Wald(1) = 8.50, Exp(B) = 3.79,
p < .005). To test the significance of the indirect effect (i.e.,
the path through the mediator), we used a bootstrapping
procedure, as recommended by Shrout and Bolger (2002).
The result of 1,000 resamples demonstrated that zero fell
outside the 95% confidence interval (CI) for the indirect
effect (95% CI low = .24; 95% CI high = 2.17), controlling
for duration of the negotiation. Thus, information exchange
mediated the effect of late versus early first offers on cre-
ative agreements, controlling for duration.
Experiment 2b
We conducted a third experiment to further document the
robustness of our effects. First, we introduced one variation
in our experimental manipulation compared with Experiment
2a. Because we had instructed the buyer to make a first offer
in the first two experiments, in Experiment 2b, we wanted to
instruct the seller to make a first offer.
Second, although the results from the first two studies
support the idea that a late (vs. early) first offer has a benefi-
cial effect on constructing creative agreements, we wanted to
provide further evidence for the proposed mechanism behind
this effect. Thus, in Experiment 2b, we tested whether greater
information exchange between the parties mediated the tim-
ing effect of first offers by using a different measure of infor-
mation exchange at the dyadic level. Specifically, we
measured how much the parties focused on exchanging
information as a proportion of the overall time spent negoti-
ating. Again, we controlled for the actual duration of the
negotiation in all analyses.
Method
Participants and Design. Participants were 202 undergraduate
students (121 women, 81 men). Dyads were randomly
assigned to one of two conditions: a late-first-offer condition
or an early-first-offer condition.
Negotiation Task. Participants performed the same negotia-
tion task as in Experiment 2a.
Experimental Manipulation. The experimental manipulations
were the same as in Experiment 2a, except for the fact that
sellers were instructed to make the first offer (instead of buy-
ers as before). Buyers were given the same control instruc-
tions as in Experiment 2a. The timing for making first offers
was the same as in Experiments 1 and 2a.
Procedure. The procedure was the same as in Experiments
1 and 2a. Participants had 20 min to read their instructions
and prepare for the negotiation. Then, they were given 20
min to negotiate face to face in dyads. After negotiating,
β = .30
**
B = 1.41
***
/ B = 1.33
**
B = 1.73
*
/ B = 1.43
^
Timing of first
offers
(late vs. early)
Information exchange
between the parties
Creative agreements
which met underlying
interests
Figure 1. Experiment 2a: Mediation by information exchange.
Note: We controlled for duration of the negotiation in all regression
analyses.
∧
p < .10, * p < .05, ** p < .01, *** p < .001.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 821
they indicated the terms of the agreement they reached, if
any. Finally, they completed a brief postnegotiation
questionnaire.
Measures
Coding of agreements. Creative performance of the dyad
was measured as a dichotomous variable on the basis of
the same coding scheme described in Experiment 2a (e.g.,
Galinsky et al., 2008; Kray et al., 2009; Kray & Haselhuhn,
2007; Maddux & Galinsky, 2009).
Duration of the negotiation. Participants were asked how
long their negotiation lasted. The two parties’ responses were
averaged into a single index (Cronbach’s α = .86).
Information exchange. Information exchange was the
hypothesized mediator for the timing effect of first offers on
creative agreements. It was measured by asking participants
“How much did you focus on exchanging information about
the diverse motivations of the two parties?” as a percent-
age of the overall time spent negotiating. The two parties’
responses were averaged into a single index (Cronbach’s
α = .74).
Supplementary measure. In this exercise, a critical piece
of information needs to be shared by the seller for the dyad
to construct a creative agreement, namely, that the seller
needs employment after selling the restaurant and complet-
ing a 2-year sailboat trip around the world (Kray & Hasel-
huhn, 2007; Maddux et al., 2008; Maddux & Galinsky,
2009). Therefore, the seller’s sharing what he or she would
do after the sailboat trip is the most critical piece of infor-
mation that can be exchanged in this negotiation (Kray &
Haselhuhn, 2007; Maddux et al., 2008; Maddux & Galinsky,
2009). Thus, we also asked sellers to what extent they shared
information about what they would do after their sailboat trip
(“As the restaurant seller, how much did you share what you
would do after your sailboat trip?”). We expected that the
seller’s sharing this critical piece of information would also
mediate the timing effect of first offers.
Manipulation checks. After the negotiation, participants
reported when the first offer was made in the negotiation
(i.e., after how many minutes; same item as in Experiments
1 and 2a). The two parties’ responses were averaged into a
single index (Cronbach’s α = .96). In addition, participants
indicated who made the first offer (same item as in Experi-
ment 2a).
Results
Manipulation Checks. In the late-first-offer condition, the first
offer was made later in the negotiation (M = 12.54 min, SD =
3.57) than in the early-first-offer condition (M = 1.65 min,
SD = .81), F(1, 96) = 416.97, p < .0001; η
P
2
= .81. Thus, the
timing of the first offer manipulation was successful. We
also checked that the seller made the first offer. This was true
for all dyads.
Creative Agreements. We examined whether a late versus
early first offer predicted whether a creative agreement was
constructed. We conducted a binary logistic regression on the
dyads’ creative agreements with timing of first offers as a
predictor. There was a significant effect of timing, such that
dyads in the late-first-offer condition were more likely to
construct a creative agreement than dyads in the early-first-
offer condition (B = 1.02, SE = .49; Wald(1) = 4.37, Exp(B)
= 2.77, p < .04). Specifically, 34.0% of dyads constructed a
creative agreement when first offers were late, whereas
15.7% did so when first offers were early.
The timing effect of first offers on creative agreements (B =
1.05, SE = .49; Wald(1) = 4.53, Exp(B) = 2.85, p < .04) held
even when adding the duration of the negotiation as a covari-
ate (ns) in a hierarchical binary logistic regression. Thus, a late
first offer increased the likelihood of reaching a creative agree-
ment, and this effect was independent of how long the negotia-
tion lasted. Note that, in this study, timing of first offers did not
significantly predict duration (p > .24). In addition, duration
did not affect creative agreements (p > .82) and was not related
to information exchange (p > .57). We return to the issue of
duration in the “General Discussion” section.
Information Exchange. Dyads focused on exchanging infor-
mation to a greater extent when first offers were late (M =
35.72%, SD = 15.89%) rather than when first offers were
early (M = 25.68%, SD = 13.30%); F(1, 100) = 11.60, p <
.001, η
P
2
= .11, controlling for duration. This effect was simi-
lar whether controlling for duration or not.
Mediation by Information Exchange. We examined whether
information exchange mediated the effect of a late versus
early first offer on constructing creative agreements. Figure 2
presents the results of the regression analyses. Importantly,
we controlled for the duration of the negotiation in all regres-
sion analyses (controlling for it or not did not affect any of the
results). As can be seen, a regression on creative agreements
was conducted with timing of first offers and information
Timing of first
offers
(late vs. early)
Information exchange
between the parties
Creative agreements
which met underlying
interests
B = 1.05
*
/ B = .72
ns
β = .33
***
B = .04
**
/ B = .04
*
Figure 2. Experiment 2b: Mediation by information exchange.
Note: We controlled for duration of the negotiation in all regression
analyses.
* p < .05, ** p < .01, *** p < .001.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

822 Personality and Social Psychology Bulletin 39(6)
exchange as simultaneous predictors. In this regression, the
timing effect of first offers was not significant (B = .72, SE =
.52; Wald(1) = 1.92, Exp(B) = 2.06, p > .16), whereas the
effect of information exchange was significant (B = .037, SE
= .016; Wald(1) = 5.23, Exp(B) = 1.04, p = .02). To test the
significance of the indirect effect (i.e., the path through the
mediator), we used a bootstrapping procedure, as recom-
mended by Shrout and Bolger (2002). The result of 1,000
resamples demonstrated that zero fell outside the 95% CI for
the indirect effect (95% CI Low = .084; 95% CI High = 1.08),
controlling for duration of the negotiation. Thus, information
exchange mediated the effect of late versus early first offers
on creative agreements, controlling for duration.
Supplementary Analysis. We also conducted a supplementary
analysis on the critical piece of information. Late first offers
increased the tendency for the seller to share the critical piece
of information, that is, what they would do after their sail-
boat trip (M = 6.78, SD = 2.84) compared with early first
offers (M = 5.37, SD = 3.45); F(1, 100) = 5.11, p < .03, η
P
2
=
.05, controlling for duration. This effect was similar whether
controlling for duration or not. We examined whether the
seller’s sharing the critical piece of information mediated the
timing effect of first offers. A bootstrapping procedure using
1,000 resamples demonstrated that indeed this was the case
(95% CI low = .019; 95% CI high = 1.23), controlling for
duration. Combined with the main analysis about the dyad’s
general focus on exchanging information, this supplemen-
tary analysis suggests that the actual content of the informa-
tion being exchanged indeed drives our effect.
General Discussion
The current research demonstrates that the timing of making
first offers can significantly affect negotiation outcomes.
Three experiments involved face-to-face interactions and
found that making a first offer later rather than earlier in a
negotiation facilitated the discovery of creative agreements.
Experiment 1 demonstrated that making a late first offer led
to final agreements that better met the parties’ underlying
interests compared with making an early first offer and a
control condition. This study also replicated the first offer
bargaining advantage—making the first offer, regardless of
timing, led negotiators to get a better outcome on the dis-
tributive issue of sale price.
Experiments 2a and 2b conceptually replicated the benefi-
cial effect of making a late first offer on creative agreements.
These experiments also demonstrated that the timing effect
on creative agreements was independent of the duration of the
negotiation, thus ruling out an important alternative explana-
tion. The last two studies further showed that the beneficial
effect of late first offers on creative agreements was mediated
by greater information exchange. Thus, a late first offer facili-
tated information exchange, and, ultimately, enabled negotia-
tors to meet their underlying, fundamental interests.
Creativity in Negotiation
Prior negotiation research has largely focused on tasks in
which the set of issues was fixed in nature (notable excep-
tions include Galinsky et al., 2008; Giacomantonio et al.,
2010; Kray et al., 2009; Kray & Haselhuhn, 2007; Maddux
& Galinsky, 2009). Indeed, prior research on first offers has
typically used tasks in which the set of issues was fixed—
often, single-issue negotiation over quantifiable issues such
as price (Galinsky et al., 2005; Galinsky & Mussweiler,
2001; Magee et al., 2007).
Departing from that prior research, the current work
explored negotiations that had more creative potential to sat-
isfy negotiators’ underlying interests. In our studies, people
had leeway to add meaningful ideas and issues to the nego-
tiation, thus discovering hidden solutions that creatively inte-
grated their interests. It is important to note that the ability to
add issues and construct creative agreements is quite com-
mon in actual conflict and negotiation (Bazerman & Neale,
1992; Fisher & Ury, 1981). Furthermore, thinking creatively
beyond existing issues and uncovering underlying interests
is an important way to expand the pie in negotiation, thus a
key standard of negotiators’ efficiency (Bazerman & Neale,
1992; Fisher & Ury, 1981; Giacomantonio et al., 2010;
Loewenstein & Howell, 2010).
Motivated Information Exchange in Negotiation
The current research also speaks to the literature on moti-
vated information exchange and processing in negotiations
and joint decisions. Specifically, the current results support
the theory that motivated information exchange and process-
ing improves joint outcomes in negotiation (De Dreu,
Beersma, Stroebe, & Euwema, 2006; De Dreu et al., 2000;
De Dreu et al., 2008; Sinaceur, 2010; Ten Velden et al., 2010;
Thompson & Hastie, 1990) in that making a later first offer
proved to be a critical antecedent of information exchange.
Our work extends this theory in two ways: (a) by identify-
ing a new, behavioral determinant of information exchange,
that is, when to make the first offer; and (b) by suggesting
that the exchange of information also needs to be studied in
contexts where efficiency resides in uncovering underlying
interests. By focusing on creative agreements (i.e., agree-
ments that uncover underlying interests), the current work
departs from prior work on motivated information exchange,
which examined trading-off between issues within fixed sets
of issues.
In this way, later first offers seem to provide a potent pos-
sibility of focusing on exchanging information. This result is
consistent with the notion that parties often come to negotia-
tion with the expectation that the pie is fixed (Bazerman &
Neale, 1992; Thompson & Hastie, 1990) but that this expec-
tation is especially strong early in the negotiation (De Dreu
et al., 2000; Thompson & Hastie, 1990). In this context,
making a later first offer may give a chance for the parties to
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 823
revise their competitive biases before engaging in a detailed
discussion of offers, thus opening more and exchange more
information. In this way, our results are consistent with prior
research that showed that judging and acting with patience
rather than impatience (e.g., having a lower need for cogni-
tive closure) makes negotiators more open, less susceptible
to biases, and more likely to move away from existing posi-
tions (De Dreu, Koole, & Oldersma, 1999), which, in turn,
improves joint outcomes for the dyad (De Dreu, 2003;
Sinaceur, 2010; Ten Velden et al., 2010).
Timing Matters in Negotiation
The current findings also support a stage-based perspective
on negotiation. Negotiation theorists have long argued that
problem solving, or the free discussion and generation of
new options through information exchange about priorities,
is a crucial stage in negotiation, but one that comes late in the
process (Morris & Keltner, 2000; Pruitt, 1981; Pruitt &
Rubin, 1986). Typically, parties are reluctant to exchange
information early in the negotiation (Morris & Keltner, 2000;
Pruitt, 1981; Pruitt & Rubin, 1986), consistent with the
fixed-pie expectation that they bring early to the table
(Bazerman & Neale, 1992; De Dreu et al., 2000; Thompson
& Hastie, 1990). Thus, by delaying making first offers, nego-
tiators facilitate information exchange before stating their
positions (see Morris & Keltner, 2000; Pruitt, 1981).
Importantly, Experiments 2a and 2b demonstrated that the
beneficial effect of late first offers was independent of the
duration of the negotiation. Thus, the current results resonate
with recent research that showed that the timing of strategies
generally affects information exchange and joint outcomes
(Lount, Zhong, Sivanathan, & Murnighan, 2008; Moore,
Kurtzberg, Thompson, & Morris, 1999; Swaab et al., 2011).
Combined with this recent research, the present findings sug-
gest that, in conflict and negotiation, the efficiency of strate-
gies relies not only on their nature but also on when they are
executed.
Offers and Anchoring
The current results qualify the abundant literature on first
offers (e.g., Galinsky et al., 2005; Galinsky & Mussweiler,
2001; Magee et al., 2007; Moran & Ritov, 2002) and anchor-
ing (e.g., Galinsky et al., 2002; Ritov, 1996) in negotiations.
Anchoring through first offers is among the most potent
strategies in negotiations (e.g., Galinsky & Mussweiler,
2001; Magee et al., 2007) by providing “a powerful tool to
influence the outcome of a negotiation for one’s advantage”
(Galinsky & Mussweiler, 2001, p. 665). Thus far, first offers
had been found to backfire only when recipients feel offended
(Schweinsberg, Ku, Wang, & Pillutla, 2012) or think about
counterarguments (Maaravi et al., 2011). Departing from this
prior research by focusing on the creativity of agreements,
the current studies document for the first time that the timing
of first offers is critical in contexts where adding novel issues
and “thinking outside the box” creates important value for
both parties. Specifically, late first offers lead to more cre-
ative outcomes that better meet parties’ underlying interests,
compared with a control (Experiment 1) and early first offers
(Experiments 1-2).
The current research still supports the bargaining advan-
tage provided by making the first offer. Making the first offer
led to a better distributive outcome in Experiment 1 regard-
less of the timing of that first offer. Thus, it is important for
negotiators to recognize whether they are part of a negotia-
tion that has the potential to add issues or is a pure distribu-
tive situation in which there are no additional issues that can
satisfy underlying interests. Of course, negotiators and their
fixed-pie beliefs vastly underestimate the creative possibili-
ties that exist at the bargaining table (Bazerman & Neale,
1992; Thompson & Hastie, 1990).
Limitations and Future Research
The current research is the first to investigate the timing effect
of first offers on creative agreements. However, this basic
effect is likely to be subject to several important boundary
conditions. First, an important limitation relates to what might
happen in a negotiation based on a fixed set of issues where
additional, creative issues cannot be added (see Bazerman &
Neale, 1992). In this situation, the benefits of late first offers
may not hold when value creation depends on trading-off
between extant issues rather than adding creative issues.
Thus, it will be important to test whether the late-first-offer
effect only affects negotiations that are rather “open” and can
include new issues discovered over the course of a negotia-
tion or whether it extends to “closed” negotiation where the
set of issues is fixed from the beginning.
Another important limitation of our studies is that we
examined negotiations that were “open” but in which price
might have seemed salient. For example, in the task used for
Experiments 2a and 2b, dyads often reach impasses because
they tend to focus only on the sale price of the restaurant
(Kray et al., 2009; Maddux & Galinsky, 2009). Therefore,
future research should examine whether our effects extend
to similarly “open” negotiations but in which price is less
salient.
In addition, another boundary condition is that we exam-
ined the timing of first offers in relatively short negotiations.
The short time frames of our studies allowed us to introduce
a high level of experimental control. Specifically, we experi-
mentally limited duration across conditions to rule out dura-
tion as an alternative explanation for our results. Accordingly,
there was relatively little variance along duration. Future
research should investigate the impact of first offers using a
longer time-window, which would allow an examination of
whether long versus short duration moderates our effects.
Perhaps the benefits of late first offers would be attenuated
when negotiations unfold over a long period of time.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

824 Personality and Social Psychology Bulletin 39(6)
Furthermore, our findings may be limited to situations in
which negotiators have fairly cooperative goals overall.
When negotiators have strong competitive goals (e.g., due to
a proself rather than prosocial orientation), they may not
become more open to exchanging information about their
interests even later in the negotiation. Thus, negotiators’
goals may not always lead to productive information
exchange and converging positions (De Dreu et al., 2006).
Therefore, it is important to note that the beneficial effect of
late first offers may be limited to negotiations that involve
relatively a priori open-minded negotiators. Future research
should examine how negotiators’ preexisting goals and
expectations moderate our effects.
Conclusion
Three experiments involved face-to-face interactions in
which negotiators could explore creative options and meet
underlying interests. Experiments 1 and 2 showed that
making a first offer later in the negotiation led to more cre-
ative agreements that better met underlying interests.
Experiments 2a and 2b further showed that the beneficial
effect of late first offers was mediated by greater informa-
tion exchange.
Our findings speak to four distinct streams of research on
conflict and negotiation. First, they resonate with the notion
that creativity is an important outcome in negotiations and
that future research may further consider what makes nego-
tiators think outside the box (Bazerman & Neale, 1992;
Giacomantonio et al., 2010; Loewenstein & Howell, 2010).
Second, our results support a motivated information-
exchange perspective on conflict and negotiation (De Dreu
et al., 2008; Ten Velden et al., 2010). Yet, they extend that
perspective to the realm of creative agreements. Third, our
results add to recent research that showed that the timing of
moves critically affects information exchange and joint out-
comes (Lount et al., 2008; Moore et al., 1999; Swaab et al.,
2011). Finally, the current studies add an important caveat to
the impact of first offers in negotiations (Galinsky &
Mussweiler, 2001; Magee et al., 2007), showing that consid-
ering the timing of first offers is critical to meet underlying
interests.
Combined with previous research, the prescription
derived from the current research may be twofold. Be the one
to make the first offer (you will most often get a better indi-
vidual outcome) but do not make it too early so that you can
exchange information about underlying interests. Making
the first offer, but making it later, lets you have your cake and
eat it too—a more creative agreement with an icing of a bet-
ter distributive outcome.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
research was funded by the INSEAD Alumni Fund (INSEAD Grant
2520-502) through an INSEAD R&D committee grant that was
awarded to the first author.
Note
1. Participants from one dyad did not answer this item, which
accounts for fewer degrees of freedom in this analysis.
References
Bazerman, M. H., & Neale, M. A. (1992). Negotiating rationally.
New York, NY: Free Press.
Bechtoldt, M. N., De Dreu, C. K. W., Nijstad, B. A., & Choi, H.-S.
(2010). Motivated information-processing, social tuning, and
group creativity. Journal of Personality and Social Psychology,
99, 622-637.
De Dreu, C. K. W. (2003). Time pressure and closing of the mind
in negotiation. Organizational Behavior and Human Decision
Processes, 91, 280-295.
De Dreu, C. K. W., Beersma, B., Stroebe, K., & Euwema, M.
(2006). Motivated information processing, strategic choice,
and the quality of negotiated agreement. Journal of Personality
and Social Psychology, 90, 927-943.
De Dreu, C. K. W., Koole, S., & Oldersma, F. (1999). On the seizing
and freezing of negotiator inferences: Need for cognitive clo-
sure moderates the use of heuristics in negotiation. Personality
and Social Psychology Bulletin, 25, 348-362.
De Dreu, C. K. W., Koole, S., & Steinel, W. (2000). Unfixing
the fixed-pie: A motivated information-processing approach
to integrative negotiation. Journal of Personality and Social
Psychology, 79, 975-987.
De Dreu, C. K. W., Nijstad, B., & Van Knippenberg, D. (2008).
Motivated information processing in group judgment and deci-
sion making. Personality and Social Psychology Review, 12,
22-49.
Fisher, R., & Ury, W. (1981). Getting to yes. New York, NY:
Penguin.
Galinsky, A. D., Leonardelli, G., Okhuysen, G., & Mussweiler, T.
(2005). Regulatory focus at the bargaining table: Promoting
distributive and integrative success. Personality and Social
Psychology Bulletin, 31, 1087-1098.
Galinsky, A. D., Maddux, W., Gilin, D., & White, J. (2008). Why
it pays to get inside the head of your opponent: The differential
effects of perspective-taking and empathy in strategic interac-
tions. Psychological Science, 19, 378-384.
Galinsky, A. D., & Mussweiler, T. (2001). First offers as anchors:
The role of perspective-taking and negotiator focus. Journal of
Personality and Social Psychology, 81, 657-669.
Galinsky, A., Mussweiler, T., & Medvec, V. (2002).
Disconnecting outcomes and evaluations: The role of nego-
tiator focus. Journal of Personality and Social Psychology,
83, 1131-1140.
Giacomantonio, M., De Dreu, C. K. W., & Mannetti, L. (2010). Now
you see it, now you don’t: Interests, issues, and psychological
distance in integrative negotiation. Journal of Personality and
Social Psychology, 98, 761-774.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from

Sinaceur et al. 825
Huber, V., & Neale, M. A. (1986). Effects of cognitive heuristics
and goals on negotiator performance and subsequent goal set-
ting. Organizational Behavior and Human Decision Processes,
38, 342-365.
Israeli–Palestinian Joint Statement at Taba. (2001, January 27).
Retrieved from http://www.mfa.gov.il/MFA/MFAArchive/
2000_2009/2001/1/Israeli-Palestinian%20Joint%20
Statement%20-%2027-Jan-2001
Kray, L. J., Galinsky, A., & Markman, K. (2009). Counterfactual
structure and learning from experience in negotiations. Journal
of Experimental Social Psychology, 45, 979-982.
Kray, L. J., & Haselhuhn, M. P. (2007). Implicit negotiation beliefs
and performance: Experimental and longitudinal evidence.
Journal of Personality and Social Psychology, 93, 49-64.
Kray, L. J., Thompson, L., & Galinsky, A. (2001). Battle of the
sexes: Gender stereotype activation in negotiations. Journal of
Personality and Social Psychology, 80, 942-958.
Loewenstein, J., & Howell, T. (2010, June). Understanding and
using what we want: Interests and exploitation in negotiations.
Paper presented at the International Association for Conflict
Management (23rd Annual Meeting), Boston, MA.
Lount, R. B., Zhong, C., Sivanathan, N., & Murnighan, K. (2008).
Getting off on the wrong foot: Restoring trust and the timing of
breach. Personality and Social Psychology Bulletin, 34, 1601-
1612.
Maaravi, Y., Ganzach, Y., & Pazy, A. (2011). Negotiation as a form
of persuasion: Arguments in first offers. Journal of Personality
and Social Psychology, 101, 245-255.
Maddux, W. W., & Galinsky, A. D. (2009). Cultural borders and
mental barriers: The relationship between living abroad and
creativity. Journal of Personality and Social Psychology, 96,
1047-1061.
Maddux, W. W., Mullen, E., & Galinsky, A. (2008). Chameleons
bake bigger pies and take bigger pieces: Strategic behav-
ioral mimicry facilitates negotiation outcomes. Journal of
Experimental Social Psychology, 44, 461-468.
Magee, J. C., Galinsky, A. D., & Gruenfeld, D. (2007). Power, pro-
pensity to negotiate, and moving first in competitive interac-
tions. Personality and Social Psychology Bulletin, 33, 200-212.
Malley, R. (2010). Middle East peace: So why have we failed?
[Interview]. Foreign Policy. Retrieved from http://www.for-
eignpolicy.com/articles/2010/04/19/so_why_have_we_failed
Miller, A. (2008). The much too promised land: America’s elusive
search for Arab-Israeli peace. New York, NY: Bantam Books.
Moore, D., Kurtzberg, T., Thompson, L., & Morris, M. W. (1999).
Long and short routes to success in electronically mediated nego-
tiations: Group affiliations and good vibrations. Organizational
Behavior and Human Decision Processes, 77, 22-43.
Moran, S., & Ritov, I. (2002). Initial perceptions in negotiations:
Evaluation and response to “logrolling” initial offers. Journal
of Behavioral Decision Making, 15, 101-124.
Morris, M. W., & Keltner, D. (2000). How emotions work: An
analysis of the social functions of emotional expression in
negotiations. Research in Organizational Behavior, 22, 1-50.
Pruitt, D. (1981). Negotiation behavior. New York, NY: Academic
Press.
Pruitt, D., & Rubin, J. (1986). Social conflict: Escalation, stale-
mate, and settlement. New York, NY: McGraw-Hill.
Ritov, I. (1996). Anchoring in a simulated competitive market
negotiation. Organizational Behavior and Human Decision
Processes, 67, 16-25.
Schweinsberg, M., Ku, G., Wang, C., & Pillutla, M. (2012). Starting
high and ending with nothing: The role of anchors and power in
negotiations. Journal of Experimental Social Psychology, 48,
226-231.
Shrout, P., & Bolger, N. (2002). Mediation in experimental and
nonexperimental studies: New procedures and recommenda-
tions. Psychological Methods, 7, 422-445.
Sinaceur, M. (2010). Suspending judgment to create value:
Suspicion and trust in negotiation. Journal of Experimental
Social Psychology, 46, 543-550.
Swaab, R., Maddux, W., & Sinaceur, M. (2011). Early words
that work: How and when virtual linguistic mimicry facili-
tates negotiation outcomes. Journal of Experimental Social
Psychology, 47, 616-621.
Ten Velden, F. S., Beersma, B., & De Dreu, C. K. W. (2010). It
takes one to tango: Epistemic motivation and information in
integrative negotiation. Personality and Social Psychology
Bulletin, 36, 1454-1466.
Thompson, L. L. (1991). Information exchange in negotiation.
Journal of Experimental Social Psychology, 27, 161-179.
Thompson, L., & Hastie, R. (1990). Social perception in negotia-
tion. Organizational Behavior and Human Decision Processes,
47, 98-123.
at INSEAD - Library on May 29, 2013psp.sagepub.comDownloaded from