See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/275699988

Inter- and Intracultural Negotiation: U.S. and Japanese Negotiators

Article in Academy of Management Journal · October 1998

DOI: 10.2307/256938

CITATIONS READS

402 4,855

2 authors, including:

Tetsushi Okumura
Toyo University
14 PUBLICATIONS 1,507 CITATIONS

SEE PROFILE

All content following this page was uploaded by Tetsushi Okumura on 02 November 2016.

The user has requested enhancement of the downloaded file.

Inter- and Intracultural Negotiation: U.S. and Japanese Negotiators
Author(s): Jeanne M. Brett and Tetsushi Okumura
Source: The Academy of Management Journal, Vol. 41, No. 5 (Oct., 1998), pp. 495-510
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/256938
Accessed: 23/02/2010 16:24

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=aom.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Journal.

http://www.jstor.org
e Academy of Management Journal
1998, Vol. 41, No. 5, 495-510.

INTER- AND INTRACULTURAL NEGOTIATION: U.S. AND

JAPANESE NEGOTIATORS
JEANNE M. BRETT
Northwestern University

TETSUSHI OKUMURA
Shiga University

In this study, we propose that culture provides scripts and schemas for negotiation.
The implications for negotiation of two cultural values, individualism/collectivism and
hierarchy/egalitarianism, are discussed. The primary hypothesis, that joint gains will
be lower in intercultural negotiations between U.S. and Japanese negotiators than in
intracultural negotiations between either U.S. or Japanese negotiators, was confirmed
with data from 30 intercultural, 47 U.S.-U.S. intracultural, and 18 Japanese-Japanese
intracultural simulated negotiations. Tests of secondary hypotheses indicated that
there was less understanding of the priorities of the other party and the utility of a
compatible issue in inter- than in intracultural negotiations. When information about
priorities was available, intercultural negotiators were less able than intracultural
negotiators to use it to generate joint gains.

In order to remain competitive, organizations are Japan and the United States are major business
increasingly engaging in international business partners (Graham & Sano, 1989). Successful nego-
ventures (Lewis, 1990). Intercultural buyer-seller tiations between Japanese and U.S. companies have
transactions and joint ventures between companies implications for the economies of both countries.
have grown exponentially during the 1990s, bring- Yet descriptions of Japanese and U.S. negotiating
ing midrange and smaller companies into interna- styles suggest substantial differences in approach
tional ventures, some for the first time (Lewis, (Graham, 1993; Kato & Kato, 1992; March, 1990)
1990). Managing these intercultural transactions re- that may affect intercultural negotiations. For ex-
quires knowing not just how to negotiate success- ample, a vice president in the Japan merchant
fully with buyers and sellers from your own cul- banking operation at Bankers Trust noted in the
ture, but also how to negotiate with buyers and New York Times Magazine that information is
sellers from other cultures. viewed as an important source of power in negoti-
Western scholars have produced substantial re- ations in both the United States and Japan (Yo-
search on negotiation over the past decade. (See shimura, 1997). U.S. negotiators, he suggested, ex-
Neale and Bazerman [1991], Pruitt and Carnevale ercise the power of information by disclosing it,
[1993], and Thompson [1997] for reviews.) There and in return, they get information from other peo-
has been some cross-cultural research comparing ple. In contrast, the Japanese exercise the power of
negotiations in different cultural contexts (Graham, information by hiding it, he noted, going on to
1993). However, research on negotiations between point out another fundamental difference between
members of different cultures, or intercultural ne- Japanese and U.S. negotiating styles: "For Japanese,
gotiation, has lagged (Leung, 1997). In this study, negotiation is usually a process of reaching a point
we developed hypotheses contrasting inter- and that is acceptable to both parties. For Americans,
intracultural negotiations and tested these hypoth- it's a competition dividing winners and losers"
eses on simulation data gathered from Japanese and (Yoshimura, 1997). Americans, he continued, often
U.S. managers negotiating interculturally and in- open negotiations at a level that is totally unaccept-
traculturally. able to the Japanese, seeing the opening offer as a
starting point, but the Japanese cannot see trust in
The authors would like to acknowledge the helpful
such behavior.
comments on a draft of this article provided by Wendi The purpose of this research was to determine
Adair, Max Bazerman, Don Moore, Anne Lytle, Margaret just how great the cultural differences between Jap-
Neale, Debra Shapiro, and Catherine Tinsley and the anese and U.S. negotiators are and whether these
financial support of the Dispute Resolution Research cultural differences interfere with the negotiation
Center, Northwestern University. of joint gains in intercultural as opposed to intrac-
495
496 Academy of Management Journal October

ultural negotiations. We chose joint gains as the contribute to the negotiation of joint gains. For
dependent variable in our study because joint gains example, there appear to be two different informa-
are relatively difficult to negotiate in any culture tion-sharing scripts-seeking and sharing informa-
(Neale & Bazerman, 1991), and we thought that tion about preferences and priorities (Olekalns,
strategies for negotiating joint gains might be dif- Smith, & Walsh, 1996; Pruitt, 1981; Weingart,
ferent across cultures. Joint gains are only one of Thompson, Bazerman, & Carroll, 1990) and heuris-
the outcomes of negotiation. However, they are an tic trial-and-error search (Pruitt & Lewis, 1975;
important outcome, since the creation of joint gains Tutzauer & Roloff, 1988)-that facilitate the nego-
can bridge a nonoverlapping bargaining zone, thus tiation of joint gains. Negotiation schemas (precon-
making an agreement possible when no agreement ceptions about negotiation) related to the negotia-
seemed likely; and, when an overlapping bargain- tion of joint gains include avoidance of premature
ing zone exists, the creation of joint gains means closure (Olekalns & Smith, 1996) and eschewal of
taking maximum advantage of available resources power (Pruitt, 1981; Pruitt & Lewis, 1975). A coop-
(Raiffa, 1982). erative motivational orientation, defined as a nego-
We begin with a review of the research identify- tiator's having goals for self and for the dyad or
ing the knowledge structures (or schemas and group (Weingart, Bennett, & Brett, 1993), and a
scripts, in the parlance of social cognition) that are mixed motive orientation, defined as expectations
associated with the negotiation of joint gains. We that preferences may not be completely opposed
propose that such knowledge structures are formed (Pinkley, Griffith, & Northcraft, 1995), in contrast to
by culture (among other influences) and develop a fixed-pie schema, defined as the assumption that
hypotheses about differences between U.S. and Jap- negotiations are necessarily distributive (Thomp-
anese schemas for negotiation. These hypotheses son & Hastie, 1990), also appear to be schemas
lead to the study's major research question,
associated with the negotiation of joint gains.
whether intercultural negotiations between U.S.
Comparative cross-cultural research document-
and Japanese negotiators are less effective in gen-
ing differences between U.S. and Japanese cultural
erating joint gains than intracultural negotiations
values (Hofstede, 1980; Schwartz, 1994) provides a
between U.S. or Japanese negotiators.
basis for developing hypotheses about how negoti-
ation scripts and schemas differ in the U.S. and
Japanese cultures. The Japanese culture is collec-
THEORY AND HYPOTHESES
tivist and hierarchical; the U.S. culture is individ-
Culture and Negotiation ualistic and egalitarian (Hofstede, 1980; Schwartz,
Culture, or a society's characteristic profile with 1994). Both of these cultural values appear to have
respect to values, norms, and institutions (Lytle, implications for the scripts and schemas that a cul-
Brett, Barsness, Tinsley, & Janssens, 1995) provides ture's members bring with them to a negotiation.
insight into the different solutions that societies Individualism versus collectivism and the self-
evolve to manage social exchanges such as negoti- interest schema. In individualist cultures, the
ation. Culture is a socially shared knowledge struc- definition of self is independent from in-group
ture, or schema, giving meaning to incoming stim- membership; in collectivist cultures, it is interde-
uli and channeling outgoing reactions (Triandis, pendent with in-group membership (Marcus &
1972). Cultural values (what is important) and Kityama, 1991; Triandis, 1989). In individualist
norms (what is appropriate) provide the members cultures, goals are independent of those of the in-
of a cultural group with schemas, or templates, for group; in collectivist cultures, goals are aligned
interpreting both the situation and the behavior of with those of the in-group (Triandis, 1989). In in-
others (Fiske & Taylor, 1991), and scripts, or se- dividualist cultures, there is an emphasis on per-
quences of appropriate social action (Shank & Abel- sonal needs; in collectivist cultures, the emphasis
son, 1977). Cultural institutions provide contexts is on social obligations (Triandis, 1989). The link-
for negotiations. Negotiation schemas are cognitive age of goals to self as opposed to the collective and
warehouses of information and expectations about the emphasis on personal needs as opposed to so-
negotiation (Thompson, 1997). Negotiation scripts cial obligations suggest that individualists should
are subsets of negotiation schemas. They store ac- be more self-interested in negotiations than collec-
tion plans in the form of behavioral sequences. tivists.
These plans can be drawn upon for enacting sche-
mas. Hypothesis 1. The cultural value of individu-
Prior research, carried out within the U.S. cul- alism versus collectivism will be related to a
ture, has identified several scripts and schemas that high self-interest negotiation schema.
1998 Brett and Okumura 497

Prior research has shown that when all negotia- but that Japanese negotiators hide information. In
tors in a dyad or group subscribe to a self-interest her face-negotiation theory, Ting-Toomey (1988)
schema, or an individualistic social or motivational proposed that individualists communicate directly
orientation, they may risk impasse; however, when and collectivists, indirectly. When Ohbuchi and
they do reach agreements, their goals appear to Takahashi (1994) asked Japanese and American
motivate them to realize high joint gains (Weingart people to describe their conflict management be-
et al., 1993). Huber and Neale (1986) also found havior, the Japanese said they used more indirect
that joint gains were facilitated when both negoti- methods, such as suggestions, ingratiation, impres-
ators in a dyad had moderate to difficult goals. sion management, and appeasement; the Ameri-
Hierarchy versus egalitarianism and the power cans said they used more direct methods, such as
schema. The cultural value hierarchy versus egal- persuasion, bargaining, and compromise. Thus, al-
itarianism has implications for how power is per- though it may seem to a U.S. negotiator that a
ceived in a culture. In hierarchical cultures, there is Japanese negotiator does not engage in information
a preference for differentiated social status. Social sharing, it may only be that the two cultures' scripts
status implies social power in a variety of contexts, for information sharing are incompatible. Further-
including negotiations. Low-status members of a more, the individualist negotiator whose own style
society are expected to concede to high-status is direct may not know how to glean information
members, who in turn have a social responsibility from the collectivist's indirect communication
to look out for the needs of the lower-status mem- script.
bers (Leung, 1997). In intercultural negotiations between U.S. and
Social status differences exist in egalitarian cul- Japanese negotiators, if scripts for information shar-
tures, but people are less receptive to power differ- ing are different, there should be less understand-
ences in egalitarian societies than in hierarchical ing than there is in intracultural negotiations,
ones (Leung, 1997). Social status may not automat- where scripts are similar. If Japanese negotiators
ically convey negotiating power in egalitarian soci- follow their culturally appropriate script and
eties, because status differences may be down- search for and provide information indirectly, U.S.
played, in keeping with the cultural value of negotiators may not be able to interpret the infor-
egalitarianism. Egalitarians expect equal engage- mation conveyed indirectly. U.S. negotiators in in-
ment in social intercourse, but those from hierar- tercultural negotiations therefore may end up un-
chical cultures have unidirectional expectations derstanding less about the preferences and
(Leung, 1997). In an egalitarian culture, a party's priorities of their Japanese opponents than U.S.
negotiating power may be tied to the best alterna- negotiators in intracultural negotiations, whose op-
tive to a negotiated agreement (BATNA) and may ponents are likely to share a similar information
therefore vary from one negotiation to another. In a script.
hierarchical culture, power is associated with the
party's status in the social structure. Since this Hypothesis 3. U.S. negotiators in intercultural
status is unlikely to change drastically from one negotiations with Japanese negotiators will
negotiation to another, power in hierarchical cul- have less understanding of their opponents'
tures may be viewed as fixed. priorities and preferences than U.S. negotia-
The relative emphasis on power in hierarchical tors in intracultural negotiations.
cultures, compared to egalitarian cultures, suggests
that power will be a more important schema for Japanese negotiators in intercultural negotia-
negotiators in hierarchical cultures than in egalitar- tions should understand the direct communica-
ian cultures. tions of U.S. negotiators regarding their own pref-
erences and priorities. However, if the Japanese
Hypothesis 2. The cultural value of hierarchy
negotiators do not reciprocate with direct infor-
versus egalitarianism will be related to a power
mation sharing, the U.S. negotiators may provide
schema for negotiation.
less information when negotiating cross-cultur-
Scripts for information exchange in negotia- ally than when negotiating with same-culture
tion. Descriptive accounts (March, 1990; Yo- partners.
shimura, 1997), theory (Ting-Toomey, 1988), and
some empirical research (Ohbuchi & Takahashi, Hypothesis 4. Japanese negotiators in intercul-
1994) suggest that the scripts U.S. and Japanese tural negotiations with U.S. negotiators will
negotiators use to gather information are quite dif- have less understanding of their opponents'
ferent. Yoshimura (1997) observed that U.S. nego- priorities and preferences than Japanese nego-
tiators share information in order to get information tiators in intracultural negotiations.
498 Academy of Management Journal October

Intercultural Negotiations and Joint Gains that incompatible negotiation schemas and scripts
embedded in cultural values can be easily changed
U.S. negotiators do not always reach optimal
or even adjusted during the course of a single ne-
agreements (Neale & Bazerman, 1991; Pruitt &
gotiation. Thus, we expected that joint gains would
Carnevale, 1993). A comparative, cross-cultural ne-
be lower in intercultural than in intracultural ne-
gotiation study that used a three-issue exercise
gotiations.
with integrative potential (Pruitt, 1981) showed
that Japanese pairs received the highest joint profits Hypothesis 5. Intercultural negotiators will re-
of any national group; U.S. negotiators' joint profits alize lower joint gains than intracultural nego-
were in the middle of the range (Graham, 1993). We tiators.
expected that relatively compatible cultural values
and negotiation schemas and scripts regarding self- We expected cultural differences in schemas and
interest, power, and information sharing would fa- scripts to impact information sharing in intercul-
cilitate intracultural negotiators' reaching better tural negotiations in such a way that intercultural
agreements than intercultural negotiators. negotiators would be less likely than intracultural
When negotiators' schemas and scripts are in- negotiators to identify optimal integrative trade-
compatible owing to culture or other reasons, their offs. We also thought that cultural differences
negotiation process may be ineffective and their would limit heuristic trial-and-error search for bet-
negotiated outcomes suboptimal. It may be as ter alternatives and contribute to premature clo-
though two actors in the same play have two dif- sure. This reasoning suggested intercultural nego-
ferent scripts and two different sets of stage direc- tiators would be less likely than intracultural
tions, each contributed by a different playwright. negotiators to include a compatible but optional
Although intercultural negotiators may still reach issue in their agreements.
agreements, their agreements, compared to those of Low joint gains may result from negotiators' hav-
intracultural negotiators, may be suboptimal in that ing incompatible schemas and scripts that lead to a
opportunities for joint gains are missed. lack of information sharing or a lack of motivation
The research on norm development (Betten- to continue searching for better alternatives. For
hausen & Murnighan, 1985) and conflict frames example, if a negotiator from a collectivist culture
(Pinkley, 1990; Putnam & Holmer, 1992) suggests uses indirect communication and a negotiator from
that negotiators whose norms or frames do not an individualist culture uses direct information
match may have difficulty reaching agreement. Bet- sharing, the "disconnect" in information-sharing
tenhausen and Murnighan (1985) proposed that scripts may result in low information sharing and
when parties define a situation (such as a negotia- low joint gains.
tion) similarly but have different scripts, their ini- Hypothesis 6. When intercultural negotiators'
tial interaction may go smoothly, but resolving dif- scripts and schemas are incompatible, joint
ferences later may be problematic, unless they gains will be lower than they will be when
negotiate their scripts and take a common approach intercultural negotiators' scripts and schemas
to the task. The research and theorizing about con- are compatible.
flict frames suggest that disputants with similar
frames or frames that converge are likely to reach Intercultural negotiators may anticipate that their
agreement but that disputants whose frames are approaches to negotiation will be incompatible, or
very different may fail to reach agreement (Drake & they may sense the incompatibility once a negoti-
Donohue, 1994; Pinkley & Northcraft, 1994).1 The ation begins and try to make adjustments. If their
implication of the norm development and conflict adjustments are successful, intercultural negotia-
frames literatures for our research on intercultural tors should be as capable of generating high joint
negotiation is that when cultural differences result gains as intracultural negotiators, nullifying Hy-
in different or conflicting negotiation scripts and pothesis 6.
schemas, a negotiated agreement may be less than
optimal, unless schemas and scripts can them-
METHODS
selves be negotiated. However, it seems unlikely
Simulation
1 A conflict frame is a perceptual set or orientation that The data were collected in conjunction with the
leads disputants to focus on some characteristics of a simulated negotiation of a buyer-seller transaction.
conflict while ignoring others (Deutsch, 1975) and that The simulation, "Cartoon," is based on the exercise
invokes particular schemas and scripts (Pinkley & North- "Working Women" (Tenbrunsel & Bazerman,
craft, 1994). 1995). In "Cartoon," the seller is a major film pro-
1998 Brett and Okumura 499

duction company and the buyer is an independent wide. (For further information about the worldwide
television station in a large metropolitan area. The syndication of cartoons, see Mifflin [1995].)
seller wishes to syndicate (sell the rerun rights for) Two versions of each role were produced: (1) a
a 100-episode cartoon, Ultra Rangers. The issues U.S. buyer (a U.S. television station) and a Japanese
include the price per episode, the number of times buyer (a Japanese television station) and (2) a U.S.
each episode will be allowed to run, the financing seller (a U.S. film company) and a Japanese seller (a
arrangement, and whether another cartoon, Strums, Japanese film company). The U.S. and Japanese
is to be added to the deal. The alternatives are buyers received the same substantive information,
quantified, as is the BATNA. There is the potential as did the U.S. and Japanese sellers. Participants,
for an integrative trade-off between the number of who were American and Japanese managers, al-
runs, which is more important to the buyer than to ways played the role of a party from their own
the seller, and financing, which is more important culture. In the intercultural simulation, half the
to the seller than to the buyer. Strums is a compat- teams had a U.S. buyer (played by a U.S. partici-
ible issue; inclusion generates value for both par- pant) and a Japanese seller (played by a Japanese
ties, although that value must still be distributed. participant); the other half of the teams had a Jap-
Table 1 shows the positions of each party on the anese buyer (played by a Japanese participant) and
issues. Table 1 also shows the explicit information a U.S. seller (played by a U.S. participant). In the
negotiators were given about the value to them of intracultural simulations, U.S. buyers negotiated
their alternative deal (BATNA). with U.S. sellers, and Japanese buyers negotiated
The program in the original exercise, "Working with Japanese sellers.
Women," is a sitcom. We decided to change the
program to a cartoon, because cartoons are popular
Participants
television shows in both the United States and
Japan. Furthermore, film companies in both coun- To obtain the intercultural data set, we invited
tries produce cartoons and syndicate them world- U.S. and Japanese managers to participate in a one-
day executive program on intercultural negotiation.
U.S. participants were selected from the alumni of
TABLE 1
a major Midwest business school who lived in the
"Cartoon": Positions of the Parties
greater metropolitan area close to the school and
Issue T.V. Station Film Company worked in finance, banking, manufacturing, or
trading. Japanese participants were identified as
Revenue $ 8,400,000 n.a. follows: First, 210 Japanese companies with offices
Price per episode
in the metropolitan area were identified. The top
Limit $ 60,000 $ 35,000
Aspiration $ 30,000 $ 70,000
local manager was sent a letter describing the in-
Runs-per-episode adjustment tercultural negotiation program and asking him (all
4 $(1,680,000) $ 500,000 were men) to select a Japanese manager whose day-
5 (840,000) 250,000 to-day business for the company was conducted in
6 0 0
English. The invitation made it clear that the pro-
7 840,000 (250,000)
8 1,680,000 (500,000)
gram would be conducted in English. Once a man-
Financing savings/cost ager was nominated for the program, he (all were
Year 1 10% -20% men) was sent information about the program (the
Year 2 20 -35 time and location, etc.), as well as notification that
Year 3 30 -50
the program would be conducted in English. We
Year 4 40 -60
Year 5 50 -70 emphasized that the program would be conducted
StrumSa in English in order to discourage Japanese manag-
Reservation price $ 20,000 $ 10,000 ers who might be uncomfortable negotiating in En-
Ratingsb glish from attending.
6-7 20% 10% Participants providing the intracultural U.S.-U.S.
7-8 50 10
8-9 10 10 data participated in the study as the first negotia-
9-10 10 50 tion exercise in their master's-level course on ne-
10-11 10 20 gotiations. Data from pairs of foreign nationals who
Alternative deal were in the course were not included in the analy-
Valuec $3.0 $2.5
sis. Men were paired with men and women with
a
This was the additional cartoon that was available. women. Since there were no outcome differences
b
The percentages given are estimated likelihoods of a range. between the male and female pairs, the latter were
c Values are millions of dollars. included in the data set.
500 Academy of Management Journal October

Participants providing the intracultural Japa- sulted with each other before answering any
nese-Japanese data set were full-time managers question that had not been anticipated. At the end
who were taking part in a management training of the hour, the preparation partners were split up,
program paid for by their companies. All were men. and each was assigned a negotiation partner. Nego-
There were 60 participants in the intercultural tiations were all one-on-one. Negotiation assign-
negotiation sample, 30 American managers and 30 ments were made so that buyer preparation part-
Japanese foreign assignees; 94 in the U.S. sample; ners were not assigned to seller preparation
and 36 in the Japanese-Japanese sample. Both in- partners. Thus, no two buyer-seller negotiation dy-
tercultural data and intracultural U.S.-U.S. data ads consisted of preparation partners. Participants
were collected at two different times approximately were given one and one-half hours to negotiate. At
one year apart. Intracultural Japanese-Japanese data the end of the negotiation, they completed a form
were collected at two different times in the inter- describing their agreement, if any, and a question-
vening year. There were no significant differences naire with biographical questions and questions
on any measure for any group between the first and about what they did during the preparation and
second data collections. negotiation, their goals, and the other party's prior-
Participants in the U.S.-Japanese sample were ities. All questionnaires were provided in English
older (mean, 38 years) than those in the Japanese- and in Japanese, as appropriate. All Japanese ma-
Japanese sample (mean, 27 years) and in one of the terials were translated and retranslated by two na-
two classes in the U.S. sample (mean, 28 years). tive-speaking, bilingual Japanese nationals. After
The average age of the participants in the other U.S. all data had been collected, results were shown to
class was 37 years. U.S. participants in the inter- the group, and the exercise and the negotiation
cultural negotiation were likely to have had previ- experiences were discussed.
ous negotiation training (64%), because there was a
negotiation training program at the school. Japa-
nese participants in the intercultural negotiation
Measures
were less likely to have had prior negotiation train-
ing (50%). Participants in the U.S.-U.S. and Japa- Dependent variable. The primary dependent
nese-Japanese samples took part in the simulation variable was joint gains. Joint gains were calculated
prior to negotiation training. as follows: Negotiators' net revenue was calculated
by subtracting their BATNA value from the gross
revenue accruing to them from the agreement they
Procedures
had negotiated. Joint gains were calculated by add-
We wrote a standard introduction to the negoti- ing the buyer's net to the seller's net. Appendixes A
ation exercise and used it each time data were and B provide examples of net revenue calculations
collected. This introduction provided an outline of for a buyer and a seller. Only one group failed to
activities and timing and described "Cartoon," the reach an agreement. That group was not included
exercise. Participants were given a short question- in the data set. Maximum possible joint gains were
naire asking about negotiation norms. They were $5.08 million.
assigned to play the role of buyer or seller and Data also were collected on the elements of the
given confidential information for their role. In the agreement: the price paid for the cartoon, the num-
intercultural negotiations, role information was ber of runs, whether or not there were financing
provided in English for the U.S. participants and in terms, and whether Strums was included in the
Japanese and English for the Japanese participants. agreement. It was also possible for negotiators to
In the intracultural negotiations, role information reach a contingent agreement based on the ratings
was provided in each participant's own language. that the cartoon generated in syndication.
Everyone was given a same-role partner and one Independent variables. Independent variables
hour to prepare for the negotiation. Japanese par- were type of negotiation (intercultural versus in-
ticipants in the intercultural negotiation were told tracultural) and culture (U.S. or Japan).
they might prepare in the Japanese language, al- Cultural values. We measured cultural values
though the negotiations would be conducted in prior to preparation and negotiation using items
English. Participants in the Japanese-Japanese from Schwartz's (1994) survey of values. This sur-
teams prepared and negotiated in Japanese. During vey, although perhaps not so well known as Hof-
the one-hour preparation time, the researchers (one stede's (1980) work, is superior in several ways. It
of whom is bilingual in Japanese and English) cir- is based on a conceptualization of values; it was
culated to answer questions. We had prepared stan- developed with systematic sampling, measure-
dard answers to questions in advance and con- ment, and analysis techniques; and perhaps most
1998 Brett and Okumura 501

important, its normative data are recent, collected the parties' alternatives to negotiating with each
in the late 1980s and early 1990s. other (BATNA; Fisher, Ury, & Patton, 1991).
Individualism versus collectivism was measured Immediately after the negotiation, before partici-
using seven items from Schwartz's (1994) survey of pants had shared their results, they answered a
values. Items included choosing own goals (select- series of questions focusing on their preparation.
ing own purposes), capable (competent, effective, These were framed by the following: "During the
efficient), successful (achieving goals), intelligent preparation session, how much did you think
(logical, thinking), self-respect (belief in one's own about. . . ?" ("a lot," "some," "a little,99"not much at
worth), ambitious (hardworking, aspiring), and in- all"). These items factored into two dimensions
dependent (self-reliant, self-sufficient). A high that we used as further measures of power and
score indicated individualism (a = .80). information schemas. Four items were summed
Hierarchy versus egalitarianism was measured into a power scale (a = .82), and 12 items were
with six items from Schwartz's (1994) survey of summed into an information-planning scale (a
values. Items included social power (control over .80). These items are presented in Appendix C.
others, dominance), authority (the right to lead or In the postnegotiation questionnaire, we asked
command), preserving my public image (protecting negotiators to rank the issues in the negotiation
my "face"), wealth (material possessions, money), with respect to their importance to the other party.
social recognition (respect, approval by others), These data were used as indicators of negotiators'
and influential (having an impact on people and scripts. Negotiators were given a list of issues in the
events). A high score indicated hierarchy (a = .77). order in which they were discussed in the case:
Negotiation schemas and scripts. Negotiation price, runs, financing, Strums, and other issues (for
schemas were conceptualized as perceptual sets or example, contingent arrangements). Our interest
was in the ranking of runs and financing, the inte-
cognitive predispositions. They were measured by
grative issues, and Strums, the compatible issue. As
questions asked prior to the preparation and nego-
can be seen in Table 1, runs were of greater value to
tiation. Participants answered a series of questions
the buyer than the seller, since the buyer wanted
framed by the words "What is appropriate behavior
the right to show the cartoon as frequently as pos-
in negotiation?" Forty-five items measured on five-
sible, thereby gaining greater advertising revenue.
point Likert-style response formats anchored by
Table 1 also shows that financing was of greater
"strongly disagree" and "strongly agree" factored
value to the seller than the buyer. The seller pre-
into three dimensions. We used the scales formed
ferred to be paid up front.
from summing the items loading on these dimen- Each issue received a rank between 1 and 6. We
sions as measurements of three negotiation sche- assigned a 6 if an issue was not ranked by a partic-
mas: self-interest, power, and information. The ipant. Categories were collapsed for the purpose of
items composing the self-interest scale did not hold data analysis. For runs and financing, we collapsed
up during reliability analysis, so the scale was trun- rankings of 1 and 2 into one category and rankings
cated into a single item: "It's appropriate to satisfy of 3-6 into another category. This coding was
your own needs." The distributive tactics scale, a based on the importance of issues as presented in
measurement of power, contained five items, in- Table 1. Sellers, for example, should have rated
cluding asking for sympathy, making the first offer, runs as the first or second most important issue to
making counteroffers, avoiding public disputes, buyers, along with price, and buyers should have
and compromising (a = .61). The information-shar- rated runs as the third, fourth, or fifth most impor-
ing scale contained five items, including sharing tant issue to sellers. The coding for Strums was 1 if
information with the other party, collaborating it received a rank of 1-4 and 0 if it received a rank
with the other party, engaging in a give-and-take of 5 or was not ranked. By collapsing categories, we
exchange, sharing information when the other were able to reduce the "ipsative" problem of
party shares, and trying to build a relationship with ranked data, whereby the ranks of the later-ranked
the other party (a = .76). items are dependent on those of the earlier-ranked
In another measurement of schemas, we assessed items. All issues had more than one opportunity to
expectations about factors affecting the outcome of be placed in a rank. Collapsing categories was also
a negotiation. We asked how strongly participants consistent with the nature of the exercise, in which
agreed or disagreed (five-point scale) with the state- there was one important distributive issue. Negoti-
ment that the outcome of a negotiation was a func- ators would not necessarily place their opponent's
tion of (1) how thoroughly the parties shared infor- trade-off issue ahead of the distributive issue.
mation, (2) how powerful the parties' companies Measures for ruling out alternative explana-
were, (3) the parties' roles (buyer, seller), and (4) tions. We asked participants to describe what went
502 Academy of Management Joural October

on during the negotiations by rating 24 items on and joint gains as the dependent variable. The unit
five-point Likert scales ranging from "strongly dis- of analysis was intercultural dyads.
agree" to "strongly agree." These items factored
into two dimensions. We labeled the first dimen-
sion "frustration." Eight items loaded on this di- RESULTS
mension (a = .81; see Appendix C). We labeled the Cultural Group, Cultural Values, and Negotiation
second dimension "cooperation/reciprocity." Six Schemas
items loaded on this dimension (a = .80; see Ap-
pendix C). The samples of U.S. and Japanese negotiators had
cultural values similar to those that prior research
has shown to be characteristic of those cultures.
U.S. negotiators (k = 39.45, s.d. = 4.80) were sig-
Analysis nificantly more individualistic than Japanese nego-
The unit of analysis for testing our hypotheses tiators (k = 35.86, s.d. = 6.40, F1, 176 = 17.15, p '
about the relationships between cultural group, .01). Japanese negotiators (k = 23.87, s.d. = 7.39)
cultural values, and negotiation schemas was the were significantly more hierarchical than U.S. ne-
individual. We used analysis of variance (ANOVA) gotiators (k = 21.01, s.d. = 6.58, F1 175 = 7.03, p <
to conduct a sampling check to determine that Jap- .01).
anese participants were less individualistic and These cultural group differences extended to the
more hierarchical than U.S. participants. ANOVA self-interest and power negotiation schemas. U.S.
was also used to test for differences between U.S. negotiators (k = 4.16, s.d. = 0.72) were signifi-
and Japanese negotiators with respect to the self- cantly more likely than Japanese negotiators (k =
interest schema and the power schema. We used 3.53, s.d. = 0.82) to espouse a self-interest negoti-
correlations to test Hypotheses 1 and 2, which are ation schema (F1, 183 = 29.35, p < .01). The Japa-
about the relationships between cultural values nese embraced a power schema with respect to
and negotiation schemas. Chi-squares were used to distributive tactics (kj = 18.86, s.d. = 2.51; XU.s =
test Hypotheses 3 and 4, which concern the relative 15.27, s.d. = 2.66, F1 178 = 7743, p ' .01) mea-
accuracy of inter- and intracultural negotiators' sured in the prenegotiation questionnaire. The Jap-
knowledge about the importance of issues. anese also focused more on power in their prepa-
The unit of analysis for testing the joint gains ration sessions than U.S. negotiators did (k- =
hypothesis (Hypothesis 5) was the dyad. Multivar- 10.69, s.d. = 2.68; U.s. = 9.31, s.d. = 2.98, F1, 172 =
iate analysis of variance (MANOVA) was used to 9.39, p ' .01). As anticipated, U.S. negotiators em-
test differences between intra- and intercultural braced the concept of BATNA as power in the
outcomes. negotiation more strongly than the Japanese (kJ =
To test Hypothesis 6, concerning joint gains and 3.82, s.d. = 0.61; kus 4.26, s.d. = 0.75; F1160
the incompatibility of intercultural negotiators' 10.91, p c .01). However, there were no differences
cultural values and negotiation schemas, we used between U.S. and Japanese negotiators with respect
the dyad as the unit of analysis. To measure incom- to their views as to the importance of status in the
patibility, we did the following: For each variable form of either buyer or seller role in the negotiation
that distinguished between Japanese and U.S. ne- or the status of the company. There were no differ-
gotiators, we divided participants into three groups ences between U.S. and Japanese negotiators with
based on the variable's grand mean in the total respect to schemas for information, whether mea-
sample of U.S. and Japanese participants. The sured prior to the preparation session, as a norm for
groups were (1) more than half a standard deviation information sharing or as importance to the out-
above the grand mean, (2) within half a standard come, or after the negotiation, in terms of what was
deviation of the grand mean, and (3) more than half discussed during the preparation session.
a standard deviation below the grand mean. We
categorized dyads instead of computing difference
Cultural Values and Negotiation Schemas
scores in order to distinguish among dyads with
high compatibility, low compatibility, and moder- Table 2 presents the correlations between cul-
ate compatibility, for which different joint gains tural group, cultural values, and measures of self-
might be anticipated. (For example, one might an- interest, power, and information schemas for ne-
ticipate high joint gains from U.S. dyads that have gotiation. These correlations confirm Hypothesis
high compatibility on self-interest, but not from 1: individualists endorsed self-interest in negoti-
U.S. dyads that have low compatibility on self- ations. The correlations also confirm Hypothesis
interest.) We then ran an ANOVA with six groups 2: negotiators with hierarchical values endorsed
1998 Brett and Okumura 503

TABLE 2
Correlations between Culture and Negotiation Schemas
Variable 1 2 3 4 5 6 7 8 9 10 11

1. Culture
2. Individualism/collectivism -.30**
3. Hierarchy/egalitarianism .20* * .31 * *
4. Self-interest -.37** .22** -.14
5. Distributive tactics .55** -.18* .20** -.13
6. Power planning .22** .06 .23** -.08 .06
7. BATNA a -.25** .11 -.04 .28** .04 .04
8. Company power -.10 .10 .06 .15 .02 .29** .21**
9. Role power .10 -.18* -.08 .11 .14 .09 .10 .30**
10. Information sharing .07 .00 .05 .15* .28** -.06 .31** .03 .04
11. Information planning .05 .15* .04 .04 .01 .21** -.13 -.08 .08 -.06
12. Information power .09 .07 .09 -.04 .02 -.02 .05 .07 -.03 .35** -.05
a Best alternative to a
negotiated agreement.
*p . 05
** p ' .01

distributive tactics and reported spending signif- Information in Inter- and Intracultural
icant time discussing power in the preparation Negotiations
session.
The results suggest that U.S. and Japanese nego-
We used regression analysis to test whether the
relationships between cultural group and sche- tiators may use different scripts for information
mas were mediated by cultural values. These sharing in negotiations. Hypotheses 3 and 4 focus
analyses showed that the cultural value of indi- on perceptions of the importance to the other party
vidualism versus collectivism partially ac- of the integrative issues, runs and financing, and
counted for the relationship between cultural the compatible issue, Strums. The buyers should
group and a self-interest schema. Adjusted R2s have understood that financing was more impor-
for the models containing different combinations tant to the sellers than runs, and the sellers should
of variables were as follows: individualism/col- have understood that runs were more important
lectivism with self-interest, .05; culture with -self- than financing to the buyers. To test these predic-
interest, .13; individualism/collectivism and cul- tions, we first broke the sample into buyers and
ture with self-interest, .17. The cultural value of sellers. Then, we compared U.S. inter- and intra-
hierarchy versus egalitarianism partially ac- cultural negotiators and Japanese inter- and intra-
counted for the relationship between distributive cultural negotiators, so as to be able to identify
tactics and cultural group, with R2s as follows: asymmetries in information exchange if such oc-
hierarchy/egalitarianism with distributive tac- curred. The pattern proposed in Hypothesis 3 ap-
tics, .04; culture with distributive tactics, .30; peared clearly and significantly with respect to
hierarchy/egalitarianism and culture, .30. Hierar- U.S. sellers. Forty-one percent of the U.S. sellers
chy versus egalitarianism also partially ac- engaged in intercultural negotiations ranked runs
counted for the relationship between power as the first or second most important issue to the
planning and cultural group: hierarchy/egalitari- buyer, compared to 74 percent of the U.S. sellers in
anism with power planning, .05; culture with intracultural negotiations (X21 = 5.36, p ' .02). A
power planning, .04; hierarchy/egalitarianism result not supporting the asymmetric prediction of
and culture with power planning, .08. All of Hypothesis 4 was that Japanese sellers understood
these relationships were significant, indicating the importance of runs to the buyer whether they
that at least some of the relationship between were negotiating intraculturally with Japanese buy-
cultural group and negotiation schema could be ers or interculturally with U.S. buyers. These data
accounted for by cultural values. However, in all indicate a clear asymmetry in the information con-
cases, the cultural group-negotiation schema re- veyed from Japanese buyer to U.S. seller in the
lationship was substantially greater than what intercultural negotiations. Additional significant
could be accounted for by a cultural value, sug- evidence supporting Hypothesis 3 is the finding
gesting that negotiation schemas are likely to be that 41 percent of the U.S. sellers in intercultural
multidetermined. negotiations ranked financing as the issue with
504 Academy of Management Journal October

lower value to buyers (ranked 3-5), compared with intercultural negotiation failed to understand the
72 percent of the U.S. sellers in intracultural nego- value of Strums, confirming both Hypothesis 3 and
tiations (X21= 4.74, p ' .03). Again, there were no Hypothesis 4. Sixty-three percent of the U.S. inter-
significant differences in the proportions of Japa- cultural negotiators, compared to 88 percent of the
nese intra- and intercultural sellers who ranked U.S. intracultural negotiators, ranked Strums as
financing as of low value to buyers. one of the top four issues in the negotiation (X21=
As predicted by Hypothesis 3, U.S. buyers in 8.69, p c .01). Likewise, 53 percent of the Japanese
intercultural negotiations had a tendency to mis- intercultural negotiators, compared with 92 per-
judge the importance of runs to the Japanese sell- cent of the Japanese intracultural negotiators,
ers. Forty-three percent of the U.S. buyers con- ranked Strums as one of the top four issues in the
sidered the issue ranked first or second by the negotiation (X21 = 10.55, p c .01). The proportions
sellers, compared with 33 percent of the U.S. of U.S. and Japanese intercultural negotiators rank-
buyers in intracultural negotiations. This pattern ing Strums as one of the top four issues in the
was in the expected direction but was not signif- negotiation were not significantly different. This
icant. However, U.S. buyers in intercultural ne- pattern of data suggests that, with regard to Strums,
gotiations were as aware of the importance of there does not seem to be an asymmetry in percep-
financing to sellers as U.S. buyers in intracultural tion between intercultural U.S. and Japanese nego-
negotiations. tiators, simply too little understanding of the im-
The pattern of findings with respect to the inte- portance of this issue.
grative issues is consistent with the prediction of
Hypothesis 3 but inconsistent with the prediction
Intercultural Negotiation and Joint Gains
of Hypothesis 4. Full information seemed to go
from the U.S. negotiator to the Japanese negotiator, Intercultural U.S.-Japanese negotiations resulted
but less information went in the opposite direction. in lower joint gains than intracultural U.S. and
U.S. negotiators were learning less about integra- Japanese negotiations, confirming Hypothesis 5.
tive issues in inter- than in intracultural negotia- Table 3 shows that intercultural dyads averaged
tions. Japanese negotiators were learning as much about $3.47 million in joint gains, compared with
about the integrative issues in inter- as in intracul- $4.32 million for intracultural negotiators (F1 93
tural negotiations. - 17.61, p ' .01). There were no significant
There were no differences in the proportions of differences in joint gains between the U.S.-U.S.
U.S. buyers and sellers in intracultural negotiations intracultural negotiators, who averaged $4.33
ranking runs and financing high and low and in the million in joint gains (s.d. = $0.81 million), and
proportions of Japanese buyers and sellers in in- the Japanese-Japanese intracultural negotiators,
tracultural negotiations making these rankings. who averaged $4.29 million in joint gains (s.d. =
This pattern suggests that information exchange $0.80 million). Controlling for age showed the
was similar in U.S. and Japanese intracultural ne- following: age, which was correlated with joint
gotiations. gains (r = -.22, p ' .05), accounted for 5 percent
However, a different pattern emerged with re- of the variance in joint gains; type of negotiation
spect to the compatible issue-the second cartoon, (inter versus intra) accounted for 16 percent of
Strums. Both U.S. and Japanese negotiators in the the variance in joint gains; and together, age and

TABLE 3
Results of Inter- and Intracultural Negotiationsa
Intercultural Intracultural

Issue Mean s.d. Mean s.d. F1,93 p

Joint gainsb $3.47 $1.14 $4.32 $0.8 17.61 .01

Price $49,042 $6,957 $51,685 $9,197 n.s.
Runs 6.47 1.25 7.47 0.81 20.92 .01
Financing 0.73 0.45 0.68 0.47 n.s.
Strums 0.57 0.50 0.85 0.36 9.42 .01
Contingent contract 0.17 0.38 0.12 0.33 n.s.

a Multivariate F589 4.6, p - .01. For the intercultural negotiations, n 30; for the intracultural negotiations, n = 65.
b
Values are millions of dollars.
1998 Brett and Okumura 505

type of negotiation accounted for 16 percent of DISCUSSION

the variance in joint gains. Thus, age contributed
The results of this study suggest that intercul-
no unique variance to the prediction of joint
tural negotiations between U.S. and Japanese
gains, and it was not correlated with joint gains managers are less likely to realize joint gains than
within the inter- or the intracultural group. Our intracultural negotiations between Japanese man-
cooperation and frustration scales, measured af- agers or intracultural negotiations between U.S.
ter the negotiation, were used to control for stron- managers. There are several explanations for this
ger in-group/out-group effects in intercultural finding that are associated with differences in cul-
than intracultural negotiations. These scales tural values and the schemas of U.S. and Japanese
were not related to type of negotiation (inter or negotiators.
intra) or to joint gains.
There were several ways to increase joint gains in
"Cartoon." Negotiators could make trade-offs be- Information
tween financing and the number of runs per epi-
The information-processing approach to negotia-
sode. Parties could increase their joint gains by
tions articulated by theorists (Bazerman & Neale,
including Strums in the agreement. The MANOVA
1991; Fisher et al., 1991; Lax & Sebenius, 1986) and
analysis reported in the lower portion of Table 3 demonstrated by empiricists (Neale & Bazerman,
indicates that the difference in joint gains between 1991; Pinkley et al., 1995; Pruitt, 1981; Weingart et
intercultural and intracultural dyads was due to al., 1990) to generate joint gains is not, according to
intracultural dyads' negotiating more runs for the this research, uniquely normative in the U.S. cul-
buyer and intracultural dyads' greater likelihood of ture. Both Japanese and U.S. managers ascribed to a
including the purchase of Strums in the package. schema for information sharing. However, the data
(Multivariate F5, 89 = 4.60, p ' .01.) The optimal suggest they were less able to enact information-
trade-off between runs and financing was eight sharing scripts in intercultural negotiations than in
runs and zero financing. Our simple dichotomous intracultural ones. The data suggest that in the case
measure of whether or not there was financing in- of the compatible issue, including the cartoon
dicated no significant difference between groups. Strums in the deal, many intercultural negotiators
Although there may have been some differences simply did not realize the issue's value to the other
between inter- and intracultural groups in the side. Despite its obvious value to their own side,
amount of financing, which would be reflected in intercultural negotiators did not pursue Strums to
the overall difference in joint gains, neither the the extent that intracultural negotiators did. In de-
inter- nor the intracultural negotiators found the briefing, we were given a variety of responses to the
optimal trade of no financing with eight runs very question, "Why didn't you include Strums?", in-
often. There were also no differences between in- cluding this answer from a Japanese participant: "I
ter- and intracultural groups in the price paid for saw this as the first negotiation in a potential long
the cartoon or in the inclusion of a contingency term-relationship. I left Strums for the future." Sev-
clause. eral U.S. participants said, "I raised it and he didn't
Hypothesis 6 proposes that when intercultural seem interested, so I dropped it." Other U.S. and
Japanese participants admitted never even discuss-
negotiators' schemas are incompatible, low joint
ing Strums.
gains will ensue. The results of ANOVAs testing
Negotiators who create less value than others
this hypothesis were not significant. There were
typically have less understanding of the other par-
numerous intercultural dyads with incompatible
ty's priorities (Thompson, 1991). Trends in the data
negotiation schemas. However, their joint gains
suggest that, with the exception of buyers with
were not significantly lower than, for example, respect to financing, either the Japanese negotiators
those of the intercultural dyads whose negotiation engaged in intercultural negotiations were not com-
schemas were compatible. There are several possi- municating their priorities, or their U.S. partners
ble explanations for this finding. First, recall that were not listening. In contrast, the Japanese nego-
the intercultural dyads negotiated lower joint gains tiators in intercultural negotiations were able to
than the intracultural dyads. Thus, within-group judge their partners' priorities as well as the U.S.
variance was restricted. Second, with only 30 in- negotiators in intracultural negotiations. However,
tracultural dyads and six types of compatibility, the this knowledge did not seem to translate into joint
number of dyads within a type was small, and gains equivalent to those realized in either U.S. or
within-cell variance on the dependent variable was Japanese intracultural negotiations. These findings
often large. are conlsistent with the expectation that Japanese
506 Academy of Management Joural October

and U.S. negotiators will use different scripts to norms of distributive tactics more strongly than the
communicate in negotiations and suggest a major U.S. negotiators. It is not totally clear from the data
communications hurdle for intercultural negotia- exactly what the Japanese negotiators viewed as the
tors. basis of the power. We asked four questions in the
Another information-sharing hurdle in intercul- prenegotiation questionnaire about factors contrib-
tural negotiations was premature closure and too uting to outcomes in negotiations, expecting the
narrow a focus. Strums, the compatible issue, was Japanese negotiators to indicate company and role
more likely to be ignored in intercultural than in- (buyer versus seller) more strongly, as suggested by
tracultural negotiations. Since Strums was in- Graham's research (1983, 1993), and the U.S. nego-
cluded equally often in the Japanese and U.S. in- tiators to indicate alternatives, or BATNA, more
tracultural negotiations, culture per se cannot be an strongly. The expected difference between the U.S.
explanation. It appears that it is the interaction of and Japanese negotiators in their perceptions of the
two cultures at the negotiating table that results in importance of alternatives was significant. There
premature closure and too narrow a focus. It is was also a tendency for Japanese negotiators to rate
possible that negotiating in a second language and role as a more important factor than the U.S. nego-
with someone using a second language may have tiators. There was a reasonable amount of variance
contributed to premature closure. However, the in- on this item among the Japanese, some of whom
tercultural negotiators were not time constrained. told us in the postnegotiation discussion that the
Many finished before time was up. All the Japanese norm of deference to buyers had fallen in the face of
negotiators in the intercultural program were fluent open markets. Of course, these participants were
English speakers, and neither the Japanese nor the expatriates who generally produced and/or sold
U.S. intercultural negotiators expressed signifi- Japanese products in the United States. It is impor-
cantly more frustration than the intracultural nego- tant for future research to collect data on bases of
tiators. power from Japanese negotiators in Japan (which
Consistent with our interpretation of premature we were unable to do in this study), as well as from
closure and too narrow a focus in interpreting the Japanese negotiators in the U.S.
results regarding Strums is the evidence regarding Nevertheless, the data on power suggest that the
the use of information about the integrative issue, Japanese may not value BATNA as power in the
financing. Similar proportions of buyers under- same way as U.S. negotiators and that, in intercul-
stood the importance of financing to the sellers in tural negotiations, this difference may contribute to
the intercultural and intracultural negotiations. a low level of joint gains. In postnegotiation discus-
However, the intercultural buyers were unable to sions, we were told that because the Japanese have
use that information to create gains. The financing such a long history of having to sell below cost in
issue was somewhat more complex than the runs intercultural negotiations, BATNA was a point to
issue in that it required more math to see how try to reach, not a point from which to begin nego-
different alternatives would work out for each tiations. This suggests that BATNA served as a low
party. Thus, to succeed in creating joint gains with anchor for some Japanese negotiating intercultur-
the financing issue required some commitment to ally and therefore contributed to premature closure
heuristic trial-and-error evaluation of alternatives. of discussion of options and the relatively low level
All participants had calculators, and the math was of joint gains in the intercultural negotiations as
not difficult. But although equal proportions of compared with the intracultural negotiations.
buyers in the inter- and intracultural negotiations
knew of the importance of financing to the seller,
Self-Interest
and of course all the sellers knew how important
financing was to themselves, the intercultural ne- Another factor in the intercultural negotiations
gotiators were not as successful as the intracultural between U.S. and Japanese negotiators that appears
negotiators in creating these gains. to have affected the level of joint gains was the
difference in the two cultures' focus on self-inter-
est. The U.S. negotiators were much more individ-
Power
ualistic and focused on self-interest than the Japa-
Power was particularly important to the Japanese nese negotiators. Two individualistic, self-
negotiators, as we predicted on the basis of their interested negotiators may be able to stimulate each
expected hierarchical cultural values. The Japanese other to achieve high goals. This conclusion is con-
negotiators indicated paying significantly more at- sistent with research on the effects of goals in ne-
tention to power in their negotiation preparation gotiation (Huber & Neale, 1986, 1987; Pinkley,
than did the U.S. negotiators. They also ascribed to Neale, & Bennett, 1994). Negotiators with easy or
1998 Brett and Okumura 507

no goals are more likely to reach compromise agree- in all intercultural negotiations. Third, it suggests
ments than integrative ones (Huber & Neale, 1987). that incompatibilities in negotiation scripts and
However, when self-interest is mismatched, one schemas interfere with the creation of joint gains in
negotiator's goals are more easily met than the oth- intercultural negotiations between U.S. and Japa-
er's. The first negotiator may not be motivated to nese managers. Although we did not find direct
continue the negotiation once her or his goals have evidence that incompatibility of negotiation sche-
been met, and so the dyad reaches premature clo- mas and scripts resulted in lower joint gains in
sure. intercultural negotiations, we believe that this hy-
pothesis should be tested in future research with
larger samples. In addition, it seems likely that the
Contributions to Theory
fundamental approaches of U.S. and Japanese ne-
In this study, we suggested that cultures differ gotiators differ in ways that were not reflected in
with respect to schemas and scripts for negotiating our analysis of the incompatibility of scripts and
behavior. Further, we suggest that when negotia- schemas. The Japanese negotiators' relative empha-
tors' scripts and schemas are incompatible, it may sis on power and indirect information sharing is
be difficult for them to negotiate joint gains. We very different from the U.S. negotiators' eschewal
developed this theoretical perspective by articulat- of power and focus on direct information sharing.
ing the implications for negotiations of two cultural In this respect, the study contributes to the theoret-
values: individualism versus collectivism and hier- ical understanding of negotiation scripts that are
archy versus egalitarianism. We proposed that characteristic of U.S. and Japanese negotiators.
when negotiators from cultures differing on these
values meet, their opportunities for joint gains may
Limitations of the Study and Avenues for Further
be limited by informational disutilities, including
Research
inefficient information sharing and premature clo-
sure of the search for alternatives. Our empirical The results demonstrate a set of associations be-
data suggest that, in addition to experiencing these tween cultural group, cultural values, and mea-
two disutilities, intercultural negotiators who have sures of negotiation schemas. They also demon-
information may be unable to use it effectively to strate that intercultural negotiations between U.S.
generate joint gains. Within the context of the in- and Japanese negotiators are less likely to generate
tercultural negotiation studied here, U.S. and Jap- joint gains than intracultural negotiations in either
anese negotiators held different views of power in culture. A series of questions concerning the gen-
negotiation. When negotiators cannot agree on who eralizability and validity of these findings merits
is more powerful, it is extremely difficult for them further discussion and research.
to reach agreement (Ury, Brett, & Goldberg, 1988). Will the intercultural results generalize to Japa-
When they don't even agree as to what constitutes nese-U.S. intercultural negotiations when Japanese
power, they may be unable to use power as a basis negotiators have had little experience working with
for reaching agreement. Westerners? Since the cultural values and norma-
Finally, in the U.S.-Japanese intercultural nego- tive negotiation schemas of the intercultural Japa-
tiation context, it appears that the clash of individ- nese negotiators were not significantly different
ualism and hierarchy may be particularly dysfunc- from those of the intracultural Japanese negotiators,
tional when social interaction requires negotiation. we anticipate equal or greater difficulty in negoti-
The negotiation scripts or patterns of behavior ating joint gains in intercultural negotiations in
stimulated by concern for self and concern for so- which the Japanese negotiators have less Western
cial power are fundamentally different, even experience than those in our intercultural sample.
though both represent distributive negotiation Will intercultural negotiations always be less
schemas. However, the negotiation scripts gener- likely to generate joint gains than intracultural ne-
ated by hierarchical and individualistic values do gotiations? We think this will depend on the fit of
not provide a common ground on which intercul- the cultures, and determining such fit in turn re-
tural negotiators can meet. quires an analysis of cultural differences in negoti-
This study contributes to theory at three levels. ation schemas and scripts. Different scripts for the
First, it provides a model for developing culture- same schema need not always be incompatible. For
specific hypotheses about intercultural negotia- example, altruists and cooperators from different
tions. Second, it identifies basic elements of nego- cultures can work together quite nicely (Messick,
tiation-information sharing, premature closure of 1991). However, this study suggests that conver-
search, information utilization, and the meaning gence with one or the other culture's focus may be
and use of power-that are potentially problematic necessary if high joint gains are to be negotiated.
508 Academy of Management Journal October

Our findings in this area are necessarily tentative tives on international industriallorganizational psy-
because of limitations owing to sample size, and chology: 640-675. San Francisco: Jossey-Bass.
they merit further empirical investigation and the- Lewis, J. D. 1990. Partnerships for profit. New York:
oretical reflection. Schemas and scripts appropriate Free Press.
for intracultural negotiations but inappropriate for Lytle, A. M., Brett, J. M., Barsness, Z. I., Tinsley, C. H., &
intercultural negotiations may be impossible to Janssens, M. 1995. A paradigm for confirmatory
turn on and off at will. If effective intercultural cross-cultural research in organizational behavior. In
negotiations require a common negotiation script, L. L. Cummings & B. M. Staw (Eds.), Research in
how should negotiators determine whose script organizational behavior, vol. 17: 167-214. Green-
should prevail? wich, CT: JAI Press.
March, R. M. 1990. The Japanese negotiator. New York:
Kodansha International.
REFERENCES
Marcus, H. R., & Kitayama, S. 1991. Culture and the self:
Bazerman, M. H., & Neale, M. A. 1992. Negotiating ra- Implications for cognition, emotion, and motivation.
tionally. New York: Free Press. Psychological Review, 98: 224-253.
Bettenhausen, K., & Murnighan, J. K. 1985. The emer- Messick, D. 1991. On the evolution of group based altru-
gence of norms in competitive decision-making ism. In R. Saltine (Ed.), Game equilibrium models I:
groups. Administrative Science Quarterly, 30: 350- Evolution and game dynamics: 304-328. Berlin:
372. Springer-Verlag.
Drake, L. E., & Donohue, W. A. 1994. Issue development Mifflin, L. 1995. Can the Flintstones fly in Figi? New
as negotiated order in conflict. Paper presented to York Times, November 27: Dl.
the annual meeting of the International Association
Neale, M. A., & Bazerman, M. H. 1991. Negotiator cog-
of Conflict Management, Eugene, OR.
nition and rationality. New York: Free Press.
Fisher, R., Ury, W., & Patton, B. 1991. Getting to yes:
Olekalns, M., & Smith, P. L. 1996. Strategic sequences
Negotiating agreement without giving in. New
and outcome optimality in negotiation. Paper pre-
York: Penguin Books.
sented at the annual meeting of the Academy of
Fiske, S. T., & Taylor, S. E. 1984. Social cognition. Read- Management, Boston.
ing, MA: Addison-Wesley.
Olekalns, M., Smith, P. L., & Walsh, T. 1996. The process
Graham, J. L. 1983. Business negotiations in Japan, Bra- of negotiating: Strategies, timing and outcomes. Or-
zil, and the United States. Journal of International ganizational Behavior and Human Decision Pro-
Business Studies, 14(1): 47-62. cesses, 68: 68-77.
Graham, J. L. 1993. The Japanese negotiation style: Char- Ohbuchi, K., & Takahashi, Y. 1994. Cultural styles of con-
acteristics of a distinct approach. Negotiation Jour- flict management in Japanese and Americans: Passiv-
nal, 9: 123-140. ity, covertness, and effectiveness of strategies. Journal
Graham, J. L., & Sano, Y. 1989. Smart bargaining: Doing of Applied Social Psychology, 24: 1345-1366.
business with the Japanese. Cambridge, MA: Ball- Pinkley, R. L. 1990. Dimensions of conflict frame: Dispu-
inger. tant interpretations of conflict. Journal of Applied
Hofstede, G. 1980. Culture's consequences: Interna- Psychology, 75: 117-126.
tional differences in work-related values. New- Pinkley, R. L., Griffith, T. L., & Northcraft, G. B. 1995.
bury Park, CA: Sage. "Fixed pie" a la mode: Information availability, in-
Huber, V. L., & Neale, M. A. 1986. Effects of cognitive formation processing, and the negotiation of subop-
heuristics and goals on negotiator performance and timal agreements. Organizational Behavior and
subsequent goal setting. Organizational Behavior Human Decision Processes, 62: 101-112.
and Human Decision Processes, 38: 342-365. Pinkley, R. L., Neale, M. A., & Bennett, R. J. 1994. The
Huber, V. L., & Neale, M. A. 1987. Effects of self and impact of alternative to settlement in dyadic negoti-
competitor goals on performance in an interdepen- ation. Organizational Behavior and Human Deci-
dent bargaining task. Journal of Applied Psychol- sion Processes, 57: 97-116.
ogy, 72: 197-203. Pinkley, R., & Northcraft, G. B. 1994. Conflict frames of
Kato, H., & Kato, J. S. 1992. Understanding and working reference: Implications for dispute processes and
with the Japanese business world. Englewood outcomes. Academy of Management Journal, 37:
Cliffs, NJ: Prentice-Hall. 193-205.
Lax, D. A., & Sebenius, J. K. 1986. The manager as ne- Pruitt, D. G. 1981. Negotiation behavior. New York: Ac-
gotiator. New York: Free Press. ademic Press.
Leung, K. 1997. Negotiation and reward associations across Pruitt, D. G., & Carnevale, P. J. 1993. Negotiation in so-
cultures. In P. C. Earley &M. Erez (Eds.), Newperspec- cial conflict. Pacific Grove, CA: Brooks/Cole.
1998 Brett and Okumura 509

Pruitt, D. G., & Lewis, S. 1975. Development of integra- APPENDIX A

tive solutions in bilateral negotiation. Journal of Example for the Buyer: Net Value Calculation
Personality and Social Psychology, 31: 621-633.
Assume that you have reached a tentative agreement
Putnam, L., & Holmer, M. 1992. Framing, reframing, and
for the purchase of Ultra Rangers that involves the fol-
issue development. In L. Putnam & M. Roloff (Eds.),
lowing terms:
Communication and negotiation: 128-155. New-
bury Park, CA: Sage. * purchase price: $60,000 per episode
* runs/episode: 7
Raiffa, H. 1982. The art and science of negotiation.
* financial terms:
Cambridge, MA: Harvard University Press.
YR 0 80%
Schwartz, S. H. 1994. Beyond individualism/collectiv- YR 1 10%
ism: New cultural dimensions of values. In U. Kim, YR 2 10%
H. C. Triandis, & G. Yoon (Eds.), Individualism and The calculation of the net value of this agreement is
collectivism: 85-117. London: Sage. described below.
Shank, R.C., & Abelson, R. P. 1977. Scripts, plans, goals
1. Expected Revenue from the Cartoon $8,400,000
and understanding: An inquiry into human knowl-
edge structures. Hillsdale, NJ: Earlbaum. Runs/Episode Adjustment $840,000
Net Value of the Cartoon $9,240,000
Tenbrunsel, A. E., & Bazerman. M. H. 1995. Working
2. Price of the Cartoon $6,000,000
women. In J. M. Brett (Ed.), Dispute Resolution Re-
Price/Episode x 100 episodes
search Center teaching manual materials. Evan- $60,000 X 100 = $6,000,000
ston, IL: Northwestern University.
3. Payment Savings $ 180,000
Thompson, L. 1991. Information exchange in negotia-
tion. Journal of Experimental Social Psychology, YR Payment Savings (%) Savings ($)
27: 161-179.
Thompson, L. 1997. The mind and the heart of the 0 .80 * $6M = $4,800,000 0% $ 0
negotiator. Upper Saddle River, NJ: Prentice-Hall. 1 .10 * $6M = $ 600,000 10% $ 60,000
2 .10 * $6M = $ 600,000 20% $120,000
Thompson, L., & Hastie, R. 1990. Social perception in
negotiation. Organizational Behavior and Human Total $180,000
Decision Processes, 47: 98-123.
4. Net Price of the Cartoon $5,820,000
Ting-Toomey, S. 1988. Intercultural conflict styles: A
Price of the Cartoon - Payment Savings
face-negotiation theory. In Y. Kim & W. Gudykunst
$6,000,000 - $180,000 = $5,820,000
(Eds.), Theories in intercultural communication:
213-235. Newbury Park, CA: Sage. 5. Other Terms of the Agreement $ 0

Triandis, H. C. 1972. The analysis of subjective culture. 6. Net Profit of the Cartoon $3,420,000
Net Value - Net Price + Other Terms
New York: Wiley.
$9,240,000 - $5,820,000 + $0 = $3,420,000
Triandis, H. C. 1989. Cross-cultural studies of individu- 7. Value of the Alternative Deal $3,000,000
alism and collectivism. In J. Berman (Ed.), Nebraska
8. NET VALUE OF THE BARGAINING $ 420,000
symposium on motivation: 41-133. Lincoln: Univer-
AGREEMENT
sity of Nebraska Press. Net profit - Alt. Deal
Tutzauer, F., & Roloff, M. E. 1988. Communication pro- $3,420,000 - $3,000,000 = $420,000
cesses leading to integrative agreements: Three paths to
jointbenefits. CommunicationResearch, 15: 360-380.
APPENDIX B
Ury, W. L., Brett, J. M., & Goldberg, S. B. 1988. Getting
disputes resolved. San Francisco: Jossey-Bass. Example for the Seller: Net Value Calculation
Weingart, L. R., Bennett, R. J., & Brett, J. M. 1993. The Assume that you have reached an agreement on the
impact of consideration of issues and motivational sale of Ultra Rangers that specifies the following terms:
orientation on group negotiation process and out- * purchase price: $50,000
come. Journal of Applied Psychology, 78: 504-517.
* runs/episode: 5
Weingart, L. R., Thompson, L. L., Bazerman, M. H., & * financial terms:
Carroll, J. S. 1990. Tactical behavior and negotiation YR 0 50%
outcomes. International Journal of Conflict Man- YR 1 30%
agement, 1: 7-31. YR 2 20%

Yoshimura, N. 1997. Interview with Noboru Yoshimura. The calculation of the net value of this agreement is
New York Times Magazine, June 8: 68. described below.
 510 Academy of Management joural October

1. Program Revenue $5,000,000 Power Analysis

Fee/Episode X 100 = $50,000 X 100 = $5,000,000 During the preparation session, how much did you think
2. Financing Cost $ 650,000 about ...a
How advantageous your negotiating position was
YR Payment Cost (%) Cost ($) How powerful your company was
How advantageous the other party's position was
0 .5 * $5,000,000 = $2,500,000 0% $ 0 How powerful the other party's company was
1 .3 * $5,000,000 = $1,500,000 20% $300,000 Frustration
2 .2 * $5,000,000 = $1,000,000 35% $350,000
How strongly do you agree or disagree?
Total $650,000
During the negotiation ...
3. Program Revenue less Financing Cost $4,350,000 Silence stimulated concessions
I felt frustrated
$5,000,000 - $650,000 = $4,350,000
I was often silent in the negotiation
4. Runs/Episode Adjustment $ 250,000 I wished I had the services of a mediator or third party
5. Other Pieces of Agreement $ 0 I wished I could consult with my preparation partner
I was confused about what the other party wanted
6. Net Revenue $4,600,000
$4,350,000 + $250,000 + $0 = $4,600,000
I had to make too many concessions
The other party was often silent
7. Value of the Alternative Deal $2,500,000
Cooperation/Reciprocity
8. NET VALUE OF THE BARGAINING $2,100,000
AGREEMENT How strongly do you agree or disagree?
Net Revenue - Value of the Alternative Deal During the negotiation ...
$4,600,000 - $2,500,000 = $2,100,000
The other party reciprocated information (offered in-
formation in return for information)
The other party reciprocated offers (made a counter
offer in return for an offer)
The other party cooperated
APPENDIX C
There was a lot of joint problem solving
Questionnaire Items I trusted the other party
I had to do all the work in the negotiation (reverse-
Information Planning coded)
During the preparation session, how much did you think a
Possible responses were "a lot," "some," "a little,"
about ... and "not much at all."
What information you would like to get from the other
Jeanne M. Brett (Ph.D., University of Illinois) is the
party DeWitt W. Buchanan, Jr., Professor of Dispute Resolution
What information not to share with the other party and Organizations and the director of the Dispute Reso-
What information to share with the other party lution Research Center at the J. L. Kellogg Graduate
What your priorities were School of Management, Northwestern University. She is
How the other party might act in the negotiation currently studying work and family issues and intercul-
Setting an agenda for the negotiation tural negotiation processes.
Opening offer
Tetsushi Okumura is an associate professor of manage-
What arguments you might use to support your ment and negotiation in the Faculty of Economics at
position Shiga University and a vice director of the Japan Institute
What you might say to defend your position of Negotiation. He finished a Ph.D. program in manage-
The range of settlements you would accept ment at Waseda University. His current research focuses
Multiple options for settlement on cross-cultural negotiations, negotiated social orders in
The order in which you would make concessions and among organizations, and strategic corporate alliances.

View publication stats